<%BANNER%>

Ecological Effects of Roads: Theory, Analysis, Management, and Planning Considerations

Permanent Link: http://ufdc.ufl.edu/UFE0002308/00001

Material Information

Title: Ecological Effects of Roads: Theory, Analysis, Management, and Planning Considerations
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0002308:00001

Permanent Link: http://ufdc.ufl.edu/UFE0002308/00001

Material Information

Title: Ecological Effects of Roads: Theory, Analysis, Management, and Planning Considerations
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0002308:00001


This item has the following downloads:


Full Text

PAGE 1

ECOLOGICAL EFFECTS OF ROADS: THEORY, ANALYSIS, MANAGEMENT, AND PLANNING CONSIDERATIONS By DANIEL JOSEPH SMITH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Daniel Joseph Smith

PAGE 3

ACKNOWLEDGMENTS I extend my greatest appreciation and honor to three mentors who played principal roles in the development of my career in ecology: Larry Harris, Gary Evink, and Richard Forman. Their knowledge of and passion for the subject gave me the inspiration to pursue and complete my goals. Most importantly, I would like to thank my parents for their love and encouragement during the difficult times. Their unwavering faith in my abilities allowed me to keep moving forward. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................................................................................. iii LIST OF TABLES ............................................................................................................ vii LIST OF FIGURES ........................................................................................................... ix ABSTRACT ....................................................................................................................... xv CHAPTER 1 LAND DEVELOPMENT IN FLORIDA AND EFFORTS TO REDUCE ROAD-RELATED WILDLIFE MORTALITY ........................................................... 1 Infrastructure and Population Growth in Florida ..................................................... 4 Origin of Road Networks and Opportunities for Development ......................... 4 Recent Population Growth and Infrastructure Needs ........................................ 5 Highway Construction ....................................................................................... 6 Land-Use Change ............................................................................................... 8 Projects Designed to Increase Permeability of Roads in Florida ........................... 10 The Everglades, Alligator Alley, and the Florida Panther ............................... 10 Habitat Connectivity in Central Florida for the Florida Black Bear ................ 13 Road Mortality of Herpetofauna in Payne's Prairie State Preserve ................. 18 2 PRIORITIZING ECOLOGICAL HOTSPOTS ON FLORIDA HIGHWAYS .......... 23 Introduction ............................................................................................................ 23 Methods .................................................................................................................. 23 Configuration of Data ...................................................................................... 28 Modeling Process ............................................................................................. 37 Results .................................................................................................................... 40 Discussion .............................................................................................................. 59 Determinants of Model Priorities ..................................................................... 59 Data Limitations and Development of other Criteria ...................................... 64 Model Priorities ............................................................................................... 65 Conclusion ............................................................................................................. 68 iv

PAGE 5

3 EXISTING INFRASTRUCTURE IN FLORIDA ...................................................... 70 Introduction ............................................................................................................ 70 Parameters for Field-Site Evaluation and Infrastructure Inventory ................. 70 Determining Appropriate Measures for Mitigation ......................................... 71 Site Inventories, Assessments, and Recommendations ......................................... 72 Measures of Model Accuracy, Site Inventory and Ecological Characterization .............................................................................................. 72 Recommended Actions to Improve Functional Habitat Connectivity: Retrofitting Highway Hotspots ...................................................................... 96 Value of Existing Infrastructure ........................................................................... 114 Determining Bridge-Replacement Costs ....................................................... 114 Infrastructure Investments for Wildlife ......................................................... 120 Statewide Bridge Expansion Program ................................................................. 124 4 MONITORING WILDLIFE USE AND DETERMINING STANDARDS FOR CULVERT DESIGN ................................................................................................ 128 Introduction .......................................................................................................... 128 Study Description and Design ....................................................................... 128 Methods and Study Area ...................................................................................... 131 Results .................................................................................................................. 137 Assumptions for Multivariate Logistic Regression Analysis ........................ 137 General Findings ............................................................................................ 142 Findings by Faunal Groups ............................................................................ 147 Discussion ............................................................................................................ 179 Comparison of Structure Use among Faunal Groups .................................... 180 Crossing Structures as Primary Habitat Features ........................................... 188 Road-kills and Culvert Avoidance ................................................................. 189 Effects of Light Penetration and Traffic Noise .............................................. 190 Current Research Benefits and Future Directions ......................................... 191 5 INTEGRATING LANDSCAPE ECOLOGY THEORY WITH TRANSPORTATION PLANNING AND POLICY ................................................ 192 Theoretical and Conceptual Basis for Infrastructure Ecology ............................. 192 Evolution and Biogeographic Principles ....................................................... 194 Landscape Ecology ........................................................................................ 199 Growth and Roads ................................................................................................ 204 Negative Ecological Consequences of Roads ................................................ 206 Road Impacts and Planning in Florida ........................................................... 209 Biodiversity Conservation and Transportation Planning and Policy ................... 209 Planning for Habitat Connectivity ................................................................. 210 Transportation-related Environmental Policies ............................................. 213 Considerations for Site Design that Enhance Use By Wildlife ..................... 217 Social Values, Economic Priorities and Future Research Needs ......................... 222 v

PAGE 6

What is the Economic Equation for Success? ................................................ 223 Land Ownership, Conservation Easements, and Highway-Greenway Interface Zones ............................................................................................. 225 Future Planning and Research to Integrate Transportation and Conservation Objectives in Florida .............................................................. 227 6 LITERATURE REVIEW ......................................................................................... 229 Ecological Effects of Roads ................................................................................. 229 Vehicle-Wildlife Collisions ........................................................................... 230 Habitat Loss, Fragmentation, and Isolation ................................................... 232 Density of Roads ............................................................................................ 234 Negative Edge Effects .................................................................................... 235 Alteration of Ecological Processes and Changes in Community Composition ................................................................................................. 237 Dispersal of Toxic Substances and Environmental Pollution ........................ 240 Floodplain and Hydrologic Alterations, Effects of Erosion, and Chemical Runoff .......................................................................................... 242 Effects of Increased Human Access .............................................................. 244 APPENDIX A QUESTIONNAIRE REGARDING WILDLIFE MORTALITY ON ROADS ....... 249 B INTERACTIVE MODEL FOR IDENTIFYING AND PRIORITIZING ECOLOGICAL HOTSPOTS ON HIGHWAYS ...................................................... 269 C RECOMMENDED CORRECTIVE ACTIONS AT SPECIFIC SITES IN FLORIDA TO IMPROVE PERMEABILITY OF ROADS FOR WILDLIFE ........ 296 REFERENCES ................................................................................................................ 319 BIOGRAPHICAL SKETCH ........................................................................................... 346 vi

PAGE 7

LIST OF TABLES Table page 2-1. Grid values for the Florida Department of Transportation (FDOT) priority model ......................................................................................................... 27 2-2. Number of contiguous prioritized road segments by FDOT district ...................... 40 2-3. Significant prioritized road segments by FDOT district ........................................ 48 2-4. Road projects identified at highly ranked, ecological hotspots on highways ......... 57 3-1. Typical bridge replacement construction costs .................................................... 117 3-2. Old bridge dimensions and costs for extension .................................................... 119 3-3. Summary of existing bridges evaluated for corrective measures ......................... 121 3-4. Summary of existing bridges on priority list evaluated for corrective measures ............................................................................................ 123 3-5. Scheduled bridge replacement construction projects, 1999-2004 ........................ 125 4-1. Faunal orders predicted to use culverts ................................................................ 130 4-2. Mammal body weight categories .......................................................................... 131 4-3. Land-cover classes ................................................................................................ 132 4-4. Replicates found for each structure type by land-cover class .............................. 139 4-5. Univariate measures and distributions of explanatory variables .......................... 140 4-6. Events, nonevents, and road-kills recorded at culverts and bridges ..................... 143 4-7. Faunal groups ....................................................................................................... 148 4-8. Species recorded and group assignments ............................................................. 149 4-9. Levels of use by faunal groups for common parameters of all study sites ........... 182 vii

PAGE 8

4-10. Levels of use by faunal groups for structural parameters of culverts ................... 185 4-11. Levels of use by faunal groups for structural parameters of bridges .................... 188 6-1. Ecological effects of roads ................................................................................... 230 viii

PAGE 9

LIST OF FIGURES Figure page 1-1. Florida road network in 1923 ................................................................................... 2 1-2. Lane km and centerline km for State-maintained roads in Florida, 1980 to 1997 ... 6 1-3. Kilometers of paved roads in Florida from 1947 to 1995 ........................................ 7 1-4. Florida's forests and wetlands from 1936-94 ............................................................ 9 1-5. Roadless areas in Florida ........................................................................................ 10 1-6. Location of the I-75 (Alligator Alley) road corridor through south Florida .......... 11 1-7. One of the functional underpasses constructed on I-75 (Alligator Alley) in the Big Cypress National Preserve in south Florida ........................................ 13 1-8. Current distribution of the Florida black bear ........................................................ 14 1-9. Wekiva Basin Geopark Okefenokee Swamp National Wildlife Refuge corridor ................................................................................................................. 15 1-10. Ocala National Forest region including location of black-bear road-kills on major highways in central Florida ........................................................................ 16 1-11. Precast concrete culvert constructed on SR 46 between Wekiva River State Park and Seminole State Forest in central Florida ............... 17 1-12. Paynes Prairie location map .................................................................................. 19 1-13. Concrete barrier wall along U.S. 441 in Paynes Prairie State Preserve ................ 20 2-1. Components of biological features dataset ............................................................. 26 2-2. Digital elevation model for FDOT District 2 ......................................................... 29 2-3. Hydrography for FDOT District 2 .......................................................................... 31 2-4. Reclassified land cover/land use for FDOT District 2 ........................................... 32 ix

PAGE 10

2-5. Element occurrences and hotspots for FDOT District 2 ......................................... 33 2-6. Documented road-kills/FDOT road projects for FDOT District 2 ......................... 35 2-7. Conservation planning programs for FDOT District 2 ........................................... 36 2-8. Flow chart of analysis algorithm ............................................................................ 38 2-9. Florida Department of Transportation maintenance districts ................................. 39 2-10a. Road segment priorities in FDOT District 1 ........................................................... 41 2-10b. Road segment priorities in FDOT District 2 .......................................................... 42 2-10c. Road segment priorities in FDOT District 3 ........................................................... 43 2-10d. Road segment priorities in FDOT District 4 .......................................................... 44 2-10e. Road segment priorities in FDOT District 5 ........................................................... 45 2-10f. Road segment priorities in FDOT District 6 ........................................................... 46 2-10g. Road segment priorities in FDOT District 7 .......................................................... 47 2-11. Number of significant priority sites by route number ............................................ 49 2-12a. Road segment priorities and greenways in panhandle Florida ............................... 50 2-12b. Road segment priorities and greenways in peninsular Florida ............................... 51 2-13a. Black-bear road-kills and greenways in panhandle Florida .................................... 52 2-13b. Black bear and Florida panther road-kills and greenways in peninsular Florida ................................................................................................. 53 2-14a. Road projects and greenways in panhandle Florida ............................................... 54 2-14b. Road projects and greenways in peninsular Florida ............................................... 55 2-15. Human population growth rates in Florida from 1990 2000 ............................... 66 2-16. Identification of intense land uses .......................................................................... 67 2-17. Identification of severe slope/ravine areas typical of northwest Florida creek systems ................................................................................................................. 68 x

PAGE 11

3-1. One of the many canals adjacent to roadways in south Florida ............................. 74 3-2. One of the many off-road access points ................................................................. 74 3-3. Spread of invasive plants ........................................................................................ 75 3-4. Active recreation in resource areas ......................................................................... 75 3-5. Percent of the mean number of field sites measured on a scale of human disturbance ................................................................................................ 76 3-6. District 1 field verification sites ............................................................................. 77 3-7. District 2 field verification sites ............................................................................. 81 3-8. District 3 field verification sites ............................................................................. 84 3-9. District 4 field verification sites ............................................................................. 87 3-10. District 5 field verification sites ............................................................................. 90 3-11. District 6 field verification sites ............................................................................. 93 3-12. District 7 field verification sites ............................................................................. 95 3-13a. Bridge over stream on US 41 near Collier-Seminole State Park ............................ 97 3-13b. New bridge on SR 72 over Horse Creek, a tributary of the Peace River ............... 98 3-14. One of the wildlife underpasses constructed on I-75 in the Big Cypress National Preserve ............................................................................. 99 3-15. Wildlife Exclusion Fence on SR 29 in Collier County ........................................ 100 3-16. Blind curve on State Road 29 in Collier County .................................................. 100 3-17a. Depression site on SR 100 in the Etonia Creek area in Putnam County .............. 101 3-17b. Side view of existing culvert ................................................................................ 102 3-17c. Box culvert at ephemeral creek site on SR 100 in the Etonia Creek area in Putnam County ................................................................. 102 3-18. Bridge approach on State Road 100 at Palmetto Branch in Putnam County ....... 103 3-19a. State Road 206 crossing through the Fish Swamp area ........................................ 104 xi

PAGE 12

3-19b. Concrete box culvert present at unknown creek at the Fish Swamp site ............. 105 3-20. State Road 24 crossing through the Cedar Key Scrub State Preserve .................. 106 3-21. U.S. Highway 98 crossing through Tyndall Air Force Base in Bay County ........ 107 3-22. State Road 40 crossing through Ocala National Forest ........................................ 109 3-23. Wekiva River conservation lands for sale ............................................................ 110 3-24a. U.S. Highway 1 crossing through Crocodile Lake National Wildlife Refuge ..... 111 3-24b. A juvenile American crocodile encountered during field surveys ....................... 112 3-25. U.S. 1 passing between Cupon Bight and Big Pine Key ...................................... 113 3-26. Construction of replacement bridge ..................................................................... 118 3-27. Wildlife bridge extension on SR 46 over the Wekiva River in central Florida .... 118 4-1. Variety of existing structures that may function as wildlife underpasses ............ 129 4-2. Site preparation at track monitoring stations ........................................................ 135 4-3. Remote infrared camera system fastened to bridge pilings .................................. 136 4-4. Bridges and culverts monitored for wildlife use (March to December 2001) ...... 137 4-5. Bridges and culverts monitored for wildlife use (July 2002 to April 2003) ......... 138 4-6. Number of tracks recorded at culverts and bridges .............................................. 144 4-7. Number of photographs taken at culverts and bridges ......................................... 144 4-8. Unidentified tracks recorded at culverts and bridges ........................................... 145 4-9. Number of visits when no tracks were found at culverts and bridges .................. 145 4-10. Number of visits when sites were flooded or washed out .................................... 146 4-11. Road-kills recorded from 3/01 to 12/01 and 7/02 to 4/03 .................................... 146 5-1. The human footprint ............................................................................................. 192 xii

PAGE 13

5-2. Federal lands in the coterminous U.S. .................................................................. 193 5-3. Proliferation of roads across the United States ..................................................... 194 5-4. Hierarchy of theories/concepts for landscape-level studies ................................. 195 5-5. Hierarchy of relevant theory and conservation applications ................................ 196 5-6. Life zones of Florida as determined by latitude and climate ................................ 197 5-7. An example of the peninsula effect ...................................................................... 198 5-8. Ocala to Okeefenokee landscape linkage ............................................................. 199 5-9. A theoretical and conceptual framework for infrastructure ecology .................... 201 5-10. Landform, road alignment and black-bear road-kill trends .................................. 202 5-11. Highway speeds, land use and black-bear road-kill trends .................................. 203 5-12. Land cover and telemetry locations of radio-collared bears and black-bear road-kill trends ................................................................................... 204 5-13. The continual cycle of development and road construction ................................. 205 5-14. Radial or axial development ................................................................................. 206 5-15. Cumulative impacts of roads on wildlife and wildlands ...................................... 207 5-16. Highway mortality of key deer in Florida ............................................................ 208 5-17. Road density on Big Pine Key and Florida Key Deer National Wildlife Refuge .................................................................................................................. 208 5-18. Greenway-highway interfaces .............................................................................. 212 5-19. The Marjorie Harris Carr, Cross-Florida Greenway ............................................ 213 5-20. Florida 2020 Transportation Plan ......................................................................... 215 5-21. The Orlando-Orange County Expressway Authority (OOCEA) 2025 Transportation Plan for the greater Orlando area ....................................... 216 xiii

PAGE 14

5-22. Cross-Florida Barge Canal overpass .................................................................... 217 5-23. Existing highway structures and potential sites for wildlife crossings ................. 220 5-24. An example of a multi-cell culvert in northwest Gainesville, Florida ................. 221 5-25. Effects of vegetation and substrate at wildlife underpasses ................................. 222 5-26. Identifying sites for retrofits ................................................................................. 223 5-27. A conceptual matrix to further fine-tune ability to establish priorities for restoring connectivity/constructing wildlife underpasses .................................... 228 6-1. Comparison of Florida black-bear road-kills and average daily vehicle trips ...... 232 6-2. Lake Jackson and Little Lake Jackson wildlife crossing ...................................... 239 6-3. Greater Ocala National Forest ecosystem ............................................................ 248 xiv

PAGE 15

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ECOLOGICAL EFFECTS OF ROADS: THEORY, ANALYSIS, MANAGEMENT, AND PLANNING CONSIDERATIONS By Daniel Joseph Smith December 2003 Chair: Joseph M. Schaefer Cochair: Larry D. Harris Major Department: Wildlife Ecology and Conservation Human population growth and land development over the last century has resulted in widespread habitat loss, fragmentation, and increased wildlife mortality on highways. Since 1936, growth in Florida has resulted in the loss of 54% of herbaceous wetlands and 22% of forested lands. Concurrent with increased land development, hard-surface roads are being constructed at the rate of 8.8 km per day. Interagency coordination in the 1990s toward establishing a statewide greenways and trails network prompted State transportation officials to look at greenway-highway interfaces where conflicts are likely to occur. First, a computer algorithm was developed to evaluate and prioritize the need for wildlife crossing structures or underpasses on state roads. A rule-based decision-support model was developed using geographic information systems (GIS). Of 15,644 road segments prioritized by the model, 81% were located in designated greenways. Chronic road-kill sites, focal species hotspots, greenway linkages, listed species presence, and xv

PAGE 16

strategic habitat conservation areas strongly influenced results. Ninety-five scheduled road construction projects coincided with high-priority road segments. Second, field inventories were conducted to determine mitigation needed at 1,232 highly ranked, ecological hotspots on highways. Features documented included parameters of existing bridges and culverts, characteristics of roadways, description of surrounding landscape features, and signs of present animal use. Mitigation measures recommended ranged from installation of underpasses or culverts to minor measures such as fencing, landscaping, signage, and speed restriction. Third, a 2-year monitoring study was conducted to determine the capacity of existing highway drainage-structure designs to function as wildlife passages. Fifty-five different organisms were identified from 47,955 records at 290 field sites. Twenty structural, environmental, and ecological factors affecting use or avoidance of highway-crossing structures by six faunal groups (birds, carnivores, herpetofauna, mesomammals, small mammals, and ungulates) were evaluated through logistic regression analysis. Significance of factors varied by faunal group. Lastly, landscape ecology principles applied to conservation and transportation planning are discussed. Gradients such as topographic relief can be used to design and situate underpasses that conform to on-site ecological flows and landscape patterns. Road-corridor design should include sufficient right-of-way clearance and reduced curve severity and slope steepness to increase driver visibility. Appropriate native vegetation at entry points to underpasses protects species either intolerant of open areas or subject to ambush predators. Fencing or other barriers should be used to direct wildlife toward the crossings and away from the road surface; and to restrict human access. xvi

PAGE 17

CHAPTER 1 LAND DEVELOPMENT IN FLORIDA AND EFFORTS TO REDUCE ROAD-RELATED WILDLIFE MORTALITY Roads are an important element for human mobility and the transport of goods and services. Before 1900, an extensive system of trails and simple roads already existed in the United States; of these, only 4% had improved surfaces (FHA 1979). As mass production of the automobile began in 1906 (FHA 1979), paved road construction became an important step in expanding commerce and land development of the United States. The Federal Aid Road Act of 1916 was the first major State-Federal cooperative funding program instituted for road improvement and expansion. In 1914, of 413,601 km of rural roads, only 22,531 km were paved (FHA 1979); yet by 1930, there were 1,116,885 km of paved rural roads (ODOT 2003). Currently, there are over 4,828,032 km of paved rural roads in the coterminous United States (FHWA 1996). The timeline for road construction in Florida was much the same as for the rest of the country. Construction of the first transcontinental highway in America (the Spanish Trail) began in Florida in the early 17 th century (McDonald 1937). Today the Spanish Trail is a major highway (US 90) connecting Florida to California, a distance of approximately 3,921 km. Yet in 1923, only 189 km of US 90 (total length in Florida, 764 km) was hard-surfaced, located primarily around Pensacola and between Jacksonville and Lake City. Without exception, few roads in Florida were hard-surfaced in 1923 (Figure 1-1). More recent growth trends for roads in Florida include construction of hard-surface 1

PAGE 18

2 roads at a rate of 8.8 km per day (FDOT 1998); and increase in road density from 0.28 km/km 2 in 1947 to 1.31 km/km 2 in 1995 (FHWA 1996). Figure 1-1. Florida road network in 1923. Four classes of road surfaces existed in 1923, hard-surfaced, semi-hard-surfaced, graded, and ungraded. As shown, most hard-surfaced roads occurred in and around towns and cities, few rural roads were paved. In 1918 there were less than 13 km paved; yet by 1936 over 11,690 km were paved (McDonald 1937). The driving factor affecting land-use change and development is human population increase. According to the American Planning Association (APA 1995), Florida population growth has been increasing, on average, at 3.2% per year since 1950 (approx. 245,000/yr). As the population burgeons, the number and size of urban centers

PAGE 19

3 expand. Poorly planned urban sprawl erodes and threatens remaining natural areas through conversion to urban and intense agricultural uses. Human population growth and land development over the last century has resulted in widespread habitat loss (Kautz 1992), fragmentation (Harris and Silva-Lopez 1992), and increased wildlife mortality on highways (Gilbert 1996, Smith 1996). The consequence of these impacts to native species diversity can be staggering. In most cases, species more adaptable to human change in the landscape (habitat generalists, midsize carnivores, and omnivores) flourish, while habitat specialists and larger, carnivorous forms decline (Crooks and Soule 1999, Rogers and Caro 1998, and Harris and Gallagher 1989). One approach that mitigates the negative effects of expanding human development is the creation of greenways or ecological networks consisting of an integrated system of habitat corridors, core habitat reserves and naturally isolated areas. Such "green" infrastructure can maintain or restore necessary ecological processes (e.g., animal and plant migration and dispersal, hydrologic flows, fire management regimes) to once-contiguous natural landscapes (Bennett 1999, Dobson et al. 1998, Csuti 1991, Harris and Scheck 1991, Harris and Gallagher 1989, Eisenberg 1986, and Forman 1983). Creative use of linear parks and greenways (open space) can also restore natural amenities to urban areas and provide buffers to protect surrounding natural systems (Smith and Hellmund 1993, and Little 1990). While greenways provide a means to restore ecological function to large-scale habitat conservation systems, roads that act as significant barriers to most ecological processes intersect most. Proper management of these greenway-highway interfaces is necessary for maintaining the integrity of an ecological network. The Florida

PAGE 20

4 Department of Transportation (FDOT) recognized the importance of these interfaces and funded a 7-yr research program to identify and prioritize needs; and to develop programs and policies to counter the negative ecological impacts of roads. Infrastructure and Population Growth in Florida Origin of Road Networks and Opportunities for Development Major expansion of settlements in Florida began with the transfer of the peninsula from Spain to the United States in 1821. One of the primary reasons for acquiring control of the territory was to regulate Seminole Indian attacks on American colonists (McDonald 1937). This brought about the construction of an integrated road system by the military during the Seminole Indian Wars. Several forts (that were forerunners to current Florida cities) were constructed across the State at various nodes of the military road network (McDonald 1937). During the U.S. Civil War, Florida (with its many seaports) served as a transportation hub for food and supplies for the Confederate Army (McDonald 1937). Many of Florida's roads and railroads were left in a condition of disrepair following the civil war. A period of reconstruction led by Henry Flagler and H.B. Plant occurred (in the late 1800s and early 1900s) that resulted in the widespread expansion of existing railroads (McDonald 1937). Construction of roads for automobiles began in earnest in 1915, with the creation of the first State Road Department (Kendrick 1964). By 1935, the population in Florida had reached 1,602,268 (McDonald 1937). In 1918 there were 7,598 km of improved state highways, yet only 13 km were paved. By 1936, 19,312 km of state roads existed and 11,690 km were hard-surfaced (McDonald 1937). The increase in construction of hard-surface roads during this period was due largely to

PAGE 21

5 depression-era public programs designed to create jobs for the unemployed (Sautter 1986). Recent Population Growth and Infrastructure Needs Population growth and land development have produced steady and increasing concerns about the declining quality of the environment and natural resources of Florida. Florida is the 4 th most populous state (behind California, Texas, and New York) (BEBR 1998). Population growth between 1990 and 2000 was 24% (USCB 2000). Twenty-eight new residents move to Florida every hour; the 3 rd highest growth rate in the U.S. The population rose from 13 to 16 million between 1990 and 2000; and is projected to reach 18.5 million by 2015 (USCB 2000). Additionally, Florida receives an estimated 59 million tourists annually (FDOT 2000). Fifty percent of the tourists visiting Florida arrive by automobile (FDOT 1994). Population density in Florida is approximately 103 persons/km 2 the 10 th highest in the U.S. (BEBR 1998). The intensity of our dependence on automobiles is illustrated by the following demographic facts: the average Florida household has 2.74 persons (BEBR 1998) and 2.15 vehicles (FHWA 1996). With 12 million private registrations, nearly all adult-aged residents own an automobile (3 rd most in the U.S.) (FHWA 1996). The impact of the large number of vehicles traveling on Florida's roads manifests itself through increased traffic density. Vehicle kilometers driven on State roads in Florida increased some 90% between 1980 and 1997, nearly doubling from 194.8 million km to 372.1 million km, respectively (FDOT 1998). Florida's traffic volume is the 3 rd highest in the U.S. (FHWA 1996).

PAGE 22

6 Highway Construction Corresponding to increases in traffic are increases in highway construction to meet demand. From 1980 to 1997, centerline km of State-maintained roads increased from 18,251 to 19,315 (5.2%) and lane km increased from 52,170 to 62,793 (20.4%) (FDOT 1998). Over this period, lane km have increased at a rate 14.4 times greater than centerline km (Figure 1-2), indicating more expansion of existing roads to multi-lane highways than construction of new roads. 50,00055,00060,00065,00070,00018,00018,50019,00019,50020,000Lane kmCenterline km Figure 1-2. Lane km and centerline km for State-maintained roads in Florida, 1980 1997. Lane km represent the construction of additional lanes to existing highways. Centerline km represent new highway construction. The graph depicts a greater increase in expansion of existing roads over the construction of new roads. (Data source: Florida Department of Transportation 1998) Kilometers of paved roads in Florida have increased 5-fold since 1947 (Figure 1-3). Florida ranks 10 th and 13 th in the U.S. for total length of roads (182,735 km) and road density (1.31 km/km 2 ), respectively (FHWA 1996). Urban roads have increased from a share of 16% in 1947 to 43% in 1995, whereas rural roads have declined from 84% in 1947 to 57% in 1995. In actual kilometers, both categories nearly doubled from

PAGE 23

7 1964 to 1976, though subsequently rural roads have declined (-3,542 km) while urban roads continue to rise (+29,463 km). Figure 1-3. Kilometers of paved roads in Florida from 1947 to 1995. Overall highway construction has increased five-fold since 1947. Urban roads in 1995 are 16 times greater than that of 1947. Kilometers of rural roads have also increased since 1947, but have remained relatively constant since 1976. These trends reflect early agricultural development and later the rapid transformation of rural areas to urban land-uses. (Data sources: Florida Statistical Abstracts (1976, 1988, 1996) and Federal Highway Administration (1947, 1964) 1947 1956 1964197619861995 urbanruraltotal 28.0 57.2 83.2 156.5 159.6 182.7 23.5 39.0 58.8 108.4 108.0 104.8 4.5 18.2 24.5 48.1 51.5 77.9 0 25 50 75 100 125 150 175 km (x 1000) Two causal factors are at work here. First, the chart does not reflect the amount of rural land that has been converted to urban land-uses over this period; resulting in rural roads being reclassified as urban roads. Therefore kilometers of urban roads are increasing because of construction of new urban roads and land-use conversion. Secondly, it would suggest that for total rural-road kilometers to remain the same, new

PAGE 24

8 rural-road construction is occurring despite urban land-use conversion. Urban and rural hard-surface roads were constructed at rates of 4.45 and 5.98 km/day respectively, between 1947 and 1995. Over the 48-yr interval, rural roads increased by 81,300 km and urban increased by 43,600 km (Figure 1-3). As a result, roadless areas have decreased dramatically in size and amount. Land-Use Change Florida has a total land area of 139,697 km 2 ( making it the 26 th largest state). Although 30% of this area (38,000 km 2 ) is protected as public parks, recreation, and preserve lands, the threat of development to key areas and linkages of Florida's statewide habitat system remains. Overall, growth in the state has resulted in the loss of 54% of herbaceous wetlands and 22% of forests (Figure 1-4; Kautz 1992, Florida Water Management Districts unpublished data 1995). According to Kautz (1992), Florida has lost the equivalent of eight Apalachicola National Forests since 1950 (an area of approximately 2,400 km 2 ). This land has largely been converted to urban and agricultural lands. From 1936 to 1994, urban lands have increased from 2,800 to 19,700 km 2 a change of approximately 600% in the last 58 years (Kautz 1992, Florida Water Management Districts unpublished data 1995). Land conversion to agriculture had reached 44,000 km 2 in 1980, but now appears to be declining; another victim of urban-development pressure. Furthermore, more than 13% of Florida has been covered by pavement (Florida Water Management Districts unpublished data 1995). In summary, over 43% of Florida's original forest and wetland habitats have been converted to human-oriented land-uses (Kautz 1992).

PAGE 25

9 8 7 5 3 2 Figure 1-4. Florida's forests and wetlands from 1936-94. Forested lands declined by 22% between 1936 and 1994. Longleaf pine, accounting for 3.1 million ha (37%) of Florida's forests in 1936, has been reduced to 0.1 million ha. Marshlands accounted for 20% of the land area in 1936; there are now less than 1.3 million ha (8.8%). About 60% of the original wetland habitats have been drained for urban and agricultural uses or water management. (Modified from Kautz (1992) with Florida Water Management Districts unpublished data 1995) 14691936194919591970198019871994ha (millions) Forests Herbaceous Wetlands There are now only eight remaining areas greater than 1,000 km 2 in Florida where paved roads are absent. There are 18 additional areas greater than 500 km 2 Combined, these areas comprise 12 separate but contiguous regions (Figure 1-5). The largest is the greater EvergladesBig Cypress region (13,900 km 2 ) that consists of nine adjacent habitat blocks separated by major paved roads, canals, and flood-control structures. Seven of the 12 regions consist of public forest, park, preserve, wildlife refuge, or military lands; the remaining five are in part targeted for public acquisition. The state presently looks like a jigsaw puzzle divided into various-sized pieces of land separated by an extensive road network. This configuration jeopardizes the ability of wide-ranging wildlife (such as the Florida panther Puma concolor coryi, Florida black bear Ursus americanus floridanus, and river otter Lutra canadensis) to successfully disperse and colonize adjacent habitat areas.

PAGE 26

10 Figure 1-5. Roadless areas in Florida (includes paved roads only). Only three regions remain that have areas greater than 1,000 sq km without paved roads, the Everglades, Three Lakes Wildlife Management Area, and Appalachicola National Forest. Projects Designed to Increase Permeability of Roads in Florida This discussion will focus on case studies of road projects designed to restore landscape connectivity to fragmented ecosystems in Florida. Over the last 15 years the FDOT constructed crossing structures designed to address public concerns about automobile passenger safety, property loss, and wildlife and floodplain management (Smith and Dodd 1999, Roof and Wooding 1996, Evink 1990, and LoBuono 1988). The Everglades, Alligator Alley, and the Florida Panther Since 1917, population growth, land conversion to agriculture and urban uses, and damaging flood events resulted in construction of over 2,400 km of canals and levees for

PAGE 27

11 flood control and water supply in the Everglades (SFWMD 1992). In many instances, roads were built on levees next to the canals constructed throughout the Everglades system. These roads compound habitat alteration, fragmentation, and barrier effects caused by the canals. Interstate-75 (Alligator Alley) is one such road that fits this characterization (Figure 1-6). Photo credit: August Burghard, Ft. Lauderdale, FL Kissimmee R. Florida Ba y Figure 1-6. Location of the I-75 (Alligator Alley) road corridor through south Florida. Insetresidual effects, canals left during dredging for construction of the original highway, ca. 1960s. Alligator Alley, a 125 km highway traversing the Florida Everglades and Big Cypress Swamp (FEBCP), was constructed in the late 1960s. Not only did the route bisect and fragment the FEBCP, but borrow canals were excavated for fill dirt and for drainage purposes to control flooding of the road (Figure 1-6). These actions further altered natural hydrologic flow patterns. As Florida's population grew in the 1970s, the traffic volume on Alligator Alley increased; and FDOT proposed four-laning the highway in the 1980s.

PAGE 28

12 Also, significant road-kills were recorded including several collisions with the endangered Florida Panther (17 deaths from 1979 to 1989). With a total population estimated at 46 to 74 individuals, the loss of one to two animals per year to automobile collisions was a significant mortality factor that could not be ignored (Maehr 1997b). The Florida panther recovery plan identified road mitigation as a necessary step for the restoration of a viable population (USFWS 1995). Specific prescriptions included wildlife underpasses, warning signs, reflectors, and nighttime speed restrictions. In the late 1980s, the FDOT installed 36 underpasses on I-75 through the FEBCP. Twenty-four of these were designed and located specifically to facilitate safe travel for the Florida panther (Figure 1-7) (Villano 1993). Locations of most crossing structures were determined from telemetry data, where repeated road crossings of 18 radio-collared cats occurred; or other significant evidence (typically vegetated corridors of high ground) where repeated Florida panther presence was found (Logan and Evink 1985). All remaining underpasses were located at 1.6 km intervals; a distance thought to minimize travel distance for Florida panthers between crossings. Thirteen bridges were also constructed that provide 12.2 m wide strips of land over canals (parallel to the road) to facilitate movement by terrestrial organisms at underpass locations (Evink 1990). After 4-lane construction was completed in 1990, the road project modifications resulted in the elimination of vehicle collisions with the Florida panther on Interstate-75 (Gary Evink, FDOT; personal communication). Foster and Humphrey (1995) recorded underpass use by Florida panther, Florida black bear, bobcat Lynx rufus, white-tailed deer Odocoileus virginianus, raccoon Procyon lotor, alligator Alligator mississippiensis, and various wading birds.

PAGE 29

13 Figure 1-7. One of the functional underpasses constructed on I-75 (Alligator Alley) in the Big Cypress National Preserve in south Florida. Bridge dimensions are 36.6 m wide and 2.4 m hi g h. ( Photo credit: David Maehr Universit y of Kentuck y, Lexin g ton KY ) The amount of preplanning and research for the design and locations was credited for the effectiveness of the underpasses. Openness of the structure (openness ratio > 0.92), 3 m tall barrier fencing that runs the length of the project boundary (64 km), and native vegetation were important aspects of the underpass design (Figure 1-7) (Foster and Humphrey 1995, and Evink 1990). The success of the Interstate-75 underpasses has led to construction of six, additional, wildlife-crossing tunnels on SR 29 (a north-south arterial highway) that also crosses through primary habitat for the Florida panther. Habitat Connectivity in Central Florida for the Florida Black Bear The pattern of intensified development and road construction in Florida has forced the threatened Florida black bear into several segregated subpopulations (Figure 1-8). This isolation has created concern regarding potential inbreeding depression, founder effects, and greater impacts by natural and unnatural disasters (Maehr 1984).

PAGE 30

14 Potentially increasing numbers of Florida black bears coupled with increasing habitat loss to development leads to greater dispersal distances for males and subadult females (in search of new home range territories). In Florida, an individual 2.5 yr old male was reported traveling 126 km over large areas of non-forested land (Maehr et al. 1988). Dispersal events in and between dwindling core habitat reserves lead to increased encounters with highways, thus increasing the probability for collisions with vehicles. Concurrently, increasing traffic volumes add to the risk of collisions. Even though roads pose a serious threat to a bear's success in moving between large habitat areas, their wide-ranging behavior indicates that widely spaced populations can maintain genetic connections (if suitable habitat-corridors are present). Figure 1-8. Current distribution of the Florida black bear. Habitat loss and fragmentation have reduced the former statewide range of the black bear to seven areas of concentration. (Adapted from Florida Fish and Wildlife Conservation Commission 2000)

PAGE 31

15 An area of great concern for black-bear conservation is preserving the landscape linkages between the Wekiva River Basin (north of Orlando) and the Okefenokee Swamp National Wildlife Refuge in southern Georgia (a distance of approximately 250 km; encompassing over 750 thousand km 2 of existing and proposed conservation-lands) (Figure 1-9). Twenty-one major roads pose serious threats to the integrity of this network (as barriers to movement, and as development facilitators that can result in additional habitat loss and fragmentation). Figure 1-9. Wekiva Basin GeoPark Okefenokee Swamp National Wildlife Refuge corridor. The area within the orange border represents a major landscape linkage for black bear conservation stretching from central Florida to southern Georgia. Several major roads (shown in red) function as barriers (of variable permeability) to wildlife movement and other ecological processes. From 1976 to 1999, 275 black bear roa d -kills were recorded on these roads ( FFWCC 2000 )

PAGE 32

16 Five of these roads (all located in the Ocala National Forest (ONF) or the Wekiva River Basin; Figure 1-10) account for the majority of these deathsSR 46 (48), SR 40 (69), SR 19 (58), SR 44 (26) and CR 42 (21). The FDOT (in 1994) constructed an experimental box culvert (Figure 1-11) under SR 46 (Roof and Wooding 1996) as a prototype underpass to address the problem. In addition, a 3 m tall barrier fence extending 0.6 km west and 1.1 km east of the underpass was erected (Roof and Wooding 1996). Figure 1-10. Ocala National Forest region including location of black bear road-kills on major highways in central Florida. From 1976 to 1999 over 200 road-kills have been documented on these five ma j or roads ( FFWCC 2000 ) The aim of the project (according to Gilbert and Wooding 1994) was to reduce the number of road-kills occurring along a 4.5 km section of SR 46 (20 fatalities from 1981 to 1994); and to establish an unobstructed connection between the isolated Wekiva River population and the larger, ONF population (that consists of more than 500 individual black bears; Eason 2000).

PAGE 33

17 The first year following construction, Roof and Wooding (1996) only recorded five black bears crossing through the underpass. Just one of the 50 black bears that encountered the barrier fence found and crossed through the underpass; another crossed the road at the end of the fence; and the remainder turned around and returned to the forest (Roof and Wooding 1996). Subsequently, three vehicle collisions with black bears have occurred at the end of the barrier fencing. Vegetative cover was added near the underpass openings to increase use. Now that black bears have become acclimated to the structure, movement through the underpass occurs frequently (87 crossings from 2001 to 2002). Figure 1-11. Precast concrete culvert constructed on SR 46 between Wekiva River State Park and Seminole State Forest in central Florida. Placement coincides with high incidence of black bear road-kills (16 from 1981 to 1998) (FFWCC 2000). Dimensions are 2.4 m hi g h 7.3 m wide and 14.6 m lon g Fourteen additional black bears have died in vehicle collisions (from 1988 to 1998) in four distinct clusters beyond the fencing, associated with the SR 46 underpass

PAGE 34

18 (FFWCC 2000). Construction of additional underpasses and barrier fencing are planned to effectively retrofit the remainder of this section of highway (to reduce future road-kills and increase permeability). In 1994, plans to increase SR 40 from 2-lanes to 4-lanes through the ONF led to a comprehensive black-bear movement study. Preliminary results (1999 to 2001) of movements of 77 radio-collared individuals revealed 324 crossings of SR 40. Track studies performed on a 17.7 km section of SR 40 revealed 752 sets of bear tracks (McCown and Eason 2001). The final results of the movement study (1999 to 2003) should provide evidence for the need to increase the permeability of the major roads that currently cross through the ONF. Speculation based on previous studies (Ariza 1998, Gilbert and Wooding 1994, Wooding and Maddrey 1994, Wooding and Hardisky 1994, Brody and Pelton 1989, and Wooding and Brady 1987) suggests that increasing SR 40 to four lanes of traffic will further fragment the ONF black-bear population; by acting as an aversion zone or barrier to movement, and increasing road-kills on side roads due to increased traffic (Gilbert 1996). Population effects that may result from avoidance of the 4-lane highway include the potential creation of subpopulations with reduced gene flow; based on research conducted in North Carolina (Brody and Pelton, 1989, and Carr and Pelton 1984). Road Mortality of Herpetofauna in Payne's Prairie State Preserve Payne's Prairie (in north-central Florida) is a wet prairie trisected by two major multi-lane highways (Figure 1-12). Since the 1930s, it has been known as a significant population reservoir for many amphibians and aquatic snakes (Kauffeld 1957, Auffenberg 1956, Hellman and Telford 1956, and Carr 1937).

PAGE 35

19 Figure 1-12. Paynes Prairie location map. Paynes Prairie, a Florida state preserve near Gainesville, Florida, encompasses approximately 7,300 ha. Several inflows occur from surrounding urban and agricultural lands. The two major roads that impact the preserve are I-75 and U.S. 441 that subdivide the prairie into three large fragments. Drainage canals and spoil areas parallel to these roads cause significant disturbance to adjacent habitat. Gainesville From data collected in the 1970s, an estimated 2,753 (predominantly aquatic) snakes were being killed annually on a 3.38 km stretch of 4-lane highway (Smith 1996). Estimates from data collected in the early 1990s revealed similar mortality levels. Public concern led to the construction of a 1.1 m high barrier wall in 2000 designed to eliminate road-kills (Figure 1-13). The FDOT also installed four new round pipe culverts as an addition to the four existing box culverts to allow for faunal movement from one side of the road to the other.

PAGE 36

20 Figure 1-13. Concrete barrier wall along U.S. 441 in Paynes Prairie State Preserve. The 1.1 m high barrier is fitted with 21.83 m and 22.4 m box culverts, and 4.91 m round p i p e culverts to allow wildlife movement and water flow between p rairie fra g ments. Monitoring of these culverts was conducted pre and post construction; to determine the effectiveness of the culvert-barrier system to eliminate road-kills, yet maintain habitat connectedness. Pre (1998 to 1999) and post (2001 to 2002) construction monitoring of organisms killed on the highway resulted in counts of 3,365 and 1,992, respectively (Barichovich and Dodd 2002, Smith and Dodd 1999). Significant reductions occurred for all taxa, except frogs and mammals. Excluding hylid treefrogs (that readily climb the barrier wall), 65% of post-construction road-kills occurred beyond the extent of the barrier wall (Barichovich and Dodd 2002). Prior to construction, 25 terrestrial or semi-aquatic vertebrate species were recorded from track surveys and funnel traps using the (2.83 m) dry culverts (Smith

PAGE 37

21 and Dodd 1999). An additional 18 different terrestrial or semi-aquatic vertebrate species were recorded using the two existing dry culverts following construction (Barichovich and Dodd 2002). Only one additional terrestrial vertebrate (ring-necked snake Diadophis punctatus) was found using the new (4.91 m) round culverts and not the existing (21.83 m) dry culverts. The new culverts (located at lower elevations along the basin) were periodically flooded, and provided passage for (8) aquatic-based organisms as well as (17) terrestrial or semi-aquatic species during dry periods (Barichovich and Dodd 2002). Post-construction use of the existing (2.4 m) aquatic box culverts included (8) aquatic and (4) semi-aquatic species (Barichovich and Dodd 2002). Total diversity of terrestrial/semi-aquatic, and aquatic species found using all culverts was 43 and 18, respectively (Barichovich and Dodd 2002, Smith and Dodd 1999). Ten species (birds excluded) were found as road-kill and not encountered using culverts (1 lizard, 3 turtles, 5 snakes, and 1 mammal). This could indicate that either: these culvert designs are unsuitable for certain species; some species may occur in much lower densities (resulting in fewer opportunities to encounter culverts); or a longer sampling period is required. The culvert/barrier wall system installed on U.S. 441 across Paynes Prairie has eliminated 59% of highway road-kills and provided safe passage between prairie fragments for at least 61 species (as determined from two years of monitoring, pre and post construction) (Barichovich and Dodd 2002, Smith and Dodd 1999). Several measures are recommended to improve future designs. First, to prevent species from circumventing the barrier, the wall or a substitute structure should be extended a sufficient length beyond the target area (along the right-of-way). Second,

PAGE 38

22 increased culvert usage might be achieved by using a sinusoid-shaped barrier wall (parallel to the highway) that funnels animals into the culverts. Third, to increase permeability for all species, the number and type of passages should be increased; different tolerances (including structural parameters) and modes of mobility must be considered (multiple culvert designs and placements may be necessary in the same area to accommodate all species, e.g., a gradient of fossorial, terrestrial, semi-aquatic, and aquatic organisms). Original research supporting these suggestions is presented in Chapter 3 and 4. Culverts and underpasses serve as connections between landscapes divided by highways; and play a critical role in decreasing the barrier effect of roadways by increasing permeability for wildlife. Increased permeability results in consequent decreases in mortality (Yanes et al. 1995). These underpasses can facilitate corridors that connect spatially separated habitats and enhance the efficacy of wildlife movement throughout the landscape (Forman 1983).

PAGE 39

CHAPTER 2 PRIORITIZATION OF ECOLOGICAL HOTSPOTS ON FLORIDA HIGHWAYS Introduction Underpasses, referred to internationally as ecopassages, can reduce transportation-related wildlife mortality and restore connectivity to the landscape. With increasing requests for these projects, it has become apparent that mitigation is needed throughout the state (in conjunction with road improvement and development projects). Government efforts and public support in the 1990s for ecologically based greenways in Florida has prompted FDOT to look at highway-greenway interfaces; and the potential for implementing a wildlife underpass construction program designed to restore landscape connectivity and ecological processes at regional and statewide scales. The ability to coordinate needs for ecopassages with future highway construction projects would prove valuable toward effective and efficient use of funds for highway construction. A necessary step in this effort is the identification and prioritization of ecological hotspots on highways. This chapter represents phase I of an FDOT-sponsored research program, designed to integrate transportation planning with statewide conservation objectives. Methods Priority of roads was determined from their relative ecological impact. Ecological impact was ascertained by ranking roads according to various, existing ecological and planning criteria. The criteria were selected to identify road segments with the greatest 23

PAGE 40

24 impact on existing and proposed conservation-landstheir ecological integrity, the species that utilize them, their connectedness to adjacent habitat areas, and their attributes regarding sustainability and restoration (Smith et al. 1998). Important environmental factors (for prioritizing relative impact of roads on lands with conservation value) were derived from a survey conducted at the FDOT-sponsored Transportation-Related Wildlife Mortality Seminar in Orlando, Florida in 1996 (Appendix A). Respondents were asked to rank various criteria associated with prioritizing sites for the location of underpasses on Florida roads. Eleven elements were identified and ranked as follows: 1. Chronic road-kill sites 2. Known migration/movement routes 3. Identified hotspots of focal species 4. Landscape linkages (designated greenways) 5. Presence of listed species 6. Identified strategic habitat conservation areas 7. Riparian corridors (with potential for retrofitting existing structures) 8. Core conservation areas 9. Presence of separated required ecological resources (e.g., a forest patch and ephemeral wetland breeding area for amphibians that is separated by a highway) for a species or set of species 10. Public ownership (or in public land acquisition program) as opposed to private lands 11. Potential to be included in proposed road improvement project Note that No. 2 (known migration/movement routes), which pertains to large-scale animal-movement events such as migrating caribou, does not occur in Florida. Therefore, this criterion was modified to apply to wildlife movement patterns typical for this region (i.e., juvenile dispersal, breeding season movement, and normal home range activity). This was estimated by focusing on landscape features that typically represent likely travel routes including topographic gradients, riparian corridors, and habitat ecotones (Forman 1995, Ims 1995, Noss et al. 1994, Cross et al. 1991, Harris et al. 1991,

PAGE 41

25 Johnson et al. 1991, Noss et al. 1990, and Harris et al. 1989). Criteria No. 8 (core conservation areas) was divided between two other overlapping criteriapublic lands and strategic habitat conservation areas. In this case, the core portions of large conservation areas in Florida are either publicly owned or part of state-sponsored programs for conservation land acquisition. Lastly, No. 9 (presence of separated required ecological resources) was dropped from the analysis due to lack of available data. Criteria were evaluated according to large-scale priorities including the plan for an ecological greenways network in Florida. Forman (1995) and Noss et al. (1994) described an approach to reserve design based on principles of landscape ecology. Using this philosophy, several statewide planning datasets were obtained. Available data corresponding to these criteria included: habitat/land cover and habitat ecotones, hydrology, topography, ecological hotspots, known-species occurrence sites, known road-kill locations for Florida black bear and Florida panther, road-kill data for state and federal parks and preserves, existing public conservation-lands, strategic habitat conservation areas, areas of conservation interest, proposed public conservation-lands, proposed ecological greenways, and future road projects. An appropriate value was given to each attribute from a dataset relative to the importance of other attributes in the set. Datasets were grouped into six categories: biological features, landscape features, road-kills, planning, infrastructure, and public conservation lands. An example of the process is shown (Figure 2-1) for the biological features category. The use of data categories reduced redundancy. Table 2-1 displays the categories, the criteria, and associated attributes with assigned base values. The survey rank and multipliers reflect priority rankings from the questionnaire (Appendix A).

PAGE 42

26 Note: Relative ranking of criteria increases in ascending order *Federal Status Figure 2-1. Components of biological features dataset. This flow chart provides a graphic dis p la y of construction of the biolo g ical features g rid and relative im p ortance of attribute values. b b i i o o l l o o g g i i c c a a l l f f e e a a t t u u r r e e s e e n n d d a a n n g g e e r r e e d d * 7 7 + + f f o o c c a a l l s s p p e e c c i i e e s 5 5 6 6 f f o o c c a a l l s s p p e e c c i i e e s 3 3 4 4 f f o o c c a a l l s s p p e e c c i i e e s s s s s c c * o o r r b b i i r r d d r r o o o o k k e e r r i i e e s s F F G G F F W W F F C C h h o o t t s s p p o o t t s t t h h r r e e a a t t e e n n e e d d * F F N N A A I I e e l l e e m m e e n n t t o o c c c c u u r r r r e e n n c c e e s s s s s s s Cell-based (raster) modeling (Arcview Spatial Analyst; Environmental Systems Research Institute, Inc., Redlands, California) was used to analyze and combine datasets and determine priority rankings. The resolution (cell size) used in the analysis was 100 m. Although certain data were available at higher resolution (30 m), 100 m was used to accommodate limited data storage capability and microprocessor speed. This resolution and scale was satisfactory to meet United States Geological Survey (USGS) national mapping and data accuracy standards. State roads were buffered on each side by 600 m; consistent with the distance determined by other studies as the primary negative edge-effect-zone of highways (Fagan et al. 1999, Yahner et al. 1997, Forman 1995, Rodgers et al. 1995, Brody et al. 1989, Harris 1988a, Yahner 1988, Kroodsma 1987, Wilcove et al. 1986, Wilcove 1985, Carr et al. 1984, and Ferris 1979). The road buffer also serves as an analysis mask to eliminate unnecessary data that would slow computer processing.

PAGE 43

27 Table 2-1. Grid values for FDOT priority model Category Criteria Base value Survey rank Multiplier Landscape Gradients 2 8 Features Topography: slopes and ridges (greater than 36m elevation) 2 Ecotones (natural lands greater than 40ha) 2 Riparian 7 4 Streams/Lakes in natural habitats 4 Canals in natural habitats 3 Streams/Lakes/Canals in urban/agriculture lands 2 GFC Habitat/Land Cover n/a* 3 Xeric Habitats 4 Wetland Habitats/Hardwood Hammocks 3 Silvicultural/Mixed Pine and Hardwoods 2 Biological GFC Hotspots 3 7 Features 7+ species 4 5 6 species 3 3 4 species 2 FL Element Occurrence (listed species locations) 5 5 Endangered 4 Threatened 3 Species Special Concern/Bird Rookery 2 Road-kill Road-kill 1 9 Listed Species (black bears, panther, key deer) 4 State Parks 2 Planning Strategic Habitat Conservation Areas 5 5 High (Clan98{proposed}, FWC-SHCA{proposed}, FNAI{A,B}, TNCERC{Priority}) 4 Low (FNAI{C}, TNCERC{Interest}) 2 Greenway Final Rankings (linkages) 4 6 High Priority 4 Medium Priority 3 3 7 Final Rankings 2 Public Public Lands 9 2 Clan98 (existing) 4 Infrastructure Road Projects 10 1 Proposed, Bridge Replacements 4 Existing 2 This criterion was not identified in the survey

PAGE 44

28 Configuration of Data Original data sources were manipulated to conform to the criteria specified by responses to the questionnaire. Nominal and ratio data types that represent ranked criteria were converted to interval data measurements so that basic calculations could be performed in the GIS model. Carr et al. (1999) described this process; where individual data layers, with descriptive attributes, are transformed into values on an interval scale to provide a measure of utility (utility assignment). Specific conversions are explained below for each criterion; and assigned base values, as mentioned previously, are shown in Table 2-1. Landscape gradients Potential wildlife movement, travel, or dispersal routes were based on topographic relief and ecotonal-habitat movement-corridors (Forman 1995). USGS digital elevation models (floating-point grids based on 1-m contours; spatial resolution m) were used to determine topographic ridges or plateaus. In Florida, low-elevation areas were considered equivalent to aquatic features including wetlands; and were therefore identified through those criteria. A floating range of elevation was used to determine significant topographic zones for each FDOT district; based on the differential change in elevation from district to district across the state (Figure 2-2). Range of elevation (m) for each district is shown below: District 1 to 88 District 2 to 76 District 3 to 117 District 4 to 24 District 5 to 68 District 6 to 5 District 7 to 91

PAGE 45

29 3203264Km Figure 2-2. Digital elevation model (USGS 90 m) for FDOT District 2. To p o g ra p hic rid g es for this district were defined as 36 m or g reater. Topographic ridges for each District were assigned as follows: District 1-3, 7greater than 36 m District 5greater than 30 m District 4, 6n/a, insignificant Ecotones were derived from the Florida Fish and Wildlife Conservation Commission (FFWCC) land cover/habitat dataset. This dataset was based on Landsat satellite imagery at 30 m resolution, and included 22 basic types that were consolidated into the following 9 categories: Coastal resources Agriculture, grasslands, open prairie, and shrub and brushlands Pinelands, and mixed hardwood and pine forests Xeric habitats Hardwood hammocks and forests Open wetlands, and shrub wetlands

PAGE 46

30 Forested wetlands Open water Barren, urban, and exotic communities These habitat types were reclassified to reduce variety (noise) in the data, and to limit ecotones to transitions between major land-cover types. Six basic types were created: Urban, open water, and coastal resources Agricultural classes Pinelands, and mixed hardwood-pine forests Xeric habitats Forested wetlands, and hardwood hammocks Open wetlands, and shrub swamps After the reclassification process, neighborhood statistics were conducted. By executing a focal majority test (5x5 cell neighborhood) increased grouping was achieved that eliminated outlier cells. After which, a focal variety function (3x3-cell neighborhood) was performed to find margins between adjacent differing, habitat types. The resulting grid represented primary habitat ecotones. All areas with a value of one, indicating like adjacent habitat, were eliminated; the remaining values represent those areas where differing habitat types meet. The grid was then converted to a shapefile to define the relative size (area) of identified ecotones. Since the focus was on large-scale movement, anything smaller than 40 ha (approximately 100 acres) was deleted. The remaining areas were converted to grid format for use in analysis. In many cases, ecotones coincide with the location of riparian corridors; the other significant landscape feature used for dispersal, travel, or movement by wildlife. This landscape feature was determined from USGS hydrography. Riparian systems The USGS 1:24,000 scale hydrography database (line and polygon coverages converted to grids and combined) was used to identify certain hydrologic features

PAGE 47

31 including streams, lakes, and canals (Figure 2-3). These features were considered important as potential travel corridors or destinations. Base values were assigned according to importance as travel corridors and quality of habitat (Table 2-1). 32 0 3 2 6 4 K m Figure 2-3. Hydrography (USGS 1:24,000) for FDOT District 2. Open water features (e.g., lakes, streams, canals) were used from this dataset; wetlands were derived from water management district 1995 lan d -cover data. Habitat/land cover The FFWCC land cover/habitat dataset was based on Landsat satellite imagery at 30 m resolution and included 22 basic types that were consolidated into the following 4 categories (Figure 2-4): 1) Urban, barren, exotic communities, shrub and brushlands, and agricultural classes 2) Pinelands, and mixed hardwood-pine forests 3) Hardwood and tropical hammocks, forested, open and shrub wetlands, open water, and coastal resources 4) Xeric habitatsdry prairie, sandhill, and scrub

PAGE 48

32 32 0 3 2 6 4 K m Figure 2-4. Reclassified land cover/land use for FDOT District 2. Categories of development threat in descending order: xeric 4, wetlands 3, silviculture/mixed hardwood-pine forests 2, agriculture/urban 1. (Data source: Water Management Districts 1995) Category four, xeric habitats, was ranked the highest due to scarcity and high rate of conversion to development. The classes in category three were next highest due to presence of high biodiversity levels and use as wildlife refuge and travel corridors. Category two contains areas with good habitat value or potential for restoration; and was therefore ranked third. Category one includes all other land-cover types, considered to be low priority areas for wildlife habitat. Identified hotspots Hotspots were determined by the FFWCC. This is a data set (grid) representing biological diversity; created by aggregation of predictive habitat maps for wading birds,

PAGE 49

33 important natural communities, and 44 focal species (Figure 2-5). Categories of values that were used include: 3 4 focal species present 5 6 focal species present 7 + focal species present 32 0 32 64 K m Figure 2-5. Element occurrences (FNAI)/hotspots (FFWCC) for FDOT District 2. Element occurrences include federal and state listed species, bird rookeries and migratory bird congregations; hotspots are shown in four categories according to number of focal species present. Base values were assigned according to the number of focal species present (Table 2-1). For more information on the determination of hotspots, refer to Cox et al. (1994). Known locations of listed species State and federally listed species, bird rookeries, and migratory bird congregation areas were extracted from the Florida Natural Areas Inventory (FNAI) element occurrence database (Figure 2-5). Only records precise to 1,850 m or less were included

PAGE 50

34 in the analysis. Base values were set consistent with designated protection levels (Table 2-1). Each species location was buffered according to home range distances documented in the Florida Gap Analysis Project bibliography (FGAP 1998). The home range of the closest related species was used in cases where home range information was not available. A minimum home range of 100 m 2 was used consistent with the models minimum cell resolution. Many bird species were buffered in line with documented negative edge-effect-distance (when home range information was unavailable). Chronic road-kill sites Road-kill locations of two focal species (Florida black bear and Florida panther) were buffered by 150 m; requisite to perform analysis at the minimum 100 m cell resolution. Segments of state roads, passing through or adjacent to state parks, with documented road-kill problems were also buffered by 150 m (Crocodile Lake and Florida Key Deer National Wildlife Refuges in District 6 received the highest base value because of the presence of endangered species). Figure 2-6 displays these features for FDOT District 2. Base values were assigned according to relative significance (Table 2-1). Road projects Road projects identified in FDOT 5-Year Work Programs (1996 to 2000, 1999 to 2004) included the following categories (Figure 2-6): New construction, and bridge replacement Road widening, reconstruction, and additional lanes Road project locations were buffered by 150 m. Base values were applied according to significance and opportunity for mitigation. New roads and bridge replacements provide greater opportunity to reduce environmental impact through appropriate planning and the

PAGE 51

35 absence of constraints from existing facilities; and were therefore ranked higher than existing road projects. Figure 2-6. Documented road-kills/road projects for FDOT District 2. Road-kill data include that recorded by the FFWCC and state park personnel; Road p ro j ects are those scheduled from 1999 2004. 32 0 3 2 6 4 K m Public conservation lands Public conservation lands were selected from the 1998 Florida Department of Environmental Protection (FDEP), Conservation and Recreation Lands-Preservation 2000 (CARL/P2000) conservation-lands database. All jurisdictions were combined into one data layer; and assigned one base value reflecting public ownership (Figure 2-7). Strategic habitat conservation areas (SHCA) SHCAs were derived from various datasets including CARL/P2000, FFWCC proposed SHCAs, FNAI areas of conservation interest (ACI) categoriesA, B, and C,

PAGE 52

36 and the Nature Conservancy (TNC) ecological-resource-charette results. These datasets all reflect efforts from different groups to identify unprotected natural resource areas that warrant some level of protection. Figure 2-7. Conservation planning programs for FDOT District 2. These include FFWCC strategic habitat conservation areas, FNAI areas of conservation interest, CARL/P2000 existing and proposed conservation-lands, Florida statewide greenways planning project, and the Nature Conservancy planning charette. 32 0 32 64 K m Base values were divided into two categorieshigh and low priority. The high category includes CARL/P2000 proposed conservation lands, FFWCC proposed SHCAs, FNAI: ACI categories A and B, and TNC priority conservation areas. The low category includes FNAI-ACI category C and TNC conservation interest areas. These datasets possessed considerable overlap; and therefore were combined into one dataset to reduce redundancy (Figure 2-7). Base values are shown in Table 2-1.

PAGE 53

37 Greenway linkages Greenway linkages came from the FDEP, Florida greenway network, final priority rankings (Figure 2-7). Priority reflects the relative value to provide regionalor local-scale connections, and need for significant restoration. Regional-scale linkages included those connecting areas of national or statewide ecological significance. Local linkages were considered secondary connections that link local conservation areas with regional conservation areas. Base values for each type of linkage were derived from final priority rankings (Table 2-1): High priority Medium priority Category 3 7 final rankings (low priority) Modeling Process A rule-based model that applies user-determined weights to input-data-layers to generate suitability surfaces (Aspinall 1993) was developed in Arcview Spatial Analyst (refer to Appendix B for specific model parameters and development). Datasets were grouped into categories; weights for each category were averaged values (rounded up) of individual criteria from the questionnaire (Table 2-1; Appendix A): Biological features 7 Landscape features 6 Infrastructure 1 Public lands 3 (includes ranking of core conservation areas, multiplier = 3) Planning 5 Road-kill 9 The algorithm combines, sorts, assigns weights, and calculates information associated with each criterion (Figure 2-8). The result is a final priority layer based on user-input weights and base values for each criterion.

PAGE 54

iinnff r raasstt r ruuccttuu r ree ppuu b blliicc llaann d dss rrooaadd--kkiillllss ppllaannnniinngg sstt r raatteeggiieess llaann d dssccaappee ffeeaattuu r reess individual cell values criteria data layers combined categories for weighting b biioollooggiiccaall ffeeaattuu r reess Figure 2-8. Flow chart of analysis algorithm. Individual cell attributes for each data layer are assigned relative values. Like category data layers are combined and assigned appropriate weights. Weighted priority layer is the result of individual cell calculations for each data layer multiplied by its associated wei g htin g and added to the other la y ers. wweeii g ghhttee d d pp r riioo r riittiieess FFGGFF W WFFCC hhoott--ssppoottss FFNNAAII eellee m meenntt ooccccuu r r r reenncceess 33--44 ffooccaall ssppeecciieess 55--66 ffooccaall ssppeecciieess 77++ ffooccaall ssppeecciieess 38

PAGE 55

39 State road-project programming and scheduling in Florida is divided into seven districts within the state. State road-project budgeting is also apportioned according to these seven FDOT districts. As a matter of logistics, it was appropriate to perform the analyses according to these districts (Figure 2-9). Therefore, results of this analysis will reflect rankings or priorities by individual FDOT districts, not the State as a whole. Calculated priority layers for each FDOT district exhibited different value ranges. Reclassifying each cell value from one to seven (one being the highest value and seven being the lowest value) standardized final priority layers for each district. Priorities were assigned by dividing the total score by equal 20 unit intervals (i.e., the highest 20 values = 1, second highest 20 values = 2, third highest 20 values = 3...). Cells with scores of one to three were considered highly ranked. Figure 2-9. Florida Department of Transportation maintenance districts.

PAGE 56

40 Results The priorities determined by the model indicate significant focus on nationally and regionally significant conservation areas and riparian corridors. The number of road segments identified as priority areas varies between districts. In this study, a road segment is any continuous section of road that contains the same cell values from the analysis results. Identified priority road segments are displayed in Figures 2-10a through 2-10g, according to relative ranking for each FDOT district. Table 2-2 contains the number of road segments in each district for priority levels one to five. Total overall scores ranged from 84 to 320. The district with the greatest range of values (89 300) was District 6; the district with the smallest range in values (84 203) was District 4. District 6 had the fewest priority one road segments (3) and District 4 had the most (22). The district with the most sites ranked three or higher was District 4; the district with the fewest sites ranked three or higher was District 6. Table 2-2. Number of contiguous prioritized road segments by FDOT district FDOT district Priority 1 Priority 2 Priority 3 Priority 4 Priority 5 Total One 10 46 81 161 697 995 Two 4 7 25 156 1,402 1,594 Three 19 39 59 475 2,230 2,822 Four 22 190 653 1,386 2,193 4,444 Five 7 53 127 236 1,653 2,076 Six 3 8 25 50 57 143 Seven 7 6 48 539 1,904 2,504 Total 72 349 1,018 4,069 10,136 15,644 Table 2-3 contains the most significant sites in each district including the criteria that contributed their ranking. Roads that contain the most significant priority sites (scores of 1 5) are shown in Figure 2-11. Additional detail can be found in the final project report (Smith et al. 1998).

PAGE 57

41 Figure 2-10a. Road segment priorities in FDOT District 1. Rankings increase from low to hi g h as red shadin g becomes darkest.

PAGE 58

42 Figure 2-10b. Road segment priorities in FDOT District 2. Rankings increase from low to hi g h as re d s ha di n g beco mes d arkest

PAGE 59

43 Figure 2-10c. Road segment priorities in FDOT District 3. Rankings increase from low to hi g h as re d s ha di n g beco mes d ar kest

PAGE 60

44 Figure 2-10d. Road segment priorities in FDOT District 4. Rankings increase from low to high as red shading becomes darkest.

PAGE 61

45 Figure 2-10e. Road segment priorities in FDOT District 5. Rankings increase from low to high as red shading becomes darkest.

PAGE 62

46 Figure 2-10f. Road segment priorities in FDOT District 6. Rankings increase from low to hi g h as re d s ha di n g beco mes d arkest

PAGE 63

47 Figure 2-10g. Road segment priorities in FDOT District 7. Rankings increase from low to high as red shading becomes darkest.

PAGE 64

48 Table 2-3. Significant prioritized road segments by FDOT district FDOT district Highways Conservation areas Contributing criteria One SR 29, US 41 Everglades NP/Big Cypress NP flp, blb, rdkl, lsp, hsp, pub, hab US 27 from SR 70 south to Fisheating Creek blb, rdkl, lsp, hsp, shca, grn, flp I-75, US 41, SR 72, SR 776 Myakka River conservation area rdkl, lsp, hsp SR 60, SR 70, SR 700 Lake Wales Ridge and Kissimmee R SOR lsp, hsp I-75, US 41, SR 31 CM Webb WMA hab, pub, shca Two SR 19 Ocala NF at the Oklawaha River rdkl, rip, hab, pub SR 20, SR 100 Etonia Creek Carl rip, grn, shca, hsp SR 21, SR 16, SR 230 Camp Blanding pub, grn, lsp, hsp SR 121 Lake Butler WMA pub, grn, lsp, hsp SR 2, US 441, I-10, US 90 surrounding Osceola NF pub, hsp, lsp, rdkl US 441, I-75, SR 20, US 301 Paynes Prairie and Prairie, Orange and Lochloosa creeks rdkl, rip, shca, grn, pub, rdpr SR 105, SR 107, SR A1A, US 1, SR 301 Timuacan National Preserve pub, rip Florida Turnpike Extension Goethe State Forest rdpr, rcw, lsp, pub, hab SR 24 Cedar Key Scrub State Preserve hsp, lsp, hab, pub SR 20, SR 55, US 27 Aucilla, Wacissa and Econfina rivers rip, rdkl Three SR 20, SR 65, SR 377, SR 61, SR 267, SR 71, US 98 Apalachicola National Forest/St. Marks NWR and Aucilla WMA area rcw, rdkl, pub, rip, hsp, lsp SR 83, SR 85, SR 87, SR 189, SR 4, US 90, I-10 Eglin AFB/Blackwater Creek SF area rdkl, lsp, hsp US 98 Topsail Hill and Henderson SRA pub, hsp, lsp, hab Four US 27, I-75, SR 710, SR 786, SR 70, SR 76 within the WCAs, Dupuis Reserve, Cypress Creek CARL, East Coast Buffer SOR, JW Corbett WMA and Pal Mar CARL areas grn, pub, shca, lsp, hsp SR 60, SR 91 Blue Cypress and Fort Drum conservation areas grn, pub, shca I-95, US 1, SR A1A St. Sebastian River Buffer SP, N. Fork of St. Lucie River, and Indian River Lagoon Blueway lsp, grn, rip Five SR 19, SR 40, US 17 Ocala National Forest area and Lake George WMD project blb, rdkl, lsp, pub, hsp, rdpr US 92, I-4 Tiger Bay State Forest grn, pub SR 46, SR 44, SR 417 Wekiva River Basin Ocala greenway, Little Econlockhatchee and St. Johns River grn, rip, rdkl, shca, rdpr I-95, SR 46, SR 407, SR 50, SR 520, SR 528 Seminole Ranch/Tosahatchee SR pub, grn, rip, hsp, shca SR 44, I-75, SR 50 Withlacoochee R. and Green Swamp rip, grn I-95 St. Sebastian River Buffer SP and Indian River Lagoon Blueway rip, pub, shca, lsp Six US 1 between Key Largo and the Everglades lsp, rdkl, hsp, pub, amcr, rdpr US 1 Prop. addition to the Florida Key Deer NWR on Big Pine Key rdkl, flkd, lsp, shca, rdpr Seven US 19 adjacent to the Chassahowitzka NWR and Annutteliga Hammock rdkl, hsp, lsp, pub, shca the Suncoast Parkway adjacent to the Chassahowitzka NWR and Annutteliga Hammock hsp, lsp, pub, shca, rdpr

PAGE 65

49 Table 2-3 continued FDOT District Highways Conservation Areas Contributing Criteria Seven I-75, I-275, SR 54, US 41 through the Cypress Creek and Hillsborough River corridors rip, grn, pub I-75, SR 43 Little Manatee and Alafia Rivers SOR rip, grn Note: Abbreviations for criteria are as follows: American crocodile (amcr), black bear (blb), Florida key deer (flkd), Florida panther (flp), habitat/land cover (hab), hotspots (hsp), greenway linkages (grn), listed species locations (lsp), public lands (pub), red-cockaded woodpecker (rcw), riparian system (rip), road-kills (rdkl), road projects (rdpr), and SHCAs (shca). I-95 024681012141618I-75US 98US 1I-10SR 20US 41US 90Fl TpikeSun PkwyUS 441SR 50SR 44 Figure 2-11. Number of significant priority sites by route number. Explanation of results is most revealing with reference to the major contributing criteria used in the analysis. Regional greenways within the State that are intersected by high priority road segments (Figure 2-12a and 2-12b) include: The south Florida ecosystem upper St. Johns River basin greenway, The middle St. Johns River basin Ocala National Forest greenway system, Ocala National Forest Osceola National Forest/Pinhook Swamp/Okefenokee Swamp greenway, Osceola National Forest Apalachicola National Forest greenway system, Green Swamp/Withlacoochee State Forest Big Bend conservation area greenway system, The south Florida ecosystem Green Swamp greenway system, and Eglin AFB Blackwater Creek State Forest greenway.

PAGE 66

50 Conecuh National Forest, Alabama Eglin Air Force Base Apalachicola National Forest Osceola National Fores t Ocala National Forest Okefenokee National Wildlife Refuge, Georgia Figure 2-12a. Road segment priorities and greenways in panhandle Florida. Prioritized ecological hotspots are shown in black for FDOT districts 2 and 3. These are overlaid upon existing and proposedstate conservation-lands and priority greenway linkages.

PAGE 67

51 Big Bend Green Swam p Greenwa y Ocala National Forest Middle St. Johns Basin Greenway South Florida Ecosystem Green Swamp Greenway Everglades Middle St. Johns Basin Greenway Figure 2-12b. Road segment priorities and greenways in peninsular Florida. Prioritized ecological hotspots are shown in black for FDOT districts 1, 4, 5, 6 and 7. These are overlaid upon existing and proposed state conservation-lands and priority greenway linkages.

PAGE 68

52 Osceola National Fores t Eglin Air F o r ce Ba se Apalachicola National Forest Aucilla Ri ve r Ba s in Rice Creek Ar e a Figure 2-13a. Black bear road-kills and greenways in panhandle Florida. Black bear road-kills, 1976-1999, are shown in black for FDOT districts 2 and 3. These are overlaid upon existing and p ro p osed state conservation-lands and p riorit y g reenwa y linka g es.

PAGE 69

53 Ocala National Forest Chassowitzka National Wildlife Refuge Tiger Bay State Fores t Fisheating Creek Area Big Cypress National Preserve, Fakhatchee Strand State Preserve Figure 2-13b. Black bear and Florida panther road-kills and greenways in peninsular Florida. Black bear, 1976-1999 (in black), and Florida panther, 1973-1996 (in red), road-kills are shown in for FDOT districts 1, 4, 5, 6 and 7. These are overlaid upon existing and proposed state conservation-lands and priority greenway linkages.

PAGE 70

54 Jacksonville Tallahassee Pensacola Gainesville Figure 2-14a. Road projects and greenways in panhandle Florida. State-funded road projects through 2004 are shown in red for FDOT districts 2 and 3. These are overlaid upon existing and proposed state conservation-lands and priority greenway linkages.

PAGE 71

55 Tampa-St. Petersburg Miami-Ft. Lauderdale Orlando Ft. Myers Daytona Bch Figure 2-14b. Road projects and greenways in peninsular Florida. State-funded road projects through 2004 are shown in red for FDOT districts 1, 4, 5, 6 and 7. These are overlaid upon existing and proposed state conservation-lands and priority greenway linkages.

PAGE 72

56 Of 15,644 road segments identified with priorities one to five, 12,828 (82%) were associated with identified greenways; and 4,019 (26%) were located within existing conservation lands. Of these road segments, 27 were ranked 1 st 150 were ranked 2 nd and 389 were ranked 3 rd On proposed CARL project lands there were 2,469 (16%) prioritized road segments. Numbers of significantly ranked, road segments on proposed conservation lands include: 13 (1 st ), 63 (2 nd ), and 181 (3 rd ). Major riparian systems where multiple priority ecological hotspots on road segments were identified include the following rivers and their tributaries: St. Johns Suwannee Aucilla/Wacissa Withlacoochee Peace Myakka Kissimmee Apalachicola Choctawatchee Escambia Yellow St. Marks Black-bear road-kills significantly influenced results at the statewide level. This was due to the high weight placed on the criterion and the wide distribution of the species. The most significant priority sites where Florida black-bear road-kills were located (Figure 2-13a and 2-13b) included: SR 19 and SR 40 in Ocala National Forest (NF) SR 44 and SR 46 crossing the Ocala Wekiva habitat corridor US 441 adjacent to Osceola NF US 41, I-75 and SR 29 in Big Cypress National Preserve US 19 adjacent to Chassahowitzka NWR US 27, SR 59 and US 98 in Jefferson and Wakulla counties US 27 in Highlands county SR 85 and SR 87 in Eglin AFB

PAGE 73

57 SR 65 and US 319 in Apalachicola NF SR 71 in Gulf county Ninety-five major, state-sponsored, road projects coincide with highly ranked ecological hotspots on highways (Figures 2-14a and 2-14b; Table 2-4). These projects were identified from FDOT district workplans for 1996 2000 and 1999 2004; and include 46 bridge replacements/new construction, 43 road expansions/reconstruction, and 6 new roads (FDOT 1999, 1996). Table 2-4. Road projects identified at highly ranked, ecological hotspots on highways FDOT District Route No. Ecological Feature Project Type 1 I-4 Green Swamp Multi-lane Reconstruction 1 SR 29 Big Cypress NP, Fakahatchee Strand SP Install Box Culvert Wildlife Underpasses 1 SR 29 Canal, S of Fisheating Creek SOR Bridge Replacement 1 SR 29 Canal, S of Fisheating Creek SOR Bridge Replacement 1 SR 29 Canal, S of Fisheating Creek SOR Bridge Replacement 1 SR 29 Canal, S of Fisheating Creek SOR Bridge Replacement 1 SR 29 south of CR 847, Corkscrew Regional Ecosystem Bridge Replacement, Road Reconstruction 1 SR 60 Kissimmee River SOR Bridge Replacement 1 SR 70 Horse Creek Bridge Replacement 1 SR 70 Horse Creek Bridge Replacement 1 SR 776 Myakka River Corridor Add Lanes/Reconstruction 1 SR 951 Rookery Bay Add Lanes, New Bridge 1 US 41 Big Cypress NP Bridge Replacement 2 Chaffee-Branen Field Rd Jennings SF, Black Creek Area New Road Construction 2 FL Turnpike Ext. Goethe SF New Road Construction 2 I-95 Julington/Durbin CARL,12-mile Swamp, Fish Swamp, Pellicer Creek Add Lanes/Resurfacing 2 I-95 Timucuan EP, Nassau River Corridor Add Lanes/Reconstruction 2 SR 105/SR AIA Ft. George Inlet, Timucuan EP Bridge Replacement 2 SR 121 Santa Fe River Corridor Bridge Replacement 2 SR 20 Fowler's Prairie Add Lanes, PD & E / EMO Study 2 SR 20 Lochloosa Creek, Prairie Creek, Newnans Lake CARL Add Lanes/Reconstruction 2 SR 20 Rice Creek, Etonia Creek CARL Bridge Replacement 2 SR 207 Deep Creek CARL, Fish Swamp, Moultrie Creek, Trestle Bay Swamp Add Lanes/Reconstruction 2 SR 47 Santa Fe River Corridor Bridge Replacement 2 SR 500 Devils Hammock WMA to Watermelon Pond Add Lanes/Reconstruction 2 SR 9A Gum Swamp, Big Island Swamp New Road Construction

PAGE 74

58 Table 2-4. continued Route No. Ecological Feature FDOT District Project Type 2 US 17 Lofton Creek Bridge Replacement 2 US 17 Nassau River Salt Marsh Bridge Replacement 2 US 17 Nassau River Salt Marsh Bridge Replacement 2 US 301 Timucuan EP, Cary SF Add Lanes/Reconstruction 2 US 90 Intra-coastal Waterway Bridge Replacement 2 US 90 Pablo Creek Add Lanes/Reconstruction 3 SR 71 Cypress Creek Bridge Replacement 3 SR 71 West Arm Creek Bridge Replacement 3 SR 83 Eglin AFB Add Lanes/Reconstruction 3 SR 87 Clear Creek Add Lanes/Reconstruction 3 SR 87 Eglin AFB, Yellow River Add Lanes/Reconstruction 3 US 27 Ochlocknee River Bridge Replacement 3 US 27 Ochlocknee River Corridor Bridge Replacement 3 US 90 Choctowatchee River Bridge Replacement 3 US 98 St. Marks River Bridge Replacement 3 US 98 Topsail Hill SP, Pt. Washington SF Add Lanes/Reconstruction 4 SR 60 Blue Cypress Conservation Area Add Lanes/Reconstruction 4 SR 70 Cypress Creek Branch Bridge Replacement 4 SR 70 Cypress Creek CARL Bridge Replacement 4 SR 70 Cypress Creek CARL Add Lanes/Reconstruction 4 SR 70 Cypress Creek CARL Bridge Replacement 4 SR 70 Relief Canal, Cypress Creek CARL Bridge Replacement 4 SR 70 Relief Canal, Cypress Creek CARL Bridge Replacement 4 SR 76 South Fork St. Lucie River Add Lanes/Reconstruction 4 SR 811 Lake Worth Creek Add Lanes/Reconstruction 4 US 1 North Savannas, Indian River Blueway Add Lanes/Reconstruction 5 FL Turnpike Ext. Marion/Levy 1 SOR, Cross-FL Greenway, Gum Slough New Road Construction 5 I-4 Reedy Creek Add Lanes 5 I-4 St. Johns River AP Bridge Replacement 5 I-4 St. Johns River AP Add Lanes 5 I-4 Tomoka River Area Add Lanes 5 I-95 Pellicer Creek AP, Graham Swamp Add Lanes 5 SR 100 Creek to Crescent Lake Bridge Replacement 5 SR 100 Flagler County Greenway, Black Point Swamp Add Lanes 5 SR 11 Middle Haw Creek Bridge Replacement 5 SR 19 Palatlakaha Creek, Lake County WCA Bridge Replacement 5 SR 40 Ocala NF Add Lanes 5 SR 40 Silver River SP, Ocala NF Add Lanes 5 SR 429 Wekiva Basin GeoPark New Road Construction 5 SR 44 CSX RR, Spruce Creek Bridge Replacement 5 SR 44 Deep Creek Swamp, Tiger Bay, Spruce Creek Swamp, Bicentennial Conservation Park Add Lanes

PAGE 75

59 Table 2-4. continued FDOT District Route No. Ecological Feature Project Type 5 SR 44 Lake Panasoffkee SOR, Gum Slough SOR, Withlacoochee River Add Lanes 5 SR 46 Lake Jessup Bridge Replacement 5 SR 520 St. Johns River, Jim Creek, Tosahatchee Creek/SP, Seminole Ranch Add Lanes 5 SR 60 C-54 Canal, Ft. Drum Conservation Area Bridge Replacement 5 SR 60 Kissimmee River SOR Bridge Replacement 5 US 1 Korona Canal, Tomoka River Area Bridge Replacement 5 US 1 Tomoka River Area Bridge Replacement 5 US 17/92 Reedy Creek Bridge Replacement 5 US 192 Reedy Creek Add Lanes 5 US 192 Three Forks SOR Conservation Area, St. Johns River, Crabgrass Creek Add Lanes 5 US 441 Lake County WCA, Lake Harris, Lake Eustis Add Lanes 6 US 1 Big Pine Key, Coupon Bight Add Lanes 6 US 1 Everglades NP, Crocodile Lake NWR, Jewfish Creek Add Lanes, Bridge Replacement, Multi-lane Reconstruction 7 I-275 Big island Gap, Tampa Bay Add Lanes/Reconstruction 7 SR 200 Jordan Ranch, Withlacoochee River Add Lanes/Reconstruction 7 SR 39 Blackwater Creek SOR Bridge Replacement 7 SR 44 Moccosin Slough, Withlacoochee River Add Lanes/Reconstruction, Bridge Replacement 7 SR 50 Withlacoochee River Bridge Replacement 7 SR 52 Starkey WMA, Pithlachascotee River Add Lanes/Reconstruction 7 SR 54 Cypress Creek Add Lanes/Reconstruction 7 Suncoast Pwy Annuteliga Hammock New Road Construction 7 US 19/98 Cross-FL Greenway New Bridge 7 US 41 Pasco 1 SOR, Five-mile Creek Add Lanes/Reconstruction 7 US 98 Annuteliga Hammock Add Lanes/Reconstruction Discussion Determinants of Model Priorities Five primary factors were instrumental in the model. The following discussion highlights their importance and explains their role in statewide conservation planning. Greenways The Florida Statewide Greenways Planning Project (Carr et al. 1999) was based on several criteria used in this model (e.g., riparian linkages, core areas of conservation, existing and proposed public conservation-lands, SHCAs, and land cover/land use).

PAGE 76

60 When priorities one to five from each district are compared to the greenways final priority results (Figure 2-12a and 2-12b; also see Smith et al. 1998), it is apparent that ecological greenways were an important element in identifying road segments in areas of high ecological value. Roads intersect most ecological greenways. As part of the design of a functionally integrated ecological network, many of these intersections will require some type of mitigation measures such as wildlife underpasses. Many of these sites are identified as high priorities in this analysis (Table 2-3; Figure 2-12a and 2-12b). Much discussion surrounding costs to construct underpasses has occurred. The common conclusion being that too few monetary resources exist to install underpasses at all of these sites. The purpose of this research was to formulate a prioritization method to determine those sites in greatest need (so that limited resources could be used most efficiently). Even with prioritization, the multiple sites selected probably require more funding than is available to construct the number of wildlife underpasses needed; and that keeps up with land acquisition proposals and encroaching development. As much as 2.28 million acres have been proposed for purchase under various state land acquisition programs (FGP 1999). FDOT is reluctant to provide funding for construction of wildlife underpasses on private lands when future land-use is uncertain. In these cases, costs for construction of mitigation structures (such as wildlife underpasses) should be included as part of the total purchase costs; as part of the ecological network plan for Florida. This could expand construction costs across multiple funding sources. For example, FDOT could focus on existing public conservation-lands and riparian corridors when allocating their resources for underpass

PAGE 77

61 construction. Whereas, FDEP could schedule funds for underpasses on acquired CARL lands in conjunction with the purchase of the property. Such an approach could help facilitate a quicker realization of the greenways visionan interconnected ecological network that serves as a statewide conservation system for Florida. Riparian systems Riparian systems were considered important movement corridors in this project. Several studies have investigated and described the function of riparian systems as habitat and movement corridors (Rosenberg et al. 1997, Baschak et al. 1994, Malanson 1993, Noss 1993, van Buuren et al. 1993, Schaefer et al. 1992, Harris 1985, and Forman 1983). In the Netherlands, riparian corridors are considered vital ecological linkages for movement of species (Lammers et al. 1996). In Florida, the Wekiva and Suwannee Rivers have been identified as critical greenway linkages by multiple authors (Noss 1993, Smith 1993, and Harris 1988b). Riparian corridors are natural features that provide functional connectivity between landscape elements. Sufficiently wide riparian systems contain gradients of ecosystems (from aquatic to xeric habitat types) that can facilitate movement by various habitat dependent species (Harris et al. 1996). Riparian systems are represented in four individual criteria used in the analysis: riparian, habitat/land cover, SHCAs, and greenway linkages. As is the case with the greenway criterion, riparian systems are an integral part of Floridas statewide conservation system. They act as refuges and travel corridors, and provide sources of food and shelter for various mammals, herpetofauna, fish, and birds (Darveau et al. 1995, Spackman et al. 1995, van Zadelhoff and Lammers 1995, Noss 1993, Smith 1993, Schaefer et al. 1992, and Dodd 1990).

PAGE 78

62 Terrestrial connections along these river corridors are essential at road intersections (to provide connectivity for terrestrial vertebrates). It is imperative that bridge replacements be programmed to include accommodations for terrestrial connections adjacent and parallel to the watercourse. These connections should include native vegetation consistent with the present community type. The St. Johns, Suwannee, Peace, Withlacoochee, and Apalachicola river systems are part of major statewide greenway linkages. Functional terrestrial underpasses are necessary where roads intersect these. Some already have adequate structures that serve as underpasses; examples include I-10 at the Suwannee River, I-10 at the Apalachicola River, SR 40 and SR 19 at the Ocklawaha River, US 90 and I-10 at the Little St. Mary's River, US 129 at the Suwannee River, and US 301 at the Santa Fe River. Others may require expansion or other minor modifications to the structure; examples include I-75, US 441, and US 27 at the Santa Fe River, US 301 at Orange Creek, and US 19 at the Suwannee River. Hotspots Location of focal species hotspots and listed species element occurrences typically signify high priority wildlife conservation areas. The hotspots dataset also acts as the basis for the FFWCC SHCAs used in this analysis. The hotspots dataset was created by overlaying habitat maps of 44 focal species and subdividing the composite map into three broad categories: 1) 3 4 species, 2) 5 6 species, and 3) 7 + species (Cox et al. 1994). These categories represent areas where existing habitat conditions can support the aforementioned number of focal species. Cox et al. (1994) described category one as typical of large forested habitat areas that serve

PAGE 79

63 many large-scale landscape functions (such as maintenance of air and water quality, and faunal movement between preserves). Category two and three serve as habitat areas for generalists as well as habitat-specific species. According to Cox et al. (1994), the latter two categories require special attention in conservation planning, since many of these areas are critical for the survival of certain rare species. Biodiversity hotspots played a major role in final rankings of road segments. Presence of hotspots and/or inclusion in SHCAs coincides with the location of most major core conservation areas. Thus, most of these locations are either in public ownership or proposed for acquisition. Updated versions of the hotspots and land-cover datasets that were recently released, were not available at the time this research was conducted. Road-kills Road-kill was included in the analysis as an independent category. It was originally included in the biological features category, and only showed a strong presence when combined in certain cells with hotspots and listed species element occurrences; therefore diminishing the significance of the road-kill data. Being the highest ranked criterion, it was decided that chronic road-kill sites should be separated from other biological data. When road-kill sites are clustered or contain multiple kills, they deserve extra consideration relating to their potential importance as travel routes for listed species. This dataset needs to be expanded to include more listed and non-listed species. The quality and accuracy of the existing data was good regarding the three listed species availableFlorida black bear, Florida panther, and Florida key deer.

PAGE 80

64 Road projects It is important to identify significant ecological hotspots that correspond with planned road projects. Through these opportunities, construction of wildlife underpasses or other mitigation measures can be facilitated. Such pre-planning can reduce costs incurred when engineers must retrofit existing roads. Road projects are conducted over a series of phases: 1) environmental assessment, 2) planning, design, and engineering, 3) rightofway development, 3) road construction, and 4) maintenance. Ideally, information from this study would be most useful if provided prior to phase two, when mitigation measures could be included as part of the original highway design. With this in mind, time is a critical factor for implementation. Road projects used in this analysis (completed in 1999) were identified from FDOT workplans from 1996 2004 (Table 2-4), and are therefore almost all completed at this writing. This means that the process must be updated repeatedly to maintain a current listing of proposed road projects. Data Limitations and Development of other Criteria Certain limitations of available data affected the capability of the model to derive more accurate results. For instance: Accuracy of listed species locations could be improved with new inventories Land-cover data were from 1994 satellite data; a more current version that is being compiled was not complete at the time of this study With availability of more recent land-cover data the SHCAs, hotspots, and habitat ecotones (based on the 1994 land-cover data) could be updated Road-kill databases could be expanded to include other species such as white-tailed deer and certain listed species not currently available Future iterations of this application should include a few additional criteria that might be developed from existing and new data sources to further refine results. Two potential new criteria utilize previously unavailable census and land-use data to identify

PAGE 81

65 at-risk, high-growth areas and high-intensity land-uses. Three other improvements might include additional components to existing criteria. These improvements include: Establishing a new criterion that utilizes 1990 and 2000 census-block data to identify high-growth areas (Figure 2-15) and future land-use to determine areas at greatest risk for habitat loss Establishing a new criterion that identifies and demotes areas consisting of high-intensity land-uses (Figure 2-16) from three indicators census (housing-unit density), updated land cover (urban, mining, intense agriculture, and exotic plant communities) and USGS 1:24,000 roads (high road densities) Utilizing 1:24,000 hydrography and updated land cover to identify ephemeral areas in high quality habitat areas, adjacent to road right-of-ways, to apply criterion no. 9 for herpetofauna Identifying severe slope and ravine areas, typical of northwest Florida creek systems, and other critical features from USGS topography, DEMs, hydrography, and updated land-cover data (Figure 2-17) Conducting more in-depth application of topographic and habitat gradient information (e.g., slope/aspect and habitat juxtaposition) to improve the ability to predict likely movement corridors; in the current analysis its application was limited to two parameters (topographic highs and lows and large-scale habitat ecotones) Model Priorities The process used in this model reduces variance in final scores by combining criteria into categories. Final scores were slanted toward road-kill, planning, biological and landscape features. This was due to higher weightings for road-kill (listed species road-kill sites), planning (SHCAs and greenways), biological features (hotspots, road-kill and listed species locations), and landscape features (gradients, riparian, and habitat/land cover); and lower weightings for public (public lands), and infrastructure (road projects). As a decision-support model, these preferences were cognizantly made. Although the latter two categories may have had some bearing on final results, most highly ranked cells are due to high values from the former four categories. Since the intent was to focus on biological components of natural landscapes, the process seemed to work as expected.

PAGE 82

66 Figure 2-15. Human population growth rates in Florida from 1990 2000. Census tract boundaries from the 1990 and 2000 U.S. Census were normalized so that population growth for the decade could be determined at the census tract level. To identify areas at high risk to development, it is suggested that future land-use data from local comprehensive plans (currently unavailable post 2000) be combined with census information. However, certain limitations are of interest. By combining criteria, a reduction in variance occurs and the intonation caused by individual criteria from each category is lost. In other words, the impact of each individual criterion is tempered by the other criteria within the respective group.

PAGE 83

67 Figure 2-16. Identification of intense land-uses. Land uses shown here consist of four categories that include all urban lands, intensive agriculture lands (e.g., vegetable crops, sugar cane, citrus), and mining lands. This figure displays 1994 Water Management District (WMD) land-use, this data is currently being updated to 1999 by individual WMDs. Housing unit data from the U.S. Census and road densities from the USGS 1:24,000 roads should be combined to create a development intensity data layer that could serve as a criterion to reduce the priority of these areas for road mitigation for conservation purposes. For example, each criterion was weighted individually for the initial trials of the model; the consequence that results were skewed to whichever criterion had the highest weighting. The original test run resulted in high values for road-kill sites because they were ranked the highest in the survey questionnaire. Contrarily, the effect of a single criterion on site rankings is diminished in the grouped version of the model (presented here). This was most appropriate, since the goal was to design the model so that one criterion could not overshadow all others. As a result, the model has greater balance and selects and ranks road segments with the highest ecological significance.

PAGE 84

68 Figure 2-17. Identification of severe slope/ravine areas typical of northwest Florida creek systems. Natural isolation of steephead creek corridors was created from erosion over geologic time. Construction of certain road designs (using fill instead of bridging ravines) in this area may result in range severance of local salamander populations that travel along t hese creek corridors. Exceptions to the groupings were public lands, road projects, and road-kills that still function independently. Public lands and road projects did not fit into the other group classifications; and therefore were placed in independent categories. The justification for an independent road-kill category was explained above. Conclusion The results pinpoint efforts to alleviate the impacts of highways and highway traffic by targeting adjacent, environmentally sensitive areas at specific highway locations. Through identification and prioritization of ecological hotspots on highways,

PAGE 85

69 the FDOT can program mitigation measures to counter negative impacts on wildlife and wildlife habitat. Potential for more accurate results exist with future refinements to the priority algorithm and updates to existing data layers. The interactive Arcview scripts provide a simple platform (Appendix B) for end users (e.g., transportation planners) to update the findings and focus on specific areas of interest. Special attention should be given to those priority sites identified that include proposed road projects and suitable existing structures.

PAGE 86

CHAPTER 3 EXISTING INFRASTRUCTURE IN FLORIDA Introduction The information presented here represents phase II of the research conducted for FDOT (designed to integrate transportation planning with statewide conservation objectives). Phase I (Chapter 2) began with the development of the aforementioned algorithm that identified and prioritized ecological hotspots on roads. Tasks performed for phase II included: 1) field verification of the computer model results; 2) inventory and ecological characterization of roadway segments identified in the prioritization model (levels 1 3; and in special cases, levels 4 5); and 3) evaluation of wildlife-passage functions of existing structures and recommendations for mitigation/restoration. Parameters for Field Site Evaluation and Infrastructure Inventory Site characterization provides valuable information that enhances efforts to implement greenway projects, by evaluating the need for underpass connectors or other less intensive retrofits. It increases FDOT's ability to quickly address mitigation measures needed at high-priority, ecological hotspots on highways. Ground-truth surveys identified several features that included presence of existing structures (bridges, culverts, etc.), their dimensions and composition, roadway characteristics (ROW width, number of lanes, width of paved surface, etc.), description of surrounding landscape features, identity of associated aquatic features, and signs of present animal use. 70

PAGE 87

71 Determining Appropriate Measures for Mitigation Inventory and evaluation of specific ecological characteristics of prioritized highway segments was used to provide general recommendations, for the appropriate type and location of mitigation measures (e.g., wildlife underpasses, fencing, bridge widening, culvert designs, special r-o-w designs, plantings, signs, and speed controls). At many sites, suitable bridges or culverts already exist that require only minor directional fencing or vegetative plantings (to enhance use by wildlife and to provide connectivity between adjacent areas). Examples include the existence of wide, floodplain bridges constructed at stream intersections and abandoned, railway bridges within existing conservation areas. Several factors should be considered in the design of roadways and wildlife crossing structures. Topographic relief can be used toward enhanced design and placement of underpasses that follow ecological flows and landscape patterns. Proper management of right-of-way and adjacent vegetation is an important factor in the design of wildlife crossing sites (Clevenger and Waltho 2000, Hewitt et.al. 1998, Tewes and Blanton 1998, Friedman 1997, Land and Lotz 1996, Roof and Wooding 1996, and Singer and Doherty 1985). Right-of-way and road corridor design should include sufficient clearance distance from the roadway, and reduction in severity of curves and slopes of hills, to increase driver visibility. Appropriate native vegetation at entry points to underpasses provides cover for species intolerant of open areas, and safety to species susceptible to ambush predators. Clevenger and Waltho (2000) identified proximity to human activity or influence, as the most critical factor in determining species performance ratios at wildlife underpasses in Banff National Park, Alberta, Canada.

PAGE 88

72 Site Inventories, Assessments and Recommendations Each field site was identified through previous GIS modeling (Chapter 2) as high priority interfaces (ecological hotspots) between state highways and critical state and regional conservation resources. The field study involved visits to 1,232 sites statewide. Recommendations for improvements were provided for 140 separate road segments containing prioritized sites. Measures of Model Accuracy, Site Inventory and Ecological Characterization A qualitative assessment was performed that compares land-cover characteristics collected from field surveys to satellite land-cover data (Water Management Districts 1995) used in the modeling process. Accuracy of the data used in the prioritization process (Chapter 2) was evaluated to assess the frequency that the model correctly predicted locations of ecological hotspots on highways. State level For the entire state, 15,644 road segments were prioritized (Chapter 4); field surveys included 989 sites (sample size equal to 6% of those sites ranked 1 5) from 65 of 67 counties. An additional 243 sites visited were unranked in the model, but included significant physical features (e.g., bridges or other potential wildlife crossing structures) that were not considered in the prioritization process. Land cover, derived from 1994 Landsat satellite data, was compared with recorded landscape characteristics for the 989 field sites statewide. Site verification revealed 115 misclassifications in the development classes. Sixty sites should have included urban (15) and exotic communities (46). Another 55 sites did not include rural residential; however, building density was likely too low, in most cases (e.g., single

PAGE 89

73 buildings sparsely arranged), to be detected at the 30 m resolution of the satellite data, or were simply classified as agriculture (that includes the farmhouse and associated buildings, etc.). Because of limitations in resolution these latter discrepancies were not included in the data assessment. Therefore 61 misclassifications from 989 sites represent a 6% error rate in land-use representation. Further analysis was conducted to determine the potential effect of this error rate on the prioritization process. A comparison matrix (comprised of site rankings and scores for nine land-use features) that reflects extent of human disturbance and habitat degradation was created to measure accuracy of those sites selected as priorities. Land-use features considered and their respective scores include: man-made canals (Figure 3-1), railroads, power-lines, fences, access roads (Figure 3-2), or trails (1); presence of invasive plants (Figure 3-3), local human-use parks (Figure 3-4), low-intensity agriculturepasture and pine plantation (3); low-density developmentsparse residential, single retail buildings, etc. (5); high-intensity agriculturemining operations, row crops, citrus, etc. (7); and high-density developmentstrip commercial, planned unit developments, industrial parks, etc. (10). For purposes of this analysis, any disturbance value of 4 or less was considered minor; and therefore was an acceptable error level in the prioritization process. Percentage-of-disturbance scores equal-to or less-than 4, for each priority level, are shown below: priority-one% (n = 49) priority-two% (n = 96) priority-three% (n = 144) priority-four% (n = 222) priority-five% (n = 478)

PAGE 90

74 Figure 3-1. One of the many canals adjacent to roadways in south Florida. Primary function of these canals is water management and flood control. They also compound the barrier and edge effects caused by roads for many species of wildlife. Figure 3-2. One of the many off-road access points. Throughout the state many primitive and substandard roads are present that provide access to public and private conservation areas. Some of these roads are managed for, but many are not. These elements add to potential disturbance for wildlife and ecological processes.

PAGE 91

75 Figure 3-3. Spread of invasive plants. Roads are commonly known to advance the spread of invasive plants. In south Florida, three species are now quite common, Brazilian p e pp er ( shown above ) Melaleuca and Australian p ine. Figure 3-4. Active recreation in resource areas. In many conservation areas, public access parks are provided for outdoor recreation. Many of these are located near roads to provide boat access. Depending on intensity of human use allowed, disturbance to adjacent area can be substantial.

PAGE 92

76 Figure 3-5 displays percent of the mean number of field sites measured on a scale of human disturbance (from 17 to 0; higher to lesser disturbance), for all model priority rankings of one to five. Based on field site disturbance scores of 4 or less, the prioritization process would, on average, accurately identify ecological hotspots on highways 87% of the time, with a 95% confidence level. 0510152025303540455055606517131211109876543210Disturbance Score% Note: Error bars represent 95% confidence intervals Figure 3-5. Percent of the mean number of field sites measured on a scale of human disturbance (17 = highest disturbance, 0 = lowest disturbance). A disturbance score of four or lower would be considered minor. As expected little difference in disturbance scores is evident between priority rankings for field sites. Of 989 total sites, on average, 10% (97) had scores g reater than 5 and 87% (856) had scores less than 5. The statewide field inventory and characterization of ecological hotspots on highways (Schaefer and Smith 1999) is presented according to FDOT districts (Figure 2-9).

PAGE 93

77 District one An inventory of 222 sites on 22 different highways was conducted in District 1 (Figure 3-6). Of the locations surveyed, 6 were priority-one sites, 25 were priority-two sites, 13 were priority-three sites, 43 were priority-four sites, and 82 were priority-five sites. Fifty-three of the sites inventoried were not ranked. Existing structures found among these sites included 5 ecopassages, 78 bridges, 37 concrete box culverts, 35 concrete pipes, and 4 corrugated steel pipes. Sixty-three sites did not have any drainage-control structures. Of all sites checked, 59 were associated with the Big Cypress Area (includes the Big Cypress National Preserve, Florida Panther National Wildlife Refuge, Fakahatchee Strand State Preserve and Golden Gate Estates CARL), 29 cross the Peace River or its tributaries, 26 were within existing or proposed conservation-lands associated with the Myakka River, 14 were associated with C.M. Webb Wildlife Management Area, 10 were within the Lake Wales Ridge conservation area, 6 were adjacent to Collier-Seminole State Park, 6 were found within Charlotte Harbor Flatwoods CARL area, and 3 each were adjacent to Highlands Hammock State Park, Fisheating Creek, and the Kissimmee River. The 6 priority-one sites surveyed contained the following adverse characteristics: 5 of 6 sites had adjacent man-made canals, one site exhibited invasive Brazilian pepper Schinus terebinthefolius, and another was adjacent to some strip development and a golf course. It should be noted that the canals were water control structures within Big Cypress National Preserve and Florida Panther National Wildlife Refuge. Signs of animal activity included multiple black-bear road-kills at 2 sites, 6 sites had single black-bear road-kills, and another site had one Florida panther and one black-bear road-kill.

PAGE 94

78 Figure 3-6. District 1 field verification sites. Of the 995 ecological hotspots on roads identified in the priority modeling process, 222 field sites were inventoried and landscape characterization performed. Recommendations for corrective measures necessary to maximize permeability for wildlife were provided on the numbered locations displayed (Appendix C).

PAGE 95

79 Characteristics negatively influencing ecological conditions at the 25 priority-two sites included: adjacent man-made canals at 18 sites, substantial development at one site, some low-intensity grazing lands at another site, presence of Brazilian pepper or Melaleuca Melaleuca quinquinervia at 5 sites, off-road vehicle access at 4 sites, and open trash receptacles at one site that has a road-side picnic area. Three sites had multiple black-bear road-kills, 15 had single black-bear road-kills, and another 3 sites had records of Florida panther road-kills. Habitat quality, other than the exceptions mentioned above, was quite good. Twenty-one of 25 sites are within the Big Cypress area. On-site factors counterproductive to conservation efforts at the 13 priority-three sites include: presence of parallel canals at 8 sites, Brazilian pepper or other invasive plants at 2 sites, access roads at 3 sites, scattered rural-development at 3 sites, minor pasture lands at 2 sites and citrus at one other site. Two sites contained multiple black-bear road-kills, 7 had single bear road deaths, one site had 2 road-killed Florida panthers, and an unidentified dead snake was found at another. The 43 priority-four sites visited contained the following adverse characteristics: 18 sites had adjacent man-made canals, 9 sites exhibited presence of invasive plant species, 5 sites had substantial residential and/or commercial development, 2 sites had single, combined business-residences nearby, and 4 sites had off-road vehicle access. Twelve sites had single, black-bear road-kills, one site had three bears killed, and another site had an unidentified dead turtle on the road. The 82 priority-five sites included the following negative influences: 13 sites had adjacent man-made canals, 7 sites contained the invasive plants Brazilian pepper and Melaleuca, 12 sites had some low-intensity pasture, 1 site had citrus development and 3

PAGE 96

80 sites had pine plantations, 8 sites had significant development adjacent or nearby including 2 housing developments, 3 mining operations and a prison, 5 sites had minor development sites including a marina, 2 local parks, a cemetery and a tourist shop, and 3 sites had off-road vehicle access. Seven sites had black-bear road-kills, 1 site had two dead alligators and one dead aquatic turtle, a dead gopher tortoise was found on the road at another site, 2 sites had road-killed snakes; one was identified as a red rat snake. District two Two hundred and eighty-eight sites were surveyed on 35 different roads (Figure 3-7). Number of field sites visited by model priorities of one to five in District 2 was 5, 5, 15, 51 and 143 respectively. An additional 69 significant, but unranked sites were also inventoried. Structures of various size, present at these sites include: 110 bridges, 89 concrete box culverts, 33 concrete pipes, 2 corrugated steel pipes, and 1 highway overpass. Fifty-three sites surveyed did not have any structures. Majority of those sites inventoried were associated with conservation lands or riparian systems. A list of field sites by conservation area would include Alapaha and Withlacoochee Rivers (8), Aucilla River (7), Camp Blanding Military Training Site and Black Creek (13), Ocala National Forest and Cross Florida Greenway (8), Osceola National Forest (11), Jennings State Forest (6), Cedar Key Scrub State Preserve (6), Etonia Creek conservation area (7), Timucuan National Preserve area (9), Suwannee River Buffers (26), Pumpkin Swamp, Waccasassa River, Devils and Gulf Hammock area (19), Lochloosa Wildlife Management Area, Newnans Lake and Paynes Prairie State Preserve (39), Santa Fe River system (15), and the Steinhatchee River system (16).

PAGE 97

81 Figure 3-7. District 2 field verification sites. Of the 1,594 road segments identified in the priority modeling process, 288 field sites were inventoried and landscape characterization performed. Recommendations for corrective measures necessary to maximize permeability for wildlife were provided on the numbered locations displayed (Appendix C).

PAGE 98

82 Characteristics negatively influencing ecological conditions at the 5 priority-one sites were minimal. Some planted-pine areas were identified at all 5 sites and off-road vehicle access at 1 site. Each of the 5 sites included records of black-bear road-kills. Hardwood, cypress swamp and pine flatwood communities dominated all of these sites. The 5 priority-two sites visited contained the following adverse characteristics: 1 site had a steep-walled canal (Cross-Florida Barge Canal), 1 site had a single residence, 3 sites had patches of planted pine, and 1 site had off-road vehicle access. Each of these sites included black-bear road-kill records. The 14 priority-three sites surveyed contained the following adverse characteristics: 6 sites had parcels of pine plantation, 2 sites had scattered rural-residential, and another had a government-facilities building and a microwave tower. Habitat was high quality and certain sites consisted of sandhill and hardwood and cypress swamp communities. Eleven sites had black-bear road-deaths and one site had a dead, striped mud turtle. For the 52 sites in category four, on-site factors contrary to conservation efforts include: 5 sites disturbed by access trails or dirt roads, railroad tracks at 3 sites, a powerline corridor at another site, scattered rural-development at 4 sites, 16 sites contained some planted pine, and minor pasture or crop lands at 4 sites. Ten black-bear road-kills occurred at 10 sites, one cottonmouth Agkistrodon piscivorous was sighted at another site, and multiple species of herpetofauna and small mammals were recorded in a research project at 2 other sites. The 147 priority-five sites included the following negative influences: 3 sites had adjacent man-made canals, 1 site had an adjacent active railroad, 2 sites had some low

PAGE 99

83 intensity pasture, 38 sites consisted partially of pine plantations, 5 sites have significant development adjacent or nearby (e.g., 1 housing development, 1 mobile home park, 2 marinas / fish camps, and 1 quarry), 15 sites have minor development (e.g., sparse rural-residential, 3 local parks, and a motel), and 5 sites have off-road vehicle access. One site had a black-bear death, an alligator was sighted at another site, and hog and deer bones were found at 2 other sites. Lastly, multiple species of herpetofauna and small mammals were sighted at 2 other sites. District three An inventory of 304 sites on 24 different highways was conducted in District 3 (Figure 3-8). Of the locations surveyed, 15 were priority-one sites, 13 were priority-two sites, 29 were priority-three sites, 64 were priority-four sites, and 107 were priority-five sites. Seventy-six additional sites surveyed were not ranked. Existing structures found among these sites included: 70 bridges, 49 concrete box culverts, 90 concrete pipes, and 5 corrugated steel pipes. Ninety sites did not have any drainage-control structures. Of all sites checked, 52 were associated with the Apalachicola River system (includes the Apalachicola National Forest, Tates Hell State Forest, and Apalachicola River and tributaries), 11 were located in the Aucilla River Wildlife Management Area, 18 were in Blackwater River State Forest, 13 cross the Choctawatchee River or its tributaries, 4 were in existing or proposed conservation-lands associated with the Econfina River aquifer recharge area, 54 were located in Eglin AFB and the Yellow River area, 14 are along the Ochlockonee River corridor, 5 are in the Tall Timbers Red Hills area, 15 were associated with Pt. Washington and Topsail Hill, 15 were within the Wakulla Springs St. Marks National Wildlife Refuge area, and 5 were in Tyndall AFB.

PAGE 100

84 Figure 3-8. District 3 field verification sites. Of the 2,822 road segments identified in the priority modeling process, 304 field sites were inventoried and landscape characterization performed. Recommendations for corrective measures necessary to maximize permeability for wildlife were p rovided on the numbered locations dis p la y ed ( A pp endix C )

PAGE 101

85 The 15 priority-one sites surveyed contained the following adverse characteristics: 4 sites had off-road vehicle access, 3 sites had clearcuts on pine plantations, and another was adjacent to a ranch and a tourist curio shop. Ecological communities included longleaf pine-turkey oak, hardwood forest, cypress swamp, and bottomland hardwoods. Recorded animal activity included multiple black-bear road-kills at 4 sites; and 10 sites had single black-bear road-kills. Characteristics negatively influencing ecological conditions at the 13 priority-two sites were minimal. Some planted-pine areas were identified at 7 sites, off-road vehicle access was found at 4 sites, one site had a fire tower, and another had a single house. Each of the 13 sites included records of black-bear road-kills. Hardwood and cypress swamp, and pine flatwoods were the dominant communities. The 29 priority-three sites included the following negative influences: 1 site had an adjacent active railroad, 12 sites consisted at least partially of commercial silviculture, 3 sites had minor development (e.g., one house, a boat ramp, and a shooting range), and 8 sites had off-road vehicle access. Typical native communities included sandhill, pine flatwoods, bottomland hardwoods, and several other wetland types. Two sites had multiple black-bear deaths; 22 others were sites of single mortalities from vehicle collisions. The 64 priority-four sites visited contained the following adverse characteristics: 3 sites had railroads, 5 sites had boat ramps and/or small parks, 13 sites had parcels of planted pine, 1 site had a borrow pit, 2 sites had significant development (e.g., a shopping center and residential areas), 3 other sites had scattered rural-residential, and 5 sites had off-road vehicle access. Ecological communities include hardwood and cypress swamp,

PAGE 102

86 hardwood hammock, and longleaf pine forest. Thirteen sites had black-bear mortalities from vehicle collisions; 2 of these contain multiple deaths. Other road-kills found include a black vulture and a cottonmouth. Characteristics negatively influencing ecological conditions at the 107 priority-five sites included: an adjacent man-made canal at 1 site, active railroads at 6 sites, scattered rural-development at 10 sites, pine plantations at 8 sites, off-road vehicle access at 13 sites, and boat ramps and / or recreation areas at 7 sites. Many of these land uses occurred at the same sites. Five sites had black-bear road-kills, and one site had a river otter road-kill. District four Characteristics of 77 field sites were documented on 11 different roads in District 4 (Figure 3-9). Number of field sites evaluated by model priorities (in parentheses) included 7(1), 18(2), 26(3), 13(4), and 10(5). Three additional but unranked sites were also inventoried. Structures of various size present at these sites included: 28 bridges, 27 concrete box culverts, 9 concrete pipes, and 1 corrugated steel pipe. Thirteen sites surveyed did not have any structures. The number of field sites by conservation area are as follows: Blue Cypress conservation area (4), Cypress Creek SOR and Loxahatchee RiverCARL, Preserve, and National Wildlife Refuge (15), Everglades buffer area (3), Ft. Drum conservation area (4), Indian River County Preserve and Padgett Marsh (4), Jonathon Dickinson State Preserve (4), Pal Mar, J.W. Corbett and Pratt-Whitney conservation areas (18), water conservation areas 2a and 3a (10), St. Sebastian River State Preserve (4).

PAGE 103

87 Figure 3-9. District 4 field verification sites. Of the 4,444 road segments identified in the priority modeling process, 77 field sites were inventoried and landscape characterization performed. Recommendations for corrective measures necessary to maximize permeability for wildlife were provided on the numbered locations displayed (Appendix C).

PAGE 104

88 The 7 priority-one sites included the following negative influences: 1 site had an adjacent man-made canal, 1 site had an adjacent active railroad, and 4 sites had invasive Melaleuca patches. Ecological communities present at these sites included sand pine scrub, coastal dunes, sawgrass marsh, cypress and hardwood swamp, hardwood hammock, and pine flatwoods. No signs of animal presence were found during visits to these sites. The 18 priority-two sites visited contained the following adverse characteristics: 3 sites had adjacent man-made canals, 3 sites exhibited presence of invasive plant species (Melaleuca or Brazilian pepper), 1 site had substantial residential and improved pasture, 1 site was partially composed of citrus, and 4 sites had active railroad lines. One unidentified dead snake was found at one field site. For the 26 sites in category three, on-site factors contrary to conservation efforts included: 11 sites disturbed by canals, some residential development and a local park at 1 site, 10 sites contained citrus groves, invasive Melaleuca was evident at 1 site, and another site had cattle grazing. Primary habitat at these sites was pine flatwoods, and flatwoods prairie with isolated cypress and hardwood swamps, and wet prairie. The 13 priority-four sites visited contained the following adverse characteristics: 8 sites had canals, 1 site had an airboat marina, 3 sites had patches of invasive plants (i.e., Melaleuca, Brazilian pepper, and Australian pine), and 3 sites had cattle grazing. Ecological communities include hardwood and cypress swamp, hardwood hammock, and sawgrass marsh. Characteristics negatively influencing ecological conditions at the 10 priority-five sites included: citrus development at 3 sites; man-made canals associated with all 10

PAGE 105

89 sites; one site had a construction-debris landfill, an aggregate mine, and a turfgrass farm; invasive plants at 2 sites; and another had a boat ramp and local park. One site in Blue Cypress conservation area had multiple road-killed turtles; and another site had a road-killed bobcat. District five An inventory of 278 sites on 24 different highways was conducted in District 5 (Figure 3-10). Of the locations surveyed, 9 were priority-one sites, 28 were priority-two sites, 44 were priority-three sites, 30 were priority-four sites, and 120 were priority-five sites. Forty-seven of the sites inventoried were not ranked. Existing structures found among these sites included 66 bridges, 90 concrete box culverts, 47 concrete pipes, and 2 corrugated steel pipes. Seventy-four sites did not have any drainage-control structures. Major conservation areas associated with field sites included: Three-Lakes Wildlife Management Area (19), Lake George CARL, Haw Creeks and Deep Creek (21), Econ-St. Johns Ecosystem including Seminole Ranch and Tosohatchee State Preserve (35), Micco Scrub (8), N. Indian River CARL (9), Ocala National Forest (47), Tiger Bay and Tomoka River (16), and the Wekiva River area (16). The 9 priority-one sites surveyed contained few adverse characteristics: 3 sites had off-road vehicle access and 1 site had a local park. Ecological communities included sand pine-xeric oak scrub, hardwood forest, and cypress swamp. Recorded animal activity included multiple black-bear road-kills at 7 sites and 2 sites had single black-bear road-kills. The 28 priority-two sites surveyed contained the following adverse characteristics: 6 sites had parcels of pine plantation, 2 sites had scattered

PAGE 106

90 Figure 3-10. District 5 field verification sites. Of the 2,076 road segments identified in the priority modeling process, 278 field sites were inventoried and landscape characterization performed. Recommendations for corrective measures necessary to maximize permeability for wildlife were provided on the numbered locations displayed (Appendix C).

PAGE 107

91 rural-residential, one had some pasture, 4 sites had access roads, and 2 sites had railroad lines. Habitat was high quality and certain sites consisted of sandhill, xeric oak and sand pine scrub, and hardwood and cypress swamp communities. Twenty-seven sites had black-bear road-deaths, 19 of them had multiple deaths; and one site had a dead corn snake. The 44 priority-three sites included the following negative influences: 2 sites had adjacent active railroads, 2 sites consisted at least partially of agricultural fields, 4 sites had minor development (e.g., scattered residential, one horse farm, and one small business), and 15 sites had off-road vehicle access, mostly forest-service roads. Eighteen sites had multiple black-bear deaths; 17 others were sites of single mortalities from vehicle collisions. One site had multiple snake road-kills. For the 30 sites in category four, on-site factors contrary to conservation efforts included: 3 sites each disturbed by access roads and canals, powerlines at 1 site, 3 sites had minor development (e.g., a toll booth, scattered rural-residential, and a fish camp), 1 site each contained some planted pine and minor pasture. Multiple black-bear road-kills occurred at 1 site and single deaths occurred at 2 sites; and one unidentified turtle was found at another site. The 120 priority-five sites visited contained the following adverse characteristics: 1 site each had an adjacent man-made canal or powerline corridor, 4 sites had active railroads, 2 sites exhibited presence of Brazilian pepper, 2 sites had substantial residential and/or commercial development, 3 sites had rural residential, one site had a single house, 2 sites had marinas, one site each had a rail-trail and a local park, and 7 sites had off-road vehicle access. Agriculture present at field sites included 7 with pasture, 4 agricultural

PAGE 108

92 fields, and 1 pine plantation. Three sites had single black-bear road-kills, 1 site had a road-killed bobcat, and 3 sites had multiple species road mortalities that included meso-mammals, snakes, a Black Vulture and a Barred Owl. At one site a river otter crossed the road; and 2 other sites had alligators in adjacent ditches. Swallowtail Kites, Wood Storks, and Caracara were also sighted at three sites. District six Eighteen sites were surveyed on U.S. Highway 1 (Figure 3-11). Number of field sites visited, by model priorities of one to five, was 1, 5, 5, 3, and 2 respectively. Two other significant but unranked sites were also inventoried. Structures present at these sites included 2 drawbridges and 2 corrugated steel pipes. Fourteen sites surveyed did not have any structures. Conservation areas where all field sites were located include Everglades National Park, Crocodile Lake National Wildlife Refuge, Key Deer National Wildlife Refuge Cupon Bight CARL, and the Southern Glades SOR. The 18 prioritized sites included the following negative influences: 1 site had an adjacent man-made canal, 1 site had an access road, 2 sites had adjacent intensive development (e.g., a hotel and marina, and other strip commercial businesses), and 5 sites had invasive Brazilian pepper and/or Australian pine. Ecological communities present at these sites included mangrove, salt marsh and tropical hardwood hammock. Signs of wildlife found included road-kills at separate sites: 3 juvenile alligators, 1 juvenile crocodile, and recorded occurrence of multiple key-deer deaths from vehicle collisions. District seven Characteristics of 46 field sites were documented on 12 different roads in District 7 (Figure 3-12). Number of field sites evaluated by model priorities (in parentheses)

PAGE 109

93 Figure 3-11. District 6 field verification sites. Of the 143 road segments identified in the priority modeling process, 18 field sites were inventoried and landscape characterization performed. Recommendations for corrective measures necessary to maximize permeability for wildlife were provided on the numbered locations displayed Appendix C).

PAGE 110

94 included 6(1), 1(2), 12(3), 17(4), and 9(5). One additional but unranked site was also inventoried. Structures of various size present at these sites included 21 bridges, 4 concrete box culverts, 9 concrete pipes, and 1 corrugated steel pipe. Eleven sites surveyed did not have any structures. A list of field sites by conservation area would include Annutteliga Hammock CARL and Chassahowitzka Wildlife Management Area (9), Withlacoochee River and tributaries and Withlacoochee State Forest (10), Hillsborough River State Park and Green Swamp SOR (7), and Little Manatee River SOR (5). The 7 priority-one and priority-two sites surveyed contained the following adverse characteristics: 1 site had a junkyard, 1 site had an abandoned house, and 5 sites had access roads. Habitat was high quality and most sites consisted of longleaf pine sandhill or hardwood hammock and swamp. No signs of wildlife activity were found. For the 12 sites in category three, on-site factors contrary to conservation efforts included: 2 sites each disturbed by access roads and railroads, 2 sites had scattered rural-residential, 2 sites had substantial development (e.g., a racetrack and a golf club), and 3 sites contained some planted pine. Typical native-plant communities included wet and dry prairie, and hardwood hammock and swamp. Multiple hog road-kills occurred at 1 field site. The 17 priority-four sites included the following negative influences: 1 site consisted partially of citrus groves, 2 sites had significant urban development, 2 sites had minor development (e.g., a park and canoe launch, and one house and a small business), and 2 sites had recreational trail access. Dominant habitat types were mixed hardwood and cypress swamp. A road-killed turkey vulture was found at one site.

PAGE 111

95 Figure 3-12. District 7 field verification sites. Of the 2,504 road segments identified in the priority modeling process, 46 field sites were inventoried and landscape characterization performed. Recommendations for corrective measures necessary to maximize permeability for wildlife were provided on the numbered locations displayed (Appendix C).

PAGE 112

96 Characteristics negatively influencing ecological conditions at the 9 priority-five sites included: urban development at 1 site, access roads associated with 4 sites, one site had a man-made canal, and another site had scattered residential. Dominant habitat types were pine flatwoods, hardwood hammock, and hardwood and cypress swamp. Wood storks were seen at one site. Recommended Actions to Improve Functional Habitat Connectivity: Retrofitting Highway Hotspots The data presented below includes those road segments from each FDOT district with the greatest need for on-site mitigation. The locations discussed are not in any rank order of importance. Further ranking of these sites is recommended through discussion with various state agency officials, researchers, and other parties with expertise on highway-wildlife issues. Such discussion must address factors associated with species-specific needs, large-scale conservation objectives, and economic limitations; and be able to weigh any such factors that conflict. The sites identified herein are generally associated with large-scale or regional, conservation objectives. As such, there may be localized situations beyond the scope of this research that require greater immediate attention. District one Seventy-two sites on ten different highways were evaluated in District 1 (Appendix C, Figure 3-6). Existing structures found among these sites included 21 bridges, 11 concrete box culverts, 12 concrete pipes and 2 steel corrugated pipes. Twenty-six sites did not have any drainage-control passageways.

PAGE 113

97 Of all sites evaluated, 32 were associated with the Big Cypress Area (includes the Big Cypress National Preserve, Florida Panther National Wildlife Refuge, Fakahatchee Strand State Preserve, Collier-Seminole State Park and Golden Gate Estates CARL). Appendix C lists recommendations for 11 locations on three roads: US 41, SR 29, and I-75. Suggested measures include expansion of existing bridges (Figures 3-13a and 3-13b), construction of new ecopassages, installation of barrier fencing, (Figures 3-14 and 3-15) and straightening blind curves to increase driver visibility (Figure 3-16 displays one such instance). Figure 3-13a. Bridge over stream on US 41 near Collier-Seminole State Park. This is an example of an old floodway bridge that only allows for water flow. When scheduled for replacement, retrofits should include an extension with an adjacent terrestrial connection along the stream under the bridge. Black bear vehicle collisions have been recorded at this particular site. Lengthening this brid g e could correct the conflict. Twelve sites were located along the Lake Wales Ridge and Fisheating Creek on US 27, US 98, and SR 66 (Appendix C). Critical needs for this area were associated with

PAGE 114

98 the local black-bear population, and included recommendations to secure protected habitat-corridors and construct underpasses. Figure 3-13b. New bridge on SR 72 over Horse Creek, a tributary of the Peace River. This recent bridge replacement includes a wide terrestrial connection adjacent to the stream; and allows for wildlife movement under the b rid g e that reduces o pp ortunities for vehicle collisions. Twenty-four sites were within existing or proposed conservation-lands associated with the Myakka River (Appendix C). Recommendations for improvements on I-75, US 41, and SR 72 included purchase of adjacent resource areas and expansion of existing structures. Appendix C details scheduled bridge replacements and recommends enhancements for wildlife use. Regarding the Peace and Kissimmee rivers, work programs should include longer bridge spans when replacements are scheduled; examples include the bridges on U.S. 17 over the Peace River at Zolfo Springs (Figure 3-6, Site No. 33) and SR 60 over the Kissimmee River (Figure 3-6, Site No. 196). Bridge replacements within these riparian

PAGE 115

99 corridors should include terrestrial linkages above the high water mark that accommodate movement by wildlife restricted to dry habitats. Figure 3-14. One of the wildlife underpasses constructed on I-75 in the Big Cypress National Preserve. The directional / barrier fence is 3 m high chain link with barbed wire. The native vegetation provides open character and good visibilit y for lar g er wildlife and lower p rotective cover for smaller or g anisms. Development along the Lake Hancock Saddle Creek Green Swamp greenway severely threatens remaining habitat areas. To realize the objective of the Statewide Greenways Project to provide linkages across Interstate 4, immediate action is necessary to secure protection for the corridor. FDOT should consider constructing new wildlife underpasses and incorporating wildlife considerations along creek bridges on I if this linkage is established. District two Fifty-eight sites on 18 different highways were evaluated in District 2 (Appendix C, Figure 3-7). Existing structures found at these sites included 23 bridges, 14 concrete box culverts, and 7 concrete pipes. Fourteen sites had no drainage-control passageways.

PAGE 116

100 Figure 3-15. Wildlife Exclusion Fence on SR 29 in Collier County. The 3 m high chain link barrier fencing with barbed wire was designed to prevent Florida panthers from entering the roadway. The fence runs parallel to the roadway for the entire section of road between two wildlife underpasses. Figure 3-16. Blind curve on State Road 29 in Collier County. These elements in road corridor design present special risks to wildlife attempting to cross the road. The inability of motorists to see what is around the curve on high-speed rural roads lead to man y vehicle wildlife collisions.

PAGE 117

101 The Ocala National Forest Cross-Florida Greenway Etonia Creek CARL region contained 9 sites on 3 roads (SR 19, SR 20, and SR 100) evaluated for retrofits (Appendix C). Suggested measures include lengthening existing bridges, replacing small pipes with larger box culverts (Figures 3-17a, 3-17b and 3-17c), constructing new ecopassages, permeating bridge approaches (berms) with additional culverts (Figure 3-18), and installing barrier fencing. Figure 3-17a. Depression site on SR 100 in the Etonia Creek CARL area in Putnam County. This site has approximately 2.5 m clearance between the road surface and the floor of the existing pipe culvert. Opportunity exists to retrofit this site with a larger structure to accommodate safe travel for larger wildlife such as b lack bear and deer. Appendix C provides information on a priority-one site on SR 206 in the Fish Swamp WMA and Twelve Mile Swamp CARL area. A small culvert at the site provides insufficient permeability for a high incidence wetland corridor (Figure 3-19a and 3-19b). Road-kills included black bear and multiple herpetofauna. Recommended corrective measures include a large animal ecopassage and exclusionary barrier system.

PAGE 118

102 Figure 3-17b. Side view of existing culvert. Dimensions of this concrete pipe are 0.6 m wide x 21.6 m long. Situated at the bottom of the depression shown in Figure 3-17a, erosion has partially blocked the structure; and combined with the small size leaves little opportunity for movement by wildlife. Figure 3-17c. Box culvert at ephemeral creek site on SR 100 in the Etonia Creek CARL area in Putnam County. This site has similar topographic relief to that shown in Figure 3-17a; only in this case a three cell concrete box culvert (each cell is 2.1 m wide x 1.8 m tall x 21.6 m long) was installed. Most times this structure is at least partially dry and allows for movement by larger wildlife species.

PAGE 119

103 Figure 3-18. Bridge approach on State Road 100 at Palmetto Branch in Putnam County. When construction of longer bridge spans is not possible, approaches such as the one shown here, consisting of fill, should be perforated with smaller culverts to increase landscape connectivity and provide greater opportunity for safe movement by wildlife. Four sites on US 90 and I-10 were located in the St. Marys River corridor Cary and Jennings State forests area (Appendix C). Two bridges scheduled for replacement need to accommodate passage by black bears, and barrier fencing needs to be erected at two other sites. Camp Blanding Military Training Site is the location of 7 sites on SR 21 and SR 16 (Appendix C). Addition of two wildlife passages for black bears on SR 16 west of Black Creek and wildlife fencing; sites on SR 21 need retrofits with larger ecopassages. Appendix C contains recommendations for 6 sites on SR 20, SR 26, and US 301, located in the Lochloosa Forest conservation easement Newnans Lake CARL Cross Florida Greenway area. Suggested improvements include replacing the Lochloosa and Orange Creek bridges with longer and higher spans (similar to that constructed on SR 46 at

PAGE 120

104 Wekiva River). Other measures needed include elimination of a blind curve and retrofitting small box culverts with larger ecopassages. Figure 3-19a. State Road 206 crossing through the Fish Swamp CARL area. This highway greenway interface was the highest ranked site in St. Johns County, it forms the linkage between Twelve-mile Swamp to the north, St. Johns River to the west, and Pellicer Creek and the Flagler County Greenway to the south. It is a known movement corridor for black bears as documented b y collisions with vehicles. On SR 24 in Cedar Key Scrub State Preserve, a 6.5 km road segment requires investigation on road mortality and barrier effects to scrub fauna and elimination of off-road vehicle disturbance (Appendix C, Figure 3-20). Six sites were located on US 98 and US 27 in the area from the Aucilla River WMA to the Econfina River corridor (Appendix C). Corrective measures recommended include retrofitting ineffective existing structures with more permeable designs and adding wildlife fencing. The Pinhook Swamp Osceola National Forest Suwannee River area contains 11 sites on three roads, US 441, SR 2, and I-10 (Appendix C). Suggestions to increase permeability at these sites include bridge replacements with designed ecopassages and installation of new underpasses and

PAGE 121

105 fencing to prevent black-bear road-kills. Appendix C contains recommendations for scheduled bridge replacements on SR 51 and US 301. Figure 3-19b. Concrete Box Culvert Present at Unknown Creek at the Fish Swamp Site. This small aquatic cross drain is obviously inadequate to accommodate large species such as black bear and white-tailed deer. Other road projects that require investigation for wildlife retrofits include: 1) Interstate 95 (add lanes) over the Nassau River corridor (Figure 3-7, Site Nos. 388 and 389), Twelve Mile Swamp and Pellicer Creek (Figure 3-7, Site No. 202), 2) U.S. 17, St. Mary's River bridge replacement, 3) ALT U.S. 27 (add lanes) crossing through the Devils Hammock Waccasassa Flats Greenway, and 4) the Florida Turnpike extension adjacent to Goethe State Forest. District three In District 3, 73 sites on 12 different highways were evaluated (Appendix C, Figure 3-8). Existing structures found among these sites included 6 bridges, 9 concrete box culverts, and 25 concrete pipes. Drainage-control structures were not present at 33 sites.

PAGE 122

106 Figure 3-20. State Road 24 crossing through the Cedar Key Scrub State Preserve. Damage to adjacent habitat occurs from all terrain vehicles that access the preserve area from the highway. Fourteen sites were located in the Aucilla Wildlife and Water Management Area on US 27 and US 98 (Appendix C). Corrective measures recommended include replacing existing substandard structures with underpasses (accompanied by high-level fencing for black bears). The Wakulla Springs State Park St. Marks NWR area has 7 sites on two roads, SR 267 and US 98 (Appendix C). Suggested improvements include purchase of adjacent CARL lands (to establish a protected corridor) and installation of underpasses. Appendix C contains recommendations for improvements to 23 sites on US 98, SR 71, and SR 65 in the Apalachicola National Forest Tates Hell SF / CARL Apalachicola River WMD project areas. Three separate measures are suggested for these sites: realign roads to reduce severity of curves, close certain access roads, and install underpass/high-level fence systems. Tyndall AFB has one site on US 98 where multiple

PAGE 123

107 black-bear deaths have occurred (Appendix C). Considerations include side road removal and improvements to increase driver visibility (Figure 3-21). Figure 3-21. U.S. Highway 98 crossing through Tyndall Air Force Base in Bay County. Multiple black bear road-kills have occurred at this site. Although exact circumstances of the collisions are unknown, probable causes are high speed, reduced driver visibility (due to proximity of dense vegetation in the verge and the curve in the road ), and increased traffic from the intersection. Eglin Air Force Base contains 19 sites on SR 85, SR 87, SR 123, and I-10 (Appendix C). Primary measures needed to improve permeability at these sites include reducing curve severity, perforating high roadbeds with box culverts, and constructing wildlife underpasses on I-10 along the Yellow River Ravines CARL corridor. Appendix C contains recommendations for improvements at other greenway linkages and scheduled road projects on SR 20, SR 22, US 98 and US 331.

PAGE 124

108 District four Eleven sites on 2 different highways were reviewed in District 4 (Appendix C, Figure 3-9). Existing structures found among these sites included 6 bridges and 2concrete box culverts. Three sites did not have any drainage-control passageways. Seven sites, located on SR 60, are associated with the Blue Cypress SOR, Kissimmee River SOR, and Ft. Drum Marsh WMD lands (Appendix C). Corrective measures needed at these sites include perforating high roadbed areas with box culverts, lengthening existing bridges scheduled for replacement, and construction of new underpasses. Appendix C contains data on 4 sites within the Cypress Creek CARL area. Recommendations include scheduled replacement of existing low-level bridges and box culverts with underpasses and exclusionary fencing for wildlife. District five Evaluations were performed on 104 sites on 11 different highways in District 5 (Appendix C, Figure 3-10). Existing structures included 12 bridges, 31 concrete box culverts, 17 concrete pipes, and 1 corrugated steel pipe. Forty-four sites did not have any drainage-control passageways. Thirty-nine sites are in the Ocala National Forest on SR 19, SR 40, and CR 42 (Appendix C). Primary impacts to address are chronic road-kill sites of black bears due to blind spots for drivers created by hills and curves (Figure 3-22). Corrections include straightening and leveling the road alignment and installing wildlife underpass/fence systems. Appendix C contains information on 5 sites on SR 44 within the Lake Woodruff NWR, Hontoon Island State Park, and Wekiva River Basin Ocala greenway SOR lands.

PAGE 125

109 Figure 3-22. State Road 40 crossing through Ocala National Forest. Multiple black bear road-kills have occurred at this site. Although precise circumstances of the collisions are unknown, probable causes are high speed, reduced driver visibility (due to hills and curves in the road), and increased traffic levels near the entrance to Juni p er S p rin g s Recreation Area. Recommendations include securing conservation easements and adding underpasses in strategic locations to maintain the connection between Ocala NF and the Wekiva River Basin. State Road 46 that bisects the Wekiva River State Park and Seminole State Forest has 10 sites where retrofits are suggested (Appendix C). Purchasing lands, or securing conservation easements south of SR 46 on the western side of the Wekiva River is essential to preserve linkages along the river (Figure 3-23). Other measures include eliminating S curves, constructing additional underpasses, and extending high-level fences. Appendix C provides data for 5 sites on US 17 and SR 40 in the Lake George State Forest and Haw Creek WMD project area. Corrective measures, such as purchasing

PAGE 126

110 development rights and replacing small culverts with large mammal underpasses, are needed to maintain connection to the Ocala National Forest and St. Johns River. Figure 3-23. Wekiva River conservation lands for sale. Probably more critical than retrofitting roads is the need to secure valuable conservation linkages before they are developed. This river frontage is one corner of the recently constructed black bear passage on SR 46. If developed it could render the road mitigation meaningless and a waste of approximately $1 million in taxpa y er dollars. Appendix C includes 18 sites on I-4, US 92, SR 44, and SR 415 in Tiger Bay State Forest, Deep Creek Swamp, and other proposed greenway lands north of the St. John's River. Existing small culverts (at these sites) need to be replaced with large underpasses to increase functional connectivity for wildlife, especially regarding Interstate 4. Establishing conservation corridors is critical in the rapidly developing SR 415 area. Several programmed road projects in District 5 should include retrofits for wildlife along important riparian corridors.

PAGE 127

111 District six Seventeen sites on U.S. Highway 1 were evaluated in District 6 (Appendix C, Figure 3-11). Existing structures found among these sites included 2 draw bridges and 2 corrugated steel pipes. Fourteen sites did not have any drainage-control structures. Sixteen of these sites are within Everglades National Park, Model Lands Basin and Southern Glades SOR, or the Crocodile Lake NWR (Appendix C). Recommendations include perforating the roadbed and installing several aquatic culverts to accommodate movement by the endangered American crocodile (Figure 3-24a and 3-24b). Figure 3-24a. U.S. Highway 1 crossing through Crocodile Lake National Wildlife Refuge. The highway bisects critical mangrove and tidal salt flats and marshes used by the endangered American crocodile. Only two 1.5 m wide corrugated steel pipe culverts exist along the 10 km of highway between the mainland and the Jewfish Creek Brid g e on Ke y Lar g o. Appendix C contains information on the other sites on US 1, which are associated with one of the following: Big Pine Key, Key Deer National Wildlife Refuge, Cupon

PAGE 128

112 Bight CARL project, Bahia Honda-, Long Key-, and John PennekampState Parks. Top priority is acquisition of the Cupon Bight habitat area (for the endangered Florida Key deer) and retrofits to US 1 (to eliminate the road mortality problem) (Figure 3-25). Figure 3-24b. A juvenile American crocodile encountered during field surveys. A consequence of the road barrier is collisions with vehicles when the crocodilians attem p t to cross from one side to the other. District seven In District 7, 29 sites on 11 different highways were assessed (Appendix C, Figure 3-12). Existing structures at these sites included 9 bridges, 3 concrete box culverts, 6 concrete pipes, and 1 corrugated steel pipe. Ten sites did not have any drainage-control passageways. Appendix C includes field sites on three roads (US 19, US 98, and the Suncoast Parkway) that cross through the Chassahowitzka NWR / WMA and Annuteliga Hammock CARL project. Suggested improvements include increasing driver visibility (either through vegetation management near intersections, or removal of access roads and curves in highways) and installation of wildlife passages designed for black bears.

PAGE 129

113 Figure 3-25. U.S. 1 passing between Cupon Bight and Big Pine Key. This is a critical section of road that the Florida key deer use to move between the two remaining primary habitat areas. Over 50 individuals die annually on this 3 km section of road from collisions with vehicles. The Withlacoochee State Forest provides the setting for 6 sites on 3 roads: US 41, SR 50, and US 301 (Appendix C). Two issues to be addressed include securing habitat connections and fragments that remain unprotected (between existing forest lands and the Chassahowitzka NWR), and replacement of existing culverts with larger suitable wildlife underpasses. Appendix C lists 6 sites on SR 52, SR 54, and I-75 within the Cypress

PAGE 130

114 Creek corridor, Pasco SOR lands, and lower Hillsborough River WMD lands. Suggested improvements include installation of larger box culverts (to increase permeability at wetland sites), and coordinating wildlife considerations with scheduled bridge replacements. Five sites of concern in the Hillsborough State Park, Hillsborough River corridor (SOR), and Blackwater Creek area are located on US 301 and SR 39 (Appendix C). Corrective measures needed include a new wildlife underpass and a longer span installed at one scheduled bridge replacement location. Other suggested highway improvements for District 7 sites on US 19, SR 200, SR 44, SR 52, US 41, and the Suncoast Parkway are summarized in Appendix C. These sites are associated either with the Withlacoochee River or the Starkey Wilderness Area. Value of Existing Infrastructure Certain existing highway structures can serve as ecopassages and provide significant benefits over new structures including: Minimum modifications or construction necessary New construction cost savings Many existing structures are effective for facilitating wildlife movements/dispersal (Chapter 4) Value of existing bridges that function as wildlife ecopassages and costs to retrofit those structures that require expansion (to provide the same ecological benefits) can be quantified. Recent bridge replacement costs were compiled to establish baseline values for construction materials. The average rates were then used to estimate costs to implement wildlife retrofits. Determining Bridge Replacement Costs An analysis was performed for 15 bridges recently constructed or bidded out. Structure variation for calculations included 2and 4lane highways and singleand

PAGE 131

115 twinspan construction. Bridge replacements generally include six project components: 1) temporary traffic control measures during construction, 2) bridge approach construction, 3) old structure removal, 4) earthwork and site preparation, 5) new bridge construction, and 6) permanent traffic control and safety measures. Costs for traffic control measures can vary widely, depending upon the traffic volume and adjacent land use. Bridge approach costs also have high variability, depending on specific site needs associated with topography and adjacent land use. Old structure removal is considered a fixed cost in project estimates (i.e., demolition), and varies depending on the volume of material to be removed. For the other two components, earthwork and the new structure, standardized values could be determined. The objective was to provide approximate costs of the structure sections that enable existing bridges to serve as wildlife ecopassages; therefore many of the project components with high variability (previously discussed) were omitted. Costs for removal of fill and the new structure addition were separated to generate a standard cost/m 2 (to determine an estimate for bridges inventoried during field surveys; as part of the statewide ecological hotspots on highways analysis). Table 3-1 provides information on costs generated from new bridge construction and earthwork for 15 previous projects. Calculations reveal an average of $7.54/m 3 for fill removal/addition on 15 typical bridge replacement projects. Construction costs for the bridge structures were split according to highand lowlevel structure types. Examples of high-level bridges encountered for this research included railroad bridges and navigable river bridges. Low-level bridges are most typical for aquatic features where vertical clearance is primarily designed to accommodate flood stages. Costs differ

PAGE 132

116 because of different structural configuration. Substructure support is much greater for high-level bridges thus costs are generally higher. FDOT officials have estimated average construction costs as follows: low-level bridge spanning river floodplain: $484/m 2 and high-level railroad bridge: $700/m 2 (Keith Campbell-FDOT, Personal Communication 2000). Since no high-level structure replacements were identified, they were not included in calculations shown in Table 3-1. An average cost of $702.19/m 2 was calculated for low-level bridges. This calculation is significantly higher than the verbal estimate provided by FDOT for low-level bridges. Table 3-2 provides information on estimated costs to add extensions (to provide terrestrial linkages for wildlife movement) to the existing bridges from Table 3-1. Calculations made in these tables are explained with the following examples: US 90 crossing the Little St. Mary's River (Figure 3-26), and SR 46 crossing the Wekiva River (Figure 3-27). The dimensions of the new structure over the L. St. Marys River on US 90 were 120 m x 27 m with a clearance of approximately 3 m. Construction bids (n = 7 bidders) for this structure averaged $545/m 2 a total cost of approximately $1.77 million (FDOT 1997). The original bridge replaced had dimensions of 102 m x 9 m. Thus length and width each increased by 18 m; width was increased to accommodate two additional lanes of traffic. Given the topographic configuration of the site, the fill removed for the new bridge would have been transported in from off-site to build the original elevated approaches. Unit costs multiplied by the area of the bridge extension segment (162 m 2 ) results in added costs of approximately $113,755.

PAGE 133

Table 3-1. Typical bridge replacement construction costs Road (configuration) Feature Year Bids (#)/ Actual Fill Amount (m 3 ) Total Fill Cost Fill Cost/m 3 Length Width Height Area (m 2 ) Structure Cost Structure Cost/m 2 Structure Cost by Height (m-m 2 ) US 1 (4 lane, 2 spans) Tomoka River 1999 a 9,850 89,151 9.05 107 30 5 3,210 1,757,192 547.41 109.48 US 1 (4 lane, 1 span) Turkey Creek 1997 a 14,400 132,700 9.22 67 37 7 2,479 1,562,624 630.34 90.05 US 1 (4 lane, 2 spans) Goat Creek 2001 b (9) 5,779 76,172 13.18 42 25 6 1,034 1,084,068 1,048.26 174.71 US 90 (4 lane, 1 span) Little St. Mary's Creek 1998 b (7) 276,327 625,868 2.26 120 27 3 3,240 1,766,449 545.20 181.73 SR 20 (4 lane, 2 spans) Lochloosa Creek 2002 a 557,983 2,260,779 4.05 73 26 2.8 1,898 819,866 431.96 154.27 SR 20 (2 lane) Rice Creek 2000 a 52,138 552,766 10.60 152 13 2.3 1,976 999,219 505.68 219.86 SR 46 (2 lane) Wekiva River 1997 a 15,194 70,453 4.64 171 14 3 2,445 919,904 376.19 125.40 SR 40 (2 lane) Tomoka River 1998 a 109,430 576,507 5.27 110 13 5 1,430 1,098,700 768.32 153.66 US 192 (2 lane) Reedy Creek (3) 1999 a 82,722 597,836 7.23 680 14 7 9,520 4,940,323 518.94 74.13 US 301 (2 lane) Shady Brook 1999 a 783 8,294 10.59 36 14 7 504 368,700 731.55 104.51 US 301 (2 lane) Jumper Creek 1999 a 623 4,933 7.92 23 14 3 322 260,574 809.24 323.69 SR 19 (2 lane) Palatlakaha Cr. 2000 a 10,696 95,068 8.89 44 13 3 592 421,082 710.95 284.38 SR 11 (2 lane) Haw Creeks (3) 2000 b (7) 23,442 225,502 9.62 172 14 4 2,477 1,850,583 747.17 186.79 SR 60 (2 lane) Kissimmee R. 2001 b (5) 259,797 1,146,036 4.41 182 16 6 2,912 2,776,482 953.46 158.91 SR 520 (2 lane) St. Johns River 1999 a 133,390 821,989 6.16 108 14 4 1,512 1,826,660 1,208.11 302.03 Average Costfill work $7.54 Average Coststructure $702.19 $176.24 117

PAGE 134

118 Figure 3-26. Construction of replacement bridge. New bridge built over the Little St. Marys River, on US Highway 90 near MacClenny in 1999. Figure 3-27. Wildlife bridge extension on SR 46 over the Wekiva River in central Florida. Structure design includes a 74 m terrestrial linkage along the river to facilitate safe movement for the Florida black bear. (T. Gilbert, Florida Fish and Wildlife Conservation Commission 1999, PersonalCommunication. )

PAGE 135

119 Table 3-2. Old bridge dimensions and costs for extension Road (configuration) Feature Length Width Area (m 2 ) Added Length Cost/m 2 Cost for Added Length US 1 (4 lane, 2 spans) Tomoka River 91.5 9.6 878 16 547.41 81,455 US 1 (4 lane, 2 spans) Turkey Creek 47 19.4 912 20 630.34 244,574 US 1 (4 lane, 2 spans) Goat Creek 23.8 19.1 455 18 1048.26 358,389 US 90 (2lane) L. St. Mary's Cr. 102 9 918 18 545.20 88,322 SR 20 (2 lane) Lochloosa Cr. 38 14 532 35 431.96 211,660 SR 20 (2 lane) Rice Creek 32 8 256 120 505.68 485,453 SR 46 (2 lane) Wekiva River 78 9 702 93 376.19 314,873 SR 40 (2 lane) Tomoka River n/a n/a n/a n/a 768.32 n/a US 192 (2 lane) Reedy Creek (3) 114 12 1,368 566 518.94 3,524,651 US 301 (2 lane) Shady Brook 31 7.2 223 5 731.55 26,336 US 301 (2 lane) Jumper Creek 23 7.2 166 0 809.24 SR 19 (2 lane) Palatlakaha Cr. 30 9 270 14 710.95 90,860 SR 11 (2 lane) Haw Creeks (3) 138 7.7 1,063 34 747.17 195,608 SR 60 (2 lane) Kissimmee R. 182 10 1,820 0 953.46 SR 520 (2 lane) St. Johns River 85 9 765 23 1208.11 250,078 Average Coststructure $702.19 Another cost comparison was performed on the low-level bridge replacement project on SR 46 over the Wekiva River in central Florida. This bridge was extended to incorporate a terrestrial wildlife underpass adjacent to the watercourse, primarily for use by black bears. The dimensions of the new structure are 171 m x 14.3 m with a clearance height of approximately 3 m. Bridge construction costs totaled $919,904 or $376/m 2 (FDOT 2000). Portions of the original bridge approaches had to be removed to allow clearance for the wildlife underpass. Expenditures for earth removal (quantity of fill = 15,194 m 3 ) associated with these approaches totaled $70,453 or $4.6/m 3 (FDOT 2000). The original bridge that this structure replaced measured 78 m x 9 m. Given the topographic configuration of the site, the fill removed for the new bridge would have been transported in from off-site to build the original elevated approaches. Therefore, assuming similar costs for the earthwork, the total costs to build the original bridge (based on the known cost for the new bridge, $376/m 2 ) would be approximately

PAGE 136

120 $264,000. Cost differential between constructing the shorter floodway bridge at this site and the extended bridge would be around $315,000. If this type of investment is being considered for each underpass, the location and modification of existing structures (at identified priority road segments) could result in substantial savings to the State for retrofits on roadways (to restore landscape connectivity between adjacent conservation lands). Infrastructure Investments for Wildlife Value of existing bridges (that function as wildlife ecopassages) was calculated from the field inventory conducted as part of the site verification and recommendation phase of the statewide prioritization process of highway greenway interfaces, a.k.a. ecological hotspots on highways. Many structures that require retrofits (to provide terrestrial linkages under the structure for wildlife movement) were also identified in the inventory. As part of this analysis, bridge extensions are proposed and approximate construction costs estimated. Table 3-3 summarizes calculations of fixed capital for the bridges listed in the field inventory. For each bridge to provide suitable permeability for terrestrial wildlife, differing corrective measures are needed (dependent on context and structural dimensions). The five categories of retrofits include: 1) bridges over canals within sawgrass marsh; 2) bridges that are suitable in their existing condition, provided that a percentage of the rip-rap is replaced by other scour preventing measures conducive to wildlife movement; 3) bridges that need extensions to provide a terrestrial linkage; 4) bridges with low clearance that require replacement by higher structures to accommodate larger wildlife; and 5) bridges that are suitable in their existing condition. Parameters

PAGE 137

121 include estimated value of existing structures and added costs for proposed retrofits, including cost for fill and excavation work and added structural work. Table 3-3. Summary of existing bridges evaluated for corrective measures Structure-Retrofit Type No. Existing Structure Surface Area (m 2 ) Estimated Existing Value Fill Amount (m 3 ) Fill Cost Added Surface Area (m 2 ) Structure Cost Total Added Cost canals in marsh 12 13,335 $ 9,363,431 $ $ $ length extension 102 117,190 $ 82,289,991 435,975 $3,287,251 44,889 $31,520,598 $34,807,850 rubble removal/replacement 5 4,633 $ 3,253,442 $ $ $ replacement/height addition 52 24,037 $ 16,878,730 27,175 $ 204,897 55,320 $38,844,885 $51,346,820 suitable in existing condition 201 573,641 $401,434,758 $ $ $ Total 372 732,836 $513,220,353 463,150 $3,492,149 100,209 $70,365,484 $86,154,669 Values for Calculations: Existing Structure Cost $702.19/m 2 Fill Cost $7.54/m 3 Increased Length (25m) and Height (minimum 3m), Added Structure Height Cost $176.24/m-m 2 Note: One suspension bridge was included in the suitable in existing condition category, although surface area was included, value information was not available. Capital investment in existing structures currently suitable as wildlife ecopassages (201) was estimated at $401.4 million. This estimate includes the value for the surface area of the bridge only. Calculations were performed for 102 bridges that require extensions to provide terrestrial passage. Estimated existing value of these structures was $82.3 million. Approximate cost to construct 25 m extensions for the 102 bridges was $34.8 million. The cost to extend each bridge was determined by estimating: 1) the cubic meters of approach fill to remove, and 2) the square meters of surface structure to add for 25 m of additional length. These amounts were multiplied by the estimated cost for structure work of $702.19/m 2 and fill work of $7.54/m 3 Additional expenses would be expected for traffic control and approach reconstruction. There were 52 bridges valued at $16.9 million identified that require increased height to provide for passage by larger wildlife. Costs for this work include replacement value and estimated cost for structure and fill work to increase the length by 25 m and the clearance height to 3 m. Cost for each square meter in surface area per added meter in

PAGE 138

122 height ($176.24/m-m 2 ) was the averaged value from the known bridge costs shown in Table 3-1. The sum of the replacement value, cost for fill work, and construction cost for increased length and height give the approximate cost to increase each bridge by 25 m in length and to 3 m in clearance height. Total cost to retrofit the 52 bridges is estimated at $51.3 million. Again, additional expenses would be expected for traffic control, old structure removal, and approach reconstruction. Twelve bridges were constructed over canals located in sawgrass marsh. These structures were entirely surrounded by aquatic habitat. Calculating modification cost for providing terrestrial linkages was therefore unnecessary. Approximate investment value of these bridges was $9.4 million (structure and fill work only). Five bridges were of sufficient length and height to act as ecopassages; however, substrate under these structures consisted of coarse rip-rap (heavy rock and rubble) used to prevent scouring around pilings. Rip-rap renders the area impassable for many species, especially ungulates and other large wildlife. Existing capital investment (structure and fill work only) in these five bridges was estimated at $3.3 million. The process necessary to improve the functionality of these bridges as wildlife underpasses involves removal of the riprap and replacement with other material (e.g., concrete or tiles). Cost to perform this work was unavailable, but considered here to be minor compared to replacement and retrofit costs previously discussed. In summary, approximate construction costs to retrofit the 154 bridges identified in the field inventory would be $86 million. Additional expenses associated with these projects would include traffic control measures and approach reconstruction. Fencing

PAGE 139

123 and landscape work is needed at all terrestrial sites identified (360) to improve their function as wildlife underpasses. A subset of the bridges evaluated for corrective measures above are those bridges located at priority sites included in the prior section, Recommended Actions to Improve Functional Habitat Connectivity: Retrofitting Highway Hotspots. Table 3-4 contains a summary of the 79 bridges from priority locations evaluated in the inventory. Of the 79 bridges, 24 existing structures (with an estimated value of $29 million) were considered functional in their existing condition as ecopassages. Seventeen bridge replacements were identified to increase clearance for larger wildlife present at each location. Structure replacement was estimated at $13.8 million, excluding traffic control measures and approach reconstruction. Bridge extensions were recommended for 34 of these bridges. Existing value of these structures was estimated at $23.1 million and added cost to extend the length of the bridges 25 m was estimated at $9.4 million. Two bridges require removal and replacement of coarse rubble with more permeable anti-scouring materials. Finally, two other bridges were identified in sawgrass marsh context. Table 3-4. Summary of existing bridges on priority list evaluated for corrective measures Structure-Retrofit Type No. Existing Structure Surface Area (m 2 ) Estimated Existing Value Fill Amount (m 3 ) Fill Cost Added Surface Area (m 2 ) Structure Cost Total Added Cost canals in marsh 2 1,084 $ 760,911 $ $ $ rubble removal/replacement 2 1,171 $ 821,972 $ $ $ length extension 34 32,222 $ 23,096,032 115,396 $ 870,088 12,146 $ 8,529,013 $ 9,399,101 replacement/height addition 17 6,568 $ 4,611,652 7,724 $ 58,242 16,705 $11,730,064 $13,754,971 existing condition suitable 24 41,247 $ 28,963,475 $ $ $ Total 79 82,292 $ 58,254,042 123,121 $ 928,330 28,851 $20,259,077 $23,154,072 Values for Calculations: Existing Structure Cost $702.19/m 2 Fill Cost $7.54/m 3 Increased Length (25m) and Height (minimum 3m), Added Structure Height Cost $176.24/m-m 2 Note: One suspension bridge was included in the suitable in existing condition category, although surface area was included, value information was not available.

PAGE 140

124 Of the 372 bridges evaluated in field surveys, 21% (79) were located at high priority locations. Capital investment for provision of functional wildlife underpasses at these sites is approximately $58.3 million; an additional $23.2 million is needed to retrofit those structures below functional standards. Existing bridges represent only 22% (79 of 364) of those sites identified for corrective measures. Although this assessment only considered bridges, similar cost analyses need to be performed for smaller structures such as pipe and box culverts; and other measures needed to prevent wildlife vehicle conflicts and increase highway permeability to wildlife. Statewide Bridge Expansion Program Over the last few years, the FDOT has been conducting an unofficial program to restore connectivity of riparian systems in conjunction with scheduled bridge replacements and upgrades. Through consultation with the Florida Fish and Wildlife Conservation Commission, existing bridges scheduled for replacement are examined to assess their present impact on floodplain and riparian habitat connectivity (Terry Gilbert-FFWCC, Personal Communication). Replacement structures within valuable riparian habitat areas are then designed to include terrestrial or floodplain linkages under the structure, not just provide for the floodway of the watercourse. This is a valuable approach that can help restore habitat connectivity to wildlife, such as the river otter and Florida black bear. Roads have been documented as an impediment to movements of river otters, usually resulting in highway mortality (Kruuk and Conroy 1991, D. Smith unpublished data). The first reported construction of a bridge in Florida, which included provisions for terrestrial wildlife movement along a stream corridor, was the replacement and expansion of the SR 46 bridge over the Wekiva River

PAGE 141

125 in 1997 (T. Gilbert-FFWCC, Personal Communication). The original bridge (consisting of a narrow passage for the floodway of the river) was replaced and expanded to a 171 m span that includes a 74 m terrestrial corridor, to accommodate movement by the Florida black bear (Figure 3-27). One hundred and three bridge construction projects were identified in existing and proposed conservation areas (Table 3-5). Many of these sites were ranked in the prioritization process and several were not. These projects were expected to cost approximately $414 million and include new bridges, replacements, and widenings (FDOT 1999). A key to mitigation of impacts of highways and automobile traffic on wildlife populations and ecologically sensitive areas is the programming of road projects and identification of existing structures. It is important to identify ecological hotspots on highways that correspond with planned road projects. Through these opportunities, construction of wildlife underpasses or other mitigation measures can be programmed into the proposed road project. Such pre-planning can reduce costs incurred when engineers must retrofit existing roads. Table 3-5. Scheduled bridge replacement construction projects, 1999-2004 FDOT District County Route No. Project Type Highest Rank Let Date Length (m) Environmental Feature 1 Charlotte I-75 Widen Bridge nr 00-01 2411 Peace River Corridor 1 Collier SR 951 New Bridge 3 97-? 488 Rookery Bay 1 Collier SR 951 Bridge Replacement nr 03-? 2313 Rookery Bay 1 Collier US 41 Bridge Replacement 3 00-? 67 Big Cypress NP 1 Collier US 41 Bridge Replacement nr 00-? 13 Big Cypress NP 1 Collier US 41 Bridge Replacement nr 96-99 13 Big Cypress NP 1 Collier US 41 Bridge Replacement nr 00-? 10 Big Cypress NP 1 Collier US 41 Bridge Replacement nr 00-? 10 Big Cypress NP 1 Collier US 41 Bridge Replacement nr 00-? 18 Big Cypress NP 1 Collier US 41 Bridge Replacement nr 00-? 10 Big Cypress NP 1 DeSoto SR 31 Bridge Replacement nr 96-99 79 Prairie Creek 1 DeSoto SR 70 Bridge Replacement nr 98-99 69 Horse Creek 1 DeSoto SR 70 Bridge Replacement 5 98-99 69 Horse Creek 1 DeSoto SR 70 Bridge Replacement 5 98-99 106 Horse Creek 1 DeSoto US 17 Bridge Replacement nr 98-99 28 Thornton Branch 1 Glades SR 29 Bridge Replacement 5 96-99 10 Canal, S of Fisheating Creek SOR 1 Glades SR 29 Bridge Replacement 5 96-99 13 Canal, S of Fisheating Creek SOR 1 Glades SR 29 Bridge Replacement 5 96-99 45 Canal, S of Fisheating Creek SOR

PAGE 142

126 Table 3-5. continued FDOT District County Route No. Project Type Highest Rank Let Date Length (m) Environmental Feature 1 Glades SR 29 Bridge Replacement 5 96-99 15 Canal, S of Fisheating Creek SOR 1 Hardee SR 636 Bridge Replacement nr 96-99 31 Peace River Corridor 1 Hardee SR 636 Bridge Replacement nr 96-99 66 Peace River Corridor 1 Hardee SR 64 Bridge Replacement nr 96-99 76 Charlie Creek 1 Hardee SR 66 Widen Bridge nr 96-99 37 Charlie Creek 1 Hardee SR 66 Widen Bridge nr 96-99 61 Charlie Creek 1 Hardee US 17 Bridge Replacement nr 96-00 92 Peace River Corridor 1 Hardee US 17 Bridge Replacement nr 96-99 50 Charlie Creek 1 Hendry SR 80 Bridge Replacement nr 98-99 37 Caloosahatchee Ecoscape, Long Hammock Cr. 1 Highlands SR 70 Widen Bridge 3 98-99 48 Fisheating Creek 1 Manatee SR 64 Bridge Replacement nr 99-00 132 Palma Sola Bay 1 Manatee SR 64 Bridge Replacement nr 99-00 43 Perico Bay 1 Manatee US 41 Widen Bridge nr 96-00 115 Terra Ceia Bay 1 Polk SR 60 Bridge Replacement nr 96-99 55 Alafia River 1 Polk SR 60 Bridge Replacement nr 96-99 69 Peace Creek 1 Polk SR 60 Bridge Replacement nr 96-99 27 Poley Creek 1 Polk SR 60 Bridge Replacement 5 00-01 963 Kissimmee River SOR 1 Polk US 92 Widen Bridge 5 96-00 50 Saddle Creek 1 Polk US 92 Widen Bridge 5 96-00 44 Lake Parker Drainage 2 Alachua SR 121 Bridge Replacement 5 00-01 1439 Santa Fe River Corridor 2 Duval SR 105 Bridge Replacement 5 00-03 410 Ft. George Inlet, Timucuan EP 2 Duval US 90 Bridge Replacement 5 00 1659 Intra-coastal Waterway 2 Duval US 90 Bridge Replacement nr 00-01 206 Brandy Branch 2 Gilchrist SR 47 Bridge Replacement 5 00 92 Santa Fe River Corridor 2 Lafayette CR 53 Bridge Replacement n/a 96-00 96 Mill Creek 2 Lafayette CR 53 Bridge Replacement n/a 00 216 4-mile Creek 2 Madison CR 150 Bridge Replacement n/a 96-99 39 Little Aucilla River 2 Madison US 90 Bridge Replacement nr 00 29 Suwannee Greenway 2 Nassau CR 121 Bridge Replacement n/a 96-00 202 Brandy Branch 2 Nassau CR 121 Bridge Replacement n/a 96-00 11 Mill Creek 2 Nassau US 17 Bridge Replacement nr 96-03 173 St. Marys River 2 Nassau US 17 Bridge Replacement 5 00 62 Nassau River Salt Marsh 2 Nassau US 17 Bridge Replacement 5 00 41 Nassau River Salt Marsh 2 Nassau US 17 Bridge Replacement 5 00 10 Lofton Creek 2 Nassau US 17 Bridge Replacement nr 00 19 McQueen Creek 2 Nassau US 17 Bridge Replacement nr 00 16 McQueen Swamp 2 Nassau US 90 Bridge Replacement nr 00 153 Deep Creek 2 Nassau US 90 Bridge Replacement nr 00 162 SCLRR Bridge 2 Putnam SR 20 Bridge Replacement nr 00 223 Sweetwater Creek 2 Putnam SR 20 Bridge Replacement 5 00 296 Rice Creek, Etonia CARL 2 St. Johns CR 13 Bridge Replacement n/a 00-04 127 Deep Creek 3 Escambia US 90 Bridge Replacement nr 96-01 499 Escambia River 3 Gadsden US 27 Bridge Replacement 4 98-99 199 Ochlocknee River Corridor 3 Gadsden US 27 Bridge Replacement 4 98-99 394 Ochlocknee River 3 Gulf SR 71 Bridge Replacement 4 96-01 77 Cypress Creek 3 Gulf SR 71 Bridge Replacement 5 96-01 66 West Arm Creek 3 Jackson SR 166 Bridge Replacement nr 96-01 169 Chipola River 3 Liberty CR 2224 Bridge Replacement n/a 99-00 247 Big Creek, North of Apalachicola NF 3 Okaloosa SR 85 Bridge Replacement nr 98-99 91 Eglin AFB, Boggy Bayou 3 Santa Rosa US 90 Bridge Replacement nr 96-01 82 Macavis Bayou 3 Santa Rosa US 90 Bridge Replacement nr 96-01 59 CSXRR Bridge 3 Wakulla US 98 Bridge Replacement 5 96-99 98 St. Marks River 3 Washington US 90 Bridge Replacement 4 98-99 764 Choctowatchee River 3 Washington US 90 Bridge Replacement nr 96-99 123 Cypress Slough 4 Brevard SR 520 Bridge Replacement nr 00-04 1444 Banana River

PAGE 143

127 Table 3-5. continued FDOT District County Route No. Project Type Highest Rank Let Date Length (m) Environmental Feature 4 Brevard SR 528 Bridge Replacement nr 00-04 1449 Indian River 4 Martin SR A1A New Bridge nr 98-99 809 St. Lucie River 4 Martin SR A1A Bridge Replacement nr 96-03 496 Indian River 4 St. Lucie SR 70 Bridge Replacement 5 97-01 10 Cypress Creek 4 St. Lucie SR 70 Bridge Replacement 5 97-01 18 Cypress Creek Branch 4 St. Lucie SR 70 Bridge Replacement 3 97-01 10 Relief Canal 4 St. Lucie SR 70 Bridge Replacement 3 97-01 10 Relief Canal 4 St. Lucie SR 70 Bridge Replacement 1 97-01 15 Cypress Creek 5 Flagler SR 100 Bridge Replacement 5 00-01 322 Creek to Crescent Lake 5 Flagler SR 11 Bridge Replacement nr 96-00 31 Sweetwater Branch 5 Flagler SR 11 Bridge Replacement nr 96-00 22 Canal 5 Flagler SR 11 Bridge Replacement nr 00 30 Black Branch 5 Flagler SR 11 Bridge Replacement 5 00 37 Middle Haw Creek 5 Lake SR 19 Bridge Replacement 5 98-00 28 Palatlakaha Creek 5 Lake US 27 Bridge Replacement nr 96-01 28 Lake Denham Lake Harris 5 Osceola SR 60 Bridge Replacement 5 96-01 148 Kissimmee River SOR 5 Osceola SR 60 Bridge Replacement nr 96-99 19 Blanket Bay Slough 5 Osceola SR 60 Bridge Replacement 5 96-99 40 C-54 Canal 5 Osceola US 17/92 Bridge Replacement 5 97-01 47 Reedy Creek 5 Osceola US 17/92 Bridge Replacement nr 97-01 35 Reedy Creek 5 Osceola US 17/92 Bridge Replacement nr 97-01 42 Reedy Creek 5 Seminole SR 46 Bridge Replacement 5 98-? 129 Lake Jessup 5 Volusia I-4 Bridge Replacement 3 00-04 5635 St. Johns River 5 Volusia SR 415 Widen Bridge 5 98-99 13 Alamana Canal 5 Volusia SR 44 Bridge Replacement 5 97-? 27 CSX RR, Spruce Creek 5 Volusia US 1 Bridge Replacement 5 96-00 103 Tomoka River Area 5 Volusia US 1 Bridge Replacement 5 96-00 79 Korona Canal 6 Monroe US 1 Bridge Replacement 4 97-02 66 Jewfish Creek 7 Citrus US 19/98 New Bridge 5 96-01 1910 Cross-FL Greenway 7 Hillsborough SR 39 Bridge Replacement 3 96-01 439 Blackwater Creek

PAGE 144

CHAPTER 4 MONITORING WILDLIFE USE AND DETERMINING STANDARDS FOR CULVERT DESIGN Introduction The FDOT has recently taken a proactive approach to construction of wildlife underpasses (e.g., Alligator Alley for Florida panthers, SR 46 for black bears). Certain issues need to be addressed prior to implementation of a statewide program designed to incorporate these provisions in all FDOT highway projects. One of these is the fact that little empirical research has been performed to correlate what species will use what type of crossing structure. Another issue involves cost of implementation. Design of new structures specifically for wildlife use is expensive and not cost effective. A preferred approach would be to use existing designs oriented toward hydraulic function, but serve dual purposes as wildlife passages and drainage structures (potential examples are provided in Figure 4-1). Study Description and Design Structural characteristics that may affect use by various faunal classes include composition, shape, texture, light penetration, moisture, and dimension (width, length, and height). Certain studies (Krikowski 1989, and Langton 1989) have shown importance of light and moisture on use by amphibian species. Size of structures has also been evaluated (Clevenger et al. 2001, Rodriguez et al. 1996, Yanes et al. 1995, and Hunt et al. 1987). 128

PAGE 145

129 Figure 4-1. Variety of existing structures that may function as wildlife underpasses.

PAGE 146

130 Some general rules of thumb have been proposed by landscape ecologists (i.e., bigger is better, and more is preferable to less) with regard to size and number of culverts per distance interval, in areas known for wildlife crossings. The biggest argument against using these rules of thumb, however, is cost effectiveness. Transportation engineers are interested in promoting protection for wildlife and public safety but at optimum efficiency, which involves minimizing costs. With this point in mind, it is therefore necessary to set standards based on the structural characteristics outlined above. The following research provides one approach to classification based on structural characteristics and suitability of use by various wildlife taxa. Objectives of this study include: 1) Identifying specific structure characteristics that enhance or detract from their suitability as wildlife crossing structures 2) Developing standards for culvert design based on relationships between criteria such as culvert type and configuration, landscape context (including species-habitat associations), and species preference This experiment evaluated culvert use by wildlife as a function of body size and environmental preference. Field data was analyzed by aquatic or terrestrial habitat preferences of each taxon or body size group. A conceptual list of faunal categories found in Florida by environmental preference is presented below (Table 4-1). Table 4-1. Faunal orders predicted to use culverts Aquatic (moist conditions) Terrestrial (dry conditions) Urodela (salamanders) Testudinata (gopher tortoise & box turtle) Anura (frogs) Squamata (lizards & snakes) Testudinata (aquatic turtles) Insectivora (moles & shrews) Crocodilia Rodentia Squamata (aquatic snakes) Carnivora Carnivora (river otters) Artiodactyla (deer)

PAGE 147

131 Alternatively groups could be established by the following body weight categories (Table 4-2). Table 4-2. Mammal body weight categories Body weight Smallest Largest 3 1000 g Southwest shrew Fox squirrel 1000 6000 g Swamp rabbit Red fox 6000 15,000 g Otter Nutria 15,000 30,000 g Bobcat Beaver 30,000 179,500 g Florida panther Black bear Note: A listing of herpetofaunal body weights was not available; however, the same approach would apply as with mammals above). Individuals from the faunal classes above (that were monitored using culverts) were also evaluated to determine preference according to structural characteristics. Three shapes of culverts existround, oval, and rectangular. Length of culverts is dependent on roadway width that most commonly varies from 8 m (2 lane) to 60 m (6 lane). Round and oval culvert width ranges from 30 cm to 3.8 m (FDOT 1989). Box culverts have great variability in dimension ranging from 0.30 m to 2.4 m (tall) and 0.30 to 7.3 m (wide); and can be constructed with single or multiple openings or cells (FDOT 1989). Width, length, and height are likely correlated in association with the "tunnel effect" that inhibits animals from passing through culverts. This results from an animals inability to sufficiently see a substantial area or destination at the other end of the culvert. As length increases, width must be increased to reduce the tunnel effect. Others have attempted to evaluate the impact of tunnel effect by developing an openness index (W x H)/L (Rodriguez et al. 1996, and Yanes et al. 1995). Methods and Study Area The study design included selection of replicates of two structure types (culverts and bridges) from 7 different land-cover classes. Using bridge and culvert categories

PAGE 148

132 reduced variance for explanatory structural variables. Study sites were randomly selected (using SAS software, SAS Institute, Cary, N.C.) from a database of previously surveyed structures (Chapter 3) that included information on structural and contextual parameters. A minimum of five replicates each of bridges and culverts were selected from each habitat/land-cover class (Table 4-3) at the 210 m level. In some cases, randomly selected sites had to be substituted because of flooding, construction, restricted access, or other uncontrollable factors. 2 Table 4-3. Land-cover classes Cate g or y Descri p tion 1 urban mining 2 dry prairie sand pine scrub/forest sandhill xeric oak scrub/forest 3 pinelands 4 cypress swamp hardwood swamp bay swamp shrub swamp cutover wetland forest bottomland hardwood forest freshwater marsh & wet prairie open water 5 grassland/pasture/agriculture intensive agriculture bare soil/clearcut 6 shrub/brush (range) land 7 mixed pine-hardwood forest hardwood hammock & forest Structures were monitored either twice weekly using track beds, or on a continual basis using remote, infrared-camera equipment. The area surrounding culvert entrances was also monitored. This design afforded the ability to evaluate culvert avoidance by particular species or faunal groups. Road-kills were also recorded at each study site. All

PAGE 149

133 species were recorded 50 m in each direction from crossing structures, while large mammals and alligators were recorded beyond 50 m. General soil moisture (5-point Likert response scale: very wet, wet, moist, dry, very dry) within each culvert was recorded during each visit. Height of right-of-way vegetation was recorded once monthly. Weekly rainfall and minimum and maximum temperatures for each site were acquired from National Climate Data Center (NCDC) weather stations closest to each monitoring site (maximum distance of 57 km). Benefits of the design included: 1) monitoring in the natural setting of those species present, 2) documenting simultaneously the effect of structural, environmental and habitat variables on effectiveness as wildlife passageways, 3) evaluating overlapping species-habitat associations to develop generalizations of culvert sizes needed in particular habitat types (e.g., Florida black bears use of multiple habitat types would overlap many smaller species that exist within the black bear's rangeumbrella effect), 4) identifying certain sites containing small culverts that are nonfunctional for larger species present in the area, and 5) identifying species that are "culvert avoiders". Track beds were prepared by clearing vegetation and debris (Figure 4-2), tilling and loosening existing soil, and when necessary applying substrate additives (e.g., builders sand) to ensure readable tracks. Problems encountered at various sites included vegetation overgrowth and mowing, rain, flooding, erosion, and washouts (that in some cases interrupted monitoring efforts or required repeated site preparation). Remote infrared-camera equipment used at camera sites included Trailmaster Models 550 and 1500, and the Camtrakker Wildlife-Pro Model. The Model 1500 is an active-sensor device that has a transmitter and a receiver. The Model 1500 system works by orienting the single infrared-beam, emitted by the transmitter, directly toward the receiver. When the path of the beam is obstructed, the camera is triggered and a photograph is taken. The range of the Model 1500 is approximately 33 m.

PAGE 150

134 before after Figure 4-2. Site preparation at track monitoring stations. Steps consisted of 1) Clearing vegetation, 2) Preparing soil (e.g., tilling and loosening), and 3) Applying soil additives (e.g., builders sand) when needed to ensure readable tracks. Frequent problems included flooding, erosion, and washouts.

PAGE 151

135 The Model 550 and the Wildlife-Pro are passive-sensor devices. The Model 550 emits an infrared array that consists of several beams, directed outward at equal interval angles that form a semicircle. The range of the infrared array is approximately 20 m. When an object interrupts two or more of the infrared rays, the camera is triggered and a photograph is taken. The Wildlife-Pro emits a cone-shaped infrared-beam with a range of approximately 20 m. When an object enters the path of the infrared beam, the camera is triggered and a photograph is taken. Timers on the sensors were set to function on a 24 hour cycle whenever possible. For sites where human activity was prevalent during the day, timers were set for nighttime hours only. Film and batteries were checked every two to three weeks depending on the amount of wildlife traffic at each site. To protect the equipment from the elements and to prevent theft, custom lock boxes were constructed out of military ammunition boxes. Figure 4-3 displays the remote-camera system attached to a bridge. Each lock box was fastened to bridge pilings by Tapcon concrete screws and steel cables to deter theft. The Model 1500 system was aligned approximately 30 to 35 cm above ground level. The other models were setup at various positions and angled toward the target area, depending on site characteristics. Statistics were performed (using SAS) on numeric variables for each monitoring site. These included basic-univariate-statistical measures, tests for normality, and distribution tables and plots. Multiple-logistic regression was performed to determine significance of numeric and class variables governing species use, recorded at each monitoring location. Logistic discrimination performs a logistic regression on the categorical variable, and assigns population membership to observations based on the associated explanatory variables (Khatree and Naik 2000). Frequency distributions for

PAGE 152

136 significant explanatory variables identified in logistic regression models, and thresholds for use by each faunal group were generated. Figure 4-3. Remote infrared camera system fastened to bridge pilings. This particular example is a Trailmaster 1500 active sensor; above is the transmitter unit, and to the right is the receiver unit and camera inside a customized security box. Independent variables used in the analyses included 34 factors: number of traffic lanes, right-of-way clearance, right-of-way gap, median width (on divided highways), AADT (annual-average-daily traffic), structuretype, composition, substrate, shape, size, and openness, distance to next nearest culvert, habitatdiversity and dominant types (primary and secondary) at 30, 210, 1020, and 2040 m 2 presence of humans and domestic predators, soil moisture, right-of-way vegetation type and height, rainfall, and minimum and maximum temperatures. The dependent variable measured for each taxon was whether animals that approached culvert entrances, passed through or not. Study sites were established in three regions in the state: central Florida (District 5), northeast Florida (District 2), and panhandle Florida (District 3). Culvert monitoring was conducted from March 2001 to December 2001 and July 2002 to March 2003. For the first study period, 22 camera sites and 247 track sites were monitored (Figure 4-4);

PAGE 153

137 during the second study period, 19 camera sites and 85 track sites were monitored (Figure 4-5). Of these, 61 sites were monitored during both study periods. Figure 4-4. Bridges and culverts monitored for wildlife use (March to December 2001). Sites were checked for tracks or monitored by infrared remote cameras on 37 roads in 32 counties. Results Results from field data collection and statistical analyses include three separate components. First, the selection of site replicates is presented, including univariate measures for each site and associated explanatory variables. Second, general findings and trends are discussed. Third, data reduction using multivariate-logistic regression and frequency distribution analysis are performed on field data. Assumptions for Multivariate Logistic Regression Analysis Four assumptions were addressed for discriminant multivariate-logistic regression: independent random sampling, multivariate normality, equality of variance-covariance matrices, and singularity and multicollinearity of explanatory variables.

PAGE 154

138 Figure 4-5. Bridges and culverts monitored for wildlife use (July 2002 to April 2003). Sites were chec k ed for tracks or monitored b y infrared remote cameras on 17 roads in 14 counties. Suitable replicates were randomly drawn from a sample of 1,232 field sites (Chapter 3) from across the State. Logistics (e.g., level of funding and staffing for the project) reduced the study area to north-central Florida and the eastern half of the panhandle (Figures 4-4 and 4-5). Coordinating driving time with twice-weekly site visits narrowed the sites available to routes that were efficient for timely and consistent data collection. The result included 290 bridge and culvert sites from the various land-cover types (Table 4-4). Enough culvert sites were found to provide more than the minimum 5 replicates for each primary land-cover class at the 210 m 2 scale, and included a total of 223 culvert monitoring sites. For bridge sites, suitable replicates were found for all primary land-cover classes except for classes 1 (urban/mining) and 2 (xeric-based land-covers).

PAGE 155

139 To obtain sufficient bridge replicates in these land-cover classes, secondary land-cover was included; and resulted in 12 additional sites for class 1, and 2 additional sites for class 2. Table 4-4. Replicates found for each structure type by land-cover class Structure Primary land-cover at the 210 m 2 scale t yp e 1 2 3 4 6 7 8 Total Culverts 12 18 77 27 24 52 13 223 Bridges 1 3 17 24 6 9 7 67 Total 13 21 94 51 30 61 20 290 Notes: 1) shaded cells indicate insufficient replicates for the primary land-cover class, secondary land-cover was included to provide 12 additional sites suitable as replicates for bridges in class 1. Two additional sites had the same secondary land-cover for bridges in class 2 that resulted in availability of five replicates for this category. 2) Table 4-3 provides land-cover descriptions. Sample size, mean, standard deviation, and normality of numerical structural, environmental, and ecological explanatory variables for the 290 monitoring sites are shown in Table 4-5. The variables in the table represent 14 static and 4 dynamic factors. The 14 static measures had the same value for the duration of the study. Normality curves were not generated for number of lanes, structure number, and the 4 habitat diversity factors that consisted of integer values; mean and standard deviation was rounded to the nearest whole number. The 4 dynamic variables that were measured throughout the study includedrainfall, minimum and maximum temperature, and right-of-way vegetation height. The other 16 explanatory variables included were non-numericprimary and secondary habitat (30, 210, 1020, 2040 m 2 ), water feature type, vegetation type, soil moisture, and other structural characteristics (i.e., type, composition, shape, substrate, and number).

PAGE 156

140 Table 4-5. Univariate measures and distributions of explanatory variables Explanatory variable n Mean Standard deviation Goodness-of-fit test for normalit y (KolmogorovSmirnov) All sites (n = 290) Number of lanes 290 3 1 Road clearance width, m 290 39.07 20.80 p < 0.01 Gap width, m 269 6.87 4.84 p = 0.058* Median width, m 290 14.08 5.57 p < 0.01 AADT 275 7165 6916 p > 0.014* Distance between structures, m 278 921.23 580.11 p > 0.15* Habitat diversity 30 m 2 290 3 1 Habitat diversity 210 m 2 290 3 1 Habitat diversity 1020 m 2 290 6 1 Habitat diversity 2040 m 2 290 6 1 R-o-w vegetation height, cm 13,060 48.47 31.02 p < 0.01 Average minimum temperature, C 15,548 15.18 6.08 p < 0.01 Average maximum temperature, C 15,558 27.47 4.74 p < 0.01 Rainfall, ml 15,671 2.65 4.07 p < 0.01 Bridges (n = 67) Openness index (w x h / l) 67 12.95 12.07 p > 0.15* Width (w), m 67 61.94 41.99 p > 0.15* Length (l), m 67 20.11 13.79 p > 0.15** Height (h), m 67 3.77 2.79 p = 0.117** Culverts (n = 223) Structure number 223 1 1 Openness index (w x h / l) 223 0.077 0.12 p < 0.01 Width (w), m 223 1.39 0.99 p < 0.01 Length (l), m 223 25.23 13.21 p < 0.01 Height (h), m 223 0.93 0.48 p < 0.01 lognormal data conversion ** exponential data conversion Note the significant difference in the means of openness index, width, and height for bridges and culverts (Table 4-5). This high variability explains why the two structure types were evaluated separately. Plots and goodness-of-fit tests for univariate normality of explanatory variables for bridges (n = 67) and culverts (n = 223) revealed skewed or peaked distributions. Lognormal and exponential data transformation and removal of extreme observations was used to fit normal data curves. Goodness-of-fit was corrected

PAGE 157

141 following data transformation for openness index, width, length, and height on bridge sites and AADT, gap width, and distance for all sites (Table 4-5). Although goodness-of-fit values (Kolmogorov-Smirnov Test) were not significantly changed from data transformation, lognormal curves were the closest approximation for culvertopenness, width, length, and height. Rainfall most closely fit an exponential curve, whereas right-of-way vegetation height was most similar to a lognormal distribution. Distribution curves of temperature parameters were not significantly improved through data transformation and/or removal of extreme observations, so original values were used. Although these did not strictly meet univariate normal requirements, they did not grossly violate them either (e.g., minimally elevated skewness and kurtosis values). They were therefore included in the model using the most appropriate data transformation method. Logistic regression still provides a robust method when multivariate distributions are nonnormal (McGarigal et al. 2000). Levenes test of homogeneity generated in SAS from one-way ANOVA performed on explanatory variables revealed equality for within-group variance for all groups (F statistic, all groupsp < 0.0001, ungulatesp < 0.004) except the alligator-aquatic mammal group. Significant overlap was evident between certain explanatory variables. To adhere to singularity and multicollinearity rules for use of logistic regression, and to improve fit of models, it was necessary to remove these. Elements of right-of-way width were expressed in three different variablesnumber of lanes, median width, clearance, and gap width (pairwise analysisPearsons correlation coefficient, p < 0.0001). The first three were removed from the analysis, as focus was most appropriately placed on gap

PAGE 158

142 width (the distance wildlife must traverse between adjacent habitat and the entrance of the crossing structure). Multicollinearity (Pearsons correlation coefficient, p = 0.0002) was also apparent between the openness index and individual structural measurements (i.e., width, length, and height). Recent studies (Bertwhistle 2003, and Garrett and Gordon 2003) reveal that the openness index may overlook the importance of specific structural dimensions to use by certain wildlife; as the mathematical function (width x height / length) can be manipulated in each of the three dimensions to derive the same index value. Due to this, the openness index was not included in the regressions; even so, its significance was included in the discussion. Concern for potential correlation between soil moisture (categorical) and rainfall (continuous) variables resulted in removal of soil moisture (the least precise of the two variables) from the analysis. Logistic regression models were optimized using backward stepwise processing to select the number of independent factors (determined by Akaike's Information Criterion (AIC) and -2 log L model fit criteria in SAS). Additional explanatory variables were removed following initial iterations of the regressions because of insignificance. These included: structure composition, water feature, habitat diversity (1020 and 2040 m 2 resolution), and dominant habitat types (primary and secondary at 1020 and 2040 m 2 resolution, and secondary at 30, 210, 1020, and 2040 m 2 resolution). This reduced the number of explanatory variables used in logistic regression models to 20. General Findings Over the course of 10 months in 2001 and 9 months in 2002-03, nearly 48,000 records were collected from monitoring sites (Table 4-6). These data were divided into three categories: events, nonevents, and road-kills. Events include records of wildlife

PAGE 159

143 movement, either through or in proximity to culverts and bridges (evidenced either by photographs or tracks). Among these events, certain records could not be identified and were classified as such. Nonevents include three different types of records: 1) none (refers to site visits when no tracks were present), 2) washouts (site encounters with washed-out track-beds or flooding following rain events), and 3) vehicle downtimes, thunderstorms, or other circumstances when data collection could not conducted (these are not reported here). Table 4-6. Events, nonevents and road-kills recorded at culverts and bridges Total Events Tracks Photographs Unidentified Nonevents Washouts None Roadkills Total 47,955 36,870 33,678 852 2,340 10,822 3,835 6,987 263 Percent 77 70 2 5 23 8 15 0.5 # sites 290 278 39 211 127 243 125 Yr 02-03 21,400 18,607 17,557 515 535 2,650 2,523 127 143 Percent 87 82 2 3 13 12 1 0.3 # sites 97 67 18 43 81 31 44 Yr 01 26,555 18,263 16,121 337 1,805 8,172 1,312 6,860 120 Percent 69 61 1 7 31 5 26 0.25 # sites 254 211 21 168 46 212 81 Overall, track records (n = 33,678) accounted for 70% of data collected (Figure 4-6); photographs accounted for only 2% of the data (Figure 4-7), and more or less served as a verification method for identification of tracks at certain sites. Five percent of records collected could not be identified (Figure 4-8) and were not used in statistical analyses. No tracks were found at monitoring sites 6,987 times (Figure 4-9), and washouts occurred 3,835 times (Figure 4-10). Road-kills recorded included 263 individual vertebrates, accounting for less than one percent of the data collected (Figure 4-11). The routine for field site visits (twice per week) proved inadequate to effectively estimate road-kill counts on road segments (where structures were monitored for tracks). This limited the ability to evaluate culvert avoidance by specific wildlife.

PAGE 160

144 Figure 4-6. Number of tracks recorded at culverts and bridges (3/01 to 12/01 and 7/02 to 4/03). Sites where larger dots are shown had more records for two possible reasonsmost were included in both study periods, and some correspond to bridge locations that consistently had more wildlife use. Figure 4-7. Number of photographs taken at culverts and bridges (3/01 to 12/01 and 7/02 to 4/03). N umbers dis p la y ed onl y re p resent p hoto g ra p hs p ositivel y identified.

PAGE 161

145 Figure 4-8. Unidentified tracks recorded at culverts and bridges (3/01 to 12/01 and 7/02 to 4/03). Sites where larger dots are shown had more records for two possible reasonsmost were included in both study periods, and some correspond to bridge locations that consistently had more wildlife use. Figure 4-9. Number of visits when no tracks were found at culverts and bridges (3/01 to 12/01 and 7/02 to 4/03). The central peninsula and panhandle were only monitored during the first study period.

PAGE 162

146 Figure 4-10. Number of visits when sites were flooded or washed out (3/01 to 12/01 and 7/02 to 4/03). The central peninsula and panhandle were only monitored during the first study period. Annual-average rainfall was higher during the second monitoring period and corresponds to the location and number of flood events shown. Most flooded sites contained surface water features. Figure 4-11. Road-kills recorded from 3/01 to 12/01 and 7/02 to 4/03. On average, less than two road-kills were recorded at each site. The central peninsula and panhandle were only monitored during the first stud y p eriod.

PAGE 163

147 Certain field sites produced exceptional results. Nine sites were flooded and impassable on average for 58% (42 of 72) of site visits, and only included road-kill records (n = 19). No tracks or photographs were recorded at 24 separate sites. In 2001 and 2002-03, 16,458 (48%) and 18,072 (52%) tracks and photographs, respectively were recorded. A total of 14,751 track records were recorded from 55 bridge sites, an average of 268 per site. Culvert sites produced 18,927 track records from 222 sites (site average = 85). Findings by Faunal Groups For all monitoring sites, 41 different organisms or categories were recorded. To perform statistical analysis, these were consolidated into 14 general groups (Table 4-7). Faunal groups were determined according to taxa and body size (discussed previously), mode of movement, and environmental preference. Logistics of monitoring and resource availability restricted the focus of the study to terrestrial organisms, the exception being those aquatic dependents (e.g., river otter, alligator) that used terrestrial areas for movement (either adjacent to aquatic systems or to get from one water source to the next). Faunal groups of interest included: meso-mammal, carnivore, bird, ungulate, herpetofauna, and small mammal. Statistics regarding use or avoidance of culverts and bridges were performed on these six categories. The other eight categories recorded were either discarded or used in the analysis of the other categories of interest. The category none referred to site visits when no tracks were detected; and were not applicable to the analysis unless a particular site revealed no movement for the course of the study. These were reviewed separately to examine why no movement was evident. The category alligator and aquatic

PAGE 164

148 mammal did not produce enough samples (n = 25 and n = 41, respectively) to perform multivariate statistics. Recording tracks in the category arthropod proved too intensive a task at the scale of the study and frequency of site visits. The importance of this groups use of connecting structures along roads is recognized; however, a separate study would be necessary to properly evaluate this faunal group. Records of presence and use by domestics and humans were applied as explanatory variables for the other categories. The category other represented those that could not be identified, and therefore could not be included in the regression analysis; additionally n/a represents failed attempts to perform monitoring duties and were also ignored in the analysis. Table 4-7. Faunal groups Group name Group code none 0 alligator 1 meso-mammal 2 arthropod 3 carnivore 4 bird 5 ungulate 6 domestic 7 herpetofauna 8 human 9 small mammal 10 aquatic mammal 11 other 12 n/a 30 The different organisms recorded and their respective group assignments are shown in Table 4-8. Forty-four were identified to the species level, two to the family level, five to the order/suborder level, four to the class level, and 4 other types. Results are presented for each faunal group by structure type (bridge or culvert) and the three groups of explanatory variablesstructural, environmental, and ecological.

PAGE 165

149 Table 4-8. Species recorded and group assignments Group name Code Common name Scientific name none 0 none alligator 1 alligator A lligator mississippiensis meso-mammal 2 armadillo D asypus novemcinctus meso-mammal 2 opossum D idelphis marsupialis meso-mammal 2 raccoon P rocyon loto r meso-mammal 2 skunk M ephitis sp., Spilogale sp. arthropod 3 insect Insecta arthropod 3 spider Arachnida arthropod 3 centipede Chilopoda carnivore 4 bear Ursus americanus carnivore 4 bobcat L ynx rufus carnivore 4 coyote Canis latrans carnivore 4 fox Urocyon sp., Vulpes sp. bird 5 wild turkey M eleagris gallopavo bird 5 great blue heron A rdea herodias bird 5 common bobwhite Colinus virginianus bird 5 wading bird Ardeidae bird 5 perching bird Passeriformes ungulate 6 white-tailed deer Odocoileus virginianus ungulate 6 wild pig / hog Sus scrofa domestic 7 domestic cat F elis catus domestic 7 domestic dog Canis familiaris herpetofauna 8 frog / toad Anura herpetofauna 8 s. leopard frog R ana utricularia herpetofauna 8 southern toad B ufo terrestris herpetofauna 8 green tree frog H yla cinerea herpetofauna 8 Cuban tree frog Osteopilus septentrionalis herpetofauna 8 lizard Squamata: Lacertilia herpetofauna 8 6-lined racerunner Cnemidophorus sexlineatus herpetofauna 8 5-lined skink E umeces spp. herpetofauna 8 anole A nolis spp. herpetofauna 8 fence lizard Sceloporus undulatus herpetofauna 8 snake Squamata: Serpentes herpetofauna 8 cottonmouth Ag kistrodon piscivorus herpetofauna 8 timber rattlesnake Crotalus horridus herpetofauna 8 s. black racer Coluber constrictor herpetofauna 8 yellow rat snake E laphe obsoleta herpetofauna 8 e. garter snake Thamnophis sirtalis herpetofauna 8 turtle / tortoise Testudines herpetofauna 8 snapping turtle Chelydra serpentina herpetofauna 8 alligator snapper M acroclemys temminckii

PAGE 166

150 Table 4-8. continued Group name Code Common name Scientific name herpetofauna 8 Florida cooter P seudemys floridana herpetofauna 8 striped mud turtle K inosternon baurii herpetofauna 8 box turtle Terrapene carolina human 9 human H omo sapiens sm mammal 10 rabbit Sylvilagus spp. sm mammal 10 eastern cottontail Sylvilagus floridana sm mammal 10 mouse Cricetidae sm mammal 10 rat Cricetidae sm mammal 10 Fl. water rat N eofiber alleni sm mammal 10 mole Scalopus aquaticus sm mammal 10 gray squirrel Sciuris carolinensis aquatic mammal 11 river otter L utra canadensis aquatic mammal 11 beaver Castor canadensis other 12 livestock Bovidae other 12 mammal Mammalia other 12 other other 12 unknown n/a 30 n/a Note: Table excludes recorded road-kill species that are discussed separately. Meso-mammal use of bridges Meso-mammal presence at 58 bridge sites included records (n = 6,480) of five different species: armadillo Dasypus novemcinctus, Virginia opossum Didelphis marsupialis, raccoon Procyon lotor, and skunk Mephitis sp. or Spilogale sp. Logistic regression model. The model was significant in predicting use of bridges by meso-mammals: Wald X 2 : 390.01 (df, p <.0001), Nagelkerkes R 2 (0.4005), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 14.52 (df, p = 0.0014). Eleven of 21 factors were found significant (p < 0.03) by the model in predicting use of bridges by meso-mammals. Percentage of movement under bridges correctly predicted by the model was 91.1%. Response to structural characteristics. Six structural parametersgap width (distance between bridge entrance and adjacent habitat), annual-average-daily traffic

PAGE 167

151 (annual-average of number of vehicles-per-day), bridge width (width of bridge opening), bridge length (distance that organisms had to travel to pass through the structure), bridge height, and distance to next nearest crossing structurewere significant in the logistic regression model for meso-mammal use of bridges. The majority of meso-mammals using bridges (75.76%, n = 4,687 of 6,186) were recorded when gap width was 8.5 m or less; presence but not passage (84.69%, n = 249 of 294) was also associated with gap widths of 8.5 m or less. Gap width (distance from adjacent habitat to bridge entrance) ranged from 0 31.1 m. When AADT was greater than 7,500, passage under bridges by meso-mammals (n = 6,179) occurred only 13.9% during the period monitored. The highest percentage of passages by meso-mammals (28.39%) were recorded when AADT was at 250 vehicles-per-day. Presence recorded near bridges, but not crossing from one side to the other, primarily occurred (61.56%, 181 of 294) when AADT was 3,000. Range of AADT on roads where meso-mammals were recorded was 250 to 69,374. Of 6,186 records of meso-mammals using bridges, three peaks occurred when bridge width (width of bridge opening) was 19 60 m (89.09%). Sixty-two percent of records of presence but not passage (n = 182 of 294) occurred at two bridge widths, 41 and 60 m. Where meso-mammals were recorded, bridge widths ranged from 7.3 200 m. Of 6,186 records of meso-mammals using bridges, most occurred (75.45%) when bridge length was less than 15 m. Presence but not passage (n = 294) was also associated predominantly with bridge lengths less than 15 m (84.69%). Bridge length used by meso-mammals ranged from 6.4 80 m. Of 6,186 records of meso-mammals using bridges, most occurred (92.65%) when bridge height

PAGE 168

152 was between 1.5 to 3.7 m. Presence but not passage (n = 294) was associated primarily (75.17%) with bridge heights of 3.1 3.7 m. Bridge height ranged from 1.2 15 m. When distance between structures was greater than 1,000 m, presence (but not passage under) bridges by meso-mammals (n = 294) occurred 76.53% during the period monitored. Records of passage under bridges (n = 6,174) were relatively even across all distances. Range of distance between structures was 130 to 3,450 m. Response to environmental characteristics. Recorded passage under bridges (n = 6,186) occurred 73.68% of the time during fall and summer months. High passage rates occurred for all seasons for meso-mammals (minimum of 88.85% in winter). Of 6,480 that approached bridge entrances, 6,186 passed through to the other side. When average maximum weekly temperature was 20 degrees C or higher, meso-mammals passing under bridges occurred 91.12% (n = 5,637 of 6,186) of the time recorded. Presence at (but not passage under) bridges showed an opposite trend regarding high temperatures; 88.44% of these records occurred when temperatures were 27 degrees C or less. Average maximum weekly temperature at bridges used by meso-mammals ranged between 12 and 34 degrees C. Response to ecological characteristics. Four ecological characteristicshabitat diversity at 30 m 2 resolution, primary habitat type at 210 m 2 resolution, right-of-way vegetation height, and presence of domestic predatorswere significant in the logistic regression model for meso-mammal use of bridges. Meso-mammals recorded using (n = 6,186) and present at (n = 294) bridges were greatest (80.71% and 90.14%, respectively) when 3 land-cover types existed at the 30 m 2 cell resolution. Habitat diversity ranged from 1 to 4 land-cover types at the 30 m 2 scale.

PAGE 169

153 Passage under bridges by meso-mammals (n = 6,186) was the highest for 210 m 2 scale land-cover classes 3 (pinelands, 18.96%), 4 (wetlands, 50.74%), and 6 (shrub and brushlands, 16.86%). Presence at (but not passage under) bridges (n = 294) was greatest for land-cover classes 4 (58.84%) and 5 (grasslands and agriculture, 23.47%). Passage rates were high (minimum of 86.71%) for all classes (n = 7). Meso-mammal movement under bridges (n = 585) was recorded most frequently (86.67%) when vegetation height was between 30 and 61 cm. Right-of-way vegetation height at monitoring sites ranged from 0 to 150 cm. Use by meso-mammals was greatest (n = 4,841 of 6,186, 78.25%) when recorded presence of domestic predators was 3 or less. Contrary to this observation, use and occurrence of meso-mammals was high when number of domestic predators recorded was 32 and 104 (32.77% and 6.66%, and 104.80% and 10.05%, respectively). This demonstrates an anomaly to the general downward trend in meso-mammal use as a result of increased presence of domestic predators. Presence of domestic predators ranged from 0 104 records at bridges used by meso-mammals. Meso-mammal use of culverts Presence of meso-mammals at 162 culvert sites included records (n = 5,957) of five different species: armadillo Dasypus novemcinctus, Virginia opossum Didelphis marsupialis, raccoon Procyon lotor, and skunk Mephitis sp. or Spilogale sp. Logistic regression model. The model was significant in predicting use of culverts by meso-mammals: Wald X 2 : 577.04 (df, p <.0001), Nagelkerkes R 2 (0.3447), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 61.55 (df, p < 0.0001). Note that R 2 is less significant for meso-mammals using culverts, than for

PAGE 170

154 meso-mammals using bridges. Ten of 21 factors were found significant (p < 0.03) by the model in predicting use of culverts by meso-mammals. Percentage of movement through culverts correctly predicted by the model was 82.4%. Response to structural characteristics. Five structural characteristics were significant in multiple logistic regression analysis. The majority of meso-mammals using culverts (77.44%, n = 4,271 of 5,515) were recorded when gap width was 7.5 m or less; presence but not passage (52.72%, n = 232 of 442) was frequently associated with gap widths of 8.5 m or more. Gap width (distance from adjacent habitat to culvert entrance) ranged from 0 39.3 m. When AADT was from 3,200 to 8,400, records of passage through culverts by meso-mammals (n = 5,472) occurred 76.9% of the time. Presence recorded near culverts, but not crossing from one side to the other, was primarily recorded (76.02%) when AADT was 6,400 or higher. Range of AADT on roads where meso-mammals were recorded was 250 to 42,500. For meso-mammals using culverts (n = 5,515), most occurred (93.36%) when culvert width was between the range of 0.5 to 3.5 m. Highest percentage of passage (19.37%) occurred at 3.1 m width. Presence but not passage (n = 442) was also associated predominantly (98.64%) with culvert widths of 0.5 to 3.5 m. Highest percentage of presence but not passage (42.08%) occurred at 0.9 m width. Width of culverts used by meso-mammals ranged from 0.3 4.3 m. Seventeen meso-mammals were recorded passing through culverts at the minimum width monitored of 0.3 m. Of 5,515 records of meso-mammals using culverts, most occurred (87.86%) when culvert

PAGE 171

155 height was between 0.6 to 1.5 m. Presence but not passage (n = 442) was also associated (89.83%) with culvert heights of 0.6 m to 1.5 m. Culvert height ranged from 0.3 3.7 m. Passage (n = 5,515) by meso-mammals through culverts occurred relatively even across all distance intervals between crossing structures, with the exception of 950 m (16.77%). Presence (n = 442) by meso-mammals at culvert entrances was also recorded evenly, with one peak at 1,330 m (18.78%). Range of distance between structures was 110 to 5,750 m. Response to environmental characteristics. Effects of three environmental measures identified from the logistic process are reported. Recorded passage through culverts (n = 5,515) was high during spring (26.24%), summer (33.42%) and fall (26.59%) months. Presence at (but not passage through) culverts (n = 442) was recorded mostly in fall and summer months (81.67%). Of meso-mammals passing through culverts (n = 5,489) and present at culvert entrances (n = 440), most occurred (83.41% and 65.22%, respectively) when average maximum weekly temperature was 24 degrees C or higher. When average maximum weekly temperature was from 26 to 33 degrees C, 70.91% of movement through culverts was recorded. Average maximum weekly temperature at monitoring sites used by meso-mammals ranged between 12 and 35 degrees C. Of 5,492 records of meso-mammals using culverts, most occurred (92.43%) when rainfall was 6 ml or less; and presence but not passage was also recorded predominantly (90.23%, n = 398 of 441) when rainfall was 6 ml or less. Average weekly rainfall at monitoring sites used by meso-mammals ranged between 0 and 25 ml.

PAGE 172

156 Response to ecological characteristics. Primary habitat type at 210 m 2 scale and influence of presence of humans and domestic predators were identified as significant factors regarding culvert use by meso-mammals. Passage through culverts (n = 5,515) by meso-mammals was the highest for 210 m 2 scale land-cover class 3 (pinelands, 22.41%) and lowest for class 2 (hardwood forests, 4.1%). Presence at (but not passage through) culverts (n = 442) was greatest for land-cover classes 3 (52.26%) and 4 (16.06%). Passage rates were high (minimum of 84.25%) for all classes (n = 7). Use by meso-mammals decreased, when presence of humans increased at monitored sites. When no humans were recorded, use of culverts (n = 5,515) occurred 90.79% of the time. Presence at (but no passage through) the culvert also was greatest (90.95%, n = 402 of 442) when no human presence was recorded. Presence of humans ranged from 0 4 records at sites used by meso-mammals. When presence of domestic predators increased at monitored sites, use by meso-mammals also decreased. Use of culverts (n = 5,515) occurred 76.28% of the time when one or less domestic predators were recorded. Similarly, presence at (but no passage through) the culvert was greatest (95.25%, n = 421 of 442) when recorded domestic predator presence was less than 2. Presence of domestic predators ranged from 0 4 records at culverts frequented by meso-mammals. Carnivore use of bridges Carnivore presence at 34 bridge sites included records (n = 384) of five species: black bear Ursus americanus, coyote Canis latrans, bobcat Lynx rufus, and fox Urocyon sp. or Vulpes sp.

PAGE 173

157 Logistic regression model. The model was significant in two of three tests for predicting use of bridges by carnivores: Wald X 2 : 45.30 (df, p <.0001), Nagelkerkes R 2 (0.5375), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 12.79 (df, p = 0.1193). Other combinations of explanatory variables were used in model simulations without improvement to the score for goodness-of-fit test. Partitions created for the goodness-of-fit test encountered too many zeros (for observed values in the nonuse category) and predicted 26% false negatives. Thus the model fit is questionable. Only 31 records were documented for nonuse of bridges, likely accounting for the weak score for the goodness-of-fit test. Only 6 of 21 factors were found significant (p < 0.05) by the model in predicting use of bridges by carnivores. Percentage of movement under bridges correctly predicted by the model was 90.9%. Response to structural characteristics. Two significant structural parameters are described here that were identified by the regression model. Records of carnivores using bridges (n = 353) occurred most often (92.91%) when bridge height was less than 4 m. Presence but not passage (n = 31) was also associated predominantly (83.88%) with bridge heights less than 4 m. Height of monitored bridges used by carnivores ranged from 1.5 15 m. Of 353 records of carnivores using bridges, two peaks occurred when distance to next nearest crossing structure was either 575 600 m (41.77%), or greater than 1,500 m (34.81%). Presence but not passage (n = 31) was notable regarding one distance, 1,050 m (53.33% of occurrences). Where carnivores were found, distance between structures ranged from 200 3,450 m.

PAGE 174

158 Response to environmental characteristics. Season, temperature, and rainfall were all identified as significant factors associated with carnivore activity recorded near bridges. Carnivore passage under bridges (n = 353) occurred mostly during summer (25.21%), fall (36.83%), and winter (23.23%) months. Upon approach to bridge entrances carnivore passage rates were high for all seasons, 353 of 384 (91.93%). Presence at (but not passage under) bridges (n = 31) was recorded most frequently in fall and winter (80.64%). Of 341 records of carnivores passing under bridges, most occurred (72.14%) when average, maximum, weekly temperature was 24 degrees C or higher. Presence at, (but not passage under) bridges (n = 31) showed a reverse trend where most movement (80.63%) was recorded when average, maximum, weekly temperature was 25 degrees C or less. Average, maximum, weekly temperature at these sites ranged between 12 and 34 degrees C. Of 341 records of carnivores using bridges, most occurred (87.38%) when rainfall was 5 ml or less; and presence but not passage was also recorded predominantly (96.77%, n = 30 of 31) when rainfall was 5 ml or less. Average weekly rainfall at monitoring sites where carnivores were recorded ranged between 0 and 18 ml. Response to ecological characteristics. Only domestic predators were identified in the regression analysis as a significant factor. When presence of domestic predators increased at monitored sites, use by carnivores decreased. Passage by carnivores (n = 302 of 353, 85.55%) coincided with 6 or less records of domestic predators. Approximately 51% of carnivores, that approached but did not pass under bridges, were

PAGE 175

159 recorded at sites where 104 domestic predators had been recorded. Presence of domestic predators ranged from 0 104 records at sites monitored. Carnivore use of culverts Records (n = 315) of carnivores at 51 culvert sites included five species: black bear Ursus americanus, coyote Canis latrans, bobcat Lynx rufus, and fox Urocyon sp. or Vulpes sp. Logistic regression model. The model was significant in two of three tests for predicting use of culverts by carnivores: Wald X 2 : 59.95 (df, p <.0001), Nagelkerkes R 2 (0.6462), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 12.74 (df, p = 0.1211). Other combinations of explanatory variables were used in model simulations without improvement to the score for goodness-of-fit test. Partitions created for the goodness-of-fit test encountered too many zeros (for observed values in the nonuse category) and predicted 24% false negatives. Thus the model fit is questionable. Only 57 records were documented for nonuse of bridges, likely accounting for the weak score for the goodness-of-fit test. Only 5 of 21 factors were found significant (p < 0.02) by the model in predicting use of culverts by carnivores. Percentage of movement through culverts correctly predicted by the model was 91.9%. Response to structural characteristics. Four structural factors were identified in regression analysis. Carnivore movement (n = 258) through culverts was greater the less culvert cells were available (i.e., 1.85%, 2.55%, 3.58%). Carnivore presence (n = 57) at culvert entrances was highest for single and three cell configurations (50.88% and 42.11%, respectively).

PAGE 176

160 When AADT was from 3,200 to 11,000, records of passage through culverts by carnivores (n = 258) occurred 90.32% of the time. Presence recorded near culverts, but not crossing from one side to the other, was primarily recorded (84.23%) when AADT was 6,900 or higher. Range of AADT on roads where carnivores were recorded was 250 to 31,500. When culvert width was greater than 1 m, most records of carnivores using culverts occurred (n = 217 of 258; 84.3%). Highest percentage of passage (36.78%) occurred at 2.4 m width. Presence but not passage (n = 57) increased as culvert widths decreased (94.43% occurrence at widths less than 2 m). Highest frequency (53.7%) of presence (but not passage) occurred at 0.9 m width. Culvert widths where carnivores were recorded ranged from 0.3 3.7 m. Of 258 records of carnivores using culverts, most occurred (86.03%) when distance to next nearest crossing structure was 875 m or more. Presence but not passage (n = 57) was notable regarding distances from 900 to 1,360 m (65% of occurrences). Where carnivores were found, distance between structures ranged from 125 5,750 m. Response to ecological characteristics. When presence of domestic predators increased at culvert sites, use by carnivores decreased. Most passage by carnivores (92.15%, n = 238 of 258) coincided with 2 or less records of domestic predators. Presence of domestic predators ranged only from 0 4 records at culvert sites where carnivores were recorded. Avian use of bridges Avian presence at 27 bridge sites included records (n = 208) of three specieswild turkey Meleagris gallopavo, common bobwhite Colinus virginianus, and great blue

PAGE 177

161 heron Ardea herodias; one family of wading birdsArdeidae (herons and egrets); and one orderPasseriformes (songbirds or perching birds). Logistic regression model. The model was significant for predicting use of bridges by birds: Wald X 2 : 22.27 (df, p <.0001), Nagelkerkes R 2 (0.5386), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 33.28 (df, p < 0.0001). Only 5 of 21 factors were found significant (p < 0.01) by the model in predicting use of bridges by birds. Percentage of movement under bridges correctly predicted by the model was 91%. Response to structural characteristics. Three structural factors were identified in regression analysis for birds and bridges. The majority (75.68%) of birds using bridges (n = 177) were recorded when gap width was 8.5 m or less; presence but not passage was also associated mostly (83.87%, n = 26 of 31) with gap widths of 8.5 m or less. Gap width (distance from adjacent habitat to bridge entrance) ranged from 5.3 28.4 m. Records of birds using bridges occurred most often (81.89%) when bridge width was less than 50 m. Presence but not passage (n = 31) was associated principally (80.65%) with bridge widths of 38 to 60 m. Bridge width of sites monitored ranged from 7.3 200 m. Of 177 records of birds using bridges, most occurred (92.64%) when bridge height was between 1.5 to 3.7 m. Presence but not passage (n = 31) was associated primarily (16.13%) with bridge heights of 3.1 m (74.19%) and 7.3 to 7.6 m. Bridge height ranged from 1.2 7.6 m. Response to environmental characteristics. Only average, minimum, weekly temperature was a significant factor from logistic regression analysis. Records of birds using bridges (n = 165) were distributed relatively even; average, minimum, weekly

PAGE 178

162 temperature at monitoring sites ranged between -2 and 23 degrees C. For birds only recorded at bridge entrances (n = 30), most occurred (86.66%) when average, minimum, weekly temperature was 10 degrees C or lower. Response to ecological characteristics. Habitat diversity at 30 m 2 resolution was the single significant factor identified in logistic regression. Birds recorded (n = 177) using bridges were greatest (89.93%) when 3 land-cover types were present at the 30 m 2 cell resolution. Similarly, presence of birds (n = 31) near bridge entrances was greatest (93.55%) when 3 land-cover types were present. Only 1 record occurred at sites where the maximum 4 land-cover types existed. Avian use of culverts Avian presence from 16 culvert sites included records (n = 115) of one specieswild turkey Meleagris gallopavo; one family of wading birdsArdeidae (herons and egrets); and one orderPasseriformes (songbirds or perching birds). Logistic regression model. The model was significant in two of three tests for predicting use of culverts by birds: Wald X 2 : 11.59 (df, p= 0.003), Nagelkerkes R 2 (0.8582), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 0.51 (df, p = 0.99). Large sample sizes are preferred for conducting logistic regression analysis and thus the small sample size (n = 115) likely accounts for the poor score for the goodness-of-fit test. Accuracy of the model is questionable. Only 2 of 21 factors were found significant (p < 0.01) by the model in predicting use of culverts by birds. Percentage of movement through culverts correctly predicted by the model was 96.2%. Response to structural characteristics. Culvert width was the only structural measure of importance in the logistic model for birds and culverts. Of 81 records of birds

PAGE 179

163 using culverts, nearly all occurred (98.76%) when culvert width was greater than 2 m. Presence but not passage (n = 34) was more variable (e.g., 0.6 m.35% and 2.4 m47.06%). No birds were recorded passing through culverts less than 1.5 m wide. Culvert width ranged from 0.3 3.1 m. Response to environmental characteristics. Birds passing through culverts (n = 81) occurred most often (75.31%) when average, maximum, weekly temperature was 25 degrees C or lower. Bird presence at culvert entrances (n = 33) frequently occurred (75.75%) when average, maximum, weekly temperature was greater than 20 degrees C. Average, maximum, weekly temperature at monitoring sites ranged between 13 and 34 degrees C. Ungulate use of bridges The ungulate group included records (n = 522) of two specieswhite-tailed deer Odocoileus virginianus, and wild pig/hog Sus scrofafrom 41 bridge sites. Logistic regression model. The model was significant in one of three tests for predicting use of bridges by ungulates: Wald X 2 : 17.96 (df, p =.0013), Nagelkerkes R 2 (0.2610), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 5.42 (df, p = 0.71). Large sample sizes are preferred for conducting logistic regression analysis; although we had over 500 records of ungulates crossing under bridges, only 13 records were documented for nonuse of bridges, likely accounting for the poor scores for the goodness-of-fit and R 2 tests. Accuracy of the model is questionable. Only 4 of 21 factors were found significant (p < 0.04) by the model in predicting use of bridges by ungulates. Percentage of movement under bridges correctly predicted by the model was 79.6%.

PAGE 180

164 Response to structural characteristics. Bridge height was the only significant structural factor identified in logistic regression analysis. Of 509 records of ungulates using bridges, most occurred (84.87%) when bridge height was between 2.1 and 6.1 m. Presence but not passage (n = 13) occurred mainly (76.92%) with bridge heights of 3.1 to 3.7 m. Bridge height ranged from 1.5 9.2 m. Response to ecological characteristics. Three significant factors were identified from logistic regression analysis: habitat diversity at 30 and 210 m 2 scales and frequency of presence of domestic predators. Ungulates recorded (n = 509) using bridges were greatest (67.39%) when 3 land-cover types were present at the 30 m 2 cell resolution. Bridges with 2 land-cover types had the second highest number of ungulate use (23.18%). Presence of ungulates (n = 13) near bridge entrances was recorded only when 2 3 land-cover types were present. Habitat diversity ranged from 1 to 4 land-cover types present at the 30 m 2 scale. Most movement under bridges by ungulates (84.03%) occurred at sites where 4 or less land-cover groups were present at the 210 m 2 level. Presence of ungulates (n = 13) near bridge entrances was recorded when 3 6 land-cover types were present. Habitat diversity ranged from 1 to 7 distinct land-cover groups. When presence of domestic predators increased at monitored sites, use by ungulates decreased. Bridge use (90.17%, n = 459 of 509) coincided with 6 or less records of domestic predators. When 2 or less domestic predators were recorded, predator use and presence was 73.08% and 84.61%, respectively. Records of domestic predators ranged from 0 104 records at sites monitored.

PAGE 181

165 Ungulate use of culverts Two species of ungulates (n = 43)white-tailed deer Odocoileus virginianus, and wild pig/hog Sus scrofawere recorded at 19 culvert sites. Due to small sample size, logistic regression analysis could not be performed for the ungulate-culvert group. Even so, thresholds of culvert use by ungulates are included in the discussion. Herpetofaunal use of bridges Herptile presence at 28 bridge sites included records (n = 1,989) of 15 speciessouthern leopard frog Rana utricularia, southern toad Bufo terrestris, green tree frog Hyla cinerea, Cuban tree frog Osteopilus septentrionalis, six-lined racerunner Cnemidophorus sexlineatus, anole Anolis spp., fence lizard Sceloporus undulatus, cottonmouth Agkistrodon piscivorus, timber rattlesnake Crotalus horridus, southern black racer Coluber constrictor, yellow rat snake Elaphe obsoleta, alligator snapping turtle Macroclemys temminckii, snapping turtle Chelydra serpentina, Florida cooter Pseudemys floridana and Florida box turtle Terrapene carolina bauri; and four orders/subordersfrog/toad Anura, lizard Squamata: Lacertilia, snake Squamata: Serpentes, and turtle/tortoise Testudines. Logistic regression model. The model was highly significant for predicting use of bridges by herpetofauna: Wald X 2 : 210.92 (df, p <.0001), Nagelkerkes R 2 (0.7472), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 30.62 (df, p < 0.0001). Ten of 21 factors were found significant (p < 0.03) by the model in predicting use of bridges by herpetofauna. Percentage of movement under bridges correctly predicted by the model was 92.9%.

PAGE 182

166 Response to structural characteristics. Five structural variables were found significant in the logistic regression: gap width, AADT, bridge width, bridge length, and bridge height. Of 848 records of herpetofauna using bridges, most occurred (82.54%) when gap width was 7.9 to 8.5 m. Records of presence but not passage (94.66%, n = 1,080 of 1,141) were also associated with gap widths of 7.9 to 8.5 m. Passage rate at these gap widths was 64.8%. Gap width (distance from adjacent habitat to bridge entrance) ranged from 4.9 16.5 m. When AADT was greater than 6,400, passage under bridges by herpetofauna (n = 848) occurred only 11.09% during the period monitored. Most passage occurrences by herpetofauna (39.27%, and 34.67%, respectively) were recorded when AADT was at 2 separate levels, 250 and 6,400. Presence recorded near bridges, but not crossing from one side to the other, was primarily recorded (88.96%, 1,015 of 1,141) when AADT was 6,400. At 6,400 AADT, passage rate was only 22.46%. Range of AADT on roads where herptiles were recorded was 250 to 24,000. Passage under bridges by herptiles occurred largely (82.32%) when bridge width was less than 40 m. Presence but not passage (n = 1,141) was associated chiefly (89%) with 38 m bridge width. Bridge width of sites monitored ranged from 19.2 152.5 m. Of 848 records of herpetofauna using bridges, the majority occurred (82.45%) when bridge length (distance that organisms had to travel to pass through the structure) was 8.5 m or less. Presence but not passage (n = 1,141) was also associated predominantly (94.66%) with bridge lengths less than 8.5 m. Bridge lengths at sites used by herptiles ranged from 7.3 38.4 m. Of 848 records of herpetofauna using bridges, most occurred when bridge height was between two size ranges, 1.5 to 1.8 m (48.94%) and 3.1 to 3.7 m

PAGE 183

167 (47.52%). Presence but not passage (n = 1,141) was associated primarily (98.25%) with bridge heights of 3.1 to 3.7 m. In this case, bridge height ranged from 1.2 7.3 m. Response to environmental characteristics. Four significant environmental parameters were reported in the logistic regression model and include seasonality, minimum and maximum temperature, and rainfall. Passage under bridges (n = 848) and presence at (but not passage under) bridges (n = 1,141) by herpetofauna occurred 77.6% and 86.59% of the time during spring and summer months. The highest passage rate was recorded in fall (154 of 204, 75.49%), and the highest number of passages occurred in spring (383 of 848, 45.17%). Of 840 records of herpetofauna passing under bridges most occurred (90.82%) when average, minimum, weekly temperature was 10 degrees C or higher. Presence at (but not passage under) bridges showed the same trend (93.86% occurred at average, minimum, weekly temperatures of 10 degrees C and above). Average, minimum, weekly temperature at monitoring sites ranged between 1 and 24 degrees C. Records (n = 840) of herpetofauna passing under bridges occurred most often (89.17%) when average, maximum, weekly temperature was 25 degrees C or higher. When average, maximum, weekly temperature was from 26 to 28 degrees C, 45.59% of movement under bridges was recorded. Presence at (but not passage under) bridges (n = 1,141) were similar, as 78.43% occurred when average, maximum, weekly temperature was 25 degrees C or higher. Average, maximum, weekly temperature at monitoring sites ranged between 15 and 34 degrees C. Of 843 records of herpetofauna using bridges, nearly all occurred (90.75%) when rainfall was 5 ml or less; and presence (but not passage) was also recorded principally

PAGE 184

168 (85.19%, n = 972 of 1,141) when rainfall was 5 ml or less. Average weekly rainfall at monitoring sites ranged between 0 and 14 ml. Response to ecological characteristics. Three important response variables were identified from logistic regression analysis. These include right-of-way vegetation height, primary habitat at 30 m 2 resolution, and human presence. Herpetofaunal movement under bridges (n = 731) was recorded most frequently (77.98%) when vegetation height was between 31 and 50 cm. Right-of-way vegetation height at monitoring sites ranged from 5 to 122 cm. Bridges where herptiles were recorded occurred in only two 30 m 2 scale primary habitat types, 4 (wetlands) and 7 (hardwood forests). Higher passage rates for herpetofauna were recorded at bridges in habitat type 4 (86.54%) than type 7 (23.58%). Use of bridges (n = 848) occurred 52.47% of the time when 5 or less humans were recorded. To the contrary, another peak in culvert use (43.04%) occurred when 73 or more humans were recorded. Presence of humans ranged from 0 110 records at sites monitored. Similar to the latter, 94.31% (n = 1,076 of 1,141) of records of herptile presence at bridge entrances occurred when 73 or more humans were recorded over the course of the study. Herpetofaunal use of culverts Presence of herptiles at 106 culvert sites included records (n = 1,489) of twelve speciessouthern leopard frog Rana utricularia, green tree frog Hyla cinerea, six-lined racerunner Cnemidophorus sexlineatus, five-lined skink Eumeces spp., anole Anolis spp., fence lizard Sceloporus undulatus, cottonmouth Agkistrodon piscivorus, southern black racer Coluber constrictor, eastern garter snake Thamnophis sirtalis, striped mud turtle

PAGE 185

169 Kinosternon baurii, snapping turtle Chelydra serpentina, and alligator snapping turtle Macroclemys temminckii; and four orders/subordersfrog/toad Anura, lizard Squamata: Lacertilia, snake Squamata: Serpentes, and turtle/tortoise Testudines. Logistic regression model. The model was highly significant for predicting use of culverts by herpetofauna: Wald X 2 : 330.78 (df, p <.0001), Nagelkerkes R 2 (0.7730), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 49.68 (df, p < 0.0001). Twelve of 21 factors were found significant (p < 0.05) by the model in predicting use of culverts by herpetofauna. Percentage of movement through culverts correctly predicted by the model was 94.9%. Response to structural characteristics. Six structural factors were found significant by the model of logistic regression: gap width, AADT, culvert width, height, and number of cells, and distance to next nearest structure. Of 580 records of herpetofauna using culverts, most occurred (80.84%) when gap width was 3.7 to 7.3 m. Presence but not passage was associated principally (97.8%, n = 889 of 909) with gap widths less than 9 m. Gap width ranged from 0 39.3 m. When AADT was 8,400 or greater, passage through culverts by herpetofauna (n = 580) occurred only 7.91% during the period monitored. Most passage occurrences by herpetofauna were recorded when AADT was at 4 separate levels: 250, 3,900, 6,500 and 8,100 (16.72%, 14.14%, 13.10% and 16.72%, respectively). Presence near culverts, but not crossing from one side to the other, was most frequently recorded when AADT was 3900 (21.35%) and 7,700 (25.06%). Range of AADT on roads at sites monitored was 250 to 42,500.

PAGE 186

170 Herpetofauna were recorded using culverts most frequent (84.15%) when culvert width was 1.5 m or wider. Presence but not passage (n = 909) was associated largely (95.34%) with culvert widths less than 1 m. Culvert width ranged from 0.3 3.7 m. Of 580 records of herptiles using culverts, most occurred (95.34%) when culvert height ranged between 0.6 and 1.5 m. Presence but not passage (n = 909) was associated predominantly (83.71%) with culvert heights less than 1 m. Culvert height ranged from 0.3 3.4 m. Passage rate (72%) through culverts was greatest when multiples of 2 3 cells were present; passage rate at sites with 1 or 4 cells was less than 25%. Presence at (but not passage through) culverts was recorded mostly (82.84%) with single cell culverts. Records of passage through culverts (n = 580) by herpetofauna coincided most often (72.07%) with distances between structures of 875 to 1,225 m. Range of distance between structures was 125 to 3,150 m. Response to environmental characteristics. Three environmental factors of significance to herptiles using culverts include seasonality, and minimum and maximum temperature. Recorded passage through culverts (n = 580) was high during spring (43.28%) and summer (41.55%) months. Presence at (but not passage through) culverts (n = 909) was recorded mostly (73.82%) in summer months. Only ten records of herpetofauna were recorded in the winter months. Regarding herpetofauna passing through culverts (n = 578) or present at culvert entrances (n = 908), most occurred (97.23% and 99.45%, respectively) when average, minimum, weekly temperature was 10 degrees C or higher. Average, minimum, weekly temperature at monitoring sites ranged between 0 and 24 degrees C. Of herpetofauna

PAGE 187

171 passing through culverts (n = 578) and present at culvert entrances (n = 908), nearly all occurred (94.29% and 95.71%, respectively) when average, maximum, weekly temperature was 26 degrees C or higher. Average, maximum, weekly temperature at monitoring sites ranged between 15 and 35 degrees C. Response to ecological characteristics. Three different ecological parameters were significant in logistic regression analysis. These include primary habitat at 210 m 2 scale, and right-of-way vegetation type and height. Passage through culverts (n = 580) by herpetofauna was the highest for 210 m 2 scale land-cover classes 3 (pinelands, 18.28%), 4 (wetlands, 14.31%), 6 (shrub and brushlands, 37.93%), and 7 (hardwood forests, 17.93%). Presence at (but not passage through) culverts (n = 909) was greatest for land-cover classes 2 (xeric lands, 29.70%), 3 (18.81%), 4 (16.94%) and 6 (15.07%). Passage rates were greatest for classes 6 (220 of 357, 61.62%) and 7 (104 of 124, 83.37%). Movement through culverts (n = 580) occurred 80.69% of the time when herbaceous groundcover (either alone or combined with barren areas or shrubs) was present. Presence at but not movement through culverts (n = 909) was also significant (91.08%), still showing preference for some level of herbaceous groundcover presence. Herpetofaunal movement through culverts (n = 580) was recorded most frequently (86.93%) when vegetation height was greater than 30 cm. Right-of-way vegetation height at monitoring sites ranged from 0 to 183 cm.

PAGE 188

172 Small mammal use of bridges The small mammal group included records (n = 5,142) of three speciesrabbit Sylvilagus spp., gray squirrel Sciuris carolinensis, and eastern mole Scalopus aquaticus; and members of one familyCricetidae (mice, rats, voles) from 39 bridge sites. Logistic regression model. The model was highly significant for predicting use of bridges by small mammals: Wald X 2 : 445.43 (df, p <.0001), Nagelkerkes R 2 (0.8126), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 43.79 (df, p < 0.0001). Twelve of 21 factors were found significant (p < 0.05) by the model in predicting use of bridges by small mammals. Percentage of movement under bridges correctly predicted by the model was 96.8%. Response to structural characteristics. Six structural variables were identified in the logistic regression model that includes: gap width, AADT, bridge width, length and height, and distance to next nearest structure. Of 618 records of small mammals using bridges, nearly all occurred (90.3%) when gap width was 8.5 m or less; and presence but not passage was also associated predominantly (92.27%, n = 4,174 of 4,524) with gap widths of 8.5 m or less. Gap width ranged from 3.1 16.5 m. When AADT was greater than 6,400, passage under bridges by small mammals occurred only 6.31% (39 of 618) during the period monitored. Most passage occurrences by small mammals were recorded (51.78%, 320 of 618) when AADT was only 250. Presence recorded near bridges, but not crossing from one side to the other, was significantly reduced (12.45%, 563 of 4,255) when AADT was higher than 9,900. Range of AADT on roads at sites monitored was 250 to 42,054.

PAGE 189

173 Of 618 records of small mammals using bridges, most occurred (88.02%) when bridge width was less than 50 m. Presence but not passage (n = 4,524) was associated principally (89%) with bridge widths of 60 m or less. Bridge width of sites monitored ranged from 19.2 200 m. Small mammals using bridges were recorded primarily (89.33%) when bridge length was less than 14 m. Almost 70% of bridge use occurred when length was 8.2 m or less. Presence but not passage (n = 4,524) was also associated predominantly (92.29%) with bridge lengths less than 14 m. Bridge length ranged from 7.3 41.5 m. Of 618 records of small mammals using bridges, most occurred when bridge height was between two size ranges, 1.5 to 2.4 m (64.56%) and 3.1 to 3.7 m (33.01%). Presence but not passage (n = 4,524) was also associated predominantly with bridge heights of 1.5 to 2.4 m (19.12%) and 3.1 to 3.7 m (80.59%). Bridge height ranged from 3.1 7.6 m. When distance between structures was greater than 1,000 m, passage under bridges by small mammals occurred only 25.56% (158 of 618) during the period monitored. Most passage occurrences by small mammals (55.97%, 346 of 618) were recorded when distance was less than 700 m. Range of distance between structures was 130 to 3,450 m. Response to environmental characteristics. Three environmental parameters of significance include seasonality, and minimum and maximum temperature. Passage under bridges (n = 618) occurred 86.25% of the time during spring and summer months. Spring months exhibited the highest passage rate, where 375 of 1,231 (30.46%) that approached bridge entrances passed through to the other side. Passage upon approach to

PAGE 190

174 bridge entrances for other seasons was less than 13%. Presence at (but not passage under) bridges (n = 4,524) showed no significant seasonal differences. Of 618 records of small mammals passing under bridges most occurred (95.31%) when average, minimum, weekly temperature was 10 degrees C or higher. Presence at (but not passage under) bridges showed no significant difference regarding low temperature. Average, minimum, weekly temperature at monitoring sites ranged between -1 and 24 degrees C. Small mammal passage under bridges, occurred most often (94.5%) when average, maximum, weekly temperature was 25 degrees C or higher. When average, maximum, weekly temperature was from 26 to 28 degrees C, 62.31% of movement under the bridge was recorded. Presence at (but not passage under) bridges showed no significant difference regarding high temperature. Average, maximum, weekly temperature at monitoring sites ranged between 12 and 34 degrees C. Response to ecological characteristics. Four significant factors were found from logistic regression analysis: right-of-way vegetation type and height, and frequency of disturbance from humans and domestic predators. Movement under bridges (n = 618) occurred 95.63% of the time when herbaceous groundcover (either alone or combined with barren areas or shrubs) was present. Presence at (but not movement under) bridges (n = 4,524) was less significant (75.55%); but still showed preference for some level of herbaceous groundcover presence. Small mammal movement under bridges (n = 585) was recorded most frequently (88.54%) when vegetation height was between 30 and 50 cm. Right-of-way vegetation height at monitoring sites ranged from 0 to 122 cm. When presence of humans increased at monitored sites, use by small mammals decreased. Use of culverts (n = 618) occurred 55.18% of the time when 1 or less humans

PAGE 191

175 were recorded; 68.61% of the time when 5 or less humans were recorded; and 83.66% of the time when 27 or less humans were recorded. Presence of humans ranged from 0 110 records at sites monitored. A reverse trend existed with presence at (but no passage under) bridges, where 72.86% (n = 3,296 of 4,524) occurred when 16 or more humans were recorded over the course of the study. When presence of domestic predators increased at monitored sites, use by small mammals decreased. Use (88.9%) and occurrence (82.28%) coincided with 3 or less records of domestic predators. Presence of domestic predators ranged from 0 104 records at sites monitored. Small mammal use of culverts The small mammal group that used culverts included records (n = 10,656) of four specieseastern cottontail Sylvilagus floridanus, rabbit Sylvilagus spp., Florida water rat Neofiber alleni, and eastern mole Scalopus aquaticus; and members of one familyCricetidae (mice, rats, voles) from 108 monitoring sites. Logistic regression model. The model was highly significant for 2 of 3 tests in predicting use of culverts by small mammals: Wald X 2 : 480.74 (df, p <.0001), Nagelkerkes R 2 (0.3914), and Hosmer and Lemeshow Goodness-of-Fit Test, X 2 : 35.95 (df, p < 0.0001). Note that R 2 is less significant for small mammals using culverts than in the previous model for small mammals using bridges. Fifteen of 21 factors were found significant (p < 0.05) by the model in predicting use of culverts by small mammals. Percentage of movement through culverts correctly predicted by the model was 86.6%. Response to structural characteristics. The four explanatory variables selected by the logistic regression model were AADT, number of culverts cells, culvert width and length. Records of passage through culverts (n = 270) by small mammals occurred

PAGE 192

176 primarily (87.05%) at sites where AADT was less than 8,000 vehicles-per-day; and most occurrences without passage through the culvert (69.63%, n = 7,232 of 10,386) were recorded at sites with AADTs of 5,800 to 9,900. Annual-average-daily traffic for sites where small mammals were recorded ranged from 250 to 29,500. Passage rate (88.89%) through culverts was greatest when multiples of four cells were present; sites with 1 to 3 cells were negligible (approximately 1 3%). Presence at (but not passage through) culverts was recorded mostly (50.33%) with single cell culverts. Of 270 records of small mammals using culverts, most occurred (90.73%) when culvert width was from 0.6 to 3.1 m. Presence but not passage (n = 10,386) was associated predominantly with culvert widths of 0.5 to 1.2 m and 2.4 m (50.81% and 32.34%, respectively). Culvert width ranged from 0.3 4.3 m. Culvert use by small mammals was primarily recorded (84.8%) when culvert length was 22 m or less. Presence near culvert entrances (n = 10,386) was largely connected (88.4%) with culvert lengths of 14 to 22 m. Culvert length ranged from 9.2 49.4 m. Response to environmental characteristics. Four significant environmental factors include seasonality, minimum and maximum temperature, and rainfall. Passage through culverts (n = 270) occurred 73.7% of the time during spring and summer months. Presence at (but not passage through) culverts (n = 10,386) was recorded mostly (64.46%) in fall and spring months. Regarding small mammals passing through culverts (n = 270) or present at culvert entrances (n = 10,386), most occurred (90.74% and 90.94%, respectively) when average, minimum, weekly temperature was 5 degrees C or higher. Average, minimum, weekly temperature at monitoring sites ranged between -1 and 24 degrees C. Of small mammals

PAGE 193

177 passing through culverts (n = 270) and present at culvert entrances (n = 10,386), most occurred (76.29% and 74.38%, respectively) when average, maximum, weekly temperature was 25 degrees C or higher. When average, maximum, weekly temperature varied from 26 to 32 degrees C, 62.95% of movement through culverts was recorded. Average, maximum, weekly temperature at monitoring sites ranged between 12 and 35 degrees C. Of 270 records of small mammals using culverts, most occurred (94.82%) when rainfall was 6 ml or less; and presence but not passage was also recorded predominantly (93.88%) when rainfall was 6 ml or less. Average weekly rainfall at monitoring sites ranged between 0 and 22 ml. Response to ecological characteristics. Seven different ecological variables were significant by logistic regression analysis. These include primary habitat at 30 and 210 m 2 scales, right-of-way vegetation type and height, habitat diversity at 30 and 210 m 2 scales, and frequency of disturbance from human presence and domestic predators. Passage through culverts (n = 270) by small mammals was the highest for 30 m 2 scale land-cover classes 2 (xeric lands, 14.81%), 3 (pinelands, 25.56%), 4 (wetlands, 18.52%), and 7 (hardwood forests, 38.15%). Presence at (but not passage through) culverts (n = 10,386) was also greatest for land-cover classes 2 (17.56%), 3 (21.54%), 4 (27.68%), and 7 (27.65%). Passage through culverts (n = 270) by small mammals was the highest for 210 m 2 scale land-cover classes 3 (22.59%), 4 (17.04%), 5 (grasslands and agriculture, 14.44%), and 6 (shrub and brushlands, 24.81%). Presence at (but not passage through) culverts (n = 10,386) was greatest for land-cover classes 3 (22.33%), 4 (21.26%), and 6 (22.67%).

PAGE 194

178 Movement through culverts (n = 270) occurred 97.41% of the time when herbaceous groundcover (either alone or combined with barren areas or shrubs) was present. Presence at but no movement through culverts (n = 10,386) was also significant (91.28%); still showing preference for some level of herbaceous groundcover presence. Small mammal movement through culverts (n = 260) was recorded most frequently (73.44%) when vegetation height was between 20 and 50 cm. Right-of-way vegetation height at monitoring sites ranged from 0 to 208 cm. Small mammal use of culverts was highest (65.56%) when habitat diversity comprised 3 different land-covers at the 30 m 2 cell resolution. When 2 land-cover types were present, passage through culverts was recorded 75 times (27.78%). Similar percentages occurred (3.79%, 2.42%) for those individuals that were present at culvert entrances only. Insufficient records occurred at sites where the maximum 5-land-cover types existed. Small mammals recorded using culverts was highest (40%) with habitat diversity comprised of 3 different land-covers at the 210 m 2 cell resolution. When 2 land-cover types were present, passage through culverts was recorded 75 times (27.78%). Similar percentages occurred (3.29%, 2.33%) for those individuals that were present at culvert entrances only. Insufficient records occurred at sites where the maximum 6 land-cover types existed. When presence of humans increased at monitored sites, use by small mammals decreased. Use of culverts (n = 270) occurred 90.37% of the time when no humans were recorded. Similarly, presence at (but no passage through) the culvert was greatest (78.87%, n = 8,191 of 10,386) when no human presence was recorded. Presence of humans ranged from 0 4 records at sites monitored where small mammals were

PAGE 195

179 recorded. Small mammals preferred sites where domestic predators were absent, whether present at or passing through culverts. Most occurred when domestic predator records were zero (63.97% and 59.26%, respectively) or one (15.97% and 27.04%, respectively). The range of records of domestic predators, where small mammals occurred, was only 0 4 at sites monitored. Alligator and aquatic mammal use of culverts and bridges A consolidated group of aquatic dependents included three speciesalligator Alligator mississippiensis (n = 25), river otter Lutra canadensis (n = 31), and beaver Castor canadensis (n = 5) from 15 culvert and 13 bridge sites. Even when combined into one group, sample size was too small to perform logistic regression. Discussion Several studies have demonstrated wildlife movement through culverts and bridges under roadways. Most case studies have shown use by specific species of specific structures, e.g. Florida pantherI-75 bridges, Florida black bear.4 x 7.3 m culvert, bobcat and raccoon.8 x 1.8 m (Hewitt et al. 1998, Norman et al. 1998, and Boarman et al. 1996). Other studies involved use of wildlife ecopassages designed for large targeted species (Clevenger and Waltho 2000, Foster and Humphrey 1996, and Roof et al. 1996). Yet smaller concrete culverts and tunnels originally designed for drainage under roadways have also been used as wildlife passages by a wide variety of small to medium size mammals (Rodriguez et al. 1996, Yanes et al. 1995, and Hunt et al. 1987) and many species of amphibians (Brehm 1989, Dexel 1989, and Norden 1990). All these studies agree that culverts are useful mitigation measures for movement of

PAGE 196

180 animals under roadways; however, none have provided comprehensive evaluations of effectiveness according to a wide distribution of structure sizes and contexts. The objective of this study was broader; specifically, it was set up to determine design standards for drainage culverts and bridges (to enhance and improve use by a wide variety of wildlife). Yet additionally it addressed larger issues, such as improving overall landscape connectivity of important large-scale ecological linkages. This structure classification system would serve as a reference for transportation agencies when programming mitigation measures for wildlife mortality on highways (based on type of species present and landscape context). Comparison of Structure Use Among Faunal Groups Six distinct faunal groups were used for analysis. Similarities and differences regarding use of crossing structures are evident among the groups. Table 4-9 summarizes findings of common parameters for all structures by each faunal group. Three thresholds for each parameter represent liberal to conservative measures of passage success by each faunal group. Thresholds for contextual parameters To sustain 90% of crossings made by each faunal group, only 3.7 m of open right-of-way separated the structure entrances from the adjacent habitat (Table 4-9). As the level of sustainability is decreased, differences occur among groups for maximum gap distance. At 75%, the gap width increases to 5.1 and 7.7 m for small mammals and ungulates, respectively. Gap width was statistically significant for the smaller species recorded using crossing structures: meso-mammals, small mammals, herpetofauna, and birds. A distance of 3.7 to 5.1 m from adjacent habitat to structure entrances is recommended to promote high levels of movement for all species.

PAGE 197

181 The gap width can be mitigated in part by right-of-way vegetation type and height. All groups preferred presence of herbaceous vegetation, and in some cases addition of shrubs. Right-of-way vegetation height generally consisted of three minimum thresholds among the faunal groups: 5 cm for birds and ungulates, 10 cm for carnivores, and 20 cm for meso-mammals, herptiles, and small mammals (Table 4-9). Maximum height preferences varied from 50 to 92 cm. Right-of-way vegetation height was a significant factor only for small mammals and herpetofauna. This would coincide with the need for cover from larger mammalian predators and birds of prey. Given specific exceptions, at least 20 cm groundcover should be maintained at all crossing sites; with addition of larger shrubs strategically placed to provide cover for larger, sensitive species, such as white-tailed deer. An example of one exception would include xeric habitats, where open sandy areas are preferred by certain species. The best recommendation regarding type of right-of-way vegetation would be the use of plant species that are the same as those in adjacent habitat, with slightly greater cover near entrances to crossing structures (to provide security for prey species). Distance between structures was a significant factor regarding use by meso-mammals, carnivores, herpetofauna, and small mammals. A maximum distance of 200 to 250 m between crossing structures was necessary to sustain 90% passage for all groups, except carnivores. The maximum recommended distance is 325 m within core conservation areas and habitat corridors, corresponding to at least 75% use by small mammals. This distance is similar to that recommended by Clevenger et al. (2001).

PAGE 198

182 Table 4-9. Levels of use by faunal groups for common parameters of all study sites Threshholds: 90% (upper), 75% (middle), 60% (lower) Mesomammals (n = 11,701) Carnivores (n = 611) Birds (n = 258) Ungulates (n = 573) Herpetofauna (n = 1,428) Small mammals (n = 888) gap width (distance 3.7 3.7 3.7 5.8 3.7 3.7 to habitat), m 5.3 5.8 6.8 7.7 6.4 5.1 5.8 7 7.9 8.1 7.8 7.9 AADT, annual avg. 250 3,200 250 3,400 250 250 vehicles / day 3,000 6,400 3,000 6,100 250 250 5,000 6,500 3,200 6,709 3,900 1,900 distance between 250 585 250 200 260 250 structures, m 560 600 600 600 625 325 775 875 775 670 875 585 human presence, 4 3 3 3 3 1 no. / year. 46 7 10 7 17 8 69 30 46 17 46 46 domestic predators, 1 1 1 1 1 1 no. / year 1 1 1 1 1 1 20 3 1 4 1 20 Preferences row vegetation type herbaceous herbaceous herb shrub herb shrubherb shrub herb shrub row vegetation height, cm 20 70, 82% 10 75, 79% 5 60, 85% 5 91, 78% 20 92, 79% 20 50, 85% soil moisture and precipitation dry moistdry moistdry dry moist dry dry habitat diversity 30m2 3 3 3 3 3 2 3 primary habitat 30m2 4, 8 4, 7, 8 4, 8 4, 7, 8 4, 8 4, 8 habitat diversity 210m2 2 4 1 4 3 4 3 4 2 4 3 4 primary habitat 210m2 3, 4, 7 3, 4, 6 3, 4, 7 3, 6, 7 3, 4, 7 4, 7 season fall, summerwinter spring, summer spring Notes: Preferences are based on a minimum of 70% structure use for each faunal group. Table 4-3 provides habitat class descriptions. Crossing structures located in areas with three land-cover types at the 30 and 210 m2 scales were most frequently used by all groups. All groups used structures most often located in wetlands (4) or hardwood forests (8) at the 30 m2 scale. Slight differences were apparent at the 210 m2 scale, with crossing structures in pinelands (3), wetlands (4), and shrub and brushlands (7) most commonly used by five of six faunal groups. Habitat type and diversity (30 m2) were significant factors in logistic regression for

PAGE 199

183 meso-mammals, small mammals, ungulates, and herptiles. This likely corresponds to use of road verges for foraging and the crossing structures as shelter. Surrounding habitat type and diversity (210 m2) were significant factors for meso-mammals, ungulates, herptiles, and small mammals. Management and mitigation for the species most frequently encountered in this study should generally focus on improving habitat diversity along road right-of-ways and adjacent land-cover classes 4 (wetlands) and 8 (hardwood forests). Specifically, issues regarding habitat should be addressed on a site-by-site basis. Thresholds for disturbance parameters Three factors can be considered measures of tolerance by faunal groups to disturbance, specifically AADT (traffic level), human presence, and presence of domestic predators. Meso-mammals, birds, small mammals, and herptiles made 90% of crossings when traffic levels were 250 or fewer vehicles-per-day (Table 4-9). Carnivores and ungulates were more tolerant, with 90% of structure use occurring when over 3,000 vehicles were present per day. While tolerance of traffic increased for 75% structure use by meso-mammals and birds (AADT = 3,000), and carnivores and ungulates (AADT > 6,000); small mammals and herptiles were more sensitive, as 250 vehicles daily was still the maximum traffic level that could sustain measured use of culverts of 75%. Herptiles appear most sensitive to traffic, which is most likely associated with traffic noise; other groups appear more tolerant. Annual-average-daily traffic was statistically significant for all groups, except birds and ungulates. Traffic level was an important factor in culvert avoidance by wildlife in other studies: Banff National Park, Alberta, Canada (Clevenger et al. 2001), and Denali National Park, Alaska (Yost and Wright 2001). Sites where

PAGE 200

184 herptiles and small mammals are present require more stringent restrictions or mitigation for traffic and traffic-related noise. Based on the results presented here, none of the groups could sustain more than 60% use of structures when daily traffic levels were over 6,709. Presence of humans was measured simply as the number of people (or signs of people) recorded over the 19 months of monitoring; therefore, representing the relative human impact that recorded wildlife would tolerate (Table 4-9; values in table were converted to number per year). Greatest number of humans recorded at any single site was 104 over 19 months (1 every 5.5 days). This factor was found significant in regression analysis for meso-mammals, herptiles, and small mammals; however, small mammals, ungulates, and carnivores appeared most sensitive to frequency of human presence (probably reacting to persistence of human scent). Meso-mammals were the least inhibited (75% structure use with 73 total records of human presence, 1 every 8 days). Ninety percent of structure use by all groups, except small mammals, occurred when only 4 6 humans were encountered (1 every 95 days; excludes track recorders). This finding echoes that of Clevenger and Waltho (2000), who found that wolves avoided culverts near areas of human activity. Grizzly bears also displayed avoidance of humans associated with logging activities (McClellan and Shackleton 1988). Certainly, the data here indicate that to maximize use by most wildlife, human presence needs to be restricted at crossing sites. All groups can be considered sensitive to activity by domestic predators near crossing structures (Table 4-9; values in table were converted to number per year). Presence of domestic predators was as high as 110 over 19 months (1 every 5 days).

PAGE 201

185 Ninety percent use of crossing structures by each group occurred when 1 or less domestic predators were present over 19 months. Carnivores, birds, ungulates, and herptiles only used crossing structures at a 60% rate when 2 6 domestic predators were encountered during the course of the study. Fencing or other devices should be used to restrict access to crossing sites from the road by domestic predators. Thresholds for structural parameters All groups showed a preference for dry soil conditions within structures (Table 4-9), which was statistically significant for meso-mammals, carnivores, herpetofauna, and small mammals; however, caution should be used regarding herpetofauna because some bias in sampling could affect this measure. During rainy periods, tracks were either hard to read or lost from washouts of track-beds, and thunderstorms restricted activities by field technicians. Though commonly found lizard-species seem to prefer dry conditions; intuitively, the movement of amphibian species should increase with wetter conditions. Although openness was not included in the logistic regression, it is still useful as a reference for general size requirements for each faunal group. Ninety percent of use by meso-mammals, carnivores, ungulates, and small mammals occurred in structures with openness index values of approximately 0.40 (Table 4-10). Birds preferred larger openness values (0.86) and herpetofauna smaller openness values (0.28) for 90% of culvert use recorded. A breakdown of the openness index into its components may be more informative.

PAGE 202

186 Table 4-10. Levels of use by faunal groups for structural parameters of culverts Threshholds: 90% (upper), 75% (middle), 60% (lower) Mesomammals (n = 5,515) Carnivores (n = 258) Birds (n = 81) Ungulates (n = 32) Herpetofauna (n = 580) Small mammals (n = 270) 0.43 0.41 0.86 0.41 0.28 0.42 openness index value 0.28 0.23 0.43 0.25 0.28 0.18 (w x h / l) 0.18 0.23 0.23 0.25 0.23 0.08 3.4 3.1 3.1 3.7 3.4 3.1 width, m 3.1 2.7 3 3 3.1 2.7 2.7 2.4 > 2.4 3 2.7 1.5 11 12.8 11 12.8 11.6 11.6 length, m 14 14.5 12 13.7 14.5 13.7 18 14.5 14.6 19.2 14.5 18.3 1.7 1.8 3.4 3.7 1.5 1.5 height, m 1.5 1.5 1.5 3.7 1.4 1.4 1.4 1.5 1.4 3.4 1.4 1.2 Preferences structure number, # trend trend trend 1 2 trend 1 structure shape rectangularrectangular rectangularrectangularrectangular rectangular structure substrate dirt dirt dirt dirt Response to culvert width was similar for all faunal groups (Table 4-10). All groups exhibited 90% usage when culvert width was 3 m or greater and 60% usage at 2.4 2.7 m, except small mammals (1.5 m). Culvert width was a significant factor in all regression models except for ungulates. The low level of variability among groups of differing body size indicates that minimum width may play a role in passage success, but in a general sense. Thus to maintain high passage rates (75%) for all species (especially carnivores) at least 2.7 m width for new structures is encouraged (for comparison, 3 5 m was recommended for white-tailed deer by Norman et al. 1998). Note that of 63 Florida black bears recorded, only 5 approached culverts, and only three of these actually crossed through from one side of the road to the other. The culverts used by these individuals had minimum openness index values of 0.23 and heights of 1.5 m. Culvert height displayed greater variance with regard to use among faunal groups (Table 4-10). Culvert height was a significant factor for meso-mammals, herpetofauna,

PAGE 203

187 and small mammals. The latter two groups were more abundant using culverts 1.5 m or lower, presumably because larger predators are more hesitant to use them. All but three individuals in the carnivore group that used culverts either were bobcats, coyotes, or foxes. For areas where only smaller prey are abundant, lower heights should be used. For large carnivores and ungulates, 3 m minimum height should be used. Ungulates did not use culverts with any significant frequency (n = 32 over 19 months). Other studies (Garrett and Gordon 2003) indicate that preference for higher openness values may limit usefulness of standard drainage culverts for deer. Foster and Humphrey (1995) found use of 2.1 m high bridges by ungulates and large carnivores; however, these structures had openness values of 0.92 1.12. Culvert use among groups was similar with regard to culvert length (Table 4-10). Frequency of usage was 90% for all groups if culvert length was 11 m or less. Due to road construction standards, length of culvert is difficult to change; it is entirely dependent on the width of the road itself. Openness index can be used as a mechanism to adjust for length. As a road becomes wider, culvert length increases and openness necessarily goes down. To maintain the same openness value, width and or height of the culvert must be increased. Other influential factors associated with culverts include number of culverts (e.g., single and multi-cell units), culvert shape, and substrate. Preferences for all groups were for dirt or soil substrate and rectangular shape. Number of structures was more variable among groups; and statistically significant for only carnivores, herptiles, and small mammals. These three groups preferred single to multi-cell configurations.

PAGE 204

188 Influence of structural factors of bridges on faunal groups is shown in Table 4-11. Differences between results for bridges and culverts reflect the significant differences between the two structure types. Bridges were much less limiting on use by faunal groups. All three dimensions (width, length and height) were significant factors for small mammals and meso-mammals, width and height for birds, width for herptiles, and height for ungulates and carnivores. When comparing bridges to culverts, only bridge height is relatively limiting (because height is similar for the two structure types). Specifically, for locations used by carnivores (especially Florida black bear) and white-tailed deer, a minimum height of 3.5 m is recommended. Predicting use as a result of structure width and length is not as clear. For the groups mentioned above with significant maximum likelihood scores for bridge width, most are likely a result of increased habitat availability. Table 4-11. Levels of use by faunal groups for structural parameters of bridges Threshholds: 90% (upper), 75% (middle), 60% (lower) Mesomammals (n = 6,186) Carnivores (n = 353) Birds (n = 177) Ungulates (n = 541) Herpetofauna (n = 848) Small mammals (n = 618) 17.89 21.35 17.08 25.42 19 17.08 openness index value 14.38 17.08 17.08 16.37 17.08 12.2 (w x h / l) 9.66 15.25 9.66 13.14 17.08 8 82.4 96.1 60.4 152.5 54.9 54.9 width, m 54.9 60.4 41.2 82.4 38.4 40.3 41.2 41.2 38.4 58 38.4 32 7.3 8.2 7.3 8.2 7.3 7.3 length, m 7.3 8.5 7.3 9.2 7.3 7.3 8.2 12 8.2 11.9 8.2 7.3 3.7 3.7 3.7 6.1 3.7 3.7 height, m 3.1 3.7 3.7 4.6 3.7 3.1 3.1 3.1 3.1 3.7 3.1 1.8 Crossing Structures as Primary Habitat Features Two groups, herpetofauna and small mammals included species that frequented culverts as part of their primary habitat rather than as movement corridors. These

PAGE 205

189 included lizards (n = 2,635), primarily six-lined racerunner and five-lined skink; and rodents (n = 15,637), i.e., various mice and rats. Lizards commonly used culvert and bridge faces for sunning areas; and mice used crevices and cracks in the structure for shelter. When controlled for these species, each groups presence is reduced significantly (other herptiles, n = 843, other small mammals, n = 161). Additional analysis that splits these groups into relevant categories is recommended to separate primary habitat needs of local species from species with larger movement requirements (e.g., snakes and turtles). Meso-mammals, especially raccoons used water collection areas at culvert and bridge sites for foraging areas for amphibians or trash from motorists and fishermen. Specific planning for movement under roads by this group is probably unnecessary because they are already quite successful using existing configurations without need for modification. Road-kills and Culvert Avoidance Although frequency of site visits was sufficient for tracks and photographs, twice per week was inadequate to evaluate quantity and types of road-kills occurring on adjacent road segments. Persistence of road-kill depends on factors such as weather, traffic density, frequency of road cleanup or maintenance, and depredation rates by scavengers. Studies of culvert avoidance should include road-kill information; without a complete complement of these data, inferences about avoidance by certain species could not be made. Additionally, surveys of abundance in habitat areas adjacent to monitoring sites may provide evidence of other species present that do not approach structure entrances where track beds were situated. Two studies (Clevenger et al. 2001, and Yanes et al. 1995) performed abundance surveys that provided better information regarding

PAGE 206

190 avoidance, and allowed for comparisons between culvert use frequency and total population size. Effects of Light Penetration and Traffic Noise Certain characteristics such as light and moisture content have been evaluated for amphibians (Krikowski 1989, and Langton et al. 1989). Amount of light available within culverts can help counter tunnel effects. Certain amphibian species would not enter when sufficient light was not present (Krikowski 1989). Light availability within drainage culverts has tremendous variability based on two factors: size of culvert opening that affects quantity of light and depth of penetration; and the length of the culvert, where quantity of light decreases as distance from the opening increases. Another potential vector affecting light availability occurs on divided highways that incorporate median drainage grates allowing light penetration at the center of the culvert. Although this study did not evaluate the influence of light availability on structure use, visibility is of obvious importance to many species that were recorded. Most specifically, openness index values are related to the amount of light penetration. This was inherently reflected in the results, where most species preferred higher openness values. Noise is another influential factor that has been investigated by other researchers (Clevenger et al. 2001, and Reijnen et al. 1995). These studies found significant road avoidance associated with the effects of traffic noise, particularly for small carnivores and birds. A direct relationship between traffic noise and traffic volume would indicate a similar relationship between traffic volume and avoidance by species sensitive to noise. This relationship is depicted in this study by the precipitous drop in use of culverts as traffic levels increase beyond 6,000 vehicles-per-day.

PAGE 207

191 Current Research Benefits and Future Directions Design standards for drainage structures (that promote use by wildlife within certain landscape/habitat contexts) should improve efficiency for retrofitting existing roads, or construction of new roads with appropriate cost-effective ecopassages. Determining the utility of existing structures, within right-of-ways at identified high-priority, highway-greenway interface zones, also provides opportunity for significant cost savings (associated with adapting these sites for optimal connectivity for wildlife). This research should enhance the ability of transportation agencies to effectively address wildlife issues at public hearings; and to quickly implement the appropriate mitigation measures needed at identified high-priority, highway-greenway interface zones. Future research needs include effects of moisture, temperature and light, and efficacy of drainage culverts to facilitate movement by amphibians in the southeast. Similar research needs to be performed on aquatic culverts; to assess connectivity for fish and other aquatic obligates in the many streams in Florida (potentially obstructed by roads). Significant research in this area has been conducted in the northwest (Ruediger 2001, and Carey 1996). Lastly, wildlife movement, in areas where few drainage structures exist (e.g., sandhill and scrub communities), needs to be investigated to assess impacts of road-kills on population levels (i.e., the need for ecopassages to improve habitat connectivity). Potential study sites include Guana River State Park, Goldhead Branch State Park, Marjorie Carr Cross Florida Greenway Ross Prairie State Forest, and parts of Ocala National Forest; where park staff has collected significant road-kill data.

PAGE 208

192 CHAPTER 5 INTEGRATING LANDSCAPE ECOLOGY THEORY WITH TRANSPORTATION PLANNING AND POLICY Theoretical and Conceptual Basis for Infrastructure Ecology In 2002, the Wildlife Conservation Society presented the results of a study (Sanderson et al. 2002) that explored the extent of human influence over the globe. The project was called the human footprint. In the United States, virtually no area has been untouched (Figure 5-1). Large, federally protected lands (Figure 5-2) are the only areas that appear minimally affected. These figures chronologically correspond to clearing of eastern native forests by European settlers in the 1800s and early 1900s for agriculture and urban development (Williams 1990). Figure 5-1. The human footprint. The human footprint analysis conducted by Sanderson et al. (2002) indicates that the eastern half of the coterminous U.S. is strongly influenced by one or more of the following factors: human population density (> 1 person / km2), < 15 km from a road or major river, consists of urban or agricultural land-uses, < 2 km from a settlement or railway, and/or produces enough light to be visible from an orbiting satellite at ni g ht. Map reprinted courtesy of the Wildlife Conservation Society, Bronx, N.Y.

PAGE 209

193 Roads played a significant role during this early period of development, and expanded immensely throughout the 20th century. Paved roads in the U.S. now exceed 6.3 million km (FHWA 2002a; Figure 5-3). This network of infrastructure has received greater attention in the last two decades regarding the extent of degradation to environmental quality and integrity of ecological systems. Conferences and meetings in the United States and Europe have drawn international attention; with the goal of improving communication between policymakers, engineers, and ecologists to develop scientifically based solutions to the ecological impacts of roads. The National Research Council, Transportation Research Board produced two documents (NRC 1997, 2002) meant to generate interest for scientific research into ecological effects of roads. Two major books on transportation impacts have been published: Environmental Impacts of Railways (Carpenter 1994) and Figure 5-2. Federal lands in the coterminous U.S. Location of the majority of remaining large protected areas mirror the results from the analysis for the human footprint in Figure 5-1; and chronologically correspond to clearing of native forests in the east by European settlers in the 1800s and early 1900s for agriculture and urban development (Williams 1990). (Data source: USGS 2000)

PAGE 210

194 Road Ecology (Forman et al. 2003). The following discussion offers a theoretical link between road impact studies and landscape-level ecological analysis. Evolution and Biogeographic Principles A top-down scientific approach to infrastructure ecology begins with a solid foundation in evolution and biogeographic principles. A brief review of this approach illustrates associations between general ecological principles and road impact studies. Three tiers of interrelated hierarchies are shown in Figure 5-4. Horizontally, the three elements represent system processes, patterns and vectors of stability or change. Vertically, the scale of study decreases from top to bottom. On the top level are evolution, biogeography, and biodiversity conservation. While the theory of evolution provides an explanation of changein successive generations of organisms (Allaby 1994, page 148), biogeography explains the pattern and distribution of organisms in Figure 5-3. Proliferation of roads across the United States. Near the beginning of the 21st century, over 6.3 million km of paved roads have been constructed. (Data source: FHWA 2002, and USGS 1996 )

PAGE 211

195 space and time. Though described at large temporal and spatial scales, the patterns at this upper level are determined in large part by the levels below acting on smaller scales, and through interaction among the factors on the same level. For example, evolutionary descent or divergence of certain species (e.g., yellowand redshafted flicker along the frontal range of the Rockies; Cox and Moore 1980) is brought about by biogeographical phenomena such as physical (e.g., geographical barriers), biological (e.g., competitive exclusion) and climatic (e.g., latitudinal zones) differentiation. The understanding of evolution is brought about by ecological studies of natural systems, communities, populations, and species that involve internal processes and/or interactions among species and their environment at appropriate biogeographic scales (e.g., among specific faunal groups or plant communities). Roads can affect Figure 5-4. Hierarchy of theories/concepts for landscape-level studies.

PAGE 212

196 evolutionary-scale phenomena such as multi-generational movement and migration patterns. Examples include fragmentation of once-contiguous habitat areas used by different stages in the life cycle of amphibian populations (Langton 1989a); or obstruction of stream reaches used by migratory fish (Carey 1996). Biodiversity conservation, now considered a discipline of its own, asserts to provide measures of protection for organisms and ecological systems across these large scales. In a world where human habitation and development threaten the future existence of many species and unique natural systems, conservation biology incorporates many principles. Such as adaptive and ecosystem management, aimed at protecting ecological components of entire ecosystems; and using techniques that allow for management changes according to response stimuli that indicate ecosystem health. Figure 5-5 provides an illustration of hierarchical linkages between planning and management applications and their respective ecological scale. Note how connections are formed between these levels. Florida can provide a relevant example to the application of these principles. Figure 5-5. Hierarchy of relevant theory and conservation applications. Current theoretical approaches such as island biogeography and metapopulation ecology not only apply to studies of naturally occurring distributions, but management for human induced conditions as well (e.g., forest fragmentation caused by roads and urban development). Evolution Re g ional Plannin g Island BiogeographyIsolated Protected Areas Metapopulations Dis j unct Tenuous Stochastic Events Extirpation/ ReintroductionNatural Connectivity RestorableConnectivit y Theory and Concept Management or Planning Strategies

PAGE 213

197 Florida is a peninsula characterized by warm latitudes and enriched waters of the Gulf of Mexico and Atlantic Gulf Stream that provide a unique environment, supporting both temperate and subtropical zones (Figure 5-6). A biogeographic, insular phenomenon (caused by the physical dimension or shape of the peninsula) occurs where higher rates of extinction and lower rates of immigration are expected as one proceeds toward the tip of a peninsula (Engstrom 1993). This generally results in a reduction in species richness (Figure 5-7), although Busack and Hedges (1984) found that other local factors also might affect species richness on peninsulas (e.g., latitude, microclimate, and vegetation types). Attention toward human-induced global warming and climate change has pointed to one consequence particularly important in Florida, accelerated sea level rise. A recent study (Williams et al. 1999) conducted on tree islands along the Gulf Coast has Figure 5-6. Life zones of Florida as determined by latitude and climate.Rapid transformationof the transition region to subtropical moist forest is predicted with minor increases in temperature and reductions in precipitation due to human-accelerated global warming (Harris and Cropper 1992). (Map Source: Larry Harris, University of Florida, Gainesville FL )

PAGE 214

198 determined that advancing sea levels have begun to reduce coastal extent of non salt-tolerant species. The result is an inland shift of non salt-tolerant species. The shift in faunal and floral ranges is a natural biogeographic response to climate change. This has been demonstrated through global climate change models and paleobotany studies that show the latitudinal shifts of dominant plant communities (Iverson and Prasad 1998). These range shifts would probably proceed uninhibited on an undeveloped peninsula (despite findings by some studies that project human-induced global warming occurring more rapidly than many plant-animal associations can adapt, resulting in multiple extinctions; e.g., Kirtlands Warbler-jack pine forests, see Brown and Lomolino 1998, and Hobbs and Hopkins 1991). In Florida, habitat fragmentation (caused by expanding development and road systems) threatens to sever the network of habitat Figure 5-7. An example of the peninsula effect. Proceeding from the panhandle to the tip of the peninsula of Florida, land bird diversity, based on breeding bird surveys, changes from 70 to 20 species. (Adapted from Robertson and Kushlan 1974 )

PAGE 215

199 linkages necessary to allow generational transfer of complete biotas inland and northward (Harris and Cropper 1992, and Hobbs and Hopkins 1991; Figure 5-8). Landscape Ecology Landscape ecology provides an approach to ecological research that combines the use of scientific principles from biogeography, geomorphology, community ecology, land-use planning, etc. with GIS, mathematical modeling, and spatial analysis techniques to examine natural phenomena within and across differing temporal and spatial scales. A key feature to this approach is that the role of humans and their impact on ecological systems is considered in the development of scientific solutions to ecological problems and better landscape management practices. Fundamental issues for study include hierarchical arrangement, scale, landscape-level patterns and processes among organisms inland/northward migration paths Figure 5-8. Ocala to Okeefenokee landscape linkage. The arrows represent potential linkages necessary to support species movement associated with global climate chan g e.

PAGE 216

200 and ecosystems. Applications of concepts are entwined with reserve design, fragmentation, corridors, and landscape connectivity. Different approaches to landscape ecology are illustrated by the variation in definitions. One such definition is concerned with the causes and effects of heterogeneity, independent of scale (Pickett et al. 1995); rather than focused at a specific spatial extent, as is the case with Forman (1995, page 13), a mosaic where the mix of local ecosystems or land uses is repeated in similar form over a kilometers-wide area. The larger, landscape approach to the study and protection of ecological systems has gained added acceptance in scientific circles, as well as national-level, land management agencies. A National Research Council committee established conservation priorities for federal agencies, based on the protection of landscape-level ecological processes (NRC 1993). Their approach incorporated provisions for large-scale habitat connectivity and multi-agency coordination for land planning and management. Jerry Franklin (1993, page 202) of the University of Washington, when recalling his experience with the National Forest Service, northern spotted owl, and old growth forests of the northwest, said we cannot even come close to attaining our goal of preserving biodiversitysustainability, if we continue to focus our effortson species. Larger-scale approachesat the levels of ecosystems and landscapesare the only way to conserve the overwhelming massthe millions of speciesof existing biodiversity. A conceptual approach to infrastructure ecology based on landscape ecology principles is presented in Figure 5-9. At the top of the diagram are concepts of biogeographic pattern and the bottom of the diagram presents the empirical data that might be associated with road impact studies. Landscape-level applications are

PAGE 217

201 scientifically based solutions derived from indices (e.g., Diversity = f (topographic relief)) and methods appropriate to evaluate pattern and process from empirical data (applicable to interactions between roads and ecological systems and species). The Florida black bear in the Ocala National Forest provides an example of a landscape approach to addressing road impacts. A significant impact study (Chapter 1, page 18) was conducted regarding the proposed 4-laning of SR 40 that essentially divides the forest in half. Track data of road crossings by black bears were compared with locations of road-kills. Figure 5-10 displays a uniform distribution of crossing sites, as determined by track counts. No preference is apparent with regard to crossing specific sections of the road along the approximate 20 km length of road checked from 1999 to 2001. Therefore an analysis was conducted to evaluate configuration of the road corridor Conce p tual Em p irical Data Landsca p e Ecolo gy A pp lications1. Island biogeography theory 2. Diversity = 1/latitude 3. Diversity = 1/altitude 4. Peninsula effect 5. Diversity = f (topographic relief) Figure 5-9. A theoretical and conceptual framework for infrastructure ecology. 1. Existing right-of-ways 2. Land use/ land cover types 3. Hydrologic/ topographic features 4. Listed species 5. Road-kills

PAGE 218

202 itself (e.g., topographic relief and curvature). Topographic contours were isolated for this section of road; and an elevational profile was generated. As revealed in Figure 5-10, there appears to be a significant connection between slope of the road and curve severity, and collision sites. This would indicate that driver visibility might be the primary factor in location of collisions with black bears. Before concluding this as the primary cause; however, it may be important to examine potential effects between collision locations and speed limit, traffic density, habitat, land use, and distribution and density of black bears. Figure 5-11 provides an examination of speed limit, traffic density, and intensive land-uses on collision locations. Note that speed limit reductions coincide with intensive land-use areas, where few if any collisions have occurred. Traffic density averaged 12,000 vehicles/day within developed 0204060(m) State Road 40 Elevation Profile Curved Sectio n Sloped Section Location : Ocala National Forest Figure 5-10. Landform, road alignment and black-bear road-kill trends. Road-kills occur more frequently where curves and slopes of the road are more severe. (Data sources: FFWCC 2001, and FDO T 2 000 )

PAGE 219

203 areas and 4,800 in undeveloped areas. It would appear that none of these are factors affecting location of collisions. Lastly, a look at habitat use and distribution of black bears (from telemetry data) may reveal some pattern. Figure 5-12 shows that while black bears were located frequently in the area along the section of road where chronic road-kill sites are located; they were also found in equal numbers farther to the west, on the ridge where few road-kills occurred. Thus, from this analysis, the best indicators of chronic road-kill sites are road configuration and topography, which impair driver visibility. While this example was only used to illustrate potential landscape analysis methods, and was descriptive in nature; most of these factors could have been quantified to perform statistical analyses (e.g., slope, extent of curve severity, and location frequencies). Figure 5-11. Highway speeds, land use and black bear road-kill trends. These factors appear to be unrelated to location of chronic road-kill sites. (Data sources: FFWCC 2001, FDOT 2000, and Water Management Districts 1995)

PAGE 220

204 Growth and Roads In a free-enterprise market economy, growth generates prosperity. Although the merits of this economic system or the inherent disparities in wealth associated with it are not discussed, it is important to recognize how this growth may impact natural systems. Particularly, what is relevant is the real estate industry and government-stimulated economic methods. Development of land increases the inherent economic value of said property making it more attractive as real estate and as investment capital, thus attracting more development. This is sometimes referred to as development clustering or mutual proximity, where businesses find it economically beneficial to locate near other like-businesses (Greer and Farrell 1988). Governments conduct public improvement Figure 5-12. Land cover and telemetry locations of radio-collared bears and black bear road-kill trends. Density of radio locations of black bears appear to be unrelated to chronic road-kill locations. ( Data sources: FFWCC 2001, and Water Mana g ement Districts 1995 )

PAGE 221

205 projects to stimulate economic growth in a community; projects that include road building and improvements, and designations of land through zoning or land-use regulations (Chapin and Kaiser 1979). While these measures are meant to increase quality of life and economic prosperity for those that live in a community, when growth is misguided or left unchecked, it can rapidly spiral out of control, resulting in detrimental effects to ecological systems (Stolzenberg 1991). This continuous cycle, where roads and development in turn promote each other, is illustrated in Figure 5-13. What occurs first is the linear expansion of cities, as new businesses are located outward from central business districts (CBDs) along major arterial streets (Greer and Farrell 1988; Figure 5-14). Second, interstitial areas between these major arterial streets are later filled in by residential and additional commercial development. This is described as a radial or axial development pattern (Greer and Farrell 1988). The process continues along these vectors as long as it is economically viable (refer to discussion on Von Thunens model of land utilization, natural zoning, and bid-rent curves; Greer and Farrell 1988, pages 55 62). Figure 5-13. The continual cycle of development and road construction. New Road Construction Increases Intensity and Expansion of Land Development Increase in Development Promotes Construction of New Roads

PAGE 222

206 Negative Ecological Consequences of Roads The negative ecological consequences of roads are examined from a landscape perspective (for more detail, see Chapter 6). They include: direct and secondary road mortality of wildlife; spread of invasive non-native species; toxic substances and soil erosion; disturbance by humans and domestic predators from increased access to otherwise remote areas; and habitat loss, fragmentation, species aversion, genetic isolation, population and species loss. A relation can be depicted for the extent of each of these impacts in space and time (Figure 5-15). The distance into adjacent natural areas affected by these and other ecological effects have been estimated (Forman 1995). Figure 5-14. Radial or axial development. A linear growth pattern that begins adjacent to the CBD and extends out along major arterial roads (Greer and Farrell 1988). Suburban development later takes place within the interstitial buffer areas.

PAGE 223

207 Few studies have determined impacts of road mortality on wildlife population persistence. Certain conditions may allow wildlife to sustain their population levels despite significant road mortality. An example is the Florida Key deer, geographically restricted to Big Pine Key and adjacent islands. Road mortality exceeded 80 in 1997 (Figure 5-16), yet at the same time population levels were estimated to remain relatively constant (approximately 250 300; Calvo and Silvy 1996). The greatest threat to population levels of the Florida Key deer is habitat loss and fragmentation from human development activities (Lopez et al. 2003, and USFWS 1985; Figure 5-17). Road density on the island is 41.2 km/km2. To protect the remaining area, an ecosystem management approach was adopted that included development moratoriums and land-use plans, Figure 5-15. Cumulative impacts of roads on wildlife and wildlands. Severity in extent and cumulative nature of impacts become magnified with increased time and space. Time S p ace

PAGE 224

208 construction of underpasses and highway improvements, high profile signage, public education, and targeted land acquisitions (USFWS 1985). Figure 5-17. Road density on Big Pine Key and Florida Key Deer National Wildlife Refuge. Road density on the island was 41.2 km/km2 in 1997. MiamiLEGENDUS 1 Figure 5-16. Highway mortality of key deer in Florida. 010203040506070809085868788899091929394959697 Source: USDI, 1998.Year

PAGE 225

209 Road Impacts and Planning in Florida Human population growth and the level of road impacts are best explained in terms of their statistical levels: Human population growth new residents/hr (USCB 2000) Road construction.8 km of new road constructed/day (FDOT 1998) Highway trafficapprox. 372.1 million vehicle km/yr (FDOT 1998) Fuel consumption.1 billion liters of gasoline/yr (FDOT 1998) Conveyances million registered private vehicles (FHWA 1996) These levels rank among the highest nationwide (Chapter 1). Transportation planning is the principle means of addressing the impacts of roads, and the need for additional roads to accommodate the burgeoning population and associated increase in traffic. In Florida, road projects are programmed on three time horizons, annually, 5 years and 25 years (discussed in a later section). These time horizons correspond to: 1) project implementation, 2) project engineering and contract bidding, and 3) plan review and impact assessment, respectively. Environmental review most appropriately occurs under 3 (above); and must occur by the 2nd time horizon (above) to affect change in engineering plans. At the annual level, it is normally too late to provide input concerning environmental impacts, unless unique circumstances exist. Given these process guidelines, transportation and conservation planners, working together, appears the most effective way to appropriately address ecological issues associated with proposed road projects. Biodiversity Conservation and Transportation Planning and Policy Coordination between transportation agencies and resource conservation agencies is essential to the development of effective policies that protect ecological systems, while

PAGE 226

210 simultaneously providing safe and efficient transportation systems. Significant efforts in this direction began in Florida in the early 1990s. Planning for Habitat Connectivity Along with the encroachment of roads and development, isolation and fragmentation of ecosystems and associated wildlife populations has occurred with varying detrimental effects, as previously discussed (see also Chapter 6). Harris (1985) presented one solution to this phenomenonestablishing conservation corridors or landscape linkages to connect large conservation areas. Numerous conservation biologists have since discussed the potential positive and negative aspects of corridors to landscape reintegration (Bennett 1999,1991,berloff and Cox 1987, and Noss and Harris 1986). Land bridges connecting continents, according to paleontologist records, provide evidence of use of corridors by entire faunas (Harris 1985). Physical connectors of sufficient size to promote genetic flow in two directions of full and balanced plant and animal communities are necessary to maintain intended ecosystem functions and processes (Noss and Cooperider 1994, Noss 1991, and Harris 1985). Standards below this objective may serve only to increase the spread of weedy species, and maintain the isolation of interior specialists and those requiring large territorial boundaries (Harris 1985; also see Beier and Noss 1998,berloff and Cox 1987). Eisenberg (1986) explains that these linkages may not increase population size, but would increase the likelihood that recolonization of locally extinct species could occur. In addition, such connections would protect against the weakening effects of inbreeding (Eisenberg 1986). Beier and Noss 1998, Dobson et.al. 1998, Csuti Soul 1991, Noss 1987, Sim Dobson et.al. 1998, and Sim

PAGE 227

211 These corridors must have sufficient width to maintain interior habitat qualities that would enhance use by threatened area-sensitive species (Noss 1983; see also Noss and Cooperider 1994, and Soul 1991). Design of these landscape connections; however, are thwarted by the presence of roads within and surrounding habitat islands. These roads act as barriers to dispersal and function as wildlife killing machines (a term coined by Walt Thomson of the Florida Department of Environmental Protection; in Harris and Gallagher 1989). Studies of reserve network design recognize road density as a critical indicator for evaluating system integrity (Noss 1995). Roads, as a barrier to animal movement, are considered one of the six major determinants of functional connectivity (Noss and Cooperider 1994). Recommended designs now illustrate use of wildlife crossings to permeate transport facilities (Noss 1995). The Statewide, Greenways Network Plan was designed to provide guidance for conserving valuable natural resources of Florida; and to restore connectivity between core conservation reserves and other isolated conservation areas. As defined by the University of Florida Greenways Planning Team (FGPT), "a greenway is a corridor of protected open space that is managed for conservation and/or recreation" (FGP 1999, page 1). They function as linkages between parks and nature reserves to create an interconnected system. The FGPT followed a regional, landscape approach designed to establish a plan for an ecologically functional, statewide habitat system (Hoctor et.al. 2000). The recommended greenways plan includes 9.31illion ha (57% of the state). Of this area, 63% is public land, CARL or SOR proposals, or open water (FGP 1999). These lands include many large natural hubs (core conservation areas) and connecting linkages (Figure 5-18). Within the core conservation areas and associated linkages m

PAGE 228

212 identified in the greenways plan are many intersections with roads. The Marjorie Harris Carr, Cross-Florida Greenway, a State Recreation and Conservation Area, provides a fitting example of the possible extent for conflicts with approximately 24 major road intersections (Figure 5-19). As part of the design of a functionally integrated, ecological network, many of these road intersections will require some type of mitigation measures, such as wildlife underpasses. The use of highway crossing structures at intersections with greenway linkages (habitat corridors) offers a method to reduce transportation-related, wildlife mortality and restore connectivity to the landscape. It has provided a realistic opportunity for Figure 5-18. Greenway-highway interfaces. The map displays important linkages necessary to maintain connections between remaining primary conservation areas for wildlife populations, including the Florida panther and black bear. Major roads that intersect greenway linkages are shown in orange. Many of these act as semi-permeable barriers to wildlife movement and alter various landscape-level ecological processes.

PAGE 229

213 ecologists and engineers to reduce the negative effects of roads, by restoring natural processes (e.g., wildlife movement and migration, flood, and fire) to a semblance of conditions prior to fragmentation of the landscape. The objective of the Florida Greenways Program is to establish an ecological network of green infrastructure whereby the aforementioned processes can occur across landscapes throughout the state. This program is closely linked with other programs with similar goalsPreservation 2000/Florida Forever 2010 and FFWCC Gap Analysis. Transportation-related Environmental Policies As a result of public opinion about major transportation projects in the 1980s and early 1990s, including those discussed in Chapter 1, Florida politicians and Illustration Credit: Larry Harris, University of Florida, Gainesville, FL. Figure 5-19. The Marjorie Harris Carr, Cross-Florida Greenway. Spanning over 140 km in length, there are 24 major road intersections to plan for in maintaining connectivity throughout the linear park.

PAGE 230

214 administrators began adjusting policies to address wildlife protection. On September 22, 1993, FDOT Secretary Ben Watts established the following policy: "FDOT will coordinate with the State Comprehensive Greenways Program to properly locate new transportation facilities such that the least possible impact upon corridors of native habitat occur. In addition, wildlife crossing structures will be installed on existing facilities where impacts on wildlife are determined to be necessary. These issues will be evaluated in coordination with other agencies and the public" (Watts 1993, Page 1). Subsequent to this policy, FDOT's mission statement in the 2020 Florida Transportation Plan (FTP) included the language "sustaining the quality of our environment" (FDOT 1995). From Goal 5 of the 2020 Florida Transportation Plan, travel choices to insure mobility, sustain the quality of the environment, preserve community values, and reduce energy consumption, the FDOT seeks strategies to reduce urban sprawl and encourage transit use. Conservation strategies include support for ecosystem management through mitigation banking and greenway planning, and increased use of native plantings on right-of-ways (FDOT 1994). The 2020 FTP (long-range transportation plan) is reviewed and revised periodically (approximately every 5 yrs) to keep current with changes in public policy and transportation needs. The ongoing planning process and public involvement stages for the 2020 FTP provides critical opportunities for environmental planning and inclusion of ecological designs to reduce impacts of highways on Florida's shrinking natural ecosystems.

PAGE 231

215 Figure 5-20 displays the proposed road projects in the Florida Transportation Plan through the year 2020. Five particular projects potentially impact important greenway linkages identified in the Statewide Greenways Network Plan (discussed in more detail in the last section). These include: the 4-laning of SR 40 through Ocala National Forest, the outer Jacksonville beltway, the Florida Turnpike extension, the northwest section of the Orlando beltway, and the Suncoast parkway. The latter four, most importantly, possess the potential to dramatically increase development in currently rural areas. State Road 40 has the potential to increase fragmentation in the Ocala National Forest by magnifying the size of the current barrier. The Orlando-Orange County, Expressway Authority (OOCEA), an autonomous political authority created by state legislative act (initiated by the City of Orlando and Orange County), proposes to build new expressways to link the City of Orlando with Figure 5-20. Florida 2020 Transportation Plan. This plan serves as a blueprint for the future development of Florida's landscapes. It provides the basis for all decisions related to the provision of state transportation facilities. (Adapted from 2020 Florida Transportation Plan; FDOT 1995) Jacksonville Beltwa y (loop 2) SR 40 four-lane widening project Florida Turnpike extension Orlando Beltway (NW) Suncoast Parkway

PAGE 232

216 major population centers and statewide transportation corridors in an eight county region of central Florida (Figure 5-21). Though the project locations have been proposed; no environmental assessments have been performed to evaluate the location of the planned corridors, before being published in government-sponsored newsletters (OOCEA 2000). Ecological assessments should be conducted, at least generally, before maps are published proposing new routes that may cause tremendous environmental impacts. The model presented in Chapter 2 and Appendix B was designed specifically for this purpose, to provide quick identification of good and poor locations for alternative road corridors. Preliminary results can be discussed among officials and followed up by rapid site assessments prior to proposal to the public. Orland o Leesburg MelbourneDaytonaBeachLakeland Figure 5-21. The Orlando Orange County, Expressway Authority (OOCEA) 2025 Transportation Plan for the greater Orlando area. The proposed plan would link an eight county region of central Florida with high-speed expressways and toll roads. (Adapted from Expressway 2025 Master Plan ; OOCEA 2003 )

PAGE 233

217 Considerations for Site Design that Enhance Use By Wildlife Landscape ecology principles can be used to integrate transportation and conservation planning. An ideal meeting of road and greenway networks would look something like the engineering marvels constructed as part of the original Cross Florida Barge Canal Project (Figure 5-22). This type of structure probably allows movement of complete faunal and floral assemblages. More economically viable and innovative solutions are required however. Due to the large amount of carnage occurring on our highways, many methods and devices have been tested and used to help prevent wildlife-vehicle collisions. Examples include wildlife underpasses, tunnels or ecoducts, barrier fencing, creation of new habitat areas, warning signs, human intervention and assistance, temporary road closures, headlight reflectors, and sound devices and infrared sensors for vehicles. Figure 5-22. Cross-Florida Barge Canal overpass. Marion County Road 316 passes high above the Oklawaha River floodplain in Ocala National Forest. ( Photo Credit: Larr y Harris, Universit y of Florida, Gainesville, FL )

PAGE 234

218 Vegetative screeningblock headlights, diminish traffic noise and provide cover Underpasses are used in south Florida by the Florida panther and other wildlife that cross Interstate-75 (Foster and Humphrey 1995). Other species that underpasses were constructed for include mountain goats Oreamnos americanus (Singer and Doherty 1985), black bears (Roof and Wooding 1996), deer (Kuennen 1989, Ford 1980, and Reed et al. 1975), mountain pygmy possums Burramys parvus (Mansergh and Scotts 1989), and badgers Taxidea taxus (Ratcliffe 1974). Smaller animals (such as toads and salamanders) were shown to use tunnels or culverts (Norden 1990, Langton 1989b, Tyning 1989, and Ryser and Grossenbacher 1989). Barrier fences can prevent deer from entering the roadway (Day 1990, Stadler 1987, and Feldhamer et al. 1986), but likely intensify the fragmenting effects of roads. Headlight reflectors, warning whistles, and motorist signage have provided inconsistent performance at best in reducing vehicle collisions with wildlife (Pafko and Kovach 1996, Reeve and Anderson 1993, Romin and Dalton 1992, Schafer and Penland 1985, and Pojar et al. 1975). Most research indicates that ecopassage-barrier fence systems are the most effective method for increasing road permeability while simultaneously reducing wildlife-vehicle collisions. Effective vs. ineffective design, correct vs. incorrect placement at heavily used crossing sites, or good maintenance (that provides containment) vs. poor maintenance (that allows circumvention) on fencing or barriers can be keys to success or failure of ecopassages. The following list of parameters can be used in practice to improve permeability of highways and increase connectivity of adjacent conservation areas: Context Sensitivityvegetation consistent with surrounding habitat Environmental variabilityprovide for terrestrial passage at semi-aquatic sites during periods of high water levels

PAGE 235

219 Directional fencingfunnel wildlife through passages and away from road surface Bermingreduce effects of traffic noise and lights Topographyroad should be designed to fit into the landscape (e.g., minimize alteration in slope of underpass/ overpass approaches) Substrateconsistent with adjacent area Lightingreduce tunnel effects by increasing openness value (height*width/length) and providing light penetration in medians of divided highways Human presencereduce human access associated with crossing sites Economic road construction practices commonly include cut and fill techniques to overcome problems associated with uneven ground. While in areas of minimal topographic relief, this technique may be suitable for constructing level roadbeds; when relief is more significant, more ecologically sound practices should be used (Figure 5-23) to increase landscape connectivity. If we must build roads, instead of simply placing them on the landscape, they should be designed to fit into the landscape to minimize barrier effects and maximize landscape connectivity. Effects of local topography and scouring (e.g., formation of depressions at culvert entrances), and seasonal and interannual variation in precipitation were shown in Chapter 4 to prevent functionality of certain culverts. Under dry conditions use by wildlife was recorded, yet under wet conditions flooding of depressions prevented access by terrestrial organisms. Road maintenance should include routine filling of depressions or employ methods to prevent persistent pooling of water at culvert entrances. One means of addressing this issue is the construction of multi-cell units. Multi-cell culverts could be designed at multiple elevations to provide dry passage at various water stages (Figure 5-24); alternatively in Europe some culvert designs include an elevated sill alongside creek drainages to provide adjacent passage for terrestrial species (Veenbaas and Brandjes 1999).

PAGE 236

220 The efficacy of underpasses is largely determined by site characteristics outlined above. Vegetation and substrate are two such characteristics that can be controlled through management practices. Wildlife movement monitoring efforts (discussed in Chapter 4) identified the lack of use of structures with substrate consisting of coarse rip-rap (large rocks or concrete designed to prevent scouring and erosion) and avoidance of open non-vegetated structures by black bears. Figure 5-25 provides examples of good and poor design features. Where scouring effects are severe, a soil base is insufficient to Figure 5-23. Existing highway structures and potential sites for wildlife crossings. A. An example (Florida Turnpike) where topographic relief could be used to construct functional overpasses and underpasses for connecting proposed greenways instead of following standard fill practices for road-building. B. An underpass that was constructed over a stream reach and ravine on SR 123 in Eglin Air Force Base, Florida provides an example of ecologically sound roadb uildin g p ractices. A B

PAGE 237

221 maintain structural integrity of bridge pilings; however, substitute materials could be applied such as tiles, asphalt or concrete paving around pilings and below the bridge. The primary objective of efforts to retrofit existing roads or plan for new roads should be to identify appropriate measures at the proper locations (Figure 5-26). An integrated transportation/conservation planning approach that incorporates landscape ecology principles and uses tools such as GIS can achieve this objective. The research presented here has endeavored to follow this approach. Figure 5-24. An example of a multi-cell culvert in northwest Gainesville, Florida. In two separate years of monitoring, this culvert experienced dry and wet periods. Wet periods significantly reduced use by wildlife.

PAGE 238

222 Social and Economic Priorities and Future Research Needs As an annual comparison of investment, Florida spends $5.8 billion (FHWA 2002c) for highways versus $1.3 billion (BEBR 2001) for conservation and natural resource management (4.5:1 differential in expenditures). Only $300 million is dedicated A. US 17 bridge over Deep Creek in Volusia County. Dimensions: 51 m wide (includes 25 m terrestrial connection, no recorded large animal movement), 14 m long, 2 m high Grass-regularly mowed, coarse rubble substrate (inset) Wildlife bridge extension on SR 46 over the Wekiva River in central Florida. Dimensions: 162 m wide (includes 74 m terrestrial connection regularly used by black bears, deer, etc.), 14 m long, 3 m high N ative trees/shrubs, cla y -loam Figure 5-25. Effects of vegetation and substrate at wildlife underpasses. Through wildlife movement monitoring efforts, these examples showed that well landscaped sites with soft ground were used much more frequently. B A A in set

PAGE 239

223 to conservation land acquisition. This reflects a significant preference by society to devote public resources toward transportation rather than conservation. Each year in the U.S., over one million deer-vehicle collisions occur that cost $1.1 billion in property damage, and cause an estimated 29,000 human injuries and 211 human fatalities (Conover et al. 1995). Although estimates of vehicle collisions with elk, moose, black bear, etc. were not available, they likely are similar in extent of damage but fewer in number. These values demonstrate the need, from a human perspective (economic and social), to increase priorities for conservation strategies designed to reduce these impacts. What is the Economic Equation for Success? Short of a needed shift in social and economic priorities, only frugal and creative mechanisms are available to reduce impacts of roads on conservation areas. In 1992, the Figure 5-26. Identifying sites for retrofits (increasing permeability of roadways). Can we identify what type structures should be where? Example of wrong structure at the wrong spot. In this case road-killed deer were recorded (A) where a frequently used deer trail (B) crossed a highway fitted with a small double pipe culvert (C). A C B

PAGE 240

224 FDOT had an inventory of 5,771 bridges (FDOT 1992). Due to limited resources, the FDOT annually carries a backlog of more than 1,000 of these bridges in need of repair and 350 in need of replacement. With appropriation of funds for these projects, significant opportunities would exist for enhancement to designs to increase habitat connectivity. A summary of FHWA transportation funding could shed light on mechanisms for partial funding of road mitigation at greenway intersections. A total of $1.68 billion in Federal funds will be allocated annually for the next five years for transportation improvements in the State of Florida (FDOT 1999). Over this 5-yr period, $45 million annually is dedicated to the TEA-21 program for transportation system enhancements in 12 eligible categories (Gary Evink-FDOT, Personal Communication). A new category now eligible for these funds is mitigation for environmental impacts on wildlife. A major stipulation; however, is that TEA-21 funds can only be used when the project is linked to highway mortality of wildlife. Administratively, TEA-21 funds are distributed to individual FDOT operational districts; and local government Metropolitan Planning Organizations (MPOs) determine their application. The primary use, to date, has been for alternative transportation enhancement (e.g., bicycle trails, pedestrian or recreational trails and walks, landscaping and aesthetics). The first project funded by TEA-21 monies for wildlife impacts was a Florida panther underpass on SR 29 in south Florida. Legal mandates for wetland protection have existed in Florida and the United States since the 1970s. Required under Section 404 of the U.S. Clean Water Act and additional state regulations regarding wetland protection, the FDOT must provide compensation or replace wetlands destroyed as a result of road construction. The

PAGE 241

225 standard practice followed by the FDOT is provision of funds to regional water management districts that are primarily used in wetland mitigation banking programs (P. Southall-FDOT, Personal Communication). Similar measures are needed for rapidly disappearing native upland communities that act as critical linkages in an integrated ecological network. To mitigate fragmentation of intact native upland communities by new highway corridors, transportation agencies should be required to provide underpasses that maintain connectivity between severed habitat areas. An economic program designed to improve functional connectivity of ecological networks must take advantage of available funding sources through all programs associated with negative impacts of highways such as wetland mitigation, river protection and restoration (e.g., Save Our Rivers, FDOT bridge replacements), TEA-21 enhancements, and conservation land acquisition (e.g., CARL, Florida Communities Trust). Land Ownership, Conservation Easements and Highway-Greenway Interface Zones State agencies have been reluctant to spend public funds for underpass structures on roads adjacent to private lands, when long-term use of the property for conservation is not guaranteed (Gary Evink-FDOT, Personal Communication). Although uncertain future use of land is a legitimate concern; private lands that currently function as commercial forest land, livestock grazing, or other low-intensity agricultural uses serve as valuable assets as landscape linkages between public conservation areas. It is important to identify highway-greenway interfaces within private lands as potential sites for mitigation. Perhaps, if these areas were designated as part of a statewide, habitat

PAGE 242

226 conservation network, land development could be directed into other less critical areas; and prevent urban sprawl and strip development along roads in rural areas. A mitigation strategy should not only include construction of underpasses or other means designed to reduce fragmentation of habitat linkages, but land-use controls or conservation designations as well. As property values along roads soar it has become increasingly difficult to justify outright purchase of land (fee simple acquisition) to further conservation goals. Opposite from outright purchase are zoning and land-use restrictions; however, these are unpopular with landowners and subject to change under political transitions. And, therefore, are less effective than perpetual land conservation devices (Wright 1994). Alternative strategies that are becoming more attractive include control of partial interests of land (i.e., conservation easements) throughdonated development rights (DDRs), purchased development rights (PDRs), and transference of development rights (Wiebe 1997). The use of PDRs, the most widely used option, began in earnest in the 1930s when the U.S. National Park Service established the Blue Ridge and Natchez Trace parkways; purchasing adjacent development rights to preserve views of scenic open space (Haapoja 1994). The State of Wisconsin also exercised this method to preserve vistas along the Great River Road in the 1950s (Wright 1993). As of 1993, only nine states have conservation easement programs protecting approximately 83,000 ha (Wright 1993). In Florida, PDRs have been used to establish conservation easements on commercial forest lands in critical floodplain management and aquifer recharge areas. The use of DDRs by conservation trust organizations has been more successful resulting in the protection of over 1.1 million ha of private lands from development

PAGE 243

227 (Wright 1993). Over 57% of the State of Florida has been identified as part of the Florida statewide ecological network. Of those lands identified, 22% are in private ownership with little to no development restrictions. The use of DDRs and PDRs can play a critical role in protecting these private lands from development; aside from costly fee-simple acquisition purchases that remove land from the property tax rolls. These techniques establish conservation easements, yet provide multiple benefits to the property owner by preserving the land's current economic use, reducing property tax assessments through devaluation of the land, and include either a cash payment (for sale of development rights) or income tax deductions (qualified as charitable contributions) for relinquishing development rights in perpetuity (Haapoja 1994). Considering that conservation easements were deemed appropriate for protecting scenic vistas in the 1930s, they are just as essential now for maintaining critical linkages at greenway-highway interfaces for an ecological network. Future Planning and Research to Integrate Transportation and Conservation Objectives in Florida More intensive wildlife movement studies of key indicator species should be conducted to analyze behavior associated with encounters with roadways. Planners and policymakers could use this information to make better decisions regarding road corridor locations, road construction methods and designs, and integration of road and ecological networks. Specifically, linking the results of the ecological hotspots on highways model (Chapter 2) and the Florida statewide greenways model would support the goal to conserve Floridas remaining biodiversity. This goal can be facilitated by using the model results to program mitigation with future road projects, and by using the inventory

PAGE 244

228 of existing infrastructure (Chapter 3) at prioritized greenway-highway interfaces to evaluate the present permeability of existing roads. The information generated from the study on wildlife use of existing bridges and culverts (Chapter 4) should be used toward the development of design standards for new wildlife crossing structures. Capability to coordinate greenway priorities with road crossing priorities could be further strengthened by modifying the prioritization model (Chapter 2, Appendix B) to include the ability to assess development pressure. A matrix that plots the intensity of development pressure against a gradient of strategic importance could further fine-tune the prioritization process; both with regard to road projects and acquisition of critical habitat-corridors (Figure 5-27). Evaluation Criteria regional greenway systems strategic habitat conservation areas ecologically-based design vs. engineering band-aids existing highway structures sheet flow, floodplains, riparian zones chronic road-kill sites threatened and endangered species charismatic species public relations value develo p mental p ressure Figure 5-27. A conceptual matrix to further fine-tune ability to establish priorities for restoring connectivity/constructing wildlife underpasses.

PAGE 245

CHAPTER 6 LITERATURE REVIEW Ecological Effects of Roads Highways and motor vehicles increase mobility of humans, but often at a high cost to wildlife. Construction of roads results in fragmentation of wildlife habitat (Hall et al. 1994, Salisbury 1993, Harris and Silva-Lopez 1992, Andrews 1990, Mech et al. 1988, and Dickman 1987), creation of barriers to wildlife movement and dispersal (Clarke et al, 1998, Madsen 1996, Brody and Pelton 1989, Harris and Gallagher 1989, and Wilkins 1982), negative edge effects (Lindenmayer et al. 1999, Gibbs 1998, and Mader 1984), and increased mortality of species attempting to cross these roads (Gilbert and Wooding 1994, Cristoffer 1991, Stadler 1987, Wooding and Brady 1987, Oxley et al. 1974, and Sargeant and Forbes 1973). General reviews and case studies of these effects can be found in Evink et al. (1999), Evink et al. (1998), Forman and Alexander (1998), Spellerberg (1998), Evink et al. (1996), and Bennett (1991). These and other pronounced ecological effects not only occur at the interface between highways and habitats, but also may extend thousands of meters beyond (Forman 1995). Direct and indirect impacts of roads to landscape-level ecological processes and species composition can be separated into four categories: traffic, isolation and barrier effects, edge effects, and human access. Each road effect can be further characterized by the specific relationships between ecological patterns and processes (Table 6-1). 229

PAGE 246

230 Table 6-1. Ecological Effects of Roads Effects Direct Impacts Indirect Impacts Traffic (Pattern) Cause direct mortality through wildlife-vehicle collisions (Process) that can reduce population levels, alter age structure and reduce long-term viability (Fahrig et al. 1995) (Pattern) Attract carrion feeding species to road-kills that can, in turn, lead to additional vehicle-related mortality (Bedford and Griffiths 1995, and Harding 1986) (Process) that could impact predator-prey relationships and population dynamics of predator species Isolation/ Barrier (Pattern) Create effective barriers/ filters to movement for many species, (Process) thus altering normal migratory, foraging and mating behavior (Harris and Scheck 1991) (Pattern) Create fragmented habitat mosaics consisting of isolated habitat islands among dense road networks and urban development (Process) that can lead to genetic isolation and decay (Mader 1984), and loss of populations and species (Harris 1984) Edge (Pattern) Facilitate the rapid spread of invasive weedy species (Process) that often out-compete native species, and are not as subject to predation (Stiles and Jones 1998, Wilcox 1989, and Verkaar 1988) and thus displace native species (Pattern) Alter adjacent habitat through drainage modifications, erosion and siltation (Process) that may affect vegetation structure, wildlife composition, reproduction and movement success (Gunn and Sein 2000, and Southall 1991) (Pattern) Act as passive repositories for human-generated pollutants (Process) that degrade habitat quality and increase the threat of toxic poisoning (Muskett and Jones 1980, and Leedy 1975) (Pattern) Create habitat ecotones (subject to altered environmental gradients of light, moisture, noise, temperature, etc.) (Process) that negatively effect breeding of area sensitive and forest interior species; generally characterized by increased predation and parasitism by opportunistic species Human Access (Pattern) Create greater human access to remote areas, (Process) thus facilitating commercial resource extraction and habitat loss through land conversion and development (Pattern) Increase access to remote resource areas for hunting, poaching and recreation (Process) that may alter wildlife population dynamics (Noss and Harris 1986, and Holbrook and Vaughn 1985) Vehicle-Wildlife Collisions Numerous authors have documented wildlife fatalities on roads due to vehicle collisions (McRae 1997, Bruinderink and Hazebroek 1996, Romin and Bissonette 1996, Smith 1996, Belant 1995, Cristoffer 1991, Harris and Scheck 1991, Day 1990, Dodd et al. 1989, Harris and Gallagher 1989, Lalo 1987, Stadler 1987, and Wooding and Brady 1987). Noteworthy examples of road-kills would include impacts on endangered species and on biological phenomena in general (e.g., amphibian migrations). Several instances

PAGE 247

231 of massive amphibian road-kills are reported in Europe, New England, and Florida where large overnight migrations occur (Barichovich and Dodd 2002, Smith and Dodd 1999, Jackson 1996, Smith 1996, and Langton 1989a). The Florida panther, Florida black bear, Florida key deer Odocoileus virginianus clavium, and American crocodile Crocodylus acutus serve as critical examples of highway mortality on populations of large threatened and endangered species. Automobile-related deaths of the Florida panther gained international attention in the 1970s with the planned extension/expansion of I-75 (Alligator Alley). Several studies discuss issues associated with the highway project and Florida panthers in south Florida, including installation and effectiveness of wildlife underpasses (Logan and Evink 1985, Foster and Humphrey 1995, and LoBuono 1988). From 1976 to 1999, 729 black bears died from vehicle collisions on Florida highways (FFWCC 2000). Annual mortality levels from vehicle collisions continue to grow every year. In 1998, 88 black-bear deaths were recorded on Florida highways (FFWCC 2000). There is a direct positive relation between black-bear road-kills and traffic volume on 2-lane highways (Figure 6-1). The population of Florida key deer (limited mainly to Big Pine Key in South Florida) averaged more than 50 deaths/yr from vehicle collisions between 1985 and 1997 (USDI 1998). The only population of American crocodiles in the United States also has been impacted by road fatalities. Gaby (1987) has shown that 46% of human-related mortality of crocodiles in Florida is from vehicle collisions. The threat is further exacerbated when the exceedingly low population levels are considered. There are an estimated 46 to 74 Florida panthers (Maehr 1997); only seven

PAGE 248

232core populations of Florida black bears, with some 2,000 to 2,500 individuals (Eason and O'Meara 2000, and Walt McCown, personal communication); 250 to 300 Florida key deer (Calvo and Silvy 1996); and approximately 800 American crocodiles (Sharp 1999). y = 17.331x 49.57R2 = 0.9102051015202530354045500123456Avg Daily Vehicle Trips (x 1000)Road-kills Habitat Loss, Fragmentation, and Isolation Habitat isolation and fragmentation by linear structures such as roads may have drastic effects on biological diversity. Human influence threatens native biological diversity through loss of species from genetic inbreeding, elimination of large uninterrupted habitat, and invasion of alien species (Forman and Alexander 1998, Andrews 1990, and Harris and Gallagher 1989). The rapid fragmentation of landscapes by roads and urbanization results in the loss of normal dispersion patterns (Harris and Scheck 1991, and Andrews 1990). Landscape character has switched from one of humans in a natural landscape matrix to one of natural areas in a human-dominated Figure 6-1. Comparison of Florida black bear road-kills and average daily vehicle trips (ADVTs). As traffic volume increases black bear-vehicle collisions increase. (Data source: Florida Fish and Wildlife Conservation Commission 1994 )

PAGE 249

233 landscape. Dickman (1988) investigated mammalian species richness as affected by proximity to habitat-fragmenting forces, such as roads and urban development. He found that small patches were unsuitable for long-term maintenance of species diversity, as mammalian species richness declined rapidly with decreasing distance to roads, buildings, and so on. Areas of insufficient size cause local extinctions and inbreeding depression for species with large area requirements, due to isolation of gene pools (Dickman 1988). It was also found that specialists were usually excluded and generalists retained when patches were less than 0.65 ha (Dickman 1988). Maehr and Cox (1995) demonstrated the importance of habitat contiguity for the Florida panther in South Florida. By correlating landscape features with radio-telemetry point locations, they were able to show that panthers preferred large forested habitat areas and avoided open cover types. Size of occupied suitable habitat patches averaged 20,816 ha (Maehr and Cox 1995). Additional analysis showed that only six adequate roadless areas of this size remain in Florida. Four of these areas, all located in south Florida, are contiguous, but separated by major highways and canals. Furthermore, collisions with automobiles represent the principal means of human-related mortality for the Florida panther (Maehr et al. 1991). Successful dispersal and expansion of the current range of the Florida panther will depend on provision of large contiguous forested habitat areas. Likelihood of patch occupancy by black bears in Florida also appears to increase as patch size increases (Hellgren and Maehr 1992). In fact, bears used public preserves of less than 100,000 ha only when adjacent to larger occupied tracts of land (Hellgren and Maehr 1992). Key management issues for Florida panther and black bear include

PAGE 250

234 increasing permeability of roads and reducing habitat loss and fragmentation through cooperative land management schemes with private landowners. Density of Roads One of the primary effects of fragmentation on the presence of certain wildlife populations is road density. Several studies have examined the effect of road density, with conclusions varying depending on the species involved. The consensus among studies of wolves Canis lupus in the Great Lakes region, was that wolves require areas with road densities less than 0.54 to 0.69 km/km2 to sustain viable populations (Mech 1989, Mech et al. 1988, Jensen et al. 1986, and Theil 1985). Lyon (1983) presented evidence that elk Cervus canadensis use of habitat areas in Montana, Idaho and Washington reduced to 25% when road densities were greater than 3.11 to 3.73 km/km2. Road density was also determined to be a critical factor in habitat suitability for the moor frog Rana arvalis in Europe (Vos and Chardon 1998). Roads appear to have an effect on habitat use by the Florida panther, especially with regard to den locations. Maehr (1997) found that resident females without kittens use habitat areas near roads, while those with kittens prefer den locations greater than one km from roads. This suggests that habitat areas with roads may be suitable Florida panther habitat provided significant road permeability is afforded between marginal road-inclusive areas and larger roadless denning-habitat areas. Extent of habitat fragmentation and isolation caused by increasing road densities in developing areas is critical for determining the persistence of disturbance-sensitive species. Based on numerous studies that have documented these impacts for various species (Bowers and Matter 1997, Rodgers and Smith 1996, Van Dyke et al. 1986, and

PAGE 251

235 Opdam et al. 1985), the best prescription for retaining disturbance-sensitive species would include conservation reserve designs that maintain contiguous large reserves. Ideally these reserves would contain wetland-upland gradients of suitable habitat (high landscape structural complexity preferred) and network connections that contain similar characteristics and that minimize negative edge effects. Negative Edge Effects The high diversity of wildlife associated with edges and ecotones has been widely accepted as a fundamental principle in ecology (Harris 1988). Historically, this was considered a positive effect and resulted in prescribed strategies for increasing edge habitat to increase game populations (Paton 1994). However, it is now recognized that the management strategy of increasing edge habitat can be detrimental to many edge-sensitive species. The attractiveness of road verges has been noted for numerous birds, rodents, and deer; all species that prefer open fields or maintained grassy areas for foraging or nesting (Canaday 1996, Paton 1994, Rich 1994, and Harris and Gallagher 1989). Marini et al. (1995) and Harris (1988) used the term ecological trap to refer to the attractive yet destructive effects of edge habitat on traditionally forest-interior nesting birds. Although forest-interior nesting birds occupy edges, the greater presence of nest predators and parasites (e.g., brown-headed cowbird, Molothrus ater) there renders the habitat less suitable (Harris and Gallagher 1989, and Harris 1988). Increased edge can impact interior species by increasing predation from edge predators as far as 300 to 600 m into a forested patch (Norse et al. 1986). Disruption of normal behavior, communication skills, and mating success can also result, as was shown for nocturnal frogs by artificial lighting

PAGE 252

236 (Buchanan 1993) and for anurans and birds by traffic and other technogenous noise pollution (Ilichev et al. 1995, and Barrass 1985). Reijnen et al. (1997, 1995) attributed forest breeding birds aversion to road verges to reduced habitat quality (primarily caused by traffic noise; and to a much lesser extent, visual disturbance or pollutants). These studies examined the effect of proximity to roads on breeding-bird density and found a 60% reduction in species diversity in plots adjacent to roads. The threshold where bird densities decline was defined as the distance from the highway where traffic noise is 42 decibels (dB) or higher (Reijnen et al. 1995). Yet other studies have found a notable variation in negative edge effects on breeding-bird nesting success, depending on the species (Marini et al. 1995); nest location preferences (e.g., clearcut, forest, and grassland) (Rudnicki et al. 1993); or type of edge (e.g., landscape structural differences, abrupt v. gradual edges) (Suarez et al. 1997). Fahrig et al. (1995) demonstrated the reduction in anuran densities as traffic intensity increases. These studies suggest significant complexity and variability in edge effects according to landscape structure, species composition, and physical attributes associated with traffic density and roadway type. Widening of highways increases the potential magnification of negative edge effects, including the absence or reduction of species sensitive to noise or visual disturbance and increased presence of predators and weedy species (e.g., birds, rodents, and omnivores) (Forman and Alexander 1998; Gibbs 1998; Reijnen et al. 1997, 1995; Bennett 1991; Andrews 1990; Garland 1984; Adams and Geis 1983; Wilkins 1982; Ferris 1979; Kozel and Fleharty 1979; and Oxley et al. 1974). In summary, roads seem to increase the richness of species that are competitively advantaged in disturbed

PAGE 253

237 environments, while decreasing the abundance and richness of area-sensitive or forest-interior species. Alteration of Ecological Processes and Changes in Community Composition The creation of edge, caused by roads, agriculture, silviculture, and urbanization results in severe reductions of native biological value by artificially introducing abiotic and biotic forces (Harris and Scheck 1991). Roads act as barriers; and also alter the intensity of light, wind speed, temperature, humidity, evaporation rates, and noise levels that cause certain species to avoid road verges (Forman and Alexander 1998, Harris and Scheck 1991). Light penetration and structural macrohabitat variables (such as canopy cover, litter cover, and presence of stumps and snags) were critical in determining distance effects of roads on presence of amphibians (Demaynadier and Hunter 1998). These results present valuable implications for intensive forest-management schemes that typically remove much of this structure and their integral function in the persistence of amphibian populations. Roads also may act as significant barriers to important ecological processes such as animal and plant dispersal and migration. For instance, Langton (1989b) and Tyning (1989) documented the obstruction of normal migratory patterns of amphibian populations by roads. In Denmark, Hels and Buchwald (2001) calculated mortality rates of amphibians on minor (0.34 to 0.61 at 3,200 vehicles/day) and major (0.89 to 0.98 at 15,000 vehicles/day) highways. Mortality rates increased as mobility of various species decreased. For populations of slower amphibian species, high-traffic roads become an interminable barrier, effectively disrupting metamorphosis.

PAGE 254

238 U.S. Highway 27, a four-lane highway constructed in the 1960s on a raised roadbed, bisects Lake Jackson, in Leon County, Florida. A single culvert provides the only connection between the two sections of the lake (Figure 6-2). To prove the inadequacy of this culvert, Aresco (2001) recorded 7,195 turtles attempting to cross the road over 18 months. Although he rescued most of these with a temporary fence (Figure 6-2), previous road-kill surveys estimated as much as 90% mortality. In fact, roads are suspected to be a major cause of population declines for turtles throughout the eastern United States (Gibbs and Shriver 2002). Aresco (2001) suggests that severe reductions in turtle diversity and abundance could result in changes to ecosystem processes including alterations to primary and secondary productivity, plant biomass levels, species composition and food web structure. Harris and Silva-Lopez (1992) argue that faunal collapse can occur when disturbance levels are sufficient to cause fundamentally different intensities of ecological processes to prevail. An example of such collapse may result from removal or extirpation of gopher tortoises Gopherus polyphemus that provide habitat within their burrows for many other species. Such loss would reduce populations of all commensals. Divisive fragmentation on forest conservation areas caused by roads would facilitate this effect, thus altering the existing ecological processes (Harris and Silva-Lopez, 1992). Roads act as barriers or inhibitors to movement of small mammals and arthropods (Mader 1984, Wilkins 1982, Kozel and Fleharty 1979, and Oxley et al. 1974) and thus may affect species distributions and community interactions. When high-intensity roads prevent crossings by wolves and black bears (Pacquet and Callahan 1996, and Brody and

PAGE 255

239Pelton 1989) the reduction in predation rates in isolated areas could result in higher prey densities and greater herbivory pressure on plant communities (Peterson 1977). Figure 6-2. Lake Jackson and Little Lake Jackson wildlife crossing. US Highway 27 bisects Lake Jackson into two fragments connected by a single box culvert. This site exhibits exceptional numbers of migrating aquatic turtles. The barrier effect caused by the road is estimated to cause as much as 90% mortality (Aresco 2001). A temporary fence was set up to prevent turtles from being killed on the road ( inset ) .(Photo credit: Matt Aresco, Florida State University, Tallahassee FL ) Roads also act as conduits for invasive plants and disease. Many predators, scavengers, and large mammals are known to use lightly traveled roads as movement corridors (Forman and Alexander 1998, and Brody and Pelton 1989). Opportunistic predators (such as black bears, coyotes Canis latrans, and raccoons Procyon lotor) serve as dispersal vectors along roadsides for pathogens (e.g., Rabies Lyssavirus spp.), and

PAGE 256

240 exotic or invasive species (e.g., Brazilian pepper Schinus terebinthifolius). Gelbard and Belnap (2003) reported that cover and richness of invasive plants was 50% greater and richness of native plants was 30% lower adjacent to paved roads than on 4-wheel-drive trails in southern Utah. Imported red fire ants Solenopsis invicta are known to use road right-of-ways as dispersal corridors and entry points for invasion of adjacent habitat areas (Stiles and Jones 1998). In south Florida, probability of occurrence of imported red fire ants was greatest when distance to roads or urban development was less than 150 m (Forys et al. 2002). Roads can impose dramatic restrictions (related to human health and safety) on wildfires, another ecological process considered an important land-management tool (in the form of prescribed burning) for maintaining certain community types. This is especially relevant where high-volume, high-speed highways and urban sprawl are within or adjacent to conservation areas. For example, the Florida Division of Recreation and Parks must use alternative methods to control the growth of invasive woody species near two major highways that cross Paynes Prairie State Preserve (a large, wet prairie) because fire and smoke are believed to pose risks to driver safety (Southall 1990). Herbicide sprays and artificially controlled, water levels have proven marginally successful in reducing invasive woody species. Dispersal of Toxic Substances and Environmental Pollution The dispersal of motor-vehicle-produced chemicals and trace metals from roads upon road verges and adjacent habitat is well documented, probably because of the need to monitor human health risks and degradation to air and water resources (25, 19, 16 and 22 articles on this subject were found in the literature from the 1970s, 1980s, 1990s and

PAGE 257

241 2000s, respectively). Road-related substances also contribute to contamination of adjacent habitats (e.g., petroleum, asphalt debris and other particulates, r-o-w maintenance chemicals, and deicing agents). Elevated levels of nitrogen oxides near roads can promote increased dominance by a few plant species, resulting in reductions of diversity; and in some cases altering adjacent plant communities (Angold 1997, Weighell 1997). Documentation exists on roadside accumulation in soil and plants of lead, cadmium, nickel (Muskett and Jones 1980), copper, manganese, zinc (Yousufzai et al. 2001), platinum group metals (Rauch et al. 2000), polycyclic aromatic hydrocarbons (Krein and Schorer 2000), carbonyl compounds (Viskari et al. 2000), and deicing agents (Banasova 1986) disseminated by motor vehicles. Sutherland and Tolosa (2001) and Muskett and Jones (1980) examined dispersal distance of potentially toxic contaminants from roads. Both studies indicate that levels of lead, cadmium, and zinc in soil decrease with distance from the road surface. Concentration of other trace metals (e.g., nickel and copper) did not show a distinct pattern with distance from road, although anthropogenically elevated levels were apparent at certain test locations 50 to 100 m from the road, suggesting more complex factors affecting dispersion (e.g., wind, water flow, vegetation type, and seasonality). Few studies address the direct impact of these substances on specific wildlife. Those that exist indicate potential for serious health risks to wildlife. In Idahos Coeur d'Alene River Basin, Hoffman (2000) documented that ingested lead-contaminated sediment caused abnormal post-hatching development, disease, and 22% mortality of wild geese and mallards Anas platyrynchos (but see Beyer et al. 2000, regarding adult mute swans Cygnus olor exposed to toxic sediments). High concentrations of lead have

PAGE 258

242 also been reported for frogs sampled from highway drainages (Birdsall et al. 1986). A study (Harrison and Dyer 1984) on mule deer Odocoileus hemionus forage near roadways in Rocky Mountain National Park found that concentrations of lead were inversely correlated to distance from roads. It was also estimated that only 1.4% of daily intake of roadside vegetation was necessary to introduce excessive amounts of lead to mule deer diet. Floodplain and Hydrologic Alterations, Effects of Erosion and Chemical Runoff Hard-surfaced roads affect aquatic and semi-aquatic systems by altering runoff patterns, storage capacity, hydroperiods, stream velocity and depth, and sediment and chemical transport. Runoff associated with roads can increase velocity and quantity of water within riparian systems, resulting in modifications to stream morphology and stream migration within the floodplain (Winter 2001). Roads can also have negative effects on groundwater flow, reductions in percolation, and aquifer recharge. In topographically significant terrain, cutbanks in roads are known to interrupt slow moving groundwater seeps converting them to fast moving surface water (Forman and Alexander 1998). This leads to accelerated peak discharges and downstream flooding. Accelerated water flows lead to increased erosion and sediment transport; deposition of increased sediment loads downstream can adversely impact spawning areas by filling pools and gravel beds (Ruediger and Ruediger 1999). Roads also contribute to degradation of adjacent aquatic and wetland habitats through the distribution of deicing salts, agriculture-related chemicals, and heavy metals (that may cause serious health problems and reduce reproductive success rates) (Harris and Scheck 1991). Research by Findlay and Bourdages (2000) indicates that changes to

PAGE 259

243 wetland communities due to road construction may not become apparent soon after, as the effects on certain taxa may not take place for many decades. This suggests that time horizons of most highway impact studies (less than 5 yrs) are inadequate to insure protection of overall wetland biodiversity. Improved understanding of these effects and more thorough project planning are needed. Unpaved and aggregate-surface forest-roads pose significant negative ecological effects to otherwise undeveloped resource conservation areas. Of the approximate 644,000 km of national forest roads, impacts to aquatic systems include: slope and bank destabilization and erosion, suspended sediments in surface waters, alterations to natural watercourse or channelization, and changes to water levels and flow rates (Sheehy 2001, Ruediger and Ruediger 1999). Best Management Practices (BMPs) describe measures to reduce these impacts (FDOF 1987a, FDOF 1987b, Hynson et al. 1982). Management actions of particular interest are roadbed maintenance, slope and bank stabilization, and proper design and placement of culverts. Sheehy (2001) and Hynson et al. (1982) discuss methods to reduce erosion of the roadbed, banks, and slopes including: improved roadbed design and maintenance; use of buffers; and bank stabilization by vegetation, terracing, and rock-filled gabions. The dynamics of natural fluctuation in water levels and flow rates must be considered when constructing stream crossings. Crossings can be designed to provide functional passage for local fish populations during high and low water levels by using multiple culverts, sills or weirs, resting pool excavations, etc. (Ruediger and Ruediger 1999, and Hynson et al. 1982). In summary, road impacts on aquatic ecosystems may

PAGE 260

244 include increased flow rates, aquatic habitat alteration, sediment and chemical transport and redistribution, and erosion of biodiversity. Effects of Increased Human Access Four human activities derive from increased access to natural areas: commercial resource extraction (e.g., logging and mining), non-commercial resource extraction (e.g., hunting, fishing, plant and animal collecting), recreation (e.g., camping, hiking, birdwatching, canoeing), and settlement (e.g., development and agricultural activities). Roads provide primary access for resource extraction activities that often negatively impact wildlife through human disturbance and habitat destruction. For example, clearcut harvesting requires extensive road systems that increase vulnerability of carnivores, such as black bears to hunters (Brody and Stone 1987). Logging truck traffic was shown to reduce presence of female adult grizzly bears Ursus horribilis as far as 200 m from roads (Archibald et al. 1987). Clearcut logging activity can also eliminate potential nesting sites for cavity-nesting birds by removing or knocking down snags. Dead snags of various age or basal area are required to satisfy specific nest requirements for flickers Colaptes auratus, pileated Dryocopus pileatus, downy Dendrocopos plibescens, and hairy woodpeckers D. villosus (Conner et al. 1975). Nellemann and Cameron (1998) studied aversion of Caribou Rangifer tarandus to an oil-field complex near Prudoe Bay, Alaska. Their results showed that females and calves are highly sensitive to surface development (particularly initial construction of roads and related facilities). Avoidance greatly exceeded the physical footprint of the development. Over-exploitation of wildlife resources is another consequence of increased human access within habitat areas. Extensive road systems were the most important

PAGE 261

245 factor for successful hunting of lynx Lynx lynx in rural areas of Norway and turkeys Mealeagris gallopavo in the Piedmont of Virginia (Sunde 1998, Holbrook and Vaughan 1985). Both dispersing and resident black-bears in the Pisgah Bear Sanctuary of North Carolina were subjected to legal and illegal killing near roads adjacent to the Sanctuary suggesting controlled access is needed along protected areas (Powell et al. 1996). Berish and Diemer (1998) interviewed 98 snake collectors in north Florida and found that 94% searched opportunistically along roads and other human-disturbed sites. Alarmingly, many collectors suggested they would kill rattlesnakes Crotalus spp. regardless of financial gain. Construction of a 12-km forest road was cited as the reason for increased exploitation of lake trout in Michaud Lake (Gunn and Sein 2000). To maintain healthy wildlife populations, the State of Oregon instituted a cooperative road closure program to control human access within conservation areas (Hyder 1982). Successful agreements gained acceptance by users because of sufficient public education, well-maintained signage, and law enforcement presence (Hyder 1982). Elk and deer hunters (75 to 95 % favored road closures) discovered after the first few years that reduced hunting pressure resulted in greater hunting success (0.31road closures, 0.04roads open). Other benefits included reduced maintenance costs, less damage from traffic, and healthier wildlife populations. In 1982, Oregon operated 52 road closures covering 764,414 ha (Stein 1982). Access provides opportunities for recreational activities within wildlands, yet negative ecological consequences often result. Campgrounds located in or near carnivore habitat areas are intrusions into otherwise native habitat that attract bears, coyotes, foxes, etc. in search of food, often resulting in human-wildlife conflicts (Creachbaum et al.

PAGE 262

246 1998). Mace and Waller (1996) described several factors designed to reduce human-bear conflicts (visitor reductions, strategic trail placement, public education, and negative conditioning of grizzly bears). Road closings and mass transit provision have been used to reduce wildlife disturbance and habitat destruction from overcrowding problems in Yosemite (Ritter 1998) and Denali national parks (Brown 1978). Trail systems used for ecotourism, recreation, and research can also change ecological systems (e.g., increased erosion, light penetration, and disturbance from humans and domestic predators). Human traffic caused activity shifts or avoidance by sensitive species near trails, and increased numbers of unaffected or habituated species (Griffiths and van Schaik 1993). Two factors are important to understand how access for human settlement practices can degrade habitat: landscape fragmentation (discussed previously), and habitat alteration near buildings and roads. In Summit County, Colorado, Theobald et al. (1997) found that pattern of development was a stronger indicator of habitat disturbance than building density; demonstrating that traditional village planning (e.g., cluster development characterized by condensed, development areas and large, uninterrupted, open spaces) is superior to contemporary suburban development (e.g., scattered, low-density developments on individually owned, large lots) in reducing the negative impacts on wildlife habitat (Arendt 1994). Rossell and Litvaitis (1994) used harvest data for black bears and human demographic variables from 1961 to 1984 to explain the effect of urbanization on black-bear presence in townships of New Hampshire. Black-bear presence was negatively correlated to the increase in human population density and development intensity; black-bear harvest rates were positively correlated to the amount of agricultural land and the

PAGE 263

247 increase in forest road density (an indication of increased observability and human access). In Yellowstone National Park, grizzly bears exhibited reduced occupancy and inefficient foraging habits near human facilities, and displacement of subordinates by dominant individuals into inferior habitat nearer development (Mattson 1987). Roads and development within and surrounding the Ocala National Forest (ONF) threaten area-sensitive interior-species and those species requiring large uninterrupted territories (Figure 6-3). There are approximately 39,160 ha of privately owned inholdings in the ONF (this represents 19% of the total area). These inholdings create access points for the spread of common weedy species. Harris and Silva-Lopez (1992) estimated that as many as 200 species of exotic plants now inhabit the ONF, effectively changing the composition and diversity of species within the forest. Property development within forest boundaries limits the use of prescribed burning that could control the proliferation of these intrusive plants. Reduced frequency of fire in the ecosystem increases the spread of oaks that favor species such as gray squirrel Sciurus carolinensis, but reduce habitat suitability for endemic and endangered red-cockaded woodpeckers Picoides borealis (Harris and Silva-Lopez 1992). Hannah (1992) examined the effects that presence of roads and edge habitat had on habitat core areas in Ocala National Forest. Estimated penetration of edge effects into forest-interior ranged as high as 600 m. Edge habitat represented 35% of roadless forest patches, of which 30% was attributed to the presence of roads. Road-kills and human depredation of gopher tortoise has been facilitated by presence of roads and increased access (Hannah, 1992). Comparatively, in California, desert tortoises Gopherus agassizii were reduced by 60% as far as 1.6 km from the highway (Fusari 1982). Increased access

PAGE 264

248to interior areas from adjacent forest roads is also suspected as the cause for increased presence of edge-dominant predators (e.g., raccoon, opossum Didelphis virginiana, gray fox Urocyon cinereoargenteus, bobcat Lynx rufus, and skunks) (Hannah, 1992). Lastly, high road density (1.84 km/km2) increases forest-wide human presence for hunting and recreation activities and shrinks core refuge areas for large wildlife (Figure 6-3). Figure 6-3. Greater Ocala National Forest ecosystem. Human impacts on this conservation area are pervasive. Because of the intensity of use, it did not qualify in the recent deliberations to establish designated roadless areas in National Forests by the U.S. Forest Service. There are 640 km of state and county highways and 3,295 km of forest roads and trails. Over 39,350 ha of private inholdings exist either as developed areas, agriculture, military installations or vacant areas. Of these, 25% (9,663 ha) are designated for purchase as additions to public conservation areas.

PAGE 265

APPENDIX A QUESTIONNAIRE REGARDING WILDLIFE MORTALITY ON ROADS A survey was conducted at the FDOT-sponsored Transportation-Related Wildlife Mortality Seminar, May 1, 1996, in Orlando, Florida. The questionnaire consisted of two parts: 12 general questions concerning wildlife and roads, and 15 specific questions concerning selection criteria for prioritizing sites for wildlife underpasses. A 5-point Likert response format (strongly agree to strongly disagree) was used for the questionnaire. Significance of responses was evaluated according to demographic and subject groupings. Frequency distribution and statistical analysis of responses was performed using 2-way contingency tables, Chi-square, and Fishers Exact Test (ET) in SAS Analyst (SAS Institute, Cary, N.C.). Fishers ET was most applicable because of small sample size. The demographic makeup of the 65 respondents at the seminar is listed in Table A-1. Table A-1. Respondent demographics Organization/ agency type % Geographic location % Job responsibility % Area of expertise % Transportation 32 Florida 62 Research/ extension/ education 49 Biological/ ecological sciences 82 Private consultant 29 Other* 38 Regulatory enforcement/ impact assessment/ policy/compliance 26 Engineering/ transportation/ planning 18 Natural resource/ academic 39 Planning/design/engineering/ management 25 *A combined group from U.S., Canadian, and International origin. 249

PAGE 266

250 Wildlife and Transportation Issues General questions were grouped according to four factors (road density, road right-of-way characteristics, traffic, and adjacent habitat/land-use characteristics) that affect permeability of roads for wildlife movement and wildlife persistence in areas where roads are present. General Trends Seventy-five percent of respondents (n = 65) agreed or strongly agreed that road density was useful as a predictor of probable number of adverse road-wildlife interactions. It was generally agreed that aversion to roads (74%, n = 48 of 65) and roadside habitat value (63%, n = 41 of 65) were a function of road size (road right-of-way width). Based on evidence from previous studies, respondents agreed that wildlife-vehicle collisions were a function of traffic speed (67%, n = 44 of 65) and that aversion to roads was a function of traffic volume (74%, n = 48 of 65). For certain species (e.g., herpetofauna) that mobility is characteristically slow and seem unaffected by traffic intensity or road size, survey takers agreed that road width and traffic volume would be useful indicators of road-kill rates (72%, n = 47 of 65). Other general questions addressed the importance of wildlife preserves, urban woodlands, and linear habitat connectors (such as topographic ridges, riparian corridors, roadside strips, hedgerows, and fence-lines). Within urban-suburban and open-agricultural landscapes, the audience agreed that linear-habitat features (such as those listed above) were important concentration points for wildlife movement (74%, n = 48 of 65) and associated road-kills (79%, n = 51 of 65). It was overwhelmingly decided that roads bisecting preserve areas would manifest greater negative impacts on wildlife

PAGE 267

251 than roads bisecting human-dominated landscapes (90%, n = 59 of 65); and that open space areas within human-dominated landscapes, whether woodland or pasture, would exhibit more frequent road-kills (81.5%, n = 53 of 65). Response by Demographic Variables Statistical tests were performed on responses to general questions by demographic variables (Table A-1). Response by job responsibility was significant (Likelihood Ratio 2 = 7.84, p < .10; Fishers ET, p < .08) concerning certain species aversion to roads as a function of traffic volume. More negative responses came from those in regulatory (agree%, no opinion%, disagree%) and management (agree%, no opinion13%, disagree%) professions than academic professions (agree%, no opinion13%, disagree%). Job responsibility also was a factor regarding whether more structured habitats would be points of concentration for negative road-wildlife interactions in rangeor grassland species (2 = 9.45, p = .05; Fishers ET, p = .08). In this case, the data reveals the absence of negative responses from those in regulatory (agree%, no opinion%) and academic professions (agree%, no opinion%). Field of expertise was important pertaining to answers to questions no.3 and no. 9 (Exhibit A), respectively, that roadside habitat value was considered a function of road width (2 = 5.26, p = .07; Fishers ET, p = .08); and that urban wildlife would experience higher road-kill rates in urban open spaces than in adjacent developed areas (2 = 10.46, p = .01; Fishers ET, p = .01). Differences in responses to the former question revealed that those with planning or engineering backgrounds, 92% agreed; whereas those with science backgrounds, 56% agreed. Regarding the latter question, difference among

PAGE 268

252 respondents was reversed; 50% of planners and engineers agreed and 89% of scientists agreed. Wildlife Mitigation Priorities on Roads Questions concerning selection criteria, for prioritizing ecological hotspots on roads, focused on 5 specific issues: land ownership, road right-of-way and pavement width, site identification methods (e.g., field studies vs. GIS modeling), importance of preserve size, and road project programming. General Trends Four questions addressed importance of land ownership as a priority for mitigation. Respondents were asked to give priority to: 1) existing, public, wildlife reserves over private-lands or potential linkages between public, wildlife reserves; 2) critical road-kill sites, regardless of adjacent land-ownership, than to roads only in public, wildlife reserves; 3) retrofitting existing substandard structures through habitat corridors (e.g., riparian systems), regardless of adjacent land-ownership, over constructing new bridge-like structures only in existing, public, wildlife reserves (that are not associated with landscape function other than wildlife movement); and 4) retrofitting existing substandard structures through habitat corridors (e.g., riparian systems), regardless of adjacent land-ownership, over constructing new bridge-like structures in privately owned areas of high conservation value (that are not associated with landscape function other than wildlife movement). Responses to these questions were mostly contradictory. Only no. 2 above received predominantly positive responses (71% strongly agree or agree, 14% no opinion, 15% disagree). The other three questions received fairly equal positive and

PAGE 269

253 negative responses, the average reaction being no opinion. The language and length of the questions likely were too confusing, and resulted in poor performance (Exhibit A, Questions 13 16, and next section, Question 16). Two questions dealt with road size (number of lanes), asking survey takers to select from two opposing ideas: two-lane roads that produce greater numbers of road-kills; or four-lane roads that act as habitat boundaries, creating aversion zones for certain wildlife. Again responses were mixed, the average answer being no opinion. Responses favoring priority for two-lane roads averaged 44.5%, whereas responses favoring priority for four-lane roads averaged 25.5%. In this case, the simplistic nature of the questions (Exhibit A, Questions 17 and 18) may not have been appropriate to accurately gage opinion on the issue. A case-by-case approach is probably required to implement mitigation for these conditions. Choosing appropriate methods of analysis for setting mitigation priorities was the subject of three criteria-related questions. Respondents were asked questions regarding GIS modeling and remote-sensing analyses that assemble indirect evidence, and field studies that gather direct evidence of wildlife phenomena. The first question stated that data for spatial modeling must only be used in conjunction with case-by-case on-site ground-truthing for siting underpass locations. Most respondents (82%) strongly agreed or agreed with this statement. The second question asked whether GIS analysis models were inadequate and diversionary from more direct methodologies for determining sites for underpasses. Responses were generally negative (12.5% agreed or strongly agreed, 26% no opinion, 61.5% disagreed or strongly disagreed). The third question stated that radio-telemetry data or other direct evidence of species movement were the only effective

PAGE 270

254 methods for determining locations for wildlife underpasses. Answers were slanted against this statement (20% agreed or strongly agreed, 11% no opinion, 69% disagreed or strongly disagreed). Respondents generally had no opinion (11%), or disagreed or strongly disagreed (54%) regarding the irrelevance of reserve size as a criterion for determining locations for underpasses. Three questions addressed the importance of FDOT work-plans in prioritizing mitigation sites for underpass construction. Attendees were asked to respond to the following: 1) should a road be excluded from consideration, if it is not included in an area already scheduled for road improvements; 2) should the priority of a road be lower, if it is not included in an area already scheduled for road improvements; and 3) should mitigation priorities (determined by ecological criteria) be ranked without regard to their inclusion/exclusion in FDOT work-plans? Responses favored determining mitigation priorities based on ecological criteria rather than dependence upon inclusion in FDOT work-plans (Exhibit A, Questions 23 to 25). Response by Demographic Variables Statistical tests were performed on responses to questions regarding mitigation priorities by demographic variables (Table A-1). One demographic factor was significant regarding emphasis on priority to retrofit existing, substandard structures through habitat corridors (e.g., riparian systems), regardless of adjacent land-ownership, over construction of new bridge-like structures in privately owned areas of high conservation value (that are not associated with landscape function other than wildlife movement). According to geographic location (Likelihood Ratio 2 = 13.03, p < .002; Fishers ET, p <

PAGE 271

255 road improvements (Likelihood Ratio = 11.03, p < .03; Fishers ET, p < .04). In this .002) 55% of Floridians agreed whereas the majority of those from outside Florida (48%) disagreed. Three demographic factors were significant concerning whether mitigation priorities determined by ecological criteria should be ranked without regard to their inclusion/exclusion in FDOT work-plans. With reference to organization type (Likelihood Ratio 2 = 11.24, p < .03; Fishers ET, p < .05), more negative responses came from transportation agency personnel (52% agree, 19% no opinion, 29% disagree) than from consultants (84% agree, 16% no opinion) or resource organization personnel (68% agree, 16% no opinion, 16% disagree). Responses by geographic location (Likelihood Ratio 2 = 14.06, p < .001; Fishers ET, p < .003) revealed that 22.5% of Floridians disagreed, but non-Floridians either agreed (68%) or had no opinion (32%). Regarding field of expertise, more positive responses came from individuals with scientific backgrounds (74% agree, 17% no opinion, 9% disagree) than from those with planning or engineering backgrounds (50% agree, 17% no opinion, 33% disagree). Response by job responsibility was significant (Likelihood Ratio 2 = 15.71, p < .004; Fishers ET, p < .006) for setting higher priority to two-lane roads (that produce greater numbers of road-kills) than four-lane roads (that act as habitat boundaries, creating aversion zones for certain wildlife). More negative responses came from academic professions (31% agree, 25% no opinion, 44% disagree) than from regulatory (59% agree, 35% no opinion, 6% disagree) or management (75% agree, 6% no opinion, 19% disagree) professions. Job responsibility also was a factor regarding whether the priority of a road should be lower, if it is not included in an area already scheduled for 2

PAGE 272

256 case, the data reveals that while regulatory professions (24% agree, 12% no opinion, 65% disagree) gave predominantly negative responses, management/engineering (44% agree, 6% no opinion, 50% disagree) and academic (34% agree, 38% no opinion 28% disagree) professions did not lean either way. Habitat-Corridor Planning and Criteria for Setting Mitigation Priorities Two questions sought opinions on real-situation habitat-corridor planning, and identification and prioritization of selection criteria for GIS modeling of ecological hotspots on roads. Applied Habitat-Corridor Planning One specific question asked the conference attendees to respond to a real-life scenario designing a functional habitat-corridor between two existing, public, conservation areas (Exhibit A, Question 27). Answers to the question did not produce statistically significant results, although 60% of all respondents agreed that the third alternative was the most logical choice. As mentioned previously for other questions, the issue was more complex than the way it was portrayed in the question. As such, the responses simply provide expert opinion on logical use of existing information for generating methods to determine priorities for mitigation. Selection Criteria for Modeling Ecological Hotspots on Highways Respondents were asked to list and rank specific criteria for use in a GIS model that identifies and prioritizes sites for underpass construction (Exhibit A, Question 27). Answers to this question were analyzed to determine if any trends for or against any particular criteria were present according to demographic groups. Frequency procedure

PAGE 273

257 11 Road projects Road projects was performed in SAS, however sample size was too small for a chi-square test; therefore Fishers Exact Test (ET) was used. Analysis by organization type provided significant findings regarding rankings of four criteria. The criterion, road-kill, was ranked first and the criterion, road projects, was ranked last for all organization types. Significant differences were found among choices between organizational groups regarding SHCAs (Fishers ET, p < .08) and greenways (Fishers ET, p < .09). Whereas the resource management group selected SHCAs 4th most frequently (24%), transportation and consultant groups chose SHCAs 8th (24%) and 7th (32%) most often, respectively. Transportation and consultant groups ranked greenways most frequently 4th (24%) and 5th/6th (41%), respectively; and resource management organizations chose 2nd (28%) most often. No other trends were found for the other seven criteria that shifted rankings frequently, according to organizational type. An analysis of response, by geographic area of work, revealed no significant variation between groups. However, most criteria ranked the same or within one ranking of each other for both groups, with the exception of listed species, greenways, and separate resources. These results are shown in Table A-2. Table A-2. Average response by geographic area Rank Florida Other 1 Roadkill Roadkill 2 Migration route Greenway 3 Hotspots Migration route 4 Listed species Hotspots 5 Greenway SHCA 6 SHCA Separate resources 7 Riparian system Riparian system 8 Core area Core area 9 Separate resources Listed species 10 Public land Public land

PAGE 274

258 An analysis of response by area of expertise also revealed similar responses among groups. All criteria were ranked the same or within one ranking of each other for these two groups, with the exception of two criteria, greenways and SHCAs. These results are shown in Table A-3. Significant differences between these groups were found regarding ranking of criteria, road projects (Fishers ET, p < .09) and separate resources (Fishers ET, p < .07). For road projects, those with biology/ecology backgrounds chose 3rd to 9th in 20 out of 53 opportunities, while those with engineering/transportation/planning backgrounds chose 3rd to 9th in 2 out of 12 opportunities. Concerning separate resources, many different choices between groups were evident. Of particular note were lack of choices for 1st, and 4th to 7th (total of 8%) for the engineering/transportation/planning group, whereas the biology/ecology group selected these ranks 29 out of 53 times (54%). Opposite selections were also the case for rankings 2nd and 8th, as biologists/ecologists selected these 6 out of 53 times (11%); but engineers/planners chose these ranks 7 out of 12 times (58%). Table A-3. Average response by area of expertise Rank Biology/ecology Engineering/transportation/planning 1 Roadkill Migration route 2 Migration route Roadkill 3 Greenway Hotspots 4 Hotspots SHCA 5 Listed species Listed species 6 SHCA Greenway 7 Riparian system Separate resources 8 Separate resources Riparian system 9 Core area Core area 10 Public land Public land 11 Road projects Road projects Analysis according to job responsibility revealed significant differences for three criteria, road-kill (Fishers ET, p = .005), hotspots (Fishers ET, p = .09), and core areas

PAGE 275

259 (Fishers ET, p = .07). Road-kill was selected predominantly 1st or 2nd by regulatory (59%) and academic (70%) professions, yet management/engineering professions selected it 1st or 2nd only 6 out of 16 times (37%). Academic and management/engineering professions chose hotspots 1st to 3rd most often (69% and 50%, respectively) while those in regulatory professions picked it 7th to 11th (59%) most often. Median ranks for core areas by professional groups were 4th (academic), 10th (regulatory), and 9th (management/engineering). Public lands and road projects were ranked 10th and 11th, respectively by all but one group, planning/engineering/design/management. The other nine criteria shifted rankings frequently, according to job types. Demographic analysis revealed little significant bias between groups; in fact results showed remarkable consistency in rankings for road-kill (1), migration (2), public lands (10), and road projects (11). This shows strong preference for the first two criteria as primary indicators of wildlife movement travel routes (locations where construction of wildlife underpasses should receive top priority). In reference to the latter two criteria, the survey indicates that their significance in determining location of underpasses should be minimized. The other criteria exhibited more variability among demographic groups, and their significance is less clear. Summary of Results from the Questionnaire First, for general questions on wildlife-road issues, respondents acknowledged that: road density was a predictor of probable number of adverse road-wildlife interactions; aversion to roads and roadside habitat value were a function of road size; wildlife-vehicle collisions were a function of traffic speed; and aversion to roads was a function of traffic volume. It was also agreed that linear habitat features (e.g., riparian

PAGE 276

260 corridors) in agricultural areas, urban woodlands, and ecological preserves were important concentration points for wildlife movement, and therefore would exhibit more frequent road-kills. Statistical differences were evident for job responsibility and field of expertise response groups. Second, regarding specific questions focused on determining mitigation priorities, respondents favored giving priority to critical road-kill sites, regardless of adjacent land ownership. They also strongly favored the use of GIS technology and data for spatial models (to identify sites for underpasses) in conjunction with on-site ground-truthing. Lastly, the audience agreed that mitigation priorities should be based on ecological criteria, rather than dependence upon inclusion in FDOT work-plans. Third, an experts list of criteria was generated that could be used to develop a modeling protocol for prioritizing ecological hotspots on highways (Table A-5). Table A-5. Rankings of road prioritization criteria by expert opinion Rank Criterion 1 Chronic road-kill sites 2 Known migration/movement routes 3 Identified hotspots of focal species 4 Landscape linkages (designated greenways) 5 Presence of listed species 6 Identified strategic habitat conservation areas 7 Riparian corridors (with potential for retrofitting existing structures) 8 Core conservation areas 9 Presence of separate-required ecological resources (e.g., a forest patch and ephemeral wetland breeding area for amphibians that is separated by a highway) for a species or set of species 10 Public ownership (or in public land acquisition program) as opposed to private lands 11 Potential to be included in proposed road improvement project

PAGE 277

261 2. engineering/transportation/planning 0.18 Exhibit A Questionnaire Regarding Wildlife Mortality on Roads Presented at the Florida Department of Transportation Sponsored Transportation-Related Wildlife Mortality Seminar, Orlando, FL., May 1, 1996 Goal 1: To assemble and rank criteria for utilizing available information to describe and analyze landscape patterns, processes, and ecologically based greenways, animal movements and location of likely points of conflicts with existing and future highway locations and linear construction projects. Goal 2: To use these criteria for developing a GIS-compatible algorithm that will process existing spatial data layers and assist with prioritization of highway-greenway interfaces to be addressed by FDOT in district workplans for installation of wildlife underpasses or appropriate corrective measures. INSTRUCTIONS Circle the most appropriate response or give a brief statement as requested. Response percentages are shown in magenta. RESPONDENT DEMOGRAPHICS (n = 65) What type of organization do you work for? 1. transportation agency 0.32 3. private consulting firm 0.29 2. resource management/research 0.39 In what geographic area do you work? (e.g., country, and state or province) 1. Florida 0.62 2. other 0.38 What is your primary responsibility within your organization? (e.g., university professor, transportation engineer, research scientist, etc.) 1. research/extension/education 0.49 3. planning/engineering/ 0.11 2. regulatory enforcement/impact 0.26 design/management assessment/policy/compliance What is your field of expertise? (e.g., game management, transportation corridor planning, etc.) 1. biological/ecological sciences 0.82

PAGE 278

262 GENERAL QUESTIONS 1. Given that road density is a known factor of consequence to wide ranging species (e.g., wolf, elk), it follows that a simple measure of road density could be used to predict the probable number of negative road-wildlife interactions. a. strongly agree 0.09 d. disagree 0.15 b. agree 0.66 e. strongly disagree 0.02 c. no opinion 0.08 2. Given that road right-of-way width is commonly a function or derivation of road size and given that some animals (e.g., adult female elk) apparently manifest aversion to roads of any size, it follows that aversion to roads should be some function of road size. a. strongly agree 0.14 d. disagree 0.09 b. agree 0.60 e. strongly disagree 0.02 c. no opinion 0.15 3. Given that road right-of-way width is commonly a function or derivation of road size and given that some animals (e.g., white-tailed deer, love bugs) commonly manifest an attraction to managed rights-of-way, it follows that roadside habitat values should be some function of road size. a. strongly agree 0.09 d. disagree 0.26 b. agree 0.54 e. strongly disagree 0.02 c. no opinion 0.09 4. Roads in-and-of-themselves do not kill wildlife, it is usually a function of traffic. It was shown in Nebraska that wildlife road-kills are a direct function of average traffic speed (Case, 1978). Can we therefore conclude that road-kill is a function of traffic speed? a. strongly agree 0.15 d. disagree 0.25 b. agree 0.52 e. strongly disagree 0.05 c. no opinion 0.03 5. Clearly some species seem to manifest an aversion to roads, even small unpaved roads (e.g., female elk, wolves). Given that some species manifest an aversion to traffic, the logical inference is that aversion could also be a function of traffic load. a. strongly agree 0.17 d. disagree 0.12 b. agree 0.57 e. strongly disagree 0.02 c. no opinion 0.12

PAGE 279

263 6. Given that some wildlife species develop more of an aversion as roads become larger and/or carry heavier traffic loads (e.g., black bearM. Pelton), it follows logically that road size and/or traffic load is not a predictor of negative road mortality for some species. a. strongly agree 0.09 d. disagree 0.17 b. agree 0.34 e. strongly disagree 0.02 c. no opinion 0.38 7. Many species seem oblivious to size and/or traffic load (e.g., snakesPaynes Prairie, FL), therefore it follows that large roads (which have the effect of increasing the time necessary to cross the road) and/or roads with heavy traffic loads would be useful predictors of kill rate for this type of species. a. strongly agree 0.20 d. disagree 0.14 b. agree 0.52 e. strongly disagree 0.00 c. no opinion 0.14 8. Given that wildlife preserves or sanctuaries maintain larger populations of many species as compared to human-dominated landscapes, it follows that a road bisecting a preserve would manifest greater negative road-wildlife interactions than a road passing through a human-developed landscape. a. strongly agree 0.43 d. disagree 0.08 b. agree 0.47 e. strongly disagree 0.00 c. no opinion 0.02 9. Given that human-developed landscapes commonly contain open spaces such as pastures or woodlands and that wildlife commonly use these open spaces in preference to more highly developed urban and suburban areas, it follows that road-kills would be more frequent in these lesser-developed areas than in the nearby highly developed areas. a. strongly agree 0.245 d. disagree 0.14 b. agree 0.57 e. strongly disagree 0.00 c. no opinion 0.045 10. Given that human-dominated landscapes commonly contain open spaces such as ridgetops and riparian systems, and that wildlife commonly travel along these open spaces in preference to more highly developed urban and suburban areas, it follows that road-kills would be more frequent in these areas than in the nearby highly developed areas. a. strongly agree 0.28 d. disagree 0.12 b. agree 0.51 e. strongly disagree 0.00 c. no opinion 0.09

PAGE 280

264 11. In areas where little topographic relief exists, long distance wildlife movement occurs primarily along continuous linear structures such as riparian corridors, roadside strips, hedgerows or fencelines rather than across discontinuous open spaces (patches) that contain no cover and/or significant barriers or boundaries to movement. a. strongly agree 0.20 d. disagree 0.08 b. agree 0.54 e. strongly disagree 0.00 c. no opinion 0.18 12. Given that even open-land species commonly disperse and/or move the longest distances along the more densely structured available habitat, it follows that more structured habitats such as hedgerows and fencelines when intersected with highways would probably be points of negative road-wildlife interactions. a. strongly agree 0.15 d. disagree 0.03 b. agree 0.60 e. strongly disagree 0.00 c. no opinion 0.22 SELECTION CRITERIA 13. When considering factors that determine where wildlife underpasses should be constructed, it would be more important to address roads that bisect existing public wildlife reserves or sanctuaries than privately owned areas or potential linkage areas between existing public wildlife preserves or sanctuaries. a. strongly agree 0.17 d. disagree 0.36 b. agree 0.32 e. strongly disagree 0.05 c. no opinion 0.10 14. When considering factors that determine where wildlife underpasses should be constructed, it would be more important to address roads identified as critical road-kill sites for listed species (regardless of adjacent land-ownership) than roads that bisect or are adjacent to existing public wildlife reserves or sanctuaries. a. strongly agree 0.14 d. disagree 0.15 b. agree 0.57 e. strongly disagree 0.00 c. no opinion 0.14

PAGE 281

265 15. When considering factors that determine where wildlife underpasses on roads are constructed, it would be more important to retro-fit existing structures (e.g., bridges over riparian systems) with a functional terrestrial connection (regardless of adjacent land-ownership), than to construct new bridge-like structures that are unassociated with any functions other than provision of safe wildlife movement in public wildlife reserves or sanctuaries a. strongly agree 0.08 d. disagree 0.34 b. agree 0.36 e. strongly disagree 0.02 c. no opinion 0.20 16. When considering factors that determine where wildlife underpasses on roads should be constructed, it would be more important to retro-fit existing structures (e.g., bridges over riparian systems) with a functional terrestrial connection (regardless of adjacent land-ownership), than to construct new bridge-like structures that are unassociated with any functions other than provision of safe wildlife movement in privately owned areas (known to contain significant wildlife populations). a. strongly agree 0.06 d. disagree 0.22 b. agree 0.43 e. strongly disagree 0.03 c. no opinion 0.26 17. It has been discovered that in many situations two-lane roads with high traffic loading produce the greatest number of road-kills. It follows, therefore, that these roads should be addressed first when prioritizing sites for underpass construction rather than modifying a four-lane highway that may be creating it's own aversion zone (e.g., black bear, M. Pelton). a. strongly agree 0.09 d. disagree 0.28 b. agree 0.40 e. strongly disagree 0.00 c. no opinion 0.23 18. Even though it has been shown, in many situations, that two-lane roads (with high traffic volume) produce greater numbers of road-kills than certain four-lane highways, such four-lane highways (which act as severe barriers to dispersal) should be addressed first with concern to installing underpasses. a. strongly agree 0.06 d. disagree 0.40 b. agree 0.17 e. strongly disagree 0.015 c. no opinion 0.355

PAGE 282

266 19. Data layers used for spatial modeling (i.e., habitat conservation areas, hydrology, topography, land ownership, wildlife hotspots or listed-species locations) must only be used in conjunction with case by case on-site ground truthing for siting locations of underpasses. a. strongly agree 0.30 d. disagree 0.08 b. agree 0.52 e. strongly disagree 0.02 c. no opinion 0.08 20. GIS analysis models that use data layers such as habitat conservation areas, hydrology, topography, land ownership, wildlife hotspots or listed-species locations are not only inadequate but diversionary from more expedient methodologies for determining where wildlife underpasses should be located. a. strongly agree 0.02 d. disagree 0.51 b. agree 0.105 e. strongly disagree 0.105 c. no opinion 0.26 21. Radio-telemetry or other direct evidence of species movement are the only effective and realistic methods for determining where underpasses should be located to facilitate safe wildlife movement. a. strongly agree 0.03 d. disagree 0.66 b. agree 0.17 e. strongly disagree 0.03 c. no opinion 0.11 22. Size of a wildlife conservation area (e.g., 4,000 acres vs. 400,000 acres) is irrelevant for determining whether underpasses to facilitate safe wildlife movement should be constructed on roads that bisect or surround them. a. strongly agree 0.06 d. disagree 0.46 b. agree 0.29 e. strongly disagree 0.08 c. no opinion 0.11 23. Roads in areas that qualify for installation of a wildlife underpass (i.e., have high priority rating based on ecological criteria) should be excluded from consideration if the road is not included in the FDOT workplan for other road work such as new construction or existing road improvements. a. strongly agree 0.015 d. disagree 0.615 b. agree 0.03 e. strongly disagree 0.28 c. no opinion 0.06

PAGE 283

267 24. As opposed to the above question, rather than over-ruling the need for a wildlife underpass on a road that is not in the FDOT workplan, it should simply be given a lower priority, so that those underpass locations that are included in other FDOT road construction projects will receive a higher priority. a. strongly agree 0.03 d. disagree 0.31 b. agree 0.31 e. strongly disagree 0.12 c. no opinion 0.23 25. Roads in areas that qualify for installation of a wildlife underpass based on ecological criteria should be ranked without regard to whether they would be included in the FDOT workplan for other road work such as new construction, or existing road improvements. a. strongly agree 0.28 d. disagree 0.14 b. agree 0.41 e. strongly disagree 0.00 c. no opinion 0.17 26. A list of criteria for prioritizing the location of underpasses on Florida roads in order to alleviate road-kills and to provide ecological linkages might include the following (Please rank by number,2,3,etc. and add any additional criteria not included that you think is important): Avg. Rank Score Criteria 8 4.65 core conservation areas (_________ acres minimum) 1 2.58 existing site of chronic road-kills of listed species 6 4.48 existence of listed species in area 10 5.94 public ownership (or in public land acquisition program) as opposed to private lands 4 3.80 landscape linkages (designated greenways) 7 4.60 existing riparian corridors (with potential for retrofitting existing structure) 5 4.42 identified strategic habitat conservation area 3 3.60 identified hotspots of listed species 11 6.26 potential to be included in proposed road improvement project 2 3.49 existence of known migration route 9 4.69 existence of separate required ecological resources (e.g., ephemeral pond for breeding amphibians) for a species or set of species

PAGE 284

268 27. This question concerns a real-life scenario from north Florida (a segment of a linkage that would connect Ocala National Forest to Osceola National Forest): say 5 publicly owned conservation areas exist in close proximity (100 km) one is a very large wildlife refuge, (300,000 ha) (by eastern standards) a). one linkage scenario based on investment in existing public lands would suggest a linkage trajectory to the northeast. b). a second linkage scenario based on large land parcels (private ownership) including a privately owned wildlife management area, would project a northwesterly trajectory. c). a third linkage scenario was found by looking at topographic relief and hydrologic features, and it was noted that a very small gap (approx. 2 km) existed between a north flowing and a south flowing pair of streams, and this scenario would lead due north. It follows from wildlife landscape ecology that the third scenario would be more beneficial for biodiversity conservation in a rapidly changing Florida environment than the other two which only reflect human land-use and /or real estate considerations. a. strongly agree 0.09 d. disagree 0.06 b. agree 0.51 e. strongly disagree 0.02 c. no opinion 0.32 ** Note: If you have any literature that strongly supports or ** detracts from any of these statements, please list citation. Prepared and Conducted by Daniel J. Smith, Graduate Assistant Department of Wildlife Ecology and Conservation, University of Florida

PAGE 285

APPENDIX B INTERACTIVE MODEL FOR IDENTIFYING AND PRIORITIZING ECOLOGICAL HOTSPOTS ON HIGHWAYS Introduction This research included the use of Geographic Information Systems (GIS) analysis to create a model/algorithm that employs existing datasets to identify ecological hotspots on highways and evaluate need for wildlife crossing structures or underpasses (Smith et al. 1998). These hotspots could then be prioritized and ranked for use in FDOT district work-plans, in coordination with new construction and improvement projects. The process would utilize existing biological and physical parameters such as species locations, habitat, hydrology, topography, and road coverages (Smith et al. 1996). Knowledge-based and Rule-based Modeling Applications and GIS The GIS application is based on concepts used in knowledge-based and rule-based models. Knowledge-based or decision-support models apply expert opinion, scientific literature and empirical data to geographic datasets to make informed, accurate decisions, to achieve specific goals or objectives (Clevenger and Wierzchowski 2001, Maurer 1999, Carr et al. 1999, and Aspinall 1994). Rule-based models utilize data overlays to develop suitability surfaces and apply user-defined weights or scores to set priorities (Aspinall 1994, and Mcharg 1969). Variations of these models have been used for several applications: habitat suitability and population viability analysis (Akcakaya 2000, and Clark and Van Manen 1992), determining locations and priorities for wildlife 269 underpasses on highways (Clevenger and Wierzchowski 2001, and Smith et al. 1998),

PAGE 286

270 creating ecological networks (Carr et al. 1999, 1996), and environmental sensitivity assessment in transportation planning (Maurer 1999, and Herzer et al. 1991). Geographic Information Systems analysis has increased the ability and ease of performing spatial analyses at previously prohibitive large scales, such as landscape and regional levels. Use of raster or cell-based models in GIS allows application of mathematical algorithms to multiple layers of geographic data, whereby individual cell locations are evaluated based on attribute information of multiple data layers (to perform environmental suitability and sensitivity analyses). This appendix discusses development and capabilities of an independent GIS application utilizing a sample dataset, whereas, Chapter 2 discusses the actual state-level analysis presented to FDOT in 1998 (including suggested data improvements and application upgrades). Model Parameters To provide FDOT the capability to identify and prioritize important ecological hotspots on highways, a GIS application was developed using Arcview Avenue scripts (Environmental Systems Research Institute, Inc., Redlands, California) to perform the following functions: 1) utilize an assembled and formatted set of existing raster data layers (scale independent), 2) establish a utility for setting appropriate analysis extent, 3) design a utility for inputting user-defined criteria weights, 4) perform prioritization analysis based on user-provided data and weightings, 5) provide for online cursory evaluation of results, and 6) establish a platform to view spatial relationships of associated field verification data.

PAGE 287

271 This application functions as an Arcview project that is located in a common directory with associated data files and Avenue scripts. A detailed set of instructions automatically pops up when the project file is opened. The instructions text-file provides details on hardware and software requirements, and information to guide users through the prioritization process and use of the image and data viewer. The instructions.txt file can be viewed as the opening window in the project or viewed with any text viewer, prior to opening the project file in Arcview. The project includes three separate views: the first displays background data, the second displays raster data used for the prioritization process, and the third contains field data for use with the image and data viewer utility. Data Format and Assembly The first step was selection of appropriate data, second it was necessary to assign base values to attribute information in the data layer, third similar data were assembled into groups, and fourth the data were formatted so an Avenue script could recognize it. Six types of existing data were found for Florida that could be used to identify ecological hotspots along roads. These included biological features, landscape features, infrastructure improvement programs, land ownership, conservation plans, and road-kill. These data (Table 2-1) correspond to the selected criteria that were identified and ranked in the expert questionnaire discussed previously (Appendix A). Other sources and types of data could be used based on availability, user preferences, and objectives sought. All vector data must be converted to raster format for use in the model and an appropriate cell resolution must be selected. In most cases, it would be desirable to select a cell size based on the highest cell resolution of the existing datasets and the desired

PAGE 288

272scale of analysis (e.g., a regional or large-scale study would be coarser than a site-specific or small-scale study). Each data layer must be formatted according to the relative value of the inherent attribute information contained in the dataset. A description of necessary data format can be explained by the following example: Suppose a data layer provides documented locations of federally listed species (Figure B-1). A simple means of assigning base values for this type of data would be to use a sliding scale of importance (e.g., endangered = 3, threatened = 2, and species of special concern = 1; the highest value representing the most important). Once each dataset was assigned base values, they were grouped into specific categories (Table 2-1). Note that categories are determined by the user and can therefore represent any variation of data and objectives; those shown in Table 2-1 simply represent those applicable to the Florida highway analysis completed in 1998 for FDOT. Datasets Figure B-1. Documented locations of listed species. Florida Natural Areas Inventory element occurrences is a database that includes documented locations of listed s p ecies, bird rookeries, and mi g rator y bird con g re g ation areas ( FNAI 1999 )

PAGE 289

273 for each category were pooled using the combine function in Arcview. This local procedure creates a new dataset that tabulates all possible combinations for each cell location, for all datasets in the group. An excerpt from a combined dataset table illustrates this process (Table B-1). Table B-1. An example of a combined dataset table Value Count Hotspots Fleo_hr Sum 1 191066 202 2 413763 101 3 945034 000 4 5830 213 5 47894 011 6 6693 112 7 104551 303 8 195220 224 9 133868 123 10 502973 022 11 168052 325 12 4219 134 13 33659 033 14 16918 336 15 6160 235 16 5413 314 The new raster layer table contains fields of new values and counts for each combination, and the original values from each of the input layers; in this case, listed species locations, FLEO_HR (Figure B-1), and focal species hotspots, HOTSPOTS (Figure B-2). For the dataset to correctly reflect the hierarchy of base values, a numeric sum column must be added to the raster table. The value in this column is derived by adding the values of each of the original data values (e.g., focal species hotspots, HOTSPOTS, and listed species locations, FLEO_HR), shown in Table B-1. This is the field that is used in calculations by the Avenue program. For the program to recognize each combined dataset, the dataset name in the view must end with comb, an abbreviation of the wordcombination (Figure B-3). The program will only recognize

PAGE 290

274data layers with this suffix. Note that the actual dataset name does not have to be changed, only the name reflected in the active view. Analysis Extent Utility To provide an interactive platform for various users and locations, it was necessary to design an environment that allowed multiple methods for selecting an analysis extent. The analysis extent utility allows the user to select from three methods: using a pre-existing roads dataset, creating a custom rectangular-analysis-area, or creating new alternative road alignments. The utility uses the selected dataset as a mask, and excludes everything outside the selected area from the analysis. Since this process was designed to identify and prioritize ecologically sensitive areas adjacent to roads within major greenways, an additional masking parameter was added that allows the user to Figure B-2. Focal species hotspots dataset for Alachua County, FL. Based on an analysis of 44 species of conservation interest, this dataset displays areas where various numbers of species of conservation interest co-occur (e.g., 2 or less 3-4 5-6 or 7 or more ) ( FFWCC 1994 )

PAGE 291

275select for or against urban land-uses. This option provides the opportunity to narrow the focus by excluding intensively developed areas from the analysis. This utility is illustrated in examples in the application functions summary. Prioritization Algorithm After choosing an analysis extent, the prioritization process begins by selecting the criteria to set priorities, those selected are then assigned weights (multipliers) from 1 to 50. An algorithm within an Avenue script then processes the combined datasets for each category. It sorts the combinations, assigns the selected weightings, multiplies the base values for the cells in each raster layer by the appropriate weighting, and sums the weighted combinations from each raster layer (to develop a single priority layer based on Figure B-3. Biological features combined dataset for Alachua County, FL. This dataset represents the combined data from Figures B-4 and B-5. Note that the dataset name in the legend contains the suffix comb, the call word used by the Avenue scri p t to reco g nize data-la y ers ( used in the p rioritization anal y sis )

PAGE 292

276the assigned weightings of the criteria categories). This process is illustrated in examples in the application functions summary. ination of values and criteria for ta layer has been generated, simply clicking cells button, or selecting this option from user to select any individual cell (Figure generated that displays prioritized value, ted (including the associated value for that cell) (Figure B-5). Figure B-4. Viewing criteria and ranking of individual cells. Values and rankings for prioritized cells can be viewed either, by clicking the option from the menu or the toolb ar and then selectin g individual cells. View criteria but t on Cursory Evaluation of Results A simple tool was added to allow examprioritized raster cells. Once the priority daon the view criteria values and ranking for the Road Priority Analysis menu, allows the B-4). Once selected, a table is automatically count, and each criterion evalua

PAGE 293

277 Viewing Associated Field Verification Data Lastly, the application provides the means to view field data associated with the model results. An image and data viewer was created using the dialog-designer function in Arcview. Clicking on the image and data viewer button, or selecting this item from the Road Priority Analysis menu (Figure B-6), opens a button box that provides options to view aerials, photographs and data for field sites, and obtain help (Figure B-7). Clicking on the help button provides instruction on use of the utility. To activate the other buttons, a point must be selected (Figure B-8). Once activated, clicking on any button in the image and data viewer button box will display the selection (i.e., an image or data table) (Figures B-9 to B-11). The utility allows for up to four field photographs per field site. Figure B-5. Table displaying values at selected points. Once a cell is selected, a pop-up table is displayed that includes information on priority value, and individual values and rankings for each data-layer used to rank the cell.

PAGE 294

278 Figure B-7. Image and data viewerbutton-box. The button-box appears when the image and data viewer is selected. Road priority analysis pull-down menu Figure B-6. Image and data viewer. Images and data from field sites are selected and dis p la y ed b y clickin g the buttonb ar item or selectin g from the p ull-down menu.

PAGE 295

279 Figure B-9. Image and data vieweraerial. Clicking the view aerial button dis p la y s an aerial p hoto that can be ma g nified b y usin g the Arcview ma g nifier tool. Arcview magnifier tools Figure B-8. Site selection for image and data viewer. Sites are selected by clicking the cursor on a chosen point; once selected the button box becomes active. Selected point a pp ears y ellow

PAGE 296

280 Figure B-11. Image and data viewerdata file. Clicking the view data button dis p la y s the data table for the selected p oint. Figure B-10. Image and data viewerphotographs. Clicking the view photos b utton dis p la y s a field ima g e ( maximum of 4 available ) for the selected p oint. Tools for scrolling through field photos

PAGE 297

281 The user can scroll through these by using the forward and back arrows below the view photo button in the image and data viewer button box (Figure B-10). The magnifier function on the Arcview tool-bar is used to enlarge selected images. Application Functions Summary The process described above was assembled into an interactive platform so that users could run the same prioritization process on any independent dataset, at any scale. Several Avenue scripts were developed to perform four primary functions: setting the analysis extent, assigning weights and generating a priority layer, viewing criteria values and rankings for individual cells, and image and data viewing of field-verified sites. The following section provides an example for each of the three different approaches to using the prioritization model. Using an Existing Roads Buffer as an Analysis Mask An existing road coverage can be used as an analysis mask, by buffering the coverage, and converting the buffer to a raster-file (this process is described in general GIS texts). This is the process that was used in Chapter 2 to produce a statewide priority-analysis for Florida roads. This raster-file then can be used as the analysis extent. The first step in the priority process is selection of the analysis extent. The analysis extent is set either, by selecting the option from the pull-down menu, or clicking on the set analysis extent button on the tool-bar (Figure B-12). The analysis extent option-box will open and the existing grid option is selected (Figure B-13). Next, an existing file must be selected from the open theme dialog-box (Figure B-14). The selected analysis extent mask automatically displays in the view (Figure B-15).

PAGE 298

282 Figure B-13. Setting the analysis-extentoptions box. The user can select from three p ossible extents: existin g g rid, drawin g a rectan g le, or new road ali g nment. Figure B-12. Setting the analysis-extent. An analysis-extent is set by clicking the set analysis-extent button on the tool bar or by choosing the option from the p ull-down menu. The set analysis-extent b u tton

PAGE 299

283 Figure B-15. Setting the analysis-extentusing an existing file. The existing file serves as an analysis-extent and a mask, excluding outside areas from the analysis. Figure B-14. Setting the analysis-extentselecting an existing file. The add theme box allows the user to select an existing raster (grid) file.

PAGE 300

284The second step is criteria weighting, found in the Road Priority Analysis pull-down menu. The user is given the choice to include or exclude intensively developed urban-lands (Figure B-16). If no is selected, an urban-area grid that defines these areas as no data is automatically added, excluding them from the analysis. If yes is selected, then the next step in the procedure follows. The third step is selection of criteria to be weighted. Criteria layers are automatically displayed in a dialog-box, after the urban-area option request is processed. The user selects criteria layers from those displayed in the dialog-box (Figure B-17). Once selected, a new dialog-box asks the user to input criteria weights, numeric values from 1 to 50 (Figure B-18). Lastly, a file name must be provided, after which the prioritization algorithm calculates and displays the priority layer (Figure B-19). Figure B-16. Criteria weightingurban areas option. First option in the priority process is to decide whether or not to include intensive urban lands in the analysis. If no, an urban area grid that defines these areas as no data is automaticall y added, and the y are then excluded from anal y sis.

PAGE 301

285 Figure B-18. Criteria weightingassigning weights. The user may enter a multi p lier from 1 to 50 in the blank next to each criterion to reflect its im p ortance. Figure B-17. Criteria weightingselecting criteria. One or more criteria are selected for the p riorit y anal y sis b y usin g the mouse and shift ke y simultaneousl y

PAGE 302

286 Creating a Custom Analysis Extent (changing scales) The priority analysis on roads can be performed at different scales. Suppose that following the countywide analysis conducted on Alachua County (Figure B-19) that a specific site needed evaluation (due to an upcoming road construction project). In this case, State Road 24 (adjacent to Austin Cary Forest) will be used as an example. To narrow the analysis extent, the create rectangle option will be used (Figure B-13). Once this option is selected, the user draws a rectangle (with the mouse) over the area of interest (Figure B-20). A new analysis extent is automatically created and added to the view (Figure B-21). The process also automatically zooms to the new analysis extent. The same process described in the previous example can be used to generate a priority layer (Figure B-22). Figure B-19. Priority analysisfinal priority layer. Once the algorithm calculates the values for each cell, the priority layer automatically displays. The hi g hest values in the le g end and darkest shades re p resent the hi g hest p rio r ities.

PAGE 303

287 Figure B-21. Setting the analysis-extentcustom rectangle zoom-in. Once drawn the rectangle will automatically be shaded and magnified. Figure B-20. Setting the analysis-extentcustom rectangle. Users can draw a custom-sized rectangle that serves as the analysis-extent and mask.

PAGE 304

288 Selecting and Assessing New or Proposed Road Alignments The priority-analysis application provides a method for quick evaluation of alternative road alignments. The once-proposed Tampa to Jacksonville toll-road provides an example of this option (Figure B-23). In the late 1980s, the Tampa to Jacksonville toll-road was proposed as an alternative expressway to the I-95/I-4 corridor. It became a highly controversial project due to potential environmental impacts and routing through existing conservation lands; and was eventually dropped from the Florida Transportation Plan (FTP). Nevertheless, this example shows the value of the application to identify environmentally sensitive areas, within proposed road corridors. Evaluating alternative road alignments requires the user to input analysis masks that represent the proposed routes. Figure B-22. Priority analysissite specific priorities. The priority layer is limited to the customized anal y sis rectan g le and the roads la y er mask.

PAGE 305

289 The first step is to set the analysis extent (Figure B-12) and select the road alignment option (Figure B-13). A pop-up box will appear that directs the user to draw a line that represents the proposed route (Figure B-24). Once drawn, the user clicks on the next button in the pop-up box; and a new pop-up box will open that requires the user to input a buffer distance value (Figure B-25). The Avenue script generates an analysis mask of the line at the specified buffer distance (Figure B-26). Following the priority process outlined previously generates a priority layer (Figure B-27) that can be compared to other alternative alignments for the proposed road. Alternative road alignments can be evaluated simultaneously by following the same steps, drawing the potential route (Figure B-28), buffering it, and running the prioritization analysis (Figure B-29). Figure B-23. New or proposed road alignments. The alternative alignments in Alachua County for the previously proposed Jacksonville to Tampa toll road (that runs through the congested Daytona Beach and Orlando areas).

PAGE 306

290 Figure B-25. Buffering the proposed road alignment. After drawing the p otential road ali g nment the user is asked to enter a bufferin g distance value. Figure B-24. Drawing the proposed road alignment. The utility allows the user to draw a line anywhere on the map to represent a potential road alignment.

PAGE 307

291 Figure B-27. Priority areas in proposed road alignment. The priority layer is limited to the potential road alignment. Darker shades indicate sensitive areas. Figure B-26. Buffered analysis mask. The Avenue script automatically generates an analysis mask at the buffer width submitted for the proposed alignment.

PAGE 308

292 Figure B-29. Alternative proposed road alignment evaluation. Simultaneous evaluation of alternative potential alignments can be performed as shown here. Figure B-28. Adding an alternative road alignment. Alternative alignments are added by repeating the same process. Here another potential road corridor is drawn.

PAGE 309

293 Assessment of Results This application is designed as a planning tool that transportation planners and engineers can use to identify and prioritize ecological hotpsots on roads, based on relative environmental sensitivity. It can also provide preliminary assessments, regarding severity of potential impacts from new or proposed road projects. Besides the cursory evaluation shown in Figures B-7 and B-8, a more extensive evaluation can be performed on each prioritized layer by two different methods; one quantitative and one qualitative. Quantitative assessment could be performed on the input data layers that generated the sensitivity levels of each cell in the priority layer. Since raster-files (grids) are made up of individual cells with the same dimensions, area can be computed from the data table of each input data layer in the analysis. This is calculated by multiplying the count for a particular value (class) by the cell size. For example, suppose it was critical to minimize the impact of the proposed Jacksonville to Tampa toll-road on rare xeric habitats and hardwood hammocks in Alachua County. This can be determined by comparing the number of hectares of rare habitats within the two proposed alignments (Figure B-29). The landscape features-comb grid (Table B-2) contains the data from the input criteria for landscape features, including rare habitats. The values for each field in Table B-2 represent a specific class (e.g., 4 = hardwood hammocks and xeric communities, 3 = wetlands, 2 = pinelands and mixed hardwood-pine forests, 1 = open water, and 0 = agriculture, urban, and shrub and brush lands). The analysis extents for the two alignments are used as analysis masks, then the rare habitats field is selected within the Arcview map calculator; and a new rare habitat grid is calculated for the analysis mask area only.

PAGE 310

294 Each proposed corridor must be calculated separately to obtain the correct data for each area. In this case, the count (number of cells) for xeric habitats and hardwood hammocks for corridor one and two are 4,105 and 1,459 cells, respectively. Each cell is 900 m2, therefore using the following formula, hectares (ha) = # cells x cell size (m) / 10,000 m2, reveals that 369.45 ha exist for alignment one and 131.31 ha exist for alignment two. Quantitative indicators of other impacted resources of interest might include area of other rare habitat types, public conservation lands and greenways, number of creek crossings, number of housing units, etc. Table B-2. Attributes of landscape features-comb Value Count Ecotones Rare Habitats Ridges Hydrology Sum 1 162067 0 4004 2 64695 1 4005 3 6502 0 2035 4 341085 0 2002 5 1957 1 4038 6 4421 0 4037 7 848865 0 0000 8 63510 0 3003 9 100062 1 2003 10 171469 1 0001 11 13735 0 0022 12 21422 1 3004 Qualitative assessment in this type of analysis includes setting and identifying relative rankings from the priority results (Chapter 2 presents a statewide application). To compare the relative impact of the two alternatives in Figure B-29, they must have the same scale of values. The data range of values for alignment one was 0 and alignment two was 0. Each dataset was reclassified into 5 equal interval classes between 0 and 100 (Figure B-30); darkest shades of red (value = 5) indicate the most sensitive areas. A comparison of the number of cells for each class (Table B-3) shows

PAGE 311

295that alignment one has significantly greater environmental impact than alignment two, with greater counts for classes 2 to 5. Other qualitative indicators that could be used to evaluate each proposed alignment include assessing habitat quality (SHCAs) and identifying listed species, FFWCC focal species hotspots, and FFWCC priority wetlands within each corridor. Alignment 1 Ali g nment 2 Table B-3. Alternative road alignment environmental sensitivity comparison Sensitivity Alignment 1 Alignment 2 value count count Difference 5 5,019 3,547 +1,472 4 13,412 6,153 +7,259 3 19,946 11,404 +8,542 2 18,115 12,296 +5,819 1 3,196 10,890 -7,694 Figure B-30. Reclassified priority layers for alternative road alignments. Priority layers were reclassified to an equivalent scale; tables reflect quantity of prioritized sites for each alignment on a scale of 1 to 5, 5 being most important ( reflected b y darkest shades of red )

PAGE 312

296 Extend/purchase lands north of Florida Panther NWR (identified in the Florida Greenways Analysis) APPENDIX C RECOMMENDED CORRECTIVE ACTIONS AT SPECIFIC SITES IN FLORIDA TO IMPROVE PERMEABILITY OF ROADS FOR WILDLIFE Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 1 Big Cypress National Preserve, Located from the entrance to Collier-Seminole State Park, extending 0.8 km east and west Fakahatchee Strand Includes a cluster of five black-bear road-kills State Preserve and Florida Panther US 41 79-84 Replace low level aquatic bridges with medium-level height and longer span bridges and add barrier fencing National Wildlife Located from 3.5 to 5.3 km west of CR 92 Refuge Region Includes a cluster of three black-bear road-kills 85-88 Replace low level aquatic bridges with medium-level height and longer span bridges and add barrier fencing Located at Port of the Islands entrance to 0.8 km east Includes two black-bear road-kills, post 1996 Road-kills located on a curve prior to the canal bridge and one east of the canal bridge 91-94 Recommendations include removal/reduction of curve to increase visibility Located 1.4 km west of CR 841 at curve Two black-bear road-kill incidents 0.6 miles apart 106-107Upgrade bridge on curve, add fencing Located 3.7 km north of northernmost underpass from US 41 intersection Two black-bear road-kills in area SR 29 119-121Add additional underpass, complete fence enclosure Located 0.6 km north of US 41 Two black-bear road-kills within 0.3 km 114 Add additional underpass/fencing Located 4.7 km south of I-75 One Florida panther and two black-bear road-kills within 1.6 km 124-127Add additional underpass and complete fence enclosure Located 5.2 km north of second underpass north of I-75 Includes a black bear and Florida panther road-kill 139-141Add wildlife bridge at GPS point location #140, extend fencing Located 8.4 km north of second underpass north of I-75 Add additional wildlife crossing bridge and extend fencing 142-143

PAGE 313

297 culverts with larger structures capable of facilitating movement by black bear, deer, etc.; add wildlife fencing. Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 1 Big Cypress Located 6 km east of intersection with SR 951 National Preserve Includes 4 black-bear road-kills continued Upgrade box culvert to bridge/add high-level fencing I 75 132-133Purchase Western Big Cypress CARL land to north and south of site Located 1.3 km west of Everglades Blvd. Overpass Includes 3 black-bear road-kills Wildlife fence ends at point 135 Two bridges present and capable of accommodating use by black bears, etc. Extend high-level wildlife fence past Everglades Blvd. 134-135Purchase Golden Gate Estates CARL Fisheating Creek Located just north of intersection with CR 731 CARL / SOR Includes 5 black-bear road-kills (Conservation and Add underpass and fencing Recreation Lands / Save Our Rivers) lands, Archbold US 27 158 Project must include acquisition of adjacent lands that functionally connect Johnson ranch lands with greenways to the east and north of Fisheating Creek Biological Station Located 4.2 km north of CR 731 and Lake Wales Includes 4 black-bear road-kills Ridge Conservation Areas Replace existing box culvert associated with unknown creek with larger combination creek/wildlife bridge, add wildlife fencing 160 Purchase lands south of Archbold BS, adjacent to roadway, and to the east along unknown creek that connects south to Fisheating Creek Located from intersection with SR 17 going south 2 km Includes 2 black-bear road-kills Increase size of existing box culverts to accommodate larger fauna, e.g. black bear; add fencing 184-186Critical to purchase Saddle Blanket Scrub CARL and greenway corridor to the east to provide linkage between Arbuckle State Forest and the Charlie Creek SOR corridor Located 1 km east of CR 17 Includes 3 black-bear road-kills Upgrade concrete pipe creek drainage with bear crossing culvert and add fencing US 98 171 Recommendations also include purchase of adjacent Lake Wales Ridge CARL land that is currently posted for sale Located along a 4.8 km section of road west of CR 635 Two black-bear road-kills located just to the east, potential use by bears likely SR 66 173-178Recommendations include replacement of pipe and box

PAGE 314

298 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 1 Fisheating Creekcontinued SR 66 173-178Purchase greenway lands on the south side of roadway adjacent to Highlands Hammock State Park Myakka River State Park, Myakka Located within the Myakka River system beginning at the Englewood exit and extending 8 km south Prairie State Preserve, Lower Myakka system Increase permeability of roadway through area by enlarging existing box culverts associated with multiple creeks that cross the highway CARL and SOR I75 1-8 Purchase Lower Myakka SOR properties projects Located within the Myakka River system beginning at intersection of CR 777 extending 1.9 km west Increase permeability of roadway through area by replacing existing culverts with larger structures associated with creek drainages that cross the highway US 41 42-47 Purchase Taylor Ranch SOR properties SR 72 11-20 Located at 0.8 km west of Myakka SP entrance extending east for 13.7 km Upgrade existing "old" bridges with wider span structure and replace assorted pipe culverts with large wildlife culverts or bridges to increase connectivity among existing state conservation-lands in the area Other Road Projects Peace River bridge replacement, near Zolfo Springs US 17 33 Maintain open character and height of existing bridge across floodplain valley Turner River bridge replacement #030083 US 41 108 Recommend increasing bridge span and fill removal or perforation with box culverts to add terrestrial linkage under bridges and increase roadway permeability 2 Ocala National Located 0.6 km north of CR 310 Forest, Cross-Florida Greenway Extend Dunn's Creek bridge when replacement scheduled and add second bridge to north and Etonia Creek CARL SR 19 285 These measures will increase permeability of roads for herpetofauna in bisected hardwood swamp that functions as greenway linkage for Ocala NF and Cross-Florida Greenway/Etonia Creek CARL Located approximately 1 km north of Ocklawaha River Replace existing 'aquatic' box culvert with expanded medium-level bridge 287 Important crossing point along the Ocklawaha River riparian corridor Located at Rice Creek within the Etonia Creek CARL SR 20 198 Scheduled bridge replacement should include longer span medium-level bridge capable of facilitating use by the Florida black bear

PAGE 315

299 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 2 Ocala National Forestcontinued In addition, raised road bed within the Rice Creek basin should be fitted with additional wildlife passages constructed up to 2.1 km east and west of the main channel bridge Erect wildlife fence the length of the Rice Creek basin (approx. 4.8 km) Florida black-bear road-kill 2.1 km east of Rice Creek SR 20 198 Complete purchase or conservation easement of Etonia Creek CARL Located 2.1 km west of SR 21, Fowlers Prairie Replace existing box culvert with longer medium-level bridge 144 In addition, raised road bed within the basin should be fitted with additional wildlife passages SR 100 193-197Located 4.5 km east of Coral Farms Rd and extending east another 10.9 km Series of 7 black-bear road-kills in area Replace Rice Creek bridge (GPS pt. #197) with medium-level span when scheduled in FDOT bridge replacement program Replace concrete pipes (GPS pts. #193, 194, 196) with wildlife passages appropriate for Florida black bears and add large wildlife fencing Topography at these sites provides sufficient relief to install large (3 m tall) culverts without need for additional fill Complete purchase or conservation easement of Etonia Creek CARL Fish Swamp WMD Located 1.6 km east of CR 305, Fish Swamp greenway (water management Black-bear road-kill at site district) project and Twelve-Mile Replace existing small box culvert associated with unknown creek with medium-level bridge Swamp CARL SR 206 200 Add additional wildlife passages to the east to increase roadway permeability and add large wildlife fencing St. Marys River Located 4.7 and 5.1 km west of US 301, respectively corridor and Cary and Jennings State Deep creek bridge scheduled replacement, install medium-level height and longer span Forests Railroad bridge # 770002 scheduled replacement, needs to be at least medium height to accommodate Florida black-bear movement US 90 115-116Black-bear movement corridor adjacent to corresponding bridges on I 10

PAGE 316

300 Greenway Site includes one black-bear road-kill Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 2 St. Marys Rivercontinued US 90 115-116Central greenway linkage connecting Osceola NF/St. Marries River corridor and Cary SF/Jennings SF/Camp Blanding MTS needs to be targeted for CARL or SOR program Located 5.3 km east of US 301 Brandy branch bridge scheduled replacement, install medium-level height and longer span 92 Important connection between Cary and Jennings SF Located 6.4 km west of US 301 One black-bear road-kill near existing railroad bridge I10 100 Abandoned railroad bridge needs to be supplemented by wildlife fencing along roadway to ensure passage through structure by black bears Camp Blanding Military Training Located at South Fork Black creek and extending 2.7 km west. Site Black-bear road-kills at creek bridge and 0.8 miles west (GPS pt. #218) Supplement South Fork Black Creek bridge with high-level wildlife fence extending west to SR 21 SR 16 217-219Add two wildlife tunnels (bear culverts) at one mile intervals west from South Fork Black Creek bridge Located 1 and 4 km south of SR 16, respectively Replace small concrete box culvert and wooden-based bridge with medium-level long span bridges Topographically sharp valleys present SR 21 220-221Implementation of these measures will increase permeability of SR 21 that functionally divides Camp Blanding from Etonia Creek CARL 222-223Located just south of intersection CR 315 Replace two box culverts with medium-level longer span bridges Implementation of these measures will increase permeability of SR 21 that functionally divides Camp Blanding from Etonia Creek CARL Located east and west of Goldhead Branch State Park entrance 224-225Install a series of Gopher tortoise tunnels/drift fences to remedy road-kill problem documented by park officials Lochloosa Forest Conservation Lochloosa Creek bridge scheduled replacement; install longer span, medium-level structure Easement, Newnans Lake CARL and SR 20 72 Important linkage for Florida black-bear movement between Lochloosa and Austin Cary forests Cross Florida Located 1.6 km west of US 301 SR 26 155

PAGE 317

301 Two black-bear road-kills each at or near both sites Add black-bear underpass at GPS pt. #385 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 2 Lochloosa Forestcontinued Eliminate sharp curve and reconfigure intersection with dirt road SR 26 155 Area represents a travel corridor for black bears between Austin Cary, Newnans Lake Carl, Lochloosa and lands to the east of US 301 Located at Alachua/Marion county line Replace low-level existing bridge with longer span medium-level bridge that incorporates part of the upland ecotone (e.g. SR 46 Wekiva River bridge) US 301 132-133Replacement will enhance use by Florida black bear and other wetland-dependent species moving along the Orange Creek corridor Located 4.2 km north of CR 325 (Lochloosa swamp) and 1.8 km south of CR 200 (Lochloosa conservation easement), respectively Black-bear road-kill at GPS pt. # 142 Replace existing box culverts with large black-bear culverts or medium-level bridges 136, 142Purchase designated greenways to the east that functionally connect Lochloosa conservation easement with the Cross Florida Greenway Cedar Key Scrub State Preserve Includes a 7 km section of highway beginning at the Channel #4 bridge going east Increase permeability for scrub/sandhill associated herpetofauna by installing small box culverts SR 24 331-335Take measures to reduce harmful erosion along right-of-way caused by human trampling and off road vehicles Aucilla River Located 2.3 km east of CR 14 and Econfina River Wildlife Three black-bear road-kills Management Area and Econfina Replace concrete pipe with large wildlife tunnel, include fencing for 1 mile each direction River Corridor US 98 345-346Econfina River bridge should be lengthened and raised according to scheduled replacement date Aucilla river bridge at the Taylor/Jefferson county line Florida black-bear road-kill 347 Replace bridge with longer span medium-level bridge and add wildlife fencing that extends beyond floodplain Located at the Econfina River Existing bridge should be lengthened and raised according to scheduled replacement date US 27 386 Add wildlife fencing that extends beyond floodplain Located 2.7 km south of Aucilla River and 2.6 km south of CR 14, respectively 383, 385

PAGE 318

302 Adjacent to Cary State Forest and Brandy Branch Creek corridor to St. Marys River Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 2 Aucilla Rivercontinued Replace existing culvert (associated with Aucilla River tributary) at point #383 with medium-level bridge that includes upland passage beyond the creek floodway or high water mark US 27 383, 385Secure purchase of conservation easement along Aucilla River and tributaries (wmd project) Pinhook Swamp, Osceola NF and Suwannee River Points #253 and #255, located 1.6 and 4.2 km south of Georgia line, respectively, are sites of Florida black-bear deaths Point #257 and #256 are located at the intersection with CR 6 and 2 km north of CR 6, respectively Point #257 is also the site of a road-kill Florida black-bear All sites are associated with proposed SOR project that connects the Suwannee River with Pinhook Swamp, purchase should be accelerated Add black-bear passages at all sites except the intersection with CR 6 US 441 253, 255-257 CR 6 intersection should be cleared to increase driver visibility Located along an 13 km section of road beginning 2.3 km west of the St. Marys River Four road-kill black-bears Replace railroad crossing 1.4 mi west of St. Marys River with high-level bridge Replace two low-level bridges at points #247 249 with longer span medium-level bridges Add black-bear passage between two small concrete pipes located at points #245 and 246. Add high-level wildlife fencing SR 2 245-250SR 2 bisects Pinhook Swamp-Okefenokee Swamp NWR; these measures will increase the permeability of the road and reduce road-kill incidents. Abandoned railroad bridge 1.4 km north of US 129 Black-bear road-kill I75 269 Extend high-level wildlife fencing from existing bridge Other Road Projects Road widening through Mallory Swamp Potential opportunity to upgrade several low quality wood-based bridges with longer span floodplain bridges SR 51 301-302Replace two small box culverts associated with Owl Creek, 2.1 and 3.1 km south of CR 357 at Cooks Hammock US 301 add lanes and reconstruction US 301 96-97

PAGE 319

303 wildlife fencing Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 2 Other Road Projectscontinued US 301 96-97 Replace existing bridges at Brandy Branch and those 2.4 km north of CR 119 with longer span medium-level bridges capable of accommodating black-bear movement 3 Aucilla Wildlife and Located at the Aucilla River, Madison county line Water Management Includes one black-bear road-kill Area US 27 301 Extend high-level wildlife fence from existing bridge in both directions to help prevent black bears from entering traffic lanes Located 3.5 to 4.2 km west of Madison County line Three black-bear road-kills Replace small culvert with black-bear underpass, add fencing 299-300Movement corridor through Aucilla R. WMA/WMD lands and proposed greenways Located 4.5 to 5.8 km east of US 19 Includes two black-bear road-kills Replace concrete pipe with black-bear wildlife passage at GPS pt. # 297 296-298Movement corridor through Aucilla R. WMA/WMD lands and proposed greenways Located 3.1 km east of US 19 Includes two black-bear road-kills Add wildlife passage and associated high-level fencing 295 Movement corridor through Aucilla R. WMA/WMD lands and proposed greenways Located 1.4 km east of US 19 Includes five black-bear road-kills Replace concrete pipe with medium-level bridge and associated wildlife fencing 294 Movement corridor through Aucilla R. WMA/WMD lands and Red Hills conservation area Located 0.8 to 2.1 km west of the Aucilla River Three black-bear road-kills Install two black-bear culverts/wildlife bridges US 98 2-3 Include wildlife fencing extending east to Aucilla River and west around curve in road Located 3.5 km east of SR 59 on curve Site of three black-bear road-kills Add black-bear wildlife underpass and associated fencing 4 Aucilla WMA Located 1.6 km east of SR 59 Includes one black-bear road-kill 6 Replace existing box culvert with wildlife underpass, add

PAGE 320

304 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 3 Aucilla Wildlifecontinued US 98 6 Proposed greenway expansion of Aucilla WMA, recommend adding to SOR or CARL Located at intersection of SR 59 and 1.6 km west of SR 59, respectively Includes two black-bear road-kills Replace existing pipe culvert with wildlife underpass Purchase land surrounding intersection to eliminate potential for development 7, 291 Proposed greenway expansion of Aucilla WMA, recommend adding to SOR or CARL Wakulla Springs State Park and Located at the Wakulla River and 1.1 km east, respectively CARL Projects and Includes one black-bear road-kill St. Marks NWR Add lower river corridor to Wakulla CARL proposal as linkage to St. Marks NWR Secure land adjacent to Wakulla River bridge to limit development US 98 284-285Install wildlife underpass at location of bear road-kill 0.7 miles east of river Located 0.8 to 4 km west of intersection with CR 61 Extend McBride Creek bridge span when replacement scheduled Install two additional wildlife underpasses west of McBride Creek SR 267 279-283Purchase Wakulla Springs CARL to secure linkage to Apalachicola NF Apalachicola National Forest, Apalachicola Bay shoreline from intersection with SR 65 to east of US 319 Tates Hell SF-CARL Eleven black-bear road-kills and Apalachicola River WMD Project Areas Land cover ranges from urban and transitional development to native xeric oak/longleaf pine forest and coastal salt marsh Secure purchase of remaining linkages from coast to interior forest habitat and consider construction of underpasses at these sites US 98 252-259Alternatively improve visibility along curves and at intersections with vegetation management and road realignments Located from Franklin county line traveling west 2.1 km Includes three black-bear road-kills Linkage from St. Joes Bay Buffers to Apalachicola River WMD lands 207-211Replace two small box culverts with black-bear wildlife underpasses

PAGE 321

305 road, or improve visibility at intersection Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 3 Apalachicola NFcontinued US 98 207-211Add wildlife fencing 1.6 km going each direction from structures (approx. 4.8 km) Located 3.6 to 4.3 km north of US 98 Three black-bear road-kills within 0.5 km Construct black-bear underpass 3.9 km north of US 98, add high-level fencing SR 71 217-218Private forest land that represents landscape linkage from Apalachicola NF to Tyndall AFB Located from 1 km south ofCR 387 to 6.4 km north of CR 387 Six black-bear road-kills evenly spaced at 5 points approx. 1.6 km apart Construct/replace existing culverts with black-bear underpasses at above mentioned intervals/ add high-level fencing Replace wood-base bridge at Cypress Creek with medium-level longer span bridge 221-224Private forest land that represents landscape linkage from Apalachicola NF to Tyndall AFB Located 0.6 km north of Dunns Creek Black-bear road-kill near curve and dirt access road Reduce severity of curve on approach to bridge/improve visibility SR 65 247 Ecotone where Tates Hell SF reaches coastal salt marsh Located 3.1 km north of Dunns Creek Black-bear road-kill near curve Reduce severity of curve/ improve visibility with vegetation management 246 Ecotone where Tates Hell SF reaches coastal salt marsh Located 10.4 km south of Liberty county line in Apalachicola NF Black-bear road-kill near bridge Supplement bridge with high-level wildlife fencing 240 Replace bridge with longer span if high water line is above sides of bridge Located 8.5 km south of CR 12 in Apalachicola NF Black-bear road-kill Investigate effects of access road use by black bears 236 Replace pipe culverts with wildlife underpass for black bears/add fencing if unable to eliminate access road Tyndall Air Force Located 2.9 km west of AFB boundary Base Two black-bear road-kills US 98 198 Investigate impact of access road, consider removal of side

PAGE 322

306 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 3 Eglin Air Force Base Located from the Yellow River bridge traveling south 1.6 km Three black-bear road-kills Area characterized by steep banks along road bed, access roads and curves Recommend removing what access roads possible, reducing severity of curve/improve visibility and adding high-level wildlife fence from the bridge extending 3.2 km south SR 87 174-176An alternative is to perforate the high road bed with large wildlife crossing culverts 179 Located 8.9 km south of intersection with CR 184 One black-bear road-kill Steep banks within valley provides opportunity to perforate road bed with large wildlife crossing culvert Located 1.9 km south of intersection SR 85 One black-bear road-kill SR 123 122 Steep ravine has inadequate existing culvert associated with unknown creek, replace with one or two large wildlife crossing culverts that maintains terrestrial and aquatic zones Located from 0.8 to 1.8 km north of intersection SR 123 Two black-bear road-kills Replace existing small culvert with medium-level wildlife bridge SR 85 127-128Supplement bridge with high-level wildlife fencing Located from 5.8 km west of CR 4 overpass going west to 5.8 km east of SR 87 Topography characterized by high ridges and deep ravines conducive to retrofitting highway without disturbing existing road surface Includes 12 sites with small pipe or box culverts (less than 1.7 m high), replace these with large wildlife tunnels and associated fencing along right-of way and through wooded medians (a.k.a. Trans-Canada Highway) Sites of most importance are GPS point nos. 136 145. I 134-145Critical linkage (Yellow River Ravines CARL) connecting Blackwater State Forest to Eglin AFB Proposed Greenway Located 12.9 km west of SR 71, Wewahitchka Linkages and Two black-bear road-kills near curve Private Forest Remove curve in road/improve visibility Lands SR 22 225 Private forest lands that form linkages from Apalachicola NF to Eglin AFB and other conservation lands west of the Apalachicola River

PAGE 323

307 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 3 Proposed Greenway Located from 5.8 to 8.4 km east of US 231 Linkagescontinued Three black-bear road-kills, one near curve Replace existing small culvert at site of two bear deaths with black-bear underpass/ add fencing Reduce severity of curve/improve visibility with vegetation removal SR 20 49-50 Private forest lands that form linkages from Apalachicola NF to Eglin AFB and other conservation lands west of the Apalachicola River Other Road Projects From 2.6 km west of CR 30a to 0.6 km west of US 331 Road project involves additional lanes and reconstruction Road bisects Topsail Hill State Park and Point Washington State Forest US 98 180-184Project should include installation of multiple wildlife crossing structures that vary in size from deer passages to tunnels for gopher tortoises, etc. to maintain a level of connectivity between inland and shoreline habitats for native fauna US 33189 Located 4.2 km north of SR 20 Project involves new construction to convert to four lane highway This area represents one of two important greenway linkages connecting Eglin AFB to conservation lands to the east. In conjunction with road reconstruction, wildlife underpasses should be constructed 4 Blue Cypress SOR, Kissimmee Located from just west of intersection of CR 512 extending west 3.5 km River SOR and Ft. Drum Marsh Road project programmedadd lanes and reconstructioncoordinate with wildlife considerations WMD lands Considerable evidence of herpetofauna mortality on the road, e.g. several species of aquatic turtles Perforate high road bed crossing through marsh by adding multiple medium-sized cross drainage culverts from CR 512 west to 1.6 km past the c-52 canal SR 60 6-7 Road bisects the Blue Cypress Creek-Ft. Drum conservation areas Road project programmedadd lanes and reconstructioncoordinate with wildlife considerations Located 7.4 and 8.4 km west of intersection of CR 512, respectively 8-9 Lengthen existing bridge span to provide terrestrial linkage under bridge on western side along high ground

PAGE 324

308 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 4 Blue Cypress SORcontinued Perforate high road bed crossing through marsh by adding multiple medium-sized cross drainage culverts 1 mile east and west of existing bridges SR 60 8-9 Road bisects the Blue Cypress-Kissimmee River SOR, and Ft. Drum conservation areas Located 7.6 km east of Osceola county line Road project programmedadd lanes and reconstructioncoordinate with wildlife considerations Padgett marsh bridge adequate during dry periods to allow movement by non-aquatics under bridge, high water periods may eliminate movement under bridge by non-aquatics 10 Extend bridge span in conjunction with road project Located 6 and 4.4 km east of Osceola county line, respectively Add wildlife passage culverts as part of programmed road project to increase connectivity between Blue Cypress conservation area and Indian River county preserve-Kissimmee River wmd lands 11-12 Four laning will increase barrier effects of roads, wildlife passages can maintain some measure of permeability of the roadway for wildlife Cypress Creek CARL SR 70 51-54 Located from 2.3 to 4.3 km east of Okeechobee county line In conjunction with programmed road project (add lanes and reconstruction) Bridge replacements scheduled for bridge numbers 940010 940012 Replace low level bridges and box culverts with medium-level wildlife underpasses and add wildlife fencing Road bisects Cypress Creek CARL lands 5 Ocala National Located from 2.4 to 9.3 km south of CR 314 Forest Area is associated with 12 black-bear road-kills Sites of road-killed bears occur on slopes of hills or curves in the highway Possible remedies for sites at GPS points 70 74 include construction of black-bear underpasses or road reconstruction that increases driver visibility by eliminating slopes and straightening curves Recommend construction of underpass at GPS point #76, a straight section of road NW of Lake George where multiple bear road-kills have occurred SR 19 70-76 Where underpasses are recommended 3 m fencing should be erected

PAGE 325

309 Associated with 11 black-bear road-kills Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 5 Ocala National Forestcontinued Located near the access roads to Pats Island and Silver Glen Springs, respectively GPS point locations are associated with six black-bear road-kills Sites of road-killed bears occur on slopes of hills, curves in the highway, and near dirt access roads SR 19 77-78 Possible remedies include construction of black-bear underpasses/barrier fencing or road reconstruction that increases driver visibility by eliminating slopes and straightening curves Located at and north of Juniper Creek GPS point locations are associated with six black-bear road-kills Recommend Juniper Creek bridge replacement that includes a 120 150 m medium-level span capable of facilitating safe travel for black bears beneath the bridge 80-81 Bridge should be equipped with barrier fencing that extends north 1.2 km (beyond curve in highway) Located 4 km north of SR 40 Includes two black-bear road-kills Bear fatalities occurred on the upslope of a hill 82 Possible remedies include construction of black-bear underpass/barrier fencing system or road reconstruction that increases driver visibility by eliminating slope Located from 0.6 to 3.5 km south of SR 40 Associated with four black-bear road-kills 96-98 Possible remedies include construction of black-bear underpass/barrier fencing system or road reconstruction that increases driver visibility by eliminating slopes 99, 105 Located at Beakman Lake and south of CR 445a Associated with three black-bear road-kills Replace existing stream culvert with combination drainage/wildlife underpass Realign road away from northeast side of lake to eliminate curve and reduce road associated runoff to the lake Purchase forest inholdings Located from 0.2 to 1.1 km north of Forest Service rd 595 Associated with two black-bear road-kills 102-104Replace existing stream culvert with combination drainage/wildlife underpass Located from 0.8 to 3.2 km north of CR 445 100-101

PAGE 326

310 Associated with four black-bear road-kills Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 5 Ocala National Forestcontinued Recommend construction of two underpasses, one at each point Fencing should be erected that runs from 0.8 km south to 4 km north of CR 445 SR 19 100-101Purchase forest inholdings Located 1 km east of CR 314a Associated with 7 black-bear road-kills Include remedies (below) as part of road project that includes adding lanes and reconstruction Recommend construction of wildlife underpass/fencing system, fence should extend from CR 314a west 1.6 km SR 40 93 Purchase remaining inholdings that are not within development clusters Located at intersections of Forest Service roads 97 and 88 Associated with two black-bear road-kills 64, 92 Improve visibility near intersections through vegetation management, reduce speed near intersections Located from intersection with Forest Service road 65 going east 2.7 km, Juniper Spgs entrance Associated with 12 black-bear road-kills Road corridor characterized by rolling hills and curves Proposed remedy includes a combination of two underpasses and road reconstruction that increases driver visibility by eliminating slopes and curves 89-91 Erect fence that extends from SR 19 to Forest Service road 65 Located 0.4 km east of Juniper Springs entrance Associated with six black-bear road-kills Area adjacent to Juniper Prairie ecosystem Road corridor characterized by rolling hills and curves Proposed remedy includes a combination of two underpass and road reconstruction that increases driver visibility by eliminating slopes and curves 88 Erect fence that extends from SR 19 to Forest Service road 65 84-86 Located from intersection with SR 19 and west 2.4 km Associated with six black-bear road-kills Road corridor characterized by rolling hills and curves Construct two underpasses Erect fence that extends from SR 19 to Forest Service road 65 Located from 0.4 to 1.9 km east of SR 19 106-108

PAGE 327

311 Includes 11 black-bear road-kills Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 5 Ocala National Forestcontinued Road separates Wildcat Lake ecosystem from wetlands to the north Construct underpass approx. 1 km east of SR 19 SR 40 106-108Erect fencing from SR 19 intersection and extend east 2.4 km (beyond double curve) Located just west of Blue Creek Lodge road Includes seven black-bear road-kills 110-111Reduce severity of curve and install wildlife underpass/fencing Located from intersection of CR 445a to 1.8 km east Includes 17 black-bear road-kills Include remedies (below) as part of road project that includes adding lanes and reconstruction Replace box culvert at unknown creek with long span medium-level bridge that includes terrestrial connection above high water line beneath bridge Add additional underpass to west (approx. 0.6 to 1.1 km from unknown creek) 112-113Include fencing from CR 445a to 1.8 km west (Ocala NF boundary) Area parameters not included in GIS analysis. Although the analysis was restricted to state roads, it should be recognized that CR 42 accounted for 19 documented black-bear road-kills as of 1999 CR 42 n/a As such it should receive serious consideration for improvements to reduce impacts on black-bear movement Lake Woodruff Located 0.2 km west of CR 42 NWR, Hontoon Includes one black-bear road-kill Island State Park and Wekiva River Basin Ocala greenway SR 44 132 Purchase remaining private lands up to 1.6 km west to secure connection between Ocala NF/Lake Woodruff NWR and the Wekiva River Basin SOR lands Located from entrance of Royal Trails development west 4.8 km to Cassia Includes six black-bear road-kills Area includes scattered rural-development, as top priority for this site recommend purchase of land in area (undeveloped inholdings and key developed parcels) to secure quality of corridor 134-135Construct two underpasses in area (contingent on ability to tie-up future development potential of area) 136-137Located east and west of CR 44a intersection (past Blackwater Creek)

PAGE 328

312 Replace bridge with medium-level (3 m or greater) long span bridge and add wildlife fencing Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 5 Lake Woodruff ...continued SR 44 136-137Add wildlife underpasses east and west of Blackwater Creek bridge, enclose section in wildlife fencing, and lengthen bridge span when replacement scheduled Wekiva River State Located from 3.5 to 4.2 km west of CR 46a Park and Seminole Site includes seven black-bear road-kills State Forest Site occurs on double curve Straighten road and add wildlife underpass/fencing system SR 46 145-146Purchase remaining BMK ranch lands Located from intersection with CR 46a going east 2.7 km Includes eight black-bear road-kills from 1996 -99, after installation of underpass/fence Recommend additional underpass be constructed to the east of existing one and fencing be extended 141-144, 66 Purchase remaining private inholdings in area including key developed parcels Located from Wekiva River west 1.3 km Area includes eight black-bear road-kills Add wildlife underpass west of Wekiva River bridge and extend fencing from bridge west 1.6 km Top priority for the entire Wekiva River Basin area should be purchase of parceled lots for sale adjacent to the river and south of SR 46; this is a critical link from Wekiva River State Park to Seminole State Forest 139-140, 65 Other remaining private inholdings should also be purchased Lake George State Located from 1.9 to 3.9 km south of SR 40 Forest and Haw Includes five black-bear road-kills Creek WMD Project Lands Important linkage from Ocala National Forest to conservation lands east of the St. Johns River Construct one underpass 1.6 km north of Deep Creek bridge Erect fencing from Deep Creek bridge 2.4 km north US 17 114-116Replace crushed rock substrate under Deep Creek bridge Located 3.2 km west of intersection CR 4023 Creek drainage with black-bear road-kills to the east and west Replace existing culvert with longer span medium-level bridge capable of accommodating black bears SR 40 119 Purchase private inholdings or development rights to restrict future development of area Located at the Little Haw Creek bridge Black-bear road-kill 120

PAGE 329

313 include wider terrestrial links during high water periods Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 5 Lake Georgecontinued SR 40 120 Purchase Container Corp tract (WMD project) or development rights Tiger Bay State Forest, Deep Located from 0.8 km west of US 92 to 3.7 km east of SR 44 Creek and Other Includes 10 black-bear road-kills in three clusters Proposed State Greenway Lands I4 126-129Recommend replacement of existing small box culverts with wildlife bridges and installation of associated high-level wildlife fencing Located from 4.8 to 13.2 km east of CR 4101 Includes two black-bear road-kills Deep Creek and Tiger Bay systems US 92 161-166Replace existing small box culverts with wildlife bridges when replacements scheduled Located from 2.3 km west of CR 4118 to 2.6 km east of Interstate 4 Includes one black-bear road-kill and numerous small mammals and reptiles at or near wetland sites Replace existing small culverts with large wildlife bridges, add additional passages at other wetland sites and straighten alignment of road These recommendations should be incorporated into planned road project to add lanes and reconstruct highway SR 44 175-180Purchase Deep Creek property and Tiger Bay proposed additions Located from 3.5 km south of SR 44 to 2.1 km north of CR 4152 Rapidly developing area that requires serious attention to save greenway connections from the St. Johns River to Tiger Bay State Forest Purchase remaining corridors along road right-of-ways SR 415 181-182Replace existing creek culverts with long span medium-level bridge Other Road Projects SR 520 projectadd lanes and reconstruction Existing culverts should be upgraded to large wildlife culverts/bridges Existing bridges over Jim and Second creeks should be lengthened to include terrestrial connections beneath the bridge that are above high water lines SR 520 272-288Secondary culvert as part of the St. Johns floodplain needs to be upgraded to a bridge with terrestrial linkages US 192 projectadd lanes and reconstruction US 192 291-293Bridges over the St. Johns river should be lengthened to

PAGE 330

314 13.7 km section of highway Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 5 Other Road Projectscontinued US 192 291-293Additional culverts/wildlife bridges should be constructed east and west of the St. Johns River bridges to increase permeability of the highway throughout the St. Johns River prairie system Bridge replacement at the Kissimmee River SR 60 District 1: 196 New bridge should be longer to incorporate terrestrial linkages under the bridge as part of restoration of the river and establishment of unbroken riparian corridor I4 214 Interstate 4 lane additions Project should include replacement of Reedy Creek bridge that includes a lengthened and higher span creating a functional terrestrial linkage for habitat areas north and south Other box culverts west of Reedy Creek should be upgraded to wildlife bridges SR 100 projectadd lanes and reconstruction Replacement bridge at Crescent lake creek should be a longer and higher span that incorporates a terrestrial habitat linkage SR 100 148-153Box culverts should be replaced with combination terrestrial wildlife/creek drainage bridges 6 Everglades National Park, Located from intersection with CR 905 extending north to the C-111 canal Model Lands Basin, Southern Multiple road-kills of the endangered American crocodile within this section of highway Glades SOR and Crocodile Lake NWR Perforate existing roadbed with the installation of aquatic wildlife crossings, recommended minimum width of openings approx. 6 m Wider openings are preferred to minimize predator-ambush activity and to help restore natural flow patterns between Biscayne and Florida Bays US 1 1-12 Placement guidelines in this study coincide with FDOT crocodile crossing location plan (9/94 project nos. 90060-3501, 90060-3585, 87010-3509) Located from C-111 canal north to intersection with CR 997 Multiple road-kills of American alligator and numerous signs of trails used by crocodilians Generally agree with FDOT proposed wildlife crossing plan (9/94 project nos. 90060-3501, 90060-3585, 87010-3509) regarding C-111 bridge and crossing structure dimensions 13-16 Recommend four additional crossing structures within this

PAGE 331

315 reconstruction Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 6 Everglades National Parkcontinued Located from the Dade county line extending east through Shark River Slough to SR 997 Although this area was not surveyed, it did receive a low priority from the GIS analysis, and as such should be considered regarding water/wildlife management issues US 41 n/a This corridor (it is assumed) will be addressed in the Everglades restoration effort Big Pine Key, Key Deer National Located on Big Pine Key from the Spanish Harbor channel bridge extending south approx. 3.2 km Wildlife Refuge, Cupon Bight Includes multiple/annual road-kills of endangered Florida Key deer CARL Project and Bahia Honda, Long Key and John Pennekamp State Recommend installation of wildlife underpasses at 0.4 km intervals (optimum width of 4.5 6 m) and include barrier fencing of appropriate height to prevent deer from entering lanes of traffic Parks US 1 18 Top priority acquisition of remaining habitat, including purchase and restoration of key developed areas to secure corridor connections from either side of the island Crossing through Bahia Honda, Long Key and Pennekamp State Parks These sites received lower priorities, however US 1 effectively divides the two prior state parks n/a Small mammal/herpetofauna tunnels installed through these areas would increase connectivity for local fauna and reduce road-kill problems in these parks 7 Chassahowitzka NWR/WMA and Location begins 4.8 km south of US 98 and extends 9.7 km south Annuteliga Hammock CARL Six of the seven sites are associated with Florida black-bear road-kills Project All black-bear road-kills occurred at sites where dirt trails/access roads intersect the highway or on curves in the highway Recommend three possible measuresincreased vegetation management at access points, removal of access roads and reduction in curves, or installation of wildlife passages designed for black bears US 19 4-10 This section of highway splits the Chassahowitzka lands from Annuteliga hammock/Withlacoochee SF to the eastan important connection for persistence of this isolated population of black bears Located 4.4 and 7.9 km east of US 19, near the Suncoast parkway corridor US 98 2-3 Sites are within a scheduled road project4 lane

PAGE 332

316 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 7 Chassahowitzka NWRcontinued Wildlife underpasses/bridges need to be constructed at approx. 4.4, 7.2 & 10.5 km from US 19 intersection to maintain connections between Chassahowitzka NWR and Annuteliga Hammock/Withlacoochee SF US 98 2-3 This area also should be a high priority CARL purchase, given potential development pressures in the surrounding area Critical area is from CR 480 south to CR 476 New toll-expressway construction project Sun-coast Park-way n/a Strategically placed wildlife underpass bridges (five recommended) are necessary along this corridor to maintain connections between Chassahowitzka NWR and Annuteliga Hammock/Withlacoochee SF Withlacoochee Located from 2.4 km north to 0.6 miles south of CR 476 State Forest Narrow strip corridor of Withlacoochee State Forest Highest priority to purchase greenway linkages to the east and west that would establish a contiguous forest from Chassahowitka to the Green Swamp US 41 31-33 Replace existing small culverts with large wildlife underpasses Located at the little Withlacoochee river and 1 km west Existing bridge and culvert need to be replaced with medium-level long span bridges SR 50 37-38 This section of highway winds through the Withlacoochee State Forest at an important connection to conservation lands to the northwest Located at the little Withlacoochee river US 301 36 Bridge is sufficient, however spreading rural-development surrounding the area threatens this important linkage b etween tracts of the Withlacoochee State Forest, purchase of the remaining undeveloped area is critical Cypress Creek Located from 0.5 to 1.6 km west of CR 583 Corridor and Pasco SOR/WMD lands, Lower Hillsborough WMD lands SR 52 11-12 Install series of large box culverts at sites of cypress strand, pond/marsh/wet & dry prairie ecosystems to increase permeability of roadway and to facilitate safe movement of small mammals/herpetofauna through the PASCO SOR corridor that connects to Cypress Creek/Hillsborough River Located at 4.5 and 7.7 km from I-75, respectively SR 54 13-14 Multi-lane construction is occurring at the latter point as part of a new road (SR 56) that cuts through the Cypress Creek tract

PAGE 333

317 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 7 Cypress Creekcontinued SR 54 13-14 Wildlife underpasses and an expanded bridge should be included in the plan as mitigation for wetland impacts; these stipulations should also apply to SR 56 In addition the multi-cell culvert at the first GPS point should be replaced with a combination creek/wildlife bridge Located 4.4 and 2.1 km north of Interstate 275 When construction (lane additions, etc.) is planned, the two structures at these points need to be upgraded to long span medium-level bridges I75 15-16 These sites are part of the Cypress Creek-Hillsborough River linkage Hillsborough State Park, Hillsborough Located at the entrance to Hillsborough SP and north 1.9 km River corridor (SOR) and Blackwater Creek US 301 23-25 Install large wildlife underpasses at each site north and south of the Hillsborough River to increase the permeability of the roadway for wetland-dependent species SR 39 27-28 Located at Blackwater Creek Scheduled bridge replacement Install a single length span to replace both structures at medium-level height (3+ m) Other Road Projects Located at the cross-Florida barge canal Scheduled construction of additional new span US 19 1 New bridge should be similar to existing one in length and of at least medium height to maintain openness of cross-Florida greenway corridor Located at the Withlacoochee River Road project involves additional lanes and bridge construction SR 200 29 New bridge should be expanded past existing span, remove existing fill and lengthen spanespecially on east side of bridge where a continuous linkage of public conservation lands are proposed Located at the Withlacoochee River Road project involves additional lanes and possible bridge construction If bridge construction planned; remove additional approach fill and establish cut terrestrial underpass via culvert installation/ or extend bridge SR 44 30 Adjacent to Flying Eagle ranch

PAGE 334

318 Appendix C. continued Dt* Conservation Area Road Site Nos. Location, Highlights, and Recommendations 7 Other Road Sunn/a Starkey Wilderness Area Projectscontinued coast Park-way, SR The US 41 lane addition project crosses the last remaining linkage for the area to the Pasco SOR land; construction of an underpass at this location should be investigated The new Suncoast Parkway also should include on-site mitigation for fragmentation impacts, e.g., construction of multiple underpasses 52, US 41 State Road 52 lane addition project would sever a small area identified in the Florida greenways project Notes: 1) the site numbers correspond to the GPS locations shown in the respective FDOT district maps (Chapter 3, Figures 3-11 to 3-17). 2) Dt refers to FDOT maintenance district number

PAGE 335

319 Green, and S. Cousins, editors. Landscape ecology and GIS. Taylor and Francis, Inc., Bristol, PA. REFERENCES Adams, L.W. and A.D. Geis. 1983. Effects of roads on small mammals. Journal of Applied Ecology 20:403-415. Akcakaya, H.R. 2000. Conservation and management for multiple species: integrating field research and modeling into management decisions. Environmental Management 26 (supplement 1):S75-S83. Allaby, M., editor. 1994. The concise Oxford dictionary of ecology. Oxford University Press, Oxford and New York. American Planning Association (APA). 1995. Environment and natural resources. Capitol Highlights, Issue No. 95-1. A Publication of the Florida Chapter of the American Planning Association, Tallahassee, FL. Andrews, A. 1990. Fragmentation of habitat by roads and utility corridors: A review. Australian Zoologist 23:130-141. Angold, P.G. 1997. The impact of roads upon adjacent heathland vegetation: Effects on plant species composition. Journal of Applied Ecology 34:409-417. Archibald, W.R., R. Ellis and A.N. Hamilton. 1987. Responses of grizzly bears to logging truck traffic in the Kimsquit River Valley, British Columbia. International Conference on Bear Research and Management Series 7:251-257. Arendt, R. 1994. Rural by Design: Maintaining small town character. American Planning Association, Chicago, IL. Aresco, M.J. 2001. Migration patterns of turtles and other wildlife between Lake Jackson and Little Lake Jackson, Leon County, Florida: Justification for the construction of a wildlife barrier along U.S. Highway 27. A report to the Florida Department of Transportation. Department of Biological Science, Florida State University, Tallahassee, FL. Ariza, L.M. 1998. Divided they fall. New Scientist 158:21. Aspinall, R. 1993. Use of geographic information systems for interpreting land-use policy and modeling effects of land-use change. Pages 223-236 in R. Haines-Young, D.

PAGE 336

320 Auffenberg, W. 1956. A study of the fossil snakes of Florida. Ph.D. Dissertation. University of Florida, Gainesville, FL. Banasova, V. 1986. The influence of deicing salts on the soil and plants along a highway. Ekologia (CSSR) 4:315-328. Barichovich, W.J. and C.K. Dodd, Jr. 2002. The effectiveness of wildlife barriers and underpasses on U.S. Highway 441 across Paynes Prairie State Preserve, Alachua County, Florida. Phase II post-construction final report, contract no. BB-854. Florida Department of Transportation Tallahassee, FL. Barrass, A.N. 1985. The effects of highway traffic noise on the phonotactic and associated reproductive behavior of selected anurans. Ph.D. Dissertation. Vanderbilt University, Nashville, TN. Baschak, L. and R. Brown. 1994. River systems and landscape networks. Pages 179-200 in E.A. Cook and H.N. van Lier, editors. Landscape planning and ecological networks. Elsevier, Amsterdam. Bedford, G.S. and A.D. Griffiths. 1995. Road kill predation by the blackheaded python Aspidites melanocephalus. Herpetofauna 25:53. Beier, L. and S. Loe. 1992. A checklist for evaluating impacts to wildlife movement corridors. Wildlife Society Bulletin 20:434-440. Beier, P. and R. Noss. 1998. Do habitat corridors provide connectivity? Conservation Biology 12:1241-1252. Belant, J.L. 1995. Moose collisions with vehicles and trains in northeastern Minnesota. Alces 31:45-52. Bennett, A.F. 1999. Linkages in the landscape: the role of corridors and connectivity in wildlife conservation. International Union for Conservation of Nature, Forest Conservation Programme; IUCN, Gland, Switzerland and Cambridge, U.K. Bennett, A.F. 1991. Roads, roadsides, and wildlife conservation: a review. Pagesin D.A. Saunders and R.J. Hobbs, editors. Nature conservation 2: The role of corridors. Surrey Beatty and Sons, Chipping Norton, Australia. Berish, J. and E. Diemer. 1998. Characterization of rattlesnake harvest in Florida. Journal of Herpetology 32:551-557. 99-117

PAGE 337

321 Bertwhistle, J. 2003. Research into Wildlife/Vehicle Collisions in Jasper National Park. Paper presented at Making Connections, ICOET 2003, International Conference on Ecology and Transportation, August 24-29, Lake Placid, N.Y. Center for Transportation and the Environment, North Carolina State University, Raleigh, N.C. Beyer, W.N., D. Day, M.J. Melancon and L. Sileo. 2000. Toxicity of Anacostia River, Washington, D.C., U.S.A. sediment fed to mute swans Cygnus olor. Environmental Toxicology and Chemistry 19:731-735. Birdsall, C.W., C.E. Grue, and A. Anderson. 1986. Lead concentrations in bullfrog Rana catesbeiana and green frog Rana clamitans tadpoles inhabiting highway drainages. Environmental Pollution Series A. Ecological and Biological 40:233-248. Boarman, W. and M. Sazaki. 1996. Highway mortality in desert tortoises and small vertebrates: successes of barrier fences and culverts. In G. Evink, P. Garrett, D. Zeigler and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Florida Department of Transportation, Tallahassee, FL. Bowers, M.A. and S.F. Matter. 1997. Landscape ecology of mammals: relationships between density and patch size. Journal of Mammalogy 78:999-1013. Brehm, K. 1989. The acceptance of 0.2-metre tunnels by amphibians during their migration to the breeding site. Pages 29-42 in T.E.S. Langton, editor. Amphibians and roads. ACO Polymer Products Ltd. Bedfordshire, England. Brody, A.J. and M.R. Pelton. 1989. Effects of roads on black bear movements in western North Carolina. Wildlife Society Bulletin 17:5-10. Brody, A.J. and J.N. Stone. 1987. Timber harvest and black bear population dynamics in a southern Appalachian forest. International Conference on Bear Research and Management Series 7:243-250. Brown, G. 1979. Alternate transportation systems for reducing natural resources impact. Pages 257-262 in R. Ittner, D.R Potter, J.K. Agee, and S. Anschell, editors. Proceedings of the Conference on Recreational Impact on Wildlands. U.S.D.A. National Forest Service and U.S.D.I. National Park Service, Washington, D.C. Brown, J.H. and M.V. Lomolino. 1998. Biogeography, 2nd ed. Sinauer Associates, Inc., Sunderland, MA. Bruinderink, G.W.T.A.G., and E. Hazebroek. 1996. Ungulate traffic collisions in Europe. Conservation Biology 10:1059-1067. Buchanan, B.W. 1993. Effects of enhanced lighting on the behaviour of nocturnal frogs. Animal Behavior 45:893-899.

PAGE 338

322 Wildlife Agencies 38:70-77. Buchanan, J.B. 1987. Seasonality in the occurrence of long-tailed weasel road-kills. The Murrelet 68:67-68. Bureau of Economic and Business Research (BEBR). 2001. Florida statistical abstracts. University of Florida, Gainesville, FL. Bureau of Economic and Business Research (BEBR). 1998. Florida statistical abstracts. University of Florida, Gainesville, FL. Bureau of Economic and Business Research (BEBR). 1997. Florida statistical abstracts. University of Florida, Gainesville, FL. Burghard, A. 1969. Alligator alley: Florida's most controversial highway. Lanman Co., Washington D.C. Busack, S.D. and S.B. Hedges. 1984. Is the peninsula effect a red herring? American Naturalist 123:266-275. Calvo, R. and N. Silvy 1996. Key deer mortality, U.S. 1 in the Florida Keys. In G. Evink, P. Garrett, D. Zeigler, and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Florida Department of Transportation, Tallahassee, FL. Canaday, C. 1996. Loss of insectivorous birds along a gradient of human impact in Amazonia. Biological Conservation 77:63-77. Carey M. 1996. Salmon passages and other wildlife activities in Washington State. In G. Evink, P. Garrett, D. Zeigler and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Florida Department of Transportation, Tallahassee, FL. Carpenter, T.G. 1994. Environmental impacts of railways. John Wiley and Sons, Chichester, U.K. Carr, A.F. Jr. 1937. The geographic and ecological distribution of the reptiles and amphibians of Florida. Ph.D. Dissertation. University of Florida, Gainesville, FL. Carr, M.H., P.D. Zwick, T.S. Hoctor and M.A. Benedict. 1999. Final report, phase II, Florida statewide greenways planning project. Department of Landscape Architecture, University of Florida, Gainesville, FL. Carr, M.H., P.D. Zwick, T.S. Hoctor and M.A. Benedict. 1996. Final report, phase I, Florida statewide greenways planning project. Department of Landscape Architecture, University of Florida, Gainesville, FL. Carr, P.C. and M.R. Pelton. 1984. Proximity of adult female black bears to limited access roads. Proceedings of the Annual Conference of Southeastern Association of Fish and

PAGE 339

323 Chapin F.S. Jr. and E.J. Kaiser, editors. 1979. Urban land-use planning, 3rd edition. University of Illinois Press, Urbana, IL. Clark, J.D. and F.T. Van Manen. 1992. Geographic information systems and black bear habitat analyses. Pages 137-153 in Proceedings of the 11th Eastern Black Bear Workshop, Waterville Valley, N.H. New Hampshire Department of Fish and Game, Concord, N.H. Clarke, G.P., P. White and S. Harris. 1998. Effects of roads on badger Meles meles populations in southwest England. Biological Conservation 86:117-124. Clevenger, A.P., B. Chruszcz and K.E. Gunson. 2001. Drainage culverts as habitat linkages and factors affecting passage by mammals. Journal of Applied Ecology 38:1340-1349. Clevenger, A.P. and N. Waltho. 2000. Factors influencing the effectiveness of wildlife underpasses in Banff National Park, Alberta, Canada. Conservation Biology 14:47-56. Clevenger, A.P. and J. Wierzchowski. 2001. GIS-based modeling approaches to identify mitigation placements along roads. Pages 134-148 in A Time for Action: Proceedings of the International Conference on Ecology and Transportation, September 24-28, 2001, Keystone, CO. Center for Transportation and the Environment, North Carolina State University, Raleigh, N.C. Conner, R.N., R.G. Hooper, H.S. Crawford and H.S. Mosby. 1975. Woodpecker nesting habitat in cut and uncut woodlands in Virginia, U.S.A. Journal of Wildlife Management 39:144-150. Conover, M.R., W.C. Pitt, K.K. Kessler, T.J. Dubow and W.A. Sanborn. 1995. Review of human injuries, illnesses, and economic-losses caused by wildlife in the United States. Wildlife Society Bulletin 23:407-414. Cox, B. and P.D. Moore. 1980. Biogeography, an ecological and evolutionary approach, 3rd ed. John Wiley and Sons, New York, N.Y. Cox, J., R. Kautz, M. MacLaughlin, and T. Gilbert. 1994. Closing the gaps in Florida's wildlife habitat conservation system. Florida Game and Freshwater Fish Commission, Office of Environmental Services. Tallahassee, FL. Creachbaum, M.S., C. Johnson and R.H. Schmidt.1998. Living on the edge: a process for redesigning campgrounds in grizzly bear habitat. Landscape and Urban Planning 42:269-286.

PAGE 340

324 Cristoffer, C. 1991. Road mortality of northern Florida vertebrates. Florida Scientist 54:65-68. Crooks, K.R. and M.E. Soul. 1999. Mesopredator release and avifaunal extinctions in a fragmented system. Nature (London) 400:563-566. Cross, H.C., P.D. Wettin and F.M. Keenan. 1991. Corridors for wetland conservation and management? Room for conjecture. Pages 159-165 in D. Saunders and R. Hobbs, editors. Nature conservation 2: The role of corridors. Surrey Beatty and Sons, Chipping Norton, Australia. Csuti, B. 1991. Conservation corridors: countering habitat fragmentation, an introduction. Pages 81-90 in W.E. Hudson, editor. Landscape Linkages and Biodiversity. Defenders of Wildlife and Island Press, Washington, D.C. Darveau, M., P. Beauchesne, L. Belanger, J. Huot and P. Larue. 1995. Riparian forest strips as habitat for breeding birds in boreal forest. Journal of Wildlife Management 59:67-78. Davies, J.M., T.J. Roper, and D.J. Shepherdson. 1987. Seasonal distribution of road kills in the European badger. The Journal of Zoology 211:525-529. Day, D. 1990. Car-deer collisions. South Dakota Conservation Digest 57:8-11. Demaynadier, P. and M. Hunter, Jr. 1998. Effects of silvicultural edges on the distribution and abundance of amphibians in Maine. Conservation Biology 12:340-352. Dexel, R. 1989. Investigations into the protection of migrant amphibians from the threats from road traffic in the Federal Republic of Germanya summary. Pages 43-49 in T.E.S. Langton, editor. Amphibians and roads. ACO Polymer Products Ltd. Bedfordshire, England. Dickman, C.R. 1987. Habitat fragmentation and vertebrate species richness in an urban environment. Journal of Applied Ecology 24:337-351. Dobson, A., K. Ralls, M. Foster, M. Soul, D. Simberloff, D. Doak, J. Estes, L.S. Mills, D. Mattson, R. Dirzo, H. Arita, S. Ryan, E. Norse, R. Noss and D. Johns. 1998. Corridors: Reconnecting fragmented landscapes. Pages 129-170 in M.E. Soul and J. Terborgh, editors. Continental conservation: Scientific foundations of regional reserve networks. The Wildlands Project and Island Press, Washington, D.C. Dodd, C.K. Jr. 1990. Effects of habitat fragmentation on a stream-dwelling species, the flattened musk turtle Sternotherus depressus. Biological Conservation 54:33-45.

PAGE 341

325 Dodd, C.K. Jr., K.M. Enge and J.N. Stuart. 1989. Reptiles on highways in north-central Alabama, USA. Journal of Herpetology 23:197-200. Eason, T.H. 2000. Black bear status report, October 12, 2000. Division of Wildlife, Florida Fish and Wildlife Conservation Commission, Tallahassee, FL. Eason, T.H. and T.E. O'Meara. 2000. Comprehensive statewide bear management program in Florida. Page 18 in Managing wildlife in the face of urbanization, The 2000 Symposium of the Florida Chapter of the Wildlife Society, April 10-12, 2000, Daytona Beach, FL. Eisenberg, J.F. 1986. The biological basis for corridors. Pages 5-6 in ENFO 86-6. Environmental Information Center of the Florida Conservation Foundation, Inc., Winter Park, FL. Engstrom, R.T. 1993. Avian communities in Florida habitats: analysis and review. Nongame wildlife program final report. Florida Game and Fresh Water Fish Commission, Tallahassee, FL. Evink, G.L. 1990. Wildlife crossings of Florida I-75. Transportation Research Record 1279:54-59. Evink, G., P. Garrett and D. Zeigler, editors. 1999. Proceedings of the Third International Conference On Wildlife Ecology and Transportation, September 13-16, 1999, Missoula, MT. Florida Department of Transportation, Tallahassee, FL. Evink, G., P. Garrett and D. Zeigler, editors. 1998. Proceedings of the Second International Conference On Wildlife Ecology and Transportation, February 10-12, 1998, Ft. Myers, FL. Florida Department of Transportation, Tallahassee, FL. Evink, G.L., P. Garrett, D. Zeigler, and J. Berry, editors. 1996. Trends in addressing transportation-related wildlife mortality: Proceedings of the Transportation-related Wildlife Mortality Seminar, June 1996, Orlando, FL. Florida Department of Transportation, Tallahassee, FL. Fagan, W.F., R.S. Cantrell and C. Cosner. 1999. How habitat edges change species interactions. The American Naturalist 153:165-182. Fahrig, L., J.H. Pedlar, S.E. Pope, P.D. Taylor and J.F. Wegner. 1995. Effect of road traffic on amphibian density. Biological Conservation 73:177-182. Federal Highway Administration (FHA). 1979. Americas highways 1776-1976, a history of the Federal Aid Program. Federal Highway Administration, U.S. Department of Transportation, Washington, D.C.

PAGE 342

326 Federal Highway and Waterway Administration (FHWA). 2002a. Public road length 2000, miles by ownership, Table HM-10. Office of Highway Policy Information, Federal Highway and Waterway Administration, Washington, D.C. Federal Highway and Waterway Administration (FHWA). 2002b. Functional system travel 2000, annual vehicle miles, Table VM-2. Office of Highway Policy Information, Federal Highway and Waterway Administration, Washington, D.C. Federal Highway and Waterway Administration (FHWA). 2002c. Total disbursements for highways, all units of government 2000, Table HF-2. Office of Highway Policy Information, Federal Highway and Waterway Administration, Washington, D.C. Federal Highway and Waterway Administration (FHWA). 2001. State motor-vehicle registrations 2000, Table MV-1. Office of Highway Policy Information, Federal Highway and Waterway Administration, Washington, D.C. Federal Highway and Waterway Administration (FHWA). 1996. Highway Statistics Summary, 1947 to 1995. Office of Highway Information Management, Federal Highway and Waterway Administration, Washington, D.C. http://www.bts.gov/site/news/fhwa/HighwayStats-Summary95/ Feldhamer, G.A., J.E. Gates, D.M. Harman, A.J. Loranger and K.R. Dixon. 1986. Effects of interstate highway fencing on white-tailed deer activity. Journal of Wildlife Management 50:497-503. Ferris, C.R. 1979. Effects of Interstate 95 on breeding birds in northern Maine. Journal of Wildlife Management 43:421-427. Findlay, C.S. and J. Bourdages. 2000. Response time of wetland biodiversity to road construction on adjacent lands. Conservation Biology 14:86-94. Florida Department of Agriculture and Consumer Services, Division of Forestry (FDOF). 1987a. Silviculture best management practices manual. Division of Forestry, Florida Department of Agriculture and Consumer Services, Tallahassee, FL. Florida Department of Agriculture and Consumer Services, Division of Forestry (FDOF). 1987b. Management guidelines for forested wetlands in Florida. Division of Forestry, Florida Department of Agriculture and Consumer Services, Tallahassee, FL; Florida Forestry Association. Florida Department of Transportation (FDOT). 2000. The 2020 Florida Transportation Plan. Florida Department of Transportation, Office of Policy Planning, Tallahassee, FL.

PAGE 343

327 Florida Department of Transportation (FDOT). 2000. Estimate of construction for bridge no. 770003, financial project no. 24014301052-01. Florida Department of Transportation, Tallahassee, FL. Florida Department of Transportation (FDOT). 1999. Floridas tentative five year work program, July 1, 1999 through June 30, 2004. Office of Management and Budget, Program Development, Florida Department of Transportation, Tallahassee, FL. Florida Department of Transportation (FDOT). 1998. Historical report on mileage and traffic on Florida's state highway system. Florida Department of Transportation, Tallahassee, FL. Florida Department of Transportation (FDOT). 1997. Plans of proposed State Road 10, State project no. 27010-3502, Baker County, Florida. Florida Department of Transportation, Tallahassee, FL. Florida Department of Transportation (FDOT). 1996. Floridas tentative five year work program, July 1, 1996 through June 30, 2000. Office of Management and Budget, Program Development, Florida Department of Transportation, Tallahassee, FL. Florida Department of Transportation (FDOT). 1995. Connections: bringing Florida together, the 2020 Florida Transportation Plan. Florida Department of Transportation, Tallahassee, FL. Florida Department of Transportation (FDOT). 1994. Connections, bringing Florida together, the 2020 Florida Transportation Plan: Initial draft report. Florida Department of Transportation, Tallahassee, FL. Florida Department of Transportation (FDOT). 1992. Draft agency strategic plan, moving Florida toward the 21st century. Florida Department of Transportation, Tallahassee, FL. Florida Department of Transportation (FDOT). 1989. Manual of uniform minimum standards for design, construction and maintenance for streets and highways. Florida Department of Transportation, Tallahassee, FL. Florida Fish and Wildlife Conservation Commission (FFWCC). 2000. Unpublished data. Florida Fish and Wildlife Conservation Commission, Tallahassee, FL. Florida Gap Analysis Project (FGAP). 1998. Unpublished documentation bibliography of wildlife home ranges and territories. Florida Cooperative Fish and Wildlife Research Unit, Gainesville, FL.

PAGE 344

328 Florida Greenways Project (FGP). 1999. Executive summary of the Florida statewide greenways system planning project: recommendations for the physical design of a statewide greenways system. Florida Greenways Project, Department of Landscape Architecture, University of Florida, Gainesville, FL. Ford, S.G. 1980. Evaluation of highway deer kill mitigation on SIE/LAS-395. Final report no. FHWA/CA/TP-80-01. California Department of Transportation. Sacramento, CA. Forman, R.T.T. 1995. Land mosaics: The ecology of landscapes and regions. Cambridge University Press, New York, N.Y. Forman, R.T.T. 1983. Corridors in a landscape: their ecological structure and function. Ekologia (Czechoslovakia) 2:375-387. Forman, R.T.T. and L.E. Alexander. 1998. Roads and their major ecological effects. Annual Review of Ecology and Systematics 29:207-231. Forman, R.T. T. and M. Godron. 1986. Landscape Ecology. J. Wiley and Sons. NewYork, N.Y. Forman, R.T.T., D. Sperling, J.A. Bissonette, A.P. Clevenger, C.D. Cutshall, V.H. Dale, L. Fahrig, R. France, C.R. Goldman, K. Heanue, J.A. Jones, F.J. Swanson, T. Turrentine and T.C. Winter. 2003. Road ecology: Science and solutions. Island Press, Washington, D.C. Forys, E.A., C.R. Allen and D.P. Wojcik. 2002. Influence of the proximity and amount of human development and roads on the occurrence of the red imported fire ant in the lower Florida Keys. Biological Conservation 108:27-33. Foster, M.L. and S.R. Humphrey. 1995. Use of highway underpasses by Florida panthers and other wildlife. Wildlife Society Bulletin 23:95-100. Foster, M.L. and S.R. Humphrey. 1992. Effectiveness of wildlife crossings in reducing animal/auto collisions on Interstate 75, Big Cypress Swamp, Florida. Florida Game and Freshwater Fish Commission, Tallahassee, FL. Franklin, J.F. 1993. Preserving biodiversity: species, ecosystems, or landscapes? Ecological Applications 3:202-205. Friedman, D.S. 1997. Nature as infrastructure: The National Ecological Network and wildlife crossing structures in the Netherlands. Report no. 138. DLO Winand Staring Centre, Wageningen, The Netherlands.

PAGE 345

329 Fusari, M. 1982. Feasibility of a highway crossing system for desert tortoises. Contract report FHWA/CA/TP-81/1. California Department of Transportation, Sacramento, CA. Gaby R. 1987. Utilization of human-altered habitats by American crocodiles in southern Florida. In R.R. Odum, K.A. Riddleberger, and J.C. Ozier, editors. Proceedings of the 3rd Southeastern Nongame and Endangered Wildlife Symposium. University of Georgia, Athens, GA. Garland, T. Jr. and W.G. Bradley. 1984. Effects of a highway on Mojave Desert rodent populations. American Midland Naturalist 111:47-56. Garrett, P. and K. Gordon. 2003. Mule deer response to an underpass in Nugget Canyon, Wyoming. Paper presented at Making Connections, ICOET 2003, International Conference on Ecology and Transportation, August 24-29, Lake Placid, N.Y. Center for Transportation and the Environment, North Carolina State University, Raleigh, N.C. Gelbard, J.L. and J. Belnap. 2003. Roads as conduits for exotic plant invasions in a semiarid landscape. Conservation Biology 17:420-432. Gibbs, J.P. 1998. Amphibian movements in response to forest edges, roads, and streambeds in southern New England. Journal of Wildlife Management 62:584-589. Gibbs, J.P. and W.G. Shriver. 2002. Estimating the effects of road mortality on turtle populations. Conservation Biology 16:1647-1652. Gilbert, T. 1996. An overview of black bear roadkill in Florida 1976-1995. In G. Evink, P. Garrett, D. Zeigler, and J. Berry, editors. Trends in addressing transportation-related wildlife mortality. Florida Department of Transportation, Tallahassee, FL. Gilbert, T. and J. Wooding. 1994. Chronic roadkill problem areas for black bear in Florida: A preliminary report of transportation impacts on wildlife. Florida Game and Freshwater Fish Commission, Tallahassee, FL. Greer, G.E. and M.D. Farrell. 1988. Investment analysis for real estate decisions, 2nd edition. Longman Financial Services Publishing, Chicago, IL. Griffiths, M. and C.P. Van Schaik. 1993. The impact of human traffic on the abundance and activity periods of Sumatran rainforest wildlife. Conservation Biology 7:623-626. Gunn, J.M. and R. Sein. 2000. Effects of forestry roads on reproductive habitat and exploitation of lake trout (Salvelinus namaycush) in three experimental lakes. Canadian Journal of Fisheries and Aquatic Sciences 57(supplement 2):97-104.

PAGE 346

330 Haapoja, M. 1994. Conservation easements: are they for you? American Forests 100(1/2):29-32. Hall, P.M., L.D. Harris, and T.S. Hoctor. 1994. Integrating outdoor recreation and wildlife conservation with rights-of-way/greenway corridors. Pages 195-229 in K.G. Hay, editor. Greenways, wildlife and natural gas pipeline corridors: New partnerships for multiple use. The Conservation Fund, Arlington, VA. Hannah, D.A. 1992. Fragmentation in the Ocala National Forest: a GIS analysis. M.S. Thesis. School of Forest Resources and Conservation, University of Florida, Gainesville, FL. Harding, B.D. 1986. Short-eared owl mortality on roads. British Birds 79:403-404. Harris, L.D. 1988a. Edge effects and conservation of biotic diversity. Conservation Biology 2:330-332. Harris, L.D. 1988b. Landscape linkages: the dispersal corridor approach to wildlife conservation. Pages 595607 in Transactions of the 53rd North American Wildlife and Natural Resources Conference. Harris, L.D. 1985. Conservation corridors: A highway system for wildlife. ENFO 85-11. Environmental Information Center of the Florida Conservation Foundation, Inc., Winter Park, FL. Harris, L.D. 1984. The fragmented forest: Island biogeography theory and the preservation of biotic diversity. University of Chicago Press, Chicago, IL. Harris, L.D. and P.B. Gallagher. 1989. New initiatives for wildlife conservation: The need for movement corridors. Pages 11-24 in G. Mackintosh, editor. In defense of wildlife: Preserving communities and corridors. Defenders of Wildlife, Washington, D.C. Harris, L.D. and W.P. Cropper, Jr. 1992. Between the devil and the deep blue sea: Implications of climate change for Florida's fauna. Pages 309-324 in R. Peters and T. Lovejoy, editors. Global warming and biological diversity. Yale University Press, New Haven, CT. Harris, L.D., T. Hoctor, D. Maehr, and J. Sanderson. 1996. The role of networks and corridors in enhancing the value and protection of parks and equivalent areas. Pages 173-197 in R.G. Wright, editor. National parks and protected areas: Their role in environment protection. Blackwell Science, Cambridge, MA.

PAGE 347

331 Harris, L.D. and J. Scheck. 1991. From implications to applications: the dispersal corridor principle applied to the conservation of biological diversity. Pages 189-220 in D.A. Saunders and R.J. Hobbs, editors. Nature Conservation 2: The Role of Corridors. Surrey Beatty and Sons, Chipping Norton, Australia. Harris, L.D. and G. Silva-Lopez. 1992. Forest fragmentation and the conservation of biological diversity. Pages 197-237 in P. Fiedler and S. Jain, editors. Conservation biology, the theory and practice of nature conservation, preservation and management. Chapman and Hall, New York, N.Y. Harrison, P.D. and M.I. Dyer. 1984. Lead in mule deer forage in Rocky Mountain National Park, Colorado, U.S.A. Journal of Wildlife Management 48:510-517. Hellgren, E.C. and D.S. Maehr. 1992. Habitat fragmentation and black bears in the eastern United States. Pages 154-165 in E.P. Orff, editor. Proceedings of the 11th Eastern Black Bear Workshop, Waterville Valley, N.H. New Hampshire Department of Fish and Game, Concord, N.H. Hellman, R.E. and S.R. Telford. 1956. Notes on a large number of red-bellied mud snakes, Farancia a. abacura, from north central Florida. Copeia 1956:257-258. Hels, T. and E. Buchwald. 2001. The effect of road kills on amphibian populations. Biological Conservation 99:331-340. Herzer, W., W.M. Johnstone, R.L. MacDonald and C. Goodbrand. 1991. Application of GIS to the planning and preliminary design of the Vancouver Island Highway. Pages 19-28 in Proceedings of the 2nd National Specialty Conference on Civil Engineering Applications of Remote Sensing and GIS, Washington, D.C. American Society of Civil Engineers, New York, N.Y. Hewitt, D., A. Cain, V. Tuovilla, D. Shindle and M. Tewes. 1998. Impacts of an expanded highway on ocelots and bobcats in southern Texas and their preferences for highway crossings. Pages 126-134 in G. Evink, P. Garrett, D. Zeigler and J. Berry, editors. Proceedings of the International Conference on Wildlife Ecology and Transportation, Feb. 10-12, 1998, Ft. Myers, FL. Florida Department of Transportation, Tallahassee, FL. Hobbs, R.J. and A.J.M. Hopkins. 1991. The role of conservation corridors in a changing climate. Pages 281-290 in D.A. Saunders and R.J. Hobbs, editors. Nature conservation 2: The role of corridors. Surrey Beatty and Sons, Chipping Norton, Australia. Hoctor, T.S., M.H. Carr and P.D. Zwick. 2000. Identifying a linked reserve system using a regional landscape approach: the Florida ecological network. Conservation Biology 14:984-1000.

PAGE 348

332 SAS software. SAS Institute, Inc. Cary, N.C. 558 pp. Hoffman, D.J., G.H. Heinz, L. Sileo, D.J. Audet, J.K. Campbell, L.J. LeCaptain and H.H. Obrecht III. 2000. Developmental toxicity of lead-contaminated sediment in Canada geese Branta canadensis. Journal of Toxicology and Environmental Health, Part A 59:235-252. Holbrook, H.T. and M.R. Vaughan. 1985. Influence of roads on turkey mortality. Journal of Wildlife Management 49:611-614. Hunt, A., H.J. Dickens, and R.J. Whelan. 1987. Movement of mammals through tunnels under railway lines. Australian Zoology 24:89-93. Hyder, L.R. 1982. Area access management: enforcement of road closure agreements. Pages 26-27 in J.M. Peek and L. Nelson Jr., editors. Proceedings of the Forest Access Symposium, Nov 12-13, 1982, Coeur d'Alene, ID. University of Idaho, Moscow, ID. Hynson, J., P. Adamus, S. Tibbetts and R. Darnell. 1982. Handbook for protection of fish and wildlife from construction of farm and forest roads: Best management practices for building activities associated with the discharge of dredged or fill material. Project report no: FWS/OBS-82/18. Energy and Land Use Team, Office of Biological Services, Fish and Wildlife Service, U.S. Department of the Interior, Washington, D.C. Ilichev, V.D., I.I. Kamenskii and O.L. Silaeva. 1995. Ecological and technogenous factors of noise pollution of natural habitats of birds. Russian Journal of Ecology 26:345-348. Ims, R. 1995. Movement patterns related to spatial structure. Pages 85-109 in L. Hansson, L. Fahrig, and G. Merriam, editors. Mosaic landscapes and ecological processes. Chapman and Hall. London. Iverson, L.R. and A. Prasad. 1998. Predicting abundance of 80 tree species following climate change in the eastern United States. Ecological Monographs 68:465-485. Jackson, S. 1996. Underpass systems for amphibians. In G. Evink, P. Garrett, D. Zeigler and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Florida Department of Transportation, Tallahassee, FL. Kauffeld, C. 1957. Snakes and snake hunting. Hanover House. Garden City, N.Y. Kautz, R. 1992. Florida wildlife habitat: The last 50 years. Florida Wildlife 45(9/10):2-6. Kendrick, B. 1964. Florida trails to turnpikes, 1914-1964. University of Florida Press, Gainesville, FL. Khattree, R. and D.N. Naik. 2000. Multivariate data reduction and discrimination with

PAGE 349

333 Kozel, R.M. and E.D. Fleharty. 1979. Movement of rodents across roads. The Southwestern Naturalist 24:239-248. Krein, A. and M. Schorer. 2000. Road runoff pollution by polycyclic aromatic hydrocarbons and its contribution to river sediments. Water Research 34:4110-4115. Krikowski, L. 1989. The light and dark zones: two examples of tunnel and fence systems. Pages 89-91 in T. Langton, editor. Amphibians and roads. ACO Polymer Products, Bedfordshire, U.K. Kroodsma, R. 1987. Edge effect on breeding birds along power-line corridors in east Tennessee. American Midland Naturalist 118:275-283. Kruuk, H. and J.W.H. Conroy. 1991. Mortality of otters (Lutra lutra) in Shetland. Journal of Applied Ecology 28:83-94. Kuennen, T. 1989. New Jersey's I-78 preserves mountain habitat. Roads and Bridges 127(2):69-73. Lammers, G.W. and F.J. van Zadelhoff. 1996. The Dutch national ecological network. Ch 9 in Perspectives on ecological networks. P. Nowicki, G. Bennett, D. Middleton, S. Rientjes and R. Walters, editors. Publication series on man and nature, vol. 1. European Centre for Nature Conservation. Land, D. and M. Lotz. 1996. Wildlife crossing designs and use by Florida panthers and other wildlife in southwest Florida. In G. Evink, P. Garrett, D. Zeigler and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Florida Department of Transportation, Tallahassee, FL. Langton, T.E.S. 1989a. Tunnels and temperature: results from a study of a drift fence and tunnel system for amphibians at Henley-on-Thames, Buckinghamshire, England. Pages 145-150 in Langton, T.E.S., editor. Amphibians and roads. ACO Polymer Products Ltd. Bedfordshire, England. Langton, T.E.S., editor. 1989b. Amphibians and roads. ACO Polymer Products Ltd. Bedfordshire, England. Lalo, J. 1987. The problem of roadkill. American Forests 93(9/10):50-53, 72. Leedy, D.L. 1975. Highway-wildlife relationships. Vol. 1. A state of the art report. FHWA-RD-76-4. Federal Highway and Waterway Administration, Washington, D.C. Lindenmayer, DB. M.L. Pope and R.B. Cunningham. 1999. Roads and nest predation: An experimental study in a modified forest system. Emu 99:148-152.

PAGE 350

334 Little, C.E. 1990. Greenways for America. Johns Hopkins University Press, Baltimore, MD. LoBuono, J.P. 1988. Alligator Alley: Protecting natural habitat. Journal of the Florida Engineering Society 41(2):14-16. Logan, T. and G. Evink. 1985. Safer travel for the Florida panther. Naturalist 6(3/4):6-7. Lopez, R.R., M.E.P. Vieira, N.J. Silvy, P.A. Frank, S.W. Whisenant, and D.A. Jones. 2003. Survival, mortality, and life expectancy of Florida Key deer. Journal of Wildlife Management 67:34-45. Lyon, J.L. 1983. Road density models describing habitat effectiveness for elk. Journal of Forestry 80:592-595, 613. Mace, R.D. and J.S. Waller. 1996. Grizzly bear distribution and human conflicts in Jewel Basin Hiking Area, Swan Mountains, Montana. Wildlife Society Bulletin 24:461-467. Mader, H.J. 1984. Animal habitat isolation by roads and agricultural fields. Biological Conservation 29:81-96. Madsen, A.B. 1996. Otter Lutra lutra mortality in relation to traffic, and experience with newly established fauna passages at existing road bridges. Lutra 39:76-88. Maehr, D.S. 1997a. The comparative ecology of bobcat, black bear, and Florida panther in south Florida. Bulletin of the Florida Museum of Natural History 40:1-176. Maehr, D.S. 1997b. The Florida panther, life and death of a vanishing carnivore. Island Press, Washington, D.C. Maehr, D.S. 1984. Distribution of black bears in eastern North America. Page 74 in D.S. Maehr and J.R. Brady, editors. Proceedings of the 7th Eastern Workshop on Black Bear Management and Research, Homosassa, FL. Florida Game and Freshwater Fish Commission, Gainesville, FL. Maehr, D.S. and J.A. Cox. 1995. Landscape features and panthers in Florida. Conservation Biology 9:1008-1019. Maehr, D.S., E.D. Land, and M.E. Roelke. 1991. Mortality patterns of panthers in southwest Florida. Proceedings of the Annual Conference of the Southeastern Association of State Fish and Wildlife Agencies 45:201-207. Maehr, D.S., J.N. Layne, E.D. Land, J.W. McCown and J.C. Roof. 1988. Long distance movements of a Florida black bear. Florida Field Naturalist 16:1-6.

PAGE 351

335 Malanson, G.P. 1993. Riparian landscapes. Cambridge University Press. Cambridge, U.K. Mansergh, I.M. and D.J. Scotts. 1989. Habitat continuity and social organization of the mountain pygmy-possum restored by tunnel. Journal of Wildlife Management 53:701-707. Marini, M.A., S.K. Robinson and E.J. Heske. 1995. Edge effects on nest predation in the Shawnee-National-Forest, southern Illinois. Biological Conservation 74:203-213. Mattson, D.J., R.R. Knight and B.M. Blanchard. 1987. The effects of developments and primary roads on grizzly bear habitat use in Yellowstone National Park, Wyoming. International Conference on Bear Research and Management Series 7:259-273. Maurer, M. 1999. Development of a community-based, landscape-level terrestrial mitigation decision support system for transportation planners. Pages 99-109 in G. Evink, P. Garrett and D. Zeigler, editors. Proceedings of the 3rd International Conference on Wildlife Ecology and Transportation, September 13-16, 1999 in Missoula, MT. Florida Department of Transportation, Tallahassee, FL. McCown, J.W. and T.H. Eason. 2001. Black bear movements and habitat use relative to roads in Ocala National Forest: preliminary findings. Pages 397-404 in Proceedings of the International Conference on Ecology and Transportation, September 24-28, 2001, Keystone, Colorado. Center for Transportation and the Environment, North Carolina State University, Raleigh, N.C. McDonald, W.H. 1937. Highways of Florida. Division of Statewide Highway Planning Survey, State Road Department of Florida, Tallahassee, FL. McGarigal, K., S. Cushman and S. Stafford. 2000. Multivariate statistics for wildlife and ecology research. Springer-Verlag, Inc., New York, N.Y. McHarg, I. 1971. Design with nature. Doubleday-Natural History Press, Garden City, N.Y. McLellan, B.N. and D.M. Shackleton. 1988. Grizzly bears and resource-extraction industries: effects of roads on behaviour, habitat use and demography. Journal of Applied Ecology 25:451-460. Mech, L.D. 1989. Wolf population survival in an area of high road density. American Midland Naturalist 121:387-389. Mech, L.D., S.H. Harris, G.L. Radde, and W.J. Paul. 1988. Wolf distribution and road density in Minnesota. Wildlife Society Bulletin 16:85-87.

PAGE 352

336 Wildlife and Island Press, Washington, D.C. Molles, M.C. Jr. 1999. Ecology: concepts and applications. WCB McGraw-Hill, Boston, MA. Muskett, C.J., and M.P. Jones. 1980. The dispersal of lead, cadmium, and nickel from motor vehicles and effects on roadside invertebrate macrofauna. Environmental Pollution 23:231-242. National Research Council (NRC). 2002. Surface transportation environmental research: a long-term strategy. National Academy Press, Washington, D.C. National Research Council (NRC). 1997. Toward a sustainable future: Addressing the long-term effects of motor vehicle transportation on climate and ecology. National Academy Press, Washington, D.C. National Research Council (NRC). 1993. Setting priorities for land conservation. Committee on Scientific and Technical Criteria for Federal Acquisition of Lands for Conservation, Board on Environmental Studies And Toxicology, Commission on Life Sciences, National Research Council. National Academy Press, Washington, D.C. Nellemann, C. and R.D. Cameron. 1998. Cumulative impacts of an evolving oil-field complex on the distribution of calving caribou. Canadian Journal of Zoology 76:1425-1430. Norden, M. 1990. Amherst's salamander tunnels. Reptile and Amphibian Magazine 1990(9/10):38-41. Norman, T., A. Finnegan and B. Lean. 1998. The role of fauna underpasses in New South Wales. Pages 195-208 in G. Evink, P. Garrett, D. Zeigler and J. Berry, editors. Proceedings of the International Conference on Wildlife Ecology and Transportation, February 10-12. Florida Department of Transportation, Tallahassee, FL. Norse, E.A., K.L. Rosenbaum, D.S. Wilcove, B.A. Wilcox, W.H. Romme, D.W. Johnston and M.L. Stout. 1986. Conserving biological diversity in our National forests. The Wilderness Society, Washington, D.C. Noss, R.F. 1995. Maintaining ecological integrity in representative reserve networks. Discussion Paper. WWF Canada, Toronto and WWF United States, Washington, D.C. Noss, R.F. 1993. Wildlife corridors. Ch. 3 in D.S. Smith and P.C. Hellmund, editors. Ecology of Greenways: Design and Function of Linear Conservation Areas. University of Minnesota Press, St. Paul, MN. Noss, R.F. 1991. Landscape connectivity: different functions at different scales. Pages 27-39 in W.E. Hudson, editor. Landscape linkages and biodiversity. Defenders of

PAGE 353

337 Noss, R.F. 1987. Corridors in real landscapes: A reply to Simberloff and Cox. Conservation Biology 1:159-164. Noss, R.F. 1983. A regional landscape approach to maintain diversity. Bioscience 33:700-706. Noss, R.F. and A.Y. Cooperider. 1994. Saving nature's legacy: protecting and restoring biodiversity. Defenders of Wildlife and Island Press, Washington, D.C. Noss, R.F. and L.D. Harris. 1990. Habitat connectivity and the conservation of biological diversity: Florida as a case history. In Proceedings of the 1989 Society of American Foresters National Convention, September 24-27, Spokane, WA. Society of American Foresters, Bethesda, MD. Noss, R.F. and L.D. Harris. 1986. Nodes, networks, and MUMs: Preserving diversity at all scales. Environmental Management 10:299-309. Oregon Department of Transportation (ODOT). 2003. Transportation: The key to development, 1800-1940 timeline. http://www.odot.state.or.us/ssbbsrvcpublic/rm/history/~appendix_n.htm Orlando-Orange County Expressway Authority (OOCEA). 2003. Expressway 2025 master plan. http://www.expresswayauthority.com/futureplans/plan_expresswaymaster.html Orlando-Orange County Expressway Authority (OOCEA). 2000. 2025 master plan. Expressway Navigator 2000(1):2. http://www.expresswayauthority.com/pdf/Navigator_summer00.pdf Opdam, P., G. Rijsdijk and F. Hustings. 1985. Bird communities in small woods in an agricultural landscape: Effect of area and isolation. Biological Conservation 34:333-352. Oxley, D.J., M.B. Fenton and G.R. Carmody. 1974. Effects of roads on small mammals. Journal of Applied Ecology 11:51-59. Pacquet, P. and C. Callahan. 1996. Effects of linear developments on winter movements of gray wolves in the Bow River Valley of Banff National Park, Alberta. In G. Evink, P. Garrett, D. Zeigler, and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Florida Department of Transportation, Tallahassee, FL. Pafko, F. and B. Kovach. 1996. Experience with deer reflectors. In G.L. Evink, P. Garrett, D. Zeigler and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Florida Department of Transportation, Tallahassee, FL.

PAGE 354

338 Paton, P.W.C. 1994. The effect of edge on avian nest success: how strong is the evidence? Conservation Biology 8:17-26. Peterson, R.O. 1977. Wolf ecology and prey relationships on Isle Royale. Scientific Monograph Series No. 11. National Park Service, U.S. Department of the Interior, Washington, D.C. Pickett, S.T.A. and M.L. Cadenasso. 1995. Landscape ecology: spatial heterogeneity in ecological systems. Science 269:331-334. Pojar, T.M., R.A. Prosence, D.F. Freed and T.N. Woodard. 1975. Effectiveness of a lighted, animated deer crossing sign. Journal of Wildlife Management 39:87-91. Powell, R.A., J.W. Zimmerman, D.E. Seaman and J.F. Gilliam. 1996. Demographic analyses of a hunted black bear population with access to a refuge. Conservation Biology 10:224-234. Ratcliffe, E.J. 1974. Through the badger gate. G. Bell and Sons, Ltd., London, England. Rauch, S., G.M. Morrison, M. Motelica-Heino, O.F.X. Donard and M. Muris. 2000. Elemental association and fingerprinting of traffic-related metals in road sediments. Environmental Science and Technology 35:3119-3123. Reed, D.F., T.N. Woodard and T.M. Pojar. 1975. Behavioral response of mule deer to a highway underpass. Journal of Wildlife Management 39:361-367. Reeve, A.F. and S.H. Anderson. 1993. Ineffectiveness of Swareflex reflectors at reducing deer-vehicle collisions. Wildlife Society Bulletin 21:127-132. Reijnen, R., R. Foppen and C.J. Braak. 1995. The effects of car traffic on breeding bird populations in woodlands III: reduction of density in relation to the proximity of main roads. Journal of Applied Ecology 32:187-202. Reijnen, R., R. Foppen and G. Veenbaas. 1997. Disturbance by traffic of breeding birds: evaluation of the effect and considerations in planning and managing road corridors. Biodiversity Conservation 6:567-581. Rich, A., D. Dobkin and L. Niles.1994. Defining forest fragmentation by corridor width: The influence of narrow forest-dividing corridors on forest-nesting birds in southern New Jersey. Conservation Biology 8:1109-1121. Ritter, J. 1998. Tourist traffic turning park into parking lot. USA Today, August 3, 1998, Section A, Pages 1-2. Gannett Company, Inc., McLean, VA.

PAGE 355

339 Robertson, W.B. Jr. and J.A. Kushlan. 1974. The southern Florida avifauna. In Memoir 2: Environments of south Florida, present and past. Miami Geologic Society, Miami, FL. Rodgers, J.A. Jr. and H.T. Smith. 1995. Set-back distances to protect nesting bird colonies from human disturbance in Florida. Conservation Biology 9:89-99. Rodriguez, A., G. Crema and M. Delibes. 1996. Use of non-wildlife passages across a high speed railway by terrestrial vertebrates. Journal of Applied Ecology 33:1527-1540. Rogers, C.M. and M.J. Caro. 1998. Song sparrows, top carnivores and nest predation: A test of the mesopredator release hypothesis. Oecologia (Berlin) 116:227-233. Romin, L.A. and J.A. Bissonette. 1996. Temporal and spatial distribution of highway mortality of mule deer on newly constructed roads at Jordanelle Reservoir, Utah. Great Basin Naturalist 56:1-11. Romin, L.A. and L.B. Dalton. 1992. Lack of response by mule deer to wildlife warning whistles. Wildlife Society Bulletin 20:382-384. Roof, J. and J. Wooding. 1996. Evaluation of SR 46 wildlife crossing. Technical report no. 54. FLDOT FL-ER-61-96. Florida Cooperative Fish and Wildlife Research Unit, U.S. Biological Service, Gainesville, FL. Rosenberg, D.K., B.R. Noon, and E.C. Meslow. 1997. Biological corridors: form, function, and efficacy. Bioscience 47:677-687. Rossell, C.R. Jr., and J.A. Litvaitis. 1994. Application of harvest data to examine responses of black bears to land-use changes. International Conference on Bear Research and Management Series 9:275-281. Rudnicky, Tamia C. and Malcolm L. Hunter, Jr. 1993. Avian nest predation in clearcuts, forests, and edges in a forest-dominated landscape. Journal of Wildlife Management 57:358-364. Ruediger, B. 2001. High, wide and handsome: designing more effective wildlife and fish crossings for roads and highways. Pages 509-516 in A Time for Action: Proceedings of the International Conference on Ecology and Transportation, September 24-28, 2001, Keystone, CO. Center for Transportation and the Environment, North Carolina State University, Raleigh, N.C.

PAGE 356

340 Conservation Biology 1:63-71. Ruediger, R. and B. Ruediger. 1999. The effects of highways on trout and salmon rivers and streams in the western U.S. Pages 151-159 in G. Evink, P. Garrett and D. Zeigler, editors. Proceedings of the 3rd International Conference on Wildlife Ecology and Transportation, September 13-16, 1999, Missoula, MT. Florida Department of Transportation, Tallahassee, FL. Ryser, J. and K. Grossenbacher. 1989. A survey of amphibian preservation at roads in Switzerland. Pages 7-14 in T.E.S. Langton, editor. Amphibians and roads. ACO Polymer Products Ltd., Bedfordshire, England. Salisbury, R.W. 1993. The fragmentation of wildlife conservation areas by linear structures. M.S. thesis. Graduate School of the Environment. Macquarie University, Sydney, Australia. Sanderson E.W., M. Jaiteh, M.A. Levy, K.H. Redford, A.V. Wannebo, and G. Woolmer. 2002. The human footprint and the last of the wild. Bioscience 52:891-904. Sargeant, A.B., and Forbes, J.E. 1973. Mortality among birds, mammals, and certain snakes on 17 miles of Minnesota roads. The Loon 45: 4-7. Sautter, U. 1986. Government and unemployment: The use of public works before the New Deal. The Journal of American History 73:59-86. Schaefer, J.M and M.T. Brown. 1992. Designing and Protecting River Corridors for Wildlife. Rivers 3:14-27. Schaefer, J.M. and D.J. Smith. 2000. Ecological characterization of identified high priority highway-ecological interface zones including the inventory and evaluation of existing Florida Department of Transportation highway facilities within these zones. Final report, contract no. B-B120, task #1 (amendment no. 3). Florida Department of Transportation, Tallahassee, FL. Schafer, J.A. and S.T. Penland. 1985. Effectiveness of swareflex reflectors in reducing deer-vehicle accidents. Journal of Wildlife Management 49:774-776. Sharp, D. 1999. In south Florida, the return of the docile crocodile. USA Today, August 11, 1999. Gannett Company, Inc., McLean, VA. Sheehy, D.M. 2001. Reconditioning and stabilization of unpaved roads for reducing road maintainance and impacts on fisheries habitat. Pages 121-131 in A Time for Action: Proceedings of the International Conference on Ecology and Transportation, September 24-28, 2001, Keystone, CO. Center for Transportation and the Environment, North Carolina State University, Raleigh, N.C. Simberloff, D. and J. Cox. 1987. Consequences and costs of conservation corridors.

PAGE 357

341 Prairie Basin. Florida Department of Transportation, Lake City, FL. Singer, F.J. and J.L. Doherty. 1985. Managing mountain goats at a highway crossing. Wildlife Society Bulletin 13:469-477. Smith, D.J. 1999. Identification and prioritization of ecological interface zones on state highways in Florida. Pages 209-229 in G. Evink, P. Garrett and D. Zeigler, editors. Proceedings of the Third International Conference on Wildlife Ecology and Transportation, September 13-16, 1999 in Missoula, MT. Florida Department of Transportation, Tallahassee, FL. Smith, D.J. 1996. The direct and indirect impacts of highways on the vertebrates of Payne's Prairie State Preserve. Technical report no. FL-ER-62-96. Florida Department of Transportation, Tallahassee, FL. Smith, D.J. and L.D. Harris. 1998. Identification of Highway-Greenway Interface Zones. Subcontract final report phase 2 (Task 9), contract no. 97070173. Department of Urban and Regional Planning, University of Florida, Gainesville. Smith, D.J., L.D. Harris and F.J. Mazzotti. 1998. Highway wildlife relationships (development of a decision-based wildlife underpass road project prioritization model on GIS with statewide application). Final report, contract no. B-9943. Florida Department of Transportation, Tallahassee, FL. Smith, D.J., L.D. Harris and F.J. Mazzotti. 1996. A landscape approach to examining the impacts of roads on the ecological function associated with wildlife movement and movement corridors: problems and solutions. In G. Evink, P. Garrett, D. Zeigler and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Florida Department of Transportation, Tallahassee, FL. Smith, D.S. Greenway case studies. 1993. Ch. 7 in D.S. Smith and P.C. Hellmund, editors. Ecology of greenways: Design and function of linear conservation areas. University of Minnesota Press, St. Paul, MN. Smith, D.S. and P.C. Hellmund., editors. 1993. Ecology of greenways: Design and function of linear conservation areas. University of Minnesota Press, St. Paul, MN. Smith, L. and C.K. Dodd, Jr. 1999. Determination of the effectiveness of wildlife barriers and underpasses on U.S. Highway 441 across Paynes Prairie State Preserve, Alachua County, Florida. Phase I final report, contract no. BB-854. Florida Department of Transportation, Tallahassee, FL. Soul, M.E. 1991. Theory and strategy. Pages 91-104 in W.E. Hudson, editor. Landscape linkages and biodiversity. Defenders of Wildlife and Island Press, Washington, D.C. Southall, P.D. 1991. The relationship between wildlife and highways in the Payne's

PAGE 358

342 the U.S. House of Representatives. http://www.census.gov/main/www/cen2000.html. South Florida Water Management District (SFWMD). 1992. Surface water improvement plan for the Everglades; supporting information document. South Florida Water Management District, West Palm Beach, FL. Spackman, S.C. and J.W. Hughes. 1995. Assessment of minimum stream corridor width for biological conservation: species richness and distribution along mid-order streams in Vermont, USA. Biological Conservation 71:325-332. Spellerberg, I.F. 1998. Ecological effects of roads and traffic: a literature review. Global Ecology And Biogeography Letters 7:317-333. Stadler, S. 1987. Roadkills: reducing the death toll. The Animals Agenda 7(8):32-35. Stein, R.H. 1982. Oregon's cooperative road closure program. Pages 28-31 in J.M. Peek and L. Nelson Jr., editors. Proceedings of the Forest Access Symposium, Nov 12-13, 1982, Coeur d'Alene, ID. University of Idaho, Moscow, ID. Stiles, J.H. and R.H. Jones. 1998. Distribution of the red imported fire ant, Solenopsis invicta, in road and powerline habitats. Landscape Ecology 13:335-346. Stolzenburg, W. 1991. The fragment connection. Nature Conservancy 41:18-25. Suarez, A., K. Pfenning and S. Robinson. 1997. Nesting success of a disturbance-dependent songbird on different kinds of edges. Conservation Biology 11:928-935. Sunde, P., K. Overskaug and T. Kvam. 1998. Culling of lynxes Lynx lynx related to livestock predation in a heterogeneous landscape. Wildlife Biology 4:169-175. Sutherland, R.A. and C.A. Tolosa. 2001. Variation in total and extractable elements with distance from roads in an urban watershed, Honolulu, Hawaii. Water, Air and Soil Pollution 127:315-338. Tewes, M.E. and D.R. Blanton. 1998. Potential impacts of international bridges on ocelots and jaguarundis along the Rio Grande wildlife corridor. Pages 135-139 in G. Evink, P. Garrett and D. Zeigler, editors. Proceedings of the Second International Conference on Wildlife Ecology and Transportation, February 10-12, 1998, Ft. Myers, FL. Florida Department of Transportation, Tallahassee, FL. Theobald, D.M., J.R. Miller and N.T. Hobbs. 1997. Estimating the cumulative effects of development on wildlife habitat. Landscape and Urban Planning 39:25-36. Tyning, T. 1989. Amherst's tunneling amphibians. Defenders 64(9/10):20-23. United States Census Bureau (USCB). 2000. Resident population and apportionment of

PAGE 359

343 United States Department of the Interior (USDI). 1998. Unpublished data of documented road-mortality accounts of the Florida key deer. Florida Key Deer National Wildlife Refuge, Big Pine Key, FL. United States Fish and Wildlife Service (USFWS). 1995. Florida panther recovery plan, 2nd revision. U.S. Fish and Wildlife Service, Atlanta, GA. United States Fish and Wildlife Service (USFWS). 1985. Florida Key deer recovery plan, 1st revision. U.S. Fish and Wildlife Service, Atlanta, GA. van Buuren, M. and K. Kerkstra. 1993. The framework concept and the hydrological landscape structure: a new perspective in the design of multifunctional landscapes. Pages 219-243 in C. Vos and P. Opdam, editors. Landscape ecology of a stressed environment. Chapman and Hall, London. Van Dyke, F.G., R.H. Brocke, H.G. Shaw, B.B. Ackerman, T.P. Hemker and F.G. Lindzey. 1986. Reactions of mountain lions to logging and human activity. Journal of Wildlife Management 50:95-102. van Zadelhoff, E. and W. Lammers. 1995. The Dutch ecological network. Landschap 95:77-88. Veenbaas, G. and J. Brandjes. 1999. Use of fauna passages along waterways under highways. Pages 253-257 in G. Evink, P. Garrett and D. Zeigler, editors. Proceedings of the 3rd International Conference on Wildlife Ecology and Transportation, Sept. 13-16, 1999 in Missoula, MT. Florida Department of Transportation, Tallahassee, FL. Verkaar, H.J. 1988. The possible role of road verges and river dykes as corridors for the exchange of plant species between natural habitats. Pages 79-84 in R.F. Schreiber, editor. Connectivity in Landscape Ecology. Munstersche Geographische Arbeiten 29, Munster, Germany. Villano, D. 1993. New hope for the panther. Defenders 68(3/4):29-35. Viskari, E.L., M. Vartiainen and P. Pasanen. 2000. Seasonal and diurnal variation in formaldehyde and acetaldehyde concentrations along a highway in eastern Finland. Atmospheric Environment 34:917-923. Vos, C.C. and J.P. Chardon. 1998. Effects of habitat fragmentation and road density on the distribution pattern of the moor frog Rana arvalis. Journal of Applied Ecology 35:44-56.

PAGE 360

344 Wardell-Johnson, G. and J.D. Roberts. 1991. The survival status of the Geocrinia rosea (Anura: Myobatrachidae) complex in riparian corridors: biogeographical implications. Pages 167-175 in D. Saunders and R. Hobbs, editors. Nature Conservation 2: The Role of Corridors. Surrey Beatty and Sons, Chipping Norton, Australia. Watts, B. 1993. Environmental policy for state transportation facilities. EnviroNews 1:1. Weighell, T. 1997. Another roadside attraction (impact of pollution on plant diversity). New Scientist 156(2104):54. Wiebe, K. 1997. Managing public and private land through partial interests. Contemporary Economic Policy 15:35-43. Wilcove, D.S. 1985. Nest predation in forest tracts and the decline of migratory songbirds. Ecology 66:1211-1214. Wilcove, D.S., C.H. McLellan and A.P. Dobson. 1986. Habitat fragmentation in the temperate zone. Pages 237-256 in M. Soule, editor. Conservation Biology: the Science of Scarcity and Diversity. Sinauer Assoc., Sunderland, MA. Wilcox, D.A. 1989. Migration and control of purple loosestrife (Lythrum-salicaria L.) along highway corridors. Environmental Management 13:365-370. Wilkins, K.T. 1982. Highways as barriers to rodent dispersal. The Southwestern Naturalist 27:459-460. Williams K., K.C. Ewel, R.P. Stumpf, F.E. Putz, T.W. Workman. 1999. Sea-level rise and coastal forest retreat on the west coast of Florida, USA. Ecology 80:2045-2063. Williams, M. 1990. Clearing of the forests. Pages 146-168 in M.P. Conzen, editor. The making of the American landscape. Unwin Hyman, Boston, MA. Winter, T.C. 2001. Water, sediment, and chemical flow related to roads. Pages 14-15 in A Time for Action: Proceedings of the International Conference on Ecology and Transportation, September 24-28, 2001, Keystone, CO. Center for Transportation and the Environment, North Carolina State University, Raleigh, N.C. Wooding, J.B. and J.R. Brady. 1987. Black bear roadkills in Florida. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 41:438-442. Wooding, J.B. and T. Hardisky. 1994. Home range, habitat use, and mortality of black bears in north-central Florida. International Conference on Bear Research and Management Series 9:349-356.

PAGE 361

345 Wooding, J.B. and R.C. Maddrey. 1994. Impacts of roads on black bears. Pages 124-129 in M.R. Pelton, R.A. Beausoleil and F.T. van Manen, editors. Proceedings of the 12th Eastern Workshop on Black Bear Research and Management, April 2-5, 1994, Gatlinburg, TN. Wright, J.B. 1994. Designing and applying conservation easements. (Planner's Notebook). Journal of the American Planning Association 60:380-388. Wright, J.B. 1993. Conservation easements: an analysis of donated development rights. (Planner's Notebook). Journal of the American Planning Association 59:487-493. Yahner, R.H. 1988. Changes in wildlife communities near edges. Conservation Biology 2:333-339. Yahner, R.H. and C.G. Mahan. 1997. Effects of logging roads on depredation of artificial ground nests in a forested landscape. Wildlife Society Bulletin 25:158-162. Yanes, M., J.M. Velasco and F. Suarez. 1995. Permeability of roads and railways to vertebrates: the importance of culverts. Biological Conservation 71:217-222. Yost, A.C. and R.G. Wright. 2001. Moose, caribou, and grizzly bear distribution in relation to road traffic in Denali National Park, Alaska. Arctic 54:41-48. Yousufzai, A.H.K., D.R. Hashmi, F. Ahmed and K. Durrani. 2001. Heavy metal accumulation in road side vegetation of urban areas of Karachi. Pakistan Journal of Scientific and Industrial Research 44:29-32.

PAGE 362

BIOGRAPHICAL SKETCH Daniel J. Smith was born September 20, 1957. He graduated from Maynard Evans High School in Orlando, Florida in 1975, and attended the University of Florida receiving a Bachelor of Science degree in horticulture from the College of Agriculture in 1983. Continuing his education at the graduate level at the University of Florida, Daniel obtained a Master of Urban and Regional Planning degree in 1989 from the College of Architecture. After 6 years practicing environmental planning at the local government level and successfully receiving certification from the American Institute of Certified Planners, Daniel returned to graduate school to pursue a career in ecological science. After receiving a second masters degree in 1995 (Master of Forest and Resource Conservation from the College of Agriculture and Life Sciences at the University of Florida), he continued in the Department of Wildlife Ecology and Conservation at the University of Florida to pursue the Doctor of Philosophy degree; which was awarded December 2003. 346