<%BANNER%>

Urban lakes and waterbirds

University of Florida Institutional Repository

PAGE 1

URBAN LAKES AND WATERBIRDS: EFFECTS OF DEVELOPMENT ON DISTRIBUTION AND BEHAVIOR By ASHLEY H. TRAUT A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2003

PAGE 2

To Kate, for your immeasurable patience, support, and love. You inspire me beyond words. And to the birds, who fill my dreams with hope.

PAGE 3

iii ACKNOWLEDGMENTS I wish to thank the Department of Fisheries and Aquatic Sciences for their financial and logistical support of this project, and for giving me the opportunity to work with Florida LAKEWATCH. Mark Hoyer and Eric Schulz were particularly generous with their time and ideas during the development of this project and provided me with invaluable advice from different perspectives. My advisor, Mark Hostetler, deserves far more thanks than a few sentences can convey. His tireless brainstorming, seemingly endless editing, and frequent encouragement made my graduate experience a truly valuable and enjoyable experience. His blending of science with public outreach has been a great inspiration to me, and his tremendously broad range of interests and scholarly pursuits has reinforced my desire to expand my own horizons and never stop learning. I thank also my committee members, Peter Frederick and George Tanner, for supporting my efforts and freely offering di fferent points of view. I thank Peter particularly for introducing me to the world of scientific conferences and showing me the value of sharing and defending my ideas. I thank George for reminding me early on to keep it fun. Though I never did end up hanging a pole and line off the back of my boat, his advice reminded me to appreciate my surroundings even during the most trying of field days. The numerous IFAS statistical consultants who worked so valiantly to take my tangled findings and make them presentable also deserve my great thanks. Attempting to

PAGE 4

iv understand the designs and intentions of so many projects and figure out how best to convey the results is a daunting task. They handle it with admirable skill and patience. Finally, I wish to thank my parents for their quiet guidance, patience, and love. Their unconditional encouragement of my interests, no matter how far those interests have strayed, has afforded me a life of incredible discoveries and happiness.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES...........................................................................................................viii LIST OF FIGURES.............................................................................................................x ABSTRACT....................................................................................................................... xi CHAPTER 1. INTRODUCTION...........................................................................................................1 Background..................................................................................................................... 1 Urbanization and Avian Ecology.............................................................................1 Urban Lakes.............................................................................................................3 Study Area..................................................................................................................... .5 2. WATERBIRD DISTRIBUTION, SHORELINE DEVELOPMENT, AND HABITAT STRUCTURE..................................................................................................................9 Introduction................................................................................................................... ..9 Methods........................................................................................................................ .11 Bird Surveys...........................................................................................................11 Categorization of Shoreline Development.............................................................13 Categorization of Habitat Structure.......................................................................13 Littoral ha bitat..................................................................................................14 Onshore habitat................................................................................................14 Substrate...........................................................................................................15 Analyses....................................................................................................................... .15 Summer and Winter Comparisons.........................................................................15 Waterbird Habitat Use............................................................................................16 Developed versus undeveloped shoreline........................................................16 Littoral and onshore habitat associations.........................................................17 Determining overall significance.....................................................................17 Independence of significa nt habitat elements..................................................18 Results........................................................................................................................ ...19 Avian Community Composition............................................................................19 Shoreline Development and Habitat Coverage......................................................20 Developed Versus Undeveloped Shoreline Use....................................................21

PAGE 6

vi Substrate Use..........................................................................................................25 Littoral Zone Habitat Association..........................................................................25 Summer............................................................................................................26 Winter..............................................................................................................26 Onshore Habitat Association.................................................................................27 Summer............................................................................................................28 Winter..............................................................................................................29 Independence of Significan t Habitat Elements......................................................30 Summer............................................................................................................30 Winter..............................................................................................................31 Discussion..................................................................................................................... 32 Seasonal Species Composition...............................................................................32 Shoreline Development..........................................................................................33 Dominant Habitat Elements...................................................................................36 Littoral ha bitat..................................................................................................36 Onshore habitat................................................................................................37 Guild Responses to Other Habitat Elements..........................................................40 Marsh birds......................................................................................................40 Wading birds....................................................................................................41 Diving birds.....................................................................................................42 Ducks...............................................................................................................42 Management and Future Research................................................................................43 3. AVIAN BEHAVIORAL RESPONSES TO SHORELINE DEVELOPMENT............48 Introduction................................................................................................................... 48 Methods........................................................................................................................ .51 Analyses....................................................................................................................... .53 Shoreline Development..........................................................................................53 Disturbance Sensitivity..........................................................................................53 Determining Overall Significance..........................................................................54 Results........................................................................................................................ ...54 Seasonal Behavioral Observations.........................................................................54 Behavioral Associations with Shoreline Development..........................................55 Alert/flee behavior...........................................................................................55 Foraging behavior............................................................................................56 Resting behavior..............................................................................................56 Tending young behavior..................................................................................57 Active/swim behavior......................................................................................57 Human Activity......................................................................................................57 Disturbance Sensitivity..........................................................................................59 Seasonal alert/flee comparisons.......................................................................59 Inter-guild alert/flee comparisons....................................................................60 Migrant versus resident alert/flee comparisons...............................................60 Discussion..................................................................................................................... 61 Avian Responses to Shoreline Development.........................................................61 Alert/flee behavior...........................................................................................61

PAGE 7

vii Foraging and resting behavior.........................................................................62 Tending young behavior..................................................................................63 Active/swimming behavior..............................................................................65 Disturbance Sensitivity..........................................................................................66 Seasonal alert/flee response.............................................................................66 Inter-guild alert/flee comparisons....................................................................67 Migrant versus resident alert/flee response......................................................68 Management and Future Research................................................................................69 APPENDIX A. WINTER HAVEN WATER BIRD SURVEY DATA..................................................73 B. WINTER HAVEN WATERBIRD HABITAT DATA.................................................75 C. WINTER HAVEN WATERBI RD BEHAVIOR DATA..............................................84 REFERENCES..................................................................................................................89 BIOGRAPHICAL SKETCH.............................................................................................99

PAGE 8

viii LIST OF TABLES Table page 2-1. Total shoreline development and hab itat coverage of all habitat elements...............22 2-2. Overall waterbird abundance along developed and undeveloped shorelines on lakes Buckeye, Conine, Deer, and Jessie during summer 2001 and winter 2001/2002...................................................................................................23 2-3. Species abundance along developed and undeveloped shorelines during summer 2001 and winter 2001/2001......................................................................24 2-4. Waterbird guild associations with littoral habitat elements for summer 2001 (S), and winter 2001/2002 (W)..............................................................................28 2-5. Waterbird associations with onshore habitat elements by guild for summer 2001 (S), and winter 2001/2002 (W)..............................................................................30 3-1. Percent of waterbird guilds engaged in focal behaviors during summer (S) 2001 and winter (W) 2001/2002.....................................................................................55 3-2. Percent guild behavioral responses to developed (D) and undeveloped (U) shoreline for summer 2001....................................................................................58 3-3. Percent guild behavioral responses to developed (D) and undeveloped (U) shoreline for winter 2001/2002..............................................................................59 A-1. Aquatic bird species observed on lakes Buckeye (B), Conine (C), Deer (D), and Jessie (J) in Winter Haven, Flor ida, summer 2001 and winter 2001/2002............73 B-1. Wading bird habitat associations, summer 2001......................................................76 B-2. Wading bird habitat a ssociations, winter 2001/2002................................................77 B-3. Marsh birds habitat associations, summer 2001.......................................................78 B-4. Marsh bird habitat a ssociations, winter 2001-2002..................................................79 B-5. Diving bird habitat associations, summer 2001........................................................80 B-6. Diving bird habitat a ssociations, winter 2001/2002.................................................81

PAGE 9

ix B-7. Duck habitat associations, summer 2001..................................................................82 B-8. Duck habitat associations, winter 2001/2002. Expected values based on availability of habitat element on each lake...........................................................83 C-1. Data for summer guild beha vior, listed by guild and lake........................................85 C-2 Data for winter guild beha vior, listed by guild and lake............................................87

PAGE 10

x LIST OF FIGURES Figure page 1-1. Map of study lakes within Winter Haven urban area, Polk county, FL......................7 1-2. Aerial images of Winter Haven study lakes................................................................8

PAGE 11

xi Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Science URBAN LAKES AND WATERBIRDS: EFFECTS OF DEVELOPMENT ON DISTRIBUTION AND BEHAVIOR By Ashley H. Traut May 2003 Chair: Mark E. Hostetler Department: Wildlife Ecology and Conservation I studied waterbird distribution and be havior in the breeding and non-breeding seasons on four partially developed urban lakes in central Florida. I examined waterbird distributional and behavioral associations w ith developed and undeveloped shorelines, as well as distributional associations with specific elements of the littoral and onshore habitat in the urban environment. By understanding how waterbirds respond to development, and by identifying habitat elements that influence their movements and behaviors, we can provide more ecologically sound development and management practices for urban aquatic environments. A total of 38 waterbird species were observed on the four lakes over both seasons. Wading bird, marsh bird, and duck abundance was significantly greater along developed shoreline in both seasons on all lakes. Diving bird abundance was significantly greater along developed shoreline in the winter. Species richness was not associated with

PAGE 12

xii shoreline development. Species evenness was greater along undeveloped shoreline in the summer and developed shoreline in the winter. Tall emergent vegetation, open shore, lawn, and canopy appeared to be the primary habitat elements determining waterbird presence. All waterbirds were negatively associated with tall emergent vegetation on two or more lakes over both seasons, whereas wading birds, marsh birds, and ducks were positively associated with open shore in the summer, and wading birds, marsh birds, and diving birds were positively associated with lawn and canopy in the winter. Summer behavioral observations revealed that wading birds foraged significantly more along developed shoreline, and that ducks rested and tended young significantly more along developed shoreline. Winter observations revealed that marsh birds foraged significantly more along undeveloped shoreline and displayed active/swimming behavior significantly more along developed shoreline. Summer ducks and winter wading birds s howed significantly greater alert/flee behavior along undeveloped shoreline. Duck s showed significantly greater alert/flee behavior than other guilds. Overall, alert/flee behavior was seen 1.6 times more often in the winter. Winter migrants did not show great er alert/flee behavior than resident birds. Results show that a wide range of waterbirds can use urban lakes during the breeding and non-breeding seasons. Further, developed shoreline appears to be favored by many species for a variety of behaviors. However, dense stands of tall emergent vegetation along undeveloped shoreline may deter birds from using this shoreline. Greater alert/flee behavior along undeveloped shorelines may warrant the use of buffer zones to protect birds using these shorelines from undue human disturbance.

PAGE 13

1 CHAPTER 1 INTRODUCTION Background Urbanization and Avian Ecology Between 1900 and 1987 the proportion of the world’s human population living in cities rose from 14 to 50 percent (United Nations 1987). By 2050 the world’s urban population is predicted to equal 6.5 billion, equivalent to today’s entire global population (United Nations 1996). Though the majority of urbanization will occur in developing countries, seventy-eight percent of United St ates residents already reside in urban environments (Adams 1994). Strohm (1974) es timated that this process of urbanization meant the paving, building over, or drowning of over 400,000 hectares of natural habitat a year. Today, urbanization is the second most frequently cited cause of species endangerment in the United States (Czech & Krausman 1997). With this understanding comes the need for a comprehensive understanding of the ecology of urban systems and how best to manage them both for the needs of humans and for wildlife. To date, however, there is a relative scarcity of such knowledge, with many questions still unanswered about the effects of urbanization on everything from ecosystems to individual species (Cairns 1988, Niemela 1999). As a result, most urban planning and management remains focused on the impacts of urbanization on human society rather than on the issues of biodiversity (Marzluff et al. 2001). Urban ecosystems as a whole can be viewed as highly fragmented, heterogeneous landscapes dominated by buildings, roads, and pavement, and often lacking in substantial

PAGE 14

2 vegetation cover (Jokimaki 1999). The remaining vegetation composition is often greatly altered, consisting of a few favored natives and numerous exotic species (DeGraaf 1986). Further, the fragments of natural vegetation that are left relatively intact may be too small or isolated to support a healthy wildlife community (Savard et al. 2000). Urban habitats are also characterized by high levels of human -associated disturbance, such as traffic, construction, and recreation (Jokimaki 1999). These changes in structure and function can lead to a greatly modified wildlife assemblage consisting of habitat generalists, human-commensal species, and exotic species. Wildlife diversity in urban environments ultimately depends on how humans design and manage urban habitats. Many urban wildlife studies to date have focused on avian responses to development. Birds are often selected for study due to their diurnal activity patterns and relative ease of identification both by song and sight (DeGraaf & Wentworth 1981). They are also regarded as excellent indicators of stresses in an environment due to their sensitivity to change in habitat structure and composition (Savard & Falls 1982, Clergeau et al. 1998). The presence or absence of various avian species within urban areas is often associated with changes in habitat structure. Results from several studies have shown that with increased urbanization there is a shift in avian species composition, often acco mpanied with a decrease in bird species richness and diversity, and an increase in total bird density as a few human-commensal, often non-native species, such as the House Sparrow ( Passer domesticus ) and European Starling ( Sturnus vulgaris ) become very common (see Blair 1996 and Savard et al. 2000 for reviews). Decreased habitat availability, vegetative complexity, and food supply, and increased habitat fragmentation, competition, and human disturbance are examples of

PAGE 15

3 some of the mechanisms that have been identified as contributing to decreases in richness and evenness in urban bird communities (Marzluff 2001). Conversely, factors such as supplemental feeding, reduced predation, and reduced human persecution have benefited certain species in urban environments (Marzluff 2001). Species composition and richness have also varied in relation to the city and the locality within a city in which studies were conducted, with some areas showing considerable diversity, depending on local environmental conditions (Tilghman 1987, Blair 1996, Hostetler 1999, Hostetler & Holling 2000). The design and management of an area can have an appreciable effect on the distribution of birds across an urban environment. Urban Lakes The vast majority of urban bird studies have focused on passerine bird species in terrestrial habitats. Studies examining waterbirds have generally focused on marsh systems in non-urban environments, with relatively little attention being devoted to lacustrine habitats (but see Parris & Grau 1978, Whitfield & Cyrus 1978, Johnson & Montalbano 1984, Zaffke 1984, Pyrovetsi & Crivelli 1988, Edelson 1990, Hoyer & Canfield 1990, 1994). Virtually no studies have been conducted in the United States that have directly quantified waterbird responses to lakefront urbanization. Urbanized lakes often undergo similar patterns of habitat alteration as terrestrial habitats. Habitat complexity is often greatly reduced as both onshore and littoral vegetation are reduced, replaced, or removed. The clearing of these habitats by property owners is a common practice in order to improve lake views and recreational pursuits such as swimming, boating, and fishing (Guillory et al. 1979, Frayer & Hefner 1991, Bryan & Scarnecchia 1992). Such modifications are usually made without considering the potential effects on local wildlife.

PAGE 16

4 With the continued loss of wetland habitat throughout the United States (Mitsch & Gosselink 1993), the importance of lacustrine habitat to waterbirds may be increasing. Florida, home to some of the largest waterbird populations in the United States, has lost almost 50% of its wetlands over the past 200 years (Dahl 1990). Much of this loss can be attributed to urbanization, as seen in the fact that more than 84% of the state’s 16 million residents currently live in urban areas (Morris & Morris 1995). With an estimated 7,783 lakes providing potential waterbird habitat, and at the same time facing significant development pressures, Florida offers optimal conditions for studying the effects of urbanization on waterbirds and developing better management policies for urban lakes. In this project, I explored how waterbirds responded to shoreline habitat changes caused by human encroachment on urban lakes. In it, I examined both indirect and direct responses in waterbirds during the breeding and non-breeding seasons. In Chapter 2, I examine indirect responses measured by the presence or absence of birds along developed shoreline and in other specific habitat variables. It addresses the following research questions: 1. What is the composition of waterbirds found on Central Florida’s urban lakes during the breeding and non-breeding seasons? 2. Are waterbird abundance, community composition, and species richness significantly different between developed and undeveloped shorelines? 3. Which onshore and littoral habitat elements are most closely associated with the presence or absence of various waterbird guilds? In Chapter 3, I examine direct responses of waterbirds to development and human disturbance by measuring changes in waterbird behavior. Chapter 3 addresses the following research questions: 1. Are primary waterbird behaviors signifi cantly different between developed and undeveloped shoreline?

PAGE 17

5 2. Which waterbird guilds appear most sensitive to human disturbance? Study Area I conducted field research on four urban lakes within the Peace River drainage basin in and around the city of Winter Haven in Polk County, Florida (Figure 1-1). Polk County ranks fourth in lake abundance among all Florida counties, with 550 lakes within its borders (Edmiston & Myers 1983). Lakes Buckeye (28 2’ 24” N, 81 42’ 20” W), Conine (28 3’ 22” N, 81 43’ 29” W), Deer (28 1’ 32” N, 81 45’ 46” W), and Jessie (28 3’ 27” N, 81 45’ 48” W) are moderately developed lakes located along the edges of Winter Haven’s urban area. They were selected for study because each lake still had significant portions of undeveloped shoreline. Polk County Environmental Services classifies these lakes as mesotrophic or eutrophic, meaning they carry moderate to high nutrient loads based on chlorophyll, nitrogen, and phosphorus concentrations (Polk County 1999). Each lake meets Florida Class III criteria for surface water quality, designated for the propagation and maintenance of healthy, well-balanced fish and wildlife populations (Polk County 1999). Lake Buckeye was the smallest lake (29 ha) with the least developed shoreline (59%) (Figure 1-2). The majority of its undeveloped shoreline was located along the NE side of the lake and consisted of bottomland hardwoods (Florida DOT 1999). Lake Deer (47 ha) had the greatest amount of developed shoreline at 79%. Undeveloped shoreline, consisting of hardwood/conifer mix (Florida DOT 1999), was found in one contiguous stand along the SW corner of the lake. Lake Jessie (75 ha) was 69% developed with hardwood/conifer mix found along the northern shore and bottomland hardwoods dominating most of the eastern shore (Florida DOT 1999). Lake Conine was the largest lake at 96 ha and was developed along 61% of its shoreline. Three undeveloped areas,

PAGE 18

6 consisting of hardwood/conifer mix, wetland forest, and bottomland hardwood, were evenly spaced around the lake (Florida DOT 1999). Undeveloped habitats on each lake were mostly discrete forest fragments less than 20 ha in size, and surrounded by development on all sides. Urbanization consisted primarily of single-family homes, with several apartment complexes located on lakes Buckeye and Deer, and a mobile-home park and private airport located on the north side of Lake Jessie. Much of the east side of La ke Conine and the southeast corner of Lake Buckeye were within 30 m of moderately to heavily traveled roads. All lakes had shallow sloping littoral shelves (Florida LAKEWATCH 2000) with moderately diverse littoral plant communities. The vast majority of residential properties had a dock and boathouse on the water. All lakes had public access boat ramps. Lake littoral zones were markedly different along the developed and undeveloped shorelines. Undeveloped zones were often characterized by continuous dense stands of cattail ( Typha spp. ). Lake Jessie was the one exception to this, with much of its east side having sparse cattail mixed with a variety of lower emergents such as para grass ( Urochloa mutica ), torpedo grass ( Panicum repens ), and maidencane ( Panicum hemitomon ). Littoral zones along developed shorelines had a much patchier aquatic plant distribution with considerably more open water and far less cattail. Heavily developed shorelines were devoid of all aquatic vegetation. Less developed properties and areas between properties were characterized by a diverse community of low and tall emergent macrophyte species. Lake Deer was notable for its near continuous emergent zone around both the undeveloped and developed s hores, and its dense floating-leafed macrophyte coverage ( Nymphea spp. ) around more than 80% of the lake perimeter.

PAGE 19

7 Lake Jessie Lake Deer Lake Conine Lake Buckeye 3036Kilometers L a ke J es sie L a ke D ee r L a ke C on ine L a ke B uc ke ye N E W S Lake Jessie Lake Deer Lake Conine Lake Buckeye 3036Kilometers L a ke J es sie L a ke D ee r L a ke C on ine L a ke B uc ke ye N E W S Figure 1-1. Map of study lakes within Winter Haven urban area, Polk County, FL. Developed areas are in grey. Undeveloped areas are in green. Study lakes are in dark blue.

PAGE 20

8 Figure 1-2. Aerial images of Winter Haven study lakes. Clockwise from upper left are Lakes Buckeye, Deer, Jessie, and Conine. During this study lake levels were unusually low (.5 – 1 m below normal) due to a severe drought. Water levels in the summer of 2001 reached record lows on all of the lakes, exposing much of the emergent zone substrate, creating numerous open mudflats along unvegetated shoreline, and creating a 50 m peninsula along the southern shore of Lake Conine and a small mudflat island off the southern shore of Lake Deer.

PAGE 21

9 CHAPTER 2 WATERBIRD DISTRIBUTION, SHORELINE DEVELOPMENT, AND HABITAT STRUCTURE Introduction The importance of habitat structure in determining avian species composition in terrestrial systems has been well established (MacArthur & MacArthur 1961, MacArthur & Wilson 1967, Karr & Roth, 1971, Roth 1976). Though many factors have been proposed to explain avifaunal shifts in urban environments (Marzluff 2001), changes in habitat structure and composition are considered some of the primary mechanisms. Decreases in avian species richness and evenness in urban areas have been correlated with decreases in total woody vegetation volume, spatial heterogeneity, vertical structure, and plant species diversity (Lancaster & Rees 1979, DeGraaf 1986, Tilghman 1987, Marzluff & Sallabanks 1998), and increases in habitat fragmentation, habitat edge, and exotic vegetation (Soul et al. 1988, Marzluff 2001). Avian species diversity has been found to be negatively correlated with elements of the built up environment, such as housing density (Lancaster & Rees 1979, Blair 2001). The mechanisms that determine avian species distribution in aquatic systems have also received considerable attention. Broad-scale factors such as geography and climate have been shown to determine waterbird breeding and winter ranges (Weller 1999, Elphick et al. 2001). On a local scale, mechanisms such as species morphology, lake area, water depth, trophic status, and predator and prey densities have all been shown to influence the distribution of waterbird communities (Weller & Spatcher 1965, Jenni

PAGE 22

10 1969, Brown & Dinsmore 1986, Picman et al. 1993, Hoyer & Canfield 1994, Weller 1999). As in terrestrial systems, habitat structure may be one of the most important factors in determining avian community composition in aquatic systems, including physical components like water column depth, wetland substrate, shoreline, and vegetation strata (Weller 1999). Specific habitat associations are well documented for a variety of waterbirds. For example, grebes (Podicipedidae) prefer relatively shallow, wellvegetated wetlands (Weller 1999, Elphick et al. 2001). Other divers, such as the cormorants (Phalacrocoracidae) and anhinga s (Anhingidae), require shrubs, trees, or snags near the water on which to loaf and nest (Hatch & Weseloh 1999, Frederick & Siegel-Causey 2000). Many large and mid-sized wading birds are known to prefer shallow, relatively open areas for foraging (Hancock & Kushlan 1984). Marsh birds such as rails and gallinules (Rallidae) require dense emergent or floating vegetation for nesting and foraging, while coots commonly use more open water habitat (Melvin & Gibbs 1996, Weller 1999, West & Hess 2002). In urban areas, development and the alteration of both aquatic and terrestrial habitat structure may be the most important factors in determining the composition and distribution of waterbird communities. Howeve r, few studies have looked at the effects of lake-habitat structure on urban bird populations. In general, the majority of research on urban aquatic environments in the United States has focused on the effects of recreation on avian abundance, distributi on, and breeding success (Hockin et al. 1992, Knight & Gutzwiller 1995). In general, these studies have found lower abundance, reduced use of sites, and lower breeding success in areas with significant recreation (see

PAGE 23

11 Hockin et al. 1992 for review). Studies in Europe and Canada have examined the effects of shoreline cottages on diving birds (Lehtonen 1970, Bundy 1979, Andersson et al. 1980, Heimberger et al. 1983), finding reduced reproductive success and lake utilization in areas with cottages. These studies focused on a single species and did not attempt to quantify the role of habitat structure. The aim of this study was to directly examine the effects of development in urban lake environments by answering the following questions: 1. What is the composition of waterbirds found on Central Florida’s urban lakes during the breeding and non-breeding seasons? 2. Are waterbird abundance, community composition, and species richness different between developed and undeveloped shoreline? 3. Which onshore and littoral habitat elements are most closely associated with the presence or absence of various waterbird guilds? Question 1 addresses the general lack of knowledge about avian species composition on urban lakes by looking for species or groups of birds that are noticeably absent, exploring seasonal fluctuations, and comparing community composition with previous lake studies conducted in the Central Florida region (Hoyer & Canfield 1990, 1994, Roth in press). Questions 2 and 3 examine the impact of developed shoreline on the composition and distribution of waterbirds and explore the role of specific habitat elements in determining distribution patterns. Methods Bird Surveys Waterbird surveys were conducted from June 7 – August 1 of 2001 and December 8 – February 6 of 2001/2002. For this study the term ‘waterbird’ referred to species in the orders Gaviiformes, Podicipediformes, Pelecaniformes, Ciconiiformes, Anseriformes, Falconiformes, Gruiformes, Charadriiformes, or Coraciiformes observed on or feeding

PAGE 24

12 from lacustrine habitats. Each of the four lakes were surveyed a total of eight times each season by driving at minimum-wake speed around the lakes (20 – 30 m from shore) in a small motorized canoe. Both the order of lakes surveyed and the direction of travel were alternated for each day of surveying in order to account for any time-related bird movements around the lakes. Surveys were conducted within the first five hours after sunrise on mornings with little to no rain and winds less than 24 km/hr. Birds were identified by sight using 8x42 binoculars. Shoreline development and associated vegetation structure were recorded for each bird once it was identified (see description below). The location of each bird was also recorded on a 1999 aerial photograph of each lake obtained from the Florida Department of Environmental Protection (DEP) at their Land Boundary Information Systems (LABINS) website ( www.labins.org ). In order to minimize count error along undeveloped shoreline, where dense tallemergent vegetation was often encountered in the littoral zone, I drove the survey boat directly into the vegetation to flush hidden birds. This procedure was repeated every 30 – 50 m in areas of continuous vegetation. On alternate insertions, I shut off the motor and listened for calling birds for approximately two minutes. Birds that flushed from any location and landed ahead of the boat were recorded only in their original location. Birds whose origin and destination were not observed or which were observed greater than 30 m offshore or 10 m onshore were recorded for analysis of overall bird composition on a given lake, but were not included in either shoreline development or habitat analyses.

PAGE 25

13 Categorization of Shoreline Development Each lake in the study had a mixture of developed and undeveloped shoreline. Lake shoreline was categorized as either developed or undeveloped based on DEP land cover classifications for each lake. Classi fications were modified during preliminary surveys by basing categorizations only on the land cover within the first 20 m of shoreline extending away from the water on each lake. For the purposes of this study, developed shoreline referred to any continuous stretch of shoreline greater than 100 m, parallel to the edge of the lake, that had a minimum of 50% long-term habitat alteration, defined as cleared land, lawns, landscaping, buildings, and roads. Undeveloped shoreline was defined as any continuous stretch of shoreline greater than 100 m, parallel to the edge of the lake, with greater than 50% intact natural habitat, and little to no sign of regular human use. Developed and undeveloped areas were separated by a buffer of 40 m to eliminate the potential effects of convergi ng habitats. Birds recorded in these border areas were not included in either shoreline development or habitat analyses. Categorization of Habitat Structure Upon sighting each bird I recorded several different habitat elements within the littoral zone and immediate onshore zone, as well as the type of substrate that each bird was sighted on. Visual estimates were made for littoral and onshore habitat elements found in 5x20 m rectangular bands around and adjacent to each bird. The 20-meter side of the band was parallel to the shoreline. Coverage of each of eight habitat elements was classified as one of five densities: 0-5% 6-25%, 26-50%, 51-75%, or 76-100%. This was done in order to see if minimum or maximum thresholds existed at which birds selected or avoided each habitat element.

PAGE 26

14 Littoral habitat Three categories were created for littoral zone habitat: low emergent vegetation (< 1 m tall), tall emergent vegetation (> 1 m tall), and floating-leafed vegetation. For birds sighted onshore (within 10 m of the water) or less than 5 m offshore, I recorded the first 5 m of aquatic vegetation extending out from the shoreline and within 10 m of either side of the bird. For birds located greater than 5 m from shore, the 5x20 m band was centered around each bird. Since the drought caused extremely low water stages during this study, I determined the upland edge of the littoral zone by the edge of herbaceous emergent vegetation, rather than the presence of standing water. In several undeveloped areas on lakes Buckeye and Conine, backmarshes of fallen cattail and vines had developed behind dense stands of cattail, completely covering the shallows. In such cases, where the water was completely covered by vegetation and was too shallow for any diving birds to use, I considered the area onshore habitat. Onshore habitat Five categories were created for shoreline habitat: open shore (moist soil or sand), lawn (any maintained grassy area), understory (< .5 m tall), shrub (.5-3 m tall), and canopy (> 3 m tall). Immediate onshore habitat was recorded for any bird in the water within 20 m of shore and any bird onshore within 10 m of the water. For birds located in the water, I made a trajectory (perpendicular to the shoreline) from where the bird was sighted to a point on shore. For these birds and birds located onshore less than 5 m from the water, I recorded onshore habitat within a 5x20 m band placed 10 m on either side of the trajectory line or bird and extending 5 m inland from the water’s edge. For birds located greater than 5 m from the water, the onshore band was centered on the bird.

PAGE 27

15 Substrate Substrate was defined as the human structure (e.g. pier, boat, etc.) or habitat element (e.g. cattail, shallows, lawn, etc.) on or nearest to which each bird was found. For example, if a bird was observed foraging in a pocket of unvegetated shallow water in a 5x20 m area that was dominated by tall emergent vegetation, the bird’s substrate was recorded as shallow water. Analyses Waterbirds were grouped into guilds for certain analyses. Guilds were based on foraging behavior and habitat use, and included wading birds (Ardeidae, Threskiornithidae, Ciconiidae, Recurvirostr idae), marsh birds (Rallidae), surface and aerial diving birds (Podicipedidae, Phalacrocoracidae, Anhingidae, Accipitridae, Laridae), and ducks (Anatidae). Analyses were only conducted for a given guild or species where 10 or more birds (n 10) were observed within that guild or species over a season. Summer and Winter Comparisons Species counts from all lakes were summed for each season to examine overall waterbird community composition. Species richness and guild abundance on each of the lakes were calculated for each survey day and then averaged across all days and lakes each season (8 surveys per season) to determine the mean number of species and individuals observed. The percent similarity measure (Brower et al. 1990) was used to compare avian community composition between seasons in order to determine whether differences in seasonal avian distribution patterns could be explained by differences in community composition.

PAGE 28

16 The Jacaard index of community similarity (Brower et al. 1990) was used to compare community composition with that of two previous urban lake studies in Central Florida. Hoyer & Canfield (1994) examined 33 lakes, both developed and undeveloped in the north central and central regions of Fl orida. Roth (in press) examined six urban lakes in the Winter Haven area in 1991. Waterbird Habitat Use I tested for waterbird associations with developed and undeveloped shoreline and with littoral and onshore habitat elements using Chi-square goodness-of-fit tests ( = 0.05, df = 1). Developed versus undeveloped shoreline Overall abundance, guild abundance, species abundance, species richness, and species evenness were compared between areas of developed and undeveloped shoreline. Expected values were based on the proportions of developed and undeveloped shoreline. For example, if 100 wading birds were observed on a lake and 70% of the shoreline was developed, then, assuming a random distribution of birds, expected values would be 70 birds along developed shoreline and 30 along undeveloped shoreline. Overall abundance, species richness, and species evenness were compared between developed and undeveloped shorelines on a lake-by-lake basis. Each lake was analyzed separately to allow for the possibility of intra-lake variability. Guild abundance and species abundance were calculated with all lakes combined based on the results of overall abundance. Evenness was calculated using Simpson’s measure of evenness (Krebs 1998).

PAGE 29

17 Littoral and onshore habitat associations I tested for waterbird associations with littoral and onshore habitat elements by comparing guild abundance in the presence and absence of each of the habitat elements on each lake. For each sighting, a habitat element was considered absent if it was recorded as occurring in 0-5% of the 5x20 m band around a bird. A habitat element was considered present if it was recorded as occurring in greater than five percent of the 5x20 m band around a bird. Expected values were based on the percent lake coverage of each habitat element around the perimeter each lake. Habitat measures were taken during the winter 2001/2002 field season. Habitat coverage did not noticeably change between seasons except for open shoreline, which dramatically decreased from summer to winter due to rising water with the onset of fall rains. The amount of open shoreline was therefore calculated separately for each season. Habitat elements were assessed for presence/absence along the perimeter of each lake. Each littoral and shoreline habitat element was recorded as present if there was greater than 5% coverage within the first 10 m offshore or onshore, respectively. Measurements were then entered into ArcView 3.2 (ESRI 1999) to determine the exact percentage of the lakes that was covered by each habitat type. Any habitat type on a lake that occurred in less than 5% of the total littoral or shoreline zone was considered functionally absent and excluded from analysis. Determining overall significance Where analyses were conducted on a lake-by-lake basis, results were considered to have overall significance where significant same-direction associations (p 0.05) were observed on at least three lakes without a si gnificant contradictory finding on the fourth lake. In cases where habitat elements were only present on two lakes, results were

PAGE 30

18 considered to have overall significance where significant same direction associations (p 0.05) were observed on both lakes. In cases where results were not significant but suggestive of an overall pattern on three or more lakes (p 0.2 for each lake), without a contradictory finding on the fourth lake, probabilities (for lakes indicating a pattern) were combined for meta-analysis (Fisher 1958) to test for overall significance. Habitat elements that were found to be significant were examined across the five different habitat coverage densities to see if minimum or maximum thresholds existed at which birds selected or avoided each habitat type. Independence of significant habitat elements My analyses for habitat association were univariate in nature, and therefore it was difficult to detect whether birds were responding to individual habitat elements, or to combinations of habitat elements. I looked for evidence of this preference for cooccurring habitat elements by comparing the proportion of birds that were significantly associated with two or more elements, with an estimate of the actual proportion of cooccurrence of those habitats on the lake. I fi rst determined how often birds within a guild were sighted in significant positive association with two or more elements. I then used the aerial photographs and habitat estimates taken during the winter to estimate the degree of overlap among elements of interest (to nearest 1%). Since exact estimates could not be determined due to the resolution of the photographs, statistical analyses were not employed. I concluded that a guild may have been responding to a combination of habitat elements if the difference between percent actual overlap of habitats and percent of birds sighted where habitats overlapped was 50% or more. For example, if lawn and open shore habitats spatially overlapped 10% of the time, but 80% of wading birds observed were in areas where both elements were present, then I concluded that the guild

PAGE 31

19 was responding to the combination of these elements. Cross tabulations could not be used for negatively significant habitat elements since, by definition, birds were found in areas where these elements were absent. Independence of negative habitat elements was therefore estimated by looking only at the degree of actual spatial overlap around the lake. Results Avian Community Composition A total of 38 waterbird species were observed over the course of this study, including nine species listed as endangered, threatened, or of special concern in the state of Florida (Appendix A). Thirty-five species were observed on more than 10% of the surveys. When overall community composition on these lakes was compared with the two previous studies of Central Florida lakes, species similarity (Jacaard index) was 0.76 and 0.7. A similarity index of 1.0 in either case would indicate that all species were found in both studies. A total of 33 waterbird species were observed during the summer 2001 season. An average of 13.6 species and 104.1 birds were observed per lake each day. Standard deviations between lakes varied widely over both seasons. Wading birds were the most common guild in terms of species richness, making up 45% of all species observed. Ducks showed the greatest abundance, making up 39% of all birds observed. Wood Ducks ( Aix sponsa ) accounted for 96% of all ducks. Thirty-two species were observed during the winter 2001/2002 season. An average of 14.8 species and 114.4 birds were observed per lake each day. Wading birds continued to show the greatest species richness, making up 41% of all species. Several

PAGE 32

20 large flocks of migrant Double-crested Cormorants ( Phalacrocorax auritus ) accounted for diving birds showing the greatest winter abundance, making up 34% of all birds observed. Species composition between seasons was moderately similar (0.54, percent similarity measure), with the variation explained by the arrival of 10 winter migrant species (i.e., birds observed in significantly greater numbers in winter than in summer in central Florida) (Appendix A). Six of these species were diving birds. Shoreline Development and Habitat Coverage Table 2.1 gives a breakdown of the percent coverage of developed and undeveloped shorelines and habitat elements on each lake. Due to the overlap between most habitat elements, total percentages on any lake were greater than 100%. Understory, shrub, and canopy in particular had near 100% overlap along undeveloped shoreline. Littoral zones along developed shorelines were characterized by a very patchy habitat structure, with distinct clumps of low and tall emergent vegetation and large areas devoid of aquatic vegetation. Onshore habitat in developed areas was characterized by significant lawn coverage with very sparse, intermittent understory and shrub layers. Across all lakes, 90% of open shore habitat was found along developed shoreline. This habitat was typically found in conjunction with lawn habitat where onshore and littoral vegetation had been cleared. Undeveloped shorelines, on the other hand, were much more homogeneous. Tall emergent vegetation dominated much of the littoral zones. Cattail ( Typha sp .) was the dominant emergent vegetation and was presen t along 89% of undeveloped shoreline, with almost 70% of it found in dense, continuous stands. Open shore, or exposed shoreline in

PAGE 33

21 general was therefore rarely available. Onshore habitat consisted of relatively continuous low to moderate understory, and moderate to dense shrub and canopy. Tall emergent vegetation was the dominant littoral zone habitat element on three of the four lakes, being present in 47-76% of total littoral zone area. Lake Deer showed considerable tall emergent coverage, at 68%, but was also dominated by floating-leafed vegetation, which was found in 83% of the total littoral zone area. Floating-leafed vegetation was considered functionally absent on Lakes Conine and Jessie since each lake had less than 5% coverage. None of the shoreline habitat elements clearly dominated the overall shoreline habitat structure. Open shore significantly declined between seasons, dropping from 29% to 7% of available onshore area from summer to winter. On Lake Buckeye and Deer, open shore coverage dropped below 5% in the winter and therefore was considered functionally absent on those lakes. Developed Versus Undeveloped Shoreline Use During both seasons a strong association was observed between shoreline development and bird abundance, with more birds found along developed shore on all four lakes (Table 2-2). Wading birds, marsh birds, and ducks showed this positive association over both seasons (all tests: 2 39.09, p 0.0001). Diving birds showed this positive association only in the winter (all tests: 2 3.87, p 0.05). Only one lake, Lake Deer in the summer, showed an association between species richness and shoreline development. In this case, significantly more species than expected were found along undeveloped shoreline ( 2 = 7.01, p = 0.008). Since no other lakes in either season showed this pattern, an overall association between species richness and shoreline development was not established. Temporal and spatial patterns of evenness varied

PAGE 34

22 considerably among lakes. Undeveloped shoreline had greater species evenness than developed shoreline on all four lakes in the summer. Evenness indices for developed shoreline in the summer ranged from 0.15 – 0.38, and for undeveloped shoreline ranged from 0.27 – 0.68. The opposite pattern was observed in the winter, with three of the lakes having greater species evenness along developed shoreline. Evenness for developed shoreline ranged from 0.62 – 0.95, and for undeveloped shoreline from 0.51 – 0.73. In all cases, an evenness index of 1.0 would mean that all species were comprised of an equal number of individuals. Table 2-1. Total shoreline development and habitat coverage of all habitat elements. Shoreline development measures were based on first 20 m of shoreline surrounding each lake. Habitat coverage was based on 5 m deep perimeter bands for littoral and shoreline zone habitats. Buckeye Conine Deer Jessie Area (ha)%Area (ha)%Area (ha) % Area (ha)% Development Developed 2.62594.32614.1479 4.5169 Undeveloped 1.83412.74391.0921 2.0931 Onshore Habitat Canopy 0.59480.1790.5138 0.5431 Shrub 0.76621.22650.5541 0.8147 Understory 0.75611.28680.6246 0.8247 Lawn 0.52420.29150.8059 0.8247 Open Shore summer 0.41340.69370.1814 0.5833 Open Shore winter 0.032*0.31160.011* 0.169 Littoral Habitat Floating Leaf 0.44370.011*1.0683 0.053* Tall Emergent 0.69591.38760.8768 0.7847 Low Emergent 0.0650.28160.5543 0.7746 Below 5% considered absent In terms of individual species abundance patterns, results for the summer season revealed that eight of 28 species (29%) observed using shoreline habitat showed a significant positive association with developed shore (all tests: 2 5, p 0.03). In the winter, 16 of 27 species (59%) showed a significant positive association with developed

PAGE 35

23 shore (all tests: 2 5.48, p 0.02); a significant increase over the summer season ( 2 5.26, p 0.02). Winter migrants accounted for half of the species showing a positive association for developed shoreline in the winter. Table 2-2. Overall waterbird abundance along developed and undeveloped shorelines on lakes Buckeye, Conine, Deer, and Jessie during summer 2001 and winter 2001/2002. All numbers represent the sum total from eight surveys conducted each season. Lake Relative Abundance Summer Winter Developed Undeveloped Developed Undeveloped Buckeye 214* 66 195* 73 Conine 336* 133 522* 49 Deer 1209* 103 1254* 47 Jessie 693* 142 745* 103 *From Chi-square goodness-of-fit tests, significantly more birds than expected along indicated shoreline (p < 0.0001). Year-round residents that showed a significant association with developed shoreline during both the summer and winter seasons included the Snowy Egret ( Egretta thula ), Tricolored Heron ( Egretta tricolor ), White Ibis ( Eudocimus albus ), Wood Duck, Common Moorhen ( Gallinula chloropus ), Purple Gallinule ( Porphyrula martinica ), and Killdeer ( Charadrius vociferous ) (Table 2-3). Winter migrants that showed a significant association with developed shoreline in the winter included the Double-crested Cormorant ( Phalacrocorax aulitus ), Ring-necked Duck ( Aythya collaris ), American Coot ( Fulica Americana ), Ring-billed Gull ( Larus delawarensis ), Belted Kingfisher ( Ceryle alcyon ), and Fish Crow ( Corvus ossifragus ).

PAGE 36

24 Table 2-3. Species abundance along devel oped and undeveloped shorelines during summer 2001 and winter 2001/2002. Data from all lakes and survey dates are summed for each season. Species Summer Abundance Winter Abundance Developed UndevelopedDeveloped Undeveloped Diving Birds Pied-billed Grebe 1 1 20 3 D-C Cormorant 34 23 97* 9 Anhinga 145 76 97 82* Osprey 11 9 9 9 Ring-billed Gull n/a n/a 18* 0 Belted Kingfisher 2 0 24* 1 Wading Birds Least Bittern 8 8 0 3 Great Blue Heron 106 50 61 32 Great Egret 43 20 41* 5 Snowy EgretS 45* 6 17* 0 Little Blue HeronS 11 3 18* 0 Tricolored HeronS 51* 6 44* 2 Cattle Egret 3 0 66* 0 Green Heron 61 32 11 6 B-C Night Heron 1 1 0 6 White IbisS 160* 9 223* 0 Glossy Ibis 12 8 n/a n/a Wood StorkE 6 1 2 0 LimpkinS 8 1 2 0 Sandhill CraneT 8 0 2 0 Black-Necked Stilt 2 0 n/a n/a Ducks (wild) Wood Duck 1022* 89 384* 40 Mallard 56* 0 n/a n/a Blue-winged Teal n/a n/a 10 0 Ring-necked Duck n/a n/a 353* 0 Marsh Birds Rail 0 1 n/a n/a Sora 0 1 n/a n/a Purple Gallinule 127* 14 97* 5 Common Moorhen 464* 72 360* 74 American Coot 1 0 217* 0 Other Killdeer 13* 0 14* 0 S State listed as Species of Special Concern, T state listed as Threatened, E state and federally listed as Endangered. *From Chi-square goodness-of-fit tests, significantly more birds than expected along indicated shoreline (p < 0.05).

PAGE 37

25 Four species, the Rail ( Rallus sp .), Sora ( Porzina carolina ), Least Bittern ( Ixobrychus exilis ), and Black-crowned Night Heron ( Nycticorax nycticorax ), were found exclusively along undeveloped shoreline during one or both seasons, but their numbers were too small to be analyzed. Only the Anhinga ( Anhinga anhinga ) showed a significant association with undeveloped shor eline, and only during the winter season ( 2 = 12.71, p < 0.001). Substrate Use Substrate analyses offered a fine-scale measure of habitat use by depicting the immediate habitat element or structure that each bird was using. The most commonly used substrates for each guild are reported here. Forty-nine percent of all marsh birds were found in either low-emergent or floati ng-leafed vegetation over both seasons. Fifty percent of wading birds were found either in the shallows or along open shore in the summer, whereas only 11% were found in these substrates in the winter, 35% were found on piers (vs. 9% in the summer), and 19% were found in low-emergent vegetation (vs. 3% in the summer). Forty-four percent of diving birds were found on piers and pylons over both seasons, while 34% were found in trees or on logs. This division was strongly related to shoreline development. Along the developed shoreline, 63% of diving birds were actually found on piers and pylons, whereas 83% of this guild was found in trees or on logs along undeveloped shoreline. Forty percent of all ducks were observed in floating-leafed vegetation over both seasons. Littoral Zone Habitat Association Table 2-4 summarizes the significant associations found between waterbird guilds and littoral zone habitat elements for summer and winter surveys.

PAGE 38

26 Summer Tall emergent Summer analyses revealed that tall emergent vegetation was negatively associated with bird presence for marsh birds, wading birds, and ducks (all tests: 2 31.39, p 0.0001). Though not meeting the requirements for overall significance, diving birds may have also been negatively associated with this habitat element showing a significant negative association on two lakes (both tests: 2 11.31, p 0.001). Species analyses revealed that the Least Bittern was the only species that was positively associated with tall emergent vegetation ( 2 = 4.12, p = 0.04). No other habitat element in either the littoral or shoreline zones showed such a broadly consistent pattern of association. Examination of overall bird presence across the five habitat densities showed that 81% of all birds using areas with tall emergent vegetation were found in areas with less than 50% coverage of this habitat element. Floating leaf Lakes Buckeye and Deer were the only lakes with enough floatingleaf coverage (>5%) to conduct habitat analyses for this element. A strong positive association was observed between ducks and floating-leafed vegetation on these lakes (both tests: 2 9.73, p 0.002). Ducks used this habitat at all coverage densities, but 40% were found in areas with greater than 75% floating-leaf coverage. Marsh birds, wading birds, and diving birds did not app ear to respond to floating-leaf vegetation. Low emergent Results for low emergent vegetation were inconclusive for all waterbird guilds. Though individual lakes and guilds showed significant results, consistent patterns were never observed on more than two lakes. Winter Tall emergent Winter analyses revealed a negative association between tall emergent vegetation and wading bird presence on all lakes (all tests: 2 8.61, p

PAGE 39

27 0.003). Again, 81% of wading birds associating with tall emergent vegetation were found in areas with less than 50% coverage of this habitat element. Though not meeting the requirements for overall significance, marsh birds and diving birds may have also been negatively associated with this habitat, showing significant negative associations on two lakes each (all tests: 2 9.79, p 0.002). Separate species analysis showed that the Green Heron showed a positive association with tall emergent vegetation ( 2 = 7.06, p = 0.008). Floating leaf Ducks continued to show a positive association with floating-leafed vegetation in the winter on the two lakes with this habitat element (both tests: 2 14.08, p 0.001). Eighty percent of the ducks that a ssociated with floating-leafed vegetation were found in areas of greater than 50% c overage. Both wading birds and diving birds were negatively associated with floating-leafed vegetation on the two lakes (all tests: 2 11.85, p 0.001). The avoidance threshold for these birds in this habitat appeared to be 50% coverage, with only 11% of the birds found in the greater coverage densities. Marsh birds were not associated with floating-leafed vegetation. Low emergent Wading bird presence was positively associated with the presence of low emergent vegetation on three of the four lakes (all tests: 2 15.91, p 0.0001). Of the waders associating with this habitat element, 40% were found in areas with less than 25% low emergent coverage. Marsh birds, diving birds, and ducks showed no consistent patterns of association with this habitat element. Onshore Habitat Association Table 2-5 summarizes the significant and marginally significant associations found between waterbirds guilds and onshore habitat elements.

PAGE 40

28 Table 2-4. Waterbird guild associations with littoral habitat elements for summer 2001 (S), and winter 2001/2002 (W). Guild Tall Emergents Low Emergents Floating-Leafed S W S W S W Marsh -Waders --++ -Divers -Ducks -++ ++ -significant negative association on at least three lakes (p < 0.05) ++ significant positive association on at least three lakes (p < 0.05) Summer Open shore Open shore showed significant positive association on three of the four lakes with both ducks and marsh birds (all tests: 2 17.71, p 0.0001). Metaanalysis also showed a significant positive association between open shore and wading birds ( 2 = 40.3, p < 0.0001). Diving birds showed no consistent pattern of association with this element. As defined (moist soil or sand), open shore was rarely found extending beyond the first meter of shoreline. Coverage densities therefore rarely exceeded 0 – 25%, which precluded interpretation of threshold tolerances across the habitat density gradient. However, birds were found with this habitat at each of the coverage densities where they were available. Lawn Summer trends for lawn were inconsistent for all guilds but diving birds, which showed an overall significant pattern of negative association on three lakes (metaanalysis: 2 = 37.4, p < 0.0001). Diving birds were found at equally low numbers across all coverage densities for this habitat element. Understory, shrub, and canopy Neither understory nor shrub habitat showed consistent patterns of association with any of the bird guilds during the summer season. Canopy habitat showed an overall significant pattern of positive association with diving

PAGE 41

29 birds on all lakes (meta-analysis: 2 = 44.5, p < 0.0001). This association was relatively even across all canopy densities. Winter Open shore Due to higher water levels in the winter, open shoreline was greatly reduced on all lakes compared to summer availability. Open shoreline became so limited on Lakes Buckeye and Deer (<5%) that these lakes were removed from analyses. Wading birds were positively associated with this habitat on the two remaining lakes (both tests: 2 4.21, p 0.04). Open shore never occurred on more than 25% of total onshore habitat in the winter, which prevented interpretation of threshold tolerances across the habitat density gradient. None of the other guilds showed a consistent preference for or against this habitat. Lawn Lawn showed a strong overall pattern of positive association with marsh birds, wading birds, and diving birds in the winter. Both marsh birds and wading birds showed significant positive associations (all tests: 2 18.4, p 0.0001) with this habitat. Diving birds showed a significantly positive association on two lakes and a positive trend on a third lake (meta-analysis: 2 = 35.51, p < 0.0001). Fifty-eight percent of the birds associating with lawn habitat were found in areas of greater than 75% lawn coverage. Understory Understory showed a significantly positive association with diving birds (all tests: 2 4.69, p 0.03), and an overall pattern of significant positive association with marsh birds (meta-analysis: 2 = 17.98, p < 0.01). No minimum or maximum habitat density thresholds were apparent. Shrub Shrub was negatively associated with wading birds, which significantly avoided this kind of onshore habitat on three lakes (all tests: 2 6.28, p 0.012). An

PAGE 42

30 avoidance threshold was not apparent, with birds found at equally low numbers across all levels of shrub coverage. Canopy Marsh birds and diving birds showed a significant positive association with canopy coverage (all tests: 2 7.77, p 0.005). Wading birds showed a significant positive association on two lakes and a positive trend on a third lake (meta-analysis: 2 42.05, p 0.0001). No minimum or maximum habitat density thresholds were apparent, with all guilds found in all levels of canopy coverage. Table 2-5. Waterbird associations with onshore habitat elements by guild for summer 2001 (S), and winter 2001/2002 (W). Guild Open Shore Lawn Understory Shrub Canopy S W S W S W S W S W Marsh ++ ++ + ++ Waders + ++ ++ -+ Divers + ++ + ++ Ducks ++ ++/-indicates significant positive/negative associations on at least three lakes (p 0.05). +/– indicates significant overall positive/negative association (p 0.05) based on metaanalysis of trends (p 0.2) on at least three lakes. Independence of Significant Habitat Elements Summer Ducks showed a significant positive association with both floating-leafed vegetation and open shore in the summer. Looking only at the portions of the lakes covered with one or both of these habitat elements, floating-leafed vegetation overlapped with open shore 12% of the time. Twenty-five percent of all ducks sighted in these areas were found where these habitats overlapped. Thus, it did not appear that ducks were strongly responding to the combination of these elements. No other guilds showed multiple significant habitat associations in the summer.

PAGE 43

31 Winter Marsh birds Marsh birds showed a significant positive association with lawn, understory, and canopy in the winter. The most substantial differences between habitat availability and marsh bird distribution occurred in lawn/understory habitat and lawn/understory/canopy habitat. Lawn/understory habitat (without canopy) occurred on only 1% of the shoreline, while 12% of marsh birds were sighted in areas with both elements present. Similarly, lawn/understory/canopy habitat occurred on only 2% of significant habitat shoreline, while 25% of marsh birds were sighted with all three habitat elements present. Additionally, canopy habitat occurred in the presence of either lawn or understory along 98% of the shorelines having one or more of these elements. Thus, although there may be some selection of marsh birds for lawn/understory/canopy combinations, it could not be determined whether these combinations played a role in attracting marsh birds. Wading birds Wading birds showed a significant positive association with low emergent vegetation, open shore, and lawn in the winter. Overall, no combination of these habitat elements clearly attracted wading birds. The most substantial difference between overlapped habitat availability and wading bird distribution occurred in low emergent/lawn habitat, where 40% of wading birds were sighted in areas where both low emergents and lawn were present. Actual overlap of these two habitat elements was only 29%. Wading birds showed a negative association with tall emergent vegetation, floating-leaf vegetation, and shrubs in the wi nter. Wading birds appeared to show the strongest negative association with tall emergent vegetation. Looking only at spatial distribution, tall emergent vegetation overlapped with at least one other element 83% of

PAGE 44

32 the time. Therefore it could not be determined whether wading birds were avoiding this habitat element alone or a combination of these elements. Diving birds Diving birds showed a significant positive association with understory and canopy in the winter. Fifty-seve n percent of diving birds were sighted in areas where both habitat elements were present. Understory and canopy habitat overlapped along 26% of the shoreline. Thus, it could not be concluded that diving birds were selecting for the combination of these habitat elements. Discussion Seasonal Species Composition Thirty-five species were observed using these lakes on a regular basis over one or both seasons, including nine state or federally listed species. In the summer, marsh birds, wading birds, and ducks were all observed breeding on these lakes, with numerous fledgling marsh birds and ducks observed (Chapt 3). In the winter, almost one third of the species that used these lakes were winter migrants. Several of these species, like the Double-crested Cormorant, Ring-billed Gull, and Ring-necked Duck were observed in large foraging flocks using these lakes for brief periods. Other species, such as the Piedbilled Grebe, American Coot, and Belted Kingfisher established themselves more permanently on these lakes for the winter season. These findings confirm that urban lakes can sustain diverse waterbird communities during both the breeding and winter seasons, and apparently provide functional habitat for a variety of seasonal needs. The similarity in community composition with Hoyer & Canfield’s study (1994) and Roth’s study (1991) indicates that urban lakes may not just have relatively stable avian communities from one season to the next, but also from year to year. This longterm stability may be the result of the dam-controlled water levels providing more stable

PAGE 45

33 environmental conditions than are found in many natural wetlands. If so, some waterbirds may learn to rely on urban lakes in times requiring stable water levels, such as the nesting season. Shoreline Development Wading bird, marsh bird, and duck abundance were significantly greater than expected along developed shoreline on all lakes during both the summer and winter seasons. While some birds may have been undercounted along undeveloped shoreline in areas of dense cattail, the degree of difference in the number of birds found along developed and undeveloped shorelines cannot be explained by this alone. In my repeated entries into cattail stands, I saw and heard almost no waterbirds using this habitat. Roth (in press) had similar results in a study of neighboring lakes. Further, many of the more conspicuous waterbirds, such as the large and mid-sized long-legged waders, are known to avoid areas of dense tall-emergent vegetation (Smith et al. 1995, Surdick 1998, Roth in press). That almost all of these species were found in significantly greater abundance along developed shoreline supports these results, and suggests that the prevalence of cattail along undeveloped shoreline may have been a primary factor in many birds selecting developed shoreline. Knight and Cole (1995) stated that the four primary ways in which human activities can impact animals are through exploitation, pollution, disturbance, and habitat modification. Hunting is not allowed on these lakes, and, given the small size of the lakes’, pollution effects are most likely evenly dispersed between developed and undeveloped shores. This leaves disturbance and habitat modification as the two primary activities affecting waterbird patterns. Human disturbance, in general, has been widely shown to be detrimental to birds and other wildlife (Hockin et al. 1992, Carney &

PAGE 46

34 Sydeman 1999, but see Nisbet 2000). If human disturbance was the primary activity influencing the distribution of waterbirds ar ound these lakes, then, given the fact that human activity was much greater in developed areas (Chapt. 3), fewer birds than expected should have been found along developed shorelines. That the exact opposite was found suggests that there are considerable benefits to the modified habitat found along developed portions of these lakes, and that under such circumstances many waterbird species will tolerate increased levels of human disturbance. Tolerance of human disturbance, as indicated by a bird’s presence along developed shorelines, may be a sign of habituation. Habituation has been defined as “the relatively persistent waning of a response as a result of repeated stimulation which is not followed by any kind of reinforcement” (Hinde 1970). Other studies that have used similar presence/absence measures of habituation have had varying results. On separate refuge studies, Burger (1981) found that waterbirds were significantly less likely to be present when people were present at a site, whereas Kl ein at al. (1995) found that only half of the species in their study shifted away from areas of human disturbance as disturbance levels increased. Several authors though, have noted that it is common for waterbirds to habituate to moderate levels of disturbance in situations where people are regularly present but not causing any direct harm (Hockin et al. 1992, Weller 1999). As mentioned above, the dense, cattail found along undeveloped shorelines appeared to be unattractive to many waterbirds. Such habitat conditions may have limited visibility, foraging opportunities, and es cape routes, and increased vulnerability to predators. If this was the case, waterbirds may not have actually been showing a preference for developed habitat, but rather an avoidance of undeveloped habitat. When

PAGE 47

35 considering this option it is important to distinguish between undeveloped habitat and natural habitat. Though the undeveloped shorelines on these lakes had relatively undisturbed terrestrial vegetation structure, the deep, dense stands of cattail commonly dominating the littoral zones were most likely not part of the natural habitat originally found on these lakes. Rather, they were a result of the artificial eutrophication that has occurred on these lakes due to years of uncontrolled urban runoff (Gilbert 1987, Roth in press). Reestablishing a healthy, more diverse, and structurally heterogeneous aquatic plant community (i.e., “hemi-marsh”) along the undeveloped portions of these lakes might very well create a more favorable habitat than is currently available for the waterbirds on these lakes. Previous urban studies based in terrestrial habitats have documented an overall increase in avian abundance as a select few human-commensal species prosper (Blair 1996, Savard et al. 2000). This same process appeared to occur on a seasonal basis on the urban lakes in this study, as seen in the summer by the lower species evenness along developed shorelines, and the fact that the overall preference for developed shoreline was explained by just eight species. Further, more secretive species such as rails and bitterns, often found in less disturbed wetland habitats, were rarely encountered on these lakes. The combined implications are that certain waterbird species are adaptable enough to benefit from aquatic urban habitats and may actively seek them out, as species like the House Sparrow ( Passer domesticus ) and European Starling ( Sturnus vulgaris ) do in terrestrial urban habitats. Species like the Great Blue Heron or Belted Kingfisher, both found in this study, are highly adaptable in terms of habitat and diet (Butler 1992, Weller 1999), and this may explain why they are found in urban aquatic habitats. Other species,

PAGE 48

36 such as the Anhinga or Black-crowned Ni ght Heron, may be found in developed areas, but at lower numbers than in less disturbed habitats. And waterbirds such as rails or bitterns (found rarely in this study) may be entirely intolerant of development and/or disturbance and may avoid urban environments altogether when possible, even when portions of the environment are left undeveloped. Dominant Habitat Elements Tall emergent vegetation, open shore, lawn, and canopy were each associated with the distribution (presence/absence) of multiple guilds on these lakes, and were thus considered dominant habitat elements. Tall emergent vegetation had a negative overall association, whereas open shore, lawn, and canopy had positive overall associations. Littoral habitat Tall emergent vegetation All guilds were negatively associated with tall emergent vegetation on at least 50% of the lakes over both seasons. No significant positive associations were observed within a guild on any lake. Previous studies have shown similar results, with a variety of waterbird species using dense tall-emergent monocultures far less than expected by chance (Weller & Spatcher 1965, Weller & Fredrickson 1974, Kaminski & Prince 1984, Collopy & Jelks 1989, Bildstein et al. 1994, Smith et al. 1995, Surdick 1998). The most substantial stands of tall emergent vegetation were located along undeveloped shoreline. Cattail was present along 89% of total undeveloped shoreline, with almost 70% of it found in dense, continuous stands. These stands were frequently found extending over 20 m from shore into water depths that were too great for even the tallest wading birds to use. In the summer, the larger stands of cattail became so dense that even smaller marsh birds and ducks had to struggle to penetrate the vegetation when attempting to flee.

PAGE 49

37 Given its extensive coverage on many of the lakes, cattail overlapped other habitat elements quite frequently. In the case of wa ding birds, the only guild showing significant negative associations with multiple habitat elements, tall emergent vegetation (cattail) occurred in combination with other significant elements 83% of the time. Thus, it could not be conclusively determined whether birds were responding primarily to cattail or some combination of elements. However, given the abundance of previous research showing the avoidance of tall, dense, emergent vegetation by numerous waterbird species (see above), it seems likely that tall emergent vegetation, namely cattail, was indeed the dominant habitat element that birds were avoiding. Cattail was most likely present on these lakes before they were developed, but at much lower densities (Gilbert 1987, Florida DEP 1983-1992). In this study both the Least Bittern (summer) and the Green Heron (winter) showed a positive association with tall emergent vegetation. Cattail, even at fairly high densities, is considered functional, even necessary habitat for a number of waterbirds; the Least Bittern, Rail, Sora, Blackcrowned Night Heron, Common Moorhen, Pu rple Gallinule, Red-winged Blackbird ( Agelaius phoeniceus ) and Boat-tailed Grackle ( Quiscalus major ) (Terres 1991, Hoppe & Kennamer 1986, Davis 1993, Melvin & Gibbs 1996, Gibbs et al. 1992). Eliminating cattail from these lakes would therefore be detrimental to these species. However, limiting cattail coverage along undeveloped shoreline may allow for greater foraging and resting opportunities for a wider range of species. Onshore habitat Open shore Open shore had a positive association with marsh birds, wading birds, and ducks during the summer. The creation of much of this habitat, defined as moist soil or sand, was a result of the extreme drought that Florida experienced in 2001.

PAGE 50

38 Though ephemeral, open shore habitat appeared to serve as a valuable foraging area for these birds (pers. obs.). Dropping water levels exposed new foraging habitat that was not available during higher water conditions. As such, many dabbling and probing birds and small waders could take advantage of the shallow waters and exposed substrate. Of the eight species that were associated with developed shoreline in the summer, seven fit this description: Snowy Egret, Tricolored Heron, White Ibis, Wood duck, Mallard, Purple Gallinule, and Common Moorhen. All of these species are known to frequently forage in areas of relatively open shallow waters and sparse vegetation (Hancock & Kushlan 1984, Weller 1999). Given that only 10% of open shore was found along undeveloped shoreline, this habitat element may explain much of the overall preference for developed shoreline that was observed in the summer season. Lawn Lawn was positively correlated with marsh birds, wading birds, and diving birds during the winter. This habitat, an obvious indicator of development, by itself probably offered little functional value to most birds other than White Ibis, Cattle Egrets and Common moorhens, which frequented lawns for foraging. Substrate analysis showed that only five percent of all birds found along developed shoreline were directly found in lawns. Most birds were likely responding to other habitat elements of the developed shoreline located in the vicinity of lawns. For example, 40% of wading birds observed in areas where lawn was present were also found with low emergent vegetation. In general, areas with significant lawn coverage had reduced understory and shrub layers, and patchier emergent vegetation. Reduced onshore and littoral vegetation structure may have afforded birds better visibility and movement, allowing for easier detection and avoidance of predators or approaching humans. This idea is supported by previous

PAGE 51

39 studies that have suggested that many wading bird species are more vigilant and more easily disturbed in areas of dense vegetation (Smith et al. 1995, Safran et al. 2000). The patchier aquatic vegetation structure may have also provided better foraging habitat. Bildstein et al. (1994) found that patchy littora l vegetation structure allowed wading birds to feed in relatively open water while taking advantage of high fish densities in adjacent vegetated areas. Though no significant differences were observed in the proportion of birds foraging along developed and undeveloped shorelines (Chapt. 3), other measures, such as foraging times, prey selection, or strike/capture ratios, may have better shown the value of this habitat for foraging birds. Other components of developed shoreline that were associated with lawns included human-made structures such as piers, pylons, and boats. Examination of the substrates on which birds were observed revealed that 25% of all birds using developed shoreline were found on such structures. Many species used these structures for resting, probably due to the unobstructed access to water and the ease of vigilance. Diving birds offer the strongest example in that 63% of all divi ng birds along developed shoreline were found on piers or pylons. Birds such as the Tricolor ed Heron, Snowy Egret, and particularly the Belted Kingfisher were observed actively foraging from these perches. Canopy Canopy was positively associated with marsh birds and wading birds in the winter and diving birds over both s easons. Though canopy was found in greater densities along undeveloped shorelines, it was also prevalent along developed shorelines, and therefore cannot be considered an indicator of undeveloped shoreline. It can be seen as an increase in structural habitat complexity, providing birds with an added vertical layer that they could use as a refuge from disturbance. Several species of wading birds

PAGE 52

40 were observed resting in canopy. Eighty-three percent of all diving birds resting along undeveloped shoreline were found resting in trees or on logs. Casual observation suggests that these individuals may have shown lower levels of alert/flee response than birds resting closer to the water. In addition, canopy may have provided shade refuge in the summer during the heat of the day, and thermal refuge in the winter, protecting birds from winter winds. In both seasons, it ma y have also provided sunny locations for basking birds. This is especially important to Anhingas, which rely on basking to dry their feathers and maintain body temperature (Frederick & Siegel-Causey 2000). The association between marsh birds and canopy was most likely indirect. Besides canopy, marsh birds were also significantly associated with lawn and understory. Though marsh birds showed no clear association with any combination of these elements, the ecology of this guild suggests that understory may have been the primary habitat element determining their distribution. In general, many members of the rail family show a strong preference for relatively dense understory and generally weedy conditions (Terres 1991, Elphick et al 2001). Guild Responses to Other Habitat Elements Marsh birds Marsh birds also showed a positive association with understory in the winter. As mentioned above, this finding was expected given the general ecology of this guild. However, though lawn, understory, and canopy overlapped along only 2% of total significant habitat shoreline, 25% of winter marsh birds were found in this area. Thus, though understory was most likely the dominant habitat element, it can not be ruled out that marsh birds were selecting for a combination of these elements.

PAGE 53

41 Wading birds Whereas wading birds were only associated with tall emergent vegetation and open shoreline in the summer, all habitat elements but understory were found to be significant in the winter. A positive association was observed with low emergents, lawn, and canopy, whereas a negative association was observed with floating-leafed vegetation and shrubs. Such variation between seasons suggests that wading bird habitat preferences may be more seasonally dependent than other waterbird guilds, and that other habitat elements, such as water depth, or groups of habitat elements, may be associated with this guild’s within-lake habitat choices. Summer results suggest that wading birds may have been primarily responding to a combination of water depth and vegetation structure. Water depth has been frequently cited as a significant factor determining the distribution of wading birds (Hancock & Kushlan 1984, Weller 1999, Bancroft et al. 2002). In the summer, this guild was found in greatest abundance in areas of open shore. Associated littoral zones in these areas were typically dominated by open water or sparse emergent vegetation. While all lakes had shallow water margins, the predominance of emergent vegetation often prevented waders access to these areas. The lack of emergent vegetation along open shore areas, when water levels dropped, allowed wading birds to utilize these shallower waters, providing optimal foraging conditions for many of these birds (Hancock & Kushlan 1984, Breininger & Smith 1990, Bildstein et al. 1994). In the winter, in the absence of extreme drought conditions, shallow water areas became more limited. Wading birds, in turn, may have become more generalized in their habitat selection. Given that tall emergent vegetation continued to be used less than expected, and open shore, even at just 28% of summer availability, continued to be used

PAGE 54

42 more than expected, these habitats may have continued to be the primary elements to which wading birds were responding. However, the abundance of significant responses to other habitat elements suggests that this guild may have been responding to combinations rather than just individual habitat elements. One such example would be low emergent/lawn habitat. Though this habitat made up 29% of significant habitat shoreline, 40% of wading birds occurring with one or more significant habitat element were found in low emergent/lawn habitat. Diving birds Diving birds showed a negative association with lawn in the summer and a positive association with understory in the winter. The negative association with lawn in the summer is puzzling and cannot be readily explained by the ecology of these birds or by lawn’s association with other habitat elements. The positive association with understory in the summer may be the result of diving birds selecting areas where understory and canopy overlapped. Understory/canopy habitat made up 26% of significant habitat shoreline, yet 57% of diving birds were found in this area of habitat overlap. Given the ecology of these birds and the results of this study showing diving birds frequently resting in the canopy, it seems likely that they were selecting either for canopy or canopy/understory, rather than just understory alone. Ducks Besides showing a negative association with tall emergent vegetation and a positive association with open shore in the summer, ducks showed a positive association only with floating-leafed vegetation during both seasons. Although only two lakes had enough floating-leaf habitat for analysis, both lakes showed strong associations (p < 0.01). On both lakes the primary floating-leafed species was spatterdock ( Nuphar

PAGE 55

43 luteum ), a type of water lily whose leaves stand above the water. Structurally acting as a low emergent species, and occurring at high densities, this species offered exceptional cover for waterfowl, particularly Wood Ducks. Tarver et al. (1978) also noted that Spatterdock seeds may be a valuable secondary food source for waterfowl in northern Florida. Floating-leafed vegetation overla pped with open shore along 12% of significant habitat shoreline, while 25% of ducks were found where these habitats overlapped. Though not conclusive, it could not be ruled out that ducks may have also been responding to this combination of habitat elements. Management and Future Research More comprehensive and long-term research is strongly recommended for urban lake habitats. Further research needs to focus on whether these lakes are truly providing valuable habitat to these birds or whether they are actually acting as biological sinks. Though most birds in this study appeared to be healthy, and both marsh birds and ducks produced numerous young, the long-term effects of human impact, such as disturbance and pollution, need to be examined. Developed shoreline around the lakes in this study clearly provided useable habitat for a variety of waterbird species, and may have actually been selected for by some species. However, the fact that over both seasons all guilds were negatively associated with tall emergent vegetation, which was predominantly found along undeveloped shoreline, suggests that it was avoidance of this habitat element that was responsible for the significantly greater proportion of birds along developed shoreline. In smaller, discrete stands, when interspersed with other plant species and patches of open water, cattail may provide habitat for a wider range of species. Weller (1999) has noted that a wide range of wetland birds prefer such “hemi-marsh” conditions. Such interspersion

PAGE 56

44 would increase structural habitat complexity and open up shallow areas closer to shore, creating preferred foraging habitat for many wading and dabbling birds. More detailed habitat studies should be conducted to determine optimal cattail densities for different avian species or guilds, and whether such management would be a feasible option. Currently, Florida’s Bureau of Invasive Plant Management has no active management plans for this plant species. Several onshore habitat elements were significantly associated with the distribution of waterbird guilds, indicating that terrestrial habitat may play a role in habitat selection for many of these birds. Specifically, open shore, lawn, and canopy appeared to be associated with multiple guilds. As shorelines continue to be developed, much of the natural habitat is being altered or removed both by developers and property owners. Florida Law (§369.20(7), Florida Statutes) states that “no person or public agency shall control, eradicate, remove, or otherwise alter any aquatic weeds or plants in waters of the state unless a permit for such activity has been issued by the department.” There are currently no such laws protecting onshore habitat. Future studies should specifically test the degree of importance of terrestrial vegetation to waterbirds in order to determine whether this habitat needs greater protection. Until further research is conducted, managers should consider the ecology of these birds when prioritizing habitats for protection. Since very few birds were observed nesting on these lakes (Chapt. 3), preliminary management should be based on the foraging and resting behavior of these birds. Wading birds, in general, prefer shallow, relatively open littoral zones. This study also showed potential associations with open shore and lawn. These conditions are currently often available along developed

PAGE 57

45 shorelines. The fact that many of the wading bird species using these lakes were found significantly more often along developed shorelines suggests that little further management may be required for this guild. Exceptions such as the Least Bittern and Green Heron should be noted though. These species often show a preference for dense, tall emergent vegetation, which was found in greater abundance along undeveloped shorelines. Management for these species should be focused there. Marsh birds and ducks are often found in open water interspersed with mixed emergent and floating-leafed vegetation. In this study, both guilds may have selected for open shore in the summer, while marsh birds may have also been associated with lawn and understory in the winter. Many of these conditions are again being met along areas of developed shoreline, but in limited extent. Management should consider increasing low emergent and floating-leafed vegetation in developed areas lacking frequent boat traffic if increasing marsh bird and duck abundance is desired. Though nesting was not observed for these guilds, it should be noted that many marsh birds do require dense tallemergent vegetation in the breeding season. Portions of this habitat should therefore be protected along undeveloped shoreline. Many of the diving birds require larger ar eas of deeper open water for foraging, and canopy or other perching substrate for resting. Open water is currently managed for to promote lake use for fishing and other recreation. Common diving birds on these lakes, such as Anhingas and Double-crested Cormoran ts, often prefer resting areas immediately over the water. Developed shorelines, with their abundance of docks and boat houses, are thus well suited to these birds. However, diving birds were the one guild that did not have greater than expected numbers along developed shoreline. Management might

PAGE 58

46 therefore focus on protecting areas of undeveloped shoreline, as woody shoreline is more readily available there. This project, being exploratory in nature, attempted to determine whether dominant habitat elements existed that independently explained waterbird distributions. Several elements, both aquatic and terrestrial, did indeed appear to be significantly more influential. Though tests for independence were inconclusive, many of the habitat elements selected in this study may have been correlated with one another, both spatially and perhaps functionally. Future studies would do well to examine waterbird/habitat associations at broader scales (e.g., Hostetler and Holling 2000) to investigate whether birds are responding to suites of habitat elements that comprise general macro-habitats around these lakes. Duda (1987) found that 88% of Floridians enjoyed having birds near their homes and frequently engaged in activities to benefit them. Lake property owners should be informed of the abundance of waterbirds using the developed shorelines on their lakes. Many of these individuals might be very receptive to an education program on the value of shoreline habitat and the ways that homeowners can manage their property to meet their own needs as well as the needs of w ildlife. Even with current conditions, many lakeshore residents are inadvertently creating habitat conditions favorable to many waterbirds by managing their properties for more heterogeneous littoral and onshore habitat in order to have direct lake access. State laws allow residents to clear up to 50%, or 50 ft (whichever is less) of the aquatic vegetation along their shoreline to allow for boating and other recreational access. With education, residents can learn to manage

PAGE 59

47 their own properties to benefit both themselves and wildlife, and may even be able to help managers improve waterbird habitat along undeveloped shorelines. Soliciting public participation in future research, monitoring, and management efforts (i.e. citizen science) would also increase public awareness and generate a greater sense of stewardship on lands that may not have been previously thought to have much wildlife value. Citizen science can also generate valuable data sets supporting long-term monitoring efforts (Hoyer et al. 2001). The popularity of such public involvement is demonstrated by Florida LAKEWATCH, a lake-monitoring program that relies on a pool of over 1,800 trained volunteers to monitor water quality on approximately 1,200 lakes around the state. Beginning in 2001, Lakewatch incorporated aquatic-bird surveys into its monitoring program. This new wildlife component is being readily adopted by many of the volunteers and holds great promise for future research and management efforts on many of Florida’s lakes.

PAGE 60

48 CHAPTER 3 AVIAN BEHAVIORAL RESPONSES TO SHORELINE DEVELOPMENT Introduction The process of urbanization may affect birds both through changes in ecosystem processes, habitat structure, and food supply, and through changes in predation pressure, competition, and disease (Marzluff 1997). Avian responses can take two forms: changes in behavioral patterns, and changes in community composition. The majority of urban bird studies to date have examined the direct effects of development and habitat alteration on community composition by focusing on changes in avian abundance, species diversity, richness, and evenness (e.g. DeGraaf & Wentworth 1981, Tilghman 1987, Blair 1996, Clergeau et al. 1998, Jokimaki 1999, Ro ttenborn 1999). Very few studies have attempted to address how avian behavior is affected by development (As examples, see: Andersson et al. 1980, Galeotti et al. 1991, Ge lbach 1994, Fernandez-Juricic et al. 2001). Studies examining changes in avian behavioral patterns have focused primarily on the effects of human activity (e.g., Hockin et al. 1992, Knight & Gutzwiller 1995, Jozkowicz & Gorska-Klek 1996). Human disturbance has many negative impacts to wildlife and may ultimately reduce wildlife densities and diversity at both local and regional scales (Boyle & Samson 1985, Cole & Knight 1990). On the behavioral level, wildlife responses to disturbance may include attraction, avoidance, or habituation (Knight & Cole 1991). Attraction behavior, such as increased foraging and reduced wariness, can result from human actions like supplemental feeding (Knight & Temple 1995). Hunting typically results in avoidance behaviors such as increased wariness,

PAGE 61

49 altered foraging patterns, and reduced nest defense (Kenney & Knight 1992, Knight & Cole 1995). Increased flight times or increased aggression towards humans are other avoidance responses that may result from excessively close or frequent human disturbance (Knight & Temple 1986, Kahl 1991). Habituation, on the other hand, may occur in response to human disturbance when that disturbance is not associated with either a positive or negative reward (Eibl-Eibesfeldt 1970). In this case, wildlife species appear to show no signs of altered behavior in response to human presence. A wide range of avoidance behaviors have been documented for waterbirds in response to recreationists or researchers in natural areas. Examples include reduced foraging and resting periods (Owens 1977, Ka iser & Fritzell 1984, Burger and Gochfeld 1991, Skagen et al. 1991), increased nest abandonment and egg loss (Kury & Gochfeld 1975, Tremblay & Ellison 1979, Titus & van Druff 1981), discouragement of late-nesting pairs from breeding (Ellison & Cleary 1978, Tremblay & Ellison 1979), and disruption of pair bonds (Tindle 1979) and parent-offspring bonds (Oldfield 1988). Other studies have shown that human disturbance can increase vigilance (Burger and Gochfeld 1991), flushing (Vos et al. 1985), flight times (Kahl 1991, Korschgen et al. 1985), and energy expenditure by birds and reduce their overall energy intake (Belanger & Bedard 1990). Conversely, other studies have shown that in cases where humans are regularly present without posing an immediate threat of harm, waterbirds appear to habituate to some forms of disturbance (Hockin et al. 1992, Weller 1999). Some examples include New Zealand Dotterels ( Charadrius obscurus ) allowing closer human approach on highuse beaches (Lord et al. 2001); Great Crested Grebes ( Podiceps cristatus ) showing reduced flush distances in sites frequently visited by humans (Keller 1989); nesting

PAGE 62

50 Ospreys ( Pandion haliaetus ) showing greater levels of hab ituation in areas of high human activity than in more remote sites (Swenson 1979, Poole 1981); Great Blue Herons ( Ardea herodias ) habituating to the activities of fishermen boating past a heronry (Vos et al. 1985); and Greylag Geese ( Anser anser ) habituating to people walking past as long as the people remained on paths (Kuhl 1979). The objective of this portion of the study was to determine the effects of urbanized lakes on the behavioral patterns of waterbirds. The specific questions to be answered were as follows: 1. Are primary waterbird behaviors diffe rent between developed and undeveloped shoreline? 2. Which waterbirds appear most sensitive to human disturbance? Question 1 addresses the general lack of knowledge about avian behavior on urban lakes. Many studies have compared waterbird behavior in areas of high human use with areas of low human use (e.g. Burger 1981, Keller 1989, Jozkowicz 1996), but have either been based in natural habitats such as refuges or areas where disturbed and undisturbed sites were well apart from one another. Question 2 focuses specifically on avoidance behavior and sensitivity to human disturbance by examining which guilds appear most vulnerable to disturbance. Numerous studies have shown that sensitivity to human disturbance can vary considerably between species (e.g. Burg er 1981, Bratton 1990, Klein 1993, Klein et al. 1995), with several studies showing greater sensitivity specifically in winter migrants (van der Zande et al. 1980, Klein et al. 1995). Both of these hypotheses are examined.

PAGE 63

51 By understanding how waterbirds behave in urban settings and how sensitive they are to human disturbance, managers will be better equipped to balance the needs of humans and wildlife on urban lakes. Methods Behavioral surveys of waterbirds were conducted in the summer from June 7– August 1 of 2001 and in the winter from December 8– February 6 of 2001/2002. The term ‘waterbird’ referred to species in the orders Gaviiformes, Podicipediformes, Pelecaniformes, Ciconiiformes, Anseriformes, Falconiformes, Gruiformes, Charadriiformes, or Coraciiformes observed on or feeding from lacustrine habitats. Each of the four lakes were surveyed a total of eight times each season by driving at minimumwake speed around the lake perimeters (20 – 30 m from shore) in a small motorized canoe. Both the order of lakes and the direction of travel were alternated for each day of surveying. Surveys were conducted within the first five hours after sunrise on mornings with little to no rain and winds less than 24 km/hr. Lake shoreline was categorized as eith er developed or undeveloped based on DEP land cover classifications for each lake. Classi fications were modified during preliminary surveys by basing the categorization only on the land cover within the first 20 m of shoreline extending away from the water on each lake. For the purposes of this study, developed shoreline referred to any continuous expanse of shoreline greater than 100 m, parallel to the edge of the lake, that had a minimum of 50% noticeable, long-term habitat alteration, defined as cleared land, lawns, landscaping, buildings, and roads. Undeveloped shoreline was defined as any continuous stretch of shoreline greater than 100 m, parallel to the edge of the lake, with greater than 50% intact natural habitat, and little to no sign of regular human use. Developed and undeveloped areas were separated

PAGE 64

52 by a buffer of 40 m to eliminate the potential effects of converging habitats. Birds recorded in these border areas were not included in analyses. In addition, behavioral analyses were not conducted on birds if their origin or destination were not observed, or if they were observed greater than 30 m offshore or 10 m onshore. Upon sighting a bird along either developed or undeveloped shoreline, I recorded the first behavior that was observed. During preliminary surveys I found five primary behaviors: active/swimming, alert/fl eeing, foraging, resting, and tending young. Active/swimming birds included those birds th at were actively moving but not foraging or showing obvious signs of distress. Birds were classified as alert/fleeing if their attention appeared focused on the boat or another nearby disturbance, or if they vacated the location in which they were observed in response to the boat’s approach. Foraging behavior was recorded for any bird monitoring the water, stalking, actively gleaning, or consuming prey. Birds were classified as tending young any time they were observed building or sitting on a nest, or in the presence of young, regardless of any other behaviors in which they may have been engaged. This behavior was only observed in the summer. Over the course of the two seasons, these five behaviors made up 89% of all recorded observations. Nine other behaviors were also recorded over the course of both seasons (e.g. aggression, calling, preening, etc.), but made up less than 12% of all observations. Because of the dominance of the five primary behaviors, I chose to restrict analyses to these behaviors only. Disturbances other than the survey boat, such as anglers, jet skiers, water skiers, or people feeding birds, were also noted. In addition, several random resident contacts were

PAGE 65

53 made in order to estimate the use of the numerous Wood Duck ( Aix sponsa ) nest boxes observed on these lakes. Analyses Waterbirds were grouped into guilds for analyses. Guilds were based on foraging behavior and habitat use, and included wading birds (Ardeidae, Threskiornithidae, Ciconiidae, Recurvirostridae), marsh birds (Rallidae), surface and aerial diving birds (Podicipedidae, Phalacrocoracidae, Anhingi dae, Accipitridae, Laridae), and ducks (Anatidae). Analyses were conducted for the four most commonly observed behaviors within each season. For a guild to be included in analyses for a given lake, at least 25 birds had to be observed for that lake during a season. Shoreline Development I examined the effect of shoreline development on guild behavior using contingency Chi-square tests ( = 0.05, df = 1), by comparing the proportions of birds engaged in each behavior along developed and undeveloped shoreline. Expected proportions were based on the number of birds exhibiting a given behavior versus the number of birds exhibiting any other beha vior, including the four focal behaviors observed each season and all other behaviors that were observed. Separate analyses were run for each lake on the four focal behaviors observed each season. Disturbance Sensitivity Measures of waterbird sensitivity to disturbance were based on alert/flee behavior. I tested whether the number of birds exhib iting alert/flee behavior differed significantly between seasons and between guilds. I used contingency Chi-square tests ( = 0.05, df = 1) for comparisons between seasons, and 2x4 contingency Chi-square tests ( = 0.05, df = 3) for comparisons between guilds. Expected proportions were based on the number of

PAGE 66

54 birds exhibiting alert/flee behavior versus the number of birds exhibiting any other behavior. Lakes were tested individually fo r both tests, with developed and undeveloped shorelines combined. Summer and winter data were combined for comparisons between guilds. Using contingency Chi-square tests ( = 0.05, df = 1), I also tested whether winter migrant species showed greater alert/flee behavior than resident species of the same guilds. Analyses were run on winter data for all lakes combined due to low numbers on individual lakes. Determining Overall Significance For all analyses conducted on a lake-by-lake basis, behavioral patterns were considered to have overall significance where significant same-direction associations (p 0.05) were observed on at least three lakes without a significant contradictory finding on the fourth lake. In a season, if a guild was only observed with enough frequency for analysis on two lakes, then behavioral patterns were considered to have overall significance if both of the lakes had significant same-direction associations (p 0.05). In cases where results were suggestive of an overall pattern on three or more lakes (p 0.2 for each lake), but not necessarily significan t, and a contradictory finding wasn’t found on the fourth lake, probabilities (for lakes indicating a pattern) were then combined for meta-analysis (Fisher 1958). A two-tailed Fisher’s exact test ( = 0.05) was applied to all analyses where expected values dropped below five. Results Seasonal Behavioral Observations In the summer, a total of 2817 behavioral observations were recorded for the four study guilds (Table 3-1). The four most common behaviors were alert/fleeing, foraging,

PAGE 67

55 resting, and tending young. Resting was the most commonly recorded behavior, accounting for 32% of all observations. Fora ging birds accounted for 26% of behavioral observations, followed by alert/fleeing birds (18%), and birds tending young (10%). All other behaviors accounted for 14% of recorded observations. In the winter, 2438 behavioral observations were recorded for the four focal guilds (Table 3-1). Alert/fleeing behavior was mo st commonly recorded, accounting for 33% of all observations. Resting behavior was observed in 26% of all birds, followed by foraging (23%), and active/swimming (12%). Table 3-1. Percent of waterbird guilds engaged in focal behaviors during summer (S) 2001 and winter (W) 2001/2002. Guild n Alert/ Flee (%) Forage (%) Rest (%) w/ Young (%) Active/ Swim (%) S W S W S W S W S W Waders 662 530 10 10 54 35 26 5 1 2 Marsh 673 723 5 19 46 46 6 5 23 26 Divers 299 361 11 23 2 68 83 68 0 6 Ducks 1183 671 31 67 4 2 38 8 9 9 n = total number of birds observed each season Behavioral Associations with Shoreline Development Alert/flee behavior Alert/flee behavior showed an overall pattern of significant association with undeveloped shoreline for ducks in the summer (Table 3-2) and wading birds in the winter (Table 3-3). Across all lakes in the summer, 72% (n=89) of the ducks found along undeveloped shoreline displayed alert/flee behavior. On the two lakes where ducks were found with enough frequency for analysis, they displayed alert/flee behavior significantly more along undeveloped shoreline (both tests: 2 28.98, p < 0.0001). Thirty-three percent of all wading birds found along undeveloped shoreline in the winter displayed

PAGE 68

56 alert/flee behavior (n=54). Wading birds disp layed this behavior significantly more along undeveloped shoreline on three of the four lakes (all tests: 2 8.7, p 0.012). Possible trends were observed in the summer for wading birds, marsh birds, and diving birds (Table 3-2). Both wading birds and marsh birds showed increased alert/flee behavior along undeveloped shoreline on two lakes (all tests: 2 2.74, p 0.12), whereas diving birds showed increased aler t/flee behavior along developed shoreline on two lakes (both tests: 2 2.59, p 0.14). Foraging behavior Fifty-eight percent of the wading birds (n=512) observed along developed shoreline in the summer were engaged in foraging behavior (Table 3-2). Foraging wading birds showed a significant overall preference for developed shoreline on three lakes (metaanalysis: 2 = 27.07, p < 0.001). The same preference was observed in the winter, but only on two lakes, and therefore lacked overall significance. Winter marsh birds displayed a significant overall preference for foraging along undeveloped shoreline on three lakes (meta-analysis: 2 = 13.78, p = 0.03). Overall, 51% of wintering marsh birds (n=74) found along undeveloped shoreline were engaged in foraging behavior. Resting behavior In the summer, on the two lakes with sufficient duck numbers, ducks showed a significant preference for resting along devel oped shoreline (Table 3-2; both tests, 2 11.35, p < 0.001). Overall, 40% of the ducks (n=1094) observed along developed shoreline were engaged in this behavior, versus just 4% along undeveloped shoreline. Though lacking overall significance, diving birds appeared to show a weak trend of resting in greater proportions along undeveloped shoreline in the summer ( 2 2.59, p 0.14 on two lakes).

PAGE 69

57 Tending young behavior During the summer, 100% (n=111) of ducks tending young did so along developed shoreline (Table 3-2). On both lakes where ducks were observed with enough frequency for analysis (n 25), they tended young significantly more along developed shoreline (both tests: 2 4.17, p 0.04). The only wading birds observed tending young were two pairs of Green Herons ( Butorides virescens ) on Lake Buckeye. No diving birds were observed tending young on any of the lakes. Active/swim behavior During the winter, a significant overall pattern of association was observed between active/swimming marsh birds and developed shoreline (Table 3-3). Twentyeight percent (n=676) of the marsh birds found along developed shoreline were observed engaged in active/swim behavior, versus only 9% along undeveloped shoreline. This pattern was observed on all four lakes (meta-analysis: 2 = 27.98, p < 0.001). Human Activity Though not quantified, the level of human activity on or immediately around all four lakes appeared relatively low during both seasons. Given that surveys were typically conducted during the week, and that the summer drought made lake access very difficult, the amount of activity that was observed was no doubt considerably lower than at peak times such as weekends during peak fishing season. High-intensity recreational activities such as water skiing or jet skiing were not observed. Though fishing from boats appeared to be the most common recreational activity, no more than four boats were ever encountered on a lake at one time. Anglers were typically found slowly trolling along undeveloped shoreline, fishing along edges of emergent or floating-leaf vegetation, and were rarely observed disturbing or displacing waterbirds. No land-based human activity

PAGE 70

58 was observed along undeveloped shorelines. Human activity along developed shorelines included feeding ducks and marsh birds, fishi ng and relaxing on piers, fishing from boats, lawn mowing, golf cart operation, construction, a nd small aircraft departure and arrival. Several informal interviews with local residents who had Wood Duck nest boxes on their properties revealed that many of these boxes were used on a yearly basis, and frequently produced successful broods. Table 3-2. Percent guild behavioral responses to developed (D) and undeveloped (U) shoreline for summer 2001. Results listed by lake. Guild n Alert/Flee Forage Rest w/Young D U % D % U % D % U % D % U % D % U Waders Buckeye 52 29 23 14 29 41 31 41 12* 0 Conine 212 68 8 16* 71** 43 17 28* Deer 141 17 6 18* 52* 29 27 29 Jessie 107 36 7 11 52* 39 33 28 Marsh Buckeye 137 21 4 33* 39 52 3 5 33** 0 Conine 48 28 6 11 50 61 19 21 Deer 283 22 2 23* 52* 36 4 5 23 14 Jessie 113 21 2 0 37 48 4 5 34 38 Divers Buckeye 21 6 38* 0 57 100* Conine 50 38 12* 3 4 3 78 89* Deer 61 11 3 0 93 91 Jessie 57 55 14 16 82 78 Ducks Buckeye 2 7 Conine 15 0 Deer 725 50 25 60** 3 0 43** 4 10** 0 Jessie 352 32 36 88** 5 0 36** 7 12** 0 n = total number of birds observed along developed and undeveloped shoreline. Trend suggesting greater percentage of guild disp layed behavior along indicated shoreline (p 0.2). ** Significantly greater percentage of gu ild displayed behavior along indicated shoreline (p < 0.05). Dash indicates less than 25 birds observed on lake or no birds observed exhibiting behavior.

PAGE 71

59 Table 3-3. Percent guild behavioral responses to developed (D) and undeveloped (U) shoreline for winter 2001/2002. Results listed by lake. Guild n Alert/Flee Forage Rest Active/Swim D U % D % U % D % U % D % U % D % U Waders Buckeye 71 13 8 38** 61** 8 24 46* 6 8 Conine 133 14 9 36** 26 14 62* 36 1 7* Deer 127 1 4 0 28 0 66 100 2 0 Jessie 152 26 7 31** 43** 19 47 50 1 0 Marsh Buckeye 57 19 25 32 25 42* 9 11 39* 16 Conine 78 20 13 25 33 50* 9 5 41* 20 Deer 442 24 19 42** 50 46 5 4 23** 0 Jessie 99 11 8 9 49 82** 3 9 34** 0 Divers Buckeye 52 32 17 22 75 72 8 3 Conine 58 13 19 23 3 0 71 69 5 8 Deer 56 3 43 67 41 33 16 0 Jessie 99 51 15 27* 3 0 77 73 4 0 Ducks Buckeye 5 9 100 56 0 44 Conine 25 2 0 100**100**0 Deer 629 19 66 68 2 0 8 0 7 0 Jessie 88 10 90 100 10 0 n = total number of birds observed along developed and undeveloped shoreline. Trend suggesting greater percentage of guild disp layed behavior along indicated shoreline (p 0.2). ** Significantly greater percentage of gu ild displayed behavior along indicated shoreline (p < 0.05). Dash indicates less than 25 birds observed on lake or no birds observed exhibiting behavior. Disturbance Sensitivity Seasonal alert/flee comparisons Guild analyses comparing alert/flee behavior between seasons showed a significant increase in alert/flee behavior in the winter for marsh birds and ducks (all tests: 2 6.2, p 0.01). Diving birds showed a similar trend, with birds on two lakes showing significantly greater alert/flee behavi or in the winter (both tests: 2 4.73, p 0.03). The proportion of wading birds displaying alert/flee behavior did not change between seasons

PAGE 72

60 ( 2 0.11, p 0.74). Across all guilds, alert/flee behavior was observed 1.6 times more often in the winter. Inter-guild alert/flee comparisons Ducks showed a significant pattern of greater alert/flee behavior than other guilds on three lakes (all tests: 2 61.2, p < 0.0001). With all lakes combined, 32% (n=1183) of summer ducks (95% of which were Wood Ducks) displayed this behavior. Sixtyseven percent (n=787) of winter ducks disp layed alert/flee behavior. Winter ducks included Wood Ducks, Blue-winged Teal ( Anas discors ), and Ring-necked Ducks ( Aythya collaris ). In both seasons, the next closest guild was diving birds, which displayed alert/flee behavior in 11% (n=299 for summer) and 25% (n=345 for winter) of all observations. Diving birds actually displayed alert/flee behavior considerably more than results indicate. However they appeared to tolerate a much closer approach distance than ducks before displaying alert/flee behavior (pers. obs.), and thus their first observed behavior was typically something other than the alert/flee response. Ducks on the other hand, tended to become alert or flee from the boat’s approach even from substantial distances (> 50 m) (pers. obs.). Migrant versus resident alert/flee comparisons Six out of seven winter migrant species were either ducks or diving birds. Alert/flee behavior was compared between migrant and resident species of these guilds. Analyses for all lakes combined failed to show a significant difference for either guild (both tests: 2 0.13, p 0.72). Of note though was that only 1% of winter migrants were found using undeveloped shoreline.

PAGE 73

61 Discussion Avian Responses to Shoreline Development Alert/flee behavior Alert/fleeing behavior was observed signi ficantly less along developed shoreline on the majority of lakes for summer ducks and winter wading birds. Though no other guilds showed overall significance, trends suggest that this pattern was the case for all guilds over both seasons (except summer divers). The findings suggest that many of the birds on these lakes showed localized habituation to human disturbance, tolerating such disturbance only along developed shoreline. Numerous other studies have noted patterns of apparent habituation in areas where waterbirds are regularly exposed to moderate or high levels of human activity. In her study of Great Crested Grebes ( Podiceps cristatus ), Keller (1989) found reduced flush distances for birds in sites frequently visited by humans. Lord et al. (2001) found that New Zealand Dotterels ( Charadrius obscurus ) nesting on high-use beaches allowed closer approach distances before flushing than birds nesting on remote beaches. Vos et al. (1985) found that Great Blue Herons were highly sensitive to disturbance early in the nesting s eason, but otherwise appeared to habituate to repeated non-threatening activities. And Burger and Galli (1987) found that the proportion of gulls fleeing when disturbed was gr eater in areas of infrequent disturbance than in heavily disturbed areas. However, the fact that summer diving birds showed a trend of greater alert/flee behavior along de veloped shoreline emphasizes that not all species may habituate. In the previous studies mentioned above disturbed and undisturbed sites were separate and unique. In this study disturbed and undisturbed sites were juxtaposed on the same lake for each of the four lakes sampled. The findings show that individual birds

PAGE 74

62 within a guild, ducks and waders in particular, may differentiate between disturbed and undisturbed sites even at a very localized scale. Even where undeveloped shoreline made up as little as 1/5 of the total habitat, such as on Lake Deer, wading birds (summer), ducks (summer), and marsh birds (both seasons) still showed greater sensitivity to disturbance in this habitat. This implies that undeveloped shoreline meets a behavioral need of at least some individuals, and may be seen as a localized refuge from disturbance. Results suggest that perhaps even small areas of undeveloped shoreline are important in order to minimize the stress to these birds. It may be that some individuals that are less tolerant to disturbance only use undeveloped shoreline. Alternatively, those same individuals may utilize both undeveloped and developed shoreline. To determine this, individual behavioral observations were required, which were beyond the scope of this study. Foraging and resting behavior Wading birds selected developed shoreline for foraging in the summer. The removal of emergent and shoreline vegetation by property owners on these lakes most likely promoted foraging conditions favorable to these birds (Chapt. 2). Large expanses of relatively open shallow water were favored by stalking waders, such as herons and egrets, while open moist shoreline was utilized by probing birds, such as ibis. Both shallow open water and open shore were extremely limited along undeveloped shorelines (Chapt. 2). The greater overall significance observed in the summer was most likely linked to the extreme drought conditions during this study, which served to greatly increase the availability of this foraging habitat. Marsh birds selected undeveloped shoreline for foraging in the winter. Rallids, in general, have strong seasonal shifts in diet, foraging primarily on animal matter in the

PAGE 75

63 summer and plant matter in the winter (Elphick et al. 2001). This shift in diet may have facilitated a winter shift to undeveloped shoreline, where greater densities of littoral vegetation were found (Chapt. 2). Ducks showed a strong preference for resting along developed shoreline in the summer. This serves as another indication of this guild showing an increased tolerance for human disturbance along developed shoreline. That summer diving birds showed a trend in the opposite direction reiterates the idea that this guild, or least the year-round resident species in this guild, may not be as tolerant as other species. Tending young behavior During the summer, 93% of ducks that tended young did so along developed shoreline. This pattern suggests that either the value of developed shoreline habitat or the disadvantages of undeveloped shoreline habita t strongly outweighed the disadvantages of human disturbance. This was best exemplified by the Wood Duck. Despite being one of the most sensitive species in this study in terms of alert/fleeing behavior (see Disturbance Sensitivity below), Wood Ducks were reported to consistently use the next boxes located on numerous lakeshore properties. Wood Ducks are known to readily use artificial nest boxes often located in or around human-populated areas. Further, all Wood Ducks observed tending young in this study were found along developed shoreline. Previous studies have shown that areas of developed lake shoreline have significantly lower fish species richness and total abundance than natural areas (Guillory et al. 1979, Bryan & Scarnecchia 1992). Other studies have shown the importance of dense stands of emergent vegetation as breeding areas and nurseries for fish and invertebrates (Wegener et al. 1973, Barnett & Schneider 1974). Given these findings and the fact that no significant differences were found in the amount of foraging observed

PAGE 76

64 along developed and undeveloped shorelines for either marsh birds or ducks (the two guilds observed with young), it seems unlikely that the greater abundance of birds tending young along developed shoreline can be explained by better foraging opportunities. However, on several occasions lake residents were observed putting out corn and bread for the birds. If this supplemental feeding occurred on a large enough scale, this could help explain the apparent preference for developed shoreline. In a review of urban bird studies over the pa st 100 years, Marzluff (2001) found 29 studies that linked supplemental feeding with increased bird densities in developed areas. Alternative possibilities that could explain this pattern include easier predator detection and reduced natural predators along developed shorelines. The timing of this study did not coincide with the nesting season of most of these birds. Observed tending young behavior therefore consisted of interactions between parents and fledglings. Further research is needed to determine which habitats breeding birds use for nesting on these lakes. Several of the species in this study, including Common Moorhens, Purple Gallinules, Rails, Soras, Least Bitterns, and Green Herons, are either facultative or obligate emergent vegetation nesters (Terres 1991). Three of these species were found actively nesting in emergent habitat in a previous study of neighboring urban lakes (Roth, in press). Ma ny wading birds as well as the Anhinga nest colonially in shrubs or trees close to waterbodies (Hancock & Kushlan 1984, Elphick et al. 2001). Considering that both emergent vegetation and woodlands were found in considerably greater abundance along undeveloped shorelines, and that many birds are least tolerant of disturbance at the beginni ng of the nesting season (Hockin et al. 1992), it

PAGE 77

65 seems likely that many of the birds in this study would have selected undeveloped shorelines for nesting. With few exceptions, wading birds and diving birds did not appear to breed or raise young on these lakes. One possibility is that these guilds were so intolerant of human disturbance during the breeding season that they sought out altogether less disturbed habitats for nesting and rearing young. Miller (1943) suggested that distance from human activity was the most important factor in heron nest site selection. However, numerous studies have documented these guilds breeding in disturbed or developed areas (e.g. Robertson & Flood 1980, Titus & Van Druff 1981, Vos et al. 1985, Keller 1989). Further, in a similar study on neighboring deve loped lakes, Roth (in press) found heron rookeries along the edges of several of his lakes. Another possible scenario is that wading and diving birds did not breed on these lakes due to the extreme drought. Several studies have shown that in cases of extreme water regimes, waterbird nesting has been delayed (Weller & Spatcher 1965, Custer et al. 1996), abandoned (Breeden & Breeden 1982), or moved to more favorable sites (Weller 1999). This is just speculation however, and there are a variety of other factors, such as the limited amount of undeveloped habitat, and the composition of this habitat, that may have been unfavorable to these birds. Active/swimming behavior Active/swimming behavior, examined only in the winter, was seen significantly more along developed shoreline in marsh birds. Active/swim behavior can be viewed as a low-distress behavior, as opposed to alert/flee, and as such may be another indication of habituation among these birds along developed shoreline. All three of the common marsh bird species observed, the Common Moorhen, Purple Gallinule, and American

PAGE 78

66 Coot, are known to become relatively tame, and even quite bold in developed areas if left unmolested (Terres 1991, Elphick et al. 2001, West & Hess 2002). In some instances this perceived behavior may have actually been an effort to slowly move away from my boat. These cases may have represented a stage in a process of habituation, where the birds were not entirel y tolerant of disturbance, but did not feel threatened enough to rapidly flee the area. Disturbance Sensitivity Seasonal alert/flee response Results of this study revealed that waterbirds on these lakes displayed alert/flee behavior 1.6 times more often in the winter than in the summer. Though several studies have examined waterbird behavior across seasons (Burger 1981, Klein 1993, Klein et al. 1995), to my knowledge, no previous studies have attempted to quantify the degree of variation in sensitivity to disturbance between seasons. The significant increase in alert/flee behavior in marsh birds, diving birds, and ducks supports the idea that avian disturbance sensitivity has a strong seasonal component. In general, there is a significantly greater cost involved in fa iling to protect a nest and young during the breeding season than there is in failing to defend temporary foraging and loafing sites at other times of the year (Rodgers and Smith 1997). Thus, birds may be more likely to temporarily abandon a preferred foraging or resting location in the non-breeding season. However, an increase in alert/flee behavior is not necessarily an indication of increased sensitivity to disturbance. In attempting to raise and ensure the survival of young during the breeding season, birds must meet some of their greatest energy demands of the year. Such physical demands may be mirrored in the amount of stress these birds undergo. If breeding birds are facing at a minimum the same amount of stress that wintering birds are

PAGE 79

67 facing, one could conclude that they would be equally sensitive to disturbance. Since this sensitivity is not indicated through overt responses such as alert/flee behavior, these birds may be enduring high levels of physiologically manifested stress, such as increased heart rate, temperature, and blood sugar that could reduce their overall fitness (Gabrielsen & Smith 1995). The alternative, that birds are more easily stressed in the non-breeding season, suggests that urban lakes may act as an energy drain for birds during this season. Even if waterbirds are normally easily stressed in the non-breeding season, due to forces such as a fluctuating food supply or simple “flightiness,” they are more likely to encounter disturbances in an urban environment that might trigger an alert/flee response. Such a response may incur numerous negative physiological responses, increase energy expenditures, reduce foraging times, and reduce overall fitness (Knight & Cole 1995). Inter-guild alert/flee comparisons Ducks showed significantly greater alert/flee behavior than other guilds in response to the approach of the survey boat. This reinforces the results of other studies that have documented a significant variation between waterbird species in sensitivity to human disturbance (Batten 1977, Burger 1981, Vask e et al. 1983, Klein 1993, Klein et al. 1995, Rodgers & Smith 1997, Rodgers & Schwikert 2002). Though Klein et al. (1995) found migratory dabbling ducks showed the most consistent patterns of avoidance of human visitors, no other studies have found this guild to be uniformly more sensitive to disturbance than other species. Wood Ducks comprised over 80% of all duck observations in this study. Previous studies have shown that this species will readily nest in urban environments (Hepp & Bellrose 1995). The heightened alert/flee behavior observed in these birds might represent an evolving process of habituation, where the

PAGE 80

68 birds originally adopted these human-dominated environments out of necessity, but are not yet entirely tolerant of the conditions. A similar process has been suggested for both the Common Loon ( Gavia immer ) and the Great Crested Grebe (Titus & Van Druff 1981, Keller 1989). Migrant versus resident alert/flee response The proportion of winter migrants that responded to human disturbance with alert/flee behavior was not significantly greater than that of resident species. This fails to confirm the results of van der Zande et al. (1980), Burger (1981), Burger and Gochfeld (1991), and Klein et al. (1995), which consiste ntly showed migrant waterbirds displaying a greater sensitivity to human disturbance than resident birds. One possible explanation is that the winter migrants arriving on urban lakes came from summer breeding grounds that were also disturbed and thus they ha d already developed a tolerance. Previous studies (above) were conducted in more natural habitats. If the winter migrants arriving in those areas came from undisturbed breeding grounds, they may have not been habituated to human disturbance upon arrival. The previous studies also found increased disturbance sensitivity primarily in migrant waterfowl and shorebirds, whereas the majority of migrants in this study consisted of diving birds. Another explanation can be seen in studies that have shown distinct variation in sensitivity to disturbance between individuals of the same species (Klein 1993, Klein et al. 1995). If this were the case, less tolerant individuals would frequent less dist urbed areas. In this study however, only 1% of migrant birds were found along undeveloped shoreline. It may be that only highly tolerant individuals (both migrants and reside nts) are attracted to urban lakes to begin with and that they are fairly habituated to humans.

PAGE 81

69 Management and Future Research Clearly there are a wide range of factor s that are associated with the behavioral patterns of waterbirds in urban environments. Season, development and changes in habitat complexity, disturbance levels, and individual guild and species’ tolerances are just a few examples. Managing for all of these variables is neither practical nor feasible for most local or state agencies. Nor would a litany of restrictions be acceptable to local residents. Though further research is needed on a greater number and variety of lakes before specific management recommendations should be made, several general recommendations are worth considering. Waterbird behavior does appear to be associated with shoreline development, and undeveloped shoreline may serve as an important refuge for birds that are more sensitive to human disturbance or developed habitat. For example, significantly more wintering marsh birds were observed foraging on undeveloped shoreline. Further, some guilds had heightened alert/flee behavior on undeveloped shoreline, such as ducks (summer) and waders (winter), which may indicate that some portion of these populations consists of individuals that have not habituated to humans. The preservation of undeveloped shoreline, preferably in larger contiguous patches, should therefore be considered in future lake development plans. Until more detailed habitat studies are conducted in urban aquatic environments, developers should use the general ecology of the waterbird guilds that they hope to attract as a guide for selecting areas of shoreline to preserve. Examples of foraging habitat include areas of shallow, relatively open littoral zones for wading birds, open water interspersed with mixed emergent and floating-leafed vegetation for ducks and marsh birds, and larger areas of deeper open water for diving birds. Examples of nesting and

PAGE 82

70 loafing habitat include shrubby or wooded shor eline for wading and diving birds, stands of dense tall-emergent vegetation for marsh bi rds, and areas of floating-leafed vegetation for ducks. Likewise, residents wishing to increase waterbird biodiversity on urban lakes have many options available to them. Adding Wood Duck boxes on residential properties may increase nesting by this species along developed shorelines. Providing perching substrates separate from high human-use areas such as docks might reduce disturbance to resting diving birds and decrease unwanted fouling of docks. Ducks and marsh birds may also be encouraged to forage along developed shorelines by planting species of native emergent and floating-leafed vegetation t ypically found in their diet. Lists of such plants are often available through local birding groups, state wildlife agencies, or state extension offices. These resources may also be able provide residents with a better understanding of the general behavior of many of these species and how best to limit negative interactions. There has been considerable research examining the potential use of buffer zones to protect waterbirds from undue disturbance (Rodgers & Smith 1995, 1997, Rodgers & Schwikert 2002). Buffer zones are impractical though along developed shorelines where residents require lake access and desire the freedom to engage in a variety of activities on their own property. If birds do indeed habituate at a local scale along developed shorelines, then buffer zones may only be required along undeveloped portions of shoreline. Given the overall elevated alert/f lee behavior observed in the winter, buffers in the non-breeding season may be warranted. However, further research is needed to examine fluctuations in sensitivity to disturbance throughout the winter season. Further

PAGE 83

71 research is also needed to determine nesting areas around urban lakes and whether breeding season buffers are needed. Future studies should be conducted in years with less extreme water conditions and examine where exactly birds are nesting on these lakes and to what degree their tolerance of human disturbance fluctuates during the breeding season. Further research on foraging and resting behaviors is needed for waterbirds in urban environments. Previous studies have suggested that sensitive species may find a lack of adequate foraging or loafing sites as their preferred habitats become developed and disturbances increase (Skagen et al. 1991, Pfister et al. 1992). This has been corroborated, at least in terms of foraging oppor tunities, by studies that have shown lower fish species richness and abundance along areas of developed lake shoreline (Guillory et al. 1979, Bryan & Scarnecchia 1992). Though I found no difference in the proportion of birds foraging along developed and undeveloped shorelines, other measures, such as foraging times, prey selection, or strike/capture ratios, may better show the value of this habitat for foraging birds, and should be examined in the future. More comprehensive and long-term research is also strongly recommended for urban lake habitats. Much work has been devoted to passerines in terrestrial systems such as parks and the urban rural gradient (e.g. Blair 1996, Clergeau et al. 1998, Fernandez-Juricic 2001). Urban aquatic systems have received far less attention, and studies typically have been of relatively short duration; no more than a season or two. Repeated case studies are needed to determine whether these birds are indeed adapting to human-manipulated environments and whether such environments are acting as functional habitat or merely as sinks. The urban rural gradient also needs to be examined

PAGE 84

72 for lacustrine habitats, comparing waterbird behaviors across varying levels of shoreline development. Additionally, though Hoyer & Canfield (1990, 1994) showed that higher trophic state lakes are positively correlated with waterbird abundance, the popularity of oligotrophic lakes for development (Hoyer, pers. comm.) indicates that future studies should look at urban lakes covering a range of trophic states. Such studies would allow us to better understand the effects of residential development and thus better plan for limiting our impact in future growth. In general, research with potential for direct application (e.g. urban studies), should be organized to be as approachable as possible to planners and managers. Research at the guild level can be much more readily applied to management efforts. While species-level studies are critical in determining birds that appear most vulnerable to development and disturbance, managing on a species by species basis requires tremendous effort and often has conflicting priorities. Research and management that focuses at the guild level, when based on sound science, should yield greater species richness and waterbird abundance with the least amount of management. Many waterbird behavioral studies have examined entire communities at the species level, occasionally grouping birds by size or seasonal distribution. Comparing ecological functiona l groups of any kind should be considered in future studies to allow for the results to be more easily applied to management efforts.

PAGE 85

73 APPENDIX A WINTER HAVEN WATERBIRD SURVEY DATA Table A-1. Aquatic bird species observed on lakes Buckeye (B), Conine (C), Deer (D), and Jessie (J) in Winter Haven, Florida, summer 2001 and winter 2001/2002. Species Scientific Name Res1Lake BuckeyeConine Deer Jessie S W S W S W S W Diving Birds Pied-billed Grebe Podilymbus podiceps W 51988 15 1088 Brown PelicanS Pelicanus occidentalis Y 0010 0 000 D-C Cormorant Phalacrocorax autitus W 333240309 10 412564 Anhinga Anhinga anhinga Y 29685426 77 379968 Osprey Pandion haliaetus Y 7113421 11 33014 Ring-billed Gull Larus delawarensis W 033036 0 67023 Caspian Tern Sterna caspia W 0005 0 201 Forster's Tern Sterna fosteri W 0400 0 000 Least TernT Sterna antillarum S 1000 0 0140 Belted Kingfisher Ceryle alcyon W 04015 0 7210 Wading Birds Least Bittern Ixobrychus exilis S 9013 6 030 Great Blue Heron Ardea herodias Y 13246742 33 75239 Great Egret Ardea alba Y 4112611 22 182114 Snowy EgretS Egretta thula Y 132910 3 0335 Little Blue HeronS Egretta caerulea Y 1442 8 1424 Tricolored HeronS Egretta tricolor Y 782814 19 221511 Cattle Egret Bubulcus ibis Y 1106 0 31535 Green Heron Butorides virescens Y 4811241 21 3162 B-C Night Heron Nycticorax nycticorax Y 0000 0 029 White IbisS Eudocimus albus Y 73511269 60 9123144 Glossy Ibis Plegadis falcinellus S 0 0310 0 000 Wood StorkE Mycteria americana Y 0021 6 210 LimpkinS Aramus guarauna Y 4300 4 010 Sandhill CraneT Grus canadensis Y 00122 0 000 Black-Necked Stilt Himantopus mexicanus S 0020 0 000 Ducks (wild) Wood Duck Aix sponsa Y 11221527 808 334396106 Mallard Anas platyrhnchos Y 0062 0 0500 Blue-winged Teal Anas discors W 0500 0 500 Ring-necked Duck Aythya collaris W 0000 0 35700

PAGE 86

74 Appendix A, continued. Species Scientific Name Res1Lake BuckeyeConine Deer Jessie S W S W S W S W Marsh Birds Rail Rallus sp. Y 0010 0 000 Sora Porzana carolina W 0010 0 000 Purple Gallinule Porphyrula martinica Y 2520 137 9700 Common Moorhen Gallinula chloropus Y 171818373 177 20615394 American Coot Fulica americana W 00026 1 169022 Other Domestic Goose Anser domesticus Y 0000 0 02922 Domestic Duck Anas domesticus Y 5764 0 03117 Bald EagleT Haliaeetus leucocephalus W 0200 0 100 Killdeer Charadrius vociferous Y 00125 0 0211 S State listed as Species of Special Concern, T State listed as Threatened, E State and federally listed as Endangered. 1Residence: S=summer resident, W=winter migrant, Y=yearround resident.

PAGE 87

APPENDIX B WINTER HAVEN WATERBIRD HABITAT DATA

PAGE 88

76 Table B-1. Wading bird habitat associations, summer 2001. Expected values based on availability of habitat elements on each lake. Habitat Lake ObservedExpected 2 p-value+/Low E Buckeye 94.134.87 0.027*+ Conine 4039.220 1 Deer 7359.784.73 0.03*+ Jessie 4864.447.26 0.007*+ Tall E Buckeye 4447.710.53 0.467 Conine 120192.03110.55 <0.0001*Deer 5794.9245.81 <0.0001*Jessie 3265.9931.94 <0.0001*Floating Buckeye 2730.210.38 0.538 Conine n/an/an/a n/a Deer 112116.060.64 0.424 Jessie n/an/an/a n/a Shore Buckeye 2927.140.1 0.752 Conine 18992.60156.66 <0.0001*+ Deer 3819.3219.89 <0.0001*+ Jessie 5546.811.82 0.177 Lawn Buckeye 2833.861.46 0.227 Conine 6937.9029.09 <0.0001*+ Deer 7675.780 1 Jessie 6266.830.53 Understory Buckeye 3546.897.08 0.008*Conine 135153.456.48 0.01*Deer 8757.9125.94 <0.0001*+ Jessie 9165.9417.29 <0.0001*+ Shrub Buckeye 4950.140.02 0.888 Conine 12412.3025.42 <0.0001*+ Deer 6452.483.92 0.048*+ Jessie 6465.990.06 0.806 Canopy Buckeye 5338.809.29 0.002*+ Conine 3722.689.25 0.002*+ Deer 3248.268.25 0.004*Jessie 2443.5712.08 <0.001*+/Significant positive or negative association with habitat element.

PAGE 89

77 Table B-2. Wading bird habitat associations winter 2001/2002. Expected values based on availability of habitat element on each lake. Habitat Lake ObservedExpected 2 p-value+/Low E Buckeye 464.90354.84 <0.0001*+ Conine 8423.87176.28 <0.0001*+ Deer 5255.080.21 0.647 Jessie 9569.9215.91 <0.0001*+ Tall E Buckeye 3956.5412.5 <0.001*Conine 68116.8983.11 <0.0001*Deer 3687.4692.21 <0.0001*Jessie 5371.608.61 0.003*Floating Buckeye 1935.8111.85 <0.001*Conine 11.080 1 Deer 88106.9418.6 <0.0001*Jessie 12.910.69 0.406 Shore Buckeye 92.1119.75 <0.0001*+ Conine 3525.104.21 0.04*+ Deer 11.280 1 Jessie 3914.3844.64 <0.0001*+ Lawn Buckeye 7340.1344.88 <0.0001*+ Conine 7323.25122.84 <0.0001*+ Deer 11675.7851.04 <0.0001*+ Jessie 11572.52246.2 <0.0001*+ Understory Buckeye 6158.460.18 0.671 Conine 93104.103.33 0.068 Deer 8058.3714.06 <0.001*+ Jessie 5872.064.82 0.028*Shrub Buckeye 4759.426.28 0.012*Conine 4999.7972.01 <0.0001*Deer 4952.480.29 0.59 Jessie 5271.609.58 0.002*Canopy Buckeye 7445.9831.61 <0.0001*+ Conine 1813.861.05 0.306 Deer 8248.2636.76 <0.0001*+ Jessie 5847.283.2 0.074 +/Significant positive or negative association with habitat element.

PAGE 90

78 Table B-3. Marsh birds habitat associations, summer 2001. Expected values based on availability of habitat element on each lake. Habitat Lake ObservedExpected 2 p-value+/Low E Buckeye 197.8515.21 <0.0001*+ Conine 109.460 1 Deer 76126.3934.36 <0.0001*Jessie 7164.441.05 0.31 Tall E Buckeye 5690.7131.39 <0.0001*Conine 4546.300.05 0.823 Deer 148200.6942.15 <0.0001*Jessie 3065.9935.88 <0.0001*Floating Buckeye 5157.440.98 0.322 Conine 00.430 1 Deer 279245.3826.14 <0.0001*Jessie 42.680.25 0.617 Shore Buckeye 7951.5921.11 <0.0001*+ Conine 3922.3318.48 <0.0001*+ Deer 3840.600.13 0.718 Jessie 7146.8117.71 <0.0001*+ Lawn Buckeye 10064.3732.94 <0.0001*+ Conine 79.210.38 0.538 Deer 84148.5967.75 <0.0001*Jessie 5266.835.84 0.016*Understory Buckeye 7091.3512.17 <0.001*Conine 3734.480.37 0.543 Deer 108106.700.01 0.92 Jessie 11466.4163.12 <0.0001*+ Shrub Buckeye 5295.3350.5 <0.0001*Conine 4139.530.06 0.806 Deer 183102.91104.33 <0.0001*+ Jessie 5565.993.14 0.076 Canopy Buckeye 6473.772.23 0.135 Conine 95.491.81 0.179 Deer 7595.006.42 0.011*Jessie 3143.574.84 0.028*+/Significant positive or negative association with habitat element.

PAGE 91

79 Table B-4. Marsh bird habitat associations, winter 2001-2002. Expected values based on availability of habitat element on each lake. Habitat Lake ObservedExpected 2 p-value+/Low E Buckeye 384.39263.45 <0.0001*+ Conine 2215.352.92 0.087 Deer 169201.128.67 0.003*Jessie 8750.7346.46 <0.0001*+ Tall E Buckeye 4750.650.48 0.488 Conine 7075.141.19 0.275 Deer 169319.34218.35 <0.0001*Jessie 3551.959.79 0.002*Floating Buckeye 2632.081.55 0.213 Conine 00.690.05 0.823 Deer 406390.463.39 0.066 Jessie 22.110 1 Shore Buckeye 101.8532.39 <0.0001*+ Conine 916.143.26 0.071 Deer 03.192.29 0.13 Jessie 1410.341.07 0.301 Lawn Buckeye 5535.1118.4 <0.0001*+ Conine 2214.953.38 0.066 Deer 234188.8525.88 <0.0001*+ Jessie 8852.1445.59 <0.0001*+ Understory Buckeye 5851.162.02 0.155 Conine 8466.9212.68 <0.001*+ Deer 243145.46118.99 <0.0001*+ Jessie 4651.811.02 0.313 Shrub Buckeye 5752.001.02 0.313 Conine 7864.157.89 0.005*+ Deer 217130.7995.2 <0.0001*+ Jessie 3551.489.32 0.002*Canopy Buckeye 6640.2430.44 <0.0001*+ Conine 88.910.02 0.888 Deer 198120.2679.62 <0.0001*+ Jessie 4833.997.77 0.005*+ +/Significant positive or negative association with habitat element.

PAGE 92

80 Table B-5. Diving bird habitat associations, summer 2001. Expected values based on availability of habitat element on each lake. Habitat Lake ObservedExpected 2 p-value+/Low E Buckeye 71.5317.01 <0.0001*+ Conine 2112.096.92 0.009*+ Deer 3028.610.05 0.823 Jessie 4251.642.98 0.084 Tall E Buckeye 1717.670 1 Conine 4659.2011.31 <0.001*Deer 4345.430.25 0.617 Jessie 2452.8828.63 <0.0001*Floating Buckeye 511.194.61 0.032*Conine n/an/an/a n/a Deer 5355.540.44 0.507 Jessie n/an/an/a n/a Shore Buckeye 910.050.05 0.823 Conine 5028.5524.26 <0.0001*+ Deer 79.380.44 0.507 Jessie 2737.524 0.046*Lawn Buckeye 712.543.48 0.062 Conine 1411.780.29 0.59 Deer 2236.1112.57 <0.001*Jessie 2453.5629.98 <0.0001*Understory Buckeye 1816.440.18 0.671 Conine 5751.381.58 0.209 Deer 3926.909.19 0.002*+ Jessie 8653.2237 <0.0001*+ Shrub Buckeye 2118.570.53 0.467 Conine 4750.540.52 0.471 Deer 4425.8320.49 <0.0001*+ Jessie 7352.8813.68 <0.001*+ Canopy Buckeye 2614.3716.55 <0.0001*+ Conine 197.0220.63 <0.0001*+ Deer 2922.622.45 0.118 Jessie 4234.921.79 0.181 +/Significant positive or negative association with habitat element.

PAGE 93

81 Table B-6. Diving bird habitat associations winter 2001/2002. Expected values based on availability of habitat element on each lake. Habitat Lake ObservedExpected 2 p-value+/Low E Buckeye 234.8567.79 <0.0001*+ Conine 2611.7818.91 <0.0001*+ Deer 3435.870.09 0.764 Jessie 6561.240.32 0.572 Tall E Buckeye 5155.960.86 0.354 Conine 3257.6845.62 <0.0001*Deer 3456.9527.49 <0.0001*Jessie 5862.710.53 0.467 Floating Buckeye 735.4435.12 <0.0001*Conine 40.769.98 0.002*+ Deer 4969.6434.05 <0.0001*Jessie 14.151.75 0.186 Shore Buckeye 02.001.16 0.281 Conine 2112.236.69 0.01*+ Deer 00.750.08 0.777 Jessie 612.503.18 0.075 Lawn Buckeye 3938.040.01 0.92 Conine 3111.3338.24 <0.0001*+ Deer 6044.4012.59 <0.001*+ Jessie 7163.041.68 0.195 Understory Buckeye 6655.424.69 0.03+ Conine 4650.701.08 0.299 Deer 6534.2049.34 <0.0001*+ Jessie 10462.6550.38 <0.0001*+ Shrub Buckeye 6256.331.24 0.265 Conine 2948.6021.33 <0.0001*Deer 5730.7536.54 <0.0001*+ Jessie 9862.2437.54 <0.0001*+ Canopy Buckeye 8343.5966.67 <0.0001*+ Conine 176.7515.47 <0.0001*+ Deer 6928.2891.86 <0.0001*+ Jessie 6441.1017.67 <0.0001*+ +/Significant positive or negative association with habitat element.

PAGE 94

82 Table B-7. Duck habitat associations, summer 2001. Expected values based on availability of habitat element on each lake. Habitat Lake ObservedExpected 2 p-value+/Low E Buckeye 00.510 1 Conine 02.171.53 0.216 Deer 194312.9978.29 <0.0001*Jessie 71176.40114.88 <0.0001*Tall E Buckeye 35.892.36 0.124 Conine 010.6340.04 <0.0001*Deer 235496.97427.24 <0.0001*Jessie 34180.65222.25 <0.0001*Floating Buckeye 93.739.73 0.002*+ Conine n/an/an/a n/a Deer 713607.66105.79 <0.0001*+ Jessie n/an/an/a n/a Shore Buckeye 23.350.33 0.566 Conine 145.1221.59 <0.0001*+ Deer 246100.80241.53 <0.0001*+ Jessie 247128.15162.31 <0.0001*+ Lawn Buckeye 34.180.19 0.663 Conine 122.1149.08 <0.0001*+ Deer 269330.3427.46 <0.0001*Jessie 206182.965.28 0.022*+ Understory Buckeye n/an/an/a n/a Conine 49.468.03 0.005*Deer 391245.78156.63 <0.0001*+ Jessie 297181.81136.78 <0.0001*+ Shrub Buckeye 66.190 1 Conine 10.6917.95 <0.0001*+ Deer 325228.7812.17 <0.001*+ Jessie 104180.6560.34 <0.0001*Canopy Buckeye 55.750.02 0.888 Conine 01.260.51 0.475 Deer 131210.3747.46 <0.0001*Jessie 40119.2775.29 <0.0001*+/Significant positive or negative association with habitat element.

PAGE 95

83 Table B-8. Duck habitat associations, winter 2001/2002. Expected values based on availability of habitat element on each lake. Habitat Lake ObservedExpected 2 p-value+/Low E Buckeye 01.380.59 0.442 Conine 24.190.8 0.371 Deer 62277.12290.09 <0.0001*Jessie 1617.820.18 0.671 Tall E Buckeye 215.9027.49 <0.0001*Conine 2520.493.25 0.071 Deer 133440.02668.13 <0.0001*Jessie 2318.251.86 0.173 Floating Buckeye 2010.0714.08 <0.001*+ Conine 20.199.15 0.002*+ Deer 580538.0218.7 <0.0001*+ Jessie 50.7419.44 <0.0001*+ Shore Buckeye 70.3796.21 <0.0001*+ Conine 04.404.13 0.042*Deer 02.201.32 0.251 Jessie 113.6714.06 <0.001*+ Lawn Buckeye 77.520 1 Conine 254.08120.51 <0.0001*+ Deer 148130.245.61 0.018*+ Jessie 1718.490.1 0.752 Understory Buckeye 1110.960 1 Conine 2718.2511.51 <0.001*+ Deer 182100.32120.76 <0.0001*Jessie 2418.372.71 0.1 Shrub Buckeye 1111.140 1 Conine 217.5013.16 <0.001*Deer 15390.2072.93 <0.0001*+ Jessie 2218.251.09 0.296 Canopy Buckeye 188.6217.54 <0.0001*+ Conine 22.430 1 Deer 17982.94176.73 <0.0001*+ Jessie 1112.050.04 0.841 +/Significant positive or negative association with habitat element.

PAGE 96

APPENDIX C WINTER HAVEN WATERBIRD BEHAVIOR DATA

PAGE 97

85Table C-1. Data for summer guild behavior, listed by guild and lake. Results from contingency Chi-square based on the proporti on of birds engaged in each focal behavior versus the proportion e ngaged in all other (Other) behaviors. Foraging Resting w/Young Alert/Fleeing D/U Forage Other 2 Pvalue Rest Other 2 Pvalue w/ Young Other 2 Pvalue Alert/ Flee Other 2 Pvalue Waders D 15 37 16 36 6 46 12 40 Buckeye U 12 17 1.32 0.251 12 17 0.93 0.336 0 29 3.61 0.083F 4 25 1.01 0.314 D 151 61 37 175 n/a n/a n/a 17 195 Conine U 29 39 18.32 <0.0001* 19 49 3.54 0.06 n/a n/a n/a n/a 11 57 3.81 0.051 D 73 68 38 103 n/a n/a n/a 9 132 Deer U 5 12 3.04 0.081 5 12 0.04 0.78F n/a n/a n/a n/a 3 14 2.74 0.123F D 56 51 35 72 n/a n/a n/a 8 99 Jessie U 14 22 1.95 0.163 10 26 0.3 0.581 n/a n/a n/a n/a 4 32 0.46 0.497F Marsh D 54 83 4 133 45 92 6 131 Buckeye U 11 10 1.26 0.261 1 20 0.20 0.653 0 21 9.65 0.002* 7 14 28.21 <0.001F* D 24 24 9 39 n/a n/a n/a 3 45 Conine U 17 11 0.82 0.366 6 22 0.08 0.777 n/a n/a n/a n/a 3 25 0.49 0.664F D 146 137 12 271 64 219 5 278 Deer U 8 14 1.89 0.169 1 21 0.01 0.999F 3 19 0.96 0.429F 5 17 28.28 <0.001F* D 42 71 5 108 38 75 2 111 Jessie U 10 11 0.82 0.367 1 20 0.01 1F 8 13 0.16 0.692 0 20 0.36 0.999F

PAGE 98

86Summer guild behavior, continued. Foraging Resting w/Young Alert/Fleeing D/U Forage Other 2 Pvalue Rest Other 2 Pvalue w/ Young Other 2 Pvalue Alert/ Flee Other 2 Pvalue Divers D n/a n/a n/a 12 9 n/a n/a n/a 8 13 Buckeye U n/a n/a n/a n/a 6 0 3.86 0.136F n/a n/a n/a n/a 0 6 3.25 0.136F D 2 48 39 11 n/a n/a n/a 6 44 Conine U 1 37 0.12 1F 34 4 2.1 0.156 n/a n/a n/a n/a 1 37 2.59 0.135F D n/a n/a n/a 57 4 n/a n/a n/a 2 59 Deer U n/a n/a n/a n/a 10 1 0.09 0.575F n/a n/a n/a n/a 0 11 0.37 1F D n/a n/a n/a 47 10 n/a n/a n/a 8 49 Jessie U n/a n/a n/a n/a 43 12 0.32 0.569 n/a n/a n/a n/a 9 46 0.12 0.731 Ducks D n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a Buckeye U n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a D n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a Conine U n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a D 23 702 315 410 70 655 181 544 Deer U 0 50 1.64 0.39F 2 48 30.11 <0.0001* 0 50 5.31 0.018F* 30 20 28.98 <0.0001* D 18 334 125 227 41 311 125 227 Jessie U 0 32 1.72 0.383F 2 30 11.35 <0.001* 0 32 4.17 0.036F* 28 4 33.08 <0.0001* D=Developed shoreline, U=Undeveloped shoreline FP-value based on Fisher’s exact test.

PAGE 99

87Table C-2 Data for winter guild behavior, listed by guild and lake. Results from contingency Chi-square based on the proportio n of birds engaged in each focal behavior versus the proportion e ngaged in all other (Other) behaviors. Foraging Resting w/Young Alert/Fleeing D/U Forage Other 2 Pvalue Rest Other 2 Pvalue w/ Young Other 2 Pvalue Alert/ Flee Other 2 Pvalue Waders D 43 28 17 54 6 65 4 67 Buckeye U 1 12 12.31 <0.001* 6 7 2.73 0.172F 5 8 8.7 0.011F* 1 12 0.08 1F D 35 98 82 51 12 121 1 132 Conine U 2 12 0.97 0.519F 5 9 3.53 0.06 5 9 8.82 0.012F* 1 13 3.86 0.182F D 36 91 84 43 5 122 2 125 Deer U 0 1 0.39 0.999F 1 0 0.51 0.999F 0 1 0.04 1F 0 1 0.02 0.999F D 66 86 72 80 11 141 1 151 Jessie U 5 21 5.42 0.02* 13 13 0.06 0.803 8 18 12.9 0.002F* 0 26 0.17 1F Marsh D 14 43 5 52 14 43 22 35 Buckeye U 8 11 2.13 0.144 2 17 0.05 1F 6 13 0.36 0.547 3 16 3.36 0.067 D 26 52 7 71 10 68 32 46 Conine U 10 10 1.9 0.168 1 19 0.335 0.999F 5 15 1.82 0.293F 4 16 3.03 0.082 D 220 222 21 421 86 356 100 342 Deer U 11 13 0.14 0.707 1 23 1.04 0.999F 10 14 6.87 0.017F* 0 24 6.91 0.009* D 49 50 3 96 8 91 34 65 Jessie U 9 2 4.15 0.042* 1 10 1.03 0.348F 1 10 0.01 1F 0 11 5.47 0.017F*

PAGE 100

88 Winter guild behavior, continued. Foraging Resting w/Young Alert/Fleeing D/U Forage Other 2 Pvalue Rest Other 2 Pvalue w/ Young Other 2 Pvalue Alert/ Flee Other 2 Pvalue Divers D n/a n/a 39 13 9 43 4 48 Buckeye U n/a n/a n/a n/a 23 9 0.1 0.752 7 25 0.270.605 1 31 0.74 0.645F D 2 56 41 17 11 47 3 55 Conine U 0 13 0.46 0.999F 9 4 0.01 1F 3 10 0.110.711F 1 12 0.13 0.563F D n/a n/a 23 33 24 32 9 47 Deer U n/a n/a n/a n/a 1 2 0.07 1F 2 1 0.660.578F 0 3 0.57 0.999F D 3 96 76 23 15 84 4 95 Jessie U 0 51 1.89 0.287F 37 14 0..32 0.57 14 37 3.270.07 0 51 2.12 0.3F Ducks D n/a n/a n/a n/a 5 0 0 5 Buckeye U n/a n/a n/a n/a n/a n/a n/a n/a 5 4 3.110.221F 4 5 3.11 0.221F D n/a n/a n/a n/a 0 25 25 0 Conine U n/a n/a n/a n/a n/a n/a n/a n/a 2 0 27 0.003F* 0 2 27 0.003F* D 13 616 53 576 412 217 41 588 Deer U 0 19 0.4 1F 0 19 1.74 0.391F 13 6 0.070.791 0 19 1.32 0.625F D n/a n/a 9 79 79 9 n/a n/a Jessie U n/a n/a n/a n/a 0 10 1.13 0.592F 10 0 1.130.592F n/a n/a n/a n/a FP-value based on Fisher’s exact test.

PAGE 101

89 REFERENCES Adams, L.W. 1994. Urban wildlife habitats: a landscape perspective. University of Minnesota Press, Minneapolis. Andersson, A. P., P. Lindberg, S. G. Nil sson, and A. Petersson. 1980. Breeding success of the black-throated diver Gavia arctica in Swedish lakes. Var Fagelvarld 39: 85-94. Bancroft, G. T., D. E. Gawlick, and K. Rutchey. 2002. Distribution of wading birds relative to vegetation and water depths in the northern Everglades of Florida, USA. Waterbirds 25(3): 265-277. Barnett, B., and R. Schneider. 1974. Fish populations in dense submerged plant communities. Hyacinth Control Journal 12: 12-14. Batten, J. A. 1977. Sailing on reservoirs and its effect on water birds. Biological Conservation 11: 49-58. Belanger, L., and J. Bedard. 1990. Energetic cost of man-induced disturbance to staging snow geese. Journal of Wildlife Management 54(1): 36-41. Bildstein, K. L., D. E. Gawlik, D. P. Ferral, I. L. Brisbin, Jr., and G. R. Wein. 1994. Wading bird use of established and newly created reactor cooling reservoirs at the Savannah River Site, near Aiken, South Carolina, USA. Hydrobiologia 279/280: 71-82. Blair, R.B. 1996. Land use and avian species diversity along an urban gradient. Ecological Applications 6(2): 506-519. Boyle, S. A., and F. B. Samson. 1985. Effects of nonconsumptive recreation on wildlife: a review. Wildlife Society Bulletin 13: 110-116. Bratton, S. P. 1990. Boat disturbance of Cic oniiformes in Georgia estuaries. Colonial Waterbirds 13(2): 124-128. Breeden, S., and B. Breeden. 1982. The drought of 1979-1980 at Keoladeo Ghana Sanctuary, Bharatpur, Rajasthan. Journal of the Bombay Natural History Society 79: 1-37. Breininger, D. R., and R. B. Smith. 1990. Waterbird use of coastal impoundments and management implications in east-central Florida. Wetlands 10(2): 223-241.

PAGE 102

90 Brower, J. E., J. H. Zar, and C. N. von Ende. 1990. Field and laboratory methods for general ecology. Wm. C. Brown, Dubuque, Iowa. Brown, M., and J. J. Dinsmore. 1986. Implications of marsh size and isolation for marsh bird management. Journal of Wildlife Management 50: 392-397. Bryan, M. D., and D. L. Scarnecchia. 1992. Species richness, composition, and abundance of fish larvae and juveniles inhabiting natural and developed shorelines of a glacial Iowa lake. Environmental Biology of Fishes 35: 329-341. Bundy, C. 1979. Breeding and feeding observati ons on the Black-throated diver. Bird Study 26: 33-36. Burger, J. 1981. The effect of human activity on birds at a coastal bay. Biological Conservation 21: 231-241. Burger, J., and J. Galli. 1987. Factors affec ting distribution of gulls (Larus spp.) on two New Jersey coastal bays. Environmental Conservation 14: 59-65. Burger, J., and M. Gochfeld. 1991. Human activity influence and diurnal and nocturnal foraging of sanderlings ( Calidris alba ). Condor 93: 259-265. Butler, R. W. 1992. Great Blue Heron. The birds of North America: life histories for the 21st century, No. 25. American Orn ithologists Union, Washington, D.C. Cairns, J. 1988. Rehabilitating damaged ecosystems. CRC Press, Boca Raton, Florida. Carney, K. M., and W. J. Sydeman. 1999. A review of human disturbance effects on nesting colonial waterbirds. Waterbirds 22(1): 68-79. Clergeau, P., J.-P.L. Savard, G. Mennechez, and G. Falardeau. 1998. Bird abundance and diversity along an urban-rural gradient: a comparative study between two cities on different continents. Condor 100: 413-425. Cole, D. N., and R. L. Knight. 1990. Impacts of recreation on biodiversity in wilderness. Pp. 33-40 in Proceedings of the Symposium on Wilderness Areas: their impact. Utah State University, Logan. Collopy, M. W., and H. L. Jelks. 1989. Dist ribution of foraging wading birds in relation to the physical and biological characteristics of freshwater wetlands in southwest Florida. Florida Game and Fresh Water Fish Commission. Nongame Wildlife Program Final Report, Tallahassee. Custer, T. W., R. K. Hines, and C. M. Cu ster. 1996. Nest initiation and clutch size of Great Blue Heron on the Mississippi Rive r in relation to the 1993 flood. Condor 98: 181-188.

PAGE 103

91 Czech, B., and P. R. Krausman. 1997. Distribution and causation of species endangerment in the United States. Science 277: 1116-1117. Dahl, T. E. 1990. Wetlands losses in the United States, 1780’s to 1980’s. U.S. Department of the Interior, Fish and Wildlife Service, Washington, D.C. Davis, W. E., Jr. 1993. Black-crowned Night Heron. The birds of North America: life histories for the 21st century, No. 74. American Ornithologists Union, Washington, D.C. DeGraaf, R.M. 1986. Urban wildlife habitat research – application to landscape design. In: L.W. Adams and D.L. Leedy, eds. Integrating man and nature in the metropolitan environment. Proceedings of the National Symposium on Urban Wildlife. National Institute for Urban Wildlife, Columbia, MD. DeGraaf, R.M., and J.M. Wentworth. 1981. Urban bird communities and habitats in New England. Transactions of the North American Wildlife Conference 46: 396413. Duda, M. D. 1987. Floridians and wildlif e: sociological implications for wildlife conservation in Florida. Nongame Wildlife Program Technical Report, No. 2. Florida Game and Fresh Water Fish Commission, Tallahassee, Florida. Edelson, N. A. 1990. Foraging ecology of wading birds using an altered landscape in Central Florida. Master’s thesis, University of Florida, Gainesville. Edmiston, H. L., and V. B. Myers. 1983. Florida lakes: a description of lakes, their processes and means of protection. De partment of Environmental Regulation, Tallahassee, Florida. Edmiston, H. L. and V. B. Myers. 1991. Florida lakes: a description of lakes, their processes, and means of protection. Water Quality Management and Restoration, Department of Environmental Regulation, Tallahassee, Florida. Eibl-Eibesfeldt, I. 1970. Ethology: the bi ology of behavior. Holt, Rinehart, and Winston, New York. Ellison, L. N., and L. Cleary. 1978. Effects of human disturbance on breeding of double-crested cormorants. Auk 95: 510-517. Elphick, C., J. B. Dunning, Jr., D. A. Sible y. 2001. The Sibley guide to bird life and behavior. Alfred A. Knopf, New York. Fisher, R. A. 1958. Statistical methods for research workers. 13th edition. Hafner Publishing Company, Inc., New York. Florida DEP. 1983 – 1992. Annual aquatic plant surveys. Florida Department of Environmental Protection, Tallahassee, Florida.

PAGE 104

92 Florida DOT. 1999. Florida land use cover and forms classification system: handbook. Surveying and mapping office, Thematic mapping section, Tallahassee, Florida. Florida LAKEWATCH. 2000. Florida LAKEWATCH data 1986-1999. Department of Fisheries and Aquatic Sciences. University of Florida/Institute of Food of and Agricultural Sciences. Gainesville, Florida. Fernandez-Juricic, E., M. D. Jimenez, and E. Lucas. 2001. Bird tolerance to human disturbance in urban parks of Madrid (Spain): management implications. In: Marzluff, J. M., R. Bowman, and R. Donnelly, eds. Avian ecology and conservation in an urbanizing world. Kluwer Academic Publishers, Boston. Frayer, W. E., and J. M. Hefner. 1991. Florida wetlands: status and trends, 1970’s to 1980’s. U.S. Department of the Interior, U.S. Fish and Wildlife Service, Southeast Region, Atlanta, Georgia. Frederick, P. C. and D. Siegel-Causey. 2000. Anhinga. The birds of North America: life histories for the 21st century, No. 522. American Ornithologists Union, Washington, D.C. Gabrielsen, G. W., and E. N. Smith. 1995. Physiological responses of wildlife to disturbance. In: R. L. Knight and K. J. Gutzwiller, eds. Wildlife and recreationists: coexistence through management and research. Island Press, Washington, D.C. Galeotti, P., F. Morimando, and C. Violani. 1991. Feeding ecology of the Tawny owls (Strix-aluco) in urban habitats (Norther n Italy). Bolletino Di Zoologia 58(2): 143150. Gehlbach, F. R. 1994. The eastern screech owl: life history, ecology, and behavior in the suburbs and countryside. Texas A & M University Press, College Station. Gibbs, J. P., F. A. Reid, and S. M. Melvin. 1992. Least Bittern. The birds of North America: life histories for the 21st century, No. 17. American Ornithologists Union, Washington, D.C. Gilbert, K. M. 1987. The role and management of cattails in a lake ecosystem. Aquatics 9(1): 9-11. Guillory, V., M. D. Jones, and M. Rebel. 1979. A comparison of fish communities in vegetated and beach habitats. Florida Scientist 42(3): 113-122. Hancock, J., and J. A. Kushlan. 1984. The herons handbook. Harper and Row, New York. Hatch, J. J., and D. V. Weseloh. 1999. Double-crested cormorant. The birds of North America: life histories for the 21st century, No. 441. American Ornithologists Union, Washington, D.C.

PAGE 105

93 Heimberger, M., D. Euler, and J. Barr. 1983. The impact of cottage development on common loon Gavia immer reproductive success in central Ontario, Canada. Wilson Bulletin 95: 431-439. Hepp, G. R., and F. C. Bellrose. 1995. Wood Duck. The birds of North America: life histories for the 21st century, No. 169. American Ornithologists Union, Washington, D.C. Hinde, R. A. 1970. Animal behavior: a synthesis of ethology and comparative psychology. McGraw-Hill Book Co., New York. Hockin, D., M. Ounsted, M. Gorman, D. Hill, V. Keller, and M. A. Barker. 1992. Examination of the effects of disturbance on birds with reference to its importance in ecological assessments. Journal of Environmental Management 36: 253-286. Hoppe, R. T., and R. A. Kennamer. 1986. Little Blue Herons nesting among cattails. Wilson Bulletin 98(3): 482-483. Hostetler, M. E. 1999. Scale, birds, and human decisions: a potential for integrative research in urban ecosystems. Landscape and Urban Planning 45: 15-19. Hostetler, M. E., and C.S. Holling. 2000. Detecting the scales at which birds respond to structure in urban landscapes. Urban Ecosystems 4: 25-54. Hoyer, M. V., and D. E. Canfield, Jr. 1990. Limnological factors influencing bird abundance and species richness on Florida lakes. Lake and Reservoir Management 6(2): 133-141. Hoyer, M. V., and D. E. Canfield, Jr. 1994. Bird abundance and species richness on Florida lakes: influence of trophic status, lake morphology, and aquatic macrophytes. Hydrobiologia 297/280: 107-119. Hoyer, M. V., J. Winn, and D. E. Canfield, Jr. 2001. Citizen monitoring of aquatic bird populations using a Florida lake. Lake and Reservoir Management 17: 1-8. Jenni, D. A. 1969. A study of the ecology of f our species of herons during the breeding season at Lake Alice, Alachua County, Florida. Ecological Monographs 39: 245270. Johnson, F. A., and F. Montalbano. 1984. Se lection of plant communities by wintering waterfowl on Lake Okeechobeee, Florida. Journal of Wildlife Management 48: 174-178. Jokimaki, J. 1999. Occurrence of breeding bird species in urban parks: Effects of park structure and broad-scale variables. Urban Ecosystems 3: 21-34. Jozkowicz, A. Gorska-Klek, L. 1996. Activity patterns of the Mute Swans Cygnus olor wintering in rural and urban areas: a comparison. Acta Ornithologica 31: 45-51.

PAGE 106

94 Kahl, R. 1991. Boating disturbance of ca nvasbacks during migration at Lake Poygan, Wisconsin. Wildlife Society Bulletin 19: 242-248. Kaiser, M. S., and E. K. Fritzell. 1984. Effects of river recreations on green-backed heron behavior. Journal of Wildlife Management 48: 561-567. Kaminski, R. M., and H. H. Prince. 1984. Dabbling duck-habitat associations during spring in Delta Marsh, Manitoba. Journal of Wildlife Management 10: 37-50. Karr, J. R. and R. R. Roth. 1971. Vegetation structure and avian diversity in several new world areas. American Naturalist 105(945): 423-435. Keller, V. 1989. Variations in the response of Great Crested Grebes Podiceps cristatus to human disturbance – a sign of adapta tion? Biological Conservation 49: 31-45. Kenney, S. A., and R. L. Knight. 1992. Flight distances of black-billed magpies in different regimes of human density and persecution. Condor 94: 545-547. Klein, M. L. 1993. Waterbird behavioral responses to human disturbances. Wildlife Society Bulletin 21: 31-39. Klein, M. L., S. R. Humphrey, and H. F. Percival. 1995. Effects of ecotourism on distribution of waterbirds in a wildlif e refuge. Conservation Biology 9(6): 14541465. Knight, R. L., and D. N. Cole. 1991. Effects of recreational activity on wildlife in wildlands. Transactions of the North American Wildlife and Natural Resources Conference 56: 238-247. Knight, R. L., and D. N. Cole. 1995. Wildlife responses to recreationists. In: R. L. Knight and K. J. Gutzwiller, eds. Wildlife and recreationists: coexistence through management and research. Island Press, Washington, D.C. Knight, R. L., and K. J. Gutzwiller, eds. 1995. Wildlife and recreationists: coexistence through management and research. Island Press, Washington, D.C. Knight, R. L., and S. A. Temple. 1986. Nest defense in the American goldfinch. Animal Behavior 34: 887-897. Knight, R. L., and S. A. Temple. 1995. Origin of wildlife responses to recreationists. In: R. L. Knight and K. J. Gutzwiller, eds. Wildlife and recreationists: coexistence through management and research. Island Press, Washington, D.C. Korschgen, C. E., L. S. George, and W. L. Green. 1985. Disturbance of diving ducks by boaters on a migrational staging area. Wildlife Society Bulletin 13: 290-296. Krebs, C. J. 1998. Ecological methodology. Addison Wesley Longman, Inc., Menlo Park, California.

PAGE 107

95 Kuhl, J. 1979. Zum flucht-und ampassungsverhalten der Grauganse ( Anser anser ) nach untersuchungen an schleswig-holsteinisc hen gewassern. Vogelwelt 100: 217-225. Kury, C. R., and M. Gochfeld. 1975. Human interference and gull predation in cormorant colonies. Biological Conservation 8: 23-34. Lancaster, R. K., and W. E. Rees. 1979. Bird communities and the structure of urban habitats. Canadian Journal of Zoology 57: 2358-2368. Lehtonen, L. 1970. Zur Biologie des Prachttauchers, Gavia a. arctica. Annales Zoologica Fennici 7: 25-60 Lord, A., J. R. Waas, J. Innes, and M. J. Whittingham. 2001. Effects of human approaches to nests of northern New Z ealand dotterels. Biological Conservation 98: 233-240. MacArthur, R. H. and J. W. MacArthur. 1961. On bird species-diversity. Ecology 42(3): 594-598. MacArthur, R. H. and E. O. Wilson. 1967. The theory of island biogeography. Princeton University Press, Princeton, New Jersey. Marzluff, J. M. 2001. Worldwide urbanization and its effects on birds. In: J. M. Marzluff and R. Sallabanks, eds. Avian conservation: research and management. Island Press, Washington, D.C. Marzluff, J. M., R. Bowman, and R. Donnell y. 2001. A historical perspective on urban bird research: trends, terms, and approaches. In: J. M. Marzluff, R. Bowman, and R. Donnelly, eds. Avian ecology and conservation in an urbanizing world. Kluwer Academic Publishers, Boston. Marzluff, J. M., F. R. Gelbach, and D. A. Manuwal. 1998. Urban environments: influences on avifauna and challenges for the avian conservationist. In: J. M. Marzluff and R. Sallabanks, eds. Avian conservation: research and management. Island Press, Washington, D.C. Marzluff, J. M. and R. Sallabanks. 1998. Past approaches and future directions for avian conservation biology. In: Marzluff, J. M. and R. Sallabanks, eds., Avian conservation: research and management. Island Press, Washington, D.C. Melvin, S. M. and J. P. Gibbs. 1996. Sora. The birds of North America: life histories for the 21st century, No. 250. American Orn ithologists Union, Washington, D.C. Miller, R. F. 1943. The Great Blue Heron. The breeding birds of the Philadelphia region, Part II. Cassina 33: 1-23. Mitsch, W. J., and J. G. Gosselink. 1993. Wetlands, 2nd edition. Van Nostrand Reinhold, New York.

PAGE 108

96 Morris, A., and J. P. Morris. 1995. The Florida handbook: 1995-1996. The Peninsular Publishing Co., Tallahassee, Florida. Niemela, J. 1999. Ecology and urban pla nning. Biodiversity and Conservation 8: 119131. Nisbet, I. C. T. 2000. Disturbance, habitua tion, and management of waterbird colonies. Waterbirds 23(2): 312-332. Oldfield, M. L. 1988. Threatened mammals affected by human exploitation of the female-offspring bond. Conservation Biology 2: 260-274. Owens, N. W. 1977. Responses of wintering brent geese to human disturbance. Wildfowl 28: 5-14. Parris, R. W., and G. A. Grau. 1978. Feeding sites of Great Blue Herons in southwestern Lake Erie. Colonial Waterbirds 2: 110-113. Pfister, C., B. A. Harrington, and M. Lavine. 1992. The impact of human disturbance on shorebirds at a migration staging ar ea. Biological Conservation 60: 115-126. Picman, J., M. L. Milks, and M. Leptich. 1993. Patterns of predation on passerine nests in marshes; effects of water depth and distance from edge. The Auk 110(1): 8994. Polk County. 1999. Annual lake and stream report. Polk County Board of Commissioners. Environmental Services Department, Natural Resources Division. Poole, A. 1981. The effects of human disturbance on osprey reproductive success. Colonial Waterbirds 4: 20-27. Pyrovetsi, M., and A Crivelli. 1988. Habitat use by waterbirds in Prespa National Park, Greece. Biological Conservation 45: 135-153. Robertson, R. J., and N. J. Flood. 1980. Effects of recreational use of shorelines on breeding bird populations. Canadian Field Naturalist 94: 131-138. Rodgers, J. A., Jr., and S. T. Schwikert. 2002. Buffer-zone distances to protect foraging and loafing waterbirds from disturbance by personal watercraft and outboardpowered boats. Conservation Biology 16(1): 216-224. Rodgers, J. A., Jr., and H. T. Smith. 1995. Se t-back distances to protect nesting bird colonies from human disturbance in Florida. Conservation Biology 9: 89-99. Rodgers, J. A., Jr., and H. T. Smith. 1997. Bu ffer zone distances to protect foraging and loafing waterbirds from human disturbance in Florida. Wildlife Society Bulletin 25(1): 139-145.

PAGE 109

97 Roth, R. R. 1976. Spatial heterogeneity and bird species-diversity. Ecology 57(4): 773782. Rottenborn, S. C. 1999. Predicting the impacts of urbanization on riparian bird communities. Biological Conservation 88: 289-299. Safran, R. J., M. A. Colwell, C. R. Isola, and O. E. Taft. 2000. Foraging site selection by nonbreeding White-faced ibis. The Condor 102: 211-215. Savard, J.-P. L., and B.J. Falls. 1982. Influence of habitat structure on the nesting height of birds in urban areas. Canadian Journal of Zoology 59: 924-932. Savard, J.-P. L., P. Clergeau, and G. Mennechez. 2000. Biodiversity concepts and urban ecosystems. Landscape and Urban Planning 48: 131-142. Skagen, S. K., R. L. Knight, and G. H. Orians. 1991. Human disturbance of an avian scavenging guild. Ecological Applications 1: 215-225. Smith, J. P., J. R. Richardson, and M. W. Collopy. 1995. Foraging habitat selection among wading birds (Ciconiiformes) at Lake Okeechobee, Florida, in relation to hydrology and vegetative cover. Pp. 247-285 in N. G. Aumen and R. G. Wetzel, eds. Ecological studies of Lake Okeechob ee, Florida. Archive fur Hydrobiologie Special Issues Advances in Limnology, 45. Soul, M. E., D. T. Bolger, A. C. Alberts, R. Sauvajot, J. Wright. M. Sorice, and S. Hill. 1988. Reconstructed dynamics of rapid extinction of chaparral requiring birds in urban habitat islands. Conservation Biology 2: 75-92. Strohm, B. 1974. For our quality of life: a steady downhill slide. National Wildlife 12: 8-9. Surdick, J. A., Jr. 1998. Biotic and abiotic indicators of foraging site selection and foraging success of four ciconiiform speci es in the freshwater Everglades of Florida. Master’s thesis, University of Florida, Gainesville. Swenson, J. E. 1979. Factors affecting status and reproduction of ospreys in Yellowstone National Park. Journal of Wildlife Management 43: 595-601. Tarver, D. P., J. A. Rodgers, M. J. Mahler, and R. L. Lazor. 1978. Aquatic and wetland plants of Florida. Florida Department of Natural Resources, Tallahassee, Florida. Terres, J. K. 1991. The Audubon Society encyclopedia of North American birds. Wings Books, New York. Tilghman, N. G. 1987. Characteristics of urban woodlands affecting breeding bird diversity and abundance. Landscape and Urban Planning 14: 481-495. Tindle. 1979. Tourists and seabirds in the Galapagos. Oryx 15: 68-70.

PAGE 110

98 Titus, J. R., and L. W. van Druff. 1981. Response of the common loon to recreational pressure in the Boundary Waters Canoe Area, northeastern Minnesota. Wildlife Monographs 79: 3-59. Tremblay, J., and L. N. Ellison. 1979. Effects of human disturbance on breeding of black-crowned night herons. Auk 96: 364-369. United Nations. 1987. Global biodiversity assessment. Cambridge University Press, Cambridge. United Nations. 1996. World urbanization pr ospects: the 1996 revision. United Nations, New York. van der Zande, A. N., W. J. ter Keurs, and W. J. van der Weijden. 1980. The impact of roads on the densities of four bird species in an open field habitat: evidence of a long-distance effect. Biol ogical Conservation 18: 299-321. Vaske, J. J., A. R. Graefe, and F. R. Kuss. 1983. Recreation impacts: a synthesis of ecological and social research. Pp. 96-107 in Transactions of the 48th Conference on North American Wildlife and Natural Resources. Vos, D. K., R. A. Ryder, and W. D. Graul. 1985. Response of breeding great blue herons to human disturbance in northcentral Colorado. Colonial Waterbirds 8: 13-22. Wegener, W., D. Holcomb, and V. Williams. 1973. Sampling shallow water fish populations using the Wegener ring. Proceedings of the Southeastern Association of the Game and Fish Commission 27: 663-674. Weller, M. W. 1999. Wetland birds: habitat resources and conservation implications. Cambridge University Press, Cambridge. Weller, M. W., and L. H. Fredrickson. 1974. Avian ecology of a managed glacial marsh. Living Bird 12: 269-291. Weller, M. W., and S. Spatcher. 1965. Role of habitat in the distribution and abundance of marsh birds. Iowa State University Special Report 43, Ames. West, R. L. and G. K. Hess. 2002. Purple Gallinule. The birds of North America: life histories for the 21st century, No. 626. American Ornithologists Union, Washington, D.C. Whitfield, A. K. and D. P. Cyrus. 1978. Feeding succession and zonation of aquatic birds at False Bay, Lake St. Lucia. Ostrich 49: 8-15. Zaffke, M. 1984. Wading bird utilization of Lake Okeechobee marshes 1977-1981. South Florida Water Management District Technical Publication 84-9, West Palm Beach.

PAGE 111

99 BIOGRAPHICAL SKETCH Ashley Traut was born in Sacramento, California, on January 31, 1972. His tumultuous arrival, after going some time without oxygen in the womb, made for lively lectures in his father’s biochemistry class on basic metabolic pathways for many years to come. Mr. Traut attended grade school in Pleasantville, New York, and Bethesda, Maryland, and middle and high school in Baltimore, Maryland. He journeyed northward for his undergraduate education, receiving a Bachelor of Science degree in terrestrial ecology from the University of Vermont in 1994. Before tackling the formidable challenge of graduate school, Mr. Traut trundled from Maryland to Hawaii to Texas in search of gainful employment in wildlife-rela ted pursuits. His list of previously held positions includes environmental educator, wildlife rehabilitator, veterinary technician, wildlife biologist, and yes, bicycle mechanic. He looks forward to this list growing considerably over the years as he continues to explore, learn, and apply himself.


Permanent Link: http://ufdc.ufl.edu/UFE0000807/00001

Material Information

Title: Urban lakes and waterbirds : effects of development on distribution and behavior
Physical Description: xii, 99 p.
Language: English
Creator: Traut, Ashley H. ( Dissertant )
Hostetler, Mark E. ( Thesis advisor )
Frederick, Peter ( Reviewer )
Tanner, George ( Reviewer )
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2003
Copyright Date: 2003

Subjects

Subjects / Keywords: Wildlife Ecology and Conservation thesis,M.S   ( local )
Dissertations, Academic -- UF -- Wildlife Ecology and Conservation   ( local )

Notes

Abstract: Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Science URBAN LAKES AND WATERBIRDS: EFFECTS OF DEVELOPMENT ON DISTRIBUTION AND BEHAVIOR By Ashley H. Traut May 2003 Chair: Mark E. Hostetler Department: Wildlife Ecology and Conservation I studied waterbird distribution and behavior in the breeding and non-breeding seasons on four partially developed urban lakes in central Florida. I examined waterbird distributional and behavioral associations with developed and undeveloped shorelines, as well as distributional associations with specific elements of the littoral and onshore habitat in the urban environment. By understanding how waterbirds respond to development, and by identifying habitat elements that influence their movements and behaviors, we can provide more ecologically sound development and management practices for urban aquatic environments. A total of 38 waterbird species w ^^ ere observed on the four lakes over both seasons. Wading bird, marsh bird, and duck abundance was significantly greater along developed shoreline in both seasons on all lakes. Diving bird abundance was significantly greater along developed shoreline in the winter. Species richness was not associated with shoreline development. Species evenness was greater along undeveloped shoreline in the summer and developed shoreline in the winter. Tall emergent vegetation, open shore, lawn, and canopy appeared to be the primary habitat elements determining waterbird presence. All waterbirds were negatively associated with tall emergent vegetation on two or more lakes over both seasons, whereas wading birds, marsh birds, and ducks were positively associated with open shore in the summer, and wading birds, marsh birds, and diving birds were positively associated with lawn and canopy in the winter. Summer behavioral observations revealed that wading birds foraged significantly more along developed sho ^^ reline, and that ducks rested and tended young significantly more along developed shoreline. Winter observations revealed that marsh birds foraged significantly more along undeveloped shoreline and displayed active/swimming behavior significantly more along developed shoreline. Summer ducks and winter wading birds showed significantly greater alert/flee behavior along undeveloped shoreline. Ducks showed significantly greater alert/flee behavior than other guilds. Overall, alert/flee behavior was seen 1.6 times more often in the winter. Winter migrants did not show greater alert/flee behavior than resident birds. Results show that a wide range of waterbirds can use urban lakes during the breeding and non-breeding seasons. Further, developed shoreline appears to be favored by many species for a variety of behaviors. However, dense stands of tall emergent vegetation along undeveloped shoreline may deter birds from using this shoreline. Greater alert/flee behavior along undeveloped shoreli ^^ nes may warrant the use of buffer zones to protect birds using these shorelines from undue human disturbance.
Subject: behavior, development, lakes, urban, waterbirds
General Note: Title from title page of source document.
General Note: Includes vita.
Thesis: Thesis (M.S.)--University of Florida, 2003.
Bibliography: Includes bibliographical references.
General Note: Text (Electronic thesis) in PDF format.

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: UFE0000807:00001

Permanent Link: http://ufdc.ufl.edu/UFE0000807/00001

Material Information

Title: Urban lakes and waterbirds : effects of development on distribution and behavior
Physical Description: xii, 99 p.
Language: English
Creator: Traut, Ashley H. ( Dissertant )
Hostetler, Mark E. ( Thesis advisor )
Frederick, Peter ( Reviewer )
Tanner, George ( Reviewer )
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2003
Copyright Date: 2003

Subjects

Subjects / Keywords: Wildlife Ecology and Conservation thesis,M.S   ( local )
Dissertations, Academic -- UF -- Wildlife Ecology and Conservation   ( local )

Notes

Abstract: Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Science URBAN LAKES AND WATERBIRDS: EFFECTS OF DEVELOPMENT ON DISTRIBUTION AND BEHAVIOR By Ashley H. Traut May 2003 Chair: Mark E. Hostetler Department: Wildlife Ecology and Conservation I studied waterbird distribution and behavior in the breeding and non-breeding seasons on four partially developed urban lakes in central Florida. I examined waterbird distributional and behavioral associations with developed and undeveloped shorelines, as well as distributional associations with specific elements of the littoral and onshore habitat in the urban environment. By understanding how waterbirds respond to development, and by identifying habitat elements that influence their movements and behaviors, we can provide more ecologically sound development and management practices for urban aquatic environments. A total of 38 waterbird species w ^^ ere observed on the four lakes over both seasons. Wading bird, marsh bird, and duck abundance was significantly greater along developed shoreline in both seasons on all lakes. Diving bird abundance was significantly greater along developed shoreline in the winter. Species richness was not associated with shoreline development. Species evenness was greater along undeveloped shoreline in the summer and developed shoreline in the winter. Tall emergent vegetation, open shore, lawn, and canopy appeared to be the primary habitat elements determining waterbird presence. All waterbirds were negatively associated with tall emergent vegetation on two or more lakes over both seasons, whereas wading birds, marsh birds, and ducks were positively associated with open shore in the summer, and wading birds, marsh birds, and diving birds were positively associated with lawn and canopy in the winter. Summer behavioral observations revealed that wading birds foraged significantly more along developed sho ^^ reline, and that ducks rested and tended young significantly more along developed shoreline. Winter observations revealed that marsh birds foraged significantly more along undeveloped shoreline and displayed active/swimming behavior significantly more along developed shoreline. Summer ducks and winter wading birds showed significantly greater alert/flee behavior along undeveloped shoreline. Ducks showed significantly greater alert/flee behavior than other guilds. Overall, alert/flee behavior was seen 1.6 times more often in the winter. Winter migrants did not show greater alert/flee behavior than resident birds. Results show that a wide range of waterbirds can use urban lakes during the breeding and non-breeding seasons. Further, developed shoreline appears to be favored by many species for a variety of behaviors. However, dense stands of tall emergent vegetation along undeveloped shoreline may deter birds from using this shoreline. Greater alert/flee behavior along undeveloped shoreli ^^ nes may warrant the use of buffer zones to protect birds using these shorelines from undue human disturbance.
Subject: behavior, development, lakes, urban, waterbirds
General Note: Title from title page of source document.
General Note: Includes vita.
Thesis: Thesis (M.S.)--University of Florida, 2003.
Bibliography: Includes bibliographical references.
General Note: Text (Electronic thesis) in PDF format.

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: UFE0000807:00001


This item has the following downloads:


Full Text











URBAN LAKES AND WATERBIRDS: EFFECTS OF DEVELOPMENT
ON DISTRIBUTION AND BEHAVIOR














By

ASHLEY H. TRAUT


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2003





























To Kate,
for your immeasurable patience, support, and love.
You inspire me beyond words.


And to the birds,
who fill my dreams with hope.















ACKNOWLEDGMENTS

I wish to thank the Department of Fisheries and Aquatic Sciences for their financial

and logistical support of this project, and for giving me the opportunity to work with

Florida LAKEWATCH. Mark Hoyer and Eric Schulz were particularly generous with

their time and ideas during the development of this project and provided me with

invaluable advice from different perspectives.

My advisor, Mark Hostetler, deserves far more thanks than a few sentences can

convey. His tireless brainstorming, seemingly endless editing, and frequent

encouragement made my graduate experience a truly valuable and enjoyable experience.

His blending of science with public outreach has been a great inspiration to me, and his

tremendously broad range of interests and scholarly pursuits has reinforced my desire to

expand my own horizons and never stop learning.

I thank also my committee members, Peter Frederick and George Tanner, for

supporting my efforts and freely offering different points of view. I thank Peter

particularly for introducing me to the world of scientific conferences and showing me the

value of sharing and defending my ideas. I thank George for reminding me early on to

keep it fun. Though I never did end up hanging a pole and line off the back of my boat,

his advice reminded me to appreciate my surroundings even during the most trying of

field days.

The numerous IFAS statistical consultants who worked so valiantly to take my

tangled findings and make them presentable also deserve my great thanks. Attempting to









understand the designs and intentions of so many projects and figure out how best to

convey the results is a daunting task. They handle it with admirable skill and patience.

Finally, I wish to thank my parents for their quiet guidance, patience, and love.

Their unconditional encouragement of my interests, no matter how far those interests

have strayed, has afforded me a life of incredible discoveries and happiness.
















TABLE OF CONTENTS


A C K N O W L E D G M E N T S ................................................................................................. iii

LIST OF TABLES .......................................... viii

LIST O F FIG U RE S ............... .......................................... ...x.... .. .... .x

A B S T R A C T ........................................................................................................ ............ x i

CHAPTER

1. IN T R O D U C T IO N ............................................................... .... .................. ............... 1

B a c k g ro u n d ........................ ........................................................................................... 1
U rbanization and A vian E cology .......................................................................... 1
U rb an L ak e s ..................................................................................................... 3
S tu d y A re a .............................................................................................................. 5

2. WATERBIRD DISTRIBUTION, SHORELINE DEVELOPMENT, AND HABITAT
ST R U C T U R E ............................................................................... . ...................9

In tro du ctio n ...................................................................................................... ........... 9
M e th o d s......................................................................................................... . ........... 1 1
Bird Surveys ............... .. .. ............................................... 11
Categorization of Shoreline Development....................................................... 13
Categorization of Habitat Structure ................................................. 13
L ittoral habitat......................................................................................... 14
O nshore habitat .................................................................... ............... 14
S u b state ........................................................................................................... 1 5
A naly ses ...................................... .................................... .................... 15
Summer and Winter Comparisons ................................................................ 15
Waterbird Habitat Use........................... .............. 16
Developed versus undeveloped shoreline .................................................. 16
Littoral and onshore habitat associations................................................... 17
D eterm ining overall significance ................................................ ............... 17
Independence of significant habitat elements ............................................ 18
R e su lts ............... ... ............................ ............................................................. ............ 1 9
Avian Community Composition ................................................................... 19
Shoreline Development and Habitat Coverage................................................ 20
Developed Versus Undeveloped Shoreline Use .............................................. 21


v









S u b state U se .......................................................................................................... 2 5
Littoral Z one H habitat A ssociation...................................................... .............. 25
Sum m er ............................................................................................. . 26
W inter ................................................................... ......................... 26
O nshore H habitat A association .................................... ...................... .............. 27
Sum m er ............................................................................................. . 28
W inter ............................................................... ..................................29
Independence of Significant Habitat Elements.................................. .............. 30
Sum m er ............................................................................................. . 30
W inter .............. ......................................................................................3 1
Discussion .......................... .................. ...... .............. 32
Seasonal Species Com position............................................................................ 32
Shoreline D evelopm ent ....................................................................................... 33
D om inant H habitat Elem ents ................................................................................ 36
L ittoral habitat ...............................................................................................36
Onshore habitat ...........................................37
Guild Responses to Other Habitat Elements......................................... .............. 40
M arsh birds ................................................................................................ 40
W a d in g b ird s ....................................................................................................4 1
D iv in g b ird s .....................................................................................................4 2
D u c k s .............. ..... ........................................................... .................. ....... 4 2
M anagem ent and Future Research............................................................................. 43

3. AVIAN BEHAVIORAL RESPONSES TO SHORELINE DEVELOPMENT ............48

In tro d u c tio n ................................................................................................................... 4 8
M methods .............................................................. 51
A analyses ..................................................... ....... .............. 53
Shoreline D evelopm ent ....................................................................................... 53
D isturbance Sensitivity ....................................................................................... 53
D eterm ining Overall Significance......................................................... .............. 54
R e su lts ....................... ....... ................... ................................................... ........... 5 4
Seasonal Behavioral Observations ................................. 54
Behavioral Associations with Shoreline Development ......................... .............. 55
A lert/flee behavior ........................................................................................55
F oraging behavior .........................................................................................56
R testing behavior ........................................................................................... 56
Tending young behavior ...............................................................................57
A ctive/sw im behavior ................................................................................57
H um an A activity ....................................................................................... 57
D isturbance Sensitivity ....................................................................................... 59
Seasonal alert/flee com parsons ....................................................................59
Inter-guild alert/flee comparisons .................................................................60
M igrant versus resident alert/flee comparisons ............................................60
Discussion................................ ... ................ ......... .............. 61
Avian Responses to Shoreline Development...................................................... 61
Alert/flee behavior .......................... ..........................61









Foraging and resting behavior .................................................... ................ 62
T ending young behavior ..................................... ..................... ................ 63
A ctive/sw im m ing behavior......................................................... ................ 65
D isturbance Sensitivity .................................................................. .............. 66
Seasonal alert/fl ee response ........................................................ ................ 66
Inter-guild alert/flee com parisons............................................... ................ 67
Migrant versus resident alert/flee response.................................................68
M anagem ent and Future R esearch............................................................ .............. 69

APPENDIX

A. WINTER HAVEN WATERBIRD SURVEY DATA.............................................73

B. WINTER HAVEN WATERBIRD HABITAT DATA............................................75

C. WINTER HAVEN WATERBIRD BEHAVIOR DATA.........................................84

R E F E R E N C E S .................................................................................................................. 8 9

BIO GR APH ICAL SK ETCH .................................................................... ................ 99















LIST OF TABLES


Table page

2-1. Total shoreline development and habitat coverage of all habitat elements............... 22

2-2. Overall waterbird abundance along developed and undeveloped shorelines
on lakes Buckeye, Conine, Deer, and Jessie during summer 2001 and
winter 2001/2002. ........... .. .............................. ........ .... ............... 23

2-3. Species abundance along developed and undeveloped shorelines during
sum m er 2001 and w inter 2001/2001................................................. ................ 24

2-4. Waterbird guild associations with littoral habitat elements for summer 2001
(S), and w inter 2001/2002 (W ) ........................................................ ................ 28

2-5. Waterbird associations with onshore habitat elements by guild for summer 2001
(S), and w inter 2001/2002 (W ) ........................................................ ................ 30

3-1. Percent of waterbird guilds engaged in focal behaviors during summer (S) 2001
and w inter (W ) 200 1/2002 ...................................... ...................... ................ 55

3-2. Percent guild behavioral responses to developed (D) and undeveloped (U)
shoreline for sum m er 200 1 .. .................................... ...................... ................ 58

3-3. Percent guild behavioral responses to developed (D) and undeveloped (U)
shoreline for w inter 2001/2002......................................................... ................ 59

A-1. Aquatic bird species observed on lakes Buckeye (B), Conine (C), Deer (D), and
Jessie (J) in Winter Haven, Florida, summer 2001 and winter 2001/2002............73

B-1. W ading bird habitat associations, summer 2001.. ............................... ................ 76

B-2. Wading bird habitat associations, winter 2001/2002...........................................77

B-3. M arsh birds habitat associations, summ er 2001.. ................................ ................ 78

B-4. M arsh bird habitat associations, winter 2001-2002............................. ................ 79

B-5. Diving bird habitat associations, summer 2001................................... ................ 80

B-6. Diving bird habitat associations, winter 2001/2002.. .......................... ................ 81









B-7. Duck habitat associations, sum m er 2001............................................. ................ 82

B-8. Duck habitat associations, winter 2001/2002. Expected values based on
availability of habitat elem ent on each lake...................................... ................ 83

C-1. Data for summer guild behavior, listed by guild and lake...................................85

C-2 Data for winter guild behavior, listed by guild and lake ................. ..................... 87















LIST OF FIGURES


Figure page

1-1. Map of study lakes within Winter Haven urban area, Polk county, FL ...................7...

1-2. A erial im ages of W inter H aven study lakes .......................................... ...............8...















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Master of Science

URBAN LAKES AND WATERBIRDS: EFFECTS OF DEVELOPMENT
ON DISTRIBUTION AND BEHAVIOR


By

Ashley H. Traut

May 2003
Chair: Mark E. Hostetler
Department: Wildlife Ecology and Conservation

I studied waterbird distribution and behavior in the breeding and non-breeding

seasons on four partially developed urban lakes in central Florida. I examined waterbird

distributional and behavioral associations with developed and undeveloped shorelines, as

well as distributional associations with specific elements of the littoral and onshore

habitat in the urban environment. By understanding how waterbirds respond to

development, and by identifying habitat elements that influence their movements and

behaviors, we can provide more ecologically sound development and management

practices for urban aquatic environments.

A total of 38 waterbird species were observed on the four lakes over both seasons.

Wading bird, marsh bird, and duck abundance was significantly greater along developed

shoreline in both seasons on all lakes. Diving bird abundance was significantly greater

along developed shoreline in the winter. Species richness was not associated with









shoreline development. Species evenness was greater along undeveloped shoreline in the

summer and developed shoreline in the winter.

Tall emergent vegetation, open shore, lawn, and canopy appeared to be the primary

habitat elements determining waterbird presence. All waterbirds were negatively

associated with tall emergent vegetation on two or more lakes over both seasons, whereas

wading birds, marsh birds, and ducks were positively associated with open shore in the

summer, and wading birds, marsh birds, and diving birds were positively associated with

lawn and canopy in the winter.

Summer behavioral observations revealed that wading birds foraged significantly

more along developed shoreline, and that ducks rested and tended young significantly

more along developed shoreline. Winter observations revealed that marsh birds foraged

significantly more along undeveloped shoreline and displayed active/swimming behavior

significantly more along developed shoreline.

Summer ducks and winter wading birds showed significantly greater alert/flee

behavior along undeveloped shoreline. Ducks showed significantly greater alert/flee

behavior than other guilds. Overall, alert/flee behavior was seen 1.6 times more often in

the winter. Winter migrants did not show greater alert/flee behavior than resident birds.

Results show that a wide range of waterbirds can use urban lakes during the

breeding and non-breeding seasons. Further, developed shoreline appears to be favored

by many species for a variety of behaviors. However, dense stands of tall emergent

vegetation along undeveloped shoreline may deter birds from using this shoreline.

Greater alert/flee behavior along undeveloped shorelines may warrant the use of buffer

zones to protect birds using these shorelines from undue human disturbance.














CHAPTER 1
INTRODUCTION

Background

Urbanization and Avian Ecology

Between 1900 and 1987 the proportion of the world's human population living in

cities rose from 14 to 50 percent (United Nations 1987). By 2050 the world's urban

population is predicted to equal 6.5 billion, equivalent to today's entire global population

(United Nations 1996). Though the majority of urbanization will occur in developing

countries, seventy-eight percent of United States residents already reside in urban

environments (Adams 1994). Strohm (1974) estimated that this process of urbanization

meant the paving, building over, or drowning of over 400,000 hectares of natural habitat

a year. Today, urbanization is the second most frequently cited cause of species

endangerment in the United States (Czech & Krausman 1997). With this understanding

comes the need for a comprehensive understanding of the ecology of urban systems and

how best to manage them both for the needs of humans and for wildlife. To date,

however, there is a relative scarcity of such knowledge, with many questions still

unanswered about the effects of urbanization on everything from ecosystems to

individual species (Cairns 1988, Niemela 1999). As a result, most urban planning and

management remains focused on the impacts of urbanization on human society rather

than on the issues of biodiversity (Marzluff et al. 2001).

Urban ecosystems as a whole can be viewed as highly fragmented, heterogeneous

landscapes dominated by buildings, roads, and pavement, and often lacking in substantial









vegetation cover (Jokimaki 1999). The remaining vegetation composition is often greatly

altered, consisting of a few favored natives and numerous exotic species (DeGraaf 1986).

Further, the fragments of natural vegetation that are left relatively intact may be too small

or isolated to support a healthy wildlife community (Savard et al. 2000). Urban habitats

are also characterized by high levels of human-associated disturbance, such as traffic,

construction, and recreation (Jokimaki 1999). These changes in structure and function

can lead to a greatly modified wildlife assemblage consisting of habitat generalists,

human-commensal species, and exotic species. Wildlife diversity in urban environments

ultimately depends on how humans design and manage urban habitats.

Many urban wildlife studies to date have focused on avian responses to

development. Birds are often selected for study due to their diurnal activity patterns and

relative ease of identification both by song and sight (DeGraaf & Wentworth 1981).

They are also regarded as excellent indicators of stresses in an environment due to their

sensitivity to change in habitat structure and composition (Savard & Falls 1982, Clergeau

et al. 1998). The presence or absence of various avian species within urban areas is often

associated with changes in habitat structure.

Results from several studies have shown that with increased urbanization there is a

shift in avian species composition, often accompanied with a decrease in bird species

richness and diversity, and an increase in total bird density as a few human-commensal,

often non-native species, such as the House Sparrow (Passer domesticus) and European

Starling (Sturnus vulgaris) become very common (see Blair 1996 and Savard et al. 2000

for reviews). Decreased habitat availability, vegetative complexity, and food supply, and

increased habitat fragmentation, competition, and human disturbance are examples of









some of the mechanisms that have been identified as contributing to decreases in richness

and evenness in urban bird communities (Marzluff 2001). Conversely, factors such as

supplemental feeding, reduced predation, and reduced human persecution have benefited

certain species in urban environments (Marzluff 2001). Species composition and

richness have also varied in relation to the city and the locality within a city in which

studies were conducted, with some areas showing considerable diversity, depending on

local environmental conditions (Tilghman 1987, Blair 1996, Hostetler 1999, Hostetler &

Holling 2000). The design and management of an area can have an appreciable effect on

the distribution of birds across an urban environment.

Urban Lakes

The vast majority of urban bird studies have focused on passerine bird species in

terrestrial habitats. Studies examining waterbirds have generally focused on marsh

systems in non-urban environments, with relatively little attention being devoted to

lacustrine habitats (but see Parris & Grau 1978, Whitfield & Cyrus 1978, Johnson &

Montalbano 1984, Zaffke 1984, Pyrovetsi & Crivelli 1988, Edelson 1990, Hoyer &

Canfield 1990, 1994). Virtually no studies have been conducted in the United States that

have directly quantified waterbird responses to lakefront urbanization.

Urbanized lakes often undergo similar patterns of habitat alteration as terrestrial

habitats. Habitat complexity is often greatly reduced as both onshore and littoral

vegetation are reduced, replaced, or removed. The clearing of these habitats by property

owners is a common practice in order to improve lake views and recreational pursuits

such as swimming, boating, and fishing (Guillory et al. 1979, Frayer & Hefner 1991,

Bryan & Scarnecchia 1992). Such modifications are usually made without considering

the potential effects on local wildlife.









With the continued loss of wetland habitat throughout the United States (Mitsch &

Gosselink 1993), the importance of lacustrine habitat to waterbirds may be increasing.

Florida, home to some of the largest waterbird populations in the United States, has lost

almost 50% of its wetlands over the past 200 years (Dahl 1990). Much of this loss can be

attributed to urbanization, as seen in the fact that more than 84% of the state's 16 million

residents currently live in urban areas (Morris & Morris 1995). With an estimated 7,783

lakes providing potential waterbird habitat, and at the same time facing significant

development pressures, Florida offers optimal conditions for studying the effects of

urbanization on waterbirds and developing better management policies for urban lakes.

In this project, I explored how waterbirds responded to shoreline habitat changes

caused by human encroachment on urban lakes. In it, I examined both indirect and direct

responses in waterbirds during the breeding and non-breeding seasons. In Chapter 2, I

examine indirect responses measured by the presence or absence of birds along

developed shoreline and in other specific habitat variables. It addresses the following

research questions:

1. What is the composition of waterbirds found on Central Florida's urban lakes
during the breeding and non-breeding seasons?

2. Are waterbird abundance, community composition, and species richness
significantly different between developed and undeveloped shorelines?

3. Which onshore and littoral habitat elements are most closely associated with the
presence or absence of various waterbird guilds?

In Chapter 3, I examine direct responses of waterbirds to development and human

disturbance by measuring changes in waterbird behavior. Chapter 3 addresses the

following research questions:

1. Are primary waterbird behaviors significantly different between developed and
undeveloped shoreline?









2. Which waterbird guilds appear most sensitive to human disturbance?

Study Area

I conducted field research on four urban lakes within the Peace River drainage

basin in and around the city of Winter Haven in Polk County, Florida (Figure 1-1). Polk

County ranks fourth in lake abundance among all Florida counties, with 550 lakes within

its borders (Edmiston & Myers 1983). Lakes Buckeye (280 2' 24" N, 810 42' 20" W),

Conine (280 3' 22" N, 810 43' 29" W), Deer (280 1' 32" N, 810 45' 46" W), and Jessie

(280 3' 27" N, 810 45' 48" W) are moderately developed lakes located along the edges of

Winter Haven's urban area. They were selected for study because each lake still had

significant portions of undeveloped shoreline. Polk County Environmental Services

classifies these lakes as mesotrophic or eutrophic, meaning they carry moderate to high

nutrient loads based on chlorophyll, nitrogen, and phosphorus concentrations (Polk

County 1999). Each lake meets Florida Class III criteria for surface water quality,

designated for the propagation and maintenance of healthy, well-balanced fish and

wildlife populations (Polk County 1999).

Lake Buckeye was the smallest lake (29 ha) with the least developed shoreline

(59%) (Figure 1-2). The majority of its undeveloped shoreline was located along the NE

side of the lake and consisted of bottomland hardwoods (Florida DOT 1999). Lake Deer

(47 ha) had the greatest amount of developed shoreline at 79%. Undeveloped shoreline,

consisting of hardwood/conifer mix (Florida DOT 1999), was found in one contiguous

stand along the SW corner of the lake. Lake Jessie (75 ha) was 69% developed with

hardwood/conifer mix found along the northern shore and bottomland hardwoods

dominating most of the eastern shore (Florida DOT 1999). Lake Conine was the largest

lake at 96 ha and was developed along 61% of its shoreline. Three undeveloped areas,









consisting of hardwood/conifer mix, wetland forest, and bottomland hardwood, were

evenly spaced around the lake (Florida DOT 1999).

Undeveloped habitats on each lake were mostly discrete forest fragments less than

20 ha in size, and surrounded by development on all sides. Urbanization consisted

primarily of single-family homes, with several apartment complexes located on lakes

Buckeye and Deer, and a mobile-home park and private airport located on the north side

of Lake Jessie. Much of the east side of Lake Conine and the southeast corner of Lake

Buckeye were within 30 m of moderately to heavily traveled roads. All lakes had

shallow sloping littoral shelves (Florida LAKEWATCH 2000) with moderately diverse

littoral plant communities. The vast majority of residential properties had a dock and

boathouse on the water. All lakes had public access boat ramps.

Lake littoral zones were markedly different along the developed and undeveloped

shorelines. Undeveloped zones were often characterized by continuous dense stands of

cattail (Typha spp.). Lake Jessie was the one exception to this, with much of its east side

having sparse cattail mixed with a variety of lower emergents such as para grass

(Urochloa mutica), torpedo grass (Panicum repens), and maidencane (Panicum

hemitomon). Littoral zones along developed shorelines had a much patchier aquatic plant

distribution with considerably more open water and far less cattail. Heavily developed

shorelines were devoid of all aquatic vegetation. Less developed properties and areas

between properties were characterized by a diverse community of low and tall emergent

macrophyte species. Lake Deer was notable for its near continuous emergent zone

around both the undeveloped and developed shores, and its dense floating-leafed

macrophyte coverage (Nymphea spp.) around more than 80% of the lake perimeter.


















Lake Jessie








Lake Deer









3


0 3 6 Kilometers


Map of study lakes within Winter Haven urban area, Polk County, FL.
Developed areas are in grey. Undeveloped areas are in green. Study lakes are
in dark blue.


Figure 1-1.











































Figure 1-2. Aerial images of Winter Haven study lakes. Clockwise from upper left are
Lakes Buckeye, Deer, Jessie, and Conine.

During this study lake levels were unusually low (.5 1 m below normal) due to a

severe drought. Water levels in the summer of 2001 reached record lows on all of the

lakes, exposing much of the emergent zone substrate, creating numerous open mudflats

along unvegetated shoreline, and creating a 50 m peninsula along the southern shore of

Lake Conine and a small mudflat island off the southern shore of Lake Deer.














CHAPTER 2
WATERBIRD DISTRIBUTION, SHORELINE DEVELOPMENT, AND
HABITAT STRUCTURE

Introduction

The importance of habitat structure in determining avian species composition in

terrestrial systems has been well established (MacArthur & MacArthur 1961, MacArthur

& Wilson 1967, Karr & Roth, 1971, Roth 1976). Though many factors have been

proposed to explain avifaunal shifts in urban environments (Marzluff 2001), changes in

habitat structure and composition are considered some of the primary mechanisms.

Decreases in avian species richness and evenness in urban areas have been correlated

with decreases in total woody vegetation volume, spatial heterogeneity, vertical structure,

and plant species diversity (Lancaster & Rees 1979, DeGraaf 1986, Tilghman 1987,

Marzluff & Sallabanks 1998), and increases in habitat fragmentation, habitat edge, and

exotic vegetation (Soule et al. 1988, Marzluff 2001). Avian species diversity has been

found to be negatively correlated with elements of the built up environment, such as

housing density (Lancaster & Rees 1979, Blair 2001).

The mechanisms that determine avian species distribution in aquatic systems have

also received considerable attention. Broad-scale factors such as geography and climate

have been shown to determine waterbird breeding and winter ranges (Weller 1999,

Elphick et al. 2001). On a local scale, mechanisms such as species morphology, lake

area, water depth, trophic status, and predator and prey densities have all been shown to

influence the distribution of waterbird communities (Weller & Spatcher 1965, Jenni









1969, Brown & Dinsmore 1986, Picman et al. 1993, Hoyer & Canfield 1994, Weller

1999).

As in terrestrial systems, habitat structure may be one of the most important factors

in determining avian community composition in aquatic systems, including physical

components like water column depth, wetland substrate, shoreline, and vegetation strata

(Weller 1999). Specific habitat associations are well documented for a variety of

waterbirds. For example, grebes (Podicipedidae) prefer relatively shallow, well-

vegetated wetlands (Weller 1999, Elphick et al. 2001). Other divers, such as the

cormorants (Phalacrocoracidae) and anhingas (Anhingidae), require shrubs, trees, or

snags near the water on which to loaf and nest (Hatch & Weseloh 1999, Frederick &

Siegel-Causey 2000). Many large and mid-sized wading birds are known to prefer

shallow, relatively open areas for foraging (Hancock & Kushlan 1984). Marsh birds such

as rails and gallinules (Rallidae) require dense emergent or floating vegetation for nesting

and foraging, while coots commonly use more open water habitat (Melvin & Gibbs 1996,

Weller 1999, West & Hess 2002).

In urban areas, development and the alteration of both aquatic and terrestrial habitat

structure may be the most important factors in determining the composition and

distribution of waterbird communities. However, few studies have looked at the effects

of lake-habitat structure on urban bird populations. In general, the majority of research

on urban aquatic environments in the United States has focused on the effects of

recreation on avian abundance, distribution, and breeding success (Hockin et al. 1992,

Knight & Gutzwiller 1995). In general, these studies have found lower abundance,

reduced use of sites, and lower breeding success in areas with significant recreation (see









Hockin et al. 1992 for review). Studies in Europe and Canada have examined the effects

of shoreline cottages on diving birds (Lehtonen 1970, Bundy 1979, Andersson et al.

1980, Heimberger et al. 1983), finding reduced reproductive success and lake utilization

in areas with cottages. These studies focused on a single species and did not attempt to

quantify the role of habitat structure. The aim of this study was to directly examine the

effects of development in urban lake environments by answering the following questions:

1. What is the composition of waterbirds found on Central Florida's urban lakes
during the breeding and non-breeding seasons?

2. Are waterbird abundance, community composition, and species richness different
between developed and undeveloped shoreline?

3. Which onshore and littoral habitat elements are most closely associated with the
presence or absence of various waterbird guilds?

Question 1 addresses the general lack of knowledge about avian species

composition on urban lakes by looking for species or groups of birds that are noticeably

absent, exploring seasonal fluctuations, and comparing community composition with

previous lake studies conducted in the Central Florida region (Hoyer & Canfield 1990,

1994, Roth in press). Questions 2 and 3 examine the impact of developed shoreline on

the composition and distribution of waterbirds and explore the role of specific habitat

elements in determining distribution patterns.

Methods

Bird Surveys

Waterbird surveys were conducted from June 7 August 1 of 2001 and December

8 February 6 of 2001/2002. For this study the term waterbirdd' referred to species in

the orders Gaviiformes, Podicipediformes, Pelecaniformes, Ciconiiformes, Anseriformes,

Falconiformes, Gruiformes, Charadriiformes, or Coraciiformes observed on or feeding









from lacustrine habitats. Each of the four lakes were surveyed a total of eight times each

season by driving at minimum-wake speed around the lakes (20 30 m from shore) in a

small motorized canoe. Both the order of lakes surveyed and the direction of travel were

alternated for each day of surveying in order to account for any time-related bird

movements around the lakes. Surveys were conducted within the first five hours after

sunrise on mornings with little to no rain and winds less than 24 km/hr.

Birds were identified by sight using 8x42 binoculars. Shoreline development and

associated vegetation structure were recorded for each bird once it was identified (see

description below). The location of each bird was also recorded on a 1999 aerial

photograph of each lake obtained from the Florida Department of Environmental

Protection (DEP) at their Land Boundary Information Systems (LABINS) website

(www.labins.org).

In order to minimize count error along undeveloped shoreline, where dense tall-

emergent vegetation was often encountered in the littoral zone, I drove the survey boat

directly into the vegetation to flush hidden birds. This procedure was repeated every 30 -

50 m in areas of continuous vegetation. On alternate insertions, I shut off the motor and

listened for calling birds for approximately two minutes.

Birds that flushed from any location and landed ahead of the boat were recorded

only in their original location. Birds whose origin and destination were not observed or

which were observed greater than 30 m offshore or 10 m onshore were recorded for

analysis of overall bird composition on a given lake, but were not included in either

shoreline development or habitat analyses.









Categorization of Shoreline Development

Each lake in the study had a mixture of developed and undeveloped shoreline.

Lake shoreline was categorized as either developed or undeveloped based on DEP land

cover classifications for each lake. Classifications were modified during preliminary

surveys by basing categorizations only on the land cover within the first 20 m of

shoreline extending away from the water on each lake. For the purposes of this study,

developed shoreline referred to any continuous stretch of shoreline greater than 100 m,

parallel to the edge of the lake, that had a minimum of 50% long-term habitat alteration,

defined as cleared land, lawns, landscaping, buildings, and roads. Undeveloped shoreline

was defined as any continuous stretch of shoreline greater than 100 m, parallel to the

edge of the lake, with greater than 50% intact natural habitat, and little to no sign of

regular human use. Developed and undeveloped areas were separated by a buffer of 40

m to eliminate the potential effects of converging habitats. Birds recorded in these border

areas were not included in either shoreline development or habitat analyses.

Categorization of Habitat Structure

Upon sighting each bird I recorded several different habitat elements within the

littoral zone and immediate onshore zone, as well as the type of substrate that each bird

was sighted on. Visual estimates were made for littoral and onshore habitat elements

found in 5x20 m rectangular bands around and adjacent to each bird. The 20-meter side

of the band was parallel to the shoreline. Coverage of each of eight habitat elements was

classified as one of five densities: 0-5%, 6-25%, 26-50%, 51-75%, or 76-100%. This was

done in order to see if minimum or maximum thresholds existed at which birds selected

or avoided each habitat element.









Littoral habitat

Three categories were created for littoral zone habitat: low emergent vegetation (<

1 m tall), tall emergent vegetation (> 1 m tall), and floating-leafed vegetation. For birds

sighted onshore (within 10 m of the water) or less than 5 m offshore, I recorded the first 5

m of aquatic vegetation extending out from the shoreline and within 10 m of either side

of the bird. For birds located greater than 5 m from shore, the 5x20 m band was centered

around each bird. Since the drought caused extremely low water stages during this study,

I determined the upland edge of the littoral zone by the edge of herbaceous emergent

vegetation, rather than the presence of standing water. In several undeveloped areas on

lakes Buckeye and Conine, backmarshes of fallen cattail and vines had developed behind

dense stands of cattail, completely covering the shallows. In such cases, where the water

was completely covered by vegetation and was too shallow for any diving birds to use, I

considered the area onshore habitat.

Onshore habitat

Five categories were created for shoreline habitat: open shore (moist soil or sand),

lawn (any maintained grassy area), understory (< .5 m tall), shrub (.5-3 m tall), and

canopy (> 3 m tall). Immediate onshore habitat was recorded for any bird in the water

within 20 m of shore and any bird onshore within 10 m of the water. For birds located in

the water, I made a trajectory (perpendicular to the shoreline) from where the bird was

sighted to a point on shore. For these birds and birds located onshore less than 5 m from

the water, I recorded onshore habitat within a 5x20 m band placed 10 m on either side of

the trajectory line or bird and extending 5 m inland from the water's edge. For birds

located greater than 5 m from the water, the onshore band was centered on the bird.









Substrate

Substrate was defined as the human structure (e.g. pier, boat, etc.) or habitat

element (e.g. cattail, shallows, lawn, etc.) on or nearest to which each bird was found.

For example, if a bird was observed foraging in a pocket of unvegetated shallow water in

a 5x20 m area that was dominated by tall emergent vegetation, the bird's substrate was

recorded as shallow water.

Analyses

Waterbirds were grouped into guilds for certain analyses. Guilds were based on

foraging behavior and habitat use, and included wading birds (Ardeidae,

Threskiornithidae, Ciconiidae, Recurvirostridae), marsh birds (Rallidae), surface and

aerial diving birds (Podicipedidae, Phalacrocoracidae, Anhingidae, Accipitridae,

Laridae), and ducks (Anatidae). Analyses were only conducted for a given guild or

species where 10 or more birds (n > 10) were observed within that guild or species over a

season.

Summer and Winter Comparisons

Species counts from all lakes were summed for each season to examine overall

waterbird community composition. Species richness and guild abundance on each of the

lakes were calculated for each survey day and then averaged across all days and lakes

each season (8 surveys per season) to determine the mean number of species and

individuals observed.

The percent similarity measure (Brower et al. 1990) was used to compare avian

community composition between seasons in order to determine whether differences in

seasonal avian distribution patterns could be explained by differences in community

composition.









The Jacaard index of community similarity (Brower et al. 1990) was used to

compare community composition with that of two previous urban lake studies in Central

Florida. Hoyer & Canfield (1994) examined 33 lakes, both developed and undeveloped

in the north central and central regions of Florida. Roth (in press) examined six urban

lakes in the Winter Haven area in 1991.

Waterbird Habitat Use

I tested for waterbird associations with developed and undeveloped shoreline and

with littoral and onshore habitat elements using Chi-square goodness-of-fit tests (a =

0.05, df= 1).

Developed versus undeveloped shoreline

Overall abundance, guild abundance, species abundance, species richness, and

species evenness were compared between areas of developed and undeveloped shoreline.

Expected values were based on the proportions of developed and undeveloped shoreline.

For example, if 100 wading birds were observed on a lake and 70% of the shoreline was

developed, then, assuming a random distribution of birds, expected values would be 70

birds along developed shoreline and 30 along undeveloped shoreline.

Overall abundance, species richness, and species evenness were compared between

developed and undeveloped shorelines on a lake-by-lake basis. Each lake was analyzed

separately to allow for the possibility of intra-lake variability. Guild abundance and

species abundance were calculated with all lakes combined based on the results of overall

abundance. Evenness was calculated using Simpson's measure of evenness (Krebs

1998).









Littoral and onshore habitat associations

I tested for waterbird associations with littoral and onshore habitat elements by

comparing guild abundance in the presence and absence of each of the habitat elements

on each lake. For each sighting, a habitat element was considered absent if it was

recorded as occurring in 0-5% of the 5x20 m band around a bird. A habitat element was

considered present if it was recorded as occurring in greater than five percent of the 5x20

m band around a bird.

Expected values were based on the percent lake coverage of each habitat element

around the perimeter each lake. Habitat measures were taken during the winter

2001/2002 field season. Habitat coverage did not noticeably change between seasons

except for open shoreline, which dramatically decreased from summer to winter due to

rising water with the onset of fall rains. The amount of open shoreline was therefore

calculated separately for each season. Habitat elements were assessed for

presence/absence along the perimeter of each lake. Each littoral and shoreline habitat

element was recorded as present if there was greater than 5% coverage within the first 10

m offshore or onshore, respectively. Measurements were then entered into ArcView 3.2

(ESRI 1999) to determine the exact percentage of the lakes that was covered by each

habitat type. Any habitat type on a lake that occurred in less than 5% of the total littoral

or shoreline zone was considered functionally absent and excluded from analysis.

Determining overall significance

Where analyses were conducted on a lake-by-lake basis, results were considered to

have overall significance where significant same-direction associations (p < 0.05) were

observed on at least three lakes without a significant contradictory finding on the fourth

lake. In cases where habitat elements were only present on two lakes, results were









considered to have overall significance where significant same direction associations (p <

0.05) were observed on both lakes. In cases where results were not significant but

suggestive of an overall pattern on three or more lakes (p < 0.2 for each lake), without a

contradictory finding on the fourth lake, probabilities (for lakes indicating a pattern) were

combined for meta-analysis (Fisher 1958) to test for overall significance.

Habitat elements that were found to be significant were examined across the five

different habitat coverage densities to see if minimum or maximum thresholds existed at

which birds selected or avoided each habitat type.

Independence of significant habitat elements

My analyses for habitat association were univariate in nature, and therefore it was

difficult to detect whether birds were responding to individual habitat elements, or to

combinations of habitat elements. I looked for evidence of this preference for co-

occurring habitat elements by comparing the proportion of birds that were significantly

associated with two or more elements, with an estimate of the actual proportion of co-

occurrence of those habitats on the lake. I first determined how often birds within a guild

were sighted in significant positive association with two or more elements. I then used

the aerial photographs and habitat estimates taken during the winter to estimate the

degree of overlap among elements of interest (to nearest 1%). Since exact estimates

could not be determined due to the resolution of the photographs, statistical analyses were

not employed. I concluded that a guild may have been responding to a combination of

habitat elements if the difference between percent actual overlap of habitats and percent

of birds sighted where habitats overlapped was 50% or more. For example, if lawn and

open shore habitats spatially overlapped 10% of the time, but 80% of wading birds

observed were in areas where both elements were present, then I concluded that the guild









was responding to the combination of these elements. Cross tabulations could not be

used for negatively significant habitat elements since, by definition, birds were found in

areas where these elements were absent. Independence of negative habitat elements was

therefore estimated by looking only at the degree of actual spatial overlap around the

lake.

Results

Avian Community Composition

A total of 38 waterbird species were observed over the course of this study,

including nine species listed as endangered, threatened, or of special concern in the state

of Florida (Appendix A). Thirty-five species were observed on more than 10% of the

surveys.

When overall community composition on these lakes was compared with the two

previous studies of Central Florida lakes, species similarity (Jacaard index) was 0.76 and

0.7. A similarity index of 1.0 in either case would indicate that all species were found in

both studies.

A total of 33 waterbird species were observed during the summer 2001 season. An

average of 13.6 species and 104.1 birds were observed per lake each day. Standard

deviations between lakes varied widely over both seasons. Wading birds were the most

common guild in terms of species richness, making up 45% of all species observed.

Ducks showed the greatest abundance, making up 39% of all birds observed. Wood

Ducks (Aix sponsa) accounted for 96% of all ducks.

Thirty-two species were observed during the winter 2001/2002 season. An average

of 14.8 species and 114.4 birds were observed per lake each day. Wading birds

continued to show the greatest species richness, making up 41% of all species. Several









large flocks of migrant Double-crested Cormorants (Phalacrocorax auritus) accounted

for diving birds showing the greatest winter abundance, making up 34% of all birds

observed.

Species composition between seasons was moderately similar (0.54, percent

similarity measure), with the variation explained by the arrival of 10 winter migrant

species (i.e., birds observed in significantly greater numbers in winter than in summer in

central Florida) (Appendix A). Six of these species were diving birds.

Shoreline Development and Habitat Coverage

Table 2.1 gives a breakdown of the percent coverage of developed and

undeveloped shorelines and habitat elements on each lake. Due to the overlap between

most habitat elements, total percentages on any lake were greater than 100%.

Understory, shrub, and canopy in particular had near 100% overlap along undeveloped

shoreline.

Littoral zones along developed shorelines were characterized by a very patchy

habitat structure, with distinct clumps of low and tall emergent vegetation and large areas

devoid of aquatic vegetation. Onshore habitat in developed areas was characterized by

significant lawn coverage with very sparse, intermittent understory and shrub layers.

Across all lakes, 90% of open shore habitat was found along developed shoreline. This

habitat was typically found in conjunction with lawn habitat where onshore and littoral

vegetation had been cleared.

Undeveloped shorelines, on the other hand, were much more homogeneous. Tall

emergent vegetation dominated much of the littoral zones. Cattail (Typha sp.) was the

dominant emergent vegetation and was present along 89% of undeveloped shoreline, with

almost 70% of it found in dense, continuous stands. Open shore, or exposed shoreline in









general was therefore rarely available. Onshore habitat consisted of relatively continuous

low to moderate understory, and moderate to dense shrub and canopy.

Tall emergent vegetation was the dominant littoral zone habitat element on three of

the four lakes, being present in 47-76% of total littoral zone area. Lake Deer showed

considerable tall emergent coverage, at 68%, but was also dominated by floating-leafed

vegetation, which was found in 83% of the total littoral zone area. Floating-leafed

vegetation was considered functionally absent on Lakes Conine and Jessie since each

lake had less than 5% coverage.

None of the shoreline habitat elements clearly dominated the overall shoreline

habitat structure. Open shore significantly declined between seasons, dropping from 29%

to 7% of available onshore area from summer to winter. On Lake Buckeye and Deer,

open shore coverage dropped below 5% in the winter and therefore was considered

functionally absent on those lakes.

Developed Versus Undeveloped Shoreline Use

During both seasons a strong association was observed between shoreline

development and bird abundance, with more birds found along developed shore on all

four lakes (Table 2-2). Wading birds, marsh birds, and ducks showed this positive

association over both seasons (all tests: x > 39.09, p < 0.0001). Diving birds showed this

positive association only in the winter (all tests: x > 3.87, p < 0.05). Only one lake, Lake

Deer in the summer, showed an association between species richness and shoreline

development. In this case, significantly more species than expected were found along

undeveloped shoreline (x2 = 7.01, p = 0.008). Since no other lakes in either season

showed this pattern, an overall association between species richness and shoreline

development was not established. Temporal and spatial patterns of evenness varied









considerably among lakes. Undeveloped shoreline had greater species evenness than

developed shoreline on all four lakes in the summer. Evenness indices for developed

shoreline in the summer ranged from 0.15 0.38, and for undeveloped shoreline ranged

from 0.27 0.68. The opposite pattern was observed in the winter, with three of the

lakes having greater species evenness along developed shoreline. Evenness for

developed shoreline ranged from 0.62 0.95, and for undeveloped shoreline from 0.51 -

0.73. In all cases, an evenness index of 1.0 would mean that all species were comprised

of an equal number of individuals.

Table 2-1. Total shoreline development and habitat coverage of all habitat elements.
Shoreline development measures were based on first 20 m of shoreline
surrounding each lake. Habitat coverage was based on 5 m deep perimeter
bands for littoral and shoreline zone habitats.
Buckeye Conine Deer Jessie
Area (ha) % Area (ha) % Area (ha) % Area (ha) %
Development
Developed 2.62 59 4.32 61 4.14 79 4.51 69
Undeveloped 1.83 41 2.74 39 1.09 21 2.09 31
Onshore Habitat
Canopy 0.59 48 0.17 9 0.51 38 0.54 31
Shrub 0.76 62 1.22 65 0.55 41 0.81 47
Understory 0.75 61 1.28 68 0.62 46 0.82 47
Lawn 0.52 42 0.29 15 0.80 59 0.82 47
Open Shore summer 0.41 34 0.69 37 0.18 14 0.58 33
Open Shore winter 0.03 2* 0.31 16 0.01 1* 0.16 9
Littoral Habitat
Floating Leaf 0.44 37 0.01 1* 1.06 83 0.05 3*
Tall Emergent 0.69 59 1.38 76 0.87 68 0.78 47
Low Emergent 0.06 5 0.28 16 0.55 43 0.77 46
* Below 5% considered absent

In terms of individual species abundance patterns, results for the summer season

revealed that eight of 28 species (29%) observed using shoreline habitat showed a

significant positive association with developed shore (all tests: 2 > 5, p < 0.03). In the

winter, 16 of 27 species (59%) showed a significant positive association with developed









shore (all tests: x2 > 5.48, p < 0.02); a significant increase over the summer season (2 >

5.26, p < 0.02). Winter migrants accounted for half of the species showing a positive

association for developed shoreline in the winter.

Table 2-2. Overall waterbird abundance along developed and undeveloped shorelines on
lakes Buckeye, Conine, Deer, and Jessie during summer 2001 and winter
2001/2002. All numbers represent the sum total from eight surveys conducted
each season.
Lake Relative Abundance
Summer Winter
Developed Undeveloped Developed Undeveloped
Buckeye 214 66 195 73
Conine 336* 133 522* 49
Deer 1209* 103 1254* 47
Jessie 693* 142 745* 103
SFrom Chi-square goodness-of-fit tests, significantly more birds than expected along
indicated shoreline (p < 0.0001).

Year-round residents that showed a significant association with developed

shoreline during both the summer and winter seasons included the Snowy Egret (Egretta

thula), Tricolored Heron (Egretta tricolor), White Ibis (Eudocimus albus), Wood Duck,

Common Moorhen (Gallinula chloropus), Purple Gallinule (Porphyrula martinica), and

Killdeer (Charadrius vociferous) (Table 2-3). Winter migrants that showed a significant

association with developed shoreline in the winter included the Double-crested

Cormorant (Phalacrocorax aulitus), Ring-necked Duck (Aythya collaris), American Coot

(Fulica Americana), Ring-billed Gull (Larus delawarensis), Belted Kingfisher (Ceryle

alcyon), and Fish Crow (Corvus ossifragus).









Table 2-3. Species abundance along developed and undeveloped shorelines during
summer 2001 and winter 2001/2002. Data from all lakes and survey dates are
summed for each season.
Species Summer Abundance Winter Abundance
Developed Undeveloped Developed Undeveloped
Diving Birds
Pied-billed Grebe 1 1 20 3
D-C Cormorant 34 23 97* 9
Anhinga 145 76 97 82*
Osprey 11 9 9 9
Ring-billed Gull n/a n/a 18* 0
Belted Kingfisher 2 0 24* 1
Wading Birds
Least Bittern 8 8 0 3
Great Blue Heron 106 50 61 32
Great Egret 43 20 41* 5
Snowy Egrets 45* 6 17* 0
Little Blue Herons 11 3 18* 0
Tricolored Herons 51* 6 44* 2
Cattle Egret 3 0 66* 0
Green Heron 61 32 11 6
B-C Night Heron 1 1 0 6
White Ibiss 160* 9 223* 0
Glossy Ibis 12 8 n/a n/a
Wood StorkE 6 1 2 0
Limpkins 8 1 2 0
Sandhill CraneT 8 0 2 0
Black-Necked Stilt 2 0 n/a n/a
Ducks (wild)
Wood Duck 1022* 89 384* 40
Mallard 56* 0 n/a n/a
Blue-winged Teal n/a n/a 10 0
Ring-necked Duck n/a n/a 353 0
Marsh Birds
Rail 0 1 n/a n/a
Sora 0 1 n/a n/a
Purple Gallinule 127* 14 97* 5
Common Moorhen 464* 72 360* 74
American Coot 1 0 217* 0
Other
Killdeer 13' 0 14* 0
s State listed as Species of Special Concern, state listed as Threatened, E state and
federally listed as Endangered.
From Chi-square goodness-of-fit tests, significantly more birds than expected along
indicated shoreline (p < 0.05).









Four species, the Rail (Rallus sp.), Sora (Porzina carolina), Least Bittern

(Ixobrychus exilis), and Black-crowned Night Heron (Nycticorax nycticorax), were found

exclusively along undeveloped shoreline during one or both seasons, but their numbers

were too small to be analyzed. Only the Anhinga (Anhinga anhinga) showed a

significant association with undeveloped shoreline, and only during the winter season (x2

= 12.71, p < 0.001).

Substrate Use

Substrate analyses offered a fine-scale measure of habitat use by depicting the

immediate habitat element or structure that each bird was using. The most commonly

used substrates for each guild are reported here. Forty-nine percent of all marsh birds

were found in either low-emergent or floating-leafed vegetation over both seasons. Fifty

percent of wading birds were found either in the shallows or along open shore in the

summer, whereas only 11% were found in these substrates in the winter, 35% were found

on piers (vs. 9% in the summer), and 19% were found in low-emergent vegetation (vs.

3% in the summer). Forty-four percent of diving birds were found on piers and pylons

over both seasons, while 34% were found in trees or on logs. This division was strongly

related to shoreline development. Along the developed shoreline, 63% of diving birds

were actually found on piers and pylons, whereas 83% of this guild was found in trees or

on logs along undeveloped shoreline. Forty percent of all ducks were observed in

floating-leafed vegetation over both seasons.

Littoral Zone Habitat Association

Table 2-4 summarizes the significant associations found between waterbird guilds

and littoral zone habitat elements for summer and winter surveys.









Summer

Tall emergent. Summer analyses revealed that tall emergent vegetation was

negatively associated with bird presence for marsh birds, wading birds, and ducks (all

tests: x > 31.39, p < 0.0001). Though not meeting the requirements for overall

significance, diving birds may have also been negatively associated with this habitat

element, showing a significant negative association on two lakes (both tests: > 11.31,

p < 0.001). Species analyses revealed that the Least Bittern was the only species that was

positively associated with tall emergent vegetation (2 = 4.12, p = 0.04). No other habitat

element in either the littoral or shoreline zones showed such a broadly consistent pattern

of association. Examination of overall bird presence across the five habitat densities

showed that 81% of all birds using areas with tall emergent vegetation were found in

areas with less than 50% coverage of this habitat element.

Floating leaf. Lakes Buckeye and Deer were the only lakes with enough floating-

leaf coverage (>5%) to conduct habitat analyses for this element. A strong positive

association was observed between ducks and floating-leafed vegetation on these lakes

(both tests: x > 9.73, p < 0.002). Ducks used this habitat at all coverage densities, but

40% were found in areas with greater than 75% floating-leaf coverage. Marsh birds,

wading birds, and diving birds did not appear to respond to floating-leaf vegetation.

Low emergent. Results for low emergent vegetation were inconclusive for all

waterbird guilds. Though individual lakes and guilds showed significant results,

consistent patterns were never observed on more than two lakes.

Winter

Tall emergent. Winter analyses revealed a negative association between tall

emergent vegetation and wading bird presence on all lakes (all tests: x2 > 8.61, p <









0.003). Again, 81% of wading birds associating with tall emergent vegetation were

found in areas with less than 50% coverage of this habitat element. Though not meeting

the requirements for overall significance, marsh birds and diving birds may have also

been negatively associated with this habitat, showing significant negative associations on

two lakes each (all tests: x2 > 9.79, p < 0.002). Separate species analysis showed that the

Green Heron showed a positive association with tall emergent vegetation (x = 7.06, p =

0.008).

Floating leaf. Ducks continued to show a positive association with floating-leafed

vegetation in the winter on the two lakes with this habitat element (both tests: x2 > 14.08,

p < 0.001). Eighty percent of the ducks that associated with floating-leafed vegetation

were found in areas of greater than 50% coverage. Both wading birds and diving birds

were negatively associated with floating-leafed vegetation on the two lakes (all tests: x2 >

11.85, p < 0.001). The avoidance threshold for these birds in this habitat appeared to be

50% coverage, with only 11% of the birds found in the greater coverage densities. Marsh

birds were not associated with floating-leafed vegetation.

Low emergent. Wading bird presence was positively associated with the presence

of low emergent vegetation on three of the four lakes (all tests: x2 > 15.91, p < 0.0001).

Of the waders associating with this habitat element, 40% were found in areas with less

than 25% low emergent coverage. Marsh birds, diving birds, and ducks showed no

consistent patterns of association with this habitat element.

Onshore Habitat Association

Table 2-5 summarizes the significant and marginally significant associations found

between waterbirds guilds and onshore habitat elements.









Table 2-4. Waterbird guild associations with littoral habitat elements for summer 2001
(S), and winter 2001/2002 (W).
Guild Tall Emergents Low Emergents Floating-Leafed
S W S W S W
Marsh --
Waders -- ++
Divers
Ducks -- ++ ++
-- significant negative association on at least three lakes (p < 0.05)
++ significant positive association on at least three lakes (p < 0.05)

Summer

Open shore. Open shore showed significant positive association on three of the

four lakes with both ducks and marsh birds (all tests: > 17.71, p < 0.0001). Meta-

analysis also showed a significant positive association between open shore and wading

birds (x = 40.3, p < 0.0001). Diving birds showed no consistent pattern of association

with this element. As defined (moist soil or sand), open shore was rarely found

extending beyond the first meter of shoreline. Coverage densities therefore rarely

exceeded 0 25%, which precluded interpretation of threshold tolerances across the

habitat density gradient. However, birds were found with this habitat at each of the

coverage densities where they were available.

Lawn. Summer trends for lawn were inconsistent for all guilds but diving birds,

which showed an overall significant pattern of negative association on three lakes (meta-

analysis: x = 37.4, p < 0.0001). Diving birds were found at equally low numbers across

all coverage densities for this habitat element.

Understory, shrub, and canopy. Neither understory nor shrub habitat showed

consistent patterns of association with any of the bird guilds during the summer season.

Canopy habitat showed an overall significant pattern of positive association with diving









birds on all lakes (meta-analysis: 2 = 44.5, p < 0.0001). This association was relatively

even across all canopy densities.

Winter

Open shore. Due to higher water levels in the winter, open shoreline was greatly

reduced on all lakes compared to summer availability. Open shoreline became so limited

on Lakes Buckeye and Deer (<5%) that these lakes were removed from analyses.

Wading birds were positively associated with this habitat on the two remaining lakes

(both tests: 2 > 4.21, p < 0.04). Open shore never occurred on more than 25% of total

onshore habitat in the winter, which prevented interpretation of threshold tolerances

across the habitat density gradient. None of the other guilds showed a consistent

preference for or against this habitat.

Lawn. Lawn showed a strong overall pattern of positive association with marsh

birds, wading birds, and diving birds in the winter. Both marsh birds and wading birds

showed significant positive associations (all tests: x2 > 18.4, p < 0.0001) with this habitat.

Diving birds showed a significantly positive association on two lakes and a positive trend

on a third lake (meta-analysis: 2 = 35.51, p < 0.0001). Fifty-eight percent of the birds

associating with lawn habitat were found in areas of greater than 75% lawn coverage.

Understory. Understory showed a significantly positive association with diving

birds (all tests: 2 > 4.69, p < 0.03), and an overall pattern of significant positive

association with marsh birds (meta-analysis: 2 = 17.98, p < 0.01). No minimum or

maximum habitat density thresholds were apparent.

Shrub. Shrub was negatively associated with wading birds, which significantly

avoided this kind of onshore habitat on three lakes (all tests: x > 6.28, p < 0.012). An









avoidance threshold was not apparent, with birds found at equally low numbers across all

levels of shrub coverage.

Canopy. Marsh birds and diving birds showed a significant positive association

with canopy coverage (all tests: x > 7.77, p < 0.005). Wading birds showed a significant

positive association on two lakes and a positive trend on a third lake (meta-analysis: x2 >

42.05, p < 0.0001). No minimum or maximum habitat density thresholds were apparent,

with all guilds found in all levels of canopy coverage.

Table 2-5. Waterbird associations with onshore habitat elements by guild for summer
2001 (S), and winter 2001/2002 (W).
Guild Open Shore Lawn Understory Shrub Canopy
S W S W S W S W S W
Marsh ++ ++ + ++
Waders + ++ ++ -- +
Divers + ++ + ++
Ducks ++
++/-- indicates significant positive/negative associations on at least three lakes (p < 0.05).
+/- indicates significant overall positive/negative association (p < 0.05) based on meta-
analysis of trends (p < 0.2) on at least three lakes.

Independence of Significant Habitat Elements

Summer

Ducks showed a significant positive association with both floating-leafed

vegetation and open shore in the summer. Looking only at the portions of the lakes

covered with one or both of these habitat elements, floating-leafed vegetation overlapped

with open shore 12% of the time. Twenty-five percent of all ducks sighted in these areas

were found where these habitats overlapped. Thus, it did not appear that ducks were

strongly responding to the combination of these elements. No other guilds showed

multiple significant habitat associations in the summer.









Winter

Marsh birds. Marsh birds showed a significant positive association with lawn,

understory, and canopy in the winter. The most substantial differences between habitat

availability and marsh bird distribution occurred in lawn/understory habitat and

lawn/understory/canopy habitat. Lawn/understory habitat (without canopy) occurred on

only 1% of the shoreline, while 12% of marsh birds were sighted in areas with both

elements present. Similarly, lawn/understory/canopy habitat occurred on only 2% of

significant habitat shoreline, while 25% of marsh birds were sighted with all three habitat

elements present. Additionally, canopy habitat occurred in the presence of either lawn or

understory along 98% of the shorelines having one or more of these elements. Thus,

although there may be some selection of marsh birds for lawn/understory/canopy

combinations, it could not be determined whether these combinations played a role in

attracting marsh birds.

Wading birds. Wading birds showed a significant positive association with low

emergent vegetation, open shore, and lawn in the winter. Overall, no combination of

these habitat elements clearly attracted wading birds. The most substantial difference

between overlapped habitat availability and wading bird distribution occurred in low

emergent/lawn habitat, where 40% of wading birds were sighted in areas where both low

emergents and lawn were present. Actual overlap of these two habitat elements was only

29%.

Wading birds showed a negative association with tall emergent vegetation,

floating-leaf vegetation, and shrubs in the winter. Wading birds appeared to show the

strongest negative association with tall emergent vegetation. Looking only at spatial

distribution, tall emergent vegetation overlapped with at least one other element 83% of









the time. Therefore it could not be determined whether wading birds were avoiding this

habitat element alone or a combination of these elements.

Diving birds. Diving birds showed a significant positive association with

understory and canopy in the winter. Fifty-seven percent of diving birds were sighted in

areas where both habitat elements were present. Understory and canopy habitat

overlapped along 26% of the shoreline. Thus, it could not be concluded that diving birds

were selecting for the combination of these habitat elements.

Discussion

Seasonal Species Composition

Thirty-five species were observed using these lakes on a regular basis over one or

both seasons, including nine state or federally listed species. In the summer, marsh birds,

wading birds, and ducks were all observed breeding on these lakes, with numerous

fledgling marsh birds and ducks observed (Chapt. 3). In the winter, almost one third of

the species that used these lakes were winter migrants. Several of these species, like the

Double-crested Cormorant, Ring-billed Gull, and Ring-necked Duck were observed in

large foraging flocks using these lakes for brief periods. Other species, such as the Pied-

billed Grebe, American Coot, and Belted Kingfisher established themselves more

permanently on these lakes for the winter season. These findings confirm that urban

lakes can sustain diverse waterbird communities during both the breeding and winter

seasons, and apparently provide functional habitat for a variety of seasonal needs.

The similarity in community composition with Hoyer & Canfield's study (1994)

and Roth's study (1991) indicates that urban lakes may not just have relatively stable

avian communities from one season to the next, but also from year to year. This long-

term stability may be the result of the dam-controlled water levels providing more stable









environmental conditions than are found in many natural wetlands. If so, some

waterbirds may learn to rely on urban lakes in times requiring stable water levels, such as

the nesting season.

Shoreline Development

Wading bird, marsh bird, and duck abundance were significantly greater than

expected along developed shoreline on all lakes during both the summer and winter

seasons. While some birds may have been undercounted along undeveloped shoreline in

areas of dense cattail, the degree of difference in the number of birds found along

developed and undeveloped shorelines cannot be explained by this alone. In my repeated

entries into cattail stands, I saw and heard almost no waterbirds using this habitat. Roth

(in press) had similar results in a study of neighboring lakes. Further, many of the more

conspicuous waterbirds, such as the large and mid-sized long-legged waders, are known

to avoid areas of dense tall-emergent vegetation (Smith et al. 1995, Surdick 1998, Roth in

press). That almost all of these species were found in significantly greater abundance

along developed shoreline supports these results, and suggests that the prevalence of

cattail along undeveloped shoreline may have been a primary factor in many birds

selecting developed shoreline.

Knight and Cole (1995) stated that the four primary ways in which human activities

can impact animals are through exploitation, pollution, disturbance, and habitat

modification. Hunting is not allowed on these lakes, and, given the small size of the

lakes', pollution effects are most likely evenly dispersed between developed and

undeveloped shores. This leaves disturbance and habitat modification as the two primary

activities affecting waterbird patterns. Human disturbance, in general, has been widely

shown to be detrimental to birds and other wildlife (Hockin et al. 1992, Carney &









Sydeman 1999, but see Nisbet 2000). If human disturbance was the primary activity

influencing the distribution of waterbirds around these lakes, then, given the fact that

human activity was much greater in developed areas (Chapt. 3), fewer birds than

expected should have been found along developed shorelines. That the exact opposite

was found suggests that there are considerable benefits to the modified habitat found

along developed portions of these lakes, and that under such circumstances many

waterbird species will tolerate increased levels of human disturbance.

Tolerance of human disturbance, as indicated by a bird's presence along developed

shorelines, may be a sign of habituation. Habituation has been defined as "the relatively

persistent waning of a response as a result of repeated stimulation which is not followed

by any kind of reinforcement" (Hinde 1970). Other studies that have used similar

presence/absence measures of habituation have had varying results. On separate refuge

studies, Burger (1981) found that waterbirds were significantly less likely to be present

when people were present at a site, whereas Klein at al. (1995) found that only half of the

species in their study shifted away from areas of human disturbance as disturbance levels

increased. Several authors though, have noted that it is common for waterbirds to

habituate to moderate levels of disturbance in situations where people are regularly

present but not causing any direct harm (Hockin et al. 1992, Weller 1999).

As mentioned above, the dense, cattail found along undeveloped shorelines

appeared to be unattractive to many waterbirds. Such habitat conditions may have

limited visibility, foraging opportunities, and escape routes, and increased vulnerability to

predators. If this was the case, waterbirds may not have actually been showing a

preference for developed habitat, but rather an avoidance of undeveloped habitat. When









considering this option it is important to distinguish between undeveloped habitat and

natural habitat. Though the undeveloped shorelines on these lakes had relatively

undisturbed terrestrial vegetation structure, the deep, dense stands of cattail commonly

dominating the littoral zones were most likely not part of the natural habitat originally

found on these lakes. Rather, they were a result of the artificial eutrophication that has

occurred on these lakes due to years of uncontrolled urban runoff (Gilbert 1987, Roth in

press). Reestablishing a healthy, more diverse, and structurally heterogeneous aquatic

plant community (i.e., "hemi-marsh") along the undeveloped portions of these lakes

might very well create a more favorable habitat than is currently available for the

waterbirds on these lakes.

Previous urban studies based in terrestrial habitats have documented an overall

increase in avian abundance as a select few human-commensal species prosper (Blair

1996, Savard et al. 2000). This same process appeared to occur on a seasonal basis on

the urban lakes in this study, as seen in the summer by the lower species evenness along

developed shorelines, and the fact that the overall preference for developed shoreline was

explained by just eight species. Further, more secretive species such as rails and bitterns,

often found in less disturbed wetland habitats, were rarely encountered on these lakes.

The combined implications are that certain waterbird species are adaptable enough to

benefit from aquatic urban habitats and may actively seek them out, as species like the

House Sparrow (Passer domesticus) and European Starling (Sturnus vulgaris) do in

terrestrial urban habitats. Species like the Great Blue Heron or Belted Kingfisher, both

found in this study, are highly adaptable in terms of habitat and diet (Butler 1992, Weller

1999), and this may explain why they are found in urban aquatic habitats. Other species,









such as the Anhinga or Black-crowned Night Heron, may be found in developed areas,

but at lower numbers than in less disturbed habitats. And waterbirds such as rails or

bitterns (found rarely in this study) may be entirely intolerant of development and/or

disturbance and may avoid urban environments altogether when possible, even when

portions of the environment are left undeveloped.

Dominant Habitat Elements

Tall emergent vegetation, open shore, lawn, and canopy were each associated with

the distribution (presence/absence) of multiple guilds on these lakes, and were thus

considered dominant habitat elements. Tall emergent vegetation had a negative overall

association, whereas open shore, lawn, and canopy had positive overall associations.

Littoral habitat

Tall emergent vegetation. All guilds were negatively associated with tall

emergent vegetation on at least 50% of the lakes over both seasons. No significant

positive associations were observed within a guild on any lake. Previous studies have

shown similar results, with a variety of waterbird species using dense tall-emergent

monocultures far less than expected by chance (Weller & Spatcher 1965, Weller &

Fredrickson 1974, Kaminski & Prince 1984, Collopy & Jelks 1989, Bildstein et al. 1994,

Smith et al. 1995, Surdick 1998). The most substantial stands of tall emergent vegetation

were located along undeveloped shoreline. Cattail was present along 89% of total

undeveloped shoreline, with almost 70% of it found in dense, continuous stands. These

stands were frequently found extending over 20 m from shore into water depths that were

too great for even the tallest wading birds to use. In the summer, the larger stands of

cattail became so dense that even smaller marsh birds and ducks had to struggle to

penetrate the vegetation when attempting to flee.









Given its extensive coverage on many of the lakes, cattail overlapped other habitat

elements quite frequently. In the case of wading birds, the only guild showing significant

negative associations with multiple habitat elements, tall emergent vegetation (cattail)

occurred in combination with other significant elements 83% of the time. Thus, it could

not be conclusively determined whether birds were responding primarily to cattail or

some combination of elements. However, given the abundance of previous research

showing the avoidance of tall, dense, emergent vegetation by numerous waterbird species

(see above), it seems likely that tall emergent vegetation, namely cattail, was indeed the

dominant habitat element that birds were avoiding.

Cattail was most likely present on these lakes before they were developed, but at

much lower densities (Gilbert 1987, Florida DEP 1983-1992). In this study both the

Least Bittern (summer) and the Green Heron (winter) showed a positive association with

tall emergent vegetation. Cattail, even at fairly high densities, is considered functional,

even necessary habitat for a number of waterbirds; the Least Bittern, Rail, Sora, Black-

crowned Night Heron, Common Moorhen, Purple Gallinule, Red-winged Blackbird

(Agelaiusphoeniceus) and Boat-tailed Grackle (Quiscalus major) (Terres 1991, Hoppe &

Kennamer 1986, Davis 1993, Melvin & Gibbs 1996, Gibbs et al. 1992). Eliminating

cattail from these lakes would therefore be detrimental to these species. However,

limiting cattail coverage along undeveloped shoreline may allow for greater foraging and

resting opportunities for a wider range of species.

Onshore habitat

Open shore. Open shore had a positive association with marsh birds, wading

birds, and ducks during the summer. The creation of much of this habitat, defined as

moist soil or sand, was a result of the extreme drought that Florida experienced in 2001.









Though ephemeral, open shore habitat appeared to serve as a valuable foraging area for

these birds (pers. obs.). Dropping water levels exposed new foraging habitat that was not

available during higher water conditions. As such, many dabbling and probing birds and

small waders could take advantage of the shallow waters and exposed substrate. Of the

eight species that were associated with developed shoreline in the summer, seven fit this

description: Snowy Egret, Tricolored Heron, White Ibis, Wood duck, Mallard, Purple

Gallinule, and Common Moorhen. All of these species are known to frequently forage in

areas of relatively open shallow waters and sparse vegetation (Hancock & Kushlan 1984,

Weller 1999). Given that only 10% of open shore was found along undeveloped

shoreline, this habitat element may explain much of the overall preference for developed

shoreline that was observed in the summer season.

Lawn. Lawn was positively correlated with marsh birds, wading birds, and diving

birds during the winter. This habitat, an obvious indicator of development, by itself

probably offered little functional value to most birds other than White Ibis, Cattle Egrets

and Common moorhens, which frequented lawns for foraging. Substrate analysis showed

that only five percent of all birds found along developed shoreline were directly found in

lawns. Most birds were likely responding to other habitat elements of the developed

shoreline located in the vicinity of lawns. For example, 40% of wading birds observed in

areas where lawn was present were also found with low emergent vegetation. In general,

areas with significant lawn coverage had reduced understory and shrub layers, and

patchier emergent vegetation. Reduced onshore and littoral vegetation structure may

have afforded birds better visibility and movement, allowing for easier detection and

avoidance of predators or approaching humans. This idea is supported by previous









studies that have suggested that many wading bird species are more vigilant and more

easily disturbed in areas of dense vegetation (Smith et al. 1995, Safran et al. 2000). The

patchier aquatic vegetation structure may have also provided better foraging habitat.

Bildstein et al. (1994) found that patchy littoral vegetation structure allowed wading birds

to feed in relatively open water while taking advantage of high fish densities in adjacent

vegetated areas. Though no significant differences were observed in the proportion of

birds foraging along developed and undeveloped shorelines (Chapt. 3), other measures,

such as foraging times, prey selection, or strike/capture ratios, may have better shown the

value of this habitat for foraging birds.

Other components of developed shoreline that were associated with lawns included

human-made structures such as piers, pylons, and boats. Examination of the substrates

on which birds were observed revealed that 25% of all birds using developed shoreline

were found on such structures. Many species used these structures for resting, probably

due to the unobstructed access to water and the ease of vigilance. Diving birds offer the

strongest example in that 63% of all diving birds along developed shoreline were found

on piers or pylons. Birds such as the Tricolored Heron, Snowy Egret, and particularly the

Belted Kingfisher were observed actively foraging from these perches.

Canopy. Canopy was positively associated with marsh birds and wading birds in

the winter and diving birds over both seasons. Though canopy was found in greater

densities along undeveloped shorelines, it was also prevalent along developed shorelines,

and therefore cannot be considered an indicator of undeveloped shoreline. It can be seen

as an increase in structural habitat complexity, providing birds with an added vertical

layer that they could use as a refuge from disturbance. Several species of wading birds









were observed resting in canopy. Eighty-three percent of all diving birds resting along

undeveloped shoreline were found resting in trees or on logs. Casual observation

suggests that these individuals may have shown lower levels of alert/flee response than

birds resting closer to the water. In addition, canopy may have provided shade refuge in

the summer during the heat of the day, and thermal refuge in the winter, protecting birds

from winter winds. In both seasons, it may have also provided sunny locations for

basking birds. This is especially important to Anhingas, which rely on basking to dry

their feathers and maintain body temperature (Frederick & Siegel-Causey 2000).

The association between marsh birds and canopy was most likely indirect. Besides

canopy, marsh birds were also significantly associated with lawn and understory.

Though marsh birds showed no clear association with any combination of these elements,

the ecology of this guild suggests that understory may have been the primary habitat

element determining their distribution. In general, many members of the rail family show

a strong preference for relatively dense understory and generally weedy conditions

(Terres 1991, Elphick et al 2001).

Guild Responses to Other Habitat Elements

Marsh birds

Marsh birds also showed a positive association with understory in the winter. As

mentioned above, this finding was expected given the general ecology of this guild.

However, though lawn, understory, and canopy overlapped along only 2% of total

significant habitat shoreline, 25% of winter marsh birds were found in this area. Thus,

though understory was most likely the dominant habitat element, it can not be ruled out

that marsh birds were selecting for a combination of these elements.









Wading birds

Whereas wading birds were only associated with tall emergent vegetation and open

shoreline in the summer, all habitat elements but understory were found to be significant

in the winter. A positive association was observed with low emergents, lawn, and

canopy, whereas a negative association was observed with floating-leafed vegetation and

shrubs. Such variation between seasons suggests that wading bird habitat preferences

may be more seasonally dependent than other waterbird guilds, and that other habitat

elements, such as water depth, or groups of habitat elements, may be associated with this

guild's within-lake habitat choices.

Summer results suggest that wading birds may have been primarily responding to a

combination of water depth and vegetation structure. Water depth has been frequently

cited as a significant factor determining the distribution of wading birds (Hancock &

Kushlan 1984, Weller 1999, Bancroft et al. 2002). In the summer, this guild was found

in greatest abundance in areas of open shore. Associated littoral zones in these areas

were typically dominated by open water or sparse emergent vegetation. While all lakes

had shallow water margins, the predominance of emergent vegetation often prevented

waders access to these areas. The lack of emergent vegetation along open shore areas,

when water levels dropped, allowed wading birds to utilize these shallower waters,

providing optimal foraging conditions for many of these birds (Hancock & Kushlan

1984, Breininger & Smith 1990, Bildstein et al. 1994).

In the winter, in the absence of extreme drought conditions, shallow water areas

became more limited. Wading birds, in turn, may have become more generalized in their

habitat selection. Given that tall emergent vegetation continued to be used less than

expected, and open shore, even at just 28% of summer availability, continued to be used









more than expected, these habitats may have continued to be the primary elements to

which wading birds were responding. However, the abundance of significant responses

to other habitat elements suggests that this guild may have been responding to

combinations rather than just individual habitat elements. One such example would be

low emergent/lawn habitat. Though this habitat made up 29% of significant habitat

shoreline, 40% of wading birds occurring with one or more significant habitat element

were found in low emergent/lawn habitat.

Diving birds

Diving birds showed a negative association with lawn in the summer and a positive

association with understory in the winter. The negative association with lawn in the

summer is puzzling and cannot be readily explained by the ecology of these birds or by

lawn's association with other habitat elements. The positive association with understory

in the summer may be the result of diving birds selecting areas where understory and

canopy overlapped. Understory/canopy habitat made up 26% of significant habitat

shoreline, yet 57% of diving birds were found in this area of habitat overlap. Given the

ecology of these birds and the results of this study showing diving birds frequently

resting in the canopy, it seems likely that they were selecting either for canopy or

canopy/understory, rather than just understory alone.

Ducks

Besides showing a negative association with tall emergent vegetation and a positive

association with open shore in the summer, ducks showed a positive association only

with floating-leafed vegetation during both seasons. Although only two lakes had

enough floating-leaf habitat for analysis, both lakes showed strong associations (p <

0.01). On both lakes the primary floating-leafed species was spatterdock (Nuphar









luteum), a type of water lily whose leaves stand above the water. Structurally acting as a

low emergent species, and occurring at high densities, this species offered exceptional

cover for waterfowl, particularly Wood Ducks. Tarver et al. (1978) also noted that

Spatterdock seeds may be a valuable secondary food source for waterfowl in northern

Florida. Floating-leafed vegetation overlapped with open shore along 12% of significant

habitat shoreline, while 25% of ducks were found where these habitats overlapped.

Though not conclusive, it could not be ruled out that ducks may have also been

responding to this combination of habitat elements.

Management and Future Research

More comprehensive and long-term research is strongly recommended for urban

lake habitats. Further research needs to focus on whether these lakes are truly providing

valuable habitat to these birds or whether they are actually acting as biological sinks.

Though most birds in this study appeared to be healthy, and both marsh birds and ducks

produced numerous young, the long-term effects of human impact, such as disturbance

and pollution, need to be examined.

Developed shoreline around the lakes in this study clearly provided useable habitat

for a variety of waterbird species, and may have actually been selected for by some

species. However, the fact that over both seasons all guilds were negatively associated

with tall emergent vegetation, which was predominantly found along undeveloped

shoreline, suggests that it was avoidance of this habitat element that was responsible for

the significantly greater proportion of birds along developed shoreline. In smaller,

discrete stands, when interspersed with other plant species and patches of open water,

cattail may provide habitat for a wider range of species. Weller (1999) has noted that a

wide range of wetland birds prefer such "hemi-marsh" conditions. Such interspersion









would increase structural habitat complexity and open up shallow areas closer to shore,

creating preferred foraging habitat for many wading and dabbling birds. More detailed

habitat studies should be conducted to determine optimal cattail densities for different

avian species or guilds, and whether such management would be a feasible option.

Currently, Florida's Bureau of Invasive Plant Management has no active management

plans for this plant species.

Several onshore habitat elements were significantly associated with the distribution

of waterbird guilds, indicating that terrestrial habitat may play a role in habitat selection

for many of these birds. Specifically, open shore, lawn, and canopy appeared to be

associated with multiple guilds. As shorelines continue to be developed, much of the

natural habitat is being altered or removed both by developers and property owners.

Florida Law (369.20(7), Florida Statutes) states that "no person or public agency shall

control, eradicate, remove, or otherwise alter any aquatic weeds or plants in waters of the

state unless a permit for such activity has been issued by the department." There are

currently no such laws protecting onshore habitat. Future studies should specifically test

the degree of importance of terrestrial vegetation to waterbirds in order to determine

whether this habitat needs greater protection.

Until further research is conducted, managers should consider the ecology of these

birds when prioritizing habitats for protection. Since very few birds were observed

nesting on these lakes (Chapt. 3), preliminary management should be based on the

foraging and resting behavior of these birds. Wading birds, in general, prefer shallow,

relatively open littoral zones. This study also showed potential associations with open

shore and lawn. These conditions are currently often available along developed









shorelines. The fact that many of the wading bird species using these lakes were found

significantly more often along developed shorelines suggests that little further

management may be required for this guild. Exceptions such as the Least Bittern and

Green Heron should be noted though. These species often show a preference for dense,

tall emergent vegetation, which was found in greater abundance along undeveloped

shorelines. Management for these species should be focused there.

Marsh birds and ducks are often found in open water interspersed with mixed

emergent and floating-leafed vegetation. In this study, both guilds may have selected for

open shore in the summer, while marsh birds may have also been associated with lawn

and understory in the winter. Many of these conditions are again being met along areas

of developed shoreline, but in limited extent. Management should consider increasing

low emergent and floating-leafed vegetation in developed areas lacking frequent boat

traffic if increasing marsh bird and duck abundance is desired. Though nesting was not

observed for these guilds, it should be noted that many marsh birds do require dense tall-

emergent vegetation in the breeding season. Portions of this habitat should therefore be

protected along undeveloped shoreline.

Many of the diving birds require larger areas of deeper open water for foraging, and

canopy or other perching substrate for resting. Open water is currently managed for to

promote lake use for fishing and other recreation. Common diving birds on these lakes,

such as Anhingas and Double-crested Cormorants, often prefer resting areas immediately

over the water. Developed shorelines, with their abundance of docks and boat houses,

are thus well suited to these birds. However, diving birds were the one guild that did not

have greater than expected numbers along developed shoreline. Management might









therefore focus on protecting areas of undeveloped shoreline, as woody shoreline is more

readily available there.

This project, being exploratory in nature, attempted to determine whether dominant

habitat elements existed that independently explained waterbird distributions. Several

elements, both aquatic and terrestrial, did indeed appear to be significantly more

influential. Though tests for independence were inconclusive, many of the habitat

elements selected in this study may have been correlated with one another, both spatially

and perhaps functionally. Future studies would do well to examine waterbird/habitat

associations at broader scales (e.g., Hostetler and Holling 2000) to investigate whether

birds are responding to suites of habitat elements that comprise general macro-habitats

around these lakes.

Duda (1987) found that 88% of Floridians enjoyed having birds near their homes

and frequently engaged in activities to benefit them. Lake property owners should be

informed of the abundance of waterbirds using the developed shorelines on their lakes.

Many of these individuals might be very receptive to an education program on the value

of shoreline habitat and the ways that homeowners can manage their property to meet

their own needs as well as the needs of wildlife. Even with current conditions, many

lakeshore residents are inadvertently creating habitat conditions favorable to many

waterbirds by managing their properties for more heterogeneous littoral and onshore

habitat in order to have direct lake access. State laws allow residents to clear up to 50%,

or 50 ft (whichever is less) of the aquatic vegetation along their shoreline to allow for

boating and other recreational access. With education, residents can learn to manage









their own properties to benefit both themselves and wildlife, and may even be able to

help managers improve waterbird habitat along undeveloped shorelines.

Soliciting public participation in future research, monitoring, and management

efforts (i.e. citizen science) would also increase public awareness and generate a greater

sense of stewardship on lands that may not have been previously thought to have much

wildlife value. Citizen science can also generate valuable data sets supporting long-term

monitoring efforts (Hoyer et al. 2001). The popularity of such public involvement is

demonstrated by Florida LAKEWATCH, a lake-monitoring program that relies on a pool

of over 1,800 trained volunteers to monitor water quality on approximately 1,200 lakes

around the state. Beginning in 2001, Lakewatch incorporated aquatic-bird surveys into

its monitoring program. This new wildlife component is being readily adopted by many

of the volunteers and holds great promise for future research and management efforts on

many of Florida's lakes.














CHAPTER 3
AVIAN BEHAVIORAL RESPONSES TO SHORELINE DEVELOPMENT

Introduction

The process of urbanization may affect birds both through changes in ecosystem

processes, habitat structure, and food supply, and through changes in predation pressure,

competition, and disease (Marzluff 1997). Avian responses can take two forms: changes

in behavioral patterns, and changes in community composition. The majority of urban

bird studies to date have examined the direct effects of development and habitat alteration

on community composition by focusing on changes in avian abundance, species

diversity, richness, and evenness (e.g. DeGraaf & Wentworth 1981, Tilghman 1987, Blair

1996, Clergeau et al. 1998, Jokimaki 1999, Rottenborn 1999). Very few studies have

attempted to address how avian behavior is affected by development (As examples, see:

Andersson et al. 1980, Galeotti et al. 1991, Gelbach 1994, Fernandez-Juricic et al. 2001).

Studies examining changes in avian behavioral patterns have focused primarily on

the effects of human activity (e.g., Hockin et al. 1992, Knight & Gutzwiller 1995,

Jozkowicz & Gorska-Klek 1996). Human disturbance has many negative impacts to

wildlife and may ultimately reduce wildlife densities and diversity at both local and

regional scales (Boyle & Samson 1985, Cole & Knight 1990). On the behavioral level,

wildlife responses to disturbance may include attraction, avoidance, or habituation

(Knight & Cole 1991). Attraction behavior, such as increased foraging and reduced

wariness, can result from human actions like supplemental feeding (Knight & Temple

1995). Hunting typically results in avoidance behaviors such as increased wariness,









altered foraging patterns, and reduced nest defense (Kenney & Knight 1992, Knight &

Cole 1995). Increased flight times or increased aggression towards humans are other

avoidance responses that may result from excessively close or frequent human

disturbance (Knight & Temple 1986, Kahl 1991). Habituation, on the other hand, may

occur in response to human disturbance when that disturbance is not associated with

either a positive or negative reward (Eibl-Eibesfeldt 1970). In this case, wildlife species

appear to show no signs of altered behavior in response to human presence.

A wide range of avoidance behaviors have been documented for waterbirds in

response to recreationists or researchers in natural areas. Examples include reduced

foraging and resting periods (Owens 1977, Kaiser & Fritzell 1984, Burger and Gochfeld

1991, Skagen et al. 1991), increased nest abandonment and egg loss (Kury & Gochfeld

1975, Tremblay & Ellison 1979, Titus & van Druff 1981), discouragement of late-nesting

pairs from breeding (Ellison & Cleary 1978, Tremblay & Ellison 1979), and disruption of

pair bonds (Tindle 1979) and parent-offspring bonds (Oldfield 1988). Other studies have

shown that human disturbance can increase vigilance (Burger and Gochfeld 1991),

flushing (Vos et al. 1985), flight times (Kahl 1991, Korschgen et al. 1985), and energy

expenditure by birds and reduce their overall energy intake (Belanger & Bedard 1990).

Conversely, other studies have shown that in cases where humans are regularly

present without posing an immediate threat of harm, waterbirds appear to habituate to

some forms of disturbance (Hockin et al. 1992, Weller 1999). Some examples include

New Zealand Dotterels (Charadrius obscurus) allowing closer human approach on high-

use beaches (Lord et al. 2001); Great Crested Grebes (Podiceps cristatus) showing

reduced flush distances in sites frequently visited by humans (Keller 1989); nesting









Ospreys (Pandion haliaetus) showing greater levels of habituation in areas of high human

activity than in more remote sites (Swenson 1979, Poole 1981); Great Blue Herons

(Ardea herodias) habituating to the activities of fishermen boating past a heronry (Vos et

al. 1985); and Greylag Geese (Anser anser) habituating to people walking past as long as

the people remained on paths (Kuhl 1979).

The objective of this portion of the study was to determine the effects of urbanized

lakes on the behavioral patterns ofwaterbirds. The specific questions to be answered

were as follows:

1. Are primary waterbird behaviors different between developed and undeveloped
shoreline?

2. Which waterbirds appear most sensitive to human disturbance?

Question 1 addresses the general lack of knowledge about avian behavior on urban

lakes. Many studies have compared waterbird behavior in areas of high human use with

areas of low human use (e.g. Burger 1981, Keller 1989, Jozkowicz 1996), but have either

been based in natural habitats such as refuges or areas where disturbed and undisturbed

sites were well apart from one another.

Question 2 focuses specifically on avoidance behavior and sensitivity to human

disturbance by examining which guilds appear most vulnerable to disturbance.

Numerous studies have shown that sensitivity to human disturbance can vary

considerably between species (e.g. Burger 1981, Bratton 1990, Klein 1993, Klein et al.

1995), with several studies showing greater sensitivity specifically in winter migrants

(van der Zande et al. 1980, Klein et al. 1995). Both of these hypotheses are examined.









By understanding how waterbirds behave in urban settings and how sensitive they

are to human disturbance, managers will be better equipped to balance the needs of

humans and wildlife on urban lakes.

Methods

Behavioral surveys of waterbirds were conducted in the summer from June 7-

August 1 of 2001 and in the winter from December 8- February 6 of 2001/2002. The

term waterbirdd' referred to species in the orders Gaviiformes, Podicipediformes,

Pelecaniformes, Ciconiiformes, Anseriformes, Falconiformes, Gruiformes,

Charadriiformes, or Coraciiformes observed on or feeding from lacustrine habitats. Each

of the four lakes were surveyed a total of eight times each season by driving at minimum-

wake speed around the lake perimeters (20 30 m from shore) in a small motorized

canoe. Both the order of lakes and the direction of travel were alternated for each day of

surveying. Surveys were conducted within the first five hours after sunrise on mornings

with little to no rain and winds less than 24 km/hr.

Lake shoreline was categorized as either developed or undeveloped based on DEP

land cover classifications for each lake. Classifications were modified during preliminary

surveys by basing the categorization only on the land cover within the first 20 m of

shoreline extending away from the water on each lake. For the purposes of this study,

developed shoreline referred to any continuous expanse of shoreline greater than 100 m,

parallel to the edge of the lake, that had a minimum of 50% noticeable, long-term habitat

alteration, defined as cleared land, lawns, landscaping, buildings, and roads.

Undeveloped shoreline was defined as any continuous stretch of shoreline greater than

100 m, parallel to the edge of the lake, with greater than 50% intact natural habitat, and

little to no sign of regular human use. Developed and undeveloped areas were separated









by a buffer of 40 m to eliminate the potential effects of converging habitats. Birds

recorded in these border areas were not included in analyses. In addition, behavioral

analyses were not conducted on birds if their origin or destination were not observed, or

if they were observed greater than 30 m offshore or 10 m onshore.

Upon sighting a bird along either developed or undeveloped shoreline, I recorded

the first behavior that was observed. During preliminary surveys I found five primary

behaviors: active/swimming, alert/fleeing, foraging, resting, and tending young.

Active/swimming birds included those birds that were actively moving but not foraging

or showing obvious signs of distress. Birds were classified as alert/fleeing if their

attention appeared focused on the boat or another nearby disturbance, or if they vacated

the location in which they were observed in response to the boat's approach. Foraging

behavior was recorded for any bird monitoring the water, stalking, actively gleaning, or

consuming prey. Birds were classified as tending young any time they were observed

building or sitting on a nest, or in the presence of young, regardless of any other

behaviors in which they may have been engaged. This behavior was only observed in the

summer. Over the course of the two seasons, these five behaviors made up 89% of all

recorded observations. Nine other behaviors were also recorded over the course of both

seasons (e.g. aggression, calling, preening, etc.), but made up less than 12% of all

observations. Because of the dominance of the five primary behaviors, I chose to restrict

analyses to these behaviors only.

Disturbances other than the survey boat, such as anglers, jet skiers, water skiers, or

people feeding birds, were also noted. In addition, several random resident contacts were









made in order to estimate the use of the numerous Wood Duck (Aix sponsa) nest boxes

observed on these lakes.

Analyses

Waterbirds were grouped into guilds for analyses. Guilds were based on foraging

behavior and habitat use, and included wading birds (Ardeidae, Threskiornithidae,

Ciconiidae, Recurvirostridae), marsh birds (Rallidae), surface and aerial diving birds

(Podicipedidae, Phalacrocoracidae, Anhingidae, Accipitridae, Laridae), and ducks

(Anatidae). Analyses were conducted for the four most commonly observed behaviors

within each season. For a guild to be included in analyses for a given lake, at least 25

birds had to be observed for that lake during a season.

Shoreline Development

I examined the effect of shoreline development on guild behavior using

contingency Chi-square tests (a = 0.05, df = 1), by comparing the proportions of birds

engaged in each behavior along developed and undeveloped shoreline. Expected

proportions were based on the number of birds exhibiting a given behavior versus the

number of birds exhibiting any other behavior, including the four focal behaviors

observed each season and all other behaviors that were observed. Separate analyses were

run for each lake on the four focal behaviors observed each season.

Disturbance Sensitivity

Measures of waterbird sensitivity to disturbance were based on alert/flee behavior.

I tested whether the number of birds exhibiting alert/flee behavior differed significantly

between seasons and between guilds. I used contingency Chi-square tests (a = 0.05, df =

1) for comparisons between seasons, and 2x4 contingency Chi-square tests (a = 0.05, df

= 3) for comparisons between guilds. Expected proportions were based on the number of









birds exhibiting alert/flee behavior versus the number of birds exhibiting any other

behavior. Lakes were tested individually for both tests, with developed and undeveloped

shorelines combined. Summer and winter data were combined for comparisons between

guilds.

Using contingency Chi-square tests (a = 0.05, df = 1), I also tested whether winter

migrant species showed greater alert/flee behavior than resident species of the same

guilds. Analyses were run on winter data for all lakes combined due to low numbers on

individual lakes.

Determining Overall Significance

For all analyses conducted on a lake-by-lake basis, behavioral patterns were

considered to have overall significance where significant same-direction associations (p <

0.05) were observed on at least three lakes without a significant contradictory finding on

the fourth lake. In a season, if a guild was only observed with enough frequency for

analysis on two lakes, then behavioral patterns were considered to have overall

significance if both of the lakes had significant same-direction associations (p < 0.05). In

cases where results were suggestive of an overall pattern on three or more lakes (p < 0.2

for each lake), but not necessarily significant, and a contradictory finding wasn't found

on the fourth lake, probabilities (for lakes indicating a pattern) were then combined for

meta-analysis (Fisher 1958). A two-tailed Fisher's exact test (a = 0.05) was applied to all

analyses where expected values dropped below five.

Results

Seasonal Behavioral Observations

In the summer, a total of 2817 behavioral observations were recorded for the four

study guilds (Table 3-1). The four most common behaviors were alert/fleeing, foraging,









resting, and tending young. Resting was the most commonly recorded behavior,

accounting for 32% of all observations. Foraging birds accounted for 26% of behavioral

observations, followed by alert/fleeing birds (18%), and birds tending young (10%). All

other behaviors accounted for 14% of recorded observations.

In the winter, 2438 behavioral observations were recorded for the four focal guilds

(Table 3-1). Alert/fleeing behavior was most commonly recorded, accounting for 33% of

all observations. Resting behavior was observed in 26% of all birds, followed by

foraging (23%), and active/swimming (12%).

Table 3-1. Percent of waterbird guilds engaged in focal behaviors during summer (S)
2001 and winter (W) 2001/2002.
Alert/ w/ Active/
Flee Forage Rest Young Swim
Guild n (%) (%) (%) (%) (%)
S W S W S W S W S W
Waders 662 530 10 10 54 35 26 5 1 2
Marsh 673 723 5 19 46 46 6 5 23 26
Divers 299 361 11 23 2 68 83 68 0 6
Ducks 1183 671 31 67 4 2 38 8 9 9
n = total number of birds observed each season

Behavioral Associations with Shoreline Development

Alert/flee behavior

Alert/flee behavior showed an overall pattern of significant association with

undeveloped shoreline for ducks in the summer (Table 3-2) and wading birds in the

winter (Table 3-3). Across all lakes in the summer, 72% (n=89) of the ducks found along

undeveloped shoreline displayed alert/flee behavior. On the two lakes where ducks were

found with enough frequency for analysis, they displayed alert/flee behavior significantly

more along undeveloped shoreline (both tests: x > 28.98, p < 0.0001). Thirty-three

percent of all wading birds found along undeveloped shoreline in the winter displayed









alert/flee behavior (n=54). Wading birds displayed this behavior significantly more along

undeveloped shoreline on three of the four lakes (all tests: > 8.7, p < 0.012).

Possible trends were observed in the summer for wading birds, marsh birds, and

diving birds (Table 3-2). Both wading birds and marsh birds showed increased alert/flee

behavior along undeveloped shoreline on two lakes (all tests: x > 2.74, p < 0.12),

whereas diving birds showed increased alert/flee behavior along developed shoreline on

two lakes (both tests: x > 2.59, p < 0.14).

Foraging behavior

Fifty-eight percent of the wading birds (n=512) observed along developed shoreline

in the summer were engaged in foraging behavior (Table 3-2). Foraging wading birds

showed a significant overall preference for developed shoreline on three lakes (meta-

analysis: x = 27.07, p < 0.001). The same preference was observed in the winter, but

only on two lakes, and therefore lacked overall significance. Winter marsh birds

displayed a significant overall preference for foraging along undeveloped shoreline on

three lakes (meta-analysis: x = 13.78, p = 0.03). Overall, 51% of wintering marsh birds

(n=74) found along undeveloped shoreline were engaged in foraging behavior.

Resting behavior

In the summer, on the two lakes with sufficient duck numbers, ducks showed a

significant preference for resting along developed shoreline (Table 3-2; both tests, x >

11.35, p < 0.001). Overall, 40% of the ducks (n=1094) observed along developed

shoreline were engaged in this behavior, versus just 4% along undeveloped shoreline.

Though lacking overall significance, diving birds appeared to show a weak trend of

resting in greater proportions along undeveloped shoreline in the summer (x > 2.59, p <

0.14 on two lakes).









Tending young behavior

During the summer, 100% (n=l 11) of ducks tending young did so along developed

shoreline (Table 3-2). On both lakes where ducks were observed with enough frequency

for analysis (n > 25), they tended young significantly more along developed shoreline

(both tests: x > 4.17, p < 0.04). The only wading birds observed tending young were two

pairs of Green Herons (Butorides virescens) on Lake Buckeye. No diving birds were

observed tending young on any of the lakes.

Active/swim behavior

During the winter, a significant overall pattern of association was observed

between active/swimming marsh birds and developed shoreline (Table 3-3). Twenty-

eight percent (n=676) of the marsh birds found along developed shoreline were observed

engaged in active/swim behavior, versus only 9% along undeveloped shoreline. This

pattern was observed on all four lakes (meta-analysis: 2 = 27.98, p < 0.001).

Human Activity

Though not quantified, the level of human activity on or immediately around all

four lakes appeared relatively low during both seasons. Given that surveys were typically

conducted during the week, and that the summer drought made lake access very difficult,

the amount of activity that was observed was no doubt considerably lower than at peak

times such as weekends during peak fishing season. High-intensity recreational activities

such as water skiing or jet skiing were not observed. Though fishing from boats appeared

to be the most common recreational activity, no more than four boats were ever

encountered on a lake at one time. Anglers were typically found slowly trolling along

undeveloped shoreline, fishing along edges of emergent or floating-leaf vegetation, and

were rarely observed disturbing or displacing waterbirds. No land-based human activity









was observed along undeveloped shorelines. Human activity along developed shorelines

included feeding ducks and marsh birds, fishing and relaxing on piers, fishing from boats,

lawn mowing, golf cart operation, construction, and small aircraft departure and arrival.

Several informal interviews with local residents who had Wood Duck nest boxes

on their properties revealed that many of these boxes were used on a yearly basis, and

frequently produced successful broods.

Table 3-2. Percent guild behavioral responses to developed (D) and undeveloped (U)
shoreline for summer 2001. Results listed by lake.
Guild n Alert/Flee Forage Rest w/Young
% % % % % % % %
D U D U D U D U D U
Waders
Buckeye 52 29 23 14 29 41 31 41 12* 0
Conine 212 68 8 16* 71** 43 17 28* -
Deer 141 17 6 18* 52* 29 27 29 -
Jessie 107 36 7 11 52* 39 33 28 -
Marsh
Buckeye 137 21 4 33* 39 52 3 5 33** 0
Conine 48 28 6 11 50 61 19 21 -
Deer 283 22 2 23* 52* 36 4 5 23 14
Jessie 113 21 2 0 37 48 4 5 34 38
Divers
Buckeye 21 6 38* 0 57 100* -
Conine 50 38 12* 3 4 3 78 89* -
Deer 61 11 3 0 93 91 -
Jessie 57 55 14 16 82 78 -
Ducks
Buckeye 2 7 -
Conine 15 0 -
Deer 725 50 25 60** 3 0 43** 4 10** 0
Jessie 352 32 36 88** 5 0 36** 7 12** 0
n = total number of birds observed along developed and undeveloped shoreline. Trend
suggesting greater percentage of guild displayed behavior along indicated shoreline (p <
0.2). Significantly greater percentage of guild displayed behavior along indicated
shoreline (p < 0.05). Dash indicates less than 25 birds observed on lake or no birds
observed exhibiting behavior.









Table 3-3. Percent guild behavioral responses to developed (D) and undeveloped (U)
shoreline for winter 2001/2002. Results listed by lake.
Guild n Alert/Flee Forage Rest Active/Swim
% % % % % % % %
D U D U D U D U D U
Waders
Buckeye 71 13 8 38* 61* 8 24 46* 6 8
Conine 133 14 9 36** 26 14 62* 36 1 7*
Deer 127 1 4 0 28 0 66 100 2 0
Jessie 152 26 7 31** 43** 19 47 50 1 0
Marsh
Buckeye 57 19 25 32 25 42* 9 11 39* 16
Conine 78 20 13 25 33 50* 9 5 41* 20
Deer 442 24 19 42** 50 46 5 4 23** 0
Jessie 99 11 8 9 49 82** 3 9 34** 0
Divers
Buckeye 52 32 17 22 75 72 8 3
Conine 58 13 19 23 3 0 71 69 5 8
Deer 56 3 43 67 41 33 16 0
Jessie 99 51 15 27* 3 0 77 73 4 0
Ducks
Buckeye 5 9 100 56 0 44
Conine 25 2 0 100** 100** 0
Deer 629 19 66 68 2 0 8 0 7 0
Jessie 88 10 90 100 10 0 -
n = total number of birds observed along developed and undeveloped shoreline. Trend
suggesting greater percentage of guild displayed behavior along indicated shoreline (p <
0.2). ** Significantly greater percentage of guild displayed behavior along indicated
shoreline (p < 0.05). Dash indicates less than 25 birds observed on lake or no birds
observed exhibiting behavior.

Disturbance Sensitivity

Seasonal alert/flee comparisons

Guild analyses comparing alert/flee behavior between seasons showed a significant

increase in alert/flee behavior in the winter for marsh birds and ducks (all tests: x > 6.2,

p < 0.01). Diving birds showed a similar trend, with birds on two lakes showing

significantly greater alert/flee behavior in the winter (both tests: X > 4.73, p < 0.03). The

proportion of wading birds displaying alert/flee behavior did not change between seasons









(x2 > 0.11, p < 0.74). Across all guilds, alert/flee behavior was observed 1.6 times more

often in the winter.

Inter-guild alert/flee comparisons

Ducks showed a significant pattern of greater alert/flee behavior than other guilds

on three lakes (all tests: x > 61.2, p < 0.0001). With all lakes combined, 32% (n= 1183)

of summer ducks (95% of which were Wood Ducks) displayed this behavior. Sixty-

seven percent (n=787) of winter ducks displayed alert/flee behavior. Winter ducks

included Wood Ducks, Blue-winged Teal (Anas discors), and Ring-necked Ducks

(Aythya collaris). In both seasons, the next closest guild was diving birds, which

displayed alert/flee behavior in 11% (n=299 for summer) and 25% (n=345 for winter) of

all observations. Diving birds actually displayed alert/flee behavior considerably more

than results indicate. However they appeared to tolerate a much closer approach distance

than ducks before displaying alert/flee behavior (pers. obs.), and thus their first observed

behavior was typically something other than the alert/flee response. Ducks on the other

hand, tended to become alert or flee from the boat's approach even from substantial

distances (> 50 m) (pers. obs.).

Migrant versus resident alert/flee comparisons

Six out of seven winter migrant species were either ducks or diving birds.

Alert/flee behavior was compared between migrant and resident species of these guilds.

Analyses for all lakes combined failed to show a significant difference for either guild

(both tests: X 2> 0.13, p < 0.72). Of note though was that only 1% of winter migrants

were found using undeveloped shoreline.









Discussion

Avian Responses to Shoreline Development

Alert/flee behavior

Alert/fleeing behavior was observed significantly less along developed shoreline on

the majority of lakes for summer ducks and winter wading birds. Though no other guilds

showed overall significance, trends suggest that this pattern was the case for all guilds

over both seasons (except summer divers). The findings suggest that many of the birds

on these lakes showed localized habituation to human disturbance, tolerating such

disturbance only along developed shoreline. Numerous other studies have noted patterns

of apparent habituation in areas where waterbirds are regularly exposed to moderate or

high levels of human activity. In her study of Great Crested Grebes (Podiceps cristatus),

Keller (1989) found reduced flush distances for birds in sites frequently visited by

humans. Lord et al. (2001) found that New Zealand Dotterels (Charadrius obscurus)

nesting on high-use beaches allowed closer approach distances before flushing than birds

nesting on remote beaches. Vos et al. (1985) found that Great Blue Herons were highly

sensitive to disturbance early in the nesting season, but otherwise appeared to habituate to

repeated non-threatening activities. And Burger and Galli (1987) found that the

proportion of gulls fleeing when disturbed was greater in areas of infrequent disturbance

than in heavily disturbed areas. However, the fact that summer diving birds showed a

trend of greater alert/flee behavior along developed shoreline emphasizes that not all

species may habituate.

In the previous studies mentioned above, disturbed and undisturbed sites were

separate and unique. In this study disturbed and undisturbed sites were juxtaposed on the

same lake for each of the four lakes sampled. The findings show that individual birds









within a guild, ducks and waders in particular, may differentiate between disturbed and

undisturbed sites even at a very localized scale. Even where undeveloped shoreline made

up as little as 1/5 of the total habitat, such as on Lake Deer, wading birds (summer),

ducks (summer), and marsh birds (both seasons) still showed greater sensitivity to

disturbance in this habitat. This implies that undeveloped shoreline meets a behavioral

need of at least some individuals, and may be seen as a localized refuge from disturbance.

Results suggest that perhaps even small areas of undeveloped shoreline are important in

order to minimize the stress to these birds.

It may be that some individuals that are less tolerant to disturbance only use

undeveloped shoreline. Alternatively, those same individuals may utilize both

undeveloped and developed shoreline. To determine this, individual behavioral

observations were required, which were beyond the scope of this study.

Foraging and resting behavior

Wading birds selected developed shoreline for foraging in the summer. The

removal of emergent and shoreline vegetation by property owners on these lakes most

likely promoted foraging conditions favorable to these birds (Chapt. 2). Large expanses

of relatively open shallow water were favored by stalking waders, such as herons and

egrets, while open moist shoreline was utilized by probing birds, such as ibis. Both

shallow open water and open shore were extremely limited along undeveloped shorelines

(Chapt. 2). The greater overall significance observed in the summer was most likely

linked to the extreme drought conditions during this study, which served to greatly

increase the availability of this foraging habitat.

Marsh birds selected undeveloped shoreline for foraging in the winter. Rallids, in

general, have strong seasonal shifts in diet, foraging primarily on animal matter in the









summer and plant matter in the winter (Elphick et al. 2001). This shift in diet may have

facilitated a winter shift to undeveloped shoreline, where greater densities of littoral

vegetation were found (Chapt. 2).

Ducks showed a strong preference for resting along developed shoreline in the

summer. This serves as another indication of this guild showing an increased tolerance

for human disturbance along developed shoreline. That summer diving birds showed a

trend in the opposite direction reiterates the idea that this guild, or least the year-round

resident species in this guild, may not be as tolerant as other species.

Tending young behavior

During the summer, 93% of ducks that tended young did so along developed

shoreline. This pattern suggests that either the value of developed shoreline habitat or the

disadvantages of undeveloped shoreline habitat strongly outweighed the disadvantages of

human disturbance. This was best exemplified by the Wood Duck. Despite being one of

the most sensitive species in this study in terms of alert/fleeing behavior (see Disturbance

Sensitivity below), Wood Ducks were reported to consistently use the next boxes located

on numerous lakeshore properties. Wood Ducks are known to readily use artificial nest

boxes often located in or around human-populated areas. Further, all Wood Ducks

observed tending young in this study were found along developed shoreline.

Previous studies have shown that areas of developed lake shoreline have

significantly lower fish species richness and total abundance than natural areas (Guillory

et al. 1979, Bryan & Scarnecchia 1992). Other studies have shown the importance of

dense stands of emergent vegetation as breeding areas and nurseries for fish and

invertebrates (Wegener et al. 1973, Barnett & Schneider 1974). Given these findings and

the fact that no significant differences were found in the amount of foraging observed









along developed and undeveloped shorelines for either marsh birds or ducks (the two

guilds observed with young), it seems unlikely that the greater abundance of birds

tending young along developed shoreline can be explained by better foraging

opportunities. However, on several occasions lake residents were observed putting out

corn and bread for the birds. If this supplemental feeding occurred on a large enough

scale, this could help explain the apparent preference for developed shoreline. In a

review of urban bird studies over the past 100 years, Marzluff (2001) found 29 studies

that linked supplemental feeding with increased bird densities in developed areas.

Alternative possibilities that could explain this pattern include easier predator detection

and reduced natural predators along developed shorelines.

The timing of this study did not coincide with the nesting season of most of these

birds. Observed tending young behavior therefore consisted of interactions between

parents and fledglings. Further research is needed to determine which habitats breeding

birds use for nesting on these lakes. Several of the species in this study, including

Common Moorhens, Purple Gallinules, Rails, Soras, Least Bitterns, and Green Herons,

are either facultative or obligate emergent vegetation nesters (Terres 1991). Three of

these species were found actively nesting in emergent habitat in a previous study of

neighboring urban lakes (Roth, in press). Many wading birds as well as the Anhinga nest

colonially in shrubs or trees close to waterbodies (Hancock & Kushlan 1984, Elphick et

al. 2001). Considering that both emergent vegetation and woodlands were found in

considerably greater abundance along undeveloped shorelines, and that many birds are

least tolerant of disturbance at the beginning of the nesting season (Hockin et al. 1992), it









seems likely that many of the birds in this study would have selected undeveloped

shorelines for nesting.

With few exceptions, wading birds and diving birds did not appear to breed or raise

young on these lakes. One possibility is that these guilds were so intolerant of human

disturbance during the breeding season that they sought out altogether less disturbed

habitats for nesting and rearing young. Miller (1943) suggested that distance from human

activity was the most important factor in heron nest site selection. However, numerous

studies have documented these guilds breeding in disturbed or developed areas (e.g.

Robertson & Flood 1980, Titus & Van Druff 1981, Vos et al. 1985, Keller 1989).

Further, in a similar study on neighboring developed lakes, Roth (in press) found heron

rookeries along the edges of several of his lakes.

Another possible scenario is that wading and diving birds did not breed on these

lakes due to the extreme drought. Several studies have shown that in cases of extreme

water regimes, waterbird nesting has been delayed (Weller & Spatcher 1965, Custer et al.

1996), abandoned (Breeden & Breeden 1982), or moved to more favorable sites (Weller

1999). This is just speculation however, and there are a variety of other factors, such as

the limited amount of undeveloped habitat, and the composition of this habitat, that may

have been unfavorable to these birds.

Active/swimming behavior

Active/swimming behavior, examined only in the winter, was seen significantly

more along developed shoreline in marsh birds. Active/swim behavior can be viewed as

a low-distress behavior, as opposed to alert/flee, and as such may be another indication of

habituation among these birds along developed shoreline. All three of the common

marsh bird species observed, the Common Moorhen, Purple Gallinule, and American









Coot, are known to become relatively tame, and even quite bold in developed areas if left

unmolested (Terres 1991, Elphick et al. 2001, West & Hess 2002).

In some instances this perceived behavior may have actually been an effort to

slowly move away from my boat. These cases may have represented a stage in a process

of habituation, where the birds were not entirely tolerant of disturbance, but did not feel

threatened enough to rapidly flee the area.

Disturbance Sensitivity

Seasonal alert/flee response

Results of this study revealed that waterbirds on these lakes displayed alert/flee

behavior 1.6 times more often in the winter than in the summer. Though several studies

have examined waterbird behavior across seasons (Burger 1981, Klein 1993, Klein et al.

1995), to my knowledge, no previous studies have attempted to quantify the degree of

variation in sensitivity to disturbance between seasons. The significant increase in

alert/flee behavior in marsh birds, diving birds, and ducks supports the idea that avian

disturbance sensitivity has a strong seasonal component. In general, there is a

significantly greater cost involved in failing to protect a nest and young during the

breeding season than there is in failing to defend temporary foraging and loafing sites at

other times of the year (Rodgers and Smith 1997). Thus, birds may be more likely to

temporarily abandon a preferred foraging or resting location in the non-breeding season.

However, an increase in alert/flee behavior is not necessarily an indication of increased

sensitivity to disturbance. In attempting to raise and ensure the survival of young during

the breeding season, birds must meet some of their greatest energy demands of the year.

Such physical demands may be mirrored in the amount of stress these birds undergo. If

breeding birds are facing at a minimum the same amount of stress that wintering birds are









facing, one could conclude that they would be equally sensitive to disturbance. Since this

sensitivity is not indicated through overt responses such as alert/flee behavior, these birds

may be enduring high levels of physiologically manifested stress, such as increased heart

rate, temperature, and blood sugar that could reduce their overall fitness (Gabrielsen &

Smith 1995).

The alternative, that birds are more easily stressed in the non-breeding season,

suggests that urban lakes may act as an energy drain for birds during this season. Even if

waterbirds are normally easily stressed in the non-breeding season, due to forces such as

a fluctuating food supply or simple flightinesss," they are more likely to encounter

disturbances in an urban environment that might trigger an alert/flee response. Such a

response may incur numerous negative physiological responses, increase energy

expenditures, reduce foraging times, and reduce overall fitness (Knight & Cole 1995).

Inter-guild alert/flee comparisons

Ducks showed significantly greater alert/flee behavior than other guilds in response

to the approach of the survey boat. This reinforces the results of other studies that have

documented a significant variation between waterbird species in sensitivity to human

disturbance (Batten 1977, Burger 1981, Vaske et al. 1983, Klein 1993, Klein et al. 1995,

Rodgers & Smith 1997, Rodgers & Schwikert 2002). Though Klein et al. (1995) found

migratory dabbling ducks showed the most consistent patterns of avoidance of human

visitors, no other studies have found this guild to be uniformly more sensitive to

disturbance than other species. Wood Ducks comprised over 80% of all duck

observations in this study. Previous studies have shown that this species will readily nest

in urban environments (Hepp & Bellrose 1995). The heightened alert/flee behavior

observed in these birds might represent an evolving process of habituation, where the









birds originally adopted these human-dominated environments out of necessity, but are

not yet entirely tolerant of the conditions. A similar process has been suggested for both

the Common Loon (Gavia immer) and the Great Crested Grebe (Titus & Van Druff 1981,

Keller 1989).

Migrant versus resident alert/flee response

The proportion of winter migrants that responded to human disturbance with

alert/flee behavior was not significantly greater than that of resident species. This fails to

confirm the results of van der Zande et al. (1980), Burger (1981), Burger and Gochfeld

(1991), and Klein et al. (1995), which consistently showed migrant waterbirds displaying

a greater sensitivity to human disturbance than resident birds. One possible explanation

is that the winter migrants arriving on urban lakes came from summer breeding grounds

that were also disturbed and thus they had already developed a tolerance. Previous

studies (above) were conducted in more natural habitats. If the winter migrants arriving

in those areas came from undisturbed breeding grounds, they may have not been

habituated to human disturbance upon arrival. The previous studies also found increased

disturbance sensitivity primarily in migrant waterfowl and shorebirds, whereas the

majority of migrants in this study consisted of diving birds. Another explanation can be

seen in studies that have shown distinct variation in sensitivity to disturbance between

individuals of the same species (Klein 1993, Klein et al. 1995). If this were the case, less

tolerant individuals would frequent less disturbed areas. In this study however, only 1%

of migrant birds were found along undeveloped shoreline. It may be that only highly

tolerant individuals (both migrants and residents) are attracted to urban lakes to begin

with and that they are fairly habituated to humans.









Management and Future Research

Clearly there are a wide range of factors that are associated with the behavioral

patterns of waterbirds in urban environments. Season, development and changes in

habitat complexity, disturbance levels, and individual guild and species' tolerances are

just a few examples. Managing for all of these variables is neither practical nor feasible

for most local or state agencies. Nor would a litany of restrictions be acceptable to local

residents. Though further research is needed on a greater number and variety of lakes

before specific management recommendations should be made, several general

recommendations are worth considering.

Waterbird behavior does appear to be associated with shoreline development, and

undeveloped shoreline may serve as an important refuge for birds that are more sensitive

to human disturbance or developed habitat. For example, significantly more wintering

marsh birds were observed foraging on undeveloped shoreline. Further, some guilds had

heightened alert/flee behavior on undeveloped shoreline, such as ducks (summer) and

waders (winter), which may indicate that some portion of these populations consists of

individuals that have not habituated to humans. The preservation of undeveloped

shoreline, preferably in larger contiguous patches, should therefore be considered in

future lake development plans.

Until more detailed habitat studies are conducted in urban aquatic environments,

developers should use the general ecology of the waterbird guilds that they hope to attract

as a guide for selecting areas of shoreline to preserve. Examples of foraging habitat

include areas of shallow, relatively open littoral zones for wading birds, open water

interspersed with mixed emergent and floating-leafed vegetation for ducks and marsh

birds, and larger areas of deeper open water for diving birds. Examples of nesting and









loafing habitat include shrubby or wooded shoreline for wading and diving birds, stands

of dense tall-emergent vegetation for marsh birds, and areas of floating-leafed vegetation

for ducks.

Likewise, residents wishing to increase waterbird biodiversity on urban lakes have

many options available to them. Adding Wood Duck boxes on residential properties may

increase nesting by this species along developed shorelines. Providing perching

substrates separate from high human-use areas such as docks might reduce disturbance to

resting diving birds and decrease unwanted fouling of docks. Ducks and marsh birds

may also be encouraged to forage along developed shorelines by planting species of

native emergent and floating-leafed vegetation typically found in their diet. Lists of such

plants are often available through local birding groups, state wildlife agencies, or state

extension offices. These resources may also be able provide residents with a better

understanding of the general behavior of many of these species and how best to limit

negative interactions.

There has been considerable research examining the potential use of buffer zones to

protect waterbirds from undue disturbance (Rodgers & Smith 1995, 1997, Rodgers &

Schwikert 2002). Buffer zones are impractical though along developed shorelines where

residents require lake access and desire the freedom to engage in a variety of activities on

their own property. If birds do indeed habituate at a local scale along developed

shorelines, then buffer zones may only be required along undeveloped portions of

shoreline. Given the overall elevated alert/flee behavior observed in the winter, buffers

in the non-breeding season may be warranted. However, further research is needed to

examine fluctuations in sensitivity to disturbance throughout the winter season. Further









research is also needed to determine nesting areas around urban lakes and whether

breeding season buffers are needed. Future studies should be conducted in years with

less extreme water conditions and examine where exactly birds are nesting on these lakes

and to what degree their tolerance of human disturbance fluctuates during the breeding

season.

Further research on foraging and resting behaviors is needed for waterbirds in

urban environments. Previous studies have suggested that sensitive species may find a

lack of adequate foraging or loafing sites as their preferred habitats become developed

and disturbances increase (Skagen et al. 1991, Pfister et al. 1992). This has been

corroborated, at least in terms of foraging opportunities, by studies that have shown lower

fish species richness and abundance along areas of developed lake shoreline (Guillory et

al. 1979, Bryan & Scarnecchia 1992). Though I found no difference in the proportion of

birds foraging along developed and undeveloped shorelines, other measures, such as

foraging times, prey selection, or strike/capture ratios, may better show the value of this

habitat for foraging birds, and should be examined in the future.

More comprehensive and long-term research is also strongly recommended for

urban lake habitats. Much work has been devoted to passerines in terrestrial systems

such as parks and the urban rural gradient (e.g. Blair 1996, Clergeau et al. 1998,

Fernandez-Juricic 2001). Urban aquatic systems have received far less attention, and

studies typically have been of relatively short duration; no more than a season or two.

Repeated case studies are needed to determine whether these birds are indeed adapting to

human-manipulated environments and whether such environments are acting as

functional habitat or merely as sinks. The urban rural gradient also needs to be examined









for lacustrine habitats, comparing waterbird behaviors across varying levels of shoreline

development. Additionally, though Hoyer & Canfield (1990, 1994) showed that higher

trophic state lakes are positively correlated with waterbird abundance, the popularity of

oligotrophic lakes for development (Hoyer, pers. comm.) indicates that future studies

should look at urban lakes covering a range of trophic states. Such studies would allow

us to better understand the effects of residential development and thus better plan for

limiting our impact in future growth.

In general, research with potential for direct application (e.g. urban studies), should

be organized to be as approachable as possible to planners and managers. Research at the

guild level can be much more readily applied to management efforts. While species-level

studies are critical in determining birds that appear most vulnerable to development and

disturbance, managing on a species by species basis requires tremendous effort and often

has conflicting priorities. Research and management that focuses at the guild level, when

based on sound science, should yield greater species richness and waterbird abundance

with the least amount of management. Many waterbird behavioral studies have examined

entire communities at the species level, occasionally grouping birds by size or seasonal

distribution. Comparing ecological functional groups of any kind should be considered

in future studies to allow for the results to be more easily applied to management efforts.















APPENDIX A
WINTER HAVEN WATERBIRD SURVEY DATA

Table A-1. Aquatic bird species observed on lakes Buckeye (B), Conine (C), Deer (D),
and Jessie (J) in Winter Haven, Florida, summer 2001 and winter 2001/2002.
Species Scientific Name Res' Lake
Buckeye Conine Deer Jessie
S W S W S W S W


Diving Birds
Pied-billed Grebe Podilymbus podiceps
Brown Pelicans Pelicanus occidentalis
D-C Cormorant Phalacrocorax autitus
Anhinga Anhinga anhinga
Osprey Pandion haliaetus
Ring-billed Gull Larus delawarensis
Caspian Tern Sterna caspia
Forster's Tern Sterna foster
Least TernT Sterna antillarum
Belted Kingfisher Ceryle alcyon
Wading Birds
Least Bittern Ixobrychus exilis
Great Blue Heron Ardea herodias
Great Egret Ardea alba
Snowy Egrets Egretta thula
Little Blue Herons Egretta caerulea
Tricolored Herons Egretta tricolor
Cattle Egret Bubulcus ibis
Green Heron Butorides virescens
B-C Night Heron Nycticorax nycticorax
White Ibiss Eudocimus albus
Glossy Ibis Plegadisfalcinellus
Wood StorkE Mycteria americana
Limpkins Aramus guarauna
Sandhill CraneT Grus canadensis
Black-Necked Stilt Himantopus mexicanus
Ducks (wild)
Wood Duck Aix sponsa
Mallard Anas platyrhnchos
Blue-winged Teal Anas discors
Ring-necked Duck Avthva collaris


5 19 8 8 15
0 0 1 0 0
3 332 40 309 10
29 68 54 26 77
7 11 34 21 11
0 33 0 36 0
0 0 0 5 0
0 4 0 0 0
1 0 0 0 0
0 4 0 15 0

9 0 1 3 6
13 24 67 42 33
4 11 26 11 22
1 3 29 10 3
1 4 4 2 8
7 8 28 14 19
1 1 0 6 0
48 11 24 1 21
0 0 0 0 0
7 35 112 69 60
0 0 31 0 0
0 0 2 1 6
4 3 0 0 4
0 0 12 2 0
0 0 2 0 0


10 8 8
0 0 0
41 25 64
37 99 68
3 30 14
67 0 23
2 0 1
0 0 0
0 14 0
7 2 10

0 3 0
7 52 39
18 21 14
0 33 5
14 2 4
22 15 11
31 5 35
3 16 2
0 2 9
91 23 144
0 0 0
2 1 0
0 1 0
0 0 0
0 0 0


11 22 15 27 808 334 396 106
0 0 6 2 0 0 50 0
0 5 0 0 0 5 0 0
0 0 0 0 0 357 0 0









Appendix A, continued.
Species Scientific Name Res' Lake
Buckeye Conine Deer Jessie
SW SW SW SW
Marsh Birds
Rail Rallus sp. Y 0 0 1 0 0 0 0 0
Sora Porzana carolina W 0 0 1 0 0 0 0 0
Purple Gallinule Porphyrula martinica Y 2 5 2 0 137 97 0 0
Common Moorhen Gallinula chloropus Y 171 81 83 73 177 206 153 94
American Coot Fulica americana W 0 0 0 26 1 169 0 22
Other
Domestic Goose Anser domesticus Y 0 0 0 0 0 0 29 22
Domestic Duck Anas domesticus Y 5 7 6 4 0 0 31 17
Bald EagleT Haliaeetus leucocephalus W 0 2 0 0 0 1 0 0
Killdeer Charadrius vociferous Y 0 0 12 5 0 0 2 11
s State listed as Species of Special Concern, T State listed as Threatened, E State and
federally listed as Endangered. 1Residence: S=summer resident, W=winter migrant,
Y=year- round resident.















APPENDIX B
WINTER HAVEN WATERBIRD HABITAT DATA









Table B-1. Wading bird habitat associations, summer 2001. Expected values based on
availability of habitat elements on each lake.
Habitat Lake Observed Expected X p-value +/-
Low E Buckeye 9 4.13 4.87 0.027* +
Conine 40 39.22 0 1
Deer 73 59.78 4.73 0.03* +
Jessie 48 64.44 7.26 0.007* +
Tall E Buckeye 44 47.71 0.53 0.467
Conine 120 192.03 110.55 <0.0001* -
Deer 57 94.92 45.81 <0.0001* -
Jessie 32 65.99 31.94 <0.0001* -
Floating Buckeye 27 30.21 0.38 0.538
Conine n/a n/a n/a n/a
Deer 112 116.06 0.64 0.424
Jessie n/a n/a n/a n/a
Shore Buckeye 29 27.14 0.1 0.752
Conine 189 92.60 156.66 <0.0001* +
Deer 38 19.32 19.89 <0.0001* +
Jessie 55 46.81 1.82 0.177
Lawn Buckeye 28 33.86 1.46 0.227
Conine 69 37.90 29.09 <0.0001* +
Deer 76 75.78 0 1
Jessie 62 66.83 0.53
Understory Buckeye 35 46.89 7.08 0.008* -
Conine 135 153.45 6.48 0.01* -
Deer 87 57.91 25.94 <0.0001* +
Jessie 91 65.94 17.29 <0.0001* +
Shrub Buckeye 49 50.14 0.02 0.888
Conine 124 12.30 25.42 <0.0001* +
Deer 64 52.48 3.92 0.048* +
Jessie 64 65.99 0.06 0.806
Canopy Buckeye 53 38.80 9.29 0.002* +
Conine 37 22.68 9.25 0.002* +
Deer 32 48.26 8.25 0.004* -
Jessie 24 43.57 12.08 <0.001* -
+/- Significant positive or negative association with habitat element.









Table B-2. Wading bird habitat associations, winter 2001/2002. Expected values based
on availability of habitat element on each lake.
Habitat Lake Observed Expected X p-value +/-
LowE Buckeye 46 4.90 354.84 <0.0001* +
Conine 84 23.87 176.28 <0.0001* +
Deer 52 55.08 0.21 0.647
Jessie 95 69.92 15.91 <0.0001* +
Tall E Buckeye 39 56.54 12.5 <0.001* -
Conine 68 116.89 83.11 <0.0001* -
Deer 36 87.46 92.21 <0.0001* -
Jessie 53 71.60 8.61 0.003* -
Floating Buckeye 19 35.81 11.85 <0.001* -
Conine 1 1.08 0 1
Deer 88 106.94 18.6 <0.0001* -
Jessie 1 2.91 0.69 0.406
Shore Buckeye 9 2.11 19.75 <0.0001* +
Conine 35 25.10 4.21 0.04* +
Deer 1 1.28 0 1
Jessie 39 14.38 44.64 <0.0001* +
Lawn Buckeye 73 40.13 44.88 <0.0001* +
Conine 73 23.25 122.84 <0.0001* +
Deer 116 75.78 51.04 <0.0001* +
Jessie 115 72.522 46.2 <0.0001* +
Understory Buckeye 61 58.46 0.18 0.671
Conine 93 104.10 3.33 0.068
Deer 80 58.37 14.06 <0.001* +
Jessie 58 72.06 4.82 0.028* -
Shrub Buckeye 47 59.42 6.28 0.012* -
Conine 49 99.79 72.01 <0.0001* -
Deer 49 52.48 0.29 0.59
Jessie 52 71.60 9.58 0.002* -
Canopy Buckeye 74 45.98 31.61 <0.0001* +
Conine 18 13.86 1.05 0.306
Deer 82 48.26 36.76 <0.0001* +
Jessie 58 47.28 3.2 0.074
+/- Significant positive or negative association with habitat element.









Table B-3. Marsh birds habitat associations, summer 2001. Expected values based on
availability of habitat element on each lake.
Habitat Lake Observed Expected X p-value +/-
LowE Buckeye 19 7.85 15.21 <0.0001* +
Conine 10 9.46 0 1
Deer 76 126.39 34.36 <0.0001*
Jessie 71 64.44 1.05 0.31
Tall E Buckeye 56 90.71 31.39 <0.0001*
Conine 45 46.30 0.05 0.823
Deer 148 200.69 42.15 <0.0001*
Jessie 30 65.99 35.88 <0.0001*
Floating Buckeye 51 57.44 0.98 0.322
Conine 0 0.43 0 1
Deer 279 245.38 26.14 <0.0001*
Jessie 4 2.68 0.25 0.617
Shore Buckeye 79 51.59 21.11 <0.0001* +
Conine 39 22.33 18.48 <0.0001* +
Deer 38 40.60 0.13 0.718
Jessie 71 46.81 17.71 <0.0001* +
Lawn Buckeye 100 64.37 32.94 <0.0001* +
Conine 7 9.21 0.38 0.538
Deer 84 148.59 67.75 <0.0001*
Jessie 52 66.83 5.84 0.016*
Understory Buckeye 70 91.35 12.17 <0.001*
Conine 37 34.48 0.37 0.543
Deer 108 106.70 0.01 0.92
Jessie 114 66.41 63.12 <0.0001* +
Shrub Buckeye 52 95.33 50.5 <0.0001*
Conine 41 39.53 0.06 0.806
Deer 183 102.91 104.33 <0.0001* +
Jessie 55 65.99 3.14 0.076
Canopy Buckeye 64 73.77 2.23 0.135
Conine 9 5.49 1.81 0.179
Deer 75 95.00 6.42 0.011* -
Jessie 31 43.57 4.84 0.028*
+/- Significant positive or negative association with habitat element.









Table B-4. Marsh bird habitat associations, winter 2001-2002. Expected values based on
availability of habitat element on each lake.
Habitat Lake Observed Expected X p-value +/-
LowE Buckeye 38 4.39 263.45 <0.0001* +
Conine 22 15.35 2.92 0.087
Deer 169 201.12 8.67 0.003* -
Jessie 87 50.73 46.46 <0.0001* +
Tall E Buckeye 47 50.65 0.48 0.488
Conine 70 75.14 1.19 0.275
Deer 169 319.34 218.35 <0.0001* -
Jessie 35 51.95 9.79 0.002* -
Floating Buckeye 26 32.08 1.55 0.213
Conine 0 0.69 0.05 0.823
Deer 406 390.46 3.39 0.066
Jessie 2 2.11 0 1
Shore Buckeye 10 1.85 32.39 <0.0001* +
Conine 9 16.14 3.26 0.071
Deer 0 3.19 2.29 0.13
Jessie 14 10.34 1.07 0.301
Lawn Buckeye 55 35.11 18.4 <0.0001* +
Conine 22 14.95 3.38 0.066
Deer 234 188.85 25.88 <0.0001* +
Jessie 88 52.14 45.59 <0.0001* +
Understory Buckeye 58 51.16 2.02 0.155
Conine 84 66.92 12.68 <0.001* +
Deer 243 145.46 118.99 <0.0001* +
Jessie 46 51.81 1.02 0.313
Shrub Buckeye 57 52.00 1.02 0.313
Conine 78 64.15 7.89 0.005* +
Deer 217 130.79 95.2 <0.0001* +
Jessie 35 51.48 9.32 0.002* -
Canopy Buckeye 66 40.24 30.44 <0.0001* +
Conine 8 8.91 0.02 0.888
Deer 198 120.26 79.62 <0.0001* +
Jessie 48 33.99 7.77 0.005* +
+/- Significant positive or negative association with habitat element.









Table B-5. Diving bird habitat associations, summer 2001. Expected values based on
availability of habitat element on each lake.
Habitat Lake Observed Expected X p-value +/-
LowE Buckeye 7 1.53 17.01 <0.0001* +
Conine 21 12.09 6.92 0.009* +
Deer 30 28.61 0.05 0.823
Jessie 42 51.64 2.98 0.084
Tall E Buckeye 17 17.67 0 1
Conine 46 59.20 11.31 <0.001*
Deer 43 45.43 0.25 0.617
Jessie 24 52.88 28.63 <0.0001*
Floating Buckeye 5 11.19 4.61 0.032*
Conine n/a n/a n/a n/a
Deer 53 55.54 0.44 0.507
Jessie n/a n/a n/a n/a
Shore Buckeye 9 10.05 0.05 0.823
Conine 50 28.55 24.26 <0.0001* +
Deer 7 9.38 0.44 0.507
Jessie 27 37.52 4 0.046*
Lawn Buckeye 7 12.54 3.48 0.062
Conine 14 11.78 0.29 0.59
Deer 22 36.11 12.57 <0.001*
Jessie 24 53.56 29.98 <0.0001*
Understory Buckeye 18 16.44 0.18 0.671
Conine 57 51.38 1.58 0.209
Deer 39 26.90 9.19 0.002* +
Jessie 86 53.22 37 <0.0001* +
Shrub Buckeye 21 18.57 0.53 0.467
Conine 47 50.54 0.52 0.471
Deer 44 25.83 20.49 <0.0001* +
Jessie 73 52.88 13.68 <0.001* +
Canopy Buckeye 26 14.37 16.55 <0.0001* +
Conine 19 7.02 20.63 <0.0001* +
Deer 29 22.62 2.45 0.118
Jessie 42 34.92 1.79 0.181
+/- Significant positive or negative association with habitat element.









Table B-6. Diving bird habitat associations, winter 2001/2002. Expected values based on
availability of habitat element on each lake.
Habitat Lake Observed Expected X p-value +/-
LowE Buckeye 23 4.85 67.79 <0.0001* +
Conine 26 11.78 18.91 <0.0001* +
Deer 34 35.87 0.09 0.764
Jessie 65 61.24 0.32 0.572
Tall E Buckeye 51 55.96 0.86 0.354
Conine 32 57.68 45.62 <0.0001* -
Deer 34 56.95 27.49 <0.0001* -
Jessie 58 62.71 0.53 0.467
Floating Buckeye 7 35.44 35.12 <0.0001* -
Conine 4 0.76 9.98 0.002* +
Deer 49 69.64 34.05 <0.0001* -
Jessie 1 4.15 1.75 0.186
Shore Buckeye 0 2.00 1.16 0.281
Conine 21 12.23 6.69 0.01* +
Deer 0 0.75 0.08 0.777
Jessie 6 12.50 3.18 0.075
Lawn Buckeye 39 38.04 0.01 0.92
Conine 31 11.33 38.24 <0.0001* +
Deer 60 44.40 12.59 <0.001* +
Jessie 71 63.04 1.68 0.195
Understory Buckeye 66 55.42 4.69 0.03 +
Conine 46 50.70 1.08 0.299
Deer 65 34.20 49.34 <0.0001* +
Jessie 104 62.65 50.38 <0.0001* +
Shrub Buckeye 62 56.33 1.24 0.265
Conine 29 48.60 21.33 <0.0001* -
Deer 57 30.75 36.54 <0.0001* +
Jessie 98 62.24 37.54 <0.0001* +
Canopy Buckeye 83 43.59 66.67 <0.0001* +
Conine 17 6.75 15.47 <0.0001* +
Deer 69 28.28 91.86 <0.0001* +
Jessie 64 41.10 17.67 <0.0001* +
+/- Significant positive or negative association with habitat element.









Table B-7. Duck habitat associations, summer 2001. Expected values based on
availability of habitat element on each lake.
Habitat Lake Observed Expected X p-value +/-
Low E Buckeye 0 0.51 0 1
Conine 0 2.17 1.53 0.216
Deer 194 312.99 78.29 <0.0001*
Jessie 71 176.40 114.88 <0.0001*
Tall E Buckeye 3 5.89 2.36 0.124
Conine 0 10.63 40.04 <0.0001*
Deer 235 496.97 427.24 <0.0001*
Jessie 34 180.65 222.25 <0.0001*
Floating Buckeye 9 3.73 9.73 0.002* +
Conine n/a n/a n/a n/a
Deer 713 607.66 105.79 <0.0001* +
Jessie n/a n/a n/a n/a
Shore Buckeye 2 3.35 0.33 0.566
Conine 14 5.12 21.59 <0.0001* +
Deer 246 100.80 241.53 <0.0001* +
Jessie 247 128.15 162.31 <0.0001* +
Lawn Buckeye 3 4.18 0.19 0.663
Conine 12 2.11 49.08 <0.0001* +
Deer 269 330.34 27.46 <0.0001*
Jessie 206 182.96 5.28 0.022* +
Understory Buckeye n/a n/a n/a n/a
Conine 4 9.46 8.03 0.005*
Deer 391 245.78 156.63 <0.0001* +
Jessie 297 181.81 136.78 <0.0001* +
Shrub Buckeye 6 6.19 0 1
Conine 1 0.69 17.95 <0.0001* +
Deer 325 228.78 12.17 <0.001* +
Jessie 104 180.65 60.34 <0.0001*
Canopy Buckeye 5 5.75 0.02 0.888
Conine 0 1.26 0.51 0.475
Deer 131 210.37 47.46 <0.0001*
Jessie 40 119.27 75.29 <0.0001*
+/- Significant positive or negative association with habitat element.









Table B-8. Duck habitat associations, winter 2001/2002. Expected values based on
availability of habitat element on each lake.
Habitat Lake Observed Expected X p-value +/-
Low E Buckeye 0 1.38 0.59 0.442
Conine 2 4.19 0.8 0.371
Deer 62 277.12 290.09 <0.0001*
Jessie 16 17.82 0.18 0.671
Tall E Buckeye 2 15.90 27.49 <0.0001*
Conine 25 20.49 3.25 0.071
Deer 133 440.02 668.13 <0.0001*
Jessie 23 18.25 1.86 0.173
Floating Buckeye 20 10.07 14.08 <0.001* +
Conine 2 0.19 9.15 0.002* +
Deer 580 538.02 18.7 <0.0001* +
Jessie 5 0.74 19.44 <0.0001* +
Shore Buckeye 7 0.37 96.21 <0.0001* +
Conine 0 4.40 4.13 0.042*
Deer 0 2.20 1.32 0.251
Jessie 11 3.67 14.06 <0.001* +
Lawn Buckeye 7 7.52 0 1
Conine 25 4.08 120.51 <0.0001* +
Deer 148 130.24 5.61 0.018* +
Jessie 17 18.49 0.1 0.752
Understory Buckeye 11 10.96 0 1
Conine 27 18.25 11.51 <0.001* +
Deer 182 100.32 120.76 <0.0001*
Jessie 24 18.37 2.71 0.1
Shrub Buckeye 11 11.14 0 1
Conine 2 17.50 13.16 <0.001*
Deer 153 90.20 72.93 <0.0001* +
Jessie 22 18.25 1.09 0.296
Canopy Buckeye 18 8.62 17.54 <0.0001* +
Conine 2 2.43 0 1
Deer 179 82.94 176.73 <0.0001* +
Jessie 11 12.05 0.04 0.841
+/- Significant positive or negative association with habitat element.















APPENDIX C
WINTER HAVEN WATERBIRD BEHAVIOR DATA














Table C-1. Data for summer guild behavior, listed by guild and lake. Results from contingency Chi-square based on the proportion of birds engaged in
each focal behavior versus the proportion engaged in all other (Other) behaviors.
Foraging Resting w/Young Alert/Fleeing
P- P- w/ P- Alert/ P-
D/U Forage Other 2 value Rest Other 2 value Young Other Z2 value Flee Other Z2 value


Waders

Buckeye


Conine


Deer


Jessie

Marsh
00
Buckeye


Conine


Deer


Jessie


37
17 1.32 0.251
61
39 18.32 <0.0001*
68
12 3.04 0.081
51
22 1.95 0.163


83
10 1.26 0.261
24
11 0.82 0.366
137
14 1.89 0.169
71
11 0.82 0.367


16 36
12 17 0.93 0.336
37 175
19 49 3.54 0.06
38 103
5 12 0.04 0.78F
35 72
10 26 0.3 0.581


4 133
1 20 0.20 0.653
9 39
6 22 0.08 0.777
12 271
1 21 0.01 0.999F
5 108
1 20 0.01 1F


6
0
n/a
n/a
n/a
n/a
n/a
n/a


45
0
n/a
n/a
64
3
38
8


46
29 3.61 0.083'
n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a


92
21 9.65 0.002*
n/a n/a
n/a n/a n/a
219
19 0.96 0.429F
75
13 0.16 0.692


40
25 1.01 0.314
195
57 3.81 0.051
132
14 2.74 0.123'
99
32 0.46 0.497'


131
14 28.21 <0.001F*
45
25 0.49 0.664'
278
17 28.28 <0.001F*
111
20 0.36 0.999F














Summer guild behavior, continued.
Foraging Resting w/Young Alert/Fleeing
P- P- w/ P- Alert/ P-
D/U Forage Other Z2 value Rest Other 2 value Young Other Z2 value Flee Other 2 value


n/a n/a
n/a n/a n/a
48
37 0.12 1F
n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a


n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a
702
50 1.64 0.39F
334
32 1.72 0.383'


0.136


2.1 0.156


0.09 0.575'


0.32 0.569


Divers

Buckeye


Conine


Deer


Jessie

Ducks

Buckeye


Conine


Deer


Jessie


30.11 <0.0001*


11.35 <0.001*


D=Developed shoreline, U=Undeveloped shoreline
FP-value based on Fisher's exact test.


n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a


n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a
655
50 5.31 0.018'*
311
32 4.17 0.036F'*


3.25 0.136F


2.59 0.135'


0.37 IF


0.12 0.731


28.98 <0.0001*


33.08 <0.0001*


n/a
n/a
n/a
n/a
315
2
125
2














Table C-2 Data for winter guild behavior, listed by guild and lake. Results from contingency Chi-square based on the proportion of birds engaged in
each focal behavior versus the proportion engaged in all other (Other) behaviors.
Foraging Resting w/Young Alert/Fleeing
P- P- w/ P- Alert/ P-
D/U Forage Other 2 value Rest Other 2 value Young Other Z2 value Flee Other 2 value


28
12 12.31 <0.001*
98
12 0.97 0.519F
91
1 0.39 0.999F
86
21 5.42 0.02*


43
11 2.13 0.144
52
10 1.9 0.168
222
13 0.14 0.707
50
2 4.15 0.042*


54
7 2.73 0.172'
51
9 3.53 0.06
43
0 0.51 0.999F
80
13 0.06 0.803


52
17 0.05 1F
71
19 0.335 0.999F
421
23 1.04 0.999F
96
10 1.03 0.348F


6
5
12
5
5
0
11
8


14
6
10
5
86
10
8
1


65
8 8.7 0.011F*
121
9 8.82 0.012F'*
122
1 0.04 1F
141
18 12.9 0.002F'*


43
13 0.36 0.547
68
15 1.82 0.293'
356
14 6.87 0.017F'*
91
10 0.01 1F


Waders

Buckeye


Conine


Deer


Jessie

Marsh

Buckeye


Conine


Deer


Jessie


67
12 0.08 IF
132
13 3.86 0.182'
125
1 0.02 0.999F
151
26 0.17 1F


35
16 3.36 0.067
46
16 3.03 0.082
342
24 6.91 0.009*
65
11 5.47 0.017'*


22
3
32
4
100
0
34
0














Winter guild behavior, continued.
Foraging Resting w/Young Alert/Fleeing
P- P- w/ P- Alert/ P-
D/U Forage Other Z2 value Rest Other Z2 value Young Other Z2 value Flee Other Z2 value


n/a
n/a
2
0
n/a
n/a
3
0


n/a
n/a
n/a
n/a
13
0
n/a
n/a


n/a
n/a n/a
56
13 0.46
n/a
n/a n/a
96
51 1.89


n/a
n/a n/a
n/a
n/a n/a
616
19 0.4
n/a
n/a n/a


n/a


n/a


0.287'F



n/a


n/a


'P-value based on Fisher's exact test.


39 13
23 9 0.1
41 17
9 4 0.01
23 33
1 2 0.07
76 23
37 14 0..32


n/a n/a
n/a n/a n/a
n/a n/a
n/a n/a n/a
53 576
0 19 1.74
9 79
0 10 1.13


0.752


43
25 0.27 0.605
47
10 0.11 0.711'
32
1 0.66 0.578F
84
37 3.27 0.07


0
4 3.11 0.221'
25
0 27 0.003F'*
217
6 0.07 0.791
9
0 1.13 0.592'


4
1
3
1
9
0
4
0


0
4
25
0
41
0
n/a
n/a


48
31 0.74 0.645'
55
12 0.13 0.563'
47
3 0.57 0.999F
95
51 2.12 0.3'


5
5 3.11 0.221'
0
2 27 0.003'*
588
19 1.32 0.625'
n/a
n/a n/a n/a


Divers

Buckeye


Conine


Deer


Jessie

Ducks

Buckeye


Conine


Deer


Jessie