<%BANNER%>

Streambank Erosion on the Restored Lower Kissimmee River, Florida

Permanent Link: http://ufdc.ufl.edu/UFE0044170/00001

Material Information

Title: Streambank Erosion on the Restored Lower Kissimmee River, Florida What Site Factors Influence Rates?
Physical Description: 1 online resource (139 p.)
Language: english
Creator: Horan, Andrew M
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: bank -- erosion -- fluvial -- geomorphology -- resources -- restoration -- river -- water
Geography -- Dissertations, Academic -- UF
Genre: Geography thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The initial purpose of this investigation was to evaluate how different vegetative, sedimentologic, and geomorphic site factors influence erosion rates in an18-km stretch of the recently restored Kissimmee River in Florida. A modified version of Rosgen's (2001) Bank Erosion Hazard Index (BEHI) was used to characterize potential erosion severity. Fifty streambanks were measured and monitored over a nine month period, from November 2010 through August 2011. At each study site, a toe pin was installed and used as a constant point of reference for each site throughout the study. Vertical and horizontal measurements of the bank profile were taken three separate times and recorded and graphed. Bank profiles were overlaid to calculate the bank areal change and bank retreat that was lost or gained due to erosion or deposition. Sediment cores were extracted and assessed for bulk density and a grain size analysis. The five main variables Rosgen used were assigned a BEHI value and corresponding rating to each site. The streambanks displayed an inaccuracy in rating related to actual sediment loss. Erosion rates showed a slight decreasing trend downstream, with the most erosion occurring upstream, including Montsdeoca, UBX, and River Runs. A few sites downstream, however, also showed excessive erosion. Vegetation cover, root depth, and the number of different sediment layers in the bank profile were the most significant variables contributing to erosion rates. Streambanks positioned along meander bends with a radius of curvature/channel width value between 1 and 2.5 to be the most consistent indicator of excessive erosion. A time series analysis of stage height and discharge of PC62 (upstream) and PC33 (downstream) displayed much variation. The pooling effect of the S65C structure in the southern reaches tends to stabilize water levels and mitigate repeated wetting and drying of banks, causing little erosion. The findings of this study can be used to explain excessive sediment yields in the river following prolonged periods of unusually high discharges and stage heights. The results will also be used to continually modify Rosgen's BEHI method to the Kissimmee for quick and accurate bank failure ratings in the future.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Andrew M Horan.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Mossa, Joann.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044170:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044170/00001

Material Information

Title: Streambank Erosion on the Restored Lower Kissimmee River, Florida What Site Factors Influence Rates?
Physical Description: 1 online resource (139 p.)
Language: english
Creator: Horan, Andrew M
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: bank -- erosion -- fluvial -- geomorphology -- resources -- restoration -- river -- water
Geography -- Dissertations, Academic -- UF
Genre: Geography thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The initial purpose of this investigation was to evaluate how different vegetative, sedimentologic, and geomorphic site factors influence erosion rates in an18-km stretch of the recently restored Kissimmee River in Florida. A modified version of Rosgen's (2001) Bank Erosion Hazard Index (BEHI) was used to characterize potential erosion severity. Fifty streambanks were measured and monitored over a nine month period, from November 2010 through August 2011. At each study site, a toe pin was installed and used as a constant point of reference for each site throughout the study. Vertical and horizontal measurements of the bank profile were taken three separate times and recorded and graphed. Bank profiles were overlaid to calculate the bank areal change and bank retreat that was lost or gained due to erosion or deposition. Sediment cores were extracted and assessed for bulk density and a grain size analysis. The five main variables Rosgen used were assigned a BEHI value and corresponding rating to each site. The streambanks displayed an inaccuracy in rating related to actual sediment loss. Erosion rates showed a slight decreasing trend downstream, with the most erosion occurring upstream, including Montsdeoca, UBX, and River Runs. A few sites downstream, however, also showed excessive erosion. Vegetation cover, root depth, and the number of different sediment layers in the bank profile were the most significant variables contributing to erosion rates. Streambanks positioned along meander bends with a radius of curvature/channel width value between 1 and 2.5 to be the most consistent indicator of excessive erosion. A time series analysis of stage height and discharge of PC62 (upstream) and PC33 (downstream) displayed much variation. The pooling effect of the S65C structure in the southern reaches tends to stabilize water levels and mitigate repeated wetting and drying of banks, causing little erosion. The findings of this study can be used to explain excessive sediment yields in the river following prolonged periods of unusually high discharges and stage heights. The results will also be used to continually modify Rosgen's BEHI method to the Kissimmee for quick and accurate bank failure ratings in the future.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Andrew M Horan.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Mossa, Joann.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044170:00001


This item has the following downloads:


Full Text

PAGE 1

1 STREAMBANK EROSION ON THE RESTORED LOWER KISSIMMEE RIVER, FLORIDA: WHAT SITE FACTORS INFLUENCE RATES? By ANDREW MICHAEL HORAN 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 2012

PAGE 2

2 2012 Andrew Michael Horan

PAGE 3

3 To my parents, Mike and Carol Horan

PAGE 4

4 ACKNOWLEDGMENTS There are numerous individuals that helped me in this project from start to finish. I would like to thank first and foremost my graduate chair, Dr. Joann Mossa, for her easy patience, constant availability, helpful criticisms, and overall guiding hand in this project. She is paramount to my success on this thesis. I would like to thank Dr. Pete Waylen and Dr. William Wise for their help with big questions early on, and their overall assistance as part of my committee. I want to thank the South Florida Water Management District for funding the project and also Jose Valdez and Joe Koebel for being the project managers and Joann Mossa as the PI. I want to thank each of my boat drivers, Jessica Wilson, Amber Graham, Brent Anderson, Michael Cheek, Andrew Rodusky, and Therese East, for without them this project would have been infinite ly more time consuming and difficult. I want to thank my field assistants Ursula Garfield, Michael Suharmadji, Michal Jones, and Angela Bell. I want to thank my two lab assistants, Angela Bell and Hannah Herrero for help with the soil analysis. I want to t hank Dr. Tim Fik for help with statistics. I want to thank David Anderson for help in collecting hydrologic data and navigating the DBHYDRO website. I want to thank my parents, Mike and Carol, my two sisters, Natalie and Leah, and my brother, Matthew, for encouraging me throughout all the standstill times during this long and arduous process. And lastly, I want to thank Patty Griffin and Steve Earle for giving me musical guidance and equanimity during the many late nights and early mornings writing and rese arching for this project.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTI ON ................................ ................................ ................................ .... 15 Background ................................ ................................ ................................ ............. 15 Objectives and Research Questions ................................ ................................ ....... 17 Study Area ................................ ................................ ................................ .............. 18 2 LITERATURE REVIEW ................................ ................................ .......................... 22 Kissimmee River ................................ ................................ ................................ ..... 23 Impacts of Channelization ................................ ................................ ................ 23 Process of Restoration ................................ ................................ ..................... 24 Phosph orus and Nutrient Loading ................................ ................................ .... 24 Streambank Erosion Processes ................................ ................................ .............. 25 Fluvial Entrainment ................................ ................................ ........................... 26 Bank Weakening Processes ................................ ................................ ............. 27 Bank Erosion Mechanics and Types ................................ ................................ 28 Meander Bend Erosion ................................ ................................ ..................... 30 Bank Erosion Hazard Index and Framework for Kissimmee ................................ ... 31 3 METHODS AND MATERIALS ................................ ................................ ................ 36 Study Area ................................ ................................ ................................ .............. 36 Bank Erosion Obser vations, Rates, and Measures ................................ ................. 38 Bank Profile Measurements ................................ ................................ ............. 38 Kissimmee River Bank Erosion Hazard Index (BEHI) ................................ ...... 39 Bank Sediment Analysis ................................ ................................ ................... 40 Areal Change and Lateral Retreat ................................ ................................ .... 42 Geographic Information Systems (GIS) ................................ ................................ .. 43 Radius of Curvature/Channel Width Measurement ................................ .......... 44 Stage Height and Discharge Hydrograph ................................ ............................... 44 Statistical Analysis ................................ ................................ ................................ .. 45 Point Biserial Correlation Coefficient ................................ ................................ 45 Kruskal Wallis Test ................................ ................................ ........................... 46 ................................ ................................ .......... 47

PAGE 6

6 Pearson Correlation Coefficient ................................ ................................ ........ 47 4 RESULTS ................................ ................................ ................................ ............... 63 Lost Sites ................................ ................................ ................................ ................ 63 Site Characteristics ................................ ................................ ................................ 65 BEHI Model, Sediment Loss, and Lateral Bank Retreat ................................ ... 65 Evidence of Erosion ................................ ................................ ......................... 68 Sediment Characteristics ................................ ................................ .................. 69 Stage Height and Discharge ................................ ................................ ................... 70 St atistical Analysis ................................ ................................ ................................ .. 71 Geographic Information Systems (GIS) ................................ ................................ .. 74 5 DISCUSSIONS AND CONCLUSIONS ................................ ................................ ... 95 BEHI Model and Measured Erosion Rates ................................ ............................. 95 Spatial Variability of Erosion Rates ................................ ................................ ......... 96 Possible Influential Variables on Erosion Rates ................................ ...................... 98 Radius of Curvature/Channel Width ................................ ............................... 101 Stage Height (m) and Discharge (m 3 /s) ................................ .......................... 102 Further Research ................................ ................................ ................................ .. 103 APPENDIX BANK PROFILE GRAPHS ................................ ................................ ...... 106 LIST OF REFERENCES ................................ ................................ ............................. 133 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 139

PAGE 7

7 LIST OF TABLES Table page 3 1 Run and Connector name with number of erosion sites ................................ ..... 59 3 2 Four main BEHI categories and r espective index value. ................................ ... 59 3 3 Grain size analysis sieve numbers and corresponding sieve apertu res and aggregate name, in order from stacked top to bottom. ................................ ....... 60 3 4 Point biserial tests of site factors vs. each site response. ................................ ... 60 3 5 Kruskal Wallis tests of site factors vs. site response ................................ .......... 61 3 6 ................................ .... 61 3 7 Pearson Correlation tests for site factors vs. site response. ............................... 62 4 1 Kissimmee River gates wi th stage height (m) and flow (cms) for each time inspection. ................................ ................................ ................................ .......... 75 4 2 Lost sites with corresponding gross changes for recorded time steps. ............... 75 4 3 Lost sites with corresponding net changes for recorded time steps. .................. 76 4 4 Gross quantities of erosi on for each site. ................................ .......................... 77 4 5 Net quantiti es of erosion for each site ................................ ................................ 78 4 6 Mean occurrence of evidence of erosion per reach location .............................. 79 4 7 Mean values of geological variables per run name. ................................ ............ 80 4 8 values when a point biserial correlati on coefficient is applied. ...... 81 4 9 Kruskal Wallis test chi square values for the BEHI variables vs. erosional change. ................................ ................................ ................................ ............... 81 4 10 performed on geological and planform variables, an d also the total BEHI score ................................ ................................ ................... 82 4 11 Pearson Correlation Test performed on the parametric variables of bank height (cm) and distance downstream (km) ................................ ........................ 82

PAGE 8

8 LIST OF FIGURES Figure page 1 1 Map of t he Lower Kissimmee River, Florida. ................................ ...................... 21 2 1 Illustration of the process of bac kfilling in the Kissimmee River .......................... 33 2 2 Illustration of various types of bank failure ................................ .......................... 33 2 3 Illustration of the effects of pore water p ressure and bank failure ...................... 34 2 4 Site MB03L, showing mass wasting of material ................................ .................. 35 2 5 Rotational slumping of material, shown here in Red River of the North .............. 35 3 1 Overview of Lower Kissimmee with all 50 initial bank erosion sites .................... 49 3 2 Map of Micco Bluff shelter area erosion sites. ................................ .................... 50 3 3 Micco Bluff Shelter. ................................ ................................ ............................ 51 3 4 Map of Montsdeoca Run south sites, including the connector channel. ............. 52 3 5 Map of Fulford Run sites and the much smaller Strayer Run. ............................ 53 3 6 Map of River Run #1 sites. ................................ ................................ ................. 54 3 7 Map of UBX Run sites. ................................ ................................ ....................... 55 3 8 Map of Montsdeoca North Sites ................................ ................................ ......... 56 3 9 Demonstration of taking bank profile measurements. ................................ ......... 57 3 10 Example graph of bank prof ile measurements ................................ ................... 57 3 11 Kissimmee River hydrographs for both PC 62 and PC33 ................................ .. 58 4 1 Kissimmee River map showing the location of the 3 gages ............................... 83 4 2 Graphs of erosion quantities vs. BEHI sc ores of sites ................................ ........ 84 4 3 Graph of bank height (cm) vs. Distance downstream of RR01 (km) ................... 86 4 4 Graph of percentage of silt/clay content vs. distance downstream of RR01 (km) ................................ ................................ ................................ .................... 86 4 5 Graph of percentage of sand content vs. distance downstream of RR01 (km). .. 87

PAGE 9

9 4 6 Graph of bulk density values (g/cm 3 ) vs. distance downstream of RR01 (km) ... 87 4 7 Graph of D50 median grain size (mm) vs distance downstream of RR01 (km). ................................ ................................ ................................ ................... 88 4 8 For each site, graphs of the number of lithological layers vs erosion quantities. ................................ ................................ ................................ .......... 89 4 9 Hydrographs for PC62 and PC33 for time period 01/01/2008 01/01/201 2 ...... 91 4 10 Flow duration curves from 01/01/2 008 01/01/2012 ................................ ........... 92 4 11 Graphs of erosion quantities vs. radius of cur vature ratio ................................ .. 93 5 1 Correlation of % silt/clay vs. bulk density ................................ .......................... 105 A 1 RR01R ................................ ................................ ................................ .............. 106 A 2 RR02L ................................ ................................ ................................ .............. 106 A 3 RR03R ................................ ................................ ................................ .............. 107 A 4 RR04R ................................ ................................ ................................ .............. 107 A 5 RR05L ................................ ................................ ................................ .............. 108 A 6 UB01L ................................ ................................ ................................ .............. 108 A 7 UB02R ................................ ................................ ................................ .............. 109 A 8 UB03R ................................ ................................ ................................ .............. 109 A 9 UB04R ................................ ................................ ................................ .............. 110 A 10 UB05L ................................ ................................ ................................ .............. 110 A 11 UB06L ................................ ................................ ................................ .............. 111 A 12 UB07L ................................ ................................ ................................ .............. 111 A 13 UB08R ................................ ................................ ................................ .............. 112 A 14 MN01R ................................ ................................ ................................ ............. 112 A 15 MN02R ................................ ................................ ................................ ............. 113 A 16 MN03L ................................ ................................ ................................ .............. 113 A 17 MN04R ................................ ................................ ................................ ............. 114

PAGE 10

10 A 18 MN05L ................................ ................................ ................................ .............. 114 A 19 MN06R ................................ ................................ ................................ ............. 115 A 20 MN07L ................................ ................................ ................................ .............. 115 A 21 MN08R ................................ ................................ ................................ ............. 116 A 22 MN09R ................................ ................................ ................................ ............. 116 A 23 MN10R ................................ ................................ ................................ ............. 117 A 24 MN11R ................................ ................................ ................................ ............. 117 A 25 MN12L ................................ ................................ ................................ .............. 118 A 26 MN13 ................................ ................................ ................................ ................ 118 A 27 MN14R ................................ ................................ ................................ ............. 119 A 28 MN15R ................................ ................................ ................................ ............. 119 A 29 MS01R ................................ ................................ ................................ ............. 120 A 30 MS02R ................................ ................................ ................................ ............. 120 A 31 FF01L ................................ ................................ ................................ ............... 121 A 32 FF02L ................................ ................................ ................................ ............... 121 A 33 FF03R ................................ ................................ ................................ .............. 122 A 34 FF04R ................................ ................................ ................................ .............. 122 A 35 FF05L ................................ ................................ ................................ ............... 123 A 36 FF06L ................................ ................................ ................................ ............... 123 A 37 FS01L ................................ ................................ ................................ ............... 124 A 38 FS02L ................................ ................................ ................................ ............... 124 A 39 FS03R ................................ ................................ ................................ .............. 125 A 40 FS04R ................................ ................................ ................................ .............. 125 A 41 ST01R ................................ ................................ ................................ .............. 126 A 42 ST02R ................................ ................................ ................................ .............. 126

PAGE 11

11 A 43 OX01R ................................ ................................ ................................ .............. 127 A 44 OX02L ................................ ................................ ................................ .............. 127 A 45 OX03L ................................ ................................ ................................ .............. 128 A 46 MB01L ................................ ................................ ................................ .............. 128 A 47 MB02L ................................ ................................ ................................ .............. 129 A 48 MB03L ................................ ................................ ................................ .............. 129 A 49 MB04R ................................ ................................ ................................ ............. 130 A 50 MB05R ................................ ................................ ................................ ............. 130 A 51 Front of Kissimmee Bank Erosion form with BEHI variables and other field information collected. ................................ ................................ ........................ 131

PAGE 12

12 LIST OF ABBREVIATIONS BEHI Bank Erosion Hazard Index DBHYDRO Environmental Database from SFWMD EPA Environmental Protection Agency FDEP Florida Department of Environmental Protection FDOT Florida Department of Transportation GIS Geographic Information Systems LOI Loss on Ignition NBS Near Bank Stress SFWMD South Florida Water Management District SPSS Statistical Package for the Social Sciences TMDL Total Maximum Daily Load UF University of Florida USACE United States Army Corps of Engineers USGS United States Geological Survey

PAGE 13

13 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science STREAMBANK EROSION ON THE RESTORED LOWER KISSIMMEE RIVER, FLORIDA: WHAT SITE FACTORS INFLUENCE RATES? By Andrew Michael Horan May 2012 Chair: Joann Mossa Major: Geography The initial purpose of this investigation was to evaluate how different vegetative, sedimentologic, and geo morphic site factors influence erosion rates in an18 km stretch (2001) Bank Erosion Hazard Index (BEHI) was used to characterize potential erosion severity. Fifty streamban ks were measured and monitored over a nine month period, from November 2010 through August 2011. At each study site, a toe pin was installed and used as a constant point of reference for each site throughout the study. Vertical and horizontal measurements of the bank profile were taken three separate times and recorded and graphed. Bank profiles were overlaid to calculate the bank areal change and bank retreat that was lost or gained due to erosion or deposition. Sediment cores were extracted and assessed for bulk density and a grain size analysis. The five main variables Rosgen used were assigned a BEHI value and corresponding rating to each site. The streambanks displayed an inaccuracy in rating related to actual sediment loss. Erosion rates showed a sli ght decreasing trend downstream, with the most erosion occurring upstream, including Montsdeoca, UBX, and River Runs. A few sites

PAGE 14

14 downstream, however, also showed excessive erosion. Vegetation cover, root depth, and the number of different sediment layers in the bank profile were the most significant variables contributing to erosion rates. Streambanks positioned along meander bends with a radius of curvature/channel width value between 1 and 2.5 to be the most consistent indicator of excessive erosion. A time series analysis of stage height and discharge of PC62 (upstream) and PC33 (downstream) displayed much variation. The pooling effect of the S65C structure in the southern reaches tends to stabilize water levels and mitigate repeated wetting and drying of banks, causing little erosion. The findings of this study can be used to explain excessive sediment yields in the river following prolonged periods of unusually high discharges and stage heights. The results will also be used to continually modify the future.

PAGE 15

15 CHAPTER 1 INTRODUCTION Background Erosion and the a ssociated sedimentation, non point source pollution, and land loss occurs naturally in every watershed. The Environmental Protection Agency (EPA) reports that material from eroding stream banks is the leading cause of water quality problems, and correspon ds to significant sources of sediment transport into the bed material load within watersheds (EPA 2011). It is estimated that 30 80% of total sediment loading into streams is directly related to stream bank erosion from pervasive hydraulic action in the ch annel (Simon and Darby, 1999; Fox et al., 2007). Bank erosion is tied to site specific conditions and will vary accordingly along river channels as a function of sediment grain size, bank angle, cohesiveness of material, moisture content, vegetation, and v ariations in shear stress from fluctuating hydraulic processes (Thorne, 1982). The Kissimmee River, located in South Central Florida, flows from Lake Kissimmee at its headwaters, to Lake Okeechobee at its mouth (Figure 1 1). Since the Kissimmee was ditch ed and channelized into one main artery, the C 38 canal, from 1962 to 1971 (Bousquin et al 2005), the environmental impacts were so damaging that the United States Army Corps of Engineers (USACE), in conjunction with the South Florida Water Management Dis trict (SFWMD), decided that the river and its floodplain must be restored back to its historical flow regime (Koebel 1995). Recent restoration studies have concluded that the river has been found to yield increased sediment loads that may be attributed to accelerate d streambank erosion (Schenk et al 2011). Furthermore, restoration of the Kissimmee River could pose a problem with bank

PAGE 16

16 erosion and sedimentary loads as the river continues to adjust through lateral migration, channel widening, and other hydraulic processes. Increased sedimentary loads can harbor adverse effects to the Kissimmee by enhancing transport of pollutants, such as phosphorus and nitrogen, downstream and eventually into Lake Okeechobee, therefore affecting Total Maximum Daily Load s (TMDL). A TMDL is a calculation of the maximum amount of a certain pollutant that a waterbody can sustain before breaching water quality standards (water.epa.gov). The EPA has listed Total Maximum Daily Loads (TMDL) for Lake Okeechobee for phosphorus to be 140 metric tons to achieve a phosphorus concentration of 40 ppb within the lake (FDEP 2001). Lake Okeechobee has been given a designated use as a potable water supply to the local communities as well as critical habitat for endangered species such as snail kites, and recreational use for fisherman and bird hunters (FDEP 2001). Therefore, increased erosion and transport in the Kissimmee River can enhance transport of phosphorus pollutants downstream and straight into Lake Okeechobee, directly affecting its TMDL for phosphorus. As the Kissimmee interacts more with its floodplain and permits nutrient removal capabilities characteristic of pre channelized floodplain marshes (Toth 1990), quantifying sediment loads from bank erosion is paramount to understand ing the status of nonpoint source pollution into the river. Some possible site factors involved in the erosion of bank material along the Kissimmee may include shear stress following high discharges of water, slumping, rotational sliding, and bio geomorphi c influences such as invasive catfish burrows, cattle trampling, and vegetation. Cohesiveness of may amplify erosion where non cohesive banks can be more susceptible to cracking and mass wasting from subaqueous

PAGE 17

17 weathering, contributing to failed banks. In cohesive banks, tension cracks may form near the top of the bank and lead to the removal of large blocks of cohesive material and deposit them at the bank toe, allowing not only temporary scour protection but also an increased hazard of lateral bank migrat ion (Hagerty 1980). Hydraulic action forces can also be subdivided into actions of surface flow, such as waves propelled from passing motorboats, and actions occurring at the bank toe, eventually leading to undercutting and subsequent overhang of bank mate rial. To evaluate the magnitude of failed bank material into the stream it is imperative to analyze the discrete processes influencing streambank erosion under existing conditions. Objectives and Research Questions The main purpose of this study is to better understand the relative role of different vegetative, sedimentologic, and geomorphic site factors that influence bank erosion on the Kissimmee River. This is facilitated through the use of a modified version of t he Bank Erosion Hazard Index (BEHI) method (Rosgen 2001). Establishing permanent bank erosion sites along the newly restored portion of the Kissimmee River will facilitate future studies on erosion and help monitor geomorphic change until the river is full y restored. A rapid and accurate method of estimating sediment contributions from streambank erosion is a requisite component in developing a watershed budget for the Kissimmee River. Another objective is to locate areas of excess erosion in the Kissim mee River to determine if there is a spatial pattern based on erosion quantities. This information can be used by the South Florida Water Management District (SFWMD) for future studies on geomorphic monitoring, increased phosphorus loading in the watershed or for mitigation and protection of river banks. The study will also add to the increasing

PAGE 18

18 the method in relation to the Kissimmee. The following research questions were asked: Where is active bank erosion occurring and where is it most severe? How are specific vegetative, sedimentologic, and geomorphic variables tied to bank erosion rates? The results of this study will serve as a baseline of bank erosion monitoring on t he Kissimmee River for state agencies as the river continues to be restored further downstream. Evaluating and mapping sites of severe erosion hazards will act as a guide to the future natural migration of the river and a tool to apportion sediment contrib ution of streambank sediment sources to the total load transported by the river. Study Area The Kissimmee River is located in south central Florida, and flows from Lake Kissimmee at its headwaters 169 kilometers south into Lake Okeechobee (Figure 1 1). T he Kissimmee and its associated streams and small watersheds comprise the largest source of water into Lake Okeechobee and are part of the greater Everglades ecosystem. The approximate 7594 km 2 watershed can be subdivided into two geographic boundaries, th e upper and lower basins. The upper basin consists of 4135 km 2 of land area and includes approximately 24 lakes varying in size from 0.5 to 152 km 2 of water area. The lower basin, which encompasses the Kissimmee River itself, Lake Istokpoga, and other asso ciated small tributaries, consists of 1731 km 2 watershed area and flows into Lake Okeechobee. The river flows southward on a gentle slope of 0.07 km 1 from an elevation of 15.5 km at Lake Kissimmee to 4.6 m at Lake Okeechobee (Koebel 1995). The geology of the basin contains the Ocala Group and Avon Park Limstone, both composing the Floridan aquifer for the Upper Tertiary and

PAGE 19

19 Quaternary stratigraphy (Shaw and Trost 1984). The soil classification of the area is the Manatee Delray Okeelanta, a slightly acidic to neutral sandy and organic soil that is poorly drained and has a dark colored surface with loamy subsoil located more than 1 m below the surface (McCollum and Pendleton 1971). Native vegetation in the Kissimmee basin consists mostly of black willow ( Sal ix nigra) sawgrass ( Cladium jamaicense) grasses tolerant of periodic wetness (e.g. Panicum hemitomo, Rynchospora inundata) a myriad of wetland shrubs (e.g. Cephalanthus occidentalis, Ludwigia peruviana) hibiscus ( Hibiscus grandiflora) scattered live o ak hammocks ( Quercus virginiana ), and occasional punctuated areas of cabbage palms ( Sabal palmetto ) and bald cypress ( Taxodium distichum) (McCollum and Pendleton, 1971; Schenk et al. 2011). Invasive vegetation in the riparian area include the Old World cli mbing fern ( Lygodium microphyllum ), introducted in the 1960s, and Brazilian pepper ( Schinus terebinthifolius ), a South American shrub introduced in the 19 th century (Ferriter et al 2008). Invasive species on the floodplain include para grass ( Brachiaria m utica ), a plant from Africa, and Torpedo Grass ( Panicum repens ), an environmentally and economically damaging pl ant from Australia (Ferriter et al 2008). The climate of the region is typical of humid sub tropical areas, with roughly similar temporal wet a nd dry seasons accompanied by hot and humid summers and moderately cool winters. Annual precipitation ranges from 121 cm in the upper basin to 114 cm in the lower basin, and annual high and low temperatures range from 5 to 30 C (SFWMD DBHYDRO Database). F urther north in the United States, rivers that flood their banks observe high discharges in the late winter and spring (USGS 2011). In most years, the Kissimmee undergoes bankfull discharge, periodic flooding, and high flows in

PAGE 20

20 the mid to late summer month s; typically July, August, and September. The increase in flow is through strong, isolated convective thunderstorms that are capable of producing heavy precipitation along central Florida, accelerating runoff and discharge into the river. Record high flows can also be observed in late summer to early fall months when the Atlantic hurricane season is busiest and more likely to spawn tropical systems and heavy rains. By 1971, the river was separated and locked into Pools A E by six dam structures, and guided by a single channel, the C 38 canal. The area of interest for this study is the recently restored sections in Pool B/C, stretching approximately 18 km from River Run #1 and meandering southward through Micco Bluff Run to the Micco Bluff Shelter (Figure 1 1). The site was chosen because since the restoration, this is the only significant study performed on bank erosion of the restored section of the Kissimmee River. Preliminary findings have indicated that the suspended sediment loads in Kissimmee have bee n higher than expected for unknown reasons (Pearman et al. 2009) and that recent channel change found along the restored section is undoubtedly related to bank erosion (Mossa et al. 2009). Eventually part of Pool D will be restored and this study will be a framework for geomorphic change along the Kissimmee as it continues to be restored to a natural flow regime.

PAGE 21

21 Figure 1 1 The Lower Kissimmee River, Florida.

PAGE 22

22 CHAPTER 2 LITERATURE REVIEW Chapter 2 discusses and explains the major questions postulated in this study. The Kissimmee River is unique because for nearly 30 years the river was ditched and channelized into a large, deep, and wide canal. The natural hydraulic action processes, such as bank erosion, point bar formation, chann el evolution, and a meandering flow regime, became entirely nonexistent. The intended purpose of the channelization was for flood protection, but little knowledge was known at the time of unintended consequences to the beautiful and diverse ecosystem in Fl position as the main artery to the greater Everglades ecosystem. Since the restoration, there have been no studies of significance on streambank erosion, related severity to land cover and bank composition, its contribution to sus pended sediment in the channel, and the adverse effect consequences it may have on the system. The concepts and problems associated with the Kissimmee River serve to better understand the ecological significance of the study. Specific site factors are then discussed in the chapter followed by a background of the BEHI rating procedure. There are many combined and discrete processes that contribute to failed bank material. Not only is direct hydraulic action a main force of bank erosion, but influences of lan d cover and zoo geomorphologic factors cannot also be ignored. Geologic composition of the profile is crucial to the cohesiveness of material and relative severity of erosion. This chapter discusses mechanics of individual streambank erosion processes alon incorporating the different vegetative, sedimentologic, and geomorphic site factors and

PAGE 23

23 their role in bank erosion to allow a thorough understanding of how his system is applied and modified in this study. Kissimmee River Impacts of Channelization The Kissimmee River was channelized between 1962 and 1971 by the USACE as part of a federally funded project for flood control of the watershed (Toth 1992). The result being the 90 km long, 9 m deep and 64 105 m wide canal labeled C 38. The construction of 6 dams separated the system into 5 Pools, Pool A E (Figure 1 1). Toth (1992) has given the physical impacts of the channelization as twofold: The complete destruction of large portions of the river channel and drainage of the floodplain, including over 2,800 ha of wetlands water levels and modifying flow In addition to the many adverse physiological effects to c hannelization, severe biological impacts quickly followed. In the pre channelized environment, nearly 14,000 ha of broadleaf marsh, wetland shrub, and wet prairie dominated the vegetative region and provided habitat for numerous waterfowl and fish species (Milleson et al. 1980). Along with restoring wildlife habitat, part of the early drive for restoration came from concern that the channelized environment fostered rapid downstream transport of nutrient loads into Lake Okeechobee, without first being absorb ed and cycled through the floodplain marsh (Toth 1992). Continuous flooding of the drained marsh landscape in turn produces healthy growth of wetland plants that create food web pathways and colonization by small fish and other invertebrates. Seed bank gro wth is enhanced as well and becomes an essential food source for neighboring and seasonal waterfowl

PAGE 24

24 (Milleson 1976). Wetland resources can now finally be effective in sustaining this valuable ecosystem. Process of Restoration The goal of the Kissimmee Rive r Restoration Plan is to restore the ecological integrity of the basin by returning natural flow to the river system so the area can support a balanced and integrated ecological community that was entirely absent during channelization (Toth 1990). The USAC E (1991) has listed four components that are paramount to the framework of the restoration. They include monitoring of ecological, hydrologic, sedimentation, and stability indicators. When the Kissimmee was channelized, remnant channels that were used to d ivert water from the C 38 conduit no longer received the flow of water characteristic of natural river systems unconstrained by channelization (Anderson and Chamberlain 2005). The main objective in restoring the natural flow lies in the process of backfill ing the C 38 canal with sediment to continue flow through the remnant channels and eventually carving its way thro ugh the floodplain (Anderson et al 2005) (Figure 2 1). Currently, the area between the upper extent of restoration near Fort Kissimmee (Figur e 1 1) and the S65C gate have been restored. Phosphorus and Nutrient L oading The Kissimmee River is the largest source of phosphorus input into Lake Okeechobee (Jones 2005). Pre restoration studies have shown that the reestablishment of floodplain hydraul ic conditions, characteristic of the historical Kissimmee, decreased the total phosphorus and inorganic nitrogen concentrations suspended in the river by 40% (Davis 1981). This occurred because when the ecosystem is exposed to normal seasonal wet and dry c ycles, the floodplain will act as a source and sink for nutrients.

PAGE 25

25 The more permanent wetland marsh, when stressed to high nutrient capacity, will at some point reach a state of equilibrium in which nutrient outputs equal i nputs (Goldstein 1990). When the river is subjected to a natural flow with episodic high and low discharges, excess nutrients will be transported downstream in the channel, and more problematically through the adsorption of phosphorus in the suspended sediment load. Phosphorus adsorption reactions are enhanced by decreasing sediment grain size and dependent on many other factors within the soil horizon, particularly the amount of Fe oxide in the organic layer and the percentage of silt/clay (Pant and Reddy 2001; Drever 1997). There is trem endous spatial variability of soils, vegetation, and hydroperiod within the Kissimmee, and all are enmeshed to influence rates of assimilation and decomposition (Jones 2005). Therefore, increased sediment and nutrient loading into the river can most certai nly be exacerbated by the occurrence of streambank erosion. Streambank Erosion Processes Streambank erosion is important to the large scale evolution of rivers and the pathways of migration through the floodplain. Bank erosion is a multi faceted fluvial p rocess that is a result of combinational factors specific to the individual environmental setting and the force exerted on the bank by the flow of water. As the river bank holds a variable of control on the stream channel width, it also directly influences other stream processes while contributing to the sedimentary load (Ritter et al. 2006). A number of studies have been performed and have identified the major processes involved in bank erosion (Thorne 1982; Simon et al. 1999). Researchers studying the ph enomenon have decided that bank erosion is not the result of a single external force, but a series of interconnected and convoluted site specific factors coming together in an opportune

PAGE 26

26 manner. Thorne (1982) has described the process through two key proces ses: fluvial entrainment and mass wasting through the destabilization of bank material. Fluvial Entrainment Fluvial entrainment is the process of detachment and eventual removal or failure of bank material from the streambank face by flow velocities acti ng directly on the near bank environment. The process of erosion generated solely by river flow is termed hydraulic action. Failed bank material through hydraulic action is then either deposited at the bank toe, fostering accumulation and erosion protectio n, or entrained and transported downstream (Ritter et al. 2006). Many factors come into play with the process of entrainment, including the geologic composition of the bank, as well as the type, density, and root system of riparian vegetation along the cha nnel (Charlton 2008). The composition of the bank material sections off into two main types, non cohesive and cohesive. Hydraulic action can sometimes undercut the non cohesive material, leaving behind overhanging bank sections, called cantilevers, compos ed of the remaining cohesive bank. Eventually, the cantilever becomes mobilized by external fluvial forces and falls into the channel and is either deposited at the toe or transported downstream (Figure 2 2a). Thorne (1982) has also described a similar pro cess when tension cracks form vertically, beginning from the top of the bank, and continuing downward. The pulsing of water gently breaks the area where the tension crack meets ailure occurs frequently within the Kissimmee River, due to the interaction with the cohesive muck material and the relatively non cohesive sand that are most prevalent in riparian soils (Anderson 2005).

PAGE 27

27 Bank Weakening Processes Weakening and weathering pr ocesses are the impetus for bank failure and instability along alluvial channels. As frequently stated, the mechanics of failure depend on a variety of variables within the environment, but also with the geometry and physical chemical properties of the ban k itself. The factor that most influences the stability of the bank is the soil moisture content, where the strength of the material is directly related with the amount of saturation. The periodical saturation, drawdown, and re saturation of banks coincidi ng with fluctuating water levels tend to cause erosion in two ways; by reducing bank strength, and hydraulic action on the bank which slowly detaches and separates bank material (Simon and Collision 2001). As the river undergoes a period of bankfull disch arge, complete saturation of the bank occurs. This fosters a positive pore water pressure within the bank, which slowly and efficiently dislodges the cohesive sediment particles (Thorne and Osman 1988). As the stream channel elevation increases above bankf ull stage, it effectively reinforces higher pressure on the sediment grains of the bank because of the shear stress acting on it from stream power of the channel. The drawdown of water from the sediment results in increased matric suction strength between sediment grains, causing clay particles to expand and creates weakening cracks at the bank surface. The quick desiccation also negates pore pressure from the bank material, thus weakening the bank and causing failure. Figure 2 3 illustrates the process of pore water pressure from initial low channel stage to bank failure (Sass 2011). Water moves below the surface as groundwater flow and infiltration through the soil from precipitation. The interstitial spaces in the non cohesive bank material can provide nu merous pathways for the transport of water as groundwater discharge from

PAGE 28

28 the bank to the channel. This erosional process is called piping and can further loosen bank particles from cohesion, preparing the bank for failure from shear stress or gravity (Simo n and Collision 2001). Seepage through the non cohesive soil can be horizontal or vertical. Horizontal seepage can transport material away from the bank and into the channel from frictional forces, while vertical seepage can cause tension cracks at the sur face, promoting bank failure by mass wasting (Iverson et al. 2000; Wilson et al. 2007). The failure of banks generally happens in the waning period after high flow, as the rapid recession of water allows for continual saturation of banks and the hydrostati c pressures from the stream channel are released (Wilson et al. 2007). This is typically when gravitational forces take over the erosional process. Bank Erosion Mechanics and Types There are several types of bank failures and mechanisms according to the se diment composition of the bank. Other vegetative and sedimentologic variables including bank height, bank angle, soil bulk density, and the reinforcing effects of riparian vegetation must also not be ignored (Rosgen 2001). Under non cohesive banks, a gener al mass wasting of material down the slope of the bank is very common. Cohesive bank failure usually occurs along a failure plane, where shear stress exceeds shear strength (Simon et al. 1999). The failure plane can be a curved or a flat, planar surface. T his is usually on steep banks with an outward exposed face composed of cohesive sediment with vertical planar failure likely to occur when the saturated banks are no longer supported by the hydraulic force from channel (Figure 2 2b) (Thorne 1982; Charlton 2008). Non cohesive banks typically fail in a sliding mass wasting fashion, slipping down a curved slope as a result of weakened bank material (Figure 2 2c) (Thorne 1982; Charlton 2008). A combination of both cohesive and non cohesive

PAGE 29

29 sediment layers resul t in undercutting of the non cohesive layer until the cohesive layer becomes an overhang and succumbs to the force of gravity (Figure 2 2a) (Thorne 1982). When the United States Army Corps of Engineers (USACE) began the backfilling of the main channel in 1999, certain parts of the original channelized Kissimmee became connector areas. This is where a small part of the original C 38 dredged canal was not backfilled, therefore connecting river flow from one side of the previous C 38 canal to the other. The i nitial incision of the Kissimmee with the connectors led to steep banks with acute angles. Due to the scour of bed and streambank, these banks, composed of a non cohesive material, are prone to failure much easier than even prior to channel incision (Thorn e 1999). Steeper slopes tend to favor mass wasting mechanisms because of the physical characteristics of the bank and the fact that the force of gravity will eventually overcome the frictional forces of the sediment grains controlling bank stability. This leads to bank failure in a planar fashion for steep banks, whereas shallow banks fail rotationally (Thorne 1999). The Kissimmee River typically experiences mass wasting on much of the banks (Figure 2 4), especially the steep banks located in tight bendways with little to no stabilizing riparian vegetation. Rotational slumping, producing a staircase like bank (Figure 2 5) leading up to bankfull height, typically occurs on shallow banks with a gentle sloping scarp down the bed of the channel. However, a mix o f mass wasting and rotational slumping also seem to be common. Aside from the sediment composition of the bank, external biological and zoo geomorphologic influences also play a role in bank erosion specific to the Kissimmee

PAGE 30

30 River. Invasive catfish ( Pteryg oplichthys sp. ) to the Kissimmee River will burrow in the channel banks to spawn and nest their young (Nico 2009). Some of the burrows can be quite long and deep, preparing the bank for eventual failure by mass wasting. Some of the burrows can be clustered together, up to 30 38 cm in a given area, promoting bank instability. The unstable non cohesive banks, as a result of gravity and pore water pressure, will fail (Thorne and Tovey 1981). Alligator slides appear regularly as well, and help to flatten banks and give way t o mass wasting. Riparian vegetation also plays a role in bank stability by contributing to the shearing resistance of the streambank through the reinforcing effect of roots (Darby and Simon 1999). The strengthening effect of roots on the streambank is dire ctly related to the root depth and density, which is Meander Bend E rosion The raidus of curvature ratio is a geomorphic variable that can indicate a certain magnitude of erosio n occurring within the meander bend streambank. The radius of curvature ratio is a planform variable indicating how sharply curved the riverbend is situated, thus influencing direction and magnitude of most of the flow when meandering downstream. Hickin an d Nanson (1975) found that certain radius of curvature values corresponded to higher rates of channel migration. Radius of curvature ( r m /W m ) ratios between 2.0 and 3.0 displayed higher rates of channel migration than values above and below that range. Huds on and Kesel (2010), however, discovered that for the Lower Mississippi River, ratios of ~1 corresponded to the highest migration rates (m/yr). They discovered that the complexity of the floodplain sediment deposits with the combination of resistant clay p lugs prominent in this part of the Mississippi contributed to the differences in this study (Hudson and Kesel 2010) from more conventional thought on

PAGE 31

31 the subject (Hooke 1997). The meandering river will continue to adjust itself in order to reach an equilib rium in form and expenditure of energy as it carves through the floodplain downstream (Hooke 1975; Knighton 1998). Through using GIS, we can determine which ratios correspond to higher rates of bank erosion, an indicator of how much and in what direction t he meander bend is migrating. Bank Erosion Hazard Index and Framework for Kissimmee As the Kissimmee River continues to meander through the floodplain and evolve until equilibrium is achieved within the restored portion, bank erosion may be the main driver for planform change and a predictor of channel migration. Rosgen (2001) has developed a general classification of effective streambank erosion potential through his Bank Erosion Hazard Index (BEHI). The combination of a modified BEHI procedure and quantif ication of lateral migration rates will give an idea as to how much of the annual sediment yield can be apportioned to bank erosion, and whether or not this will escalate based on the calculated severity. The point system of the BEHI is based on seven cri teria. These are bank height vs. bankfull depth, bank angle, root density, root depth, surface protection, soil stratification, and particle size. These are streambank characteristics that are identified separate from the near bank velocity gradient and sh ear stress developed from the Near Bank Stress (NBS) ratings (Rosgen 2001). Four of these criteria, coupled with a thorough bank sediment analysis and bank erosion observations and type, create the modified point system developed in giving hazard ratings t o different erosion sites. The and scaled appropriately from low to extreme in potential for erosion. Athough Rosgen has included the ratio of bank height to bankfull heig ht in the original BEHI study (2001),

PAGE 32

32 this study has omitted it as a characteristic because the bank heights are low, and the river historically was above bankfull much of the year; in this case, the value is close to 1. Bank angle, however, gives an idea of whether undercutting is occurring at the bank toe and thus giving way to cantilever failure, or whether planar failure can occur due to a near 90 angle bank face. Banks of high angles are more susceptible because of the near likelihood of scouring at t he toe, thus weakening the cohesive cantilever and causing failure (Ritter et al. 2006). Root density and depth are coupled together because they both represent the reinforcing effects of riparian vegetation on the stability of streambanks. The higher the density and depth percentage, the more roots there are to provide cohesion of sediment grains and strengthens the composition of the streambank (Genet et al. 2005; Pollen 2007). The number of layers that stratify the bank are also a factor of severity. The more uniform the bank, the less likely it is to erode because seepage is kept to a minimal and a more uniform bank composition will strengthen the bank (Rosgen 2006). Increased stratification allows for more seepage and dislodging of bank material. Partic le size plays a large role in the severity of erosion, where higher particle size indicates non cohesive soil layers that are more susceptible to erosion, and decreased particle size refers to cohesive soil layers that are less susceptible to erosion and m ore likely to increase streambank strength (Rosgen 2001; 2006).

PAGE 33

33 Figure 2 1 Illustration of the process of backfilling in the Kissimmee River The channelized system is represented in the top and the restored system in the bottom. Arro ws indicate direction of flow. [Adapted from Anderson et al 200 5.] Figure 2 2. Illustration of various types of bank failure [Adapted from Ritter et al. 2006 used with permission from Waveland Press, Inc ]

PAGE 34

34 Figure 2 3. Illustration of the effects of pore w ater pressure and bank failure [Adapted from Sass 2011 used with permission from Christopher K. Sass ]

PAGE 35

35 Figure 2 4. Site MB03L, showing mass wasting of material Figure 2 5. Rotational slumping of material, shown here in Red River of the North, July, 2000. http://www.ndsu.edu/fargo_geology/mass_wasting/slumptypes.htm [Photo courtesy of North Dakota State University Geosciences Department 2000 used with permission fr om North Dakota State University ]

PAGE 36

36 CHAPTER 3 METHODS AND MATERIAL S Acquisition of field data is essential to the analysis, test of predictions, and further understanding of bank erosion in the Kissimmee River. The main framework for the methods of t his research study is a slightly modified version of the BEHI prediction system developed by Dave Rosgen (Rosgen, 1996, 2001, 2006). For the fifty sites employed in the study, the ability of a streambank to resist erosion coupled with erosion potential a re categorized on a scale from very low to extreme using the BEHI model (Rosgen 1996, 2001, 2006). Longitudinal profile measurements were taken at each site to show the specifics of erosion, such as amount of lateral migration (if any), areal changes, and the occurrence/nonoccurrence of scour/fill. Measured erosion rates from the nine months of this study are then synthesized and related to the many variables in this study and to the approximate annual sediment yield of the Kissimmee. Study Area The study area encompasses an approximately 18 km stretch of channel located in Pool B/C of the Kissimmee River, Florida (Figure 3 1). Figure 3 1 also gives an overview of the region with all fifty sites included. This stretch was chosen for a number of reasons. The Micco Bluff shelter area in Micco Bluff Run (Figure 3 2), the southernmost portion of the study, is an S shaped meander bend that has been undergoing rapid change. Recently, the actual shelter has been demolished due to land loss from bank erosion (Figur e 3 3). Further upstream, the geomorphology team from the University of Florida (UF) (Mossa et al 2010) laid out transect locations on strategically placed meander bends and straight reaches in Fulford and Montsdeoca Runs, as well as the connector locati ons. For use in future geomorphic monitoring, we

PAGE 37

37 coupled transect site locations with active bank erosion sites in both Montsdeoca and Fulford Runs, and their connectors (Figures 3 4, 5). Further upstream is where the channel has been most recently backf illed and restored. Chosen sites, mainly on meander bends and places observed to be actively eroding with a diversity of riparian vegetation, were placed in this northern section at River Run #1, UBX Run, and the most upstream corridor of Monstdeoca Run ( Figure 3 6, 7, 8). These sites were chosen mainly because they have experienced the most direct effect from the fluctuating water levels and discharge and have since incurred tipping and fallen vegetation, direct sediment removal from obvious streambank si te factors, and their position in the channel that is thought to exhibit increased erosion potential. Site names were labeled in shorthand by the run name, ascending numerical order from upstream to downstream, and given an L or R at the end to denote left or right bank location. Seven runs and two connectors were included in the study, stretching from the furthest upstream site, RR01R, at 27.5182 N and 81.2078 W, to MB05R, at 27.4333 N and 81.1391 W. Table 3 1 shows each run corresponding to a given nu mber of sites. Fifty sites were initially installed, with minor changes occurring as sites were lost during the nine months of observation. The timeframe of the study is from 11/15/2010 8/22/2011, with an intermediate measurement in April or May of 2011. Transportation up and down the river and to and from each site was provided by SFWMD boat drivers.

PAGE 38

38 Bank Erosion Observations, Rates, and Measures Bank Profile M easurements The vertical and horizontal profiling of the streambank is an integral part of the study and the key to quantifying the amount of erosion tangibly taking place at a particular site. The nature of the profile begins at the bank toe, with a toe pin used as a monument for each successive temporal measurement. The toe pins are installed by driving sections of 1.2 m long and 1.9 cm thick rebar vertically into the channel bed immediately adjacent to the streambank of interest. The pins were pounded in using a heavy mallet until approximately 30 cm. of rebar showed above the surface of the b ank toe. The first installation of these pins occurred during the week of November 15 19, 2010, when water levels averaged 11m or lower, according to the United States Geological Survey (USGS) water level monitoring station at Lorida, Florida. For the meas urements, we used a pair of flat edged survey rods and a framing level. Each toe pin served as the control point for vertical and horizontal measurements. One person held the vertical rod straight up and steady on the toe pin. I made each horizontal measur ement to keep consistent over time with values. Horizontal measurements were made in increments of 10 cm in relation to the vertical zero point, located at the top of the toe pin (Figure 3 9). Negative measures in 10 cm increments were recorded to capture the entire profile in perspective, not just from the top of the toe p in. The profiles were transcribed into Microsoft Excel graphed, and overlaid with each subsequent time step data. Figure 3 10 shows RR01R as an example of a site graphed and overlaid in Excel with the three time steps. Other bank erosion variables and characteristics were recorded on the respective BEHI measurem ent forms (Figure A 51 ).

PAGE 39

39 Kissimmee River Bank Erosion Hazard Index (BEHI) The Bank Erosion Hazard Index (BEHI) (Rosgen, 1996, 2001 ) is an integrated process involving data collected and categorized solely on field observations. Rosgen (2001) has listed categorized percentages for the four main BEHI characteristics, which are root depth, root density, surface protection, and bank angl e (degrees). Table 3 2 shows the four main categories with index ratings for each percentage category. Stratification of soil layers, particularly if it is stratified near the bankfull height, indicates high potential of erosion so the number of layers cor responds to an index value that is directly related to the number of layers. Uniform banks add one point, whereas banks composed of 5 layers will add 5 points, and so on and so forth. Rosgen (2001) suggests this because banks composed of several different lithological layers have an inherent non cohesiveness to them, thus being more likely to fail. In addition to stratification, Table 3 3 shows index values for the composition of bank materials. The index values for each bank are summed and given a categori cal rating based on the total score. Then streambank is then evaluated as having very low, low, moderate, high, very high, or extreme erosion potential. The operating and evaluating procedure for the six variables are: Root depth This is a ratio of root depth to bank height as a percentage. When viewing the entire bank face, it is the percentage of roots extending to the bottom of the bank profile out of the entire bankfull height. Roots not extending to the bottom of the bank may cause undercutting of th e bank and subsequent failure. Root density Observation based assessment of the percentage proportion of roots covering the bank surface and face. The greater the root density, the more stabilization it has on the bank and the less likely it is to fail. Surface protection Percentage of streambank surface covered by vegetation. This includes trees, shrubs, grasses, bushes, weeds, etc. The more surface protection, the more stability is given to the bank.

PAGE 40

40 Bank angle This is calculated in Excel using the graphed bank profile measu rements. This is calculated using E quation 3 1: Bank Angle = [Tan(Bank Height/Horizontal Distance at Bank Height) 1 ]*100 (3 1) Generally, the steeper the angle, the more likely it is to give way to mass wasting due to gravitational forces and shear stress. The more obtuse the angle, the more likely it is that undercutting is occurring and cantilever failure is imminent. However, this is complex because there is a relation between sediment size and cohesiv eness, with the bank angle and possible cantilever formation. Soil stratification Layers in the bank profile face can cause weakness and instability in the bank and increase the potential for bank erosion. The number of layers is the number of points add ed to the index. Bank material adjustment As previously stated, different soil types are more prone to erosion than others, therefore, a thorough grain size and sediment type analysis must be done for the streambank. For example, banks composed of sand f are worse than banks composed mostly of silt/clay. This was done by driving a 1L Shelby tube into the thickest layer, typically in the middle, and extracting it from the bank and deposited in a Ziploc bag, then transferring it back to the lab for analysis. For each bank, only one sample was taken, consisting of the sediment type most represented in the bank face. Table 3 3 has listed index values for different bank material composition. In addition to the main six characteristics taken into account for the BEHI model, the forms for each bank are also formatted to record exact location, general vegetation species land cover, a list of site factors possibly witnessed at the site, evidence of stability, and site geographic background. The alternate side of the form is oriented to show the bank profile measurements as well as a quick visual sketch of the site for accountability purposes. A picture of each site was taken at every measurement time, and the original fifty pictures were uploaded to the geo database in ArcGIS. See Figure A for the Kissimmee River BEHI and profile measurement form. Bank Sediment A nalysis Laboratory analysis on the bank sediment was performed following initial installation of the fifty erosion sites in November of 2010. Each site underwe nt three

PAGE 41

41 separate lab analyses for bulk density, grain size, and loss of ignition (LOI). Soil bulk density is an indicator of soil compaction, and reflects the stability of the soil in relation to water movement and soil aeration (Blake and Hartge 1986). H igh bulk densities typically indicate soils more prone to erosion such as sand and gravel, whereas lower bulk densities indicate soils rich in silt/clay, and are therefore, more stable. A grain size analysis is essential in determining the geologic makeup of the streambank and, more importantly, the percentage of silt/clay embedded within the bank. High silt/clay content coupled with active erosion can accelerate pollutant transport downstream by the chemical adsorption properties of silt and clay particles A grain size analysis also shows how much of the bank profile is sand, a sediment type highly prone to erosion. Loss on ignition is used as an estimate of organic matter. Soil bulk density proceeded as follows. A liter of soil must be oven dried in a met al pan at 110 C for 8 hours. The weight of the soil is then recorded and bulk density calculated in Excel. The bulk density equation is: Bulk Density = (Weight of dried soil/1000) = g/mL (3 2) The dried sediment was then put in a bag and stored until g rain size analysis could be performed. The grain size analysis proceeded as follows. Before measuring out a sample of approximately 150g, a sample splitter was used to appropriately split up the large sample evenly. At least 150g of sediment was used for the grain size analysis. Table 3 3 shows, in order from the stacked sieves, top to bottom, the sieve numbers and millimeters used in the procedure. After the sample was weighed, a mortar and pestle was used to break it down as much as possible into individ ual grains. The weighed sample was then put into the top sieve and secured in the Ro Tap, and then

PAGE 42

42 shaken for fifteen minutes. To prevent dust from escaping, a wait period of 10 minutes existed between the stop of the Ro Tap and the weighing of remaining s oil in each sieve. Repeat shakes were sometimes permitted due to sticky, stubbornly cohesive organic material present in some of the sites. The weight of sediment in each sieve was then input in Excel and calculated out to give a distribution of grain size Loss on ignition was calculated using the method from Wilke (2005). First, crucibles must be heated in a muffle furnace to 550 C for 20 minutes and then dried and to room temperature in a desiccator, and then weighed. Then 10 20 g of sieved sample is we ighed and recorded and placed in the cool muffle furnace and then slowly ignited to 550 C for 3 4 hours. Afterward, the crucibles are carefully taken out and cooled for 1 hour in a desiccator and then weighed twice. The loss of mass after ignition is: m = m 1 m 2 (3 3) Where: m 1 = mass of dried soil and crucible (g) m 2 = mass of soil ignited to 550 C and crucible (g) The loss on ignition (LOI) is calculated by: m/m d )*100 (3 4) Where: m = loss of mass after ignition (g) m d = mass of original dried soil (g) Results are then input into Excel and evaluated. Areal Change and Lateral R etreat This study was executed over a period of nine months with three consecutive measurements, spaced approximately 4 5 months apart, for each bank erosion site. With the profile plotted in an Excel graph, including each successive profile overlaid, a

PAGE 43

43 pattern of either + or areal change (cm 2 ) can be interpolated and calculated. For each time series, areal change in Excel was cal culated using the standard equation for the area of irregular polygons placed on a coordinate system: A = I (x 1 y 2 + x 2 y 3 + x 3 y 4 n y 1 ) (x 2 y 1 + x 3 y 2 + x 4 y 3 1 y n ) I / 2 (3 5) Assuming the polygon is on a coordinate system with vertices (x,y). P olygon is traced in a clockwise direction. With each successive + or areal change, total gross change and net change for each time series can be approximated. This gives an exact quantification to the amount of sediment being lost due to erosion. This figure is also used to calculate the average lateral retreat of the bank over time from erosion. Using the height of the larger profile for each time series, the average lateral re treat can be calculated using Equation 3 6: Lateral Retreat (cm) = [Net Are al Change (cm 2 )]/[Bank Height (cm)] (3 6) These calculations are necessary to the overall evaluative process of the BEHI system and to pinpoint exact hazardous locations and reaches susceptible to erosion. Geographic Information Systems (GIS) GIS is use d in the study mainly as a tool to evaluate where the most erosion is taking place spatially in light of the entire study area. A visual representation is helpful and the geodatabase associated with the study can be overlaid with newer images and updated a s the sites are re evaluated over time. The pictures of each site are uploaded into the geodatabase as a raster field so when the feature class of the erosion sites is viewed in ArcMap, a photograph of the bank profile appears adjacent to the point. Once t he BEHI is evaluated, a map can be created in ArcGIS that gives a colored rating corresponding to the BEHI category of each site, giving a thematic visual representation of erosion potential on the Kissimmee River.

PAGE 44

44 Radius of Curvature/Channel Width M easure ment GIS can also be used to calculate values of variables that are rendered difficult to measure in the field, such as the radius of curvature ratio. There is a relatively simple way to calculate this in GIS. At each site, using the circle drawing tool, I expanded the circle until the curve of the circle fit the curve of the centerline of the meander bend. Then, using the distance measure tool (m), I drew and measured a straight line to obtain an approximate diameter value, and then halved it for the radi us. Then again, using the distance measure tool, I measured the width of the channel at each bank erosion site using 2010 imagery (FDOT). The radius of curvature equation is as follows: Radius of Curvature = ( r m / W m ) (3 7) where r m = Radius of circle (m) corresponding the meander bend curvature W m = Channel width (m) from specific bank erosion site As expected, not every site is situated on a meander bend. There are several sites located along straight reaches and become a little trickier to calculate. This simply requires much larger circles and corresponds to higher ratios. Ratios in the area of 5 or higher are more likely located in straight reaches or very large and wide meander areas. Lower ratios are indicative of more tightly wound, sh arply curved meander bends. Stage Height and Discharge Hydrograph The study area can be explained hydrologically through two gage sites, PC62 and PC33. PC62 is located within the northern portion of Montsdeoca Run while PC33 is located near the Micco Bluff shelter at the southern end of the study area (Figure 3 1). The i mportance of the stage height hydrograph cannot be understated. The S65A gate releases water periodically, depending on the water levels in Lake Kissimmee to the north, thus affecting the upstream reaches with fluctuating water levels. The S65C

PAGE 45

45 gate, furth er south, stabilizes stage height and discharge around the Micco Bluff area and the reaches on the southe rn end of the study. Figure 3 11 displays the stage height and discharge for both the upstream (PC62) and downstream (PC33) gage sites, respectively. S tatistical Analysis Point Biserial Correlation C oefficient A portion of this study incorporates understanding which major site factors are statistically linked, if at all, to increased totals of gross ( ) areal change (cm 2 ) in the profile and gross ( ) la teral migration (cm) of the bank. Gross and net values were used as the dependent variables instead of net values because they are the direct indicators of sediment loss and erosion directly from the bank at the specific measurement date. Nine different si te factors were observed and recorded for the study. To see if the different occurrence/nonoccurrence of factors are statistically linked to erosion totals, a point biserial correlation coefficient ( r pb ) was applied to a Pearson correlation test and then a t test to see if the two are statistically linked. Areal change and lateral migration are both parametric, ordinal variables. Each site factor is a discrete dichotomy variable of two levels coded 1 if the site factor was observed, and 0 if not. The equati on for the point biserial correlation coefficient is: (3 8) where s n = standard deviation used for every member of population M 1 = mean value on continuous variable X for group 1 M 0 = mean value on continuous va riable X for group 2 n 1 = number of data points in group 1

PAGE 46

46 n 0 = number of data points in group 2 n = total sample size To run this statistical test, the Excel data were uploaded into SPSS Statistics 19 and coded for the various dichotomous ind ependent variables. A bivariate correlation test independently for both gross areal change and lateral migration vs. each of the independent erosion processes. Values in Pearson range from 1 to 1, with values close to 1. Values close to 1 describe variab les that are in perfect concordance, and the opposite for values close to 1. SPSS automatically calculates the t significance test with the Pearson point biserial test and gives statistical significance to values that are below 0.05. The point biserial correlation coefficient test was given for each comparison (Table 3 4). Kruskal Wallis Test The four main BEHI va riables should also be tested for significance in predicting gross areal and net change to see if these four main variables of the test would work well under Kissimmee River conditions. In this case, the dependent variables are again the gross and net eros ion rates classified as parametric and ordinal. The independent variables are the four main BEHI indicators and each of them have six different cases present that may or may not affect erosion rates. I also included, as an independent variable, the number of lithological layers within the bank profile. If the independent variables contain 3 or more independent groups (cases), the Kruskal Wallis test proves appropriate. The test statistic, H is given by Equation 3 9: H = (3 9) where R i = sum of ranks for each group

PAGE 47

47 N = total sample size n i = sample size of particular group The Kruskal Wallis test was run in SPSS 19 with the data range being from 1 6 for the independent variables, coinciding with the six catego ries present in the data. SPSS, from the calculated H value, then gives the significance for the test. The following Kruskal Wallis tests were run (Table 3 5). Other variables were not categorical, dichotomous, or parametric, a nd so geological data and radius of curvature data which displayed non parametric characteristics. Just as before, the variables were loaded into SPSS 19 and run with a sim ( 3 10) where d i = x i y i between the ranks of the two variables n = total sample size Correlation Coefficient and then observed with SPSS for significance (Table 3 6). Pearson Correlation Coefficient There are certain variables that meet parametric assumptions. If both the independent and dependent variable for the bivariate test meets par ametric assumptions, then a simple Pearson Correlation Coefficient can be run to observe any statistical significance. The Pearson Correlation Coefficient equation is calculated using E quation 3 11:

PAGE 48

48 (3 11) Where cov = the covariance of the independent (X) and the dependent (Y) The following bivariate tests were performed using the Pearson Correlation Coefficient in SPSS and observed for significance (Table 3 7)

PAGE 49

49 Figure 3 1. Overview of Lower Kissimmee with all 50 initial bank erosion sites

PAGE 50

50 Figure 3 2 Map of Micco Bluff shelter area erosion sites.

PAGE 51

51 A B Figure 3 3. Micco Bluff Shelter A) April 2010. B) November 2010.

PAGE 52

52 Figure 3 4. Map of Montsdeoca Run south sit es, including the connector channel.

PAGE 53

53 Figure 3 5. Map of Fulford Run sites and the much smaller Strayer Run.

PAGE 54

54 Figure 3 6. Map of River Run #1 sites.

PAGE 55

55 Figure 3 7. Map of UBX Run sites.

PAGE 56

56 Figure 3 8. Map of Montsdeoca North Sites

PAGE 57

57 Figure 3 9. Demonstration of taking bank profile measurements. Figure 3 10. Example graph of bank profile measurements overlaid with the three time steps. Site RR01R. -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) River Run 01R 11/17/2010 5/17/2011 8/22/2011

PAGE 58

58 A B Figure 3 11. Kissimmee River hydrographs for both PC 62 and PC33 for the time period 11/01/2010 08/31/2011. A) Stage Height (m) B) Discharge (m3/s). 10 10.5 11 11.5 12 12.5 13 13.5 14 14.5 15 Stage Height (m) Date PC62 PC33 0 5 10 15 20 25 30 35 40 45 50 Discharge (cms) Date PC62 PC33

PAGE 59

59 Table 3 1. Run and Connector name with number of erosion sites Run/Connector Name Number of Erosion Sites River Run #1 5 UBX Run 8 Montsdeoca Run 15 Montsdeoca Connector Run 2 Fulford Run 6 Fulford Connector Run 4 Strayer Run 2 Oxbow Run 3 Micco Bluff Run 5 Table 3 2. Four main BEHI categories and respective index value. The numerical scores given here differed from those used by Rosgen (2001). BEHI Category Root depth (%) Root density (%) Surface Protection (%) Bank Angle (degrees) BEHI Scores, by category Total Scores, by category Very Low Low Moderate High Very High Extreme 90 100 50 89 30 49 15 29 5 14 < 5 80 100 55 79 30 54 15 29 5 14 < 5 80 100 55 79 30 54 15 29 10 14 < 10 0 20 21 60 61 80 81 90 91 119 > 119 1 2 3 4 5 6 < 6 6 11 12 17 18 23 24 29 30 35

PAGE 60

60 Table 3 3. Grain size analysis sieve numbers and corresponding sieve apertures and aggregate name, in order from stacked top to bottom. Sieve # Sieve Aperture Aggregate Name 4 4.75 mm Fine gravel 6 3.35 mm Very fine gravel 10 2.00 mm Very fine gravel 12 1.70 mm Very coarse sand 18 1.00 mm Coarse sand 35 Medium sand 60 120 230 Fine sand Very fine sand Silt/Clay Table 3 4. Point biserial tests of site factors vs. each site response. Site Factors Site Response Undercutting Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Overhanging vegetation Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Cracking/Mass Wasting/Slumping Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Presence of a Scarp Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Exposed Roots Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Tipping/Fallen Trees Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Alligator Slide Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Invasive Catfish Burrow Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat

PAGE 61

61 Table 3 5. Kruskal Wallis tests of site factors vs. site response Site Factors Site Response Root Depth Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Root Density Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Surface Protection Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Bank Angle Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Number of Lithological Layers Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Table 3 6. factors vs. site response Site Factors Site Response % Silt/Clay Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Bulk Density Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat D50 Median Grain Size Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Radius of Curvature Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat

PAGE 62

62 Table 3 7. Pearson Correlation tests for site factors vs. site response. Site Factors Site Response % Silt/Clay Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Bulk Density Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat D50 Median Grain Size Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Radius of Curvature Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat

PAGE 63

63 CHAPTER 4 RESULTS This study implemented, measured, and evaluated a total of 50 original bank erosion sites placed spatially along nine reaches spanning approximately 18 km of the erosion pote ntial of each site to compare with actual measurements of sediment loss and bank retreat. This will give an idea as to whether or not the Kissimmee River can be rapidly and accurately categorized to a scale of erosion potential. The quantification of erosi on rates will provide an accurate representation of erosion hotspots and further a visual understanding of sediment loss in relation to the full Kissimmee River basin area. Lost Sites The water levels (m) and flow (m 3 /s) for each measurement time aid in e xplaining the loss of certain sites and the magnitude of erosion. Figure 4 1 is a map of the three gaging stations; S65A positioned north of the study location, PC33 positioned within the study location in the middle of Montsdeoca Run, and S65C located at the southern end of the study location. Table 4 1 shows the flow (m 3 /s) and stage height (m) for each site at every measurement date. High stage heights correspond to deep water, especially at some of the bank toes. This prevents easy access to measure an d locate sites that were thought to be lost but may still be intact far below the surface. Inevitably, as with most bank erosion studies, there were several sites that were washed away and lost during the course of the survey. This occurred during both tim e steps, and each time a site was lost it was replaced in the same general area as the original site, with the new GPS location noted. Out of the 50 sites implemented, one was omitted entirely from the study (FS03R) due to bad data. Eleven sites had to be

PAGE 64

64 replaced at different times. The only reaches containing lost sites were UBX, Montsdeoca, and Fulford Runs, with UBX losing the most sites. One site, UBX07, was never found during the second survey, likely due to very deep water and/or temporary sediment b urial. UBX07 was found during the third survey, giving data for that particular site that spans the 9 month length of the study. It is safe to say that the lost sites contain excessive amounts of erosion and bed scouring/filing which led to the point of re placement. Table 4 2 and 4 3 shows gross and net erosion quantities, respectively, trip. The quantities listed are for the replacement sites, not the original, and ther e were no sites that were lost more than one time. Each was recovered and known to have at least one time step of data to give an idea of the change occurring in that particular region. All of the UBX sites were lost during the second survey, coinciding wi th high water levels (Table 4 1) and the rapidly eroding bendway in which the sites are oriented. The USGS gage station at PC33 listed the water level height at approximately 11.95m for the second trip in May, about 1.12m higher than the water level at sit e implementation in November of 2010. The deep water, swift current, and approximate site location made it difficult to find the sites, if at all present, and were subsequently reassigned as replacement sites. This led us to assume those sites to have unde rgone excess erosion, thus leading to the disappearance. The Montsdeoca and Fulford lost sites occurred either during time 1 or time 2, and are also assumed to be affected by excess erosion and mass wasting at the site, leading to replacement.

PAGE 65

65 Site Chara cteristics BEHI Model, Sediment Loss, and L ateral Bank R etreat Of the original fifty sites, thirty eight bank erosion sites were used for this study, spanning approximately 18 km of channel length in the Kissimmee River. Eleven sites were lost and replaced with one omitted altogether during the study, and therefore excluded from the analysis. Over a period of nine months, bank profiling, erosion evidence assessment, and BEHI variable identification was implemented to better characterize each site in terms of erosion potential. Sediment characterization for all sites was also included. Table 4 4 lists each site and its BEHI rating, and gross erosion quantities from the results of the study. Figure 4 2 shows the correlation between the BEHI scores and erosion quantities. None of the sites experienced Extreme or Very Low BEHI ratings. There were 4 sites rated Very High, 11 rated High, 16 rated Moderate, and 7 rated Low. Very High ratings are constrained within the three northern reaches; River #1, UBX, and Mont sdeoca Runs. Alternatively, there are only 2 Low rated banks in those same reaches whereas the rest are all confined to the reaches south of the Montsdeoca Connector. The two most frequently rated banks were High and Moderate. The bank profile graphs cond ucted in excel were the framework for the erosion rates and quantities calculations (Appendix A 1 50 ). The gross areal change ( ) (cm 2 ) is the measurement of sediment loss only. Net areal change (cm 2 ) is the measurement that takes into account sediment adde d to the bank profile through stream transport, upper bank failure and deposition near the toe, or any other mechanism by which sediment can be deposited along the bank profile. Net areal change is calculated by subtracting sediment deposited at the bank p rofile from the gross areal change. The site

PAGE 66

6 6 distribution of gross areal change quantities after sorting in descending order shows disjunction among the sites. There is no absolute order in relating site areal change and reach placement. Figure 4 2 support s the fact of no significance when erosion quantities are correlated with BEHI scores. There is, however, a slight positive trend in higher BEHI scores giving higher erosion rates. Site MN11R, rated Very High, had the highest gross and net change in the ba nk profile at 11709 cm 2 and 11314 cm 2 respectively. Sites MS02R, RR03R, OX02L, and RR02L, rated as Moderate, Very High, Moderate, and Moderate, respectively, closely follow MN11R in gross areal change. Interestingly, three of the five sites with highest a real changes were rated as Moderate. With four different reaches and spatial locations represented in the five highest erosion quantities, the fact that bank erosion is tied to site specific factors and is accelerated at many different parts of the river s ystem, is validated. MB04R, rated Low, lost the smallest amount of sediment, with 405 cm 2 while the rest of the sites in the study area ranged between 1000 9000 cm 2 of gross sediment loss. Net areal change displays a different pattern (Table 4 5). The top five highest loss sites now include both OX02L (Moderate) and OX03L (Moderate), at 9862 cm 2 and 8351 cm 2 MN11R, MS02R, and RR03R are still the three sites that have lost more than 10000 cm 2 of sediment. The Strayer Run sites, ST01R (High) and ST02R (High) are relegated to the bottom five exclusively whereas with gross areal change, they were interspersed among the rest of the sites. Interestingly, these two sites were rated High, yet give so me of the lowest erosion quantities. This clearly indicates that sediment deposition is a dominant driver of bank profile change within the two Strayer Run sites, despite their High BEHI ratings. One site, FF05L (Low), had a net change value of

PAGE 67

67 1369 cm 2 indicating that sediment deposition occurs at a larger scale than bank erosion. Bank retreat takes into account the distance the bank has laterally migrated over time due to the forces of erosion. Even though it is directly related to the areal change and bank height, bank retreat can tell a different story of erosion by explaining how much the channel is migrating and in which direction the channel is shifting as it carves through the floodplain. The two sites in Oxbow Run, OX02 and OX03, both rated Moder ate, give the highest amounts of gross lateral change; at 88.1 cm and 81.1 cm, respectively (Table 4 4). MN11R (Very High) and MS02R (Moderate), two sites highest in areal changes, closely followed the two Oxbow sites. The lowest amounts of gross retreat occurred at MB04R with 4.8 cm, a site also recording the lowest amount of areal change. Fulford Run has three sites, FF03R (Moderate), FF04R (Low), and FF05L (Low), placed in the bottom ten for lowest amount of retreat with 28.3 cm, 15.5 cm, and 12.8 cm, r espectively. The rest of the sites are interspersed between the distribution, ranging between 12.8 cm and 88.1 cm. When sorted in ascending order for net lateral retreat, the values range from 17.5 at FF05L to 87.4 at OX02L (Table 4 5). MN11R and MS02R s till give high values for net retreat, keeping in line with high values and high areal changes. Net retreat is highest in OX02L and OX03L with 87.4 and 80.3, followed by MN11R, FF06L, MS02R, and RR03R, respectively. Certain runs do not claim a significance of high values. The Kissimmee River decreases in bank height (cm) as we move downstream (Figure 4 3). This characteristic can be influenced by the water level height and location to the S65A gate (Figure 1 1), which aids in regulating the flow (m 3 /s) and water level

PAGE 68

68 height (m) for the Kissimmee River. Table 4 1 gives river stage height (m) and flow (m 3 /s) for S65A, PC33, and S65C, for each date of inspection. Evidence of E rosion During the first trip in November of 2010, we noted a list of predominant sit e factors for each bank erosion site. The list covered everything that could be seen as contributing to the erosional process overall or is a result of bank weakening processes and shear stress exerted on the bank by the flow of water. Invasive catfish bur rows, characterized by long circular burrows hollowed into the bank face and thought to contribute to sediment loss, was added to the list because a large population of banks were observed to have been hollowed out to some extent by the burrows. The large population of alligators in the Kissimmee are constantly climbing and sliding up and down the short steeped banks for an easy entry into the river. We noted this is as well because it weakens the bank and makes it more susceptible to erosion. Out of the 38 sites observed consistently over the time span, the three clearly most dominant site factors and characteristics were cracking/mass wasting/slumping, presence of a scarp, and invasive catfish burrows. Table 4 6 shows the average amount of evidence noted for site location of either bendway or straight reach. Values are (n=18) and straight re ach types (n=20), and an almost similar display of erosional evidence. Undercutting is clearly more prominent in bendways than in straight reaches, with values of 0.44 and 0.15, respectively. There is more mass wasting occurring in straight reaches (0.80) than in bendways (0.67), and the invasive catfish seem to like to burrow in straight reaches (0.60) more than in bendways (0.33).

PAGE 69

69 Sediment C haracteristics There are certain sediment characteristics that are known to facilitate bank erosion more easily tha n others. Soil layers in banks containing larger sized, more coarse grained soil types serve as a pathway for water seepage, further weakening the bank and making it more susceptible to failure (Simon et al. 1999). After initial collection of bank material s in situ from the first field trip in November of 2010, several different soil characteristics were assessed in the laboratory. Table 4 7 shows values of % silt/clay, bulk density, D50 grain size, and loss on ignition for each site. Eighty six percent of sites reported values of silt/clay content as less than %10. The percentage of silt/clay gradually displays a slight trend of decreasing values as you move downstream, however the significance is negligible and the dataset seem are heterogeneous in values throughout the study area (Figure 4 4). The silt/clay content ranges from a low of 0.01% to a high of 16.67% (Table 4 7). Higher content of silts and clays correspond to a certain level of cohesiveness of material, which in turn inhibits the normal weaken ing and weathering of banks (Thorne 1999). Coarser grained sediments such as sand are thought to facilitate the weakening and weathering of banks (Thorne and Tovey 1981).Eighty four percent of sites are banks containing 90% or more of sand. Figure 4 5 show s the percentage content of sand among the sites, clearly indicating that sand is the dominant sediment type in the Kissimmee riverbanks. The soil bulk density values do not display a high variance, ranging from a minimum of 0.47 at MN06R to a maximum of 1.72 at UB08R (Table 4 7). Higher bulk density values, around 1.5 2, typically indicate sandy soils, whereas majority clayey soils are manifested in values of 1 1.2, and even sometimes <1 (Hillel 1980). Figure 4 6) shows the spatial distribution of bulk de nsity values, showing the majority falling

PAGE 70

70 between 1.2 and 1.8 g/cm 3 The distribution is relatively heterogeneous and there is no clear trend in location of higher or lower bulk density. The D50 grain size (mm) results are not unexpected, and further ill uminate the fact that the Kissimmee River banks are composed mostly of fine grained sand (Figure 4 7). Coarser grained sediments, such as sand, promote sapping and piping within the bank material through hydraulic action, and is inherently more prone to ba nk erosion and failure than its finer grained counterparts, silt and clay. The riverbank stratigraphy can also play a role in facilitating erosion. Thorne and Tovey (1981) report that the layer composed of coarser grained sediments lead the bank to erode a t a higher frequency and magnitude than other layers composed of finer grained sediments. Figure 4 8 shows the number of lithological layers in the bank composition and their effect on erosion quantities. In nearly every instance with gross (4 8a) and net (4 8b) areal change and gross (4 8c) and net (4 8d) bank retreat, banks composed of three different layers tend to contribute to higher erosion rates. The majority of bank strata in this study were composed of sand, and sand typically represented the large st extent of an individual layer within the bank. That being said, only one soil sample was extracted from each site, not one sample from each layer. The sedimentologic variables truly only describe the characteristics of the largest layer in the bank comp osition. Sta ge Height and Discharge Pearman et al (2010) has stated that follwing Tropical Storm Fay in August of 2008, the floodplain was entirely inundated and bank erosion on the scale of meters in change was observed. Therefore, Figure 4 9 displays s tage height and discharge in context with this study to the past four years for comparison with the stage height and

PAGE 71

71 discharge following Tropical Storm Fay. A qualitative look at the two sites show discharge (Figure 4 9b) to be similar in pattern for both sites yet for stage height (Figure 4 9a), PC62 wildly fluctuates while PC33 remains relatively stagnant. Flow duration curves for both sites for the period of 01/01/2008 01/01/2012 are displayed in Figure 4 10. Pearman et al. (2010) noted that when disch arge exceeds approximately 50 m 3 /s there is overbank flow. According to the flow duration curves for the past four years, this occurs 15% of the time for PC62 and 18% of the time for PC33. The study period never observed overbank flow. Since bank height de creases downstream yet discharge has been slightly higher for PC33 with little stage variation, then the channel downstream is wider and deeper than in the upstream reaches near River #1 and UBX Runs. Statistical Analysis The statistical analysis was imp lemented only on the 38 sites that reported consistent measurements for each trip. These sites retain the original toe pin installed in November of 2010, and will continue to be evaluated for change. A point biserial Pearson correlation coefficient was use d to determine if certain site factors affect gross sediment loss in the Kissimmee River. From the 16 comparisons in relation to both gross and net areal change, and gross and net bank retreat, only 2 scenarios came back significant (Table 4 8). It is interesting how very few site factors turned out to be significant in producing high or rather low bank erosion rates, possibly due to the time period being less than a year, alternative to the multi year studies established on bank erosion (Sass 201 1; Pearson 2006; Hudson 2011; Van Eps 2009). A presence of a scarp proved to be by far the most significant factor on both accounts, even significant at the 99% level with a significance value of 0.006 and 0.009 for gross and net areal change and gross ban k

PAGE 72

72 retreat, respectively (Table 4 8). It also proved significant at the 95% level for net bank retreat. This shows that when a scarp is present, it is very likely that large amounts of erosion are taking place at the particular site. The next variable remot ely close was evidence of cracking/mass wasting/slumping with a moderate correlation of 0.320, at a significance level of 90% with a sig value of 0.080. Undercutting of banks proved significant at the 90% level. Alligator slides also proved significant, h owever, due to the fact that only one site was observed to have an alligator slide, these results are negligible. All other variables displayed varying degrees of weak correlation values, with some displaying negative correlation values. Most of them are s tatistically meaningless. The Kruskal Wallis statistical test was performed on the four main BEHI variables vs. the gross and net erosion quantities. Surface protection, root depth, and the number of lithological layers in the bank composition came back wi th varying degrees of significance (Table 4 9) Surface protection proves significant at the 90% level for both gross and net areal change, but is negligible statistically for bank retreat. Root depth proves significant at the 90% level for net areal change 95% level for the gross areal change, and at the 90% level for net bank retreat. The number of lithological layers in the bank composition statistically affects erosion, proving significant at the 90% level for net areal change, and at the 95% for both g ross and net bank retreat. All other values came out, unexpectedly, as statistically meaningless. The initial Kruskal Wallis test does not give an explanation for which category of the independent variable explains best the dependent variable outcome. Pos t hoc tests must be performed. For surface protection, ratings of 4, 5, or 6 (Table 3 2) prove to be the most significant in facilitating erosion. High surface protection ratings correspond to

PAGE 73

73 lower percentages of vegetation, therefore these results are su pported by the literature which states that the less amount of surface vegetation there is, the more easily the bank is eroded due to the lack of tensile strength and cohesiveness within the soil that increased vegetation provide (Rosgen 2001). For root d epth, ratings of 5 and 3 seem to be the most significant at weakening the bank. Roots provide cohesion of the soil and can aid in diverting flows of water from continuing to weaken and weather the bank (Gyssels and Poeson 2003, Thorne 1990). High root dept h ratings correspond to a lower percentage of the bank that they extend into. A rating of 3 and 5 can be interpreted as roots that extend to less than half of the bank height. Therefore, banks with roots extending to less than half of the bank height have a greater effect on erosion than roots that extend further. Riverbanks composed of three lithological layers have the most significant effect on erosion than any other, proving significant in each post hoc test. applied to non parametric variables that describe geological characteristics and also a planform variable. I also wanted to test the correlation between the BEHI total score and the erosion quantities to check for statistical significance. Table 4 10 give the individual tests, of which none came out to be significant. This was unanticipated, and I even thought maybe there was a discrepancy in the methodology and lab work in getting the % silt/clay and bulk densi ty values. I went ahead and performed a large negative correlation ( 0.545) came up at with a very high significance value (0.000). This means that the lower the silt/ clay content, the higher the bulk density

PAGE 74

74 values for sample, which follows conventional thought that high sand content within a sample corresponds to higher bulk density values (Hillel 1980). Geographic Information Systems (GIS) GIS was used in this study to measure and evaluate the radius of curvature/channel width of each site. Table 4 7 gives the radius of curvature/channel width value for each site. Straight reaches will have a larger radius of curvature/channel width value then sharply curved bendways (Gregory and Walling 1973). There have been many studies on the amount of erosion caused by the radius of curvature values. Studies commonly report that radius of curvature values between 2.0 and 3.0 are most effective at eroding banks and restructuring a nd predicting planform changes (Begin 1981; Nanson and Hickin 1986; Hooke 1997; Thorne et al. 1997). More recently, Hudson and Kesel (2010) have reported values between 1.0 and 2.0 are most effective at eroding the banks of the Lower Mississippi River. Ero sion rates for each site with the corresponding radius of curvature value are plotted in Figure 4 9. The consistency of high erosion values correlate well with values between 1 and 2.5, with the largest amount of erosion occurring between 1.0 and 1.5. This indicates that erosion is highest among sites situated within sharply curved bendways. Hudson and Kesel (2010) have shown that the highest erosion within the Lower Mississippi River does not rely solely on the radius of curvature/channel width value, but rather on the spatial distribution of flood plain deposits, some of which help provide resistance to bank erosion and retreat. In the Kissimmee River, it seems as though the radius of curvature may play a larger role in explaining erosion quantities becaus e of the localized flow of water being higher and more forceful in certain geographic areas of the study rather than generally equalized throughout the channel length.

PAGE 75

75 Table 4 1. Kissimmee River gates with stage height (m) and flow (cms) for each time ins pection. Gate/Data type 11/17/2010 4/21/2011 5/17/2011 8/23/2011 S65A Stage (m) S65A Flow (cms) PC33 Stage (m) PC33 Flow (cms) S65C Stage (m) S65C Flow (cms) 14.2 6.2 10.8 6.1 10.8 3.0 14.2 38.8 12.0 33.2 10.4 24.4 14.1 25.4 11.6 24.9 10.3 24.3 14.1 29.1 11.6 24.7 10.7 17.5 Table 4 2. Lost sites with corresponding gross changes for recorded time steps. Site Name Gross Areal ( ) Change Time 1 (cm 2 ) Gross Areal ( ) Change Time 2 (cm 2 ) Gross Bank Retreat Time 1 (cm) Gross Bank Retreat Time 2 (cm) UB03R UB04R UB05L UB06L UB07L MN02R MN03L MN04R MN08L FF01L FF02L LOST LOST LOST LOST LOST 7655 LOST 1891 3008 LOST 4260 2079 9098 3244 1595 1406 LOST 1679 LOST LOST 777 LOST LOST LOST LOST LOST LOST 47.0 LOST 12.1 16.4 LOST 43.0 19.3 67.4 49.9 9.1 13.3 LOST 11.1 LOST LOST 7.8 LOST

PAGE 76

76 Table 4 3. Lost sites with corresponding net changes for recorded time steps. Site Name Net Areal ( ) Change Time 1 (cm 2 ) Net Areal ( ) Change Time 2 (cm 2 ) Net Bank Retreat Time 1 (cm) Net Bank Retreat Time 2 (cm) UB03R UB04R UB05L UB06L UB07L MN02R MN03L MN04R MN08L FF01L FF02L LOST LOST LOST LOST LOST 7655 LOST 709 2938 LOST 4260 2079 9097 3244 1173 1406 LOST 900 LOST LOST 588 LOST LOST LOST LOST LOST LOST 47.0 LOST 4.5 16.1 LOST 43.0 19.3 67.4 49.9 6.7 13.3 LOST 5.9 LOST LOST 5.9 LOST

PAGE 77

77 Table 4 4. Gross quantities of erosion for each site. Time 1 is Nov. 20210 approx. April/May 2011. Time 2 is April/May Aug. 2011 Site Name Gross Areal Change (cm 2 ) Time 1 Gross Areal Change (cm 2 ) Time 2 Gross Bank Retreat (cm) Time 1 Gross Bank Retreat (cm) Time 2 Total Gross Areal Change (cm 2 ) Total Gross Bank Retreat (cm) BEHI Rating RR01R 1211 2645 9.2 20.7 2856 29.9 High RR02L 4249 5139 24.9 30.1 9388 54.9 Moderate RR03R 8142 2320 48.5 13.7 10462 62.2 Very High RR04R 1934 2930 14.4 21.9 4864 36.3 Moderate RR05L 2338 4600 15.5 37.1 6938 52.6 Very High UB01L 2035 3355 13.7 22.4 5390 36.0 Low UB02R 2286 973 14.4 6.7 3259 21.0 Very High UB08R 1310 4870 6.0 22.1 6180 28.1 Moderate MN01R 5541 3044 38.5 20.6 8585 59.0 High MN05L 3649 2113 27.0 14.9 5762 41.9 High MN06R 2937 3001 20.1 20.1 5938 40.3 High MN07L 4510 2730 24.8 13.3 7240 38.1 Low MN09R 1760 4852 10.4 28.0 6612 38.4 Moderate MN10R 1859 2071 15.4 17.1 3930 32.5 Moderate MN11R 7508 4201 48.0 26.6 11709 75.0 Very High MN12L 1655 3807 18.0 37.7 5462 55.7 High MN13R 779 560 7.5 5.3 1339 12.8 High MN14R 1060 3356 7.8 24.7 4416 32.5 Moderate MN15R 3510 1143 26.6 8.7 4653 35.3 High MS01R 2802 2170 16.7 12.9 4972 29.6 High MS02R 8812 2723 50.7 14.8 11535 65.4 Moderate FF03R 2577 341 25.0 3.3 2918 28.3 Moderate FF04R 910 407 10.1 5.4 1317 15.5 Low FF05L 150 1025 1.8 11.0 1175 12.8 Low FF06L 5716 695 53.4 9.9 6411 63.3 Moderate FS01L 882 2640 5.2 14.4 3522 19.5 Low FS02L 5477 3004 34.9 18.8 8481 53.7 Moderate FS04R 2360 1845 21.1 18.5 4205 39.5 Low ST01R 474 3700 3.8 28.0 4174 31.8 High ST02R 1300 560 14.8 5.7 1860 20.5 High OX01R 142 2791 1.6 34.0 2933 35.7 Moderate OX02L 4704 5158 42.0 46.1 9862 88.1 Moderate OX03L 4525 3826 43.9 37.1 8351 81.0 Moderate MB01L 29113 750 29.4 6.9 3663 36.4 Moderate MB02L 1613 1135 10.1 7.1 2748 17.2 Moderate MB03L 428 3684 3.1 26.9 4112 30.0 Moderate MB04R 305 100 3.6 1.2 405 4.8 Low MB05R 4327 0 31.8 0.0 3991 31.8 High

PAGE 78

78 Table 4 5. Net quantities of erosion for each site. Time 1 is Nov. 2010 approx.. April/May 2011. Time 2 is April/May Aug. 2011 Site Name Net Areal Change (cm 2 ) Time 1 Net Areal Change (cm 2 ) Time 2 Net Bank Retreat (cm) Time 1 Net Bank Retreat (cm) Time 2 Total Net Areal Change (cm 2 ) Total Net Bank Retreat (cm) BEHI Rating RR01R 936 2645 7.1 20.7 3581 29.9 High RR02L 3037 2809 17.8 16.4 5846 54.9 Moderate RR03R 7976 1965 47.5 11.3 9941 62.2 Very High RR04R 754 1935 5.7 14.4 2689 36.3 Moderate RR05L 2199 4600 14.6 37.1 6799 52.6 Very High UB01L 2005 3225 13.5 21.5 5230 36 Low UB02R 2286 403 14.4 2.8 2689 21 Very High UB08R 3765 4335 17.1 19.7 570 28.1 Moderate MN01R 5117 3004 35.5 20.3 8121 59 High MN05L 3649 1608 27 11.3 5257 41.9 High MN06R 1897 2862 13 19.2 4759 40.3 High MN07L 3930 1536 21.6 7.5 5466 38.1 Low MN09R 399 4432 2.3 25.6 4033 38.4 Moderate MN10R 1710 1911 14.1 15.8 3621 32.5 Moderate MN11R 7198 4116 46.4 26.1 11314 75 Very High MN12L 1622 3807 17.6 37.7 5429 55.7 High MN13R 659 450 6.3 4.3 1108 12.8 High MN14R 905 3356 6.7 24.7 4261 32.5 Moderate MN15R 3510 943 26.6 7.1 4453 35.3 High MS01R 2276 1970 13.6 11.7 4246 29.6 High MS02R 8812 2146 50.6 11.7 10958 65.4 Moderate FF03R 2577 405 25 3.9 2173 28.3 Moderate FF04R 910 382 10.1 5 1292 15.5 Low FF05L 2394 1025 28.5 11 1369 12.8 Low FF06L 5716 691 53.4 9.9 6407 63.3 Moderate FS01L 658 1985 3.9 10.8 2643 19.5 Low FS02L 4804 1112 30.6 7 5915 53.7 Moderate FS04R 2000 1845 17.9 18.5 3845 39.5 Low ST01R 2281 3384 18.1 25.6 1103 31.8 High ST02R 931 620 10.6 6.3 311 20.5 High OX01R 158 2791 1.8 34 2633 35.7 Moderate OX02L 4704 5083 42 45.4 9787 88.1 Moderate OX03L 4445 3826 43.2 37.1 8271 81.1 Moderate MB01L 2788 390 28.2 3.6 3178 36.4 Moderate MB02L 1552 348 9.7 2.2 1204 17.2 Moderate MB03L 208 2204 1.5 16.1 2412 30 Moderate MB04R 305 100 3.6 1.2 405 4.8 Low MB05R 4327 336 31.8 2.5 3991 31.9 High

PAGE 79

79 Table 4 6. Mean occurrence of evidence of erosion per reach location. n = 18 for Bendway; n = 20 for Straight Site Factors Bendway Straight Reach Undercutting 0.44 0.15 Overhanging vegetation 0.22 0.25 Cracking/Mass Wasting/Slumping 0.67 0.80 Presence of a Scarp 0.67 0.70 Exposed Roots 0.39 0.30 Tipping/Fallen Trees 0.17 0.15 Alligator Slide 0.00 0.05 Invasive Catfish Burrow 0.33 0.60

PAGE 80

80 Table 4 7. Mean values of geological variables per run name. Site Name % Silt/Clay D50 Grain Size (mm) Bulk Density (g/cm 3 ) Loss on Ignition (% Weight) No. Lithological Layers Radius of Curvature/ channel width RR01R RR02L RR03R RR04R RR05L UB01L UB02R UB08R MN01R MN05L MN06R MN07L MN09R MN10R MN11R MN12R MN13R MN14R MN15R MS01R MS02R FF03R FF04R FF05L FF06L FS01L FS02L FS04R ST01R ST02R OX01R OX02L OX03L MB01L MB02L MB03L MB04R MB05R 5.60 1.69 16.76 5.47 3.49 0.16 16.55 4.41 0.01 8.79 5.77 4.00 0.66 5.68 9.10 7.32 8.34 4.24 11.41 4.51 9.07 7.23 5.08 2.91 4.75 2.21 3.68 3.99 12.37 12.36 6.98 1.99 0.47 0.49 2.83 7.21 1.05 3.04 0.16 0.13 0.13 0.12 0.15 0.38 0.08 0.17 0.65 0.22 0.30 0.14 0.18 0.15 0.15 0.16 0.17 0.16 0.15 0.16 0.13 0.12 0.17 0.23 0.14 0.20 0.22 0.19 0.12 0.12 0.14 0.20 0.17 0.10 0.06 0.09 0.24 0.06 1.53 1.49 0.90 1.27 1.41 1.46 0.56 1.72 1.50 1.49 0.47 1.07 1.36 1.36 0.85 0.89 0.93 1.32 0.72 1.64 1.23 1.41 1.41 0.90 0.96 1.54 1.50 0.96 1.16 1.15 0.89 1.47 1.40 1.51 1.53 1.39 1.31 1.21 3.51 0.47 11.93 2.66 3.01 7.25 15.25 3.56 12.36 3.78 48.56 4.52 7.78 5.81 21.71 35.95 23.63 11.45 30.56 2.44 2.93 2.06 24.58 1.74 26.45 1.53 10.06 15.85 4.10 4.42 15.83 1.78 0.56 14.83 1.22 14.46 1.72 2.89 2 2 3 3 2 2 2 2 2 3 3 1 2 2 3 3 2 4 3 2 2 2 2 2 2 2 2 1 2 2 1 3 3 2 2 2 1 3 2.00 1.48 2.13 3.30 4.04 2.84 1.42 5.22 1.74 5.79 6.72 4.48 4.50 6.22 1.44 2.00 2.33 5.88 2.86 1.30 1.30 2.00 7.00 7.88 4.76 2.60 2.60 2.60 2.34 2.41 7.34 1.21 1.60 5.80 1.44 1.36 0.72 1.68

PAGE 81

81 Table 4 8. values when a point biserial correlation coefficient is applied. Bivariate tests for gross erosional change vs. evidence of erosion. Independent Variables Gross Areal Change Gross Bank Retreat Net Areal Change Net Bank Retreat Undercutting 0.164 0.248*** 0.19 0 0.019 Overhanging vegetation 0.163 0.008 0.39 0.39 0 Cracking/Mass Wasting/Slumping 0.232*** 0.192 0.129 0.129 Presence of a Scarp 0.407* 0.382* 0.380* 0.361** Exposed Roots 0 .000 0.37 0 0.092 0.042 Tipping/Fallen Trees 0.38 0 0.125 0.106 0.063 Alligator Slide 0.240*** 0.312** 0.226 0.450* Invasive Catfish Burrow 0.83 0 0.024 0.07 0 0.116 Level of significance for 99%, 95%, 90%, are noted with *,**,***, respectively Table 4 9. Kruskal Wallis test chi square values for the BEHI variables vs. erosional change. Dependent Variables Root Depth (% of bank height) Root Density (%) Surface Protection (% Area) Bank Angle (degrees) Number of Lithological Layers Gross Areal ( ) Change (cm 2 ) Net Areal Chan ge (cm 2 ) Gross Bank Retreat (cm) Net Bank Retreat (cm) 8.587 9.924*** 11.766 ** 10.991*** 5.635 2.722 5.018 2.524 10.385 *** 9.258*** 6.548 6.170 5.133 5.240 2.994 3.636 5.727 7.529*** 8.332** 9.446** Level of significance for 99%, 95%, 90%, are noted with *,**,***, respectively

PAGE 82

82 Table 4 10 which none are significant. Level of significance for 99%, 95%, 90%, are noted with *,**,***, respectively Table 4 11 Pearson Correlation Test performed on the parametric variables of bank height (cm) and distance downstream (km) Listed are the values. Level of significance for 99%, 95%, 90%, are noted with *,**,***, respectively. Dependent Variables Bank Height (cm) Distance Downstream from RR01R (km) Gross Areal ( ) Change (cm 2 ) Net Areal Change (cm 2 ) Gross Bank Retreat (cm) Net Bank Retreat (cm) 0.529* 0.301** 0.142 0..063 0.349** 0.260*** 0.193 0.128 Level of significance for 99%, 95%, 90%, are noted with *,**,***, respectively De pendent Variables % Clay/Silt Bulk Density (g/L) D50 Grain Size (mm) Total BEHI Score Gross Areal ( ) Change (cm 2 ) Net Areal Change (cm 2 ) Gross Bank Retreat (cm) Net Bank Retreat (cm) 0.80 0.092 0.76 0.117 0.033 0.056 0.116 0.124 0.146 0.151 0.097 0.106 0.204 0.208 0.166 0.157

PAGE 83

83 Figure 4 1. Kissimmee River map showing the location of the 3 gages, S65A, PC33, and S65C, that were utilized for water level (m) and flow (cms) for each trip. PC33

PAGE 84

84 A B Figure 4 2. Graphs of erosion quantities vs. BEHI scores of sites A) Gross areal change (cm2). B) Net areal change (cm2). y = 152.65x + 2881.3 R = 0.073 0 2000 4000 6000 8000 10000 12000 14000 5 10 15 20 25 30 35 Gross Areal Change (cm2) BEHI Total Scores y = 184.39x + 1370.7 R = 0.0956 -2000.00 0.00 2000.00 4000.00 6000.00 8000.00 10000.00 12000.00 14000.00 5 10 15 20 25 30 35 Net Areal Change (cm2) BEHI Total Scores

PAGE 85

85 C D Figure 4 2 Continued Graphs of erosion quantities vs. BEHI scores of site s. C) Gross bank retreat (cm). D) Net bank retreat (cm). y = 0.8214x + 25.954 R = 0.0466 0 20 40 60 80 100 5 10 15 20 25 30 35 Gross Bank Retreat (cm) BEHI Total Scores y = 1.0281x + 15.616 R = 0.0532 -20 0 20 40 60 80 100 5 10 15 20 25 30 35 Net Bank Retreat (cm) BEHI Total Scores

PAGE 86

86 Figure 4 3. Graph of bank height (cm) vs. Distance downstream of RR01 (km) Figure 4 4. Graph of percentage of silt/clay content vs. distance downstream of RR01 (km) 0 50 100 150 200 250 0 5 10 15 20 Bank Height (cm) Distance downstream of RR01R (km) y = 0.1963x + 7.0241 R = 0.0521 0 2 4 6 8 10 12 14 16 18 0 2 4 6 8 10 12 14 16 18 20 Silt/Clay Content (%) Distance Downstream (km)

PAGE 87

87 Figure 4 5. Graph of percentage of sand content vs. distance downstream of RR01 (km). Figure 4 6. Graph of bulk density values (g/cm 3 ) vs. distance downstream of RR01 (km) y = 0.048x + 91.91 R = 0.0012 60 65 70 75 80 85 90 95 100 0 2 4 6 8 10 12 14 16 18 20 Snad Content (%) Distance Downstream (km) y = 0.0063x + 1.1865 R = 0.0101 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 0 2 4 6 8 10 12 14 16 18 20 Bulk Density of soils (g/cm3) Distance downstream of RR01 (km)

PAGE 88

88 Figure 4 7. Graph of D50 median grain size (mm) vs. distance downstream of RR01 (km). y = 0.0056x + 0.2147 R = 0.0739 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 2 4 6 8 10 12 14 16 18 20 D50 median grain size (mm) Distance downstream of RR01 (km)

PAGE 89

89 A B Figure 4 8. For each site, graph s of the number of lithological layers vs. erosion quantities. A) Gross areal change. B) Net areal change. y = 1375.7x + 2290.7 R = 0.1002 0 2000 4000 6000 8000 10000 12000 14000 0 1 1 2 2 3 3 4 4 5 Gross Areal Change (cm2) Number of lithological layers y = 1680.4x + 616.12 R = 0.134 -2000 0 2000 4000 6000 8000 10000 12000 0 1 2 3 4 5 Net areal change (cm2) Number of lithological layers

PAGE 90

90 C D Figure 4 8 Continued For each site, graph of the number of lithological layers vs. erosion quantities. C) Gross bank retreat. D) Net bank retreat. y = 11.112x + 14.576 R = 0.1441 0 10 20 30 40 50 60 70 80 90 100 0 1 2 3 4 5 Gross bank retreat (cm) Number of Lithological layers y = 12.967x + 3.4569 R = 0.143 -25 -5 15 35 55 75 95 0 1 1 2 2 3 3 4 4 5 Net bank retreat (cm) Number of lithological layers

PAGE 91

91 A B Figure 4 9. Hydrographs for PC62 and PC33 for time period 01/01/2008 01/01/2012. A) Stage Height. B) Discharge. 10 10.5 11 11.5 12 12.5 13 13.5 14 14.5 15 1-Jan-08 1-Mar-08 1-May-08 1-Jul-08 1-Sep-08 1-Nov-08 1-Jan-09 1-Mar-09 1-May-09 1-Jul-09 1-Sep-09 1-Nov-09 1-Jan-10 1-Mar-10 1-May-10 1-Jul-10 1-Sep-10 1-Nov-10 1-Jan-11 1-Mar-11 1-May-11 1-Jul-11 1-Sep-11 1-Nov-11 1-Jan-12 Stage Height (m) Date PC62 PC33 TROPICAL STORM FAY STUDY PERIOD 0 50 100 150 200 250 1-Jan-08 1-Mar-08 1-May-08 1-Jul-08 1-Sep-08 1-Nov-08 1-Jan-09 1-Mar-09 1-May-09 1-Jul-09 1-Sep-09 1-Nov-09 1-Jan-10 1-Mar-10 1-May-10 1-Jul-10 1-Sep-10 1-Nov-10 1-Jan-11 1-Mar-11 1-May-11 1-Jul-11 1-Sep-11 1-Nov-11 1-Jan-12 Discharge (cms) Date PC62 PC33 STUDY PERIOD TROPICAL STORM FAY

PAGE 92

92 A B Figure 4 10. Flow duration curves from 01/01/2008 01/01/2012. A) PC62. B) PC33. 0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 Discharge (cms) Percent of time that indicated discharge was equaled or exceeded 0 50 100 150 200 250 0 20 40 60 80 100 120 Discharge (cms) Percent of time that indicated discharge was equaled or exceeded

PAGE 93

93 A B Figure 4 11. Graphs of erosion quantities vs. radius of curvature rati o. Red box groups together highest values. A) Gross areal change. B) Net areal change. 0 2000 4000 6000 8000 10000 12000 14000 0 1 2 3 4 5 6 7 8 Gross Areal Change (cm2) Radius of Curvature/Channel Width Ratio -2000 0 2000 4000 6000 8000 10000 12000 14000 0 1 2 3 4 5 6 7 8 Net Areal Change (cm2) Radius of Curvature/Channel Width Ratio

PAGE 94

94 C D Figure 4 11 Continued Graphs of erosion quantities vs. radius of curvature ratio Red box shows groups of highest values. C) Gross bank retreat. D) Net bank retreat. 0 20 40 60 80 100 0 1 2 3 4 5 6 7 8 Gross Bank Retreat (cm) Radius of Curvature/Channel Width Ratio -25 -5 15 35 55 75 95 0 1 2 3 4 5 6 7 8 Net Bank Retreat (cm) Radius of Curvature/Channel Width Ratio

PAGE 95

95 CHAPTER 5 DISCUSSIONS AND CONC LUSIONS Bank erosion is a complex process involving many different variables interconnected at different moments and work to weaken and weather the bank to the point of soil erosion. These variables are notoriously active or inactive at any given time, and since erosion occurs sequentially throughout the course of a long temporal study, it is difficult to pinpoint exact site factors taking place at that exact location, furthering the weakening of bank material and facilitating bank failu re. Invariably, these interconnected and enmeshed site factors are virtually limitless in their ability to cause a bank to fail, but nevertheless there are times when an accurate perception of bank erosion can be understood. This chapter will discuss the B EHI model results and its position and significance within the Kissimmee River framework of actual measured erosion quantities and rates; certain variables that affect Kissimmee bank erosion more so than others; possible explanations for the geographic ori gin of large erosion rates; and conclusions and future research that may benefit in understanding more thoroughly bank erosion in the Kissimmee River. BEHI Model and Measured Erosion Rates Throughout the course of approximately nine months, 38 sites within 9 stream quantities. After calculating and evaluating BEHI scores for each site, gross and net areal change and bank retreat within the bank profile were calculated and compare d with the aforementioned BEHI scores and subsequent labeling. The BEHI system works relatively well in categorizing some of the sites and does not work well in categorizing other sites (Table 4 4 and 4 5). Due to the discordance and general

PAGE 96

96 disagreement among values, it is hard to distinguish a threshold of erosion values that relate to the layered BEHI ratings. In terms of gross areal change for example, RR01R is rated High with 3856 cm 2 of sediment loss, whereas RR02L is rated Moderate with 9388 cm 2 of sediment loss. This constant discordance is observed over and over again within every reach and every erosion quantity. Alternatively, the top ten sites with the most sediment loss were rated Very High, High, or Moderate, with most being Moderate. The erosion quantities (Table 4 10). Figure 4 2, however, displays a slight increasing trend in erosion quantities with BEHI total scores. This jumbling and disconnection between ratings and values, coupled with no statistical significance with erosion rates and BEHI values support the fact that reliable way of rating erosion hazards and susceptibility amongst its specific geographic locations. It is worth noting that the BEHI scores only takes into account five variables of bank composition only, instead of taking into account the spatial location of the site and other geomorphic, sedimentologic, planform, and veg etative variables. The accurate labeling of bank erosion sites for the Kissimmee River will clearly need to include more Spatial Variability of Erosion Rates Aside from utilizing the BEHI model as the framework for erosion rate prediction, measured amounts of sediment loss certainly give the most accurate description of bank erosion origin in relation to the geography of the basin. The spatial variability of erosion rates can be tied to the geographic p osition near the S65A gate, the structure that controls the amount of water being discharged into the Kissimmee River from the

PAGE 97

97 Lake Kissimmee Basin further north. Table 4 1 gives the differences in stage height and flow for three stations near the study lo cation. Originally, we expected the northern most reaches to lose the greatest amount of sediment because the area has been most recently restored to its meandering position and will continue to incise its channel until the river is able to establish som e sort of equilibrium of transporting water and sediment. As expected, River Run #1, the northern most reach of this study, contains one of the highest amounts of sediment loss. UBX Run, while for the most part gives moderate values of erosion, lost 5 site s during the course of the study, indicating immeasurable but large amounts of sediment loss. This can be explained through a multitude of different factors. Knighton (1984) states that bank erosion is exacerbated by the force of hydraulic action resulting from high discharges, particularly on non cohesive banks composed of coarser grained sediments such as sand. Therefore, it is impossible to talk about the reasons for excessive erosion in this region and ignore the fact that the geographic position is clo se to the S 65A gate that is used to release excess water stemming from the Lake Kissimmee Basin further to the north. Hydraulic action and fluctuating stage height and discharges in this region are the most likely reason for excess erosion within these r eaches. The stream power exerted from the force of the water release at the S 65A gate likely dissipates as the channel meanders further downstream, thus explaining why the far south reaches like Fulford, Micco Bluff, and Strayer, are not losing as much se diment as the reaches north of the Montsdeoca Connector. The two sites in Oxbow Run clearly are anomalies within the larger framework, as they have been calculated to undergo the most bank retreat and high areal changes. Their position along the outer bend of a sharply curved

PAGE 98

98 meander makes them more susceptible to the force of water at higher discharges. The radius of curvature and other vegetative, geomorphic, and sedimentologic variables that can affect erosion in the Kissimmee are discussed next. Possibl e Influential Variables on Erosion Rates As previously explained, a number of independent acting and sequential variables come into play when explaining the frequency and magnitude of bank erosion in any given area. The Kissimmee is unique in the fact that out of all the various BEHI, geological, planform, and processes variables tested, very few came out with any erosion in the Kissimmee River. However, a very large si gnificance value was noted when the bank contains a fluvial scarp that connects the bankfull height with any surface standing lower in the channel. Even though the channel height and discharge was much greater during the second and third visits, the origin al trip contained low channel height and discharge, and a scarp can be easily observed sloping down to an even further lower flat surface. This indicates that some significant change must have occurred between the pre existing conditions of the banks durin g the development of the floodplain surface and those that produced the fluvial scarp present at the site. In order for erosional terraces to form, lateral erosion must be the dominant process when constructing the tread, which is clearly demonstrated with the high significance value with lateral bank retreat values. Since most erosion is occurring within the reaches of Montsdeoca and to the north, it is fair to make the assumption that these sites also contain more fluvial scarps than the southern reaches; demonstrating the fact that the river is undergoing a period of channel position shifting back and forth across the floodplain, creating the flat alluvial surface that is in effect a mirrored image of the plane

PAGE 99

99 surface of the elevated floodplain. The orig in of the underlying tread, however, is not clearly understood, as most of the site factors described previously did not correlate well at all, except for a suggestive correlation between mass wasting and slumping, a typical predominant form of bank erosio n. The extent of the root depth also plays a role in bank retreat, where roots extending further down into the bank are thought to provide more tensile strength to the profile and effectively aid in inhibiting lateral bank retreat. This proved to be signi ficant following the results of the Kruskal Wallis test, and was expected due to the reasons outlined by Rosgen (2001) in explaining the thought process for the four main BEHI variables. The tensile strength of the roots combined with the compressive stren gth of the soils provide stability within the profile that may not elsewhere be supported under different bank material composition (Genet et al. 2005). In addition to root depth playing a large part in bank protection, the amount of surface protection is also significant for gross and net areal change. More surface protection provides the bank with a buffer to mitigate the effect of hydraulic action on the bank and will also likely include more roots extending further into the bank to reinforce the profile structure. Within the Kissimmee, banks with low surface protection and low root depth are the banks most prone to soil erosion. The results from the bulk density analysis proved to be entirely unexpected. Wynn and Mostaghimi (2006), among others, have f ound that bulk density of soils has been the most significant factor in determining erosion rates within a watershed. The complete randomization of the data when correlated with erosion quantities lacked any sound evidence that bulk density contributes to increased erosion rates. Furthermore, a

PAGE 100

100 correlation between the % silt/clay values and bulk density (Figure 5 1) supports conventional thought that higher bulk densities contain smaller amounts of silt and clay (Hillel 1980). Therefore, the presumption tha t the methods of bulk density and %silt/clay were somehow executed incorrectly or with haphazardness cannot be supported. The Kissimmee River just may be unaffected by bulk density when it comes to susceptibility of erosion. The geology of the bank materi al resulted in even more unexpected results. The Kissimmee is not a very heterogeneous basin geologically speaking, so the fact that most of the banks are composed of some form of sand with little organic material involved elevates the likelihood of erosio n automatically, as sand (and other coarser grained sediments) are more liable to erosion (Knighton 1998). There is also little variation among the D50 grain size (mm) results from the sifting of samples, which will give inconclusive results when correlat ed with erosion rates. The only sedimentological variable that showed any sort of significance was the amount of lithological layers present within the bank profile. Banks composed of three independent layers produced the largest amount sediment loss. The amount of layering may not be the deciding factor; rather it is the thickest and most coarse grained layer that is the one eroding away (Thorne and Tovey 1981). Since the sediment sample taken from each bank was produced from the largest layer, combined wi th every sample containing mostly sand, it is safe to say that the other layers within the bank profile may provide some sort of resistance to erosion, while the sandy layer from the middle to the top of the bank is more easily weathered. Although the bank sedimentology and stratigraphy turned out to be mostly inconclusive and ineffective in providing concrete answers to bank erosion in

PAGE 101

101 the Kissimmee, it seems as though the major evidence speaking to bank erosion stems from the radius of curvature, hydrauli c action and the rapid increase in discharge following release of water from the S65 A gate. Radius of Curvature/Channel Width Meander bend curvature is a planform variable that describes how sharply curved a meander bend is. The relationship can be seen a s representative of how and where flow is being directed and energy is dissipating within the meandering stream (Hooke, 1975). Many studies have shown that maximum erosion rates for meander bends occur when the ratio of the radius of curvature ( r m ) to chan nel width ( W m ) is between 2.0 and 3.0 (Behin 1981; Nanson and Hickin 1986; Hooke 1997; Thorne et al. 1997). The Kissimmee River undergoes higher rates of erosion with radius of curvature values between approximately 1.0 and 2.5, with the highest occurring between 1.0 and 1.5 (Figure 4 9). In the Kissimmee, sharply curved meander bends are more effective at eroding banks than less curved meander bends. While Hudson and Kesel (2010) relate deviating radius of curvature values and the variability of migration rates to complex flood plain deposits scattered through the basin, the Kissimmee River may best be explained by geographic closeness to the S65A gate to the North. UBX Run is geographically close to the S65A gate. UBX has also lost five sites throughout th e course of the study, more than any other run. Those five sites were situated around meander bends with radius of curvature values between 1.0 and 1.5, supporting the assumption that lost sites underwent excessive erosion periods. In fact, River Run #1, U BX, and Montsdeoca, the three runs closest to the gate and shown to have considerably higher stage and flow values (Table 4 1), lost more sites than anywhere else. Only one site was lost south of Montsdeoca. These lost sites were situated around

PAGE 102

102 tightly cu rved meander bends with radius of curvature values between 1.0 and 2.5. The tightness of the meander bend, in addition to being geographically close in proximity to the hydraulic action pulsing from the S65A gate seem to be the greatest indicators of where riverbanks erode the greatest in the Kissimmee River. Sta ge H eight (m) and Discharge (m 3 /s ) Since bankfull discharge never occurred, it is difficult to give a definitive answer on soil moisture variables and their effect on erosion rates if the banks and floodplain have never been inundated. Simon et al (1999) has noted that most bank erosion is observed following bankfull discharge when the river is receding and a release of pressure and matric suction weakens and weathers the bank. This can be supported by the greater erosion rates in the northern reaches where the stage fluctuates wildly whereas in the southern reaches the pooling effect caused by the S65C control structure stabilized water levels, preventing them from fluctuating periodically. Pearman et al (2010) also noted that the largest amount of suspended sediment within the channel occurred immediately following Tropical Storm Fey in August of 2008, when discharges exceeded an astounding 135 m 3 /s, more than 4 times the amount observed during th e peak discharge of this study. Also, they noted that bank erosion amounts in the range of several meters (Schenk and Hupp 2010) post Tropical Storm Fey, an amount much larger than anything experienced in this study, indicating that prolonged bankfull disc harge may be necessary for excessive amounts of sediment loss instead of from the direct force of swift flowing water and repeat wetting and drying of banks. Pearman (2010) postulated where the large amount of suspended sediment may be originating from. H e thought that if several meters of erosion was observed during extremely high discharges following intense precipitation fall, than bank erosion may be

PAGE 103

103 the cause of that increase, but only during high flows. Since this study did not experience anything cl ose to the amount of discharge undergone during Fey in 2008, it is still a factor that is yet to be scientifically determined. Further Research In conclusion, it seems as though the BEHI method developed by Rosgen (2001) is moderately reliable at defining a generalized geographic area on a scale of erosion susceptibility. The most accurate prediction of where the largest amount of erosion is occurring is determining whether or not a scarp is present, a feature known to be present in rapidly changing and inc ising channels. High amounts of erosion can also be observed around tightly curved meander bends. This will continue until the meander bend closes off and becomes a cutoff, leaving behind a small oxbow lake (Hooke 1987). The river will then continue to car ve through the alluvial floodplain, forming new meander bends and cutoffs as the channel changes and evolves. Root depth also proved to be reliable in explaining bank retreat, and this can possibly be further investigated if the study decides to take into account the presence of woody or nonwoody vegetation. Woody vegetation may be linked to extent of root depth because the woody vegetation would need more support than the nonwoody vegetation like Hibiscus grandiflora or Ludwiga peruviana two species of no nwoody plants prevalent apply even more tensile strength within the bank composition, resisting the forces of erosion (Sass 2011). The data has also supported the fact that geographic position near the S65 A gate north of River Run #1 (Figure 1 1) causes the banks to be more vulnerable to

PAGE 104

104 quick, episodic high discharge pulses from the loc k, supported from the stage height time series of PC62 (Figure 5 4a). Even though the Kissimmee reports high discharge values as far south as the Micco Bluff area, the stabilizing water levels caused by the pooling effect from the S65C structure help to mi tigate erosion. Extremely high bankfull discharges, such as those experienced immediately following Tropical Storm Fey in 2008, seem to be necessary to create large scale lateral migration of banks across the span of the entire Kissimmee River restored bas in. Just a fraction of that bankfull discharge was the highest, short lived peak experienced within this nine month study. After the last bank profile measurement trip in August of 2011, discharges in the Kissimmee exceeded 20 m 3 /s well into September, wit h much higher discharges exceeding 226 m 3 /s recorded during October 2011. These recorded discharges certainly cause the banks to overflow and spill onto the floodplain. It will be interesting to further this study and see the results of bank erosion from sustained bankfull discharge. Even during the time span of this study, 11 sites were either washed away or buried under bed deposition, and discharges never came close to bankfull. Perhaps a thorough wetting of the floodplain and subsequent recession of f low is required to more accurately predict how severe erosion rates are within the Kissimmee.

PAGE 105

105 Figure 5 1. Correlation of % silt/clay vs. bulk density y = 0.0412x + 1.4625 R = 0.3182 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 Bulk Density (g/cm2) Silt/Clay content (%)

PAGE 106

106 APPE NDIX BANK PROFILE GRAPHS Figure A 1. RR01R Figure A 2. RR02L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) River Run 01R 11/17/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) River Run 02L 11/18/2010 5/17/2011 8/22/2011

PAGE 107

107 Figure A 3 RR03R Figure A 4. RR04R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) River Run 03R 11/18/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) River Run 04R 11/18/2010 5/17/2011 8/22/2011

PAGE 108

108 Figure A 5. RR05L Figure A 6. UB01L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) River Run 05L 11/18/2010 5/18/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) UBX Run 01L 11/17/2010 5/17/2011 8/22/2011

PAGE 109

109 Figure A 7. UB02R Figure A 8. UB03R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) UBX Run 02R 11/17/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) UBX Run 03R 11/18/2010 Series2 8/22/2011

PAGE 110

110 Figure A 9. UB04R Figure A 10. UB05L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) UBX Run 04R 11/17/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) UBX Run 05L 11/17/2010 5/17/2011 8/22/2011

PAGE 111

111 Figure A 11. UB06L Figure A 12. UB07L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) UBX Run 06L 11/17/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) UBX Run 07L 11/17/2010 8/22/2011

PAGE 112

112 Figure A 13. UB08R Figure A 14. MN01R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) UBX Run 08R 11/18/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 01R 11/17/2010 5/17/2011 8/22/2011

PAGE 113

113 Figure A 15. MN02R Figure A 16. MN03L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 02R 11/17/2010 5/19/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toe pin (cm) Distance relative to toepin (cm) Montsdeoca Run 03L 11/17/2010 5/17/2011 8/22/2011

PAGE 114

114 Figure A 17. MN04R Figure A 18. MN05L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 04R 11/17/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 05L 11/17/2010 5/17/2011 8/22/2011

PAGE 115

115 Figure A 19. MN06R Figure A 20. MN07L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 06R 11/17/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 07L 11/17/2010 5/17/2011

PAGE 116

116 Figure A 21. MN08R Figure A 22. MN09R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 08R 11/17/2010 5/17/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 09R 11/17/2010 5/17/2011 8/22/2011

PAGE 117

117 Figure A 23. MN10R Figure A 24. MN11R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 10R 11/16/2010 5/18/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 11R 11/16/2010 5/18/2011 8/22/2011

PAGE 118

118 Figure A 25. MN12L Figure A 26. MN13 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 12L 11/16/2010 5/18/2011 8/22/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 13R 11/16/2010 5/18/2011 8/22/2011

PAGE 119

119 Figure A 27. MN14R Figure A 28. MN15R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 14R 11/16/2010 5/18/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Run 15R 11/16/2010 5/18/2011 8/23/2011

PAGE 120

120 Figure A 29. MS01R Figure A 30. MS02R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Connector 01RB 11/16/2010 5/18/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Montsdeoca Connector 02RB 11/16/2010 5/18/2011

PAGE 121

121 Figure A 31. FF01L Figure A 32. FF02L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height Relative to toepin (cm) Distance relative to toepin (cm) Fulford Run 01L 11/16/2010 5/18/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Fulford Run 02L 11/16/2010 5/18/2011 8/23/2011

PAGE 122

122 Figure A 33. FF03R Figure A 34. FF04R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Fulford Run 03R 11/16/2010 5/18/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Fulford Run 04R 11/16/2010 4/21/2011 8/23/2011

PAGE 123

123 Figure A 35. FF05L Figure A 36. FF06L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Fulford Run 05L 11/16/2010 4/21/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Fulford Run 06L 11/16/2010 4/21/2011

PAGE 124

124 Figure A 37. FS01L Figure A 38. FS02L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Fulford South Connector 01L 11/15/2010 4/21/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Fulford South Connector 02L 11/15/2010 4/21/2011

PAGE 125

125 Figure A 39. FS03R Figure A 40. FS04R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Fulford South Connector 03R 11/15/2010 5/18/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance from toepin (cm) Fulford South Connector 04R 11/15/2010 4/21/2011 8/23/2011

PAGE 126

126 Figure A 41. ST01R Figure A 42. ST02R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 450 Height relative to toepin (cm) Distance relative to toepin (cm) Strayer Run 01R 11/15/2010 4/21/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Strayer Run 02R 11/15/2010 4/21/2011 8/23/2011

PAGE 127

127 Figure A 43. OX01R Figure A 44. OX02L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Oxbow Run 01R 11/15/2010 5/18/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Oxbow Run 02L 11/15/2010 5/18/2011 8/23/2011

PAGE 128

128 Figure A 45. OX03L Figure A 46. MB01L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Oxbow Run 03L 11/15/2010 5/18/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Micco Bluff Run 01L 11/15/2010 5/18/2011 8/23/2011

PAGE 129

129 Figure A 47. MB02L Figure A 48. MB03L -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Micco Bluff Run 02L 11/15/2010 4/22/2011 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Micco Bluff Run 03L 11/15/2010 4/22/2011 8/23/2011

PAGE 130

130 Figure A 49. MB04R Figure A 50. MB05R -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Micco Bluff Run 04R Roots 11/15/2010 Dirt 11/15/2010 Roots 4/21/2011 Dirt 4/21/2011 Roots 8/23/2011 Dirt 8/23/2011 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 160 180 0 50 100 150 200 250 300 350 400 Height relative to toepin (cm) Distance relative to toepin (cm) Micco Bluff Run 05R 11/15/2010 4/22/2011

PAGE 131

131 KISSIMMEE RIVER BANK EROSION HAZARD INDEX (BEHI) AND BANK PROFILE MEASUREMENT FORM Site#: _____________________Date: Personnel: Location of Toe Pin (GPS Coordinates, X): Location of Toe Pin (GPS Coordinates, Y): Pool: Run Name: Land Cover: Site Background (circle): RB/LB; Bendway/Straight; Restored/Recarved/ U. Remnant; Planar/Rotational/Slab/Overhang Photographs Taken: BANK EROSION HAZARD INDEX (Circle one in each category) Bank Height/ Bankfull Height Root Depth (% of BH) Root Density (%) Surface Protection (Avg. %) Bank Angle (degrees) 1.0 1.1 90 100 80 100 80 100 0 20 1.11 1.19 50 89 55 79 55 79 21 60 1.2 1.5 30 49 30 54 30 54 61 80 1.6 2.0 15 29 15 29 15 29 81 90 2.1 2.8 5 14 5 14 10 14 91 119 >2.8 < 5 < 5 < 10 > 119 EVIDENCE OF EROSION (Circle, describe spatial extent and height of features) None Undercutting Overhanging vegetation Cracking / mass wasting / sloughing Presence of a scarp Exposed roots Tipping / fallen trees Trampling features Alligator slide Invasive catfish burrows (density, spatial characteristics) Other EVIDENCE OF STABIL ITY (Circle, describe) None Sediment deposition Stabilizing vegetation Other SOIL LAYERING (texture, color, description from top to bottom), include photos #s ______________________ Layer (t to b) Thickness (cm) Texture Color (Munsell) Uniform/Layers Comments (roots, burrows) Figure A 51. Front of Kissimmee Bank Erosion form with BEHI variables and other field information collected

PAGE 132

132 BANK PROFILE MEASUREMENTS (from toepin) Toepin height above bed______ Height above toepin (cm) Distance from toepin to bank (cm) Comments Identify and measure horizontal pin heights above toepin, below bank, and amount exposed Identify top of bank and comment on changes in lithology. profile, burrowing, etc. SKETCH: Figure A 51 Continued. Back of Kissimmee Bank Erosion form

PAGE 133

133 LIST OF REFERENCES Anderson, D.H., and Chamberlain, J.R. 2005. Impacts of channelization on the hydrology of the Kis simmee River, Florida (Chapter 2). Bousquin SG Anderson DH Williams GE Colangelo DJ (eds). Establishing a Baseline: Pre Restoration Studies of the C hannelized Kissimmee River Technical Publication ERA #432 South Fl orida Water Management District: West Palm Beach, FL. Anderson, D.H., Frei, D., Davis, W.P. 2005. River channel geomorphology of the channe lized Kissimmee River, Florida (Chapter 3). Bousquin SG, Anderson DH, Willia ms GE, Colangelo DJ (eds). Establishing a Baseline: Pre Restoration Studies of the Channelized Kissimmee River, Technical Publication ERA #432 South Fl orida Water Management District: West Palm Beach, Florida. Technical publication ERA #432. Begin, Z.B. 1 981. Stream curvature and bank erosion: A model based on the momentum equation: Journal of Geology 89 : 497 504 Birkeland, P.W. 1984. Soils and geomorphology. Oxford University Press, New York, 14 15. Blake, G.R. and K.H. Hartge. 1986. Bulk Density, in A. Klute, ed., Methods of soil analysis, part 1. Physical and mineralogical methods. Agronomy Monograph no. 9, 2 nd ed., 363 375. Bousquin, S.G., Anderson, D.H., Colangelo, D. J., and Williams, G.E. 2005. Introduction to baseline studies of the channelized Kis simm ee River (Introduction). Bousquin SG, Anderson DH, Williams GE, Colangelo DJ (eds). Establishing a Baseline: Pre Restoration Studies of the C hannelized Kissimmee River Technical Publication ERA #432 South Fl orida Water Management District: West Pal m Beach, Florida, pp. 1 15. Charlton, Ro. 2008. Fundamentals of fluvial geomorphology. Routledge, New York, NY. Davis, S.M. 1981. Mineral flux in the Boney Marsh, Ki ssimmee River. Mineral Retention in Relation to Overland Flow During the Three Year Period Following R eflooding Technical Publication #81 1. South Florida Water Management District: West Palm Beach, Florida. 54 pp. Drever, J.I. 1997. The geochemistry of natural waters: surface and groundwater environments. Prentice Hall, 3 rd ed., 436 pp. Enviro largest water quality problem. Pointer No. 1: EPA841 F 96 004A. http://water.epa.gov/polwaste/nps/outreach/ point1.cfm

PAGE 134

134 Environmental Protection Agency (EPA). 2011. Impaired waters and total maximum daily loads. Section 303(d) of the Clean Water Act. http://water.epa.gov/lawsregs/lawsguidance/c wa/tmdl/ Ferriter, A., Doren, B., Winston, R., Thayer, D., Miller, B., Thomas, B., Barrett, M., Pernas, T., Hardin, S., Lane, J., Kobza, M., Schmitz, D., Bodle, M., Toth, L., Rodgers, L., Pratt, P., Snow, S., and Goodyear, C. 2008. The Status of Nonindigen ous Species in the South Florida E nvironment (Chapter 9) 2008 So uth Florida Environmental Report #9 101 South Florida Water Management District: West Palm Beach, FL. p.9 1 Florida Department of Environmental Protection (FDEP). 2001. Total maximum daily l oad for total phosphorus Lake Okeechobee, Florida. Submitted to: U.S. Environmental Protection Agency, Region IV, Atlanta, GA. Florida Department of Transportation (FDOT). 2010. Florida aerial photography archive collection (APAC). http://www.dot.state.fl.us/surveyingandmapping/apac.shtm Fox, G. A., Wilson, G. V., Simon, A., Langendoen, E. J., Akay, O., and Fuchs, J. W. 2007 Measuring streambank erosion due to ground water seepage: Correlation to bank pore water pressure, precipitation and stream stage. Earth Surface Processes and Landforms 32 : 1558 1573. Genet, M., Stokes, A., Salin, F., Mickovski, S.B., Fourcaud, T., Dumail, J F., et al. 2005. The influence of cellulose content on tensile strength in tree roots. Plant and Soil, 278: 1 9. Goldstein, A.L. 1990. Upland demonstration project water quality impacts of agricultural land use and wetland BMPs in the Kissimmee River valley. Loftin, M.K., To th, L.A., and Obeysetera, J. (eds) Proceedings of the Kissimmee River restoration symposium, October 1988, Orlando, FL South Fl orida Water Management District: West Palm Beach, Florida. P.111 124 Gregory, K.J., and D.R. Walling. 1973. Drainage basin form and processes. A geomorphological approach. Edward Arnold, London. Grissinger, E.H. 1982. Bank erosion of cohesive materials. In Gravel bed rivers, Hey, R.D., Bathu rst, J.C., and Thorne, C.R. (eds) Wiley: Chichester, 273 287 Gurnell, A. 1997. The hydrological and geomorphological signif icance of forested floodplains. Global Ecology and Biogeography Letters 6 : 219 229. Hagerty, D.J. 1980. Multifactor analysis of bank caving along a navigable strea m In: National Waterways Roundtable Proc. U.S. Army Engineering Water Resources Support Cen ter, Institute for Water Resources IWR 80 1 :463 92.

PAGE 135

135 Hickin, E.J., and Nanson, G.C. 1975. The character of channel migration on the Beatton river, northeast British Columbia, Canada: Geological Society of America Bulletin 86 : 487 494. Hillel, D. 1980. Fundamentals of soil physics. Academic Press, New York, 413 pp. Hooke, J.M. 1987. Changes in meander morphology, in Gardiner, V., ed, International geomorphology 1986, Part I : New York, John Wiley & Sons, 591 609. Hooke, J.M. 1997. Styles of channel change in Thorne, C.R., et al., ( eds ). Applied fluvial geomorphology for river engineering and management John Wiley & Sons: New York, New York, 237 268 Hudson, P.F., and R.H. Kesel. 2000. Channel migration and meander bend curvature in the lower Mississippi R iver prior to major human modification. Geology 28 ;531 534 Iverson, R.M., Reid, M.E., Iverson, N.R., LaHusen, R.G., Logan, M., Mann, J.E., and Brien, D.L. 2000. Acute sensitivity of landslide rates to initial soil porosity. Science 290 : no. 5491, p. 513 516. Jones, B.L. 2005. Turbidity and suspended solids concentrations in the river channel. Bousquin SG, Anderson DH, Williams GE, Colangelo DJ (eds). Establishing a Baseline: Pre Restoration Studies of the C hannelized Kissimmee River Technical Publicatio n ERA #432 South Florida Water Management District, West Palm Beach, Florida. Expectation 9. Technical Publication ERA #432. Knighton, D. 1984. Fluvial forms and processes. Edward Arnold Ltd, London, United Kingdom, pp. 58 64. Koebel, J. W. 1995. An histo rical perspective on the Kissimmee River Restoration Project. Resotration Ecology 3 :149 159. McCollum, S.H., and R.F. Pendleton. 1971. Soil survey of Okeechobee County, Florida. Soil Conservation Service, U.S. Department of Agriculture, Washington, D.C., U SA. Milleson, J.F. 1976. Environmental responses to marshland reflooding in the Kissimmee River basin. South Florida Water Management District, Technical Publication #76 3. 39 pp. Milleson, J.F., Goodrich, R.L., and Van Arman, J.A. 1980. Plant communities of the Kissimmee River valley. South Florida Water Management District, Technical Publication #80 7. 42 pp.

PAGE 136

136 Mossa, J., Garfield, U., and Rasmussen, J. 2009. Channel cross section monitoring. Mossa J., Gellis A., Hupp C.R., Pearman J.L., Garfield U., Schenk E., Rasmussen J., Valdez J., Habermehl P. Geomorphic monitoring of the Kissimmee River restoration: 2006 2009 South Florida Water Management District, Final Report, West Palm Beach, Florid a. Nanson, G.C., and Hickin, E.J. 1986. A statistical analysis of bank erosion and channel migration in western Canada: Geological Society of America Bulletin 97 : 497 504 Nico, L.G., Jelks, H.L., and Tuten, T. 2009 Non native suckermouth armored catfishes in Florida: Description of nest burrows and burrow colonies with assessment of shoreline conditions. Aquatic Nuisance Species Research Program Bulletin 9 :1 1 30. Pant, H.K., and K.R. Reddy. 2001. Hydrologic influence on stability or organic phosphorus in wetland detritus. Journal of Environmental Quality 30 :668 673. Pearman, J.L., Gellis, A.C., Habermehl, P.J. 2009. Streamflow and fluvial sediment transport in Pool C, restored section of the Kissimmee River, 2007 2008. Mossa J., Gellis A., Hupp C.R., Pe arman J.L., Garfield U., Schenk E., Rasmussen J., Valdez J., Habermehl P. Geomorphic monitoring of the Kissimmee River restoration: 2006 2009 South Florida Water Management District, Final Report, West Palm Beach, Florida. Pollen, N. 2007 Temporal and spa tial variability in root reinforcement of streambanks: Accounting for soil shear strength and moisture. Catena 69 : 197 205. Ritter, D.F., Kochel, R.C., and J.R. Miller. 2006. Process geomorphology. 4 th ed. Waveland press, Long Grove, IL. Robert, Andre`. 2 003. River Processes: An Introduction to Fluvial Dynamics. Oxford University Press, London, United Kingdom, pp.62 65. Rosgen, D. 1996. Applied river morphology. 2 nd ed. Wildland Hydrology, Pagosa Springs, CO. Rosgen, D. 2001 A practical method of computing streambank erosion rate. Proceedings of the Seventh Federal Interagency Sedimentation Conference March 25 29, 2001, Reno, NV. Vol. 2. Rosgen, D. 2006. Watershed assessment of river stability and sediment supply (WARSSS). Wilkland Hydrology, Fort Collins, CO Sass, C.K. 2011. Evaluation and development of predictive streambank erosion curves dissertation).

PAGE 137

137 Schenk, E.R., Hupp, C.R., and Gellis, A. 2011. Sediment dynamics in the resto red reach of the Kissimmee River basin, Florida: a vast subtropical riparian wetland. River Research and Applications Early View. Shaw, J.E., and Trost, S.M. 1984. Hydrogeology of the Kissimmee planning area. South Florida Water Management District. South Florida Water Management District Tec hnical Publication 84 1, 237 p. Simon, A., and Collison, A.J.C. 2001. Pore water pressure effects on the detachments of cohesive streambeds: Seepage forces and matric suction Earth Surface Processes and Landforms 26( 13) : 1421 1442. Simon, A., Curini, A., Da rby, S., and Langendoen, E. J. 1999 Streambank mec hanics and the role of bank and near ban k processes in incised channels. Darby S.E., Simon A. (eds). Incised River Channels John Wiley & Sons Ltd. West Sussex, Engl and: pp. 123 152. Simon, A., and Darby, S.E.. 1999. The nature and significance of incised river channels. Darby S.E., Simon A. (eds) Incised river channels: processes, forms, engineering and management John Wiley & sons, New York: 3 18 South Florida Wate r Management District (SFWMD). DBHYDRO Environmental database. 2011. http://www.sfwmd.gov/portal/page/portal/xweb%20environmental%20monitoring/ d bhydro%20application Thorne, C.R. 1982. Processes and mechanisms of river bank erosion. In: Gravel bed rivers, R.D. Hey, J.C. Bathurst, and C.R. Thorne. John Wiley and Sons, Chichester, England, pp. 227 271. Thorne, C.R., Hey, R.D., and M.D. Newson. 1997. Applied fluvial geomorphology for river engineering and management. John Wiley & Sons, New York, 376 pp. Thorne, C.R., and Osman. A.M. 1988. Riverbank stability analysis. II. Applications. Journal of Hydraulic Engineering 114 :151 173. Thorne, C.R., and To vey, N.K. 1981. Stability of composite river banks. Earth Surface Processes and Landforms 6 :469 484 Toth, L.A. 1990. Impacts of channelization on the Kissimmee River ecosystem. Pages 47 56 in K. Loftin, L. Toth, and J. Obeysekera, editors. Proceedings of the Kissimmee River Restoration Symposium South Florida Water Management District. West Palm Beach, FL. Toth. L.A. 1992. The ecological basis of the Kissimmee River restoration plan. Division of Kissimmee and Okeechobee Systems Research, Dept. of Research South Fl orida Water Management District. West Palm Beach, Florida, pp. 1 28.

PAGE 138

138 U. S. Army Corps of Engineers. 1991. Final integrated feasibility report and environmental impact statement environmental restoration Kissimmee River, FL. U. S. Army Corps of En gineers, Jacksonville, FL. U. S. Geological Survey (USGS). 2011. Historical maps of monthly and annual streamflow conditions by water year. http://water.usgs.gov/nwc/ Van Eps, M.A., Formica, S.J., Morris, T.L., Beck, J.M., and Cotter, A.S. 2004. Using a ba nk erosion hazard index (BEHI) to estimate annual sediment loads from streambank erosion in the West Fork White River watershed. Proceedings of the 12 15 September 2004 ASAE Conference St. Paul, MN, September 12, 2004. ation number 701P0904. Wilke, B. M. 2005. Determination of chemic al and physical soil properties. Margensin R, Schinner F (eds). Manual of soil analysis Springer, Germany. Wilson, G.V., Periketi, R.K., Fox, G.A., Dabney, S.M., Shields, F.D., and Cullum, R .F. 2007. Soil properties controlling seepage erosion contributions to streambank failure. Earth Surface Processes and Landforms 32 : 447 459. Wynn, T.M., and S. Mostaghimi. 2006. The effects of vegetation and soil type on streambank erosion, southwestern Virginia, USA. Journal of the American Water Resources Association 42 : 69 82.

PAGE 139

139 BIOGRAPHICAL SKETCH Andrew Michael Horan was born in 1987 in Sarasota, Florida, and raised in Venice, Florida. He spent his undergraduate years in Atlanta, GA, at the Georgia two years, from 2006 2008. He g raduated with a B.S. degree in e arth and atmospheric s ciences in August of 2009. As an undergraduate, he also held many jobs th at varied from construction work to lifeguarding to interning at a large company as a water resources engineer. He also worked at the United States Geological Survey (USGS) in Gainesville as a GIS technician to help with funding for his graduate education. He enjoys trail running at San Felasco, biking all around Gainesville, listening to music and NPR, generally being outside, and dinner and conversation with his close friends and family.