GEOMORPHIC EVALUATION OF CROSS SECTIONS IN THE KISSIMMEE RIVER, FLORIDA, 1928 TO 1960
By LISA-ANN G. WALSH
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 2011 1
© 2011 Lisa-Ann G. Walsh
2
To my family, friends and professors, and to the curiosity that keeps us looking forward
3
ACKNOWLEDGMENTS The members of my supervisory committee (Drs. Mossa, Waylen, and Southworth) deserve the highest accolades and my grateful thanks for their expertise and support. Dr. Mossa was especially generous with her encouragement, as well as with opportunities for funding during the course of my academic career at the University of Florida. Dr. Waylen’s steadfast presence across the hall and his patience in explanations is greatly appreciated. Dr. Southworth’s and Dr. Matyas’ willingness to step in at the last minute was also a motivator. Heartfelt thanks also go to Mr. Howard Miller, of the Altamonte Springs United States Geological Survey office, for his assistance in obtaining data, and his willingness to help. My family, especially my parents, deserves a huge thank-you, for supporting my efforts to make something better of myself. My daughter and husband’s understanding of all the times I couldn’t be there is much appreciated. Included in my family would have to be my friends, Ann Angelheart and Aleta Mitchell-Tapping. They acted many evenings as my daughter’s surrogate parents. Thanks go out to Todd Hammerle, Robert Day, Joan Carter, and Pepe Garcia for their help in obtaining county maps and bridge plans. Lastly, I am grateful to Lou Toth, also of the South Florida Water Management District, as the individual who first sparked my interest in the restoration of the Kissimmee River.
4
TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 LIST OF ABBREVIATIONS ........................................................................................... 11 ABSTRACT ................................................................................................................... 12 CHAPTER 1
INTRODUCTION .................................................................................................... 13
2
STUDY AREA ......................................................................................................... 15 Location .................................................................................................................. 15 Climate.................................................................................................................... 17 The River and its History......................................................................................... 17 Geology and Stratigraphy of the Kissimmee Floodplain ......................................... 22
3
LITERATURE REVIEW .......................................................................................... 27 Stream Management and Restoration .................................................................... 27 Fluvial Geomorphology and Restoration ................................................................. 30 Discharge Prediction and Channel Geometry ......................................................... 32 Riley’s Bench Index ................................................................................................ 33 Graph Analysis ....................................................................................................... 33
4
METHODS .............................................................................................................. 34 Opening Remarks ................................................................................................... 34 Data Source, Configuration and Descriptive Statistics............................................ 34 Riley’s Bench Index ................................................................................................ 38 Graph Analysis ....................................................................................................... 39
5
RESULTS ............................................................................................................... 40 Opening Remarks ................................................................................................... 40 Riley’s Bench Index ................................................................................................ 40 Graph Analysis ....................................................................................................... 46 Time Series Graphs and Low Velocities ........................................................... 47 Geomorphic Variable Graphs ........................................................................... 56
5
6
DISCUSSION AND CONCLUSION ........................................................................ 73 Discussion .............................................................................................................. 73 Riley’s Bench Index ................................................................................................ 74 Graph Analyses ...................................................................................................... 75 Conclusion .............................................................................................................. 76 Future Direction ...................................................................................................... 77
LIST OF REFERENCES ............................................................................................... 78 BIOGRAPHICAL SKETCH ............................................................................................ 84
6
LIST OF TABLES Table
page
4-1
Kissimmee River gauging stations...................................................................... 36
4-2
Gauge datum changes. ...................................................................................... 36
4-3
Descriptive statistics for Kissimmee River data. ................................................. 37
5-1
Comparison of Riley’s bench index results ......................................................... 46
5-2
Low velocities above bankfull discharge, Lake Wales station............................. 50
5-3
Low velocities above bankfull discharge, Cornwell station. ................................ 53
5-4
Low velocities above bankfull discharge, Okeechobee station. .......................... 56
7
LIST OF FIGURES Figure
page
2-1
Location of Kissimmee River basin in Florida. .................................................... 16
2-2
Hurricane flooding, train depot, downtown Kissimmee, 1956. From State of Florida photographic archives............................................................................. 19
2-3
Lower Kissimmee River bridge soil boring locations........................................... 20
2-4
State Road 60 bridge over Kissimmee River soil borings, 1955. Florida Department of Transportation bridge design documents, 1955. ......................... 24
2-5
State Road 60 bridge over Kissimmee River soil borings, 1956. Florida Department of Transportation bridge design documents, 1955. ......................... 24
2-6
State Road 70 bridge over Kissimmee River soil borings, 1965. Florida Department of Transportation bridge design documents, 1955. ......................... 25
2-7
State Road 78 bridge over Kissimmee River soil borings, 1961. Florida Department of Transportation bridge design documents, 1955. ......................... 25
4-1
Kissimmee River gauging stations map. ............................................................. 35
5-1
Station 02269000, Kissimmee River below Lake Kissimmee at Lake Wales, Florida. Cross section measurement #322, November 9, 1960. ........................ 41
5-2
Station 02269000, Kissimmee River below Lake Kissimmee at Lake Wales, Florida. Riley’s bench index and width/depth ratio............................................. 42
5-3
Station 02272500, Kissimmee River near Cornwell, Florida. Cross section #1, December 9, 1948. ....................................................................................... 43
5-4
Station 02272500, Kissimmee River near Cornwell. Riley’s bench index and width/depth ratio ................................................................................................. 44
5-5
Station 02273000, Kissimmee River at S-65E near Okeechobee. Cross section measurement #389, October 4, 1960. .................................................... 45
5-6
Station 02273000, Kissimmee River at S-65E near Okeechobee. Riley’s bench index and width/depth ratio. ..................................................................... 46
5-7
Velocity time series graph for Station 02269000 - Kissimmee River below Lake Kissimmee at Lake Wales.......................................................................... 48
5-8
Discharge time series graph for Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales. .................................................................... 49
8
5-9
Tropical cyclone tracks across the Kissimmee River basin. ............................... 51
5-10 Velocity time series graph for Station 02272500 – Kissimmee River near Cornwell.............................................................................................................. 52 5-11 Width time series graph for Station 02272500 – Kissimmee River near Cornwell.............................................................................................................. 53 5-12 Velocity time series for Station 02273000 – Kissimmee River at S-65E near Okeechobee. ...................................................................................................... 54 5-13 Width time series Graph for Station 02273000 – Kissimmee River at S-65E near Okeechobee. .............................................................................................. 55 5-14 Station 02269000, Kissimmee River below Lake Kissimmee at Lake Wales, Florida. Cross section measurement #322, November 9, 1960. ........................ 57 5-15 Discharge versus velocity, total system, pre-channelization, Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida. ............................................................................................................... 58 5-16 Discharge versus width, total system, pre-channelization, Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida. ................. 59 5-17 Discharge versus mean depth, total system, pre-channelization, Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida. ............................................................................................................... 60 5-18 Discharge versus area, total system, pre-channelization, Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida. ................. 60 5-19 Flow duration curve. Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida. ................................................................ 62 5-20 Station 02272500, Kissimmee River near Cornwell, Florida. Cross section #1, December 9, 1948. ....................................................................................... 62 5-21 Discharge versus velocity, total system, pre-channelization, Station 02272500 – Kissimmee River near Cornwell, Florida. ........................................ 63 5-22 Discharge versus width, total system, pre-channelization, Station 02272500 – Kissimmee River near Cornwell, Florida............................................................. 64 5-23 Discharge versus mean depth, total system, pre-channelization, Station 02272500 – Kissimmee River near Cornwell, Florida. ........................................ 65 5-24 Discharge versus area, total system, pre-channelization, Station 0272500 – Kissimmee River near Cornwell, Florida............................................................. 65
9
5-25 Flow duration curve. Station 02272500 – Kissimmee River near Cornwell, Florida. ............................................................................................................... 66 5-26 Station 02273000, Kissimmee River at S-65E near Okeechobee. Cross section measurement #389, October 4, 1960. .................................................... 67 5-27 Discharge versus velocity, total system, pre-channelization, Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida.................... 68 5-28 Discharge versus width, total system, pre-channelization, Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida. ...................................... 69 5-29 Discharge versus mean depth, total system, pre-channelization, Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida.................... 69 5-30 Discharge versus area, total system, pre-channelization, Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida. ...................................... 70 5-31 Flow duration curve. Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida. ......................................................................................... 71 5-32 Comparison of bankfull discharge values and methodologies. ........................... 72
10
LIST OF ABBREVIATIONS a
area
ac
acres
cfs
cubic feet per second
cm
centimeter
ft
feet
ft/sec
feet per second
ha
hectare
in
inch
km
kilometer
KRRP
Kissimmee River Restoration Plan
KRCC
Kissimmee River Coordination Council
m
meter
m/sec
meters per second
m2
square meter
m3/sec
cubic meters per second
masl
meters above sea level
mi
mile
ft2
square feet
sq km
square kilometer
sq mi
square mile
q
discharge
v
velocity
w
width
11
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 GEOMORPHIC EVALUATION OF CROSS-SECTIONS IN THE KISSIMMEE RIVER, FLORIDA FROM 1928 TO 1960 By Lisa-Ann G. Walsh August 2011 Chair: Joann Mossa Major: Geography The Kissimmee River underwent channelization during the 1960’s. During the past two decades, river managers have begun restoring the river to a semblance of its historical counterpart. Part of the restoration effort includes flooding the river valley in a manner similar to historic temporal and spatial patterns. Bankfull discharge is the point at which a river overflows its banks, so it is crucial to know what volume of water constitutes bankfull discharge. My study utilizes various methods to compare historic bankfull discharge relationships at three stations along the Kissimmee River in Florida. Methods used include Riley’s Bench Index and graph analysis. According to Toth (1996) and Warne (1998), bankfull discharge for the Kissimmee River is a quantity of water above a range of 40 to 60 m3/sec (1,412 to 2,118 cfs). Using the techniques mentioned, my study verifies that range. While all bankfull discharges found in my study fall above the range specified by Warne (1998), estimated discharges using Riley’s Bench Index were significantly higher than the range in the literature. Graph analysis on various ranges of discharge produced results closer to those of Warne (1998).
12
CHAPTER 1 INTRODUCTION Since Europeans arrived in North America, rivers on this continent have been altered in order to support commercial and industrial needs. In Florida, the Kissimmee River was used for commercial transportation beginning in the mid 1800’s. Spurred by the natural hazards of hurricanes and drought, alteration of the central and southern Florida peninsula began with the River and Harbor Act of June 13, 1902, in which the United States Government authorized the ‘improvement and alteration” of any waterway available for transport. This Act authorized the then fledgling Army Corps of Engineers to “maintain a channel 3 feet deep by 30 feet wide” on the entire course of the Kissimmee to facilitate the transport of commercial goods (United States Army 1931). Over the course of the next 60 years, passage of subsequent River and Harbor Acts, and the Flood Control Acts of 1937, 1939, and 1941 authorized numerous other projects. In 1960, work began along the Kissimmee to construct six water control structures. These structures created five stepped pools that contained the flows of the Kissimmee within a defined channel. Previously, the river had overflowed its banks during high flows. The floodplain would remain inundated for long periods of time, often months. It is these prolonged inundations that made the Kissimmee River unique among river systems. The time of inundation allowed exchanges of nutrients and oxygen between the river and it floodplain, creating a rich habitat for wildlife. These exchanges ceased during and after channelization (Toth 1990).
13
Before the channelization project was completed in 1970, scientists and engineers alike were concerned about the impacts of the project to the floodplain and wildlife habitats. Continued environmental concern resulted in studies showing a 90% decrease in the wading bird population and decreasing levels of dissolved oxygen in the river (Toth 1990). Further evidence of habitat destruction was documented in extensive studies done in regions further south, especially in the Everglades. Scientists have documented the interconnectedness of the hydrology of all of South Florida, and recognize the Kissimmee River as the beginning of the South Florida/Everglades system (Fernald and Perdum 1998, Warne 1995). In response to negative impacts to the river and wildlife habitat, public outcry again spurred changes in bureaucratic thinking and funding. Work to remove two control structures and the backfilling of a part of the channel began in October 1999, concurrent with work to restore the Everglades. The Army Corps of Engineers and other agencies retain control of the flows within the restored portion of the river, and the flows required to inundate the floodplain are documented. Historically, bankfull discharge for the Kissimmee River is cited as being above a range of 40 to 60m3/sec at various gauging stations along the length of the river (Toth 1996, Warne 1998). This range was arrived at by “preliminary analysis” of stream gauge records between 1931 and 1961. No in-depth studies of these records have been done (Warne 1998).
My study conducts quantitative analyses of the USGS
stream gauge and flow records to compare methods to arrive at a bankfull discharge for the Kissimmee River. The analyses include Riley’s Bench Index, linear regression, and linear regression on distinct populations of discharge.
14
CHAPTER 2 STUDY AREA Location Located in the heart of Central Florida, the Kissimmee River Basin is divided into two sections; the Upper Basin and the Lower Basin. The headwaters of the Kissimmee River begin in an area known as Reedy Creek, in northwestern Osceola County, Florida. Osceola County is directly south of Orange County, in which lies the largest city in Central Florida, Orlando. Reedy Creek empties into a series of twenty-six interconnected lakes, the largest of which is Lake Tohopekaliga near the city of Kissimmee, then into several smaller lakes, culminating in Lake Kissimmee. Twenty lateral tributary sloughs also contribute flow to the river (Toth 1996). While the upper basin is more urban in nature, the lower basin remains relatively rural, consisting largely of conservation lands and cattle ranches. The largest city in the lower basin is Okeechobee. The Lower Basin of the Kissimmee begins at the southern end of Lake Kissimmee, immediately south of State Road 60. The study area begins at this point, and follows the historic river channel as it meanders south to Lake Okeechobee. The study area is located near the center of Florida (Figure 2-1), and is entirely within the Coastal Plain province. The Osceola Plain covers the northern part of the Lower Basin, and the Okeechobee Plain covers the southern part, with the Lake Wales Ridge as a western boundary (Scarlatos et al. 1990). The Lower Basin has an area of some 1,771sq km. Prior to channelization, the lower basin had over 18,000 ha of wetlands. After channelization, flooding of the historic river valley ceased, thus dramatically reducing the area of wetlands by almost 80% (Toth 1990).
15
Figure 2-1. Location of Kissimmee River basin in Florida.
16
Climate The climate within the basin is humid subtropical, as listed in Köppen Climate Types. The first rains of the water year come in late winter and early spring of the year, generated primarily by frontal systems. Between late fall and late winter, this area generally receives very little rainfall. The second peak in rainfall occurs between the months of June and October. These rains are generated by convectional processes, and usually occur daily. While the average annual rainfall for the area is approximately 121 cm per year, hurricanes and tropical storms can often cause annual rainfall amounts to exceed 203 cm (Fernald and Perdum 1998). Of the variations apparent in an analysis of rainfall data by the South Florida Water Management District, one is significant to the Kissimmee River. Throughout the state, average annual rainfall ranges between less than 1.12 m to greater than 1.63 m. However, the lowest amounts of rainfall occur within the Kissimmee River Valley and in the Florida Keys (Fernald and Perdum 1998). The River and its History The pre-channelized river is estimated at approximately 166 km in length (Scarlatos et al. 1990), and was shaped primarily by variations in discharges with an average width of 15 to 27 m and an average depth of 1 to 1.5 m (Toth 1996). Rosgen’s (1994) classifications identify the Kissimmee as a C5 stream, meaning it is a low gradient, sinuous, meandering stream, with point bars, and located in within a broad drainage basin. Warne (1998) further describes the river as having a DA5 classification in the some portions of the floodplain, because of anastomosing, a high channel width to depth ratio, and sandy substrates.
17
Modern man has significantly affected the Kissimmee since 1881, when Hamilton Disston purchased 1,618,742 ha of land in South Florida. At the time, the United States Federal government encouraged settlement of the area, sanctioning ditching and drainage to create arable land in the swampy South Florida region. The government also directed the Army Corps of Engineers to keep the Kissimmee open for navigation, as well as to create navigable channels throughout the state of Florida. Mr. Disston created the Caloosahatchee River canal linking Lake Okeechobee and the Gulf of Mexico (Fernald and Perdum 1998). The Florida State Legislature created the Everglades Drainage District in 1907. Again, the mission was to ditch and drain the swamp to create arable land. Between 1913 and 1927, 708 km of levees, 6 major canals, and 16 locks and dams were constructed to contain and direct water in South Florida (Warne 1998). After devastating hurricanes in 1926 and 1928, public outcry compelled lawmakers to create the Okeechobee Drainage District, with its jurisdiction defined as the Kissimmee River Valley and all lands lying south of it. A massive program began to dig canals that connected lakes and sloughs to Lake Okeechobee. Canals dug south of Lake Okeechobee drained water directly to the Atlantic Ocean. Provision of navigable waterways became secondary to the mission of flood control. Strong hurricanes struck South Florida in 1947 and again in 1956, causing flooding even into the upper reaches of the Kissimmee basin (Fernald and Perdum 1998). Florida State Archive photos contain photographs of downtown Kissimmee and the railroad station shortly after the hurricanes. The floodwaters reach to the top of Hamilton Disston’s railroad tracks (Figure 2-2).
18
Figure 2-2. Hurricane flooding, train depot, downtown Kissimmee, 1956. From State of Florida photographic archives.
In 1948, the Central and Southern Florida Flood Control District was created, with duties including channelizing the Kissimmee River (Fernald and Perdum 1998). The Army Corps of Engineers has been charged with making the river navigable since 1902 (Army Corps of Engineers 1953). The channelization of the river in the 1960s led to a wide variety of impacts to the entire floodplain, caused mainly by the elimination of overbank flooding. Floodplain wetlands inundated for several months at a time are one component that made the historic Kissimmee floodplain a rich wildlife nursery. The Army Corps of Engineers dredged a channel measuring 91 m wide by 9 m deep in the decade between 1962 and 1971 (Toth et al. 1993). This channel shortened the 217 km river by 64 km, to 169 km. While the historic channel was used to delineate boundaries between counties, the modern channel flows in direct lines through Osceola, Polk, Highlands, Hardee, and Okeechobee counties (Figure 2-3).
19
Figure 2-3. Lower Kissimmee River bridge soil boring locations.
20
Concerns regarding environmental damage from channel construction emerged during the early stages of the project. By 1969, the National Environmental Protection Act was passed, requiring the Central and Southern Florida Flood Control District and the Army Corps of Engineers to consider damages to the environment when making decisions regarding their projects. The Water Resources Act was passed in 1972, broadening the authority and responsibility of the Central and Southern Florida Flood Control District, and requiring the control and regulation of water supplies and their use (Fernald and Perdum 1998). When the Kissimmee overflowed, the floodplain was often inundated for months at a time. Channelization eliminated this unique aspect of the river, resulting in major changes to the Kissimmee River ecosystem. Toth documented damage to the Kissimmee River system in 1990. Of approximately 18,211 ha of wetlands in the historic system, only 4,047 ha remained. A 90% decrease in the wintering waterfowl and wading bird populations occurred. The steadily decreasing levels of dissolved oxygen decimated fish populations. A significant decrease in flora diversity was also documented. In addition, sedimentation of organic matter increased, raising muck levels along the river’s bed (Toth 1990). In 1976, the Central and Southern Florida Flood Control District became the South Florida Water Management District. And by the 1980’s, the District had begun a major effort to restore and protect the Kissimmee-Okeechobee-Everglades system (Fernald and Perdum 1998).
21
The restoration effort includes expensive land acquisition (Jones and Malone 1990) and extensive studies in the Kissimmee River valley. A 1985 pilot project and study by the South Florida Water Management District indicated that filling in the channel and allowing water to flow through the historic channel of the river would generate no harmful effects. A subsequent study by Toth, et al. in 1993 confirmed the findings and stated that re-establishment of historic flow characteristics would be required in order to improve the water quality and habitat. By 1999, Florida successfully appropriated $500 million dollars for restoration of the Everglades and its tributaries, namely the Kissimmee River. On June 19, 2000, the South Florida Water Management District and the United States Army Corps of Engineers began Phase I of the physical restoration of the Kissimmee by ceremoniously blowing up the S-65B water control structure near Lorida. Phase III of the KRRP has been completed ahead of schedule in early 2011. Water flow has been reestablished to 74 km of historic channel and that 10,360 ha of floodplain have been restored (United States Army Corps of Engineers, www.saj.usce.army.mil/Divisions/Everglades/Branches/ProjectExe/Sections/UECKLO/K RR.htm, April 2011). Geology and Stratigraphy of the Kissimmee Floodplain Much of southern Florida is underlain by Pleistocene formations of Miami Limestone, Key Largo Limestone, the Anastasia Formation, Fort Thompson Formation and Caloosahatchee Marl (Fernald and Perdum 1998). Within the Kissimmee River basin, evidence of marine terraces formed during the Pleistocene Period is found. These terraces include Wicomoco, Penholoway, Talbot, and Pamlico, with Wicomoco the oldest and Pamlico the youngest. 22
Also found in the basin are distinct geomorphic features: the Osceola Plain, Bombing Range Ridge, Okeechobee Plain, Caloosahatchee Incline, Lake Wales Ridge, Lake Okeechobee, and the Kissimmee Floodplain itself (Warne 1998). Sediments within the basin are primarily marine deposited silica sands with approximately 20% organic materials (Scarlatos et al. 1990). Indeed, Warne (1998) stated that the river is a complex product of marine shoreline sedimentation and subterranean carbonate solution and subsidence, reworked by multiple episodes of flooding. While the geologic and geomorphic processes are continuous natural processes, modern humans have intervened in these processes since the early 1800’s. First, the Kissimmee was maintained for navigation, and in the mid 1900's, the Florida Department of Transportation constructed several bridges over the Kissimmee River. Design documentation of these bridges exists. The documents include core samples taken from the areas of the river where the bridge was constructed. Core sample diagrams give a sense of the stratigraphy of the Kissimmee River bed, and may also show historic channels subsequently buried by newer sediments. Interesting to note is the layer of muck common to all the core samples taken at the time. Muck is typically comprised of decomposed vegetation and fine sediments. Its presence in large quantities may indicate a prolonged drought resulting in a vegetative die-off, or a lack of velocity sufficient to entrain it and move it downriver. However, historic stream discharge records show that floods did occur, so another cause for the presence of the muck exists. A comparison of rainfall and discharge records may yield more information. In Figures 2-4 through 2-7, channel banks are not indicated in the coring
23
results and thus are not shown in the figures. All core samples include a combination of sand, marl, and muck. Clay and shell are found in samples farther downriver.
Figure 2-4. State Road 60 bridge over Kissimmee River soil borings, 1955. Florida Department of Transportation bridge design documents, 1955.
Figure 2-5. State Road 60 bridge over Kissimmee River soil borings, 1956. Florida Department of Transportation bridge design documents, 1955.
24
Figure 2-6. State Road 70 bridge over Kissimmee River soil borings, 1965. Florida Department of Transportation bridge design documents, 1955.
Figure 2-7. State Road 78 bridge over Kissimmee River soil borings, 1961. Florida Department of Transportation bridge design documents, 1955.
Changes occur in the shape of the layers of materials as one goes downriver. Where State Road 60 now crosses the river, layers are thick and indicate channels in 25
various shapes. Farther downstream, at State Road 70, the materials include clay. The layers are thick and discontinuous suggesting numerous channel alignment changes. Warne (1998) states that differences in channel shape, such as deeper, narrower channel dimensions may be a function of the sandy substrate and densely rooted surface vegetation or decreased sediment input and increased groundwater to the overall discharge. The deeper, narrower channel shapes occur in the middle reaches of the river, rather than the headwaters or the mouth. Near the mouth of the river at Lake Okeechobee, the layers of material change to include clay and shell, but no marl. In addition, there is no discontinuity in material layers, suggesting vertical accretion, similar to that found behind a dam. The soil borings support the geology and geomorphic features discussed by Warne (1998) and Toth et al. (2002), in that the Okeechobee scarp caused a pooling effect at the mouth of the historic river.
26
CHAPTER 3 LITERATURE REVIEW Stream Management and Restoration Even before Europeans began to inhabit the United States, rivers were altered to better suit human needs for transportation, water, and energy sources. Evolution of the management of rivers has led to the ethic of control in order to either confine or use the resource. As of 1988, only 42 rivers longer than 200 km still flow free in the United States. With the exception of the Yellowstone River, all rivers in the United States longer than 1,000 km have been altered for navigation and/or hydropower. Most of the larger and high-quality streams are found in the Southern Atlantic states, with three in the state of Florida. The Choctawhatchee, Suwannee, and the St. Mary’s Rivers flow freely through the northern tier of the state (Benke 1990). The decline of fisheries, general environmental values, and habitats has prompted many authors and citizens to call for the restoration of rivers in the United States. Sparks et al. (1990) mentions that natural disturbances, restoration efforts, and manmade changes should be opportunities to test ideas. There is a broad body of literature regarding stream management. This topic has long been considered the domain of engineers, who are accustomed to accomplishing the task using structural methods such as weirs, dams, pools, and channelization. However, in the recent past, many authors have begun to question this ethic, especially in the area of stream restoration projects. Winkley and Schumm (1994) have recognized that engineering methods alone are limited with larger rivers because of their magnitude of energy. Brookes (1988) recognized alternative strategies for river management that reconcile engineering objectives with nature conservation.
27
Shields et al. (1995) noted success in the restoration of a reach of Hotophia Creek in Mississippi that combined both hard and soft engineering structures with the planting of trees along the bank. Alternative restoration techniques for the Kissimmee River are also discussed by Shen et al.. (1994) and Loftin and Obeysekera (1990). A more holistic view and the protection of ecosystems have become as important as flood control for stream management. Some authors Stanford et al. (1996) look at entire catchments to determine specific restoration techniques, and Kern (1992) recommends the consideration of the riverbed, the floodplain and the tributaries when planning a restoration. Restoration of a riverine ecosystem requires quantities of water in patterns similar to the historic flow regime. According to Dunne (1988), “geomorphology is the only means of providing the required historical background for flood and sediment control.” Obeysekera and Loftin (1990) recommend restoration of historic Kissimmee basin flow characteristics to ensure adequate supplies of water system-wide. Flood control is still an issue in most restorations. Petts (1996) champions a standardized policy for water allocation to enhance lotic and riparian systems damaged by abstraction and diversion of water below dams and reservoirs. Former Executive Director of the Sierra Club, David Brower, shares this view. He states that in restricting the flow of the Colorado River, the United States has violated the spirit of a waterdevelopment agreement with Mexico, and recommends removal of the Glen Canyon dam, since the Hoover dam would be capable of all necessary flood control. This would aid in the delivery of sufficient water for Mexico and the restoration of lotic and riparian habitats along the river (Brower 1997).
28
Citizens have also called for non-structural restoration alternatives by the scientific community. Historically, many urban streams have been channelized or confined to underground culverts. Within the last two decades, neighborhoods and local groups have come to recognize that natural streams have an amenity utility that works to increase the resale value of homes (Pollock 1989). Thus, citizens are now more supportive of working to restore urban streams to more natural forms in their communities. On a larger scale in Florida, citizens have worked with the scientific community and policymakers to secure funding for one of the largest restoration projects in the world - that of the Kissimmee River basin and the South Florida Everglades. This enormous undertaking has ecology as its basis, with the goal of ‘ecological integrity’ encompassing biological, chemical, and physical integrity (Toth 1995). Koebel (1995) writes that the United States Geological Survey issued a report in 1971 that documented environmental concerns. This report led to a Governor’s Conference on Water Management in South Florida the same year, and the creation of the Kissimmee River Coordinating Council (KRCC) in 1976. Scientists working with the South Florida Water Management District have further documented impacts (Toth 1990) to local flora and fauna populations and seek to reinstate historic flow regimes to the floodplain to reverse the degradation (Toth 1993). However, restoration of a stream that has been dramatically altered does not mean restoration of pristine conditions. Land use and hydraulic change within the watershed may preclude original conditions. Brookes (1995b) and Obeysekera and Loftin (1990) state that true restoration may be impossible due to changes in weather and climate patterns, as well as lack of historic data.
29
Creation of a river channel may have tenuous results, be cost-prohibitive, and unsustainable. Rehabilitation, that is, changes made to reflect the historic flows and channel forms as closely as possible, or enhancement of degraded systems may be more easily sustained. Costs of rehabilitation are also much less than recreation of a channel. Moreover, time scales for recovery of degraded systems varies widely. In terms of benthic macroinvertebrate fauna recovery periods, it was found that recovery could be very short if the stream was connected to another healthy ecosystem (Fuchs and Statzner 1990). However, even the most conscientious restoration or return to a meandering from a straightened channel may have a time scale of approximately 100 years, making funding and monitoring important issues to policymakers. The monitoring period for the Kissimmee is slated to be 15 years (South Florida Water Management District 1995). Kondolf (1995) and Kondolf and Micheli (1995) recommend a minimum monitoring period of 10 years for effective evaluation of any restoration. Fluvial Geomorphology and Restoration Fluvial geomorphology increasingly has a larger role in stream management. Until recently, river management practices consumed large amounts of resources both natural and man-made, and consisted of using only engineering to achieve solutions that treated only the symptoms of riverine problems, rather than addressing the cause (Sear et al. 1995). Fluvial geomorphology is evolving towards a holistic catchment management, and is discussed by more than one scientist (Newson 1995; Petts et al. 1992; Petts 1995; Gottle 1992). The demand for geomorphology in assisting with policymaking has happened as a result of the science becoming more synthetic (Graf 1996). 30
Brookes (1995a) lays out challenges and objectives for geomorphology in United Kingdom river management. Geomorphic objectives should be included in the development of design guidance, policy, management approaches, procedures, and training. The challenges include the need to develop a more professional image, as well as appropriate standards. The most formidable challenge may be that of river and floodplain restoration. Kissimmee River managers established geomorphology as critical to the success of the project by authorizing a study of the geomorphic processes necessary for the ecological diversity and integrity of the restored system (Warne 1998). The study lists the hydro-characteristics for restoration as flow duration and variability similar to historic conditions: a flow velocity of 0.24 – 0.55 m/sec within the channel, a stage-discharge relationship that is equal to bankfull when the discharge is greater than 39.6 – 56.6 m3/sec, a stage recession of less than 0.3048 m per month, and floodplain submergence periods similar to historic conditions (South Florida Water Management District 1995). Applications of geomorphic principles in stream restoration pertain mainly to channel geometry, and the study of how components of geometry affect discharge. Many modern authors, such as Park (1995), Pickup and Reiger (1979), Rhoads (1994), Warne, (1998), Wharton (1995), Kondolf (1995) and Phillips (1991) have examined geomorphic variables. However, it was Leopold and Maddock (1953) and Wolman (1955) who first looked quantitatively at how width, depth, and velocity vary with changes in discharge. Wolman (1955) found that variables change in a progressive and orderly fashion, and that discharge and suspended sediment load control the shape and longitudinal profile of an alluvial channel within the confines of local geology.
31
According to Morris (1995), any stream restoration is essentially a geomorphic activity, inasmuch as quasi-equilibrium stream channels and functional floodplains promote the greatest aquatic and terrestrial habitat diversity and represent the natural conditions under which riparian ecosystems develop. Vinson (1989) studied sediment dynamics of meandering and straight reaches in Birch Creek, Idaho. He concluded that a meandering form is better for the overall health of a stream, as well as for the inchannel and overbank distribution of fine sediments. Gore et al. (1995) state a quality stream restoration should include factors of hydrology, water quality, hard and soft bank protection, riparian vegetation, macroinvertebrates, fish habitat enhancement, planning and monitoring, as well as predicting enhancement and recovery. These authors also state that equilibrium should be reached within a reasonable amount of time, but do not define a time frame. Discharge Prediction and Channel Geometry There are qualitative ways in which hydraulic variables respond to flow changes, but my study uses both quantitative and qualitative methods. Essentially, there are infinite combinations of flow width, depth, and energy grade slope that may satisfy a given extremum (Phillips, 1991). ‘Quasi-equilibrium’ and ‘in-regime’ are terms that refer to the characteristics of a stream that fluctuates about a relatively stationary mean condition at the considered time scale (Pickup and Reiger, 1979). Bankfull width, “that stage above which discharge commences to flow over the floodplain” is one such characteristic (Riley, 1972). Estimating bankfull, or flood discharge, is currently done in several ways. These include, but are not limited to, Riley’s Bench Index, width/depth ratio and graph analysis, as well as other more sophisticated regression analyses.
32
Riley’s Bench Index In 1955, Wolman stated that bankfull discharge could be estimated by determining the minimum width/depth ratio of a series of cross-sections. However, he found this method did not yield reliable results when comparing the minimum width/depth ratio to observed floodplain-channel junctions. Riley (1972) used a bench index to determine bankfull width, which is the first point that the index goes to a maximum. The cross-section is divided into a grid, and differences in width are divided by differences in depth. The accuracy of results depends on the grid scale. Graph Analysis All methods have an innate level of uncertainty (Wohl 1998), which is increased with additional knowledge gained (Brookes et al. 1998). Pickup and Reiger (1979) state that linear regression analysis of channel geometry characteristics may be oversimplified because channel response to discharge varies according to the magnitude and frequency of discharge. Rinaldi and Johnson (1997) found that using the regression equations suggested by Leopold and Wolman (1957, 1960) yielded a set of parameters too large for a stream being studied in Maryland. They suggested that the equations be adjusted for individual regions. This suggestion was also borne out by Harvey (1969) during research on three rivers in southern England. Other investigators utilize different methods of estimation. Reinfelds (1997) used aerial photography and a limited number of channel cross-sections to reconstruct changes in bankfull width. Calculations of wavelength, radius of curvature, width/depth ratio, and sinuosity were used by Rechard and Schaefer (1984) to restore a stream to conditions similar to those before a strip mine was dug in the eastern Powder River Basin in northeastern Wyoming. 33
CHAPTER 4 METHODS Opening Remarks The first part of this chapter describes the collection and preparation of discharge summary data. Figure 4-1 illustrates the location of stream gauges in the lower Kissimmee River. Table 4-1 lists location data for the gauges, as well as the period of record. Discharge summary data indicated several datum gauge changes, and these are illustrated in Table 4-2. The latter part of the chapter discusses the methods used for analysis. Three methods are used to compare the estimated bankfull discharge at specific stations on the Kissimmee River. These include Riley’s Bench Index (Riley, 1972), minimum width/depth ratios, and then graph analysis on all channel variable measurements. Data Source, Configuration and Descriptive Statistics Original data files from the United States Geological Service (USGS) office in Altamonte Springs, Florida are used for this thesis. These files include cross section measurements and discharge summary sheets from three pre-channelization gauging stations along the Kissimmee River in central and south Florida. Data from each of the discharge summary sheets were entered into an Excel spreadsheet program in the same format as they appeared on the original files. Each variable was converted to metric units, and other fluvial geomorphic variables are computed, such as mean bed elevation and mean depth. The data were verified for correctness, and adjustments were made for gauge station datum changes according to information from USGS Water Resources Data files.
34
Figure 4-1. Kissimmee River gauging stations map.
35
Table 4-1. Kissimmee River gauging stations. Station Name Station Latitude Number Kissimmee River below Lake 02269000 N27°46’13” Kissimmee at Lake Wales Kissimmee River near Cornwell
02272500
Longitude
N27°21’52”
Record Period
W81°10’45”
1930-1969
W81°03’07”
1948-1951 1962-1964
Kissimmee River 1928-1962 at S65-E near 02273000 1964-1990 N27°13’22” W80°57’46” Okeechobee Note: In 1964, the name of station 02273000 was changed to reflect the addition of the weir structure. The gauge remained in the same location. Table 4-2. Gauge datum changes. Kissimmee River below Kissimmee River near Lake Kissimmee near Lake Cornwell Wales To 03/21/34: 06/31 – 09/30/48: no -22.9cm gauge datum changes
Kissimmee River at S65-E near Okeechobee To 04/13/49: no gauge datum changes
03/22/34 – 09/30/50: -+46.3cm
10/1/48 – 09/30/53: +7.5m
04/14/49 – 09/30/64: +41.8cm
10/01/50 – 10/31/69: +13.3m
10/01/62 – 10/01/64: no gauge datum changes
Thereafter, no gauge datum changes
In order to begin analysis, descriptive statistics are calculated for each station, and are shown in Table 4-3. All analysis methods utilize “Total System” measurements, except Riley’s Bench Index, which uses only main channel cross sections to create the index. Data from these files are graphed as time series against the variables of area, discharge, and velocity. The variables are defined as follows: •
Area, total system: a spatial measure of width multiplied by depth
•
Discharge, total system: the quantity of water flowing past the gauge, in cubic meters per second, of all channels present; found by multiplying the area by the velocity.
36
•
Velocity, total system: speed of flow past the gauge of all channels present. The period of record covered in this data set is from 1928 to 1960. Station
02268903 (Kissimmee River at S-65 near Lake Wales) was installed in 1968. This date is post channelization, making the use of data for this station irrelevant in my study. In the stream discharge records, measurements are taken at approximately monthly intervals, but may be discontinuous at times. The cause for these discontinuities is unknown, but may be related to low funding and wartime conditions. Table 4-3. Descriptive statistics for Kissimmee River data. Station Statistic Width, Mean Area, n m Depth, m m2 Kissimmee Mean 157 1.94 289 River below Lake 304 Min 23 0.17 10 Kissimmee at Lake Wales Max 974 5.40 2,601
Kissimmee River near Cornwell
Kissimmee River at S65E near Okeechobee
62
304
Velocity , m/sec 0.25
Discharge, m3/sec 31
-0.58
-34
0.48
213
Mean
93
1.16
288
0.24
64
Min
20
0.24
20
0.15
7
Max
981
5.49
1,194
0.46
380
Mean
158
1.60
264
0.35
63
Min
15
0.29
12
0.02
2
Max
953
3.37
1,905
0.65
459
Warne (1998) stated that “preliminary analyses” indicated a bankfull discharge range of greater than 40 to 60 m3/sec. The mean, minimum and maximum discharges are listed in Table 4-3. When comparing the mean discharge for the three stations, all the values are close to the range identified by Warne (1998).
37
Riley’s Bench Index To S.J. Riley (1972), bankfull is the “stage above which discharge commences to flow over a floodplain.” While searching for a reliable measure of bankfull, Riley (1972) determined that using only vegetative cover as a guide was too varied for specific locations. In addition, the use of aerial photography was too subjective to the interpreter and to environmental factors. Because it depended on the channel shape of a stream, using only the minimum width/depth ratio is characterized by Wolman, 1955 and Pickup and Warner, 1976 as unreliable. The width/depth (w/d) ratio is determined at each point where the horizontal lines and the cross-section profile intersect. The first point at which the w/d ratio goes to a minimum is considered the bankfull threshold or bankfull discharge. Riley (1972) developed his method using the cross-section of a stream during a flood event, equally divided horizontally. The difference between the width measurements and the difference between the depth measurements are divided to obtain an index. The first point at which the index reaches a maximum is considered the bankfull discharge depth. The bankfull discharge is then computed using the measurements of the cross section. That is, velocity multiplied by area equals discharge. In my study, bench indices are created for single channel cross sections at each of the gauging stations using the formula below. Cross sections used are dated prior to channel construction. The deepest and widest channel is analyzed. A 0.5 meter grid is used for Riley’s Bench Index analysis, as channels in the Kissimmee River often exceed 100 meters in width. Precision of the index relies on the grid spacing. A finer grid will note smaller, in-channel benches that may not be relevant to bankfull discharge. 38
Riley’s Bench Index Model (Riley, 1972) BI = W(i) – W(i+1) D(i) – D(i+1) where BI is the bench index, W(i) is channel width vector, ranked, and D(i) is the channel depth vector. Graph Analysis Before one can determine the proper method to use for analysis, one must understand the physical properties and behavior of the fluvial system under study. Understanding begins with graphing channel variables and reviewing the results. As Phillips (1991) stated, “there are infinite combinations of flow width, depth, and energy grade slope that may satisfy a given extremum.” Therefore, initial graphs created are those of time series of the variables Discharge, Width, Area, Velocity, Mean Depth and Stage. Subsequently, graphs of variables versus other variables are created in order to study the relationships between them and how they change given various environments.
39
CHAPTER 5 RESULTS Opening Remarks Various stream data analysis methods are utilized to determine bankfull discharge for the Kissimmee River in this study. They include Riley’s Bench Index (Riley, 1972), minimum width/depth ratios, and graph analysis. Riley’s Bench Index (Riley, 1972) determines bankfull threshold by using the first maximum of the Bench Index. Wolman (1955) and Pickup and Warner (1976) utilize the point at which the width to depth (w/d) ratio of a cross section of a stream goes to a minimum. Graphs of channel variables show distinct, and in many instances, bifurcated non-linear patterns, making linear regression invalid. Results are organized by method and then by station in downstream order. Riley’s Bench Index Cross sections of 02272500 (Kissimmee River near Cornwell) and 02273000 (Kissimmee River at S-65E near Okeechobee) that exhibit multiple channels indicating a flood condition are used for this analysis. No multiple channel cross sections are available for Station 02269000 (Kissimmee River below Lake Kissimmee near Lake Wales), and existing cross sections for this station have vertical lines at the initial and end points, suggesting no overbank flow. Stream gauge data indicate discharges for a main channel and total system where a flood condition exists, however, no total system discharges are listed for the Lake Wales cross sections, thereby supporting the assumption that only a main channel with no flood condition is measured at this station. Discharge for the Lake Wales cross section in Figure 5-1 is estimated at 119 m3/sec,
40
width is 162 m, velocity is 0.14 m/sec, and area is estimated at 859 m2. Adjusted gauge height is 16.40 m.
4.0
Bed elevation (m)
3.0 2.0 1.0 0.0 -1.0 -2.0 -3.0 -4.0 0
20
40
60 80 100 120 140 160 180 Distance from initial point (m)
Figure 5-1. Station 02269000, Kissimmee River below Lake Kissimmee at Lake Wales, Florida. Cross section measurement #322, November 9, 1960. Bankfull measures include topographic features, such as a change from a vertical bank to a horizontal flood plain or a change in the size distribution of sediments. Changes in vegetation can be used, but may be subjective, Lawlor (2004). For S.J. Riley (1972), bankfull width is indicated as the point where the Bench Index reaches its first maximum. Analysis of the November 9, 1960 cross section for station 02269000 (Kissimmee River below Lake Kissimmee near Lake Wales) shows a first maximum Bench Index number of 20, at a maximum depth of 4.0 m. The maximum and minimum width/depth ratio is 248 and 40.5, respectively. While Riley (1972) stated that minimum width/depth ratio was not a reliable measure of bankfull because it relied on the shape of the channel, note that bankfull depth is the same for both methods in Figure 5-2.
41
Variability of the bench index values may be caused by smaller benches found within the cross section of the stream. Since a width/depth ratio is easily derived from the information used in Riley’s Bench Index method, the associated graph is included with this method. It should be noted that S. J. Riley (1972) did not recommend using a
300
25
250
20
200
15
150 10
100 50
5
0
0 0.5
1
1.5
2 2.5 3 3.5 Maximum Depth, m
Bench Index
Bench Index
Width / Depth Ratio
width/depth ratio because of its dependence on the specific shape of a channel.
4
W/D Ratio
Figure 5-2. Station 02269000, Kissimmee River below Lake Kissimmee at Lake Wales, Florida. Riley’s bench index and width/depth ratio. The cross section for station 02272500 (Kissimmee River near Cornwell) in Figure -3 is dated December 9, 1948. Multiple channels are shown, and stream gauge records note a main channel and a total system, suggesting a flood condition. The channel used for this analysis is to the far left of the cross section, between 0 m and 50 m from the initial point. According to the stream gauge record, the discharge for the main channel is 70 m3/sec, with discharge for the total system estimated as 104 m3/sec. Width of the main channel is shown as 47 m, while width of the total system is 980 m. The velocity of 0.45 m/sec for the main channel far exceeds the velocity of the total
42
system of 0.16 m/sec. Area of the main channel is estimated 155 m2 while the area of the total system is estimated at 633 m2. The adjusted gauge height for this measurement is 9.75 m. 12
Bed elevation (m)
11 10 9 8 7 6 5 1000
950
900
850
800
750
700
650
600
550
500
450
400
350
300
250
200
150
100
50
0
4
Distance from initial point (m) Figure 5-3. Station 02272500, Kissimmee River near Cornwell, Florida. Cross section #1, December 9, 1948. Analysis of the December 9, 1948 cross section for station 02269000 (Kissimmee River near Cornwell), which has one of the highest discharges in the available data set at 104.4 m3/sec, shows a first maximum Bench Index number of 5, at a maximum depth of 5.0m. The maximum and minimum width/depth ratio is 18 and 9, respectively. Again, note that bankfull depth is the same for both Riley’s Bench Index and Wolman’s (1955) minimum width/depth ratio method in Figure 5-4. The large peak, or bench, in the center of the main channel caused the bench index to yield negative values. For this reason, the bench index begins above this peak, at a maximum depth of 2m. This action also results in equally stepped values in Figure
43
5-4. The channel peak feature is present in cross sections up to December 9, 1950, when it is eroded, leaving smoother, stepped benches in the channel.
Figure 5-4. Station 02272500, Kissimmee River near Cornwell. Riley’s bench index and width/depth ratio For station 02273000 (Kissimmee River at S-65E near Okeechobee) the cross section is dated October 4, 1960. Discharge estimated for the main channel is 270 m3/sec. This value is well over the range found by Warne (1998). The channel used for this analysis is located between 350 m and 450 m from the initial point. According to the stream gauge record, the discharge for the main channel is 270 m3/sec with discharge for the total system estimated as 390 m3/sec. Width of the main channel is shown as 960 m, while width of the total system is the same. The velocity of 0.29 m/sec for the main channel is similar to the velocity of the total system at 0.23 m/sec. Area of the main channel is an estimated 933 m2 while the area of the total system is estimated at 1,730 m2. The adjusted gauge height for this measurement is 8.85 m.
44
10 9 8
Bed elevation (m)
7 6 5 4 3 2 1 1000
950
900
850
800
Distance from initial point (m)
750
700
650
600
550
500
450
400
350
300
250
200
150
100
50
0
0
Figure 5-5. Station 02273000, Kissimmee River at S-65E near Okeechobee. Cross section measurement #389, October 4, 1960. The first maximum bench index number for Station 02273000 (Kissimmee River at S-65E near Okeechobee) is 213, at a maximum depth of 7.0m. The maximum width/depth ratio is 32, with a minimum of 8, as shown in Figure 5-6. Interestingly, this minimum occurs at a maximum depth of 6m, even though the Bench Index indicates a higher bankfull discharge at 7.0 m. The width/depth ratio then jumps up to its maximum value. It should be noted that the channel contained a large, anomalous peak in the center, skewing the width/depth ratio. However, all available cross sections for this station exhibit this feature. In this case, Wolman (1955) is correct in that the shape of the channel causes the width/depth ratio to be somewhat uncertain. Bench index analysis is done on the channel found at approximately 400 m from the initial point.
45
Figure 5-6. Station 02273000, Kissimmee River at S-65E near Okeechobee. Riley’s bench index and width/depth ratio. Note that both Riley’s Bench Indices and the width/depth ratio reveal the same information regarding maximum depth with the exception of the last station near Okeechobee. Channel shape is the probable cause of the discontinuity. Table 5-1 compares the results of the Riley’s Bench Index analysis.
Table 5-1. Comparison of Riley’s bench index results Station Station Riley’s Water Max Number Name, abbr. Bench Elev., W/D m Index Ratio 02269000 …near Lake Wales 20 4.0 248 02272500 …near Cornwell 5 5.0 119 02273000 …near Okeechobee 213 7.0 32
Min W/D Ratio
W, m
Q, m /sec
V, m/sec
3
41
91
91
0.14
32
48
70
0.17
9
223
62
0.29
Graph Analysis Graphs of channel variables in time series show the Kissimmee River to be mostly in equilibrium during the study period. Equilibrium in this context means regular
46
variations about a mean state (Pickup and Reiger, 1979). Deviations from this state generally occur during hurricane season or periods of drought. In Florida, the hurricane season extends from June 1 to November 30 each year. Deviations may also occur during late winter, when frontal weather systems move through the area and occasionally stall, increasing precipitation in the basin. In addition, the time series graphs are utilized to define low velocities for each variable by noting anomalies and differences from the norm in the pattern of measurements. Stations are shown in downriver order, and only Velocity and Width time series are shown for each station.
Time Series Graphs and Low Velocities Eighty-two percent (82%) of the velocity measurements for Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, are equal to or greater than 0.11 m/sec. At this velocity value, there is a distinct break in the pattern of measurements, lending credence to the choice of low velocity measurements. In Figure 5-7, note the grouping of low velocities beginning in 1957. These anomalous groupings also appear in Figure 5-8. Construction of a bridge over the Kissimmee River by the United States Army Corps of Engineers at State Road 60 began in 1956. Part of the bridge project was the creation of causeways at the south end of Lake Kissimmee where it flows into the Kissimmee River. Disturbance of the river channel or flow is a possible cause of the anomalies noted.
47
Figure 5-7. Velocity time series graph for Station 02269000 - Kissimmee River below Lake Kissimmee at Lake Wales. The Width versus Year graph in Figure 5-8 clearly shows the low velocity measurements. With the exception of the anomalous grouping of low values beginning in 1958, all the low velocity values occur above a width of 200 m. This is consistent with the idea that the majority of low velocities for the Kissimmee River occur when the river floods its main channel. The low velocities seen beginning in 1958 are indicative of bridge construction for the State Road 60 bridge further upriver, near the outflow of Lake Kissimmee. Construction included causeways at the east and west sides of the Lake Kissimmee outflow. The velocities observed during this time may be caused by channel disturbance, as the stream gauge is approximately one mile from the State Road 60 bridge.
48
1,000
Width, m
800 600 400 200 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
0
High Velocity
Year
Low Velocity
Figure 5-8. Discharge time series graph for Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales. Historical weather data also illuminates information found in graphs of geomorphic variables. Table 5-2 lists the time of year that low velocity measurements take place for the Lake Wales station. Many of the low velocity measurements can be linked to a year that a tropical cyclone moved across the Kissimmee basin.
Of the seventeen
measurement groups, ten can be linked to years that a tropical cyclone passed over the basin. The following criteria are used to determine which tropical cyclones are shown in Figure 5-9. First, the storm had to cross the Kissimmee basin. Second, the storm had to occur between the years of 1928 and 1960. The storms are classified as tropical depression, tropical storm, or hurricane. A total of thirteen storms crossed the basin during the period of record used in my study. One storm is classified as a tropical depression, two storms are classified as tropical storms, and the remaining ten storms are hurricanes.
49
Table 5-2. Low velocities above bankfull discharge, Lake Wales station. Month(s) and Year(s) of Tropical Storm or Hurricane Month(s) In Which Low Velocity, < 0.11 That Year? Storm Occurred m/sec Nov 1933 Yes Aug, Sept Dec 1933 thru Jan 1934 Yes Aug, Sept Jul thru Oct 1935 Yes Sept, Nov Jan 1938 Yes Jul, Aug, Sept 1937 Aug thru Dec 1941 Yes Oct (2) Mar 1942 No Jul 1943 No Jul 1944 No Sept 1945 Yes Sept Sept thru Dec 1946 Yes Oct (2) Jul 1947 No Mar thru Nov 1948 Yes Sept Dec 1949 Yes Sept Sept 1953 thru Jan 1954 Yes Oct Dec 1957 No Feb thru May 1958 No Apr thru Oct 1959 No National Oceanic and Atmospheric Administration, July 2011
50
Source: Florida Geographic Data Library, 2011, and National Oceanic and Atmospheric Administration, July 2011.
Figure 5-9. Tropical cyclone tracks across the Kissimmee River basin. 51
Figures 5-10 and 5-11 show the Velocity and Width time series for Station 02272500 – Kissimmee River near Cornwell, Florida. This station has fewer measurements than other stations and there is a large void of measurements between 1932 and 1949. The reason for the void is unknown, but may be due to funding, wartime issues, or lack of access to the station. In the Velocity graph, there is a cluster of measurements found at and below 0.20 m/sec. While this station has the fewest measurements, the velocity is similar to that of the other stations. Again, note the presence of low velocities at higher widths in the Width time series graph. This is also the only station showing a maximum width greater than 1,000 meters during the period of record. According to the original data set, this does not appear to be a measurement error. Of the sixty-two (62) velocity measurements used at this station, seventy-one percent (71%) are equal to or greater than 0.20 m/sec.
0.50
0.30 0.20 0.10 0.00 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
Velocity, m/sec
0.40
High Velocity
Year
Low Velocity
Figure 5-10. Velocity time series graph for Station 02272500 – Kissimmee River near Cornwell.
52
3,000
Width, m
2,500 2,000 1,500 1,000 500 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
0
High Velocity
Year
Low Velocity
Figure 5-11. Width time series graph for Station 02272500 – Kissimmee River near Cornwell.
Historical weather data also illuminates the graphs. Table 5-3 lists the time of year that low velocity measurements take place for the Lake Wales station. Many of the low velocity measurements can be linked to a year that a tropical cyclone moved across the Kissimmee basin. Of the eight low velocity measurement groupings, four can be linked to years where one or more tropical storms or hurricanes crossed the Kissimmee River basin, per the National Oceanic and Atmospheric Administration (July 2011). Table 5-3. Low velocities above bankfull discharge, Cornwell station. Month(s) and Year(s) Tropical Storm or Month(s) In Which Of Low Velocity Hurricane Storm Occurred That Year? Jun 1931 No Dec 1948 Yes Sept, Oct May 1949 No Aug thru Dec 1949 Yes Sept Jan thru Mar 1950 No Sept thru Oct 1950 Yes Sept, Oct May thru Aug 1951 No Oct 1951 Yes Oct National Oceanic and Atmospheric Administration, July 2011
53
Velocity and Width time series are shown in Figures 5-12 and 5-13 for Station 03373000 – Kissimmee River at S-65E near Okeechobee. In the velocity time series, a distinct difference in the pattern of measurements can be seen at 0.25 m/sec. In the width graph, note the low velocities observed at high width values, and the lack of low velocity values during the years 1943 to 1950. Although there were five (5) tropical storms or hurricanes that crossed the Kissimmee River basin between 1943 and 1950, they may not have generated enough precipitation to flood the river. Of the 304 velocity measurements taken at this station, seventy-one percent (71%) have a value equal to or higher than 0.25 m/sec.
Velocity, m/sec
0.70 0.60 0.50 0.40 0.30 0.20 0.10 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
0.00
Year
High Velocity
Low Velocity
Figure 5-12. Velocity time series for Station 02273000 – Kissimmee River at S-65E near Okeechobee.
54
1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
Width, m
1,000 900 800 700 600 500 400 300 200 100 0
High Velocity
Year
Low Velocity
Figure 5-13. Width time series Graph for Station 02273000 – Kissimmee River at S65E near Okeechobee. Eight of the fifteen, or 53%, of the low velocity measurement time periods for the Okeechobee station are linked with a year during which one or more tropical cyclones crossed the Kissimmee River basin. Table 5-4 lists the time of year that each low velocity measurement takes place. It then indicates whether there was a cyclone that year, and the month or months in which it occurred. The chart assists the reader in determining not only when the cyclones crossed the basin, but the possible effects increased rainfall may have on discharge and residence time of water in the Kissimmee River basin. Possible effects also include longer flood recession rates. For example, the period listed in Table 5-4 as October 1937 through January 1938 is a longer perioed of low velocity than other periods. However, there were also three cyclones across the basin that season. Less than half of the low velocities occurred outside the traditional hurricane season of June through November.
55
Table 5-4. Low velocities above bankfull discharge, Okeechobee station. Month(s) and Year(s) Tropical Storm or Month(s) In Which Of Low Velocity Hurricane Storm Occurred That Year? Feb thru May 1928 No Jul 1928 thru Jan 1929 Yes Aug (3), Sept Sept 1929 Yes Sept Feb 1930 thru Jun 1930 No Aug 1930 thru Jun 1931 Yes Sept Aug 1932 No Jul 1933 thru May 1934 Yes Aug, Sept, Oct Jun thru Jul 1934 No Sept thru Oct 1935 No Feb thru Oct 1936 Yes Jul, Aug, Sept Oct 1937 thru Jan 1938 Yes Jul, Aug, Sept Sept thru Nov 1939 No Feb thru Sept 1956 No National Oceanic and Atmospheric Administration, July 2011
Geomorphic Variable Graphs After reviewing the time series graphs, graphs of the geomorphic variables versus discharge are constructed. Patterns in the graphs reveal discrete, z-shaped or bifurcated populations. There appears to be at least three separate ways that the Kissimmee responds to changes in discharge. Extremely low velocities occur at either low or high discharges, when periods of drought or flooding occur. High velocities appear before the river has overflowed its main channel. Velocities between the two extremes occur after the river has overflowed the main channel and before it has reached maximum widths. This is consistent with the physical form of the river, as it has multiple channels in a broad, relatively flat valley and floodplain. As also happens in such a river form, multiple channels form, called an anastamosing river.
56
4.0
Bed elevation (m)
3.0 2.0 1.0 0.0 -1.0 -2.0 -3.0 -4.0 0
20
40 60 80 100 120 140 160 180 Distance from initial point (m)
Figure 5-14. Station 02269000, Kissimmee River below Lake Kissimmee at Lake Wales, Florida. Cross section measurement #322, November 9, 1960.
Figure 5-14 is shown here as reference for Figures 5-15 through 5-18. This figure is the November 9, 1960 cross section for the Lake Wales station, and does not indicate a flood condition. It is believed that the vertical lines at 10 m and 170 m from the initial point show places where measurements began and ended, but may not be representative of channel levees that would appear for bankfull discharge calculation purposes. In Figure 5-15, discharge versus velocity for the Lake Wales station is compared. The thick vertical line at 47 m3/sec discharge indicates the point at which velocity drops rapidly, indicating overbank flow. The velocity points valued 0.11 m/sec or less and at discharges greater than 47 m3/sec are indicative of flood conditions. The same value points seen at discharges less than 47 m3/sec are indicative of low flow conditions.
57
Velocity, m/sec
0.50 0.40 0.30 0.20 0.10 0.00 0
50
100
150
200
250
Discharge, m3/sec Figure 5-15. Discharge versus velocity, total system, pre-channelization, Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida.
How width changes with increasing discharge at the Lake Wales station is shown in Figure 5-16. The heavy vertical line appearing at 47 m3/sec of discharge is the point at which the main channel overflows and width increases. Width does not exceed 1,000 m, regardless of discharge during the period of record, demonstrating the limits of the valley. Width values in the horizontal pattern between 100 m and 200 m are believed to show the filling of secondary channels. Width values between 300 m and just over 800 m represent the filling of other secondary channels, while the horizontal observations above 900 m indicate all channels have been filled and the river is contained only by its floodplain. The pattern of observations in this graph is what is called a ‘z-shape’ in my study.
58
Width, m
1000 900 800 700 600 500 400 300 200 100 0 0
50
100
150
200
250
Discharge, m3/sec Figure 5-16. Discharge versus width, total system, pre-channelization, Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida.
Mean Depth versus Discharge in Figure 5-17 is one of the more clear examples of the multiple bankfull discharge mechanisms for the Kissimmee River. As discharge increases, mean depth increases. However, at 47 m3/sec, mean depth drops dramatically. It is here that the main channel overflows. Mean depth then increases slowly at greater discharges. Lower mean depths prior to bankfull discharge are the result of lower discharges, such as when the basin is experiencing a drought. Mean depths above 4 m represent the filling of the secondary channels. In Figure 5-18, area observations follow a pattern more often seen in single channel rivers. That is, area increases linearly to a certain point, then becomes more scattered after the channel overflows. Again, the break point for bankfull discharge occurs at 47 m3/sec.
59
6.00 Mean Depth, m
5.00 4.00 3.00 2.00 1.00 0.00 0
50
100
150
200
250
Discharge, m3/sec Figure 5-17. Discharge versus mean depth, total system, pre-channelization, Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida.
3000
Area, m2
2500 2000 1500 1000 500 0 0
50
100
150
200
250
Discharge, m3/sec Figure 5-18. Discharge versus area, total system, pre-channelization, Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida.
Flow duration curves describe the percentage of time that a discharge value is equaled or exceeded. They are useful for determining the capacity of water or wastewater treatment plants, the frequency of suspended sediment load, habitat
60
suitability to different magnitudes and frequency of streamflow, as well as determining the optimal withdrawal rates from water reservoirs (Vogel and Fennessey, 1994). For the Kissimmee River station at Lake Wales in Figure 5-18, bankfull discharge is shown to be 47 m3/sec. On the flow duration curve, bankfull discharge is exceeded 15% of the time of the period of record. Examining flow duration curves is important to river managers because it not only describes the flow regime, but can reveal the return period for the bankfull discharge. Once a bankfull discharge is found or calculated, it can be placed on the flow duration curve, as shown in Figure 5-19 as a horizontal line. The percent of time bankfull discharge is equaled or exceeded is shown as the vertical line at the intersection of the bankfull discharge in the graph. In this case, the percentage is 15%. One can then take the reciprocal of 15%, or 1 divided by 0.15, and get the estimated return period for bankfull discharge. At the Lake Wales station, this return period is estimated as 6.66 years. Leopold and Wolman (1953 and 1957), and Wolman (1955) describe the average return period of bankfull discharge for rivers in the United States as being five to seven years. The Kissimmee River, at this station, is consistent with that return period.
61
3
Bankfull Discharge = 47 m /sec
Figure 5-19. Flow duration curve. Station 02269000 – Kissimmee River below Lake Kissimmee near Lake Wales, Florida. 12 11
Bed elevation (m)
10 9 8 7 6 5 1000
950
900
850
800
750
700
650
600
550
500
450
400
350
300
250
200
150
100
50
0
4
Distance from initial point (m)
Figure 5-20. Station 02272500, Kissimmee River near Cornwell, Florida. Cross section #1, December 9, 1948. Figure 5-20 is shown as reference for Figures 5-21 through 5-24. This figure is the December 9, 1948 cross section for the Cornwell station. The distance from the initial
62
point is almost 1,000 m, the same as the typical width of the Kissimmee Valley, and as there are multiple channels shown, a flood condition is indicated. The period of record for this station is much shorter than the Lake Wales or Okeechobee station. It contains only sixty-two observations, whereas the other stations have over three hundred observations. While two of the measurements are taken in the early 1930’s, the remainder is grouped between the years 1948 and 1953. Because of the low number of observations for the Cornwell station in the velocity versus discharge chart in Figure 5-21, bankfull discharge is more difficult to determine by using graph analysis. It can be argued that the break comes at 40 m3/sec; however, the value becomes more apparent when examining the other variable graphs. For velocity, there is a drop in the pattern at 52 m3/sec, and the heavy black vertical line
Velocity, m/sec
marker is placed at that value.
1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 0
100
200
300
400
Discharge, m3/sec Figure 5-21. Discharge versus velocity, total system, pre-channelization, Station 02272500 – Kissimmee River near Cornwell, Florida.
63
Bankfull discharge becomes much more apparent in Figure 5-22. With a pattern similar to that found for the Lake Wales station, bankfull is again demonstrated at 52 m3/sec. This station has one measurement for width that falls well over the 1,000 m limit found at the other stations. It is dated October 14, 1953, and has a value of 2,816 m. There was one tropical depression in 1953 that crossed the Kissimmee River basin, and no hurricanes. The other width value that falls outside the norm is located at the discharge value of 250 m3/sec. The corresponding date is October 7, 1949. That year, a category 4 hurricance crossed the basin.
1200
Wdith, m
1000 800 600 400 200 0 0
50
100
150
200
250
300
Discharge, m3/sec Figure 5-22. Discharge versus width, total system, pre-channelization, Station 02272500 – Kissimmee River near Cornwell, Florida.
Mean Depth for the Cornwell station in Figure 5-23 increases linearly and then drops at just under 52 m3/sec, indicating bankfull flow. Depth then increases gradually as discharge increases. Secondary channel filling is indicated by the grouping of values between 1 m and 2.5 m.
64
Mean Depth, m
6.00 5.00 4.00 3.00 2.00 1.00 0.00 100
0
200
300
400
Discharge, m3/sec Figure 5-23. Discharge versus mean depth, total system, pre-channelization, Station 02272500 – Kissimmee River near Cornwell, Florida. In the area chart in Figure 5-24, bankfull discharge is quite easily discerned by the linear increase in area up to 40 m3/sec. Beyond that, increases in area are more scattered. The large area values correspond to October 14, 1953 for the 1,200 m2 measurement and October 7, 1949 for the 820 m2 value.
1400
Area, m2
1200 1000 800 600 400 200 0 0
100
200
300
400
Discharge, m3/sec Figure 5-24. Discharge versus area, total system, pre-channelization, Station 0272500 – Kissimmee River near Cornwell, Florida. 65
At the Cornwell station, bankfull discharge is found to be 52 m3/sec. This is indicated in Figure 5-25 by the horizontal line in the graph. The intersecting vertical line shows that the bankfull discharge is equaled or exceeded 33% of the time of the period of record. This is equivalent to a return period of 3.13 years, slightly longer than that found by Leopold and Wolman (1957). They estimated recurrence intervals at one to two years.
3
Bankfull Discharge = 52 m /sec
Figure 5-25. Flow duration curve. Station 02272500 – Kissimmee River near Cornwell, Florida.
66
10 9 8
Bed elevation (m)
7 6 5 4 3 2 1
1000
950
900
850
800
Distance from initial point (m)
750
700
650
600
550
500
450
400
350
300
250
200
150
100
50
0
0
Figure 5-26. Station 02273000, Kissimmee River at S-65E near Okeechobee. Cross section measurement #389, October 4, 1960. The cross section for the Okeechobee station in Figure 5-26 is dated October 4, 1960. It shares similar characteristics with the Cornwell cross section in Figure 5-20, measured in 1948. The width is over 950 m and there are multiple channels, again indicating a flood condition. The total system discharge measured at the Okeechobee station on October 4, 1960 is 390 m3/sec, with the main channel discharge measured as 270 m3/sec. The discharge measured on this date is one of the largest in the data set. Unlike the Cornwell station, the number of measurements used is three hundred four (304), and the period of record used in my study spans the years between 1928 and 1960. At this station, velocity changes significantly at the bankfull threshold of 86 m3/sec. It is believed by this author that the high velocities above 0.30 m/sec and at greater than
67
86 m3/sec represent secondary channel filling at this station. The lower velocity values below the bankfull threshold represent measurements at low flow, such as when the basin experiences a drought. The floodplain for the Kissimmee River is highly vegetated. Once secondary channels are filled and the water begins to flow over the floodplain, the vegetation would slow the water until discharge increases enough to negate this effect.
Figure 5-27. Discharge versus velocity, total system, pre-channelization, Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida.
Figure 5-28 exhibits the same Z-shaped pattern for width as the Lake Wales and Cornwell stations. Bankfull width is shown as 86 m3/sec, as this is where width begins to increase, rapidly climbing to the maximum width the river valley of just under 1,000 m. Horizontal width measurements as discharge increases indicate the valley filling.
68
Figure 5-28. Discharge versus width, total system, pre-channelization, Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida. Mean depth for the Okeechobee station in Figure 5-29 also has a pattern similar to the Lake Wales and Cornwell stations in that three separate groupings appear. Bankfull discharge is indicated at 86 m3/sec, where the heavy black vertical line marker appears, as this is the point where the main group of mean depth measurements changes. As the river overflows its main channel, mean depth decreases, then gradually rises as discharge increases and flows over the valley plain.
Figure 5-29. Discharge versus mean depth, total system, pre-channelization, Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida.
69
In the Discharge versus Area graph in Figure 5-30, the bankfull discharge is clearly shown at 86 m3/sec. There is a distinct break seen at this point in the area values. Area values greater than 86 m3/sec demonstrate that area increases linearly with increases in discharge. This is a characteristic not so readily apparent in other geomorphic variable graphs.
Area, m2
2000 1500 1000 500 0 0
100
200
300
400
500
Discharge, m3/sec Figure 5-30. Discharge versus area, total system, pre-channelization, Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida.
In the flow duration curve in Figure 5-31, bankfull discharge for the Okeechobee station is shown by the horizontal line in the graph. The intersecting vertical line indicates the percent of time this value is exceeded as 17.5%. Or, that only 17.5% of the measurements used at this station exceed 86 m3/sec. Translating, the 17.5% value is equivalent to a bankfull discharge return period of 5.71 years. This is consistent with Leopold and Wolman (1953 and 1957) and Wolman (1955).
70
3
Bankfull Discharge = 86 m /sec
Figure 5-31. Flow duration curve. Station 02273000 – Kissimmee River at S-65E near Okeechobee, Florida.
Figure 5-32 shows a comparison of discharge values for each methodology. Note that Warne (1998) stated the bankfull discharge range to be greater than 40 to 60 m3/sec. All methodologies yield bankfull discharge values above the range described by Warne (1998) for the Kissimmee. While Riley’s Bench Index has significantly higher values, this author believes it is because the value is derived from a single cross section at a single point in time, rather than over the entire data set. In addition, Williams (1978) suggests that bench indices may overstate bankfull discharge because the analysis grid can end farther out onto the floodplain, thus yielding higher width values. Values for Riley’s Bench Index and minimum width/depth ratio are the same. Graph analysis yields bankfull discharge values similar to that of Warne (1998), but more precise. The graph analysis also indicates that bankfull discharge increases as one moves downstream, which is consistent with the literature. The bench index and width/depth ratio contradict this.
71
Discharge, m3/sec
100 90 80 70 60 50 40 30 20 10 0
Warne, 1998 W/D Ratio Riley's Bench Index Graph Analysis
Lake Wales
Cornwell
Okeechobee
Stations, in downstream order Figure 5-32. Comparison of bankfull discharge values and methodologies.
72
CHAPTER 6 DISCUSSION AND CONCLUSION Discussion Rivers, especially anastomosing rivers, are complex systems. Replication of historic basin flow characteristics (Obeysekera and Loftin 1990) is central to the restoration effort of the Kissimmee River. Brookes, 1990, suggested alternative strategies for river management that integrate both engineering flood control objectives and nature conservation. This is useful because scientists recognize the value of restoring historic flow characteristics while retaining protection of property and life. According to Dunne, 1988, this is where applied geomorphology is best suited. Examining a river from a geomorphic view also furthers the goal of ‘ecological integrity’ that Toth championed in 1995, and Warne recognized as critical to the success of the Kissimmee restoration in 1998. This study seeks to examine pre-channelization flow characteristics in an effort to identify historic bankfull discharge by using various methods. A value or range of values for bankfull discharge is needed to understand the quantity of water necessary to flood the river valley, thus providing habitat needed for all species of flora and fauna to thrive in a manner similar to pre-channelization of the river (Toth 1993). All methods have an innate level of uncertainty (Wohl, 1998), and that uncertainty rises with increases in knowledge. Results found in analyses for the Kissimmee River bear out this axiom. Riley’s Bench Index produces bankfull discharge values different than that of Warne (1998), while Graph Analysis results are very similar.
73
Riley’s Bench Index Riley’s Bench Index is examined as a technique with which to estimate bankfull discharge (Riley 1972). It relies on physical cross sections of the river at flood flows, as well as mathematical indices to find a bankfull discharge. Bankfull width is considered to be the first point at which the index goes to a maximum. One can then use the corresponding discharge to determine bankfull discharge. Only Station 02269000, Kissimmee River below Lake Kissimmee at Lake Wales, has an available single channel cross section of the river prior to channelization, and it is not at flood stage. The remaining stations have flooded cross sections; however, they show multiple channels. In these cases, the main channel is used in the Bench Index. The main channel is verified by examining all available cross sections for consistency of location. All discharges shown in Table 5-1 satisfy the bankfull range of greater than 40 to 60 m3/sec indicated by Warne (1998). The bankfull discharges associated with cross sections used for Riley’s Bench Index are estimated as 119 m3/sec, 70 m3/sec, and 270 m3/sec, for stations 02269000 (Kissimmee River below Lake Kissimmee near Lake Wales), 02272500 (Kissimmee River near Cornwell), and 02273000 (Kissimmee River at S-65E near Okeechobee), respectively. This investigator believes the significantly higher bankfull discharge values for the Lake Wales and Okeechobee stations are due to the use of single cross sections. A single cross section yields a single value at a single point in time, while the other methods analyze multiple values over a long period of record, thus averaging or smoothing large disparities. In addition, Williams (1978) states that Riley’s Bench Index may yield higher bankfull discharges because the index grid often falls farther out onto a floodplain, resulting in larger width values.
74
The large discharge value for the Okeechobee station may also be due to the extreme width of the channel, as well as a peak found in the center of the main channel. This feature exists in all available cross sections for this station. The peak causes multiple maximums in the Bench Index analysis, as the channel widths are added together. Since the index relies on the difference between the widths divided by the differences in depths, a very wide, shallow channel would yield a high index number. Pooling effects as described by Warne (1998) and Toth et al. (2002) due to what is known as the Okeechobee scarp may also play a role in the large discharge figure.
Graph Analyses Review of discharge versus other channel variables shows the Kissimmee River system is not linear in its behavior. Multiple channels are indicated in both the discharge summary data and available cross sections. Plots of the variables show patterns such as bifurcation or a z-shape. Velocity measurements in the graphs reveal distinct populations. These populations are extracted and classified according to either high or low velocity. The majority of low velocity values occur after the Kissimmee River has overflowed its main channel. The remainder of low velocity values occurs during periods of low flow or drought. In turn, the floods that produce these low velocity values often happen during years when a tropical cyclone has crossed the river basin. The groupings revealed in the graphs correspond to filling of a main channel, flooding, and then filling and flooding of subsequent channels. This is supported by the multiple channels found within the Kissimmee River cross section data, as well as relationships between the variables, such as velocity decreasing with large increases in discharge, and the presence of high velocities within narrow ranges. 75
Utilizing the relatively simple methods described in my study, bankfull discharge and velocity is found to be within the range of values in the literature. Given the complexity of the Kissimmee River, choosing methods with which to analyze bankfull discharge is both difficult and necessary. Warne (1998) stated that bankfull discharge and velocities are lower than other similar rivers because of the large size of the channel. Other authors, Wolman (1955) and Pickup and Warner (1976), have suggested that Riley’s Bench Index would be more useful if a local maximum bench index is used. This may certainly be the case for the Kissimmee. However, Riley’s Bench Index and graph analysis of distinct populations used in this study bear out and support their utility, even in a complex system such as the Lower Kissimmee River. Conclusion In order to arrive at plausible bankfull discharge ranges as found by Warne (1998), various techniques are warranted. Williams (1978) suggests using multiple methods to examine bankfull discharge on any given stream. Since Riley’s Bench Index does not depend on regression, it may be a more useful measure of bankfull discharge in rivers such as the Kissimmee if a series of singular channels are examined. While some indices from cross sections correspond to quantities in the range of discharges for the Kissimmee found within the literature, multiple channels again pose challenges. Warne (1998) found that bankfull discharges for the Kissimmee River were greater than the range of 40 m3/sec to 60m3/sec. For Riley’s Bench Index, the values found are significantly higher than Warne’s (1998), with the exception of the Okeechobee station.
76
Since the width/depth ratio method is derived from Riley’s bench index, the bankfull discharge values found using this method are the same as the bench index. Graph analysis yielded bankfull discharges much more similar to Warne (1998). With deeper study and more advanced statistical techniques, it may be possible to examine individual channels and their respective discharges and cross sections to find bankfull values that are consistent and fall within accepted ranges. The objective of my study is to show that various methods can work in predicting bankfull discharge for the Kissimmee. The objective has been met, in that the chosen methods yielded results similar to previous authors. However, further study is recommended. Future Direction Results from discharge graphs indicate non-linearity in the data set. Because of non-linearity and multiple channels in the Kissimmee River, Riley’s Bench Index may be the best way to estimate bankfull discharge in this situation. More sophisticated quantitative methods could yield better results concerning bankfull discharges, but are beyond the scope of this study. Future researchers may wish to examine the Kissimmee River system for other reasons and in other contexts. These contexts can include flood control, water supply, salt water intrusion, groundwater inflow and biological effects. In addition, future methods could include utilizing more sophisticated regression analysis to determine bankfull discharge, investigating aerial photography and using Geographic Information Systems to examine changes in the river channel over time.
77
LIST OF REFERENCES Army Corps of Engineers, South Atlantic Division. Water resources development by the Corps of Engineers in Florida, Office of the Division engineer, Atlanta, Georgia, series-annual. 1931, 1953, 1955, 1957, 1961. Benke, Arthur, C., 1990. A perspective on America’s vanishing streams. Journal of North American benthological society, 9(1), 77-88. Brookes, Andrew, 1988. Channelized Rivers, perspectives for environmental management. John Wiley and Sons, New York, 326p.Brookes, Andrew, 1990. Restoration and enhancement of engineered river channels: some European experiences. Regulated rivers: research and management, 5, John Wiley and Sons, Ltd., New York, 45-56. Brookes, A., 1995a. Challenges and objectives for geomorphology in UK river management. Earth Surface Processes and Landforms, 20: (7), John Wiley and Sons, Ltd., West Sussex, 593-610. Brookes, Andrew, 1995b. River channel restoration: theory and practice. Changing river channels, Angela Gurnell and Geoffrey Petts, ed., John Wiley and Sons, Ltd., New York, 369-388. Brookes, A., Downs, P., Skinner, K., 1998. Uncertainty in the engineering of wildlife habitats, Journal of the chartered institution of water and environmental management, 12(1), Chartered Institution of water and environmental management, London, 25-29. Brower, David R., 1997. Let the river run through it, Sierra, 82 (Mar/Apr), 42, 43, 64. Dunne, T., 1988. Geomorphologic contributions to flood control planning. In, Flood geomorphology, R. Victor, R. Baker, C. Kochel, and P.C. Patton, eds, John Wiley and Sons, New York, 421-438. Fernald, E., Perdum, E., 1998. Water Resources Atlas of Florida. Institute of Science and Public Affairs, Florida State University, Tallahassee, Florida. 312p. Fuchs, U., Statzner, B., 1990. Time scales for the recovery potential of river communities after restoration: Lessons to be learned from smaller streams, Regulated Rivers: research and management, 5(1), John Wiley and Sons, Ltd., New York, 77-87. Gore, J.A., Bryant, F.L.; and Crawford, D.J., 1995. River and stream restoration. In Rehabilitating damaged ecosystems, John Cairns, Jr., ed., Lewis Publishers, Boca Raton, Florida, 245-275. Gottle, A., 1992. Natural design and maintenance of rivers and streams: targets, features and conclusions, Water Science and Technology, 26: (9-11), 2625-2634.
78
Graf, W.L., 1996. Geomorphology and policy for restoration of impounded American rivers: What is ‘natural’? The scientific nature of geomorphology: Proceedings of the 27th Binghamton symposium in geomorphology held 27-29 September 1996. Edited by B.L. Rhoads and C.E. Thorn, John Wiley and Sons, Ltd., 443-473. Jones, R.R., and Malone, W.F., 1990. Mission improbable - the Kissimmee River. Proceedings of Kissimmee River restoration symposium, October, 1988, South Florida Water Management District, West Palm Beach, Florida, 249-251. Kern, Klaus, 1992. Restoration of lowland rivers: the German experience. Lowland floodplain rivers: geomorphological perspectives, P.A. Carling and G.E. Petts, ed., John Wiley and Sons, Ltd., New York, 279-297. Koebel, J.W., Jr., 1995. An historical perspective on the Kissimmee River restoration project. Restoration ecology, 3(3), Society for Ecological Restoration, 149-159. Kondolf, G.M., Micheli, E.R., 1995. Evaluating stream restoration projects, Environmental management, 19(1), Springer-Verlag New York, Inc., 1-15. Kondolf, G.M., 1995. Five elements for effective evaluation of stream restoration. Restoration ecology, 3(2), Society for Ecological Restoration, 133-136. Lane, E. W., 1957. A study of the shape of channels formed by natural streams flowing in erodible material. Missouri River Division Sediment Series No. 9, U.S. Army Engineering Division Missouri River, Corps of Engineers, Omaha, Nebraska. Lawlor, Sean M., 2004. Determination of Channel-Morphology Characteristics, Bankfull Discharge, and Various Design-Peak Discharges in Western Montana. Scientific Investigations Report 2004-5263, U.S. Department of the Interior, U.S. Geological Survey, Reston, Virginia. Leopold, Luna B. and Maddock, Thomas, Jr., 1953. The hydraulic geometry of stream channels and some physiographic implications. U.S. Geological Survey Professional Paper 252. U.S. Government Printing Office, Washington, D.C., 57pp. Leopold, Luna B. and Wolman, M.G., 1957. River Channel Patterns: Braided, Meandering and Straight, U.S. Geological Survey Professional Paper 282-B, 51p. Leopold, Luna B. and Wolman, M.G., 1960. River Meanders, Geological Society of America, Bulletin, v. 71, 769-794. Loftin, M.K., and Obeysekera, J., 1990. Kissimmee River restoration - Engineering considerations in a multiobjective framework. Proceedings of Kissimmee River restoration symposium, October, 1988, South Florida Water Management District, West Palm Beach, Florida, 187-196.
79
Morris, S.E., 1995. Geomorphic aspects of stream channel restoration, Physical Geography, 16: (5), V.H. Winston and Son, Inc., Palm Beach, 444-45. National Oceanic and Atmospheric Administration, http://www.aoml.noaa.gov/general/lib/lib1/nhclib/mwreviews/mwreviews.html Newson, M.D., 1995. Fluvial geomorphology and environmental design, Changing river channels, Angela Gurnell and Geoffrey Petts, eds., John Wiley and Sons, Ltd., New York, 413-432. Obeysekera, J., and Loftin, M.K., 1990. Hydrology of the Kissimmee River basin Influence of man-made and natural changes. Proceedings of Kissimmee River restoration symposium, October, 1988, South Florida Water Management District, West Palm Beach, Florida, 211-222. Park, C.C., 1995. Channel cross-sectional change. In, Changing River Channels, A. Gurnell, and G. Petts, eds., John Wiley and Sons, Ltd., Chichester, 117-145. Petts, G.E., Large, A.R.G., Greenwood, M.T., and Bickerton, M.A., 1992. Floodplain assessment for restoration and conservation: linking hydrogeomorphology and ecology. Lowland floodplain rivers: geomorphological perspectives, G.E. Petts, and P.A. Carling, eds., John Wiley and Sons, Ltd., New York, 217-234. Petts, G.E., 1995. Changing river channels: the geographical tradition. In, Changing River Channels, A. Gurnell and G. Petts, eds., John Wiley and Sons, Ltd., Chichester, 1-23. Petts, G.E., 1996. Water allocation to protect river ecosystems, Regulated Rivers Research and Management, 12: (4-5), John Wiley and Sons, Ltd, West Sussex, 353-365. Phillips, J.D., 1991. Multiple modes of adjustment in unstable river channel crosssections, Journal of Hydrology, 123(1-2), American Society of Civil Engineers, 3949. Pickup, G., and Reiger, W.A., 1979. A conceptual model of the relationship between channel characteristics and discharge. Earth surface processes and landforms, (4), John Wiley and Sons, Ltd., New York, 37-42. Pickup, G., and Warner, R.F., 1976. Effects of hydrologic regime on magnitude and frequency of dominant discharge, Journal of hydrology, (29), Elsevier Science Publishers B.V., Amsterdam, 51-75. Pollock, Sarah, 1989. The charge of the brook brigades, Sierra, 74 (Nov/Dec), 24-28. Rechard, R.P. and Greg Schaefer, 1984. Stripmine streambed restoration using meander parameters. River meandering, American Society of Civil Engineering, New York, 306-317.
80
Reinfelds, Ivars, 1997. Reconstruction of changes in bankfull width, Applied geography, Elsevier Science Limited, Great Britain, 17(3), 203-213. Rhoads, B.L., 1994. Fluvial geomorphology, Progress in Physical Geography, 18(4), Edward Arnold Publishers, 588-608. Riley, S.J., 1972. A comparison of morphometric measures of bankfull. Journal of Hydrology, North Holland Publishing Company, Amsterdam, 17, 23-31. Rinaldi, Massimo, and Johnson, P.A., 1997. Characterization of stream meanders for stream restoration, Journal of Hydraulic Engineering, 123(6), ASCE, 567-570. Rosgen, David L., 1994. A classification of natural rivers, Wildland Hydrology, Catena 22, Pagosa Springs, CO, 169-199. Scarlatos, P.D.; Loftin, M.K.; Obeysekera, J., 1990. Restoration hydraulics: Hydraulic energy gradients along pre-project and revitalized river sections. Proceedings of Kissimmee River restoration symposium, October, 1988, South Florida Water Management District, West Palm Beach, Florida, 223-237. Sear, D.A.; Newson, M.D.; Brookes, A., 1995. Sediment-related river maintenance: the role of fluvial geomorphology, Earth surface processes and landforms, 20, John Wiley and Sons, Ltd., New York, 629-647. Shen, Hsieh Wen, Tabios , III, G., and Harder, J.A., 1994. Kissimmee River restoration study, Journal of Water Resources Planning and Management, 120: (3), American Society of Civil Engineers, 330-349. Shields, F. Doug; Cooper, Charles M.; Knight, Scott S., 1995. Experiment in stream restoration, Journal of Hydraulic Engineering, 121(6), American Society of Civil Engineers, 494-502. South Florida Water Management District, 1995. South Florida Ecosystem Restoration Plan, South Florida Water Management District, West Palm Beach, Florida, 15-25. Stanford, J.A., Ward, J.V., Liss, W.J., Frissell, C.A., Williams, R.N., Lichatowich, J.A., and Coutant, C.C., 1996. A general protocol for restoration of regulated rivers, Regulated Rivers: research and management, 12, John Wiley and Sons, Ltd., New York, 391-413. Toth, L.A., 1990. Impacts of channelization on the Kissimmee River ecosystem. Proceedings of Kissimmee River restoration symposium, October, 1988, South Florida Water Management District, West Palm Beach, Florida, 47-56. Toth, L.A., 1993. The ecological basis of the Kissimmee River restoration plan, Florida scientist, 56(1), 25-51.
81
Toth, L.A., Obeysekera, J.T.B., Perkins, W.A., and Loftin, M.K., 1993. Flow regulation and restoration of Florida’s Kissimmee River, Regulated Rivers: Research and Management, 8: (1-2), 155-166. Toth, L.A., 1995. Principles and guidelines for restoration of river/floodplain ecosystems - Kissimmee River, Florida. In Rehabilitating damaged ecosystems, John Cairns, Jr., ed., Lewis Publishers, Boca Raton, 49-73. Toth, L.A., 1996. Restoring the hydrogeomorphology of the channelized Kissimmee River, River channel restoration: guiding principles for sustainable projects. A. Brookes and F.D. Shields, Jr., eds., John Wiley and Sons, Ltd., New York, 369383. Toth, L.A., Koebel, Jr., J.W., Warne, A.G., Chamberlain, J., 2002. Implications of Reestablishing Prolonged Flood Pulse Characteristics of the Kissimmee River and Floodplain Ecosystem. In: Middleton, B.A. (Ed.), Flood Pulsing in Wetlands: Restoring the Natural Hydrologic Balance, New York: John Wiley & Sons, Inc., 191-221. United States Army Corps of Engineers, www.saj.usce.army.mil/Divisions/Everglades/Branches/ProjectExe/Sections/UECK LO/KRR.htm, April 2011 Vinson, M.R., 1989. Sediment dynamics in meandering and straight sections of a relocated stream channel. In, Restoration, creation, and management of wetland and riparian ecosystems in the American West, A symposium of the Rocky Mountain chapter of the Society of Wetland Scientists, November 14-16, 1988, Denver, Colorado, PIC Technologies and CRS Sirrine, Inc., 76-87. Vogel, Richard M. and Fennessey, Neil M., 1994. Flow-Duration Curves. I: New interpretation and confidence intervals, Journal of water resources planning and management, 120(4), July/August 1994, 485-504. Warne, A.G., 1998. Preliminary geomorphic assessment of Kissimmee River system, Florida. United States Army Corps of Engineers, Waterways experiment station, Vicksburg, Mississippi, 54p. Wharton, Geraldene, 1995. The channel-geometry method: guidelines and applications, Earth surface processes and landforms, 20, John Wiley and Sons, Ltd., New York, 649-660. Williams, Garnett P., 1978. Bank-full discharge of rivers, Water Resources Research, 14(6), United States Geological Survey, Denver, 1141-1154. Winkley, B.R., Schumm, S.A., 1994. River variability – Engineering significance. In, The variability of large alluvial rivers, Stanley A. Schumm, Brien R. Winkley, eds., American Society of Civil Engineers Press, New York, 455- 459.
82
Wohl, E.E., 1998. Uncertainty in flood estimates associated with roughness coefficient, Journal of Hydraulic Engineering, 124(2), American Society of Civil Engineers, 219-223. Wolman, M.G., 1955. The natural channel of Brandywine Creek, Pennsylvania. Geological Survey Professional Paper 271 United States Geological Survey, United States Government Printing Office, Washington, DC, 54p.
83
BIOGRAPHICAL SKETCH Lisa-Ann G. Walsh holds a Bachelor of Arts degree from the University of Florida, 1997. Her major is geography, specifically physical geography. All geographic studies focused on fluvial geomorphology. After completing coursework for her Master of Science degree at the University of Florida, Lisa-Ann began working as an Urban Planner, focusing on community issues such as historic preservation, clean water programs, corridor studies, population studies, transportation, and the utilization of geographic information systems to assist in decision-making. She encourages all municipalities to consider stream restoration within their communities. Lisa received her Master of Science from the University of Florida in the summer of 2011.
84
