<%BANNER%>

Simplified Hydrologic Modeling for Evaluating Surface Water Stage of Historically Isolated Subtropical Wetlands

Permanent Link: http://ufdc.ufl.edu/UFE0045494/00001

Material Information

Title: Simplified Hydrologic Modeling for Evaluating Surface Water Stage of Historically Isolated Subtropical Wetlands
Physical Description: 1 online resource (89 p.)
Language: english
Creator: Guan, Jing
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013

Subjects

Subjects / Keywords: hydrology -- modeling -- recession -- wetland
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Phosphorous is the main nutrient attributed to LakeOkeechobee water quality deterioration and eutrophication. Recently, the historically isolated wetlands in the watershed have been highlighted for phosphorus management based on their specific hydrologic and biogeochemical role to runoff retention. But, the P retention capacity of these wetlands has been reduced greatly from ditching and draining to increase the land area for grazing over the past century. To support managements of restoration of these isolated wetlands, the objective of this study is to develop a simplified predictive model of the dynamics of wetland water stage for historically isolated wetlands in the Lake Okeechobee Basin (LOB), FL. In this study, the recession analysis model simulated the continuous fluctuation of wetland water stage and showed good agreement with R2= 0.6 and 0.7 in two study wetlands. First, the rationale is presented for developing a simplified model of the hydrology of isolated wetlands in the LOB (Chapter 1). The hydrogeological characteristics of the study wetlands in LOB are described,along with input data for this model (Chapter 2). In Chapter 3, a traditional water budget method is introduced to simulate wetland water stage. Finally,features of the simplified recession analysis model are discussed using the above case study (Chapter 4). This study provides a foundation for a coupled hydrologic and wetland plant biomass model for isolated wetlands, which has implications related to phosphorus management in the LOB. Most importantly, the simplified hydrologic model requires few input data and parameters, which reduces the cost and time for estimating wetland hydrology.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jing Guan.
Thesis: Thesis (M.S.)--University of Florida, 2013.
Local: Adviser: Jawitz, James W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045494:00001

Permanent Link: http://ufdc.ufl.edu/UFE0045494/00001

Material Information

Title: Simplified Hydrologic Modeling for Evaluating Surface Water Stage of Historically Isolated Subtropical Wetlands
Physical Description: 1 online resource (89 p.)
Language: english
Creator: Guan, Jing
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013

Subjects

Subjects / Keywords: hydrology -- modeling -- recession -- wetland
Soil and Water Science -- Dissertations, Academic -- UF
Genre: Soil and Water Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Phosphorous is the main nutrient attributed to LakeOkeechobee water quality deterioration and eutrophication. Recently, the historically isolated wetlands in the watershed have been highlighted for phosphorus management based on their specific hydrologic and biogeochemical role to runoff retention. But, the P retention capacity of these wetlands has been reduced greatly from ditching and draining to increase the land area for grazing over the past century. To support managements of restoration of these isolated wetlands, the objective of this study is to develop a simplified predictive model of the dynamics of wetland water stage for historically isolated wetlands in the Lake Okeechobee Basin (LOB), FL. In this study, the recession analysis model simulated the continuous fluctuation of wetland water stage and showed good agreement with R2= 0.6 and 0.7 in two study wetlands. First, the rationale is presented for developing a simplified model of the hydrology of isolated wetlands in the LOB (Chapter 1). The hydrogeological characteristics of the study wetlands in LOB are described,along with input data for this model (Chapter 2). In Chapter 3, a traditional water budget method is introduced to simulate wetland water stage. Finally,features of the simplified recession analysis model are discussed using the above case study (Chapter 4). This study provides a foundation for a coupled hydrologic and wetland plant biomass model for isolated wetlands, which has implications related to phosphorus management in the LOB. Most importantly, the simplified hydrologic model requires few input data and parameters, which reduces the cost and time for estimating wetland hydrology.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jing Guan.
Thesis: Thesis (M.S.)--University of Florida, 2013.
Local: Adviser: Jawitz, James W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045494:00001


This item has the following downloads:


Full Text

PAGE 1

1 SIMP L IFIED HYDROLOGIC MODELING FOR EVALUATING SURFACE WATER STAGE OF HISTORICALLY ISOLATED SUBTROPICAL WETLANDS By JING GUAN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT O F THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Jing Guan

PAGE 3

3 To my grandparents and parents

PAGE 4

4 ACKNOWLEDGMENTS I would like to acknowledge several individuals for th eir assistance and contributions in the completion of this thesis. First, I would like to especially thank my major advisor, Dr. James W. Jawitz, for giving me the opportunity to pursue my beloved research and patience invested in me and my work, and my c ommittee members Dr. Mark W. Clark and Stefan Gerber for their advice and guidance. They were the most approachable professors I have encountered and always were around for advice. Special thanks go to my grandparents and parents for their support and enc ouragement. Special thanks go to all the friends I have met here, especially William Schmahl, Minjune Yang and Yu Fang. In particular, I would like to thank Jing Hu Jiexuan Luo and Xiaolin Liao for their invaluable wisdom they imparted on me. The frien dliness of all of the faculty, staff, and students in this department made completing this research as enjoyable as possible.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ........ 13 1.1 Research Background ................................ ................................ ...................... 13 1.1.1 The Importance of Lake Okeechobee ................................ ..................... 13 1.1.2 The Problems of Lake Okeechobee ................................ ........................ 13 1.1.3 Reasons for the Problems ................................ ................................ ....... 14 1.1.3.1 Land use ................................ ................................ ........................ 14 1.1.3.2 Soil and hydrologic characteristics ................................ ................. 15 1.1.4 Programs for Reducing Phosphorus Loads ................................ ............. 15 1.1.5 Wetland Water Retention (WWR) for Reducing Phosphorus Loads ........ 16 1.1.6 Isolated Wetland Hydrologic Model ................................ ......................... 17 1.1.6.1 Water budget model ................................ ................................ ....... 17 1.1.6.2 Recession analysis model ................................ .............................. 18 1.2 Research Motivation ................................ ................................ ......................... 19 1.3 Research Objective ................................ ................................ ........................... 20 2 MATERIAL AND METHODS ................................ ................................ ...................... 21 2.1 Site Description ................................ ................................ ................................ 21 2.2 Wetland Monitoring ................................ ................................ ........................... 22 2.2.1 Wetland Bathymetry ................................ ................................ ................ 22 2.2.2 Well Install ation ................................ ................................ ....................... 23 2.3 Metrological Data ................................ ................................ .............................. 23 2.3.1 Rainfall ................................ ................................ ................................ .... 23 2.3.2 Evapot ranspiration ................................ ................................ ................... 26 2.3.2.1 Definition of evapotranspiration ................................ ...................... 26 2.3.2.2 Reference evapotranspiration ................................ ........................ 26 2.3.2.3 SFWMD S65DWX station description ................................ ............ 29 2.3.2.4 Crop evapotranspiration ................................ ................................ 30 3 WATER BUDGE T MODEL ................................ ................................ ......................... 44

PAGE 6

6 3.1 Introduction ................................ ................................ ................................ ....... 44 3.2 Quantifying Main Components of Wetland Water Budget ................................ 44 3.2.1 Precipitation and Evapotranspiration ................................ ....................... 44 3.2.2 Groundwater Exchange ................................ ................................ ........... 45 3.2.3 Runoff ................................ ................................ ................................ ...... 46 3.2.4 Ditch flow ................................ ................................ ................................ 47 3.2. 5 Result ................................ ................................ ................................ ...... 48 3.3 Simplified Water Budget Model ................................ ................................ ......... 48 3.3.1 Simplified Climate Water Budget Model ................................ .................. 49 3.3.2 Combined Simplified Water Budget Model ................................ .............. 50 3.4 Conclusion ................................ ................................ ................................ ........ 51 4 RECESSION ANALYSIS MODEL ................................ ................................ .............. 59 4.1 Flow Recession Analysis Introduction ................................ ............................... 59 4.2 Flow Recession Analysis in Groundwater ................................ ......................... 59 4.3 Recession Analysis Method in Wetland Water Stage Analysis ......................... 62 4.3.1 Monthly Wetland Surface Water Stage Analysis ................................ ..... 63 4.3.2 Daily Wetland Water Stage Analysis ................................ ....................... 65 4.3.2.1 Result and discussion ................................ ................................ .... 66 4.3.2.2 Sensitivity analysis ................................ ................................ ......... 66 4.3.2.3 Wetland water stage prediction ................................ ...................... 67 4.4 Conclusion ................................ ................................ ................................ ........ 67 5 RECOMMENDATIONS FOR FUTURE WORK ................................ .......................... 77 5.1 Model Improveme nt ................................ ................................ .......................... 77 5.2 Model Generalization ................................ ................................ ........................ 77 5.3 Coupled Model ................................ ................................ ................................ .. 78 LIST OF R EFERENCES ................................ ................................ ............................... 79 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 89

PAGE 7

7 LIST OF TABLES Table page 2 1 Monitoring period s, wetland footprint areas, relative elevation of ditch and topographic range in study wetlands. ................................ ................................ 32 2 2 Information of 4 SFWMD weather stations. ................................ ........................ 32 2 3 The relation between weather station S65D and other 3 sites. .......................... 32 2 4 The slopes, RMSE, and determination coefficients (R 2 ) of the measured value and the estimated values ................................ ................................ ......... 33 2 5 Monthly averaged net radiation (R n ) (Unit MJ m 2 day 1 ) collected from S65DWX weather station. ................................ ................................ .................. 33 2 6 Monthly averaged reference ET 0 (Unit m) calculated with the FAO56 PM method using climate data form SFWMD S65DWX. ................................ .......... 34 4 1 The recession rate ................................ 68 4 2 The estimated recession rate ................. 68 4 3 The coefficients for linear function of ................................ ............................ 69 4 4 The recession rate of Larson wetland surface water. ................................ ......... 69

PAGE 8

8 LIST OF FIGURES Figure page 2 1 The location of Dixie Larson Ranch. ................................ ................................ ... 35 2 2 The location of two study wetlands Larson East (LE) and Larson West (LW). ... 35 2 3 Relationship between wetla nd water stage and flooded area. ............................ 36 2 4 Mini troll STP, In situ, Inc. logging pressure transducers. ................................ .. 37 2 5 PVC screened monitori ng well. ................................ ................................ ........... 37 2 6 Onset Communications Corp. Model RG3 M logging tipping buckets. ............... 38 2 7 Location of Larson Wetland and SFWMD weat her stations. .............................. 38 2 8 Setting site S65D as target site, the correlation of the measured value of target site and the estimated values. ................................ ................................ .. 39 2 9 Correlation between in situ rainfall data and weighted average based estimation when the data are available from onsite and weather stations. ......... 40 2 10 The daily rainfall rates from Mar ch 2004 to Mar ch 2006. ................................ .... 40 2 11 The daily rainfall estimation from January 2006 to October 2010. ...................... 41 2 12 Monthly net radiatio n trend line, derived from the monthly averaged net radiation from 2004 to 2010. ................................ ................................ ............... 41 2 13 Correlation of reference ET and net radiation (R n ). ................................ ............ 42 2 14 C rop ET calculated using the reference ET multipl ied by Kc during the monitoring p eriod ................................ ................................ .............................. 42 2 15 Correlation between monthly averaged wetland ET c in Larson wetland and op en water lysimeter based ET measured by Mao et al. (2002). ........................ 43 3 1 Conceptual diagram of water budget in historically isolated wetlands in the Lake Okeechobee basin. ................................ ................................ .................... 53 3 2 Conceptual diagram showing groundwater flow between wetland and upland modeled in this study (Charbeneau 2000). ................................ ......................... 53 3 3 Illustration of ditch geometr y. ................................ ................................ .............. 54 3 4 The wetland water budget based simulated wetland water stage with the measured water stage (Min et al. in process). ................................ .................... 54

PAGE 9

9 3 5 The daily wetland water level (H est ) estimation with simplified climate water budget model compare to measured wetland water level (H wet ). ........................ 55 3 6 Plot of F (..) versus daily precipitation ................................ ................................ 56 3 7 wet derived from combined simplified water wet based on measured wetland water stage. ................... 56 3 8 A linear function was regressed to describe the r esiduals of wetland water balance. ................................ ................................ ................................ .............. 57 4 1 The 6 natural drawdown events. ................................ ................................ ......... 70 4 2 During natural drawdown period, the nonlinear ex ponential recession trend of groundwater discharge. ................................ ................................ ...................... 71 4 3 The recession curve modeling of LW ( ) and LE ( ). ................................ ................................ ................................ 72 4 4 Fluctuation of monthly averaged wetland stage ................................ ................. 73 4 5 Estimated wetland surface water level (H est ) with monthly recession analysis method compared to measured wetland surface water level (H wet ) ................... 74 4 6 Estimated wetland stage (H est ) with daily recession analysis method compared to measured wetland water level (H wet ). ................................ ............. 75 4 7 Plot of Larson wetland water stage prediction from January1 2006 to October 6 2010. ................................ ................................ ................................ ............... 76

PAGE 10

10 LIST OF ABBREVIATIONS D Ditch Flow ET Evapotranspiration ET o Reference Crop Evapotranspiration ET c Crop E vapotranspiration GW Groundwater Exc hange Kc Crop Coefficient LOAP Lake Okeechobee Action Plan LOB Lake Okeechobee Basin O Runoff P Precipitation R 2 Determination Coefficient RMSE Root M ean S quare Error SFWMD South Florida Water Management District STDEV Standard Deviation TMDL Total Maxim um Daily Load WWR Wetland Water Retention

PAGE 11

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 SIMPL IFIED HYDROLOGIC MODELING FOR EVALUATING SURFACE WATER STAGE OF HISTORICALLY ISOLATED SUBTROPICAL WETLANDS By Jing Guan May 2013 Chair: James W. Jawitz Major: Soil and Water Science Phosphor us is the main nutrient attributed to Lake Okeechobee water quality deterioration and eutro phication. Recently, the historically isolated wetlands in the watershed have been highlighted for phosphor o us management based on their specific hydrologic and biogeochemical role to runoff retention. But, the p hosphor o us retention capacity of these wetla nds has been reduced greatly from ditching and draining to increase the land area for grazing over the past century. To support managements of restoration of these isolated wetlands, the objective of this study is to develop a simplified predictive model o f the dynamics of wetland water stage for historically isolated wetlands in the Lake Okeechobee Basin (LOB), FL. In this study, the recession analysis model simulated the continuous fluctuation of wetland water stage and showed good agreement with R 2 = 0.6 and 0.7 in two study wetlands. First, the rationale is presented for developing a simplified model of the hydrology of isolated wetlands in the LOB (Chapter 1). The hydrogeological characteristic s of the study wetlands in LOB are described, along with inp ut data for this model (Chapter 2). In Chapter 3, a traditional water budget method is introduced to simulate wetland water

PAGE 12

12 stage. Finally, features of the simplified recession analysis model are discussed using the above case study (Chapter 4). This study provides a foundation for a coupled hydrologic and wetland plant biomass model for isolated wetlands, which has implications related to phosphor o us management in the LOB. Most importantly, the simplified hydrologic model requires few input data and parame ters, which reduces the cost and time for estimating wetland hydrology.

PAGE 13

13 CHAPTER 1 INTRODUCTION 1.1 Research Background 1.1.1 The Importance of Lake Okeechobee Lake Okeechobee covering approximately 1890 square kilometers, is the largest freshwater lake i n the Florida, the seventh largest freshwater lake in the United States and the second largest freshwater lake contained entirely within the conterminous U.S. ( Muller 2008 ) It is the central component of the hydrologic system of central Florida ( Havens 2005 ) Lake Okeechobee holds one trillion gallons of water, serving as major water supply source to agricultural and urban user and the Everglades National Park. The Lake also contributes to flood control, navigation, recreation, and fish and wildlife protection ( SFWMD 2013 ) 1.1.2 The Problems of Lake Okeechobee The depth of Lake Okeechobee is exceptionally shallow for its large size, with an average depth of three meters ( Henkel 2010 ) In recent years, nutrient enriched runoff from surrounding watersheds has contributed to the shallow sub tropical lake moving from a naturally eutrophic state to a hyper eutrophic stat e in the 1980s ( Reddy 1995 ) Phosphorus is the most critical nutrient influencing water quality of Lake Okeechobee ( Flaig 1995b ; Havens 2005 ; McCaffery 1976 ) T he recent five year averaged (2008 2012) total p hosphorus load to the lake from whole drainage basins and atmospheric deposition is 387 metric tons, almost 2.8 times greater than the 140 metric ton Total Maximum Daily Load (TMDL) per year ( SFWMD 2012 ) TMDL is regulated in the U.S. Clean Water Act and is described as the maximum amount of certain pollution that a water body can accept while stil l meeting water quality standards ( Houck 2002 ) The

PAGE 14

14 high p hosphorus loads have contributed to the frequent and extensive noxious blue green algal blooms in the l ake. Consequently, the lake becomes more eutrophic, losing biodiversity of benthic invertebrate and causing cattail overspread. Over enrichment could seriously impair water quality and, thereby, affect water users ( SFWMD 2012 ) 1.1.3 Reasons for the Problems 1.1.3.1 Land u se The high p hosphorus loads have been largely attribute d to numerous dairies located in the northern drainage basins ( Flaig 1995b ; Goldstein 1995 ; Haan 1995 ; Reddy 1995 ) The primary routes of p hosphorus transport to the lake are through surface and subsurface from uplands to adjacent wetlands and streams, which terminally discharge into the lake ( Reddy 1995 ) The primary land uses of the watershed are cattle ranches and dairy farms. Approximately 1900 square kilometers improved pasture an d 1350 square kilometers of unimproved pasture support about 180,000 beef cattle in the watershed ( Flaig 1995a ) Beef cattle and dairies are the major revenue indus tries for LOB. By the end of June 2012, there were 170,700 cattle and calves in this area. The primary feed for dairy cows contain high phosphor o us materials ( SFWMD 2012 ) Historically, in LO B small depressional isolated wetlands were abundantly distributed, occupying 25% of the area ( McCaffery 1976 ) These isolated wetlands may play an important role in re ducing phosphorus loss from uplands to the lake ( Hiscock 2003 ; SFWMD 2013 ) But, with the development of agric ulture, many isolated wetlands have been drained by shallow ditches for grazing and crop lands. Now there are only about 615,081 acres wetlands, occupying about 17.9% of the Lake Okeechobee Basin

PAGE 15

15 ( SFWMD 2004 ) Drained pastures result in greater runoff, bringing more high phosphorous cattle manures loads to the Lake Okeechobee ( SFWMD 2012 ) 1.1.3.2 Soil and hydrologic c haracteristics The soils of LOB consist of Spodosols, Entisols, and Histosols. The majority of the soils in northern Lake Okeechobee watershed where most of the dairies and beef ranches are located are Spodosols ( USDA 1990 ) These soil s are characterized by very low phosphor o us retention capacity and high water permeability ( Allen 1987 ) Phosphorus loads to Lake Okeechobee are controlled by water table fluctuations and uplands runoff ( Gatewood 1975 ) The annual average precipitation of the LOB is 120cm ( Flaig 1995b ) During the wet season, the water table is between the Spodic horizon and the soil surface, which attributes to significant phosphorus transport in subsurface water. When the soils become saturated and then runoff is generated, high phosphorous cow manure particles move with runoff into streams and wetlands ( Reddy 1995 ) 1.1.4 Programs for Reducing P hosphorus Loads Over the last decades, in order to rehabilitate the ecosystem of Lake Okeechobee, various federal and state programs have been working cooperatively to reduce nutrient loads to Lake Okeechobee, particularly phosphorus These efforts have made some difference. I n 1993 Okeechobee Surface Water Improvement and Management Plan regulated cattle operations in the basin ( SFWMD 1993 ) In 1999, Lake Okeechobee Action Plan (LOAP) was developed by South Florida Water Management District ( SFWMD ) as part of South Florida Ecosystem Restoration ( Harvey 1999 ) However, those programs have not been sufficient to meet phosphorus load

PAGE 16

16 reduct ion target. Additional phosphorus control strategies still need to be developed to further reduce phosphorus loads to the lake. Four practical remediation methods have been proposed in LOAP to minimize phosphorus discharges to the Lake: (1) runoff reten tion from old phosphorous hot spots, (2) remediation of phosphorus rich soils, (3) rehabitation of wetlands, and (4) impoundments of stormwater flow ( SFWMD 2012 ) Among th o se, Wetland Water Retention (WWR) has shown promise for reducing phosphorus discharge to surface waters and groundwater from ranches in the LOB ( Shukla 2011 ) 1.1.5 Wetland Water Retention (WWR) for Reducing P hosphorus Loads The LOAP proposed us ing naturally occurring depressional isolated wet lands, abundantly distributed in LOB, to retain runoff for reducing phosphorus ( Harvey 1999 ) Isolated Wetlands in the LOB serve as phosphorus sinks, reducing downstream eutrophication ( Davis 1982 ) High removal rates of many nutrients have been reported for wetlands which has led to serious consideration of wetlands for phosphorus reduc tion ( Bastian 1979 ) Wetlands in the area reduce phosphorus loads via two approaches. The first is physical processes: reduction of the nutrient enriched water reaching the lake. S torm water runoff can be effectively attenuated by increasing regional water storage. Much of the water stored in isolated wetlands will be evaporated, thereby not becoming the phosphorus discharge of the lake ( Bottcher 1995 ; Harvey 1999 ) Second are biological and chemical processes A significant portion of phosphorus in wetlands is removed by macrop h ytes and algae, organic deb ri s accumulation, and soil adsorption ( Richardson 1985a ) Soil charact eristics such as amount of clay, percentage of Fe and A l oxides and Ca compounds, and pH, determine

PAGE 17

17 reactions o f phosphorus including forming Fe and Al bound P and Ca P ( Richardson 1985b ) 1.1.6 Isolated Wetland Hydrologic Model Hydrology as one of the primary controlling factors for wetlands, determines the vegetation community, soil and wetland functions through determining depth, hydroperiod a nd water flow patterns ( Hammer 1989 ) H ydrologic variations could be described by wet land w ater level ( EPA 2008 ) Ho wever, the hydrology of such wetlands is relatively poorly understood because of their highly transient hydrologic characteristics ( LaBaugh 1986 ; Lee 2009 ; Winter 2003 ) To improve management for phosphorus retention in these isolated wetlands, characteriza tion of their hydrology is needed 1.1.6.1 Water b udget m odel W ater budget model s are generally constructed to investigate wetlands hydrology ( Drexler 1999 ) The water budget model is a traditional valuable tool for assessing water availability, watershed characteristics, and water management. Water budgets provide a basis for understanding hydrologic processes of a wetland ( Gleick 1986 ) The wetland water budget should contain total amount of inflows and outflows within a wetland. Major components of the wetland water balance are precipitation (P) runoff ( O) groundwater exchange (GW) and evapotranspiration (ET) ( Carter 1996 ) Water balance models were developed by Thornthwaite ( Thornthwaite 1948 ) These models have been modified and widely applied to hydrological problems ( Xu 1996 ) Water budgets have been used in various kinds of depressional wetlands systems: it has been computed three natural Carolina bay wetlands in Bladen County, Nor th Carolina, USA ( Caldwell 2007 ) ; the mechanisms of water and solute transfer

PAGE 18

18 between the wetland and the surrounding upland in a small catchment in Saskatchewan, Canada has been evaluated with a snowmelt water budget ( Hayashi 1998 ) ; Drexler used a water budget model to quantify nutrient loading in a small peatland ( Drexler 1999 ) ; a conceptual wetland water budget model was developed to describe the interaction s in a wetland in the Swan Coastal Plain in Perth, Western Australia ( Krasnostein 2004 ) ; a water budget model was conducted in Saskatchewan, ( Parsons 2004 ) ; and Wilcox et al. has investigated a sedge fen affected by diking and ditching with a water budget approach ( Wilcox 20 06 ) However, water balance models have their shortcomings: requiring long term monitoring, difficulty in extrapolation of hydrologic components, and needing large input data and parameter collection ( Drexler 1999 ) In addition, beca use of the complexity of the wetland ecosystem, there is still a great deal of uncertainty over the hydrologic budgets ( Owen 1995 ) There are few water budget model s that are completely specific for wetlands ( Mitsch 1993 ) 1.1.6.2 Recession a nalysis m odel Process based models, like the water balance model are always constrained by requiring extensive data and parameters ( Todini 2007 ) In contrast, empirical models require fewer parameters from effective and spatial averaged estimations ( Breuer 2009 ) Due to the problems associated with monitoring and the continued data required for such water budgets model in predicting these isolated wetlands, an empirical method recession analysis model ing is proposed to investigate these isolated wetlands.

PAGE 19

19 Recession is the processes of surface wate r or groundwater discharge gradually diminishing during periods of little or no precipitation. Recession behavior is determined by the properties of the watercourse and aquifer ( Franchini 1991 ) Recession analysis is an empirical graphical method using graph to analyze hydrologic regime s R ecession analysis has a long history, Boussinesq and Maillet developed the exponential function for hydrographic quantitative analysis and proved its reasonability ( Boussine sq 1877 ; Maillet 1905 ) Flow recession analytical method could be used to quantify and model the natural stream flow and ground water discharge process. Recession analysis is popular in water resource management, such as low flow forecast for irrigation management, hydroelectric power plants, and pollutant concentration. Recession analysis as an indirect method can assist in determining watershed hydrologic characteristics ( Demuth 1993 ; Harlin 1991 ; Kachroo 1992 ; Korkmaz 1990 ; Vogel and Kroll 1992 ) Brutsaert has used the recession analysis method for estimating hydrogeological parameters in a low stream flow basin ( Brutsaert 1998 ) A stream aqu ifer recession analysis model was extended to a watershed scale to simulate low flow behavior in Massachusetts ( Vogel and Kroll 1992 ) A base flow recession curve can be used to estimate the time rate of groundwater outflow after recharge as base flow is discharged by groundwater ( Knisel 1963 ; Meyboom 1961 ) Rorabaugh and Glover have used the empirical method for groundwater discharge in an idealized, homogenous aquifer ( Glover 1964 ; Rorabaugh 1964 ) 1.2 Research Motivation Elevation of trophic level caused by non point sources of agricultural pollution may result in water quality deterioration, habitat loss and agro ecosystem destruction.

PAGE 20

20 Isolated wetlands are locate d in the lowest physiographic landscape position and represent a considerable hydrologic and nutrient store. Restoring drained wetlands in the LOB may reduce phosphorus loads to Lake Okeechobee. Understanding the primary mechanisms of phosphorous retention and restoration of these isolated wetlands within the basin are critical research directions for reducing nutrient loads to the lake. In order to explain the various wetland ecological phenomena and ecosystem processes, we need understand their hydro log y ( Price and Waddington 2000 ) However, the hydrological characteristics of subtropical isolated wetlands are complicated by seasonal and hydrologic extremes ( Parsons 2004 ) Further, not all the hydrological flowpaths can be measured easily. Therefore, there is a need for simplified models to predict and estimate h ydrologic fluctuation based on exsiting data. 1.3 Research Objective This study focuses on developing a simplified predictive model of the dynamics of wetlands water stage. This study use s a recession analysis model requiring less input data and parameter s from field to simulate the continuous fluctuation of wetland water stage. Through this hydrologic model we can obtain useful fundamental hydrologic research for future work.

PAGE 21

21 CHAPTER 2 MATERIAL AND METHODS The principal aim of this chapter is to collec t and process the required input data and parameters for the simplified hydrologic model to simulate and predict wetland water stages. This chapter consists of three parts: (1) the introduction of study wetland; (2) a description of the equipment and metho ds for wetland monitoring; and (3) methods for collecting and processing metrological data. 2.1 Site Description The study site is located at the Dixie the Lake Okeechobee watershed, Florida (USA) (Fig ure 2 1 ). Two wetlands were studied: Larson East (LE) and Larson West (LW) ( Fig ure 2 2 ). These sites are in priority sub basins that have been identified as relatively higher contributors of nutrient loads to Lake Okeechobee, compared to the surrounding lake sub basi ns ( Fl aig 1995b ) The two study isolated wetlands are within 113 hectare Dixie Larson ranch, which has been raising heifers since 1974. From 1992 to present, the ranch changed to maintained cow and calf. About 150 to 200 head of cattle rotate on the pastures a nnually. These study wetlands are historically isolated or depressional wetlands. Geographically, isolated wetlands do not exhibit surface water connectivity with rivers, lakes, oceans, or other water bodies ( Leibowitz and Nadeau 2003 ; Whigham and Jordan 2003 ) Tiner defined isolated wetland as completely surrounded by uplands consisting of non hydric soils an d non aquatic vegetation ( Tiner 2003 ) The two study isolated wetlands were drained for more pasture by an extensive network of ditc hes in the mid 1950s, but were not drained completely. Each of the wetlands has only one

PAGE 22

22 drainage ditch, and is surround ed by pastures. Isolated wetlands in the Okeechobee basin are generally less than 1m deep, appear bowl shaped, and are relatively small ( Lewis 2001 ) Larson East (LE) and Larson West (LW) are about 2.1 hectares and 2.6 hectares respectfully ( Fig ure 2 2 ). The format ion of wetlands in this region is due to sink hole karst geology. This area has low topographic relief with average slopes from 0 to 2% and shallow water tables ( Blatie 1980 ) The soil types of Larson wetlands belong to Basinger and Placid soils and those of surroundin g uplands are Myakka Immokalee Basinger fine sand, both of which are characterize d as very deep, poorly drained, rapid permeable, sandy marine sediments ( Lewis 2003 ) Wetland soils contain higher organic matter and show lower saturated hydraulic conductivity compared to upland soils. Bhadha and Jawitz discovered that sandy soil is up to 100 cm at upland, and below that loam a nd sandy clay were observed ( Bhadha and Jawitz 2010 ) The two study wetlands have similar vegetation communities. Typical vegetat ion at these wetlands includes: Juncus effuses Panicum sp. Pontedaria cordata var. lancifolia Ludwigia repens Hydrocotyle ranunculoides and Andropogon glomeratus Bahia grass ( Paspalum notatum ) is predominant species ( Dunne 2007 ) 2.2 Wetland Monitoring 2.2.1 Wetland Bathymetry The detailed topographic surveys were conducted with combination of a line of sight laser level and a hand held Garmin GPS in the two study wetlands. The elevation point data were interpolated wit h ordinary kriging. Then form wetlands elevation contours with ArcGIS (ESRI ArcMap 9.1). The boundaries of surveys were by the appearance of palm, which was approximately 20 m on upland ( Min 2010 ) During the

PAGE 23

23 monitoring period, wetlands water stages have not exceed the boundary. According to the analysis of bathymetric of two wetlands, the w etland flooded area could be estimated ( Fig ure 2 3 ). Other information from the survey is shown in Table 2 1 2.2.2 Well Installation Two years of wetland stages monitoring data were used form Mar ch 2004 to Mar ch 2006 in Larson wetlands ( Min 2010 ) Logging pressure transducers (Mini troll STP, In situ, Inc.) ( Fig ure 2 4 ) were installed in diameter 0.032 m and depth 2.5 to 3.0 m screened PVC monitoring wells at the deepest point of wetlands ( Fig ure 2 5 ). This kind of logging pressure transducers has a range of 0 to 34 kPa with 0.01% accuracy. Wetland water stages were recorded by pressur e transducers in half hour interval. Then these data were daily averaged. 2.3 Metrological Data 2.3.1 Rainfall Rainfall w as measured by logging tipping buckets (Onset Communications Corp. Model RG3 M) ( Fig ure 2 6 ). But not all gauge based data are relia ble: errors result from systematic measuring errors (evaporation loss or wind drifting) and from incomplete observations (spatial coverage, temporal gaps) ( Rudolf and Schneider 2005 ) For the physical obstructions of onsite rainfall gau ges such as flow measurement structures and unprofessional maintain influence the rainfall data. In addition, some weather factors, such as wind speed or wind direction during rainfall events, also affect the rainfall data. Sometimes the onsite instrum entation malfunctioned for electronic system failure or inlets were stuck by something periodically. Therefore, we collected rainfall data from nearby weather stations for data onsite based calibration and remaining portions. The 4

PAGE 24

24 weather stations are man aged by South Florida Water Management District (SFWMD) and surrounding the study wetlands ( Fig ure 2 7 ). The difference of observed rainfall data from 4 weather stations is relatively large, as the rainfall amounts vary spatially over Lake Okeechobee bas in. The spatial variations of rainfall amounts reduce as rainfall duration increases. Therefore, calculated methods of mean rainfall are used in rainfall estimation. There are many methods to calculate mean precipitation in a catchment including arithmeti c means method, weighted average method, Thiessen polygon method, and isohyets methods ( Eagleson 1967 ) The Thiessen polygon method is generally used in large watersheds, while weighted average method shows advantage in analyzing different kinds of precipitation events in small areas ( Brunsdon 2001 ) The weighted average method is derive d from ( 2 1 ) where P is the estimated precipitation of the area [mm], P i is the observed values of a same rain event from different weather station[mm] ,d i is distance between study wetlands and weather station[km],n is the number of weather station. In some places with mild t opographic relief and even precipitation, the simplest arithmetic means method may be sufficiently accurate The arithmetic means method is expressed as ( 2 2 )

PAGE 25

25 where P is the average precipitation of the are a [mm], P 1 P 2 P n is the observed values of a same rain event from different weather station[mm], n is the number of weather station. Rainfall data c ollection In order to minimize the errors from incomplete observations both the arithmetic mean method an d weighted average methods were evaluated. D etail information of 4 weather stations is in T able 2 2 First, one of the 4 weather stations was set as target site and the estimated value of the target site was calculated using observed data from the other 3 stations using both methods T he estimated values were compared with observed value s of target site s The relationship between weather station S65D and other 3 weather stations are shown in Table 2 3 as an example The determination coefficient (R 2 ) of t he measured value of site S65D and the estimated values with weighted average method is 0.5344, and R 2 is 0.4 with arithmetic mean method ( Fig ure 2 8 ). With the method above, Table 2 4 summarize the slope, R 2 and Root mean square error (RMSE) of 4 weathe r stations respectively. Based on Table 2 4 the weighted average method is more reliab le for rainfall estimation compar ed to the arithmetic mean. Therefore, weighted average based estimations were used to validate in situ rainfall data and fill gaps in t he measured data between Mar ch 2004 and Mar ch 2006 The weighted average method was also applied to estimate precipitation from Jan uary 2006 to Oct ober 2010 for the recession prediction model in Chapter 4. In situ rainfall data correlated well with the we ighted average based estimation (determination coefficient R 2 =0.6) ( Fig ure 2 9 ) when the data were available from onsite and weather station. The daily rainfall rates from Mar ch 2004 to Mar ch 2006 are

PAGE 26

26 shown in Fig ure 2 10 and daily rainfall rates from J an uary 2006 to Oct ober 2010 are shown in Figure 2 11 About two thirds of the annual precipitation occurs in August, September and October, which is considered the wet season. The period from November to April is the dry season, and May and October are t ransitional months. 2.3.2 Evapotranspiration This section explains the concepts and measurement of reference crop evapotranspiration (ET o ) and actual crop evapotranspiration (ET c ). 2.3.2.1 Definition of e vapotranspiration Evapotranspiration (ET) represent s the total amount of water evaporated from plants or soil surface s plus the amount of water that transpires from plants into the atmosphere as water vapor ( Cuenca 1989 ) It is difficult to measure evapotranspiration separately into evaporation and transpiration, so the term evapotranspiration is widely used in water balance. Evapotranspiration can be measured directly by lysimeters which are tank s filled with soil on which cr ops are grown in natural environment and measure the amount of water loss through evaporation and transpiration ( Evett 2012 ) Lysimeter s are widely used to study climatic effects on evapotranspiration. Climate parameters (solar radiation, air temperature, humidity, and wind speed), crop characteristics and managements are factors affecting evapotranspiration. ET is commonly calculated based on daily scale and is expressed as mm day 1 or MJ m 2 day 1 ( Valiantzas 2013 ) 2.3.2. 2 Reference e vapotranspiration Reference evapotranspiration is defined as evapotranspiration rate from a reference crop surface, commonly express as ET 0 The FAO Expert Consultation on

PAGE 27

27 Revision of FAO Methodologies for Crop Water Requirements defin ed refer ence fixed surface resistance of 70 s m 1 defined as a uniform surface of dense, actively growing vegetation having specified h eight and surface resistance, not short of soil water, and representing and expanse of at least 328 ft. of the same or similar vegetation ( Allen 2005 ) Because of the difficulty of obtaining accurate field evapotranspiration measurements, ET is commonly comp uted from weather data. C limatic parameters are the sole impact factors for ET 0 Consequently, ET 0 can be computed from climate data, without considering characteristics of crop and soil. There are a large number of empirical methods to estimate ET from cl imatic variables. Some of these methods are derived from Penman equation ( Penman 1948 ) including FAO Pe nman ( Dooren bos 1977 ) Kimberly Penman ( Wright 1982 ) Penman Mon teith ( Allen 1994 ) and the FAO Penman Monteith ( Allen 1998 ) Among them FAO 56 Penman Monteith method (FAO56 PM) is most recommended for its close approximation and combination of energy balance and aerodynamic formula ( Allen 2005 ) The FAO56 PM equation represents the process of evapotranspiration in a simple way concerning physical and physiological governing factors. It has successfully been used in estimating ET in Florida wetl and ecosystem ( Abtew 1996 ; Jacobs et al. 2002 ; Mao et al. 2002 ) The FAO56 PM method to compute reference evapotranspiration from climate data is as follows: ( 2 3 )

PAGE 28

28 where ET 0 is reference evapotranspiration [mm day 1 ]; T is mean daily air temperature at 2 m height [ ], calculated as average of minimum and maximum daily air temperatures; R n is average daily net radiation at crop surface [MJ m 2 day 1 ]; G is daily soil heat flux, which is negligible in daily ET 0 calculations since it is very small compared to R n ; u 2 is mean daily wind speed at 2 m height [ms 1 pressure curve [kPa 1 ]; e s is saturated vapor pressure on a daily time scale [kPa]; e a is actual vapor pressure on a daily time scale [kPa]; e s e a is deficit of saturation vapour is psychrometric constant [kPa C 1 ]. ( 2 4 ) where T is mean daily air temperature at 2 m height [ ], calculated as average of minimum and maximum daily air temperatures. A tmospheric pressure (P) is expressed as ( 2 5 ) where P is atmospheric pressure [kPa]; z is elevation above sea level [m]. The psychometric constant (kPa 1 ) is given by ( 2 6 ) where c p is specific heat at constant pressure, 1.013*10 3 [MJ kg 1 1 ]; P is J kg 1 ]. The saturated vapor pressure e s [kPa ] calculated as the average of minimum and maximum daily vapor pressures

PAGE 29

29 ( 2 7 ) ( 2 8 ) e s [kPa]is then derived as ( 2 9 ) e a [kPa] is actual vapor pressure calculated as a function of e s and minimum and maximum daily relative humidity: ( 2 10 ) where RH max is maximum relative humidity [%]; RH min is minimum relative humidity [%]. 2.3.2.3 SFWMD S65DWX s tation d escription All the input weather parameters were collected from the SFWMD S65DWX monitoring station, which is located 8.5 km so uth west of the Larson wetland. Input weath er parameters ( net solar radiation, wind speed, air temperature, and relative humidity ) were recorded by daily step and collected for reference ET calculation. Reference ET is calculated with the standardized FAO56 PM method in daily time step for its sup erior using climate data form SFWMD S65DWX ( Gregory 2005 ; Itenfisu et al. 2003 ) Monthly averaged net sola r radiation (R n ) from 2004 to 2010 is shown in T able 2 5 and F igure 2 1 Monthly averaged reference evapotranspiration values from 2004 to 2010 are shown in T able 2 6 Since reference ET is calculated from net solar radiation, reference ET varies with R n consistently, which could be found in Fig ure 2 13 The highest monthly averaged reference ET at SFWMD S65DWX weather

PAGE 30

30 station occurs in April and May, the lowest monthly averaged value occurred in December. 2.3.2.4 Crop e vapotranspiration Crop evapotra nspiration is the actual amount of water required by a crop for transpiration (growing and development) plus the amount of water lost from the soil evaporation ( Beard 2008 ) commonl y denoted as ET c Crop evapotranspiration is formally defined as a crop evapotranspiration under standard conditions, which is disease free, well fertilized, grown in large fields, optimum water supply, and achieving full production under certain climatic condition. Crop Evapotranspiration is different from reference evapotranspiration (ET o into atmosphere in unit of time from a short uniform height green crop, completely shading the ground, and never short ( Penman 1956 ) Crop evapotranspiration can be measured by lysimetry direc tly ( Howell 1985 ) but it takes more time and cost to obtain accurate field ET c values by lysimetry. ET c could be calculated by relating crop coefficients (Kc) with reference ET 0 Thus, crop evapotranspiration (ET c ) of certain crop can be solved by: ( 2 11 ) By mean of crop coefficient (Kc) approach, the reference evapotranspiration rates (ET 0 ) can be corrected to more accurate crop evapotranspiration rates. Crop coefficient (Kc) The crop coefficient (Kc) serve s as a differential aggregat ion of physical and physiological between crops. It is varies in time based on characteristics of certain crop, horticultural practice and its growth stages ( Doorenbos 1975 ) Under the same climate situations, differences of crop evapotranspiration caused by differences of physical (leaf structures, stomatal characteristics) and

PAGE 31

31 physiological (growing stage) among crops species. For a given crop, Kc changes with field managements (sowing, harvest) ( Jagtap and Jones 1989 ) C rop coefficient is the ratio of actual evapotranspirat ion (ET c ) to reference evapotranspiration (ET 0 ). ( 2 12 ) Currently, the most common practice for estimating daily crop evapotranspiration (ET c ) in wetlands is daily reference ET multiplie d b y monthly averaged crop coefficients ( Mao et al. 2002 ) Larson wetland is operated as a ranch with forage Bahiagrass ( Paspalum notatum ) ( Min 2010 ) As there is no t suitable information available on crop coefficient s for wetland plants, the upland Kc in the study is based on monthly averaged Kc of Bahiagrass ( Paspalum notatum ) in central Florida reported by ( Jia 2009 ) and wetland Kc is referred to monthly average Kc of open water lysimeter monitored by ( Mao et al. 2002 ) in Fort Drum Marsh located 32km northeast of the Larson Ranch. The monthly average crop coefficients Kc of wetland and upland are shown in Table 2 3 Monthly averaged Kc values decrease in winter for grass dormant and reach peak in May. ( Min 2010 ) calculated daily crop ET in Larson ranch by daily reference ET multiplying monthly averaged crop coefficient reported by ( Mao et al. 2002 ) Daily crop ET during the monitoring period (Mar ch 2004 to Mar ch 2006) ranged from 0.23 to 6.59 mm/day in the study wetlands (Figure 2 14 ). The daily wetland crop ET of the entir e wetland derives from ET c per unit area and daily basis wetland surface area, which comes from the relationship between daily measured wetland water stage and wetland area ( Fig ure 2 3 ). The monthly averaged crop ET in Larson ranch is closely correlated w ith open water lysimeter based ET monitored by ( Mao et al. 2002 ) ( Fig ure 2 15 ) with R 2 =0.96.

PAGE 32

32 Table 2 1 Monitoring periods, wetland footprint areas, relative elevation of ditch and topographic range in study wetlands. Wetland ID Monitoring Period Footprint area (ha) Depth of Ditch (m) Topographic wetland range(m) LW 03/09/2004 to 03/10/2006 2.6 0.3 0.75 LE 03/09/2004 to 03/10/2006 2.2 0.3 0.80 Table 2 2 Information of 4 SFWMD weather stations. Site ID Latitude Longitude Dista nce to Larson Wetland(km) S 191 271131.168 804545.201 24.9161 0.0016 0.0607 S 154 271238.166 805506.214 15.3441 0.0042 0.16 S 65E 271331.164 805745.217 13.3608 0.0056 0.2111 S 65D 271852.153 810122.221 8.14278 0.0151 0.5682 Table 2 3 The relation between weather station S65D and other 3 sites. Site ID Distance to S 65D (km) S 191 29.0771 0.093 S 154 15.6058 0.323 S 65E 11.5936 0.584

PAGE 33

33 Table 2 4 The slopes, RMSE, and determination coef ficients (R 2 ) of the measured value and the estimated values with weighted average method and arithmetic mean, setting 4 sites as target site respectively. Site ID Weighted Average Arithmetic Mean Slope R 2 RMSE Slope R 2 RMSE S 65D 1.17 0.53 0.31 1.1 7 0.40 0.33 S 191 0.46 0.25 0.24 0.33 0.16 0.23 S 154 0.28 0.21 0.27 0.48 0.10 0.29 S 65E 1.20 0.81 0.27 1.28 0.54 0.32 Table 2 5 Monthly averaged net radiation (R n ) (Unit MJ m 2 day 1 ) collected from S65DWX weather station from 2004 to 2010. Year 2004 2005 2006 2007 2008 2009 2010 Mean JAN 5.2 7.8 8.6 7.8 6.9 6.0 3.5 6.1 FEB 6.9 9.5 10.4 9.5 9.5 6.9 5.2 7.8 MAR 9.5 11.2 12.1 13.0 11.2 7.8 7.8 10.1 APR 11.2 14.7 13.8 13.8 13.8 10.4 10.4 12.2 MAY 13.0 14.7 14.7 14.7 13.8 10.4 12.1 13.0 JUN 14.7 13.0 13.8 14.7 15.6 12.1 13.0 13.0 JUL 16.4 16.4 13.8 15.6 14.7 12.1 12.1 13.8 AUG 13.8 15.6 14.7 16.4 13.8 11.2 10.4 13.0 SEP 8.6 12.1 13.8 13.8 13.0 10.4 7.8 10.9 OCT 11.2 10.4 12.1 8.6 10.4 7.8 8.6 9.2 NOV 7.8 8.6 8.6 8.6 8.6 4.3 7.0 DEC 6.9 7.8 6.0 6.9 7.8 3.5 5.6

PAGE 34

34 Table 2 6 Monthly averaged reference ET 0 (Unit m) calculated with the FAO56 PM method using climate data form SFWMD S65DWX from 2004 to 2010. Year 2004 2005 2006 2007 2008 2009 2 010 Jan 2.7 3.1 3.8 3.3 3.3 3.1 2.8 Feb 3.0 3.7 4.0 3.8 4.1 3.7 2.9 Mar 3.9 4.4 4.9 5.3 4.5 4.1 3.8 Apr 4.8 5.6 5.8 5.8 5.4 4.9 4.7 May 5.5 5.5 6.0 6.1 6.0 5.0 5.1 Jun 5.8 4.5 5.8 5.7 5.9 5.0 5.4 Jul 6.0 5.9 5.2 5.8 5.4 4.9 4.9 Aug 4.9 6.0 5.6 6.1 5.3 4.8 4.5 Sep 4.3 4.9 5.3 5.2 5.0 4.3 3.9 Oct 4.3 4.2 5.1 4.1 4.2 4.0 5.0 Nov 3.6 3.6 3.7 3.8 3.7 3.0 Dec 2.8 3.2 3.1 3.0 3.2 2.4

PAGE 35

35 Figure 2 1 The location of Dixie Larson Ranch. Figure 2 2 The location of two study wetlands Larson East (LE) and Larson West (LW)

PAGE 36

36 ( A ( B ) Figure 2 3 Relationship between wetland water stage and flooded area in wetland L W ( A ) & LE ( B ) ( Min 2010 )

PAGE 37

37 Figure 2 4 Mini troll STP, In situ, Inc. logging pressure transducers. Figure 2 5 PVC screened monitoring well.

PAGE 38

38 Figure 2 6 Onset Communications Corp. Model RG3 M logging tipping buckets. Figure 2 7 Location of Larson Wetland and SFWMD weather stations.

PAGE 39

39 ( A ) ( B ) Figure 2 8 Setting site S65D as target site, the correlation of the measured value of target site and the estimated values with weighted aver age method ( A ) and the estimated values with arithmetic mean method ( B ).

PAGE 40

40 Figure 2 9 Correlation between in situ rainfall data and weighted average based estimation when the data are available from onsite and weather station s. Figure 2 10 The daily rainfall rates from Mar ch 2004 to Mar ch 2006.

PAGE 41

41 Figure 2 11 The daily rainfall estimation from Jan uary 2006 to Oct ober 2010. Figure 2 12 Mont hly net radiation trend line, derived from the monthly averaged net radiation from 2004 to 2010.

PAGE 42

42 Figure 2 13 Correlation of reference ET and net radiation (R n ). Figure 2 14 C rop ET calculated u sing the reference ET (calculated by FAO56 PM method) multipl ied by Kc during the monitoring period (Mar ch 2004 to Mar ch 2006).

PAGE 43

43 Figure 2 15 Correlation between monthly averaged wetland ET c in Larson wetland (point s ) and open water lysimeter based ET measured by Mao et al. (2002) (line).

PAGE 44

44 CHAPTER 3 WATER BUDGET MODEL The purpose of this chapter is to introduce water budget models applied to isolated wetlands. This chapter consists of three parts: (1) introduction of wetland w ater budgets; (2) methods of analyzing main components of wetland water budget; (3) and evaluation of simplified water budget models by comparing with measured data from the study isolated wetland s 3.1 Introduction Water budgets provide the framework to investigate hydrological systems It calculates all sources and sinks to the wetland system ( Drexler 1999 ) A w etland water budget model describe s the interactions between a wetland and its surrounding catchment ( Caldwell 2007 ) The main components of the study wetlands water budget are as follows: ( 3 1 ) where P is rainfall, ET is evapotranspiration, O is direct flow, D is ditch flow, GW is ground the water budget ( Fig ure 3 1 ) needs to be calculated or measured respectively, requiring significant input data and parameters. 3.2 Quantifying Main Components of Wetland Water Budget According to Eq uation 3 1, the components of the study wetland water budget were measured independently as follows 3.2.1 Precipitation and Evapotranspiration The methods of collecting and calculating precipitation (P) and Evapotranspiration (ET) were desc ribed in C hapter 2.

PAGE 45

45 3.2.2 Groundwater Exchange Among all the hydrologic components in a water budget, groundwater is the most difficult to quantify because of complicated heterogeneous underground condition s Darcy's law is the most widely used groundwater flow analysis method in wetland studies ( Choi and Harvey 2000 ) As a key component regulating the wetland groundwater dynamics, soil saturated hydraulic conductivity, K (m/d), need to be decided first ly. Soil saturated conductivity could be obtained from in situ field data or laboratory soil core analys e s ( Hopmans 2005 ) The natural drawdown method, which is analogous to active pumping test used by Wise et al. ( Wise 2000 ) was used in this study to calculate K. The natural drawdown method estimates K through choosing several wetland sur face water stage natural drawdown events which are in conditions of no rainfall, no runoff, and no ditch flow. Under these conditions, wetland water stage changes only affect by ET and groundwater exchange, namely, ( Min 2010 ) Combine with radial Dupuit equation, groundwater exchange (GW) can be arranged as ( 3 2 ) ( 3 3 ) ( 3 4 ) where [L/T] is mean soil hydraulic conductivity, r wet [L] is wetl and radius, L [L] is the distance between wetland and upland monitoring well ( Fig ure 3 2 ), A wet is wetland surface water area.

PAGE 46

46 Through Eq uation 3 2 and Eq uation 3 3 with hydraulic gradient between wetland and upland during drawdown events, we can estimat e mean Then substitute and hydraulic gradient into Eq uation 3 4 to calculate GW exchange. 3.2.3 Runoff Larson wetlands have humid and well vegetated environment s with a shallow groundwater table ( Reiss 2006 ) Under these conditions, surface runoff consis t s of overland flow and exfiltrated water from saturated soil ( Manfreda 2008 ; Mishra and Singh 2004 ) The r unoff curve numbe r method (CN method), which was developed by the USDA Natural Resources Conservation Service ( Bosznay 1989 ) was used to determined amount of overland flow in study wetlands. This method is widely used for analyzing directly runoff due to its simplicity ( Mishra and Singh 2004 ) When precipitation is greater than the threshold of runoff generation (0.2 S) (Eq uation 3 5 ), direct runoff could be calculated with the CN method ( Eq uation 3 6 ). When precipitation is less than 0.2 S, exfiltrated runoff (O I ) could be c alculated with Eq uation 3 7 ( 3 5 ) (P(t) ( 3 6 ) (P(t) 0.2 S) ( 3 7 ) where Q(t) is runoff [m], P(t) is precipitation [m], S is watershed storage [m], CN is determined based on the local hydrologic soil group, land use and hydrologic condition ( USDA 1990 ) According to environmental condition of study wetlands, the CN value is 69. is area of upland with water elevation of

PAGE 47

47 is wetland area, is average of elevation of ground water at the wetland upland border. Flooded area is based on the stage area relationship (Fig ure 2 3 ). Volume based direct runoff (O D ) (m 3 /d) equals to Q(t) multiply runoff generating area. In order to calculate the O D (m 3 /d), the effective runoff gene rating a ) need to be calculated. Based on upland water budget where I(t) is daily infiltration [m], net P(t) is net precipitation (P ET up )[m]. The volume of water infiltrated in the runoff generating area is expe cted to be equivalent to soil water storage capacity, SWSC [L 3 ]. SWSC is determined by upland soil properties. Through studying a nonlinear regression ( Nachabe 2004 ) of SWSC and soil cores collected from LW, the volumetric SWSC [L 3 ] can be expressed as: ( 3 8 ) ( 3 9 ) Where x AVG and x UP are the lower and upper boundaries of the effective runoff generating areas, x AVG is the radius of wetland with H avg Thus, the upper boundary of the upland runoff generating area can be determined from Eq uation 3 9 during P > 0.2S. 3.2.4 Ditch flow Ditch f low only occurs during a small period in a year and ditch velocity was not Eq uation 3 10 and Eq uation 3 1 1 ) was used to estimate ditch flow. Water surface slope (S) was derived from daily d ifference s between H wet and Taylor Creek downstream elevation (gage station USGS 02272676). The hydraulic resistance in ditches n = 0.07 s/m 1/3 ( Bakry

PAGE 48

48 1992 ) As the cross sectional velocity is less than 0.1 m/s, seasonal variations of vegetative no have signif icant differences, so n was applied to all season in study wetlands. ( 3 10 ) ( 3 11 ) w here V is cross sectional velocity [m/s], K is conv ersion factor, n is Manning coefficient, including surface roughness and sinuosity, R n is the hydraulic radius [m], S is the water surface slope [h/L], A is the cross section area of flow [m 2 ] (Fig ure 3 3 ), P is channel perimeter [m] (Figure 3 3 ). 3.2. 5 Result The primary aim of this part is to simulate the isolated wetland water stage H wet The simulated LW and LE wetland water stage (Fig ure 3 4 ) show good agreement with the measured stage with R 2 = 0.72 and 0.54 respectively ( Min et al. in process ) The RMSE value of model calibrations was 0.08 m. 3.3 Simplified Water Budget Model C omplex models, like the wetland water budget model described above, demand a large number of pa rameters about soil hydraulic properties ( Bonell 1998 ) For foregoing reason, this study develop ed more parsimonious wetland water budget models. The Larson wetla nds offer a good test case for simpler modeling approaches. Compar ed with less impacted areas, the hydrologic respons e s of the managed pasture areas are more predictable, allowing us to use simpler, minimum calibration

PAGE 49

49 models. In the study wetlands, extens ive ditch networks increase the drainage efficien cy and the soils variations are even. 3.3.1 Simplified Climate Water Budget Model A simplified climate water balance model was developed to characterize the role of precipitation and evapotranspiration on the study isolated wetland without concerning land runoff, groundwater exchange and ditch flow. This method has been used to study shallow depressional wetlands on the Atlantic Coastal Plain ( Lide et al. 1995 ) Pitman has developed an example of the models for g enerating river flows for the South African catchments ( Pitman 1973 ) The climatic water balance is expressed as: ( 3 12 ) ( 3 13 ) w wet is change of wetland water stage [m], P is daily precipitation [m], ET is daily evapotranspiration [m], H(t+1) wet is wetland water stage at moment of t+1[m], H(t) wet is wetland water stage at moment of t [m]. Based on Eq uation 3 12 and Eq uation 3 1 3 the correlations of daily based estimated wetland water stage (H est ) and corresponding measured wetland surface water stage (H wet ) are shown in Fig ure 3 5 Nevertheless, the determination coefficient R 2 of simplified climate water budget model are approximately 0, which means the residual parts of wetland water budget (GW, O, D) are much greater than or less than zero and cannot be ignored. Obviously, the simplified climate water budget model is over simple to simulate and predict these isolated wetlands in LOB. Actually, groundwater exchange is an important component of wetland water budget in the isolated wetlands of the LOB. Acc ording to ( Min 2010 ) mean groundwater exchange was 22m 3 /d and average annual net ground water recharge is approximately

PAGE 50

50 3650m 3 / yr in study wetlands. ( P. 2007 ) also reported ditch flow contributes approximately 50% of the total flow. Therefore, the simplified climate water budget model cannot be used in these isolated wetlands on LOB. 3. 3 .2 Combined Simplified Water Budget Model In t his section, the residual of wetland water balance after accounting for P and ET are combined into linear functions of precipitation and net precipitation respectively: ( 3 14 ) ( 3 15 ) ( 3 16 ) ( 3 17 ) ( 3 18 ) where F(..) is a linear f unction used to describe the residuals of the wetland water balance, except of P and ET [m 3 ]; O is land runoff [m 3 ]; D is ditch flow [m 3 ]; GW is groundwater exchange [m 3 wet is volume change in wetland [m 3 ]; H (t) is wetland water stage at moment of t [m]; S t is wetland flood area at moment of t, derived from the stage area relationship (Fig ure 2 3 ) [m 2 ]. Through wetland water budget equation ( Eq uation 3 1) and E quati on 3 15 a linear daily precipitation function used to describe the residuals of wetland water balance could be regressed from Fig ure 3 5 : ( 3 19 ) with R = 0.0059. But the negative slope which mean s more rainfall will decrease the water stage does not seem reasonable. Substitut ing daily P into Eq uation 3 19 and then calculat ing wet with Eq uation 3 16 Fig ure 3 7 shows the correlation of

PAGE 51

51 wet derived from Eq uation 3 19 and Eq uation 3 16 wet based on measured wetland water stage, with R 2 approximate ly 0. The result demonstrates that linear daily precipitation f unction Eq uation 3 19 cannot effectively describe the residual components of theses wetland water budget s Based on the method described above, linear relations between the residuals of wetland water balance and recent 5 day precipitation (Fig ure 3 8 a ), recent 5 day net wetland precipitation (Fig ure 3 8 b ), and recent 5 day net upland precipitation (Fig ure 3 8 c ) were regressed. Same methods have also been used in recent 3 day, 7 day, 15 day and 30 day. Only the 5 day based linear function show ed rela tively higher wet with linear functions above and Eq uation 3 16 Eq uation 3 17 and Eq uation 3 18 est show ed wet based on measured wetland water stage, with R 2 approximate ly 0. Therefore, the combined simplified water budget model is not a good way to simulate the hydrology of isolated wetlands i n the LOB. 3.4 Conclusion With limited data, complex and oversimplification of wetland water budget model s are not usable for predicting study isolated wetlands hydrology in the LOB. Although in section 3.2 estimation for wetland water stage with water budget model shows a good agreement with measured data, it is difficult to accurately measure all components in t he wetland water balance and large amounts of in situ data are required There are several problems with current water budget methodologies to predict wetlands hydrology ( Koob 1999 ) Drexler concluded that there were a large of errors in all components of the wa ter budget ( Drexler 1999 ) The w ater b udget as a physical method has inherent problems, relying on measurements from in situ piezometers that may

PAGE 52

52 malfunction. Greater amount of information about soils, hydraulic data and topography must be known for accurate water budget prediction ( Basu and Merwade 2010 ) And it is a challenge to collect the accurate hydrologic data for various hydrological processes. Detailed field studies are expensive and time consuming, and may not be possible in some remote locations.

PAGE 53

53 Figure 3 1 Conceptual diagram of water budget in historically isolated wetlands in the Lake Okeechobee basin. Figure 3 2 Conceptual diagram showing groundwater flow between wetland and upland modeled in t his study ( Charbeneau 2000 )

PAGE 54

54 Figure 3 3 Illustration of di tch geometry. Figure 3 4 The wetland water budget based simulated LW(A) and LE (B) wetland water stage with the measured water stage ( Min et al. in process )

PAGE 55

55 ( A ) ( B ) Figure 3 5 The LW ( A ) and LE ( B ) daily wetland water level (H est ) estimation with simplified climate water budget model compare to measured corresponding wetland surface water level (H wet ).

PAGE 56

56 Figure 3 6 Plot of F (..) versus daily precipitation. Days with zero precipitation have been deleted. A linear daily precipitation function was regressed to describe the residuals of wetland water balance. Figure 3 7 wet derived from combined simplified water budget model Equation 3 19 & E quation 3 16 wet based on measured wetland water stage.

PAGE 57

57 Figure 3 8 Plot of F (..) versus recent 5 day precipitation ( A ), recent 5 day net wetland precipitation( B ), and recent 5 day net upland precipitation ( C ). A linear function was regressed to describe the residuals of wetland water balance. ( A ) ( B )

PAGE 58

58 ( C )

PAGE 59

59 CHAPTER 4 RECESSION ANALYSIS MODEL The principal aim of t he chapter is to simulate the Larson wetlands water stage with a recession analysis model, which is an empirical hydrologic model. This section consists of three parts: (1) introduction of the methodology of flow recession analysis; (2) application of flow recession analysis method used for groundwater flow in Larson wetlands; and (3) simulating wetland surface water stage with recession analysis model s Recession analysis method s are advantag eous for their easy mathematical manipulation. 4.1 Flow Recession Analysis Introduction Recession analysis is widely used in quantifying groundwater flow and surface water drainage ( Tallaksen 1995 ) A representation of the recession curve function of time is given in E quation 4 1. ( 4 1 ) where Q t is discharge at the moment of t [m 3 0 is the volume of discharge at beginning of recession [m 3 ]. 4.2 Flow Recession Analysis in Groundwater When surface water levels increase with precip itation, the groundwater reacts to this recharge causing an increase of the groundwater discharge and reaching a peak some time after precipitation cessation Then the groundwater recession begins due to depletion of groundwater balance, which is determine d by soil properties, climate, and vegetation ( Nichols 1994 ) This process is quite similar to surface water flow routing. There are two methods to quantify g roundwater exchange: one is the water balance approach; another is monitoring ground water movement in the vadose zone

PAGE 60

60 with the help of tracers, lysimeter and tensiometers ( Wood 1995 ) However, the above two methods are restricted for their high costs of monitoring equipment and time consu ming. For large watershed s with sub humid or humid climates, groundwater recession analysis is an alternative method ( Hoos 1990 ) This method can simulate management scenarios on groundwater exchange, groundwater stage and total water yields, which are widely used in evaluatin g proposed water resources dev elopment plan ( Boughton and Freebairn 1985 ; Kachroo 1992 ; Knisel 1963 ; Korkmaz 1990 ) In some large watershed s where variations of hydrogeology are complex, recession curve method s are used to estimate groundwater exchange di rectly ( Wittenberg and Sivapalan 1999 ) The r ecession analysis method estimating groundwater outflow is based on surface water records. Based upon Eq uation 4 0 need to be first estimated. In this study, six natural drawdown events were chosen in LW and LE respectively (Fig ure 4 1 ) to investigate volume of groundwater exchange with the water budget approach. These natural drawdown even ts met the following conditions: (1) wetland water stages are between the elevation of ditch on and wetland bottom, such that there was no ditch flow; (2) there was no rain, and thus no overland runoff; and (3) water stages in the study wetland are measure d. Therefore, only ET and groundwater exchange occur during these events. The groundwater outflow could be derived from the constrained water budget equation ( 4 2 ) w wet (m 3 ) is the daily volume change in wetland, GW (m 3 ) is daily groundwater outflow, and ET (m 3 ) is wetland evapotranspiration.

PAGE 61

61 The exponential recession trend could be found in plots of groundwater outflow versus recession date (Fig ure 4 2 ). The assessed by logarithmic transformation of E quation 4 3 as follows ( 4 3 ) w t is groundwater outflow at the moment t, Q 0 is 0 and Q t Result and d iscussion Groundwater discharge recession analysis provides information on the watershed retention characteristic s and on groundwater storage and depletion. T he Q 0 represents the soil water storage capacity and the recession coefficient characterizes the drainage velocity. The characteristic s of the aquifer can be obtained through analyzing all recorded recession limbs. The parameters the two Larson wetl ands are averaged form the groups of recession curves (Table 4 1 and Table 4 2 ). Therefore groundwater discharge can be expressed as at LW and at LE. The GW values derived from Eq uation 4 2 are considerable: GW was observed to reach up to 6279 m 3 / d at LE and 3111 m 3 / d at LW. The volume of groundwater discharge demonstrates that groundwater exchange is an important component of wetland water and chemical budgets. The event in LE, with STDEV 0.037 and 0.023 respectively (Table 4 1 and Table 4 2 ). The velocity of LW is faster than that of LE (Fig ure 4 3 ). The GW discharge volumes at the beginning of recession (Q 0 ) were considerably different between different wetlands in same period and between different periods in

PAGE 62

62 same wetland. Even though the two wetlands have quite similar ecology, hydrogeology and topography, the complexity of groundwater mechanisms translates to the extremely varia ble groundwater discharge value s The wetland ground water exchange mechanisms are very complex, which link with nearby wetlands surfac e water, shallow unconfined aquifers, local soil characters, topography, vegetation, land use as well as kinds of weather factors. The differences of Q 0 may be attribute d to seasonal variation of the storage capacity of shallow aquifers, which vary as sea sonal varying evapotranspiration and precipitation ( Neumann 2012 ) During the dry season, water stored in soil is removed by evapotranspiration, which is much greater than precipitation W hen effective precipitation come s it take s more time for gr oundwater discharge to reach its peak compar ed to the wet season. Therefore the beginning time of recession also vary with season. The seasonal difference in recession dynamics shows as faster recession rates in summer comparing to autumn or winter ( Neumann 2012 ) But these phenomena are not very significant in Larson wetland, as in summer high evapotranspiration always happens with heavy and frequent precipitation. 4.3 Recession Analysis Method in Wetland Water Stage Analysis Wetlands have some wa ter storage capacity, such that small rainfall events cannot raise water level easily and immediately (Fig ure 4 1 ). Thus, there are relatively stable exponential recession periods after heavy rains. Small rainfall events could not affect the overall expon ential recession trend. Stream channel recession curve s are generally described as ( Eq uation 4 1), which could also be applied in influent and effluent lake ( Arnold 1999 ) The water stage recessio n equation

PAGE 63

63 ( 4 4 ) is derived from volumetric recession equation ( Eq uation 4 1 ) where H t [m] is 0 [m] is the surface water stage at the beginning of recession [d]. Qian used the water stage recession analysis on water level fluctuation of Poyang Lake, China ( Qian 1995 ) At present, simulat ion and predictive model ing of wetland water stage with recess ion analysis method s has not been reported in the literature. 4.3.1 Monthly Wetland Surface Water Stage Analysis In this section the recession analysis was used to estimate monthly wetland surface water stage. To simulate monthly wetland surface water stag e, H 0 parameters need to be first investigated. H 0 is maximum of among consecutive months. As a result of rainfall and evapotranspiration could be simplified as a linear function ( 4 5 ) where is the estimated maximum monthly averaged wetland water stage with current month net precipitation, P ET is current month net precipitation, and a and b are tland surface water stage is assessed by logarithmic transformation E quation 4 6 ( 4 6 ) t is wetland water level at the moment of t, H 0 is the wetland wate 0 and H t The wetland water stage result s from superposition of several recession curves, which could be expressed as a composite surface water stage recession equation

PAGE 64

64 ( 4 7 ) events. Result s and d iscussion The coefficients of linear functions of are shown in Table 4 3 Recession rates as parameter s of the wetland hydrologic properties, w ere determine d through respectively averaging the recession rates of 5 wetland surface water recession events (Fig ure 4 4 ) at LW and LE. Based upon Eq uation 4 6, averaged recession rates are 1.80 and 1.67 at LW and LE r espectively (Table 4 4 ) The is big (Table 4 4 ), because the surface water stage of the Larson wetlands is easily affected by precipitation and evapotranspiration for their relatively small sizes compar ed to large l wetlands have quite similar drainage velocity. Fig ure 4 5 shows the correl ation between estimated wetland surface water levels (H est ) and measured corresponding wetland surface water levels (H wet ) with R 2 =0.5132 (LW) and 0.588 (LE). However, the monthly recession analysis for wetland water stage s may not be precise enough for making management decision s And the method is not suitable for predicting the groundwater level when the wetland stage descends below the ground surface Therefore, a daily based recession analysis model was developed as follow s

PAGE 65

65 4.3.2 Daily Wetland Wat er Stage Analysis It is difficult to simulate daily water stage with the proposed method in section 4.3.1, as there is no significant linear or nonlinear relation between the precipitation and H 0 In order to better understand the daily based wetland water stage, another recession analysis model was developed ( 4 8 ) ( 4 9 ) ( 4 10 ) w 0 is the wetland water level at the beginning of recession, P is daily precipitation [m], S y is specific yield. ted with the same method of section 4.3.1. Water level recession analysis is available only when there is surface water and no precipitation. change of exponential recession c urve, which could be described as the exponent derivate precipitation. If there is surface water in the wetland, the recession period stop s when the next rain comes and h 0 would cha nge to the intraday wetland water level. When wetland water stages drop below the ground y which S y also change s with groundwater level ( Sophocleous 1991 ) ( i.e., water stag e Hwet <0) ( 4 11 )

PAGE 66

66 The S y ( Johnson 1967 ) and ( Min et al. in process ) 4.3.2.1 Result and d iscussion Based on the daily recession analysis method, correlation of estimated wetland water stage and measured daily wetland water stage form Mar ch 20 04 to Mar ch 2006 is shown in Fig ure 4 6 with R 2 =0.6116 and 0.7293 in LW and LE respectively. The averaged recession rates are 0.067 with RMSE 0.019 in LW and 0.054 with RMSE 0.028. The model shows a good agreement during the first 330 day s in LW with hig h coefficient of determination, R 2 = 0.9563, and the first 300 day s in LE with R 2 =0.9557. The prediction for hydroperiod is also good with R 2 =0.91 in LW and R 2 =0.94 in LE. 4.3.2.2 Sensitivity a nalysis sensitivity analysis was performed in the two study wetlands. Daily average precipitation data were changed by 10 %, considering the uncertainties in measuring daily precipitation ( Drexler 1999 ) The P data change of 10 % was based on ( Bakry 1992 ; Kroger et al. 2008 ) Sensitivity coefficient was expressed as monitoring period (Mar ch 2004 to Mar ch 2006) was defined as Precipitation is a sensitive hydrologic factor for daily recession analysis model with sensitivity coefficients equal 67 to 12. Errors in precipitation measurements could lead to relatively large error. The impacts of small changes of rainfall pattern related to climate change would lead to a great wetland hydrologic change. Therefore, the poor correlation of the model after 300 d ay s may be attribute d to i naccurate precipitation measurements. At the beginning of the monitoring period, the precipitation

PAGE 67

67 measurements were relatively reliable and the in situ rainfall gages were in good condition. W ith time, the instruments performance may have drifted leading to poorer correlation with actual rainfall The remaining portions of precipitation data from nearby weather stations are not very accurate considering the frequent localized p recipitation in Florida. As an accumulative model ( Eq uation 4 10 ), error s in the recession analysis model are magnified with time. For maximum model accuracy, measured wetland water stage as correction data need to be inserted periodically. Based on the good agreement during the first 330 days and 300 days i n LW and LE wetland, the recommended interval is 300 days. 4.3.2.3 Wetland w ater s tage p rediction A 5 year (Jan uary 1 2006 to Oct ober 6. 2010) Larson wetland water stage prediction was made based upon the daily recession analysis model (Fig ure 4 7 ). Th e trend of water stage reflects the rainfall pattern in the area. During summer, concentrated precipitation do es not bring necessarily translate to standing surface water in the wetlands due to the increased consumptive use by evapotranspiration. 4.4 Concl usion The r ecession analysis approach appears to be a fast, simple way for estimating wetland water stage, with advantage s of fewer input data and parameter. Using o nly parameters of daily precipitation, ET, soil special yield and recession rate, wetland water stage and hydroperiod could be simulated and predicted. Regression analysis models provide an easy way to predict ungaged wetland hydrograph s

PAGE 68

68 Table 4 1 The recession rate Event NO. 1 2 3 4 5 6 Period 3/10/04 to 3/15/04 3/18/04 to 4/2/04 9/27/04 to 12/8/04 12/31/04 to 1/13/05 1/16/05 to 2/16/05 11/22/05 to 1/17/06 STDEV 0.037 0.29 0.26 0.27 0.30 0.19 0.26 Av erage 0.265 Table 4 2 The estimated recession rate Event NO. 1 2 3 4 5 6 Period 3/17/04 to 4/4/04 9/27/04 to 1/12/05 12/27/04 to 1/12/05 1/20/05 to 2/16/05 12/22/05 to 1/17/06 2/6/06 to 2/23/06 STDEV 0.023 0.15 0.137 0.11 0.10 0.16 0.13 Average 0.131

PAGE 69

69 Table 4 3 The coefficients for linear function of LW LE a 20.188 16.893 b 0.1758 0.1552 R 2 0.3725 0.3474 Table 4 4 The recession rate of Larson wetland surface water. Recession NO. 1 2 3 4 5 Average STDEV 1.35 2.50 2.84 1.07 1.27 1.80 0.72 0.96 2.597 1.90 1.21 1.74 1.67 0.57

PAGE 70

70 ( A ) ( B ) Figure 4 1 The 6 nat ural drawdown events are shown in blue bar. Fluctuation of Wetland stage and rainfall at LW ( A ) & LE ( B ). The zero on the left y axis represents the lowest bathymetric wetland elevation.

PAGE 71

71 ( A ) ( B ) Figure 4 2 During natural drawdown period, the water stage of wetland surface water, groundwater outflow, and the nonlinear exponential recession trend of groundwater discharge in LW ( A ) & LE ( B ).

PAGE 72

72 Figure 4 3 The recession curve modeling of LW ( ) and LE ( ).

PAGE 73

73 ( A ) (B) Figure 4 4 Fluctuation of monthly averaged wetland stage at LW ( A ) & LE ( B ). The 5 wetland surface water recession curves are shown in red bar. The zero on the left y axis represe nts the lowest bathymetric wetland elevation.

PAGE 74

74 ( A ) ( B ) Figure 4 5 Estimated wetland surface water level (H est ) with monthly recession analysis method compared to measured corresponding wetland surface water level (H wet ) at LW ( A ) and LE ( B ).

PAGE 75

75 ( A ) ( B ) Figure 4 6 Estimated wetland stage (H est ) with daily recession analysis method compared to measured corresponding wetland surface water level (H wet ) at LW ( A ) and LE ( B ).

PAGE 76

76 Figure 4 7 Plot of Larson wetland water stage prediction from Jan uary 1 2006 to Oct ober 6 2010. Red line is daily rainfall rate.

PAGE 77

77 CHAPTER 5 RECOMMENDATIONS FOR FUTURE WORK 5.1 Model Improvement The daily recession analysis model is not good at simulat ing groundwater stage. The groundwater part of the model needs to be improved, for the error of this model is mainly from groundwater estimation. Specific yield, which affects the negative wetland stage should be calibrated by minimizing r oot mean square error (RMSE) between measured and modeled wetland stage in future work. Precipitation measurement and statistical methods should also be improved. This recession analysis model is vulnerable to errors in precipitation data that could lead t o relatively large error s In this study distance weight averaged methods are used to estimate the precipitation. In future work, area weight averaged methods and Thies s en polygon method s would be used to increase the accuracy of precipitation estimation. In addition, precipitation in Florida is characterized by its frequent localized convectional rain in summer, while in winter the spatial variations are reduce d Therefore, an optimum estimation method adopt ing the nearest weather station data in summer, while weighted averaged surrounding several weather stations data in winter may increase the simulative accuracy of the recession analysis model. 5.2 Model Generalization The recession analysis model for simulating and predicting wetland water stage w as only tested in two small wetland s More sites should be tested with this simplified recession analysis model. The test only needs local precipitation, ET and hydrograph of monitoring period. If the generalization of the model is proved, the recession an alysis

PAGE 78

78 model would provide an easy mathematical manipulation for hydrologic simulation and prediction. 5.3 Coupled Model A coupled model which combines the recession analysis model with wetland plant biomass model could be developed for isolated wetlands in LOB in future. This coupled model may reveal implications related to phosphorus management in the Lake Okeechobee Basin. The objective of coupled model is to develop a predictive model of the dynamics of wetland biomass and accretion of soil, C and P. The model should incorporate grazing intensity, wetland water stage, and climate factors. Although vegetation nutrient storage is short term, biomass turnover supports accretion of soil and associates with retention of C and P. Sequestration of C and P an d reduction of P load to the lake are directly related to the wetland area. Hydrologic restoration of isolated wetlands in LOB to a criticial depth not only benefit s nutrient manageent but may also support ranching as well, as remaining water in the wetlan d can serves as a cooling pond for cattle. The predictive coupled hydrologic and wetland vegetation biomass model would provide information for optimizing P management and grazing intensity.

PAGE 79

79 LIST OF REFERENCES Abtew, W., 1996. Evap otranspiration measurements and modeling for three wetland systems in south Florida. Water Resources Bulletin 32(3):465 473. Allen, L. H., 1987. Dairy siting criteria and other options for wastewater managment on high water table soil. Soil Crop Sci Soc Fl a Proc 47:108 127. Allen, R. G., L.S. Pereira, D. Raes, and M. Smith. 1998. Crop evapotranspiration: guidelines for computing crop water requirements. FAO Irrig Drain Paper No 56 Rome, Italy. Allen, R. G., M. Smith, A. Pereira, and L.S. Pereira. 1994. An update for the definition of reference evapotranspiration. ICID Bull 43(2) 1 34. Allen, R. G., Walter, I.A., Elliot, R.L., Howell, T.A., Itenfisu, D., Jensen, M.E. and Snyder, R. 2005. The ASCE standardized reference evapotranspiration equation. ASC E and American Society of Civil Engineers. Arnold, J. G., Allen, P. M., 1999. Automated methods for estimating baseflow and ground water recharge from streamflow records. Journal of the American Water Resources Association 35(2):411 424 doi:10.1111/j.1752 1688.1999.tb03599.x. Bakry, M. F., Gates, T.K., Khattab, A.F., 1992. Field measured hydraulic resistance characteristics in vegetation infested canals. Journal of Irrigation and Drainage Engineering 118 (2): 256 274. Bastian, R. K. R., S.C. 1979. Aquacu lture System for Wastewater Treatment: :Seminar Proceedings and Engineering Assessment. (MCD 67, Environmental Protection Agency, Washington, D.C.). Basu, N. B. R., P. S. C.; Winzeler, H. E.; Kumar, S.; Owens, P. & V. Merwade, 2010. Parsimonious modeling o f hydrologic responses in engineered watersheds: Structural heterogeneity versus functional homogeneity. Water Resources Research 46 doi:10.1029/2009wr007803. Beard, J. B., Kenna, M.P. 2008. Water quality and quantity issues for turfgrasses in urban land scapes. Council for Agricultural Science and Technology Ames, Iowa, U.S. 298 p. Bhadha, J. H. & J. W. Jawitz, 2010. Characterizing deep soils from an impacted subtropical isolated wetland: implications for phosphorus storage. Journal of Soils and Sediments 10(3):514 525 doi:10.1007/s11368 009 0151 4. Blatie, D., 1980. Land into Water. Water into Land Florida State University Press, Tallahassee, FL.

PAGE 80

80 Bonell, M., 1998. Selected challenges in runoff generation research in forests from the hillslope to headwater drainage basin scale. Journal of the American Water Resources Association 34(4):765 785 doi:DOI 10.1111/j.1752 1688.1998.tb01514.x. Bosznay, M., 1989. Generalization of SCS curve number method. Journal of Irrigation and Drainage Engineering Asce 115(1):13 9 144. Bottcher, A. B. T., K.; Campbell, K. L., 1995. Best management practices for water quality improvement in the Lake Okeechobee watershed. Ecological Engineering 5(2 3):341 356. Boughton, W. C. & D. M. Freebairn, 1985. Hydrograph recession characteris tics of some small agricultural catchments. Aust J Soil Res 23: 373 382. Boussinesq, J., 1877. Essai sur la the orie des eaux courantes. Me moires pre sente XXIII, n o. 1: 1 680. (Cited by Hall 1968). Breuer, L., Huisman, J. A., 2009. Assessing the impact of land use change on hydrology by ensemble modeling (LUCHEM). Advances in Water Resources 32(2):127 128 doi:10.1016/j.advwatres.2008.10.010. Brunsdon, C., McClatchey J., Unwin, D. J., 2001. Spatial variations in the average rainfall altitude relationship in Great Britain: An approach using geographically weighted regression. International Journal of Climatology 21(4):455 466 doi:10.1002/joc.614. Brutsaert, W., J.P. L opez., 1998. Basin scale geohydrologic drought flow features of riparian aquifers in the southern Great Plains. Water Resources Research 34(No. 2: 233 240). Caldwell, P. V., Vepraskas, Michael J., Skaggs, R. Wayne, Gregory, James D., 2007. Simulating the w ater budgets of natural Carolina bay wetlands. Wetlands 27(4):1112 1123 doi:10.1672/0277 5212(2007)27[1112:stwbon]2.0.co;2. Carter, V., 1996. Technical Aspects of Wetlands: Wetland Hydrology, Water Quality, and Associated Functions. National Water Summary (United States Geological Survey Water supply Paper). Charbeneau, R. J., 2000. Groundwater hydraulics and pollutant transport. Prentice Hall, Upper Saddle River. Choi, J. & J. W. Harvey, 2000. Quantifying time varying ground water discharge and recharge in wetlands of the northern Florida Everglades. Wetlands 20(3):500 511 doi:10.1672/0277 5212(2000)020<0500:qtgdar>2.0.co;2. Cuenca, R. H., 1989. Irrigation system design. An engineering approach. Prentice Hall, Englewood Cliffs, NJ.

PAGE 81

81 Davis, F. E., 1982. Straz zulla grove study. In: L.B. Baldwin and A.B. Bottcher (Eds.) Nonpoint Pollution Control Technology in Florida Proceeding of the IFAS Conference, Gainesville, Florida:190 200. Demuth, S. a. H., I., 1993. Case study of regionalising base flow in SW Germany a pplying a hydrogeological index Flow Regimes from International Experimental and Network Data (FRIEND) Vol. I: Hydrological Studies(Institute of Hydrology, Wallingford, UK ). Doorenbos, J., and W. O. Pruitt. 1975 Guidelines for Prediction of Crop Water Requirements. FAO Irrig Drain(Paper No. 24. Rome, Italy.). Doorenbos, J., W. O. Pruitt, 1977. Guidelines for Prediction of Crop Water Requirements. PruittFAO Irrig Drain.(No. 24 (revised)):Rome, Italy. Drexler, J. Z. B., B.L.; DeGaetano, A.T.; Siegel D.I. 1999. Quantification of the water budget and nutrient loading in a small peatland. Journal of the American Water Resources Association 35 (4): 753 769. Dunne, E. J., Smith, J, Perkins, D.B., Clark, M.W., Jawitz, J.W., Reddy, K.R., 2007. Phosphor us storages in historically isolated wetland ecosystems and surrounding pasture uplands. Ecological Engineering 31:16 28. Eagleson, P. S., 1967. Optimum density of rainfall networks. Water Resources Research 3(4):1021 & doi:10.1029/WR003i004p01021. EPA, U. S., 2008. Methods for Evaluating Wetland Condition: Wetland Hydrology. Office of Water, US Environmental Protection Agency, Washington, DC. (EPA 822 R 08 024). Evett, S. R., Schwartz, R. C., Howell, T. A., Baumhardt, R. L., Copeland, K. S., 2012. Can wei ghing lysimeter ET represent surrounding field ET well enough to test flux station measurements of daily and sub daily ET? Advances in Water Resources 50:79 90 doi:10.1016/j.advwatres.2012.07.023. Flaig, E. G., Havens, Karl E., 1995a. Historical trends in the Lake Okeechobee ecosystem: I. Land use and nutrient loading. Archiv fuer Hydrobiologie Supplementband 107(1):1 24. Flaig, E. G., Reddy, K. R., 1995b. Fate of phosphorus in the Lake Okeechobee watershed, Florida, USA: Overview and recommendations. Ecolo gical Engineering 5(2 3):127 142 doi:10.1016/0925 8574(95)00021 6. Franchini, M., Pacciani, M., 1991. Comparative analysis of several conceptual rainfall runoff models J Hydrol, 122: 161 219. Gatewood, S. E., P.B. Bedient, 1975. Drainage density in the L ake Okeechobee drainage area. Report Division of State Planning Florida Department of Administration, Tallahassee, Florida.

PAGE 82

82 Gleick, P. H., 1986. Methods for Evaluation the Regional Hydrologic Impacts of Global Climatic Changes. Journal of Hydrology 88(1 2) :97 116 doi:10.1016/0022 1694(86)90199 x. Glover, R. E., 1964. Ground water movement: U.S. Bureau of Reclamation Engineering Monograph Series No. 31:31 34. Goldstein, A. L., Berman, W., 1995. Phosphorus management on confinement dairies in southern Florida Ecological Engineering 5(2 3):357 370 doi:10.1016/0925 8574(95)00032 1. Gregory, J. H., 2005. Stormwater Infiltration at the Scale of an Individual Residential Lot in North Central Florida Maste r's Thesis, Univ ersity of Florida, Gain es ville, Fl. Haan C. T., 1995. Fate and transport of phosphorus in the Lake Okeechobee Basin, Florida. Ecological Engineering 5(2 3):331 339 doi:10.1016/0925 8574(95)00030 5. Hammer, D. A., 1989. Constructed Wetlands for Wastewater Treatment. Lewis Publishers Chelsea, MI, USA. Harlin, J., 1991. D evelopment of a process orienred calibration scheme for theHBV hydrological model Nordic Hydrology 22(1):15 36. Harvey, R. H., K. 1999. Lake Okeechobee Action Plan. South Florida Water Management District West Palm Beach, FL:PP .1 43. Havens, K. E., Gawlik, D. E., 2005. Lake Okeechobee conceptual ecological model. Wetlands 25(4):908 925 doi:10.1672/0277 5212(2005)025[0908:locem]2.0.co;2. Hayashi, M., van der Kamp, G., Rudolph, D. L., 1998. Water and solute transfer between a prai rie wetland and adjacent uplands, 1. Water balance. Journal of Hydrology 207(1 2):42 55 doi:10.1016/s0022 1694(98)00098 5. Henkel, H. S., 2010. SOFIA Virtual Tour Lake Okeechobee. http://sofiausgsgov/ Hiscock, J. G. Thourot, C. S., Zhang, J., 2003. Phosphorus budget land use relationships for the northern Lake Okeechobee watershed, Florida. Ecological Engineering 21(1):63 74 doi:10.1016/j.ecoleng.2003.09.005. Hoos, A. B., 1990. Recharge Rates and Aquifer Characte ristics for Selected Drainage Basins in Middle and East Tennessee. USGeological Survey Water Resources Investigations Report 90 4015, 34 pp. Hopmans J. W. G. H. S., 2005 Soil water flow at different spatial scales. in Encyclopedia of Hydrologic Sciences 2. (edited by M. G. Anderson and F. J. McDonnell):pp. 1 11, John Wiley, New York.

PAGE 83

83 Houck, O. A., 2002. The Clean Water Act TMDL Program: Law, Policy, and Implementation. Howell, T. A., R.L. McCormick, C.J. Phene, 1985. Design and installation of large weigh ing lysimeters. Trans ASAE 28(1):106 112. Itenfisu, D., R. L. Elliott, R. G. Allen & I. A. Walter, 2003. Comparison of reference evapotranspiration calculations as part of the ASCE standardization effort. Journal of Irrigation and Drainage Engineering Asc e 129(6):440 448 doi:10.1061/(asce)0733 9437(2003)129:6(440). Jacobs, J. M., S. L. Mergelsberg, A. F. Lopera & D. A. Myers, 2002. Evapotrans piration from a wet prairie wetland under drought conditions:Paynes Prairie Preserve. Florida, USA Wetlands 22:374 385 Jagtap, S. S. & J. W. Jones, 1989. Stablity of crop coefficients under different climate and irrigation management practices. Irrigation Science 10(3):231 244. Jia, X., M. D. Dukes, J. M. Jacobs, 2009. Bahiagrass crop coefficients from eddy correlatio n measurements in central Flordia. Irrigation Science 28(1):5 15. Johnson, A. I., 1967. Specific yield compilation of specific yields for various materials. US Geological Survey Water Supply Paper 1662 D 74 p. Kachroo, R. K., 1992. Storage required to au gment low flow: a regional study. Hydrol Sci J 37(3): 247 261. Knisel, W. G., 1963. Baseflow recession analysis for comparison of drainage basins and geology. Journal of Geophysical Research 68(12):3649 & doi:10.1029/JZ068i012p03649. Koob, T., M. E. Barbe W. E. Hathorn. 1999. Hydrologic design consideration of constructed wetlands for urban stormwater runoff. Journal of the American Water Resources Association 35:323 331. Korkmaz, N., 1990. The estimation of groundwater recharge from spring hydrographs. Hydrol Sci J 35(2(4)): 209 217. Krasnostein, A. L., Oldham, C. E., 2004. Predicting wetland water storage. Water Resources Research 40(10) doi:10.1029/2003wr002899. Kroger, R., M. M. Hollad, M. T. Moore & C. M. Cooper, 2008. Agricultural drainage ditches mitigate phosphorus loads as a function of hydrological variability. Journal of Environmental Quality 37(1):107 113 doi:10.2134/jeq2006.0505. LaBaugh, J. W., 1986. Wetland ecosystem studies from a hydrologic perspective. Water Resources Bulletin 22 (1): 1 10.

PAGE 84

84 Lee, T. M. H., K.H.; Metz, P.A.; Sacks, L.A., 2009. Comparative hydrology, water quality, and ecology of selected natural and augmented freshwater wetlands in West Central Florida. U S Geological Survey Professional Paper 1758 152. Leibowitz, S. G & T. L. Nadeau, 2003. Isolated wetlands: State of the science and future directions. Wetlands 23(3):663 684 doi:10.1672/0277 5212(2003)023[0663:iwsafd]2.0.co;2. Lewis, D., Liudahl, K., Noble, C., Carter, L. 2003. Soil Survey of Okeechobee County, Flori da. Natural Resources Conservation Service, U S Department of Agriculture. Lewis, D. L. K. J. L., C.V. Noble, and L.J.Carter, 2001. Soil Survey of Okeechobee County, Florida. In: USDA/NRCS in cooperation with the University of Florida, I. o. F. a. A. S ., Agricultural Experimental Stations and Soil Science Department; and Florida Department of Agriculture and Consumer Services. (ed). Lide, R. F., V. G. Meentemeyer, J. E. Pinder & L. M. Beatty, 1995. Hydrology of a Carolina bay located on the upper coast al plain of western south Carolina. Wetlands 15(1):47 57. A Herman (Cited by Hall 1968). Manfreda, S., 2008. Runoff generation dynamics within a humid river basin. Natural Hazards and Earth System Sciences 8(6):1349 1357. Mao, L. M., M. J. Bergman & C. C. Tai, 2002. Evapotranspiration measurement and estimation of three wetland environments in the Upper St. Johns River Basin, Florida. Journal of the American Water Re sources Association 38(5):1271 1285 doi:10.1111/j.1752 1688.2002.tb04347.x. McCaffery, P. M., Hinkley, W.M., MacGill,R., Cherr, G.D. 1976. Report of Investigations in the Kissimmee Lake Okeechobee Watershed. Fla Depart Environ Reg Tech Series 2(2):Tallah assee. Florida. Meyboom, P., 1961. Estimating groundwater recharge from stream hydrographs. Journal of Geophysical Research 66(4):1203 & doi:10.1029/JZ066i004p01203. Min, J. H., D. B. Perkins, J. W. Jawitz & M. D. Annable, in process. Quantifying hydrolo gic pathways in subtropical depressional isolated wetlands: implications for hydrologic restoration. Hydrological Processes. Min, J. H., Perkins, D. B.,Jawitz, J. W, 2010. Wetland groundwater interactions in subtropical depressional wetlands. Wetlands 30(5 ):997 1006 doi:10.1007/s13157 010 0043 9.

PAGE 85

85 Mishra, S. K. & V. P. Singh, 2004. Validity and extension of the SCS CN method for computing infiltration and rainfall excess rates. Hydrological Processes 18(17):3323 3345 doi:10.1002/hyp.1223. Mitsch, W. T., J. G Gosselink, 1993. Wetlands. Van Nostrand Reinhold Co. New York, NY, USA. Muller, P. O., 2008. Lake Okeechobee. World Book Online Reference Center 12 Jan. 2008 Nachabe, M., Masek, C., Obeysekera, J. 2004. Observations and modeling of profile soil water storage above a shallow water table. Soil Science Society of America Journal 68: 719 724. Neumann, P. E., 2012. Shallow aquifer storage and recovery (SASR): regional management of underground water storage in hydraulically connected aquifer stream systems. A thesis submitted to Oregon State University. Nichols, W. D., 1994. Groundwater discharge by phreatophyte shrubs in the Great Basin as related to depth to groundwater. Water Resources Research 30(12):3265 3274 doi:10.1029/94wr02274. Owen, C., 1995. Water budget and flow patterns in an urban wetland. Journal of Hydrology 169:171 187. P., D. B., 2007. Water and phosphorus budgets of depressional wetlands in the Okeechobee Basin,FL. A dissertation presented to the graduate school of the university of Florida Parsons, D. F. H., M.; van der Kamp, G., 2004. Infiltration and solute transport under a seasonal wetland: bromide tracer experiments in Saskatoon, Canada. Hydrological Processes 18(11):2011 2027 doi:10.1002/hyp.1345. Penman, H. L., 1948. Natural evapor ation from open water, bare soil and grass. Proceedings of the Royal Society of London Series A, Mathematical and Physical Sciences, Vol. 193, (No. 1032.):p. 120 145. Penman, H. L., 1956. Evaporation: an introductory survey. Netherlands J Agric Sci 1:9 29, 87 97, 151 153. Pitman, W. V., 1973. A mathematical model for generating monthly river flows from meteorological data in South Africa. Report 2/73, Hydrological Research Unit, University of the Witwatersrand,Johannesburg. Price, J. S. & J M. Waddington, 2000. Advances in Canadian wetland hydrology and biogeochemistry. Hydrological Processes 14(9):1579 1589 doi:10.1002/1099 1085(20000630)14:9<1579::aid hyp76>3.0.co;2 #.

PAGE 86

86 Qian, M., 1995. On the Regulartions of Water Level Fluctuation in Poya ng Lake. Journal of Lake Sciences 7(3). Reddy, K. R., Diaz, O. A., Scinto, L. J., Agami, M., 1995. Phosphorus dynamics in selected wetlands and streams of the Lake Okeechobee Basin. Ecological Engineering 5(2 3):183 207. Reiss, K. C., 2006. Florida wetland condition index for depressional forested wetlands. Ecological Indicators 6(2):337 352 doi:10.1016/j.ecolind.2005.03.013. Richardson, C. J., 1985a. Mechanisms controlling phosphorus retention capacity in fresh water wetlands. Science 228(4706):1424 1427 d oi:10.1126/science.228.4706.1424. Richardson, C. J., 1985b. Nutrient, carbon, and water dynamics of a Titi shrub swamp ecosystem in Apalachicola, Florida. PhD diss Univ of Florida, Gainesville. Rorabaugh, M. I., 1964. Estimating changes in bank storage and ground water contribution to streamflow. International Association of Scientific Hydrology Publication 63:432 441. Rudolf, B. & U. Schneider, 2005. Calculation of gridded precipitation data for the global land surface using in situ gauge observation 2nd W orkshop of the International Precipitation Working Group. SFWMD, 1993. Update of Okeechobee SWIM Plan. South Florida Water Management District West Palm Beach, Florida(January, 1993). SFWMD, 2004. Optimization of best management practices (BMPs) for beef c attle ranching in the Lake Okeechobee Basin SFWMD, 2013. Lake Okeechobee Watershed Protection Program. South Florida Environmental Report:Chapter 8. SFWMD, F., FDACS, 2012. Lake Okeechobee Protection Program, Lake Okeechobee Protection Plan. South Florida Water Management District, West Palm Beach, FL Florida Department of Environmental Protection, Tallahassee, FL(Florida Department of Agriculture and Consumer Services, Tallahassee, FL). Shukla, S., Graham, W. Goswami, D., Knowles, J., Hodges, A., Nair, V., Christman, M., Wu, C. 2011. Evaluation of Cow calf Best Management Practices With Regards to Nutrient Discharges in the Lake Okeechobee Basin. Final report submitted to Florida Department of Agriculture and Consumer Services Tallahasee, FL, and South Florida Water Management District, West Palm Beach, FL. Sophocleous, M. A., 1991. C ombining the soilwater balance and water level fluctuation methods to estimate natural groundwater recharge practical aspects Journal of Hydrology 124(3 4):229 241 doi :10.1016/0022 1694(91)90016 b.

PAGE 87

87 Tallaksen, L. M., 1995. A review of baseflow recession analysis. Journal of Hydrology 165( No. 1 4: 349 370). Thornthwaite, C. W., 1948. An approach toward a rational classification of climate. Geogr Rev.38 (1), 55 94. Tiner, R. W., 2003. Estimated extent of geographically isolated wetlands in selected areas of the United States. Wetlands 23(3):636 652 doi:10.1672/0277 5212(2003)023[0636:eeogiw]2.0.co;2. Todini, E., 2007. Hydrol ogical catchment modeling : Past, present and fu ture. Hydrol Earth Syst Sci 11:468 482. USDA, 1990. State soil geographic database for Florida. US Department of Agriculture, Soil Conservation Service, Gainesville, Florida. Valiantzas, J. D., 2013. Simple ET0 forms of Penman's equation without wind and/o r humidity data. I: theoretical development. Journal of Irrigation and Drainage Engineering Asce 139(1):1 8 doi:10.1061/(asce)ir.1943 4774.0000520. Vogel, R. M. & C. N. Kroll, 1992. R egional geohydrologic geomorphic relationships for the estimation of low flow statistics Water Resources Research 28(9):2451 2458 doi:10.1029/92wr01007. Whigham, D. F. & T. E. Jordan, 2003. Isolated wetlands and water quality. Wetlands 23(3):541 549 doi:10.1672/0277 5212(2003)023[0541:iwawq]2.0.co;2. Wilcox, D. A., Sweatet, M. J. 2006. A water budget approach to restoring a sedge fen affected by diking and ditching. Journal of Hydrology 320(3 4): 501 517. Winter, T. C. L., J. W., 2003. Hydrologic considerations in defining isolated wetlands. Wetlands 23(3):532 540 doi:10.1672 /0277 5212(2003)023[0532:hcidiw]2.0.co;2. Wise, W. R., Annable, M.D., Walser, J.A.E., Switt, R.S., Shaw, D.T. 2000. A wetland aquifer interaction test. Journal of Hydrology 227:257 272. Wittenberg, H. & M. Sivapalan, 1999. Watershed groundwater balance estimation using streamflow recession analysis and baseflow separation. Journal of Hydrology 219(1 2):20 33 doi:10.1016/s0022 1694(99)00040 2. Wood, W. W., Sanford, W. E., 1995. Chemical and Isotopic methods for Quantifying Ground Water Recharge in a Regi onal Semi Arid Environment. Groundwater 33(3):458 468. Wright, J. L., 1982. New evapotranspiration crop coefficients. J Irrig Drain E ASCE(ASCE, 108 (IR2)): 57 74.

PAGE 88

88 Xu, C. Y., Seibert, J., Halldin, S., 1996. Regional water balance modelling in the NOPEX are a development and application of monthly water balance models. Hydrol Earth Syst Sci.

PAGE 89

89 BIOGRAPHICAL SKETCH Jing Guan was born in Qingdao, China, a beautiful coastal city. Her interest s in natural science go back to her years in secondary school, where she indulged her self in all kinds of little scientific experiments. A s the generation born in the 1980s in China, she witnessed the overpopulation, water environmental deterioration and exhaustion of fishery resource. It is her enthusiasm to natural scienc es and a sense of mission to make difference for the environment that inspires her to study Fisheries Sciences and Aquaculture in undergraduate study. Undergraduate c ourses deepened her understanding to E nvironmental S cience and brought her to a n ew realm of W ater S cience. She determined to do interdisciplinary researches on water treatments, water quality management and hydro ecology restoration. In 2012, she met Dr. James W. Jawitz and was mentored by Dr. Jawitz in So il and Water Science Department UF. T hanks to the guidance and patience of Dr. Jawitz throughout these years Jing ha s learned an incredible amount within a short time. This thesis is a result of some of the research during the time.