Citation
Protection of Floodplain Wetlands Associated with Minimum Flow and Level Development in Southwest Florida

Material Information

Title:
Protection of Floodplain Wetlands Associated with Minimum Flow and Level Development in Southwest Florida
Creator:
MUNSON, ADAM BURTON
Copyright Date:
2008

Subjects

Subjects / Keywords:
Bodies of water ( jstor )
Floodplains ( jstor )
Habitat loss ( jstor )
Rivers ( jstor )
Surface water ( jstor )
Topographical elevation ( jstor )
Water management ( jstor )
Water resources ( jstor )
Water supply ( jstor )
Wetlands ( jstor )
City of Tallahassee ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Adam Burton Munson. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
3/1/2007
Resource Identifier:
659813304 ( OCLC )

Downloads

This item has the following downloads:

munson_a ( .pdf )

munson_a_Page_36.txt

munson_a_Page_58.txt

munson_a_Page_03.txt

munson_a_Page_50.txt

munson_a_Page_75.txt

munson_a_Page_60.txt

munson_a_Page_91.txt

munson_a_Page_71.txt

munson_a_Page_40.txt

munson_a_Page_77.txt

munson_a_Page_11.txt

munson_a_Page_69.txt

munson_a_Page_16.txt

munson_a_Page_82.txt

munson_a_Page_37.txt

munson_a_Page_76.txt

munson_a_Page_02.txt

munson_a_Page_35.txt

munson_a_Page_32.txt

munson_a_Page_23.txt

munson_a_Page_19.txt

munson_a_Page_25.txt

munson_a_Page_07.txt

munson_a_Page_46.txt

munson_a_Page_57.txt

munson_a_Page_17.txt

munson_a_Page_26.txt

munson_a_Page_88.txt

munson_a_Page_61.txt

munson_a_Page_29.txt

munson_a_Page_66.txt

munson_a_Page_73.txt

munson_a_Page_54.txt

munson_a_Page_06.txt

munson_a_Page_74.txt

munson_a_Page_78.txt

munson_a_Page_12.txt

munson_a_Page_68.txt

munson_a_Page_27.txt

munson_a_Page_45.txt

munson_a_Page_52.txt

munson_a_Page_53.txt

munson_a_pdf.txt

munson_a_Page_90.txt

munson_a_Page_38.txt

munson_a_Page_49.txt

munson_a_Page_70.txt

munson_a_Page_30.txt

munson_a_Page_87.txt

munson_a_Page_08.txt

munson_a_Page_09.txt

munson_a_Page_44.txt

munson_a_Page_63.txt

munson_a_Page_15.txt

munson_a_Page_79.txt

munson_a_Page_14.txt

munson_a_Page_05.txt

munson_a_Page_43.txt

munson_a_Page_01.txt

munson_a_Page_56.txt

munson_a_Page_39.txt

munson_a_Page_85.txt

munson_a_Page_51.txt

munson_a_Page_18.txt

munson_a_Page_10.txt

munson_a_Page_47.txt

munson_a_Page_59.txt

munson_a_Page_48.txt

munson_a_Page_04.txt

munson_a_Page_83.txt

munson_a_Page_28.txt

munson_a_Page_89.txt

munson_a_Page_65.txt

munson_a_Page_81.txt

munson_a_Page_62.txt

munson_a_Page_20.txt

munson_a_Page_55.txt

munson_a_Page_86.txt

munson_a_Page_67.txt

munson_a_Page_24.txt

munson_a_Page_80.txt

munson_a_Page_34.txt

munson_a_Page_33.txt

munson_a_Page_84.txt

munson_a_Page_13.txt

munson_a_Page_42.txt

munson_a_Page_64.txt

munson_a_Page_41.txt

munson_a_Page_22.txt

munson_a_Page_21.txt

munson_a_Page_31.txt

munson_a_Page_72.txt


Full Text





PROTECTION OF FLOODPLAIN WETLANDS ASSOCIATED WITH MINIMUM FLOW
AND LEVEL DEVELOPMENT INT SOUTHWEST FLORIDA























By

ADAM BURTON MUNSON


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2006

































Copyright 2006

by

Adam Munson









ACKNOWLEDGMENTS

Professionally, I wish to express my thanks to Dr. Joseph Delfino, Dr. Charles Cichra, Dr.

Douglas Shaw and Dr. Warren Viessman for their tireless review and their continued belief in

my abilities. I thank the staff of the Ecologic Evaluation Section at the Southwest Florida Water

Management District for their efforts in developing minimum flows and levels and their constant

vigilance for new technologies and methods. Specifically, I would like to thank Marty Kelly,

Doug Leeper, and Jonathan Morales, who have all contributed to the minimum flows and levels

dialogue in Florida and never sugarcoat criticism. I also appreciate the support of the Southwest

Florida Water Management District and acknowledge that the work contained in this dissertation

is academic research and does not necessarily reflect the policies of the Southwest Florida Water

Management District.

Personally, I would like to thank my wife who reminds me that ambition without

persistence remains unfulfilled. I also thank my parents who encouraged me to pursue whatever

studies made me happy.












TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .............. ...............3.....


LI ST OF T ABLE S ............ ..... ._ ...............6...


LI ST OF FIGURE S .............. ...............7.....


AB S TRAC T ..... ._ ................. ............_........8


CHAPTER


1 INTRODUCTION ................. ...............10.......... ......


2 DETERMINING MINIMUM FLOWS AND LEVELS IN FLORIDA: A
LEGISLATIVE MANDATE ................. ...............15.................


Introducti on ................ .. .. ...... ........... ... .......... ..........1
History of Water Management in Florida (1824-1972) ................. ............... 16...........
Water Resources Act of 1972 .............. ... ... ........... .......... ..... ... ...........2
Water Management in Florida; Quality not Quantity (1972-1997) ................. ................ ..22
The 1997 Law and Water Quantity ................. ...............26.......... ...
Response to the 1997 Law ................. ...............30...............
Summary ................. ...............33.................
Conclusions............... ..............3


3 CURRENT TECHNIQUES UTRLIZED FOR THE QUANTIFICATION OF
FLOODPLAIN HABITAT LOSS ASSOCIATED WITH REDUCTIONS IN
WET SE ASON F LOW ................. ...............36.......... ......


Introducti on ................. ...............36.................
M ethodology ................. ...............40.................
Re sults ................ ...............45.................
Discussion ................. ...............48.................
Sum m ary ................. ...............5.. 1..............


4 THE UTILIZATION OF GEOGRAPHIC INFORMATION SYSTEMS IN
DETERMINING FLOODPLAIN INUNDATION FOR MFL DEVELOPMENT ................61


Introducti on ................. ...............61.................

Study Area .............. ...............63....
M ethod s .............. ...............64....
Re sults ................ ...............65.................
Discussion .................. ...............66.................
GIS to HEC-RAS .............. ...............66....
HEC-RAS to GIS .............. ...............67....













GIS in MFL Development............... ..............6
Future W ork. ............. ...... ...............68...


5 SUMMARY, CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE
W ORK ............. ...... ...............78...


Summary ............. ...... ._ ...............78...
Conclusions............... ..............8
Future Work............... ...............8 1.


LIST OF REFERENCE S ............. ...... .__ ...............84..


BIOGRAPHICAL SKETCH .............. ...............91....










LIST OF TABLES


Table page

3-1 Beginning Julian days for the Wet and Dry periods (Blocks 1 and 3) and ending date
for the Wet period at four different gage stations in the SWFWMD ................. ...............52

3-2 Mean flow requirement at the USGS Alafia River near Lithia gage to inundate
floodplain features, and percent of flow reductions associated with no more than a
15% reduction in the number of days that the required flows are equaled or exceeded
for two benchmark periods (1940 through 1969 and 1970 through 1999).............._..... ....53

3-3 Mean flow requirements at the USGS Myakka River near Sarasota gage needed to
inundate floodplain features, and percent of flow reductions associated with no more
than a 15% reduction in the number of days that the required flows are equaled or
exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999)........54

3-4 Mean flow requirements at the USGS Peace River at Arcadia gage needed to
inundate floodplain features, and percent of flow reductions associated with no more
than a 15% reduction in the number of days that the required flows are equaled or
exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999).......55

4-1 Percent exceedance flows for the Braden River near Lorraine gage ................ ...............70

4-2 Estimated acres of inundated land at different exceedance flows. ............. ...................71










LIST OF FIGURES


Figure page

2-1 Map of the State of Florida showing the three early water management districts and
the five current water management districts. ............. ...............35.....

3-1 Map of the Southwest Florida Water Management District showing the Alafia, Peace
and Myakka rivers and the USGS gage locations used in this study on each. ..................56

3-2 Percent-of-flow reductions that result in a 15% reduction in the number of days that
flows on the Alafia, middle Peace and Myakka rivers are achieved. Horizontal lines
represent the flow reduction standards chosen by the SWFWMD for each river.
Graphs are adapted from Kelly et al. 2005a, b, and c. ................... ............... 5

3-3 Percent-of-flow reductions that result in a 15% reduction in the number of days that
flows on the Alafia, middle Peace and Myakka rivers are reached. Gage flow has
been divided by the area of the watershed above the gage and power curves fit to
each rivers data. ............. ...............58.....

3-4 Percent flow reduction required to result in a 15% loss of top width plotted against
flow for the Alafia, Myakka, and middle Peace Rivers. .................. ................5

3-5 Percent of flow reduction corresponding to a 15% spatial loss (measured as the
summation of cross-section top width) plotted against the percent of flow reduction
corresponding to a 15% temporal loss (measured as a loss of days during which the
river historically reached a specific flow), for the Alafia, Myakka, and middle Peace
R ivers. ............. ...............60.....

4-1 Location map of the Braden River in southwest Florida. ................. ................ ...._.72

4-2 Location of surveyed cross-sections across the Braden River. .........__ ..... ............. ...74

4-3 Portion of the digital elevation model generated of the Braden River. ........._.... .............73

4-4 Portion of the Braden River with modeled inundation patterns from a 1% exceedance
flow ............. ...............75.....

4-5 LiDAR points used in the DEM of the Braden River ................. ......... ................76

4-6 Conceptual view of inundated area calculated by HEC-RAS and in a GIS
environment. The Shaded Blue area represents Modeled inundation patterns in GIS.
The Red trapezoid is a depiction of how HEC-RAS would estimate the inundated
area. .............. ...............77....









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PROTECTION OF FLOODPLAIN WETLANDS ASSOCIATED WITH MINIMUM FLOW
AND LEVEL DEVELOPMENT INT SOUTHWEST FLORIDA

By

Adam Burton Munson

December 2006

Chair: Joseph Delfino
Major Department: Environmental Engineering Sciences

Floodplain wetlands, associated with river systems, have been recognized as important

components of the ecologic community of river corridors. However, the importance of out-of-

bank flows to the riverine ecosystems is often not recognized by instream methods used to assess

flow requirements. Quantification of potential differences in floodplain inundation associated

with differences between minimum flow requirements and natural hydrologic regimes should be

a critical component of regulatory programs that are intended to protect or restore riverine

habitat. The Southwest Florida Water Management District uses the historic flow records to

produce a temporal measure of habitat loss when developing minimum flow requirements for

floodplain wetland protection. In this dissertation an alternative approach, employing a measure

of spatial loss by examining the change in top width at each cross section modeled under

different flow conditions, is compared to the temporal measure

Both spatial and temporal measures of habitat loss can be related to percent flow reduction

in a river channel. However, both serve as a proxy to evaluate the extent of the river-floodplain

connection. This connection has not yet been fully described on any river studied by the District

because complicated floodplain geometry is presently represented by a limited number of

floodplain cross-sections.









This dissertation concludes with an analysis of the role Geographic Information Systems

(GIS) coupled with hydraulic models can play in IVFL development by providing data

concerning floodplain inundation under different flow scenarios. Riverine ecosystems benefit

from spatial analysis since it provides data in three temporally dynamic spatial dimensions. The

coupling of a one-dimensional hydraulic model and spatial information from GIS results in a

powerful tool for relating variations in floodplain inundation with variations in river flow. When

developing minimum flows, topographic data, in a geographic information system, can provide

detailed information about riverine floodplain connection and lead to better understanding of

river floodplain interaction.









CHAPTER 1
INTTRODUCTION

For many years, surface water in Florida was drained to support development.

Channelization of rivers, development of canals, and draining of wetlands were the foundation of

early water management in Florida (Anderson and Rosendahl 1998, Egozcue 2001, Purdum et al.

1998). In more recent years, ground water use has increased as water demands have grown.

Under these conditions, the importance of natural systems and their linkage to the economy of

Florida and its potable water supply have become more evident. In more recent times, the water

related resources of the state have been offered some protection through the requirement that

minimum flows and levels (MFLs) be established.

The 1972 Florida Water Resources Act (Chapter 373, Florida Statutes (F.S.)) represents

the initial legislation calling for development of MFLs in the state of Florida. Aimed at

protecting the ecology of natural systems from significant harm" associated with water

withdrawals, MFLs represent an attempt to quantify the hydrologic requirement of aquifers,

rivers, lakes, and wetlands. Initially, MFL development was somewhat sparse because of

uncertainties in the legislative language, among other factors. It was not until 1997, when the

Florida legislature enacted several maj or changes to Florida water law in an attempt to improve

water resource planning and protection, that MFL development became a key component in

water resource regulation in the state. The five water management districts of Florida responded

to the changes in the law by prioritizing water bodies within their boundaries and developing

MFLs for water bodies as identified on each priority list.

A number of rivers in the Southwest Florida Water Management District (SWFWMD)

were considered potentially impacted by withdrawals and thus to be high value candidates for

MFL development and as such were placed on the priority list. The District subsequently









developed a multi-parameter approach to determining MFLs on the freshwater portions of these

rivers that included evaluation of loss of floodplain habitat. This action was concurrent with the

development of approaches by other water management districts in the state. Although

floodplains have long been recognized as seasonally important riverine habitat, floodplain

inundation has historically not been addressed in most minimum flow determinations (Postal and

Richter 2003). Middleton (1999) also points out that regulation of water regimes has not always

fully appreciated the interaction between floodplains and rivers via the flood pulse. Regulation

of river flows can, however, result in decreased stage fluctuations and alteration of inundation

patterns of floodplain wetlands (Poff et al. 1997, Woltemade 1997). Quantification of potential

differences in floodplain inundation associated with differences between minimum flow

requirements and natural hydrologic regimes should, therefore, be a critical component of

regulatory programs that are intended to protect or restore riverine habitat.

Compared to instream evaluations of minimum flows requirements, there has been

relatively little research on river flows necessary for meeting the requirements of floodplain

species, communities or functions. However, periodic inundation of riparian floodplains by high

flows is closely linked with the overall biological productivity of river ecosystems (Crance 1988,

Junk et al. 1989). Further evidence exists suggesting that floodplain inundation benefits fish

(Ainsle et al. 1999, Hill and Cichra 2002, Wharton et al. 1982), supports high rates of primary

production (Brinson et al. 1981, Conner and Day 1979), is critical to food webs (Gregory et al.

1991, Vannote et al. 1980) and results in development of wetland soils that are important to the

overall function of the river ecosystem (Kuensler 1989, Stanturf and Schoenholtz 1998,

Walbridge and Lockaby 1994, Wharton et al. 1982).









The Southwest Florida Water Management District (District) has developed an approach

which recognizes that fundamental to the development of MFLs is the realization that a flow

regime is necessary to protect the ecology of the river system (Hill et al. 1991, Hupalo et al.

1994, Richter et al. 1996, Stalnaker 1990). The high flow component of the flow regime is

intended to protect the connection between the floodplain and the river. The resource

management goal identified for the seasonally predictable wet period is maintenance of seasonal

hydrologic connections between the river channel and floodplain to ensure floodplain structure

and function.

Minimum flows developed for the high flow season are intended to protect ecological

resources and values associated with floodplains by maintaining hydrologic connections between

the river channel and floodplain and maintaining the natural variability of the flow regime.

These goals are quantified through use of the Hydrologic Engineering Centers-River Analysis

System (HEC-RAS) and the alteration of long-term flow records to evaluate floodplain feature

inundation patterns associated with channel-floodplain connectivity.

A temporal measure of habitat loss has been employed by the District in establishing

MFLs on freshwater river systems in southwest Florida, during the high flow season of the year.

This reduction in number of days of inundation is a temporal measure of loss based on how

many days, annually, a specified water-surface elevation was reached compared to how often

that elevation would have been reached if flow conditions had been reduced by a given amount.

Data and methods used to develop minimum flows for seasonal, high-flow periods for the

Alafia River, Myakka River, and Peace River are used in this study to illustrate this approach to

quantifying habitat change associated with temporal difference in floodplain inundation patterns.

An alternative approach, employing a measure of spatial loss by examining the change in top









width at each cross section in a hydraulic model under different flow conditions, is also explored.

The use of a temporal measure of habitat loss for establishing 1VFLs during high flows is

compared to the spatial loss of habitat for the same flow reduction. The applicability of both is

discussed and the results compared to determine which is more restrictive in terms of allowable

flow reduction.

However, as Mertes (1997) points out, floodplain vegetation development and persistence

may not, however, necessarily depend wholly on inundation from the river channel.

Groundwater seepage, hyporheic inputs, discharge from local tributaries, and precipitation can

also lead to floodplain inundation. Recent work on the upper segment of the Peace River and the

Alafia River in central Florida suggests that direct and continuous inundation of floodplain

wetlands by river flows is insufficient to account for inundation needs of the dominant species

found in the wetlands (SWFW1VD 2002).

After comparing the temporal and spatial measures of habitat loss, being employed in the

minimum flow and level process, this dissertation examines an alternative means to measuring

habitat. The use of high-density remotely gathered data to generate a digital elevation model is

explored. This terrain model is then used in the development of cross-sectional data for use in a

one-dimensional hydraulic model. Beyond this, data generated in the hydraulic model can be

processed in a Geographic Information System (GIS) environment to produce inundation maps

which are not limited to the one dimensional cross-sections. The coupling of a one-dimensional

hydraulic model and spatial information from GIS results in a powerful tool for relating

variations in floodplain inundation with variations (or reductions) in river flow.

The obj ectives of this dissertation include reviewing the history of water management in

Florida with an emphasis on decisions culminating in the 1997 legislation requiring the









development of minimum flows and levels. Attention is paid to water quantity issues and

policies and the natural and anthropogenic factors influencing them. The measures of riverine

floodplain habitat loss use in response to the minimum flows and levels mandate are also

examined and temporal loss measures are compared with spatial loss measures. Finally, a

method of examining inundation changes in a GIS system environment is presented as an

alternative to current approaches.









CHAPTER 2
DETERMINIG MINIMUM FLOWS AND LEVELS IN FLORIDA: A LEGISLATIVE
MANDATE

Introduction

On average, Florida receives 140 cm of rainfall annually (Carriker 2000, Henry 1998).

This precipitation feeds Florida' s diverse hydrologic systems and makes it the second wettest

state in the United States. Florida has more than 7,700 lakes, ranging from 0.4 ha to over

180,000 ha (Shafer et al. 1986). Thirteen major rivers originate in, or flow through, the state

with a combined mean annual discharge of nearly 1700 m3S-1 (Henry 1998). Contributing to this

total are 27 first magnitude springs that discharge a total average flow of 272 m3S-1 or more than

22 billion liters day-l (Rosenau et al., 1947). Florida also contains significant areas ofwetlands,

which at one time covered 54 percent of the state and still comprise about 30 percent of the total

land area (Henry 1998, SWFWMD 2001).

In a state with such abundant water resources, stringent water supply management may

seem unnecessary. This was the presumption for many years, as surface water systems were

drained to support development. Channelization of rivers, development of canals, and draining

of wetlands were cornerstones of early water management in Florida (Anderson and Rosendahl

1998, Egozcue 2001, Purdum et al. 1998). In more recent years, ground water use has increased

as water demands have continued to grow. Under these conditions, the importance of natural

systems and their linkage to the economy of Florida and its potable water supply has become

more evident.

The 1972 Florida Water Resources Act (Chapter 373, Florida Statutes (F.S.)) represented

the initial legislation calling for development for minimum flow and levels (MFLs) in the State.

Aimed at protecting the ecology of natural systems from significant harm" associated with

water withdrawals, MFLs represent an attempt to quantify the hydrologic requirement of










aquifers, rivers, lakes, and wetlands. Though some water management agencies did proceed

with the development of regulatory flows and levels, it was not until 1997 that legislation was

enacted which propelled MFLs from a stand-alone statute to an integral part of water resource

management and regulation.

The obj ectives of this chapter include providing a brief review of the history of water

management in Florida with an emphasis on decisions culminating in the 1997 legislation

requiring the development of minimum flows and levels. Attention is paid to water quantity

issues and policies and the natural and anthropogenic factors influencing them. Also examined

are the effects of the 1997 law on water management and the efforts that have been made to

comply with the legislative mandates.

History of Water Management in Florida (1824-1972)

In the early years of Florida's governance, there was little activity that could be said to

constitute a statewide water policy. With respect to water management and use, local

communities and landowners were free to do as they chose with little interference. Water was

abundant, often too much so for most settlers, and prior to statehood in 1845, only minor concern

seems to have been given to regional or statewide water management issues. Although not

enacted, an 1824 proposal by the legislative council calling for the construction of a canal across

Florida for ship passage represents one exception to the dearth of early large-scale water

management and planning (Purdum et al. 1998).

Florida was granted statehood in 1845 and was given 202,344 ha of land by the federal

government for "internal improvements". In 1850, the U.S. Congress additionally conveyed all

swamps and inundated lands, over 8 million ha in total, to state ownership through the Swamp

Lands Act (U. S. Congress 1911). In subsequent years, public lands granted to the state by

Congress were sold and the resulting funds were funneled through the state's Internal










Improvement Fund and used to invest in railroads and guarantee railroad bonds. However, the

outbreak of the Civil War devastated Florida' s railroads, and with the prospect of railroads

failing, future sales of public lands went to funding canals to drain the Everglades (Egozcue

2001). The legislature looked to the south Florida Everglades for reclamation and settlement,

believing these lands offered suitable areas for cultivation of sugar, rice, cotton and tobacco

(Egozcue 2001).

In 1881, the state sold four million acres of land to Hamilton Disston for $1 million. A

businessman from Philadelphia, Disston began channelization of the Caloosahatchee and the

Upper Kissimmee River basins in south and central Florida (Purdum et al. 1998). Numerous

other companies were given charters to dredge canals throughout the state and were given public

lands in compensation for their efforts. The land was subsequently sold to farmers as it was

drained. Under pressure from farmers, the legislature encouraged the creation of additional

canals and dykes to drain more land for development (Egozcue 2001). In 1883, Lake Beauclair

and Lake Apopka in central Florida were connected to reduce Lake Apopka's water elevation by

approximately one meter and enhance the farming ventures on the northern shore of the lake

(Shofner 1982). During the years leading up to the 20th century, additional canals were

completed between Lakes Apopka, Dora, Eustis, Griffin and the Ocklawaha River. These canal

proj ects reinvigorated the dream of a cross-Florida shipping canal, which had been considered

earlier but had been set aside in favor of railroad development.

In 1902, Congress heeded a United States Army Corp of Engineers (USACOE)

recommendation and a federal navigation proj ect was initiated on the Kissimmee River. Other

regional proj ects, developed to de-water the Everglades, led to the creation of the Everglades

Drainage District in 1907 (Figure 1). This was the first regional authority created by the state to










implement water policy. The District constructed six maj or canals totaling over 700 km, which

still connect Lake Okeechobee to the Gulf of Mexico and the Atlantic Ocean. As had central

Florida' s earlier trend, the drainage of land for agriculture in south Florida continued to be

encouraged. In 1913, the General Drainage Act gave landowners the right to form districts to

reclaim their lands. Over 100 such districts were created in the Everglades area; some still

operate today (Anderson and Rosendahl 1998, Purdum et al. 1998).

The 1920's saw a marked shift in state water management goals. Until that time, lands had

been drained and canals created to reclaim land and foster the development of both agriculture

and urban areas. During the 1920s, four major hurricanes hit south Florida and during the 1928

hurricane, Lake Okeechobee overflowed, killing over 2,000 people. This catastrophe led to an

increase in Federal assistance for water management projects and spurred the state to establish

the Okeechobee Flood Control District in 1929 (Anderson and Rosendahl 1998; Figure 1).

While the methods remained largely the same, the stated goal, which brought in federal help

from the USACOE, became one of flood protection, which had the ancillary benefit of providing

continued drainage for agriculture and urban development.

By the 1940s, the legislature began to act under mounting recognition that state-level

oversight of water management was necessary. In 1945, the Florida legislature created the State

Board of Conservation. The board was tasked with the protection of the state's marine, mineral

and water resources. In 1947, the legislature created the Water Survey and Research Division,

whose records were later forwarded to the Florida Geological Survey in 1955 when the division

was dissolved (Purdum et al. 1998). Political willingness to move towards increased state

oversight was matched by the public sentiment of the times. Concern over the adequacy of

environmental protection was growing as newspapers reported fish kills in lake Apopka










(Bachmann et al. 1999) and publication of The Everglades: River of Gra~ss (Douglas 1947)

raised considerable concern for the state's water resources.

As had happened earlier in the 1920s, natural disasters influenced the course of state water

policy. In 1947, two hurricanes caused substantial flooding in Miami. The Central and South

Florida Flood Control District was established in 1949 to coordinate the efforts of the federal

government' s Central and South Florida Flood Control Proj ect. The district' s plan had many

components, including flood control, water control, water conservation, prevention of salt water

intrusion into freshwater reservoirs and aquifers, preservation of fish and wildlife, improved

navigation, and pollution abatement (Maloney et al. 1968). A decade later, two floods within a

two-year period, followed by Hurricane Donna in 1960, inundated over 1 million acres of land

east of Tampa and caused $29 million in damages. The state enlisted federal aid through the

USACOE, and with a price tag of $100 million, the Four Rivers Basin Proj ect was conceived.

The Southwest Florida Water Management District (SWFWMD; Figure 1) was created in 1961

to manage the proj ect and to oversee Florida's $40.5 million share of the expenses (Blake 1980).

Clearly, flooding turned the focus and the resources of the state to flood control and

canalization and diluted the mounting desire for environmental regulation and natural systems

protection. However, the era of canalization and drainage began to come to an end as 1970 saw

a demand for a reevaluation of the Four Rivers Basin Proj ect, and work on the proj ect was halted

(Purdum et al. 1998). Under mounting public pressure and with concern that drainage of

regional surface waters was causing environmental harm, the SWFWMD requested that the

USACOE undertake an environmental impact assessment of the Four Rivers Basin Proj ect, an

action that eventually led to abandonment of the proj ect.










By the early 1970s, public demands for environmental protection were again gaining

momentum and the state was committed to implementing more comprehensive forms of water

management. Florida had been struggling with a need for statewide water management and

several divisions and boards were established as a result of committees established by various

governors. The state's executive office made environmental issues a major policy platform.

Under greater state scrutiny, the number of approved dredge-and-Hill proj ects decreased from

2,000 to 200 a year (Blake 1980, Derr 1989, Kallina 1993). Several environmental laws were

passed in 1971, including the Environmental Protection Act, which allowed Florida citizens to

sue the state when environmental laws are not enforced. However, it wasn't until 1972 that the

state Einally developed a coherent water policy through passage of the Florida Water Resources

Act.

Water Resources Act of 1972

The Florida Water Resources Act of 1972 was an important step forward for state water

policy. Previously, water had been managed regionally in a disjointed manner. Agencies had

often been created in response to recent weather events or local economies, and their policies

were so narrowly focused on one issue that other problems were often exacerbated by their

activities. However, the task of creating centralized agencies and passing the laws that would

empower them to conduct more comprehensive planning for water resources, was daunting. To

this end, Florida lawmakers turned to A M~odel Water Code (Maloney et al. 1972).

Prior to the adoption of the code, Florida adhered predominately to eastern water law.

Eastern water law is a common law riparian system, which largely restricts water rights to

landowners whose property is directly adj acent to water bodies. Further, water rights are subj ect

to reasonable use. Historically, there was a lack of administrative guidance on reasonable use,

leaving litigation, and it' s uncertain outcomes, as the only means of determining reasonableness.









This uncertainty of future water availability can discourage industrial investment in areas with no

guarantees of meeting future needs (Maloney et al. 1972).

Believing that Florida' s water resources would not be protected unless statewide control

and long-term, scientifically based planning was adopted, the drafters of the Model Water Code

stated three purposes:

to take into account the hydrologic interrelationship of all types of water resources in the
state; to provide greater certainty than is possible under a court-administered reasonable
use approach; yet to retain sufficient flexibility to make possible realistic long-range plans
for the conservation and wise use of water resources and the elimination of waste.
(Maloney et al. 1972)

To achieve these goals, multipurpose water planning was essential. The state had a long history

of single-purpose special districts. The majority of entities actually vested with regulatory or

management authority were local forms of government, including drainage districts, irrigation

districts, water supply districts, aqueduct districts, sewer districts and mosquito control districts

(Maloney et al. 1980). In contrast, the state had a short history of multi-purpose special districts;

the first having been the Central and South Florida Flood Control District whose charge was

multifaceted (Maloney et al. 1968). The state had also established, in 1961, the SWFWMD.

This was the first water resource district in the state whose name reflected the broad multifaceted

functions it was tasked with, rather than a single function, such as drainage or flood control.

To achieve multipurpose water planning at a state level (multipurpose water planning was

implemented 10 years earlier at the federal level by the Water Resource Planning Act), A Model

Water Code (Maloney et al. 1972) identified four requirements for success. First, centralized

planning was considered necessary. Lack of centralized control could lead to state programs that

could contradict each another and diminish management efficiency. Second, the planning must

be based in science; the hydrologic cycle and the effects of pollution, land use, recharge loss, and

development must all be considered as well as their interrelationships. Third, planning must









recognize relationships between water pollution and water use and these relationships must be

considered when developing long-term plans. Finally, consumptive use must be regulated

(Maloney et al. 1972).

The Model Water Code (Chapter 373, Florida Statutes) envisioned statewide planning, but

also envisioned scientific planning based on hydrologic boundaries rather than political

boundaries. This seeming contradiction was resolved by the creation of five regional water

management districts, collectively covering the state. These districts would be responsible for

local issues within their boundaries but would not have the authority to prepare independent

water-use plans (Tj oflat and Quincey 2001). Therefore, scientific research could be conducted

and district water control structures and lands could be owned and managed on a regional basis,

but a single state entity would be responsible for adopting a state water plan. The five districts

are the Northwest Florida (NWFWMD), Suwannee River (SRWMD), St. John's River

(SJRWMD), Southwest Florida (SWFWMD), and South Florida Water Management (SFWMD)

Districts (Figure 1). Statewide oversight was initially granted to the Division of Natural

Resources and now resides with the re-named Florida Department of Environmental Protection

(DEP) .

Water Management in Florida; Quality not Quantity (1972-1997)

The 1972 Water Resources Act gave broad powers to the regional Water Management

Districts (WMDs) and to the Division of Natural Resources (now, DEP). The new law allowed

for consumptive use permitted on a reasonable-beneficial use basis. Defining reasonable and

beneficial use proved to be difficult however, and before the turn of the 21s~t century some areas

of the state would be over-permitted for withdrawals. The law also required that each district

formulate a water shortage plan and establish minimum flows and levels (MFLs) for surface

waters and minimum levels for aquifers.









District MFLs programs were not, however, implemented with any sense of immediacy. A

Model Water Code (Maloney et al. 1972) had called for minimum flows and levels to establish a

"harm" threshold to protect water resources from withdrawals. As Ross (2001) points out, this

linked the MFLs program to the consumptive-use permitting program, which was designed to

prevent "harm" to the water resources of the state. However, when enacting the 1972 law, the

passage concerning MFLs was modified slightly to protect the water resources from significantt

harm". This difference is notable because although neither term is defined, the law establishes

that there is a difference between "harm" and significantt harm", and suggests some separation

of the consumptive use and MFLs programs. Ross (2001) further pointed out that the linkage

between MFLs and the State Water Use Plan, though present in the original 1972 legislation, was

removed in 1973 when the MFL requirement was moved into a stand-alone statue (F.S.

373.042). These changes from the Model Water Code reduced MFLs to a more marginal role

than was likely intended by the authors of the Code.

While water quantity policies were still evolving, water quality concerns took center stage.

The 1970s saw the state acquire 141,640 ha of environmentally sensitive lands, often for the

protection of adjacent water bodies. In 1975, the Department of Environmental Regulation

(now, DEP) was formed and given the task of controlling pollution, permitting dredge and fill

activities and supervising the WMDs. Water resource legislation in the 1980s included the 1983

Florida Water Quality Assurance Act and, in 1984, the Warren Henderson Wetlands Protection

Act, which provided additional protection for wetlands by including criteria for the evaluation of

dredge and fill permits (Purdum et al. 1998). Florida also passed the 1987 Surface Water

Improvement and Management Act. This act focused on the restoration and protection of

surface water bodies of regional significance.









The 1990s saw the continued support for public land acquisition and, by 1998, Florida had

acquired more than 849,840 ha of land for conservation purposes. Combined with lands

protected by federal and local programs or owned by private conservation groups, the total

amounted to 3 million ha or 22 % of Florida (Purdum et al. 1998). Florida also continued to

revisit water management decisions of the past, eventually passing the Everglades Forever Act,

continuing restoration work around Lake Apopka, remaining engaged in discourse regarding the

restoration of the Ocklawaha River, and with the help of the USACOE, initiated restoration of

the Kissimmee River (Anderson and Rosendahl 1998, Bachmann et al. 1999, Canfield et al.

2000, Lowe et al. 1999, Purdum et al. 1998).

Perhaps the clearest indication of a growing recognition that water quantity as well as

water quality was an important management concern comes from the difference between a

statement made in 1985 and one a decade later in 1995. In 1985, the legislature adopted the state

comprehensive plan, which stated:

Florida shall assure the availability of an adequate supply of water for all competing uses
deemed reasonable and beneficial and shall maintain the functions of natural systems and
the overall present level of surface and ground water quality. Florida shall improve and
restore the quality of waters not presently meeting water quality standards. (F.S. Chapter
187)

While this statement is reasonably strong, it is imprecise. Its only mention of water quantity is to

note that Florida shall maintain an "adequate" supply to meet the needs of consumers who's

proposed use must meet the permitting criteria of being reasonable and beneficial. It suggests

that the function of natural systems and the "overall" level of water quality shall be maintained.

This statement maintains the theme in the 1980s of protection of water quality, which it seeks to

"improve and restore." In 1995, a similar statement was made in the Florida Water Plan adopted

by DEP:









water must be managed to meet the water needs of the people while maintaining,
protecting, and improving the state's natural systems. (DEP 1995)

This statement reflects a shift in perception and understanding and acknowledges that Florida's

water resources are not boundless. It maintains that the goal of water supply is to meet the needs

of the people of the state and that this must be accomplished in conjunction with the protection

and improvement of natural systems. In this statement, no distinction is made between the need

to manage water quantity and water quality. This shift in management goals seems to be

indicative of the growing and prevailing need to address water quantity issues, which had been

playing a larger and larger role in regional politics.

Florida has large populations in relatively small areas of the state. This has led to localized

problems, especially in coastal areas where the local water supplies are most vulnerable to salt

water intrusion and surface water bodies are frequently affected by groundwater levels. Highly

localized demand raised concern in some counties that their water would be exported to more

developed counties, leaving them with degraded environmental resources and reduced ability to

enhance development. These factors, coupled with generally low surface water levels

throughout the state in the late eighties and early nineties, led to a number of law suits and

complaints being filed against many of the WMDs. These complaints from citizen groups and

local governments led to a number of actions. In 1993, a decision involving the SJRWMD found

that the water management district must establish MFLs and determined that it should be done

within a reasonable time period. At the same time, the SFWMD responded to complaints filed

against it by initiating MFL development for the Everglades. A few years later in 1996, the

Florida Legislature responded to local concerns in the SWFWMD by requiring a priority list be

established for MFL development within the district and that once developed, MFLs should be

subject to independent scientific peer review (F.S. 373.042).









In that same year, a Governor's task force on water supply and funding was convened

(Exec. Order No. 96-297). A core assumption adopted by the task force was that water demand

was increasing and that the increased demand needed to be met (Matthews and Nieto 1998).

Though the task force put forth approximately fifty recommendations, its primary theme was that

increasing the supply of available water was necessary to ensure continued population and

economic growth. This would lead to new responsibilities for all the water management districts

and new protections for the environment.

The 1997 Law and Water Quantity

The recommendations from the Governor's task force provided the building blocks for

legislation referred to as the "1997 Water Act". This law represented the first major revision of

state water policy since the Model Water Code was utilized to create Chapter 373 of the Florida

Statutes in 1972. Perhaps the best summary of the changes in policy is taken from the statute

itself, which was amended to "promote the availability of sufficient water for all existing and

future reasonable-beneficial uses and natural systems" (F.S. 373.016 (2) (d)). This statement is

significant because it denotes a fundamental shift in the state's water policy. Until this juncture,

the state's primary goal was the allocation of the existing water supply among uses and, recently,

consumptive uses and natural systems. The Districts would now be responsible for allocating

portions of available water between users and natural systems, and would also be charged with

promoting expansion of the water supply, through water resource development.

Though other topics were covered, the 1997 changes to Chapter 373 were strongly focused

on water quantity issues and consumptive uses. The new law more clearly defined the role of the

WMDs and local governments in supplying water to the public. The water management districts

were tasked with "water resource development" and technical support. "Water resource

development" was defined as the formulation and implementation of regional water resource










management strategies and includes the task of data collection, construction and operation of

structural components for flood control, surface and underground water storage, and ground

water recharge augmentation (F.S. 373.019 (19)). Local governments, or utilities were charged

with "water supply development", which was defined as design, planning, construction and

operation of the infrastructure necessary to collect and distribute water for sale, resale or end use

(F.S. 373.019 (21)).

The primary goal of this shift in policy, which aimed to increase the volume of water

available, was to avoid pitting consumptive users against each other and against natural systems

(Matthews and Nieto 1998). To accomplish these goals, the legislature stated that proj ects which

create sustainable water sources but require Einancial assistance to complete, and proj ects that

implement reuse, storage, recharge, or conservation of water be given priority funding by the

water management districts (F.S. 373.083 1 (4) (a) (3)). Further, if a priority proj ect is expected

to bring an MFL water body into compliance, it is given "first consideration" for funding

assistance (F. S. 373.0831 (4) (b)).

Along with establishing funding priority for water resource development, the 1997 law

also clarified the role of water resource planning in water management. The DEP is required to

produce the Florida Water Plan in conjunction with the WMDs, regional water supply

authorities, and others (F.S. 373.036). The Florida Water Plan includes the district Water

Management Plans that are called for in the 1997 legislation. These plans are based on a 20-year

planning horizon and are to be updated at least every Hyve years (F.S. 373.036 (2)). The plans

include, among other items, methodologies for establishing minimum flows and levels, water

supply assessments, anticipated future needs, projections of water supply adequacy, and

Regional Water Supply Plans which are developed for areas determined to have an insufficient










supply to meet the needs of proj ected reasonable-benefieial uses and to sustain the water

resources and related natural systems (F.S. 373.036 (2) (b)). This last clause is important

because it means that when the water needs of natural systems are proj ected to be inadequate

(i.e., a minimum flow or level is expected to be violated), the WMDs must implement regional

water supply planning. Incorporation of the five Water Management District water management

plans into the Florida Water Plan provides for greater consistency among statewide and regional

planning efforts.

Another important change to Chapter 373 in 1997 concerned development of minimum

flows and levels (MFLs). While the Model Water Code was developed with MFLs as an integral

part of consumptive water use permitting, Chapter 373 (F.S.) did not explicitly incorporate MFLs

into the permitting process. Some efforts had been made by the WMDs to determine

ecologically sensitive water levels for selected water bodies. For example, implementation of

the SWFWMDs Lake Levels program, had led to the adoption of management levels for nearly

four hundred lakes by 1996 (SWFWMD 1996). Yet MFLs remained an undefined and relatively

abstract concept. In large part, this dilemma stemmed from the phrase significantt harm" which

was included in the statute (F.S. 373.042). Because the language was changed from the original

"harm" standard contained in the Model Water Code, it was clear that legislators intended to

differentiate "harm" from significantt harm." Compounding the issue was the lack of guidance

provided by the legislature regarding factors that should be considered when establishing MFLs.

This is important because many of the state's surface water basins have been substantially

altered, and a return to higher historic levels in these basins would require maj or restoration

efforts and often result in the flooding of privately owned lands and structures, including homes.

Another basis for the lack of action could be attributed to uncertainty regarding the purpose of









IVFLs in the W1VDs regulatory framework. Even if significantt harm" had been defined, the

IVFLs statute remained isolated from consumptive use permitting criteria, and thus served no

immediately evident role in the district's regulatory charge.

In 1997, the Florida Legislature provided guidance regarding factors to be considered

when establishing minimum flows and levels. The language the legislature sought was one of

compromise and practicality. It sought to protect the water resources of the state from water

withdrawals that would cause significant harm. However, they acknowledged that reestablishing

historic water flows or levels was not necessarily desirable in all cases and that structural

changes in the watersheds should be considered during the development of 1VFLs (F.S. 3 73.0421

(1) (a)). They also provided exemptions for water bodies smaller than 25 acres in size, those that

have been constructed or ones that no longer serve their historic function (F.S. 373.0421 (1) (b)).

A list of both cultural and scientific factors to be considered when establishing minimum flows

or levels was developed, and incorporated into the Florida Administrative Code (Ch. 62-40.473

F.A.C.). Identified factors include recreation, Eish and wildlife habitats and passage, estuarine

resources, maintenance of freshwater storage and supply, aesthetic and scenic attributes,

filtration and absorption of nutrients, water quality, and navigation. However, determination of

the precise methodology for establishing 1VFLs was left up to each WMD and has, to date,

resulted in the development of different approaches for establishing minimum flows and levels.

The 1997 Water Act also provided criteria for actions to be implemented when a minimum

flow or level is not met or proj ected to be unmet in the next 20 years. The statute directs that the

DEP or the W1VD governing boards shall implement a recovery or prevention strategy, which

includes the development of additional water supplies and conservation concurrently with, to the

extent practical, reductions in permitted withdrawals (F.S. 373.0421 (2)). This strategy became









part of the regional water supply plan and, thus, eventually the Florida Water Plan. It is this

language, from F.S. 373.0421 (2), that explicitly gave great weight to the minimum flows and

levels developed by the water management districts. Violation of a flow or level can result in

reduction of permitted withdrawals, and by extension, denial of proposed withdrawals that are

projected to cause a violation of a minimum flow or level. Thus, the 1997 statute corrected a

variance between the minimum flows and levels requirement and consumptive water use

permitting. Matthews and Nieto (1998) pointed out that the WMDs must attempt to develop

alternative water sources in conjunction with any reductions in permitted withdrawals. They

contended that combining the obligation to engage in environmental restoration with the

requirement for water resource development should protect the environment without inequitably

reducing water use or stunting economic development. Therefore, a balance between

maximizing water resource benefits for both consumptive use and natural systems protection

may be achieved.

Response to the 1997 Law

The mandate to give priority funding to water supply proj ects has been well implemented

and, in fiscal year 2001, the five water management districts spent between 17.7 % of their total

budget in the Northwest District and 40. 1 % of their budget in the Southwest District with all

five districts committing $287 million, or 26.9 % of their combined budget on water resource

development (DEP 2000). In 2002 the total climbed to $344 million, representing 28.0 % of the

combined WMD budgets. Fiscal year 2003 budgets included approximately $372 million for

water supply activities (DEP 2003a). Funds budgeted between 2001 and 2006 are projected to

exceed $1.4 billion (DEP 2002a).

The 1997 amendments to Chapter 373 require all five water management districts to

implement regional water supply planning in areas where water supplies are inadequate to meet









the proj ected water needs on a 20-year horizon. Water demand in the state is expected to

increase from approximately 27.2 billion L day-l in 1995 to 34.8 billion L day-l by the year 2020

(DEP 2002a). Based on this projected demand, four of the five water management districts

anticipate water deficits in at least a portion of their jurisdictional boundaries and have been

required to submit regional water plans to the Florida DEP. As of August 2001, all the required

plans were complete (DEP 2002a).

While it will take years to measure the effectiveness of maj or water-resource planning

efforts, there is some evidence to suggest at least partial success. Planning and funding priorities

guide the districts towards water resource development goals, which will need to be achieved to

meet future water demands while protecting Florida aquatic ecosystems. Planning has helped

quantify the future needs and focus water supply development on cost effective solutions,

including seawater desalinization and aquifer storage and recovery. While these technologies

hold promise and are being actively developed, they have yet to fully reach their potential role in

Florida's water supply. Other strategies such as reclaimed water and conservation efforts are

currently contributing cost effective alternatives to potable water use (DEP 2002b). For

example, reclaimed water systems in Florida supplied 2.2 billion L day-l in 2002 and system

capacity is currently in excess of 4.2 billion L day-l (DEP 2003b).

While the 1997 Water Resource Act made considerable progress in clarifying the role that

MFLs are to play in state water policy, it also left many technical questions involved in the

development of MFLs unanswered. Considerable work has been done on lotic systems, such as

rivers and streams. Much of this work centers on protection of habitat by identifying necessary

flow regimes using tools such as Instream Flow Incremental Methodology (IFIM) and Physical

Habitat Simulation System (PHABSIM) (Poff et. al. 1997, Postel and Richter 2003, Richter et. al










1997). Work on MFL development for lentic systems is less extensive. Further complicating

MFL development is the need to account for both natural systems and cultural values during

development. Cultural factors, such as aesthetic and scenic attributes, are difficult to quantify

and like many natural systems values, are related to and affected by numerous other factors, such

as water quality.

While all Hyve water management districts seek, through MFLs, to prevent significant harm

to the natural systems in their region, those that have initiated the process are employing

different approaches (Ch. 40C-8, 40D-8 and 40E-8, F.A.C.). While this process has not yielded

a uniform methodology, it may be appropriate for each WMD to establish its own methods since

each faces different challenges; some need to protect water resources that are not highly

influenced by consumptive uses, while others must develop MFLs for water resources that are

severely impacted by withdrawals. As of 2002, three of the Hyve WMDs have established MFLs

on a total of 132 lakes, 9 rivers and streams, 46 wetlands, 20 aquifers, and 2 estuaries (DEP

2003c).

One of the greatest strengths of the 1997 law has been tying compliance with MFLs to

water planning, funding of water resource development and regulatory oversight of consumptive

use. In the SWFWMD, this interaction between MFLs, planning, funding and water-use

permitting is evident. Regional, public-supply wellfields in the Northern Tampa Bay area are

believed partially responsible for reducing water levels in some area water bodies. To meet

MFLs in the region, an agreement calling for a combination of permit reductions, water supply

development, and funding assistance was entered into by the SWFWMD, the regional water

supply authority (Tampa Bay Water), and its member governments. Specifically, the agreement

required ground water withdrawals at 11 regional wellfields to be reduced, in stages, from 158










mgd to 346 mgd by December 31, 2007. An intermediate reduction to 121 mgd was set for

December 31i, 2002 and has been met. To help offset reductions of wellfield pumping, 85 mgd

of new water supply will be needed. The SWFWMD is providing, following funding priority

directives, up to $183 million in funding assistance between 1995 and 2007 for alternative water

supply development (DEP 2000). Alternative source projects underway, include surface water

source development, indirect potable reuse, and seawater desalination.

Summary

The 1997 Water Act improved and enhanced the tools available to water managers and

provided needed emphasis on water quantity and consumptive-use issues (Hamann 2001). The

minimum flows and levels component asks the question, presuming we are going to harvest as

much as water as we can, "How much can we harvest without significantly harming the natural

resources of the state?" The planning requirements ask managers to determine where projected

shortages may be expected during the next 20 years, and how the shortfalls can be met with

alternative sources and conservation efforts. The priority-funding requirement attempts to assure

that the alternative sources identified in regional and state plans receive necessary economic

support.

The challenge, as we move forward, is for scientists, engineers, policy makers and citizens

to respond to the 1997 legislation with ingenuity. The groundwork for future success has been

established through the legal acknowledgement of clean water as a limited resource. When

establishing consumptive-use limits, managers must consider scientific parameters such as

absorption of nutrients and estuarine resources, as well as cultural parameters such as recreation,

aesthetics, and scenic attributes. Water resource managers and policy makers must also

determine what constitutes significantt harm", and identify the "practicable" steps needed to










develop alternative water resources to offset withdrawal reductions. Thus far, Florida has

responded to these challenges.

Conclusions

In keeping with the 1972 vision of regional water management, the 1997 legislation

retained flexibility for each of the water management districts to address regional water resource

issues. The 1997 legislative changes have directly impacted state water management by

mandating the development of MFLs and assuring that they are met by tying them to water

planning, funding of water resource development and regulatory oversight of consumptive use.

Planning has helped quantify the future needs and focus water supply development on cost-

effective solutions. The Water Management Districts' commitment to water supply planning is

evident in the district budgets, which demonstrate a steady increase in funding for water supply

proj ects. Floridians have had the resourcefulness to ask, "How much water is there"? This

means that water will no longer be a limited resource in some abstract or theoretical future, but

will have real limitations defined for its use.














Water Control Districts
Central and South Florida
Flood Control District

Everglades Drainage District
Okeechobee Flood Control District

Water Management Districts
SNWFWMD
SSFWMD
SSJRWMD
SSRWMD
SSWFWVMD

80 40 0 80


160


I L I Miles a /
Figure 2-1. Map of the State of Florida showing the three early water management districts and
the five current water management districts.









CHAPTER 3
CURRENT TECHNIQUES UTRLIZED FOR THE QUANTIFICATION OF FLOODPLAIN
HABITAT LOSS ASSOCIATED WITH REDUCTIONS IN WETSEASON FLOW

Introduction

Although floodplains have long been recognized as seasonally important riverine habitat,

floodplain inundation has historically not been addressed in most minimum flow determinations

(Postal and Richter 2003). Middleton (1999) also points out that regulation of water regimes has

not always fully appreciated the interaction between floodplains and rivers via the flood pulse.

Regulation of river flows can, however, result in decreased stage fluctuations and alteration of

inundation patterns of floodplain wetlands (Poff et al. 1997, Woltemade 1997). Quantification

of potential differences in floodplain inundation associated with differences between minimum

flow requirements and natural hydrologic regimes should, therefore, be a critical component of

regulatory programs that are intended to protect or restore riverine habitat.

Compared to instream evaluations of minimum flows requirements, there has been

relatively little research done on river flows necessary for meeting the requirements of floodplain

species, communities or functions. Junk et al. (1989) noted that the "driving force responsible

for the existence, productivity, and interactions of the maj or river-floodplain systems is the flood

pulse". Periodic inundation of riparian floodplains by high flows is closely linked with the

overall biological productivity of river ecosystems (Crance 1988, Junk et al. 1989). Many fish

and wildlife species associated with rivers utilize both instream and floodplain habitats. During

inundation of the river floodplains, the habitat and food resources available to these organisms

greatly expands (Ainsle et al. 1999, Hill and Cichra 2002, Wharton et al. 1982). Inundation

during high flows also provides water and nutrients that support high rates of primary production

in river floodplains (Brinson et al. 1981, Conner and Day 1979). This primary production

generates large amounts of organic detritus, which is critical to food webs on the floodplain and









within the river channel (Gregory et al. 1991, Vannote et al. 1980). Floodplain inundation also

contributes to other physical-chemical processes that can affect biological production, uptake and

transformation of macro-nutrients, and development of wetland soils that are important to the

overall function of the river ecosystem (Kuensler 1989, Stanturf and Schoenholtz 1998,

Walbridge and Lockaby 1994, Wharton et al. 1982).

Floodplain vegetation development and persistence may not, however, necessarily depend

wholly on inundation from the river channel. Groundwater seepage, hyporheic inputs, discharge

from local tributaries, and precipitation can also lead to floodplain inundation (Mertes 1997).

Recent work on the upper segment of the Peace River and the Alafia River in central Florida

suggested that direct and continuous inundation of floodplain wetlands by river flows is

insufficient to account for inundation needs of the dominant species found in the wetlands

(SWFWMD 2002). However, because river channel-floodplain connections are important, can

be influenced by water use, and may be a function of out-of-bank flows, it is valuable to

characterize this connectivity. As with most other floodplain-related water management issues,

technically sound development of minimum flows and levels requires such characterization.

The Southwest Florida Water Management District (District) has developed an approach

which recognizes that fundamental to the development of minimum flows and levels (MFLs) is

the realization that a flow regime is necessary to protect the ecology of the river system (Hill et

al. 1991, Hupalo et al. 1994, Richter et al. 1996, Stalnaker 1990). The recognition of a seasonal

flow regime as an essential component of this task led the District to develop minimum flow

criteria for three distinct annual-flow periods; periods of low, medium and high flows. The

District approach also acknowledges the effects of climatic oscillations on regional river flows









and includes identification of two distinct benchmark flow periods that are consistent with

effects of the Atlantic Multidecadal Oscillation (AMO) (Enfield et al. 2001, Kelly 2004).

Because the methodologies employed by the District in the low and medium flow periods

are commonly applied, consisting of components such as wetted perimeter analysis and physical

habitat simulation analysis, they are not examined in the context of this chapter. Further, the

focus of this paper remains on the connection between the floodplain and the river, which most

frequently occurs during the high flow season. The resource management goal identified for the

seasonally predictable wet period is maintenance of seasonal hydrologic connections between the

river channel and floodplain to ensure floodplain structure and function

Minimum flows developed for the high flow season are intended to protect ecological

resources and values associated with floodplains by maintaining hydrologic connections between

the river channel and floodplain and maintaining the natural variability of the flow regime.

These goals quantified through use of the Hydrologic Engineering Centers-River Analysis

System (HEC-RAS) and the alteration of long-term flow records to evaluate floodplain feature

inundation patterns associated with channel-floodplain connectivity.

The reduction in number of days of inundation is a temporal measure of loss based on how

many days, annually, a specified water-surface elevation was reached compared to how often

that elevation would have been reached if flow conditions had been reduced by a given amount.

This temporal measure of habitat loss has been employed by the District in establishing MFLs on

freshwater river systems in southwest Florida, during the high flow season of the year. It is

applied by reducing historic flows iteratively until the percent of flow reduction that results in a

15% reduction in the number of days of inundation is found. The use of a percent of flow









reduction in establishing MFLs assures protection of the natural hydrograph (Flannery et al.

2002, Kelly at al. 2005a, b, c).

In Florida, regional Water Management Districts, by virtue of their responsibility to permit

the consumptive use of water and a legislative mandate to protect water resources from

"significant harm" have been directed to establish MFLs for streams and rivers within their

boundaries (Section 373.042, Florida Statutes; Munson et al. 2005). The Southwest Florida

Water Management District has developed methods for establishing MFLs that acknowledge the

importance of seasonal flow regimes, including flows necessary for floodplain inundation. A

temporal measure of habitat loss based on changes in inundation is used to develop minimum

flows for seasonal high-flow periods. An alternative approach, employing a measure of spatial

loss, by examining the change in top width at each cross section in the model under different

flow conditions, is also explored. The use of a temporal measure of habitat loss for establishing

MFLs during high flows is compared to the spatial loss of habitat for the same flow reduction.

The applicability of both is discussed and the results compared to determine which is more

restrictive in terms of allowable flow reduction.

The Alafia River, Myakka River and Peace River are used in this chapter to illustrate the

temporal approach to quantifying habitat change associated with temporal difference in

floodplain inundation patterns (Figure 3-1). All three rivers are located in southwest Florida.

The Alafia River originates as a series of creeks that combine to form two branches of the Alafia

River know and the North Prong and the South Prong. The North prong drains a swampy area

and is characterized by low-lying wetlands. The South Prong is drains a highly mined area and

is more incised. The Alafia River downstream of the confluence is more highly incised as well.

The Alafia River above the Lithia gage (USGS# 02301500) drains approximately 335 mi^2. The









Myakka River drains approximately 229 mi^2 above the Sarasota gage (USGS# 02298830). The

Myakka River is a shallow river with broad flat floodplains. Two wide shallow lakes are near

the gage site as well as substantial grass prairies. It is the lease incised of the three rivers

studied. The Peach River at Arcadia gage (USGS# 02296750) captures approximately 1,392

mi^2 Of the Peace River watershed. The channel is well defined upstream of Arcadia and widens

downstream of Arcadia. In some areas of the river the floodplain is over a mile wide.

Methodology

The District's approach to protection of flows associated with floodplain habitats,

communities, and functions involves consideration of direct connection between the river

channel and the floodplain. As part of this process, plant communities and soils were identified

across river floodplains at a number of sites on the Alafia River, Myakka River, and Peace River

in central Florida (Figure 3-1), and periods of inundation/connection with the rivers were

reconstructed on a seasonal basis. These data were used to characterize the inundation of these

communities/soils by out of bank river flows, and to develop criteria for establishing minimum

flows for the seasonal period of high flows.

Floodplain cross-sections were selected based on the location of vegetation communities

identified from the USGS Gap Analysis Program maps within the Alafia, Myakka, and Peace

river corridors. Attention was also made of the location of shoals in the river in an attempt to

capture the hydraulic control points in the model. Eight, twelve, and ten representative

floodplain/vegetation cross-sections were established perpendicular to the river channel within

dominant National Wetland Inventory vegetation types for the Alafia, Myakka, and Peace rivers,

respectively. Cross-sections were established between the 0.5 percent exceedance levels on

either side of the river channel, based on previous determinations of the landward extent of

floodplain wetlands in the river corridor. Ground elevations were determined at 50-foot intervals










along each cross-section. Where changes in elevation were conspicuous, elevations were

surveyed more intensively (Kelly et al. 2005a b c).

To characterize forested vegetation communities along each cross-section, changes in

dominant vegetation communities were located and used to delineate boundaries between

vegetation zones. At each change in vegetation zone, plant species composition, density, basal

area and diameter at breast height (for woody vegetation with a diameter at breast height > 2.5

cm) were recorded. Soils were characterized within each vegetation zone as hydric, organic,

peat, or mineral by obtaining at least three soil cores. The cores were examined to a depth

ranging from 51-152 cm to classify the soils. Special consideration was placed on locating

elevations of the upper and lower extent of mucky soils (> 20 cm in thickness) at cross-sections

where they occurred (Berryman and Henigar 2004).

Steady-state HEC-RAS modeling was used to determine corresponding flows at a

downstream gage that would be necessary to inundate specific floodplain elevations (e.g., mean

vegetation zone and soils elevations) along the floodplain/vegetation cross-sections. Version

3.1.1 of the HEC-RAS model released by the U. S. Army Corps of Engineers Hydrologic

Engineering Center in November 2002 was utilized for all three rivers. The HEC-RAS model is

a one-dimensional hydraulic model that can be used to analyze river flows. For subcritical

flows, it operates as a typical step-back model, resolving the energy equation between adj accent

cross-sections. Profile computations begin at a cross-section with known or assumed starting

condition and proceed upstream (US Army Corps of Engineers 2001). Models for each river

were based on cross-sectional data collected by the United States Geologic Survey and the

floodplain/vegetation data collected by the District (Hammett et al. 1978, Kelly et al. 2005 a, b,

c, Murphy et al. 1978).









For development of minimum flows, the year is categorized into three distinct flow periods

or "blocks". Based on flow records for long-term United States Geological Survey (USGS) gage

sites at the Myakka River near Sarasota (USGS# 02298830), the Alafia River at Lithia (USGS#

02301500), the Peace River at Arcadia (USGS# 02296750), and the Hillsborough River at

Zephyrhills (USGS# 02303000, The Hillsborough River was used to develop the season block

for the region but is not otherwise discussed in this chapter), the seasonal blocks were

determined as periods corresponding to dates when specific exceedance flows occurred. On an

annual basis, a low-flow period, Block 1, was defined as beginning when the median daily-flow

fell below and stayed below the annual 75% exceedance flow. Block 1 was defined as ending

when the high-flow period, or Block 3, began. Block 3 was defined as beginning when the

median daily flow exceeded and stayed above the mean annual 50% exceedance flow and ending

when the flow fell below and stayed below the mean annual 50% exceedance flow. The

medium-flow period, Block 2, was defined as extending from the end of Block 3 to the

beginning of Block 1.

Between these rivers, there was little difference in the dates that each defined flow period

began and ended (Table 3-1). Block 1 is defined as beginning on Julian day 110 (April 20 in

non-leap years) and ending on Julian day 175 (June 24). Block 3 is defined as beginning on

Julian day 176 (June 25) and ending on Julian day 300 (October 27). Block 2, the medium-flow

period, extends from Julian day 301 (October 28) to Julian day 109 (April 19) of the following

calendar year (Kelly et al. 2005a, b, c). Though the District has developed MFLs for all three

blocks, this paper focuses on the criteria for developing minimum flow standards for Block 3,

which runs from June 25 to October 27 of each year.









For development of MFLs, it is necessary to identify a benchmark period from which flow

deviations may be evaluated. Often changing flow trends are presumed to be the result of

anthropogenic stresses on a system. This is possibly because variation in flow is frequently

assumed to be the result of random independently and identically distributed random variables

such as rainfall in flood risk analysis (Olsen et al. 1999). However, the effect of multidecadal

oscillation on river flow patterns must also be recognized as natural climatic variations. As

Enfield et al. (2001) observe, the Atlantic Multi-decadal Oscillation has a pronounced effect on

rainfall in the continental United States and wet-season rainfall in peninsular Florida is

negatively correlated with the Atlantic Multi-decadal Oscillation (McCabe and Wolock 2002).

Kelly (2004) examined stream flows in Florida and identified a step-trend similar in timing to the

step-trend describing the AMO (Enfield et al. 2001). This resulted in the identification of two

benchmark periods, one from 1940-1969 and one from 1970-1999, which are characterized by

relatively higher and lower wet season river flows in peninsular Florida, respectively. For

determination of minimum flows, records from both benchmark periods are analyzed, and used

to develop percent-of-flow reduction criteria. Minimum flows are determined by calculating the

percent flow reduction for each benchmark period that would result in no more than a 15%

reduction in habitat. For the Block 3 period, this means identifying the flow reduction that

would result in no more than a 15% loss in the number of days of days that floodplain features

are inundated by the river, during each of the two benchmark periods and utilizing the more

limiting of the two.

In establishing minimum flows, the District defines a 15% change in habitat availability as

a criterion for identifying significant harm, or unacceptable change. For the high-flow period or

Block 3, this change is expressed as a temporal difference in the number of days specific water









surface elevations are reached. Potential temporal change or reduction in the number of days of

direct connection between the channel/floodplain at the floodplain/vegetation cross-sections on

the Alafia, Myakka and Peace rivers was estimated through the alteration of the historic flow

record and the HEC-RAS models. Target elevations corresponding to features of interest, such

as the median elevation of cypress swamps or the top of the river banks, were identified during

the vegetative surveys. The HEC-RAS model was then used to determine the flow necessary at

the downstream gage to inundate the features with flow from the river. The HEC-RAS model

thus allows the elevations, identified at different transects, to be compared as the flow required at

the gage site, to reach the corresponding elevation at the upstream cross-section where the

elevation was identified. The daily historic record from the gage site was then examined. All

days during the flow period of interest were checked to determine if the flow of interest was

exceeded. The total number of days during each year of the historic record that the flow of

interest was reached or exceeded were then tallied. Because the HEC-RAS model estimates the

water surface elevation at cross sections, based on flows at the down stream gage, the reduction

in the total number of days a flow is exceeded is equivalent to the reduction in total number of

days that the corresponding water surface elevation is exceeded, at any cross section. The flow

record was then reduced incrementally, until the percent flow reduction required to reduce the

number of days a flow was reached or exceeded was reduced by 15 percent.

A measure of spatial loss of habitat at each cross-section was also estimated using the

HEC-RAS model output and historic flow records. For each flow profie calculated in the

model, the top width (linear distance in a cross section of water's edge to water's edge) at each

cross section is derived. The sum of top widths at all cross sections in the model is computed for

each flow profie. This provides an estimate of the area inundated. Once top width was









calculated for each profie, the relationship between top width and flow was plotted. A 15%

reduction was then made to the top width at each flow profile in each model. The generated

relationships between top width and flow were then used to calculate the flow required at the

downstream gage sites to achieve the reduced top widths. The flow needed to achieve the non-

reduced top width and the flow required to maintain the top width minus 15% were then

compared to determine the percent of flow reduction required to reduce the top width by 15

percent.

Results

The vegetative surveys on the Alafia, Myakka and Peace rivers resulted in the

identification of different features of interest on each river. For the Alafia River, elevations of

eight floodplain features were identified at each of the eight vegetative cross-sections where they

occurred (Table 3-2). For the Alafia River, the downstream gage at which flow requirements

were compared is the USGS Alafia gage at Lithia (USGS# 02301500). Flow requirements, for

inundation of each feature ranged from 22.2 m3S-1 to reach the floodplain wetted perimeter

inflection point to 64.2 m3 S-1 to reach the low bank elevation for inundation of both sides of

river floodplain. Relatively high standard deviations, for the required flows, indicate that the

flow requirements for some features differed greatly among cross-sections. Flow reduction

resulting in 15% fewer days of inundation for the floodplain features, ranged from 5% for the

low-bank elevations and highest floodplain vegetation class to 9% for the highest swamp class

and floodplain wetted perimeter inflection point. Comparison of percent-of-flow reductions,

associated with the temporal loss of feature inundation between the two benchmark periods,

indicate that the 1970 to 1999 benchmark period provided the more conservative flow reduction

(Kelly et al. 2005a).









For the Myakka River, the median elevations of six vegetative zones (i.e., Oak-Palm Wet

Hammock and Panicum Marsh) as well as six physical characteristics (i.e., Lowest Bank

Elevation to inundation both sides of river floodplain and Median elevation of hydric soils) were

identified at each of the 12 vegetative cross-sections, when they occurred. For the Myakka River

the downstream gage at which flow requirements were compared is the USGS Myakka River

near Sarasota gage (USGS# 02298830). The downstream flow requirements for each identified

feature are summarized in Table 3-3 and range from 0.9 m3 S-1 to reach the median elevation of

mixed marsh to 24.4 m3 S-1 to reach the median elevation of oak-palm wet hammock. Standard

deviations are also presented and indicate that flow requirements of some features differ greatly

among cross-sections, as noted in the Alafia River. The percentage by which reducing flow

would result in 15% fewer days of the target flow being reached is also presented. Flow

reduction resulting in 15% fewer days ranged from 8% for the median elevation of oak-palm wet

hammock and the lowest bank elevation to inundate one side of the river floodplain to 68% for

the median elevation of mixed marsh. This analysis was completed for each of the two

benchmark periods and results were similar to those found for the Alafia River (Kelly et al.

2005b).

For the middle Peace River, 10 features, including both vegetative and physical, were

utilized in characterizing the flood plain (Table 3-4). For the middle Peace River the

downstream gage, at which flow requirements, are compared is the USGS Peace River at

Arcadia gage (USGS# 02296750). The downstream flow requirements for each identified

feature are summarized in Table 3-4 and range from 66.5 m3/S to reach the mean elevation of

mucky soils to 196.8 m3/S to reach the 90th percentile elevation of wet hardwood hammock.

Standard deviations are also presented and indicate that some features' flow requirements differ









greatly among cross-sections, as was true for the Alafia and Myakka rivers. The percentage, by

which reducing flow would result in 15% fewer days of the target flow being reached, is also

presented. Flow reduction resulting in 15% fewer days ranged from 5% for lowest bank

elevation to inundate both sides of the river floodplain to 10% for the mean elevation of mucky

soils. This analysis was completed for each of the two benchmark periods and results were

similar to both the Alafia River and Myakka River (Kelly et al. 2005c).

In establishing minimum flows for the high flow season on the Alafia, middle Peace, and

Myakka rivers, Kelly et al. (2005a, b, c) did not select a single floodplain feature to protect.

Rather, they noted that higher flows might require a slightly more restrictive standard than some

of the indicators associated with low flows and that higher flows seem to consistently tend

towards a reduction between 5% and 10% (Tables 3-2 to 3-4). To further investigate limiting

factors associated with the river floodplains, plots of percent-of-flow reductions that would result

in 15% losses in the number of days for which corresponding river flows were reached were

produced for each river (Figure 3-2). The plots indicate that up to an 8%reduction in the flows

necessary to inundate floodplain features of the Alafia and middle Peace rivers, including those

features not identified, would result in no more than a 15% reduction in the number of days the

features are inundated. Similarly, it was determined that a 7% reduction in flow in the Myakka

River, during the high-season flows, would not reduce the number of days of floodplain-river

connection by more than 15%. This measure of temporal loss of habitat is central to the

approach taken by the District in establishing MFLs to protect the river-floodplain connection

during the high flow part of the year.

In examining all three rivers it is evident that as flows increase, the percentage by which

the flow can be reduced, without lowering the number of days that flow is reached, is reduced. It









is useful to normalize the flow by looking at flow per watershed area. Percent-of-flow

reductions that would result in 15% losses of the number of days that river flows reached a given

flow plotted against flow per watershed area, are presented in Figure 3-3. Power trends were fit

to each curve. The trend lines and the data points for all three rivers are similar when plotted

together (Figure 3-3).

Utilizing the HEC-RAS model, the spatial loss associated with decreased flows was

determined. Specifically, the flow reductions, resulting in a 15% loss of top width, were

calculated. Percent of flow reductions, required to achieve a 15% loss of spatial habitat, were

plotted versus flow per square mile of watershed above the gage and again power trends were fit

to the data (Figure 3-4). For each river, the relationship between spatial loss and flow resembles

the relationship between temporal loss and flow, in so far as, at low flow, a high percentage of

flow reduction is required to result in a 15% decrease in top width. The fraction of flow required

to effect a 15% loss in top width, is reduced as flow increases until trending slightly upwards at

the highest modeled flows.

However, plots of the percent of flow reduction required to effect a 15% loss of spatial

habitat versus the percent of flow reduction required to effect a 15% loss of temporal habitat

indicate that for the range of flows examined, temporal loss is a more restrictive measure of

habitat loss on these three rivers than spatial loss (Figure 3-5). A 15% loss of temporal habitat is

associated with less of a flow reduction than a 15% loss of spatial habitat, as estimated by top

width.

Discussion

The goal of an IVFL determination is to protect the aquatic resource from significant harm

due to water withdrawals. An IVFL was broadly defined in the enacting legislation as "the limit

at which further withdrawals would be significantly harmful to the water resources or ecology of









the area." (F.S. Chapter 373.042, Munson et al. 2005). What constitutes "significant harm" was

not defined. The District has identified loss of flows associated with fish passage and

maximization of stream bottom habitat exposure as significantly harmful to river ecosystems.

Significant harm can also be defined as quantifiable reductions in the amount of available habitat

(Gore et al. 2002).

Determining the amount of habitat loss, or deviation from a benchmark, that a system is

capable of withstanding is based on professional judgment. In establishing MFLs, the

SWFWMD recognized that the Instream Flow Incremental Methodology (IFIM) involves a

negotiated threshold to be used as an acceptable measure of habitat loss (Bovee et al. 1998).

Gore et al. (2002) note that instream flow analysts often consider a loss of more than 15%

habitat, as compared to undisturbed or current conditions, to be a significant impact on a

population or assemblage when employing Physical Habitat Simulation (PHABSIM) analysis.

With some exceptions (e.g., loss of fish passage or wetted perimeter inflection point), there are

few clearly delineated break points which can be relied upon to judge when significant harm"

occurs. Hill and Cichra (2002) noted that loss of habitat in many cases occurs incrementally as

flows decline, often without a clear inflection point or threshold.

The District employed a threshold of a 15% change in habitat availability as a measure of

significant harm for the purpose of MFL development. Although the District utilized a 15%

change in habitat availability as a measure of unacceptable loss, percentage changes employed

for other instream flow determinations have ranged from 10% to 33%. For example, in reference

to the use of PHAB SIM, Dunbar et al. (1998) noted that an alternative approach is to select a

flow that provides protection to 80% of the habitat, which is equivalent to a 20% loss. Jowett

(1993) used a guideline of one-third loss of existing habitat at naturally occurring low flows, but









acknowledged that no methodology exists for the selection of a percentage loss of "natural"

habitat, which could be considered acceptable. The state of Texas utilized a target decrease of

less then 20% of the historic average of habitat area in establishing an inflow requirement for

Matagorda Bay (Texas Parks and Wildlife 2005).

For establishment of minimum river flows, the District has proposed the use of a percent of

flow approach, which identifies the percentage by which a flow may be reduced before the river

system is significantly harmed. This approach preserves natural flow patterns by protecting the

inherent variability of natural flow regimes. The approach accounts for flow seasonality, by

identifying three distinct annual flow seasons and identifying different habitat measures to be

protected during each season. Two distinct benchmark periods, with characteristically higher

and lower flows were used to analyze all measures of habitat loss to assure protection during

both phases of the climate cycle associated with the north Atlantic Ocean.

In examining the loss of temporal habitat on the three rivers, the comparison in Figure 3-3

suggests that these three rivers respond in a similar fashion to reduced flows. This could be

interpreted as the three rivers having similar flow distribution patterns during the wet season and

thus showing similar responses to declining flows. This is probable because rainfall drives wet-

season flows and since these three rivers are all located in central Florida, it might be expected

that some similarity in watershed-wide rainfall over the period of record exist (at least 60 years

in all three cases). The use of spatial loss instead of temporal loss did not provide a similar

response among rivers (Figures 3-3 and 3-4). Though somewhat similar, there is considerably

more variation in the plots of spatial loss than those of temporal loss. This is likely due to the

addition of morphology into the calculation, albeit implicitly. While rainfall drives the wet-

season flow distribution, floodplain morphology also plays a direct role in calculating loss.









In comparing the flow reduction required to produce a 15% spatial loss with the flow

reduction required to produce a 15% temporal loss, it was found that a 15% temporal loss

occurred consistently at a smaller reduction in flow (Figure 3-5). This does not specify in an

ecological sense how the two different measures relate though it does indicate that, on these three

rivers, the temporal measure is more conservative with respect to protection of natural flows.

Summary

For determination of MFLs the District has chosen the measure of temporal loss as a

measure of habitat change. This analysis requires that the relation between the target elevation at

a specific point in the river corridor and the flow at a long-tem gage is reasonably well

understood. It also requires long-term gage data. The spatial analysis requires the acquisition of

cross-sections and the development of a model to calculate the habitat loss. The resolution of the

cross-sections and the use of top-width as a proxy for inundated area limit the accuracy of the

analysis. For rivers currently being studied for MFL development improved topographical data

(i.e., LiDAR) is being collected. This will allow the development of digital elevation models and

improved spatial analysis of floodplain inundation. Comparisons of these improved methods

with the current methods should be performed. As instream flow professionals we are often

asked to protect the environment from harm. However, ultimately what is measured is change

and how much change constitutes harm must be determined. Therefore, we must know how

different measures of change relate to each other.









Table 3-1. Beginning Julian days for the Wet and Dry periods (Blocks 1 and 3) and ending date
for the Wet period at four different gage stations in the SWFWMD.


Begin Block 1 Begin Block 3 End Block 3

Alafia at Lithia 106 175 296
Hillsborough at Zephyrhills 112 176 296
Myakka at Sarasota 115 181 306
Peace at Arcadia 110 174 299
Mean 110 176 300









Table 3-2 Mean flow requirement at the USGS Alafia River near Lithia gage needed to inundate
floodplain features, and percent of flow reductions associated with no more than a
15% reduction in the number of days that the required flows are equaled or exceeded
for two benchmark periods (1940 through 1969 and 1970 through 1999).


Floodplain Feature Identified for
Development of Minimum Flows and
Levels for the Alafia River


Mean (SD)
Flow
Requirements
(m3/S)


Percent-of-
Flow
Reduction
from 1940
through 1969


Percent-of-Flow
Reduction from
1970 through
1999


Low bank elevation
Low bank elevation for inundation of
both sides of river floodplain
Highest floodplain vegetation class
Mean elevation of swamp classes
Highest swamp class
Floodplain wetted perimeter inflection
pomnt
Mean elevation of hydric soils
Highest elevation of hydric soils


33 (17)
64 (18)

101 (9)
28 (12)
42 (22)
22 (8)

29 (11)
58 (36)










Table 3-3. Mean flow requirements at the USGS Myakka River near Sarasota gage needed to
inundate floodplain features, and percent of flow reductions associated with no more
than a 15% reduction in the number of days that the required flows are equaled or
exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999).


Floodplain Feature Identified for
Development of Minimum Flows and
Levels for the Myakka River


Lowest bank elevation to inundate one
side of the river floodplain
Lowest bank elevation to inundation
both sides of river floodplain
Median elevation of oak-palm wet
hammock
Median elevation of oak-popash wet
hammock
Median elevation of popash swamp
Median elevation of paragrass marsh
Median elevation of mixed marsh
Median elevation of panicum marsh
Median elevation of mucky soils
Median elevation of hydric soils
First maj or low inflection point on
wetted perimeter
First major high inflection point on
wetted perimeter


Mean (SD)
Flow
Requirements
(m3/S)

9 (8)

16 (19)

24 (12)

13 (5)

10 (3)
9 (4)
1 (1)
18 (12)
20 (16)
18 (13)
13 (8)

23 (26)


Percent-of-
Flow
Reduction
from 1940
through 1969
15

16

13

16

20
21
72
16
15
16
17

15


Percent-of-Flow
Reduction from
1970 through
1999

8

11

8

9

15
15
68
11
10
11
11

9









Table 3-4. Mean flow requirements at the USGS Peace River at Arcadia gage needed to
inundate floodplain features, and percent of flow reductions associated with no more
than a 15% reduction in the number of days that the required flows are equaled or
exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999).


Floodplain Feature Identified for
Development of Minimum Flows and
Levels for the middle Peace River


Lowest bank elevation to inundate one
side of the river floodplain
Lowest bank elevation to inundate both
sides of river floodplain
90th Percentile elevation of river terrace
vegetation
90th Percentile elevation of wet hardwood
hammock
Mean elevation of river terrace
vegetation
Mean elevation of wet hardwood
hammock
Mean elevation of cypress swamp
Mean elevation of hardwood swamp
Mean elevation of mucky soils


Mean (SD)
Flow
Requirements
(m3/S)


Percent-of-
Flow
Reduction
from 1940
throughl969
7

6

7

7

8

7


Percent-of-Flow
Reduction from
1970 through
1999


115 (78)

158 (57)

123 (47)

197 (72)

112 (66)

160 (61)

82 (25)
83 (34)
66 (31)
























































Figure 3-1. Map of the Southwest Florida Water Management District showing the Alafia, Peace
and Myakka rivers and the USGS gage locations used in this study on each.






56






















































































500 1000 1500 2000
Flow at USGS IVyakka near Sarasota Gage (cfs)


2500


Figure 3-2. Percent-of-flow reductions that result in a 15% reduction in the number of days that
flows on the Alafia, middle Peace, and Myakka rivers are reached. Horizontal lines

represent the flow reduction standards chosen by the SWFWMD for each river.

Graphs are adapted from Kelly et al. 2005a, b, and c.


t*


~*


0 500 1000 1500 2000 2500
Flow at USGS Alafia River at Lithia gage (cfs)


3000 3500


6000 70


00


0 1000 2000 3000 4000 5000
Flow at USGS Peace River near Arcadia Gage (cfs)


-


1 30

4
~ 25
c
6
F~P 20
e!
B
~15
CE

5610



d 5
h o














100

90 i *Peace R2 =0.9358
'80 = lfa R2 =0.9611

'70
Myakka R2 =0.9215
~60

50

S40

S30






0 2 4 6 8 10 12 14
Flow (cfs/mi2)

Figure 3-3. Percent-of-flow reductions that result in a 15% reduction in the number of days that
flows on the Alafia, middle Peace, and Myakka rivers are reached. Gage flow has
been divided by the area of the watershed above the gage and power curves fit to the
data for each river.































0 2 4 6 8
Flow (cfslmi2)


10 12 14


Figure 3-4.


Percent flow reduction required to result in a 15% loss of top width plotted against
flow for the Alafia, Myakka, and middle Peace rivers.














S100
8 0
S70
60
50
40
30



0 20 40 60 80 100
Te mporal pe rce nt of flow red auction



Mlyakka

100
90 *

60
50



4 0 2 0 60 8 0







50 *C

20
10

0 20 40 60 80 100
Temporal Percent of flow reduction



Figure 35. Perc ntoflowd reductin orsodn oa1%spta os(esrdah

su mto ofcosscin o it) lotdaantth ecn fflwrdcin
corsodn toa1%tmoa os maue sals f asdrn hc h
rie hitrclyrace pcfc lwfrteAlfa yka admdl ec
nvers.


Alafia









CHAPTER 4
THE UTILIZATION OF GEOGRAPHIC INFORMATION SYSTEMS IN DETERMINING
FLOODPLAIN INUNDATION FOR MFL DEVELOPMENT

Introduction

In its response to the 1997 law requiring the development of minimum flows and levels

(MFLs) in the state of Florida, the Southwest Florida Water Management District (District) has

developed minimum flows and levels for multiple freshwater rivers (Munson et al. 2005). As

part of its MFL development, the District has acknowledged the importance of the natural flow

regime and attempted to preserve it through the use of a seasonally specific percent of flow

reduction approach (Flannery 2002). The use of seasonally specific criteria to identify

acceptable percent of flow reductions to be applied to each of a low, middle, and high flow

period, addresses both the cyclical nature of the annual flow pattern and the natural variation of

this pattern. Further, the use of multiple benchmark periods has served to recognize that longer-

term climatic cycles effect flow variation on a multidecadal basis (Kelly 2004).

By establishing a seasonally specific, block approach to the development of MFLs, the

District required the development of criteria for each block that identified an appropriate percent

of flow reduction to be used during each of the three seasonal blocks. Development of criteria to

be applied during the low flow block and the medium flow block was aided by methods currently

employed in the establishment of instream flow requirements in other areas of the state.

Specifically, the use of wetted perimeter, PHABSIM, and availability of snag habitat were

employed in the development of low and medium flow criteria (Kelly et al. a,b,c 2005).

Methods suitable for developing criteria for high season flow protection were less well

documented and few examples were available.

Floodplain wetlands associated with river systems have been recognized as an important

component of the river corridors ecologic community (Hughes and Rood 2003, Likens and









Bormann 1974). However, for many years, the importance of out-of-bank flows to the riverine

ecosystems was not recognized by instream methods used to assess flow requirements

(Middleton 1999). Methods such as the Tennant method or the use of wetted perimeter can be

serviceable in the protection of the lowest flows. However, they deal predominately with

between bank analyses and offer little to no protection for high flows, though both methods have

been widely applied (Gippel and Stewardson 1998, Jowett 1997, Postal and Ricther 2003).

While these methods offer valuable information for regulating the low end of the flow regime, it

is necessary to measure and assess the flows required at the high end of the flow regime as well.

Munson and Delfino (2007) state that quantification of potential differences in floodplain

inundation associated with differences between minimum flow requirements and natural

hydrologic regimes should be a critical component of regulatory programs that are intended to

protect or restore riverine habitat. The District has employed a method for establishing MFLs

that acknowledges the importance of seasonal flow regimes, including flows necessary for

floodplain inundation. The method employs the historic flow records and a temporal measure of

habitat loss based on the reduction in the number of days specified flows are reached under

various flow reduction scenarios. An alternative approach, employing a measure of spatial loss

that examined the change in top width at each cross section modeled under different flow

conditions, was previously explored in this dissertation (Munson and Delfino 2007).

Both spatial and temporal measures of habitat loss can be related to percent flow reduction

in a river channel. However, as previously discussed, both serve as proxies to evaluate the extent

of the river-floodplain connection. This connection has not yet been fully described on any river

studied by the District because complicated floodplain geometry is being represented by a

limited number of historic floodplain cross-sections derived from earlier studies. Though these









cross-sections represented the best available data, their limitations were recognized, especially

with regards to in-channel accuracy.

This chapter describes the role that Geographic Information Systems (GIS), coupled with

hydraulic models, can play in MFL development by providing high quality data concerning

floodplain inundation under different flow scenarios. As Overton (2005) points out, riverine

ecosystems benefit from spatial analysis since it provides temporally dynamic data in three

spatial dimensions. This coupling of a one-dimensional hydraulic model and spatial information

from GIS results in a powerful tool for relating variations in floodplain inundation with

variations (or reductions) in river flow.

Study Area

The Braden River is the largest tributary of the Manatee River located in Southwest

Florida (Figure 4-1). It drains approximately 215 km2 and is composed of three distinct

segments. The most downstream segment is an approximately 10-km long estuarine reach,

which j oins with the Manatee River near its mouth. Immediately upstream of this is another

approximately 10-km reach of river impounded by a broad-crested weir, creating the Ward Lake

reservoir. Above this is a 13-km segment of naturally flowing incised channel (DelCharco and

Lewelling 1997).

The Braden River was selected for this study by virtue of its listing on the MFL priority list

and because its size made it an efficient option for the utilization of new technology. The river

has three USGS gage locations within the study reach. These include, from upstream to

downstream, the Braden River at Lorraine (USGS# 02300029), the Braden River near Lorraine

(USGS# 023 00032), and the Braden River at Linger Lodge (USGS# 023 0003 58).

Flows in the Braden River are less substantial than other rivers in the area. In part, this is

due to the watershed, though development and attenuation in the watershed may play a role.









Flows do not exceed 100 ofs until higher than the 10% exceedance flow. This, combined with

the incised nature of the channel and well-drained sandy soils, results in the expectation that

there is little floodplain inundation except under the highest flows.

Methods

For development of an accurate digital elevation model (DEM), three-dimensional

topographic data are required. For the Braden River, these data were gathered through the use of

Light detection and Ranging (LiDAR). In 2005, a qualified photogrammetric firm conducted the

LiDAR and orthophotography acquisition for the Braden River proj ect. The flights utilized an

ALS40 LiDAR system flying at 5000 ft with a 30-degree field of view and 20% side overlap.

Acquisition of LiDAR data used a 2-m post spacing interval, digital one-foot orthophotographs

and 3D breakline features necessary to meet a one-foot contour interval product. The vertical

accuracy of the data was specified at 10 cm in homogeneous, unambiguous terrain. All LiDAR

data were collected using the North American Vertical Datum of 1988 (NAVD88).

For the Braden River, a Triangular Irregular Network (TIN) was used as the digital

elevation model. The TIN was generated from a combination of LiDAR data, break lines, and a

limited number of surveyed cross-sections (Figure 4-2). The TIN was then used to generate a

series of cross sections for export into the Hydrologic Engineers Centers-River Analysis System

(HEC-RAS; Figure 4-3). Known water surface elevations were used as downstream boundary

conditions and a rating curve, supplied by USGS, was used to calibrate to the USGS near

Lorraine gage (USGS# 02300032). All elevations, associated with USGS gages, were converted

to a NAVD88 standard when necessary. The model was considered calibrated when calculated

values of water surface elevation were within 0.5 ft of the rated value.

Once the model was complete, data from the model were imported back into the GIS

environment. Essentially, these data consist of the cross-section locations and ground elevations










from the model and water surface elevations associated with different flow profies. In the case

of the Braden River, the model generated flow profies for twenty-one evenly distributed percent

exceedance flows. For each of these flow profiles, a water surface elevation was calculated at

each cross-section and represented in GIS as a series of lines at each cross-section with different

elevation. So, each specified flow corresponds in GIS to a series of lines with elevation from

which a second DEM can be generated in the form of a TIN. This TIN becomes a TIN of

calculated water surface elevation based on the model.

Once both terrain surface elevation and water surface elevation models are generated,

analysis can be performed in a GIS environment, which clips the water surface elevation model

to the terrain model. The use of an extension to the GIS software titled GEO-RAS automates

this process, resulting in a raster surface. A raster surface is a form of digital elevation model

which uses a matrix of same size squares with each one having an elevation based on the

elevation of the data points within its bounds. Raster cells do not slope, and thus all the area

falling within a cell is assigned the same elevation. The new DEM, generated by subtracting the

land surface from the water surface, is a water surface model where each cell represents a depth

of water over the land surface (Figure 4-4). This allows for the estimation floodplain inundation

in a river corridor based on three dimensional floodplain topology rather than two-dimensional

cross-sections and intervening channel length.

Results

This study utilized a combination of LiDAR data and survey data to generate a DEM of the

freshwater portion of the Braden River. This DEM provides useful information for the

development of cross-sections in the hydraulic model. Further, the output from the model is

used to generate a water surface profie in GIS and to analyze inundation patterns.









The LiDAR points, used to generate the model, were numerous and dense in the upland

area of the digital elevation model. However, points were sparser in the forested portion of the

floodplain wetland and in the river corridor itself. A typical bend in the river and the lack of

corresponding LiDAR data within the floodplain are shown in Figure 4-5. This trend continued

for the length of the study corridor.

In the downstream half of the study area, break lines were added along each bank. This

results in an approximation of the floodplain area by defining the upland portion with LiDAR

and the river's edge with a hard break line. This is useful in the TIN because it keeps the river

channel well defined, even in the absence of high-density LiDAR data near the river. However,

the more upstream portion of the study lacked these break lines and the river channel is more

poorly confined in this portion of the DEM. Survey data also provided localized improvement of

the topography TIN and resulted in better development of the DEM in areas where survey data

augmented the LiDAR data. However, these data resulted in only localized improvements in the

topography TIN.

The hydraulic model estimates an inundated area by calculating the area of a trapezoid.

The trapezoid is bound by the water surface extent at two adj acent cross-sections. The distance

between cross-sections is defined in the model. In a GIS environment, it is possible to estimate

the inundated area between cross-sections by calculating the area where the water surface DEM

is higher then the topography DEM (Figure 4-6). The inundated area calculated by GIS and

HEC-RAS between two cross-sections is presented in Table 4-2.

Discussion

GIS to HEC-RAS

The use of LiDAR to augment survey data was initially viewed as a device for

supplementing partial floodplain survey work. The concept was that cross-sections could be









established manually across the river and then, subsequently, those cross sections could be

lengthened, and potentially, additional cross-sections could be added, which are generated from

the LiDAR data. This was not the case with the Braden River.

Cross-sectional data aligned well with LiDAR-derived data where overlap occurred. The

generated DEM was used to create cross-section lines, which were imported into HEC-RAS.

The DEM allowed surveyed cross-sections to be lengthened and extended beyond the floodplain.

However, the quality of the DEM in areas where the LiDAR data were sparse and survey data

did not exist, resulted in cross-sections which were too poorly defined within the floodplain and

river channel to be useful in a hydraulic model.

It should be noted that this was due not to inaccuracy in the data but to a localized lack of

data. Relatively few additional data points would be needed to significantly improve the

accuracy of the DEM. The acquisition of a thalwag alone would add considerable accuracy to

the DEM. It should also be noted that the LiDAR data for the Braden River corridor were

among the first collected in the SWFWMD and that specifications were far less rigorous than

more recently collected data. Preliminary results from the Withlacoochee River indicate that 4-

foot post spacing and newer systems are penetrating the floodplain canopy and providing

regularly spaced data. The data in the floodplain is less dense than the non-canopy areas flown,

but significantly improved over the Braden River data.

HEC-RAS to GIS

The utilization of output from HEC-RAS to model floodplain inundation were useful but

fell short of creating really meaningful results. The focus of future studies should be on the

generation of a high quality DEM of the terrain. This can be accomplished by supplementing the

LiDAR data with hydrographic survey data. Hydrographic surveys can be accomplished through

both manual and electronic means. A combination of sonar, survey and global positioning









satellite (GPS) methods can be used to define a thalwag and provide credible sub-surface data to

be used in conjunction with LiDAR. As noted above, LiDAR capabilities have increased

significantly in the past few years as increased demand has resulted in increased precision. It

should be noted that because the water surface elevation TIN is limited in geographic range by

the extent of the cross-sections, and because the bounding polygon, generated by the GIS

extension, is defined by the cross-sections, those cross-sections should be chosen carefully.

HEC-RAS offers approximations of inundation based on trapezoidal areas (Figure 4-6). It

should be noted that Figure 4-6 is conceptual since HEC-RAS uses channel length and not

geographic distance as the distance between cross-sections. However, inundation, in a broader,

more complex floodplain than the Braden River, would not necessarily be well modeled in this

manner without a significant number of cross-sections to capture variability in the floodplain.

The GIS and HEC-RAS estimates, for inundated area in the section of the Braden River shown

in Figure 4-6, are presented in Table 4-2.

GIS in MFL Development

It has already been noted that historically instream-flow requirements have been focused

largely on the low-flow portion of the flow regime. This is probably due in part to the

complexity of floodplain hydrology. As Mertes (1997) noted, floodplain inundation does not

rely solely on overbank flows from the associated river. GIS provides the opportunity, when

coupled with high quality data, to analyze floodplain inundation more accurately than was

previously practical. Initially, this can improve MFL development by allowing more accurate

assessment of habitat changes associated with changes in flow.

Future Work

The purpose of this research was to generate a hydraulic model to be used in the

development of a MFL for the Braden River and also to determine the usefulness of LiDAR data









and the role they can play in MFL determinations. The Braden River LiDAR data represent

approximately 26 km2 Of area. Generated as part of a larger flight plan, the data generally cost

approximately $700/ km2 to collect and process, though it can vary depending on the size and

shape of the area to be flown. Data collection has begun on the Withlacoochee River, which

includes approximately 3 10 km2 Of necessary data. Prior to an investment of that size, it was

prudent to determine to what extent the data was useful and what subsurface data are necessary

to generate high quality results.

This proj ect demonstrated that LiDAR, combined with ground surveys, can be used to

generate cross-sections for the development of a hydraulic model. However, the LiDAR is best

used to supplement floodplain portions of a cross section. LiDAR alone results in insufficient

instream data to generate an accurate cross-section. This research also demonstrated that where

LiDAR data are sufficiently augmented with high quality instream and below canopy data, GIS

can be used to predict inundation patterns and depths under different flow conditions.

LiDAR data for the Withlacoochee River will be coupled with a hydrographic survey to

produce a highly accurate DEM for the length of the study corridor. The hydrographic survey

will include a combination of electronic data collected with sonar and manually surveyed data.

A thalwag of the river will also be included. The Withlacoochee River study will be the first

maj or river in Florida to have a highly accurate DEM generated which will be used in future

studies of the river including MFL development.













Table 4-1. Percent exceedance flows for the Braden River near Lorraine gage (USGS#
02300032).


Percent Exceedance
(%)
99
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
1


Exceedance flows
(cfs)
0.67
1.02
1.40
1.73
2.04
2.47
2.87
3.35
4.09
5.12
6.22
7.45
10.25
13.59
17.95
25.16
36.36
54.41
89.53
188.82
647.47











Table 4-2. Estimated acres of inundated land at different exceedance flows between two Braden
River cross-sections shown in Figure 4-6.



Percent GIS Results HEC-RAS Results
Exceedance (acres) (acres)
90%/Ex 1.12 1.26
50%/Ex 1.14 1.26
25%/Ex 1.33 1.30
10%/Ex 1.95 1.49
1%/Ex 5.49 7.10
























































Figure 4-1. Location map of the Braden River in southwest Florida.







































Legend
SBradenXS
ti n2
Elevation
64 986 73 730
56 241 64 986
S47 497 56 241
38.752 -47.497
30.008 -38.752
S21 263 30 008
12 519 21 263
3 774 12 519
-4.970 -3.774




I I I I I Feet
0 1625 3250 6500

Figure 4-2. Portion of the digital elevation model generated of the Braden River, Florida.











































Figure 4-3. Location of surveyed cross-sections across the B~raden River, F`lorda.




































arida with modeled inundation patterns from a 1%


gure 4-4. Portion of the Braden River,
exceedance flow.





































gure 4-3. Livax points usea in tne univi or tne Ltracen Klver, rlonaa. Livax points are
yellow dots, which are numerous enough to show as dark areas on the map. Empty
areas are where data was determined not to represent the ground elevation (i.e.,
Houses and canopied areas where the LiDAR did not penetrate to the ground.


































Figure 4-6. Conceptual view of inundated area calculated by HEC-RAS and in a GIS
environment. The shaded blue area represents inundation patterns modeled in GIS.
The red trapezoid is a depiction of how HEC-RAS would estimate the inundated area.









CHAPTER 5
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK

Summary

This dissertation reviewed the water management policies in Florida, which eventually led

to a legislative mandate for MFL (minimum flows and levels) development. The Hyve water

management districts, tasked with MFL development, have responded using a variety of

methods. These methods include utilization of multi-parameter approaches, including both

hydrologic and biological components, as well as recognition of public water supply needs. As

of the publication of the 2005 MFL priority list, minimum flows or levels had been established

for 242 water bodies in four of the Hyve water management districts. The fifth had an established

priority list identifying water bodies within its boundaries for which MFLs were in the process of

being established.

The development of the minimum flow standards used to protect riverine ecosystems from

reductions in the highest part of the flow regime was the focus of this dissertation, because it is

the least understood. Water managers frequently utilize methods described in the literature for

the protection of in-channel flows. These included the use of physical habitat simulation

analysis, Eish passage, and lowest wetted perimeter inflection point in determining appropriate

flow standards. River/floodplain interaction is less well understood in terms of establishing flow

requirements for protection of the riverine floodplain ecosystem.

Two methods of measuring habitat loss in floodplain wetlands were evaluated. The use of

a temporal measure of habitat loss as a criterion to develop minimum flows for protection of

riverine floodplains was examined for three rivers in Florida. The appropriateness of the

temporal measure was compared to a spatial measure of habitat loss, and an alternative method










using GIS for measuring floodplain losses, associated with water withdrawals, was presented and

discussed.

The temporal loss of habitat measure is intended to provide protection of the riverine

ecosystems by limiting the reduction in the number of days of connection between the river and

the floodplain. Research showed that the relationship between the flow and the percent-of-flow

reductions required to generate a 15% decrease in the number of days a certain flow is reached

annually is similar for all three rivers studied. The similarity among rivers is highlighted when

flows are plotted in terms of flow per watershed area.

This research also utilized an alternative approach to determining habitat loss by

employing a spatial measure of habitat, derived from the HEC-RAS model. This method

summed the lengths of inundation along each cross-section in the hydraulic model as a proxy of

area of inundation. This was done because it was important to examine if different measures of

habitat relate differently to changes in flow. Further, if they did respond differently to changes in

flow, it was necessary to characterize which measure would be limiting in terms of flow

reduction if a 15% loss of habitat criteria was applied. The reductions in flow required to

generate a 15% loss of inundated width were calculated and the results for different flows per

watershed area for all three rivers were plotted.

Calculations were preformed to relate habitat types to specific periods of inundation.

Wetland characteristics were examined and the amount of flow reduction that would result in a

15% loss of inundation was determined. The results for each wetland characteristic among the

cross-sections of a specific river varied widely. Vegetative cross-sections were selected based on

location of shoals within the river and vegetation communities. Cross-sections were selected

which were characteristic of the river corridor. However, as it became evident that it was









difficult to characterize a riverine floodplain with a limited number of cross-sections, alternatives

to linear measures of spatial habitat were sought.

Utilizing LiDAR data, gathered in the Braden River basin, a digital terrain model of the

Braden River corridor was generated. This was done to examine how useful digital terrain

models might be in examining floodplain wetland inundation in a GIS environment, rather then

simply along linear cross-sections. The accuracy of the digital terrain model was limited by the

quality of the LiDAR data, which were among the earliest such data collected in southwest

Florida. Inaccuracies were due mainly to the lack of bare earth data, collected in the canopied

floodplain.

Cross-sections from the digital terrain model were imported into the HER-RAS hydraulic

model and water surface profiles were calculated. These profiles were then exported into GIS

where a digital elevation model of the water surface was created. The water surface map was

then layered with the terrain map and the extent of inundation at different flows was assessed.

This was done to develop a better measure of spatial habitat loss, which could ultimately lead to

a better understanding of the inundation requirements of riverine floodplains.

Conclusions

* The 1997 legislative mandate to develop minimum flows and levels in Florida has placed
the state among the leaders of forward-looking water resources protection in the United
States.

* Minimum flow requirements, for protection of riverine wetland protection, are
conceptually understood but poorly quantified.

* The measure of temporal loss among rivers in this study are highly similar due to
similarities in flow characteristics of the rivers and do not account, directly, for differences
in floodplain morphology.

* Abrupt changes in slope, when plotting spatial measures of habitat loss, are likely
associated with variations in channel and floodplain morphology, and not accounted for
with the temporal measure of habitat loss.










* The use of a temporal measure to assess habitat loss is appropriate since floodplain
characteristics, at least in part, are derived from fluvial processes.

* LiDAR is a useful tool in the development of digital terrain models.

* Digital terrain models should be used, in conjunction with hydraulic models, to estimate a
spatial measure of habitat loss.

* GIS offers the opportunity to better estimate inundation in the floodplain as a result of
over-bank flow, and thus spatial habitat loss, as a result of changes in river flow.

* The use of GIS in determining spatial loss is preferable to the use of cross-sectional data
because the three dimensional nature of the spatial data resolves some issues of
characterizing the river corridor with a limited number of cross-sections.

* Hydrology is currently the dominant factor in establishing high-flow MFLs.

* MFLs are intended to protect the aquatic biology and ecology of the river system, though
to date, biology and ecology have only been minor components in the development of
high-flow protection standards.

* Current methods of MFL development do not differentiate flow needs based on wetland
habitat type.

* Better modeling ability, derived from GIS, should allow improved understanding of
floodplain community structure relative to flow.

Future Work

This dissertation has shown that the MFL requirements for riverine floodplains wetlands

are poorly understood. The conceptual importance of over-bank flows has been well

documented but not well quantified. This research pointed out that habitat and change in habitat

can be measured in multiple ways and it explored the use of new technology in measuring habitat

extent. These measures suggest future studies that should be conducted, given the information

presented in this dissertation.

Spatial loss in river floodplains is inconsistent from river to river while temporal loss is

similar among the three rivers studied and both should be considered in MFL development.

Thus, both spatial and temporal measures should be examined in future MFL determinations.









Flow reductions, on rivers studied in this dissertation, consistently resulted in higher

habitat loss when measured by the temporal measure of habitat loss rather than the spatial

measure of habitat loss. It will be important to determine if this is a consistent finding for other

rivers in the area.

If this relationship proves to be inconsistent, then both the spatial and temporal methods

should be applied to each river and the more restrictive one utilized. This will assure that no

more than a 15% loss of habitat would be allowed due to water withdrawals, whether habitat is

measured spatially or temporally. It is recommended that in the future, the flow reductions from

the temporal analysis be evaluated in GIS to determine the corresponding spatial loss of habitat.

If the resulting spatial loss is calculated to exceed the temporal loss, consideration should be

given to this loss when determining the MFL.

Use of GIS as a spatial modeling tool provides the means for better understanding of the

river floodplain interaction. Inundation maps combined with vegetation maps (i.e., National

Wetlands Inventory Maps) may provide a better understanding of the inundation requirements of

various wetland types and these relations should be investigated.

It is necessary to measure not only the accuracy of the MFL, but also to evaluate the

accuracy of some of the underlying assumptions, and thus the effectiveness of the MFLs. This

requires the implementation of biological monitoring and a long-term commitment to evaluate

the effectiveness of MFLs. This should begin with the monitoring of the wetland extent at

various locations in each river system. This could be spatially mapped. Changes over time

could be noted and correlated to changes in inundation as a result of changes in flow.

The LiDAR data, used to generate the Braden River digital elevation model, had gaps in

the canopied area. Current LiDAR technology has improved and these improved data should be










coupled with ground surveys to supplement areas of sparse LiDAR data. This combination of

remote and ground surveys can result in a dense enough network of points to generate an

accurate and detailed digital elevation model of the river/floodplain system. These techniques

should be pursued and incorporated into future MFL determinations.










LIST OF REFERENCES


Ainsle, W.B., B.A. Pruitt, R.D. Smith, T.H. Roberts, E.J. Sparks and M. Miller. 1999. A
regional guidebook for assessing the functions of low gradient riverine wetlands in western
Kentucky. U.S. Army Corps of Engineers Waterways Experiment Station. Technical Report
WRP-DE-17, Vicksburg, Mississippi.

Anderson, D.L. and P.C. Rosendahl 1998. Development and management of land/water
resources: the Everglades, agriculture, and south Florida. Journal of the American Water
Resources Association 34: 235-248 pp.

Bachmann, R. W., M. V. Hoyer and D. E. Canfield, Jr. 1999. The restoration of lake Apopka in
relation to alternative stable states. Hydrobiologia 394:219-232.

Berryman and Henigar. 2004. Characterization of wetland vegetation communities and hydric
soils along the middle Peace River. Prepared for the Southwest Florida Water Management
District. Brooksville, FL.

Blake, N. M. 1980. Land into water-water into land. University Press of Florida, Tallahassee, FL.
344 pp.

Bovee, K.D., B.L. Lamb, J.M. Bartholow, C.B. Stalnaker, J. Taylor and J. Hendrickson. 1998.
Stream habitat analysis using the instream flow incremental methodology. U.S. Geological
Survey, Biological Resources Division Information and Technology Report USGS/BRD-1998-
0004.

Brinson, M.M., B.L. Swift, R.C. Plantico and J.S. Barclay. 1981. Riparian ecosystems: their
ecology and status. U.S. Fish and Wildlife Service, Biological Services Program Report
FWS/OBS-81/17, Washington, D.C.

Canfield, D. E., Jr., R. W. Bachmann and M. V. Hoyer, 2000. A management alternative for
Lake Apopka. Lake and Reservoir Management 16: 205-221.

Carriker, R. R., 2000. Florida's water: supply, use, and public policy, Institute of Food and
Agricultural Science, University of Florida, Gainesville, FL.

Conner, W.H. and J.W. Day. 1976. Productivity and composition of a bald cypress-water tupelo
site and a bottomland hardwood site in a Louisiana swamp. American Journal of Botany 63:
1354-1364.

Crance, J.H. 1988. Relationships between palustrine forested wetlands of forested riparian
floodplains and fishery resources: a review. U.S. Fish and Wildlife Service, Biological Report
88(32), Washington, D.C.

DelCharco, M. J., and B. R. Lewelling. 1997. Hydrologic description of the Braden River
Watershed, West-Central Florida. United States Geological Survey Open Report 96-634.










DEP (Department of Environmental Protection). 1995. Florida water plan 1995. Florida
Department of Environmental Protection, Office of Water Policy, Tallahassee, FL.

DEP (Department of Environmental Protection). 2000. Florida's water supply: will there be
enough water in 2020, DEP's annual status report on regional water supply planning, Florida
Department of Environmental Protection, Office of Water Policy, Tallahassee, FL. 19 pp.

DEP (Department of Environmental Protection). 2002a. Implementing regional water supply
plans: is progress being made, DEP's annual status report on regional water supply planning.
Florida Department of Environmental Protection, Tallahassee, FL. 25 pp.

DEP (Department of Environmental Protection). 2002b. Florida water conservation initiative.
Florida Department of Environmental Protection, Tallahassee, FL.

DEP (Department of Environmental Protection). 2003a. 2003, Annual status report on regional
water supply planning. Florida Department of Environmental Protection, Tallahassee, FL.

DEP (Department of Environmental Protection). 2003b. 2002 Reuse inventory. Florida
Department of Environmental Protection, Tallahassee, FL.

DEP (Department of Environmental Protection). 2003c. 2002 Florida water plan; 2002 Annual
Progress Report, Florida Department of Environmental Protection, Tallahassee, FL.

Derr, M. 1989. Some kind of paradise. William Morrow and Co., New York, NY. 416 pp.

Douglas, M. S. 1947. The Everglades; river of grass. Rinehart, New York, NY. 487 pp.

Dunbar, M.J., A. Gustard, M.C. Acreman and C.R. Elliott. 1998. Overseas approaches to setting
river flow objectives. Institute of Hydrology. R&D Technical Report W6-161. Oxon, England.
83 pp.

Egozcue, D. K. 2001. The history of water in the Tampa Bay area. Master's Thesis, Department
of Geography, University of South Florida, Tampa, FL. 126 pp.

Enfield, D., A. Mestas-Nunez and P. Trimble. 2001. The atlantic multidecadal oscillation and
its relation to rainfall and river flows in the continental U.S. Geophysical Research Letters
28(10): 2077-2080.

Flannery, M.S., E. Peebles and R. T. Montgomery. 2002. A percent-of-flow approach for
managing reductions of freshwater inflows from unimpounded rivers to southwest Florida
esturies. Estuaries 25: 1318-1332.

Gippel, C. J. and M. J. Stewardson. 1998. Use of wetted perimeter in defining minimum
environmental flows. Regulated Rivers 14: 53-67










Gore, J. A., C. Dahm and C. Klimas. 2002. A review of "Upper Peace River: an analysis of
minimum flows and levels". Prepared for the Southwest Florida Water Management District.
Brooksville, FL. http://www. swfwmd. state. fl.us/documents/

Gregory, S.V., F.J. Swanson, W.A. McKee and K.W. Cummins. 1991. An ecosystem
perspective on riparian zones. Bioscience 41: 540-551.

Hamann, R. 2001. Consumptive use permitting criteria. In: Florida environmental and land use
law, The Florida Bar, Tallahassee, FL.

Hammett, K.M., J.F. Turner, Jr. and W.R. Murphy, Jr. 1978. Magnitude and frequency of
flooding on the Myakka River, Southwest Florida. Department of the Interior. United States
Geological Survey. Water-Resources Investigations, Open-File Report 78-65, Tallahassee, FL.

Henry, J. A. 1998. Weather and climate. Pp. 16-37 in Water resources atlas of Florida. E. A.
Fernald and E. D. Purdum (eds.). Institute of Science and Public Affairs, Florida State
University, Tallahassee, FL.

Hill, J. E. and C.E. Cichra. 2002. Minimum flows and levels criteria development. Evaluation
of the importance of water depth and frequency of water levels / flows on fish population
dynamics, literature review and summary: the effects of water levels on fish populations.
Institute of Food and Agricultural Sciences, Department of Fisheries and Aquatic Sciences.
University of Florida. Gainesville, FL. 40 pp.

Hill, M.T., W.S. Platts and R.L. Beschta. 1991. Ecological and geological concepts for instream
and out-of-channel flow requirements. Rivers 2: 198-210.

Hughes, F. M. R. and S. B. Rood. 2003. Allocation of river flows for restoration of floodplain
forest ecosystems: A review of approaches and their applicability in Europe. Environmental
Management 32(1): 12-33

Hupalo, R., C. Neubauer, L. Keenan, D. Clapp and E. Lowe. 1994. Establishment of minimum
flows and levels for the Wekiva River system. Technical Publication SJ94-1. St. Johns River
Water Management District, Palatka, FL.

Jowett, I.G. 1993. Minimum flow requirements for instream habitat in Welllington rivers. NZ
Freshwater Miscellaneous Report No. 63. National Institute of Water and Atmospheric Research,
Christchurch, New Zealand. 33 pp.

Jowett, I.G. 1997. Instream flow methods: A comparison of approaches. Regulated Rivers 13:
115-127

Junk, W. P., P.B. Bayley and R.E. Sparks. 1989. The flood pulse concept in river-floodplain
systems. Proceedings of the international large river symposium. Special Publication of the
Canadian Journal of Fisheries and Aquatic Sciences 106: 1 10-127










Kallina, E. F. 1993. Claude Kirk and the politics of confrontation. University Press of Florida,
Gainesville, FL. 253 pp.

Kelly, M.H. 2004. Florida river flow patterns and the atlantic multidecadal oscillation. Ecologic
Evaluation Section. Southwest Florida Water Management District. Brooksville, FL. 80 pp. +
appendix http://www.swfwmd. state.fl. us/documents/

Kelly, M. H., A. B. Munson, J. Morales and D. L. Leeper. 2005a. Alafia River minimum flows
and levels; freshwater segment including Lithia and Buckhorn Springs. Ecologic Evaluation
Section. Southwest Florida Water Management District, Brooksville, FL. 188 pp + appendix
http://www. swfwmd. state. fl.us/documents/

Kelly, M. H., A. B. Munson, J. Morales and D. L. Leeper. 2005b. Proposed minimum flows and
levels for the upper segment of the Myakka River, from Myakka City to SR 72. Ecologic
Evaluation Section. Southwest Florida Water Management District, Brooksville, FL. 144 pp +
appendix http://www.swfwmd. state. fl.us/documents/

Kelly, M. H., A. B. Munson, J. Morales and D. L. Leeper. 2005c. Proposed minimum flows for
the middle segment of the Peace River, from Zolfo Springs to Arcadia. Ecologic Evaluation
Section. Southwest Florida Water Management District, Brooksville, FL. 177 pp + appendix
http://www. swfwmd. state. fl.us/documents/

Kuensler, E.J. 1989. Values of forested wetlands as filters for sediments and nutrients. Pp. 85-
96 In: D.D. Hook and R. Lea (eds.), Proceedings of the Symposium: the forested wetlands of the
United States. USDA Forest Service, Southeastern Forest Experimental Station, General
Technical Report SE-50.

Likens, G. E. and F. H. Bormann. 1974. Linkages between terrestrial and aquatic ecosystems.
Bioscience 24(8): 447-456.

Lowe, E. F., L. E. Battoe, M. Coveney and D. Stites. 1999. Setting water quality goals for
restoration of Lake Apopka: inferring past conditions. Lake and Reservoir Management 15: 103-
120.

Maloney, F. E., S. J. Plager and F. N. Baldwin. 1968. Water law and administration: the Florida
experience. University of Florida Press, Gainesville, FL. 488 pp.

Maloney, F. E., R. C. Ausness and J. S. Morris. 1972. A Model Water Code. University of
Florida Press, Gainesville, FL. 373 pp.

Maloney, F. E., S. J. Plager, R. C. Ausness and B. D. E. Canter. 1980. Florida water law. Water
Resources Research Center, University of Florida, Gainesville, FL. 762 pp.

Matthews, F. E. and G. E. Nieto. 1998. Florida water policy: a twenty-five year mid-course
correction. Florida State University Law Review 25: 365-390.










McCabe, G. and D. Wolock. 2002. A step increase in streamflow in the conterminous United
States. Geophysical Research Letters 29: 2185-2188.

Mertes, L. A. K. 1997. Documentation and significance of the perirheic zone on inundated
floodplains. Water Resources Research 33(7): 1749-1762.

Middleton, B. 1999. Wetland restoration: flood pulsing and disturbance dynamics. John Wiley
and Sons, Inc. New York, NY. 388 pp.

Munson, A. B., and J. J. Delfino. 2007. Minimum wet-season flows and levels in southwest
Florida rivers. Journal of the American Water Resources Association. Accepted for publication.

Munson, A. B., J. J. Delfino and D. A. Leeper. 2005. Determining minimum levels: the Florida
experience. Journal of the American Water Resources Association. 41(1):1-10.

Murphy, W. R., K. M. Hammett and C. V. Reeter. 1978. Flood profiles for Peace River, south-
central Florida. U.S. Geological Survey, Water resource Investigation 78-57, Tallahassee, FL.

Olsen, J. R., J. R. Stedinger, N. C. Matalas and E. Z. Stakhiv. 1999. Climate variability and
flood frequency estimation for the upper Mississippi and lower Missouri rivers. Journal of the
American Water Resources Association. 35 (6):1509-1523.

I. C. Overton. 2005. Modeling floodplain inundation on a regulated river: integrating GIS,
remote sensing and hydrological models. Regulated Rivers 21: 991-1001.

Poff, N. L., D. Allan, M.B. Bain, J.R. Darr, K.L. Perstegaard, B.D. Richter, R.E. Sparks and J.C.
Stromberg. 1997. The natural flow regime; A paradigm for river conservation and restoration.
Bioscience 47: 769-784.

Postel, S. and B. D. Richter. 2003. Rivers for life; managing water for people and nature. Island
Press, Washington D. C. 253 pp.

Purdum, E. D., L. C. Burney and T. M. Swihart. 1998. History of water management, In: Water
Resources Atlas of Florida, E. A. Fernald and E. D. Purdum (Editors). Institute of Science and
Public Affairs, Florida State University, Tallahassee, FL.

Richter, B.D., J.V. Baumgartner, J. Powell and D.P. Braun. 1996. A method for assessing
hydrologic alteration within ecosystems. Conservation Biology 10: 1163-1174.

Richter, B.D., J.V. Baumgartner, R. Wigington and D.P Braun. 1997. How much water does a
river need? Freshwater Biology 37: 231-249.

Rosenau, J. C., G. L. Faulkner, C. W. Hendry Jr. and R. W. Hull. 1947. Springs of Florida.
Florida Geologic Survey, Tallahassee, FL.

Ross, C. I. 2001. Minimum flows and levels. In: Florida environmental and land use law, The










Florida Bar, Tallahassee, FL.


Shafer, M. D., R. E. Dickinson, J. P. Heaney and W. C. Huber. 1986. Gazetteer of Florida
lakes. Water Resources Research Center, University of Florida and United States Geological
Survey, Gainesville, FL. 256 pp.

Shofner, J. H. 1982. History of Apopka and northwest Orange County, Florida. Apopka
Historical Society, Apopka, FL. 357 pp.

Stalnaker, C.B. 1990. Minimum flow is a myth. Pp. 31-33 In: M.B. Bain (ed.), Ecology and
assessment of warmwater streams: workshop synopsis. Biological Report 90(5). U.S. Fish and
Wildlife Service, Washington, D.C.

Stalnaker, C., B.L. Lamb, J. Henriksen, K. Bovee and J. Bartholow. 1995. The instream flow
incremental methodology: a primer for IFIM. Biological Report 29. U.S. Department of the
Interior, National Biological Service, Washington, D.C. 46 pp.

Stanturf, J.A. and S.H. Schoenholtz. 1998. Soils and landforms in southern forested wetlands
In: Ecology and management, M.G. Messina and W.H. Conner (eds.) CRC Press-Lewis
Publishers, Boca Raton, FL.

SWFWMD (Southwest Florida Water Management District). 1996. Lakes level program annual
report. Southwest Florida Water Management District, Brooksville, FL.

SWFWMD (Southwest Florida Water Management District). 2001. Regional water supply plan.
Southwest Florida Water Management District, Brooksville, FL.

SWFWMD (Southwest Florida Water Management District). 2002. Upper Peace River: an
analysis of minimum flows and levels. Southwest Florida Water Management District,
Brooksville, FL.

Texas Parks and Wildlife. 2005. Freshwater inflow needs of the Matagorda Bay system.
http://www.tpwd. state .tx. us/l andwater/water/con servati on/coastal/fre shwater/m atagorda/
Accessed on October 12, 2005.

Tjoflat, M. P. and I. K. Quincey. 2001. Florida water planning In: Florida environmental and
land use law, The Florida Bar, Tallahassee, FL.

U.S. Army Corps of Engineers. 2001. HEC-RAS river analysis system user's manual. US Army
Corps of Engineers, Davis, CA.

U.S. Congress. 1911. Everglades of Florida. 62nd Congress. Doc. 89, Washington, D.C.

Vannote, R.L., G.W. Minshall and K.W. Cummins. 1980. The river continuum concept.
Canadian Journal of Fisheries and Aquatic Sciences 37: 130-137.










Walbridge, M.R. and B.G. Lockaby. 1994. Effect of forest management of biogeochemical
functions in southern forested wetlands. Wetlands 11: 417-439.

Wharton, C.H., W.M. Kitchens, E.C. Pendleton and T.W. Sipe. 1982. The ecology of
bottomland hardwood swamps of the southeast: a community profile. U. S. Fish and Wildlife
Service FWX/OBS-81/37, 133 pp.

Woltemade, C. J. 1997. Water level management opportunities for ecological benefit, Pool 5
Mississippi River. Journal of the American Water Resources Association 33(2): 443-454.









BIOGRAPHICAL SKETCH

Born and raised in southwest Florida, Adam Munson has spent his academic and

professional career working to protect local water bodies. A graduate of Vivian Gaither High

School in north Tampa, Adam attended the University of Florida as an undergraduate maj oring

in mechanical engineering. Graduating with a B.S.M.E. in 1994, Adam's interest in natural

systems and science led him to the Department of Fisheries and Aquatic Sciences at the

University of Florida, from which he graduated in May of 1999 with a M. S. specializing in

limnology. After a brief period as ecosystem manager for the St. Marks and Ochlockonee River

basins in North Florida, Adam returned to the southwest Florida area where he evaluates

instream flow requirements for the development of minimum flows and levels.




Full Text

PAGE 1

1 PROTECTION OF FLOODPLAIN WETLANDS ASSOCIATED WITH MINIMUM FLOW AND LEVEL DEVELOPMENT IN SOUTHWEST FLORIDA By ADAM BURTON MUNSON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

2 Copyright 2006 by Adam Munson

PAGE 3

3 ACKNOWLEDGMENTS Professionally, I wish to express my thanks to Dr. Joseph Delfino, Dr. Charles Cichra, Dr. Douglas Shaw and Dr. Warren Viessman for their tireless review and thei r continued belief in my abilities. I thank the staff of the Ecologic Evaluation Section at the Southwest Florida Water Management District for their efforts in develo ping minimum flows and levels and their constant vigilance for new technologies and methods. Sp ecifically, I would like to thank Marty Kelly, Doug Leeper, and Jonathan Morales, who have a ll contributed to the mi nimum flows and levels dialogue in Florida and never suga rcoat criticism. I also appreci ate the support of the Southwest Florida Water Management District and acknowledge that the work contained in this dissertation is academic research and does not necessarily refl ect the policies of the Southwest Florida Water Management District. Personally, I would like to thank my wife who reminds me that ambition without persistence remains unfulfilled. I also thank my parents who encouraged me to pursue whatever studies made me happy.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........6 LIST OF FIGURES................................................................................................................ .........7 ABSTRACT....................................................................................................................... ..............8 CHAPTER 1 INTRODUCTION................................................................................................................. .10 2 DETERMINING MINIMUM FLOWS A ND LEVELS IN FLORIDA: A LEGISLATIVE MANDATE..................................................................................................15 Introduction................................................................................................................... ..........15 History of Water Management in Florida (1824-1972)..........................................................16 Water Resources Act of 1972.................................................................................................20 Water Management in Florida; Quality not Quantity (1972-1997)........................................22 The 1997 Law and Water Quantity........................................................................................26 Response to the 1997 Law......................................................................................................3 0 Summary........................................................................................................................ .........33 Conclusions.................................................................................................................... .........34 3 CURRENT TECHNIQUES UTILIZED FOR THE QUANTIFICATION OF FLOODPLAIN HABITAT LOSS ASSO CIATED WITH REDUCTIONS IN WETSEASON FLOW............................................................................................................36 Introduction................................................................................................................... ..........36 Methodology.................................................................................................................... .......40 Results........................................................................................................................ .............45 Discussion..................................................................................................................... ..........48 Summary........................................................................................................................ .........51 4 THE UTILIZATION OF GEOGRAPH IC INFORMATION SYSTEMS IN DETERMINING FLOODPLAIN INUNDATION FOR MFL DEVELOPMENT................61 Introduction................................................................................................................... ..........61 Study Area..................................................................................................................... .........63 Methods........................................................................................................................ ..........64 Results........................................................................................................................ .............65 Discussion..................................................................................................................... ..........66 GIS to HEC-RAS............................................................................................................66 HEC-RAS to GIS............................................................................................................67

PAGE 5

5 GIS in MFL Development...............................................................................................68 Future Work.................................................................................................................... .68 5 SUMMARY, CONCLUSIONS AND RE COMMENDATIONS FOR FUTURE WORK........................................................................................................................... .........78 Summary........................................................................................................................ .........78 Conclusions.................................................................................................................... .........80 Future Work.................................................................................................................... ........81 LIST OF REFERENCES............................................................................................................. ..84 BIOGRAPHICAL SKETCH.........................................................................................................91

PAGE 6

6 LIST OF TABLES Table page 3-1 Beginning Julian days for the Wet and Dry periods (Blocks 1 and 3) and ending date for the Wet period at four different gage stations in the SWFWMD.................................52 3-2 Mean flow requirement at the USGS Alaf ia River near Lithia gage to inundate floodplain features, and percen t of flow reductions associ ated with no more than a 15% reduction in the number of days that the required flows are equaled or exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999)........................53 3-3 Mean flow requirements at the USGS Mya kka River near Sarasota gage needed to inundate floodplain features, and percent of flow reductions associated with no more than a 15% reduction in the number of days that the required flows are equaled or exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999)........54 3-4 Mean flow requirements at the USGS Peace River at Arcadia gage needed to inundate floodplain features, and percent of flow reductions associated with no more than a 15% reduction in the number of days that the required flows are equaled or exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999)........55 4-1 Percent exceedance flows for the Br aden River near Lorraine gage.................................70 4-2 Estimated acres of inundated land at different exceedance flows.....................................71

PAGE 7

7 LIST OF FIGURES Figure page 2-1 Map of the State of Florida showing the three early water management districts and the five current water management districts......................................................................35 3-1 Map of the Southwest Florida Water Mana gement District show ing the Alafia, Peace and Myakka rivers and the USGS gage locations used in this study on each...................56 3-2 Percent-of-flow reductions th at result in a 15% reduction in the number of days that flows on the Alafia, middle Peace and Myakka rivers are achieved. Horizontal lines represent the flow reduction standards chos en by the SWFWMD for each river. Graphs are adapted from Ke lly et al. 2005a, b, and c........................................................57 3-3 Percent-of-flow reductions th at result in a 15% reduction in the number of days that flows on the Alafia, middle Peace and Myakka rivers are reached. Gage flow has been divided by the area of the watershed above the gage and pow er curves fit to each rivers data............................................................................................................... ...58 3-4 Percent flow reduction required to result in a 15% loss of top width plotted against flow for the Alafia, Myakka, and middle Peace Rivers.....................................................59 3-5 Percent of flow reduction corresponding to a 15% spatial loss (measured as the summation of cross-section top width) plot ted against the percen t of flow reduction corresponding to a 15% temporal loss (meas ured as a loss of days during which the river historically reached a specific flow), for the Alafia, Myakka, and middle Peace Rivers......................................................................................................................... ........60 4-1 Location map of the Braden River in southwest Florida...................................................72 4-2 Location of surveyed cross-se ctions across the Braden River...........................................74 4-3 Portion of the digital elevation m odel generated of the Braden River..............................73 4-4 Portion of the Braden River with modele d inundation patterns fr om a 1% exceedance flow........................................................................................................................... .........75 4-5 LiDAR points used in the DEM of the Braden River........................................................76 4-6 Conceptual view of inundated area calculated by HEC-RAS and in a GIS environment. The Shaded Blue area represen ts Modeled inundation patterns in GIS. The Red trapezoid is a depiction of how HEC-RAS would estimate the inundated area........................................................................................................................... ..........77

PAGE 8

8 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PROTECTION OF FLOODPLAIN WETLANDS ASSOCIATED WITH MINIMUM FLOW AND LEVEL DEVELOPMENT IN SOUTHWEST FLORIDA By Adam Burton Munson December 2006 Chair: Joseph Delfino Major Department: Environmental Engineering Sciences Floodplain wetlands, associated wi th river systems, have been recognized as important components of the ecologic community of river corridors. However, the importance of out-ofbank flows to the riverine ecosystems is often no t recognized by instream methods used to assess flow requirements. Quantification of potentia l differences in floodplai n inundation associated with differences between mini mum flow requirements and natu ral hydrologic regimes should be a critical component of regulatory programs that are intended to protect or restore riverine habitat. The Southwest Florida Water Management District uses the hist oric flow records to produce a temporal measure of habitat loss wh en developing minimum flow requirements for floodplain wetland protection. In th is dissertation an alternativ e approach, employing a measure of spatial loss by examining the change in top width at each cross section modeled under different flow conditions, is compared to the temporal measure Both spatial and temporal measures of habitat loss can be related to percent flow reduction in a river channel. However, both serve as a pr oxy to evaluate the exte nt of the river-floodplain connection. This connection has no t yet been fully described on a ny river studied by the District because complicated floodplain geometry is presently represented by a limited number of floodplain cross-sections.

PAGE 9

9 This dissertation concludes with an analysis of the role Geographic Information Systems (GIS) coupled with hydraulic models can pl ay in MFL developm ent by providing data concerning floodplain inundation unde r different flow scenarios. Riverine ecosystems benefit from spatial analysis since it provides data in three temporally dynamic spatial dimensions. The coupling of a one-dimensional hydraulic model a nd spatial information from GIS results in a powerful tool for relating variati ons in floodplain inundation with va riations in river flow. When developing minimum flows, topographic data, in a geographic information system, can provide detailed information about riverine floodplain connection and lead to better understanding of river floodplain interaction.

PAGE 10

10 CHAPTER 1 INTRODUCTION For many years, surface water in Florida was drained to support development. Channelization of rivers, development of canals, and draining of wetlands were the foundation of early water management in Florida (Anderson and Rosendahl 1998, Egozcue 2001, Purdum et al. 1998). In more recent years, ground water use has increased as water demands have grown. Under these conditions, the importance of natural systems and their linkage to the economy of Florida and its potable water supply have become more evident. In more recent times, the water related resources of the state have been offered some prot ection through the requirement that minimum flows and levels (MFLs) be established. The 1972 Florida Water Resources Act (Chapter 373, Florida Statutes (F.S.)) represents the initial legislation calling fo r development of MFLs in the state of Florida. Aimed at protecting the ecology of natura l systems from "significant ha rm" associated with water withdrawals, MFLs represent an attempt to quantify the hydrologic requi rement of aquifers, rivers, lakes, and wetlands. Initially, MFL development was somewhat sparse because of uncertainties in the legislative language, among other factors. It was not until 1997, when the Florida legislature enacted severa l major changes to Florida water law in an attempt to improve water resource planning and protection, that MFL development became a key component in water resource regulation in the state. The five water management districts of Florida responded to the changes in the law by prioritizing wate r bodies within their boundaries and developing MFLs for water bodies as identified on each priority list. A number of rivers in the Southwest Florida Water Mana gement District (SWFWMD) were considered potentially impacted by withdraw als and thus to be high value candidates for MFL development and as such were placed on the priority list. The Di strict subsequently

PAGE 11

11 developed a multi-parameter approach to determining MFLs on the freshwater portions of these rivers that included evaluation of loss of floodplai n habitat. This action was concurrent with the development of approaches by other water mana gement districts in the state. Although floodplains have long been recognized as seas onally important riverine habitat, floodplain inundation has historically not been addressed in most minimum fl ow determinations (Postal and Richter 2003). Middleton (1999) also points out th at regulation of water regimes has not always fully appreciated the interaction between floodplai ns and rivers via the flood pulse. Regulation of river flows can, however, resu lt in decreased stage fluctuati ons and alteration of inundation patterns of floodplain wetlands (Poff et al. 1997, Woltemade 1997). Quantification of potential differences in floodplain inunda tion associated with differe nces between minimum flow requirements and natural hydrologic regimes shoul d, therefore, be a critical component of regulatory programs that are intended to protect or restore riverine habitat. Compared to instream evaluations of mi nimum flows requirements, there has been relatively little research on ri ver flows necessary for meeting the requirements of floodplain species, communities or functions. However, pe riodic inundation of riparian floodplains by high flows is closely linked with the overall biologi cal productivity of rive r ecosystems (Crance 1988, Junk et al. 1989). Further evidence exists sugge sting that floodplain i nundation benefits fish (Ainsle et al. 1999, Hill and Cichra 2002, Wharton et al. 1982), supports high rates of primary production (Brinson et al. 1981, Conne r and Day 1979), is critical to food webs (Gregory et al. 1991, Vannote et al. 1980) and results in development of wetland soils that are important to the overall function of the river ecosystem (Kuensler 1989, Stanturf and Schoenholtz 1998, Walbridge and Lockaby 1994, Wharton et al. 1982).

PAGE 12

12 The Southwest Florida Water Management Distri ct (District) has developed an approach which recognizes that fundamental to the develo pment of MFLs is the realization that a flow regime is necessary to protect the ecology of the river system (Hill et al. 1991, Hupalo et al. 1994, Richter et al. 1996, Stalnaker 1990). The hi gh flow component of the flow regime is intended to protect the connec tion between the floodplain and the river. The resource management goal identified for the seasonally pred ictable wet period is maintenance of seasonal hydrologic connections between the river channe l and floodplain to ensure floodplain structure and function. Minimum flows developed for the high flow season are intended to protect ecological resources and values associated with floodplains by maintaining hydrologic connections between the river channel and floodplain and maintaining the natural vari ability of the flow regime. These goals are quantified thr ough use of the Hydrologic Engin eering Centers-River Analysis System (HEC-RAS) and the alteration of long-te rm flow records to evaluate floodplain feature inundation patterns associated with channel-floodplain connectivity. A temporal measure of habitat loss has b een employed by the Dist rict in establishing MFLs on freshwater river systems in southwest Fl orida, during the high flow season of the year. This reduction in number of days of inundation is a temporal measure of loss based on how many days, annually, a specified water-surface el evation was reached compared to how often that elevation would have been reached if flow conditions had been reduced by a given amount. Data and methods used to develop minimum fl ows for seasonal, high-flow periods for the Alafia River, Myakka River, and Peace River are us ed in this study to illust rate this approach to quantifying habitat change associ ated with temporal difference in floodplain inundation patterns. An alternative approach, employing a measure of spatial loss by examining the change in top

PAGE 13

13 width at each cross section in a hydraulic model unde r different flow conditions is also explored. The use of a temporal measure of habitat lo ss for establishing MFLs during high flows is compared to the spatial loss of habitat for the same flow reducti on. The applicability of both is discussed and the results compared to determine wh ich is more restrictive in terms of allowable flow reduction. However, as Mertes (1997) points out, fl oodplain vegetation development and persistence may not, however, necessarily depend wholly on inundation from the river channel. Groundwater seepage, hyporheic in puts, discharge from local trib utaries, and precipitation can also lead to floodplain inundation. Recent work on the upper segment of the Peace River and the Alafia River in central Florida suggests that direct and continuous inundation of floodplain wetlands by river flows is insufficient to acc ount for inundation needs of the dominant species found in the wetlands (SWFWMD 2002). After comparing the temporal and spatial meas ures of habitat loss, being employed in the minimum flow and level process, this dissertatio n examines an alternative means to measuring habitat. The use of high-density remotely gathered data to generate a digital elevation model is explored. This terrain model is then used in the development of cross-sectional data for use in a one-dimensional hydraulic model. Beyond this, data generated in the hydraulic model can be processed in a Geographic Information System (GIS) environment to produce inundation maps which are not limited to the one dimensional cro ss-sections. The coupling of a one-dimensional hydraulic model and spatial information from GIS results in a powerful tool for relating variations in floodplain inundation with va riations (or reductions) in river flow. The objectives of this disserta tion include reviewing the hist ory of water management in Florida with an emphasis on decisions culmin ating in the 1997 legislation requiring the

PAGE 14

14 development of minimum flows and levels. Atte ntion is paid to water quantity issues and policies and the natural and anthr opogenic factors influencing them The measures of riverine floodplain habitat loss use in response to th e minimum flows and levels mandate are also examined and temporal loss measures are compared with spatial loss measures. Finally, a method of examining inundation changes in a GIS system environment is presented as an alternative to current approaches.

PAGE 15

15 CHAPTER 2 DETERMINING MINIMUM FLOWS AND LEV ELS IN FLORIDA: A LEGISLATIVE MANDATE Introduction On average, Florida receives 140 cm of rainfall annually (Car riker 2000, Henry 1998). This precipitation feeds FloridaÂ’s diverse hydrol ogic systems and makes it the second wettest state in the United States. Florida has more than 7,700 lakes, ranging from 0.4 ha to over 180,000 ha (Shafer et al. 1986). Thirteen major ri vers originate in, or flow through, the state with a combined mean annua l discharge of nearly 1700 m3s-1 (Henry 1998). Contributing to this total are 27 first magnitude springs that discharge a total average flow of 272 m3s-1 or more than 22 billion liters day-1 (Rosenau et al., 1947). Florida also contains significant areas of wetlands, which at one time covered 54 percent of the state and still comprise about 30 percent of the total land area (Henry 1998, SWFWMD 2001). In a state with such abundant water res ources, stringent water supply management may seem unnecessary. This was the presumption fo r many years, as surface water systems were drained to support development. Channelization of rivers, development of canals, and draining of wetlands were cornerstones of early water management in Florida (Anderson and Rosendahl 1998, Egozcue 2001, Purdum et al. 1998). In more recent years, ground water use has increased as water demands have continued to grow. U nder these conditions, the importance of natural systems and their linkage to the economy of Fl orida and its potable wa ter supply has become more evident. The 1972 Florida Water Resources Act (Chapter 373, Florida Statutes (F.S.)) represented the initial legislation calling for development for minimum flow and levels (MFLs) in the State. Aimed at protecting the ecology of natural syst ems from "significant harm" associated with water withdrawals, MFLs represent an atte mpt to quantify the hydrologic requirement of

PAGE 16

16 aquifers, rivers, lakes, and wetlands. Though so me water management agencies did proceed with the development of regulatory flows and leve ls, it was not until 1997 that legislation was enacted which propelled MFLs from a stand-alone statute to an integral part of water resource management and regulation. The objectives of this chapter include providi ng a brief review of the history of water management in Florida with an emphasis on decisions culminating in the 1997 legislation requiring the development of mini mum flows and levels. Attenti on is paid to water quantity issues and policies and the natural and anthropoge nic factors influencing them. Also examined are the effects of the 1997 law on water manageme nt and the efforts that have been made to comply with the legislative mandates. History of Water Management in Florida (1824-1972) In the early years of Florida's governance, ther e was little activity that could be said to constitute a statewide water policy. With respect to wate r management and use, local communities and landowners were free to do as they chose with little interference. Water was abundant, often too much so for most settlers, and prior to statehood in 1845, only minor concern seems to have been given to regional or stat ewide water management issues. Although not enacted, an 1824 proposal by the legi slative council calling for the construction of a canal across Florida for ship passage represents one excepti on to the dearth of early large-scale water management and planning (Purdum et al. 1998). Florida was granted statehood in 1845 and was given 202,344 ha of land by the federal government for “internal improvements”. In 1850, the U.S. Congress additionally conveyed all swamps and inundated lands, over 8 million ha in total, to state owne rship through the Swamp Lands Act (U.S. Congress 1911). In subsequent years, public lands granted to the state by Congress were sold and the resulting funds were funneled through the state's Internal

PAGE 17

17 Improvement Fund and used to invest in railroa ds and guarantee railroad bonds. However, the outbreak of the Civil War devastated FloridaÂ’s ra ilroads, and with the prospect of railroads failing, future sales of public lands went to funding canals to drain the Everglades (Egozcue 2001). The legislature looked to the south Florid a Everglades for reclamation and settlement, believing these lands offered suitable areas for cultivation of sugar, rice, cotton and tobacco (Egozcue 2001). In 1881, the state sold four million acres of land to Hamilton Di sston for $1 million. A businessman from Philadelphia, Disston began channelization of the Caloosahatchee and the Upper Kissimmee River basins in south and centr al Florida (Purdum et al. 1998). Numerous other companies were given charters to dredge canals throughout the stat e and were given public lands in compensation for their efforts. The la nd was subsequently sold to farmers as it was drained. Under pressure from farmers, the le gislature encouraged the creation of additional canals and dykes to drain more land for deve lopment (Egozcue 2001). In 1883, Lake Beauclair and Lake Apopka in central Florida were connect ed to reduce Lake Apopka 's water elevation by approximately one meter and enhance the farming ventures on the northern shore of the lake (Shofner 1982). During the years leading up to the 20th century, additional canals were completed between Lakes Apopka, Dora, Eustis, Gr iffin and the Ocklawaha River. These canal projects reinvigorated the dream of a cross-Flor ida shipping canal, which had been considered earlier but had been set aside in favor of railroad development. In 1902, Congress heeded a United States Army Corp of Engineers (USACOE) recommendation and a federal navi gation project was initiated on the Kissimmee River. Other regional projects, developed to de-water the Ever glades, led to the creation of the Everglades Drainage District in 1907 (Figure 1). This was the first regional authority created by the state to

PAGE 18

18 implement water policy. The District construc ted six major canals to taling over 700 km, which still connect Lake Okeechobee to the Gulf of Mexico and the A tlantic Ocean. As had central FloridaÂ’s earlier trend, the draina ge of land for agriculture in south Florida continued to be encouraged. In 1913, the General Drainage Act ga ve landowners the right to form districts to reclaim their lands. Over 100 such districts were created in the Evergl ades area; some still operate today (Anderson and Rosendahl 1998, Purdum et al. 1998). The 1920's saw a marked shift in state water ma nagement goals. Until that time, lands had been drained and canals created to reclaim land and foster the developm ent of both agriculture and urban areas. During the 1920s, four major hurricanes hit south Florida and during the 1928 hurricane, Lake Okeechobee overflowed, killing ove r 2,000 people. This catastrophe led to an increase in Federal assistance for water manageme nt projects and spurred the state to establish the Okeechobee Flood Control District in 1929 (Anderson and Rosendahl 1998; Figure 1). While the methods remained largely the same, th e stated goal, which brought in federal help from the USACOE, became one of flood protecti on, which had the ancillary benefit of providing continued drainage for agriculture and urban development. By the 1940s, the legislature began to act under mounting recogniti on that state-level oversight of water management was necessary. In 1945, the Florida legisl ature created the State Board of Conservation. The board was tasked with the protection of the stateÂ’s marine, mineral and water resources. In 1947, the legislature created the Water Survey and Research Division, whose records were later forwarded to the Flor ida Geological Survey in 1955 when the division was dissolved (Purdum et al. 1998). Political willingness to move towards increased state oversight was matched by the public sentiment of the times. Concern over the adequacy of environmental protection was growing as newspa pers reported fish kills in lake Apopka

PAGE 19

19 (Bachmann et al. 1999) and publication of The Everglades: River of Grass (Douglas 1947) raised considerable concern for the state's water resources. As had happened earlier in the 1920s natural disasters influenced the course of state water policy. In 1947, two hurricanes cau sed substantial flooding in Miam i. The Central and South Florida Flood Control District wa s established in 1949 to coordina te the efforts of the federal governmentÂ’s Central and South Florida Flood Cont rol Project. The districtÂ’s plan had many components, including flood control, water contro l, water conservation, prevention of salt water intrusion into freshwater reserv oirs and aquifers, pr eservation of fish and wildlife, improved navigation, and pollution abatement (Maloney et al. 1968). A decade later, two floods within a two-year period, followed by Hurricane Donna in 1960, inundated over 1 million acres of land east of Tampa and caused $29 million in damages. The state enlisted federal aid through the USACOE, and with a price tag of $100 million, th e Four Rivers Basin Project was conceived. The Southwest Florida Water Management Dist rict (SWFWMD; Figure 1) was created in 1961 to manage the project and to oversee Florida' s $40.5 million share of the expenses (Blake 1980). Clearly, flooding turned the focus and the re sources of the state to flood control and canalization and diluted the mounting desire fo r environmental regulation and natural systems protection. However, the era of canalization and drainage began to come to an end as 1970 saw a demand for a reevaluation of the Four Rivers Basin Project, and work on the project was halted (Purdum et al. 1998). Under mounting public pre ssure and with concern that drainage of regional surface waters was causing environmen tal harm, the SWFWMD requested that the USACOE undertake an environmental impact asse ssment of the Four Rivers Basin Project, an action that eventually led to abandonment of the project.

PAGE 20

20 By the early 1970s, public demands for envi ronmental protection were again gaining momentum and the state was committed to impl ementing more comprehensive forms of water management. Florida had been struggling with a need for statewide water management and several divisions and boards were established as a result of committees established by various governors. The stateÂ’s executive office made e nvironmental issues a major policy platform. Under greater state scrutiny, the number of appr oved dredge-and-fill projects decreased from 2,000 to 200 a year (Blake 1980, Derr 1989, Kallina 1993 ). Several environmental laws were passed in 1971, including the Envi ronmental Protection Act, whic h allowed Florida citizens to sue the state when environmental laws are not en forced. However, it wasnÂ’t until 1972 that the state finally developed a cohere nt water policy through passage of the Florida Water Resources Act. Water Resources Act of 1972 The Florida Water Resources Act of 1972 was an important step forward for state water policy. Previously, water had been managed regi onally in a disjointed manner. Agencies had often been created in response to recent weathe r events or local economies, and their policies were so narrowly focused on one issue that other problems were often exacerbated by their activities. However, the task of creating centralized agencies and passing the laws that would empower them to conduct more comprehensive planning for water resources, was daunting. To this end, Florida lawmakers turned to A Model Water Code (Maloney et al. 1972). Prior to the adoption of the code, Florida ad hered predominately to eastern water law. Eastern water law is a common la w riparian system, which largel y restricts water rights to landowners whose property is directly adjacent to water bodies. Further, water rights are subject to reasonable use. Historically, there was a l ack of administrative guid ance on reasonable use, leaving litigation, and itÂ’s uncertain outcomes, as the only means of determining reasonableness.

PAGE 21

21 This uncertainty of future water availability can discourage industrial inve stment in areas with no guarantees of meeting future ne eds (Maloney et al. 1972). Believing that FloridaÂ’s water resources would not be protected unless statewide control and long-term, scientifically ba sed planning was adopted, the draf ters of the Model Water Code stated three purposes: to take into account the hydrologic interrelations hip of all types of water resources in the state; to provide greater certainty than is possible under a court-administered reasonable use approach; yet to retain sufficient flexibil ity to make possible re alistic long-range plans for the conservation and wise use of water resources and the elimination of waste. (Maloney et al. 1972) To achieve these goals, multipurpose water planning was essential. The state had a long history of single-purpose special districts. The majority of entities actually vested with regulatory or management authority were local forms of govern ment, including drainage districts, irrigation districts, water supply districts, aqueduct districts, sewer district s and mosquito control districts (Maloney et al. 1980). In contrast the state had a short history of multi-purpose special districts; the first having been the Centra l and South Florida Flood Contro l District whose charge was multifaceted (Maloney et al. 1968). The state had also establ ished, in 1961, the SWFWMD. This was the first water resource district in the state whose name reflected the broad multifaceted functions it was tasked with, ra ther than a single function, such as drainage or flood control. To achieve multipurpose water planning at a state level (multipurpose water planning was implemented 10 years earlier at the federal le vel by the Water Resource Planning Act), A Model Water Code (Maloney et al. 1972) identified four requirements for success. First, centralized planning was considered necessary. Lack of centralized control c ould lead to state programs that could contradict each another and diminish management effici ency. Second, the planning must be based in science; the hydrologi c cycle and the effects of polluti on, land use, recharge loss, and development must all be considered as well as their interrelationships. Third, planning must

PAGE 22

22 recognize relationships between water pollution a nd water use and these re lationships must be considered when developing long-term plans. Finally, consumptive use must be regulated (Maloney et al. 1972). The Model Water Code (Chapter 373, Florida Statutes) envisioned statewide planning, but also envisioned scientific planning based on hydrologic boundaries ra ther than political boundaries. This seeming contradiction was reso lved by the creation of five regional water management districts, collectively covering the st ate. These districts would be responsible for local issues within their boundari es but would not have the aut hority to prepare independent water-use plans (Tjoflat and Quincey 2001). Ther efore, scientific research could be conducted and district water control struct ures and lands could be owned and managed on a regional basis, but a single state entity would be responsible for adopting a st ate water plan. The five districts are the Northwest Florida (NWFWMD), Su wannee River (SRWMD), St. JohnÂ’s River (SJRWMD), Southwest Florida (SWFWMD), and South Florida Water Management (SFWMD) Districts (Figure 1). Statewid e oversight was initia lly granted to the Division of Natural Resources and now resides with the re-named Florida Department of Environmental Protection (DEP). Water Management in Florida; Qu ality not Quantity (1972-1997) The 1972 Water Resources Act gave broad pow ers to the regional Water Management Districts (WMDs) and to the Di vision of Natural Resources ( now, DEP). The new law allowed for consumptive use permitted on a reasonable-beneficial use basis. Defining reasonable and beneficial use proved to be difficult however, and before the turn of the 21st century some areas of the state would be over-permitted for withdrawal s. The law also required that each district formulate a water shortage plan and establish minimum flows a nd levels (MFLs) for surface waters and minimum levels for aquifers.

PAGE 23

23 District MFLs programs were not, however, im plemented with any sense of immediacy. A Model Water Code (Maloney et al. 1972) had called for minimum flows and levels to establish a “harm” threshold to protect water resources from withdrawals. As Ross (2001) points out, this linked the MFLs program to the consumptiveuse permitting program, which was designed to prevent “harm” to the water resources of the st ate. However, when enacting the 1972 law, the passage concerning MFLs was modified slightly to protect the water resour ces from “significant harm”. This difference is notable because alt hough neither term is defined, the law establishes that there is a difference between "harm" and "s ignificant harm", and suggests some separation of the consumptive use and MFLs programs. Ro ss (2001) further pointed out that the linkage between MFLs and the State Water Use Plan, tho ugh present in the original 1972 legislation, was removed in 1973 when the MFL requirement was moved into a stand-alone statue (F.S. 373.042). These changes from the Model Water Code reduced MFLs to a more marginal role than was likely intended by the authors of the Code. While water quantity policies were still evolvi ng, water quality concerns took center stage. The 1970s saw the state acquire 141,640 ha of e nvironmentally sensitive lands, often for the protection of adjacent water bodies. In 1975, the Department of Environmental Regulation (now, DEP) was formed and given the task of controlling pollution, pe rmitting dredge and fill activities and supervisin g the WMDs. Water resource legisla tion in the 1980s included the 1983 Florida Water Quality Assurance Act and, in 1984, the Warren Henderson Wetlands Protection Act, which provided additional protection for wetlands by including criteria for the evaluation of dredge and fill permits (Purdum et al. 1998). Florida also passed the 1987 Surface Water Improvement and Management Act. This act focused on the restorat ion and protection of surface water bodies of regional significance.

PAGE 24

24 The 1990s saw the continued support for pub lic land acquisition and, by 1998, Florida had acquired more than 849,840 ha of land for cons ervation purposes. Combined with lands protected by federal and local programs or owned by private conservation groups, the total amounted to 3 million ha or 22 % of Florida (Pur dum et al. 1998). Florida also continued to revisit water management decisions of the past, eventually passing the Everglades Forever Act, continuing restoration work around Lake Apopka, re maining engaged in discourse regarding the restoration of the Ocklawaha River, and with th e help of the USACOE, in itiated restoration of the Kissimmee River (Anderson and Rosendahl 1 998, Bachmann et al. 1999, Canfield et al. 2000, Lowe et al. 1999, Purdum et al. 1998). Perhaps the clearest indicati on of a growing recognition that water quantity as well as water quality was an important management concern comes from the difference between a statement made in 1985 and one a decade later in 1995. In 1985, the legislature adopted the state comprehensive plan, which stated: Florida shall assure the availability of an ad equate supply of water for all competing uses deemed reasonable and beneficial and shall ma intain the functions of natural systems and the overall present level of surface and ground water quality. Florida shall improve and restore the quality of waters not presently meeting water quality standards. (F.S. Chapter 187) While this statement is reasonabl y strong, it is imprecise. Its only mention of water quantity is to note that Florida shall maintain an “adequate” supply to meet the needs of consumers who's proposed use must meet the permitting criteria of being reasonable and beneficial. It suggests that the function of natural systems and the “overa ll” level of water quality shall be maintained. This statement maintains the theme in the 1980s of protection of water qua lity, which it seeks to “improve and restore.” In 1995, a similar statemen t was made in the Florida Water Plan adopted by DEP:

PAGE 25

25 water must be managed to meet the wate r needs of the peopl e while maintaining, protecting, and improving the stateÂ’ s natural systems. (DEP 1995) This statement reflects a shift in perception a nd understanding and ackno wledges that Florida's water resources are not boundless. It maintains that the goal of wa ter supply is to meet the needs of the people of the state and that this must be accomplished in conjunc tion with the protection and improvement of natural systems. In this st atement, no distinction is made between the need to manage water quantity and water quality. This shift in management goals seems to be indicative of the growing and prevailing need to address water quantity issues, which had been playing a larger and larger role in regional politics. Florida has large populations in relatively small ar eas of the state. This has led to localized problems, especially in coastal areas where the local water supplies are most vulnerable to salt water intrusion and surface water bodies are freq uently affected by groundwater levels. Highly localized demand raised concern in some countie s that their water woul d be exported to more developed counties, leaving them with degraded environmental resources and reduced ability to enhance development. These factors, coupled with generally low surface water levels throughout the state in the late eighties and early nineties, led to a number of law suits and complaints being filed against many of the WMDs. These complaints from citizen groups and local governments led to a number of actions. In 1993, a decision invol ving the SJRWMD found that the water management district must establ ish MFLs and determined that it should be done within a reasonable time period. At the same time, the SFWMD responded to complaints filed against it by initiating MFL development for the Everglades. A few years later in 1996, the Florida Legislature responded to local concerns in the SWFWMD by requiring a priority list be established for MFL development within the dist rict and that once developed, MFLs should be subject to independent scient ific peer review (F.S. 373.042).

PAGE 26

26 In that same year, a Governor's task force on water supply and funding was convened (Exec. Order No. 96-297). A core assumption a dopted by the task force was that water demand was increasing and that the increased demand ne eded to be met (Matthews and Nieto 1998). Though the task force put forth approximately fift y recommendations, its primary theme was that increasing the supply of available water was necessary to ensure continued population and economic growth. This would lead to new responsi bilities for all the water management districts and new protections for the environment. The 1997 Law and Water Quantity The recommendations from the Governor's ta sk force provided the building blocks for legislation referred to as the “ 1997 Water Act”. This law represen ted the first major revision of state water policy since the Model Water Code was utilized to create Ch apter 373 of the Florida Statutes in 1972. Perhaps the be st summary of the changes in po licy is taken from the statute itself, which was amended to “promote the availability of sufficient water for all existing and future reasonable-beneficial uses and natural sy stems" (F.S. 373.016 (2) (d)). This statement is significant because it denotes a fundamental shift in the state’s water policy. Until this juncture, the state's primary goal was the allocation of the existing water supply among uses and, recently, consumptive uses and natural systems. The Di stricts would now be responsible for allocating portions of available water between users and natu ral systems, and would also be charged with promoting expansion of the water suppl y, through water reso urce development. Though other topics were covered, the 1997 cha nges to Chapter 373 were strongly focused on water quantity issues and cons umptive uses. The new law more clearly defined the role of the WMDs and local governments in supplying water to the public. The water management districts were tasked with "water res ource development" and technical support. "Water resource development" was defined as the formulation and implementation of regional water resource

PAGE 27

27 management strategies and includes the task of data collection, cons truction and operation of structural components for flood control, su rface and underground water storage, and ground water recharge augmentation (F.S. 373.019 (19)). Local governments, or utilities were charged with "water supply development", which was defined as design, planning, construction and operation of the infrastructure necessary to collect and distribute water for sale, resale or end use (F.S. 373.019 (21)). The primary goal of this shift in policy, which aimed to increase the volume of water available, was to avoid pitting consumptive user s against each other and against natural systems (Matthews and Nieto 1998). To acc omplish these goals, the legislatur e stated that projects which create sustainable water sources but require financial assistance to complete, and projects that implement reuse, storage, recharge, or conserva tion of water be given priority funding by the water management districts (F.S. 373.0831 (4) (a) (3)). Further, if a priority project is expected to bring an MFL water body into compliance, it is given "first cons ideration" for funding assistance (F.S. 373.0831 (4) (b)). Along with establishing funding priority fo r water resource development, the 1997 law also clarified the role of wate r resource planning in water management. The DEP is required to produce the Florida Water Plan in conjunc tion with the WMDs, regional water supply authorities, and others (F.S. 373.036). The Florida Water Plan includes the district Water Management Plans that are called for in the 1997 legisl ation. These plans are based on a 20-year planning horizon and are to be updated at least every five years (F.S. 373.036 (2)). The plans include, among other items, met hodologies for establishing minimum flows and levels, water supply assessments, anticipated future needs, projections of water supply adequacy, and Regional Water Supply Plans whic h are developed for areas determ ined to have an insufficient

PAGE 28

28 supply to meet the needs of projected reasonabl e-beneficial uses and to sustain the water resources and related natural systems (F.S. 373.036 (2) (b)). This last clause is important because it means that when the water needs of na tural systems are projected to be inadequate (i.e., a minimum flow or level is expected to be violated), th e WMDs must implement regional water supply planning. Incorporat ion of the five Water Management District water management plans into the Florida Water Plan provides fo r greater consistency am ong statewide and regional planning efforts. Another important change to Chapter 373 in 1997 concerned development of minimum flows and levels (MFLs). While the Model Water Code was developed with MFLs as an integral part of consumptive water use permitting, Chapter 373 (F.S.) did not explicitly incorporate MFLs into the permitting process. Some efforts had been made by the WMDs to determine ecologically sensitive water leve ls for selected water bodies. For example, implementation of the SWFWMDs Lake Levels program, had led to the adoption of management levels for nearly four hundred lakes by 1996 (SWFWMD 1996). Yet MFLs remained an undefined and relatively abstract concept. In large part, this dilemm a stemmed from the phrase "significant harm" which was included in the statute (F.S. 373.042). Because the language was changed from the original "harm" standard contained in th e Model Water Code, it was clear that legislators intended to differentiate "harm" from "significant harm." Compounding the issue wa s the lack of guidance provided by the legislature regardi ng factors that should be considered when establishing MFLs. This is important because many of the state's surface water basins have been substantially altered, and a return to higher historic levels in these basins would re quire major restoration efforts and often result in the flooding of privat ely owned lands and struct ures, including homes. Another basis for the lack of action could be attr ibuted to uncertainty re garding the purpose of

PAGE 29

29 MFLs in the WMDs regulatory framework. Even if "significant harm" had been defined, the MFLs statute remained isolated from consump tive use permitting criteria, and thus served no immediately evident role in th e district's regulatory charge. In 1997, the Florida Legislature provided guida nce regarding factors to be considered when establishing minimum flows and levels. The language the legislature sought was one of compromise and practicality. It sought to protect the water res ources of the state from water withdrawals that would cause si gnificant harm. Howe ver, they acknowledged that reestablishing historic water flows or levels was not necessar ily desirable in all cases and that structural changes in the watersheds should be consider ed during the development of MFLs (F.S. 373.0421 (1) (a)). They also provided exemptions for wate r bodies smaller than 25 acr es in size, those that have been constructed or ones that no longer se rve their historic function (F.S. 373.0421 (1) (b)). A list of both cultural and scientific factors to be considered when es tablishing minimum flows or levels was developed, and incorporated into the Florida Admini strative Code (Ch. 62-40.473 F.A.C.). Identified factors include recreation, fi sh and wildlife habitats and passage, estuarine resources, maintenance of freshwater storag e and supply, aesthetic a nd scenic attributes, filtration and absorption of nutrients, water qua lity, and navigation. However, determination of the precise methodology for establishing MFLs wa s left up to each WMD and has, to date, resulted in the development of different approa ches for establishing minimum flows and levels. The 1997 Water Act also provided criteria for actions to be implemented when a minimum flow or level is not met or projec ted to be unmet in the next 20 y ears. The statute directs that the DEP or the WMD governing boards shall implemen t a recovery or prevention strategy, which includes the development of additional water supp lies and conservation concurrently with, to the extent practical, reductions in permitted withdraw als (F.S. 373.0421 (2)). This strategy became

PAGE 30

30 part of the regional water supply plan and, thus, eventually the Fl orida Water Plan. It is this language, from F.S. 373.0421 (2), that explicitly gave great weight to the minimum flows and levels developed by the water management districts. Violation of a flow or level can result in reduction of permitted withdrawals, and by extens ion, denial of proposed withdrawals that are projected to cause a violation of a minimum flow or level. Thus, the 1997 statute corrected a variance between the minimum flows and leve ls requirement and consumptive water use permitting. Matthews and Nieto (1998) pointed out that the WMDs must attempt to develop alternative water sources in conjunction with any reductions in permitted withdrawals. They contended that combining the obligation to en gage in environmental restoration with the requirement for water resource development shoul d protect the environment without inequitably reducing water use or stunting economic deve lopment. Therefore, a balance between maximizing water resource benefits for both consumptive use and natural systems protection may be achieved. Response to the 1997 Law The mandate to give priority funding to wa ter supply projects has been well implemented and, in fiscal year 2001, the five water manageme nt districts spent betw een 17.7 % of their total budget in the Northwest District and 40.1 % of their budget in the Southwest District with all five districts committing $287 million, or 26.9 % of their combined budget on water resource development (DEP 2000). In 2002 the total clim bed to $344 million, representing 28.0 % of the combined WMD budgets. Fiscal year 2003 bu dgets included approximately $372 million for water supply activities (DEP 2003a). Funds budgeted between 2001 and 2006 are projected to exceed $1.4 billion (DEP 2002a). The 1997 amendments to Chapter 373 require all five water management districts to implement regional water supply planning in areas where water supplies are inadequate to meet

PAGE 31

31 the projected water needs on a 20-year horizon. Water demand in the state is expected to increase from approximately 27.2 billion L day-1 in 1995 to 34.8 billion L day-1 by the year 2020 (DEP 2002a). Based on this projected demand, f our of the five water management districts anticipate water deficits in at least a portion of th eir jurisdictional boundaries and have been required to submit regional water plans to the Florida DEP. As of August 2001, all the required plans were complete (DEP 2002a). While it will take years to measure the eff ectiveness of major wate r-resource planning efforts, there is some evidence to suggest at leas t partial success. Planni ng and funding priorities guide the districts towards water resource developm ent goals, which will need to be achieved to meet future water demands while protecting Flor ida aquatic ecosystems. Planning has helped quantify the future needs and focus water supp ly development on cost effective solutions, including seawater desalinizati on and aquifer storage and recove ry. While these technologies hold promise and are being actively developed, they have yet to fully reach their potential role in Florida's water supply. Other strategies such as reclaimed water and conservation efforts are currently contributing cost e ffective alternatives to potab le water use (DEP 2002b). For example, reclaimed water systems in Florida supplied 2.2 billion L day-1 in 2002 and system capacity is currently in excess of 4.2 billion L day-1 (DEP 2003b). While the 1997 Water Resource Act made consider able progress in clarifying the role that MFLs are to play in state water policy, it also left many technical questions involved in the development of MFLs unanswered. Considerable work has been done on lo tic systems, such as rivers and streams. Much of this work cente rs on protection of habita t by identifying necessary flow regimes using tools such as Instream Fl ow Incremental Methodolog y (IFIM) and Physical Habitat Simulation System (PHABSIM) (Poff et. al. 1997, Postel and Richter 2003, Richter et. al

PAGE 32

32 1997). Work on MFL development for lentic system s is less extensive. Further complicating MFL development is the need to account for both natural systems and cultural values during development. Cultural factors, such as aesthetic and scenic attributes, are difficult to quantify and like many natural systems values, are related to and affected by numerous other factors, such as water quality. While all five water management districts s eek, through MFLs, to prev ent significant harm to the natural systems in their region, those that have initiated th e process are employing different approaches (Ch. 40C-8, 40D-8 and 40E-8, F.A.C.). While this process has not yielded a uniform methodology, it may be appropriate for each WMD to establish its own methods since each faces different challenges; some need to protect water resources that are not highly influenced by consumptive uses, while others mu st develop MFLs for water resources that are severely impacted by withdrawals. As of 2002, th ree of the five WMDs have established MFLs on a total of 132 lakes, 9 rivers and streams, 46 wetlands, 20 aquifers, and 2 estuaries (DEP 2003c). One of the greatest strengths of the 1997 la w has been tying compliance with MFLs to water planning, funding of water resource development and regulat ory oversight of consumptive use. In the SWFWMD, this interaction between MFLs, planning, funding and water-use permitting is evident. Regional, public-supply wellfields in the Northern Tampa Bay area are believed partially responsible for reducing water levels in some area water bodies. To meet MFLs in the region, an agreement calling for a combination of permit reductions, water supply development, and funding assistance was ente red into by the SWFWMD the regional water supply authority (Tampa Bay Water), and its me mber governments. Specifically, the agreement required ground water withdrawals at 11 regional wellfields to be reduced, in stages, from 158

PAGE 33

33 mgd to 346 mgd by December 31, 2007. An inte rmediate reduction to 121 mgd was set for December 31, 2002 and has been met. To help offset reductions of wellfield pumping, 85 mgd of new water supply will be needed. The SW FWMD is providing, following funding priority directives, up to $183 million in funding assistan ce between 1995 and 2007 for alternative water supply development (DEP 2000). Alternative source projects underway, include surface water source development, indirect potabl e reuse, and seawater desalination. Summary The 1997 Water Act improved and enhanced the tools available to water managers and provided needed emphasis on water quantity and c onsumptive-use issues (Hamann 2001). The minimum flows and levels component asks the question, presuming we are going to harvest as much as water as we can, "How much can we harvest without significantly harming the natural resources of the state?" The pl anning requirements ask managers to determine where projected shortages may be expected during the next 20 ye ars, and how the shortfalls can be met with alternative sources and conservati on efforts. The priority-funding re quirement attempts to assure that the alternative sources identified in regi onal and state plans receive necessary economic support. The challenge, as we move forward, is for scie ntists, engineers, polic y makers and citizens to respond to the 1997 legislation with ingenuity The groundwork for future success has been established through the legal acknowledgement of clean water as a limited resource. When establishing consumptive-use limits, managers mu st consider scientific parameters such as absorption of nutrients and estuarine resources, as well as cultural parameters such as recreation, aesthetics, and scenic attribut es. Water resource managers and policy makers must also determine what constitutes "significant harm", and identify the "practicable" steps needed to

PAGE 34

34 develop alternative water resources to offset w ithdrawal reductions. Thus far, Florida has responded to these challenges. Conclusions In keeping with the 1972 vision of regi onal water management, the 1997 legislation retained flexibility for each of the water manage ment districts to addre ss regional water resource issues. The 1997 legislative changes have di rectly impacted state water management by mandating the development of MFLs and assuring that they are met by tying them to water planning, funding of water resource development and regulatory oversight of consumptive use. Planning has helped quantify the future need s and focus water supply development on costeffective solutions. The Water Management Dist ricts' commitment to water supply planning is evident in the district budgets, which demonstr ate a steady increase in funding for water supply projects. Floridians have had the resourcefulness to ask, "How much water is there"? This means that water will no longer be a limited resour ce in some abstract or theoretical future, but will have real limitations defined for its use.

PAGE 35

35 SFWMD SJRWMD NWFWMD SWFWMD SRWMDWater Control Districts Central and South Florida Flood Control District Everglades Drainage District Okeechobee Flood Control District NWFWMD SFWMD SJRWMD SRWMD SWFWMD 80 0 80 160 40 MilesWater Management Districts Figure 2-1. Map of the State of Florida showing the three early water management districts and the five current water management districts.

PAGE 36

36 CHAPTER 3 CURRENT TECHNIQUES UTILIZED FOR TH E QUANTIFICATION OF FLOODPLAIN HABITAT LOSS ASSOCIATED WITH RE DUCTIONS IN WETSEASON FLOW Introduction Although floodplains have long been recognized as seasonally important riverine habitat, floodplain inundation has historically not been addressed in most minimum flow determinations (Postal and Richter 2003). Middl eton (1999) also points out that regulation of water regimes has not always fully appreciated the interaction be tween floodplains and rivers via the flood pulse. Regulation of river flows can, how ever, result in decreased stage fluctuations and alteration of inundation patterns of floodplain wetlands (Poff et al. 1997, Woltemade 1997). Quantification of potential differences in fl oodplain inundation associated with differences between minimum flow requirements and natural hydrologic regimes s hould, therefore, be a critical component of regulatory programs that are intended to protect or restore riverine habitat. Compared to instream evaluations of mi nimum flows requirements, there has been relatively little research done on river flows necessary for meeti ng the requirements of floodplain species, communities or functions. Junk et al. (1 989) noted that the “driving force responsible for the existence, productivity, and interactions of the major river-floodplain systems is the flood pulse”. Periodic inundation of riparian floodplai ns by high flows is closely linked with the overall biological productivity of river ecosyst ems (Crance 1988, Junk et al. 1989). Many fish and wildlife species associated with rivers ut ilize both instream and fl oodplain habitats. During inundation of the river floodplains, the habitat and food resources available to these organisms greatly expands (Ainsle et al. 1999, Hill and Cichra 2002, Whar ton et al. 1982). Inundation during high flows also provides wa ter and nutrients that support high rates of primary production in river floodplains (Brinson et al. 1981, Conner and Day 1979). This primary production generates large amounts of organic detritus, which is critical to food webs on the floodplain and

PAGE 37

37 within the river channe l (Gregory et al. 1991, Vannot e et al. 1980). Flo odplain inundation also contributes to other physical-che mical processes that can affect biological production, uptake and transformation of macro-nutrients, and developmen t of wetland soils that are important to the overall function of the river ecosystem (Kuensler 1989, Stanturf and Schoenholtz 1998, Walbridge and Lockaby 1994, Wharton et al. 1982). Floodplain vegetation development and persis tence may not, however, necessarily depend wholly on inundation from the river channel. Gr oundwater seepage, hyporh eic inputs, discharge from local tributaries, and pr ecipitation can also lead to fl oodplain inundation (Mertes 1997). Recent work on the upper segment of the Peace Rive r and the Alafia River in central Florida suggested that direct and continuous inundation of floodplai n wetlands by river flows is insufficient to account for inunda tion needs of the dominant sp ecies found in the wetlands (SWFWMD 2002). However, because river cha nnel-floodplain connections are important, can be influenced by water use, and may be a func tion of out-of-bank flows, it is valuable to characterize this connectivity. As with most other floodplain-related water management issues, technically sound development of minimum flows and levels re quires such characterization. The Southwest Florida Water Management Distri ct (District) has developed an approach which recognizes that fundamental to the develo pment of minimum flows and levels (MFLs) is the realization that a flow regime is necessary to protect the ecology of the river system (Hill et al. 1991, Hupalo et al. 1994, Rich ter et al. 1996, Stalnaker 1990). The recognition of a seasonal flow regime as an essential component of this task led the District to develop minimum flow criteria for three distinct annual-flow periods ; periods of low, medium and high flows. The District approach also acknowledge s the effects of climatic oscill ations on regional river flows

PAGE 38

38 and includes identification of tw o distinct benchmark flow peri ods that are consistent with effects of the Atlantic Multid ecadal Oscillation (AMO) (Enfie ld et al. 2001, Kelly 2004). Because the methodologies employed by the Distri ct in the low and medium flow periods are commonly applied, consisting of components su ch as wetted perimeter analysis and physical habitat simulation analysis, they are not examined in the context of this chapter. Further, the focus of this paper remains on the connection be tween the floodplain and the river, which most frequently occurs during the high flow season. The resource management goal identified for the seasonally predictable wet period is maintenan ce of seasonal hydrologic connections between the river channel and floodplain to ensu re floodplain structure and function Minimum flows developed for the high flow season are intended to protect ecological resources and values associated with floodplains by maintaining hydrologic connections between the river channel and floodplain and maintaining the natural vari ability of the flow regime. These goals quantified through us e of the Hydrologic Engineer ing Centers-River Analysis System (HEC-RAS) and the alteration of long-te rm flow records to evaluate floodplain feature inundation patterns associated with channel-floodplain connectivity. The reduction in number of days of inundation is a temporal measure of loss based on how many days, annually, a specified water-surface el evation was reached compared to how often that elevation would have been reached if flow conditions had been reduced by a given amount. This temporal measure of habitat loss has been employed by the District in establishing MFLs on freshwater river systems in southwest Florida, during the high flow season of the year. It is applied by reducing historic flows iteratively until the pe rcent of flow reducti on that results in a 15% reduction in the number of days of inunda tion is found. The use of a percent of flow

PAGE 39

39 reduction in establishing MFLs a ssures protection of the natural hydrograph (Flannery et al. 2002, Kelly at al. 2005a, b, c). In Florida, regional Water Management Districts, by virtue of their re sponsibility to permit the consumptive use of water and a legislativ e mandate to protect water resources from “significant harm" have been directed to esta blish MFLs for streams a nd rivers within their boundaries (Section 373.042, Florida Statutes; Munson et al. 2005) The Southwest Florida Water Management District has developed methods for establishing MFLs that acknowledge the importance of seasonal flow regimes, includ ing flows necessary for floodplain inundation. A temporal measure of habitat loss based on cha nges in inundation is used to develop minimum flows for seasonal high-flow periods. An altern ative approach, employing a measure of spatial loss, by examining the change in top width at each cross section in the model under different flow conditions, is also explore d. The use of a temporal measur e of habitat loss for establishing MFLs during high flows is compared to the spatia l loss of habitat for th e same flow reduction. The applicability of both is discussed and the results compared to determine which is more restrictive in te rms of allowable flow reduction. The Alafia River, Myakka River and Peace River are used in this chapter to illustrate the temporal approach to quantifyi ng habitat change associated with temporal difference in floodplain inundation patterns (Figure 3-1). All thr ee rivers are located in southwest Florida. The Alafia River originates as a series of creeks that combine to form two branches of the Alafia River know and the North Prong and the South Prong. The North prong drains a swampy area and is characterized by low-lying wetlands. Th e South Prong is drains a highly mined area and is more incised. The Alafia River downstream of the confluence is more highly incised as well. The Alafia River above the Lithia gage (USGS# 02301500) drains approximately 335 mi^2. The

PAGE 40

40 Myakka River drains approximately 229 mi^2 above the Sarasota gage (USGS# 02298830). The Myakka River is a shallow river with broad flat floodplains. Tw o wide shallow lakes are near the gage site as well as substantial grass prairies It is the lease inci sed of the three rivers studied. The Peach River at Arcadia gage (USGS# 02296750) captures approximately 1,392 mi^2 of the Peace River watershed. The channel is well defined upstream of Arcadia and widens downstream of Arcadia. In some areas of th e river the floodplain is over a mile wide. Methodology The District's approach to protection of flows associat ed with floodplain habitats, communities, and functions involves considera tion of direct connect ion between the river channel and the floodplain. As part of this pro cess, plant communities and soils were identified across river floodplains at a numbe r of sites on the Alafia River, Myakka River, and Peace River in central Florida (Figure 3-1), and periods of inundation/connection with the rivers were reconstructed on a seasonal basis. These data we re used to characterize the inundation of these communities/soils by out of bank river flows, and to develop criteria for establishing minimum flows for the seasonal period of high flows Floodplain cross-sections were selected ba sed on the location of vegetation communities identified from the USGS Gap Analysis Progr am maps within the Alafia, Myakka, and Peace river corridors. Attention was al so made of the location of shoals in the river in an attempt to capture the hydraulic control poin ts in the model. Eight, tw elve, and ten representative floodplain/vegetation cross-sections were established perpendicula r to the river channel within dominant National Wetland Inventory vegetation t ypes for the Alafia, Myakka, and Peace rivers, respectively. Cross-sections were establishe d between the 0.5 percent exceedance levels on either side of the river channel, based on prev ious determinations of the landward extent of floodplain wetlands in the river corridor. Ground elev ations were determined at 50-foot intervals

PAGE 41

41 along each cross-section. Where changes in el evation were conspicuous, elevations were surveyed more intensively (Kelly et al. 2005a b c). To characterize forested vegetation commun ities along each cross-section, changes in dominant vegetation communities were located and used to delineate boundaries between vegetation zones. At each change in vegeta tion zone, plant species composition, density, basal area and diameter at breast height (for woody ve getation with a diameter at breast height > 2.5 cm) were recorded. Soils were characterized within each vegetation zone as hydric, organic, peat, or mineral by obtaining at least three soil cores. The cores were examined to a depth ranging from 51-152 cm to classify the soils. Special considerati on was placed on locating elevations of the upper and lower extent of mucky soils (> 20 cm in thickness) at cross-sections where they occurred (Berryman and Henigar 2004). Steady-state HEC-RAS modeli ng was used to determine corresponding flows at a downstream gage that would be necessary to inun date specific floodplain elevations (e.g., mean vegetation zone and soils elevations) along the floodplain/vegetation cro ss-sections. Version 3.1.1 of the HEC-RAS model released by the U. S. Army Corps of Engineers Hydrologic Engineering Center in November 2002 was utilized for all three rivers. The HEC-RAS model is a one-dimensional hydraulic model that can be us ed to analyze river flows. For subcritical flows, it operates as a typical step-back model, resolving the energy equation between adjacent cross-sections. Profile computations begin at a cross-section with known or assumed starting condition and proceed upstream (US Army Corp s of Engineers 2001). Models for each river were based on cross-sectional data collected by the United States Geologic Survey and the floodplain/vegetation data collec ted by the District (Hammett et al. 1978, Kelly et al. 2005 a, b, c, Murphy et al. 1978).

PAGE 42

42 For development of minimum flows, the year is categorized into three distinct flow periods or "blocks". Based on flow records for long-term United States Geological Survey (USGS) gage sites at the Myakka River near Sarasota (USGS# 02298830), the Alaf ia River at Lithia (USGS# 02301500), the Peace River at Arcadia (USGS# 02296750), and the Hillsborough River at Zephyrhills (USGS# 02303000, The Hillsborough Rive r was used to develop the season block for the region but is not otherwise discussed in this chapter), the seasonal blocks were determined as periods corresponding to dates wh en specific exceedance flows occurred. On an annual basis, a low-flow period, Block 1, was defi ned as beginning when the median daily-flow fell below and stayed below the annual 75% ex ceedance flow. Block 1 was defined as ending when the high-flow period, or Block 3, began. Block 3 was defined as beginning when the median daily flow exceeded and stayed above the mean annual 50% exceedance flow and ending when the flow fell below and stayed below the mean annual 50% exceedance flow. The medium-flow period, Block 2, was defined as extending from the end of Block 3 to the beginning of Block 1. Between these rivers, there was little differen ce in the dates that each defined flow period began and ended (Table 3-1). Block 1 is defined as beginnin g on Julian day 110 (April 20 in non-leap years) and ending on Julian day 175 (June 24). Block 3 is defined as beginning on Julian day 176 (June 25) and ending on Julian da y 300 (October 27). Block 2, the medium-flow period, extends from Julian day 301 (October 28) to Julian day 109 (April 19) of the following calendar year (Kelly et al. 2005a, b, c). Though th e District has developed MFLs for all three blocks, this paper focuses on th e criteria for developing minimu m flow standards for Block 3, which runs from June 25 to October 27 of each year.

PAGE 43

43 For development of MFLs, it is necessary to identify a benchmark period from which flow deviations may be evaluated. Often changing fl ow trends are presumed to be the result of anthropogenic stresses on a system. This is possi bly because variation in flow is frequently assumed to be the result of random independently and identically distributed random variables such as rainfall in flood risk analysis (Olsen et al. 1999). However, the effect of multidecadal oscillation on river flow pattern s must also be recognized as natural climatic variations. As Enfield et al. (2001) observe, the Atlantic Mult i-decadal Oscillation has a pronounced effect on rainfall in the continental United States and wet-season rainfall in peninsular Florida is negatively correlated with the Atlantic Multidecadal Oscillation (McC abe and Wolock 2002). Kelly (2004) examined stream flow s in Florida and identified a st ep-trend similar in timing to the step-trend describing the AMO (Enfield et al. 2001) This resulted in the identification of two benchmark periods, one from 1940-1969 and one from 1970-1999, which are characterized by relatively higher and lower wet season river flows in peninsular Florida, respectively. For determination of minimum flows, records from both benchmark periods are analyzed, and used to develop percent-of-flow reduc tion criteria. Minimum flows ar e determined by calculating the percent flow reduction for each benchmark peri od that would result in no more than a 15% reduction in habitat. For the Block 3 period, th is means identifying the flow reduction that would result in no more than a 15% loss in the nu mber of days of days that floodplain features are inundated by the river, during each of the two benchmark periods and utilizing the more limiting of the two. In establishing minimum flows, th e District defines a 15% change in habitat availability as a criterion for identifying significant harm, or unacceptable change. For the high-flow period or Block 3, this change is expressed as a temporal difference in the number of days specific water

PAGE 44

44 surface elevations are reached. Po tential temporal change or reduction in the number of days of direct connection between the ch annel/floodplain at the floodplain /vegetation cross-sections on the Alafia, Myakka and Peace rivers was estimated through the alteration of the historic flow record and the HEC-RAS models. Target eleva tions corresponding to featur es of interest, such as the median elevation of cypress swamps or the top of the river banks, were identified during the vegetative surveys. The HEC-RAS model was th en used to determine the flow necessary at the downstream gage to inundate the features w ith flow from the river. The HEC-RAS model thus allows the elevations, identifi ed at different transect s, to be compared as the flow required at the gage site, to reach the co rresponding elevation at the upstream cross-section where the elevation was identified. The daily historic reco rd from the gage site was then examined. All days during the flow period of interest were chec ked to determine if the flow of interest was exceeded. The total number of days during each year of the histor ic record that the flow of interest was reached or exceeded were then ta llied. Because the HEC-RAS model estimates the water surface elevation at cross sections, based on flows at the down stream gage, the reduction in the total number of days a fl ow is exceeded is equivalent to the reduction in total number of days that the corresponding water surface elevation is exceeded, at any cross section. The flow record was then reduced incrementally, until the percent flow reduction required to reduce the number of days a flow was reached or exceeded was reduced by 15 percent. A measure of spatial loss of habitat at each cross-section was also estimated using the HEC-RAS model output and historic flow records. For each flow profile calculated in the model, the top width (linear distance in a cross s ection of water's edge to water's edge) at each cross section is derived. The sum of top widths at all cross sectio ns in the model is computed for each flow profile. This provides an estimat e of the area inundated. Once top width was

PAGE 45

45 calculated for each profile, the relationship betw een top width and flow was plotted. A 15% reduction was then made to the t op width at each flow profile in each model. The generated relationships between top width and flow were th en used to calculate the flow required at the downstream gage sites to achieve the reduced top widths. The flow needed to achieve the nonreduced top width and the flow required to ma intain the top width minus 15% were then compared to determine the percent of flow reduction required to reduce the top width by 15 percent. Results The vegetative surveys on the Alafia, Myakka and Peace rivers resulted in the identification of different features of interest on each river. For the Alafia River, elevations of eight floodplain features were identified at each of the eight vegetative cross-sections where they occurred (Table 3-2). For the Alafia River, the downstream gage at which flow requirements were compared is the USGS Alafia gage at Lithia (USGS# 02301500). Flow requirements, for inundation of each feature ranged from 22.2 m3s-1 to reach the floodplain wetted perimeter inflection point to 64.2 m3 s-1 to reach the low bank elevation for inundation of both sides of river floodplain. Relatively high st andard deviations, for the requi red flows, indicate that the flow requirements for some f eatures differed greatly among cr oss-sections. Flow reduction resulting in 15% fewer days of inundation for th e floodplain features, ranged from 5% for the low-bank elevations and highest floodplain vegetation class to 9% for the highest swamp class and floodplain wetted perimeter inflection point. Comparison of per cent-of-flow reductions, associated with the temporal loss of featur e inundation between the two benchmark periods, indicate that the 1970 to 1999 benchmark period pr ovided the more conser vative flow reduction (Kelly et al. 2005a).

PAGE 46

46 For the Myakka River, the median elevations of six vegetative zones (i.e., Oak-Palm Wet Hammock and Panicum Marsh) as well as six physical characterist ics (i.e., Lowest Bank Elevation to inundation both sides of river floodplain and Median elevation of hydric soils) were identified at each of the 12 vegetative cross-secti ons, when they occurred. For the Myakka River the downstream gage at which flow requirement s were compared is the USGS Myakka River near Sarasota gage (USGS# 02298830). The downstream flow requirements for each identified feature are summarized in Table 3-3 and range from 0.9 m3 s-1 to reach the median elevation of mixed marsh to 24.4 m3 s-1 to reach the median elevation of oak-palm wet hammock. Standard deviations are also presented and indicate that fl ow requirements of some features differ greatly among cross-sections, as noted in the Alafia Ri ver. The percentage by which reducing flow would result in 15% fewer days of the target flow being reached is also presented. Flow reduction resulting in 15% fewer days ranged from 8% for the median elevation of oak-palm wet hammock and the lowest bank elevation to inundat e one side of the rive r floodplain to 68% for the median elevation of mixed marsh. This analysis was completed for each of the two benchmark periods and results were similar to those found for the Alafia River (Kelly et al. 2005b). For the middle Peace River, 10 features, incl uding both vegetative and physical, were utilized in characterizing the flood plain (Table 3-4). For the middle Peace River the downstream gage, at which flow requirements, are compared is the USGS Peace River at Arcadia gage (USGS# 02296750). The downstr eam flow requirements for each identified feature are summarized in Table 3-4 and range from 66.5 m3/s to reach the mean elevation of mucky soils to 196.8 m3/s to reach the 90th percentile elevation of wet hardwood hammock. Standard deviations are also pr esented and indicate that some features' flow requirements differ

PAGE 47

47 greatly among cross-sections, as was true for th e Alafia and Myakka rivers. The percentage, by which reducing flow would result in 15% fewer days of the target flow being reached, is also presented. Flow reduction resu lting in 15% fewer days ranged from 5% for lowest bank elevation to inundate both sides of the river floodplai n to 10% for the mean elevation of mucky soils. This analysis was completed for each of the two benchmark periods and results were similar to both the Alafia River a nd Myakka River (Kelly et al. 2005c). In establishing minimum flows for the high flow season on the Alafia, middle Peace, and Myakka rivers, Kelly et al. (2005a b, c) did not select a single floodplain fe ature to protect. Rather, they noted that higher flow s might require a slightly more restrictive standard than some of the indicators associated w ith low flows and that higher fl ows seem to consistently tend towards a reduction between 5% and 10% (Tables 3-2 to 3-4). To further investigate limiting factors associated with the rive r floodplains, plots of percent-of -flow reductions that would result in 15% losses in the number of days for wh ich corresponding river flows were reached were produced for each river (Figure 3-2). The plots indicate that up to an 8%reduction in the flows necessary to inundate floodplain features of th e Alafia and middle Peace rivers, including those features not identified, would re sult in no more than a 15% reduc tion in the number of days the features are inundated. Similarly, it was determined that a 7% reduction in flow in the Myakka River, during the high-season flow s, would not reduce the number of days of floodplain-river connection by more than 15%. Th is measure of temporal loss of habitat is central to the approach taken by the District in establishing MFLs to protec t the river-floodplain connection during the high flow part of the year. In examining all three rivers it is evident th at as flows increase, the percentage by which the flow can be reduced, without lo wering the number of days that fl ow is reached, is reduced. It

PAGE 48

48 is useful to normalize the flow by looking at flow per watershed area. Percent-of-flow reductions that would result in 15 % losses of the number of days that river flows reached a given flow plotted against flow per wa tershed area, are presented in Fi gure 3-3. Power trends were fit to each curve. The trend lines and the data points for all three rivers are similar when plotted together (Figure 3-3). Utilizing the HEC-RAS model, the spatial lo ss associated with decreased flows was determined. Specifically, the flow reductions resulting in a 15% loss of top width, were calculated. Percent of flow reduc tions, required to achieve a 15% loss of spatial habitat, were plotted versus flow per square mile of watershe d above the gage and again power trends were fit to the data (Figure 3-4). For each river, the re lationship between spatial loss and flow resembles the relationship between temporal loss and flow, in so far as, at low flow, a high percentage of flow reduction is required to re sult in a 15% decrease in top width. The frac tion of flow required to effect a 15% loss in top width, is reduced as flow increases until trending slightly upwards at the highest modeled flows. However, plots of the percent of flow reduc tion required to effect a 15% loss of spatial habitat versus the percent of flow reduction requ ired to effect a 15% lo ss of temporal habitat indicate that for the ra nge of flows examined, temporal loss is a more restrictive measure of habitat loss on these three rivers than spatial loss (Figure 3-5). A 15% loss of temporal habitat is associated with less of a flow reduction than a 15% loss of spatia l habitat, as estimated by top width. Discussion The goal of an MFL determination is to protect the aquatic resource fr om significant harm due to water withdrawals. An MFL was broadly de fined in the enacting legi slation as "the limit at which further withdrawals would be significantly harmful to the water resources or ecology of

PAGE 49

49 the area." (F.S. Chapter 373.042, Munson et al. 2005). What constitutes "significant harm" was not defined. The District has identified loss of flows associated with fish passage and maximization of stream bottom habitat exposure as significantly harmful to river ecosystems. Significant harm can also be defined as quantifiabl e reductions in the amount of available habitat (Gore et al. 2002). Determining the amount of habitat loss, or de viation from a benchmark, that a system is capable of withstanding is based on professi onal judgment. In es tablishing MFLs, the SWFWMD recognized that the Instream Flow Incremental Methodology (IFIM) involves a negotiated threshold to be used as an acceptabl e measure of habitat loss (Bovee et al. 1998). Gore et al. (2002) note that instream flow anal ysts often consider a loss of more than 15% habitat, as compared to undisturbed or curren t conditions, to be a significant impact on a population or assemblage when employing Physi cal Habitat Simulation (PHABSIM) analysis. With some exceptions (e.g., loss of fish passage or wetted perimeter inflection point), there are few clearly delineated break point s which can be relied upon to judge when "significant harm" occurs. Hill and Cichra (2002) not ed that loss of habitat in many cases occurs incrementally as flows decline, often without a cl ear inflection point or threshold. The District employed a threshold of a 15% chan ge in habitat availability as a measure of significant harm for the purpose of MFL development. Although the District utilized a 15% change in habitat availability as a measure of unacceptable loss, percentage changes employed for other instream flow determinations have rang ed from 10% to 33%. For example, in reference to the use of PHABSIM, Dunbar et al. (1998) noted that an alternative appr oach is to select a flow that provides protection to 80% of the habitat, which is e quivalent to a 20% loss. Jowett (1993) used a guideline of one-third loss of existing habitat at naturally occurring low flows, but

PAGE 50

50 acknowledged that no methodology ex ists for the selection of a percentage loss of "natural" habitat, which could be considered acceptable. Th e state of Texas utilized a target decrease of less then 20% of the historic av erage of habitat area in establishing an inflow requirement for Matagorda Bay (Texas Pa rks and Wildlife 2005). For establishment of minimum ri ver flows, the District has pr oposed the use of a percent of flow approach, which identifies th e percentage by which a flow ma y be reduced before the river system is significantly harmed. This approach preserves natural flow patterns by protecting the inherent variability of natural flow regimes. The approach accounts for flow seasonality, by identifying three distinct annual flow seasons an d identifying different habitat measures to be protected during each season. Tw o distinct benchmark periods, with characteristically higher and lower flows were used to analyze all measur es of habitat loss to a ssure protection during both phases of the climate cycle associat ed with the north Atlantic Ocean. In examining the loss of temporal habitat on the three rivers, the comparison in Figure 3-3 suggests that these three rivers respond in a similar fashion to reduced flows. This could be interpreted as the three rivers having similar flow distribution patterns during the wet season and thus showing similar responses to declining flows. This is pr obable because rainfall drives wetseason flows and since these three rivers are all lo cated in central Florida, it might be expected that some similarity in watershed-wide rainfall ov er the period of record exist (at least 60 years in all three cases). The use of spatial loss in stead of temporal loss di d not provide a similar response among rivers (Figures 3-3 and 3-4). T hough somewhat similar, there is considerably more variation in the plots of spatial loss than th ose of temporal loss. Th is is likely due to the addition of morphology into the calculation, albe it implicitly. While rainfall drives the wetseason flow distribution, floodplai n morphology also plays a direct role in calculating loss.

PAGE 51

51 In comparing the flow reducti on required to produce a 15% spatial loss with the flow reduction required to produce a 15 % temporal loss, it was f ound that a 15% temporal loss occurred consistently at a smaller reduction in fl ow (Figure 3-5). This does not specify in an ecological sense how the two different measures re late though it does indicat e that, on these three rivers, the temporal measure is mo re conservative with respect to protection of natural flows. Summary For determination of MFLs the District has chosen the measure of temporal loss as a measure of habitat change. This analysis requires that the relation between the target elevation at a specific point in the river corridor and the flow at a long-tem gage is reasonably well understood. It also requires long-t erm gage data. The spatial anal ysis requires the acquisition of cross-sections and the development of a model to calculate the habi tat loss. The resolution of the cross-sections and the use of top-width as a proxy for inundated area limit the accuracy of the analysis. For rivers currently being studied for MFL developm ent improved topographical data (i.e., LiDAR) is being collected. This will allow the development of digital elevation models and improved spatial analysis of floodplain inundatio n. Comparisons of these improved methods with the current methods should be performed. As instream fl ow professionals we are often asked to protect the environment from harm. Howe ver, ultimately what is measured is change and how much change constitutes harm must be determined. Therefore, we must know how different measures of change relate to each other.

PAGE 52

52 Table 3-1. Beginning Julian days for the Wet and Dry periods (Blocks 1 and 3) and ending date for the Wet period at four different gage stations in the SWFWMD. Begin Block 1 Begin Block 3 End Block 3 Alafia at Lithia 106 175 296 Hillsborough at Zephyrhills 112 176 296 Myakka at Sarasota 115 181 306 Peace at Arcadia 110 174 299 Mean 110 176 300

PAGE 53

53 Table 3-2 Mean flow requirement at the USGS Al afia River near Lithia gage needed to inundate floodplain features, and percen t of flow reductions associ ated with no more than a 15% reduction in the number of days that the required flows are equaled or exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999). Floodplain Feature Identified for Development of Minimum Flows and Levels for the Alafia River Mean (SD) Flow Requirements (m3/s) Percent-ofFlow Reduction from 1940 through 1969 Percent-of-Flow Reduction from 1970 through 1999 Low bank elevation 33 (17) 10 5 Low bank elevation for inundation of both sides of river floodplain 64 (18) 6 5 Highest floodplain vegetation class 101 (9) 9 5 Mean elevation of swamp classes 28 (12) 10 7 Highest swamp class 42 (22) 9 9 Floodplain wetted perimeter inflection point 22 (8) 12 9 Mean elevation of hydric soils 29 (11) 9 8 Highest elevation of hydric soils 58 (36) 5 7

PAGE 54

54 Table 3-3. Mean flow requirement s at the USGS Myakka River near Sarasota gage needed to inundate floodplain features, and percent of flow reductions associated with no more than a 15% reduction in the number of days that the required flows are equaled or exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999). Floodplain Feature Identified for Development of Minimum Flows and Levels for the Myakka River Mean (SD) Flow Requirements (m3/s) Percent-ofFlow Reduction from 1940 through 1969 Percent-of-Flow Reduction from 1970 through 1999 Lowest bank elevation to inundate one side of the river floodplain 9 (8) 15 8 Lowest bank elevation to inundation both sides of river floodplain 16 (19) 16 11 Median elevation of oak-palm wet hammock 24 (12) 13 8 Median elevation of oak-popash wet hammock 13 (5) 16 9 Median elevation of popash swamp 10 (3) 20 15 Median elevation of paragrass marsh 9 (4) 21 15 Median elevation of mixed marsh 1 (1) 72 68 Median elevation of panicum marsh 18 (12) 16 11 Median elevation of mucky soils 20 (16) 15 10 Median elevation of hydric soils 18 (13) 16 11 First major low inflection point on wetted perimeter 13 (8) 17 11 First major high inflection point on wetted perimeter 23 (26) 15 9

PAGE 55

55 Table 3-4. Mean flow requirements at the US GS Peace River at Arcadia gage needed to inundate floodplain features, and percent of flow reductions associated with no more than a 15% reduction in the number of days that the required flows are equaled or exceeded for two benchmark periods (1940 through 1969 and 1970 through 1999). Floodplain Feature Identified for Development of Minimum Flows and Levels for the middle Peace River Mean (SD) Flow Requirements (m3/s) Percent-ofFlow Reduction from 1940 through1969 Percent-of-Flow Reduction from 1970 through 1999 Lowest bank elevation to inundate one side of the river floodplain 115 (78) 7 8 Lowest bank elevation to inundate both sides of river floodplain 158 (57) 6 5 90th Percentile elevati on of river terrace vegetation 123 (47) 7 7 90th Percentile elevat ion of wet hardwood hammock 197 (72) 7 6 Mean elevation of river terrace vegetation 112 (66) 8 8 Mean elevation of wet hardwood hammock 160 (61) 7 6 Mean elevation of cypress swamp 82 (25) 12 8 Mean elevation of hardwood swamp 83 (34) 12 7 Mean elevation of mucky soils 66 (31) 13 10

PAGE 56

56 Figure 3-1. Map of the Southwest Florida Water Management District s howing the Alafia, Peace and Myakka rivers and the USGS gage locations used in this study on each.

PAGE 57

57 0 5 10 15 20 25 30 05001000150020002500 Flow at USGS Myakka near Sarasota Gage (cfs)Flow reduction (%) resulting in a 15% reduction in the number o f days the flow is reached 0 5 10 15 20 25 30 0500100015002000250030003500 Flow at USGS Alafia River at Lithia gage (cfs)Flow reduction (%) resulting in a 15% reduction in the number o f days the flow is reached 0 5 10 15 20 25 30 01000200030004000500060007000 Flow at USGS Peace River near Arcadia Gage (cfs)Flow reduction (%) resulting in a 15% reduction in the number o f days the flow is reached Figure 3-2. Percent-of-flow reductio ns that result in a 15% reducti on in the number of days that flows on the Alafia, middle Peace, and Myakka rivers are reached. Horizontal lines represent the flow reduction standards chos en by the SWFWMD for each river. Graphs are adapted from Ke lly et al. 2005a, b, and c.

PAGE 58

58 R2 = 0.9358 R2 = 0.9611 R2 = 0.9215 0 10 20 30 40 50 60 70 80 90 100 02468101214 Flow (cfs/mi2)Temporal percent of flow reduction Peace Alafia Myakka Figure 3-3. Percent-of-flow reductio ns that result in a 15% reducti on in the number of days that flows on the Alafia, middle Peace, and Mya kka rivers are reache d. Gage flow has been divided by the area of the watershed a bove the gage and power curves fit to the data for each river.

PAGE 59

59 R2 = 0.6042 R2 = 0.7499 R2 = 0.6404 0 10 20 30 40 50 60 70 80 90 100 02468101214Flow (cfs/mi2)Spatial percent flow reduction Peace Alafia Myakka Figure 3-4. Percent flow reducti on required to result in a 15% lo ss of top width plotted against flow for the Alafia, Myakka, and middle Peace rivers.

PAGE 60

60 Alafia0 10 20 30 40 50 60 70 80 90 100 020406080100Temporal percent of flow reduction Spatial percent of flow reduction Myakka0 10 20 30 40 50 60 70 80 90 100 020406080100Temporal Percent of flow reduction Spatial percent of flow reduction Middle Peace0 10 20 30 40 50 60 70 80 90 100 020406080100Temporal Percent of flow reduction Spatial percent of flow reduction Figure 3-5. Percent of flow reduction, correspon ding to a 15% spatial loss (measured as the summation of cross-section top width), plotte d against the percent of flow reduction, corresponding to a 15% temporal loss (meas ured as a loss of days during which the river historically reached a specific flow), for the Alafia, Myakka, and middle Peace rivers.

PAGE 61

61 CHAPTER 4 THE UTILIZATION OF GEOGRAPHIC IN FORMATION SYSTEMS IN DETERMINING FLOODPLAIN INUNDATION FOR MFL DEVELOPMENT Introduction In its response to the 1997 law requiring the development of minimum flows and levels (MFLs) in the state of Florida, the Southwest Fl orida Water Management Di strict (District) has developed minimum flows and levels for multiple freshwater rivers (Munson et al. 2005). As part of its MFL development, the District ha s acknowledged the importan ce of the natural flow regime and attempted to preserve it through the use of a seasonally sp ecific percent of flow reduction approach (Flannery 2002). The use of seasonally specific criteria to identify acceptable percent of flow reductions to be applied to each of a low, middle, and high flow period, addresses both the cyclical nature of the annual flow pattern and the natural variation of this pattern. Further, the use of multiple benchmark periods has served to recognize that longerterm climatic cycles effect flow variation on a multidecadal basis (Kelly 2004). By establishing a seasonally specific, block approach to the development of MFLs, the District required the development of criteria for each block that identified an appropriate percent of flow reduction to be used during each of the three seasonal blocks. Deve lopment of criteria to be applied during the low flow block and the me dium flow block was aided by methods currently employed in the establishment of instream flow requirements in other areas of the state. Specifically, the use of wetted perimeter, PHABSI M, and availability of snag habitat were employed in the development of low and medium flow criteria (Kelly et al. a,b,c 2005). Methods suitable for developi ng criteria for high season fl ow protection were less well documented and few examples were available. Floodplain wetlands associated with river syst ems have been recognized as an important component of the river corridors ecologi c community (Hughes and Rood 2003, Likens and

PAGE 62

62 Bormann 1974). However, for many years, the im portance of out-of-bank flows to the riverine ecosystems was not recognized by instream me thods used to assess flow requirements (Middleton 1999). Methods such as the Tennant method or the use of wetted perimeter can be serviceable in the protection of the lowest flow s. However, they deal predominately with between bank analyses and offer little to no pr otection for high flows, though both methods have been widely applied (Gippel and Stewards on 1998, Jowett 1997, Postal and Ricther 2003). While these methods offer valuable information fo r regulating the low end of the flow regime, it is necessary to measure and assess the flows require d at the high end of the flow regime as well. Munson and Delfino (2007) state that quantifi cation of potential differences in floodplain inundation associated with differences betw een minimum flow requirements and natural hydrologic regimes should be a critical component of regulatory programs that are intended to protect or restore riverine habi tat. The District has employed a method for establishing MFLs that acknowledges the importance of seasonal fl ow regimes, including flows necessary for floodplain inundation. The method empl oys the historic flow records and a temporal measure of habitat loss based on the reduction in the numbe r of days specified flows are reached under various flow reduction scenarios. An alternativ e approach, employing a measure of spatial loss that examined the change in top width at each cross section modeled under different flow conditions, was previously explored in th is dissertation (Munson and Delfino 2007). Both spatial and temporal measures of habitat loss can be related to percent flow reduction in a river channel. However, as previously disc ussed, both serve as proxies to evaluate the extent of the river-floodplain connection. This connection has not yet been fully described on any river studied by the District because complicated fl oodplain geometry is being represented by a limited number of historic floodplain cross-sectio ns derived from earlier studies. Though these

PAGE 63

63 cross-sections represented the best available da ta, their limitations were recognized, especially with regards to in-channel accuracy. This chapter describes the role that Geogra phic Information Systems (GIS), coupled with hydraulic models, can play in MFL developmen t by providing high quality data concerning floodplain inundation under different flow scenarios. As Overton (2005) points out, riverine ecosystems benefit from spatial analysis since it provides temporally dynamic data in three spatial dimensions. This coupling of a one-dim ensional hydraulic model and spatial information from GIS results in a powerful tool for re lating variations in floodplain inundation with variations (or reducti ons) in river flow. Study Area The Braden River is the largest tributary of the Manatee River located in Southwest Florida (Figure 4-1). It drains approximately 215 km2 and is composed of three distinct segments. The most downstream segment is an approximately 10-km long estuarine reach, which joins with the Manatee River near its mo uth. Immediately upstream of this is another approximately 10-km reach of river impounded by a broad-crested weir, creating the Ward Lake reservoir. Above this is a 13-km segment of naturally flowing incised channel (DelCharco and Lewelling 1997). The Braden River was selected for this study by virtue of its listing on the MFL priority list and because its size made it an efficient option for the utilization of new technology. The river has three USGS gage locations within the study reach. These include, from upstream to downstream, the Braden River at Lorraine (USGS# 02300029), the Braden River near Lorraine (USGS# 02300032), and the Braden River at Linger Lodge (USGS# 023000358). Flows in the Braden River are less substantial than other rivers in the ar ea. In part, this is due to the watershed, though development and attenuation in the watershed may play a role.

PAGE 64

64 Flows do not exceed 100 cfs until higher than th e 10% exceedance flow. This, combined with the incised nature of the channel and well-draine d sandy soils, results in the expectation that there is little floodplain inundati on except under the highest flows. Methods For development of an accurate digital elevation model (DEM), three-dimensional topographic data are required. For the Braden River, these data we re gathered through the use of Light detection and Ranging (LiDAR). In 2005, a qualified photogrammetric firm conducted the LiDAR and orthophotography acquisition for the Braden River project. The flights utilized an ALS40 LiDAR system flying at 5000 ft with a 30 -degree field of view and 20% side overlap. Acquisition of LiDAR data used a 2-m post spacing interval, di gital one-foot orthophotographs and 3D breakline features necessary to meet a one-foot contour interval product. The vertical accuracy of the data was specified at 10 cm in homogeneous, unambiguous terrain. All LiDAR data were collected using the North Am erican Vertical Datum of 1988 (NAVD88). For the Braden River, a Tria ngular Irregular Network (TIN ) was used as the digital elevation model. The TIN was generated from a combination of LiDAR data, break lines, and a limited number of surveyed cross-sections (Figure 4-2). The TIN was then used to generate a series of cross sections for export into the H ydrologic Engineers Centers-River Analysis System (HEC-RAS; Figure 4-3). Known water surface el evations were used as downstream boundary conditions and a rating curve, supplied by USGS, was used to calibrate to the USGS near Lorraine gage (USGS# 02300032) All elevations, associated w ith USGS gages, were converted to a NAVD88 standard when necessary. The model was considered calibrated when calculated values of water surface elevation were within 0.5 ft of the rated value. Once the model was complete, data from the model were imported back into the GIS environment. Essentially, these data consist of the cross-section locations and ground elevations

PAGE 65

65 from the model and water surface elevations associat ed with different flow profiles. In the case of the Braden River, the model generated flow pr ofiles for twenty-one even ly distributed percent exceedance flows. For each of these flow prof iles, a water surface elevation was calculated at each cross-section and represented in GIS as a seri es of lines at each cros s-section with different elevation. So, each specified flow corresponds in GIS to a series of lines with elevation from which a second DEM can be generated in the fo rm of a TIN. This TIN becomes a TIN of calculated water surface elevation based on the model. Once both terrain surface elev ation and water surface elevat ion models are generated, analysis can be performed in a GIS environment, which clips the water surface elevation model to the terrain model. The use of an extensi on to the GIS software titled GEO-RAS automates this process, resulting in a rast er surface. A raster surface is a form of digital elevation model which uses a matrix of same size squares with each one having an elevation based on the elevation of the data points with in its bounds. Raster cells do not slope, and thus all the area falling within a cell is assigned the same eleva tion. The new DEM, generated by subtracting the land surface from the water surface, is a water su rface model where each cell represents a depth of water over the land surface (Figure 4-4). Th is allows for the esti mation floodplain inundation in a river corridor based on thr ee dimensional floodplain topology rather than two-dimensional cross-sections and inte rvening channel length. Results This study utilized a combinati on of LiDAR data and survey da ta to generate a DEM of the freshwater portion of the Braden River. This DEM provides useful information for the development of cross-sections in the hydraulic m odel. Further, the output from the model is used to generate a water surface profile in GIS and to analyze inundation patterns.

PAGE 66

66 The LiDAR points, used to generate the mode l, were numerous and dense in the upland area of the digital elevation model. However, poi nts were sparser in the forested portion of the floodplain wetland and in the river corridor itself. A typical bend in the river and the lack of corresponding LiDAR data within the floodplain are shown in Figur e 4-5. This trend continued for the length of the study corridor. In the downstream half of the study area, br eak lines were added along each bank. This results in an approximation of the floodplain area by defining the upland portion with LiDAR and the riverÂ’s edge with a hard break line. This is useful in the TIN because it keeps the river channel well defined, even in the absence of high-d ensity LiDAR data near the river. However, the more upstream portion of the study lacked th ese break lines and the river channel is more poorly confined in this portion of the DEM. Su rvey data also provided localized improvement of the topography TIN and resulted in better develo pment of the DEM in areas where survey data augmented the LiDAR data. However, these data resulted in only localiz ed improvements in the topography TIN. The hydraulic model estimates an inundated ar ea by calculating the area of a trapezoid. The trapezoid is bound by the water surface extent at two adjacent cross-s ections. The distance between cross-sections is defined in the model. In a GIS envir onment, it is possible to estimate the inundated area between cross-sections by ca lculating the area where the water surface DEM is higher then the topography DEM (Figure 46). The inundated area calculated by GIS and HEC-RAS between two cross-sectio ns is presented in Table 4-2. Discussion GIS to HEC-RAS The use of LiDAR to augment survey data was initially viewed as a device for supplementing partial floodplain survey work. Th e concept was that cro ss-sections could be

PAGE 67

67 established manually across the river and then, subsequently, those cro ss sections could be lengthened, and potentially, additional cross-secti ons could be added, which are generated from the LiDAR data. This was not th e case with the Braden River. Cross-sectional data aligned well with LiDAR-d erived data where overlap occurred. The generated DEM was used to create cross-secti on lines, which were imported into HEC-RAS. The DEM allowed surveyed cross-sections to be lengthened and exte nded beyond the floodplain. However, the quality of the DEM in areas where the LiDAR data were sparse and survey data did not exist, resulted in cro ss-sections which were too poorly defined within the floodplain and river channel to be useful in a hydraulic model. It should be noted that this was due not to inaccu racy in the data but to a localized lack of data. Relatively few additional data points would be needed to significantly improve the accuracy of the DEM. The acquisition of a thalwag alone would add considerable accuracy to the DEM. It should also be noted that the Li DAR data for the Braden River corridor were among the first collected in the SW FWMD and that specifications were far less rigorous than more recently collected data. Preliminary results from the Withlacoochee River indicate that 4foot post spacing and newer systems are pe netrating the floodplai n canopy and providing regularly spaced data. The data in the floodplai n is less dense than the non-canopy areas flown, but significantly improved ove r the Braden River data. HEC-RAS to GIS The utilization of output from HEC-RAS to model floodplain inundati on were useful but fell short of creating really mean ingful results. The focus of future studies should be on the generation of a high quality DEM of the terrain. This can be accomplished by supplementing the LiDAR data with hydrographic survey data. Hydrographic surveys can be accomplished through both manual and electronic means. A combinat ion of sonar, survey and global positioning

PAGE 68

68 satellite (GPS) methods can be used to define a thalwag and provide credible sub-surface data to be used in conjunction with LiDAR. As not ed above, LiDAR capabilities have increased significantly in the past few years as increased demand has resulted in increased precision. It should be noted that because the water surface elevation TIN is limited in geographic range by the extent of the cross-sections, and b ecause the bounding polygon, generated by the GIS extension, is defined by the cross-sections, thos e cross-sections should be chosen carefully. HEC-RAS offers approximations of inundation based on trapezoidal areas (Figure 4-6). It should be noted that Figure 4-6 is conceptual since HEC-RAS uses ch annel length and not geographic distance as the distan ce between cross-sections. Howe ver, inundation, in a broader, more complex floodplain than the Braden River, would not necessarily be well modeled in this manner without a significant number of cross-sec tions to capture variability in the floodplain. The GIS and HEC-RAS estimates, for inundated area in the section of the Braden River shown in Figure 4-6, are presented in Table 4-2. GIS in MFL Development It has already been noted that historically instream-flow requirements have been focused largely on the low-flow portion of the flow regi me. This is probably due in part to the complexity of floodplain hydrology. As Mert es (1997) noted, floodpl ain inundation does not rely solely on overbank flows from the associ ated river. GIS provi des the opportunity, when coupled with high quality data, to analyze floodplain inundation more accurately than was previously practical. Initially, this can impr ove MFL development by allowing more accurate assessment of habitat changes asso ciated with changes in flow. Future Work The purpose of this research was to generate a hydraulic model to be used in the development of a MFL for the Braden River and al so to determine the usefulness of LiDAR data

PAGE 69

69 and the role they can play in MFL determina tions. The Braden River LiDAR data represent approximately 26 km2 of area. Generated as part of a larg er flight plan, the data generally cost approximately $700/ km2 to collect and process, though it can vary depending on the size and shape of the area to be flown. Data collec tion has begun on the Withlacoochee River, which includes approximately 310 km2 of necessary data. Prior to an investment of that size, it was prudent to determine to what extent the data wa s useful and what subsurface data are necessary to generate high quality results. This project demonstrated that LiDAR, combined with ground surveys, can be used to generate cross-sections for the development of a hydraulic model. However, the LiDAR is best used to supplement floodplain portions of a cros s section. LiDAR alone results in insufficient instream data to generate an accu rate cross-section. This research also demonstrated that where LiDAR data are sufficiently augmented with high quality instream and below canopy data, GIS can be used to predict inunda tion patterns and depths unde r different flow conditions. LiDAR data for the Withlacoochee River will be coupled with a hydrographic survey to produce a highly accurate DEM for the length of the study corridor. The hydrographic survey will include a combination of electronic data coll ected with sonar and manually surveyed data. A thalwag of the river will also be included. The Withlacoochee River study will be the first major river in Florida to have a highly accurate DEM generated which will be used in future studies of the river including MFL development.

PAGE 70

70 Table 4-1. Percent exceedance flows for the Braden River near Lorraine gage (USGS# 02300032). Percent Exceedance Exceedance flows (%) (cfs) 99 0.67 95 1.02 90 1.40 85 1.73 80 2.04 75 2.47 70 2.87 65 3.35 60 4.09 55 5.12 50 6.22 45 7.45 40 10.25 35 13.59 30 17.95 25 25.16 20 36.36 15 54.41 10 89.53 5 188.82 1 647.47

PAGE 71

71 Table 4-2. Estimated acres of inundated land at different exceedance flows between two Braden River cross-sections shown in Figure 4-6. Percent GIS Results HEC-RAS Results Exceedance (acres) (acres) 90%Ex 1.12 1.26 50%Ex 1.14 1.26 25%Ex 1.33 1.30 10%Ex 1.95 1.49 1%Ex 5.49 7.10

PAGE 72

72 Braden R ive rManat e e R i ver Figure 4-1. Location map of the Brad en River in southwest Florida.

PAGE 73

73 Figure 4-2. Portion of the digi tal elevation model generated of the Braden River, Florida.

PAGE 74

74 Figure 4-3. Location of surveyed cross-sec tions across the Braden River, Florida.

PAGE 75

75 Figure 4-4. Portion of the Braden River, Flor ida with modeled inundation patterns from a 1% exceedance flow.

PAGE 76

76 Figure 4-5. LiDAR points used in the DEM of the Braden River, Florida. LiDAR points are yellow dots, which are numerous enough to show as dark areas on the map. Empty areas are where data was determined not to represent the ground elevation (i.e., Houses and canopied areas where the Li DAR did not penetrate to the ground.

PAGE 77

77 Figure 4-6. Conceptual view of inundated area calculated by HEC-RAS and in a GIS environment. The shaded blue area repres ents inundation patterns modeled in GIS. The red trapezoid is a depiction of how HEC-RAS would estimate the inundated area.

PAGE 78

78 CHAPTER 5 SUMMARY, CONCLUSIONS AND RECO MMENDATIONS FOR FUTURE WORK Summary This dissertation reviewed the water management policies in Florida, which eventually led to a legislative mandate for MFL (minimum flows and levels) development. The five water management districts, tasked with MFL deve lopment, have responded using a variety of methods. These methods include utilization of multi-parameter appr oaches, including both hydrologic and biological components, as well as r ecognition of public water supply needs. As of the publication of the 2005 MFL priority list, minimum flows or levels had been established for 242 water bodies in four of the five water mana gement districts. The fifth had an established priority list identifying wa ter bodies within its bounda ries for which MFLs were in the process of being established. The development of the minimum flow standard s used to protect riverine ecosystems from reductions in the highest part of the flow regime was the focus of this dissertation, because it is the least understood. Water managers frequently utilize methods describe d in the literature for the protection of in-channel flows. These included the use of physical habitat simulation analysis, fish passage, and lowest wetted perime ter inflection point in determining appropriate flow standards. River/floodplain interaction is less well understood in terms of establishing flow requirements for protection of the riverine floodplain ecosystem. Two methods of measuring hab itat loss in floodplain wetlands were evaluated. The use of a temporal measure of habitat lo ss as a criterion to develop minimum flows for protection of riverine floodplains was examin ed for three rivers in Florid a. The appropriateness of the temporal measure was compared to a spatial meas ure of habitat loss, a nd an alternative method

PAGE 79

79 using GIS for measuring floodplain losses, associat ed with water withdrawals, was presented and discussed. The temporal loss of habitat measure is inte nded to provide protection of the riverine ecosystems by limiting the reduction in the number of days of connection between the river and the floodplain. Research showed that the relation ship between the flow and the percent-of-flow reductions required to generate a 15% decrease in the number of days a certain flow is reached annually is similar for all three rivers studied. The similarity among rivers is highlighted when flows are plotted in terms of flow per watershed area. This research also utilized an alternat ive approach to determining habitat loss by employing a spatial measure of habitat, deri ved from the HEC-RAS model. This method summed the lengths of inundation along each crosssection in the hydraulic model as a proxy of area of inundation. This was done because it was im portant to examine if different measures of habitat relate differently to change s in flow. Further, if they di d respond differently to changes in flow, it was necessary to characterize which measure would be limiting in terms of flow reduction if a 15% loss of habitat criteria was applied. The reductions in flow required to generate a 15% loss of inundated width were calc ulated and the results for different flows per watershed area for all thr ee rivers were plotted. Calculations were preformed to relate habi tat types to specific periods of inundation. Wetland characteristics were examined and the am ount of flow reduction that would result in a 15% loss of inundation was determined. The resu lts for each wetland characteristic among the cross-sections of a specific river varied widely. Vegetative crosssections were selected based on location of shoals within the river and vegetatio n communities. Cross-sections were selected which were characteristic of the river corridor However, as it became evident that it was

PAGE 80

80 difficult to characterize a riverine floodplain with a limited number of cross-sections, alternatives to linear measures of sp atial habitat were sought. Utilizing LiDAR data, gathered in the Braden River basin, a digital terrain model of the Braden River corridor was generated. This was done to examine how useful digital terrain models might be in examining floodplain wetland inundation in a GIS environment, rather then simply along linear cross-sections. The accuracy of the digital terrain model was limited by the quality of the LiDAR data, which were among th e earliest such data co llected in southwest Florida. Inaccuracies were due mainly to the la ck of bare earth data, collected in the canopied floodplain. Cross-sections from the digi tal terrain model were imported into the HER-RAS hydraulic model and water surface profiles were calculated. These profile s were then exported into GIS where a digital elevation model of the water surface was created. Th e water surface map was then layered with the terrain map and the extent of inundation at differe nt flows was assessed. This was done to develop a better measure of spatial ha bitat loss, which coul d ultimately lead to a better understanding of the inundation re quirements of riverine floodplains. Conclusions The 1997 legislative mandate to develop minimu m flows and levels in Florida has placed the state among the leaders of forward-looking water resour ces protection in the United States. Minimum flow requirements, for protect ion of riverine we tland protection, are conceptually understood but poorly quantified. The measure of temporal loss among rivers in this study are highly similar due to similarities in flow characteri stics of the rivers and do not account, directly, for differences in floodplain morphology. Abrupt changes in slope, when plotting spat ial measures of habitat loss, are likely associated with variations in channel and floodplain morphology, and not accounted for with the temporal measure of habitat loss.

PAGE 81

81 The use of a temporal measure to assess habitat loss is approp riate since floodplain characteristics, at least in part, are derived from fluvial processes. LiDAR is a useful tool in the deve lopment of digital terrain models. Digital terrain models should be used, in conj unction with hydraulic models, to estimate a spatial measure of habitat loss. GIS offers the opportunity to better estimate inundation in the floodplain as a result of over-bank flow, and thus spatial habitat loss, as a result of ch anges in river flow. The use of GIS in determining spatial loss is pr eferable to the use of cross-sectional data because the three dimensional nature of th e spatial data resolves some issues of characterizing the river corridor with a limited number of cross-sections. Hydrology is currently the dominant fact or in establishing high-flow MFLs. MFLs are intended to protect the aquatic biology and ecology of the river system, though to date, biology and ecology have only been minor components in the development of high-flow protection standards. Current methods of MFL development do not differentiate flow needs based on wetland habitat type. Better modeling ability, derived from GIS, should allow improved understanding of floodplain community structure relative to flow. Future Work This dissertation has shown that the MFL re quirements for riverine floodplains wetlands are poorly understood. The conceptual im portance of over-bank flows has been well documented but not well quantified. This research pointed out that habitat and change in habitat can be measured in multiple ways and it explored the use of new technology in measuring habitat extent. These measures suggest future studie s that should be conducte d, given the information presented in this dissertation. Spatial loss in river floodplains is inconsistent from river to river while temporal loss is similar among the three rivers studied and both should be consid ered in MFL development. Thus, both spatial and temporal measures should be examined in future MFL determinations.

PAGE 82

82 Flow reductions, on rivers studied in this di ssertation, consistently resulted in higher habitat loss when measured by the temporal measure of habita t loss rather than the spatial measure of habitat loss. It will be important to determine if this is a consistent finding for other rivers in the area. If this relationship proves to be inconsistent, then both th e spatial and temporal methods should be applied to each river and the more rest rictive one utilized. This will assure that no more than a 15% loss of habitat would be allowed due to water w ithdrawals, whether habitat is measured spatially or temporally. It is recomme nded that in the future, the flow reductions from the temporal analysis be evaluated in GIS to de termine the corresponding spatial loss of habitat. If the resulting spatial loss is calculated to ex ceed the temporal loss, consideration should be given to this loss when determining the MFL. Use of GIS as a spatial mode ling tool provides the means fo r better understanding of the river floodplain interaction. Inundation maps combined with vegetation maps (i.e., National Wetlands Inventory Maps) may provide a better unde rstanding of the inunda tion requirements of various wetland types and these re lations should be investigated. It is necessary to measure not only the accur acy of the MFL, but also to evaluate the accuracy of some of the underlying assumptions, a nd thus the effectiveness of the MFLs. This requires the implementation of biological mon itoring and a long-term co mmitment to evaluate the effectiveness of MFLs. This should begi n with the monitoring of the wetland extent at various locations in each river system. This could be spatially mapped. Changes over time could be noted and correlated to changes in inundation as a result of changes in flow. The LiDAR data, used to generate the Braden River digital elevation model, had gaps in the canopied area. Current LiDAR technology ha s improved and these improved data should be

PAGE 83

83 coupled with ground surveys to supplement areas of sparse LiDAR data. This combination of remote and ground surveys can result in a dens e enough network of points to generate an accurate and detailed digital elevation model of the river/floodplain system. These techniques should be pursued and incorporated into future MFL determinations.

PAGE 84

84 LIST OF REFERENCES Ainsle, W.B., B.A. Pruitt, R.D. Smith, T.H. R oberts, E.J. Sparks and M. Miller. 1999. A regional guidebook for assessing the functions of low gradient ri verine wetlands in western Kentucky. U.S. Army Corps of Engineers Wate rways Experiment Station. Technical Report WRP-DE-17, Vicksburg, Mississippi. Anderson, D.L. and P.C. Rosendahl 1998. Development and management of land/water resources: the Everglades, agri culture, and south Florida. Jo urnal of the American Water Resources Association 34: 235-248 pp. Bachmann, R. W., M. V. Hoyer and D. E. Canfie ld, Jr. 1999. The restoration of lake Apopka in relation to alternative stable states. Hydrobiologia 394:219-232. Berryman and Henigar. 2004. Characterization of wetland vegetation communities and hydric soils along the middle Peace River. Prepared for the Southwest Florida Water Management District. Brooksville, FL. Blake, N. M. 1980. Land into water-water into land. University Press of Florida, Tallahassee, FL. 344 pp. Bovee, K.D., B.L. Lamb, J.M. Bartholow, C.B. Stalnaker, J. Taylor and J. Hendrickson. 1998. Stream habitat analysis using the instream flow incremental met hodology. U.S. Geological Survey, Biological Resources Division Info rmation and Technology Report USGS/BRD-19980004. Brinson, M.M., B.L. Swift, R.C. Plantico and J. S. Barclay. 1981. Riparian ecosystems: their ecology and status. U.S. Fish and Wildlife Se rvice, Biological Services Program Report FWS/OBS-81/17, Washington, D.C. Canfield, D. E., Jr., R. W. Bachmann and M. V. Hoyer, 2000. A management alternative for Lake Apopka. Lake and Rese rvoir Management 16: 205-221. Carriker, R. R., 2000. Florida's water: supply, use, and public policy, Institute of Food and Agricultural Science, University of Florida, Gainesville, FL. Conner, W.H. and J.W. Day. 1976. Productivity a nd composition of a bald cypress-water tupelo site and a bottomland hardwood site in a Loui siana swamp. American Journal of Botany 63: 1354-1364. Crance, J.H. 1988. Relationships between palust rine forested wetlands of forested riparian floodplains and fishery resources: a review. U.S. Fish and Wild life Service, Biological Report 88(32), Washington, D.C. DelCharco, M. J., and B. R. Lewelling. 1997. H ydrologic description of the Braden River Watershed, West-Central Florida. United St ates Geological Survey Open Report 96-634.

PAGE 85

85 DEP (Department of Environmental Protec tion). 1995. Florida water plan 1995. Florida Department of Environmental Protection, Office of Water Policy, Tallahassee, FL. DEP (Department of Environmenta l Protection). 2000. Florida's wa ter supply: will there be enough water in 2020, DEP's annual status repor t on regional water supply planning, Florida Department of Environmental Protection, O ffice of Water Policy, Tallahassee, FL. 19 pp. DEP (Department of Environmental Protection) 2002a. Implementing regional water supply plans: is progress being made, DEP's annual status report on regional water supply planning. Florida Department of Environmenta l Protection, Tallahassee, FL. 25 pp. DEP (Department of Environmental Protection). 2002b. Florida water conservation initiative. Florida Department of Environmen tal Protection, Tallahassee, FL. DEP (Department of Environm ental Protection). 2003a. 2003, Annual status report on regional water supply planning Florida Department of Environm ental Protection, Tallahassee, FL. DEP (Department of Environmental Protec tion). 2003b. 2002 Reuse inventory. Florida Department of Environmental Protection, Tallahassee, FL. DEP (Department of Environmental Protection) 2003c. 2002 Florida water plan; 2002 Annual Progress Report, Florida Department of Environmental Protection, Tallahassee, FL. Derr, M. 1989. Some kind of paradise. William Morrow and Co., New York, NY. 416 pp. Douglas, M. S. 1947. The Everglades; river of grass. Rinehart, New York, NY. 487 pp. Dunbar, M.J., A. Gustard, M.C. Acreman and C.R. Elliott. 1998. Overseas approaches to setting river flow objectives. Institute of Hydr ology. R&D Technical Report W6-161. Oxon, England. 83 pp. Egozcue, D. K. 2001. The history of water in the Tampa Bay area. Master's Thesis, Department of Geography, University of South Florida, Tampa, FL. 126 pp. Enfield, D., A. Mestas-Nunez and P. Trimble. 2001. The atlantic multidecadal oscillation and its relation to rainfall and river flows in th e continental U.S. Geophysical Research Letters 28(10): 2077-2080. Flannery, M.S., E. Peebles and R. T. Mont gomery. 2002. A percent-of-flow approach for managing reductions of freshwater inflows fr om unimpounded rivers to southwest Florida esturies. Estuaries 25: 1318-1332. Gippel, C. J. and M. J. Stewardson. 1998. Use of wetted perimeter in defining minimum environmental flows. Regulated Rivers 14: 53-67

PAGE 86

86 Gore, J. A., C. Dahm and C. Klimas. 2002. A review of "Upper Peace Ri ver: an analysis of minimum flows and levels". Prepared for the Southwest Florida Water Management District. Brooksville, FL. http://www.sw fwmd.state.fl.us/documents/ Gregory, S.V., F.J. Swanson, W.A. McKee a nd K.W. Cummins. 1991. An ecosystem perspective on riparian z ones. Bioscience 41: 540-551. Hamann, R. 2001. Consumptive use permitting criteria. In: Florida environmental and land use law, The Florida Bar, Tallahassee, FL. Hammett, K.M., J.F. Turner, Jr. and W.R. Mu rphy, Jr. 1978. Magnitu de and frequency of flooding on the Myakka River, Sout hwest Florida. Department of the Interior. United States Geological Survey. Water-Resources Investigati ons, Open-File Report 78-65, Tallahassee, FL. Henry, J. A. 1998. Weather and climate. Pp. 1637 in Water resources atlas of Florida. E. A. Fernald and E. D. Purdum (eds.). Institute of Science and Public Affairs, Florida State University, Tallahassee, FL. Hill, J. E. and C.E. Cichra. 2002. Minimum flow s and levels criteria development. Evaluation of the importance of water depth and frequenc y of water levels / flows on fish population dynamics, literature review and summary: the e ffects of water levels on fish populations. Institute of Food and Agricultural Sciences, De partment of Fisheries and Aquatic Sciences. University of Florida. Gainesville, FL. 40 pp. Hill, M.T., W.S. Platts and R.L. Beschta. 1991. Ecological and geological concepts for instream and out-of-channel flow requi rements. Rivers 2: 198-210. Hughes, F. M. R. and S. B. Rood. 2003. Allocat ion of river flows for restoration of floodplain forest ecosystems: A review of approaches a nd their applicability in Europe. Environmental Management 32(1): 12-33 Hupalo, R., C. Neubauer, L. Keenan, D. Cla pp and E. Lowe. 1994. Establishment of minimum flows and levels for the Wekiva River system. Technical Publication SJ 94-1. St. Johns River Water Management District, Palatka, FL. Jowett, I.G. 1993. Minimum flow requirements for instream ha bitat in Welllington rivers. NZ Freshwater Miscellaneous Report No. 63. National Institute of Water and Atmospheric Research, Christchurch, New Zealand. 33 pp. Jowett, I.G. 1997. Instream flow methods: A co mparison of approaches. Regulated Rivers 13: 115-127 Junk, W. P., P.B. Bayley and R.E. Sparks. 1989. The flood pulse concept in river-floodplain systems. Proceedings of the international larg e river symposium. Special Publication of the Canadian Journal of Fisheries and Aquatic Sciences 106: 110-127

PAGE 87

87 Kallina, E. F. 1993. Claude Kirk and the politics of confrontation. University Press of Florida, Gainesville, FL. 253 pp. Kelly, M.H. 2004. Florida river flow patterns a nd the atlantic multidecada l oscillation. Ecologic Evaluation Section. Southwest Florida Water Mana gement District. Brooksville, FL. 80 pp. + appendix http://www.swfwmd.state.fl.us/documents/ Kelly, M. H., A. B. Munson, J. Morales and D. L. Leeper. 2005a. Alaf ia River minimum flows and levels; freshwater segment including L ithia and Buckhorn Springs. Ecologic Evaluation Section. Southwest Florida Water Management District, Brooksville, FL. 188 pp + appendix http://www.swfwmd.state.fl.us/documents/ Kelly, M. H., A. B. Munson, J. Morales and D. L. Leeper. 2005b. Proposed minimum flows and levels for the upper segment of the Myakka Ri ver, from Myakka City to SR 72. Ecologic Evaluation Section. Southwest Florida Water Ma nagement District, Brooksville, FL. 144 pp + appendix http://www.swfwmd.state.fl.us/documents/ Kelly, M. H., A. B. Munson, J. Morales and D. L. Leeper. 2005c. Proposed minimum flows for the middle segment of the Peace River, from Zolfo Springs to Arcadia. Ecologic Evaluation Section. Southwest Florida Water Management District, Brooksville, FL. 177 pp + appendix http://www.swfwmd.state.fl.us/documents/ Kuensler, E.J. 1989. Values of forested wetlands as filters for sediments and nutrients. Pp. 8596 In: D.D. Hook and R. Lea (eds.), Proceedings of the Symposium: the forested wetlands of the United States. USDA Forest Service, Southeas tern Forest Experimental Station, General Technical Report SE-50. Likens, G. E. and F. H. Bormann. 1974. Linkage s between terrestrial an d aquatic ecosystems. Bioscience 24(8): 447-456. Lowe, E. F., L. E. Battoe, M. Coveney and D. Stites. 1999. Setting water quality goals for restoration of Lake Apopka: in ferring past conditions. Lake an d Reservoir Management 15: 103120. Maloney, F. E., S. J. Plager and F. N. Baldwi n. 1968. Water law and admi nistration: the Florida experience. University of Florida Press, Gainesville, FL. 488 pp. Maloney, F. E., R. C. Ausness and J. S. Morri s. 1972. A Model Water Code. University of Florida Press, Gainesville, FL. 373 pp. Maloney, F. E., S. J. Plager, R. C. Ausness and B. D. E. Canter. 1980. Florida water law. Water Resources Research Center, University of Florida, Gainesville, FL. 762 pp. Matthews, F. E. and G. E. Nieto. 1998. Florid a water policy: a twenty-five year mid-course correction. Florida State Univer sity Law Review 25: 365-390.

PAGE 88

88 McCabe, G. and D. Wolock. 2002. A step increa se in streamflow in the conterminous United States. Geophysical Rese arch Letters 29: 2185-2188. Mertes, L. A. K. 1997. Documentation and si gnificance of the perirh eic zone on inundated floodplains. Water Resources Research 33(7): 1749-1762. Middleton, B. 1999. Wetland rest oration: flood pulsing and dist urbance dynamics. John Wiley and Sons, Inc. New York, NY. 388 pp. Munson, A. B., and J. J. Delfino. 2007. Minimu m wet-season flows and levels in southwest Florida rivers. Journal of the American Water Resources Association. Accepted for publication. Munson, A. B., J. J. Delfino and D. A. Leeper 2005. Determining mini mum levels: the Florida experience. Journal of the American Water Resources Association. 41(1):1-10. Murphy, W. R., K. M. Hammett and C. V. Reeter. 1978. Flood profiles for Peace River, southcentral Florida. U.S. Geological Survey, Wate r resource Investigation 78-57, Tallahassee, FL. Olsen, J. R., J. R. Stedinger, N. C. Matalas and E. Z. Stakhiv. 1999. Climate variability and flood frequency estimation for the upper Mississippi and lower Missouri rivers. Journal of the American Water Resources Association. 35 (6):1509-1523. I. C. Overton. 2005. Modeling floodplain inunda tion on a regulated rive r: integrating GIS, remote sensing and hydrological mode ls. Regulated Rivers 21: 991-1001. Poff, N. L., D. Allan, M.B. Bain, J.R. Darr, K.L. Perstegaard, B.D. Richter, R.E. Sparks and J.C. Stromberg. 1997. The natural flow regime; A para digm for river conservation and restoration. Bioscience 47: 769-784. Postel, S. and B. D. Richter. 2003. Rivers for life; managing water for people and nature. Island Press, Washington D. C. 253 pp. Purdum, E. D., L. C. Burney and T. M. Swihar t. 1998. History of water management, In: Water Resources Atlas of Florida, E. A. Fernald and E. D. Purdum (Editors). Institute of Science and Public Affairs, Florida State University, Tallahassee, FL. Richter, B.D., J.V. Baumgartner, J. Powe ll and D.P. Braun. 1996. A method for assessing hydrologic alteration within ecosystem s. Conservation Biology 10: 1163-1174. Richter, B.D., J.V. Baumgartner, R. Wigingt on and D.P Braun. 1997. How much water does a river need? Freshw ater Biology 37: 231-249. Rosenau, J. C., G. L. Faulkner, C. W. Hendry Jr and R. W. Hull. 1947. Springs of Florida. Florida Geologic Survey, Tallahassee, FL. Ross, C. I. 2001. Minimum flow s and levels. In: Florida envir onmental and land use law, The

PAGE 89

89 Florida Bar, Tallahassee, FL. Shafer, M. D., R. E. Dickinson, J. P. Heaney and W. C. Huber. 1986. Gazetteer of Florida lakes. Water Resources Research Center, University of Florid a and United States Geological Survey, Gainesville, FL. 256 pp. Shofner, J. H. 1982. History of Apopka a nd northwest Orange County, Florida. Apopka Historical Society, Apopka, FL. 357 pp. Stalnaker, C.B. 1990. Minimum flow is a myt h. Pp. 31-33 In: M.B. Bain (ed.), Ecology and assessment of warmwater streams: workshop synops is. Biological Report 90(5). U.S. Fish and Wildlife Service, Washington, D.C. Stalnaker, C., B.L. Lamb, J. Henriksen, K. Bovee and J. Bartholow. 1995. The instream flow incremental methodology: a primer for IFIM. Bi ological Report 29. U.S. Department of the Interior, National Biological Serv ice, Washington, D.C. 46 pp. Stanturf, J.A. and S.H. Schoenholtz. 1998. Soils and landforms in southern forested wetlands In: Ecology and management, M.G. Messina and W.H. Conner (e ds.) CRC Press-Lewis Publishers, Boca Raton, FL. SWFWMD (Southwest Florida Water Management District). 1996. Lakes level program annual report. Southwest Florida Water Mana gement District, Brooksville, FL. SWFWMD (Southwest Florida Water Management District). 2001. Regional water supply plan. Southwest Florida Water Manageme nt District, Br ooksville, FL. SWFWMD (Southwest Florida Water Management District). 2002. Upper Peace River: an analysis of minimum flows and levels. Sout hwest Florida Water Management District, Brooksville, FL. Texas Parks and Wildlife. 2005. Freshwater inflow needs of the Matagorda Bay system. http://www.tpwd.state.tx.us/la ndwater/water/conservation/coas tal/freshwater/matagorda/ Accessed on October 12, 2005. Tjoflat, M. P. and I. K. Quincey. 2001. Florid a water planning In: Florida environmental and land use law, The Florida Bar, Tallahassee, FL. U.S. Army Corps of Engineers. 2001. HEC-RAS river analysis system user's manual. US Army Corps of Engineers, Davis, CA. U.S. Congress. 1911. Everglades of Florida. 62nd Congress. Doc. 89, Washington, D.C. Vannote, R.L., G.W. Minshall and K.W. Cummins. 1980. The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37: 130-137.

PAGE 90

90 Walbridge, M.R. and B.G. Lockaby. 1994. Effect of forest management of biogeochemical functions in southern forested wetlands. Wetlands 11: 417-439. Wharton, C.H., W.M. Kitchens, E.C. Pendl eton and T.W. Sipe. 1982. The ecology of bottomland hardwood swamps of the southeast: a community profile. U.S. Fish and Wildlife Service FWX/OBS-81/37, 133 pp. Woltemade, C. J. 1997. Water level management opportunities for ecological benefit, Pool 5 Mississippi River. Journal of the American Water Resour ces Association 33(2): 443-454.

PAGE 91

91 BIOGRAPHICAL SKETCH Born and raised in southwest Florida, Adam Munson has spent his academic and professional career working to pr otect local water bodies. A gra duate of Vivian Gaither High School in north Tampa, Adam attended the Univer sity of Florida as an undergraduate majoring in mechanical engineering. Graduating with a B.S.M.E. in 1994, AdamÂ’s interest in natural systems and science led him to the Department of Fisheries and Aquatic Sciences at the University of Florida, from which he gradua ted in May of 1999 with a M.S. specializing in limnology. After a brief period as ecosystem ma nager for the St. Marks and Ochlockonee River basins in North Florida, Adam returned to the southwest Florida area where he evaluates instream flow requirements for the development of minimum flows and levels.