Citation
Dissolved Oxygen as Related to Minimum Flows and Levels

Material Information

Title:
Dissolved Oxygen as Related to Minimum Flows and Levels
Creator:
Hood, Jason L
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (68 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Soil and Water Science
Committee Chair:
Clark, Mark W.
Committee Members:
Jawitz, James W.
Gore, James
Graduation Date:
12/14/2007

Subjects

Subjects / Keywords:
Correlations ( jstor )
Froude number ( jstor )
Human geography ( jstor )
Oxygen ( jstor )
Rain ( jstor )
River water ( jstor )
Rivers ( jstor )
Water depth ( jstor )
Water flow ( jstor )
Water quality ( jstor )
Soil and Water Science -- Dissertations, Academic -- UF
aeration, dissolved, do, fish, florida, flows, levels, mfl, minimum, oxygen, passage, river, riverine, shoal, water
City of Tampa ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Soil and Water Science thesis, M.S.

Notes

Abstract:
Dissolved oxygen (DO) levels are a significant factor used to determine overall ecosystem health in aquatic systems. This thesis presents an evaluation of the effects of low flow conditions on DO levels in several warm-water lotic systems of west-central Florida. DO concentrations were measured at sites on the Withlacoochee, Anclote, Alafia, and Hillsborough rivers to evaluate how flow across shoals affects this water quality parameter. Correlations between shoal water depth and dissolved oxygen concentrations were used to examine how the ?minimum flows and levels? regulatory criteria, currently used by the Southwest Florida Water Management District (SWFWMD) may influence DO levels within the studied rivers. Peer review of a recent minimum flows and levels report for the Middle Peace River by the Southwest Florida Water Management District (SWFWMD) questioned the effect of low flow conditions on dissolved oxygen concentration. The peer review panel pointed out that the fish passage criteria currently used by the SWFWMD was originally developed for migratory salmonoids in cool, well oxygenated rivers and streams in the western United States. The current criteria requires a minimum of 0.6 ft (18.3 cm) of water passing over the lowest elevation of shoals which is also presumed to supply sufficient dissolved oxygen concentrations for fish. The panel suggested that a study be conducted to determine if the current criteria is applicable to warm-water rivers and streams of Florida, and to determine the level at which low flows negatively effect water quality, primarily dissolved oxygen. Data for this study were collected over a two year period under medium (which for this study include discharges that are exceeded 35 to 70 percent of the time) and low (which for this study include discharges that are exceeded 70 to 100 percent of the time) flow conditions. During periods of medium flow, instantaneous measurements were taken above, at, and below selected shoals. Instantaneous measurements were taken during medium flow conditions to provide DO data during non-stressed periods. There was not enough equipment to place data-loggers at each site for the entire two years. As river stages decreased and neared the 18.3 cm threshold, data-loggers were deployed above and below the shoals. DO concentrations were correlated to maximum shoal depth and compared to state standards and historical data. DO levels were also evaluated as a function of water moving across shoals utilizing Manning's N coefficients and Froude numbers. Shoal depth, and thereby flow, showed a wide range of correlations with dissolved oxygen for the rivers studied. Findings also indicated that the current Florida Department of Environmental Protection (FDEP) dissolved oxygen criteria of 5 mg/l is not being met at the 18.3 cm threshold for one of the four rivers studied. Application of the measured stage ? DO correlation to historical stage data for each river indicated that the 18.3 cm threshold in the past has maintained dissolved oxygen levels above 5 mg/l 90% of the time in three of the four studied rivers. In summary, this study demonstrated that additional factors, possibly rainfall, watershed land uses, and seasonality, may be influencing dissolved oxygen concentrations in these warm-water systems and that depth, although a factor controlling dissolved oxygen concentrations, is not the only variable that should considered when trying to maintain DO levels. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2007.
Local:
Adviser: Clark, Mark W.
Statement of Responsibility:
by Jason L Hood.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Hood, Jason L. Permission granted to the University of Florida to digitize, archive and distribute 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.
Classification:
LD1780 2007 ( lcc )

Downloads

This item has the following downloads:


Full Text





flow river, the water molecules are turning over slowly allowing a small percentage of the water

volume to come in contact with the atmosphere. In areas with turbulence, man-made or naturally

occurring, many more water molecules interact with the atmosphere bringing the water column

more rapidly into equilibrium with the air.

Reaeration models are utilized to predict the rates at which oxygen is dissolved into the

water column. Variables common to these models include water surface tension, molecular

diffusion, kinematic viscosity, discharge, velocity, stream slope, gravitational acceleration,

Manning's roughness coefficient, depth, surface area, Froude number, and temperature (Gualteri

et al 2002). Recent studies of reaeration models have shown many of these variable to be

redundant leading to the combination or elimination of certain variables in the analyses

(Ramakar et al 2001). Findings of a recent study using eighteen reaeration models to compare

predicted DO changes to actual changes found that the formula (Parkhurst and Pomeroy 1972),

K2 = 23(1 + .17Fr2)(SV).375 H

where:

Fr = Froude number
V = friction velocity
S = Slope
H = depth

was the most accurate (Ramakar et al 2004).

Naturally occurring shoals and shallow runs cause the water column to turn over, exposing

a higher portion of the water molecules to make contact with the atmosphere. A shoal is a

shallow feature of the river channel that acts to control the flow of water, at least under low flow

conditions. This control point holds back the flow of water causing a higher water surface

elevation on the upstream side of the shoal than if the river bottom followed a smooth fall.

Because the atmosphere generally has higher oxygen content than the water, the dissolved










Table 3-1. Summary Statistics for DO data collected at the USGS gauge at Anclote River Near
Elfer's, FL.
Statistic Value
mean (mg/L) 5.03
median (mg/L) 5.07
min (mg/L) 1.20
max (mg/L) 10.40
90% exceedance (mg/L) 3.20
% of readings below 5mg/L 48%
% of readings below 4mg/L 28%
% of readings below 3mg/L 11%
% of readings below 2mg/L 3%


0 n--
Jan-60


Jan-70 Jan-80 Jan-90 Jan-00


Jan-10


Figure 3-7. Historical DO concentrations for Anclote River near Elfers, FL (USGS).









(Ramakar et al 2004). Manning's N values were selected for use in this study due to the

accepted use of them by instream flow scientists. Although both Froude and Reynolds numbers

are both widely used in reaeration models to characterize flow conditions and both take depth,

and velocity into account, Froude numbers were chosen for this study due to the findings of

Ramakar et al (2004) that indicated that reaeration model that predicted reaeration most

accurately utilized Froude numbers in the calculation (Ramakar 2004).

As part of the primary data analysis for this study, determinations were made of the

maximum shoal depth needed to meet two distinct DO concentration criteria. The first criterion

was the FDEP standard for Class III waterbodies of 5 mg/L. The second criteria utilized to

determine the maximum depth necessary at the shoal was DO concentration based on river-

specific historical data. DO concentration data were retrieved from USGS, FDEP and District

databases for evaluation of historical or background conditions in the four studied rivers.

Dissolved oxygen, rainfall, stage, and discharge data have, and continues to be collected by

many agencies and organizations. External DO data gathered from the USGS at or near study

sites were all collected in-situ during daylight hours. Due to these data only being collected

during daylight hours, which is when DO levels reach their daily highs, considerations must be

made when comparing historic values with 24 hour logging data collected for this study.

A secondary analysis of the data looked at the physical component of water moving across

a shoal in an effort to quantify DO change from upstream to downstream of shoals. Output data

from the PHABSIM model were used to determine average Froude number and Manning's N for

the three rivers previously studied by the District. Manning's N values are calculated by an

algorithm in the PHABSIM system and represent friction as a result of surface roughness and

sinuosity. These values are produced using the equation (Chow 1959):









Site Specific Alternative Criterion (SSAC) and have been established for areas in Florida such as

the Everglades Marsh area (Weaver, 1998) and the Lower St. Johns River (Brooks, 2006).

In addition to SSAC development, the Florida Department of Environmental Protection

(FDEP) recently completed data collection for a statewide water quality investigation. One of

the primary goals of this investigation is to evaluate current DO criteria used on Florida's

waterbodies. The data from this investigation will be available during the latter part of 2007

(Russell Frydenborg FDEP, July 24, personal communication). How this investigation will

affect the current dissolved oxygen criteria has yet to be decided.

DO levels typically follow a diurnal cycle, increasing during the day due to sunlight

increasing photosynthetic activities and decreasing during the night due to respiration (Tadessa et

al 2004). DO levels typically peak during the mid-afternoon hours and reach their minimum

during the hours just before sunrise. Depressed DO conditions can also occur during daylight

hours during cloudy days when available sunlight for photosynthesis is limited and respiration

demands for oxygen exceed oxygen supply (Richard and Moss, 2002).

In addition to oxygen being introduced into the water by photosynthetic plants,

atmospheric oxygen is also fluxing across the air water interface. Based principally on water

temperature and atmospheric pressure, a dynamic equilibrium between oxygen concentration the

water column and that in the air is balanced. When there is excess DO in the water column due

to production or a rise in water temperature, DO concentration in the water can become saturated

and oxygen is released to the atmosphere. If water temperatures fall or oxygen is consumed, DO

is less than saturated in the water column and oxygen diffuses from the atmosphere into the

water column. Because of the slow diffusion rate of oxygen in the water, water movement can

significant effect the rate of oxygen exchange across the air water interface. In a lake or low-











10

8


I-



2 2

0I I
J F M A M J J A S O N D
Month

Figure 2-11 Rainfall comparison for study year and 92 year average for the Alafia watershed.


A Hillhbnro .g h G i S ere HboroIh d D Site A 2 ____ 10 D 'IDmetr


Figure 2-12. Hillsborough River DO site.









entire range of stages. Second, lowest DO levels generally occur when stage first starts to rise at

the beginning of the rainy season. This is to be expected due to large numbers of riverine

wetlands and the large, low-relief flood plain at, and upstream, of the study site. It appears these

early-season rain events flush high BOD waters and detrital material from surrounding wetlands

and swamps into the river causing increased oxygen demand and depressing DO concentrations.

A comparison between historical discharge and discharge during the study period shows the lack

of the typical rainy season for the studied years (figure 3-19).

Withlacoochee River

Continuous Measurement Data

One site was selected for data collection on Withlacoochee River, which is just

downstream of a wide spot in the river named Silver Lake (figure 2-14). Only data-logging data

was collected at the Withlacoochee River site for six and a half months, from November 9, 2006

through June 28, 2007. During this period, the maximum depth at the shoal ranged from 0.0 cm

to 40 cm (figure 3-20 & 3-21). DO levels ranged from 0 to nearly 20 mg/1. A plot of maximum

shoal depth and DO at this site shows a good correlation (figure 3-22).

Comparison with FDEP Class III DO Criteria

Dissolved Oxygen concentration above the shoal remained above the 5 mg/1 FDEP

standard until the maximum shoal depth dropped slightly below the 18.3 cm threshold currently

used by the SWFWMD for fish passage. DO below the shoal fell below the 5 mg/1 standard

when the maximum shoal depth was slightly above 24 cm. At this site, and none of the others,

there is an average loss of DO across the shoal. It is speculated that this was due to the

downstream data-logger being placed in a small pool off to the side of the main channel. The

downstream data-logger was placed off to the side because of the large number of airboats

navigating the river. Probe was placed in the only safe place downstream of the shoal so that it











0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
10O1(


3a2007


4128207 (11712007 ?&2007 9'251


18
16
14
12
10

6
4
2
0
2007


tudy Q n Q

Figure 3-19. Mean discharge (USGS) and discharge during study period for Withlacoochee
River near Croom.


25
depth

-20
a3


1 15 -

0
1 10


S5


0
10/10/06 11/29/06 1/18/07 39./07

U.S-


45.00

-40.00

35.00

30.00

25.00

20.00

15.00

10.00

5.00

0.00
4/28/07 6117/07 86/07

-Max Depth


Figure 3-20. Time series analysis of DO concentration (upstream (U.S.) of shoal), maximum
shoal depth, and rainfall data collected at Withlacoochee River site.


Iw % _


1120308 M129=6 ilia 182007

































To my beautiful wife and my parents who's love and support enable me to reach my goals









TABLE OF CONTENTS


Page

A C K N O W L E D G M E N T S ..............................................................................................................4

L IST O F T A B L E S ................................ ..........................

LIST OF FIGURES .................................. .. ..... ..... ................. .8

A B S T R A C T ......... ....................... ............................................................ 1 1

CHAPTER

1 INTRODUCTION ............... .............................. ............................. 13

Rationale ........................ .. .......... ........... ........ .... ................... 13
Background Information on Dissolved Oxygen and Hydrologic Conditions .......................14

2 M A TER IA L S A N D M ETH O D S ........................................ ............................................19

Site Selection ............................................................... 19
Study Location and Site D escriptions.......................................................... ............... 20
A n c lo te R iv e r ............................................................................................................. 2 0
A lafia R iver ............. ......... ................................................................ ..... 2 1
Hillsborough River .............................................. .................22
Withlacoochee River ............................................. ...................23
D ata C o lle ctio n ................................................................................................................. 2 4
D ata A n aly sis ................................................................................................. ...............2 5

3 RESULTS AND DISCUSSION .................................................................. 37

A n clo te R iv er ................................................................ ..................................... .. 3 7
D iscrete M easurem ent D ata ................................................... .............................. 37
Continuous M easurem ent D ata ........................................ ................. ............... 37
Comparison with FDEP Class III DO Criteria ........................................................... 38
Comparison with Historic DO Measurements ...................................... ............... 39
A lafia R iv er ................................................................. ................4 0
D iscrete M easurem ent D ata .................................................. .............................. 40
Continuous M easurem ent D ata ........................................ ................. ............... 40
Comparison with FDEP Class III DO Criteria ............................................................41
Comparison with Historic DO Measurements ...................................... ............... 41
H illsb orou gh R iv er ........................................................................4 1
D iscrete M easurem ent D ata .................................................. .............................. 41
Continuous M easurem ent D ata ........................................ ................. ............... 42
Comparison with FDEP Class III DO Criteria...........................................................42
Comparison with Historic DO Measurements ...................................... ............... 42









LIST OF REFERENCES


Anerson, T.H. and Taylor, G.T. 2001. Nutrient Pulses, Plankton Blooms, and Seasonal Hypoxia
in Western Long Island Sound. Estuaries 24(2):228-243.

Bosch, D., R. Lowrance, and G. Vellidis. 2002. Dissolved Oxygen Concentrations in Three
Coastal Plain Watersheds: Implication for TMDLs. American Society of Agricultural and
Biological Engineers, St. Joseph, Michigan. http://www.asabe.org, Accessed February
2007.

Brooks, J. 2006. ERC Adoption of Dissolved Oxygen SSAC for the Lower St. Johns River.
Florida Department of Environmental Protection, Tallahassee, Florida.
http://floridadep.net/northeast/stjohns/TMDL/docs/May06/ERCLSJR%20DO%20SSAC
ERCMEETING.pdf, Accessed May 2007.

Chow, V.T. 1959. Open-channel Hydraulics. McGraw Hill. New York, New York.

Gualtieri, C., Gualtieri, P. and Doria, G.P. 2002. Dimensionless Analysis of Reaeration Rate in
Streams. Journal of Environmental Engineering 28(1): 12-21.

Hood, J. 2007. Standardized Methods Utilized by the Ecologic Evaluation Section for Collection
and Management of Physical Habitat Simulation Model and Instream Habitat Data.
Southwest Florida Water Management District, Brooksville, Florida.

Jain, SK and Jha, R. 2004. The Froude Number Stick; an Evaluation. River Research and
Application 20(1):99-102.

Lakewatch. 2003. A Beginner's Guide to Water Management Fish Kills (Information Circular
107). Lakewatch/University of Florida, Gainesville, Florida.
http://lakewatch.ifas.ufl.edu/circpdffolder/fish_kill_LR.pdf, Accessed May 2007.

Moog, D.B. and Jirka, G.H. 1999. Stream Reaeration in Nonuniform Flow: Macroroughness
Enhancement. Journal of Hydraulic Engineering 125(1):11-16.

Parkhurst, J.D. and Pomeroy, R.D. 1972. Oxygen Absoption in Streams. Journal of the Sanitary
Engineering Division, American Society of Civil Engineers 98(1):101-124.

Ramaker, J., Ojha, C.S.P. and Bhatia, K.K.S. 2001. Refinement of Predictive Reaeration
Equations for a Typical Indian River. Hydrological Processes 15(6):1047-1060.

Ramaker, J., Ojha, C.S.P. and Bhatia, K.K.S. 2004. A Supplemental Approach for Estimating
Reaeration Rate Coefficients. Hydrological Processes 18(1):65-79.

Rhinesmith, P. and R. Smith. 2001. Wysong-Coogler Water Conservation Structure
Environmental Monitoring Report FDEP Permit Authorization Number 09-177432-001.
Southwest Florida Water Management District, Brooksville, Florida.










0.7

0.6

0.5

0.4

0.3

0.2

0.1
0


0.001 0.002


0.003 0.004 0.005

Ave. Froude #


Figure 3-13. Correlation between DO change across shoal and Froude number for Alafia River
site.


Table 3-3. Summary Statistics for DO data collected at the USGS gauge at
Springs.


Alafia River at Lithia


Statistic Value
mean (mg/L) 7.2
min (mg/L) 3.5
max (mg/L) 14
90% exceedance (mg/L) 5.2
% of readings below 5 mg/L 0.03
% of readings below 4 mg/L 0.01
% of readings below 3 mg/L 0
% of readings below 2 mg/L 0

Table 3-4. Statistical analyses for DO data collected at the USGS gauge at Alafia River at Lithia
Springs indicating a significant increasing trend in DO (SWFWMD 2005).
Parameter Residual
Residual Median n p Value Intercept Slope
Dissolved
Oxygen -0.057 194 0 0.514 -0.00002


_y = 6.2203x + 0.1863
SR2 = 0.0031


0.006


0.007


0.008









Richard, A. and A. Moss. 2002. Plant Management in Florida Waters; All you want to know
about Florida's lakes, rivers, springs, marshes, swamps, and canals. University of Florida,
Gainesville, Florida and Florida Department of Environmental Protection, Tallahassee,
Florida. http://aquatl.ifas.ufl.edu/guide/oxygen.html, Accesses May 2007.

Tadessa, I., Green, FB, Puhakka, JA. 2004. Seasonal and Diurnal Variations of Temperature, pH,
and Dissolved Oxygen in Advanced Integrated Wastewater Pond System (R) Treating
Tannery Effluent. Water Research 38(3):645-654.

United States Environmental Protection Agency. 2004. Total Maximum Daily Load (TMDL) for
Dissolved Oxygen (DO) in Brooker Creek (WIBID #1474). United States Environmental
Protection Agency, Region 4, Atlanta, Georgia.

United States Geological Survey(USGS)(1). 2007. National Water Information System: Web
Interface. United States Geological Survey, Tampa, Florida.
http://waterdata.usgs.gov/fl/nwis/nwisman/?siteno=02310000&agency_cd=USGS,
Accessed May 2007.

United States Geological Survey(USGS)(2). 2007. National Water Information System: Web
Interface. United States Geological Survey, Tampa, Florida.
http://waterdata.usgs.gov/fl/nwis/nwisman/?siteno=02301500&agency_cd=USGS,
Accessed May 2007.

United States Geological Survey(USGS)(3). 2007. National Water Information System: Web
Interface. United States Geological Survey, Tampa, Florida.
http://waterdata.usgs.gov/fl/nwis/uv/?siteno=02303000&PARAmetercd=00065,00060,
Accessed May 2007.

United States Geological Survey(USGS)(4). 2007. National Water Information System: Web
Interface. United States Geological Survey, Tampa, Florida.
http://waterdata.usgs.gov/fl/nwis/uv/?siteno=02312500&PARAmetercd=00065,00060,
Accessed May 2007.

Weaver, K. 2003. An Alternative Water Quality Criterion for Everglades Dissolved Oxygen.
Florida Department of Environmental Protection, Tallahassee, Florida.
http://www.floridadep.org/water/wqssp/everglades/docs/DDO_SSAC_Public_Workshop_
Augl9.pdf, Accessed May 2007.























2-
0-
60 110 160 210 260 310 360 410 460 510 560
Stage (cm)


Figure 3-14. Correlation between stage and DO concentrations for Alafia River near Litia, FL
(USGS) historical data.


-20 0 20 40 60 80 100
River Distance (m)
-.--Max Shoal Depth 167 cm -.-Max. Shoal Depth 124 cm
Max. Shoal Depth 109 cm


Figure 3-15. Hillsborough River site discrete measurement DO data.









to approximately 8 mg/1. Overnight, when DO plummeted to anoxic levels in the water coming

from Silver Lake, the opposite occurred where DO concentrations were elevated to around 5

mg/1. In this situation, the shoal acted as a buffering mechanism to moderate extreme levels

created in the lake.

Comparison with Historic DO Measurements

USGS has been taking water quality measurements at their Withlacoochee River near

Croom gauging station since the late 1960's and data collected during this study mirror historical

data (Appendix A). This site is just downstream of the site used for this study. Results of a

Kendall's tau analysis on residuals for DO regressed against flow indicate no significant trend in

dissolved oxygen over the period of record (table 3-6).

USGS collected 122 DO measurements between 1967 and 2005. Of the measurements

collected by the USGS, 59% were below the 5 mg/1 FDEP standard and the historical range is

from 1.6 to 5.7 mg/1 (Table 3-7). DO upstream of the study shoal remained within the historic

range until the maximum shoal depth dropped to approximately 7 cm. Downstream DO

concentration dropped below the 5 mg/1 threshold when the maximum shoal depth dropped

below 23.2 cm. It is again speculated that this variability is due to the positioning of the

downstream data-logger and is assumed to be invalid data for downstream representation. A

correlation between stage and DO concentration for the Withlacoochee River historical data

indicates that low DO concentrations occur at all stages (figure 3-26).

It should be noted that the average time of day the USGS water quality parameters were

collected at the Withlacoochee River near Croom was 12:27 P.M. (with a minimum time of 7:30

A.M. and a maximum time of 6:30 P.M.). At this time of day, due to diurnal cycles, DO

concentrations are nearing their daily maximum. When data from the data-logging portion of

this study, collected during mid-day hours are analyzed, the minimum DO concentration is 2.52










Table 3-5. Statistical analysis for D.O. data collected at the USGS gauge at Hillsborough River
near Zephyrhills (SWFWMD 2005).

Parameter Residual
Residual Median n p Value Intercept Slope

Dissolved
Oxygen -0.0787 259 0.97 -0.0988 0.00003


E 0
w 9
8
o 7
Zm6
E
C5


03

2
0
Q0


81- -$ &y &0 -19
Q5 & Z' Q5
NtN
11' cv \11V
tNL
'^ */ ^*O
^ x<^^>

-- Stage


D.O. (ave daily)


Figure 3-18. Time series analysis of USGS stage and DO continuous data for Hillsborough
River near Zephyrhills.









oxygen levels in the water flowing through shoals are usually raised as a result of diffusion

across the water/atmosphere interface.

This study is driven by three main hypotheses:

1) DO concentrations are maintained within the river specific historical DO ranges when
flows necessary to maintain an 18.3 cm maximum shoal depth are met or exceeded.

2) DO concentrations are maintained above the FDEP Class III waterbody standard of 5
mg/1 when flows necessary to maintain an 18.3 cm maximum shoal depth are met or
exceeded.

3) Physical shoal characteristics represented by Froude numbers and Manning's N values
influence reaeration of water passing over studied shoals.









this historic low DO concentration is determined to be naturally occurring, a SSAC for DO that

is lower than state wide criterion should be considered as it will likely be more protective of least

impacted conditions and a more attainable DO concentration.

Comparison with Historic DO Measurements

The USGS and the SWFWMD have been taking water quality measurements at their

Anclote River near Elfers, FL gauging station since 1962 and have collected 278 DO

measurements (figure 3-7 and Appendix A). This USGS site is 3.1 km downstream of the site

used for this study but is still in the upper portion of the watershed. Results of a Kendall's tau

analysis on residuals for water quality parameters regressed against flow indicate a significant

increasing trend in DO concentration over the period of record (table 3-2). Historical DO range

is 1.20 to 10.4 mg/1. For 48% of measurements, DO concentrations were below 5 mg/1 and 28%

of measurements were below 4 mg/1 (Table 3-2). A correlation between DO and stage for the

historical data collected at this USGS site indicates that most occurrences of DO concentrations

below 4 mg/1 correspond to low flow periods (figure 3-8).

At the Anclote River site (figure 3-4), it is apparent that DO drops as water depth drops.

However, as discussed in Chapter 1, as a shoal water depth becomes shallower, water becomes

more oxygenated as it moves across the shoal. DO concentration in the Anclote River

occasionally falls below 2 mg/1 and recognizing the oxygenating ability of shoals (last few days

presented in Figure 3-3), it is speculated that waters downstream of shoals are providing a refuge

to mobile organism and may ensure the survival of a percentage of non-mobile organisms during

critical periods. At the 18.3 cm threshold, currently used by the SWFWMD when setting MFLs,

the DO concentration is being elevated as much as 0.8 mg/1 as water passes over the shoal. The

DO concentration fell below the historical range (1.2 mg/1) when the maximum shoal depth

























J F M A M J J A S O N D
Month
Figure 2-3. Typical hydrograph for Anclote River (Based on USGS data from 1945 2004).


30

& 29.5

29
28.5

28

27.5 ,,,,,,,
0 2 4 6 8 10 12 14 16 18
Hzontal Distance from Left Bank toRiat Bank (m)

Figure 2-4. Anclote River Site 1 cross section (location corresponds to X=0 in figure 2-5).


5.6
S 5.5

S 3" 3- s h "



4.8 -
4.7 ,
-4 -2 0 2 4 6 8 10
LorOitrunal Distance from Shoal (m)

Figure 2-5. Anclote River Site 1 mid-channel profile.









CHAPTER 1
INTRODUCTION

Rationale

The five water management districts of the State of Florida are tasked with permitting

consumptive water use within their boundaries and protecting water resources from significant

harm. One component of this regulatory charge involves establishment of Minimum Flows and

Levels (MFLs) for streams and rivers. As currently defined by the statute, "the minimum flow

for a given watercourse shall be the limit at which further withdraws would be significantly

harmful to the water resources or ecology of the area" (Section 373.042, Florida Statutes).

For the development of MFLs in the Southwest Florida Water Management District

(District), low flow thresholds, based on fish passage depth and wetted perimeter inflection

points, are used to establish short-term compliance standards. The District has used 18.3 cm

water depth at shoal sites as sufficient to provide for fish passage. Low flow thresholds based on

this fish passage depth are developed to maintain natural patterns of river-flow continuity and

thereby allow for longitudinal movement of biota within the river segment. The thresholds are

also assumed to be sufficient for continued recreational use of the river and to minimize negative

water quality or physical attributes associated with stagnant pool conditions.

Peer review of a recent Minimum Flows and Levels report for the Middle Peace River by

the Southwest Florida Water Management District (SWFWMD) questioned the assumption that

a 0.6 foot (18.3 cm) shoal depth provided sufficient dissolved oxygen concentration for fish

health (SWFWMD 2005). The review panel pointed out that the fish passage criteria was

originally developed for migratory salmonoids in cool, well oxygenated rivers and streams in the

western United States and may not suitable for warm slow moving lotic systems of western

Florida.









CHAPTER 4
CONCLUSIONS

The purpose of this study was to test the three hypotheses discusses in Chapter 1. The first

two hypotheses required the analysis of the maximum shoal depths needed to maintain DO levels

within the river-specific historical range and above the FDEP 5 mg/1 standard. The most

restrictive site in relation to the FDEP standard was on the Anclote River which indicated that

maximum shoal depth needs to remain above 38.1 cm (table 4-1). For the historical record

criteria, the Anclote River site was again the most restrictive and suggested that maximum shoal

depth needs to remain greater than 19.8 cm.

Correlations between DO and maximum shoal depth were good for most of the studied

sites. In the rivers studied, DO concentrations have been observed well above the 5 mg/1 criteria

during low flow conditions (maximum shoal depth less than 18.3 cm threshold) and below 5

mg/1 during higher flows indicating that other factors may be influencing DO concentrations as

well. One possible factor influencing DO levels is rainfall that may introduce high BOD

materials or pollutants which can lead to depressed DO concentrations. Findings of this study

disprove both my hypothesis that DO concentrations would remain within the river-specific

historical range and my hypothesis that DO concentrations would remain above the FDEP, 5

mg/1 standard when flows necessary to maintain a maximum shoal depth of 18.3 cm are met or

exceeded. Although both of these criteria were met for a portion of the studied site and may

serve as a guidance level, factors beyond the scope of this study indicate that DO should be

monitored in each system when establishing minimum depth criteria.

The third hypothesis required analysis of the physical aspect of water passing over shoals

by attempting to correlate Manning's N and Froude values to increase in DO from upstream to

downstream of the shoals. In all cases, no correlation was indicated by the data, disproving this










APPENDIX A
A COMPARISON BETWEEN HISTORICAL AND STUDY DISSOLVED OXYGEN


Anclote River
Historical vs. Study Period D.O. (mglL) Statistics 12
12 0
130
8.0
6.0
4.0


Min


* Historical


Max Mean
Study


Alafia River
Historical vs. Study Period D.O. (rg/L) Sttistics


m Historical




Historical













* Historical


Max Mean
Study


Withlacoochee River
vs Study Period D.O. (rngL) Statistics
--- -- -


20.0


Mean

a Study


Figure A-1. Comparison between historical and study dissolved oxygen.











20
depth

15 -



l10.



5



0
10110/06 11/29/06 1/18/07


4500
40.00
35 00

30,00

25.00 i


15.00
10.00
5.00

S7 6 0.00
3/9,'07 428.107 6/17/07 8U107


D.S. M ax. Depth


Figure 3-21. Time series analysis of DO concentration (downstream (D.S.) of shoal), maximum
shoal depth, and rainfall data collected at Withlacoochee River site.


0 -
0.00


10.00 20.00 30.00


40.00


Max. Shoal Depth (cm)
Figure 3-22. Correlation between Maximum Shoal Depth and Dissolved Oxygen at
Withlacoochee River Site.

































A 'Irrlc:le I U Jt V rAthe:ro r DL2 Ste 2 1 10


Figure 2-16. Withlacoochee River DO site.


615

10


J F M A M J J
Month


A S O N D


Figure 2-17. Hydrograph for Withlacoochee River (Based on USGS data from 1939-2007).









mg/1 (Table 3-8), indicating that DO concentrations collected during this study were within the

historical range, even when the shoal was dry.


IP
Se
a

M


-20


-10 0 10 20 30


River Distnce (Mn

-*-- Ix. Shoal Depth 162 cm --a- Max. Shal Depth 127 cm


Figure 3-1. Anclote River Site discrete measurement DO data.


22

20
22 ;----------------------------

0 18

9 16
S6 y= -0.908x + 22.995
0. 14 R2 = 0.6227
E
12

In


Dissolved Oxygen (mg/L)


Figure 3-2. Water Temperature compared to Dissolved Oxygen concentrations for Anclote
River site (discrete measurements).









W ithlacoochee River ............................. ......... .. ...... ..... .. ............43
Continuous M easurem ent D ata ........................................ ................. ............... 43
Comparison with FDEP Class III DO Criteria ............................................................43
Comparison with Historic DO Measurements ...................................... ............... 45

4 C O N C L U SIO N S ................. ......... ................................ .......... ........ ..... .... ...... .. 62

APPENDIX A COMPARISON BETWEEN HISTORICAL AND STUDY DISSOLVED
O X Y G E N ...........................................................................................64

L IST O F R E F E R E N C E S .............................................................................. ...........................66

B IO G R A PH IC A L SK E T C H .............................................................................. .....................68




























Figure 3-23. Picture depicting the downstream end of Silver Lake on the Withlacoochee River
without choking of aquatic vegetation and algae (Photo by R. Gant, SWFWMD).


Figure 3-24. Picture depicting the downstream end of Silver Lake on the Withlacoochee River
choked with aquatic vegetation and algae (Photo by R. Gant, SWFWMD).









Study Location and Site Descriptions

Rivers selected for research included the Withlacoochee, Alafia, Anclote, and

Hillsborough in west-central Florida (figure 2-1). Study sites on each river were in Hernando,

Hillsborough, and Pasco Counties. The studied rivers include clear water to dark water systems

with small to large floodplains.

Anclote River

Anclote River is a small river originating in central Pasco County. This river meanders

through Starkey Wellfield / Wilderness Park, passes under Starkey Boulevard, where it enters a

highly developed residential area, and eventually discharges into the Gulf of Mexico (figure 2-2).

Anclote River has a contributing drainage area of approximately 188 km2 at the studied site

(USGS, 2007(1)) and exhibits a relatively low relief over its 42 km. Flow records for the

USGS's gauging station at Elfers are from 1945 until current. Average discharge for the period

of record is 1.88 m3/sec (USGS, 2007(1)), with the majority of flow occurring during the rainy

season from June through September (figure 2-3). Minimum and maximum daily averages for

discharge were 0.023 and 105 m3/sec, respectively, and both occurred during the early 1960's

(USGS, 2007(1)).

The site studied on the Anclote River was called Site 1. This site is located approximately

one mile upstream of the Starkey Boulevard bridge (figure 2-2) at 28.22182N and -82.63435W

(decimal degrees format).

This site consists of a shoal area with a pool upstream and a run downstream. The bottom

contour of the shoal perpendicular to flow is a gentle slope (figure 2-4) with a sand and snag

substrate. Average depth in the pool for the entire study period was approximately 122 cm and

average depth in the run was approximately 76.2 cm. Average maximum depth at the shoal for

the entire study period was 50.3 cm (depicted in Figure 2-5).

















SLI

s
t Ul ---- __ _* =
r
e

m

<-shoal >>flw>>

7


150


River Distance (m)

Figure 3-9. Alafia River site discrete measurement DO data.


DatemTime


--Deoth


E
25.00

*D
C
15.00 0
aU


- Rain


Figure 3-10. Time series analysis of DO concentration (both upstream (US) and downstream
(DS)), maximum shoal depth, and rainfall data collected at Alafia River site.


-Max. Shoal
Depth 146 cm
-u-Max. Shoal
Depth 159 cm
Max. Shoal
Depth 193 cm
Max. Shoal
Depth 241 cm


-50


100


Si1


.fl


---D.S


--U.S









3-2 Water Temperature compared to DO concentrations for Anclote River site (discrete
m easurem ents). ............................................................................46

3-3 Time series analysis of DO concentration (both upstream (US) and downstream
(DS)), maximum shoal depth, and rainfall data collected at Anclote River site. ..............47

3-4 Maximum shoal water depth compared to DO concentration for Anclote River site. ......47

3-5 Correlation between DO change across shoal and Froude number for Anclote River
site ........ ........ ............................................................................. 4 8

3-6 Correlation between DO change across shoal and Manning's N for Anclote River
site ........ ........ ............................................................................. 4 8

3-7 Historical DO concentrations for Anclote River near Elfers, FL (USGS). .....................49

3-8 Correlation between stage and DO concentrations for Anclote River near Elfers, FL
(U SGS) historical data. ................................... .. .. ......... .. ............50

3-9 Alafia River site discrete measurement DO data................. .................. ...... ......... 51

3-10 Time series analysis of DO concentration (both upstream (US) and downstream
(DS)), maximum shoal depth, and rainfall data collected at Alafia River site ................51

3-11 Maximum shoal water depth compared to DO concentration for Alafia River site..........52

3-12 Correlation between DO change across shoal and Manning's N for Alafia River site......52

3-13 Correlation between DO change across shoal and Froude number for Alafia River
site ........ ........ ............................................................................. 5 3

3-14 Correlation between stage and DO concentrations for Alafia River near Litia, FL
(U SGS) historical data. ................................... .. .. ........ .. ............54

3-15 Hillsborough River site discrete measurement DO data................................................. 54

3-16 Time series analysis of DO concentration (both upstream (US) and downstream
(DS)), maximum shoal depth, and rainfall data collected at Hillsborough River site.......55

3-17 Maximum shoal water depth compared to DO concentration for Hillsborough River
site ........ ........ ............................................................................. 5 5

3-18 Time series analysis of USGS stage and DO continuous data for Hillsborough River
near Z ephyrhills. ............................................................................56

3-19 Mean discharge (USGS) and discharge during study period for Withlacoochee River
n e ar C ro o m ............................................................ .............. 5 7









deployed 10m upstream and 30m downstream of shoals at similar depths and were set-up to log

water quality data every 30 minutes. Data-loggers were deployed for varying periods of time

based on flow conditions. For the purpose of data-logging, the YSI multi-probes were tethered

inside a 10.2 cm diameter PVC pipe with holes drilled to allow free flow of water. The PVC

pipe was bolted on top of a concrete block to stabilize and elevate the probe (figure 2-21) 18.3

cm above the bottom of the river bottom.

Parameters collected during data-logging phase include pH, specific conductance,

temperature, dissolved oxygen concentration, percent saturation of dissolved oxygen, and depth.

Each of these parameters was logged at half hour intervals. Data from multi-probes were

collected weekly, at which time the multi-probes where calibrated and any necessary

maintenance was performed.

Although each multi-probe logs depth, maximum depths at each studied shoal were

measured on each field visit. These collected values were used to verify data collected by the

multi-probe as well as making data corrections in the occurrence of drift. Logged depth values

were also analyzed to determine if the probes had been tampered with and, if so, which data

needed to be discarded.

Data Analysis

Data were analyzed in several different ways in an effort to determine physical factors

affecting DO concentrations. Primary analysis for this study was the determination of the

correlation between DO concentrations and maximum shoal depth. In order to compare the DO

change as water passed over the shoal area to the physical characteristics of each of the shoal

areas, Manning's n values were compared to change in DO from upstream to downstream of the

shoals. The majority of reaeration equations use a variable to quantify the roughness of the

streambed. These variables include Manning's n, Reynolds numbers, friction velocity and others




















7 4

6
14.00


y = -0.0032x + 8.2205
R2 = 0.0001


I 0 ./
*-^- -- o -
,o,


* *


19.00


24.00


29.00


34.00


Max. Shoal Depth (cm)

Figure 3-11. Maximum shoal water depth compared to DO concentration for Alafia River site.


0.7
0.6
) 0.5
E
e 0.4
| 0.3
S0.2
d
0 -2
ci


0.03


0.035


0.04


0.045


0.05


Manning's N


Figure 3-12. Correlation between DO change across shoal and Manning's N for Alafia River
site.


_y = -6.5515x + 0.4793
R2 = 0.0324







I a









Q =(R2/3S1/2/n)A

Where:

Q = discharge
A = cross-sectional area of flow
n = coefficient of roughness
R = hydraulic radius
S = slope of the water surface.

In river systems, the higher the Manning's n value, the higher the likelihood of water

column turnover. Froude numbers (Fr) are also produced using the PHABSIM system and

describe flow conditions (Table 2-1). Values are produced by the equation (Jain, SK and Jha, R

2004)):

Fr = (V)2 / (gD)

Where:

V = velocity
g = gravitational constant
D = hydraulic depth.

Along with Manning's n, Froude numbers have been shown to be an important variable in

the calculation of reaeration rates (Ramakar et al, 2004). Froude numbers quantify the energy at

the water/atmosphere interface and indicate levels of air entrainment (Moog and Jirka, 1999).

These factors were plotted against change in dissolved oxygen, from upstream to

downstream, to determine what, if any, relationship exists under low flow conditions. These

numbers were averaged for each site for incremental flow and stage conditions for comparison

with the collected DO data. Withlacoochee River has not been analyzed using the PHABSIM

and therefore lacks Froude numbers, Manning's N values and subsequent comparisons.







































J F M A M J J A S O N D

Month


Figure 2-13. Hydrograph for Hillsborough River (Based on USGS data from 1940 2007).


3.5
0 3
2.5
LU 2
1.5
U) 1

0.5
0
0 5 10 15 20 25 30
Horizontal Distance from Left Bankto Right Bank (m)


Figure 2-14. Hillsborough River site shoal cross section (location corresponds to X=0 in figure

2-15).


0 0.4

S)' 0.2
LU .a
Ma 0

E -0.2

-0.4
n


-40 -30 -20 -10 0 10

Longitudinal Dstance from Shoal (m)


Figure 2-15. Hillsborough River site channel profile.


20 30


> FLOW-'


s
h
o
a
I









LIST OF FIGURES


Figure p e

2-1 L location of four study sites................................................................................................28

2-2 Site M ap for Anclote River ......................................................................... 28

2-3 Typical hydrograph for Anclote River (Based on USGS data from 1945 2004)............29

2-4 Anclote River Site 1 cross section (location corresponds to X=0 in figure 2-5). ..............29

2-5 Anclote River Site 1 mid-channel profile ........................................................................ 29

2-6 Rainfall comparison for study year and 92 year average for the Anclote watershed........30

2-7 Alafia River DO site. ................................... ................. ............ 30

2-8 Typical hydrograph for Alafia River (Based on USGS data from 1932 2007). ............31

2-9 Alafia River site shoal cross section (location corresponds to X=0 in figure 2-10)..........31

2-10 A lafi a R iver site channel profile. ................................................................... ............31

2-11 Rainfall comparison for study year and 92 year average for the Alafia watershed..........32

2-12 H illsborough R iver D O site. ..................................................................... ...................32

2-13 Hydrograph for Hillsborough River (Based on USGS data from 1940 2007). .............33

2-14 Hillsborough River site shoal cross section (location corresponds to X=0 in figure 2-
15). ............................................................................. 3 3

2-15 Hillsborough River site channel profile ........................................ ......................... 33

2-17 Hydrograph for Withlacoochee River (Based on USGS data from 1939-2007). ..............34

2-18 Withlacoochee River site cross section (location corresponds to X=0 in figure 2-19). ....35

2-19 W ithlacoochee River site channel profile. ........................................ ....... ............... 35

2-20 Rainfall comparison for study year and 92 year average for the Withlacoochee River
B asin .......................................................... ................................... 3 5

2-21 Picture of a YSI multi-probe mounted in PVC tube ................ ............................... 36

3-1 Anclote River Site discrete measurement DO data..........................................................46









dropped below 19.8 cm. Depths this shallow or shallower occurred 5% of the time based on 60

years of USGS discharge data.

Alafia River

Discrete Measurement Data

One site was selected for data collection on Alafia River (figure 2-6). Data for Alafia

River were collected from March through April in 2006. Discrete measurement data were

collected on four occasions for this site (figure 3-9) with maximum shoal depths ranging from

146 cm to 241 cm. DO concentration during this range of water depths was from 7.5 mg/1 to 10

mg/1. Discrete measurements suggest no significant effect of the shoal on DO concentration at

least within this range of water depths.

Continuous Measurement Data

For the Alafia River site, data-logging data were collected for one month. During this

period, the maximum depth at the shoal fluctuated between a 15.2 cm and 30.5 cm (figure 3-10).

DO levels ranged from 6.5 to nearly 11 mg/1. There was no significant correlation to shoal water

depth and DO concentration (figure 3-11). This lack of correlation held true even when diurnal

variability in DO concentrations due to photosynthesis was taken into account by comparing

only data from the same time each day. At this site, even during low flow conditions, where the

maximum depth was below the threshold for fish passage (18.3 cm), DO concentrations never

fell below 5 mg/1 or outside the historical range..

When plotted against DO change across the shoal, neither Manning's N (figure 3-12) nor

Froude number (figure 3-13) appear to have any relationship with DO. It is speculated that

under the narrow range of flows analyzed for this study, there is not a wide enough variation in

Manning's N values to make a meaningful comparison. With all of the Froude numbers being









It is hypothesized that the 18.3 cm fish passage criterion currently utilized by the

SWFWMD is sufficient to meet the present DO criteria. This thesis will also attempt to compare

DO concentrations under extreme low flow conditions to river-specific, historical data and

analyze the oxygenation of water passing over shoals.

Background Information on Dissolved Oxygen and Hydrologic Conditions

DO levels are influenced by physical, chemical, and biological factors. Oxygen is

introduced into riverine systems through diffusion across the air-water interface and by

photosynthetic activity of aquatic plants. Oxygen is removed from rivers by diffusion across the

air-water interface, by chemical oxidative processes, and through respiration of river biota. In

systems with large hydraulic storage areas, DO levels are often depressed during the rainy season

due to the inflow of detrital material and other bio-available carbon sources and stimulating

biological activity (Rhinesmith and Smith, 2001 and Anderson and Taylor, 2001). This in turn

can result in elevated biological oxygen demand in the water column and depressed oxygen

levels.

In other systems, DO levels may be depressed during low flow conditions and elevated

during high flow conditions. A United States Environmental Protection Agency (EPA) study of

Georgia coastal plain watersheds attempted to correlate DO levels to temperature, pH,

conductivity, oxidation-reduction potential, turbidity, and flow. The only statistically significant

response suggested that low DO levels occurred during periods of low flow (Bosch et al., 2002).

A study of Brooker Creek in Hillsborough County, Florida agreed with the findings of

Rhinesmith and Smith (2001) that DO levels are suppressed when low DO waters, stored in

riverine wetlands, are flushed into the river system (United States Environmental Protection

Agency, 2004) Because much of the flow in Brooker Creek is derived from overland flow from

agriculture areas (United States Environmental Protection Agency, 2004) it was theorized that










Anclote River
Average Rainfall I Study Year Rainfall (cm)


30.00

20.00

10.00

0.00
Jan Mar May Jui Sep Nov

Study Year m 92 Year Ave


Figure 2-6. Rainfall comparison for study year and 92 year average for the Anclote watershed.


-afia CO Sit A Aafa L-;SS St
Figure 2-7. Alafia River DO site.


A I MwemI









BIOGRAPHICAL SKETCH

Jason was born in Brooksville, Florida in 1972. He began working for the Southwest

Florida Water Management District in 1989 and graduated from high school in 1990. Attending

the University of South Florida and Saint Leo University off and on, while pursuing his career,

he graduated with a Bachelor of Science in environmental science from Saint Leo University in

2005. In 2006, he began his master's degree at the University of Florida. He is currently an

Environmental Scientist with the Southwest Florida Water Management District where he plans

to continue his career.









LIST OF TABLES


Table page

2-1 F roude num ber descriptions.................................................................... .....................36

3-1 Summary Statistics for DO data collected at the USGS gauge at Anclote River Near
Elfer's, FL .......................... .... .............. .........................................49

3-2 Statistical analyses for DO data collected at the USGS gauge at Anclote River Near
Elfers, FL indicating a significant decreasing trend in DO. ............................................50

3-3 Summary Statistics for DO data collected at the USGS gauge at Alafia River at
L ith ia S p rin g s....................................................................... 5 3

3-4 Statistical analyses for DO data collected at the USGS gauge at Alafia River at Lithia
Springs indicating a significant increasing trend in DO. ................................................. 53

3-5 Statistical analysis for D.O. data collected at the USGS gauge at Hillsborough River
near Z ephyrhills. ............................................................................56

3-6 Statistical analysis of DO data for the USGS gauge at Withlacoochee River near
Croom indicating no significant trend. ........................................ ......................... 60

3-7 Statistical analysis for DO concentration data collected at the USGS gauge at
W ithlacoochee River near Croom, FL. ........................................ .......................... 60

3-8 Statistical analysis for DO concentration data collected at the Withlacoochee near
Silver L ake Site (m id-day only)............................................... .............................. 61

4-1 Sum m ary of findings for all studied rivers ......................................................................63

A-i Numerical comparison between historical and study dissolved oxygen. ........................65









CHAPTER 2
MATERIALS AND METHODS

Site Selection

Site selection for this study was completed in two ways. Sites for three rivers were

selected from sites previously studied by the District. The site on the fourth river was selected

using the same method that the District utilized to pick previously studied sites. Prior to

initiation of this DO study, the District had completed data collection efforts supporting

development ofMFLs on the Anclote, Alafia, and Hillsborough rivers. One previously utilized

District site was selected on each of these three rivers. On the Withlacoochee River, a site was

selected using the same site selection criteria. These four rivers were selected due to variability

in their size, water chemistry, water color and watershed type.

One aspect of the MFL process is analyzing representative control points throughout the

stretch of the river. Data collected on these control points are used to model usable habitat under

a full range of flow conditions using a Physical Habitat Simulation model (PHABSIM). These

control points, also referred to as shoals, are selected based on the following criteria (Hood,

2007):

* sites should be representative of the river reach for which minimum flows are to be
developed

* sites should be reasonably accessible

* sites should typically include a pool and a run upstream of the shoal (in situations where
this is not possible, a pool, run or shoal can be selected individually

* sites should, if possible, not include significant bends in the river channel

* sites should show no evidence of braiding during high flow

* sites should be located between established river gauging stations.









ACKNOWLEDGMENTS

I would like to thank Dr. Mark Clark for his time and dedication in helping me achieve this

goal. His knowledge and his enthusiasm for the sciences make Dr. Clark an invaluable resource.

I would also like to thank the rest of my committee (Dr. James Gore and Dr. James Jawitz) for

their guidance. Thanks go to William (BJ) Grant, Don Hampton, Courtney James, and Tammy

Schmaltz for their dedication in assisting me in the collection of field data. I would like the

thank Dr. Martin Kelly (SWFWMD), Dr. Adam Munson (SWFWMD), Dr. Jonathan Morales

(SWFWMD), and Doug Leeper (SWFWMD) for their guidance and support.









the suppressed DO levels may be a result of elevated nutrient levels. Upon analysis it was

discovered that nutrient levels in the creek were below the statewide averages for nitrate/nitrite,

total phosphorous, and ammonia (United States Environmental Protection Agency, 2004).

Drainage of low DO waters from surrounding from surrounding wetlands combined with the

water stagnation caused by low stream gradient were determined to be the primary causes of the

suppressed DO concentrations.

Water quality plays a major role in the overall health of river ecosystems. For example,

different fish species exhibit varying levels of tolerance to elevated temperatures and depressed

DO levels which are two of the major water quality factors affecting Florida rivers. Fish require

DO levels of about 5 milligrams per liter (mg/1) for optimal health and can begin gasping and

dying when levels reach 1 mg/1 (Lakewatch, 2003). The Florida state standard for DO in

predominantly fresh water systems is that DO levels "shall not be less than 5.0 mg/1 and normal

daily and seasonal fluctuations above these levels shall be maintained" (EPA 62-302.530). This

standard is applied to all Class III waterbodies of the State of Florida. Class III waterbody is a

state designation for waters used for "recreation and propagation and maintenance of a healthy,

well-balanced population of fish and wildlife" (Florida Administrative Code 62.302.5.0). Many

Florida rivers do not meet this standard due to natural background conditions. In systems that

have been determined to have background conditions that do not meet these criteria, a regulatory

solution was established by the Florida Department of Environmental Protection (FDEP). It

states that "dissolved oxygen levels that are attributable to natural background conditions or

man-induced conditions which cannot be controlled or abated may be established as alternative

dissolved oxygen criteria" (EPA 62-302.500(2)(f)). These modified regulations are known as











Wliblcaachee DJD v5.M3 Depth a Shoal


Cdeall


II


14, ej








0


--- wefaa


Figure 3-25. Depiction of moderation of DO concentration by shoal at Withlacoochee River
site.


Table 3-6.


Statistical analysis of DO data for the USGS gauge
Croom indicating no significant trend


at Withlacoochee River near


Parameter Residual
Residual Median N p Value Intercept Slope

Dissolved
Oxygen -0.21696 121 0.837 0.0047 -0.00001



Table 3-7. Statistical analysis for DO concentration data collected at the USGS gauge at
Withlacoochee River near Croom, FL.
Statistic Value
mean (mg/L) 5.65
min (mg/L) 1.6
max (mg/L) 10.2
90% exceedance (mg/L) 2.6
% of readings below 5 mg/L 59%
% of readings below 4 mg/L 25%
% of readings below 3 mg/L 11%
% of readings below 2 mg/L 4%


S&XS


BsolvCr


S;1BC


S'lE.W


SM07'C?


- OC17l











8.5

8-

7.5
T-

7

6.5

6-

5.5
-JB


" -" ^


- U.S.


Date


- D.S.


k


-- epth


70.-0 0
60.00 E

50.00
40.-0
30.00 5
20.00 a
10.00 .
0.00
0.00


b


tw
J0.
^'3


- Rain


Figure 3-16. Time series analysis of DO concentration (both upstream (US) and downstream
(DS)), maximum shoal depth, and rainfall data collected at Hillsborough River site.


8.5


I- 8
E
C
0 7.5

0
-0

o
Co
o 6.5


6


50 55
Max. Shoal Depth (cm)


Figure 3-17. Maximum shoal water depth compared to DO concentration for Hillsborough River
site.









Although the Anclote River site shows a general increase in DO concentration from

upstream to downstream of the studied shoal, physical shoal parameters, Froude number (figure

3-5) and Manning's N number (figure 3-6) do not show a correlation with DO concentration

change across the shoal.

Water depth and DO levels fell for the Anclote River site during the study period except

for a few days after a rainfall event in late February. During the rainfall event, an estimated 0.66

cm of rain was received in the area. The rise in DO after a rainfall is quite common due to input

of well oxygenated water. However, if Anclote River had a more substantial floodplain or if

rainfall had been more substantial, results may have been reversed due to inflow of water with

elevated biological oxygen demands (BOD).

Comparison with FDEP Class III DO Criteria

Based on FDEP standards for DO on Class III waterbodies, DO of Anclote River must

maintain a minimum concentration of 5 mg/1. Using this criterion and data collected during this

study, it appears that the minimum shoal depth that can be reached before DO levels do not meet

the standard would be approximately 38.1 cm. Flow at which this depth is maintained is 0.12

m3/sec. This flow is met or exceeded 73% of the time (based on 60 years of USGS discharge

data).

The EPA and DEP allow review of site-specific DO standards when sufficient pre-

disturbance data are available to contest the applicability of the statewide 5 mg/1 standard. These

exceptions are referred to as site specific alternative criteria (SSAC) and are set by analyzing

pre-impact datasets and or datasets from similar non-impacted waterbodies within the region and

by determining the minimum criteria needed by the waterbodies' most sensitive organisms.

Although a SSAC has not been established for the Anclote River, historical data shows that this

waterbody has DO concentrations regularly below the EPA Class III standard (Table 3-1). If










Table A-i. Numerical comparison between historical and study dissolved oxygen.
River Anclote Alafia Withlacoochee
Historical Min 1.2 3.5 1.6
Historical Max 10.4 14.0 10.2
Historical
Mean 5.0 7.2 5.7
Study Min 1.1 6.5 0.1
Study Max 9.2 10.5 17.7
Study Mean 5.3 8.0 7.4














V



6
lu /I----------------------












J M A M J J A S O N D
Month

Figure 2-8. Typical hydrograph for Alafia River (Based on USGS data from 1932 2007).





S. 4









Figure 2-9. Alafia River site shoal cross section (location corresponds to X=0 in figure 2-10).
Figure 2-9. Alafia River site shoal cross section (location corresponds to X 0 in figure 2-10).


0.46


.22
-6 0.32
La
J. o,15
01
0.05
0


0 10
LorgitL1iWJ Dsj t7 frtam ShoEd {rro


Figure 2-10. Alafia River site channel profile.


r
h

I
L


-30









(figure 2-10). Average maximum depth at the shoal for the entire study period was 20.7 cm.

Flora in the area surrounding the shoal includes Taxodium distichum and Salix spp in and within

9 meters (m) of the river channel with higher elevations dominated by upland vegetation

including Quercus virginiana, Quercus laurifolia, Pinus elliottii, wild Citrus sinesis, and Serona

repens.

Rainfall from the beginning of 2006 through the end of May was 47% below average

(figure 2-11). Rainfall for the entire year of 2006 was 10% below average.

Hillsborough River

Hillsborough River is a medium-large sized river originating in Green Swamp in Pasco and

Polk counties (figure 2-12). Green Swamp is an estimated 2253 km2 (Hills. River WMP 2000)

and is also the headwaters of the Peace, Oklawaha, and Withlacoochee Rivers. Hillsborough

River travels approximately 87 km before emptying into Hillsborough Bay (figure 2-1). The

study site is located in northeastern Hillsborough county in Hillsborough River State Park

(821343.3W 280902.1N).

Hillsborough River has a contributing drainage area of approximately 748 km2 at the

studied site (Hillsborough River WMP 2000). Flow record for the USGS's gauging station near

Zephyrhills is from 1939 until present. This USGS site also had a dissolved oxygen probe from

2001 until 2003. Minimum and maximum daily averages for discharge were 0.76 (2000) and

348 (1960) m3/sec respectively (USGS(3)). Average discharge for the period of record is 6.9

m3/sec (USGS(3)) with the majority of flow occurring during the rainy season, from summer

June through September (figure 2-13).

The bottom contour of the shoal perpendicular to flow consists of high banks and a bottom

comprised of limerock outcroppings and log jams (figure 2-14). Site consists of shoal area with









warm-water rivers and streams of Florida, and to determine the level at which low flows

negatively effect water quality, primarily dissolved oxygen.

Data for this study were collected over a two year period under medium (which for this

study include discharges that are exceeded 35 to 70 percent of the time) and low (which for this

study include discharges that are exceeded 70 to 100 percent of the time) flow conditions.

During periods of medium flow, instantaneous measurements were taken above, at, and below

selected shoals. Instantaneous measurements were taken during medium flow conditions to

provide DO data during non-stressed periods. There was not enough equipment to place data-

loggers at each site for the entire two years. As river stages decreased and neared the 18.3 cm

threshold, data-loggers were deployed above and below the shoals. DO concentrations were

correlated to maximum shoal depth and compared to state standards and historical data. DO

levels were also evaluated as a function of water moving across shoals utilizing Manning's N

coefficients and Froude numbers.

Shoal depth, and thereby flow, showed a wide range of correlations with dissolved oxygen

for the rivers studied. Findings also indicated that the current Florida Department of

Environmental Protection (FDEP) dissolved oxygen criteria of 5 mg/1 is not being met at the 18.3

cm threshold for one of the four rivers studied. Application of the measured stage DO

correlation to historical stage data for each river indicated that the 18.3 cm threshold in the past

has maintained dissolved oxygen levels above 5 mg/1 90% of the time in three of the four studied

rivers. In summary, this study demonstrated that additional factors, possibly rainfall, watershed

land uses, and seasonality, may be influencing dissolved oxygen concentrations in these warm-

water systems and that depth, although a factor controlling dissolved oxygen concentrations, is

not the only variable that should considered when trying to maintain DO levels.






















4-


2


200
200


3003 400 5GO 600 700
Stage (cm)


Figure 3-8. Correlation between stage and DO concentrations for Anclote River near Elfers, FL
(USGS) historical data.


Table 3-2. Statistical analyses for DO data collected at the USGS gauge at Anclote River Near
Elfers, FL indicating a significant decreasing trend in DO.

Parameter Residual p
Residual Median n Value Intercept Slope
Dissolved
Oxygen -0.0524 199 0 4.4665 0.0002














1.5

1

0.5

0

-0.5


0 0.005 0.01


0.015 0.02 0.025 0.03


Froude Number

Figure 3-5. Correlation between DO change across shoal and Froude number for Anclote River
site.


1.5

1

0.5

0


-0.5 -
0.04


0.045 0.05 0.055


0.06


Manning's N
Figure 3-6. Correlation between DO change across shoal and Manning's N for Anclote River
site.


y= -6.4193x + 0.2028
R2 = 0.078
*





-









3-20 Time series analysis of DO concentration (upstream (U.S.) of shoal), maximum
shoal depth, and rainfall data collected at Withlacoochee River site. ............................57

3-21 Time series analysis of DO concentration (downstream (D.S.) of shoal), maximum
shoal depth, and rainfall data collected at Withlacoochee River site. ............................58

3-22 Correlation between Maximum Shoal Depth and Dissolved Oxygen at
W ithlacoochee R iver Site. ........................................................................ ...................58

3-23 Picture depicting the downstream end of Silver Lake on the Withlacoochee River
without choking of aquatic vegetation and algae (Photo by R. Gant, SWFWMD) ..........59

3-24 Picture depicting the downstream end of Silver Lake on the Withlacoochee River
choked with aquatic vegetation and algae (Photo by R. Gant, SWFWMD) ....................59

3-25 Depiction of moderation of DO concentration by shoal at Withlacoochee River site.......60

3-26 Correlation between stage and DO concentrations for Hillsborough River near
Croom (U SG S) historical data. ................................................ ............................... 61

A-1 Comparison between historical and study dissolved oxygen. ........................................64











10.5-
9.5
8.5-
7.5 -
6.5
5.5
4.5
3.5
2.5
1.5
0.5
i;


k


e"


Date


----- U.S.


--D.S.


--Depth


100.0
90.0 1
80.0
70.0
60.0 po
50.0 .
40.0
30.0
depth 20.0 0
10.0
0.0




-Rainfall


Figure 3-3. Time series analysis of DO concentration (both upstream (US) and downstream
(DS)), maximum shoal depth, and rainfall data collected at Anclote River site.


10

8
y = 0.1218x 0.4638
R2 = 0.8712
E
d 4
d

2

0


Max. Shoal Depth (cm)


Figure 3-4. Maximum shoal water depth compared to DO concentration for Anclote River site.









































Figure 2-21. Picture of a YSI multi-probe mounted in PVC tube.

Table 2-1. Froude number descriptions.

Fr < 1 Subcritical Flow. Streaming flow where disturbances can travel upstream

Fr= 1 Flow is critical.
Supercritical Flow. Shooting flow where downstream disturbances cannot be felt
Fr> 1 upstream









DISSOLVED OXYGEN AS RELATED TO
MINIMUM FLOWS AND LEVELS




















By

JASON HOOD


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

UNIVERSITY OF FLORIDA

2007









well below the critical threshold, air entrainment was negligible under all of the flow conditions

measured.

Rainfall data indicated that there was only one rain event during the period of data

collection. Less than 0.25 cm of rain fell on the area on April 9, 2006 and did not appear to have

any impact on the DO concentration.

Comparison with FDEP Class III DO Criteria

Although the maximum shoal depth fell below the 18.3 cm fish passage criteria during the

data collection period, DO remained above the 5 mg/1 FDEP standard for the entire study period.

Comparison with Historic DO Measurements

USGS has been taking water quality measurements at Alafia River at Lithia Springs

gauging station since the late 1960's and data collected for this study mirror historical data quite

well (table 3-3 and Appendix A). This site is 4 km downstream of the site used for this study but

is still in the upper portion of the watershed. Results of a Kendall's tau analysis on residuals for

DO regressed against flow indicate a statistically significant downward trend in DO over the

period of record (table 3-4). Even with the downward trend, Alafia River typically exhibits DO

levels above 5 mg/1 even under low flow conditions (table 3-3). A correlation between historical

stage and DO concentrations showed that all DO values below 4 mg/1 occurred during low flow

periods (figure 3-14). All DO concentration measurements during the study period were within

the historical range.

Hillsborough River

Discrete Measurement Data

Discrete measurement data were collected at Hillsborough River State Park for

approximately three weeks beginning in mid June, 2006. Discrete measurement data from the









CHAPTER 3
RESULTS AND DISCUSSION

Data were collected on Anclote, Alafia, Hillsborough, and Withlacoochee Rivers. At the

first three rivers mentioned above, data were collected by discrete measurements and by data-

logging. At Withlacoochee River, data were only collected by data-logging. Data herein are

presented by river.

Anclote River

Discrete Measurement Data

The site on the Anclote River was named Anclote Site #1. Discrete measurement data for

this site were collected when the maximum shoal depth was at 127 cm and at 162 cm. DO during

discrete measurements ranged from 6.7 mg/1 to 8.0 mg/1. Discrete measurement data illustrated

an increase in DO concentration under lower water levels than under higher water levels (figure

3-1). When an average for all DO concentration and temperature measurements collected during

the discrete measurement portion of the study are compared, there is a good correlation (R2

0.6227) (figure 3-2).

Continuous Measurement Data

For the data-logging portion of the Anclote River data collection, data were collected for

40 days. During this period the maximum depth at the shoal dropped from nearly 183 cm to a

minimum depth of around 18.3 cm (figure 3-3). During this period there was a very good (R2

0.8712) correlation between maximum shoal depth and DO concentration (figure 3-4). Further

investigation of continuous data indicated that the last three days of data for Anclote Site #1,

when water depth over the shoal was at its shallowest, had the greatest increase in DO

concentration between upstream and downstream data-loggers.











S 0.7
0.6
Luj 0.5

E 0.3
0.2
0.1
0
0 20 40 60 80 100 120
Horizontal Distance from Left Bank to Right Bank (m)


Figure 2-18. Withlacoochee River site cross section (location corresponds to X=0 in figure 2-
19).


0.5
0
LU 0


E .1
= -1, ,


-200 -150 -100 -50 0 50 100 150
Longitudinal Distance fi Com Scal: r.I


Figure 2-19. Withlacoochee River site channel profile.



Withlacoochee River Average Rainfall/ Period of
Study Rainfall (cm)
25

20

S15 ,/ -.
I 10 ,"----.--" ---
10 \
5 ,- .- ---"

0 I I I I I I I I

1 2 3 4 5 6 7 8 9 10 11 12
Month

---- 2006 2007 -*-92YearAve


Figure 2-20 Rainfall comparison for study year and 92 year average for the Withlacoochee River
Basin.


S
h
FLCIW
I









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

DISSOLVED OXYGEN AS RELATED TO
MINIMUM FLOWS AND LEVELS

By

Jason Hood

December 2007

Chair: Mark Clark
Major: Soil and Water Science

Dissolved oxygen (DO) levels are a significant factor used to determine overall ecosystem

health in aquatic systems. This thesis presents an evaluation of the effects of low flow

conditions on DO levels in several warm-water lotic systems of west-central Florida. DO

concentrations were measured at sites on the Withlacoochee, Anclote, Alafia, and Hillsborough

rivers to evaluate how flow across shoals affects this water quality parameter. Correlations

between shoal water depth and dissolved oxygen concentrations were used to examine how the

"minimum flows and levels" regulatory criteria, currently used by the Southwest Florida Water

Management District (SWFWMD) may influence DO levels within the studied rivers.

Peer review of a recent minimum flows and levels report for the Middle Peace River by the

Southwest Florida Water Management District (SWFWMD) questioned the effect of low flow

conditions on dissolved oxygen concentration. The peer review panel pointed out that the fish

passage criteria currently used by the SWFWMD was originally developed for migratory

salmonoids in cool, well oxygenated rivers and streams in the western United States. The

current criteria requires a minimum of 0.6 ft (18.3 cm) of water passing over the lowest elevation

of shoals which is also presumed to supply sufficient dissolved oxygen concentrations for fish.

The panel suggested that a study be conducted to determine if the current criteria is applicable to










hypothesis and indicating that other factors may be controlling DO change across the shoals. It

is speculated that, due to the narrow range of flow conditions evaluated (low flow only), there

was not enough variability in Manning's N to sufficiently assess this relationship. It is also

speculated that, due to the very low Froude numbers typically observed in these systems, air

entrainment was negligible in all rivers studied. If any of the sites had approached or exceeded

critical flow, air entrainment may have been better correlated with DO increases across the

studied shoals.

This study shows that DO concentrations are moderated by shoals, aerating water that

enters the shoal with low DO levels and de-gassing water that enters the pool super-saturated

with DO. This process was most notable on the Withlacoochee River as the river became highly

influenced by submerged aquatic vegetation and algal blooms.

Table 4-1. Summary of findings for all studied rivers
Percentage of
Shoal Water Depth Time Needed
Necessary to Maintain Flow Conditions
5.0 mg/L FDEP Corresponding Met (Based on
River Standard (cm) Flow (m3/sec) POR)
Anclote 38.1 .12 73%
Alafia < 18.3 .62 99%
Hillsborough Met during entire study n/a n/a
Withlacoochee 16.8 2.6 85%

Shoal Water Depth Percentage of
Necessary to Maintain Time Needed
DO Concentrations Flow Conditions
within Historical Range Corresponding Met (Based on
River (cm) Flow (m3/sec) POR)
Anclote 19.8 .08 95%
Alafia < 18.3 .62 99%
Hillsborough Met during entire study n/a n/a
Withlacoochee 7.0 .02 99%









would not be crushed by one of the airboats. After the study was complete, discrete DO

measurements were taken in the main channel and off to the side where the downstream data-

logger had been placed. Each of the three times that this comparison was made, DO

concentration was between 2 and 3 mg/1 lower where the data-logger had been placed than in the

main channel. For this reason, the downstream data-logger data are not believed to be a good

representation of DO concentrations downstream of the studied shoal.

Both upstream and downstream sampling locations exhibited periods with DO levels above

5 mg/1 when the maximum shoal depth was well below the 18.3cm threshold. Conversely, both

locations had periods of DO below the 5 mg/1 threshold when the maximum shoal depth was

greater than 18.3 cm. Due to poor correlations between water depth at the shoal and DO

concentration, it appears that factors other than water depth have a significant effect to DO

levels.

It should be noted that during data-logging at this site, Silver Lake changed from an initial

condition of moderate to low levels of aquatic vegetation (primarily Hydrilla) and algae to one

that was almost completely clogged with aquatic vegetation and algae (figures 3-23 and 3-24).

The resulting effect on release and uptake of 02 in the water column becomes apparent in diurnal

cycles for the upstream data-logger towards the end of the study (figure 3-25).

Due to high and low DO concentrations occurring in Silver Lake, an interesting

phenomenon was recorded at the Withlacoochee River site starting in early May, 2007 when

flows were at their lowest. During this period, widely fluctuating DO levels in the water coming

out of Silver Lake were moderated by the shoal (figure 3-25). During the daytime, when DO

levels coming from Silver Lake often reached super saturated concentrations nearing 20 mg/1,

oxygen was released to the atmosphere as it crossed the shallow shoal bringing DO levels down







































2007 Jason Lamar Hood









site illustrated that water passing over the shoal is aerated to a greater extent under low flow

conditions than under higher flow conditions (figure 3-15).

Continuous Measurement Data

During the data-logging period, the maximum depth at the shoal fluctuated around 52 cm.

DO levels ranged from 6 to nearly 8 mg/1 (figure 3-16). DO concentrations were generally

higher downstream of the shoal than upstream indicating the ability of the shoal to oxygenate

water flowing over the shoal. During the study period, there appears to be no correlation

between depth and DO (figure 3-17). In an effort to remove the diurnal effect from the data, a

comparison was made between depth and DO for only measurements logged at noon each day.

Results of that correlation were even weaker. Due to flows being in the medium range, based on

historical flow data, and the rather long anticipated time before the river would drop to the

desired study range, data-loggers were pulled out and deployed at another site. There were

hopes of returning the equipment at a later date, but scheduling and flow conditions never

allowed.

Comparison with FDEP Class III DO Criteria

DO concentrations during the entire study period remained above the 5 mg/1 FDEP

standard.

Comparison with Historic DO Measurements

USGS has been taking water quality measurements at Hillsborough River near Zephyrhills

gauging station since the mid 1970's. This site is just downstream of the site used for this study.

Results of a Kendall's tau analysis on residuals for DO regressed against flow indicate no

significant trend in dissolved oxygen over the period of record (table 3-5). DO was measured by

USGS at this gauging station from 2001 through 2003. A plot of their data from this period

points out two interesting facts (figure 3-18). First, the DO levels remain relatively high over the















10


8-



4 *


.* *
2 -


0
0 50 100 150 200 250 300 350
Stage (cm)

Figure 3-26. Correlation between stage and DO concentrations for Hillsborough River near
Croom (USGS) historical data.


Table 3-8. Statistical analysis for DO concentration data collected at the Withlacoochee near
Silver Lake Site (mid-day only).
Statistic Value
min (mg/L) 2.52
max (mg/L) 14.39
mean (mg/L) 8.72









Anclote River in this area is generally incised having banks averaging approximately 244

cm with some areas dropping much lower where side channels enter and exit the main river. The

Substrate is predominately sand with scattered exposed limerock and has a moderate amount of

woody debris and snags.

Flora in the area is dominated by Taxodium distichum and Salix spp. in river bottom and

transitions to pine flatwood ecosystem, dominated by Pinus elliottii, Serona repens, Quercus

virginiana, Quercus laurifolia, and Vaccinium corymbosum outside of the incised channel.

During the study period the rainfall for the watershed was 65% below the 92 year average

(Figure 2-6). Rainfall for the year the study was conducted was 19% below the 92 year average.

Alafia River

Alafia River is a medium sized river originating in eastern Hillsborough and western Polk

counties and is fed by two major tributaries, the North and South Prongs (figure 2-7). The river

runs approximately 30.6 km through quickly developing urban and industrial areas before

emptying into Tampa Bay.

Alafia River has a contributing drainage area of approximately 868 km2 at the studied site

(USGS, 2007(2)). The flow record for the USGS's gauging station at Lithia is from 1932 until

current. Average discharge for the period of record is 10.53 m3/sec (USGS, 2007 (2)) with the

majority of flow occurring during the rainy season, from June through September (figure 2-8).

Minimum and maximum daily averages for discharge were .91 (1945) and 139.2 (1959) m3/sec

respectively (USGS, 2007(2)).

The study site on the Alafia River was named ALA2 and is located a short distance

downstream of the confluence of North and South Prongs (820831.6W 275151.6N). Bottom

contour of the shoal perpendicular to flow consists of high banks and a bottom lined with sand

and snags (figure 2-9). Site consists of a shoal area with a run upstream and a run downstream







































Figure 2-1. Location of four study sites.


F ArrInre n- r Sir A Irrjnc- Uiors e i- A

Figure 2-2. Site Map for Anclote River.


E KicmalrmI









Adjacent uplands are a scrub-shrub habitat dominated by Quercus laurifolia, Quercus laevis,

Quercus marilandica, and Serona repens.

Rainfall at the site during the data-logging portion of the study was below the 92 year

average for all but one month (figure 2-20).

Data Collection

Equipment used for data collection included YSI 600XLM multi-probes, survey level

instruments (Topcon G4), and level rods. Multi-probes were used for discrete water quality

measurements and unattended data-logging of water quality parameters and field computers were

used to download logged data. Survey levels and rods were used to determine elevations

(relative or NGVD) of site cross sections and thalwags. The data collection period ran from

November 30, 2005 through June 29, 2007 and generally occurred on one river at a time due to

equipment restraints.

Data collection was carried out in two distinct manners. During periods of medium to high

flow conditions, discrete, instantaneous measurements of pH, specific conductance, temperature,

depth, dissolved oxygen (mg/L), and dissolved oxygen (%). Each time data were collected for a

particular site, readings were taken at numerous locations in relation to the shoal. The first two

measurements at each site also included sampling for field parameters one meter from each

stream bank and in the middle of the stream. After analyzing data from the first two visits to

each site it was determined that mid-depth, mid-channel measurements sufficiently represent the

in-situ conditions. An attempt was made to collect these data at approximately the same time of

day (around noon) on each visit to minimize variation due to the diurnal fluctuation of

temperature and dissolved oxygen.

During periods when water levels dropped to the point that maximum depth over a shoal

was near the presumed critical level (18.3 cm fish passage threshold), multi-probes were









a run upstream and a run downstream (figure 2-15). Average maximum depth at the shoal for

the entire study period was 52.7 cm.

Due to relatively frequent flooding of Hillsborough River and low relief of surrounding

terrain, the floodplain in the area of the shoal is quite expansive. Many side channels parallel the

main river and are inundated under various levels of flooding. Flora within a hundred yards of

river consists of Taxodium distichum and Salix spp. in the lower areas and Quercus virginiana,

Quercus laurifolia, Pinus elliottii, wild Citrus sinesis, Serona repens, Liquidambar styraciflua,

and Sabalpalmetto at higher elevations.

Withlacoochee River

Withlacoochee River is a medium-large sized river originating in Green Swamp. This

river has a contributing drainage area of approximately 2098 km2 at the studied site (USGS (4)).

Study site is located on the northeast edge of Hernando County (821307.2W 283456.6N) with

the river marking the boundary between Hernando and Sumter (figure 2-16). Flow records for

the United States Geological Surveys gauging station near Croom are from 1939 until present.

Average discharge for the period of record is 12.7 m3/sec (USGS(4)) with the majority of flow

occurring during the rainy season from summer through late fall (figure 2-17). Minimum and

maximum daily averages for discharge were .13 (1981) and 244 (1960) m3/sec respectively.

The bottom contour of the shoal perpendicular to flow consists of low, gently sloping

banks and a bottom of sand and submerged aquatic vegetation (say) (figure 2-18). Site consists

of shoal area with a run upstream and a pool downstream (figure 2-19). Average maximum

depth at shoal for the entire study period was 22.3 cm.

Gently sloping banks and low relief of surrounding terrain lead to a very wide floodplain.

Within the river bottom, flora is dominated by Hydrilla verticillata, Salix spp., and Faxinus

caroliniana. Surrounding floodplain is dominated by Taxodium distichum and Acer rubrum.




Full Text

PAGE 1

1 DISSOLVED OXYGEN AS RELATED TO MINIMUM FLOWS AND LEVELS By JASON HOOD A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Jason Lamar Hood

PAGE 3

3 To my beautiful wife and my parents who's l ove and support enable me to reach my goals

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank Dr. Mark Clark for his tim e and dedication in helping me achieve this goal. His knowledge and his enthus iasm for the sciences make Dr. Clark an invaluable resource. I would also like to than k the rest of my committee (Dr. James Gore and Dr. James Jawitz) for their guidance. Thanks go to William (BJ) Gr ant, Don Hampton, Courtney James, and Tammy Schmaltz for their dedication in assisting me in the collection of field da ta. I would like the thank Dr. Martin Kelly (SWFWMD), Dr. Adam Munson (SWFWMD), Dr. Jonathan Morales (SWFWMD), and Doug Leeper (SWFWMD) for their guidance and support.

PAGE 5

5 TABLE OF CONTENTS Page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 ABSTRACT...................................................................................................................................11 CHAP TER 1 INTRODUCTION..................................................................................................................13 Rationale.................................................................................................................................13 Background Information on Dissolved Oxygen and Hydrologic Conditions........................ 14 2 MATERIALS AND METHODS........................................................................................... 19 Site Selection..........................................................................................................................19 Study Location and Site Descriptions..................................................................................... 20 Anclote River...................................................................................................................20 Alafia River.....................................................................................................................21 Hillsborough River.......................................................................................................... 22 Withlacoochee River....................................................................................................... 23 Data Collection.......................................................................................................................24 Data Analysis..........................................................................................................................25 3 RESULTS AND DISCUSSION............................................................................................. 37 Anclote River..........................................................................................................................37 Discrete Measurement Data............................................................................................ 37 Continuous Measurement Data....................................................................................... 37 Comparison with FDEP Class III DO Criteria................................................................ 38 Comparison with Historic DO Measurements................................................................ 39 Alafia River............................................................................................................................40 Discrete Measurement Data............................................................................................ 40 Continuous Measurement Data....................................................................................... 40 Comparison with FDEP Class III DO Criteria................................................................ 41 Comparison with Historic DO Measurements................................................................ 41 Hillsborough River............................................................................................................. ....41 Discrete Measurement Data............................................................................................ 41 Continuous Measurement Data....................................................................................... 42 Comparison with FDEP Class III DO Criteria................................................................ 42 Comparison with Historic DO Measurements................................................................ 42

PAGE 6

6 Withlacoochee River............................................................................................................ ..43 Continuous Measurement Data....................................................................................... 43 Comparison with FDEP Class III DO Criteria................................................................ 43 Comparison with Historic DO Measurements................................................................ 45 4 CONCLUSIONS.................................................................................................................... 62 APPENDIX A COMPARISON BETWEEN HISTORICAL AND STUDY DISSOLVED OXYGEN ...............................................................................................................................64 LIST OF REFERENCES...............................................................................................................66 BIOGRAPHICAL SKETCH.........................................................................................................68

PAGE 7

7 LIST OF TABLES Table page 2-1 Froude number descriptions...............................................................................................36 3-1 Summary Statistics for DO data collected at the USGS gauge at Anclo te River Near Elfer's, FL...........................................................................................................................49 3-2 Statistical analyses for DO data collected at the USGS gauge at Anclo te River Near Elfers, FL indicating a significant decreasing trend in DO...............................................50 3-3 Summary Statistics for DO data collected at the USGS gauge at Alafia R iver at Lithia Springs.....................................................................................................................53 3-4 Statistical analyses for DO data collected at the USGS gauge at Alafia R iver at Lithia Springs indicating a signifi cant increasing trend in DO.................................................... 53 3-5 Statistical analysis for D.O. data colle cted at the USGS gaug e at Hillsborough River near Zephyrhills. .............................................................................................................. ..56 3-6 Statistical analysis of DO data for the USGS gauge at W ithlacoochee River near Croom indicating no si gnificant trend...............................................................................60 3-7 Statistical analysis for DO concentrati on data collected at the USGS ga uge at Withlacoochee River near Croom, FL............................................................................... 60 3-8 Statistical analysis for DO concentration data collected at the W ithlacoochee near Silver Lake Site (mid-day only)......................................................................................... 61 4-1 Summary of findings for all studied rivers ........................................................................ 63 A-1 Numerical comparison between hist orical and study dissolved oxygen. .......................... 65

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Location of four study sites............................................................................................... .28 2-2 Site Map for Anclote River................................................................................................28 2-3 Typical hydrograph for Anclote River (Based on USGS data from 1945 2004)............ 29 2-4 Anclote River Site 1 cross section (lo cation corresponds to X=0 in figure 2-5). .............. 29 2-5 Anclote River Site 1 mid-channel profile.......................................................................... 29 2-6 Rainfall comparison for study year and 92 year average for the Anclote watershed. ....... 30 2-7 Alafia River DO site..........................................................................................................30 2-8 Typical hydrograph for Alafia Rive r (Based on USGS data from 1932 2007)..............31 2-9 Alafia River site shoal cross section (l ocation corresponds to X=0 in figure 2-10). ......... 31 2-10 Alafia River site channel profile........................................................................................ 31 2-11 Rainfall comparison for study year and 92 year average for the Alafia watershed. .......... 32 2-12 Hillsborough River DO site............................................................................................... 32 2-13 Hydrograph for Hillsborough River (Bas ed on USGS data from 1940 2007)............... 33 2-14 Hillsborough River site shoal cross secti on (loc ation corresponds to X=0 in figure 215)......................................................................................................................................33 2-15 Hillsboroug h River site channel profile ............................................................................33 2-17 Hydrograph for Withlacoochee River (Based on USGS data from 1939-2007)............... 34 2-18 Withlacoochee River site cross section (location corres ponds to X=0 in figure 2-19). .... 35 2-19 Withlacoochee River site channel profile.......................................................................... 35 2-20 Rainfall comparison for study year and 92 year average for the Withlacoochee River Basin..................................................................................................................................35 2-21 Picture of a YSI multi-probe mounted in PVC tube.......................................................... 36 3-1 Anclote River Site discrete m easurement DO data............................................................46

PAGE 9

9 3-2 Water Temperature compared to DO concen trations for Anclote River site (disc rete measurements).................................................................................................................. .46 3-3 Time series analysis of DO concentr ation (both upstream (US) and downstream (DS)), maximum shoal depth, and rainfall da ta collected at Anclote River site............... 47 3-4 Maximum shoal water depth compared to DO concentration f or Anclote River site....... 47 3-5 Correlation between DO change across s hoal and Froude number for Anclote River site. .......................................................................................................................... ...........48 3-6 Correlation between DO change across shoal and Manning' s N for Anclote River site........................................................................................................................... ...........48 3-7 Historical DO concentrations for An clote River near El fers, FL (USGS). ....................... 49 3-8 Correlation between stage and DO concentra tions for Anclote River near Elfers, FL (USGS) historical data. ......................................................................................................50 3-9 Alafia River site disc rete m easurement DO data............................................................... 51 3-10 Time series analysis of DO concentr ation (both upstream (US) and downstream (DS)), maximum shoal depth, and rainfall da ta collected at Al afia River site.................. 51 3-11 Maximum shoal water depth compared to DO concentration for Alafia River site.......... 52 3-12 Correlation between DO change across shoa l and Manning' s N for Alafia River site...... 52 3-13 Correlation between DO change across s hoal and Froude number for Alafia River site. .......................................................................................................................... ...........53 3-14 Correlation between stage and DO concentra tions for Alafia River near Litia, FL (USGS) historical data. ......................................................................................................54 3-15 Hillsborough River site discrete m easurement DO data.................................................... 54 3-16 Time series analysis of DO concentr ation (both upstream (US) and downstream (DS)), maximum shoal depth, and rainfall da ta collected at Hillsborough River site....... 55 3-17 Maximum shoal water depth compared to DO concentration for Hillsborough River site. .......................................................................................................................... ...........55 3-18 Time series analysis of USGS stag e and DO continuous data for Hillsborough River near Zephyrhills. .............................................................................................................. ..56 3-19 Mean discharge (USGS) and discharge during study period for W ithlacoochee River near Croom..................................................................................................................... ....57

PAGE 10

10 3-20 Time series analysis of DO concentrat ion (upstream (U.S.) of shoal), maximum shoal depth, and rainfall data collect ed at Withlacoochee River site................................ 57 3-21 Time series analysis of DO concentrat ion (downstream (D.S.) of shoal), maximum shoal depth, and rainfall data collect ed at Withlacoochee River site................................ 58 3-22 Correlation between Maximum Shoa l Depth and Dissolved O xygen at Withlacoochee River Site.................................................................................................. 58 3-23 Picture depicting the downstream end of Silver L ake on the Withlacoochee River without choking of aquatic vegetation and algae (Photo by R. Gant, SWFWMD)........... 59 3-24 Picture depicting the downstream end of Silver L ake on the Withlacoochee River choked with aquatic vegetation and al gae (Photo by R. Gant, SWFWMD)...................... 59 3-25 Depiction of moderation of DO concentra tion by shoal at W ithlacoochee River site....... 60 3-26 Correlation between stage and DO con centration s for Hillsborough River near Croom (USGS) historical data........................................................................................... 61 A-1 Comparison between histori cal and study dissolved oxygen. ........................................... 64

PAGE 11

11 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DISSOLVED OXYGEN AS RELATED TO MINIMUM FLOWS AND LEVELS By Jason Hood December 2007 Chair: Mark Clark Major: Soil and Water Science Dissolved oxygen (DO) levels ar e a significant factor used to determine overall ecosystem health in aquatic systems. This thesis pres ents an evaluati on of the effects of low flow conditions on DO levels in several warm-water lotic systems of westcentral Florida. DO concentrations were measured at sites on the Withlacoochee, Anclote, Alafia, and Hillsborough rivers to evaluate how flow across shoals affects this water qu ality parameter. Correlations between shoal water depth and dissolved oxygen c oncentrations were used to examine how the minimum flows and levels regulatory criteria, currently used by the Southwest Florida Water Management District (SWFWMD) may influence DO levels with in the studied rivers. Peer review of a recent minimum flows and levels report for the Middle Peace River by the Southwest Florida Water Management District (SWFWMD) questioned the effect of low flow conditions on dissolved oxygen concentration. The peer review panel pointed out that the fish passage criteria currently used by the SWFWMD was origina lly developed for migratory salmonoids in cool, well oxygenated rivers and st reams in the western United States. The current criteria requires a minimum of 0.6 ft (18.3 cm ) of water passing over the lowest elevation of shoals which is also presumed to supply su fficient dissolved oxygen concentrations for fish. The panel suggested that a study be conducted to determine if the curr ent criteria is applicable to

PAGE 12

12 warm-water rivers and streams of Florida, a nd to determine the level at which low flows negatively effect water quality, primarily dissolved oxygen. Data for this study were collected over a tw o year period under medium (which for this study include discharges that are exceeded 35 to 70 percent of the time) and low (which for this study include discharges that are exceeded 70 to 100 percent of the tim e) flow conditions. During periods of medium flow, instantaneous measurements were taken above, at, and below selected shoals. Instantaneous measurements were taken during medium flow conditions to provide DO data during non-stresse d periods. There was not e nough equipment to place dataloggers at each site for the enti re two years. As river stages decreased and neared the 18.3 cm threshold, data-loggers were depl oyed above and below the shoals. DO concentrations were correlated to maximum shoal depth and compared to state standards and historical data. DO levels were also evaluated as a function of water moving across shoa ls utilizing Manning's N coefficients and Froude numbers. Shoal depth, and thereby flow, showed a wide range of correlations with dissolved oxygen for the rivers studied. Findi ngs also indicated that the cu rrent Florida Department of Environmental Protection (FDEP) dissolved oxygen cr iteria of 5 mg/l is not being met at the 18.3 cm threshold for one of the four rivers stud ied. Application of th e measured stage DO correlation to historical stage data for each river indicated that the 18.3 cm threshold in the past has maintained dissolved oxygen leve ls above 5 mg/l 90% of the time in three of the four studied rivers. In summary, this study demonstrated that additional factors, possibly rainfall, watershed land uses, and seasonality, may be influencing di ssolved oxygen concentrations in these warmwater systems and that depth, although a factor controlling dissolved oxygen concentrations, is not the only variable that should considered when trying to maintain DO levels.

PAGE 13

13 CHAPTER 1 INTRODUCTION Rationale The f ive water management districts of the State of Florida are tasked with permitting consumptive water use within their boundaries an d protecting water resources from significant harm. One component of this regulatory charge involves esta blishment of Minimum Flows and Levels (MFLs) for streams and rivers. As curren tly defined by the statute, the minimum flow for a given watercourse shall be the limit at which further withdraws would be significantly harmful to the water resources or ecology of the area (Section 373.042, Florida Statutes). For the development of MFLs in the Sout hwest Florida Water Management District (District), low flow thresholds, based on fish passage depth and wetted perimeter inflection points, are used to establish short-term complia nce standards. The District has used 18.3 cm water depth at shoal sites as suffi cient to provide for fish passage. Low flow thresholds based on this fish passage depth are developed to maintain natural patterns of river-flow continuity and thereby allow for longitudinal movement of biota within the river segment. The thresholds are also assumed to be sufficient for continued recr eational use of the river and to minimize negative water quality or physical attributes associat ed with stagnant pool conditions. Peer review of a recent Minimum Flows and Levels report for the Middle Peace River by the Southwest Florida Water Management Distri ct (SWFWMD) questioned the assumption that a 0.6 foot (18.3 cm) shoal depth provided sufficient dissolved oxygen concentration for fish health (SWFWMD 2005). The review panel point ed out that the fish passage criteria was originally developed for migratory salmonoids in cool, well oxygenated rivers and streams in the western United States and may not suitable for warm slow m oving lotic systems of western Florida.

PAGE 14

14 It is hypothesized that the 18.3 cm fish pa ssage criterion curren tly utilized by the SWFWMD is sufficient to meet the present DO criteria. This thesis will also attempt to compare DO concentrations under extreme low flow conditions to river-specific, historical data and analyze the oxygenation of water passing over shoals. Background Information on Dissolved Oxygen and Hydrologic Conditions DO levels are influenced by physical, chem i cal, and biological factors. Oxygen is introduced into riverine systems through diffu sion across the air-water interface and by photosynthetic activity of aquatic plants. Oxygen is removed from rivers by diffusion across the air-water interface, by chemical oxidative processes, and through respiration of river biota. In systems with large hydraulic storage areas, DO leve ls are often depressed during the rainy season due to the inflow of detrital material and other bio-availabl e carbon sources and stimulating biological activity (Rhinesmith and Smith, 2001 a nd Anderson and Taylor, 2001). This in turn can result in elevated biologi cal oxygen demand in the water column and depressed oxygen levels. In other systems, DO levels may be depre ssed during low flow conditions and elevated during high flow conditions. A United States Environmental Protection Agency (EPA) study of Georgia coastal plain watersheds attempted to correlate DO levels to temperature, pH, conductivity, oxidation-reduction potential, turbidity, and flow. Th e only statistically significant response suggested that low DO levels occurred during periods of low fl ow (Bosch et al., 2002). A study of Brooker Creek in Hillsborough Count y, Florida agreed with the findings of Rhinesmith and Smith (2001) that DO levels ar e suppressed when low DO waters, stored in riverine wetlands, are flushed in to the river system (United St ates Environmental Protection Agency, 2004) Because much of the flow in Br ooker Creek is derived from overland flow from agriculture areas (United States Environmenta l Protection Agency, 2004) it was theorized that

PAGE 15

15 the suppressed DO levels may be a result of elev ated nutrient levels. Upon analysis it was discovered that nutrient levels in the creek were below the statewid e averages for nitrate/nitrite, total phosphorous, and ammonia (United States Environmental Protection Agency, 2004). Drainage of low DO waters from surrounding from surrounding wetlands combined with the water stagnation caused by low stream gradient were determined to be the primary causes of the suppressed DO concentrations. Water quality plays a major role in the overall health of river ecosystems. For example, different fish species exhibit vary ing levels of tolerance to elev ated temperatures and depressed DO levels which are two of the major water quality factors affecting Florida rivers. Fish require DO levels of about 5 milligrams per liter (mg/l) for optimal health and can begin gasping and dying when levels reach 1 mg/l (Lakewatch, 200 3). The Florida state standard for DO in predominantly fresh water systems is that DO le vels shall not be less than 5.0 mg/l and normal daily and seasonal fluctuations above these levels shall be ma intained (EPA 62-302.530). This standard is applied to all Class III waterbodies of the State of Florida. Class III waterbody is a state designation for waters used for "recrea tion and propagation and maintenance of a healthy, well-balanced population of fish and wildlife" (Florida Administrative Code 62.302.5.0). Many Florida rivers do not meet this standard due to natural background conditions. In systems that have been determined to have background conditions that do not meet thes e criteria, a regulatory solution was established by the Florida Departme nt of Environmental Protection (FDEP). It states that dissolved oxygen levels that are attributable to natural background conditions or man-induced conditions which cannot be controlled or abated may be esta blished as alternative dissolved oxygen criteria (EPA 62-302.500(2)(f)). These modified regulations are known as

PAGE 16

16 Site Specific Alternative Criterion (SSAC) and have been established for areas in Florida such as the Everglades Marsh area (Weaver, 1998) a nd the Lower St. Johns River (Brooks, 2006). In addition to SSAC development, the Florid a Department of Environmental Protection (FDEP) recently completed data collection for a statewide water quality investigation. One of the primary goals of this investigation is to evaluate current DO criteria used on Floridas waterbodies. The data from this investigation w ill be available during the latter part of 2007 (Russell Frydenborg FDEP, July 24, personal co mmunication). How this investigation will affect the current dissolved oxygen criteria has yet to be decided. DO levels typically follow a diurnal cycle, increasing during the day due to sunlight increasing photosynthetic activities and decreasing during the night due to respiration (Tadessa et al 2004). DO levels typically peak during th e mid-afternoon hours and reach their minimum during the hours just before sunrise. Depre ssed DO conditions can also occur during daylight hours during cloudy days when available sunlight for photosynthesis is limited and respiration demands for oxygen exceed oxygen supply (Richard and Moss, 2002) In addition to oxygen being introduced in to the water by phot osynthetic plants, atmospheric oxygen is also fluxing across the air wa ter interface. Based principally on water temperature and atmospheric pressure, a dynamic equilibrium between oxygen concentration the water column and that in the air is balanced. When there is excess DO in the water column due to production or a rise in water temperature, DO concentration in the water can become saturated and oxygen is released to the atmosphere. If wa ter temperatures fall or oxygen is consumed, DO is less than saturated in the water column and oxygen diffuses from the atmosphere into the water column. Because of the slow diffusion ra te of oxygen in the water, water movement can significant effect the rate of oxyge n exchange across the air water in terface. In a lake or low-

PAGE 17

17 flow river, the water molecules are turning over slowly allowing a small pe rcentage of the water volume to come in contact with the atmosphere. In areas with turbulence man-made or naturally occurring, many more water molecules interact w ith the atmosphere bringing the water column more rapidly into equilibrium with the air. Reaeration models are utilized to predict the rates at which oxygen is dissolved into the water column. Variables common to these models include water surface tension, molecular diffusion, kinematic viscosity, discharge, veloc ity, stream slope, grav itational acceleration, Manning's roughness coefficient, depth, surface area, Froude number, and temperature (Gualteri et al 2002). Recent studies of reaeration models have shown many of these variable to be redundant leading to the combination or elimin ation of certain variab les in the analyses (Ramakar et al 2001). Findings of a recent study using eighteen reaeration models to compare predicted DO changes to actual changes found that the formula (Parkhurst and Pomeroy 1972), K2 = 23(1 + .17Fr2)(SV).375 H-1 where: Fr = Froude number V = friction velocity S = Slope H = depth was the most accurate (Ramakar et al 2004). Naturally occurring shoals and sh allow runs cause the water column to turn over, exposing a higher portion of the water molecules to make contact with the atmosphere. A shoal is a shallow feature of the river channel that acts to control the flow of water, at least under low flow conditions. This control point holds back th e flow of water causing a higher water surface elevation on the upstream side of the shoal than if the river bottom followed a smooth fall. Because the atmosphere generally has higher oxy gen content than the water, the dissolved

PAGE 18

18 oxygen levels in the water flowing through shoals are usually raised as a result of diffusion across the water/atmosphere interface. This study is driven by three main hypotheses: 1) DO concentrations are maintained within the river specific hist orical DO ranges when flows necessary to maintain an 18.3 cm maximum shoal depth are met or exceeded. 2) DO concentrations are maintained above the FDEP Class III waterbody standard of 5 mg/l when flows necessary to maintain an 18.3 cm maximum shoal depth are met or exceeded. 3) Physical shoal characteristics represente d by Froude numbers and Manning's N values influence reaeration of water passing over studied shoals.

PAGE 19

19 CHAPTER 2 MATERIALS AND METHODS Site Selection Site se lection for this study was completed in two ways. Sites for three rivers were selected from sites previously studied by the Dist rict. The site on the fourth river was selected using the same method that the Di strict utilized to pick previ ously studied sites. Prior to initiation of this DO study, the District had completed data collection efforts supporting development of MFLs on the Anclote, Alafia, and Hillsborough rivers. One previously utilized District site was selected on each of these three rivers. On the Withlacoochee River, a site was selected using the same site selection criteria. Th ese four rivers were selected due to variability in their size, water chemistry, water color and watershed type. One aspect of the MFL process is analyzi ng representative contro l points throughout the stretch of the river. Data coll ected on these control points are used to model usable habitat under a full range of flow conditions using a Physical Habitat Simulation model (PHABSIM). These control points, also referred to as shoals, are selected based on the following criteria (Hood, 2007): sites should be representative of the river reach for which minimum flows are to be developed sites should be reasonably accessible sites should typically include a pool and a run upstream of the shoal (in situations where this is not possible, a pool, run or shoal can be selected individually sites should, if possible, not include significant bends in the river channel sites should show no evidence of braiding during high flow sites should be located between es tablished river gauging stations.

PAGE 20

20 Study Location and Site Descriptions Rivers selected for research included th e W ithlacoochee, Alafia, Anclote, and Hillsborough in west-central Florida (figure 2-1) Study sites on each river were in Hernando, Hillsborough, and Pasco Counties. The studied rivers include clear water to dark water systems with small to large floodplains. Anclote River Anclote Riv er is a small river originating in central Pasco County. This river meanders through Starkey Wellfield / Wilderness Park, pass es under Starkey Boulevard, where it enters a highly developed residential area, an d eventually discharges into th e Gulf of Mexico (figure 2-2). Anclote River has a contributing drai nage area of approximately 188 km2 at the studied site (USGS, 2007(1)) and exhibits a relatively low relief over its 42 km. Flow records for the USGSs gauging station at Elfers are from 1945 un til current. Average discharge for the period of record is 1.88 m3/sec (USGS, 2007(1)), with the major ity of flow occurri ng during the rainy season from June through September (figure 23). Minimum and maximum daily averages for discharge were 0.023 and 105 m3/sec, respectively, and both o ccurred during the early 1960s (USGS, 2007(1)). The site studied on the Anclote River was called Site 1. This site is located approximately one mile upstream of the Starkey Boulevard bridge (figure 2-2) at 28.22182N and -82.63435W (decimal degrees format). This site consists of a shoal area with a pool upstream and a run downstream. The bottom contour of the shoal perpendicular to flow is a gentle slope (figure 2-4) with a sand and snag substrate. Average depth in the pool for the entire study period was approximately 122 cm and average depth in the run was approximately 76.2 cm. Average maximum depth at the shoal for the entire study period was 50.3 cm (depicted in Figure 2-5).

PAGE 21

21 Anclote River in this area is generally in cised having banks aver aging approximately 244 cm with some areas dropping much lower where side channels enter and exit the main river. The Substrate is predominately sand with scattered exposed limerock and has a moderate amount of woody debris and snags. Flora in the area is dominated by Taxodium distichum and Salix spp. in river bottom and transitions to pine flatwood ecosystem, dominated by Pinus elliottii, Serona repens, Quercus virginiana, Quercus laurifolia, and Vaccinium corymbosum outside of the incised channel. During the study period the rainfall for the wa tershed was 65% below the 92 year average (Figure 2-6). Rainfall for the year the study was conducted was 19% below the 92 year average. Alafia River Alafia River is a m edium sized river origin ating in eastern Hillsborough and western Polk counties and is fed by two major tr ibutaries, the North and South Prongs (figure 2-7). The river runs approximately 30.6 km through quickly developing urban and industrial areas before emptying into Tampa Bay. Alafia River has a contributing drainage area of approximately 868 km2 at the studied site (USGS, 2007(2)). The flow record for the USGSs gauging station at Lithia is from 1932 until current. Average discharge for the period of record is 10.53 m3/sec (USGS, 2007 (2)) with the majority of flow occurring dur ing the rainy season, from June through September (figure 2-8). Minimum and maximum daily averages for discharge were .91 (1945) and 139.2 (1959) m3/sec respectively (USGS, 2007(2)). The study site on the Alafia River was named ALA2 and is located a short distance downstream of the confluence of North and South Prongs (820831.6W 275151.6N). Bottom contour of the shoal perpendicular to flow cons ists of high banks and a bottom lined with sand and snags (figure 2-9). Site consists of a s hoal area with a run upstream and a run downstream

PAGE 22

22 (figure 2-10). Average maximum depth at the shoal for the entire st udy period was 20.7 cm. Flora in the area surrounding the shoal includes Taxodium distichum and Salix spp in and within 9 meters (m) of the river channel with hi gher elevations dominated by upland vegetation including Quercus virginiana, Quercus laurifolia Pinus elliottii, wild Citrus sinesis, and Serona repens Rainfall from the beginning of 2006 through the end of May was 47% below average (figure 2-11). Rainfall for the enti re year of 2006 was 10% below average. Hillsborough River Hillsboroug h River is a medium-large sized rive r originating in Green Swamp in Pasco and Polk counties (figure 2-12). Green Swamp is an estimated 2253 km2 (Hills. River WMP 2000) and is also the headwaters of the Peace, Ok lawaha, and Withlacoochee Rivers. Hillsborough River travels approximately 87 km before empt ying into Hillsborough Bay (figure 2-1). The study site is located in nort heastern Hillsborough county in Hillsborough River State Park (821343.3W 280902.1N). Hillsborough River has a contributing dr ainage area of approximately 748 km2 at the studied site (Hillsborough River WMP 2000). Flow record for the USGSs gauging station near Zephyrhills is from 1939 until present. This US GS site also had a dissolved oxygen probe from 2001 until 2003. Minimum and maximum daily averages for discharge were 0.76 (2000) and 348 (1960) m3/sec respectively (USGS(3)). Average discharge for the period of record is 6.9 m3/sec (USGS(3)) with the major ity of flow occurring during the rainy season, from summer June through September (figure 2-13). The bottom contour of the shoal perpendicular to flow consists of high banks and a bottom comprised of limerock outcroppings and log jams (fi gure 2-14). Site consis ts of shoal area with

PAGE 23

23 a run upstream and a run downstream (figure 2-15) Average maximum depth at the shoal for the entire study period was 52.7 cm. Due to relatively frequent flooding of Hills borough River and low relief of surrounding terrain, the floodplain in the area of the shoal is quite expansive. Many side channels parallel the main river and are inundated unde r various levels of flooding. Fl ora within a hundred yards of river consists of Taxodium distichum and Salix spp. in the lower areas and Quercus virginiana, Quercus laurifolia Pinus elliottii wild Citrus sinesis Serona repens Liquidambar styraciflua and Sabal palmetto at higher elevations. Withlacoochee River W ithlacoochee River is a medium-large sized river originating in Green Swamp. This river has a contributing drainage area of approximately 2098 km2 at the studied site (USGS (4)). Study site is located on the northeast e dge of Hernando County (821307.2W 283456.6N) with the river marking the boundary between Hernando and Sumter (figure 2-16). Flow records for the United States Geological Su rveys gauging station near Croom are from 1939 until present. Average discharge for the period of record is 12.7 m3/sec (USGS(4)) with the majority of flow occurring during the rain y season from summer through late fall (figure 2-17). Minimum and maximum daily averages for discharge were .13 (1981) and 244 (1960) m3/sec respectively. The bottom contour of the shoal perpendicular to flow consists of low, gently sloping banks and a bottom of sand and submerged aquatic vegetation (sav) (figure 2-18). Site consists of shoal area with a run upstream and a pool downstream (figure 2-19). Average maximum depth at shoal for the entire study period was 22.3 cm. Gently sloping banks and low relief of surr ounding terrain lead to a very wide floodplain. Within the river bottom, flora is dominated by Hydrilla verticillata Salix spp., and Faxinus caroliniana. Surrounding floodplain is dominated by Taxodium distichum and Acer rubrum.

PAGE 24

24 Adjacent uplands are a scrub-shrub habitat dominated by Quercus laurifolia, Quercus laevis, Quercus marilandica, and Serona repens. Rainfall at the site during the data-logging portion of the study was below the 92 year average for all but one month (figure 2-20). Data Collection Equipm ent used for data collection incl uded YSI 600XLM multi-probes, survey level instruments (Topcon G4), and level rods. Multiprobes were used for discrete water quality measurements and unattended data-l ogging of water quality paramete rs and field computers were used to download logged data. Survey levels and rods were used to determine elevations (relative or NGVD) of site cro ss sections and thalwags. The data collection period ran from November 30, 2005 through June 29, 2007 and generally occurred on one river at a time due to equipment restraints. Data collection was carried out in two distinct manners. During periods of medium to high flow conditions, discrete, instantaneous measurem ents of pH, specific conductance, temperature, depth, dissolved oxygen (mg/L), and dissolved oxygen (%). Each time data were collected for a particular site, readings were ta ken at numerous locations in relati on to the shoal. The first two measurements at each site also included samp ling for field parameters one meter from each stream bank and in the middle of the stream. Afte r analyzing data from the first two visits to each site it was determined that mid-depth, mid-channel measurements sufficiently represent the in-situ conditions. An attempt wa s made to collect these data at approximately the same time of day (around noon) on each visit to minimize vari ation due to the diurnal fluctuation of temperature and dissolved oxygen. During periods when water levels dropped to the point that maximum depth over a shoal was near the presumed critical level (18.3 cm fish passage threshold), multi-probes were

PAGE 25

25 deployed 10m upstream and 30m downstream of shoa ls at similar depths and were set-up to log water quality data every 30 minutes. Data-loggers were deployed for varying periods of time based on flow conditions. For the purpose of data-logging, the YSI multi-probes were tethered inside a 10.2 cm diameter PVC pipe with holes drilled to allow free flow of water. The PVC pipe was bolted on top of a concrete block to stabilize and elevate th e probe (figure 2-21) 18.3 cm above the bottom of the river bottom. Parameters collected during data-logging phase include pH, specific conductance, temperature, dissolved oxygen concentration, pe rcent saturation of dissolved oxygen, and depth. Each of these parameters was logged at half hour intervals. Data from multi-probes were collected weekly, at which time the multi-probes where calibrated and any necessary maintenance was performed. Although each multi-probe logs depth, maximu m depths at each studied shoal were measured on each field visit. These collected values were used to verify data collected by the multi-probe as well as making data corrections in the occurrence of drift. Logged depth values were also analyzed to determine if the probes ha d been tampered with and, if so, which data needed to be discarded. Data Analysis Data were analyzed in severa l different ways in an effort to determ ine physical factors affecting DO concentrations. Primary analysis for this study was the determination of the correlation between DO concentrations and maximu m shoal depth. In order to compare the DO change as water passed over the shoal area to the physical characteristics of each of the shoal areas, Manning's n values were compared to ch ange in DO from upstream to downstream of the shoals. The majority of reaeration equations use a variable to quan tify the roughness of the streambed. These variables include Manning's n, Re ynolds numbers, friction velocity and others

PAGE 26

26 (Ramakar et al 2004). Mannings N values were selected for use in this study due to the accepted use of them by instream flow scient ists. Although both Froude and Reynolds numbers are both widely used in reaeration models to characterize flow conditions and both take depth, and velocity into account, Froude numbers were chosen for this study due to the findings of Ramakar et al (2004) that indi cated that reaeration model th at predicted reaeration most accurately utilized Froude numbers in the calculation (Ramakar 2004). As part of the primary data analysis for th is study, determinations were made of the maximum shoal depth needed to meet two distinct DO concentration criteria. The first criterion was the FDEP standard for Class III waterbodies of 5 mg/L. The second criteria utilized to determine the maximum depth necessary at th e shoal was DO concentration based on riverspecific historical data. DO concentration data were retrie ved from USGS, FDEP and District databases for evaluation of historical or background conditions in the four studied rivers. Dissolved oxygen, rainfall, stage, and discharge data have, and continues to be collected by many agencies and organizations. External DO da ta gathered from the USGS at or near study sites were all collected in-situ during daylight hours. Due to these data only being collected during daylight hours, which is when DO levels r each their daily highs, considerations must be made when comparing historic values with 24 hour logging data collected for this study. A secondary analysis of the da ta looked at the physical component of water moving across a shoal in an effort to quantify DO change from upstream to downstream of shoals. Output data from the PHABSIM model were used to determ ine average Froude number and Manning's N for the three rivers previously studied by the Dist rict. Manning's N values are calculated by an algorithm in the PHABSIM system and represent friction as a result of surface roughness and sinuosity. These values are produ ced using the equation (Chow 1959):

PAGE 27

27 Q = (R2/3S1/2/n)A Where: Q = discharge A = cross-sectional area of flow n = coefficient of roughness R = hydraulic radius S = slope of the water surface. In river systems, the higher the Manning's n value, the higher the likelihood of water column turnover. Froude numbers (Fr) are al so produced using the PHABSIM system and describe flow conditions (Table 2-1). Values are produced by the equation (Jain, SK and Jha, R 2004)): Fr = (V)2 / (gD) Where: V = velocity g = gravitational constant D = hydraulic depth. Along with Manning's n, Froude numbers have been shown to be an important variable in the calculation of reaeration rates (Ramakar et al, 2004). Froude numbers quantify the energy at the water/atmosphere interface and indicate levels of air entrainment (Moog and Jirka, 1999). These factors were plotted against change in dissol ved oxygen, from upstream to downstream, to determine what, if any, relationship exists under low flow conditions. These numbers were averaged for each site for increm ental flow and stage conditions for comparison with the collected DO data. Withlacoochee River has not been analyzed using the PHABSIM and therefore lacks Froude numbers, Manning's N values and subsequent comparisons.

PAGE 28

28 Figure 2-1. Location of four study sites. Figure 2-2. Site Map for Anclote River.

PAGE 29

29 Figure 2-3. Typical hydrograph for Anclote River (Based on USGS data from 1945 2004). Figure 2-4. Anclote River Site 1 cross section (location corr esponds to X=0 in figure 2-5). Figure 2-5. Anclote River Si te 1 mid-channel profile.

PAGE 30

30 Figure 2-6. Rainfall comparison for study year and 92 year aver age for the Anclote watershed. Figure 2-7. Alafia River DO site.

PAGE 31

31 Figure 2-8. Typical hydrogra ph for Alafia River (Based on USGS data from 1932 2007). Figure 2-9. Alafia River site shoal cross sect ion (location corresponds to X=0 in figure 2-10). Figure 2-10. Alafia River site channel profile.

PAGE 32

32 Figure 2-11 Rainfall comparison for study year and 92 year aver age for the Alafia watershed. Figure 2-12. Hillsborough River DO site.

PAGE 33

33 Figure 2-13. Hydrograph for Hillsborough Rive r (Based on USGS data from 1940 2007). 0 0.5 1 1.5 2 2.5 3 3.5 0 5 10 15 20 25 30 Horizontal Distance from Left Bank to Ri g ht Bank ( m ) Land Surface Elevation (m-relative) Figure 2-14. Hillsborough River site shoal cross section (location corresponds to X=0 in figure 2-15). -0.6 -0.4 -0.2 0 0.2 0.4 0.6 -40-30-20-10010203040 Lon g itudinal Distance from Shoal ( m ) Land Surface Elevation (m-relative) s h o a l >>FLOW>> Figure 2-15. Hillsborough River site channel profile.

PAGE 34

34 Figure 2-16. Withlacoochee River DO site. 0 5 10 15 20 25 30 JFMAMJJASOND MonthAverage Monthl y Q (cms) Figure 2-17. Hydrograph for Withlacoochee River (Based on USGS data from 1939-2007).

PAGE 35

35 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 02 04 06 08 01 0 01 2 0 Horizontal Distance from Left Bank to Right Bank (m)Land Surface Elevation (m relative) Figure 2-18. Withlacoochee River site cross s ection (location correspond s to X=0 in figure 219). Figure 2-19. Withlacoochee River site channel profile. Figure 2-20 Rainfall comparison fo r study year and 92 year averag e for the Withlacoochee River Basin.

PAGE 36

36 Figure 2-21. Picture of a YSI multi-probe mounted in PVC tube. Table 2-1. Froude number descriptions. Fr < 1 Subcritical Flow. Streaming flow where disturbances can travel upstream Fr = 1 Flow is critical. Fr > 1 Supercritical Flow. Shooting flow wher e downstream disturbances cannot be felt upstream

PAGE 37

37 CHAPTER 3 RESULTS AND DISCUSSION Data were collected on Anclote, Alafia, H illsbo rough, and Withlacoochee Rivers. At the first three rivers mentioned above, data were collected by discrete measurements and by datalogging. At Withlacoochee River, data were onl y collected by data-logging Data herein are presented by river. Anclote River Discrete Measurement Data The site on the Anclote River was nam ed Anclot e Site #1. Discrete measurement data for this site were collected when the maximum s hoal depth was at 127 cm and at 162 cm. DO during discrete measurements ranged from 6.7 mg/l to 8. 0 mg/l. Discrete measurement data illustrated an increase in DO concentration und er lower water levels than unde r higher water levels (figure 3-1). When an average for all DO concentrati on and temperature measurements collected during the discrete measurement portion of the study are compared, there is a good correlation (R2 = 0.6227) (figure 3-2). Continuous Measurement Data For the data-logging portion of the Anclote Rive r data collection, data were collected f or 40 days. During this period the maximum depth at the shoal dropped from nearly 183 cm to a minimum depth of around 18.3 cm (figure 3-3). During this period there was a very good (R2 = 0.8712) correlation between maximum shoal depth and DO concentration (fi gure 3-4). Further investigation of continuous data indicated that the last three da ys of data for Anclote Site #1, when water depth over the shoal was at its sh allowest, had the greatest increase in DO concentration between upstream and downstream data-loggers.

PAGE 38

38 Although the Anclote River site shows a gene ral increase in DO concentration from upstream to downstream of the studied shoal, ph ysical shoal parameters, Froude number (figure 3-5) and Mannings N number (figure 3-6) do no t show a correlation with DO concentration change across the shoal. Water depth and DO levels fell for the Anclot e River site during th e study period except for a few days after a rainfall event in late Febr uary. During the rainfall event, an estimated 0.66 cm of rain was received in the area. The rise in DO after a rainfall is quite common due to input of well oxygenated water. However, if Anclote River had a more substa ntial floodplain or if rainfall had been more substantial, results may ha ve been reversed due to inflow of water with elevated biological oxygen demands (BOD). Comparison with FDEP Class III DO Criteria Based on FDEP standards for DO on Class III waterbodies, DO of Anclote River m ust maintain a minimum concentration of 5 mg/l. Using this criterion and data collected during this study, it appears that the minimum sh oal depth that can be reached before DO levels do not meet the standard would be approximately 38.1 cm. Fl ow at which this depth is maintained is 0.12 m3/sec. This flow is met or exceeded 73% of the time (based on 60 years of USGS discharge data). The EPA and DEP allow review of site-s pecific DO standards when sufficient predisturbance data are available to contest the applicability of the statewide 5 mg/l standard. These exceptions are referred to as site specific al ternative criteria (SSAC) and are set by analyzing pre-impact datasets and or datasets from simila r non-impacted waterbodies within the region and by determining the minimum criteria needed by the waterbodies' most sensitive organisms. Although a SSAC has not been establis hed for the Anclote River, historical data shows that this waterbody has DO concentrations regularly below th e EPA Class III standard (Table 3-1). If

PAGE 39

39 this historic low DO concentration is determined to be naturally occurring, a SSAC for DO that is lower than state wide criterion should be consid ered as it will likely be more protective of least impacted conditions and a more attainable DO concentration. Comparison with Historic DO Measurements The USGS a nd the SWFWMD have been ta king water quality measurements at their Anclote River near Elfers, FL gauging st ation since 1962 and have collected 278 DO measurements (figure 3-7 and Appendix A). This USGS site is 3.1 km downstream of the site used for this study but is still in the upper portion of the wate rshed. Results of a Kendall's tau analysis on residuals for water quality parameters regressed against flow indicate a significant increasing trend in DO concentration over the period of record (table 3-2). Historical DO range is 1.20 to 10.4 mg/l. For 48% of measurements DO concentrations were below 5 mg/l and 28% of measurements were below 4 mg/l (Table 3-2). A correlation between DO and stage for the historical data collected at this USGS site indi cates that most occurrences of DO concentrations below 4 mg/l correspond to low flow periods (figure 3-8). At the Anclote River site (figur e 3-4), it is apparent that DO drops as water depth drops. However, as discussed in Chapter 1, as a shoa l water depth becomes shallower, water becomes more oxygenated as it moves across the shoal. DO concentration in the Anclote River occasionally falls below 2 mg/l and recognizing th e oxygenating ability of shoals (last few days presented in Figure 3-3), it is sp eculated that waters downstream of shoals are providing a refuge to mobile organism and may ensure the survival of a percentage of non-mobile organisms during critical periods. At the 18.3 cm threshold, currently used by the SWFWMD when setting MFLs, the DO concentration is being elevated as much as 0.8 mg/l as water passes over the shoal. The DO concentration fell below the historical ra nge (1.2 mg/l) when the maximum shoal depth

PAGE 40

40 dropped below 19.8 cm. Depths this shallow or shallower occurred 5% of the time based on 60 years of USGS discharge data. Alafia River Discrete Measurement Data One site was selected for data collection on Al afia River (figure 2-6). Data for Alafia River were collected from March through Apr il in 2006. Discrete m easurement data were collected on four occasions for this site (figur e 3-9) with maximum shoal depths ranging from 146 cm to 241 cm. DO concentrat ion during this range of water depths was from 7.5 mg/l to 10 mg/l. Discrete measurements suggest no significant effect of the shoal on DO concentration at least within this range of water depths. Continuous Measurement Data For the Alafia River site, data-logging data were collected for one m onth. During this period, the maximum depth at the shoal fluctuated between a 15.2 cm and 30.5 cm (figure 3-10). DO levels ranged from 6.5 to nearly 11 mg/l. There was no significant correlation to shoal water depth and DO concentration (figure 3-11). This lack of correlati on held true even when diurnal variability in DO concentrations due to photos ynthesis was taken into account by comparing only data from the same time each day. At this site, even during low flow conditions, where the maximum depth was below the threshold for fish passage (18.3 cm), DO concentrations never fell below 5 mg/l or outside the historical range.. When plotted against DO change across the shoal, neither Manning's N (figure 3-12) nor Froude number (figure 3-13) appear to have any relationship with DO. It is speculated that under the narrow range of flows analyzed for this study, there is not a wide enough variation in Manning's N values to make a meaningful comp arison. With all of the Froude numbers being

PAGE 41

41 well below the critical threshold, air entrainment was negligible under all of the flow conditions measured. Rainfall data indicated that there was only one rain event during the period of data collection. Less than 0.25 cm of rain fell on th e area on April 9, 2006 and di d not appear to have any impact on the DO concentration. Comparison with FDEP Class III DO Criteria Although the m aximum shoal depth fell below th e 18.3 cm fish passage criteria during the data collection period, DO remained above the 5 mg /l FDEP standard for th e entire study period. Comparison with Historic DO Measurements USGS has been taking water qu ality m easurements at Alafia River at Lithia Springs gauging station since the late 1960's and data collect ed for this study mirror historical data quite well (table 3-3 and Appendix A). This site is 4 km downstream of the site used for this study but is still in the upper portion of the watershed. Re sults of a Kendall's tau analysis on residuals for DO regressed against flow indicate a statistica lly significant downward trend in DO over the period of record (table 3-4). Even with the downward trend, Alafia Rive r typically exhibits DO levels above 5 mg/l even under low flow conditions (table 3-3). A correlation between historical stage and DO concentrations showed that all DO values below 4 mg/l occurred during low flow periods (figure 3-14). All DO c oncentration measurements duri ng the study period were within the historical range. Hillsborough River Discrete Measurement Data Discrete m easurement data were collect ed at Hillsborough River State Park for approximately three weeks beginning in mid June, 2006. Discrete measurement data from the

PAGE 42

42 site illustrated that water passi ng over the shoal is aerated to a greater extent under low flow conditions than under higher flow conditions (figure 3-15). Continuous Measurement Data During the data-logging period, the m aximum depth at the shoal fluctuated around 52 cm. DO levels ranged from 6 to nearly 8 mg/l (fi gure 3-16). DO concentrations were generally higher downstream of the shoal than upstream indi cating the ability of th e shoal to oxygenate water flowing over the shoal. During the st udy period, there appears to be no correlation between depth and DO (figure 3-17). In an effort to remove the diurnal effect from the data, a comparison was made between depth and DO for only measurements logged at noon each day. Results of that correlation were even weaker. Du e to flows being in the medium range, based on historical flow data, and the rather long antic ipated time before the river would drop to the desired study range, data-loggers were pulled out and deployed at another site. There were hopes of returning the equipment at a later da te, but scheduling and flow conditions never allowed. Comparison with FDEP Class III DO Criteria DO concentrations during the entire study period rem ained above the 5 mg/l FDEP standard. Comparison with Historic DO Measurements USGS has been tak ing water quality measurem ents at Hillsborough River near Zephyrhills gauging station since the mid 1970's. This site is just downstream of the site used for this study. Results of a Kendall's tau analysis on residua ls for DO regressed against flow indicate no significant trend in dissolved oxygen over the period of record (table 3-5). DO was measured by USGS at this gauging station from 2001 through 2003. A plot of their data from this period points out two interesting facts (f igure 3-18). First, the DO levels remain relatively high over the

PAGE 43

43 entire range of stages. Second, lo west DO levels generally occur when stage first starts to rise at the beginning of the rainy season. This is to be expected due to large numbers of riverine wetlands and the large, low-relief flood plain at, and upstream, of the study site. It appears these early-season rain events flush high BOD waters and detrital material from surrounding wetlands and swamps into the river causing increased oxyg en demand and depressing DO concentrations. A comparison between historical discharge and discharge during th e study period shows the lack of the typical rainy season for the studied years (figure 3-19). Withlacoochee River Continuous Measurement Data One site was selected for data collecti on on Withlacoochee River, which is just downstream of a wide spot in th e river named Silver Lake (figur e 2-14). Only data-logging data was collected at the Withlacoochee River site for six and a half months, from November 9, 2006 through June 28, 2007. During this period, the maxi mum depth at the shoal ranged from 0.0 cm to 40 cm (figure 3-20 & 3-21). DO levels ranged fr om 0 to nearly 20 mg/l. A plot of maximum shoal depth and DO at this site show s a good correlation (figure 3-22). Comparison with FDEP Class III DO Criteria Dissolved O xygen concentration above the shoal remained above the 5 mg/l FDEP standard until the maximum shoal depth dropped slightly below the 18.3 cm threshold currently used by the SWFWMD for fish passage. DO below the shoal fell below the 5 mg/l standard when the maximum shoal depth was slightly above 24 cm. At this site, and none of the others, there is an average loss of DO acr oss the shoal. It is speculated that this was due to the downstream data-logger being placed in a small pool off to the side of the main channel. The downstream data-logger was placed off to the side because of the large number of airboats navigating the river. Probe was placed in the only safe place downstream of the shoal so that it

PAGE 44

44 would not be crushed by one of the airboats After the study was complete, discrete DO measurements were taken in the main channel and off to the side wh ere the downstream datalogger had been placed. Each of the thr ee times that this comparison was made, DO concentration was between 2 and 3 mg/l lower wher e the data-logger had been placed than in the main channel. For this reason, the downstream data-logger data are not believed to be a good representation of DO concentrations downstream of the studied shoal. Both upstream and downstream sampling locations exhibited periods with DO levels above 5 mg/l when the maximum shoal depth was well below the 18.3cm threshold. Conversely, both locations had periods of DO below the 5 mg/l threshold when the maximum shoal depth was greater than 18.3 cm. Due to poor correlations between water depth at the shoal and DO concentration, it appears that f actors other than water depth have a significant effect to DO levels. It should be noted that during da ta-logging at this site, Silver Lake changed from an initial condition of moderate to low levels of aquatic vegetation (primarily Hydrilla ) and algae to one that was almost completely clogged with aquati c vegetation and algae (f igures 3-23 and 3-24). The resulting effect on release and uptake of O2 in the water column becomes apparent in diurnal cycles for the upstream data-logger towards the end of the study (figure 3-25). Due to high and low DO concentrations o ccurring in Silver Lake, an interesting phenomenon was recorded at the Withlacooch ee River site starting in early May, 2007 when flows were at their lowest. During this period, widely fluctuating DO levels in the water coming out of Silver Lake were moderated by the s hoal (figure 3-25). During the daytime, when DO levels coming from Silver Lake often reached su per saturated concentrations nearing 20 mg/l, oxygen was released to the atmosphere as it cr ossed the shallow shoal bringing DO levels down

PAGE 45

45 to approximately 8 mg/l. Overnight, when DO pl ummeted to anoxic levels in the water coming from Silver Lake, the opposite occurred where DO concentrations were elevated to around 5 mg/l. In this situati on, the shoal acted as a buffering mech anism to moderate extreme levels created in the lake. Comparison with Historic DO Measurements USGS has been taking water qu ality m easurements at thei r Withlacoochee River near Croom gauging station since the la te 1960's and data collected duri ng this study mirror historical data (Appendix A). This site is just downstream of the site used for this study. Results of a Kendall's tau analysis on residuals for DO regresse d against flow indicate no significant trend in dissolved oxygen over the period of record (table 3-6). USGS collected 122 DO measurements betw een 1967 and 2005. Of the measurements collected by the USGS, 59% were below the 5 mg/l FDEP standard and the historical range is from 1.6 to 5.7 mg/l (Table 3-7). DO upstream of the study shoal remained within the historic range until the maximum shoal depth dropped to approximately 7 cm. Downstream DO concentration dropped below the 5 mg/l threshold when the maximum shoal depth dropped below 23.2 cm. It is again speculated that this variability is due to the positioning of the downstream data-logger and is assumed to be in valid data for downstr eam representation. A correlation between stage and DO concentration fo r the Withlacoochee River historical data indicates that low DO concentrations occur at all stages (figure 3-26). It should be noted that the av erage time of day the USGS water quality parameters were collected at the Withlacoochee River near Cr oom was 12:27 P.M. (with a minimum time of 7:30 A.M. and a maximum time of 6:30 P.M.). At this time of day, due to diurnal cycles, DO concentrations are nearing thei r daily maximum. When data from the data-logging portion of this study, collected during midday hours are analyzed, the mini mum DO concentration is 2.52

PAGE 46

46 mg/l (Table 3-8), indicating that DO concentratio ns collected during this study were within the historical range, even when the shoal was dry. Figure 3-1. Anclote River Site discrete measurement DO data. y = -0.908x + 22.995 R2 = 0.622710 12 14 16 18 20 22 23456789 Dissolved Oxygen (mg/L)Temperature (C) Figure 3-2. Water Temperature compared to Dissolved Oxygen concentrations for Anclote River site (discrete measurements).

PAGE 47

47 Figure 3-3. Time series anal ysis of DO concentration (bot h upstream (US) and downstream (DS)), maximum shoal depth, and rainfall data collected at Anclote River site. y = 0.1218x 0.4638 R2 = 0.87120 2 4 6 8 10 15253545556575 Max. Shoal Depth (cm)D.O. (mg/l) Figure 3-4. Maximum shoal water depth compared to DO concentration for Anclote River site.

PAGE 48

48 y = -6.4193x + 0.2028 R2 = 0.078-0.5 0 0.5 1 1.5 2 00.0050.010.0150.020.0250.03Froude NumberD.O. Change (mg/l) Figure 3-5. Correlation between DO change acro ss shoal and Froude number for Anclote River site. y = -10.553x + 0.6647 R2 = 0.0543-0.5 0 0.5 1 1.5 2 0.040.045 0.050.055 0.06Manning's ND.O. Change (mg/L) Figure 3-6. Correlation between DO change ac ross shoal and Manning's N for Anclote River site.

PAGE 49

49 Table 3-1. Summary Statistics fo r DO data collected at the USGS gauge at Anclote River Near Elfer's, FL. Statistic Value mean (mg/L) 5.03 median (mg/L) 5.07 min (mg/L) 1.20 max (mg/L) 10.40 90% exceedance (mg/L) 3.20 % of readings below 5mg/L 48% % of readings below 4mg/L 28% % of readings below 3mg/L 11% % of readings below 2mg/L 3% 0 2 4 6 8 10 12 Jan-60Jan-70Jan-80Jan-90Jan-00Jan-10D.O. (mg/L) Figure 3-7. Historical DO concentrations for Anclote River near Elfers, FL (USGS).

PAGE 50

50 Figure 3-8. Correlation between stage and DO concentrations for Anclote River near Elfers, FL (USGS) historical data. Table 3-2. Statistical analyses for DO data collected at the USGS gauge at Anclote River Near Elfers, FL indicating a significant decreasing trend in DO. Parameter Residual Residual Median n p Value InterceptSlope Dissolved Oxygen -0.0524 199 0 4.4665 0.0002

PAGE 51

51 Figure 3-9. Alafia River site discrete measurement DO data. Figure 3-10. Time series analysis of DO concentration (both upstream (US) and downstream (DS)), maximum shoal depth, and rainfall data collected at Alafia River site.

PAGE 52

52 y = -0.0032x + 8.2205 R2 = 0.00016 7 8 9 10 11 14.0019.0024.0029.0034.00 Max. Shoal Depth (cm)D.O. (mg/l) Figure 3-11. Maximum shoal water depth compared to DO concentration for Alafia River site. y = -6.5515x + 0.4793 R2 = 0.0324 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.030.0350.040.0450.05 Manning's ND.O. Change (mg/L) Figure 3-12. Correlation between DO change across shoal and Manning's N for Alafia River site.

PAGE 53

53 y = 6.2203x + 0.1863 R2 = 0.0031 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 00.0010.0020.0030.0040.0050.0060.0070.008 Ave. Froude #D.O. Change (mg/L) Figure 3-13. Correlation between DO change acro ss shoal and Froude number for Alafia River site. Table 3-3. Summary Statistics for DO data collected at the USGS ga uge at Alafia River at Lithia Springs. Statistic Value mean (mg/L) 7.2 min (mg/L) 3.5 max (mg/L) 14 90% exceedance (mg/L) 5.2 % of readings below 5 mg/L 0.03 % of readings below 4 mg/L 0.01 % of readings below 3 mg/L 0 % of readings below 2 mg/L 0 Table 3-4. Statistical analyses fo r DO data collected at the USGS gauge at Alafia River at Lithia Springs indicating a significant in creasing trend in DO (SWFWMD 2005). Parameter Residual Residual Median n p ValueIntercept Slope Dissolved Oxygen -0.057 194 0 0.514 -0.00002

PAGE 54

54 0 2 4 6 8 10 12 14 16 60110160210260310360410460510560 Stage (cm)DO (mg/l) Figure 3-14. Correlation between stage and DO con centrations for Alafia River near Litia, FL (USGS) historical data. Figure 3-15. Hillsborough River site discrete measurement DO data.

PAGE 55

55 Figure 3-16. Time series analysis of DO concentration (both upstream (US) and downstream (DS)), maximum shoal depth, and rainfall da ta collected at Hillsborough River site. y = -0.0093x + 7.8127 R2 = 0.00276 6.5 7 7.5 8 8.5 45 50 55 60 Max. Shoal Depth (cm)Dissolved Oxygen (mg/l ) Figure 3-17. Maximum shoal water depth compar ed to DO concentration for Hillsborough River site.

PAGE 56

56 Table 3-5. Statistical analysis for D.O. data collected at th e USGS gauge at Hillsborough River near Zephyrhills (SWFWMD 2005). Parameter Residual Residual Median n p ValueIntercept Slope Dissolved Oxygen -0.0787 259 0.97 -0.0988 0.00003 0 1 2 3 4 5 6 7 8 9 101 0 / 1 / 2 0 0 0 4 / 1 9 / 2 0 0 1 1 1 / 5 / 2 0 0 1 5 / 2 4 / 2 0 0 2 1 2 / 1 0 / 2 0 0 2 6 / 2 8 / 2 0 0 3 1 / 1 4 / 2 0 0 4Dissolved Oxygen (mg/l) & Stage (m) Stage D.O. (ave daily) Figure 3-18. Time series analysis of US GS stage and DO continuous data for Hillsborough River near Zephyrhills.

PAGE 57

57 Figure 3-19. Mean discharge (USGS) and discharge during study period for Withlacoochee River near Croom. Figure 3-20. Time series analysis of DO concentration (upstream (U.S.) of shoal), maximum shoal depth, and rainfall data collected at Withlacoochee River site.

PAGE 58

58 Figure 3-21. Time series analysis of DO concen tration (downstream (D.S.) of shoal), maximum shoal depth, and rainfall data collected at Withlacoochee River site. y = 0.2122x + 3.136 R2 = 0.5051 0 5 10 15 20 0.00 10.00 20.00 30.00 40.00 Max. Shoal Depth (cm)Dissolved Oxygen (mg/L) Figure 3-22. Correlation between Maximu m Shoal Depth and Dissolved Oxygen at Withlacoochee River Site.

PAGE 59

59 Figure 3-23. Picture depicting the downstream end of Silver Lake on the Withlacoochee River without choking of aquatic vegetation and algae (Photo by R. Gant, SWFWMD). Figure 3-24. Picture depicting the downstream end of Silver Lake on the Withlacoochee River choked with aquatic vegetation and al gae (Photo by R. Gant, SWFWMD).

PAGE 60

60 Figure 3-25. Depiction of moderation of DO concentration by shoal at Withlacoochee River site. Table 3-6. Statistical analysis of DO data for the USGS gauge at Withlacoochee River near Croom indicating no significant trend. Parameter Residual Residual Median N p ValueInterceptSlope Dissolved Oxygen -0.21696 121 0.837 0.0047 -0.00001 Table 3-7. Statistical analysis for DO concentr ation data collected at the USGS gauge at Withlacoochee River near Croom, FL. Statistic Value mean (mg/L) 5.65 min (mg/L) 1.6 max (mg/L) 10.2 90% exceedance (mg/L) 2.6 % of readings below 5 mg/L 59% % of readings below 4 mg/L 25% % of readings below 3 mg/L 11% % of readings below 2 mg/L 4%

PAGE 61

61 0 2 4 6 8 10 12 0 50 100150200250300350 Sta g e ( cm ) DO (mg/l) Figure 3-26. Correlation between stage and DO concentrations for Hillsborough River near Croom (USGS) historical data. Table 3-8. Statistical analysis for DO concentrat ion data collected at the Withlacoochee near Silver Lake Site (mid-day only). Statistic Value min (mg/L) 2.52 max (mg/L) 14.39 mean (mg/L) 8.72

PAGE 62

62 CHAPTER 4 CONCLUSIONS The purpose of this study was to test the thr ee hypotheses discusses in Chapter 1. The first two hypotheses required the analysis of the m aximum shoal depths needed to maintain DO levels within the river-specific historical range and above the FDEP 5 mg/l standard. The most restrictive site in relation to the FDEP standard was on the Anclote River which indicated that maximum shoal depth needs to remain above 38.1 cm (table 4-1). For the historical record criteria, the Anclote River site was again the mo st restrictive and suggested that maximum shoal depth needs to remain greater than 19.8 cm. Correlations between DO and maximum shoal depth were good for most of the studied sites. In the rivers studied, DO concentrations have been obs erved well above the 5 mg/l criteria during low flow conditions (maximum shoal dept h less than 18.3 cm threshold) and below 5 mg/l during higher flows indicating that other factors may be infl uencing DO concentrations as well. One possible factor influencing DO leve ls is rainfall that may introduce high BOD materials or pollutants which can lead to depr essed DO concentrations. Findings of this study disprove both my hypothesis that DO concentra tions would remain within the river-specific historical range and my hypothe sis that DO concentrations would remain above the FDEP, 5 mg/l standard when flows necessary to maintain a maximum shoal depth of 18.3 cm are met or exceeded. Although both of these criteria were me t for a portion of the studied site and may serve as a guidance level, fact ors beyond the scope of this study indicate that DO should be monitored in each system when establishing minimum depth criteria. The third hypothesis required an alysis of the physical aspect of water passing over shoals by attempting to correlate Manning's N and Froude va lues to increase in DO from upstream to downstream of the shoals. In all cases, no correl ation was indicated by th e data, disproving this

PAGE 63

63 hypothesis and indicating that other factors may be controlling DO change across the shoals. It is speculated that, due to the narrow range of flow conditions evaluated (low flow only), there was not enough variability in Manning's N to suff iciently assess this relationship. It is also speculated that, due to the very low Froude numbe rs typically observed in these systems, air entrainment was negligible in all rivers studied. If any of the sites had approached or exceeded critical flow, air entrainment may have been better correlated with DO increases across the studied shoals. This study shows that DO concentrations ar e moderated by shoals, aerating water that enters the shoal with low DO levels and de-gassi ng water that enters th e pool super-saturated with DO. This process was most notable on the Withlacoochee River as the river became highly influenced by submerged aquatic vegetation and algal blooms. Table 4-1. Summary of findi ngs for all studied rivers River Shoal Water Depth Necessary to Maintain 5.0 mg/L FDEP Standard (cm) Corresponding Flow (m3/sec) Percentage of Time Needed Flow Conditions Met (Based on POR) Anclote 38.1 .12 73% Alafia < 18.3 .62 99% Hillsborough Met during en tire study n/a n/a Withlacoochee 16.8 2.6 85% River Shoal Water Depth Necessary to Maintain DO Concentrations within Historical Range (cm) Corresponding Flow (m3/sec) Percentage of Time Needed Flow Conditions Met (Based on POR) Anclote 19.8 .08 95% Alafia < 18.3 .62 99% Hillsborough Met during en tire study n/a n/a Withlacoochee 7.0 .02 99%

PAGE 64

64 APPENDIX A A COMPARISON BETWEEN HISTORICAL AND STUDY DISSOLVED OXYGEN Figure A-1. Comparison between hi storical and study dissolved oxygen.

PAGE 65

65 Table A-1. Numerical comp arison between historical and study dissolved oxygen. River Anclote Alafia Withlacoochee Historical Min 1.2 3.5 1.6 Historical Max 10.4 14.0 10.2 Historical Mean 5.0 7.2 5.7 Study Min 1.1 6.5 0.1 Study Max 9.2 10.5 17.7 Study Mean 5.3 8.0 7.4

PAGE 66

66 LIST OF REFERENCES Anerson, T.H. and Taylor, G.T. 2001. Nutrient Pulses, Plankton Blooms, and Seasonal Hypoxia in W estern Long Island Sound. Estuaries 24(2):228-243. Bosch, D., R. Lowrance, and G. Vellidis. 2002. Dissolved Oxygen Concentrations in Three Coastal Plain Watersheds: Implication for TM DLs. American Society of Agricultural and Biological Engineers, St. Joseph, Michigan. http://www.asabe.org, Accessed February 2007. Brooks, J. 2006. ERC Adoption of Dissolved Oxygen SSAC for the Lower St. Johns River. Florida Departm ent of Environmenta l Protection, Tallahassee, Florida. http://floridadep.net/northeast/stjohns/ TMDL/docs/May06/E RC_LSJR%20DO%20SSAC_ ERCMEETING.pdf, Accessed May 2007. Chow, V.T. 1959. Open-channel Hydraulics McGraw Hill. New York, New York. Gualtieri, C., Gualtieri, P. and Doria, G.P. 2002. Dimensionless Analysis of Reaeration Rate in Streams. Journal of Environm ental Engineering 28(1):12-21. Hood, J. 2007. Standardized Methods Utilized by th e Ecologic Evaluation Section for Collection and Management of Physical Habitat Simu lation Model and Instream Habitat Data. Southwest Florida Water Management District, Brooksvi lle, Florida. Jain, SK and Jha, R. 2004. The Froude Number Stick; an Evaluation. River Research and Application 20(1):99-102. Lakewatch. 2003. A Beginner's Guide to Water Mana gement Fish Kills (Information Circular 107). Lakewatch/University of Fl orida, Gainesville, Florida. http://lakewatch.ifas.ufl.edu/circpd ffolder/fish_kill_LR.pdf, Accessed May 2007. Moog, D.B. and Jirka, G. H. 1999. Stream R eaeration in Nonuniform Flow: Macroroughness Enhancement. Journal of Hydr aulic Engineering 125(1):11-16. Parkhurst, J.D. and Pome roy, R.D. 1972. Oxygen Absoption in St reams. Journal of the Sanitary Engineering Division, American Societ y of Civil Engineers 98(1):101-124. Ramaker, J., Ojha, C.S.P. and Bhatia, K.K.S. 2001. Refinement of Predictive Reaeration Equations for a Typical Indian River. Hydrological Processes 15(6):1047-1060. Ramaker, J., Ojha, C.S.P. and Bhatia, K.K.S. 2004. A Supplemental Approach for Estimating Reaeration Rate Coefficients. H ydrological Processes 18(1):65-79. Rhinesmith, P. and R. Smith. 2001. Wysong-Coogler Water Conservation Structure Environmental Monitoring Report FDEP Permit Authorization Number 09-177432-001. Southwest Florida Water Management District, Brooksvi lle, Florida.

PAGE 67

67 Richard, A. and A. Moss. 2002. Plant Management in Florida Waters; All you want to know about Florida's lakes, rivers, springs, marshes, swamps, and canals. University of Florida, Gainesville, Florida and Florida Department of Environmental Protection, Tallahassee, Florida. http://aquat1.ifas.ufl.edu/guide/oxygen.html, Accesses May 2007. Tadessa, I., Green, FB, Puhakka, JA. 2004. Seasonal a nd Diurnal Variations of Te mperature, pH, and Dissolved Oxygen in Advanced Integrat ed Wastewater Pond System (R) Treating Tannery Effluent. Water Research 38(3):645-654. United States Environmental Protection Agenc y. 2004. Total Maximum Daily Load (TMDL) for Dissolved Oxygen (DO) in Brooker Creek (W IBID #1474). United States Environmental Protection Agency, Region 4, Atlanta, Georgia. United States Geological Survey(USGS)(1). 2007. National Water Information System: Web Interface. United States Geologi cal Survey, Tampa, Florida. http://waterdata.usgs.gov/fl/nwis/nw ism an/?site_no=02310000&agency_cd=USGS, Accessed May 2007. United States Geological Survey(USGS)(2). 2007. National Water Information System: Web Interface. United States Geologi cal Survey, Tampa, Florida. http://waterdata.usgs.gov/fl/nwis/nw isman/?site_no=02301500&agency_cd=USGS, Accessed May 2007. United States Geological Survey(USGS)(3). 2007. National Water Information System: Web Interface. United States Geologi cal Survey, Tampa, Florida. http://waterdata.usgs.gov/fl/nwis/ uv/?site_no=02303000&PARAmeter_cd=00065,00060, Accessed May 2007. United States Geological Survey(USGS)(4). 2007. National Water Information System: Web Interface. United States Geologi cal Survey, Tampa, Florida. http://waterdata.usgs.gov/fl/nwis/ uv/?site_no=02312500&PARAm eter_cd=00065,00060, Accessed May 2007. Weaver, K. 2003. An Alternative Water Quality Criterion for Everglades Dissolved Oxygen. Florida Department of Environmenta l Protection, Tallahassee, Florida. http://www.floridadep.org/water/wqssp /everglades/docs/DDO_SSAC_Public_W orkshop_ Aug19.pdf, Accessed May 2007.

PAGE 68

68 BIOGRAPHICAL SKETCH Jason was born in Brooksville, Florida in 1972. He began working for the Southwest Florida W ater Management District in 1989 a nd graduated from high school in 1990. Attending the University of South Florida and Saint Leo University off and on, while pursuing his career, he graduated with a Bachelor of Science in envi ronmental science from Saint Leo University in 2005. In 2006, he began his master's degree at the Un iversity of Florida. He is currently an Environmental Scientist with the Southwest Flor ida Water Management District where he plans to continue his career.