Title Page
 Table of Contents
 Executive Summary
 Annotated Literature Review
 Stormwater, Pollutant Loads and...
 Spatial Modeling of Nitrogen Loading...
 Spatial Model of Total Phosphorus...
 Appendix A: Energy Language...
 Appendix B: TP Model Algorithm
 Appendix C: Emergy Calculation...
 Appendix D: Secondary Coverage...
 Appendix E: SubBasin Character...
 References Cited
 Department of Environmental Protection...
 Proposal: Development of a Spatial...

Spatial Modeling of Landscape Development Intensity and Water Quality in the St. Marks River Watershed
Full Citation
Permanent Link: http://ufdc.ufl.edu/AA00004017/00001
 Material Information
Title: Spatial Modeling of Landscape Development Intensity and Water Quality in the St. Marks River Watershed
Physical Description: Report
Language: English
Creator: Brown, Mark T.
Parker, Neal
Foley, Alan
Publisher: Center for Wetlands
Publication Date: 1998
Subjects / Keywords: GIS
landscape development intensity (LDI)
water quality
spatial modeling
Spatial Coverage: United States -- Florida -- Wakulla -- St. Marks -- St. Marks River
Coordinates: 30.11 x -84.1
General Note: 246 Pages
 Record Information
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: AA00004017:00001


This item has the following downloads:

( PDF )

Table of Contents
    Title Page
        Page i
    Table of Contents
        Page ii
    Executive Summary
        Page ES-1
        Page ES-2
        Page ES-3
        Page ES-4
        Page ES-5
        Page ES-6
        Page ES-7
        Page ES-8
        Page ES-9
        Page ES-10
        Page ES-11
        Page ES-12
        Page ES-13
        Page ES-14
        Page ES-15
        Page ES-16
    Annotated Literature Review
        Page 1-1
        Page 1-2
        Page 1-3
        Page 1-4
        Page 1-5
        Page 1-6
        Page 1-7
        Page 1-8
        Page 1-9
        Page 1-10
        Page 1-11
        Page 1-12
        Page 1-13
        Page 1-14
        Page 1-15
        Page 1-16
        Page 1-17
        Page 1-18
        Page 1-19
        Page 1-20
        Page 1-21
        Page 1-22
        Page 1-23
        Page 1-24
        Page 1-25
        Page 1-26
        Page 1-27
        Page 1-28
        Page 1-29
        Page 1-30
        Page 1-31
        Page 1-32
    Stormwater, Pollutant Loads and Management in the Lafayette and Munson Basins
        Page 2-1
        Page 2-2
        Page 2-3
        Page 2-4
        Page 2-5
        Page 2-6
        Page 2-7
        Page 2-8
        Page 2-9
        Page 2-10
        Page 2-11
        Page 2-12
        Page 2-13
        Page 2-14
        Page 2-15
        Page 2-16
        Page 2-17
        Page 2-18
        Page 2-19
        Page 2-20
        Page 2-21
        Page 2-22
        Page 2-23
        Page 2-24
        Page 2-25
        Page 2-26
        Page 2-27
        Page 2-28
        Page 2-29
        Page 2-30
        Page 2-31
        Page 2-32
    Spatial Modeling of Nitrogen Loading to a Surficial Aquifer from Residential On-site Sewage Disposal Systems in Wakulla County, Florida
        Page 3-1
        Page 3-2
        Page 3-3
        Page 3-4
        Page 3-5
        Page 3-6
        Page 3-7
        Page 3-8
        Page 3-9
        Page 3-10
        Page 3-11
        Page 3-12
        Page 3-13
        Page 3-14
        Page 3-15
        Page 3-16
        Page 3-17
        Page 3-18
        Page 3-19
        Page 3-20
        Page 3-21
    Spatial Model of Total Phosphorus Loading and Landscape Development Intensity in the St. Marks River Watershed
        Page 4-0
        Page 4-1
        Page 4-2
        Page 4-3
        Page 4-4
        Page 4-5
        Page 4-6
        Page 4-7
        Page 4-8
        Page 4-9
        Page 4-10
        Page 4-11
        Page 4-12
        Page 4-13
        Page 4-14
        Page 4-15
        Page 4-16
        Page 4-17
        Page 4-18
        Page 4-19
        Page 4-20
        Page 4-21
        Page 4-22
        Page 4-23
        Page 4-24
        Page 4-25
        Page 4-26
        Page 4-26a
        Page 4-27
        Page 4-28
        Page 4-29
        Page 4-30
        Page 4-31
        Page 4-32
        Page 4-33
        Page 4-34
        Page 4-35
        Page 4-36
        Page 4-37
        Page 4-38
        Page 4-39
        Page 4-40
        Page 4-41
        Page 4-42
        Page 4-43
        Page 4-44
        Page 4-45
        Page 4-46
        Page 4-47
        Page 4-48
        Page 4-49
        Page 4-50
        Page 4-51
        Page 4-51a
        Page 4-52
        Page 4-53
        Page 4-54
        Page 4-55
        Page 4-56
        Page 4-57
        Page 4-58
        Page 4-59
        Page 4-60
        Page 4-61
        Page 4-62
        Page 4-63
        Page 4-64
        Page 4-65
        Page 4-66
        Page 4-67
        Page 4-68
        Page 4-69
        Page 4-70
        Page 4-71
        Page 4-72
        Page 4-73
        Page 4-74
        Page 4-75
        Page 4-76
        Page 4-77
        Page 4-78
        Page 4-79
        Page 4-80
        Page 4-81
        Page 4-82
        Page 4-82a
        Page 4-83
        Page 4-84
        Page 4-85
        Page 4-86
        Page 4-87
    Appendix A: Energy Language Symbols
        Page 4-88
    Appendix B: TP Model Algorithm
        Page 4-89
    Appendix C: Emergy Calculations
        Page 4-90
        Page 4-91
        Page 4-92
        Page 4-93
        Page 4-94
        Page 4-95
    Appendix D: Secondary Coverages
        Page 4-96
        Page 4-97
        Page 4-98
        Page 4-99
        Page 4-100
        Page 4-101
        Page 4-102
        Page 4-103
        Page 4-104
    Appendix E: SubBasin Characteristics
        Page 4-105
        Page 4-106
        Page 4-107
        Page 4-108
        Page 4-109
        Page 4-110
        Page 4-111
        Page 4-112
        Page 4-113
        Page 4-114
        Page 4-115
        Page 4-116
        Page 4-117
    References Cited
        Page 4-118
        Page 4-119
        Page 4-120
        Page 4-121
        Page 4-122
    Department of Environmental Protection Presentations
        Page 5-0
        Page 5-1
        Page 5-2
        Page 5-3
        Page 5-4
        Page 5-5
        Page 5-6
        Page 5-7
        Page 5-8
        Page 5-9
        Page 5-10
        Page 5-11
    Proposal: Development of a Spatial Model of Pollutant Loading and Water Quality for Florida Watersheds
        Page 6-1
        Page 6-2
        Page 6-3
        Page 6-4
        Page 6-5
Full Text

Final Project Report

Spatial Modeling of Landscape Development Intensity
And Water Quality in the St Marks River Watershed
DEP Contract #GW138

to the
Department of Environmental Protection
Bureau of Water Resources Protection


Mark T. Brown, Neal Parker, and Alan Foley

Center for Wetlands
Department of Environmental Engineering Sciences
University of Florida
Gainesville, FL 32611
(352) 392-2424

14 September, 1998


Table of Contents

Executive Summary
Project Narrative
Report Summary

Chapter 1: Literature Review and Annotated Bibliography

Chapter 2: Stormwater, Pollutant Loads and Management in the Lafayette and Munson
Basins. Mark Brown and Neal Parker

Chapter 3. Spatial Modeling of Nitrogen Loading to a Surficial Aquifer from Residential
On-site Sewage Disposal Systems in Wakulla County, Florida. Alan Foley

Chapter 4. Spatial Models of total Phosphorus Loading and Landscape Development
Intensity in the St Marks River Watershed. Neal Parker

Chapter 5. Department of Environmental Protection Presentations

Chapter 6. PROPOSAL: Development of a Spatial Model of Pollutant Loading and
Water Quality for Florida Watersheds


Spatial Modeling of Landscape Development Intensity
And Water Quality in the St Marks River Watershed
DEP Contract #GW138


Organization of the Report

The report is organized in chapters. Each chapter is self contained, having individually
numbered pages, figures, and tables. In addition, each chapter contains a bibliography
and appendices as required. The first chapter is an annotated bibliography of literature
relevant to the St Marks Watershed and spatial and pollutant modeling. The second
chapter reports on pollutant loading of the Lake Lafayette and Lake Munson sub-basins.
In the third chapter a spatial model of nitrogen in surficial ground waters of Wakulla
County is reported. The fourth chapter gives the results of the work conducted to relate
Landscape Development Intensity (LDI) to spatial loading of Total Phosphorus within
the St Marks. In the fifth chapter copies of slides used in a presentation given at the DEP
Twin Tower offices in May, 1998 are provided. Finally, the sixth chapter is a proposal
for continuation of work on spatial modeling and LDI's incorporating a statewide data
base, expanding model parameters, and using statistical tests to validate model

Project Narrative

Research began in the late spring of 1997. Graduate students Neal Parker and
Alan Foley spent several days within the St Marks Watershed traveling from top to
bottom and throughout the basin learning first hand about its development patterns,
drainage networks, and ecological systems. Additionally they spent several days
retrieving literature from state and local agencies. On a separate occasion, Mark Brown
traversed the basin for a two day period in June, 1997. In April and May, 1997 Mark
Brown attended several meetings at DEP in Tallahassee discussing the scope and timing
of the project and participating in kickoff presentations for the St. Marks Watershed
The goal of the initial phase of our project was to support the publication of the
document summarizing the watershed planning process for the St. Marks Basin. As a
result, the first 4 months of the project were aimed at providing this support. First an
annotated bibliography of relevant literature was complied. Second, simulations and
analysis of water quality in the Lafayette and Munson sub-basins were completed, and
third nitrogen loading resulting from septic tanks in areas near Wakulla Springs were
evaluated. The reports that resulted from these investigations are included in this final
report as Chapters 1, 2 and 3.
The goal of the overall project was to further develop Landscape Development
Intensity (LDI) indices and demonstrate their link to stormwater quality using the St.
Marks Watershed. Since water quality data within the St. Marks were few and far

Draft... 9/14/98


between, developing links between LDI's and stormwater quality could not be
accomplished using existing data. As a result it was necessary for Neal Parker to develop
a GIS based spatial model for predicting stormwater pollutant loading in all sub-basins of
the watershed. This model was an outgrowth of our initial work on the Lafayette and
Munson sub-basins and is reported in Chapter 4 of this report.
Work progressed on the spatial model and the Wakulla County nitrogen loading
model through the winter and spring of 1998 and we made a presentation In Tallahassee
of our results to date in May, 1998. A copy of the PowerPoint presentation is included as
Chapter 5. Since the May presentation, we have finalized the spatial simulation model
and LDI analysis and the Wakulla County nitrogen loading model, taking several extra
months since the end of the contract to finalize these reports.
As our worked progressed through the summer of 1998, it was apparent that a
larger effort was needed to validate the pollutant loading model and LDI relationships
that were explored with this initial investigation. Validation of the model was hampered
by a lack of data within the St Marks basin and as a result water quality correlations to
LDI's were somewhat tenuous. The final chapter (Chapter 6) in this report is a proposal
for continuation of work on spatial modeling and LDI's incorporating a statewide data
base, expanding model parameters, and using statistical tests to validate model

Project Summary

In this section the goals and results of each of the studies conducted during this
project are summarized and major points are extracted and summarized from each of the
chapter reports.

Chapter 1: Literature Review and Annotated Bibliography
Neal Parker and Alan Foley
The literature on land use based pollutant loading, pollutant loading models, and
especially data sources for the St marks basin were reviewed and an annotated
bibliography compiled. The bibliography contains over 130 entries.

Chapter 2: Stormwater, Pollutant Loads and Management in the Lafayette and
Munson Basins. Mark Brown and Neal Parker

The effects of spatial distributions of land uses on pollutant loading received by
surface water features were modeled using spatial models that incorporated land use and
topography. Lake Lafayette and Lake Munson were divided into sub-basins to evaluate
the various sub-basin contributions to each water body. In addition, pollutant loads were
modeled for major water bodies and closed basins within each of the larger watersheds to
provide perspective on areas of concern.
As a means of understanding loss of "basin function", pollutant loads were
modeled for three time periods for each of the basins: past, present, and future, and then
compared. Stormwater management options including Best Management Practices



(BMP's), and restoration of historic wetlands were tested with the models for the present
and future conditions to evaluate their effectiveness in reducing pollutant loads.

Pollutant Loading and Percent Impervious Surface

The amount of impervious surface within watersheds is related to the intensity of
human activity and as a result, is a good predictor of stormwater water quality. The
graph in Figure I shows pollutant load in the sub-basins of Lake Lafayette and Lake
Munson watersheds compared to impervious surface. As the graph shows, "impervious-
ness" is strongly correlated to pollutant load. Numerous other studies across the United
States have suggested that impervious surface may be a very good predictor of
stormwater quality and the health of downstream waterbodies.
Sub-basins of the Lafayette and Munson watersheds were ranked based on their
imperviousness for the present land use conditions, and for future land uses based on
maps provided by Leon County Planning (Figures 2 and 3). Based on previous studies by
others which suggested that imperviousness was related to ecosystem health we
concluded that at the present time 7 sub-basins out of the total number of 16 sub-basins
within the Lafayette and Munson watersheds have sufficient impervious surface (greater
than 30%) to warrant serious concern for the ecological health of surface water bodies.
Further, using land use projections we suggested that 11 of the 16 sub-basins. Will have
imperviousness greater than 30% raising concern for ecological health of water bodies
within these basins.

Pollutant Loading
Using a GIS based spatial model, annual pollutant loads for each of the sub-basins
in the Lafayette and Munson watersheds were modeled. The graphs in Figures 4 and 5
show comparisons of annual pollutant loading by sub-basin for the past, present and
future conditions for the Lafayette and Munson Basins. The shortest bars in the graph are
for the natural landscape, averaging about 0.4 lbs/acre*yr-1 (0.45 kg/ha*yr-1). Urbanized
areas have about 3 times these background loads (1.4 lbs/acre*yr-1 [1.48 kg/ha*yr-1).
The biggest changes from present conditions to future conditions are found in the
Lafayette Basin where sub-basins exhibit annual pollutant load increases of between 50
to 80%. The increases in the Munson basin between the present and future condition are
much smaller, with only one basin exhibiting a 75% increase in annual load. The
remaining basins all appear to exhibit increase of between 5 and 15%.

Pollutant Transfer
A second GIS based simulation model was developed that used an overland flow
algorithm to converge and concentrate runoff. Annual pollutant load for the past, present,
and future conditions in the Lake Lafayette and Lake Munson basins were generated
using the model. Pollutant loads are summed along flow paths so that total load at any
point in the drainage basin could be read from the resulting maps. Among the most
significant simulation results in the Lake Lafayette basin the model simulated:
a present day increase of 350% over historic pollutant loads in the lower reach
of Alford Arm Branch.



Relationship Between Percent Imperviousness and Average Phosphorus Loading




I 11

0 10 20 30 40 50 61
Percent Imperviousness

Figure E-1. Impervious surface vs. Phosphorus Load in lake Lafayette and Lake
Munson sub-basins

Draft...9/14/98 ES-4



S C3 IT 3I2 ] C3 19 j ** l
C 3 i 2Z I % Imperviousness s Iperviousness
Ji_ i El 11 l -13 44 c ss


1 6 K

"- .... ) ',

s 9 6 -

f 8 7, '-1 -

a 2,500 ,]00f MI.t. i Jt f

Gilberts Pond Outlet
Roberts Pond Outlet
Alford Arm
Buck Lake Outlet

5. Unnamed Run
6. Lake Lafayette Drain
7. Unnamed Slough
8. Unnamed Run
9. Mall Drainage Area

Figure E-2. Maps of Impervious surface in Lake Lafayette sub-basins. Top left is
present condition, top right is based on future land use.




C j 17 E3 is
E D 0 I m .*'
l ia % Imperviousness
Cj i10

Li M0 *1 % Imperviousness
ii >

J< 1 1

S2 _

7 5


e ',4*00 5.ll00 MBIIC

1. Unnamed Run
2. Godby Ditch
3. Central Drainage Ditch
4. St. Augustine Branch

S. East Drainage Ditch
6. Munson Slough
7. Bradford Brook

Figure E-3. Maps of Impervious surface in Lake Munson sub-basins. Top left is
present condition, top right is based on future land use.



Total Phosphorus Average Loads for the Lake Lafayette Basin

1.8 E Past D Present l Future

1.6 ---






1 2 3 4 5 6 7 8 9
Basin Number

Figure E-4. Simulated phosphorus loads for the past, present, and future in Lake
Lafayette sub-basins


Total Phosphorus Average Loads for the Lake Munson Basin

0.6- -- -



1 2 3 4 5 6 7
Basin Number

Figure E-5. Simulated phosphorus loads for the past, present, and future in Lake
Munson sub-basins

Draft... 9/14/98

a 450% increase between the historic annual load and the present load in the
Tallahassee Drainage inflow to Upper Lake Lafayette,
an 80% increase in annual load to Lake McBride in the future. and
a 70% increase for Killearn Lakes .

Significant simulation results in the Lake Munson basin included:

little change in the pollutant load at the inflow to Bradford Lake between the
past, present and future.
an increase in pollutant load of 367% from past to present day at the inflow to
Munson Slough with increases in the future of 30% over present day loads,
an increase in the Northern Drainage Area of 233% from past to present
increasing another 30% over current loads in the future.

Stormwater Management Alternatives

The GIS based simulation model was used to compared several stormwater management
alternatives within the Lake Lafayette and Lake Munson basins. The graphs in Figures 6
and 7 show changes in pollutant loading at several locations in the basins and the effects
of three management alternatives. In the first alternative, BMP's (street sweeping and
swales) are used to reduce pollutant loads by 10% in residential areas and 20% in
commercial areas. In the second alternative, wetlands are reconstructed throughout the
watersheds to replace wetlands lost over the years to development And the third
alternative is to combine BMP's with the wetlands alternative. BMP's, an effective
means of reducing some loads, appeared to reduce loads by about 10 15% in urbanized
basins. The largest increase in water quality resulted from the wetlands reconstruction
alternative, lowering pollutant loads in some cases by more than 50%. The combined
approach, provided additional improvement in reducing total load.

Chapter 3. Spatial Modeling of Nitrogen Loading to a Surficial Aquifer from
Residential On-site Sewage Disposal Systems in Wakulla County, Florida.
Alan Foley

This study investigated spatial modeling of nitrogen loading to the surficial aquifer
resulting from septic tanks Wakulla County. The study focused on five hydrologic sub-
basins surrounding Wakulla Springs State Park and the Wakulla River. The area is
rapidly urbanizing and has a high potential for groundwater degradation.
A raster-based geographic information system was used to manipulate digital map layers
to solve a one-dimensional analytical equation over two-dimensional space using a cell
resolution of 30m X 30m. A coarse digital elevation model (DEM) was created by
interpolation between 1:100,000 ft scale contours and the assumption was made that the
phreatic surface within the surficial aquifer was generally a reflection of the land surface



Total Phosphonus Loads for Several
Management Altemativs at Silr A in the Lake Lafayeltt Basin

Total Phosphorus Loading for Several
Management Altrnamives at Site Bin Lake Lafayeloe Basin

I o

....l irl~u 3


uli __sa

IJ^-_____ ______________
*I i ----------------

i ^aa


Total Phosphorus Loading for Several
Management Alternatives at Sit. C In the Lak Lafayette Basin


Total Phosphorus Loading for Several
Management Alltmatves at Site D in th Lake Lafayette Baln


Figure E-6. Change in simulated phosphorus loads that result from BMP's and

increases in wetland areas absorbing runoff in Lake Lafayette sub-basins

Draft.. 9/14/98

Pn .


Totai Ptophto Loads lor Sov
Mangmannt Alhltoan r a Site A In lt Lak* Munmsn Basin

Toa Phosphois Ladr for $"OS
Msanagn AttMalatvu. at Sit. B in th LUak Mouson Basin

Total Phosphorus Loads for Snal
MUlwnamant AIItmalite at Sit C In lhe L.l Muson Basin

Figure E-7. Change in simulated phosphorus loads that result from BMP's and increases

in wetland areas absorbing runoff in Lake Munson sub-basins

Draft...9/14/98 ES-


A general soils map of the area was used to generate soil hydraulic conductivity and soil
carbon content coverages that were necessary for the simulation.
For the most part the simulation results show that the majority of N03O
attenuation in the soil occurs within the first 30 to 90 meters surrounding the land use that
generates the nitrogen load. Concentrations generally drop to 2 mg NO3/1 within 30 to
90 m's of the septic tank source. Figure 8 shows a detail of the nitrogen model output,
where the source of nitrogen is in the upper right hand corer. Evident is the drop in
nitrogen concentrations within 1 to 3 cells from the source boundary.
The results confirmed that the high spatial and temporal variability of governing
parameters particularly hydrology poses a significant challenge to modeling
subsurface chemical dynamics. Isolating and modeling the effect of a single of nitrogen
input source, like septic tanks, within a karst system requires either broad generalizations
or a considerable amount of data. Model calibration was difficult due to the lack of and
or questionable validity of ground water quality data.

Chapter 4. Spatial Models of total Phosphorus Loading and Landscape
Development Intensity in the St Marks River Watershed. Neal Parker

A spatial pollutant load model was developed for total phosphorus (TP) and then
model results were correlated with measures of landscape development intensity (LDI) in
the St. Marks Watershed. A main feature of the model was to account for the mitigating
effects of distance upon TP in stormwaters delivered to water bodies. in the spatial
model, and (2) use model results to examine the relationship between development
intensity and TP for sub-basins within the St. Marks Watershed.
The model used a geographic information system (GIS) with an overland flow
algorithm to predict TP for every 100 meter by 100 meter cell within the watershed. The
function of TP absorbed by the landscape with distance controlled the amount of TP that
entered each stream. Each cell in the watershed was assigned a literature determined TP
loading value dependent upon land use. Calibration against existing data from the
watershed was used to specify TP uptake coefficients.
A best fit of modeled with observed TP concentrations (r2 = 0.48) was found by
using two distance decay factors, one for locations in the northern half of the watershed
(-0.2) and another for the southern half (-0.3). These regions had distinct differences in
surface geology and soil types. All combinations of distance decay factors were modeled
until a best fit between predicted and observed TP concentrations at 16 locations within
the basin were found. Both linear and exponential decay functions were modeled. Final
results suggested that the exponential function produced best results, and that lands
within 200 300 meters of surface water bodies were most influential in determining TP
loads to the water bodies. Modeled TP concentrations ranged from background levels of
0.01 mg/I to 0.49 mg/l compared with a range of 0.01 mg/1 to 0.87 mg/1 for observed
data. Model validation was hampered by the lack of available date in the St Marks
Five LDI measures were developed from spatial coverages including two physical
and three emergy indices. Emergy (spelled with an m) is the energy of one kind required
directly and indirectly to make a product. The physical indices were percent impervious




- Is

Figure E-8. Detail of the simulated nitrogen plume resulting from a septic tank
source in the upper right hand corer. The map at the top shows the Wakulla
County study area.

Draft... 9/14/98


%iw ".

surface and a weighted land use intensity index. The emergy indices were total and
developed emergy flow per area (empower density) and a ratio of nonrenewable to
renewable inputs to the landscape (environmental loading ratio). These five LDIs were
correlated with modeled TP loads for sub-basins of the St. Marks Watershed (Figure 9).
The imperviousness LDI exhibited the strongest correlation with TP (r2 = 0.74
above 10% imperviousness. Weighted land use LDI had the second best fit (r2= 0.67)
above an LDI of about 3.0. Graphs of LDI's vs. TP showed that developed land area
(urban and/or agriculture) had to exceed about 25% to 30% of basin land area before TP
in surface water bodies was detectable above natural background fluctuations .
In summary several important conclusions were made regarding pollutant load
modeling and the relationship between LDIs and TP within the St. Marks Watershed.
1. Pollutant load models can be successfully developed that aggregate many
parameters including soils, topography, and imperviousness into a decay
coefficient, vastly simplifying the model algorithm and data requirements.
2. The pollutant load model suggested that the most important contributions
of pollutant loads come from developed lands within between 200 and 300
meters of surface water bodies.
3. Management efforts may be best focused at locations where the spatial TP
model predicted significantly higher TP concentrations than observed
concentrations. These may be locations where development induced
impacts are just beginning to occur.
4. It appears that there is a development threshold above which TP loads in
surface waters are higher than natural background variability. Above this
threshold, the LDI's have stronger correlations with TP loads. These
thresholds occurred at 10% imperviousness for the imperviousness LDI,
3.5 for the weighted land use LDI, 50E15 sej/ha/yr for the total and
developed empower LDIs, and 20 for the ELR LDI. For intensities above
the LDI threshold, the imperviousness LDI had the highest correlation
with TP loads (r = 0.60).
5. Overall, the model suggested that TP loads become apparent above
background levels when the area of development exceeds 30% of the total
area contributing stormwater runoff to surface waters..

Chapter 5. DEP Presentations
A half-day presentation of project results was conducted in Tallahassee on May 1,
1998. The presentation included theory and principles of spatial analysis and spatial
modeling and then we presented results of efforts to model pollutant loading in the St
Marks Basin and application of the LDI to resulting model output.

Chapter 6. PROPOSAL: Development of a Spatial Model of Pollutant Loading and
Water Quality for Florida Watersheds

The next step in developing a GIS based spatial model of pollutant loading is to
statistically validate the model used in the St Marks analysis with water quality data from





I: 3

3. da

Figure E-9. Landscape Development Intensity indices graphed against simulated
phosphorus loads in the St. Marks Watershed. a) Imperviousness, b) weighted land use,
c)total empower density, d) developed empower density, e) environmental loading ratio.

Draft...9/14/98 ES-


around the state. We propose to select a large number of basins (stratified based on
ecoregion.) that have good water quality data, and use statistical methods to validate the
model, extending the number of parameters included as well as refining the spatial
modeling algorithm.
It is our belief that efforts to screen Florida's surface waters for present and future
quality problems can be significantly enhanced through the use of a relatively simple, yet
sophisticated spatial model in a GIS environment. Rather than detailed computer
simulation packages that require hundreds of coefficient s and months of validation
before they can be applied to a single basin, the raster based spatial simulation requires
few coefficients, and minor amounts of time. The complex simulation packages are good
for detailed analysis and where legal challenges dictate scientific scrutiny. The proposed
spatial model is needed to screen Florida's waters, and provide needed macro-scale
information for the effective allocation of resources for management. The model will be
beneficial in the following applications:

1. Estimate degree of impairment of water bodies
2. Spatially locate areas of impairment and identify sources of non point source
pollutant generation
3. Direct field monitoring where: (a) the model predicts impairment and (b) where
uncertainties are the greatest. In other words, suggest monitoring plan based on
model output.
4. Predict future impairment based on changes in land use/land cover
5. Model could be used as the 305B model to estimate % of state water bodies that
are within compliance

Draft... 9/14/98


Chapter 1
Annotated Literature Review

Neal Parker and Alan Foley

St. Marks

Bartel, R., et. al. 1991. Lake Munson Basin Plan City of Tallahassee and Leon County
Stormwater Management Plan Vol. II. NWFWMD.

This report (Volume II) represents a description of the Lake Munson basin stormwater management plan
developed by the Northwest Florida Management District in cooperation with Leon County and the City of
Tallahassee. The flooding and water quality problems in four major basins in the study area, which
included the Lake Munson, Lake Lafayette, Lake Jackson, and Fred George Sink basins. The technical
details, which included the use of hydrologic models as a tool for plan development, have been described in
a technical report (Volume VI) for this study. An overall summary of the problems identified and the
recommended plan for the Lake Munson basin may be found in the Executive Summary (Volume I) of this

Bartel, R., et. al. 1991. Stormwater Management Plan for the City of Tallahassee and
Leon County Vol. IV Technical Report. NWFWMD.

This publication represents Volume VI of a series of six reports to document the Stormwater Management
Plan for the City of Tallahassee and Leon County. It is provided as a guide for understanding the
techniques, data, and principles used for the development of basin management plans in the project area.
As such, it is mainly intended for the technical reader having some familiarity with hydrologic analyses and
for City and county staff who may be involved in making recommendations to actually implement the basin
management plans and updating the plan in the future.

Bartel, Ronald L. and A.E. Maristany, P.E. 1989. Wetlands and Stormwater
Management A Case Study of Lake Munson Part II: Impacts of Sediment and Water
Quality. American Water Resources Association. pp. 231-246.

In early 1950, a 255-acre tract of cypress and gum swamp in southern Leon County, Florida was converted
from a wetland into Lake Munson. The purpose for this shallow impoundment was to alleviated
downstream flooding problems. Unfortunately, the area never reached its full potential as a recreational
water body or aquatic habitat due to severe water quality impacts on the lake. The inadvertent discharge of
wastewater effluents and stormwater have been the major cause for the degradation of this water body, to
the point that in 1982, it was ranked on the trophic state index as the seventh most degraded lake in Florida.
In 1984, the wastewater effluents were diverted to a land application system and lake water quality has
improved. However the lake continues to suffer algal blooms, fish kills, depressed oxygen levels, and high
nutrient and bacterial levels. Restoring ecologically sound conditions to the lake and developing its
potential use as a recreation area will require upstream treatment of stormwater and possibly the removal of
bottom sediment. Ironically, results show that upstream treatment may require a wetlands area at least the
size of Lake Munson.

Bartodziej, W., A.J. Leslie. 1997. Water Hyacinth as a Biological Indicator of Water
Quality. FDEP, TSS #97-100, Tall. FL.

Draft... 9/14/98

Water Hyacinth production was measured against water quality in four north Florida mesotrophic lakes to
see if plant growth related to nutrient enrichment. Being a large, conspicuous plant, water hyacinth may be
an ideal biological indicator of nutrient enrichment if generalizations can be made about its response to
water-column phosphorus and nitrogen concentrations in certain lake types.

Burton, T.M., Turner, R.R., and R.C. Harriss. 1977. Nutrient Export from Three North
Florida Watersheds in Contrasting Land Use. Symposium Papers: Watershed Research
in Eastern North America Volume II. Edgewater, MD. pp. 323-341.

Exports of nutrients from a forested-agricultural, a suburban, and an urban watershed in north Florida were
measured from 1973 to 1975. The most significant impact of urbanization has been the change in temporal
distribution and quantity of total and dissolved P exports. The urban watershed exports 16 times more total
P than does the forested-agricultural watershed with the suburban watershed intermediate. Further, 98% of
this total P is exported in quickflow (stormflow) on the urban watershed while only 53% is exported in
quickflow on the forested-agricultural watershed. Dissolved P exports are relatively low from all 3
watershed with the primary effect of urbanization being the temporal changes in export, with 77% of the
urban exports being in quickflow compared to 35% for the forested-agricultural watershed. Exports of
inorganic N were highest from the suburban watershed, probably as a consequence of septic tank drainage;
exports of N03-N were 6 times higher for the suburban and 3 times higher for the urban, sewered
watershed compared to the forested-agricultural watershed. Si and CI were also monitored. Mechanisms
controlling export of P, N, Cl, and Si are emphasized.

Burton, T.M., Turner, R.R., and R.C. Harriss. 1977. Descriptive Hydrology of Three
North Florida Watersheds in Contrasting Land Use. Symposium Papers: Watershed
Research in Eastern North America Volume II. Edgewater, MD. pp. 211-224.

Results of hydrologic studies on three adjacent watersheds in north Florida representing respectively urban
(792 ha), suburban (430 ha) and forested-agricultural (611 ha) land uses support theory and the findings of
others that urbanization (1) increases storm peak flows, (2) increases the ratio of quickflow volume to
delayed flow volume and (3) increases annual runoff. Total runoff losses from the forested-agricultural,
suburban and urban basins during the period July 1973 to June 1975 were respectively, 31.2, 38.1, and 48.3
cm of which 39%, 51%, and 82% were quickflow. Ratios of total quickflow volume to total precipitation
over the study period were 0.05, 0.08, and 0.16 for the forested-agricultural, suburban and urban
watersheds respectively, comparison of individual storm hydrographs also revealed striking contrasts in
the relative magnitude and temporal distribution of streamflow from these watersheds. Maximum
discharges during the study period were 0.19 m^3/sec/km^2 (forested-agricultural), 0.75 m^3/sec/km^2
(suburban) and 2.37 m^3/sec/km^2 (urban). When combined with results of contemporaneous water
quality studies on the same watersheds these hydrologic findings offer an ideal example of some
hydrochemical consequences of urbanization on a small watershed scale.

Clemens, Linda A. 1988. Ambient ground water quality in northwest Florida Part II: A
Case Study in Regional Ground Water Monitoring Wakulla Springs, Wakulla County,
Florida. Havana, FL: NWFWMD. Water Resources Special Report 88-1.

Water samples were taken from within the caverns of Wakulla Springs. Sample water quality did not seem
to vary considerably for different locations within the spring caverns.

Florida Department of Environmental Protection. 1996 Water-Quality Assessment for the
State of Florida: Section 305(b) Main Report. Bureau of Water Resources Protection,
Division of Water Facilities, FDEP, Tallahassee, FL.

Draft... 9/14/98

This report covers water-quality information for the state with a section on basic facts, ecology, and human
impacts within the St. Marks River Basin. The two main problem areas noted within the report include
Munson Slough and Rattlesnake Branch. Munson Slough drains portions of the Tallahassee urban area and
historically received wastewater from small package plants. A partial solution was to divert Tallahassee's
treatment plant effluent to land-spreading operation. The second problem area is the St. Marks River
downstream from Rattlesnake Branch (lower three to four miles) which received effluent from Seminole
Refining corporation and Purdom Power Plant. Near the confluence with the St. Marks, the Wakulla
receives nutrients from Boggy Branch. Lake Lafayette may also be an additional problem area which is
still under investigation.

Florida. Bureau of Geology. 1972. Environmental geology and hydrology, Tallahassee
area, Florida. Tallahassee, FL.

There is a distinct escarpment that separates the coastal lowlands from the Tallahassee Hills which ranges
in elevation from sea level to approximately 260 feet. The lowland soils are sandy (immediate infiltration
of rain) hence surface runoff is minimal. In the Tallahassee Hills, soils are clayey and several tens of feet
thick, over permeable limestone promoting local surface drainage from the hills. Lake Lafayette is one of
the headwater tributaries of the St. Marks River. The surface layer is ancient Miocene-Pliocene delta plain.
Dissolution of the limestone around Tallahassee has created the three largest lake basins: Lake Jackson,
Lake lamonia, and Lake Lafayette. The St. Marks River has a poorly defined channel north of natural
bridge. Additional water from the Floridan Aquifer is added to the river at natural bridge. Cores taken in
the area reveal the following layers from shallowest to deepest: Miccosukee, Hawthorn, St. Marks,
Suwannee Limestone, and Crystal River Formation. The deeper layers are limestones and dolomites.

Hendry, C.W. Jr., and C.R. Sproul. 1966. Geology and Ground-Water Resources of Leon
County, Florida. Tallahassee, Published for The Florida Geological Survey.

The physiography of the gulf coast is generally divided into the highlands and lowlands which are further
divided into the delta plain, tertiary highlands, terraced coastal and river valley lowlands. Leon County
specifically contains the northern highlands, gulf coastal lowlands, and the river valley lowlands. The
Tallahassee Hills are tertiary in age and deltaic in origin. The Miocene-Pliocene delta plain is continually.
dissolved by streams and modified by subsurface solution. The Hills are composed of a heterogeneous mix
of yellow-orange clays, silts, and sands that are weakly cemented. Soils are loamy and support much
vegetation and sediments. The impermeable nature of these soils gives rise to small wet weather ponds and
lakes. The gulf coastal lowlands are affected by Pleistocene erosion and deposition. Terraces have been
created as a result of fluctuating sea level. Two major units exist in the Gulf Coastal Lowlands: the
Appalachicola Coastal Lowlands and Woodville Karst Plain. The Appalachicola Coastal Lowlands,
located mostly within the Appalachicola National Forest, are underlain by thick elastic deposits creating
flat, sandy surfaces marked by shallow bays, few poorly defined creeks. The underlying sand and clay is
less than eighty feet thick. During the rainy season, the area is swampy resulting from the water table
approaching the surface. The Woodville Karst Plain ranges from zero to sixty feet in elevation.
Characterized by loose quartz sands over a limestone substrata, the topography is one of sinkholes and sand
dunes. Vegetation mostly consists of pines, black-jack, turkey oak, and cypress and bays in lower areas.
Most streams disappear below ground. The St. Marks River Valley has a poorly perceptible flood plain
valley due to the stream flowing upon slightly incised bedrock with a thin veneer of loose quartz sand.
During high waters, Lake Lafayette contributes to St. Marks River. The limestone of this region becomes
soluble with atmospheric C02 and organic acids. The dense dolomite is least soluble while the porous
calcitic limestones are most readily soluble.

Highley, Bradley A. et al. 1994. Recent sediments of the St. Marks River Coast,
Northwest Florida, a low-energy, sediment-starved estuary. Abstracts with Programs -
Geological Society of America. 26(4):20.

Draft... 9/14/98

The St. Marks river of northwest Florida drains parts of the central panhandle of northwestern Florida, and
a small area in southwestern Georgia. It traverses nearly 56.3 kilometers through a watershed of 1711
square kilometers. The slow-moving river carries little sediment and terminates in Apalachee Bay, a low-
energy embayment in the northeastern most gulf of Mexico. The coastal region is characterized by
mudflats, seagrass beds, and an absence of sandy beaches and barrier islands. Clastic sediments of the
coast and shelf rest on a shallow-dipping carbonate platform. The upper surface of the platform is locally
karstic. As a result, like other rivers in this region of northwest Florida, the St. Marks watershed is marked
by sinkholes and disappearing streams. The fact that the river travels underground through part of its lower
watershed serves to trap or sieve some of its plastic load. In the estuary, the undulating karst topography
causes the estuarine sediments to vary in thickness from 0 to 4+ meters. In places, in both the estuary and
lower river valley, the Tertiary carbonate units are exposed at the surface. The concave shape of the
coastline and its orientations with respect to prevailing winds result in low average wave energy.
Sedimentation is therefore controlled by riverine and tidal forces. The relatively low energy conditions
result in good preservation of the sedimentary record in the St. Marks estuary. A suite of sediment cores
has been collected in the lower river, estuary and adjacent Gulf of Mexico. Lead-210 dating results
indicate a slow average sedimentation rate (-lmm/yr). Investigation of sedimentation rates and sediment
characteristics (grain size trends and clay mineralogy) over time in the St. Marks estuary indicates that
sedimentologic conditions in this low-energy environment have been relatively stable during the recent
geologic history of the estuary.

Katz, B.G. and A.F. Choquette 1991. Aqueous geochemistry of the sand-and-gravel
aquifer, Northwest Florida. Ground Water. 29(1):47-55.

The aqueous geochemistry of the sand-and-gravel aquifer in northwest Florida was characterized as part of
the Florida Ground-Water Quality Monitoring Network Program, a multiagency cooperative study
delineating baseline and/or background water quality for the major aquifer systems throughout the State.
The aquifer is the principal source of water in northwest Florida and consists predominantly of quartz sand
with smaller amounts ofandesine, chlorite, calcite, kaolinite, and illite.

Livingston, Robert J., ed. 1991. The Rivers of Florida. New York: Springer-Verlag,

Discusses generalities of many Florida rivers. Discusses rivers of the northern gulf in some detail with
mention to the St. Marks River and its underlying strata.

Macesich, M. and O.J. Kenneth. 1989. Uranium isotopic study of Wakulla Springs.
Geological Society of America, Southeastern Section, 38th annual meeting. 21(3):49.

Uranium isotopes have been used as natural tracers in many hydrologic regimes. The identities of differing
water sources are discriminated by their uranium (U238) concentrations and their isotopic activity ratios
(U234/U238). This method was used to identify the multiple sources of water contributing to Wakulla
Springs, a first magnitude spring located within the Woodville Karst Plain and flows through the Floridan
Aquifer. Wakulla Springs has the greatest known range of discharge of all Florida springs but averages
390 cubic feet per second. Six samples were taken from different deep natural tributary conduits, from as
much as 4600 feet from the spring mouth (up to 270 feet in depth). Surface and local water samples were
tested for their uranium signatures. Preliminary results suggest that the conduits which feed the spring have
both shallow and deep sources. There is a noticeable temporal isotopic variation in the spring water
suggesting a varying degree of contribution among the conduits. Comparison of surface discharge and
deep water samples suggest that a shallow seep component may also be involved.

Maristany, Agustin, et al. 1988. Water Quality Evaluation of Lake Munson Leon
County Florida. Water Resource Assessment 88-1. NWFWMD.

Draft... 9/14/98

This report details the history of Lake Munson's water quality since its formation from a cypress swamp in
1950. Many water quality parameters were looked at and how the lakes condition has deteriorated as a
result of increased levels in some of these parameters. Finally, restoration techniques that would give
greatest improvements to the lake are recommended.

Maristany, Agustin E. and R.L. Bartel. 1989. Wetlands and Stormwater Management:
A Case Study of Lake Munson Part I: Long-Term Treatment Efficiencies. American
Water Resources Association. pp. 215-229.

The use of wetlands or wet detention ponds for stormwater management represents a relatively new
approach which has been successfully applied in recent years to address water quality problems in urban
areas. Since most systems have been in operation for only a few years, questions have been raised
concerning their long-term performance. It has been speculated that once these systems reach a state of
dynamic equilibrium, nutrient removal may decline due to the reduced nutrient uptake of a mature
ecosystem. This paper sheds some light on the subject based on a recent study by the Northwest Florida
Water Management District of a 255-acre wetland/lake system which has received wastewater effluent and
stormwater discharges for over 30 years. Nutrient and pollutant removal rates were estimated for a wide
range of parameters based on concurrent sampling of stormwater inflows, outflows, and lake water quality.
Long-term removal rates for Lake Munson compared favorably with rates reported for relatively new

Rupert, F. and Steve Spencer. 1988. Geology of Wakulla County, Florida. Bulletin No.
60. Tallahassee, Florida Geological Survey.

Wakulla County is located in the Gulf Coastal Lowlands amid poorly drained pine flatwoods, swamps, and
river basins. The average slope of this region is about four feet per mile. There are about five relict marine
beach ridges based on topographical elevation due to changing sea elevation (from shore inland): Silver
Bluff Terrace, Pamlico Terrace, Talbot Terrace, Penholoway Terrace, and Wicomico Terrace. The
Woodville Karst Plain is bounded on the west by Appalachicola Coastal Lowlands. Less than twenty feet
of quartz sand lies on the karstic St. Marks and Suwannee Limestone. The St. Marks River headwaters is
located in the Tallahassee Hills of eastern Leon County. Seven major springs exist in Wakulla County:
Indian Springs, Kini, Newport, Panacea Mineral, River Sink, Wakulla, and Spring Creek.

Stephens, D.W., B. Waddell, and J.B. Miller. 1988. Reconnaissance Investigation of
Water Quality, Bottom Sediment, and Biota Associated with Irrigation Drainage in the
Middle Green River Basin, Utah, 1986-1987. U.S. Geological Survey Water-Resource
Investigations Report 88-4011. Salt Lake City, Utah.

Reconnaissance of wildlife areas in the middle Green River basin of Utah was conducted during 1986 and
1987 to determine whether irrigation drainage has caused, or has the potential to cause significant harmful
effects on human health, fish, and wildlife, or may adversely affect the suitability of water for beneficial
uses. Studies at Stewart Lake Waterfowl Management Area and Ouray National Wildlife Refuge indicated
that concentrations of boron, selenium, and zinc in water, bottom sediment, and biological tissue were
sufficiently large to be harmful to fish and wildlife, and to adversely affect pesticides in surface water
generally were less than established standards with the exception of gross alpha radiation, which exceeded
by factors of three to five times the standard of 15 picocuries per liter in water from two of the drains
discharging into Stewart Lake.

Swanson, Sloan, and Chernets. 1996. Lake Lafayette Management: A Report Outlining
Lake Shore, In-Lake, and Land Use Management Proposals. Dept. of Growth and
Environmental Management, Tallahassee, FL.

Draft... 9/14/98

This report addresses the degradation of Lake Lafayette by urban stormwater and point source discharge.
Localities around the lake of greatest concern are highlighted. The report stresses the need for an
immediate and comprehensive management program for Lake Lafayette. The offered management
program consists of strategies for land use controls, in-lake management and clean-up, and subsequent
research and monitoring.

Turner, R.R., et al. 1975. The Effect of Urban Land Use on Nutrient and Suspended-
Solids Export from North Florida Watersheds. Florida State University, Tallahassee
Dept. of Oceanography. in 'Mineral Cycling in Southeastern Ecosystems', (Conf-
740513), p.868-888.

Two watersheds of similar size, geomorphology, and pedology representing forested-agricultural and
residential-commercial (urban) land use were hydrologically instrumented to obtain comparative nutrient
and suspended-solids export data. Constituents measured included suspended solids, total dissolved solids,
dissolved silicon, and dissolved nutrients (nitrogen and phosphorus). Intensive hydrochemical analysis of
runoff from 13 storms in the urban watershed and 8 storms in the forested watershed demonstrated a strong
contrast in the magnitude and temporal distribution of nutrient and suspended-solids concentrations and
exports. Suspended-solids concentration and export were directly dependent on stream discharge.
Although concentrations of dissolved constituents were generally inversely dependent on stream discharge,
export of dissolved constituents were directly dependent on stream discharge. Higher cumulative stream
discharge in the urban watershed thus exported higher total storm loads of all constituents except dissolved
silicon. Exports of suspended solids, total dissolved solids, and all the dissolved nitrogen species from the
urban watershed were higher than the higher volume of stream discharge in this watershed might otherwise
indicate, suggesting significant additional sources of these constituents in the urban watershed. Exports of
dissolved phosphorus from the urban watershed were also higher but near what the higher volume of
stream discharge might indicate. Exports of dissolved silicon from the urban watershed were lower than
from the forested watershed despite the higher volume of stream discharge in the urban watershed.
Observed differences in exports were related to the changes in hydrology associated with urban
development, i.e., in streamflow rate, total volume of stream discharge, and the relative significance of
various pathways of water movement, as well as to increased diffuse anthropogenic inputs in the urban
watershed. Comparison of material loads exported by storm flow and low flow in each watershed
suggested increased significance of storm events in materials export in the urban watershed.

Winchester, J.W., et al. 1995. Atmospheric deposition and hydrogeologic flow of
nitrogen in northern Florida watersheds. Geochimica et Cosmochimica Acta.

Atmospheric wet and dry deposition ("acid rain") appears to be the principal source of nitrogen in twelve
northern Florida watersheds that range from Pensacola to Gainesville (Escambia to Alachua Counties).
The study was based on statistical analysis of chemical concentrations measured for more than ten years in
weekly rainfall samples of the National Atmospheric Deposition Program, NADP, and more than twenty
years of river water samples of the US Geological Survey, USGS. River fluxes of total dissolved nitrogen
average close to the atmospheric deposition fluxes of nitrate and ammonium ions. Factor analysis was
applied to the datasets to resolve principal components: (1) in atmospheric data, that distinguish air
pollution nitrate and sulfate from sea salt sodium and chloride, and (2) in surface water data, that
distinguish ground water Ca, Mg, and silica from metioric water nitrate and sulfate. Relationships within
the sets of measured concentration data suggest that, following atmospheric deposition, inorganic nitrogen
undergoes biogeochemical transformation within the watersheds, which results in inorganic nitrogen being
transformed to organic forms. River concentration ratios N/P in the watersheds are high, averaging twice
the Redfield mole ration N/P=16 for aquatic plant nutrients. The results indicate that excess dissolved
nitrogen could be temporarily recycled in the watersheds but not retained, so that it could eventually flow
to the coastal zone where N may be a limiting nutrient for marine plants. Chemical interactions of meteoric
water within the watersheds depend on geologic, hydrologic, and biogeochemical processes and are
certainly complex. However, in one watershed that is geologically the simplest, separate statistical

Draft... 9/14/98

analyses of river water composition during high and low flow conditions show nitrate and sulfate to be
correlated during high flow, but not during low flow, providing further evidence for an atmospheric
nitrogen source and watershed transformation after deposition.

Wooten, Nicholas, et. al. July 1991. Lake Lafayette Basin Plan Stormwater
Management Plan Vol. IV. NWFWMD.

This report (Volume IV) represents a description of the Lake Lafayette Basin Stormwater Management
Plan developed by the Northwest Florida Management District in Cooperation with Leon County and the
City of Tallahassee, Florida. The plan developed represents the culmination of a Five-Year Study to
identify flooding and water quality problems in four major basins in the study area which included the Lake
Munson, Lake Lafayette, Lake Jackson, and Fred George Sink Basins. The technical details, which
included the use of hydrologic models as a tool for plan development have been described in a technical
report (Volume VI) for this study. An overall summary of the problems identified and the recommended
plan for the Lake Lafayette Basin may be found in the Executive Summary (Volume I) of this study.
Yon, J. Williams. 1966. Geology ofJefferson County, Florida. Tallahassee, Published for the Florida
Geological Survey. This book provides a detailed description on the geomorphology of Jefferson County.


Alfoldi, Laszlo, 1983. Movement and Interaction of Nitrates and Pesticides in the
Vegetation Cover Soil Groundwater Rock System. Environmental Geology vol. 5 no.
1, pp. 19-52.

Review of solute dynamics primarily in the vadose zone. Divides soil into three areas. (1) Near the root
zone constituent transformation system. (2) Between the root zone and water table constituent
transportation system. (3) In the water table constituent accumulation system. Water provides horizontal
transportation in area (3). In aquifers under natural conditions the lateral movement of water will not be
significant, excluding the flow through karstic formations, and consequently the infiltrated pollution
propagates only slowly and accumulates relatively rapidly. As a result of water intake (pumpage)
operations, flow will turn toward a well with accelerating velocities, and pollution may rapidly reach the
place of water intake.

Anderson, LJ and H Kristiansen, 1984. Nitrate in Groundwater and Surface Water
Related to Land Use in the Karup Basin, Denmark. Environmental Geology, vol. 5, no.
5, pp. 207-212.

Investigation and analysis of the content and distribution of nitrate in the groundwater and surface water in
the Karup Basin area. Soil profiles indicate (1) an upper oxidation zone where nitrate is present and iron is
absent and (2) a lower reduction zone where iron is present but nitrate is absent. In cultivated areas with
high nitrate concentrations, the oxidation zone is found at depths of 8-15m below the groundwater table. In
forested areas, the oxidized nitrate zone is thinner, 5-7 m, and has a low nitrate content.

Asbury, C.E. and E.T. Oaksford. 1997. A comparison of drainage basin nutrient inputs
with instream nutrient loads for seven rivers in Georgia and Florida, 1986-90. U.S.
Geological Survey, Water-Resources Investigation Report 97-4006.

Instream nutrient loads of the Altamaha, Suwannee, St. Johns, Satilla, Ogeechee, Withlacoochee, and
Ochlockonee River basins were computed and compared with nutrient inputs for each basin for the period
1986-90. Nutrient constituents that were considered included nitrate, ammonia, organic nitrogen, and total
phosphorus. Sources of nutrients considered for this analysis included atmospheric deposition, fertilizer,
animal waste, wastewater-treatment plant discharge, and septic discharge. Although instream nutrient

Draft... 9/14/98

loads constitute only one of the various pathways nutrients may take in leaving a river basin, only a
relatively small part of nutrient input to the basin leaves the basin in stream discharge for the major coastal
rivers examined in this study. The actual amount of nutrient transported in a river basin depends on the
ways in which nutrients are physically handled, geographically distributed, and chemically assimilated
within a river basin.

Ayres and Associates, July, 1989. Onsite Sewage Disposal System Research in Florida-
Performance Monitoring and Ground Water Quality Impacts of OSDS's in Subdivision
Developments. Report to the State of Florida Department of Health and Rehabilitative
Services. Tampa, Florida.

Progress report presenting the first results of field monitoring of ground water below unsewered
subdivisions and monitoring of the performance of individual OSDS's. Ground water was monitored
beneath four specific subdivisions in four different hydrogeologic regimes. The study sites were selected to
represent subdivisions developed under the requirements of the 1983 revisions to Chapter 10D-6, Florida
Administrative Code, Standards for Onsite Sewage Disposal Systems. The study areas were also intended
to reflect the various hydrogeologic regimes found in Florida. Findings revealed that septic tank effluent
(STE) in Florida was similar in character to effluent generated in other areas of the U.S. Notable
exceptions were the lower concentrations of most constituents in septage and the higher suspended solids in
STE. These may be attributed to the comparatively higher temperatures of both waste streams. Ground
water quality in the vicinity of relatively new subdivisions (Less than 20 years old) served by individual
OSDS's had not suffered substantial widespread contamination. Localized areas of potential impact were
observed. If subdivision impacts are to occur, they may take decades to manifest themselves due to the low
ground water seepage velocities at the test sites. Report contains a synopsis of prior research as well as
details of the methods and results of the investigation.

Ayres and Associates, March, 1993. Onsite Sewage Disposal System Research in
Florida- The Capability of Fine Sandy Soil for Septic Tank Effluent Treatment: A Field
Investigation at an In-Situ Lysimeter Facility in Florida. Report to the State of Florida
Department of Health and Rehabilitative Services. Tampa, Florida.

Investigation into the treatment capabilities of fine sandy soils common to Florida under controlled
experimental conditions in the field. A unique field lysimeter research facility was designed, constructed,
and operated under controlled conditions. Key variables investigated were the thickness of unsaturated soil
below the wastewater infiltration system and the hydraulic loading rate of effluent to the system. The
report describes lysimeter site selection and characterization, investigative methodology, and the results of
the lysimeter facility monitoring for the first six months of operation. Preliminary results showed
substantial attenuation of key parameters related to septic tank effluent (STE) treatment in the fine sandy
soils. Total phosphorous removal was high. There were indications that the phosphorous capacity of the
soil was being approached. It appeared that phosphorous removal was effected negatively by less
unsaturated zone travel and greater hydraulic loading. Total organic carbon reduction was on the order of
80 percent. Total Kjeldahl nitrogen reduction was in excess of 97 percent indicating almost complete
nitrification of STE nitrogen. The nitrate produced by nitrification was transported through the system in
relatively high concentrations.

Ayres Associates, 1987. The Impact of Florida's Growth on the Use of On-Site Sewage
Disposal Systems. Report to the Florida Department of Health and Rehabilitative
Services. Tampa, Florida.

Summaries of the major soil types in Florida and their distribution; critical characteristics of the soils
affecting their ability to accept and treat on-site disposal system (OSDS) effluent; soil types supporting
most of the current and future OSDS installations in Florida; and the density and geographic distribution of
OSDS designs most commonly used in Florida.Results indicate a continuing trend of high OSDS use in

Draft... 9/14/98

urban fringes and the increasing use of soils with severe limitations for proper operation of conventional
OSDS designs. Much of the suitable soil in many counties has been developed and there is increasing
pressure to utilize alternative OSDS designs on inadequate soils. The characteristics that are expected to
limit the ability of these soils are saturated conditions under and around drainfields due to a high ground
water table and excessive permeability of sandy soils.

Ayres Associates, February, 1993. An Investigation of the Surface Water
ContaminationPotential From On-Site Sewage Disposal Systems (OSDS) in the Turkey
Creek Sub-basin of the Indian River Lagoon. Report to St. Johns River Water
Management District and the Florida Department of Health and Rehabilitative Services.
Tampa, Florida.

Study conducted as part of the State of Florida's Surface Water Improvement and Management Program.
The study assessed the impact of several existing OSDS on water quality in adjacent canals that eventually
drain to the Indian River Lagoon. Ground water and surface water quality investigations were conducted
around two existing OSDS and one control area. Pollutant loading to the ground water and adjacent
drainage canals was estimated. Ground water flow characteristics and seepage rates into the canals were
determined. Recommended estimation of a preliminary nutrient budget to ascertain whether nutrient
reduction techniques for OSDS should be initiated.

Ayres Associates, March, 1993. Onsite Sewage Disposal System Research in Florida:An
Evaluation of Current Onsite Sewage Disposal System (OSDS) Practices in Florida.
Report to the Florida Department of Health and Rehabilitative Services. Tampa, Florida.

Chapter 1 Overview of the onsite wastewater system regulatory program in Florida. Evaluation of onsite
wastewater treatment system performance. Summary of OSDS performance monitoring in Florida.
Technical guidelines for site evaluation, system design, construction, operation, and maintenance. Review
of regulations and enforcement. Recommendations for overall improvement of systems and regulations.

Bartholomew, WV and FE Clark (eds.), 1965. Soil Nitrogen. American Society of
Agronomy. Madison Wisconsin.

Text reviewing nitrogen geochemistry.

Battoe, LE and EF Lowe, 1992. Acidification of Lake Annie, Highlands Co., FL. Water,
Air, and Soil Pollution 65:69-80.

Lake Annie is a clear-water seepage lake in south-central Florida, remote from significant pollutant
sources. It is suggested that the lake's acidification was a threshold phenomenon wherein, following
depletion of the watershed's buffering capacity, acidification of the lake was rapid.

Berndt, MP, 1990. Sources and Distribution of Nitrate in Groundwater at a Farmed Field
Irrigated with Sewage Treatment-Plant Effluent, Tallahassee, Florida. US Geological
Survey Water Resources Investigations Report 90-4006. Tallahassee, Florida.

Effluent from a secondary sewage-treatment plant and fertilizers containing inorganic nitrogen were
applied in conjunction with the operation of a commercial farm. Water samples indicated that conversion
of nitrogen species to nitrate was complete before the nitrogen-enriched water reached the water table.
Water samples from monitoring wells inside the sprayfield had nitrate concentrations in excess of the
drinking water standard of 10 mg/L. Samples from wells outside of the sprayfield had background levels
of nitrate. Isotopic analysis indicated that nitrogen contributions from fertilizers was significant in some
test areas.

Draft... 9/14/98

Bicki, Thomas J. et al, 1984. Impact of On-Site Sewage Disposal Systems on Surface
and Ground Water Quality. Report to Florida Department of Health and Rehabilitative
Services, Institute of Food and Agricultural Sciences. Gainesville, Florida.

Brief description of on-site disposal systems. Characterization of septic tank effluent constituents of
concern nitrogen, phosphorous, chlorides, sulfates, sodium, detergent MBAS, toxic organic, bacteria, and
viruses. Review of constituent fate and transport, water quality surveys, ground water monitoring studies,
lysimeter, sand filter, and column studies. Summary ofon-site sewage disposal system density, ground
water resources and research needs in Florida. Extensive list of references.

Boggess, C.F. 1994. The Biogeoeconomics of Phosphorus in the Kissimmee Valley.
PhD dissertation, Environmental Engineering Sciences. Univ. of Florida, Gainesville.
234 pp.

This study tested ways of relating a chemical cycle to a regional economy, using phosphorus in the
northern drainage basin of Lake Okeechobee, Florida as an example. Among the methods used were a
mass balance approach to phosphorus budgeting, dynamic modeling for runoff simulation, cost
effectiveness for economic analysis, and emergy evaluation techniques. Spatial data on land use and
management practices were organized using a geographic information system. Phosphorus management
scenarios were evaluated and compared in terms of their ability to meet alternative goals for the region:
physical (i.e., target phosphorus load reduction to Lake Okeechobee); economic (i.e., minimize cost of
phosphorus reduction); and energetic (i.e., maximize regional empower). Five principles for the new field
of biogeoeconomics were proposed for managing at the interface between an elemental cycle, its role in the
environment, and its economic use to enhance the self-organizing properties of the landscape.

Bradley, PM, CM Aelion, and DA Vroblesky, 1992. Influence of environmental factors
on denitrification in sediment contaminated with JP-4 jet fuel. Ground Water, vol. 30,
no. 6: 843-848.

Attempted to identify factors likely to influence microbial activity under denitrifying conditions in a
shallow aquifer contaminated by an 83,000 gallon jet fuel spill. Examined the fate of amended NO3, the
effect ofpH, NO3 and P04 on denitrification, and the variability of denitrification in sediments collected at
the site. Denitrification rates were at least 38% lower at pH=4 than observed at pH=7.

Brooks, R.P. et. al. 1989. A Methodology for Biological Monitoring of Cumulative
Impacts on Wetland, Stream, and Riparian Components of Watersheds. Paper presented
and submitted for inclusion in the proceedings of the International Symposium:
Wetlands and River Corridor Management, Charleton, SC.

Biotic communities were compared between two watersheds in the Ridge and Valley Province of central
Pennsylvania. An undisturbed watershed served as a reference for comparisons against a similar watershed
that was disturbed by agricultural and developmental activities.

Byron, E.R. 1989. Land-use and water quality in tributary streams of Lake Tahoe,
California-Nevada. Journal of Environmental Quality. 18:84.

Concluded that land use usually affects water quality to a greater degree than geomorphology or soil types
of the drainages. Long-term average nutrient flux originating from non-point sources closely reflects
intensity and location of development in a watershed. Disturbances that affects the most errosive areas of
watershed has the greatest affect on water quality.

Draft... 9/14/98


Canter, Larry W, 1996. Nitrates in groundwater. Lewis Publishers. Boca Raton,

Excellent text. Reviews nitrogen dynamics in groundwater, sources, fate, pollution control, modeling, and

Correll, D.L., T.E. Jordan, and D.E. Weller. 1992. Nutrient flux in a landscape: effects
of coastal land use and terrestrial community mosaic on nutrient transport to coastal
waters. Estuaries. 15:431-442.

Long-term interdisciplinary studies of the Rhode River estuary and its watershed in the mid-Atlantic
coastal plain of North America have measured fluxes of nitrogen and phosphorus fractions through the
hydrologically-linked ecosystems of this landscape. These ecosystems are upland forest, cropland, and
pasture; streamside riparian forests; floodplain swamps; tidal brackish marshes and mudflats; and an
estuarine embayment. Croplands discharged far more nitrogen per hectare in runoff than did forest and
pastures. However, riparian deciduous hardwood forest bordering the cropland removed over 80 percent of
the nitrate and total phosphorus in overland flows and about 85 percent of the nitrate in shallow
groundwater drainage from cropland. Nevertheless, nutrient discharges from riparian forests downslope
from croplands still exceeded discharges from pastures and other forests. The atomic ratio of nitrogen to
phosphorus discharged from the watersheds into the estuary was about 9 for total nutrients and 6 for
inorganic nutrient fractions. Such a low N:P ratio would promote nitrogen rather than phosphorus
limitation ofphytoplankton growth in the estuary. Estuarine tidal marshes trapped particulate nutrients and
released dissolved nutrients. Subtidal mudflats in the upper estuary trapped particulate P, released
dissolved phosphate, and consumed nitrate. This resulted in a decrease in the ratio of dissolved inorganic
N:P in the estuary. However, the upper estuary was a major sink for total phosphorus due to sediment
accretion in the subtidal area. bulk precipitation accounted for 31 percent of the total nongaseous nitrogen
influx to the landscape, while farming accounted for 69 percent. Forty-six percent of the total non-gaseous
nitrogen influx was removed as farm products, 53 percent either accumulated in the watershed or was lost
in gaseous forms, and 1 percent entered the Rhode River. Of the total phosphorus influx to the landscape, 7
percent was from bulk precipitation and 93 percent was from farming. Forty-five percent of the total
phosphorus influx was removed as farm products, 48 percent accumulated in the watershed, and 7 percent
entered the Rhode River. These nitrogen and phosphorus discharges into the Rhode River, although a
small fraction of total loading in the watershed, were large enough to cause seriously overenriched
conditions in the upper estuary.

Degen, MB, RB Reneau, Jr., C Hagedor, and DC Martens, 1991. Denitrification in
Onsite Wastewater Treatment and Disposal Systems. Virginia Water Resources
Research Center Bulletin 171. Blacksburg, Virginia.

Study evaluating the effects of effluent type, effluent loading rate, dosing interval, and temperature on
denitrification in onsite wastewater treatment and disposal systems. From the study, a model was
developed that predicted the mean nitrous oxide production for each combination of the experimental
treatments. The results of the study and the model indicate that denitrification can be enhanced in
OSWTDS's by the application of anaerobic effluent at the Virginia Department of Health's recommended
effluent loading rate to surface soil horizons using a 48-hour dosing interval.

Dillon, P.J., and W.B. Kirschner. 1975. The effects of geology and land use on the
export of phosphorus from watersheds. Water Research. 9:135-148.

The export of total phosphorus from 34 watersheds in Southern Ontario was measured over a 20-month
period. The annual average export for igneous watersheds (i.e., those of the Canadian Shield) that were
forested was 4.8 mg/m2/yr, significantly different from the average (11.0 mg/m^2/yr) for watersheds that
included pasture as well as forest. Similarly, on sedimentary rock, the mean export from forested

Draft... 9/14/98

watersheds (10.7 mg/m^2/yr) differed significantly from those with forest and pasture (28.8 mg/m^2/yr).
The difference between watersheds of different geology but similar land use were also highly significant.
Additional data from the literature supported our conclusions. Other forested igneous watersheds of
plutonic origin averaged 4.2 mg/m^2/yr of total phosphorus exported; forested igneous watersheds of
volcanic origin, however, averaged 72 mg/mA2/yr. The overall average export from each type of watershed
as classified by geology and land use was very similar to that for the same classification found in our study.
The effects of agriculture and urbanization were to greatly increase the total phosphorus exported. Wide
ranges of values probably reflet the intensity of land use.

Eisenreich, SJ (ed.), 1981. Atmospheric Pollutants in Natural Waters. Ann Arbor
Science Publishers Inc. Ann Arbor, Michigan.

Text reviewing all aspects of atmospheric pollution. Discussions of nitrogen and phosphorous deposition
in Florida, modeling of atmospheric removal processes, metals in the atmosphere, and anthropogenic

Famworth, E. G. et al. 1979. Impacts of Sediment and Nutrients on Biota in Surface
Waters of the United States. Environmental Research Laboratory, Office of Research
and Development, U.S. EPA. Athens, Georgia. 134 pp.

A review of research on the impacts of sediment, nitrogen, and phosphorus on aquatic biota was performed
to determine the influences of sediment and nutrients on biota, to suggest directions for future research, and
to provide suggestions for management of freshwater systems across the United States. This report is
divided into two sections. The first section provides an organization and background information to enable
incorporation of large amounts of available information and allow assessment of impacts at several
hierarchical levels. Included are a hierarchical scheme which is the foundation of the analytical study; a
regional analysis of the concentrations of sediment, nitrogen, and phosphorus in surface waters; a review of
biotic impact assessment approaches; and a review of modelling of sediment and nutrient impacts. The
second section reviews the impacts of sediment, nitrogen, and phosphorus on biota, integrates this
information into the hierarchical scheme developed in the first section, and shows how the hierarchical
scheme can be used for impact analysis.

Florida Geological Survey, 1988. Geology of Wakulla County, Florida. Florida
Geological Survey Bulletin No. 60. Tallahassee, Florida.

General overview of the geology and mineral resources of Wakulla County based on existing literature and
well data on file at the Florida Geological Survey.

Frissel, MJ, and JA van Veen (eds.), 1981. Simulation of Nitrogen Behavior of Soil-
Plant Systems: papers of a workshop, models for the behavior of nitrogen in soil and
uptake by plants, comparison between different approaches. Center for Agricultural
Publishing and Documentation. Wageningen, Netherlands.

Compilation of papers from a conference.

Groffman PM, Jaworski, NA, 1990. Watershed Nitrogen Management "Upper Potomac
River Basin Case Study." Report to USEPA. Narragansett, RI.

The authors developed a watershed nitrogen mass balance. Results indicated that the highest nitrogen
export occurred during high surface water flows. Monthly loading can vary by a factor of three, therefore
annual mass balances are required to evaluate best management practices (BMP). BMP's must maintain
their integrity throughout the year since most nitrogen transport is during high-flow periods. Current
BMP's do not strongly affect nitrogen storage mechanisms within agricultural fields.

Draft... 9/14/98

Groffman, Peter M. Ecology of Nitrification and Denitrification in Soil Evaluated at
Scales Relevant to Atmospheric Chemistry. In Rogers, JE and WB Whitman (eds.)
Microbial Production and Consumption of Greenhouse Gases: Methane, Nitrogen
Oxides, and Halomethanes. American Society for Microbiology. Washington DC.

Organismal scale processes are well understood translation of knowledge to useful information applicable
at a larger scale is difficult. Plant community patterns are often strongly related to nitrogen availability,
and different plant communities should thus exhibit distinct patterns of nitrification. At the landscape
scale, soil type and plant community type are useful conceptual regulators of denitrification. Soil texture
and drainage are strong controllers of oxygen availability and indirectly regulate nitrate supply through
their influence on nitrification. Similarly, plant community type affects nitrate supply by controlling the
nitrification rate. Both soil type and plant community type have strong effects on the decomposition of
plant material and thus influence carbon supply to denitrifiers. Since plant community composition
differences are the integrative product of the same ecological factors that influence microbial trace gas
fluxes (water and nutrient availability), remote sensing of plant variables should be useful for large-scale
soil-atmosphere gas exchange studies.

Groffman, Peter M. and James M Tiedje, 1989. Denitrification in north temperate forest
soils: relationships between denitrification and environmental factors at the landscape
scale. Soil Biology and Biochemistry Vol. 21 No. 5, pp. 621-626.

Relationships between annual denitrification nitrogen loss and soil physical and biological factors were
investigated in nine north temperate forest soils of different texture and drainage classes. Soil texture was
analyzed numerically by using percentage of sand as a variable. Soil wetness was quantified by a
continuous drainage index function. Found that most of the variability (86%) in annual nitrogen loss to
denitrification was explained with a multiple regression model using soil texture (%sand) and soil drainage
index as predictor variables. Analysis at the landscape scale was necessary to derive this relationship.
Denitrification enzyme activity (DEA) and DEA-to-biomass C ratio (Strongest predictor observed)
accounted for up to 96% of the variation in annual denitrification nitrogen loss (ratio was higher in poorly-
drained soils than well-drained soils. Percentage sand and soil wetness could also account for a large
proportion of the annual variation in denitrification rates among sites (r2 = 0.86).

Groffnan, Peter M. and James M Tiedje, 1989. Denitrification in north temperate forest
soils: relationships between denitrification and environmental factors at the landscape
scale. Soil Biology and Biochemistry Vol. 21 No. 5, pp. 621-626.

A report on different aspects of the above mentioned study. Found that denitrification activity was highest
in the spring and fall, and lowest in the summer in Michigan. Over 80% of the annual nitrogen loss to
denitrification occurred during brief (3-6 week) periods of high activity in the spring and fall. Rates of
denitrification during these periods exceeded 0.5 kg N/ha/day in some soils. Estimates of annual nitrogen
loss to denitrification ranged from <1 kg N/ha/yr. in a well-drained sand soil to over 40 kg N/ha/yr. in a
poorly drained clay loam soil. Lack of available nitrate was the primary factor limiting denitrification in
summer, but available carbon was probably occasionally limiting, especially in the well-drained soils.

Heede, B.H. 1985. Interactions Between Streamside Vegetation and Stream Dynamics.
Symposium paper presented at Riparian Ecosystems and Their Management:
Reconciling Conflicting Uses. Tucson, Arizona. pp. 54-57.

Interrelationships between vegetation and hydrologic processes in riparian ecosystems must be considered
by managers before they attempt to alter these natural systems. A 5-year experiment demonstrated that
logs that fall across the channel from streamside forests dissipate flow energy, maintain channel stability,
decrease bedload movement, and increase water quality.

Draft... 9/14/98


Hendry, MJ, RW Gillham, and JA Cherry, 1983. An integrated approach to
hydrogeologic investigations a case history. Journal of Hydrology, vol. 63: 211-232.

Study addressing the occurrence and migration of nitrate in a thin (3-7 m) phreatic sand aquifer beneath
agricultural land. High nitrate concentrations were found at shallow depths, while concentrations at greater
depths were close to zero. Through the application of physical hydrogeologic methods of investigation,
geochemical studies, environmental isotope studies, numerical simulations of groundwater flow and solute
transport, it was shown that denitrification was the principal cause of the nitrate distribution. The study
showed that denitrification can be an important process in groundwaters at a regional scale.

Hirsch, R.M., W.M. Alley, and W.G. Wilber. 1988. Concepts for a National Water-
Quality Assessment Program. U.S. Geological Survey Circular 1021.

Outlines three major goals of the National Water-Quality Assessment Program. Goals include providing a
nationally consistent description of current water-quality conditions for a large part of the Nation's water
resources, define long-term trends (or lack of trends) in water quality, and identify, describe, and explain,
as possible, the major factors that affect observed water-quality conditions and trends.

Hubbard, R.K. 1990. Dissolved and Suspended Solids Transport from Coastal Plain
Watersheds. Journal of Environmental Quality. 19:413-420.

Excessive amounts of dissolved or suspended solids in surface runoff or base flow may degrade the quality
of streams, lakes, or other water bodies. Loads of dissolved and suspended solids in streamflow reflect the
quality of water entering via surface runoff or base flow. This study was conducted to determine the
concentrations and loads of dissolved and suspended solids in Coastal Plain streamflow; to examine
relationships between concentrations, loads, and flow rate; and to determine overall streamflow water
quality for these parameters. Dissolved solids and suspended sediment concentrations were determined on
a weekly or high-flow storm event streamflow samples collected at gaging stations on three subwatersheds
of the Little River Watersheds located near Tifton, GA. Dissolved solids concentrations ranged from 19 to
159 mg/L, and generally decreased as per unit area instantaneous discharge rate increased. Suspended
sediment concentrations ranged from 1 to 137 mg/L, and generally increased as per unit area instantaneous
discharge rate increased. Regression analyses showed good relations between log transforms of both
dissolved solids load and suspended sediment load, vs. total monthly runoff. Mean suspended sediment
concentrations during high-flow events were greater than means from the overall data set, while mean
concentrations of dissolved solids from these events were reduced relative to the overall data set. The
study showed that dissolved solids are the major component of total solids in coastal Plain streamflow. The
mean dissolved and suspended sediment concentrations during the study were 67, 60, and 51 mg/L and 14,
17, and 14 mg/L for Watersheds B, F, and K, respectively. Overall, the study showed that, as measured on
these watersheds, Coastal Plain streamflow is of good quality in terms of both dissolved and suspended
solids. This good quality may reflect land-use practices designed to prevent soil erosion, but primarily
reflects the Coastal Plain landform shape, which causes sediments eroded from the uplands to be deposited
in the riparian zone before they can enter streamflow.

Izuno, F.T., et al., 1991. Phosphorus concentrations in drainage water in the Everglades
Agricultural Area. Journal of Environmental Quality. 20:608-619.

Phosphorus in drainage water leaving the Everglades Agricultural Area (EAA) in southern Florida is
alleged to be contributing to the accelerated eutrophication of Lake Okeechobee and the degradation of the
Water Conservation Areas and the Everglades National Park ecosystems. Agricultural "best management
practices" (BMPs) offer a means for achieving reductions in P in drainage water. Prior to developing and
implementing BMPs, it is necessary to establish baseline EAA P concentrations. Baseline total P (TP) and
total dissolved P (TDP) concentrations for various crop and field conditions in the EAA were determined.
Thirty-six 0.7-ha plots were installed at four locations. Average TP and TDP concentrations were derived

Draft... 9/14/98

from 6 to 30 drainage events for each of five conditions between November 1988 and December 1989:
sugurcane, radish, and cabbage production fields, flooded fallow fields, and drained fallow fields. Baseline
TP and TDP concentrations for main farm canals and rainfall were also determined. Average TP
concentrations ranged rom 0.25 mg/L for radishes to 1.03 mg/L during the drain-down of flooded fallow
plots. Total dissolved P concentrations ranged from 48 to 80% of TP. Main farm canal TP concentrations
averaged 0.16 mg/L. Total P concentrations in rainfall averaged 0.07 mg/L. Total P in drainage water
during 1989 for sugurcane, cabbage, and drained fallow fields were 0.72, 1.38, and 0.59 kg/ha,
respectively. During the radish season, drainage water TP loading was 0.8 kg/ha. Flooded fallow fields
after radishes yielded a TP loading rate of 3.82 kg/ha. Total P loading to the fields from rainfall averaged
0.7 kg/ha. Total dissolved P loading rates ranged from 25 to 60% of TP. Potential areas for BMP
development and implementation for P concentration and loading reduction in the EAA include drainage
rate, volume, and timing management, fertilizer use reduction, and enhanced crop rotation strategies.

Jaworski, N.A., et al. 1992. A watershed nitrogen and phosphorus balance: The upper
Potomac River Basin. Estuaries. 15(1):83-95.

Nitrogen and phosphorus mass balances were estimated for the portion of the Potomac River basin
watershed located above Washington, D.C. The total nitrogen (N) balance included seven input source
terms, six sinks, and one "change-in-storage" term, but was simplified to five input terms and three output
terms. The phosphorus (P) balance had four input and three output terms. The estimated balances are
based on watershed data from seven information sources. Major sources of nitrogen are animal waste and
atmospheric deposition. The major sources of phosphorus are are animal waste fertilizer. The major sink
for nitrogen is combined denitrification, volatilization, and change-in-storage. The major sink for
phosphorus is change-in-storage. River exports of N and P were 17% and 8%, respectively, of the total N
and P inputs. Over 60% of the N and P were volatilized or stored. The major input and output terms on the
budget are estimated from direct measurements, but the change-in-storage term is calculated by difference.
The factors regulating retention and storage processes are discussed and research needs are identified.

Johengen, T.H., A.M. Beeton, and D.W. Rice. 1989. Evaluating the effectiveness of best
management practices to reduce agricultural nonpoint source pollution. Lake and
Reservoir Management. 5:63-70.

The Saline Valley project is one of 20 national projects sponsored by the U.S. Department of Agriculture
(USDA) under the Rural Clean Water Program (RCWP) to evaluate methods of controlling agricultural
nonpoint source pollution. The goals of this project were (1) to evaluate whether a voluntary approach
using cost-share incentives would produce adequate participation by local farmers and (2) to reduce
phosphorus loads from the area by 40 percent. Water quality has been monitored since 1981 using weekly
grab samples and flow measurements. Trends in empirical relationships between concentration and
discharge at three sampling stations were used to examine the effectiveness of best management practices
(BMP). These relationships were highly variable among the sub-basins and years, and did not appear to
correlate with areal estimates of BMP implementation. Overall, low participation within the project area
hindered the ability to quantify changes in water quality resulting from BMP implementation and prevented
the project from meeting its phoshorus reduction goals.

Jones, GW, Dr. SB Upchurch, KM Champion, 1996. Origin of nitrate in Ground Water
Discharging from Rainbow Springs, Marion County, Florida. Southwest Florida Water
Management District. Brooksville, Florida.

Report from an investigation to determine the sources of increasing nitrate levels in Rainbow Springs.
Historic data indicated a possible 20 fold increase in nitrate concentrations during the past 40 years. The
investigation involved delineating the areas where nitrate is entering the aquifer system, identifying the
land uses that are contributing the nitrate, and determining what can be done to slow or reverse the
nitrification of ground water in the area. The three greatest sources of nitrate were determined to be pasture
fertilization, horses, and cattle. Development-related nitrogen sources such as septic tanks, sewage, and

Draft... 9/14/98

residential turf and golf courses were determined to be minor contributors of nitrogen to the study area. It
was concluded that nothing could be done to reduce current nitrate loading from the springs to the river due
to the lag-time between nitrogen application and flow from the springs. The effects of reduced nitrate
loading to the land may not be evident in water quality for a decade or more. Development of best
management practices was recommended.

Jones, R.C. and B.H. Holmes. 1985. Effects of Land Use Practices on Water Resources in
Virginia. Bulletin 144, Virginia Water Resources Reserach Center, Virginia Polytechnic
Institute and State University, Blacksburg, Virginia.

This study reviews the relationship between land use and water resources in Virginia. It examines three
major land uses in the state--agriculture, urban, and forestry activities. For each land use, the relevant
literature and state management programs are reviewed. In addition, the report outlines research needs in
each area.

Kirkner and Associates, Inc, 1987. Risk Assessment of On-Site Sewage Disposal
Systems for Selected Florida Hydrologic Regions. Report to the Florida Department of
Health and Rehabilitative Services. Lake Wales, Florida.

Preliminary report complimenting the work of Ayres and Associates, 1987. Describes techniques used to
select and monitor high-density subdivisions utilizing on-site disposal systems (OSDS). Uncertainty and
sensitivity analysis techniques were applied to eight hydrologic regions distributed throughout Florida.

Korom, SF, 1992. Natural denitrification in the saturated zone: a review. Water
Resources Research, vol. 28, no. 6: 1657-1668.

Review synthesizing published literature on natural aquifer denitrification. Discusses microbial processes
and environmental requirements. Suggests guidelines for future research.

Lawrence, SJ, 1996. Nitrate and Ammonia in Shallow Ground Water, Carson City Urban
Area, Nevada, 1989. US Geological Survey Water Resources Investigations Report 96-
4224. Carson City, Nevada.

A network of 26 wells at 20 sites was established to investigate groundwater quality beneath the oldest and
most developed part of the Carson City urban area. Nitrate and ammonia concentrations were positively
correlated with several other solute concentrations. Contamination sources might be nitrogen-based
fertilizers, septic systems, or leaky municipal sewer lines. Nitrification and denitrification control nitrate
and ammonia concentrations beneath the study area.

Lowrance, R., et al., 1984. Riparian forests as nutrient filters in agricultural watersheds.
BioScience. 34:374-377.

Riparian (streamside) vegetation may help control transport of sediments and chemicals to stream channels.
Studies of a coastal plain agricultural watershed showed that riparian forest ecosystems are excellent
nutrient sinks and buffer the nutrient discharge from surrounding agroecosystems. Nutrient uptake and
removal by soil and vegetation in the riparian forest ecosystem prevented outputs from agricultural uplands
from reaching the stream channel. The riparian ecosystem can apparently serve as both a short- and long-
term nutrient filter and sink if trees are harvested periodically to ensure a net uptake of nutrients.

Lowrance, RR and HB Pionke, 1989. Transformation and movement of nitrate in
aquifer systems. In Follett RF (ed.). Nitrogen Management and Ground Water
Protection. Elsevier, New York.

Draft... 9/14/98


Reviews processes leading to changes in N03 concentration in groundwater and discusses case studies.In
general, N03 reaching shallow aquifer systems, especially those with rapidly fluctuating water tables, has a
good chance for removal by denitrification or uptake by deeply rooted vegetation. Drawdown of a deeper
unconfined aquifer can cause inflow of water with higher N03 concentrations from a shallower aquifer.

Magette, W.L. et al. 1983. Wastewater Treatment in Soil as a Function of Residence
Time in the Root Zone. Bulletin 137, Virginia Water Resources Research Center,
Virginia Polytechnic Institute and State University, Blacksburg, Virginia.

A laboratory study was conducted to determine nitrogen removal rates from a land-applied wastewater as a
function of the length of time the wastewater remained in the root zone. A digital simulation model was
used as an aid in describing soil-water (and wastewater) movement through the root zone under wet
conditions (i.e., root zone 50-75 percent saturated). A procedure was developed to predict the rate and
volume of drainage as a function of initial soil moisture content, amount of liquid applied, and time after
liquid application. An exact relationship between nitrogen removals and wastewater residence time in the
root zone could not be developed. However, removals of up to 95 percent of applied NH4+-N were
observed in an 18-cm-deep root zone when residence times were as short as 2 hr.

Meals, D.W. 1993. Assessing nonpoint phosphorus control in the LaPlatte River
Watershed. Lakes and Reservoir Management. 7:197-207.

Phosphorus loading from agricultural activities such as manure and fertilizer applications often contributes
to eutrophication of surface waters. The primary goal of the LaPlatte River Watershed Project in
northwestern Vermont was to reduce phosphorus loading from farmland through implementation of best
management practices (BMPs). Eleven years of monitoring did not show a dramatic decrease in
phosphorus concentration or load from the watershed. However, analysis controlling for hydrologic
variability suggested significant decreases in phosphorus load from some subwatersheds following
completion of the land treatment program. Post-BMP phosphorus load reductions of 26-44% (0.01-0.14
kg/ha/yr) were estimated using a paired regression technique that accounts for discharge differences
between years. Phosphorus export was reduced under most circumstances, except under the highest runoff
conditions, suggesting that the capacity of the land treatment system to control phosphorus may have been
exceeded occasionally. Observed phosphorus reductions in treated watersheds appeared to be related to the
degree of treatment afte a minimum threshold level of land treatment had been achieved.

Miller, WL, 1992. Hydrogeology and Migration of Septic Tank Effluent in the Surficial
Aquifer System in the Northern Midlands Area, Palm Beach County, Florida. US
Geological Survey Water Resources Investigations Report 91-4175. Tallahassee, Florida.

The northern Midlands area in Palm Beach County is an area of expected residential growth, but its flat
topography, poor drainage, and near-surface marl layers retard rainfall infiltration and cause frequent
flooding. Tests at three septic tank sites showed traces of effluent in groundwater (38-92 feet from the
septic tank outlets) and that near-surface marl layers greatly impede the downward migration of the effluent
in the surficial aquifer system throughout the northern midlands.

Northwest Florida Water Management District, February, 1994. Non-point Source
Assessment: Deer Point Lake Watershed. Water Resources Special Report 93-6.
Havana, Florida.

Evaluation of existing and potential pollution contributions to Deer Point Lake from various non-point
sources. Estimated pollutant-loading rates from non-point sources. Ranked water quality parameter
loading and calculated cumulative rank indices to provide an overall indication of potential water quality
impacts. Methodology integrated GIS, satellite imagery, and land use/cover maps to model existing

Draft... 9/14/98

development patterns. The project methodology was intended to provide a framework for the development
and implementation of pollution load reduction goals, total maximum daily loads, best management
practices, land development regulations, land preservation and acquisition water quality protection, and
watershed management goals.

Pederson, JK, PL Bjerg, and TH Christensen, 1991. Correlation of nitrate profiles with
groundwater and sediment characteristics in a shallow sandy aquifer. Journal of
hydrology, vol. 124: 263-277.

Develops a characterization of sediment profiles according to total reduction capacity (TRC), that
quantifies the total amount of reduced compounds in aquifer material. A distinct increase of TRC was
observed at and below the oxidation-reduction front. The TRC profiles showed that some reduction
capacity is still present above the oxidation-reduction front and is apparently able to support denitrification.

Peterson, GH and AD McLaren (eds.), 1967. Soil Biochemistry. Marcel Dekker Inc.
New York.

Text covering soil chemical dynamics.

Pitt, R. and M. Bozeman. 1982. Sources of Urban Runoff Pollution and Its Effects on an
Urban Creek. United States Environmental Protection Agency. EPA-600/S2-82-090.

Sources and impacts of urban runoff were studied for the Coyote Creek near San Jose, California. The 3-
year monitoring study included three tasks: (1) identifying and describing important sources of urban
runoff pollutants; (2) describing the effects of those pollutants on water and sediment quality, aquatic
organisms, and associated beneficial uses of the creek; and (3) assessing potential measures for controlling
the problem pollutants in urban runoff. Results indicated that various urban runoff constituents (especially
organic and heavy metals) may be responsible for many of the adverse biological conditions observed in
Coyote Creek. But adequate control of pollutants would require extremely high removals that would be
difficult as well as costly to achieve.

Postma, FB, AJ Gold, and GW Loomis, 1992. Nutrient and microbial movement from
seasonally-used septic systems. Journal of Environmental Health, vol. 55, no. 2.

Seasonal occupancy may promote the transmission of contaminants to groundwater due to incomplete
formation of a biological clogging mat in soil adsorbtion systems. Groundwater surrounding three
seasonally-used septic systems was monitored to determine the movement and attenuation of nitrogen,
phosphorous, fecal coliforms, and Clostridiumperfringens. The septic systems showed inadequate
attenuation of these parameters. Biological clogging mats were not found when the systems were evaluated
at the end of summer occupancy. Siting seasonally-used shoreline septic systems may require improved
effluent distribution to achieve wastewater renovation.

Puckett, L.J. 1995. Identifying the major sources of nutrient water pollution.
Environmental Science and Technology. 29:408A-14A.

Atmospheric nitrogen may be a very important source of nutrient contamination. It is generally not enough
to know the land use alone, the type and intensity of that use are equally important.

Reddy, Konda Rameshwer, 1976. Nitrification denitrification reactions in flooded soil.
Ph.D. dissertation, Louisiana State University and Agricultural and Mechanical College.

Draft... 9/14/98


Research results and review of nitrogen dynamics in soil and groundwater under fluctuating hydrologic

Reneau et al, 1989. Fate and Transport of Biological and Inorganic Contaminants.
Journal of Environmental Quality, vol. 18, Apr-Jun 1989. Review of contaminant
dynamics in the subsurface environment especially as related to on-site disposal

Denitrification may be significant in soils with restricted drainage if conditions are also adequate for
nitrification of NH4+ before it reaches the anaerobic zone and if an adequate C source is available.
Residual soil organic matter is probably not a satisfactory long- term energy source for denitrification.
Data indicates that in well-drained soils where nitrification occurs immediately below the OSDS,
denitrification does not adequately remove N03- and the most probable mechanism for reducing N
concentrations is dilution.

Reynoldson, T.B. Jr., and H.R. Hamilton. 1982. Spatial heterogeneity in whole lake
sediments towards a loading estimate. Hydrobiologia 91:235-240.

Studies of nutrient loadings, to shallow culturally eutrophied Alberta lakes, suggest internal inputs are
significant. In this regard, estimation of bottom sediment P loads to Lake Wabamun were examined.

Riekerk, H and Korhnak, LV, 1992. Rainfall and Runoff Chemistry of Florida
PineFlatwoods. Water, Air, and Soil Pollution 65: 59-68.

Rainfall chemistry was monitored for ten years, and correlated with runoff chemistry of an undisturbed 140
ha pine-cypress flatwoods watershed in Florida. The seasonal variation in pH showed a minimum during
the summer months. Levels ofnitrate-N and phosphate-P showed a significant decrease while Ca showed a
significant increase in runoff over time. The pine-cypress flatwoods ecosystem appeared to absorb most of
the acidity N, and P while losing only a little of the bases. Florida's sandhill lakes are poorly buffered and
sensitive to acidification by rainwater of pH 4.7 or less, but forest lands are less vulnerable as the tree
canopy and soil offers better buffering capacity, and the low productivity becomes improved by
atmospheric nutrient inputs.

Robertson, WD, JA Cherry, and EA Sudicky, 1991. Ground-water contamination from
two small septic systems on sand aquifers. Ground Water, vol. 29, no. 1.

Study conducted at two single-family homes located on shallow unconfined sand aquifers in Ontario. As a
result of low transverse dispersion in the aquifer, mobile plume solutes such as nitrate and sodium occurred
at more than 50 percent of the source concentrations 130 m downgradient from the septic system. Almost
complete nitrate attenuation was observed within the last 2 m of the plume flowpath before discharge to an
adjacent river. This was attributed to denitrification occurring within organic matter-enriched riverbed
sediments. Concluded that, for many unconfined sand aquifers, the minimum distance-to-well regulations
for permitting septic systems in most parts of North America should not be expected to be adequately
protective of well-water quality in situations where contaminants are not attenuated by chemical or
microbiological processes.

Rosswall, T, 1982. Microbiological regulation of the biogeochemical nitrogen cycle.
Plant and Soil, vol. 67: 15-34.

Discusses some aspects of the individual microbiological processes in the nitrogen cycle and their
importance for an efficient management of agroecosystems. The influence of abiotic factors such as

Draft... 9/14/98

oxygen concentration, inorganic nitrogen concentration, and pH is discussed in relation to the different

Sawhney, B.L. 1977. Predicting phosphate movement through soil columns. Journal of
Environmental Quality. 6:86-89.

To assess the potential pollution of ground water with P from septic tank drainfields, sorption capacities of
various soils were determined over an extended period of time and related to P movement through soil
columns using solutions having P concentrations similar to waste waters. The amounts ofP sorbed by fine
sandy loam (fsl) and silt loam (sil) soil columns before breakthrough occurred were approximately equal to
the sorption capacities determined from isotherms obtained over a sufficiently long reaction time of about
200 hours. In Merrimac fsl, breakthrough occurred after about 50 pore volumes of waste water had passed
through the column while about 100 pore volumes passed through Buxton sil before the breakthrough
occurred. Following breakthrough, concentration ofP in the effluent continued to increase and approached
the influent concentration after several hundred pore volumes of effluent had passed through the columns.
The results, thus, suggest that while most deep soils should effectively remove P from waste water, ground
water under drainfields installed in soils of low P sorption capacity after prolonged use may contain
undesirably large concentrations of P.

Starr, JL and BL Sawhney, 1980. Movement of nitrogen and carbon from a septic system
drainfield. Water, Air, and Soil Pollution, vol. 13: 113-123.

A two year study of vertical and horizontal movement of nitrogen and carbon from a septic system
drainfield in Connecticut. Concluded that 20 to 25% of total effluent nitrogen will be mineralized, even in
unusually wet years. The nitrogen will be nitrified and will move with infiltrating water to the
groundwater. Calculated that under continuous use and with 5-households / ha, the nitrate transported to
groundwater aquifers below the sandy, well drained septic systems is about 35 kg / ha.

Stephens, D.W., B. Waddell, and J.B. Miller. 1988. Reconnaissance Investigation of
Water Quality, Bottom Sediment, and Biota Associated with Irrigation Drainage in the
Middle Green River Basin, Utah, 1986-1987.

U.S. Geological Survey Water-Resource Investigations Report 88-4011. Salt Lake City, Utah.
Reconnaissance of wildlife areas in the middle Green River basin of Utah was conducted during 1986 and
1987 to determine whether irrigation drainage has caused, or has the potential to cause significant harmful
effects on human health, fish, and wildlife, or may adversely affect the suitability of water for beneficial
uses. Studies at Stewart Lake Waterfowl Management Area and Ouray National Wildlife Refuge indicated
that concentrations of boron, selenium, and zinc in water, bottom sediment, and biological tissue were
sufficiently large to be harmful to fish and wildlife, and to adversely affect pesticides in surface water
generally were less than established standards with the exception of gross alpha radiation, which exceeded
by factors of three to five times the standard of 15 picocuries per liter in water from two of the drains
discharging into Stewart Lake.

Sweets, P Roger, 1992. Diatom Paleolimnological Evidence for Lake Acidification in
the Trail Ridge Region of Florida. Water, Air, and Soil Pollution 65: 43-57.

Paper reviews the basic data on north Florida lakes from PIRLA I (Paleoecological Investigation of Recent
Lake Acidification) and adds data inferred from Florida Panhandle lakes analyzed during PIRLA I. Of the
lakes determined to have acidified, there is evidence for the existence of natural acidification processes.
Hydrological processes may play a large role in determining which lakes and areas are susceptible to
anthropogenic acidification. Climatic differences are a likely exacerbating factor. Individual lakes are
being affected according to amounts of acidic deposition, the number of years deposition has been received,
soil types, bedrock type, flushing rates, and assorted limnological variables such as productivity, DOC, and
competing anthropogenic influences. Past lakewater chemistries were inferred from dated cores. Such

Draft... 9/14/98


studies have provided strong evidence for acidification resulting from acidic deposition in a small subset of
lakes from various regions of the US. This subset is defined geographically by different acidic deposition
rates and bedrock types, and by the many limnological factors that contribute to lake pH and buffering
capacity. Lakes in the Panhandle did not show evidence of recent anthropogenic acidification though
only a few lakes were studied.

Tippett, John. 1993. Linking land use to water quality. Water Environment and
Technology. 5:17+.

A GIS was used to determine land use patterns of five riparian buffer zones at increasing distances from a
stream. Multiple linear regression was used to relate soluble reactive phosphorus and nitrate-nitrogen at
each sampling station to watershed area and land use patterns in each buffer zone. Four seasonal equations
were developed for each of the five buffer zones.

Trudell, MR, RW Gillham, and JA Cherry, 1986. An in-situ study of the occurrence and
rate of denitrification in a shallow unconfined sand aquifer. Journal of Hydrology, vol.

An in-situ injection experiment was conducted using a specially designed injection-withdrawal-sampling
drive point. Nitrate and a conservative tracer (bromide) were added to natural groundwater and injected at
3m depth into a shallow, unconfined aquifer. The relative changes in concentration over time were
observed. The organic carbon source required for denitrification is either dissolved organic carbon or soil
organic carbon. Soil organic carbon, at 0.08-0.16% by weight, is adequate to denitrify large amounts of
nitrate. The measured rate ofdenitrification ranged from 0.0078 to 0.13 g NO3-N/m3/hr.

Tsai, Yuong-How, 1989. Factors Affecting Denitrification Kinetics in Selected Florida
Soils. Master of Science Thesis. Gainesville, Florida.

Study of denitrification rates under anaerobic conditions for 24 soil horizons. The soil horizons differed
widely in pH and organic carbon content. Soil horizons were selected from six soil profiles representing
several poorly-drained soil series in Florida. Denitrification rates were determined by measuring the rate of
nitrate disappearance or the rate of nitrous oxide production. Linear correlation analysis showed that zero-
order denitrification values were significantly related to total or soluble organic carbon content and with
nitrate-nitrogen content. High denitrification rates were obtained in surface horizons. In subsoil horizons,
denitrification rates were relatively low due to low organic matter content, low microbial population, and
pH induced toxicities (Possibly high aluminum concentrations).

US Environmental Protection Agency, 1977. Alternatives for Small
WastewaterTreatment Systems: On-Site Disposal / Septage Treatment and Disposal.
EPA Technology Transfer Seminar Publication.

Identification of community needs and suitability for non-central wastewater facilities. Description of
wastewater characteristics and soil treatment capabilities. Estimation of soil infiltrative and percolative
capacities. Summary of on-site treatment and disposal system alternatives and alternative selection.
Overview of septage generation, treatment, and disposal.

Draft... 9/14/98

USGS, 1994. Nitrate in Ground Water and Spring Water Near Four Dairy Farms in
North Florida, 1990-93. Water Resources Investigations Report 94-4162. Tallahassee,

Investigation of nitrate in groundwater. Analysis of groundwater parameters as indication of type and
extent of microbiological activity in aquifer particularly denitrification.

USGS, 1996. Hydrogeologic Investigation and Simulation of Ground-Water Flow in the
Upper Floridan Aquifer of North-Central Florida and Southwestern Georgia and
Delineation of Contributing Areas for Selected City of Tallahassee, Florida, Water
Supply Wells. USGS Water-Resources Investigations Report 95-4296. Tallahassee,

Results of an investigation of the part of the Upper Floridan aquifer that underlies Tallahassee and Leon
County, Florida, and the surrounding counties in North-Central Florida and southwestern Georgia.
Previously collected hydrogeologic data was used in conjunction with a computer model to characterize
ground-water flow in the study area. Computer simulation was performed using two USGS software
packages: MODFLOW a modular three-dimensional finite-difference ground-water flow model and
MODPATH a particle-tracking program. Contributing areas for five City of Tallahassee water-supply
wells were delineated. Computer simulation results were compared to analytical methods.

USGS, 1996. National Water Quality Assessment of the Georgia-Florida Coastal Plain
Study Unit Water Withdrawals and Treated Wastewater Discharges, 1990. Water-
Resources Investigations Report 95-4084. Tallahassee, Florida.

Compilation of data by county on water use in the Georgia-Florida Coastal Plain Study Unit.

USGS. Relation of Nitrate Concentrations in Ground Water to Well Depth, Well Use,
and Land Use in Franklin Township, Gloucester County, New Jersey, 1970-85. Water
Resources Investigations Report 94-4174. USGS. West Trenton, New Jersey.

Research report on nitrate dynamics in an unconfined aquifer system in New Jersey. Primary sources of
nitrogen were leachate from on-site disposal systems, runoff from animal feedlots, and leachate from
nitrogen fertilizers. Nitrate concentration increased with the percentage of developed land in a well's
buffer zone. Nitrate concentrations tend to decrease with well depth. Deep wells contain older water -
nitrate concentrations are more likely to represent past rather than present use.

Vighi, M. 1991. Phosphorus loads from selected watersheds in the drainage area of the
Northern Adriatic Sea. Journal of Environmental Quality. 20:439-44.

The Po Valley is one of the most productive agricultural areas in Europe and P losses from fertilizers are
often accused of being among the main factors responsible for eutrophication of the Northern Adriatic Sea.
to quantify nonpoint phosphorus loads in this area, 15 small watersheds were studied. Thirteen watersheds
were in the intensive agricultural area near the coast and two watersheds were in the forested mountains.
Land use in the watersheds was carefully examined and P loads from various sources were theoretically
evaluated and experimentally measured. The results indicate fertilization does not increase the losses of P
through leaching from the coastal soils, where the measured release were in the range 0.03 to 0.21 kg-
P/ha/yr with a mean value of about 0.1 kg-P/ha/yr. There is, however, a greater loss of P through soil
erosion from the mountain watersheds (0.6 kg/ha/yr). It can be concluded that the control of point sources

Draft... 9/14/98

must take priority over nonpoint sources in efforts to reduce accelerated eutrophication of the Northern
Adriatic Sea.

Walker, WG, J Bouma, DR Keeney, and PG Olcott, 1973. Nitrogen tansformations
during subsurface disposal of septic tank effluent in sands: II Ground water quality.
Journal of Environmental Quality, vol. 2, no.4: 521-525.

Groundwater samples were analyzed to establish patterns of nitrogen enrichment in the groundwater around
seepage beds and to evaluate system performance in sands in terms of nitrogen removal. The data obtained
suggested that, in sands, the only active mechanism of lowering the nitrate content is by dilution with
uncontaminated groundwater possibly requiring relatively large areas or low septic tank densities.

Waller, BG, B Howie, and CR Causaras, 1987. Effluent Migration from Septic Tank
Systems in Two Different Lithologies, Broward County, Florida. US Geological Survey
Water Resources Investigations report 87-4075. Tallahassee, Florida.

Two septic tank test sites, one in sand and one in limestone, were analyzed for effluent migration.
Monitoring wells were sampled for a 16 month period. Results were graphically presented. Variances in
hydrologic regimes posed difficulties to generalizations of results. Dilution appeared to be the main factor
in reduction of nitrate concentrations.

Weier, KL and JW Gillham, 1986. Effect of acidity on denitrification and nitrous oxide
evolution from Atlantic Coastal Plain soils. Soil Science Society of America Journal,
vol. 50:1202-1205.

The effect of acidity on denitrification and nitrous oxide production in six soils from the Atlantic Coastal
Plain was estimated using laboratory incubations of flooded soil for periods up to 21 days. Increased
denitrification effects were associated with a decrease in cidity in all soils most of the effect occurred
above pH = 6.5. Nitrous oxide evolved increased with increasing acidity with a maximum at pH <= 5.8.
Atlantic coastal Plain soils generally have a pH < 5.8.

Whigham, D.F. et al. 1988. Impacts of Freshwater Wetlands on Water Quality: A
Landscape Perspective. Environmental Management 12(5):663-671.

Suggest that a landscape approach might be useful in evaluating the effects of cumulative impacts on
freshwater wetlands. The reason for using this approach is that most watersheds contain more than one
wetland, and effects on water quality depend on the types of wetlands and their position in the landscape.
Riparian areas that border uplands appear to be important sites for nitrogen processing and retention of
large sediment particles. Fine particles associated with high concentrations of phosphorus are retained in
downstream wetlands, where flow rates are slowed and where the surface water passes through plant litter.
Riverine systems also may play an important role in processing nutrients, primarily during flooding events.
Lacustrine wetlands appear to have the least impact on water quality, due to the small ratio of vegetated
surface to open water. Examples are given of changes that occurred when the hydrology of a Maryland
floodplain was altered.

Yates, MV, 1985. Septic Tank Density and Ground-Water Contamination. Ground
Water, vol. 23, no. 5.

Reviews literature regarding septic tank density and waterborne disease in several states. Concludes that
the single most important means of limiting groundwater contamination by septic tanks is to restrict the
density of these systems in an area.

Draft... 9/14/98


Adamus, C.L. and M.J. Bergman. 1995. Estimating Nonpoint Source Pollution Loads
with a GIS Screening Model. Water Resource Bulletin, AWWA 31(4):647-655.

The St. Johns River Water Management District (SJRWMD) is using a Geographic Information System
(GIS) screening model to estimate annual nonpoint source pollution loads to surface waters and determine
nonpoint source pollution problem areas within the SJRWMD. The model is a significant improvement
over current practice because it is contained entirely within the district's GIS software, resulting in greater
flexibility and efficiency, and useful visualization capabilities. Model inputs consist of five spatial data
layers, runoffcoefficients, mean runoff concentrations, and stormwater treatment efficiencies. The spatial
data layers are: existing land use, future land use, soils, rainfall, and hydrologic boundaries. These data
layers are processed using the analytical capabilities of a cell-based GIS. Model output consists of seven
spatial data layers: runoff, total nitrogen, total phosphorous, suspended solids, biochemical oxygen
demand, lead, and zinc. Model output can be examined visually or summarized numerically by drainage
basin. Results are reported for only one of the SJRWMD's ten major drainage basins, the lower St. Johns
River basin. The model was created to serve a major planning effort at the SJRWMD; results are being
actively used to address nonpoint source pollution problems.

Bacon, PE, 1995. Nitrogen Fertilization in the Environment. Marcel-Dekker, Inc. New

Agriculturally-oriented reference text that reviews the interactions between nitrogen and the ecosystem and
presents simulation models.

Caussade, B. and M. Pratt, 1990. Transport modelling in watersheds. Ecological
Modelling, 52: 135-179.

Reviews modeling concepts and methodologies as related to watershed-scale modeling. Points out model
requirements for various temporal and spatial scales. Includes a model review.

CDM, 199??. Master Stormwater Management Plan, City of Tallahassee, Florida.
Camp, Dresser, and McKee, Tallahassee, FL.

Cushing, C.E. et. al. 1983. Relationships among chemical, physical, and biological
indices along river continue based on multivariate analyses. Arch. Hydrobiol. 98(3):317-

A variety of multivariate analyses were applied to chemical, physical, and biological data from 16 stream
sites to explore the usefulness of these factors in possible stream classification systems and to test
hypotheses of the River Continuum Concept.

DeVantier, B.A., and A.D. Feldman. 1993. Review of GIS applications in hydrologic
modeling. Journal of Water Resources Planning and Management. 119:246-261.

Geographic information systems (GIS) provide a digital representation of watershed characteristics used in
hydrologic modeling. This paper summarizes past efforts and current trends in using digital terrain models
and GIS to perform hydrologic analyses. There methods of geographic information storage are discussed:
raster or grid, triangulated irregular network, and contour-based line networks. The computational,
geographic, and hydrologic aspects of each data-storage method are analyzed. The us of remotely sensed

Draft... 9/14/98

data in GIS and hydrologic modeling is reviewed. Lumped parameter, physic-based, and hybrid
approaches to hydrologic modeling are discussed with respect to their geographic data inputs. Finally,
several applications areas (e.g., floodplain hydrology, and erosion prediction) for GIS hydrology are

Domenico, PA and FW Schwartz, 1998. Physical and Chemical Hydrogeology. John
Wiley and Sons, Inc. New York.

Hydrological reference text. Discussions of fundamentals of groundwater flow, contaminant transport,
remediation, risk assessment, and modeling.

Donigian, A.S. and W.C. Huber. 1991. Modeling of Nonpoint Source Water Quality in
Urban and Non-Urban Areas. EPA/600/3-91/039.

Nonpoint source assessment procedures and modeling techniques are reviewed and discussed for both
urban and non-urban land areas. Detailed reviews of specific methodologies and models are presented,
along with overview discussions focusing on urban methods and models, and on non-urban (primarily
agricultural) methods and models. Simple procedures, such as constant concentration, regression,
statistical, and loading function approaches are described, along with complex models such as SWMM,
HSPF, STORM, CREAMS, SWRRB, and others. Brief case studies of ongoing and recently completed
modeling efforts are described. Recommendations for nonpoint runoff quality modeling are presented to
elucidate expected directions of future modeling effort.

Eckersten, H -Gardenas Jansson, 1992. Modelling seasonal nitrogen, carbon, water
and heat dynamics of the Solling spruce stand. Ecological Modelling 83: 119 -129.

The authors coupled two mechanistic models: SOIL and SOILN to generate a more comprehensive
ecological model. The driving variables in the SOIL model were air temperature, precipitation, wind
speed, vapor pressure, and global radiation all of which were obtained from the nearest meteorological
station. SOIL also considered soil water tension. The driving variables in SOILN were soil water
flows/contents, soil temperature, and the ratio between actual and potential transpiration. Problems were
encountered in model calibration.

ESRI, 1992. Cell-based modeling with GRID. Environmental Systems Research
Institute, Inc., Redlands, California.

User manual for GRID application.

Follett, RF (ed.), 1989. Nitrogen Management and Ground Water Protection. Elsevier,
New York.

Text that address nitrogen and groundwater concerns from an agricultural perspective. Contains chapters
reviewing nitrogen transport mechanisms, groundwater concerns, fertilizer management, and modeling
methodologies and concerns.

Frimpter, MH, JJ Donohue, IV, and MV Rapacz, 1990. A mass-balance model for
predicting nitrate in ground water. New England Water Works Association, vol. 54, no.

Development of a model to be used for prediction of nitrate concentrations in public-supply wells under
steady-state conditions. Predicted concentrations are derived by calculating the concentration that results

Draft... 9/14/98 1-25

from the total weight of nitrogen and total volume of water entering the zone of contribution to the well.
Not a spatial model.

Grayson, RB., I.D. Moore, and T.A. McMahon. 1992. Physically based hydrologic
modeling 2. Is the concept realistic? Water Resources Research. 26:2659-2666.

Future directions for physically based, distributed-parameter models intended for use as hydrologic
components of sediment and nutrient transport models are discussed. The attraction of these models is their
potential to provide information about the flow characteristics at points within catchments, but current
representations in process-based models are often too crude to enable accurate, a priori application to
predictive problems. The difficulties relate to both the perception of model capabilities and the
fundamental assumptions and algorithms used in the models. In addition, the scale of measurement for
many parameters is often not compatible with their use in hydrological models. The most appropriate uses
of process-based, distributed-parameter models are to assist in the analysis of data, to test hypotheses in
conjunction with field studies, to improve our understanding of processes and their interactions and to
identify areas of poor understanding in our process descriptions. The misperception that model complexity
is positively correlated with confidence in the results is exacerbated by the lack of full and frank discussion
of a model's capability/limitations and reticence to publish poor results. This may ultimately diminish the
opportunity to advance understanding of natural processes because the managers of research resources are
given the impression that the answers are already known and are being provided by models. Model
development is often not carried out in conjunction with field programs designed to test complex models,
so the link with reality is lost.

Hantzsche, NN, and EJ Finnemore, 1992. Predicting Ground-Water Nitrate-Nitrogen
Impacts. Ground Water vol. 30, no. 4.

Review of literature concerning the contribution and fate of nitrogen beneath septic tank disposal fields.
Simplified methods are developed for estimating long-term groundwater nitrate increases on an area-wide
basis. Predicted values are compared with actual monitoring data for three California communities to
verify the reasonableness of the suggested methods. The simple model does not account for spatial

Harper, Harvey H., 1992. Estimation of Stormwater Loading Rate Parameters for Central
and South Florida. Environmental Research and Design, Inc., Orlando, Florida.

Provides loading and concentration data for various landuses in Florida.

Hart, R.L. ed. 1993. Management Guidelines and Goals for the Myakka River Basin.
Florida Department of Environmental Regulation, Office of Coastal Management.

The objective of the Myakka River Basin Project was to provide a technical basis for management goals
and recommendations that would protect the natural resources of the Myakka River and its estuary,
Charlotte Harbor. The major task for the third year was to conduct a geographic information system (GIS)
analysis and use the results to develop management goals and recommendations. The Myakka River drains
a watershed of approximately 1,559 km^2. Much of the watershed consists of rural uses and publicly
owned lands. Water quality of the river is generally good. However, population growth projections for the
region and concern over the potential environmental impacts associated with growth require planning to
protect river resources from future degradation. A GIS-based computer model was developed to illustrate
how models can be used for projecting the runoff and chemical loadings to the Myakka River as a result of
changes in land use. Three scenarios were modeled for a subbasin which is generally undeveloped at this
time. The scenarios included urban development as projected by the County Planning Department with no
preservation areas; urban development with preservation of wetland and hammock areas; and urban
development with preservation of wetland, hammocks, and a 220-foot shoreline buffer.

Draft... 9/14/98

He, C., J.F. Riggs, and Y. Kang. 1993. Integration of Geographic Information Systems
and a Computer Model to Evaluate Impacts of Agricultural Runoff on Water Quality.
Water Resources bulletin, AWRA. 29(6):891-900.

This study integrates an Agricultural Non-Point source Pollution Model (AGNPS), the Geographic
Resource Analysis Support System (GRASS) and GRASS WATERWORKS (a hydrologic modeling tool
box being developed at the Michigan State University Center for Remote Sensing) to evaluate the impact of
agricultural runoff on water quality in the Cass River, a subwatershed of Saginaw Bay. AGNPS is used to
estimate the amounts, origin, and distribution of sediment, nitrogen (N), and phosphorus (P) in the
watershed. GRASS and GRASS WATERWORKS are used to generate parameters needed for AGNPS
from digital maps, which include soil association, land use, watershed boundaries, water features, and
digital elevation. Outputs of the model include spatially distributed estimates of volume and peak runoff,
overland and channel erosion, sediment yields, and concentrations of nitrogen and phosphorus.
Management scenarios are explored in the AGNPS model to minimize sedimentation and nutrient loading.
Scenarios evaluated include variations in crop cover, tillage methods, and other agricultural management
practices. In addition, areas vulnerable to erosion are identified for best management practices.

Heidtke, T.M. and M.T. Auer. 1993. Application of a GIS-based Nonpoint Source
Nutrient Loading Model for Assessment of Land Development Scenarios and Water
Quality in Owasco Lake, New York. Water Science and Technology. 28(3-5):595-604.

The magnitude and water quality implications of nonpoint source phosphorus loadings to Owasco Lake
(New York) are evaluated through the application of a methodology which links geographic characteristics,
long-term average runoff loads and a set of critical lakeside water quality response parameters. The
approach utilizes the Universal Soil Loss Equation together with empirical loading functions to derive
representative phosphorus export coefficients for the local drainage system. Cumulative loadings from
individual subbasins within the watershed serve as input to a simple water quality model of Owasco Lake,
showing the expected lake response in terms of average total phosphorus concentration, trophic state, water
transparency, and minimum hypolimnetic dissolved oxygen concentration. The methodology facilitates
easy andrapid assessment of general watershed management and development scenarios of interest. A
unique aspect of the approach is its dependence upon descriptive data supplied by a Geographic
Information System (GIS) to establish the coincidence of specific land use, soil texture and surface slope
attributes within each of the hydrologic sub-basins comprising the overall watershed. The GIS-generated
attribute matrices provide a much more accurate depiction of critical geographic characteristics known to
impact nonpoint source runoff loadings, thereby improving the reliability of current and projected
phosphorus loads to Owasco Lake.

James, C.R., et al. 1995. Vulnerability Zone Identification: A Watershed Management
Tool for Protecting Reservoir Water Quality. Proc. 22 Annual Confr. Integr. Water Res.
Plan 21 Century. ASCE. p.293-296.

As part of an extensive watershed management project, an approach was developed for integrating the
potential impacts of watershed physical characteristics on raw water quality. This approach uses the
capabilities of a geographic information system (GIS) database to produce water quality vulnerability zone
maps which, in turn, can be used to develop watershed management strategies. Therefore, these
vulnerability zone maps can be used by utilities to proactively manage the watersheds so that source
protection becomes an effective first barrier to water quality degradation and to ultimately improve the
water quality. An overview of the conceived approach used to produce the water quality vulnerability zone
maps is documented here.

Jeton, A.E., and J.L. Smith. 1993. Development of Watershed Models for Two Sierra
Nevada Basins Using a Geographic Information System. Water Resources Bulletin.

Draft.. 9/14/98

Techniques were developed using vector and raster data in a geographic information system (GIS) to define
the spatial variability of watershed characteristics in the north-central Sierra Nevada of California and
Nevada and to assist in computing model input parameters. The U.S. Geological Survey's Precipitation-
Runoff Modeling System, a physically based, distributed-parameter watershed model, simulates runofffor
a basin by partitioning a watershed into areas that each have a homogeneous hydrologic response to
precipitation or snowmelt. These land units, known as hydrologic-response units (HRU's), are
characterized according to physical properties, such as altitude, slope, aspect, a land cover, soils, and
geology, and climate patterns. Digital data were used to develop a GIS data base and HRU classification
for the American River and Carson River basins. The following criteria are used in delineating HRU's: (1)
Data layers are hydrologically significant and have a resolution appropriate to the watershed's natural
spatial variability, (2) the technique for delineating HRU's accommodates different classification criteria
and is reproducible, and (3) HRU's are not limited by hydrographic-subbasin boundaries. HRU's so
defined are spatially noncontiguous. The result is an objective, efficient methodology for characterizing a
watershed and for delineating HRU's. Also, digital data can be analyzed and transformed to assist in
defining parameters and in calibrating the model.

Kalkhoff, S.J. 1993. Using a Geographic Information System to Determine the Relation
Between stream Quality and Geology in the Roberts Creek Watershed, Clayton County,
Iowa. Water Resources Bulletin. 29(6):989-996.

A geographic information system (GIS) was used to determine the relation between the stream-water
quality and underlying geology in Roberts Creek watershed, Clayton County, Iowa, for base-flow
conditions during the spring and summer of 1988-90. Geologic, stream, basin and subbasin boundaries,
and water quality sampling-site coverages were created by digitizing available maps. A contour coverage
was created from digital linegraph data. The areal extent of geologic units subcropping in each subbasin
was quantified with GIS, and the results then were output and joined with the discharge and water-quality
data for statistical analyses. Illustrations showing the geology of the study area and the results of the study
were prepared using GIS. By using GIS and a statistical software package, a weak but statistically
significant relation was found between the water temperature, pH, and nitrogen concentrations in Roberts
Creek and the underlying geology during base-flow conditions.

Lahlou, M. et al. May 1996. Better Assessment Science Integrating Point and Nonpoint
Sources (BASINS) Version 1.0 Users Manual. Office of Water (4305) EPA-823-R-96-

The EPA BASINS model is a watershed modeling package designed to integrate real data, and simulations
with a spatial GIS framework. The model can be broken into two portions. The first is the
Assessment/Planning Module which allows quick evaluation of selected areas, organize information and
display results. Three scales of analysis can be performed including a Target or regional analysis, Assess at
the watershed level, and Data Mining located at the station level. The second portion is the Modeling
Module. This examines impacts of pollutant loadings from point and non-point sources. Three different
models are contained within the Modeling Module. The NPSM (non-point source model) simulates
nonpoint source runoff, pollution loadings and dissolved oxygen levels in runoff for selected runoff.
QUAL2E is a I-D steady-state water quality and eutrophication model that allows fate and transport
modeling for point and nonpoint source loadings. TOXIROUTE is a screening level stream routing model
that performs simple dilution/decay calculations under mean or low-flow conditions for a stream system
within a given watershed.

Levine, D.A., and W.W. Jones. 1990. Modeling phosphorus loading to three Indiana
reservoirs: a geographic information system approach. Lake and Reservoir
Management. 6:81-91.

Draft... 9/14/98 1-28

This paper describes a geographic information system (GIS) approach to modeling the effects of distance to
water and slope angle on external phosphorus loading to lakes. A raster GIS database was created that
included land use, topography, soils, and watershed boundaries for three Indiana reservoirs (Lakes
Kickapoo, Lenape, and Shakamak). Three sets of phosphorus export coefficients were selected and
assigned to each cell according to land use. Linear filters were designed and applied to the export
coefficients such that areas nearest the water and with the steepest slopes would contribute the greatest
amount of phosphorus relative to the initial phosphorus export coefficient. These filtered coefficients were
used to calculate aerial phosphorus loading and in-lake phosphorus concentrations. The predicted
concentrations were within 5 ug/L of the observed phosphorus concentration in Lake Shakamak for three
modeling scenarios, and within 22 ug/L of the observed concentration in Lake Lenape for three scenarios.
Predictions of phosphorus concentrations in Lake Kickapoo were consistently low (35 percent to 95
percent). This may have resulted from the complex hydrology of Lake Kickapoo or the inability to
accurately model the physical processes as intended. The GIS system was useful for modeling the effects
of distance and slope on phosphorus loading and for providing data highlighting critical management areas.

McElroy, A.D. et. al., 1976. Loading Functions for Assessment of Water Pollution from
Nonpoint Sources. EPA-600/2-76-151, U.S. Environmental Protection Agency,
Washington, D.C., USA.

This is a user's handbook that provides two basic functions. First, it presents loading functions together
with the methodologies for their use. Second, it presents some of the needed data, provides references to
other sources of data, and suggests approaches for generation of data when available data ane inadequate.
A corollary function consists of assessments of the adequacies of functions and their supporting inventories
of data, and an assessment as well of the extent to which pollutants and nonpoint sources are adequately

Mehran, M, J Noorishad, and KK Tanji, 1984. A numerical technique for simulation of
the effect of soil nitrogen transport and transformations on groundwater contamination.
Environmental Geology, vol. 5, no. 4, 213-218.

Development of a model intended as a tool for long-term prediction of the impact of agricultural activities
on aquifer systems and evaluation of management alternatives. Nitrate pollution potential in groundwater
is predicted using two numerical models. One model is for the vadose zone and includes transport by
dispersion and convection of mobile species of nitrogen, ammonium ion exchange, first order nitrogen
transformations, and plant uptake of nitrogen. The other model is for the aquifer system where transport of
nitrate is assumed to be affected only by dispersion-convection phenomena. A simple hypothetical
example problem is solved

Moosburner, GJ and EF Wood, 1980. Management model for controlling nitrate
contamination in the New Jersey Pine Barrens aquifer. Water Resources Bulletin, vol.
16, no. 6: 971-978.

Application of a land use management model to Jackson Township of the New Jersey Pine Barrens. The
model consisted of a simulation model for the transport of nitrates from septic tank systems through the
aquifer and a multiobjective, goal programming optimization model to determine population density
restrictions using planning population projections. Results showed that growth may have to be curtailed in
areas of Jackson Township in order to maintain acceptable nitrate concentrations in groundwater.

Pastor, J and WM Post, 1986. Influence of climate, soil moisture, and succession on
forest carbon and nitrogen cycles. Biogeochemistry, vol. 2: 3-27.

Report on a computer simulation developed to assemble a model ecosystem that links abiotic and biotic
processes through equations that predict decomposition processes, actual evapotranspiration, soil water

Draft... 9/14/98 1 -,

balance, nutrient uptake, growth of trees, and light penetration through the canopy. The model can make
accurate quantitative predictions ofbiomass accumulation, nitrogen availability, soil humus development
and net primary production for forests in eastern North America.

Rifai, H.S. et al., 1993. Getting to the nonpoint source with GIS. Civil Engineering.

As part of the Galveston Bay National Estuary Program in Texas, engineers have characterized the
nonpoint pollution sources that are poisoning the bay. A geographic information system has helped them
with extensive mapping-based calculations.

Smith, Richard A, Gregory E Schwarz, and Richard B Alexander. Regional
interpretation of water-quality monitoring data. Water Resources Research, vol. 33, no.
12, pp. 2781-2798.

Paper describes a model SPARROW Spatially referenced regressions on watershed attributes. The
method was designed to overcome problems in data interpretation caused by factors that complicate
regional water quality assessments: sparseness of sampling locations due to cost constraints, spatial biases
in the sampling network, and drainage basin heterogeneity. The method is used to estimate the proportion
of watershedss in the conterminous United States with outflow total phosphorous concentrations less than
0.1 mg/L and to classify cataloging units according to total nitrogen yield.

Soranno, P.A., S.L. Hubler, and R. Carpenter. 1996. Phosphorus Loads to Surface
Waters: A Simple Model to Account for Spatial Pattern of Land Use. Ecological
Applications 6(3):865-878.

This research models nonpoint-source phosphorus (P) loading from land to surface waters using a simple
model that accounts for spatial pattern in topography and land use using geographic information system
(GIS) databases. They estimated areas of the watershed that strongly contributed to P loading by
approximating overland flow, and modeled annual P loading by fitting three parameters to data obtained by
stream monitoring. The model was calibrated using P loading data from two years of contrasting annual
precipitation for Lake Mendota, a Wisconsin eutrophic lake in a watershed dominated by agriculture and
urban lands. Land-use scenarios were developed to estimate annual P loading from pre-settlement and
future land uses. As much as half of the Lake Mendota watershed did not contribute significantly to annual
P loading. The greatest contribution to loading came from a heterogeneous riparian corridor that varied in
width from 0.1 km to -6 km depending on topography and runoff conditions. They estimated that loading
from pre-settlement land use was one-sixth of the loading from present land use. A future scenario,
representing an 80% increase in existing urban land (from 9 to 16% of total watershed area, which would
be reached in 30 yr with current landuse trends), showed only modest increases in annual P loading but
possible significant effects on water quality. If the watershed were to become entirely urbanized, P loading
to the lake would double and potential effects on water quality would be severe. Changes in P loading
were strongest with conversion of undisturbed vegetated lands, especially riparian areas, to either urban or
agricultural uses. Variability in total annual rainfall led to variability in the riparian area that affects P
loading, with implications for policies intended to control nonpoint nutrient inputs.

Tim, U.S. and R. Jolly. 1994. Evaluating Agricultural Nonpoint-Source Pollution Using
Integrated Geographic Information Systems and Hydrologic/Water Quality Model.
Journal of Environmental Quality. 23(1):25-35.

Considerable progress has been made in developing physically based, distributed parameter,
hydrologic/water quality (H/WQ) models for planning and control of nonpoint-source pollution. The
widespread use of these models is often constrained by the excessive and time-consuming input data
demands and the lack of computing efficiencies necessary for iterative simulation of alternative

Draft... 9/14/98

management strategies. Recent developments in geographic information systems (GIS) provide techniques
for handling large amounts of spatial data for modeling nonpoint-source pollution problems. Because a
GIS can be used to combine information from several sources to form an array of model input data and to
examine any combinations of spatial input/output data, it represents a highly effective tool for H/WQ
modeling. This paper describes the integration of a distributed-parameter model (AGNPS) with a GIS
(ARC/INFO) to examine nonpoint sources of pollution in an agricultural watershed. The ARC/INFO GIS
provided the tools to generate and spatially organize the disparate data to support modeling, while the
AGNPS model was used to predict several water quality variables including soil erosion and sedimentation
within a watershed. The integrated system was used to evaluate the effectiveness of several alternative
management strategies in reducing sediment pollution in a 417-ha watershed located in southern Iowa. The
implementation of vegetative filter strips and contour buffer (grass) strips resulted in a 41 and 47%
reduction in sediment yield at the watershed outlet, respectively. In addition, when the integrated system
was used, the combination of the above management strategies resulted in a 71% reduction in sediment
yield. In general, the study demonstrated the utility of integrating a simulation model with GIS for
nonpoint-source pollution control and planning. Such techniques can help characterize the diffuse sources
of pollution at the landscape level.

Vieux, B.E. and Scott Needham. 1993. Nonpoint-Pollution Model Sensitivity to Grid-
Cell Size. Journal of Water Resources Planning and Management. 119(2).

Nonpoint-pollution models estimate loadings of chemicals, sediment, and nutrients that degrade water
quality. Before controls can be implemented, location and severity of pollution must be identified in the
watershed basin. Geographic information systems (GISs) are computer-automated, data management
systems simplifying the input, organization, analysis, and mapping of spatial information. Because
nonpoint-pollution models simulate distributed watershed basin processes, a heterogeneous and complex
land surface must be divided into computational elements such as grid cells. Model parameters can be
derived from each grid cell directly from maps using GIS. Cell size selection, if arbitrarily determined
though, yields ambiguous if not erroneous results. This paper investigates the effects of cell size selection
through a sensitivity analysis of input parameters for the nonpoint-pollution model, Agricultural Nonpoint
source Pollution Model (AGNPS), using a GIS for a small research watershed. Model grid-cell sizes were
found to be the most important factor affecting sediment yield. As the grid-cell sizes increase, stream
meanders are short-circuited. The shortened stream lengths cause sediment yield to increase by as much as

Winchester, John W, and Ji-Meng Fu, 1992. Atmospheric Deposition of Nitrate and Its
Transport to the Apalachicola Bay Estuary in Florida. Water, Air, and Soil Pollution 65:

Estimation of Nitrate deposition based on statistical analysis. Used weekly data from five National Acid
Deposition Program sites within the watershed and river water chemical data from the USGS. Other
surface sources of nitrate and chemical transformations were not fully quantified. Atmospheric deposition
appeared to be sufficient to account for essentially all the dissolved nitrate and ammonium and total organic
nitrogen flow in the river.


Odum, H.T., C. Diamond, M.T. Brown. 1987c. Energy Analysis Overview of the
Mississippi River Basin. Report to The Cousteau Society. Center for Wetlands
(Publication 87-1), Univ. of Florida, Gainesville. 107pp.

Draft... 9/14/98

Energy flow, quality, and embodied energy enables one to quantify and compare resource uses and
determining the development strategies that maximize energetic of both human and natural systems in an
increasing low energy world. Water cycles of the Earth are so important in the organization of the
landscape, river basins form a natural unit for understanding, predicting, and planning for the future. Soil
reserves within the basin were considered the most valuable long-range resource. Diking and channeling
caused much of these reserves to be lost to the sea.

Romitelli, S. 1997. Energy Analysis of Watersheds. PhD Dissertation, Environmental
Engineering Sciences. Univ. of Florida, Gainesville. 292 pp.

This research uses a new approach to study the organization of watersheds and to provide insight for their
management. It evaluates work done by water energies on the landscape and explores an hypothesis that
"self-organizing watersheds couple the geopotential and chemical potential energy use to maximize
biological and geological production". Work of the mountains was measured by the geopotential energy
use and related to work on terrestrial productivity of valleys measured by the chemical potential energy
evapotranspired. Using data on rainfall and river flow data and topographic geographic information, spatial
and temporal energy analysis and EMERGY evaluations were performed for six Brazilian watersheds of
the Ribeira de Iguape River basin, and for the Coweeta River basin in North Carolina. EMERGY is the
energy of one kind used directly and indirectly to make a product or service. Maps and graphs included the
water energies used, empower, and river transformities. Transformity is EMERGY per unit of energy.
Water Resources Council. 1979. A Unified National Program for Flood Plain Management. United States
Water Resources Council. This report (1) sets forth a conceptual framework for floodplain management,
(2) identifies available management strategies and tools for reducing the risk of flood loss, minimizing the
impact of floods on human safety, health, and welfare, and restoring and preserving natural and beneficial
floodplain values; (3) assesses the implementation capability of existing Federal and State agencies and
programs; and (4) makes recommendations for achieving "A Unified National Program for Floodplain
Management Although drafted as the result of a Federal initiative, the concepts and strategies of this
report are presented from a national perspective and offer guidance applicable to all governmental and
nongovernmental interests.

Whitfield, Douglas F. 1993. Emergy Basis for Urban Land Use Patterns In Jacksonville,
Florida. Thesis, University of Florida, Gainesville, FL. 212 pp.

Over the last fifty years, American cities have experienced a large proportion of their growth in the
suburbs. In many cities, residential and commercial activities are more dominant in the suburbs than in
downtown. This suburban land use pattern has changed the level of resources used to connect and support
urban functions. This study used Jacksonville, Florida to identify the major resources shaping the
organization of urban systems. The relationship between land use patterns and urban infrastructure,
resource consumption in transportation, and environmental contributions was investigated. Resources were
evaluated with emergy, an energy based unit of value, to place all resource contributions, both natural and
human, on a common basis. Emergy analysis represents a donor based measure of value in contrast to a
user based approach used in economics.

Draft... 9/14/98

Chapter 2
Stormwater, Pollutant Loads and Management
in the Lafayette and Munson Basins

Mark T. Brown and Neal Parker


Non-point sources of pollution have increasingly become the focus of attention as point
sources such as the outfalls from sewage treatment and industrial plants have been eliminated from
surface water bodies. Over the past decade, while great improvement of surface water quality was
achieved with elimination of direct discharge of wastes, non-point sources became more obvious
as an important contributor to surface water quality degradation.
In this study, two watersheds within the St. Marks watershed, the Lake Lafayette and Lake
Munson watersheds are looked at in detail (Figure 2-1). The two watersheds are similar, yet
exhibit differences that make the analysis of their stormwater, pollutant loads and the changes
expected in the future instructive. Lake Lafayette basin has urban development along its western
edge, with rural development and agricultural uses in the eastern and northern portions of the
watershed. Lake Munson, on the other hand, is more developed. With the city of Tallahassee
occupying most of its northern sub-basins, Lake Munson receives almost 2/3's of it total inflows
from heavily urbanized areas. In all, the Lake Lafayette basin is less developed, but significant
expansion of urban uses is occurring and anticipated in the future, while Lake Munson is more
developed. The contrasts between an urbanized watershed and one that is beginning to urbanize,
provide an interesting test of management alternatives. In the Lake Munson basin, reductions in
pollutant loading will require retrofitting an already urbanized watershed, while in the northern
portions of the Lake Lafayette basin, innovative development options that include wetlands and
special development buffers for waterbodies can be tested.

Description of the Study Areas
The Lake Lafayette watershed (Figure 2-2) encompasses approximately 80 square miles of
land in the northwest quadrant of the St. Marks Basin. The Lake is situated in the lower reaches of
the watershed and has been impounded for years in three places. These impoundments basically
divide the lake into 4 sections: Upper Lafayette, Piney Z, Lower Lafayette, and Alford Arm .
Alford Arm receives surface water from the northern portions of the drainage basin which is
predominately rural and contains a large amount of storage in natural depressions and man-made
ponds, as well as numerous closed basins. Lake Lafayette receives drainage from areas
immediately surrounding the lake and a major tributary called Northeast Drainage Ditch that drains
the heavily urbanized areas to the west.
Adjacent to the Lake Lafayette drainage basin is the Lake Munson basin, encompassing
approximately 69 square miles (Figure 2-3). The heavily urbanized areas of Tallahassee in the
northern portion of the basin contributes a significant amount of the total drainage through what are
called the West, Central, and East Drainage Ditches. The western portions of the basin are
dominated by the Apalachicola National Forest which drains into Bradford Brook to the Bradford
Lake Chain, eventually to Munson Slough. There are numerous closed basins and Bartel et al.
(1991) estimate that about 25% of the basin does not contribute stormwater to lake Munson. In the
past, several sewage treatment plants discharged to surface water, but these discharges were
eliminated in 1984.

Draft... 9/14/98

Munson Basin



Capital Circle

St MPas River

0 5,000 10,000 Metres

Figure 2-1. Map of the St. Marks Watershed showing Lake Lafayette and Lake Munson

Draft... 9/14/98 2-2

Figure 2-2. Maps of the land use and land cover in the Lake Lafayette Basin for the past
(top left), present (top right) and future (bottom).


1 Res. Low Density
2 Res. Medium Density
3 Res. High Density
4 Transportation/Utilities
5 Commercial
6 Industrial
7 Extractive
8 Institutional
9 Recreation
I 10 Cropland/Pasture
11 Upland Forest/Silviculture
S 12 Wetlands
13 Streams & Lakes

Figure 2-3. Maps of the land use and land cover in the Lake Munson Basin for the past (top
left), present (top right) and future (bottom).

Draft... 9/14/98

S1 Res. Low Density
2 Res. Med. Density
| 3 Mixed Use
4 Transp./Utilities
5 Commercial
6 Industrial
7 Extractive
8 Institutional
9 Recreation
1 I 10 Cropland/Pasture
11 Upland
12 Wetlands
S13 Streams/Lakes

Plan of Study
To better understand the effects of spatial distributions of land uses on pollutant loading,
spatial models that used land use and topography were developed that modeled pollutant load
received by surface water features. Lake Lafayette and Lake Munson were divided into sub-basins
to evaluate the various sub-basin contributions to each water body. In addition, pollutant loads
were modeled for major water bodies and closed basins within each of the larger watersheds to
provide perspective on areas of concern.
As a means of understanding loss of "basin function", pollutant loads were modeled for
three time periods for each of the basins: past, present, and future, and then compared.
Stormwater management options including Best Management Practices (BMP's), and restoration
of historic wetlands were tested with the models for the present and future conditions to evaluate
their effectiveness in reducing pollutant loads.


In this study two different methods were used to model pollutant load using Geographic
Information System (GIS) data layers. The first uses a fine resolution "Drain Model" for pollutant
transfer that predicts yearly average load for all locations within a basin. The second uses a
distance modified basin scale pollutant load model that sums potential load for sub-basins.
Comparisons were made between past conditions assuming no development, present conditions
based on 1989 land use / land cover, and for future conditions based on the Tallahassee-Leon
County 2010 Comprehensive Plan.
In both models, total phosphorus (TP) is used as the constituent of concern. Other
pollutants could have been used, by multiplying by their areal loading rates (Table 2-1). In
previous studies (Brown and Tilley, 1995, Tilley and Brown, 1998) and in the Stormwater Water
Management Plans produced by the NWFWMD (Bartel, et al, 1992) it has been determined that
few significant differences in relative loading exist between constituents, thus in essence, TP can
be thought of as an indicator for most other constituents of stormwater.

The Drain Model was used to model pollutant transfer, so that annual pollutant load could
be estimated for any point in the watershed. Thus pollutant load delivered by several streams or
ditches to a lake can be evaluated separately, and management actions taken accordingly. In like
manner, the annual pollutant load from stormwater overland flow to sinks and smaller lakes can be
determined as the sum of the flows entering through the lake or sink edge.
Figure 2-4 is a diagram of the steps in the GIS framework necessary to model pollutant
transfer from land uses to surface water bodies. The model requires as inputs a coverage1 of the
surface water bodies, a topography coverage (DEM), and Land use / land cover. The surface water
bodies and topography coverages were combined to produce a third coverage which depicts flow
paths. Essentially, the flow path coverage shows the direction and volume of pollutant transfer
from any point within the watershed "downhill" to the water bodies (or sink holes). A coverage of
Land Use / Land Cover was used in combination with Areal Stormwater Pollutant Loading Rates
(Table 2-1) to generate a Pollutant Loading coverage. The Pollutant Loading coverage was, in turn,
combined, or drained through the Flow Path Coverage to generate the Pollutant Transfer coverage.
Each cell in the Pollutant Transfer coverage has the total yearly pollutant load received from
"uphill" which is passed on to cells "downhill." Thus by reading the value in any cell, one is
reading the accumulated pollutant load in that cell.

S A coverage is GIS terminology that refers to a thematic, GIS map. Usually maps are paper
copies of GIS data, and several coverages can be combined to make one map.


Drain Model

~" ^ Information for
Decision Making

Figure 2-4. Diagram of the steps in the GIS framework for the DRAIN Model of pollutant
transfer from land uses to surface water bodies


Table 2-1. Areal Stormwater Pollutant Loading Rates (after Harper, 1995)

Land Use Category Areal Loading TP

Low density residential 1.53
Single family residential 2.84
Multi-family residential 8.24
Low intensity commercial 3.11
High intensity commercial 9.39
Industrial 5.94
Highways 6.32
Agriculture (pasture) 4.2
Agriculture (general) 2.64
Open Space 1.22
Mining 1.35
Wetland 0.00
Open water 0.00

Draft.. 9/14/98

Distance Modified Pollutant Load Model (DMPL Model)
This model was used to evaluate sub-basin annual pollutant loading to water bodies.
Generally, the pollutant load flowing to water bodies is a function of the intensity of activity in a
watershed and the physical characteristics of the watershed. Most pollutant load models deal with
watersheds as single dimensional space, treating all pollutant loads the same regardless of their
location within a watershed. With the increased availability of spatial data, it is possible to model
pollutant loading spatially; based not only on slope, but also on location, or distance from the
water body. The farther a source of pollutant generation is from a water body, the more natural
"treatment" may occur in the intervening space and thus the lower the actual load received by the
water body.
The Distance Modified Pollutant Load Model was developed to determine total pollutant
load based on topography, "effective" distance from the water body, land use/land cover, and areal
pollutant loading rates (Table 2-1). Figure 2-5 is a diagram of the steps in the GIS framework
necessary to model distance modified pollutant load for sub-basins. Required inputs to the model
are water bodies, topography (DEM's), and land use / land cover. The waterbody and topography
coverages are combined to generate a coverage that measures effective distance from waterbodies.
Effective distance is a slope modified straight line distance, so that steeper slopes have shorter
effective distance for the same straight line distance. All locations within a watershed are assigned
a unique distance to the nearest downhill water body, based on the relationship graphed in Figure
2-6. The coverage of land use / land cover is multiplied by aerial pollutant loading rates to create a
pollutant generation coverage where each land use contains the unit area, annual pollutant
generation. The pollutant generation coverage is multiplied by the effective distance coverage to
produce the annual distance modified pollutant loading coverage. Summing by sub-basins provides
total annual loads for each sub-basin.

Alternative Management Schemes
Three stormwater management schemes were tested using the Drain Model to determine
how present and future pollutant loads might be reduced by management practices and ecological
engineering. First, Best Management Practices (BMP's) were tested by assuming pollutant load
reductions of 10% for residential land uses and 20% for commercial/ industrial land uses. BMP's
might consist of street sweeping, the use of roadside swales instead of curb and gutter, and regular
cleaning and maintenance of stormwater conveyance systems. The second management scheme
consisted of a wetlands reconstruction alternative. For this alternative, all historic wetlands were
reconstructed in the basins, regardless of their locations, or what present day structures or
infrastructure might occupy the site. Finally the third alternative that was tested combined BMP's
with wetlands reconstruction.
The objective of modeling these alternative management schemes was to evaluate the effects
of management and ecological engineering on pollutant load. The BMP alternative represents a
conservative approach and a conservative estimate of pollutant reduction, assuming current
technologies with minor investments in stormwater management. The wetlands alternative
represents, in some respects, a radical approach to stormwater management, by reconstructing and
rehabilitating historic wetlands, regardless of their location, so they will function as stormwater
treatment areas. In a sense, this alternative indicates how much basin function has been lost
through years of development and engineering of stormwater systems that was concerned mainly
with quickly and efficiently removing stormwater




Pollutant Load Model

Total Pollutant

Figure 2-5. is a diagram of the steps in the GIS framework for the Distance Modified
Pollutant Load Model (DMPL Model) of distance modified pollutant load for sub-basins.

Draft... 9114/98

Graph of Distance Algorithm





< 0.4



0 10 20 30 40 50 60 70 80 90 100
Distance from Waterbody (Relative Units)

Figure 2-6. Graph of the distance algorithm used to assign all locations within a watershed
a unique distance to the nearest downhill water body.

Draft... 9/14/98



Stormwater in the St. Marks Basin
Stormwater is the water that runs off lands during rainfall events. When graphed against
time, the amount of stormwater runoff from a watershed exhibits a curve called a "runoff
hydrograph." Two such hydrographs are depicted in Figure 2-7. The two graphs show how the
amount and timing of runoff following a storm event changes when a watershed is developed. As a
watershed becomes more urbanized, stormwater runoff increases in amount and the speed at which
it flows off the land is increased as well.The lower graph is a hydrograph for an undeveloped
watershed, showing the slow increase in runoff after a rainstorm, with volume peaking about 3
hours after it begins to rain. The total area under the curve is the total volume of runoff that results
from the rainstorm. The higher graph is the hydrograph for the same watershed but with an
increase in impervious surface that has resulted from urbanization. Two things are apparent. First,
the timing of peak runoff has been changed. The peak comes 2 hours earlier than in the
undeveloped watershed. Second the amount of runoff has increased (area under the second graph
is larger than the area under the first graph) because the area of impervious surface forces more
water to runoff instead of percolating into soils.

Pollutant Loading and Percent Impervious Surface
As water runs off the land it carries with it many kinds of materials of varying sizes. Larger
things typically can be seen in drainage ditches, streams, and along the rivers and lakes of the St.
Marks watershed as the "jetsam and flotsam" of wood, plastic, cloth, bottles, cans, etc.
Stormwater carries sediments, or soil particles, that have been eroded from the watershed. It also
carries many materials and chemicals that are invisible to the eye because their are dissolved, or are
so small that they go unnoticed without the aid of a magnifying glass. All of these things might be
considered "pollutants", that is, too much of something in the wrong place at the wrong time.
However, while the big things carried by stormwater are the most noticeable, and often get the
attention of the public to focus on clean up of a water body, it is the smallest of these...the
dissolved solids, chemicals, and organic matter that cause the most trouble.
The things carried by stormwater are often called its pollutant load and depend on the type
of land the water flows over. In watersheds that are covered by forests, water running off the land
during storms might have a small amount of chemicals like soil nutrients, some organic
compounds that are washed from the soil surface, and bits and pieces of decaying vegetation.
Usually, without development, a watershed exports very little eroded soil, because the vegetative
cover is good protection against erosion, and acts to filter sediments if some should begin to
move.The volume of water flowing off undeveloped watersheds is slowed down by the vegetation
and allowed to percolate into soils, reducing the erosive ability of runoff and minimizing velocity.
When developed, with increasing amounts of impervious surface, the amount of water and the
speed at which it runs off increase.These factors have profound influence on the pollutant load that
stormwater carries.
Because stormwaters often end up in surface water bodies, the load of pollutants it carries
is of interest to the public. Water bodies that receive stormwaters often exhibit unacceptable
characteristics when polluted. Not only is the total load important, but the types of materials and
chemicals that are carried by stormwater are important. Experience has shown that stormwater
from urban areas contains, fertilizers, metals, and sediments, as well as a myriad of other
chemicals in smaller amounts. Stormwaters from agricultural lands contain fertilizers, pesticides,
sediments, and lessor amounts of metals. Which of these pollutant is of most concern, depends in
large part on their concentration. Usually, in urban watersheds, the effects of phosphorus in
stormwaters are more exaggerated than other chemicals. However, sediments can often be a major
problem. In agricultural watersheds, often phosphorus is a problem, but pesticides can be
extremely important because of the large effects that result form such small concentrations. Typical
constituents of stormwater from urban watersheds are given in Table 2-2.
The amount of impervious surface within watersheds is related to the intensity of human
activity and as a result, is a good predictor of stormwater water quality. The impervious surface in

Draft... 9/14/98





After Development


0 2 4 6 8 10 12

Time (hours)

Figure 2-7. Stormwater hydrographs for a developed and undeveloped watershed
showing how the amount and timing of runoff following a storm event changes when a
watershed is developed.

Draft.. 9/14/98


Table 2-2. Characteristics of Urban Stormwaters (from Corbitt, 1990)



Biological Oxygen Demand
Total Organic Carbon
Chemical Oxygen Demand
Suspended Solids
Total Solids
Volatile total solids
Settleable solids
Organic N
Total P04

1 700 mg/l
1 150 mg/l
5 3,100 mg/l
2 11,300 mg/l
200 14,600 mg/1
12 1,600 mg/1
0.5 5,4000 mg/1
0.01 16 mg/1
0.1 2.5 mg/I
0.01 1.5 mg/1
0.1 10 mg/l
0.1 125 mg/1
2 25,000 mg/l
0 110mg/l
0 0.2 mg/1
0 1.9 mg/l



itself is not the cause of poor water quality, but is a contributing factor, since increased impervious
surface translates into increased runoff and decreased ability for pollutant processing in soils. But
more importantly, as impervious surface area increases in watersheds, it indicates that intensity of
human activity increases as well. Human presence and all the activities associated with modem life
generate by-products, many of which lay around on the ground just waiting to be swept along with
stormwater runoff as it is directed toward downstream locations. Oils and metals from
automobiles, fertilizer runoff, and chemicals from combustion of fossil fuels are the most prevalent
of these by-products. During rainstorms these materials and chemicals are washed from the ground
surface, along with loose soil particles, organic debris, etc and deposited in streams and rivers.
When the amount of impervious surface within watershed areas is graphed against pollutant
load as in Figure 2-8, very often there is a strong relationship between the two. In this graph,
pollutant load in the sub-basins of Lake Lafayette and Lake Munson watersheds are compared to
impervious surface. As the graph shows, "impervious-ness" is strongly correlated to pollutant
load. Numerous other studies across the United States have shown consistently that there is a
strong correlation between impervious-ness of drainage basins and the health of the receiving
surface water body (Klein 1979; Griffin, 1980; Schueler, 1987; Todd, 1989; Schueler, 1992;
Booth and Reinfelt, 1993; and Schueler, 1994). Thus impervious surface may be a very good
predictor of stormwater quality and the health of downstream waterbodies.
A study of water quality in Wisconsin surface water bodies has shown that there is a shift
in the health of waterbodies receiving stormwaters when impervious-ness exceeds about 10% -
20% (Schueler, 1994). Water bodies receiving stormwaters from basins with between 10% and
20% imperviousness exhibited lower indices of ecosystem health than basins with less impervious
surface. As the percent imperviousness increased the indices decreased. Schueler (1992) suggested
that when imperviousness reached about 30% of basin surface area, water bodies receiving
stormwaters exhibited degraded conditions. In another study of Wisconsin streams Wang et al.
(1997) showed that indices of ecosystem health (they used Index of Biotic Integrity [IBI]) were
strongly correlated with land use in watersheds. They found that when agricultural uses exceeded
50% of total basin area, there was a marked decline in the IBI. In addition their study reinforced
the findings of Schueler (1994) that urban uses greater than 10% 20% of basin area significantly
lowered IBI's for basin streams.
In our study of the Lafayette and Munson watersheds, we ranked sub-basins based on their
imperviousness for the present land use conditions, and for future land uses based on maps
provided by Leon County Planning. Figures 2-9 and 2-10 show the sub-basins ranked by
imperviousness for Lake Lafayette and Lake Munson Basins respectively. The lightest gray
represents the lowest percent impervious surface, while the darkest gray has highest impervious
surface area. In the Lafayette Basin (Figure 2-9a), 4 basins had percent imperviousness less than
20% (basins 1,2,3, and 9), and only basin No.1 had percent imperviousness less than 10%. In the
Munson Basin (Figure 2-10a), two sub-basins had imperviousness equalt to or less than 20%
(Sub-basins 6 and 7), while four sub-basins (2,3,4 and 5) had greater than 30% imperviousness
Based on the previous studies by others which suggested that imperviousness was related
to ecosystem health (Schueler, 1992,1994; Wang et al.,1997), we might conclude that at the
present time 7 sub-basins out of the total number of 16 sub-basins within the Lafayette and
Munson watersheds have sufficient impervious surface (greater than 30%) to warrant serious
concern for the ecological health of surface water bodies within those sub-basins. Further, 6 sub-
basins within these two watersheds have sufficient areas of impervious surface (10% 20%
imperviousness) to raise concern for the ecological health of their water bodies.
When imperviousness is mapped for the future land use condition as in Figures 2-9b and 2-
10b, 14 basins have percent impervious surface greater than 30%, while no sub-basins have less
than 10%. Clearly, there is cause for concern. As a predictor of ecological health, the increase in
percent impervious surface within the Lafayette and Munson watersheds provides an indicator that
out look for the health of surface water bodies within these watersheds is questionable.



Relationship Between Percent Imperviousness and Average Phosphorus Loading

E 1.5

a. 1



0 10 20 30 40 50 6
Percent Imperviousness

Figure 2-8. Pollutant load vs. impervious surface in the sub-basins of Lake Lafayette and
Lake Munson watersheds showing a strong correlation between "impervious-ness" and
pollutant load.

Draft...9/14/98 2-15


0 2, 0 5W ^

g r- 17 [ L 32
S13 I 22 3s % Imperviousness
S14 0 27 i 43

1. Gilberts Pond Outlet unm"""u "
2. Roberts Pond Outlet 6. Lake Lafayette Drain
3. Alford Arm 7. Unnamed Slough
4. Buck Lake Outlet 8. Unnamed Run
9. Mall Drainage Area

I- is r v7 u1 ss
II as n s53 s % Imperviousness
r44 J 55

Fig. 9

Figure 2-9. Maps of the sub-basins of Lake Lafayette Basin ranked by imperviousness for
the present condition (top left) and the future (top right). The lightest gray represents the
lowest percent impervious surface, while the darkest gray has highest impervious surface

Draft... 9/14/98


M 13 [E 35
- O 20E- 48
z Z 49 % Imperviousness

1. Unnamed Run
2. Godby Ditch
3. Central Drainage Ditch
4. St. Augustine Branch

- 21 5s s6
S30 6s1 "% Imperviousness
[ 48 6

5. East Drainage Ditch
6. Munson Slough
7. Bradford Brook

Figure 2-10. Maps of the sub-basins of Lake Munson Basin ranked by imperviousness for
the present condition (top left) and the future (top right). The lightest gray represents the
lowest percent impervious surface, while the darkest gray has highest impervious surface

Draft... 9/14/98


Pollutant Loading
Using the DMPL Model, annual pollutant loads for each of the sub-basins in the Lafayette
and Munson watersheds were generated. The maps in Figures 2-11 and 2-12 show the spatial
distributions of pollutant loading for the past, present and future. For comparative purposes the
map values are given as load per area of sub-basin (kg/halyr). In this way the spatial generation of
pollutant load can be compared. The Lake Lafayette and Lake Munson sub-basins are numbered as
follows (Figures 2-1 land 2-12):
Lake Lafayette Sub-basins

1) Gilberts Pond 6) Lake Lafayette Drain
2) Roberts Pond 7) Un-named Slough
3) Alford Arm 8) Un-named Run
4) Buck Lake 9) Mall Drainage Area
5) Un-named Run

Lake Munson Sub-basins

1) Un-named Run 5) East Drainage Ditch
2) Godby Ditch 6) Munson Slough
3) Central Drainage 7) Bradford Brook
4) St. Augustine Branch

The maps show that future development in the Lafayette basin is moving northward, as the
higher pollutant loads move from the southern sub-basins to the northern basins (Figure 2-11).
Sub-basins within the presently urbanized area of Tallahassee have total annual pollutant loads of
about 1.5 lbs/acre*yr-1 (1.68 kg/ha**yrl), while Gilberts Pond and Alford Arm sub-basins (sub-
basins 2 and 3) have annual loads of about 0.7 lbs/acre*yr-1 (0.78 kg/ha*yr-1). Future annual
pollutant loads are significantly higher in the Alford Arm Basin (sub-basin 3) averaging about 1.37
lbs/acre*yr-1 (1.53 kg/ha*yr1)
Development in the Munson basin is moving toward the southeast although the changes
between the present annual pollutant loads and those of the future are not as significant as the
changes in the Lafayette Basin. The Bradford Brook Sub-basin ( sub-basin 7) has annual pollutant
loads of about 0.7 lbs/acre*yr-1 (0.78 kg/ha*yr-1) presently, increasing to about 0.8 lbs/acre*yr-1
(0.9 kg/ha* *yr1) in the future. There is not much increase from the present to the future in total
annual pollutant load in the Tallahassee sub-basins (sub-basins 2,3,4, and 5), since these areas are
nearly fully developed.
The graphs in Figures 2-13 and 2-14 show comparisons of annual pollutant loading by
sub-basin for the past, present and future conditions for the Lafayette and Munson Basins. The
shortest bars in the graph are for the natural landscape, averaging about 0.4 lbs/acre*yr-1 (0.45
kg/ha*yr-1). Urbanized areas have about 3 times these background loads (1.4 lbs/acre*yr-1 [1.48
kg/ha*yr-1]). The biggest change from present conditions to future conditions are found in the
Lafayette Basin where the moderately developed sub-basins exhibit annual pollutant load increases
of between 50 to 80%. The increases in the Munson basin between the present and future condition
are much smaller, with only one basin (Un-named Run #1) exhibiting a 75% increase in annual
load. The remaining basins all appear to exhibit increase of between 5 and 15%.

Pollutant Transfer
To better understand the spatial distribution of annual pollutant loads requires taking a
closer look at the Lafayette and Munson Basins. The Drain Model uses an overland flow algorithm



..... S.000 14.,...

Sa n m 1.o TP (lbs/ac/yr) TP (lbs/ac/yr)

o.86 1022 La 1.77

Gilberts Pond Outlet
Roberts Pond Outlet
Alford Arm
Buck Lake Outlet

5. Unnamed Run
6. Lake Lafayette Drain
7. Unnamed Slough
8. Unnamed Run
9. Mall Drainage Area

Figure 2-11.Maps of Lake Lafayette Basin showing the spatial distributions of pollutant
loading for the present (top left) and future (top right). For comparative purposes the map
values are given as load per area of sub-basin (lb/acre*yr 1).

Draft... 9/14/98


9 0.57 1.34 06? I 64
O0.77 1 .s rP (Ibs/ac/yr) 0a s W 1.6i TP (Ibs/ac/yr)
1.00 1.75 1.04 I 1.78

1. Unnamed Run
2. Godby Ditch
3. Central Drainage Diitch
4. St. Augustine Branch

5. East Drainage Ditch
6. Munson Slough
7. Bradford Brook

Figure 2-12.Maps of Lake Munson Basin showing the spatial distributions of pollutant
loading for the present (top left) and future (top right). For comparative purposes the map
values are given as load per area of sub-basin (lb/acre*yr-l).

Draft... 9/14/98


Total Phosphorus Average Loads for the Lake Lafayette Basin


1.8 l Past E Present E3 Future


2 3 4 5
Basin Number

Figure 2-13. Comparison of annual pollutant loads by sub-basin for the past, present and future
conditions for the Lake Lafayette Basin.



Total Phosphorus Average Loads for the Lake Munson Basin

1 2 3 4 5 6 7
Basin Number

Figure 2-14. Comparison of annual pollutant loads by sub-basin for the past, present and
future conditions for the Lake Munson Basin.


Draft... 9/14/98

to converge and concentrate runoff much as water actually runs off land. The maps in Figures 2-15
and 2-16 depict annual pollutant load for past condition, present condition, and future condition in
the Lake Lafayette and Lake Munson basins that were generated using the Drain Model. Darker
areas indicate highest total pollutant load, while the lighter areas indicate lowest overall pollutant
load. Evident in the maps is a dendritic pattern of pollutant convergence and concentration.
Pollutant loads are summed along flow paths so that total load at any point in the drainage basin
can be read from the maps. The graphs in Figures 2-17 and 2-18 show concentrations at several
points within each of the basins for the three time periods. Figure 2-17 shows the annual pollutant
loads in Lake Lafayette Basin at the following four locations: A) lower reach of Alfrod Arm
Branch, B) Tallahassee Drainage inflow to Upper Lake Lafayette. C) Lake McBride, D) Killeam
Lakes,. Figure 2-18 graphs the annual pollutant load at the following 3 locations in the Lake
Munson Basin: A) Bradford Brook inflow to Bradford Lake, B) Northern Drainage Area (at the
confluence of Gun Creek and West Drainage Ditch), and C) inflow to Lake Munson.
One of the most significant changes in the Lafayette Basin (Figure 2-17) is the 350%
increase in annual pollutant load over historic loads in the lower reach of Alford Arm where
because of the drainage basin size and increasing urbanization the total loads are very large. While
there was a 450% increase between the historic annual load and the present load in the Tallahassee
Drainage inflow to Upper Lake Lafayette the change from present to future is much smaller (15%)
because the area contributing runoff is mostly built out. The model predicts an 80% increase in
annual load to Lake McBride in the future and a 70% increase for Killearn Lakes.
Annual pollutant loads at the three locations within the Munson basin are shown in Figure
2-18. There is little change in the pollutant load at the inflow to Bradford Lake between the past,
present and future. The most significant increase in the basin is at the inflow to Munson Slough
where the increase from past to present was about 367%. The model predicts that the annual
pollutant load will increase again in the future about 30% over current loads. The increase in the
Northern Drainage Area was about 233% from past to present. The model predicts that annual
pollutant loads at this location will increase another 30% over current loads

Ecological and Hydrological Function
Ecological function describes the normal properties and processes that an ecosystem
exhibits over the course of time. Since the climate is not constant, but varies from day to day and
season to season, ecosystem processes have to be thought of as average conditions. Sometimes
ecosystem functions are called ecosystem services because often they are not "things" so much as
they are processes that might be exploited. Probably the single most important ecosystem
processes is primary production, since all other properties and processes occur as a result of the
fact that an ecosystem is a living system that provides services as part of its everyday processing of
energy and materials.
Watersheds like the St. Marks are a collection of ecosystems and human dominated
systems. Ecosystem function of a watershed is the collection of properties and processes of the
ecosystems of the watershed; for instance, a basin's ability to cycle nutrients, store water, provide
erosion protection, and provide wildlife habitat. We might say that ecological function is 100%
when fully natural (having no urbanization or agriculture uses) and potentially decreases as
urbanization and agricultural uses increase.
Hydrologic function of a watershed is the sum total of properties and processes of water
moving within a watershed. It is such things as storage and timing of water flows, watershed
nutrients dynamics, and the erosion and sedimentation cycle driven by water. Again, we might say
that hydrologic function is 100% when fully natural and potentially decreases as the watershed is
urbanized, or converted to agricultural uses.
Measuring ecological or hydrologic function is no easy matter. The loss of ecological
function might be measured by computing Indices of Ecological Health for all ecosystems within a
basin, or by simply measuring the land areas of natural undisturbed lands verses urban and
agricultural lands and equating decreases in indices or natural land area as loss of ecological
function. Loss of hydrologic function might be measured by changes in stream hydrographs that

Draft... 9/14/98

Figure 2-15. Maps depicting annual pollutant load for past condition (top left), present
condition (top right), and future condition (bottom) in the Lake Lafayette Basin generated
using the Drain Model. Darker areas indicate highest total pollutant load, while the lighter
areas indicate lowest overall pollutant load.

Draft... 9/14/98

Figure 2-16. Maps depicting annual pollutant load for past condition (top left), present
condition (top right), and future condition (bottom) in the Lake Munson Basin generated
using the Drain Model. Darker areas indicate highest total pollutant load, while the lighter
areas indicate lowest overall pollutant load.

Draft...9/14/98 2-25

Total Phosphorus Loading for Four Locations within the Lake Lafayette Basin


8000 --



Past Present Future

Figure 2-17. Graphs of the annual pollutant loads in Lake Lafayette Basin for the three
time periods at four locations: A) lower reach of Alfrod Arm Branch, B) Tallahassee
Drainage inflow to Upper Lake Lafayette, C) Lake McBride, D) Killean Lakes.



Total Phosphorus Loading for
Three Locations in the Lake Munson Watershed

40000 ,




. 20000




Past Present Future

Figure 2-18. Graphs of the annual pollutant load for the three time periods at 3 locations in
the Lake Munson Basin: A) Bradford Brook inflow to Bradford Lake, B) Northern
Drainage Area (at the confluence of Gun Creek and West Drainage Ditch), and C) inflow to
Lake Munson.

Draft...9/14/98 2-27

indicate loss of water storage capacity, changes in erosion and sedimentation rates, or changes in
pollutant loads. Thus loss of function is relative; measured against some given condition.
Stream flow and the materials that it carries can be a good indicator of both hydrologic and
ecological function. If a watershed is ecologically healthy, and hydrologically healthy, waters
should flow without destructive quantities and timing and should contain normal amounts of
sediments and nutrients. Changes in ecological function or hydrologic function should show up in
the quantities and timing of water and in the constituents carried by water. So we might speak of
the sum total of ecological and hydrological function as "Basin Function."
Comparing pollutant loading in the present condition with that of the past as in Figures 2-
13 and 2-14 and Figures 2-17 and 2-18 provides a metric against which we might judge loss of
basin function.. Potential losses can be measured by comparing past with future conditions
assuming build out. The loss of basin function is shown as the increase in pollutant loading.
Basins where present pollutant loadings are less than 50% greater than historic loads, while having
lost some function are still functioning at reasonable levels (Roberts Pond Sub-basin in Lafayette
Basin and Bradford Brook Sub-basin in Munson Basin). The remaining basins have greater than
100% change between historic loads and present loads; some as much as 450% increases. These
basins have lost most hydrologic functions and will require serious efforts to ecologically engineer
them to approach historic levels. In the following section, we suggest several alternatives for
stormwater management that could increase basin hydrologic function. We use the Drain Model to
test the effectiveness of these management alternatives.

Stormwater Management Alternatives
Water flows down hill. This simple truth is the basis for most stormwatermanagement
systems. Until very recently, stormwater management consisted of collecting and efficiently
conducting stormwater runoff to the lowest point within a watershed. The lowest point often was a
stream, river, or lake. The concern of stormwater managers was to get water off the land to
minimize flooding. In the early days, when developed areas were still small and undeveloped areas
large, this technique of stormwater management worked. Stormwaters were quickly and efficiently
routed off developed lands through constructed ditches and deposited in downstream water bodies.
However, as the amount of impervious surface within watersheds increased, the amount of runoff
greatly increased and the pollutant load it carried increased as well. Receiving waterbodies began to
show signs of stress; lakes became eutrophic, rivers exhibited fish kills and extreme sediment
erosion and deposition, and ultimately coastal estuaries showed signs of toxic waste buildup and
bacterial contamination.
In some respects, stormwater has become over managed. That is to say that through
efficient removal of ever increasing volumes of stormwater carrying increasing pollutant loads,
down stream waterbodies have suffered. The stormwater was managed, but the system was not.
The time has come to revisit the concept of stormwater management, rethinking its goals, and
redesigning stormwater management systems. Much work has already been done in developing
and administering stormwater management regulations that have begun to reverse the trends of
declining stormwater quality. New developments are required to retain and treat stormwaters, thus
reducing quantity of runoff and improving quality. However, there is much stormwater runoff that
does not benefit from these new regulations because it flows from areas that were developed prior
to adoption of stormwater regulations. It is these older areas that often present the most difficult
stormwater problems to solve.
The graphs in Figures 2-19 and 2-20 show changes in pollutant loading at several locations
in the Lafayette and Munson basins and the effects of three management alternatives. In the first
alternative, BMP's are used to reduce pollutant loads by 10% in residential areas and 20% in
commercial areas. In the second alternative, wetlands are reconstructed throughout the watersheds
to replace wetlands lost over the years to development. And the third alternative is to combine
BMP's with the wetlands alternative. In almost every case, BMP's have minor effect on reducing
pollutant load. On the other hand re-establishing wetlands throughout these basins can have a
significant impact on water quality, reducing it, in some cases, to near historic levels. We
recognize the radical nature of this proposal, and suggest it more in a comparative mode. It



Total Phosphorus Loads for Sevral
Managmnont Altnnatives at Sit. A in the Lak* Lafayette Bastn

Tow Phostphors Ladw to Samwr
MoMlgenten Alteatlna t Site B In LaLfayftt Bain


aooot- ____---- -- ----- -

matn --- ------ -- ----
1. -.--1 -- I----

mm ---
I -

lot Ph),.r -
IPhM r'mswa --

ToWt Pholphonus Loading for Soveral
tUnagor nt ANematives at Site C In te Lake Laefyett Basin

Total Phosphonas LNading for S-nv]al
Mnagem nt Altmatni at Site 0 in Iti Lake Lafayette Basin


.ow -- - - I_- n
.-B lS ______ _-___ ____
,-o --- -o .11

Pa-a Pr Fu- l Pa Pe

Figure 2-19. Graphs of the changes in pollutant loading at several locations in the Lake
Lafayette Basin and the effects of three management alternatives: 1) the use of Best
Management Practices (BMP's) that reduce pollutant loads by 10% in residential areas and
20% in commercial areas, 2). wetlands reconstruction throughout the watersheds to replace
wetlands lost over the years to development, and 3) combination of BMP's with the
wetlands alternative.

Draft... 9/14/98


U3-- -


.Olill ir*r JID
`IYOJ ~,~Hu~l ---- -~ ~- -



Total Phosphorut Loads for Sovral
Mangmn lAl[tmat at St A in th. Lako Munon Basin

200t-- d----- I .

2S0--- -


Toal Phophmom Loads for Svaal
Mmoag.mtnt AfornatlUs Site B In th Lake MUurn Bin

12000 -----

Dam. --- --- --

ToOU Phosphous Loads for Svral
Managmnt AaternatIs at Sile C in th Lake Munron Basin

0- NODAOn.
.001- ---- .....-- _______s_ ___
goat iaai I-- ij "~

Figure 2-20. Graphs of the changes in pollutant loading at several locations in the Lake
Munson Basin and the effects of three management alternatives: 1) the use of Best
Management Practices (BMP's) that reduce pollutant loads by 10% in residential areas and
20% in commercial areas, 2). wetlands reconstruction throughout the watersheds to replace
wetlands lost over the years to development, and 3) combination of BMP's with the
wetlands alternative.

Draft... 9/14/98


essentially says, that with more sensitive development, protecting wetlands and storing water
within watersheds instead of diverting it downstream, significant losses of basin function can be
Figure 19 shows graphs of annual pollutant load at Sites A through D in the Lafayette basin
(Figure 15). The historic pollutant load at each location is given first in each graph, and then loads
for the present and future conditions are given. The first bar in each time period is the load
predicted by the Drain Model with no management alternatives. The second and third bars result
from instituting BMP's and the wetlands reconstruction alternatives respectively. The fourth bar
represents total load when both BMP's and wetlands are combined as a management alternative.
BMP's while an effective means of reducing some loads, appears to reduce loads by about 10 -
15% in urbanized basins. The largest increase in water quality resulted form the wetlands
reconstruction alternative, lowering pollutant loads in some cases by more than 50%. The
combined approach, provides additional improvement in reducing total load.
Pollutant loads at sites A through C in the Munson Basin (Figure 16) are given in Figure
20. The largest improvement in total load is achieved at Site B (the northern drainage area) where
reconstruction of historic wetlands provided a 40% decrease in pollutant load. When the combined
management alternative was tested, the improvement was almost 60%. This is a moderately
developed basin that had relatively large historic areas of wetlands
Its important to note that these management alternatives do not lower pollutant loads to
historic levels, but certainly should provide significant increases in water quality. We are quite
aware that the reconstruction of historic wetlands is extremely difficult and probably highly
unlikely, given the present amount of urban infra-structure throughout these basins. However, the
alternative shows the impacts of their loss on basin hydrologic function, and the difference that
wetland sensitive development could make today and in the future. The same level of treatment
might be attained through the construction of wetland stormwater treatment areas within basins in
locations where it is possible to do so, although finding suitable vacant property in basins that are
nearly completely developed may be difficult at best. In these situations, plans now for the
purchase of sites in future years that are suitable because of under utilization or pollutant status
(Brown Fields) may be extremely desirable. In all, it is apparent that BMP's alone will not provide
the level of treatment necessary to improve today's nonpoint source pollution problems in
urbanizing areas, leaving the future in serious doubt. Even effective programs of ecological
engineering, may find it difficult to provide the level of treatment necessary to improve the health
of water bodies in urban landscapes.

Draft... 9/14/98


Bartel, R., R. Arteaga, N. Wooten, F.B. Ard, and A.T. Benoit, 1991a. Lake Munson Basin Plan
City of Tallahassee and Leon County Stormwater Management Plan Vol. II. NWFWMD.

Bartel, R., R. Arteaga, N. Wooten, F.B. Ard, and A.T. Benoit, 1991b. StormwaterManagement
Plan for the City of Tallahassee and Leon County Vol. IV Technical Report. NWFWMD.

Booth, D.B. and Reinfelt, L.E. 1993. Consequences of urbanization on aquatic systems-
measured effects, degradation thresholds, and corrective strategies. In Proceedings of the
watershed '93 Conference, Alexandria, Va. March 1993 1,3:114-6

Brown, M.T. and D. Tilley. 1995. SOUTH DADE WATERSHED PROJECT: Data Inventory
and Compilation, Evaluation of Wetland Stormwater Requirements, and Partial Ranking
of Drainage Basins. A Research Report to the South Florida Water Management District
and the Center for Community Design, University of Miami. Center for Wetlands,
University of Florida, Gainesville, FL.

Corbitt, R.A. (ed) 1990. Standard handbook of Environmental Engineering. McGraw-Hill, Inc,
New York.

Harper, H. H., 1994. Estimation of Stormwater Loading Rate Parameters for Central and South
Florida. Environmental Research and Design, Inc., Orlando, Florida.

Klein, R.D. 1979. Urbanization and Stream quality impairment. Water Resources Bull. 15:948-

Griffin, D.M.1980. Analysis of non-point pollution export from small catchments. J. Water. Pol.
Control Fed. 60,1:95-108

Schueler, T. 1987. Controlling urban runoff: A practical manual for planning and designing urban
BMP's. Publication #87703 of the Washington Metro. Council of Governments.

Schueler, T. 1992. Mitigating the adverse impacts of urbanization on streams. In Kumble and
Schueler (eds). Watershed Restoration Sourcebook Publication #92701 of the
Washington Metro. Council of Governments.

Schueler, T. 1994. The importance of imperviousness. Watershed Protection Techniques 1:100-

Tilley, D.R. and M.T. Brown. 1998. Wetland networks for stormwater management in subtropical
urban watersheds. EcologicalEngineering. [In press].

Todd, D.A. 1989. Impact of Land use and nonpoint source loads on lake quality. J. of Env. Eng.
115, 3:633-649.

Wang, L., J. Lyons, P. Kanehl, and R. Gatti.1997. Influences of watershed land use on habitat
quality and biotic integrity in Wisconsin streams. Fisheries 22, 6:6-12.

Draft... 9/14/98 2-32

Chapter 3:
Spatial Modeling of Nitrogen Loading to a Surficial Aquifer from Residential
On-site Sewage Disposal Systems in Wakulla County, Florida.

Alan Foley


This study developed a spatial modeling technique within a GIS environment to predict
the fate of nitrogen loading from septic tanks to the surficial aquifer in Wakulla County (Figure
1). The study focused on five hydrologic sub-basins surrounding Wakulla Springs State Park
and the Wakulla River. This area is rapidly urbanizing and has a high potential for groundwater
degradation. The model was used to evaslaute nitrogen loading that resulted from present
development patterns and future development patterns based on the Wakulla County Future
Land Use Plan and assuming buildout to zoned densities (Figure 2)


The estimated population of Wakulla County in 1985 was 13,159 people. The projected
county population in 2005 is 20,000 people (Ayres and Associates, 1987). This population
increase will proportionally increase potable water and wastewater demands.
The majority of residences in Wakulla County are un-sewered and therefore utilize on-
site sewage disposal systems (OSDS's) for wastewater treatment. It is estimated that OSDS's
provide about 70% of Wakulla County's wastewater treatment, (Ayres and Associates, 1987).
OSDS's dispose of wastewater using a variety of components and configurations, the most
common being the septic tank/soil absorption system (STSA), (Kirkner and Associates, 1987).
Physical, chemical, and biological processes in septic tanks and soil provide treatment of
wastewater. Soil adsorbtion of chemicals in septic tank effluent is probably the major
contaminant removal mechanism in the subsurface environment (Wilhelm et al., 1994, Canter
and Knox, 1984). Adequate contaminant removal may not be possible in some areas of Wakulla
County, (FDEP, 1979). STSA's have a direct influence on water quality in the surficial aquifer
and may influence water quality in the deeper Floridan Aquifer drinking water source for most
of Wakulla County (L6wrance and Pionke, 1989, Kirkner and Assoc., 1987).
Increased STSA density has been related to groundwater degradation, (USGS, 1995,
Yates, 1989). STSA induced groundwater degradation can be characterized by high nitrate
(N03) and bacteria concentrations in addition to potentially significant amounts of organic
contaminants. Septic tank failure is commonly due to exceedance of soil effluent absorption
capacity. Soils with high permeability such as those in the Woodville Karst Plain can be
rapidly overloaded with organic and inorganic chemicals and Canter and Knox, 1984). Current
STSA design parameters and regulations may not provide for adequate protection of
groundwater quality (Bicki, et al., 1984). Elevated N03- concentrations have been detected at
Wakulla Springs State Park and in some private wells (FDH, 1998, Hand, 1998).


Figure 1. Location map of the Wakulla County Study Area. Shown are: (1) a portion of
the Apalachcola National Forest (the dark green are in the western portion of the study
area), (2) Wakulla Springs State Park (purple area in the center of the study area), (3) St
Marks National Wildlife Area (lighter purple area in the southern tip of the study area),
and (4) present zoneing classes (see Figure 2 for details).


Present Land Use

Incorporated City
Urban 1, 2 DU/ ac.
Rural 2, 1 DU / 2 ac.
Rural 1, 1 DU / 5 ac.
Zoned Agriculture / Silviculture
Wakulla Springs State Park
Apalachicola National Forest
St. Marks National Wildlife

Future Land Use

Figure 2. Present and future land use in the study area.


I i

Throughout much of the populated areas of the county there is no confining unit
overlying the Floridan aquifer the primary source of drinking water in the county, and in much
of northern Florida. The eastern half of Wakulla County is located on the Woodville Karst Plain
- a groundwater discharge area with characteristic karst features such as springs and sinkholes.
The limestone aquifer in this area is covered by a thin layer of fine, unconsolidated quartz sand
and clay that generally is less than 20 feet thick (SCS, 1991).
Water in the sand and clay layer overlying the St. Marks formation is free to rise and fall
and is referred to as the surficial aquifer. Ground water recharge in this area is derived mostly
from precipitation. The Wakulla County Soil Survey describes nearly all of the soil types
present as severely limited in regard to the placement of septic tank adsorption fields (FDEP,
1979, SCS, 1991). As the county continues to build out at rural densities, management
agencies are challenged to accommodate growth while preserving the water resources and
environmental features in Wakulla County.
The tight integration between hydrology, geology, and biology requires a
multidisciplinary management approach where geographic information systems (GIS) may
provide a means to work across disciplinary lines in problem analysis. The compilation and
standardization of digital geographically referenced data sets enables spatial modeling which
may aid in land use decisions and efficient allocation of management resources.

Review of the Literature

Regulation of nitrogen in drinking water
The US Environmental Protection Agency has set drinking water maximum
contamination levels (MCL's) for two inorganic, nitrogen-containing compounds, nitrite (NO;2),
and nitrate (NO3). The current MCL for NO2- is 1 mg/L (as nitrogen). The current MCL for
NO3 is 10 mg/L (as nitrogen). A third inorganic compound, ammonia (NH4) is on the
Drinking Water Priority List (DWPL). The DWPL contains substances that are known or
anticipated to occur in public water systems and that may require regulation under the Safe
Water Drinking Act, (AWWA, 1992).
NO2 may combine with amines in the stomach to form potentially carcinogenic
nitrosamines. High NO3 levels in drinking water can cause methemoglobinemia (Blue baby
syndrome). Nitrate is converted to nitrite (NO;) in the intestines. The NO; combines with
hemoglobin in the blood to form methemoglobin. Methemoglobin does not transport oxygen in
the blood. High levels of methemoglobin can cause death due to asphyxiation. Babies and cattle
are especially susceptible to this condition. The current drinking water limit for NO; is a
concentration of 10 mg/L.

Nitrogen in the environment
Nitrogen containing compounds can be introduced to the environment as solid, liquid, or
gaseous forms. USGS chemical analyses of atmospheric precipitation yielded low nitrate
concentrations relative to concentrations in groundwater, (USGS, 1996a). Atmospheric
deposition of nitrogen was not considered in this analysis.
Agricultural practices such as crop fertilization and dairy farming can introduce large
quantities of nitrogen into the environment. Approximately 2 percent of the land area in Wakulla


County is used for agriculture, predominantly for the production of row crops and as pasture,
(SCS, 1991).
Treated wastewater from domestic and commercial OSDS's is discharged directly to the
subsurface environment. OSDS effluent is the most frequently reported source of groundwater
contamination, (Miller,1980).
Groundwater monitoring studies and laboratory column studies indicate that
approximately 20 -40% of nitrogen in STSA effluent may be adsorbed or otherwise removed
before the effluent reaches the groundwater, (Bicki, et al, 1984).

Nitrogen chemistry in soil Several chemical processes govern nitrogen transportation and
distribution in the subsurface environment. These processes convert nitrogen from organic to
inorganic forms and include: ammonification, volatilization, nitrification, denitrification,
dissimilatory nitrate reduction to ammonium, adsorption, and biological uptake, (Bicki et al,
1984). Denitrification is the only process that may serve as a major nitrogen sink, the other
processes temporarily immobilize nitrogen, (Korom, 1992).

Ammonification Ammonification mineralizationn) is the conversion of organic nitrogen
compounds such as proteins and amino acids to inorganic compounds such as ammonia (NH3),
ammonium (NH4), nitrite (NO2), and nitrate (N03). Ammonification is performed by a wide
variety of organisms (bacteria, actinomcetes, and fungi) and follows several reaction pathways.
Ammonification typically refers to complete mineralization when all organic nitrogen is
converted to NH4 (Bitton, 1994; Bartholomew, 1965).

Volatilization NH4' is the predominant reduced nitrogen form in acidic and neutral pH aquatic
environments. As pH increases, NH3 becomes more abundant and is released as a gas.
Volatilization does not remove significant amounts of nitrogen from OSDS effluent due to the
fairly neutral range of effluent pH (7.0 8.5). In addition, subsurface disposal systems do not
provide adequate air-water contact for appreciable amounts of NH3 to volatilize (Bicki et al,
1984). When conditions are favorable, NH, is oxidized to NO2- or N03 before its assimilation
by microbes (Bartholomew, 1965).

Nitrification Nitrification is the oxidation of NHI to N03. Nitrification occurs in the
temperature range of 15-35 C and is favored in wastewater effluents with a low biological
oxygen demand (BOD) and a high NH4C content, (Bitton, 1994). Nitrifying bacteria require an
aerobic environment (oxygen concentration greater than 1.0 mg/L), and sufficient alkalinity to
neutralize the hydronium ions produced during the oxidation of NH4+ (pH of 7.2-8.4 is optimal
for nitrification). Acidic soil conditions will prevent nitrification, (Bartholomew, 1965). High
nitrification rates may adversely affect denitrification by lowering pH, (Reneau et al., 1989).
The two-step nitrification reaction is performed by two groups of nitrifying bacteria. The
Nitrosomonas group of bacteria converts NH4 to NO2. The Nitrobacter bacteria group then
transforms the NO; to NO;, (Bicki et al., 1984; Bitton, 1994). NO3 is very soluble and
chemically inactive under aerobic conditions. Once in the groundwater N03 is highly mobile.
Reduction of NO; concentrations in an aquifer will occur due to dilution or denitrification
(Bicki, et al., 1984).


Denitrification The denitrification process reduces NO3- to gaseous N20 or N2. These gases
will diffuse to the atmosphere, resulting in a loss of nitrogen from the subsurface system.
Denitrification requires nitrogen oxides (NO3-, NO2, NO, and NzO), bacteria with the metabolic
capacity to perform denitrification, anaerobic conditions, and suitable electron donors, (Korom,
1992). Chemical denitrification in shallow groundwater due to the oxidation of ferrous iron may
occur, (Lowrance and Pionke, 1989).
Denitrification is a two step process; NO3- is converted to NO;, which is then converted
to N20 or N2, (Bitton, 1994). Nitrogen must be in an oxidized form for denitrification to take
place. Nitrification provides the oxidized nitrogen compounds for denitrification. Aerobic
nitrification must precede anaerobic denitrification in the soil.
The bacteria that carry out the denitrification process are facultative anaerobes -
organisms capable of survival with or without 02. Necessary anaerobic environments are present
under saturated conditions. Organic matter (OM) is the most commonly used electron donor.
Bacteria use 02 to oxidize the OM until oxygen supplies are depleted. Facultative anaerobes
then switch to using NO3- as an oxidizing agent (electron acceptor). An increase in 02 levels will
cause the bacteria to return to using O0 instead of NO3- to oxidize OM, (Korom, 1992).
Denitrification rates depend on the types of bacteria present locally and whether they are
actively denitrifying. Local denitrification rates need to be determined on an individual basis,
(Korom, 1992). Consistent nitrification-denitrification processes are rare in a subsurface system
due to the variability of the water table.

DNRA Dissimilatory nitrate reduction to ammonium (DNRA), like denitrification, also occurs
under anaerobic conditions. DNRA tends to conserve nitrogen within the soil system, while
denitrification tends to export nitrogen from the soil system. DNRA is favored when N03
supplies are limiting. Denitrification is favored when OM (electron donor) supplies are limiting.
Though the inorganic NH4 ion is not water-soluble under aerobic conditions, it may later
undergo nitrification to produce NO,3. Saturated soil conditions may create an anaerobic
environment in which NH4 can be leached to the ground water, (Korom, 1992). Anaerobic and
aerobic conditions may alternate at a location due to water table fluctuations.

Adsorbtion Soil adsorbtion of NH4 can be significant under certain conditions. The factors
influencing NH4+ adsorption are: the number of cation exchange sites in the soil, the affinity of
those sites for NH4+, site saturation with NH4, and the composition of the effluent. Other cations
present in the effluent may outcompete NH4 for exchange sites, (Bicki et al, 1984). Soils with a
higher content of clay tend to adsorb more NH4. Studies referenced in Bicki et al, 1984, found
that NH4 adsorbtion ranged from 2 mg/100 grams of sandy soil to 100 mg/100 grams of fine-
textured soil with 30% clay content.
Saturation of cation exchange sites in the soil underlying an OSDS may occur. Cations in
the OSDS effluent would no longer be adsorbed and would pass through the septic field and into
the groundwater. Adsorbed NH4+ can be desorbed and nitrified if reaeration occurs due to water
table fluctuation, (Bicki et al, 1984).

Biological uptake Biological uptake of nitrogen by microbes and plants results in nitrogen
immobilization for the life of the assimilating organism. The low carbon/nitrogen ratio of OSDS
effluent limits microbial uptake. The amount of nitrogen generated by OSDS's typically exceeds


the capability of local plant uptake, (Bicki et al, 1984). Much of the nitrogen may not be
available for plant use if hydraulic loadings cause the highly soluble NO,3 to be leached below
plant roots, (EPA, 1977).

Septic tanks operation
Septic tank effluent carries a variety of organic and inorganic chemicals into the
subsurface environment. Chemical components are retained in the soil, assimilated biologically
or transported by groundwater movement. Satisfactory effluent treatment depends on numerous
factors including:
characteristics of the wastewater
design of system components
construction techniques
rate of hydraulic loading
age of system
periodic maintenance
morphological, chemical, and physical properties of the soil

STSA's can provide varying levels of wastewater treatment. Proper STSA functioning is
achieved only if a sufficient volume of aerobic, unsaturated soil is available to absorb the volume
of effluent. It should be noted that a "properly" functioning septic system discharges significant
quantities of NO3- into the subsurface environment, (Bicki et al., 1984, Wilhelm et al., 1994).
Most STSA constructions provide two zones where reduction/oxidation redoxx) reactions
occur. The first redox zone is the anaerobic septic tank. Most of the nitrogen (75-85%, Bicki et
al., 1984) is released from the septic tank in the reduced NH' form. Septic tank retention times
average about three days. Effluent is distributed to the drainfield with a typical areal loading of
one to five cm/day.
The second redox zone is the aerobic drainfield. The typical O2 demand exerted by
wastewater ranges from 400 to 1500 mg/1. NH4 oxidation in the drainfield results in N03
concentrations roughly two to seven times the drinking water limit (10 mg/l as nitrogen
equivalent to 45 mg/1 as NO3-). Almost complete oxidation of NH4+ can occur within 1m of the
distribution pipes and within a few hours exposure to oxygen (Bicki et al, 1984, Walker et al.,
1973). The predicted nitrogen specie under drainfields are: nitrate in sandy soils, a mixture of
nitrate and ammonium in silt loams, and ammonium in clays, (Bicki et al., 1984).
The largest changes in wastewater composition occur in the drainfield. As NH4 is
oxidized to N03, organic carbon in wastewater is oxidized to CO,. CO2 production can lower
pH if the CO2 remains in solution. A lowered pH may cause significant dissolution of calcium
carbonate (CaCO3) over an extended period, (Wilhelm et al., 1994).
If the native sediments in the drainfield are finer than the gravel surrounding the
distribution pipes, a two to five cm thick mat will form beneath the gravel. This "biomat" is
generally moist and has a high 0, demand, resulting in local anaerobic conditions. Biomat


formation will alter the hydraulic conductivity in the drainfield. Drainfield specifications
account for the permeability of the natural soil, but not for the permeability of the biomat.
Some STSA configurations provide a third redox zone in which denitrification occurs
under anaerobic conditions. Most sites do not have the geology necessary for a third redox zone.
Denitrification requires an approximate one to one ratio between organic carbon and N03-.
Organic carbon in the wastewater will have oxidized by this point. Long term denitrification
requires significant amounts of available organic carbon, (Wilhelm et al., 1994).
STSA hydraulic loading and precipitation recharge will cause downward movement of
any mobile constituents. N03 contamination tends to remain as a distinct plume due to the
typically laminar flow of groundwater, (Ayres and Associates, 1993). Denitrification in the
saturated zone has not been found to be significant. It is thought that the organic carbon present
is recalcitrant and therefore unavailable for microbial utilization.
Limited denitrification and low dispersivity in the saturated zone can result in locally
elevated concentrations of N03-. Harman et al. (1996) delineated 110 m of a septic tank plume
in an unconfined sand aquifer. NO3- concentrations exceeded drinking water standards over the
entire 110 m. Postma et al. (1992) detected average concentrations in the 20 50 mg N03 /1
range, with a peak concentration of 115 mg N03 /1 2 m downgradient from a drainfield receiving
effluent with 100 mg NO3- /1. In a laboratory study Willman et al. (1981) found that 77-92% of
an initial 178 mg NH4I/ was converted to NO3 in 60 cm soil columns containing limestone/shale
sand. Limestone columns with 0, 3, 6, and 12% clay reached a steady-state concentration near
150 mg N0311 (Willman et al., 1981). It is estimated that 20-40% of the nitrogen is removed
from the effluent as it percolates through the soil (Bicki et al., 1984, Ayres and Associates,

Septic tank regulation
To prevent groundwater contamination, current regulations (FL Administrative Code
Chap. 10D-6, 1985) specify a 24" separation distance between the bottom of the adsorbtion
system and the seasonal high groundwater. A 24" separation distance may be inadequate. In-
situ GW monitoring studies and lab column studies support a minimum depth of 36". STSA's
may impact more severely in some locations than in others due to differences in land use, soil-
water-landscape relations, STSA density, and recharge capability of local aquifers, (Bicki, et al,

Assimilative capacity of the environment
Environmental assimilation of STSA effluent is a function of hydrology, geology, and
biology. The hydrologic and biologic components will exhibit some degree of seasonal
variation. Water provides both N03' transport and moisture to create saturated, anaerobic soil
conditions. Water table fluctuations will cause alternating aerobic/anaerobic regimes.
Denitrification may be enhanced or hindered depending on the frequency and amplitude of water
table fluctuation. Greater groundwater velocities may enhance dilution of an N03- plume.
Natural processes governed by the vegetative cover on the ground surface can affect
contaminant concentrations in a phreatic aquifer system, such as the surficial aquifer in Wakulla
County. The amount of rainfall recharge to groundwater will be affected by the plant
evapotranspiration and by soil permeability. The quantity and quality of vegetative litter


deposited on the surface may affect soil characteristics such as permeability and chemical
Biologically, an overabundance of N03 may cause eutrophication of water bodies and
shifts in species composition. Microbes are typically carbon limited in the saturated zone. In
some areas such as organic riparian soils microbes may be NO,3 limited. Increases in N03'
loading may result in development of a microbial community with the ability to denitrify large
amounts of NO,3. The presence of denitrifying bacteria does not necessarily indicate the
occurrence of denitrification, (USGS, 1994).
The geologic composition of the subsurface will determine the assimilative capacity of
the soils. Soils with higher clay content will have a larger cation exchange capacity (CEC) and
will sequester a greater amount of NH4 than more sandy soils. Willman et al. (1981) found that
laboratory columns of limestone sand retained NH4, for four to nine weeks. CEC can eventually
be exceeded and the soil will no longer retard movement of NH4+, (Canter, 1996).
Wakulla County is in the Gulf Coastal Lowlands physiographic province. The county
can be further subdivided into the Woodville Karst Plain and the Apalachicola Coastal
Lowlands. The study area for this project is located in the Woodville Karst Plain. The
sediments at the surface and near the surface are made up of quartz sand that is not more than 7
m thick. The sands overlie a karstic, early Miocene limestone, (SCS, 1991). Soils in the
Woodville Karst Plain do not appear to contain the necessary components for significant
Ayres and Associates (1987) determined that 65% of the soils in Wakulla County had
severe to very severe soil limitations for OSDS use. Only 5% had slight limitations. The soils
within the five sub-basin study area could generally be considered to be severely limited for
OSDS use, (SCS, 1991).

Groundwater models can be generally classified in two categories flow and solute
transport models. Flow models are concerned with groundwater movement while solute
transport models describe the movement of constituents within the water. These two general
categories can be further subdivided as either analytical or numerical models.
Analytical models describe the water behavior in an aquifer with differential equations
derived from basic principles such as the laws of continuity and conservation of energy (Canter
and Knox, 1984). The application of analytical models in a heterogeneous environment is
limited by their inability to account for parameter variability.
Numerical models approximate analytical solutions as a system of algebraic equations, or
alternatively by simulating transport through the spread of a large number of moving reference
particles (Domenico and Schwartz, 1997). Numerical approaches can accommodate parameter
variation and thus allow modeling of more complex geometries in multiple dimensions.

Contaminant transport models usually begin with some form of mass transport equation:

mass inflow rate mass outflow rate = change in mass storage with time

The mass transport equation is then modified to account for processes such as chemical
reactions, biological uptake, and molecular and mechanical dispersion.



Analytical model

One-dimensional mass transport involving a first-order kinetic reaction can be described as:

C/C0 = (1/2) exp {(x/2acx)[1-(1+(4Xac)/v)"2]} erfc [(x-vt(4Xc,)/v)`a/2(axvt)'2]

where: C = constituent concentration (mg/L);
Co = constituent concentration at time to= 0 (mg/L);
x = distance (m);
= longitudinal dispersion coefficient (m);
X = decay constant (1/day);
v = groundwater velocity (m/day);
t = time (day);
erfc is the complimentary error function.

The above equation approaches a steady state as the argument of erfc approaches negative two.
Assuming a continuous source, the steady-state solution to the above equation is:

C/Co = exp {(x/2a,)[l-(l+(4ta,)/v)"/2]} (2)

Using data compiled by Gelhar et al (1992), an ac of 10m was selected to characterize
longitudinal dispersion in the surficial aquifer.

Spatial implementation

A raster-based geographic information system (GIS) was used to manipulate digital map
layers to solve the above one-dimensional analytical equation over two-dimensional space. A
cell resolution of 30m X 30m was used for all maps.
A coarse digital elevation model (DEM) was created by interpolation between 1:100,000
ft scale contours (See Figure 3). The interpolation results were modified to reflect an average
north-south slope of 4 feet per mile (SCS, 1991).
The average slope within each cell of the DEM coverage was determined. It was
assumed that the phreatic surface within the surficial aquifer is generally a subdued reflection of
the land surface (USGS, 1991). Surface slope in this case can be interpreted as the change in
hydraulic head with respect to distance 6h/6x.
A general soils map of the area was used to generate soil hydraulic conductivity and soil
carbon content coverages (See Figure 3). Darcian velocities (q) within the surficial aquifer were
determined using Darcy's Law:

Draft... 9/14/98


Digital Elevation Map of the Study Area

Elevation in meters

Study Area Soil Characteristics

Soil Drainage and Type:

[ 1 Excessive to moderate sandy soils
[ I Moderate to excessive sandy soils
SModerate sandy soils
Poor to moderate sandy soils
Poor sandy soils
Poor to very poor sandy soils
Very poor organic soils

Figure 3. Digital Elevation Model (top map), and Soils coverage (bottom map) of the
study area.

Draft...9/14/98 311


q = (-K/n) (6h/6x)

where: K = hydraulic conductivity (m/day);
n = effective soil porosity dimensionlesss ratio).

Assuming an average porosity of 0.3, Darcy's Law was solved spatially through algebraic
manipulation of the polygon layers of hydraulic conductivities and surface slope map. The
operation resulted in a map with each cell representing the Darcian velocity based on the
pertinent parameters at the cell's location. This map was used as "v" in the analytical equation.
Additional simulations were conducted with velocities double and half the initial "v" value.
Water surface slope (DEM) and hydraulic conductivity layers were used to generate
flowpaths radiating away from the residential land use polygons. Flowpaths were generated
from every cell on the perimeter of a residential land use polygon. The resulting flowpath
coverage contained linearly distributed values increasing away from the land use polygons.
Flowpath values were used as linear x coordinates along which the one-dimensional equation
was applied. Flowpaths were immediately adjacent to each other, thus allowing a two-
dimensional simulation with a one-dimensional equation.
Research has shown that denitrification rates can be correlated with aoil organic carbon
content, (Groffman and Tiedje, 1989, Tsai, 1989, Pratt et al., 1979, Yan, 1995). The soil carbon
content layer was entered into a model of denitrification rates for sandy Florida soils (Tsai, 1989)
to develop an initial estimate of potential denitrification rates for the soils in the study area. The
potential denitrification rates were compared against an average initial NO3- concentration to
develop a coverage representing X in the analytical equation. Several simulations were
conducted using a range of values for k.
Accurate estimation of longitudinal dispersion is not possible without field studies. An
average longitudinal dispersion coefficient of 10 m was selected based on data compiled by
Gelhar et al. (1992). The land use polygons were simulated as an areal source (Mooseburer and
Wood, 1980) with an initial N03- concentration of 140 mg/1. Simulation under these conditions
represents a steady-state scenario. It is important to note that the output maps depict potential
maximum concentrations within each cell.


Figure 4 presents simulation results obtained using parameter values believed to be
representative of conditions in the study area. The gradation of concentrations is not apparent at
this map scale. Figure 5 shows details of NO3- attenuation in cells on the parameter of an area
that is a source of N from septic tanks. The flow is right to left from the upper right hand corer.
The majority of NO3- attenuation occurs within the first few cells surrounding the land use
Figure 6 is a bar graph showing the fraction of the study area having N03- concentrations
within given ranges. Six categories are presented in the graph legend Present/Future -
low/median/high. The low, median, and high terms refer to the flow rate used in obtaining the
results. The median flow rate is the one believed to best characterize conditions in the study
area. The low flow rate is half of the median and the high flow rate is double the median. Note



Potential Maximum N03-
Concentrations Under Present
Zoning Plan

Unaffected by Residential Septic Tanks
m 1 2 mg N03 / I
70 mg N03 / I
S 140 mg NO3 /I

Potential Maximum N03-
Concentrations Under Future
Zoning Plan

Figure 4. Spatial simulation results of N03O loading the surficial aquifer from septic
tanks. Top map shows present conditions and the bottom map shows conditions based on
future land use.




Figure 5. Details of the spatial simulation of nitrogen loading and attenuation of NO3- in
an area adjacent to a septic tank. The upper right hand corer is the effluent source and
flow is right to left..

Draft... 9/14/98


Figure 7: Effect of Variations in Denitrification Rate

50 M Present-high
I Future-high
40 0 Present-median
orU Future-median
0o U Present-law
-30 O Future-low



140 105-139.9 70-104.9 35-69.9 2-34.9 <2 >101 <140 All
Concentration Range (mgIl)

Figure 6. Graph of the percent of the study area influenced by various NO;
concentrations when uptake rate is held constant and ground water flow rate is varied one
order of magnitude above and below median flow rate.



that percent area affected increases with increasing flow rate. Figure 7 presents results from
simulations holding flow rate constant ( median flow rate) and varying the NO3- uptake rates.
The median uptake is believed to best characterize conditions in the study area. The low uptake
rate is an order of magnitude lower than the median and the high uptake rate is an order of
magnitude greater.


The simulation results in Figure 4 contain some anomalies resulting from the input data.
As explained below, use of the DEM as a phreatic surface generated questionable groundwater
velocities and isolated flowpaths in some small areas. The most striking feature of the maps is
the extent of affected areas in the future scenario.
Ambient N03 concentrations in this area are believed to be 2 mg N03/1 or less (USGS,
1994, 1996a). The majority of the yellow areas in.the maps represent ambient concentrations.
Figure 5 provides a closer look at a plume segment to illustrate the gradation not visible on the
maps. Initial concentrations of 140 mg N3-/l are found under the land use polygons only. The
percent area affected by 140 mg N0311 is equivalent to the percent area zoned for residential
development (See Figures 6, and 7).
The majority of NO3- attenuation occurs within the first few cells surrounding the land
use polygons. Concentrations generally drop to 2 mg N0311 within 30 to 90 m's. In Figure 6,
the fraction area affected generally increases with an increase in flow velocity. Percent area
affected by 2 mg N03/1 decreases with increasing flow velocities under the future zoning plan.
The decrease at this concentration range is attributable to increases in other concentration ranges.
In the greater-than-10 / less-than-140 mg N03/1 range the "Present-low velocity" shows a larger
fraction affected area than the "Future-low velocity." The difference is due to a larger fraction of
residential area and therefore a larger fraction being affected by 140 mg NO3/..
Variations in NO3- uptake rate show as similar trend as with variations in flow rate. No
variations were made between the relative uptake rates of the soil types. Higher degrees of
variation in uptake rates between soil types may exist. More variation was expected due to the
spatial heterogeneity of the soils. Use of a more detailed soils coverage may result in greater
The model results give an indication of potential STSA NO3- loading to the surficial
aquifer system and the effects of variation in flow and N03 uptake rates. Further modifications
could allow the model to be used as a screening tool to highlight areas at potential risk of
contamination. Modifications would require extensive data sets and a comprehensive
hydrological analysis.
Use of a DEM to represent the phreatic surface is adequate for a first attempt at
modeling. DEM derived values will under- and overestimate groundwater slope depending on
local conditions. A karst aquifer system, such as that in the Woodville Karst Plain, has an
intimate relationship between surface and groundwater. Both flow regimes need to be
considered for a full analysis.
The springs and sinkholes present in the study area create highly variable flow patterns
that shift in response to storm events. Karst aquifers often have discrete groundwater basins that
receive recharge from the land surface through sinkholes and sinking streams. Basin divides are
determined by highs in the water table. These boundaries may shift with the water table. Each



Figure 6: Impact of Flowrate Variations on % Total Area Affected

B Future-high

140 105-139.9 70-104.9 35-69.9 2-34.9 <2 >101<140 All
Concentration range (mgIl)

Figure 7. Graph of the percent of the study area influenced by various NO3
concentrations when groundwater flow rate is held constant and nitrogen uptake rate is
varied one order of magnitude above and below median uptake.



groundwater basin typically drains to a spring or group of springs. Groundwater basins usually
extend beyond surface basin boundaries and groundwater flow may not be in the same direction
as surface water flow (White 1993).
The estimates of N03O uptake used in the model are unverified. Wide fluctuations in the
Wakulla County water table may result in significant changes in denitrification rates on a
seasonal basis. NO3- reaching shallow aquifer systems, especially those with rapidly fluctuating
water tables, has a good chance for removal by denitrification or uptake by deeply rooted
vegetation (Lowrance and Pionke, 1989). Data are not available to verify the model. Local
testing is necessary to characterize nitrogen dynamics within the soil system.
The high spatial and temporal variability of governing parameters particularly
hydrology poses a significant challenge to modeling subsurface chemical dynamics in Wakulla
County. Isolating and modeling the effect of one nitrogen input (e.g. STSA's) within a karst
system requires either broad generalizations or a considerable amount of data. Canter and Knox
(1984) described the application of groundwater models to septic tanks as "disappointing and
frustrating." Model calibration is difficult due to the lack or questionable validity of input data.
Traditional hydrologic modeling such as the approach used in this study may not be efficient.
Intensive data gathering and complex modeling may yield no better than general results.
Hydrologic simulations are adapted to a larger time scale (annual) then physicochemical
processes (days) (Caussade and Pratt, 1990). It does not seem possible to quantify N03-
contamination at a 30m X 30m cell size without a large amount of data gathering. Work has
begun on a more general modeling approach that will allow the use of minimal data sets
derived from satellite images to evaluate land use decisions.

Draft... 9/14/98


University of Florida Home Page
© 2004 - 2011 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated May 24, 2011 - Version 3.0.0 - mvs