United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
Research and Development
EPA-600/S3-84-078 Sept 1984
Project Summary
Receiving Water Quality
Database for Testing of
Mathematical Models
Wayne C. Huber, David F. Maclntyre, and James P. Heaney
Many mathematical models exist for
simulation of quantity and quality
parameters of receiving waters. Such
models are frequently used in the
evaluation of effects on receiving
waters of pollution control alternatives
such as advanced waste treatment and
non-point source runoff abatement
practices. Data for testing of such
models, however, are hard to obtain.
This project has assembled detailed
data sets, sufficient for model calibration
and verification, for four rivers, two
lakes and one estuary: Otter Creek,
Vermont; Winooski River, Vermont;
Chattahoochee River, Georgia; Lower
Fox River, Wisconsin; Lake Okeechobee,
Florida; Lake Jackson, Florida; Potomac
Estuary, Maryland and Virginia. The
data—contained in a report, on magnetic
tapes and in addenda—include physical
descriptions (e.g., reach lengths, cross
sections), hydrologic and hydraulic
information, inflows and outflows,
pollutant loads, and in-stream concen-
trations.
This Project Summary was developed
by EPA's Environmental Research
Laboratory, Athens, GA, to announce
key findings of the research project that
is fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
Properly formulated and operated mathe-
matical models, when coupled with
appropriate data for calibration and
verification, are tools of tremendous
importance as aids in decision making for
maintenance of receiving water quality.
For example, models can aid in the
evaluation of effects on receiving waters
of advanced waste treatment and non-
point source runoff controls.
This project focused on the collection of
data for proper validation of mathematical
representations of actual receiving water
processes as well as for calibration
(parameter adjustment) and verification
(a check on previous parameter adjust-
ments using new data) of models. The
results are documented in the project
report, with most of the data points
available on magnetic tapes. The project
focused on use of only a few good, well-
documented sites, rather than inclusion
of several sites for which only sketchy
documentation would be available.
Data Sources
Many different groups were approached
for data during the course of the project,
of which most possessed candidate data
sets. Major contributions were made by
several offices of the U.S. Environmental
Protection Agency (EPA), by the Geologi-
cal Survey, U.S Department of Interior
(USGS), and by the Corps of Engineers,
U.S. Army. Other contributors included
several state "environmental regulation"
departments, river basin commissions,
councils of governments, water manage-
ment districts, universities, consultants,
municipalities and the National Council
for Air and Stream Improvement. Not all
of these groups possessed complete data
sets, but many contributed information to
add to other data sets.
Site Characteristics
The seven sites included in the database
are discussed briefly below, and their
characteristics are summarized in Table 1
(rivers), Table 2 (lakes) and Table 3
(Potomac Estuary).
-------
Table 1.
Characteristics of the Selected River Locations
River
Name
Otter
Creek
Length
of
Study 7010 No of
Section Flow Point Parameters
(miles) (cfsj Sources Measured
21 79 5 Temperature. D O , SODs.
/VO3-/VO2-/V, NH3-N.
TKN
Approximate
Frequency &
Duration of
Measurements
4 hourly
(for 3
days)
No of
No of Independent
Stations Data Sets Other Data****
22 2 Point source BOD,
NO3-NOs, NHs, TKN
loads Stream bed
profile Daily
precipitation +
max & mm air
temperatures
Upper
Wmoosk/
River
Chattahoo-
chee River
7.6
60
43
980*
Lower
Fox
River
39
950
33
Temperature, D 0 , BOD5. 4 hourly 12
NOz-NOi-N. NH3- (for 2-
N, TKN (Ortho-P 3 days)
Total-P. Chloro-
phyll a, one data
set only)
Temperature. D 0 . BOD^ 1-9 per 31
pH, Total-N, Organ- day (for
ic-N, NH3-N, NOt- 1-4 days)
N. Jotal-P. Ortho-P,
Fecal Strep, Trace
metals. Suspended
solids**
Temperature, D 0. BODs. Daily (for 49
Secchi depth. Organ- 1 day)
ic N. NOyNOi-N.
Total-P, Soluble-P,
% volatile solids***
Point source BOD,
NO3-NO}-N, NH3-N
TKN loads Stream
bed profile Daily
precipitation +
max mm air temperatures
River bed profile.
41 river cross-
sections 41
Mannings Coeffs
limited land use
data.
49 mean cross-
sectional depths.
River bed pro-
file Point
source BOD loads
Table 2.
Characteristics of the Selected Lake Locations
Lake Okeechobee
Lake Jackson
"Estimate Flow is regulated
**Some parameters not measured at all stations
***Not all measured in all surveys Some surveys contain only temperature & D 0
****Measured flow rates are available for all except the Lower-Fox River, which was estimated Measured and/or estimated flow velocities are
available for all rivers
In general, suitable data for rivers and
streams are plentiful; these data were the
easiest to obtain for the project, and
several alternative data locations exist
The four selected river sites, which were
among the best documented of those
encountered during the project, were
chosen so that both small and large rivers
would be represented.
Although lake data are numerous, well-
documented comprehensive studies are
not. The two lakes selected were chosen
primarily because of their comprehensive
nature and proximity to the University of
Florida, which made it possible to obtain
the necessary ancillary information Most
other comprehensive lake studies (e g ,
Lake George in New York) are not
concisely documented and/or have non-
computerized data sets.
Sites for estuaries and bays have
similar problems to those for lakes The
selected site (the Potomac Estuary)
possesses an enormous history of studies
and data, but has the advantage of a
recent, computerized database Although
the Delaware Estuary also has a long
history of water quantity and quality
studies, it has not received as much
recent attention as the Potomac, and its
Watershed Area*
(sq miles)
Lake Area
(sq miles)
Mean Depth
(feet)
Trophic State
Residence Time
(years)
Parameters
Measured
4,600
706
92
Eutrophic
1 0
Temperature, D O , Specific Conduc-
tance, pH, Secchi Depth, Tur-
422
625**
56**
Mesotrophic
0 7*«
Temperature, pH, alkalinity, tur-
bidity, suspended solids.
bidity. Color, Total Suspended
Solids, Ortho-P, Total-P, NOt-
N. N02-N. NO3-N, NHA-N, TKN-
NHt-N, Total N. Total Fe. A~
Alkalinity
Secchi Depth, Specific
Conductance, Color,
NO^-N, NH3-N, Ortho-P, Total-
P, Total dissolved P, D 0,
(+ some chloride & sulfate)
Approximate
Frequency &
Duration of
Measurements
Number of
Stations
Biweekly to monthly (for 7
years)
8 & 40
Monthly (4 studies covering
10 years)
10 (not in same position
for all studies)
*1 square mile - 640 acres.
* These figures are based on a stage of 87 ft-MSL In recent years the stage level has varied
considerably
2
-------
Table 3.
Characteristics of Potomac Estuary Location
Length (milesj
Average Flow (cfs)
Point Sources
Non-Point Sources
Parameters
Frequency and
Duration
Number of Stations
Other Data
117
10.000
13 POTW's monitored and
estimated
CSO's monitored in D C,
others estimated
Temperature, D 0 , Salinity,
BODs. Nutrients, misc
1968-1981, intensive,
1979-80, weekly plus
some storm event and
diet
25 EPA, 34 USGS
Miscellaneous cross sections,
meteorological,
navigation, maps, etc
data are mostly contained in the STORE!
files During the course of this project it
was found that documentation of even
one estuary of the degree of complexity of
the Potomac was a large task, hence, only
one estuary site was included
Otter Creek, Vermont
Otter Creek is a stream in the Champlam
Valley in western Vermont It is about 100
miles long and empties into Lake Cham-
plain. Intensive surveys during the low-
flow conditions on August 1 -3, 1977, and
August 1-3, 1978, were performed as
part of a wasteload allocation study by the
State of Vermont Agency of Environmental
Conservation on a 21 -mile segment of
the stream
Upper Winooski River,
Vermont
The Winooski River flows from Wash-
ington County westwards through Mont-
peher to Lake Champlam It is about 90
miles long and has a drainage area of
1080 square miles. The study area
contains 3.4 miles of the Stevens Branch
immediately upstream of its junction with
the Winooski River, and 4.2 miles of the
Winooski River from just above its
junction with the Stevens Branch down-
stream through Montpelier This section
of the Winooski has two small tributaries
in addition to the Stevens Branch
The studies on this river also were
performed as part of a wasteload allocation
study by the State of Vermont Agency of
Environmental Conservation. Intensive
water quality surveys were performed
under low-flow conditions on August 22-
24, 1978, and July 9-11, 1979.
Chattahoochee River, Georgia
The Chattahoochee River flows south-
wards from the mountains of north
Georgia to Lake Semmole on the Georgia-
Florida border The section used in this
study is a 43-mile segment from Atlanta
downstream to Whitesburg In addition to
effluent from seven sewage treatment
plants, this segment receives runoff from
urban and cultivated areas Several small
tributaries enter the river Extensive
water quality data are available on this
segment of the Chattahoochee River,
collected by the USGS, State of Georgia
and others. The main difficulty is to
choose a cohesive data set Four low flow
studies from 1976 and 1977 that have been
used for model testing by the USGS and
others were selected
Lower Fox River, Wisconsin
The Lower Fox River is 38.9 miles long
and flows from Lake Wmnebagoto Green
Bay. The river is heavily utilized, receiving
effluent from 32 sources, including 13
sewage treatment plants Five small
tributaries enter the river and water is
withdrawn at 15 points (mostly for
industrial use). The data presented in the
full report are drawn from a wasteload
allocation study bythe Wisconsin Depart-
ment of Natural Resources from 1 972 to
1977
Lake Okeechobee, Florida
Lake Okeechobee is situated in south
Florida, north of the Everglades. With an
area of 706 square miles, it is the second
largest freshwater lake in the United
States. This eutrophic lake is surrounded
by a large dike to protect surrounding
areas from flooding during a hurricane.
All inflows and outflows are controlled as
they pass through the dike so that the lake
level can be regulated. The water budget
for the lake is not well determined,
however, because of difficulties in
calculating the amount of precipitation
and interactions with ground water
Lake water quality was monitored
extensively from 1973 to 1980 by the
South Florida Water Management District,
which is responsible for regulation of the
lake. Input-output, systems, and complex
hydrodynamic models have been applied
to Lake Okeechobee The dynamics of
nutrient cycles in the lake have been
investigated, and some spatially lumped
models for nitrogen and phosphorus have
been developed.
Lake Jackson, Florida
Lake Jackson is situated on the
outskirts of Tallahassee in northwest
Florida. This mesotrophic lake has an
area of 4,000 acres and is situated in a
watershed of 27,500 acres. The lake is
largely flat bottomed, and few areas are
deeper than 14 feet. There are no exit
channels from the lake, so that the only
inputs are rainfall and runoff, and the
only outlets are evaporation and ground-
water recharge. The hydrologic history
shows wide fluctuations in the lake level
in response to annual rainfall. The data
presented in the full report are based on
studies by several Florida agencies from
1971 to 1981
Potomac Estuary, Washington,
D.C.
The Potomac Estuary extends 117 miles
from Chain Bridge in Washington, D.C., to
Chesapeake Bay. The estuary is well
mixed vertically so that saline wedge
effects rarely occur. Mathematical model-
ing of the Potomac Estuary was begun in
the 1960s by predecessor agencies to the
EPA, and many programs of data collec-
tion have been reported The selected
period (1979-1981) includes intensive
and synoptic studies sponsored by the
USGS, EPA and Washington, D.C.,
Council of Governments Modeling acti-
vity on the Potomac is also extensive
Other Locations
Alternative data locations also dis-
cussed in the full report include' Willa-
mette River, Oregon; Arkansas River,
Colorado; Ouachita River, Arkansas and
Louisiana, Lake George, New York,
Onondaga Lake, New York, Delaware
Estuary, and San Francisco Bay.
Database Format
Site descriptions, maps, pollutant
sources, rate constants, etc., are given in
the full report as much as is possible.
Measured receiving water quality data
values are presented on magnetic tapes
Modeling data for several sites are also
3
-------
included on the magnetic tape available
from the EPA's Environmental Research
Laboratory, Athens, GA. In some instances
(eg, the Fox River) the modeling data
also serve to document point and non-
point source loads to the receiving water.
In a few instances, some useful but bulky
information (e.g., stream cross sections)
is available as an addendum to the full
report This information has been retained
in files at the University of Florida
Sufficiency of Project Data for
Model Testing
Can the information supplied for the
seven sites by this project be used by
itself for model testing? Probably not.
Considering that most sites are documented
with multiple reports of hundreds of
pages, it is unrealistic to assume that all
the information anyone would need
about a particular site could be included
in a single report. Modelers will want to
obtain some of the references listed for a
site in order to obtain needed information,
although it is intended that the material
presented in this project could certainly
initiate a modeling study. In addition to
the site summaries and references, the
primary value of this project is the
presentation of the voluminous m-stream
data in a machine readable format on
magnetic tapes This should eliminate a
considerable task of most modeling
projects.
W. C. Huber, D. F. Maclntyre, and J. P. Heaney are with the University of Florida,
Gainesville, FL 32611.
T. O. Barnwell, Jr., is the EPA Project Officer (see below).
The complete report, entitled "Receiving Water Quality Database for Testing of
Mathematical Models." (Order No. PB 84-220 300; Cost: $23.50, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
Athens, GA 30613
U S GOVERNMENT PRINTING OFFICE, 1984 — 759-016/7802
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
•' I
-.'.'!•>!"' 14
------- |