Region III Library
Environmental Protection Agency
SAMPLING PROGRAM FOR
NUTRIENT TRANSPORT INTO SUBESTUARIES
OF THE
CHESAPEAKE BAY
prepared for
Environmental Research Laboratory
Environmental Protection Agency
Athens, Georgia
Chesapeake Bay Program
Region III, EPA
State of Maryland
State of Virginia
prepared by
Hydrocornp Inc.
1502 Page rill Road
Palo Alto, California 94304
Julv 1978
EPA Report Collection
information Resource Center
US EPA Region 3
Philadelphia, PA 19107
-------
-------
L.0 INTRODUCTION
This report describes the preliminary design of sampling programs to
characterize nonpoint pollutant transport into selected subestuaries of
Chesapeake Bay. The proposed sampling programs are part of a comprehensive
eutrophication study of the Hay sponsored through the cooperative efforts of
the U.S. Environmental Protection Agency and the States of Maryland and
Virginia. Specifications for the sampling program described herein have been
coordinated with EPA, the States of Maryland and Virginia, and the
consultant that is responsible for the sampling program on the subestuaries.
Data collected from the nonpoint sampling programs, in conjunction with data
from estuarine sampling programs, will be the basis for calibration and
verification of nonpoint source, stream, and estuarine water quality models.
The data from the selected tributaries and subestuaries with representative
soils and lard use will be used to extrapolate model parameters for the
entire region. The models will be the tools for analyzing nonpoint nutrient
loadings and the impacts of land use changes and management practices on the
eutrophication of Chesapeake Bay.
The preliminary design of the nonpoint sampling program includes the
recommended period and frequency of sampling, constituents to be analyzed,
sampling techniques, and guidelines.for selection of sampling locations.
These guidelines refer to the characteristics (size, soils, land use) of the
sampled watersheds to insure representation of nonpoint loadings. Prior to
implementation, a detailed design is needed to select specific watershed
sites, sampling equipment, analytical techniques, and quality control
specifications. The Chesapeake Bay Program has developed standard quality
control specifications; analogous standards are needed for sampling
equipment and analytical techniques to guarantee valid comparisons of data
:o]lected at the different sampling sites.
-------
-------
2.0 SELECTED WATERSHED AND SU3ESTUARISS
The watersheds and subestuaries selected for the nonpoint sampling program
include the Chester and Patuxent Rivers in Maryland and the Cccoquan,
PC quo son, and Ware Rivers in Virginia. Figure 1 shows the location of these
watersheds and their outlets to Chesapeake Bay.
The Chester River is located on the eastern shore of Chesapeake Bay and
drains an area of 440 sq mi. The watershed is predominantly agriculture
•with Coastal Plain soils and mostly tidal streams. The subestuary has
experienced shellfish and aquatic plant problems, and is undergoing studies
by the State of Maryland.
The Patuxent River is the largest of the chosen watersheds covering an area
of 930 sq mi on the western shore of the Bay. The watershed includes upland
streams, reservoirs, tidal streams, and a long estuary. Both Piedmont and
Coastal Plain soils are represented in a variety of changing land uses.
Portions of the watershed are heavily urbanized since it is located on the
urban fringe of the Washington Metropolitan area. The river experiences
high sediment and nutrient loads that impact the shellfish and finfish
industries. Extensive studies are being conducted by various departments of
the State of Maryland.
The Poquoson and Ware Rivers are the smallest of the selected watersheds
draining areas of 21 sq mi and 67 sq mi, respectively, on the southwestern
shore of Chesapeake Bay. The Poquoson is a branching estuary with mostly
marsh, forest, and agricultural lands and some rapidly urbanizing areas.
The Ware River watershed also is mostly tidal with forest and agriculture
land use. Residential development is minimal. Both rivers experience high
coliform counts and problems with septic tank waste. Shellfish beds are
":''03-r-d in both subestuaries. 'The Poqucson River is undergoing studies by
"-- -'.'rrjinia State Water Control Branch and other state agencies.
-------
-------
Washingtcxi DC V
'
Atlantic
Ocean
10 miles-
Ficrure 1. Locations of test watershed;
Chesapeake
Bay Study
Hyd
r o c o m p
-------
-------
The Cccoquan River drains 575 sq mi, including a large fresh water
reservoir, before entering the Potomac River. The Piedmont soils of the
Occoquan support several distinct land uses including forest, agriculture,
high density residential, and commercial. Numerous point and nonpoint
pollutant sources are present in the watershed which has been the subject of
an extensive 2CS study including water quality sampling and modeling.
-------
-------
3.0 NCNPOINT SAMPLING PROGRAM DESIGN
A summary description of the recommended sampling program is provided in
Table 1. The primary goal of the program is to characterize the nonpoint
nutrient transport into subestuaries of Chesapeake Bay as part of the
eutrophication study of the Bay. Although the resulting data will be used
for other studies and analyses, the program was specifically designed for
this purpose. Each of the major topics in Table I is discussed below.
3.1 Sampling Equipment
The central components of the sampling program are automated samplers
designed to collect either a single volume-integrated sample or discrete
samples by incremental charges in stage. The choice of a specific sampler
will be made in the final program design. Continuous recording flow (stage)
gages at each site will provide a continuous flow record and indicate the
time period (or periods) when samples are obtained. Each site will include
at least two continuous recording rain gages located at or near the sampling
site and upstream within the drainage area of the site.
3.2 Land Use Sampling Site
Twelve sampling sites are recommended in each state distributed among the
five watersheds and specific land uses as described in Table 1.- Each
sampling site should drain an area of at least 200 acres but not more than
1000 acres to insure that pollutant loads measured at the sites are derived
from nonpoint sources alone without the interference of point sources and
Lnstream reactions. Each site should drain one dominant land use to allow
extrapolation of pollutant loadings and model parameters.
'""^ major land uses represented by the sar.pling sites include urban
'v"'~id< r.tiol, ccr^nercial, industrial), forescs, pastures, idle land,
-------
-------
Table 1. SUMMARY DESCRIPTION OF THE SAMPLING PROGRAM FOR NONPOTNT SOURCE
TRANSPORT INTO CHESAPEAKE BAY
Equipment: " Automatic samplers actuated by stage
" Continuous recording flow (stage) meters
* Recording rain gages at sampling site or upstream
Watershed/Land Use Sampling Sites:
Watershed Land Use Sampling Sites
Chester River, MD - 6 sites 1 forest, 1 urban, 1 idle, 2 cultivated
cropland (conservation and conventional tillage) ,
1" active pasture
Patuxent River, MD - 6 sites 3 urban (residential, commercial, and indus-
trial) , 1 forest, 1 pasture or idle, 1
cultivated cropland
Occoquan River, VA - 6 sites 1 forest, 1 idle, 2 cultivated cropland
(conservation and conventional tillage), 1
pasture, 1 construction
Poquoson River, VA - 3 sites 2 urban (residential, commercial) , 1 rural or
idle
Ware River, VA - 3 sites 1 forest, 1 cultivated cropland, 1 urban
Sampling Site Size: 200 acres to 1000 acres
Downstream sites: One downstream (main stem) sampling station of sane design for
each river.
Constituents: Nitrogen - Arrmonia, Nitrate/Nitrite, Total Kjeldahl Nitrogen (or
Organic Nitrogen)
Phosphorus - Orthophospfete, Total Phosphorus
Carbon - Total Organic Carbon, Carbonate, Bicarbonate
Other - Total Suspended Solids, Total Solids (suspended and
settleable), Carbonaceous BOD (20 days) , COD (optional),
DO (optional) , Water Temperature (optional)
All nitrogen and phosphorus constituents measured oh. filtered and unfiltered . :
samples.
Sampling Period and Frequency:
Two-year sampling period for all storms at each site.
Continuous flow and rainfall record for all storms.
Voiurs-integrated sa,~ple (by sampling equipment or in laboratory) for all
storms for constituent analysis.
"otailed sampling for pollufcograph definition for 2 to 5 storms per year.
Bimonthly grab sampling of low flows, analyzed for above constituents.
'nventory and selected sampling of point sources for characterization.
-------
-------
cultivated cropland (conservation and conventional tillage), and
construction. The two dominant soil classifications in the five watersheds
are the sandy Coastal Plain soils (Ware, Patuxent, Cnester, Poquoson) and
the Piedmont soils (Occoquan, Patuxent) with high silt and clay content.
Thus, the sampling sites were distributed so that the major land use and.
soils combinations are represented. In the watersheds where septic tank
problems exist (for example, Poquoson, Ware) the urban or idle land sampling
sites should be chosen to include septic tank areas. The specific sites
within each watershed will be selected in the final program design.
3.3 Downstream Sites
At least one sampling station is needed on the main stem of each river
downstream of the lard use sites. This station will be at a point that
drains nixed land uses ar.d will serve to integrate the separate nonpoint
source loads from each land use. For the nontidal portions of the larger
watersheds (that is, Occoquan, Patuxent), two or three stations will be
needed to quantify instream reactions. However, existing sampling stations
*•
or previous data might be used for these purposes. In the Poquoson River
Basin, the site should be replaced with a reservoir sampling site if the
detention of the Harwoods Mill Reservoir is significant with respect to the
entire travel time through the basin.
3.4 Constituents
The constituents to be analyzed are listed in Table 1. Nitrogenr phosphorus,
and carbon constituents were selected because the primary emphasis of
this study is eutrophication. Since Total Kjeldahl Nitrogen (TKN) is the
r.Mrr. of ammonia and organic nitrogen, the TKN test will be used for organic
-.'Brecon ,-ir.c? the 3r-T.cn ia ccr.por.ent will be analyzed separately. All
shoschorus constituents will be analyzed on both filtered and
-------
-------
unfiltered samples to determine the total and dissolved amounts.
Other recommended constituents include solids, BOD, COD, DO, and water
temperature. Total suspended solids and total solids (suspended and
settleable) are usually simulated by nonpoint models and are often the basis
for simulating other pollutants. Total solids refers to the total
particulate or sediment content, and particle size analysis (sand, silt,
clay fractions) may be needed if detailed sediment modeling is executed.
DO, carbonaceous BOD, and water temperature are conventional oxygen related
constituents that should be measured; DO and water temperature are optional
at the watershed sampling sites, but they should be measured at the downstream
sampling stations. COD is usually a more reliable measure of oxygen demand
so it is included as an optional check on BCD values. COD is not often
simulated by water quality models.
Other constituents, such as coliforms and heavy metals may be added to the
list at the discretion of the states if the data are to be used for analysis
of problems other than eutrcphication.
3.5 Sampling Period and Frequency
The sampling program is recommended for a continous 2-year period, sampling
all storms at each site. Accompanying continuous rainfall and flow records
are required. A volume-integration method of sampling is recommended to
provide a single composite sample for each storm event at each site. The
volume integration can be done manually in the laboratory or automatically
by the field equipment, depending on the laboratory and equipment
specifications as determined in the final design. The Chesapeake Bay region
will usually experience 20-25 storms per year requiring samples.
In addition to the volume-integrated samples, detailed samplirg for two to
five £-;or;7,s oer vear is needed at each site to define the shape of the
-------
-------
pollutograph. This will require two to three samples collected on the
rising limb of the hydrcgraph and two to three samples on the falling limb.
Stores sampled in detail should also have volume-integrated samples to
insure the accuracy and reliability of the volume-integration procedure.
Storms for detailed sampling should occur in different seasons, and should
folio-// different antecedent dry periods varying from 1 day to 1 to 2 weeks-
This will allow investigation of varying pollutant accumulation and seasonal
characteristics.
Bimonthly sampling of low flows analyzed for the same constituents is needed
to characterize pollutant contributions from groundwater. Three to four days
of stable flows should occur prior to sampling to insure a low flow
condition. This sampling may not be possible on some small watershed sites
that: experience only storm-related runoff without a continuous baseflow.
Sampling major point source discharges will be required if sufficient data
to characterize the discharge are not available on the constituents listed
in Table 1. Minor point sources discharges can be characterized from
industry type, manufacturing process, population (municipal effluent), etc.
-------
-------
t.C GUIDELINES FOR PROGRAM MODIFICATIONS
Tr.e reccm-ended sampling program should be reviewed and integrated with
sxls-irg or anticipated programs in each watershed to eliminate any
Duplication of effort. Sampling stations operated by other state or federal
Agencies (for example, USGS, Forest Service) could replace or supplement the
proposed station if the required constituents are analyzed. Any current
'.-st rumen, ted sites should be primary candidates for the sampling sites as
establishing new sites for this study.
[f budgetary considerations require a reduction in scope, reducing the
v_r.~-er of sampling sites is better than shortening the length of the
;arrpllng period. If sites are eliminated, all important land use/soils
:crb :. nations should be represented in the reduced program in order to
Drovice a basis for extrapolation of parameters to the entire Chesapeake Bay
:eg Lcn. Additional economies can be realized by restricting detailed storm
•;pcllutcgrach) sampling to selected sites (as opposed to all sites) with
-epr~39ntative land use and soil characteristics.
Ihe recommended sampling program provides a comprehensive frame'.vork for
•retaining the nonpoint source data needed for the proposed eutrophication
study of Chesapeake Bay. Reductions in the proposed design should be
rinirized to the extent possible. As mentioned previously, the final
croc ran design is needed prior to implementation to select specific
•vatershed sites, sampling equipment, and laboratory and field procedures.
This final design should include input from the sponsoring federal and state
a-ger.ci.es, the modeling consultant, and the contractors for sampling and
laboratory analyses.
-------
------- |