Establishing a Chesapeake Bay
Nontidal Watershed
Water-Quality Network

September 2004

Prepared by the Chesapeake Bay Program's
Nontidal Water Quality Monitoring Workgroup

Chesapeake Bay Program

A Watershed


-------
Executive Summary

The Chesapeake Bay Program has committed to meeting water-quality criteria
(dissolved oxygen, water clarity and chlorophyll a) in the Bay and its tidal tributaries, by
2010. To achieve these criteria, the Chesapeake Bay Program's partners are
implementing management actions, through the tributary strategy process, to reduce
nutrients and sediments from entering the Chesapeake Bay watershed. The Chesapeake
Bay Nontidal Watershed Water-Quality Network is a critical tool for measuring the
nutrient and sediment concentrations and loads in the watershed and for assessing water-
quality changes and progress toward meeting water-quality criteria in the Chesapeake
Bay by 2010. Therefore, the Chesapeake Bay Program's Nontidal Water Quality
Monitoring Workgroup is designing a network for the Chesapeake Bay watershed. The
network is building from and integrating existing State and Federal (USGS and EPA)
monitoring programs.

The objectives of the Chesapeake Bay Nontidal Watershed Water-Quality Network are
to: (1) measure and assess the status and trends of nutrient and sediment concentrations
and loads in the tributary strategy basins across the watershed, (2) help assess the factors
affecting nutrient and sediment status and trends, and (3) improve calibration and
verification of partners' watershed models.

The Chesapeake Bay Nontidal Watershed Water-Quality Network will be designed so
that data are collected within tributary strategy basins and therefore, meet the objectives
of the network. The goal is to have all stations meet the requirement for a "load" station.
At a load station there will be a stream gage, 20 samples a year will be collected over a
range of flow, including storms, and samples will be analyzed for total nitrogen, total
phosphorus, and sediment. Since funding is currently not available to meet the
requirements for load stations in all areas, some stations will initially be implemented to
meet the minimum requirements to compute trends (stream-flow and monthly samples
will be collected at these stations). The workgroup has agreed upon procedures to ensure
data comparability between stations. The procedures include "routine" (monthly)
samples and storm samples and laboratory analyses.

To develop a list of stations that would be included in the Chesapeake Bay Nontidal
Watershed Water-Quality Network, existing stations were first evaluated to determine if
they could be useful as part of the network. The Nontidal Water Quality Monitoring
Workgroup developed a list of 188 candidate stations that included existing stations (115
stations) and locations where new stations (73) were recommended. The initial selection
of stations in the network was focused on the streams/rivers draining the tributary
strategy basin segments. Each jurisdiction prioritized stations in consultation with State
Tributary Strategy Coordinators and the Chesapeake Bay Program's watershed
monitoring and modeling staff. In total, 87 stations have been selected for
implementation in order to meet the trend and load objectives of the network.

Multiple sources of funding will be needed to implement the network. However, the
primary approach is to utilize and enhance existing water-quality monitoring and stream-

2


-------
gage programs. In general, a station in the network will cost about $45,000 a year to
operate. With an initial network of nearly 100 stations, total operation cost per year will
be near $4,500,000. The Nontidal Water Quality Monitoring Workgroup has been able to
utilize existing ambient water-quality monitoring programs and stream gages to cover
most of the network's implementation costs. Through discussions with EPA Regions 2
and 3 as well as each State's Water-Quality Program Coordinator, decisions were made
to directly consider the Chesapeake Bay Nontidal Watershed Water-Quality Network as
they revise their own water-quality monitoring strategies and networks that are funded
with 106 grants. Additional funding was secured through the states abilities to shift some
resources from other monitoring activities to the network, and through the Chesapeake
Bay Program ($175,000) and the USGS ($100,000 one-time Congressional earmark).

To help implement the current stations and enhance the Chesapeake Bay Nontidal
Watershed Water-Quality Network in the future, a Memorandum of Understanding will
be signed by the States and Federal Agencies (EPA and USGS). As more stations are
considered for the network, a similar strategy of utilizing multiple funding sources will be
followed. Stations will be added to ensure that the network represents the different
watershed characteristics and the range of nutrient and sediment sources.

Data from the Chesapeake Bay Nontidal Watershed Water-Quality Network will be
interpreted to provide several "indicators" of water-quality conditions in the watershed.
The indicators will be related to efforts to meet water-quality standards in the Chesapeake
Bay by reducing nutrients and sediments in the Chesapeake Bay watershed. The
"indicators" need to be used in conjunction with the Watershed Model Progress Runs,
and the interpretation of tidal monitoring data, to provide a group of indicators that can be
used to assess status and trends of water quality in the Chesapeake Bay and its watershed.

3


-------
Acknowledgements

The Chesapeake Bay Chesapeake Bay Nontidal Watershed Water-Quality Network was
designed by members of the Chesapeake Bay Program's Nontidal Water Quality
Monitoring Workgroup.

Scott Phillips - Chair, U.S. Geological Survey - Baltimore

Steve Preston - U.S. Geological Survey/Chesapeake Bay Program Office

Lisa Bowen - Chesapeake Research Consortium

Carlton Haywood - Interstate Commission on the Potomac River Basin

Mary Ellen Ley - U.S. Geological Survey/Chesapeake Bay Program Office

John Brakebill - U.S. Geological Survey - Baltimore

Ricky Bahner - Interstate Commission on the Potomac River Basin/Chesapeake

Bay Program Office
Bruce Michael - Maryland Department of Natural Resources
Paul Miller - Maryland Department of Natural Resources
Rick Hoffman - Virginia Department of Environmental Quality
Don Smith - Virginia Department of Environmental Quality
Richard Shertzer - Pennsylvania Department of Environmental Protection
Mike Langland - U.S. Geological Survey - Pennsylvania
Kevin McGonigal - Susquehanna River Basin Commission
Hassan Mirsajadi - Delaware Department of Natural Resources and

Environmental Control
Matt Monroe - West Virginia Department of Agriculture
John Wirts - West Virginia Department of Environmental Protection
Ron Entringer - New York Department of Environmental Conservation
Gary Shenk - U.S. Environmental Protection Agency - Chesapeake Bay Program
Office

Emery Cleaves - Maryland Geological Survey

Steve Preston and John Brakebill led the approach for station selection and development
of candidate stations. Funding for implementation of the network has come from
multiple sources including State and Federal water-quality monitoring programs, the
Chesapeake Bay Program, and the USGS.

4


-------
Need for Network

The Chesapeake Bay Program partners have committed to meet Bay specific water-
quality criteria (dissolved oxygen, water clarity and chlorophyll a) in the Bay and its tidal
tributaries by 2010. Through the tributary strategy process, the Chesapeake Bay Program
partners are implementing management actions throughout the six-state 64,000 square
mile watershed to reduce nutrient and sediment loads in order to restore Chesapeake Bay
water quality. The Chesapeake Bay Nontidal Watershed Monitoring Network is critical
to measuring local stream and river ambient nutrient and sediment concentrations and
loads to help track water-quality improvements and to progress toward meeting the
resultant reduced loads from tributary strategy basins. Therefore, the Chesapeake Bay
Program's Nontidal Water Quality Monitoring Workgroup is designing a network for the
nontidal Chesapeake Bay watershed, building from and integrating existing State and
Federal - U.S. Geological Survey (USGS) Environmental Protection Agency (U.S. EPA)
- stream-flow gage and water-quality monitoring programs.

The existing monitoring programs do not necessarily meet the Chesapeake Bay
restoration needs because they were designed separately by each agency for other
purposes. For example, each State operates a stream water-quality monitoring program
independently for its own objectives. Usually there is little coordination among the
States and federal agencies doing monitoring in the watershed. As a result, there can be
situations where two agencies monitor the same stream without knowledge of each
other's efforts. In other cases, information can be lost for multiple jurisdictions when one
agency makes a decision regarding termination of data collection on streams that extend
across borders. Greater coordination of all nontidal monitoring in the Chesapeake Bay
watershed is needed in order to improve efficiency for all jurisdictions. The Chesapeake
Bay Nontidal Watershed Water-Quality Network, described above, will help provide the
basis for that coordination.

Objectives of Network

The three primary objectives of the Chesapeake Bay Nontidal Watershed Water-Quality
Network are to:

1)	Measure and assess the status and trends of nutrient and sediment
concentrations and loads in the tributary strategy basins across the watershed;

2)	Assess the factors affecting nutrient and sediment status and trends; and

3)	Improve calibration and verification of partners' watershed models.

In order to meet these objectives, data from the network will be interpreted to provide the
Chesapeake Bay Program partners with the:

Status of water-quality conditions in the watershed;

5


-------
Trends in nutrient and sediment concentrations and stream-flow at
key stations in the tributary strategy basins characterizing loads
from the upstream tributary strategy basin;

Annual nutrient and sediment loads at a subset of these key
stations;

Information that can be used with other data to determine the
factors affecting the observed trends in concentration and load;

Information to improve calibration and verification of the partners'
and individual jurisdiction's watershed models; and

Indicators that can be used to communicate progress towards
reducing load caps allocated to each tributary strategy basin, water-
quality improvement in local streams and rivers, and progress
towards meeting the basinwide cap load goal necessary for
attaining water-quality standards in the Chesapeake Bay.

Station Requirements for Meeting Objectives

The Chesapeake Bay Nontidal Watershed Water Quality Network was designed so data
are collected within individual tributary strategy basins to meet the objectives of the
network. The goal is to have all stations meet the requirement for a "load" station, which
requires high-flow data collection (see below). Since funding is currently not available to
meet the requirements for load stations in all areas, some stations will initially be
implemented to meet the minimum requirements for a "trend" station. These stations will
be used to compute trends in concentration and flow. Data from both of these station
types, and in particular load stations, will be used to improve watershed models. The
load and trend data, along with data sets on nutrient and sediment sources, Best
Management Practices (BMPs), land-use changes and watershed characteristics will be
used to help assess the factors affecting local stream and river nutrient and sediment
concentrations, flow and the resultant loads to downstream waters. The observed
concentration/flow trends and calculated load data will help the tributary strategy teams
to: 1) assess their progress toward meeting cap load allocations; 2) evaluate the
effectiveness of the implementation of the tributary strategies to improve water-quality of
local streams; and 3) determine if tributary strategy implementation in the watersheds
will result in achievement of water-quality standards in the Bay.

Requirements for "Trend" Stations

Trend stations will generate the data necessary to determine if changes in in-stream
nutrient and sediment concentrations are occurring over time. Trends will be determined
for "flow-adjusted" concentration and stream/river flow. Having these two types of
trends will provide information to help managers, scientists and tributary strategy
stakeholders understand: 1) changes in stream-flow that are potentially due to land use

6


-------
changes; 2) changes in nutrient and sediment concentrations over time; and 3) the
separate influences of hydrology and on-the-ground management actions on the observed
water-quality changes. At least five years of monthly ambient concentration and
stream/river data are needed to compute trends in concentration and flow.

To be included as a trend station within the basinwide network, a station must meet the
following criteria:

1.	The station must be associated with a stream-flow gage to allow
computation of trends. Water-quality sampling must be conducted
close to the stream-flow gage so that the water-quality and discharge
information are comparable.

2.	Samples need to be collected at least monthly over a five-year time
span.

3.	At a minimum, the samples should be analyzed for total nitrogen, total
phosphorous and total suspended solids.

4.	Samples should be collected using methods to ensure they represent water-quality
conditions at the station.

Requirements for "Load" Stations

Load stations will generate the concentrations and flow data necessary to quantify the
amount of nutrients and sediment leaving tributary strategy basins as loads to either the
next downstream tributary strategy basin or to Chesapeake Bay tidal waters. Time series
data records from these stations will also be used to calibrate watershed models. In order
to compute loads, water-quality samples need to be collected over a range of stream/river
flow (including storms) because of the change in concentration during storms. This is
especially important for the phosphorus and sediment parameters. In addition to the
water-quality samples, continuous stream/river flow data are needed to apply the
statistical tools used to estimate stream loads. At least three years of concentration and
flow data are needed to compute loads.

To be included as a load station within the network, a station must meet the following
criteria:

1.	The station must be associated with a stream-flow gage to allow
computation of load (the product of flow times concentration).

Water-quality sampling must be conducted close to the stream
gage so that the water-quality and discharge information are
comparable.

2.	A total of 20 samples should be collected each year, including 12
monthly samples and eight storm samples based on an average
flow year (the actual number of samples could be less during low
flow years and more during high flow years as funds are available).

3.	In general, the storm samples should attempt to target at least four
different storms (one each season). More than one sample can be
collected per storm but no more than one per day (this is a
requirement of the statistical load estimation program).

7


-------
4.	At a minimum, the samples should be analyzed for total nitrogen,
total phosphorous, ammonium, nitrate, phosphate and total
suspended solids. Storm samples should also include suspended
sediment and particle size.

5.	For watershed modeling, it is recommended that samples also be
analyzed for the recommended parameters list in Table 2.

6.	Samples should be collected to ensure that they represent water-
quality conditions at the station.

Station Selection Process

To develop the list of stations included in the Chesapeake Bay Nontidal Watershed
Water-Quality Network, existing stations were first evaluated to determine if they would
directly contribute to meeting objectives of the basinwide network. Stream-flow gage
locations and water-quality stations were initially evaluated separately and then combined
to search for stations where water-quality data were collected in close proximity to a
stream-flow gage. The station selection process was used to determine which stations
met the criteria described above for trend and load stations. The combination of existing
stations that met basic trend or load criteria and proposed new stations provided a list of
candidate stations. If all of the stations on the list were implemented, the stated network
objectives and underlying management information needs would be met. Since funding
is not currently available to support operation of all the candidate stations, they were
prioritized for implementation during the initial year. Gaps in the coverage of that initial
network were identified and new stations were proposed on the streams and rivers in
those areas. The identified gaps helped to highlight a few of the remaining station
selection issues.

Stream Gage Locations

Stream gages are usually established and operated by the U.S. Geological Survey (USGS)
alone or through cooperative agreements. At each gage, continuous data records of
stream-flow are collected and are often reported as mean-daily discharge (i.e. - in cubic
feet per second). These data are used for various purposes depending on the gage, but are
critical for estimating loads of water-quality constituents, calibrating watershed models
and for understanding processes affecting water-quality. A full list of historical stream-
gage locations was initially compiled from the USGS National Water Information System
(NWIS) database. Stream gages were operated at 703 locations in the Chesapeake Bay
watershed over some period of time in the historical record (1950-present). Of those, 319
were in operation in 2000 and were considered active for the purpose of the initial station
screening (Figure 1).

The network of existing stream gages is quite valuable for the purposes of the
Chesapeake Bay Nontidal Watershed Water-Quality Network. The operation of stream-
flow gages represents a significant amount of the cost of monitoring at a load or trend
station. Significant savings can be realized if the water-quality data collection can be
added where an existing stream-flow gage is already funded and in operation. Thus, the

8


-------
population of existing stream-flow gages represents stations that can be monitored for
loads and/or trends most cost effectively.

Figure 1. A total of 703 historical and 319 currently active stream gages have
been operated by the U.S. Geological Survey in the Chesapeake Bay
watershed since 1950.

9


-------
Water-quality Monitoring Locations

Existing nontidal water-quality monitoring stations were initially compiled from a water-
quality database that was developed for the Chesapeake Bay Program's Nutrient
Subcommittee (Langland et al. 1995). The water-quality database includes data collected
from more than 1,700 stations in the Chesapeake Bay watershed during the period 1972
to 2003. Many of the stations were sampled for less than the minimum three years and
were eliminated from consideration. In many other cases, sampling ended long before
the current time frame (1985 to present) and no recent data were available. A total of 641
stations were found be part of an active monitoring program in 2001 and were considered
currently active (Figure 2).

10


-------
Water Quality
Monitoring Network

Active Water Quality

Historical Water Quality

Figure 2. The locations of the more than 1700 historical and 641 currently active
water-quality monitoring stations in the Chesapeake Bay watershed.

11


-------
The second step of the screening process involved merging the list of existing water-
quality monitoring stations with that of the stream gage locations, to determine how many
of both station types were co-located (as defined by the network operators) (Figure 3).
Of the 641 water-quality monitoring stations that were determined to be current, only 161
stations were co-located with a stream gage. In some cases, the locations of the water-
quality sampling station and stream gage were separated by some distance, but were still
considered as co-located by the agencies that conducted the monitoring. The distance
separating the two types of data collection was sometimes substantial (i.e. - more than 10
miles), because of logistical factors that affect one or both of the two types of
measurement. Spatial separation of where stream-flow is measured and water-quality
data is collected can cause errors in load and trend assessment, if there are substantial
differences in the drainage area characterized by the two stations. Each of the stations
that are geographically separated was evaluated to see if they should be considered co-
located. Once the list of co-located stations was defined, each station on the list was
evaluated to determine if the sampling frequency and parameter criteria were met for the
trend and/or load objectives.

12


-------
Water Quality
Monitoring Network

* Active Water Quality with Active Streamflow _
0 Active Water Quality

Figure 3. Active water-quality monitoring stations (n=641) and
active water-quality stations that are co-located with a
stream gage (n= 161).

13


-------
Evaluation Results

A minimal frequency criterion of 30 samples over three years was applied, with the idea
that stations with that sampling frequency and data record could be upgraded with
minimal effort. Usually a frequency of more than 30 samples over three years implies at
least a monthly sampling program. Such monitoring would only require storm sampling
to meet the load frequency criterion and would meet the trend frequency criterion if the
sampling was continuous over a record length of five years. Stations that did not meet
the minimal frequency criteria were usually those with semi-monthly sampling programs
(i.e., 18 samples over three years). Of the 161 co-located stations, only 94 met minimal
frequency -30 samples over three years- and parameter -total phosphorous and suspended
solids- requirements (Figure 4).

As a final step, the Delaware, Maryland, New York, Pennsylvania, Virginia and West
Virginia state water-quality management agencies were consulted, to determine if any
changes in their networks occurred since the end of the period of the water-quality
database (2000). A total of 21 additional stations were identified that were either started
near the end of the period of record or were in the States of West Virginia and Delaware,
which were not included in the original development of the data base. In all cases, those
21 additional stations were both co-located with a stream gage and met the minimal
frequency and parameter requirements. Thus, there were a total of 115 existing stations
that could be considered for the initial Bay Nontidal Watershed Water-Quality Network
(Figure 4).

14


-------
Water Quality
Monitoring Network

Figure 4. Locations of the 182 active water-quality monitoring
stations that are co-located with a stream gage and the
subset of 115 stations, which meet minimal frequency and
parameter criteria.

15


-------
New monitoring locations to address information gaps

The initial Chesapeake Bay Nontidal Watershed Water-Quality Network, based on
existing stations, did not provide all of the information needed to meet the network's
three primary objectives. Some tributary strategy basins were completely unrepresented
while others did not have monitoring at the appropriate locations. To address these
needs, new stations were proposed wherever the following types of stream locations were
not monitored:

1.	Outlets of major streams/rivers draining tributary strategy basins;

2.	Areas within the tributary strategy basins that have the highest nutrient load
delivery to the Chesapeake Bay and its tidal tributaries; and/or

3.	Areas with relatively low density of monitoring stations.

Outlets of major streams/rivers were targeted in order to provide information regarding
loading from tributary strategy basins and for tracking trends in water quality. Such
information can be used to track progress in water-quality improvement as on-the-ground
management actions are implemented. It should be noted that some tributary strategy
basins are drained by multiple streams/rivers (e.g., western shore of the lower
Susquehanna). In those cases, all stream s/ri vers with mean annual discharge above 50
cubic feet per second were targeted for monitoring. Areas that have the highest nutrient
delivery to the Bay were targeted in order to track the areas with the greatest loads and,
therefore, the greatest potential for load reduction. Areas of high loading were identified
using the output from the USGS SPARROW nitrogen model (Preston and Brakebill,
1999), which provides consistent estimates of nutrient loading for the entire Chesapeake
Bay watershed. Finally, areas of low monitoring density were targeted in order to
identify previously unknown sources of nutrients and provide additional monitoring in
the largest drainages for tracking local water-quality trends.

Using the three decision rules described above for identifying additional monitoring
stations, 73 new stations were identified where monitoring information was needed to
fully meet the basinwide network's three primary objectives. A total of 191 stations were
identified as needed for the Chesapeake Bay Nontidal Watershed Water-Quality
Network. All of these stations will provide data valuable for tracking progress within the
respective tributary strategy basin. However, it was clear that funding would not be
available for immediate operation and maintenance at all stations. The list of 191 stations
was considered to be a list of candidate stations from which a subset would be selected
for implementation (Figure 5). Each of the six watershed states was given the option to
prioritize the candidate stations within its jurisdiction for implementation.

16


-------
Initial Candidate Sites for a
Non-Tidal Monitoring Network

	Tributary Basins

A New Proposed Sites

a Minimum Criterion

SPARROW

Medium

Figure 5. Existing and proposed new stations that form the list of
"candidate" stations for inclusion in the Chesapeake Bay
nontidal water-quality monitoring network. Tributary
boundaries are shown. The areas of low, middle and high
nutrient loading, estimated based on the USGS SPARROW
model are indicated by the gray shading.

17


-------
Remaining Network Station Selection Issues

The list of candidate stations primarily addresses the objective of measuring and
assessing the status and trends of nutrient and sediment concentrations and loads in the
tributary strategy basins. It is not known at this time how well the candidate stations
specifically address the objective of improving the calibration and verification of
partners' watershed models. To best support watershed modeling, selected stations
should be fully representative of the range of geographic features such as land use,
physiography and climatography. There has not yet been any evaluation of the
representativeness of the candidate stations. To be fully supportive of the network's
watershed modeling objective, an evaluation of the representativeness of the initial
network is needed.

The initial network design was focused on addressing the more immediate management
objective to track progress of load reductions to support tributary strategy efforts. The
Chesapeake Bay Program's Nontidal Water Quality Monitoring Workgroup plans to
evaluate the representativeness of the stations within the initial network design. Obvious
gaps in representativeness will be addressed by changing the location of candidate
stations or adding new stations to the initial network (based upon available funds) to
better capture the geographic characteristics that are unrepresented.

Implementation of Network

Once the list of candidate stations was established, each State prioritized and finalized the
stations that could be implemented within the available funding constraints. Several of
the stations had limitations regarding their sampling and the resulting analyses.

Therefore, in some cases, available funds were dedicated toward resolving these
limitations within existing stations as opposed to establishing new stations.

Funding

Much of the water-quality monitoring is supported by U.S. EPA Clean Water Act section
106 funding. Because this data collection serves the broader needs of the States, there
was limited flexibility to adapt these monitoring stations to also meet the basinwide
network's objectives. However, each State reviewed its existing water-quality
monitoring stations for any possible redundancies and inefficiencies. By shifting some of
the existing water-quality monitoring activities, some 106 funds were made available to
help support enhancements to other existing stations such as initiation of storm-event
sampling or to help establish new stations.

Multiple sources of funding will be needed to fully implement the network, but the initial
focus is on utilizing and enhancing existing monitoring programs. A station in the
network would cost about $45,000 a year to operate. With an initial network of nearly
100 stations, total operation cost per year will be near $4,500,000. The network design is
based on the utilization of existing ambient water-quality monitoring programs and
stream gages to cover the majority of the implementation cost. In addition, the network

18


-------
design effort included discussions with EPA Regions 2 and 3, and State Water-quality
Program Coordinators to directly consider the Chesapeake Bay Nontidal Watershed
Water-Quality Network as the States revise their water-quality monitoring strategies
funded by 106 grants.

Two new sources of funding were identified to help implementation. First, the USGS,
through a congressional "earmark," allocated $100,000 to help support network
implementation. It is unclear if these funds will continue in the future and so these funds
were dedicated primarily toward the establishment of new stream gages, which would be
a one-time cost. The second source of funding was from the Chesapeake Bay Program
through the Monitoring and Analysis Subcommittee. The Chesapeake Bay Program
allocated $175,000 to support State efforts to establish the needed monitoring. These
funds were allocated among the States roughly by the amount of drainage area that each
jurisdiction has in the Bay watershed. This funding allocation approach was considered
equitable by the Nontidal Water Quality Monitoring Workgroup, since larger drainages
would theoretically require more monitoring.

Each State representative on the workgroup was charged with the responsibility of
selecting stations for implementation. The selections were made within the constraints of
funding that could be made available through improving existing network efficiency and
the new funds from the USGS and U.S. EPA Chesapeake Bay Program Office. Based on
the work done by each State, 87 stations have tentatively been identified for
implementation or enhancement to meet the network's load and /or trend objectives
(Figure 6, Table 1). A small number of those 87 stations are still being evaluated to see if
problems such as geographic separation between water-quality sampling station and
stream-flow gage can be rectified. However, most stations are slated for full
implementation beginning in October 2004.

19


-------
Initial Non-Tidal
Monitoring Network

A Load

Figure 6. The 87 load and trend stations selected for initial

implementation of Chesapeake Bay Nontidal Watershed
Water-Quality Network. A subset of 13 stations are still
being evaluated to determine if the water-quality
monitoring can be co-located with the designated stream
gage. The tributary strategy basins are shown to illustrate
the network coverage for these basins.

20


-------
Table 1. List of 87 stations selected for initial implementation of the Chesapeake Bay
Watershed Nontidal Water-Quality Monitoring Network.

STATE

SITE NAME

MAP-ID

STREAM
GAGE ID

WATER QUALITY
ID

LOAD

TREND

DE

NANTICOKE RIVER NEAR BRIDGEVILLE

1

01487000

304191



X

DE

MARSHY HOPE CR

2

01488500

302031



X

MD

CHOPTANK RIVER NEAR GREENSBORO

3

01491000

01491000

X

X

MD

TUCKAHOE R

4

01491500

TUK0133

X

X

MD

BIG ELK CR

5

01495000

BEL0043

X

X

MD

SUSQUEHANNA RIVER AT CONOWINGO

21

01578310

01578310

X

X

MD

GUNPOWDER FALLS AT HOFFMANVILLE

22

01581810

GUN0476



X

MD

GUNPOWDER FALLS AT GLENCOE

23

01582500

GUN0258



X

MD

NORTH BRANCH PATAPSCO RIVER AT CEDARHURST

24

01586000

NPA0165



X

MD

PATAPSCO RIVER AT HOLLOFIELD

25

01589000

PAT0285



X

MD

GWYNNS FALLS AT VILLA NOVA

26

01589300

GWN0115



X

MD

JONES FALLS AT SORRENTO

27

01589440

JON0184



X

MD

PATUXENT RIVER NEAR UNITY

28

01591000

PTX0972



X

MD

PATUXENT RIV NEAR LAUREL

29

01592500

PTX0809



X

MD

PATUXENT RIVER NEAR BOWIE

30

01594440

01594440

X

X

MD

WESTERN BRANCH AT UPPER MARLBORO

31

01594526

WXT0045



X

MD

GEORGES CREEK AT FRANKLIN

33

01599000

GE00009

X

X

MD

POTOMAC RIVER AT PAW PAW

37

01610000

POT2766



X

MD

POTOMAC RIVER AT HANCOCK

39

01613000

POT2386



X

MD

POTOMAC RIVER AT SHEPHERDSTOWN

43

01618000

POT1830



X

MD

ANTIETAM CREEK NEAR SHARPSBURG

44

01619500

ANT0044



X

MD

CATOCTIN CREEK NEAR MIDDLETOWN

50

01637500

CAC0148



X

MD

POTOMAC RIVER AT POINT OF ROCKS

52

01638500

POT1595

X

X

MD

MONOCACY RIVER AT BRIDGEPORT

53

01639000

MON0528

X

X

MD

BIG PIPE CREEKAT BRUCEVILLE

54

01639500

BPC0035



X

MD

MONOCACY RIVER AT JUG BRIDGE NEAR FREDERICK

55

01643000

MON0155



X

MD

POTOMAC RIVER NEAR WASH DC LITTLE FALLS PUMP

57

01646500

01646580

X

X

MD

NE BRANCH ANACOSTIA RIVER AT RIVERDALE

58

01649500

01649500

X

X

MD

NW BRANCH ANACOSTIA RIVER NEAR HYATTSVILLE

59

01651000

01651000

X

X

MD

MATTAWOMAN CR

62

01658000

01658000

X

X

MD

WILLS CREEK NEAR CUMBERLAND

107

01601500

WIL0013



X

MD

ROCK CREEK AT SHERRILL DRIVE WASHINGTON

108

01648000

RCM0111



X

NY

SUSQUEHANNA R AT WAVERLEY

100

01515000

-9999

X

X

NY

CHEMUNG R

102

01531000

-9999

X

X

PA

TIOGA RIVER AT TIOGA JUNCTION

6

01518700

WQN0324

X

X

PA

SUSQUHANNA RIVER AT TOWANDA

7

01531500

WQN0305

X

X

PA

SUSQUEHANNA RIVER AT DANVILLE

8

01540500

WQN0301

X

X

PA

SUSQUEHANNA R, W BR AT KARTHUS

9

01542500

WQN0404

X

X

PA

BALD EAGLE CR AT CASTANEA

10

01548085

WQN0445

X

X

PA

SUSQUEHANNA R W BR AT JERSEY SHORE

11

01549760

WQN0448

X

X

PA

WEST BRANCH SUSQUEHANNA RIVER AT LEWISBURG

12

01553500

WQN0401

X

X

PA

PENNS CREEKAT PENNS CREEK

13

01555000

WQN0229

X

X

PA

RAYSTOWN BRANCH JUNIATA RIVER AT SAXTON

14

01562000

WQN0223

X

X

PA

JUNIATA RIVER AT NEWPORT

15

01567000

WQN0214

X

X

PA

CONODOGUINET CR NEAR HOGESTOWN

16

01570000

WQN0271

X

X

PA

SWATARA CR NEAR HERSHEY

17

01573560

WQN0272

X

X

PA

WEST CONEWAGO CREEK NEAR MANCHESTER

18

01574000

WQN0210

X

X

PA

SUSQUEHANNA RIVER AT MARIETTA

19

01576000

WQN0201

X

X

PA

CONESTOGA CR NR CONESTOGA

20

01576754

WQN0273

X

X

21


-------
Table 1. Continued







STREAM

WATER QUALITY





STATE

SITE NAME

MAP-ID

GAGE ID

ID

LOAD

TREND

PA

LICKING CR

40

01613500

WQN0509

X

X

PA

CONOCOCHEAGUE CREEKAT FAIRVIEW

41

01614500

WQN501

X

X

PA

COWANESQUE RIVER NEAR LAWRENCEVILLE

101

01520000

WQN0320

X

X

PA

SUSQUEHANNA RIVER NEAR WILKES-BARRE

103

01536500

WQN0302

X

X

PA

SHERMAN CR

104

01568000

WQN0279

X

X

PA

YELLOW BREECHES CR

105

01571500

WQN0212

X

X

PA

PEQUEACR

106

01576787

WQN0204

X

X

VA

SOUTH RIVER NEAR WAYNESBORO

45

01626000

1 BSTH027.85



X

VA

S F SHENANDOAH RIVER

46

01628500

1 BSSF100.10



X

VA

S F SHENANDOAH RIVER AT FRONT ROYAL

47

01631000

1 BSSF003.56

X

X

VA

N F SHENANDOAH RIVER NEAR STRASBURG

48

01634000

1 BNFS010.34

X

X

VA

CATOCTIN CREEKAT TAYLORSTOWN

51

01638480

1 ACAX004.57



X

VA

GOOSE CREEK NEAR LEESBURG

56

01644000

1AGOO011.23



X

VA

ACCOTINK CREEK NEAR ANNANDALE

60

01654000

1AACO014.57



X

VA

CEDAR RUN NEAR CATLETT

61

01656000

1ACER016.46



X

VA

RAPPAHANNOCK RIVER AT REMINGTON

63

01664000

3-RPP147.10



X

VA

RAP 1 DAN RIVER NEAR RUCKERSVILLE

64

01665500

3-RAP066.54



X

VA

ROBINSON RIVER NEAR LOCUST DALE

65

01666500

3-ROB001.90



X

VA

RAP 1 DAN RIVER NEAR CULPEPER

66

01667500

3-RAP030.21

X

X

VA

RAPPAHANNOCK RIVER NEAR FREDERICKSBURG

67

01668000

3-RPP113.37

X

X

VA

CAT POINT CR NEAR MONTROSS

68

01668500

3-CAT011.62



X

VA

PISCATAWAY CR NEAR TAPPAHANNOCK

69

01669000

3-PIS009.24



X

VA

DRAGON SWAMP AT MASCOTT

70

01669520

7-DRN010.48



X

VA

NORTH ANNA RIVER AT HART CORNER NEAR DOSWELL

71

01671020

8-NAR005.42



X

VA

LITTLE RIVER NEAR DOSWELL

72

01671100

8-LTL009.54



X

VA

PAMUNKEY RIVER NEAR HANOVER

73

01673000

8-PMK082.34

X

X

VA

PO RIVER NEAR SPOTSYLVANIA

74

01673800

8-POR008.97



X

VA

MATTAPONI RIVER NEAR BOWLING GREEN

75

01674000

8-MPN094.79



X

VA

MATTAPONI RIVER NEAR BEULAHVILLE

76

01674500

8-MPN054.17

X

X

VA

BACK CREEK NEAR MOUNTAIN GROVE

77

02011500

2-BCC004.71



X

VA

BULLPASTURE RIVER AT WILLIAMSVILLE

78

02015700

2-BLP000.79



X

VA

CALFPASTURE RIVER ABOVE MILL CREEKAT GOSHEN

79

02020500

2-CFP004.67



X

VA

MAURY RIVER NEAR BUENA VISTA

80

02024000

2-MRY014.78



X

VA

JAMES RIVER AT BENT CREEK

81

02026000

2-JMS229.14



X

VA

PINEY RIVER AT PINEY RIVER

82

02027500

2-PNY005.29



X

VA

MECHUMS RIVER NEAR WHITE HALL

83

02031000

2-MCM005.12



X

VA

JAMES RIVER AT CARTERSVILLE

84

02035000

2-JMS157.28

X

X

VA

APPOMATTOX RIVER AT FARMVILLE

85

02039500

2-APP110.93



X

VA

DEEP CREEK NEAR MANNBORO

86

02041000

2-DPC005.20



X

VA

APPOMATTOX RIVER AT MATOACA

87

02041650

2-APP016.38

X

X

VA

JAMES R

109

02025500

2-JMS279.41

X

X

VA

RIVANNA RIVER AT PALMYRA

110

02034000

2-RVN015.97

X

X

VA

JAMES RIVER NEAR RICHMOND

111

02037500

2-JMS117.35

X

X

VA

CHICKAHOMINY RIVER NEAR PROVIDENCE FORGE

112

02042500

2-CHK032.77



X

WV

STONEY RIVER NEAR MT. STORM

32

01595200

NB-SR1

X

X

WV

PATTERSON CR

34

01604500

NB-PC1

X

X

WV

POTOMAC R S BR

35

01608500

SOU0004

X

X

WV

BACK CR

36

-9999

PT-BC1

X

X

WV

CACAPON RIVER NR GREAT CACAPON

38

01611500

PC-000-006.0

X

X

WV

OPEQUON CREEK NEAR MARTAINSBURG

42

01616500

PT-OP2

X

X

WV

SHENANDOAH R

49

01636500

SR-SH3

X

X

22


-------
Initial Evaluation of Existing Water-Quality Stations

Only the minimal frequency and parameter criteria were initially applied to the evaluation
of stations. The intent was that selected stations could be upgraded to meet the load-
station criteria. However, this initial evaluation approach identified several limitations of
the existing stations. The limitations include:

1.	The majority of existing stations do not collect samples during storms
preventing proper determination of nutrients and sediment loads;

2.	The samples at some stations are not analyzed for a complete list of
parameters; and

3.	The determination of sediment concentration is not consistent at many
stations.

At the majority of existing stations samples are not collected during storms preventing
proper determination of nutrients and sediment loads. Only a small number of stations
have samples collected over a range of stream-flow conditions (including storms), which
is needed to properly compute a load. This is particularly important for determination of
total phosphorous and sediment loads. Additionally, some of the existing nontidal
monitoring programs have recently decreased their sample collection to less than once a
month. This decrease will impact the ability to compute and detect trends in
concentration and flow.

Samples at some stations are not analyzed for total nitrogen and total phosphorous. Some
nontidal monitoring programs are only sampling for one species of nitrogen, such as
nitrate, and a total nitrogen value cannot be determined. At other stations, both total
nitrogen and total phosphorous are not being determined. An assessment of the
parameters being analyzed by each of the jurisdictions and Federal agencies is presented
in Table 2.

The determination of sediment is not consistent at many stations. There are two
measures of sediment being conducted in the Chesapeake Bay watershed: total suspended
solids and suspended sediment. One study by Gray et al. 2000 has shown that at higher
concentration levels, during storms or when sand is greater than 25 percent of dry mass,
total suspended solids analysis does not provide an accurate measure of suspended
sediment in a river. In addition, many nontidal monitoring programs do not sample
during storm events when most sediment is in transport so sediment concentrations over a
range of flow are not being collected at many stations in the watershed. Finally, during
storm events some sediment in transport will also include sand in addition to silt and clay.
A "sand-fine" or particle-size analysis to document the relative amounts of each is not
conducted at the vast majority of stations in the watershed.

23


-------
Development of Comparable Methods

A goal of the Chesapeake Bay Nontidal Water-Quality Network is to evaluate conditions
across the watershed. Because the monitoring network is comprised of selected stations
from eight different State and federal monitoring programs, it is important that the
Chesapeake Bay Program is able to combine and interpret the data sets with a high
degree of confidence. To achieve this, participating agencies will analyze a core set of
parameters using comparable sample collection and laboratory analysis procedures, as
well as comparable data submission and data management practices. The Nontidal Water
Quality Monitoring Workgroup will continue improving the comparability and
consistency of data collection activities as the network is implemented and refined.

Collection and Analytical Procedures

The Nontidal Water Quality Monitoring Workgroup has agreed on sample collection and
laboratory analysis procedures that will ensure data comparability among all of the
stations in the network. The procedures include: 1) "routine" (monthly) samples, 2)
storm samples and 3) laboratory methods.

Environmental agencies collect routine surface water samples using two different
approaches: either a grab sample is taken from the mid-channel of the stream, or several
depth-integrated samples are collected across the stream and composited. The latter
method is preferred because concentration gradients often occur vertically and
horizontally across the channel. Grab sampling however, is less expensive.

Those agencies collecting grab samples will work towards collecting a depth-integrated
sample from a well-mixed point in the stream. USGS will assist the states in assessing
the sampling stations to evaluate the representativeness of their collection points.

Representative storm samples are critical for accurate sediment and total phosphorus load
estimations. For storm event sampling, agencies have agreed to collect depth-integrated
samples composited across the transect of a stream, following USGS guidelines. State
agencies will either contract with USGS or the Susquehanna River Basin Commission to
collect storm samples, or will work with USGS staff to develop comparable State
procedures. It is expected that many of the stations will require an on-station review to
determine station-specific procedures for obtaining representative samples.

The laboratory methods used by the eight agencies participating in the Chesapeake Bay
Nontidal Watershed Water-Quality Network are listed in Table 2. The analytical
methods for nitrogen, phosphorus and total suspended solids are comparable. For
suspended sediment and particle size analyses (storm events only), agencies will either
contract with the USGS Sediment Laboratory in Kentucky, or develop identical methods
in their State laboratories. Semi-annual, inter-laboratory comparison samples will be
tested by the laboratories to demonstrate the accuracy of their data and document the
comparability of these methodologies.

24


-------
Table 2. Sample collection and analytical methods of the participating agencies in the
Chesapeake Bay Nontidal Water-Quality Monitoring Network.

1. Required Parameters

USGS
River Input

SRBC

PADEP

VDEQ

MDNR

DNREC

WVDA

NYSDEC

Total Nitrogen

PN +TDN

alk. persulfate
SM 4500-Norg D

alk. persulfate
SM 4500-Norg D

alk. persulfate
SM 4500-Norg D

TKN+N023
351.2

TKN+N023
351.2

TKN+N023
351.1

TKN+N023
351.2

Ammonium (dissolved)

90

350.1

350.1

(unfiltered) 350.1

350.1

350.1

350.1

350.1

Nitrate + Nitrite (dissolved)

353.2

353.2

353.2

(unfiltered) 353.2

(unfiltered) 353.2

353.2

353.2

353.2

Total Phosphorus

365.1

365.1

365.1

365.4

365.4

365.4

365.2

365.1

Phosphate (dissolved)

365.1

365.1

365.1

(unfiltered) 365.1

(unfiltered) 365.1

365.1



365.1

Total Suspended Solids

SM 2540 D

USGS-I-3765-85

USGS 1-3765

USGS-I-3765-85

SM 2540 D

160.2

SM 2540 D

160.2

Suspended Sediment (storms

ASTM D3977C

X

X

ASTM D3977C

ASTM D3977C

N/A

ASTM D3977C

X

%Sand/Fine Particles (storms

X

X

X

X

X

N/A

X

X

II. Recommended Parameters (for watershed model and source assessments)



Dissolved Oxygen (field)





YSI meter

Insitu / minisonde

Hydrolab

360.1

SM 4500-0 G

Hydrolab

Temperature (field)

X

Thermistor

Thermistor

Insitu / minisonde

Hydrolab

170.1

170.1

Hydrolab

Total Dissolved Phosphorus

365.1

365.1

365.1











Total Dissolved Nitrogen

alk. persulfate

SM 4500-Norg D

SM 4500-Norg D











Chlorophyll a (corrected)

SM 20th ed







SM 20th ed

SM 20th ed





Total Organic Carbon (or PC)

(PC) 440.0

persulfate IR
SM 5310D

persulfate IR
SM 5310D



combustion IR
SM 5310 B

415.1





Dissolved Organic Carbon

UV persulfate/IR









415.1





Volatile Suspended Solids

SM 2540





USGS-I-3765-85









III. Sampling Design

Number of Trend-only Sites

0

0

0

27

20

2

0

0

Number of Load/Trend Sites

9

6

21

11

10

0

7

2

Routine Sampling Frequency

12/yr.

12/yr.

12/yr.

12/yr.

12/yr.

12/yr.

12/yr.

12/yr. (6 SRBC)

Storm Sampling Frequency

10-20 days/yr.

8 days/yr.

8 days/yr.

8 days/yr.

8 days/yr.

N/A

8 days/yr.

8 days/yr.

Sample Type (routine)

cross section,
depth integrated
composite

cross section,
depth integrated
composite

depth integrated
composite
across transect

mid-channel,
surface grab
(bucket)

mid-channel,
surface grab
(bucket)

mid-channel,
below the
surface

mid-channel,
mid-depth grab

cross section,
depth integrated
composite

Sample Type (storms)

depth integrated
composite
across transect
(isokinetic)

depth integrated
composite
across transect
(isokinetic)

depth integrated
composite
across transect

depth integrated
composite
across transect
(isokinetic)

To be
established

N/A

To be
established

depth integrated
composite
across transect
(isokinetic)

Abbreviations:

DNREC	Delaware Natural Resources and Environmental Control

MDNR	Maryland Department of Natural Resources

NYSDEC	New York State Department of Environmental Conservation

PADEP	Pennsylvania Department of Environmental Protection

SRBC	Susquehanna River Basin Commission

USGS	U.S. Geological Survey

VDEQ	Virginia Department of Environmental Quality

25


-------
Data Submission and Data Management Procedures

The data collected as part of the Chesapeake Bay Nontidal Watershed Water-Quality
Network will be added to the Chesapeake Bay Program water-quality database, which is
part of the Chesapeake Information Management System (CIMS). These data will pass
the same level of quality assurance used with the current Chesapeake Bay Program
database and will be made available to the public through the CIMS "Data Hub" on the
Chesapeake Bay Program website at www.chesapeakebav.net.

Data management for this project will have two phases. First, the nontidal SAS database
(1970-2003), now in use for data analysis, will be reformatted and added to the
Chesapeake Bay Program water-quality relational database. Then, SAS data prior tol984
will be reformatted after the 1984 through 2003 data are completed. This work will be
completed at the Chesapeake Bay Program's Data Center in Annapolis, Maryland. Then,
the 2004 and future monitoring data will be posted through a CIMS networked website or
submitted electronically to the Chesapeake Bay Program on a yearly or more frequent,
basis through either the Client to Access Web Service or the DUQAT Online Submission
Tool.

Direct Data Submission:

DUQAT (Data Upload and Quality Assurance Tool) is the current tool for electronic data
submissions to the Chesapeake Bay Program water-quality database. Using this method,
the dataset must be formatted in ACCESS 97 tables and manually submitted to the online
DUQAT software. The dataset is processed overnight, and a quality assurance report is
available for the submitter to review the next day. After each dataset passes all the
quality assurance checks and is approved by the Chesapeake Bay Program Water-quality
Data Manager, it is added to the Chesapeake Bay Program water-quality database.

The data submission will be a database composed of three tables: 1) WQEVENT,
2) WQDATA and 3) WQCRUISES. The WQ EVENT table describes information
pertaining to the sampling event for each station and date; the WQ DATA table contains
the measured value and other information for each parameter at a particular station, date,
and sample depth; and, the WQ CRUISES table assigns a monitoring cruise ID to each
monthly cruise or cruises and records the agency, program, project and the start and end
dates for the cruise. The DUQAT software will run approximately 70 quality assurance
checks on the submitted nontidal data. These checks include comparing database codes
such as stations, parameter names, and methods to the database lookup tables.

Create Client to Access Web Service:

To use the second option, the Client to Access Web Service, the Chesapeake Bay
Program Data Center staff will develop a web service that will accept and import data to
the water-quality database. The web service is a self-describing, self-contained, modular
unit of application logic that provides some business functionality to other applications
through an internet connection. Applications access web services via web protocols and
data formats, such as HTTP and XML, without respect to how each web service is
implemented. There is not an associated web page, as the client application can use the

26


-------
web service like a function. The data submitters will develop clients that will send data
from their data holdings to the Chesapeake Bay Program web service. This option will
require developing a client for each data submitter and a new web service. However,
once these programs are in place, data will not have to be manually formatted and
submitted to the Chesapeake Bay Program.

Currently, four data submitters are using one of these two options to submit tidal and
nontidal monitoring data to the Chesapeake Bay Program. The Chesapeake Bay Program
Office staff will work with the other data submitters to arrive at the most practical
solution for data submission. The process of acquiring these data will be a collaborative
effort between the submitter and the Chesapeake Bay Program. In addition, each data
submitter will be responsible for the creation and yearly update of a metadata record
using the CIMS Online Metadata Entry Tool (COMET).

The quality assurance checks run by the Chesapeake Bay Program web service will be
limited to verifying that the data fit into the database without causing errors. There will
be checks for duplicate records and for properly formatted XML files being sent to the
Chesapeake Bay Program web service. The data submitter will be responsible for
running their own in-house quality assurance checks prior to the transfer of these data.

Interpretation of Results from the Network

Data from the Chesapeake Bay Nontidal Water-Quality Network, which includes the
river-input stations, will be processed to provide several "indicators" of water-quality
conditions through the Chesapeake Bay watershed. All of the proposed indicators relate
to meeting water-quality standards in the Chesapeake Bay through efforts to reduce
nutrients and sediment loads from the watershed. The indicators that are being
considered for interpretation of the nontidal water-quality data include:

1.	Water-quality status.

2.	Yield of nutrients and sediment.

3.	Statistically defined temporal trend in stream-flow.

4.	Statistically defined temporal trend in flow-adjusted concentration.

5.	Graphical illustration of temporal changes in estimated load.

The importance of indicators for the water-quality status and yield of nutrients and
sediments indicators is relatively straightforward. Trend in flow-adjusted concentration
attempts to detect a trend in concentration by adjusting for the influence of flow and
season. This indicator will be a valuable diagnostic tool to look at the change in
concentration that would be due mostly to reduction of nutrient and sediment sources.
Trends in stream-flow will also be an important diagnostic tool to assess its influence on
concentration and load. Finally, changes in estimated load will be an indicator for annual
delivery of load to the Chesapeake Bay and also change in load over time for selected
stations in the watershed.

27


-------
References

Gray, J.R., G.D.Glysson, L.M. Turcios and G.E. Schwarz, 2000. Comparability of
suspended-sediment concentration and total suspended solids data. U.S. Geological
Survey Water Resources Investigations Report 00-4191, 20 p.

Langland, M.J., P.J. Leitman, and S.J. Hoffman, 1995. Synthesis of nutrient and
sediment data for watersheds within the Chesapeake Bay drainage basin. U.S.
Geological Survey Water-Resources Investigations Report 95-4233, 121 p.

Preston, S.D. and J.W. Brakebill, 1999. Application of spatially referenced regression
modeling for the Chesapeake Bay Watershed. U.S. Geological Survey Water-Resources
Investigations Report 99-4054, 12 p.

28


-------