EPA Office of Water, Washington, DC. TMDL Program Results Analysis Fact Sheet # EPA841-F-09-009, December 2009
Fact Sheet:
Watershed Boundaries for the 2002 Impaired Waters Baseline National Geospatial Dataset
Background: the Clean Water Act, impaired waters and TMDLs
The goal of the Clean Water Act (CWA) is "to restore and maintain the chemical, physical, and biological integrity of the
Nation's waters". Under section 303(d) of the CWA, states, territories, and authorized tribes, collectively referred to in the
Act and here as "states," are required to develop lists of impaired waters every two years. A state's 303(d) impaired
waters list is comprised of all waters where required pollution controls are not sufficient to attain or maintain applicable
water quality standards. The law requires that states establish a prioritized schedule for waters on the lists and develop
Total Maximum Daily Loads (TMDLs). A TMDL is a calculation of the maximum amount of a pollutant that a waterbody
can receive and still meet water quality standards, and an allocation of the load reduction needed from various sources of
the pollutant. TMDLs are technical documents that summarize the analysis and lay the groundwork for beginning to plan
restoration. Over 40,000 TMDLs have been developed, and the number is continually increasing.
EPA's Assessment and TMDL Tracking and
Implementation System (ATTAINS) is a national
compilation of states' 303(d) listing and TMDL
development information, spanning several years of
tracking over 40,000 impaired waters. ATTAINS
impaired waters data are publicly available online in
tabular format (www.epa.qov/waters/ir) and as
geographic information systems (CIS) datasets (Figure
1) that are available for download from EPA at
http://epamap32.epa.gov/radims/.
Figure 1: GIS data on impaired waters, available online from EPA
Geospatial data on impaired waters and their watersheds
National geospatial datasets on impaired waters are produced and periodically updated by EPA using state-re ported data.
The GIS versions of state 303(d) lists are provided to EPA by states after the lists are approved and finalized. EPA
compiles the state datasets, reconciles the differences in format among state data, and indexes the data to the National
Hydrography Dataset Plus (NHDP/us) to provide a nationally consistent reference layer. The indexed datasets are based
on information taken from the Reach Address Database (RAD) and are housed in EPA's Waters Assessment, Tracking,
and Environmental Results (WATERS) environment. They include the 2002 Impaired Waters Baseline National
Geospatial Dataset containing impaired waters as of the 2002 baseline reporting year, and the 303(d) Listed Impaired
Waters National Geospatial Dataset that
includes more recent state data (from variable
dates) available as of the date of extraction.
The RAD also provides dynamic access to
individual state or watershed-level data
downloads as new data become available.
This fact sheet provides information about the
Watershed Boundaries for the 2002 Impaired
Waters Baseline National Geospatial Dataset
that was developed by EPA due to widespread
interest in GIS data on impaired waters and their
watersheds. Figure 2 is an example from this
dataset, which is available for public download
through the WATERS website. As a medium-
resolution dataset similar to the other national
NHDP/us products, these watershed boundaries
are appropriate for approximating the land
use/land cover, slope, and other general
characteristics of numerous impaired
watersheds across a state, region, or the nation.
Figure 2: Watershed boundary for a 2002 baseline impaired water
Water Body: Twin Grove Branch
-------
EPA Office of Water, Washington, DC. TMDL Program Results Analysis Fact Sheet # EPA841-F-09-009, December 2009
How the impaired watershed boundaries were delineated
Watershed delineation can be extremely labor-intensive when performed manually. Manual creation of a national
impaired watersheds boundary dataset would have been cost-prohibitive, but the availability of nationally consistent
source datasets and automated basin delineation tools made it feasible. The necessary data included the national 2002
baseline impaired waters geospatial data and the NHDP/us national catchments dataset.
Initially, all the geospatial information on impaired water body segments listed through 2002 was extracted from the RAD
to a geospatial data layer. The downstream "pour point" was then identified for each impaired segment based on
NHDP/us flow information. The pour point provided the basis for defining the watershed boundary as all up-gradient lands
and waters draining to that point. Where larger water bodies such as rivers were listed in segments (i.e., each segment is
one impaired water), a separate downstream pour point and watershed boundary were delineated for each segment.
Thus, the larger watershed boundaries in this dataset often contain smaller watersheds. For example, a large impaired
river's watershed in this dataset also contains the smaller watersheds of several of its impaired tributary streams.
The NHDP/us catchments national dataset (see figure 3) was used in the next step of automated boundary delineation.
The catchments dataset was produced by integrating
elevation data with NHD surface waters data in order to
obtain a delineation of a surface drainage unit for every
surface water segment in the entire NHD. Some
catchments are true watersheds, some are "flow-through"
drainage areas with true watersheds farther upstream.
These catchments have been further related to one
another by flow path information. Thus, starting with any
point in the flow network, one can navigate upstream to
determine all the catchments that together comprise the
watershed for that point. In the case of impaired waters,
the multiple catchments upstream of each pour point were
then dissolved into single watersheds with a single, outer
watershed boundary (figure 4).
Figure 4: watershed boundary (black) of an impaired water
(orange) delineated by aggregating upstream catchments
Figure 3: the NHDPIus catchments dataset consists of surface
drainage units delineated at each tributary confluence
Once the watersheds for each impaired water in the 2002
baseline were delineated, several attributes about each
watershed were compiled in tabular format. The List ID
that uniquely identifies each impaired water in the 2002
baseline dataset was used in this attribute table as the
Basin ID for its watershed. Other fields included
watershed area; whether the impaired water was a linear,
areal, or point data feature; and the number of additional
impaired waters occurring upstream in the watershed. The
presence of the List ID in the watersheds dataset enables
users to link to other attributes such as waterbody name,
causes of impairment, and whether a TMDL has been
completed and made available online.
Limitations of the impaired watershed boundaries dataset
The automated delineation process used to create this national dataset efficiently generated watershed boundaries for
most, but not all, impaired waters. The delineation process was attempted on all 39,567 impaired waters (including point,
linear and areal features) in the 2002 baseline extracted from the RAD. 32,605 unique pour points were identified relative
to these waters. This is a smaller number mainly because two or more impaired waters (such as one impaired tributary
and a mainstem segment ending at the confluence) frequently share one pour point. For a variety of reasons, including
absence of a downstream pour point or ambiguity of pour point information, the watersheds for many additional waters
could not be delineated. In total, 28,350 delineated watersheds were compiled in this national dataset.
For more about TMDLs and impaired waters data, visit:
EPA TMDL home: www.epa.qov/owow/tmdl
WATERS: http://www.epa.gov/waters/about/index.html
ATTAINS:
Data Downloads:
TMDL Results Analysis:
www.epa.gov/waters/ir
http://epamap32.epa.gov/radims/
www.epa.gov/owow/tmdl/results
------- |