StreamCat Updates

September 9, 2021 - Added all NLCD 2019 and Impervious 2019 years

April 20, 2021 - Added Monitoring Trends in Burn Severity (MTBS)

December 4, 2020 - Added all NLCD 2016 and Impervious 2016 years

• There was change to flow table connections in Hydroregions 2 and 13 that altered watershed areas slightly for a number of COMIDS. All the metrics tables were rebuilt as an update and to accommodate these differences in these two hydroregions.

June 6, 2020 - Fix to CanalDensity in Hydroregion 18 / California

• There was a systematic error in canal density values for Hydroregion 18 and California which has been fixed in the CanalDensity_Region18.zip and CanalDensity_CA.zip files

February 12, 2020 - Added NLCD 2016

November 6, 2018 - Revised versions of ICI/IWI

The indices of catchment and watershed integrity (ICI/IWI, respectively) were revised Johnson et al. (2019). Revisions include:

• The StreamCat metrics used in the calculation of ICI and IWI functional components (versions 2 and 2.1) were standardized by the same maximum value (usually catchment maximum). Doing so places the indices on the same scale such that an ICI and IWI with the same value (e.g., ICI = IWI = 0.5) both have the same levels of human-related land use in each.
• The chemical component of the ICI and IWI (CCHEM and WCHEM, respectively) were revised (version 2.1) based on empirical relationships between landscape stressors and an instream water chemistry index developed with field measurements from the 2008-2009 National Rivers and Streams Assessment. Revisions of CCHEM and WCHEM automatically result in revisions of the ICI and IWI, respectively. See Johnson et al. (2019) for details.
• Revised indices are in the tables ICI_IWI_v2.1_(RegionID).csv and ICI_IWI_v2.1_(StateID).csv.
• We’ve also published ICI_IWI_v2_(RegionID).csv and ICI_IWI_v2_(StateID).csv results which ONLY show the standardization by same maximum value (usually catchment maximum) in first bullet BUT NOT the revision to chemical component (second bullet).
• The revised version of ICI and IWI (v2.1) is highly recommended over v.2, which is published simply for transparency.

October 30, 2018 - Update Wetness Index

The Composite Topographic Index (CTI) generated by EnviroAtlas used to generate the Wetness Index in StreamCat was recently updated. The Wetness Index in StreamCat was re-calculated using this updated, corrected CTI raster from EnviroAtlas.

April 10, 2018 - LakeCat released

LakeCat, a companion dataset to StreamCat, is now available online.

• LakeCat data
• LakeCat article

November 14, 2017 - Indices of watershed and catchment integrity added to StreamCat

• The conceptual framework of watershed integrity was developed by Flotemersch et al. (2016).
• Flotemersch et al. (2016) defined watershed integrity as “the capacity of a watershed to support and maintain the full range of ecological processes and functions essential to the sustainability of biodiversity and of the watershed resources and services provided to society”.
• This definition was operationalized by first identifying six key functions of watersheds -
  • Hydrologic regulation (HYD)
  • Regulation of water chemistry (CHEM)
  • Sediment regulation (SED)
  • Hydrologic connectivity (CONN)
  • Temperature regulation (TEMP)
  • Habitat provision (HABT)
• Stressors of each of these six functions were identified along with geospatial layers that could provide estimates of these stressors within watersheds.
• Thornbrugh et al. (2017) used negative linear relationships to estimate the loss of integrity of each function in response to stressors.
• The index of watershed integrity (IWI) is the product of the six functional scores (see Flotemersch et al. 2016 for details).
• This approach was expanded to the concept of catchments (see geospatial framework and terms) by Thornbrugh et al. (2017).
• New metrics include:
  • Indices of watershed and catchment integrity
    • ICI_IWI_-RegionID-.csv
    • ICI_IWI_-StateID-.csv
• The code used to produce these tables is available.
• Columns in tables:
  • COMID - Unique identifier for each stream segment, and associated catchment, in NHDPlusV2.
  • CatAreaSqKm - Area of local catchment measured in square kilometers.
  • WsAreaSqKm - Area of the local catchment and accumulated upstream catchments composing the entire watershed (Ws) measured in square kilometers.
  • CatPctFull - Minimum % of catchment that overlaps with landscape layer across all predictors used to calculate catchment component and index values. Accounts for areas of NoData pixels and international borders.
  • WsPctFull - Minimum % of watershed that overlaps with landscape layer across all predictors used to calculate watershed component and index values. Accounts for areas of NoData pixels and international borders.
  • CHYD - Hydrologic regulation component score calculated using catchment metrics.
  • CCHEM - Regulation of water chemistry component score calculated using catchment metrics.
  • CSED - Sediment regulation component score calculated using catchment metrics.
  • CCONN - Hydrologic connectivity component score calculated using catchment metrics.
  • CTEMP - Temperature regulation component score calculated using catchment metrics.
  • CHABT - Habitat provision component score calculated using catchment metrics.
  • ICI - Index of catchment integrity.
  • WHYD - Hydrologic regulation component score calculated using watershed metrics.
  • WCHEM - Regulation of water chemistry component score calculated using watershed metrics.
  • WSED - Sediment regulation component score calculated using watershed metrics.
  • WCONN - Hydrologic connectivity component score calculated using watershed metrics.
  • WTEMP - Temperature regulation component score calculated using watershed metrics.
  • WHABT - Habitat provision component score calculated using watershed metrics.
  • IWI - Index of watershed integrity.
• Component scores of the ICI and IWI are based on the following StreamCat metrics: 

September 25, 2017 NLCD 2001 added to StreamCat (full watershed and riparian buffers)

September 08, 2017 - Predicted biological condition added to StreamCat

This dataset consists of predicted probabilities of good biological condition based on the US EPA 2008/2009 National Rivers and Streams Assessment (NRSA) benthic invertebrate multimetric index (BMMI).

The development of the NRSA BMMI is documented in the 2008/2009 NRSA Report and by Stoddard et al. (2008).

These assessed sites were then paired with StreamCat data and a random forest model was developed and applied back to the StreamCat data to produce the prediction probability of being in good biological condition for about 1.1 million segments that were part of the NRSA sampling frame. Model development is documented in Fox et al. (2017) and Hill et al. (2017).

The resulting tables have the standard columns that are normally distributed with StreamCat tables. However, CatPctFull and WsPctFull represent the minimum data completeness of geospatial data within the catchment and watershed metrics used in modeling (i.e., across all predictors used in the model). For example, if 50% of the area of a catchment intersected with NoData values of a geospatial layer and all other geospatial layers for that catchment were complete (100%), then the CatPctFull value reported for that catchment is 50%.

• The code used to produce these tables is available.
• New tables include:
  • NRSA_PredictedBioCondition_-RegionID-.csv
  • NRSA_PredictedBioCondition_-StateID-.csv
• Columns within tables:
  • COMID - Unique identifier for each stream segment, and associated catchment, in NHDPlusV2.
  • CatAreaSqKm - Area of local catchment measured in square kilometers.
  • WsAreaSqKm - Area of the local catchment and accumulated upstream catchments composing the entire watershed (Ws) measured in square kilometers.
  • CatPctFull - Minimum % of catchment that overlaps with landscape layer across all predictors used to model biological condition. Accounts for areas of NoData pixels and international borders.
  • WsPctFull - Minimum % of watershed that overlaps with landscape layer across all predictors used to model biological condition. Accounts for areas of NoData pixels and international borders.
  • NRSA_Frame - A column indicating whether the NHDPlusV2 stream segment was included as a candidate site for sampling during the 2008/2009 NRSA. Predicted conditions are restricted to only those stream segments that were part of this frame.
    • Include - Included in the NRSA sampling frame and may have a predicted biological condition if StreamCat metrics are available for this stream segment.
    • Exclude - Not included as part of the NRSA sampling frame.
  • NARS_Region - National Aquatic Resource Surveys (NARS) reporting region. The NRSA developed a BMMI for each of the 9 NARS regions and a model was developed, and predictions made, separately for each region.
    • CPL - Coastal Plains
    • NAP - Northern Appalachia
    • NLP - Northern Plains
    • SAP - Southern Appalachia
    • SPL - Southern Plains
    • TPL - Temperate Plains
    • UMW - Upper Midwest
    • WMT - Western Mountains
    • XER - Xeric
  • prG_BMMI - Predicted probability that a stream segment is in good biological condition based on a random forest model of the NRSA BMMI.

• Additional documentation in updated Data Dictionary (HTML|EXCEL) and Variable List

August 31, 2017 - Added 2011 NLCD imperviousness to StreamCat and name corrections for some variables

July 11, 2017 - Corrected NonAgIntrodManagVeg

• The ‘NonAgIntrodManagVeg’ tables on StreamCat were incorrectly using all non-native LANDFIRE existing vegetation types rather than only the non-native vegetation types excluding primarily agricultural vegetation types. This has been corrected in both the NonAgIntrodManagVeg and NonAgIntrodManagVegRipBuf100 tables for both hydro-regions and states.

March 3, 2017 - Reference-condition stream temperature predictions added to StreamCat

• The models of Hill et al. (2013) were applied to a set of specially-produced StreamCat metrics to predict reference-condition mean summer (July-August), mean winter (January-February), and mean annual (January-December) stream temperatures (MSST, MWST, and MAST, respectively).
• The original models contained a predictor variable (stream slope) that was developed with DEM-based, synthetic stream networks. We could not replicate this variable with the StreamCat approach. We excluded this variable from the models but analyses suggest there was a minimal effect on overall predicted temperatures. Full documentation and code used to apply these predictions to NHDPlusV21 stream segments through StreamCat predictors is available.
• Unlike previous StreamCat metrics that provide summaries of local catchment and watershed-level landscape data, these tables represent predictions made to individual NHDPlusV21 stream segments. Therefore, metric names do not use “Cat” and “Ws” suffixes as is the case with other StreamCat data.
• These models used a set of least-disturbed USGS stations and sites to model stream temperatures from a set of landscape metrics.
• These predictions were made to provide estimates of reference-condition stream temperatures in support of the 2008-2009 and 2013-2014 (forthcoming) National Rivers and Streams Assessments.
• Predictions were constrained to only those stream segments that were within the range of predictor values used to develop the temperature models. Values at stream segments with landscape information outside of the range of calibration data were set to NA within tables.
• The StreamCat Data Dictionary and Variable List have been updated with descriptions of these stream temperature predictions that are now available for download. See metadata links below for additional documentation.
• New metrics include:
  • Mean summer stream temperature for yrs. 2008 (MSST_2008), 2009 (MSST_2009), 2013 (MSST_2013), and 2014 (MSST_2014) - MetaData
    • RefStreamTempPred_-RegionID-.zip
    • RefStreamTempPred_-StateID-.zip
  • Mean winter stream temperature for yrs. 2008 (MWST_2008), 2009 (MWST_2009), 2013 (MWST_2013), and 2014 (MWST_2014) - MetaData
    • RefStreamTempPred_-RegionID-.zip
    • RefStreamTempPred_-StateID-.zip
  • Mean annual stream temperature for yrs. 2008 (MAST_2008), 2009 (MAST_2009), 2013 (MAST_2013), and 2014 (MAST_2014) - MetaData
    • RefStreamTempPred_-RegionID-.zip
    • RefStreamTempPred_-StateID-.zip

Jan. 25, 2017 - Compression of StreamCat tables on FTP

• To reduce the size of StreamCat tables, we have zipped all HydroRegion and State tables
• The tables are still flat, comma-delimited text files but compression reduced their size on disk to ~1/3 of their original size

Jan. 25, 2017 - New catchment and watershed metrics added to StreamCat

• The StreamCat Data Dictionary and Variable List have been updated with descriptions of new catchment and watershed metrics that are now available for download. See metadata links below for additional documentation
• Some metrics were developed to support modeling of biological condition of the 2008-2009 National Rivers and Streams Assessment and were, therefore, derived for those specific years (e.g., PRISM climate data from 2008-2009)
• Note that within certain HydroRegions and States, it is possible for certain metrics to have all 0s. These metrics include coal mines, wildfire perimeters, and forest loss
• New metrics include:
  • Coal mines - MetaData
    • CoalMines_-RegionID-.csv
    • CoalMines_-StateID-.csv
  • Agricultural-based nitrogen - MetaData
    • AgriculturalNitrogen_-RegionID-.csv
    • AgriculturalNitrogen_-StateID-.csv
  • Housing-unit density within riparian buffer - MetaData
    • USCensus2010RipBuf100_-RegionID-.csv
    • USCensus2010RipBuf100_-StateID-.csv
  • Wildfire perimeters by year (2000-2010) - MetaData
    • FirePerimeters_-RegionID-.csv
    • FirePerimeters_-StateID-.csv
  • Wildfire perimeters by year (2000-2010) within riparian buffers - MetaData
    • FirePerimetersRipBuf100_-RegionID-.csv
    • FirePerimetersRipBuf100_-StateID-.csv
  • Forest loss by year (2001-2013) - MetaData
    • ForestLossByYear_-RegionID-.csv
    • ForestLossByYear_-StateID-.csv
  • Forest loss by year (2000-2013) within riparian buffers - MetaData
    • ForestLossByYearRipBuf100_-RegionID-.csv
    • ForestLossByYearRipBuf100_-StateID-.csv
  • Non-native, non-agricultural vegetation - MetaData
    • NonAgIntrodManagVeg_-RegionID-.csv
    • NonAgIntrodManagVeg_-StateID-.csv
  • Non-native, non-agricultural vegetation (riparian buffer) - MetaData
    • NonAgIntrodManagVegRipBuf100_-RegionID-.csv
    • NonAgIntrodManagVegRipBuf100_-StateID-.csv
  • Mean air temperature (PRISM Yrs. 2008-2009) - MetaData
    • PRISM_2008_2009_-RegionID-.csv
    • PRISM_2008_2009_-StateID-.csv
  • Mean precipitation (PRISM Yrs. 2008-2009) - MetaData
    • PRISM_2008_2009_-RegionID-.csv
    • PRISM_2008_2009_-StateID-.csv
  • Topographic wetness index - MetaData
    • WetIndx_-RegionID-.csv
    • WetIndx_-StateID-.csv

November 2, 2016 - Modified names of PRISM climate tables and variables

• To more clearly reflect the time period of the data, we changed the names of the PRISM climate tables in StreamCat
• Affected tables:
  • PRISM_-Region ID-.csv changed to PRISM_1981_2000_-Region ID-.csv
  • PRISM_-Region ID-.csv changed to PRISM_1981_2000_-State ID-.csv
• Affected columns (both HydroRegion and State tables):
  • PrecipCat changed to Precip8110Cat
  • TmaxCat changed to Tmax8110Cat
  • TmeanCat changed to Tmean8110Cat
  • TminCat changed to Tmin8110Cat
  • PrecipWs changed to Precip8110Ws
  • TmaxWs changed to Tmax8110Ws
  • TmeanWs changed to Tmean8110Ws
  • TminWs changed to Tmin8110Ws

October 11, 2016 - Olson and Hawkins (2014) geochemical and geophysical characteristics

• StreamCat now contains catchment and watershed summaries of a suite of raster layers depicting the geochemical and geophysical attributes of surface and near-surface geology from Olsen and Hawkins (2014). These layers include the following attributes:
  • Geochemical:
    1. Percent of lithological aluminum oxide (Al2O3)
    2. Percent of lithological calcium oxide (CaO)
    3. Percent of lithological ferric oxide (Fe2O3)
    4. Percent of lithological potassium oxide (K2O)
    5. Percent of lithological magnesium oxide (MgO)
    6. Percent of lithological sodium oxide (Na2O)
    7. Percent of lithological nitrogen (N)
    8. Percent of lithological phosphorus pentoxide (P2O5)
    9. Percent of lithological silicon dioxide (SiO2)
    10. Percent of lithological sulfur (S)
    11. Percent of lithological silicon dioxide (SiO2)
  • Geophysical:
    1. Lithological hydraulic conductivity (micrometers per second)
    2. Lithological compressive strength, measured as uniaxial compressive strength (megaPascals)
• Because of differing occurrences of NoData pixels in several of the geochemical and geophysical rasters, we created four tables so that the percent of pixels within catchments and watersheds with data could be consistently reported for each set. A similar approach was used when summarizing the STATSGO soils rasters for StreamCat.
• New tables:
  • Al2O3, CaO, Fe2O3, K2O, MgO, P2O5, S, and SiO2:
    • GeoChemPhys1_-Region ID-.csv
    • GeoChemPhys1_-State ID-.csv
  • N:
    • GeoChemPhys2_-Region ID-.csv
    • GeoChemPhys2_-State ID-.csv
  • Hydraulic conductivity:
    • GeoChemPhys3_-Region ID-.csv
    • GeoChemPhys3_-State ID-.csv
  • Uniaxial compressive strength:
    • GeoChemPhys4_-Region ID-.csv
    • GeoChemPhys4_-State ID-.csv
• Documentation:
  • Metadata
  • Data Dictionary (HTML|EXCEL)
  • Variable List

June 30, 2016 - Added NABD to StreamCat | Updated NLCD2006RipBuf100 | Corrected Superfund and Mines

• The National Anthropogenic Barriers Dataset (NABD) was summarized and added to StreamCat
  • New tables include:
    • NABD_-RegionID-.csv
    • NABD_-StateID-.csv
  • Metadata are available here:
    • NABD
  • Additional documentation in updated Data Dictionary (HTML|EXCEL) and Variable List
• NLCD2006RipBuf100 tables were updated to be consistent with NLCD2011RipBuf100. These tables originally provided the areas and percent completeness of full catchments (CatAreaSqKm|CatPctFull) and watersheds (WsAreaSqKm|WsPctFull). They have now been updated to contain the area and percent completeness of data within the 100-m riparian buffers within catchments (CatAreaSqKmRp100|CatPctFullRp100) and watershed (WsAreaSqKmRp100|WsPctFullRp100).
  • Affected tables:
  • NLCD2006RipBuf100_-RegionID-.csv
  • NLCD2006RipBuf100_-StateID-.csv
• The Variable List has been corrected for SuperfundDensCat, SuperfundDensWs, MinesDensCat, and MineDensWs. These entries incorrectly stated that the tables contained information on the density of Superfund sites and mines within the 100-m riparian buffer. As of May 3, 2016 (see entry below), riparian data were moved to separate tables.

June 13, 2016 - Error in metric names - NLCD 2011 riparian buffers

• Column names in the 2011 NLCD riparian buffer tables incorrectly contained 2006 (e.g., PctOw2006CatRp100 rather than PctOw2011CatRp100)
• The data in these tables have been verified to contain data from NLCD 2011
• Variable names have been corrected to contain 2011
• Error included tables:
  • NLCD2011RipBuf100_-RegionID-.csv
  • NLCD2011RipBuf100_-StateID-.csv

May 4, 2016 - Added NLCD 2011 within full watershed and 100-m buffers

• Land cover and land use from the National Land Cover Dataset (2011) are now summarized and available for watersheds and within 100-m riparian buffers
• New tables include:
  • NLCD2011_-RegionID-.csv
  • NLCD2011_-StateID-.csv
  • NLCD2011RipBuf100_-RegionID-.csv
  • NLCD2011RipBuf100_-StateID-.csv
• Metadata are available here:
  • NLCD 2011
  • NLCD 2011 - Riparian Buffer
• Additional documentation in updated Data Dictionary (HTML|EXCEL) and Variable List

May 3, 2016 - Separation of riparian data and update of PctFull values

• Until now, riparian features (e.g., mine density in riparian buffers) were often included with tables reporting on full (non-riparian) watershed features
• Riparian features are now separated from these tables into their own set of tables
  • Names of new riparian tables are included in the Variable List
  • New tables include:
    • EPA_FRSRipBuf100_-RegionID-.csv
    • EPA_FRSRipBuf100_-StateID-.csv
    • MinesRipBuf100_-RegionID-.csv
    • MinesRipBuf100_-StateID-.csv
    • USCensus2010RipBuf100_-RegionID-.csv
    • USCensus2010RipBuf100_-StateID-.csv
    • ImperviousSurfaces2006RipBuf100_-RegionID-.csv
    • ImperviousSurfaces2006RipBuf100_-StateID-.csv
    • RoadDensityRipBuf100_-RegionID-.csv
    • RoadDensityRipBuf100_-StateID-.csv
• In addition to being placed in their own tables, the percent completeness of these data within catchments and watersheds (i.e., CatPctFull and WsPctFull, respectively) will be reported for the riparian buffers only.
  • For example, in the previous version of data PctFull value for the catchment marked with ’*’ in the image below would not have distinguished between catchment and riparian values
  • In the new tables, the PctFull of the riparian buffer (pink) is now reported and would be 100% in the image below
  • The CatPctFull would still be <100% in non-riparian tables
  • New riparian tables report the areas of the riparian buffer within catchments and watershed (i.e., CatAreaSqKmRp100 and WsAreaSqKmRp100, respectively) rather than full areas of catchments and watersheds

• A small percentage of NHDPlusV2 catchments are entirely outside of the USA borders
• In these cases, CatPctFull and WsPctFull values for point data (e.g., mines) were incorrectly reported as 100%
• These data have been corrected to be 0% in these cases
• Error included tables:
  • Mines_-RegionID-.csv
  • Mines_-StateID-.csv
  • EPA_FRS_-RegionID-.csv
  • EPA_FRS_-StateID-.csv
  • Dams_-RegionID-.csv
  • Dams_-StateID-.csv
  • NABD_-RegionID-.csv
  • NABD_-StateID-.csv

Mar. 28, 2016 - StreamCat GitHub page and archived code

• The code to generate StreamCat metrics has been substantially modified to improve processing time and workflow
• Code to create the initial set of StreamCat data has been archived
• The latest working code is now housed on the EPA StreamCat GitHub repository

Mar. 18, 2016 - New catchment and watershed metrics added to StreamCat

• The StreamCat Data Dictionary and Variable List have been updated with descriptions of new catchment and watershed metrics that are now available for download. See metadata links below for additional documentation
• New metrics include:
  • Soil erodibility (STATSGO) - MetaData
    • Kffact_-RegionID-.csv
    • Kffact_-StateID-.csv
  • USGS base-flow index - MetaData
    • BFI_-RegionID-.csv
    • BFI_-StateID-.csv
  • National Atmospheric Deposition Program (NADP) - Metadata
    • Ammonium
      • NH4_-RegionID-.csv
      • NH4_-StateID-.csv
    • Nitrate
      • NO3_-RegionID-.csv
      • NO3_-StateID-.csv
  • 30-year climate normals (PRISM 1981-2010) - MetaData
    • Annual minimum temperature
      • Tmin_-RegionID-.csv
      • Tmin_-StateID-.csv
    • Annual maximum temperature
      • Tmax_-RegionID-.csv
      • Tmax_-StateID-.csv
    • Annual mean temperature
      • Tmean_-RegionID-.csv
      • Tmean_-StateID-.csv
    • Annual total precipitation
      • Precip_-RegionID-.csv
      • Precip_-StateID-.csv
  • 2012 National Anthropogenic Barrier Dataset (NABD) - MetaData
    • NABD_-RegionID-.csv
    • NABD_-StateID-.csv

Mar. 7, 2016 - Road density modified to include values in water bodies

• The raster used to calculate road density contained NAs in water bodies (see image below). In some regions, such as the Upper Midwest, these raster NAs resulted in numerous NA values in StreamCat
• In addition, it was decided that water bodies should contain the underlying road density raster values rather than NAs to more accurately represent road densities within local catchments and watersheds
• Modified tables include:
  • RoadDensity_-RegionID-.csv
  • RoadDensity_-StateID-.csv

Feb. 2, 2016 - Dams, Mines, and EPA_FRS data corrected

• An error in calculating point densities (and volumes in the case of dams) affected records with CatPctFull and WsPctFull values < 100%
  • These records were off by 2 decimal places
• Error included tables:
  • Dams_-RegionID-.csv
  • Dams_-StateID-.csv
  • Mines_-RegionID-.csv
  • Mines_-StateID-.csv
  • EPA_FRS-RegionID-.csv
  • EPA_FRS-StateID-.csv
• All tables updated to have correct densities and volumes for these records

Jan. 20, 2016 - Elevation data corrected

• Values in some elevation tables were in meters * 0.1, but were reported in documentation as meters
• Error included tables:
  • Elevation_-RegionID-.csv
  • Elevation_-StateID-.csv
• All tables updated to be in meters

Dec. 9, 2015 - ‘schema.ini’ files added to State and HydroRegion folders

• ArcMap can have trouble interpreting data types (e.g., long vs. float) of table columns when reading StreamCat ‘.csv’ files.
• To ensure correct interpretation of column types by ArcMap, users can download the ‘schema.ini’ files and store them in the same directory as the StreamCat tables:
  • State schema.ini
  • HydroRegion schema.ini

Dec. 8, 2015 - StreamCat JAWRA article published online

Article titled: The Stream-Catchment (StreamCat) Dataset: A Database of Watershed Metrics for the Conterminous United States published by the Journal of the American Water Resources Association

Sept. 28, 2015 - StreamCat Dataset published online

The EPA’s URL gateway to the StreamCat FTP site is now available