The StreamCat Dataset provides summaries of natural and anthropogenic landscape features for ~2.65 million streams, and their associated catchments, within the conterminous USA. This dataset is associated with the following publication: Hill, R.A., M. Weber , S. Leibowitz , T. Olsen , and D.J. Thornbrugh. The Stream-Catchment (StreamCat) Dataset: A database of watershed metrics for the conterminous USA. JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION. American Water Resources Association, Middleburg, VA, USA, 9, (2015).

This dataset consists of predicted probabilities of good biological condition based in the US EPA 2008/2009 National Rivers and Streams Assessment (NRSA). NRSA assesses the biological condition of rivers and streams using several approaches, including a benthic invertebrate multimetric index (BMMI). The development of the NRSA BMMI is documented in the 2008/2009 NRSA Report (https://www.epa.gov/national-aquatic-resource-surveys/national-rivers-and-streams-assessment-2008-2009-results) and by Stoddard et al. (2008) (http://www.bioone.org/doi/abs/10.1899/08-053.1). This assessment resulted in the classification of 1,380 streams as being in good or poor biological condition. These sites were paired with StreamCat data and a random forest model was developed to predict the probable condition of streams based on the binary response of condition to catchment and watershed features. This model was then applied to NHDPlusV2 stream segments that were within the NRSA sampling frame, i.e., streams that were candidates for sampling during the 2008/2009 NRSA (~1.1 million stream segments). Model development was documented in Fox et al. (2017) (https://link.springer.com/article/10.1007/s10661-017-6025-0) and Hill et al. (2017)(http://onlinelibrary.wiley.com/doi/10.1002/eap.1617/full).

This dataset represents the Index of Watershed Integrity / Index of Catchment Integrity (IWI/ICI) within individual local NHDPlusV2 catchments and upstream, contributing watersheds based on 23 other StreamCat metrics. The Index of Watershed Integrity (IWI) is based on first order approximations of relationships between stressors and six watershed functions: hydrologic regulation, regulation of water chemistry, sediment regulation, hydrologic connectivity, temperature regulation, and habitat provision. Link to paper: https://doi.org/10.1016/j.ecolind.2017.10.070 The Index of Watershed Integrity / Index of Catchment Integrity (IWI/ICI) were summarized to produce local catchment-level and watershed-level metrics as a continuous data type.

This dataset represents the elevation values within individual local NHDPlusV2 catchments and upstream, contributing watersheds based on the National Elevation Dataset (see Data Sources for links to NHDPlusV2 data and NED data). NHDPlusV2 records NED snapshot dates as follows: August 2010 - VPU04 February 2011 - VPUs 05, 06 June 2011 - VPU 17 August 2011 - VPUs 07, 10L, 10U, 11, 18 December 2011 - VPUs 01, 02, 03N, 03S, 03W, 08, 09, 12, 13, 14, 15, 16. The elevation characteristics were summarized to produce local catchment-level and watershed-level metrics as a continuous data type.

This dataset represents the soil characteristic within individual, local NHDPlusV2 catchments and upstream, contributing watersheds based on the STATSGO landscape rasters. Attributes of the landscape layer were calculated for every local NHDPlusV2 catchment and accumulated to provide watershed-level metrics. This data set is derived from the STATSGO landscape rasters for the conterminous USA. Individual rasters (Landscape Layers) of depth to bedrock (rckdep), organic material (om), percent clay (clay), percent sand (sand), permeability (perm), soil erodibility (KFFACT/KFACT), and water table depth (wtdep) were used to calculate soil characteristics for each NHDPlusV2 catchment. The soil characteristics were summarized to produce local catchment-level and watershed-level metrics as a continuous data type. The STATSGO data are distributed in two sets, STATSGO_Set1 and STATSGO_Set2, based on common NoData locations in each set of soil GIS layers.

This dataset represents data derived from the NLCD dataset and the National Hydrography Dataset version 2.1(NHDPlusV2) (see Data Sources for links to NHDPlusV2 data and NLCD). Attributes were calculated for every local NHDPlusV2 catchment and accumulated watershed to provide watershed-level metrics for classes within the NLCD. This data set is derived from the NLCD raster composed of 16 of the modified Anderson land cover classes (categorical data type) for the conterminous USA (excluding the four Alaska-specific land cover classes). Additional agriculture on slope metrics were derived using slope based on National elevation DEMs delivered with NHDplusV2 for agriculture NLCD classes. The NLCD raster was produced based on a decision-tree classification of 2001, 2004, 2006, 2008, 2011, 2013, 2016, and 2019 Landsat satellite data (see Data Structure and Attribute Information for a description of each metric). This dataset will include additional years as they become available.

This dataset represents geochemical or geophysical attributes in surface or near surface geology within individual, local NHDPlusV2 catchments and upstream, contributing watersheds. Attributes of the landscape layer were calculated for every local NHDPlusV2 catchment and accumulated to provide watershed-level metric. For information regarding how the Landscape layers were created see https://www.sciencebase.gov/catalog/item/53481333e4b06f6ce034aae7. Landscape Layers are partitioned into 4 tables based on the location of no-data cells within their rasters to correctly reflect the PctFull attributes within each table.

This dataset represents the percent of non-agricultural, non-native vegetation based on LANDFIRE existing vegetation type (EVT) for a 30-m grid cell within individual local NHDPlusV2 catchments and upstream, contributing watersheds based on field reference data and Landsat, elevation, and ancillary data. EVTs are mapped using decision tree models, field data, Landsat imagery, elevation, and biophysical gradient data. Decision tree models are developed separately for each of the three lifeforms -tree, shrub, and herbaceous and are then used to generate lifeform specific EVT layers. The LF-GAP Map Units Descriptions provide descriptions for each LF EVT including species, distribution and classification information. Vegetation map units are primarily derived from NatureServe's Ecological Systems classification, alliances of the U.S. National Vegetation Classification (USNVC), and the National Land Cover Database and LF specific types. LANDFIRE EVT groups were reclassified into introduced managed vegetation cover where EVT_GP = (701,702,703,704,705,706,707,708,709,711,731).

This dataset represents density of total fresh surface-water withdrawals in agricultural land within individual, local NHDPlusV2 catchments and upstream, contributing watersheds. Measured as L/day as described in DOI: 10.1016/j.scitotenv.2020.137661

This dataset represents data derived from the NLCD dataset and the National Hydrography Dataset version 2.1(NHDPlusV2) (see Data Sources for links to NHDPlusV2 data and NLCD). Attributes were calculated for every local NHDPlusV2 catchment and accumulated upstream catchments to provide watershed-level metrics for imperviousness values within the NLCD. This data set is derived from the NLCD Impervious Surfaces raster, which describes percent imperviousness (continuous data type). Values indicate the degree to which the area is composed of impervious anthropogenic materials (e.g., parking surfaces, roads, building roofs). This raster was produced based on a decision-tree classification of 2001, 2004, 2006, 2008, 2011, 2013, 2016, and 2019 Landsat satellite data. This dataset will include additional years as they become available.

This dataset represents the mine density within individual, local NHDPlusV2 catchments and upstream, contributing watersheds based on mine plants and operations monitored by the USGS National Minerals Information Center. The National Minerals Information Center canvasses the nonfuel mining and mineral-processing industry in the United States for data on mineral production, consumption, recycling, stocks, and shipments. Mine plants and operations for commodities are expressed as points in a shapefile that was downloaded from the USGS directly. The (mines / catchment) were summarized and accumulated into watersheds to produce local catchment-level and watershed-level metrics as a point data type.

This dataset represents the estimated density of georeferenced sites within individual, local NHDPlusV2 catchments and upstream, contributing watersheds based on the EPA's Facility Registry Services (FRS) geodatabase. Attributes of the landscape layer were calculated for every local NHDPlusV2 catchment and then accumulated to provide watershed-level metrics. The FRS geodatabase is a collection of point locations of facilities or sites subject to environmental regulation. TRI, NPDES, and Superfund sites were extracted individually to summarize for each in the resulting .csv. (see Data Sources for links to NHDPlusV2 data and FRS data) The (site locations / catchment) were summarized and accumulated into watersheds to produce local catchment-level and watershed-level metrics as a points data type (see Data Structure and Attribute Information for a description of each metric).

This dataset represents the characterization of global forest extent and change by year from 2001 through 2013 within individual local NHDPlusV2 catchments and upstream, contributing watersheds based on the Global Forest Change 2000-2013. These data are based on global tree cover loss for the period from 2001 to 2013 at a spatial resolution of 30m. The analysis used to create the landscape layer is based on Landsat data. Forest loss was defined as a stand-replacement disturbance or the complete removal of tree cover canopy at the Landsat pixel scale. This landscape layer is a disaggregation of total forest loss to annual time scales. Encoded as either 0 (no loss) or else a value in the range 1, representing loss detected primarily in the year 2000-2013, respectively. The forest loss by year characteristics (%) were summarized to produce local catchment-level and watershed-level metrics as a continuous data type.

This dataset represents predicted channel widths and depths from Doyle et al. 2023. Values include: Predicted wetted width: distance of the water\u2019s edge from left to right bank, Predicted thalweg depth: deepest point in the channel cross section from the bottom substrate to the water surface, Predicted bankfull width: distance from left to right bank at bankfull stage where the potential water height would spill outside of the channel and into the floodplain, and Predicted bankfull depth: thalweg depth plus bankfull height, which is the height from the water surface to the bankfull stage.

This dataset represents density of septic systems within individual, local NHDPlusV2 catchments and upstream, contributing watersheds. The data is based on the 1990 U.S. Census.

This dataset represents deposition estimates of nutrients within individual local NHDPlusV2 catchments and upstream, contributing watersheds based on the National Atmospheric Deposition Program. The National Trends Network provides long-term records of precipitation chemistry across the United States. Individual rasters describe ammonium, nitrate, inorganic nitrogen, and average sulfur/nitrogen deposition per year. See Source Info for links to NADP. The nitrogen and sulfur characteristics (kg N/ha/yr) were summarized to produce local catchment-level and watershed-level metrics as a continuous data type.

This dataset represents Nitrogen from rock weathering (kg/ km2) within AOI

Publication of EPA’s Nutrient Inventory is a critical step towards thorough mapping and accounting of sources of N and P to US landscapes. However, summaries of nutrients within accumulative watersheds are needed to develop accurate watershed-level nutrient budgets and relate landscape inputs to instream nutrient concentrations. This subproduct will accumulate the Nutrient Inventory across available years for all streams and lakes within the medium resolution National Hydrography Dataset Plus version 2 (NHDPlus), i.e., 2.6 million stream segments and nearly 400,000 lakes across the conterminous US. These data will allow OW to easily and rapidly identify the dominant sources of N or P for any stream segment or lake in the US. Further, these data will be made accessible through the EPA’s StreamCat and LakeCat datasets and a soon-to-be released online database and an application programming interface (API). This database and API will make nutrient watershed accumulations readily accessible and easily integrated by a variety of OW programs and tools. Finally, the accumulated nutrient data will serve as the basis for a multiple proposed StRAP subproducts and models in SSWR.401, SSWR.404, and SSWR.405. These data will contribute directly to OW, region, and state efforts to identify and reduce non-point nutrient sources. Having spatially explicit data about nutrient sources and loads can help target and inform restoration and conservation efforts, as well as more formal TMDLs, nutrient reduction plans, and groundwater management approaches. This subproduct will produce a database of accumulated nutrient values for at least 2.6 million stream segments and 400,000 lakes of the medium resolution National Hydrography Dataset Plus version 2 (NHDPlus). These data will be made accessible through the StreamCat and LakeCat datasets. They will also be made available as an online database with application programming interface (API) that will facilitate data acquisition and use by OW and state partners. This database will provide a state-of-the science accounting of nutrient sources that drain to all streams and lakes in the conterminous US. It will allow EPA and state partners to identify dominant sources of N and P to individual waterbodies and will greatly facilitate nutrient reduction strategies and planning.

This dataset represents the density of road and stream crossings within individual, local NHDPlusV2 catchments and upstream, contributing watersheds. Attributes of the landscape layer were calculated for every local NHDPlusV2 catchment and then accumulated to provide watershed-level metrics. The landscape layer (raster) was developed by James Falcone of the USGS. US Census TIGER 2000 line files of roads and the NHDPlusV1 line files of all streams were converted to 30-meter grids where the presence of a street or stream was a 1 and everything else a 0. These were intersected and anything that was a 1 in both grids is the result. The density of road and stream crossings (crossings / square kilometer) were summarized to produce local catchment-level and watershed-level metrics as a continuous data type.

This dataset represents the estimated surface water runoff within individual, local NHDPlusV2 catchments and upstream, contributing watersheds. Attributes of the landscape layer were calculated for every local NHDPlusV2 catchment and then accumulated to provide watershed-level metrics.(see Data Sources for links to NHDPlusV2 data and metadata) The landscape layer (raster) was developed with a water-balance model developed by Dave Wolock of the USGS and is detailed further in the paper "Independent effects of temperature and precipitation on modeled runoff in the conterminous United States". McCabe and Wolock[2011] Runoff is defined as the flow per unit area delivered to streams and rivers in units of millimeters per month. The runoff estimates were summarized to produce local catchment-level and watershed-level metrics as a continuous data type.