D83991AF-31B5-4B03-8218-268E70F7D3EB eng utf8 dataset dataset James Wickham U.S. Environmental Protection Agency, Office of Research and Development, Center for Public Health and Environmental Assessment 919.541.3077 109 TW Alexander Dr. Durham NC 27709 US wickham.james@epa.gov pointOfContact 2020-12-03 ISO 19139 Geographic Information - Metadata - Implementation Specification 2007 EPSG:4326 NLCD 2016 Land Cover Reference Data, Conterminous United States 2020-11-23 publication The U.S. Geological Survey (USGS), in partnership with several federal agencies, has developed and released five National Land Cover Database (NLCD) products over the past two decades: NLCD 1992, 2001, 2006, 2011 and 2016. U.S. EPA leads the accuracy assessment of NLCD land cover products in coordination with USGS and SUNY-ESF. The posted dataset includes map labels and reference labels for 4629 sample locations (pixels) from the NLCD 2016 database. The sample pixels were selected using a stratified random design based on the dual 2011 – 2016 map labels. The data can be used to re-create the accuracy results reported in a forthcoming paper and in support of other applications. Accuracy results are reported for the 2011 and 2016 land cover products in the NLCD2016 database and for selected 2011-2006 changes. U.S. Environmental Protection Agency James Wickham U.S. Environmental Protection Agency, Office of Research and Development, Center for Public Health and Environmental Assessment 919.541.3077 109 TW Alexander Dr. Durham NC 27709 US wickham.james@epa.gov publisher United States United States place EPA Place Names 2015-01-31 publication 020:072 020:072 theme Federal Program Inventory 2013-09-16 publication Climate Land Climate Environment Land theme EPA GIS Keyword Thesaurus 2007-11-02 publication 010 digital spatial data GIS Image processing ImageryBaseMapEarthCover Land cover Land Use Land Cover Theme National Geospatial Data Asset U.S. Geological Survey (USGS) 010 digital spatial data GIS Image processing ImageryBaseMapEarthCover Land cover Land Use Land Cover Theme National Geospatial Data Asset NGDA U.S. Geological Survey (USGS) theme User 2020-11-23 publication None eng utf8 -127.700452 -66.103258 23.360434 51.42191 2020-11-25T00:00:00 Corner Coordinates (center of pixel, projection meters) Upper Left Corner: -2493045 meters(X), 3310005 meters(Y) Lower Right Corner: -177285 meters(X), 2342655 meters(Y) https://edg.epa.gov/EPADataCommons/Public/ORD/EnviroAtlas/NLCD2016RefData.zip download download https://www.mrlc.gov/data information download dataset NLCD2016 reference data collection and analysis process: NLCD2016 2011 and 2016 land cover components were combined in a GIS and the duals map labels were used to define a map of 21 strata. Stratum 1 -Water, no change; Stratum 2 - Open urban, no change; Stratum 3 - Low intensity urban, no change; Stratum 4 - Medium intensity urban, no change; Stratum 5 - High intensity urban, no change; Stratum 6 - Barren, no change; Stratum 7 - Deciduous forest, no change; Stratum 8 - Evergreen forest, no change; Stratum 9 - Mixed forest, no change; Stratum 10 - Shrubland, no change; Stratum 11 - Grassland, no change; Stratum 12 - Pasture, no change; Stratum 13 - Cropland, no change; Stratum 14 - Woody wetlands, no change; Stratum 15 - Herbaceous wetlands, no change; Stratum 16 - Catch-all stratum (all 2011─2016 change pixels not in strata 17─21; Stratum 17 - Urban gain, not urban (2011) to urban (2016); Stratum 18 - Forest loss, forest (2011) to not forest (2016) except forest to urban; Stratum 19 - Forest gain; not forest (2011) to forest (2016); Stratum 20 - Agriculture (Ag) loss, Ag (2011) to not Ag (2016) except Ag to forest or urban; Stratum 21 - Ag gain, not Ag (2011) to Ag (2016), except for forest to Ag. The strata map was converted to 21 individual binary maps (e.g., stratum 1; nodata) to generate a random number map of the stratum; pixels with random numbers corresponding to the number of target samples were selected. Selected sample pixels were converted to points for each stratum, and stratum points were then mosaicked into a single map and their order was randomized. Sample points (pixels) were provided to photointerpreters to collect reference data from Google Earth imagery. Reference data collection was conducted by two interpreters who have collected reference data for all previous NLCD accuracy assessments and who have been involved in the production of all previous NLCD databases. Reference data collection was done in a blind format (i.e., collection was done without knowledge of map classification). Reference data collection was also supported by weekly web-enabled conference calls to promote consistency among interpreters and resolve labeling for samples whose interpretation was difficult.