Typical American suburban community with model homes.

Sep 15, 2015 • You have likely heard phrases such as, “Welcome to the Anthropocene.” More and more we hear the term “anthropocene” used to describe the current epoch of our planet when humankind has had a profound impact on Earth. This month, the U.S. Geological Survey has released a report and dataset on anthropogenic land use trends in the U.S. between 1974 and 2012.

The 60-m raster datasets cover the conterminous U.S. for five time periods: 1974, 1982, 1992, 2002, and 2012. Lead author James Falcone started with the 2011 Landsat-based National Land Cover Dataset (recoded to land uses) and then mapped backwards in time how those land uses had changed. The dataset focuses on how humans use the land (which correlates with the land’s economic function). These datasets were compiled using an impressively broad array of existing data sources including:

+ NLCD 2011, April 2014 version, 30 meters (m) (Jin and others, 2013) **Landsat-based**
+ USDA Census of Agriculture, 1974–2012, State and county-level data (U.S. Department of Agriculture, 2014a)
+ U.S. Census 2010 hden, provided by GeoLytics Inc., block group scale (GeoLytics, 2012)
+ Spatial Analysis for Conservation and Sustainability (SILVIS) Lab housing unit density, 1970–2000, partial block group scale (Hammer and others, 2004)
+ History Database of the Global Environment (HYDE) 3.1 1970–2005 decadal cropland and pasture 9-km data (History Database of the Global Environment, 2013)
+ USDA/USGS soil capability class, 1:250k (Baker and Capel, 2011)
+ Raster dataset based on STATSGO data processed from Baker and Capel (2011)
+ USGS National Elevation Data (NED)-derived slope, 100 m (U.S. Geological Survey, 2014a)
+ USGS Landfire (Landscape Fire and Rescue Management Planning Tools), 1999–2012 annual disturbance grids, 30 m (U.S. Geological Survey, 2014b) **Landsat-based**
+ National Inventory of Dams (NID) point locations, 2013 (U.S. Army Corps of Engineers, 2013)
+ Dam removal point locations, 1912–2013 (American Rivers, 2014)
+ NLCD 2006 and 2001 land cover/use, 30 m: (Fry and others, 2011; Homer and others, 2007; U.S. Geological Survey, 2014c) **Landsat-based**
+ NLCD “Enhanced” 1992 (NLCDe 1992), 30-m landcover/use (Nakagaki and others, 2007) **Landsat-based**
+ USGS Geographic Information Retrieval and Analysis System (GIRAS) 1970s land cover/use (Price and others, 2007)
+ Protected Areas Database, PADUS_CBI_Edition V2 (Conservation Biology Institute, 2013)
+ USGS historical oil/gas well locations
+ U.S. Environmental Protection Agency (EPA) point locations of mining, landfill, metal, and nonmetal processing sites, and oil/gas exploration sites, from Envirofacts MultiSystem Query (U.S. Environmental Protection Agency, 2014a)
+ NHD (National Hydrography Dataset) 1:24k water bodies (U.S. Geological Survey, 2012) and NHDP- lus V2 streamlines (U.S. Environmental Protection Agency, 2014b)
+ USDA NRI (U.S. Department of Agriculture, 2013), 1982–2012 and ERS (U.S. Department of Agriculture, 2014b), 1974–2007: State-level statistics and USGS LC Trends (Sleeter and others, 2013), 1973–2000, ecoregion statistics
+ USGS Land Cover Trends **Landsat-based**
+ USDA Cropland Data Layer (CDL) 2009, 56 m, updated version, downloaded November 2014 (U.S. Department of Agriculture, 2014c) **Uses Landsat data**
+ Homeland Security Infrastructure Program (HSIP) 2012 land use polygons (n ~ 100,000), lines, and points (Homeland Security Infrastructure Program, 2012)
+ Esri USA landmark (important land use features) polygons (n ~ 80,000) and lines (Esri, 2014a)
+ David Theobald national land use dataset (Theobald, 2014)
+ Census 2000 roads, provided by GeoLytics, Inc. (GeoLytics, 2001)
+ USGS 1992–2001 30-m Retrofit Land Cover Change Product (U.S. Geological Survey, 2014c) **Landsat-based**
+ Massachusetts State historical land use for 1971, 1985, and 1999, 1:25,000 (Massachusetts Office of Geographic Information (MassGIS); State of Massachusetts, 2014)
+ U.S. Department of Commerce Census Bureau block-level employment data (U.S. Department of Commerce, 2014)
+ U.S. Department of Commerce Census Bureau 2010 block boundaries (U.S. Department of Commerce, 2012)
+ In-house derived land use polygons database (n ~ 45,000), in-house-derived conservation areas change data (n ~ 1,400), and in-house derived reservoir change data polygons (n ~ 2,900)

Report’s abstract:
This dataset provides a U.S. national 60-meter, 19-class mapping of anthropogenic land uses for five time periods: 1974, 1982, 1992, 2002, and 2012. The 2012 dataset is based on a slightly modified version of the National Land Cover Database 2011 (NLCD 2011) that was recoded to a schema of land uses, and mapped back in time to develop datasets for the four earlier eras. The time periods coincide with U.S. Department of Agriculture (USDA) Census of Agriculture data collection years.

Changes are derived from (a) known changes in water bodies from reservoir construction or removal; (b) housing unit density changes; (c) regional mining/extraction trends; (d) for 1999–2012, timber and forestry activity based on U.S. Geological Survey (USGS) Landscape Fire and Resource Management Planning Tools (Landfire) data; (e) county-level USDA Census of Agriculture change in cultivated land; and (f) establishment dates of major conservation areas.

The data are compared to several other published studies and datasets as validation. Caveats are provided about limitations of the data for some classes. The work was completed as part of the USGS National Water-Quality Assessment (NAWQA) Program and termed the NAWQA Wall-to-Wall Anthropogenic Land Use Trends (NWALT) dataset. The associated datasets include five 60-meter geospatial rasters showing anthropogenic land use for the years 1974, 1982, 1992, 2002, and 2012, and 14 rasters showing the annual extent of timber clearcutting and harvest from 1999 to 2012.

Go to the report:
+ U.S. conterminous wall-to-wall anthropogenic land use trends (NWALT), 1974–2012