A Comparison of Raster-Based Forestland Data in Cropland Data Layer and the National Land Cover Database

The National Agricultural Statistics Service, the statistical arm of US Department Agriculture, and Multi-Resolution Land Characteristics Consortium, a group federal agencies, collect publish several land-use land-cover data sets. aim this study is to analyze consistency forestland estimates based on two widely used, publicly available products: Land-Cover Database (NLCD) Cropland Data Layer (CDL). Both remote-sensing-based products provide raster-formatted categorization at spatial resolution 30 m. Although processing yearly published CDL non-agricultural less frequently updated NLCD, large-area mapping between these datasets has not been assessed. To assess similarities differences CDL- NLCD-based mappings for state North Carolina, we overlay years 2011 2016 in ArcMap 10.5.1 location attributes matched mismatched forestland. We find that mismatch relatively smaller areas where forests occupy larger shares total land, relative when compared 2016. also large portion attributable dynamics re-growth periodically harvested otherwise disturbed forests. Our results underscore need holistic approach preparation, attribution, accuracy performing high-scale map-based analyses using each products.