Conventional and fuzzy accuracy assessment of the New York Gap Analysis Project land cover map

The accuracy of a regional-scale thematic map of land cover was assessed using conventional and fuzzy methods. The mapping process integrated Landsat Thematic Mapper (TM) data with expert knowledge to map land cover for a 12 million-ha area. Accuracy assessment was based on a stratified random sample of 113 sampling areas. Paired observed and predicted land cover types for 9745 polygons for conventional accuracy and 933 polygons for fuzzy accuracy were collected. Overall map accuracies using conventional methods ranged from 74% to 42% for maps with increasing taxonomic resolution. Fuzzy map accuracies were assessed at low and high taxonomic resolutions resulting in an improvement in map accuracy of 19% and 23%, respectively. The nature, magnitude, and frequency of errors associated with mapping land cover types for large land areas at different levels of taxonomic resolution are reported, and two accuracy assessment methods are compared with respect to information content and opportunities for improving map quality.