You are here:
Completion of the National Land Cover Database (NLCD) 1992-2001 Land Cover Change Retrofit Product
Fry, J. A., M. J. Coan, C. G. Homer, D. K. Meyer, AND J. D. WICKHAM. Completion of the National Land Cover Database (NLCD) 1992-2001 Land Cover Change Retrofit Product. U.S. Geological Survey, Reston, VA, 2009.
The Multi-Resolution Land Characteristics Consortium (MRLC; http://www.mrlc.gov) has supported the development of two national land cover products including the National Land Cover Dataset 1992, and the National Land Cover Database (NLCD) 2001. Many users would like to compare these two products to derive land cover change, but new improvements in mapping methodology, input data, and minor mapping legend modification for NLCD 2001 have made direct comparison between NLCD 1992 and NLCD 2001 inadvisable. Direct comparison of these two independently created land cover products is not recommended and will typically result in changes that are artifacts of different production methods rather than true change. Users that pursue direct comparison are likely to discover that differences in the methodology used to produce the two products overwhelm actual differences related to land cover change. The NLCD design team has recognized that these legacy issues exist and developed a retrofit change product to accommodate direct comparison of NLCD 1992 and NLCD 2001. The goal of this project was to design and implement an operational solution to change analysis, based on 30-m resolution Landsat imagery that minimized error from direct comparison of the original land cover products (Vogelmann, and others, 2001; Homer and others, 2004; Homer and others, 2007), and provided invariant, standardized, and consistent land cover change information across all mapping zones in the conterminous United States. The essential design requirements were that the process was to be rigorous, automated, affordable, and fast. The product presented here, called the NLCD 1992–2001 Land Cover Change Retrofit (NLCD–LCCR) product, provides more accurate and tenable estimates of land cover change at the Anderson Level I classification scale (Anderson and others, 1976).
The Multi-Resolution Land Characteristics Consortium has supported the development of two national digital land cover products: the National Land Cover Dataset (NLCD) 1992 and National Land Cover Database (NLCD) 2001. Substantial differences in imagery, legends, and methods between these two land cover products must be overcome in order to support direct comparison. The NLCD 1992–2001 Land Cover Change Retrofit product was developed to provide more accurate and useful land cover change data than would be possible by direct comparison of NLCD 1992 and NLCD 2001. For the change analysis method to be both national in scale and timely, implementation required production across many Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETM+) path/rows simultaneously. To meet these requirements, a hybrid change analysis process was developed to incorporate both post-classification comparison and specialized ratio differencing change analysis techniques. At a resolution of 30 meters, the completed NLCD 1992–2001 Land Cover Change Retrofit product contains unchanged pixels from the NLCD 2001 land cover dataset that have been cross-walked to a modified Anderson Level I class code, and changed pixels labeled with a “from-to” class code. Analysis of the results for the conterminous United States indicated that about 3 percent of the land cover dataset changed between 1992 and 2001.