Download (PDF 800 KB): https://research.fs.usda.gov/download/treesearch/34984.pdf
Abstract
An accuracy assessment method involving double sampling and the multivariate composite estimator has been used to validate the prototype seasonal land cover characteristics database of the conterminous United States. The database consists of 159 land cover classes, classified using time series of 1990 1-km satellite data and augmented with ancillary data including terrain, climate, and ecological definitions. Reference data from more than 2,200 1 km2 plots were independently collected; data included descriptions of dominant land cover types. Validity of the database is presented in terms of conditional probabilities and variability estimates. This case study demonstrates the usefulness of the method in describing the reference composition of the mapped categories, handling different classification schemes between the database and reference, and comparing various regrouping complexities. The validation results support using the database as a vegetation baseline in large scale environmental simulation models.
Parent Publication
Citation
Zhu, Zhiliang; Ohlen, Donald O.; Czaplewski, Raymond L.; Burgan, Robert E. 1996. Alternative method to validate the seasonal land cover regions of the conterminous United States. In: Mowrer, H. Todd; Czaplewski, Raymond L.; Hamre, R. H., tech coords. Spatial Accuracy Assessment in Natural Resources and Environmental Sciences: Second International Symposium. May 21-23, 1996. General Technical Report RM-GTR-277. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station. p. 409-418.
