ABSTRACT
California’s hardwood rangeland resource was historically characterized by low use and low value. Hardwood stocking has declined, as have the number of acres of hardwood rangeland as the resource has been converted to industrial, residential, and intensive agricultural uses. Both the 1981 and 1990 projects were concerned with mapping the extent, type, and condition of California hardwood rangeland. Despite known spatial autocorrelation in the distribution of hardwood rangeland cover types, clusters of sample polygons were selected for both photo-interpreted and field-visited sites because of the travel and set-up time cost savings resulting from the use of clusters of polygons. The preliminary accuracy assessment sampling design anticipated that the 1981 vegetation type polygon coverage could be used as the sampling units for accuracy assessment of both the 1981 and the 1990 maps. Substantially more acres of hardwood rangeland exist than reported in the 1981 photo-based mapping project.
