We examined the use of coarse resolution land cover data (USGS LUDA) to accurately discriminate ecoregions and landscape-scale features important to biodiversity monitoring and management. We used land cover composition and landscape indices, correlation and principal components analysis, and comparison with finer-grained Landsat TM data, to assess how well LUDA data discriminate changing patterns across an agriculture-forest gradient in Minnesota, U.S.A. We found LUDA data to be most accurate at general class levels of agriculture and forest dominance (Anderson Level I), but inconsistent and limited in ecotonal areas of the gradient and within forested portions of the study region at finer classes (Anderson Level II). We expected LUDA to over-represent major (matrix) cover types and under-represent minor types, but this was not consistent with all classes. 1) Land cover types respond individualistically across the gradient, changing landscape grain as well as their spatial distribution and abundance. 2) Agriculture is not over-represented where it is the dominant land cover type, but forest is over-represented where it is dominant. 3) Individual forest types are under-represented in an open land matrix. 4) Within forested areas, mixed deciduous-coniferous forest is over-represented by several orders of magnitude and the separate conifer and hardwood types under-represented. Across gradual, transitional agriculture-forest areas, LUDA cover class dominance changes abruptly in a stair-step fashion. In general, rare cover types that are discrete, such as forest in agriculture or wetlands or water in forest, are more accurately represented than cover classes having lower contrast with the matrix. Northward across the gradient, important changes in the proportions of conifer and deciduous forest mixtures occur at scales not discriminated by LUDA data. Results suggest that finer-grained data are needed to map within-state ecoregions and discriminate important landscape characteristics. LUDA data, or similar coarse resolution data sources, should be used with caution and the biases fully understood before being applied in regional landscape management.
- Forest ecosystems
- Land cover data
- Landsat TM
- Principal components analysis