An evaluation of VEGETATION-1 imagery for broad-scale landscape mapping of Russia: effects of resolution on landscape pattern

The problem of efficient use of multi-scale data for land-cover and landscape mapping has already attracted considerable attention in landscape ecology and some other disciplines. Over the last few decades, however, with the development of satellite remote sensing techniques, the questions of efficient planetary and macro-regional ecological mapping and modeling and integration of multi-scale data have become especially pertinent. The purpose of this study is to test the suitability of the coarse-resolution VEGETATION/SPOT imagery for landscape mapping. Effects of changing spatial resolution on land-cover proportion estimates were examined in 16 different landscapes using spatially degraded high-resolution imagery from the Russian satellite, RESURS-F. Comparison of simulated coarse-resolution data and original images and ancillary data for sites with different landscape patterns, showed strong scale dependence of the landscape characteristics (land-cover proportion, fragmentation, patchiness) in different case study areas, such as sub-boreal forest, sub-urban areas, and an agricultural region in the steppe zone.