Optimal resolution of soil properties maps varies according to their geographical extent and location

The important development of digital soil mapping (DSM) these last decades has led to a large number maps properties with increasingly finer raster size. Map resolution is mostly determined by expert knowledge or matching the existing data, while scale recognized as major issue. Using pH and C/N ratio describing surface horizon forest soils estimated bioindication, we evaluated effect changes on model map performance for different geographical extents. 40,663 plots from national inventory 25 environmental variables calculated at eight spatial levels (50, 100, 250, 500, 1000, 8000, 16,000, 50,000 m), modeled mapped over vast diversified area 91,000 km2 in north-east France. models highlighted importance geology variations, lesser extent stand type, climate topography, slight influence data predictor selection. On contrary, accuracy decreased significantly above 1000 m resolution. Significant differences were observed according location size extent. Globally, more heterogeneous characteristics smaller extent, better fine performed. In addition, aggregation fine-resolution coarser cell improved compared direct use coarse-resolution predictors. impact varies procedure, local environment, should be DSM studies optimize accuracy.