Setting the Flow Accumulation Threshold Based on Environmental and Morphologic Features to Extract River Networks from Digital Elevation Models

Determining the flow accumulation threshold (FAT) is a key task in extraction of river networks from digital elevation models (DEMs). Several methods have been developed to extract Digital Elevation Models. However, few studies considered geomorphologic complexity FAT estimation and network extraction. Recent estimated influencing factors’ impacts on length or drainage density without considering anthropogenic landscape patterns. This study contributes two methods. The first method explores statistical association between 47 tentative explanatory factors. Specifically, multi-source data, including meteorologic, vegetation, anthropogenic, landscape, lithology, topologic characteristics are incorporated into density-FAT model basins with complex topographic environmental characteristics. Non-negative matrix factorization (NMF) was employed evaluate predictive performance. second exploits fractal geometry theory estimate at regional scale, that is, whose large areal extent precludes use basin-wide representative regression predictors. paper’s methodology applied data acquired for Hubei Qinghai Provinces, China, 2001 through 2018 systematically tested visual criteria. Our results reveal local features useful within context relatively small spatial scales establish importance properly choosing multifractal exhibits more accurate extracting than box-counting scale.