《Table 1 The environmental predictors used in the machine learning methodologies》

  提示：宽带有限、当前游客访问压缩模式

本系列图表出处文件名：随高清版一同展现

《Modelling and mapping soil erosion potential in China》

1. 获取 高清版本忘记账户？点击这里登录

1. 下载图表忘记账户？点击这里登录

1) DEM，digital elevation model；MrVBF，multi resolution index of valley bottom flatness；NDVI，the normalized difference vegetation index；NPP，net primary productivity；SOC，soil organic carbon.2) SRTM，Shuttle Radar Topography Mission；CMA，China Meteorologi

In addition，geographically weighted regression（GWR）（Xu et al.2015；Ma et al.2017） was applied to downscale R factor to 1-km spatial resolution.Some key steps were described in Ma et al.（2017）and Teng et al.（2017）.The environmental variables，including terrain attributes，climatic variables，vegetation and soil（Table 1），were chosen in the downscaling procedure.