Mapping grazing intensity using remote sensing in the Xilingol steppe region, Inner Mongolia, China

Satellite remote sensing has been used in many fields but less commonly for monitoring grazing intensity (GI). In this study, grassland above-ground biomass (AGB) was simulated by multispectral reflectance derived from an artificial neural network (ANN) to investigate GI. One-way analysis of variance (ANOVA) and frequency histograms from sample plots were used to determine the thresholds of AGB for four levels of GI characterized as ungrazed, lightly grazed, moderately grazed and heavily grazed. The distribution of GI in the Xilingol steppe region of Inner Mongolia, China, was then mapped from the remotely sensed grassland AGB. This study could be used as a guide for management interventions and grassland restoration.