ABSTRACT

ABSTRACT:   Based on the daily rainfall data of 84 meteorological stations from 1980 to 2009 in Zhejiang Province, the power function model was used to calculate rainfall erosivity. The frequency analysis method, the Mann–Kendall test method and the cumulative deviation method were used to mine data changes. And the spatial variation characteristic was analyzed by the radial basis function interpolation method. The results showed that the relationship of distribution area in different periods was R l (low rainfall erosivity) > R e (extreme rainfall erosivity) > R m (medium rainfall erosivity) > R h (high rainfall erosivity). The area mutation of the R l and R m were detected in the early 21st century. And in 1980s, the area under R h and R e also mutated. The variation process of rainfall erosivity in the 30 years can be divided into three successive stages. In spatial distribution, central lower grade rainfall erosivity gradually spread into marginal higher grade rainfall erosivity.