New gridded dataset of rainfall erosivity (1950–2020) on the Tibetan Plateau

Abstract

Abstract. The risk of water erosion on the Tibetan Plateau (TP), a typical fragile ecological area, is increasing with climate change. A rainfall erosivity map is useful for understanding the spatiotemporal pattern of rainfall erosivity and identifying hot spots of soil erosion. This study generates an annual gridded rainfall erosivity dataset on a 0.25∘ grid for the TP in 1950–2020. The 1 min precipitation observations at 1787 weather stations for 7 years and 0.25∘ hourly European Center for Medium-Range Weather Forecasts Reanalysis 5 (ERA5) precipitation data for 71 years are employed in this study. Our results indicate that the ERA5-based estimates have a marked tendency to underestimate annual rainfall erosivity when compared to the station-based estimates, because of the systematic biases of ERA5 precipitation data including the large underestimation of the maximum contiguous 30 min peak intensity and relatively slight overestimation of event erosive precipitation amounts. The multiplier factor map over the TP, which was generated by the inverse distance-weighted method based on the relative changes between the available station-based annual rainfall erosivity grid values and the corresponding ERA5-based values, was employed to correct the ERA5-based annual rainfall erosivity and then reconstruct the annual rainfall erosivity dataset. The multiyear average correction coefficient over the TP between the station-based annual rainfall erosivity values and the newly released data is 0.67. In addition, the probability density and various quantile values of the new data are generally consistent with the station-based values. The data offer a view of large-scale spatiotemporal variability in the rainfall erosivity and address the growing need for information to predict rainfall-induced hazards over the TP. The dataset is available from the National Tibetan Plateau/Third Pole Environment Data Center (https://doi.org/10.11888/Terre.tpdc.271833; Chen, 2021).