A dataset of spring sandstorm distribution on the Mongolian Plateau (2000-2021)

Abstract

The Mongolian Plateau is one of the cradles of sandstorms in Asia, among which the Gobi region of Mongolia and the desert region of western Inner Mongolia are to blame for the frequent sandstorms in China and Mongolia in recent years. First, based on meteorological observation data about sandstorms and text data of news events, we collected 76 sandstorm events on the Mongolian Plateau from 2000 to 2021 in the study. Second, we downloaded MODIS data of corresponding spatiotemporal regions, which were screened and processed in combination with the cloud size. And we preprocessed the MODIS data for geometric correction with MRTK plugin in the ENVI software. Then, we constructed the DSDI of dust index and extracted the spatial distribution information of dust to obtain the of spring sandstorm distribution of Mongolian Plateau. With the ArcGIS software for random selection of verification points, we used the station record, text data, image enhancement, visual interpretation to complete the accuracy evaluation. The overall classification accuracy reaches 85.24% with a Kappa coefficient of 0.7636. This dataset can directly reflect the spatial distribution of sandstorms on the Mongolian Plateau, which can provide detailed and reliable data support for the control of sandstorm disasters. At the same time, it can also back up decision-making process of controlling the risk of sandstorm disasters on the Mongolian Plateau.