Spatial-temporal seamless identification model of dust storm weather based on Himawari 8 geostationary meteorological satellite
The identification and monitoring of dust storm weather are the key technical issues for disaster warning and management.This study fully explores the potential of the data from the Himawari 8 geostationary meteorological satellite and constructs a spatiotemporal seamless identification model for dust storm weather.The model simultaneously utilizes four thermal infrared bands,namely 8.6um,10.4um,12.3um,and 7.3um,which are the most sensitive to sand and dust.It is not affected by the alternation of day and night,and achieves seamless monitoring of time and space every 10 minutes.Using 274 dust storm weather monitoring data from fixed stations of the National Meteorological Administration,the accuracy of the model was verified to be 90.14%.A model was used to extract the hourly distribution of dust storm weather from March to May 2018.A total of 6.11 million square kilometers nationwide were affected by dust storm weather,with an average of 0.64 days affected.From the perspective of time distribution characteristics,the highest range and number of days affected by dust storm weather are in April,followed by March and May.From the perspective of spatial distribution,the total land area affected by dust storm weather exceeding 10 days is 769500 square kilometers,with Xinjiang,Gansu,and Inner Mongolia accounting for 92.92%.These regions are the main areas in China where the desertification lands are distributed widely.The model was used to monitor the origin and process of the March 22,2023 strong dust storm.Results showed that the dust storm mainly originated in Mongolia and central Inner Mongolia,it cumulatively impacted the areas approximate 2.233 million square kilometers including 15 provinces(districts,cities).The model has been applied to the monitoring of dust storm weather from March to May 2018 to 2022,playing an important role in disaster assessment and emergency response.
万方数据
维普
