基于天气系统自动识别的新疆牧区雪灾遥感监测与预警研究

天气系统识别对天气预报业务具有重要意义，是研究气象灾害及其预警模型的理论前提。然而，目前的天气系统识别主要依靠气象站观测数据和专家经验，较难满足天气预报与灾害预警对时效性、客观性和准确性的要求。针对该问题，本项目结合3S技术与机器学习方法，实现基于卫星遥感的天气系统自动识别，并在此基础上发展高时空分辨率的牧区雪灾遥感监测与预警模型。内容包括：（1）利用风云四号卫星遥感资料、地面气象观测资料、数值预报以及CLDAS大气驱动场资料，实现天气系统遥感自动识别。（2）建立高时空分辨率雪灾遥感监测平台，动态监测雪灾相关致灾因子、承灾体以及孕灾环境。（3）结合新疆牧区自然地理与社会经济背景，根据天气系统、中期数值预报以及临近天气预报，建立分阶段雪灾遥感预警模型，从而提高牧区雪灾预警时效性。最终可为新疆牧区雪灾监测与预警工作提供理论支撑与应用示范，切实提高新疆牧区雪灾的防灾减灾能力，并向其他地区推广。

Weather system recognition is very important to weather forecasting service, and is the theory prerequisite to the study of meteorological disaster early-warning system. However, the existing methods of weather system recognition were mainly based on ground weather observations data and expertise, and it's difficult to meet the timeliness, objectivity and accuracy demand of weather forecasting and disaster warning systems. Aiming at this problem, in this project, we will develop a novel weather system automatic recognition method based on satellite remote sensing, by using the 3S technique and machine learning method, and on this basis, develop a remote sensing monitoring and early warning system of snow disaster in pastoral area. The main studies include: (1) Using the Fengyun-4 remote sensing, ground weather observations data, numerical weather forecast and the CLDAS forcing data to achieve weather system automatic recognition with high spatial and temporal resolution. (2) Develop a novel remote sensing monitoring system of snow disaster, which can been using for dynamic monitoring the disaster environment, disaster drivers and disaster bearers of snow disaster. (3)Develop a multistage early warning model for snow disaster, and improve its effectiveness and accuracy by fusion of the weather system recognition, medium-range numerical weather prediction, short-range weather forecasting and the geography and society-economy background of Xinjiang pastureland. For the ultimate goal, this project will provide theoretical support for snow disaster monitoring and early warning in Xinjiang pastureland, improve the level of disaster prevention and mitigation, and gradually extended to other regions.