深度学习网络在降水相态判识和预报中的应用

利用1996-2015年中国的高空探测资料和地面观测数据,挑选发生降水的数十万个样本将其分为降雨和降雪两类事件,抽象为二分类问题,采用深度学习网络技术构建降水相态判识模型,并用2016 2017年的数据进行测试检验,针对2018年1月下旬中国一次大范围雨雪天气过程进行个例检验,在此基础上探讨了深度学习网络在降水相态判识...

Application of Deep Learning Method to Discrimination and Forecasting of Precipitation Type

This paper applies deep learning method on establishing a model to discriminate the precipitation type.Hundreds of thousands of precipitation samples obtained from sounding and observation data of China from 1996 to 2015 were divided into rain and snow events.The 2016-2017 data were tested,and a case test was conducted on a rain and snow weather process over China in late January 2018.Furthermore,the application of deep learning method to discrimination and forecasting of precipitation type was discussed.The main conclusions are as follows.Discrimination accuracy of the model is 98.2%,which is more improved than the traditional index threshold method.TS scores of rain and snow are 97.4%and 94.4%,false discriminate rates are 1.7%and 2.0%,and omission rates are 1.0%and 3.7%,respectively.The case test denotes that the model discrimination results based on the observation data are basically consistent with the observation data.The ECMWF precipitation type products and the model results also have a good forecast performance for precipitation type over China.Compared with ECMWF,the model forecast for rain and snow separating line is more consistent with observation.The test results show that the discrimination model established in this paper can extract the key features of precipitation type of rain and snow.The application of deep learning method to discrimination and forecasting of precipitation type is feasible and advantageous.Thus,this method could provide important technical support for the objective identification and prediction of precipitation type.