Rainstorm Indicators During Meiyu Period Over Zhoushan Based on Random Forest and Logistic Regression
This paper analyzes the spatial and temporal distribution characteristics and weather situation characteristics of rainstorm during the Meiyu period(June to July)in Zhoushan from 2010 to 2020.Based on ERA5 reanalysis data and the random forest algorithm,the correlation between eight physical quantities including 700 hPa wind speed and direction,850 hPa wind speed and direction,850 hPa temperature and dew point temperature,500 hPa temperature,and strong weather threat index.Filter out the top six physical quantities to establish a prediction equation,through the method of logistic regression,the rainstorm prediction model of Zhoushan Meiyu period can be obtained,which can be used to judge whether rainstorm occurs,and can provide a reference basis for the probability of rainstorm occurring in Zhoushan Meiyu period.
Meiyu Period over Zhoushanrainstorm indicatorsrandom forestlogistic regression
万方数据
维普
