Study On Statistical Analysis And Forecasting Method Of Precipitation In Yantai Area

Predictions of the magnitude, region and tendency of precipitation have always been the important and difficult aspects in the precipitation forecast. According to the analysis and research of1980-2010precipitation observation data in Yantai, the amount of precipitation has been increasing as year by year while the number of days of precipitation was decreasing. Based on the study of the temporal-spatial and seasonal variation of precipitation, a prediction of the trend of precipitation in the future years has been made, and has a significant guideline in future precipitation forecast.Moreover, this paper provides the comparison and analysis on statistics of weather conditions of38precipitation cases occurred from March2012to November2012in Yantai, based on T639L60numerical forecast products and the actual precipitation observation data. Generally circulation patterns affecting precipitation in Yantai could fall into seven categories:one-trough-one-ridge, two-trough-one-ridge, spring-to-summer-transition, summer-to-autumn-transition, latitudinal, longitudinal, and mobile, the latter two patterns being most likely ones to produce torrential rain. And it is concluded that main weather systems affecting precipitation in Yantai include cyclones, shear lines, cold fronts, subtropical highs, typhoons, cold vortexes, and transverse troughs.Based on precipitation formation mechanism, the relationship between T639numerical products—various kinds of physical quantity, such as relative humidity, vapor flux divergence, vorticity, vertical velocity, and precipitation was analyzed and studied in this paper. The study reveals:for precipitation to happen in Yantai, air humidity must reach70%or above; when precipitation occurs in Yantai, vapor flux divergence value at700hpa is often minus and it is equal to or less than-4×10-8g·cm-2hpa-1·s-1; vorticity at500hpa is equal to or greater than4×10-5·s-1; vertical velocity is most highly correlated with precipitation and whether its value is less than-12×10-2pa·s-1, it indicates the likelihood of rain; both a plus velocity center and a minus velocity center are appear in the vertical velocity field during precipitation, and the region with the highest values may have the highest amount of precipitation.By the application of correlation analysis, a number of factors highly correlative to precipitation were selected, a multiple regression equation was obtained based on the Model Output Statistics (MOS) and a precipitation forecast model was established in combination with specific circulation features with its applicability verified, which is over75%accuracy.