基于信息扩散理论的闽清大暴雨雨量风险预测

Rainfall Risk Prediction of Heavy Rainstorm Based on Information Diffusion Theory in Minqing

为加强闽清县大暴雨雨量风险预测能力,减小常规统计方法对雨量预测结果的误差,利用1971-2010年共18个大暴雨过程的雨量数据,基于信息扩散技术分别对台风和非台风(包括过程性强降水和强对流强降水)引起的大暴雨进行雨量风险预测。结果表明:24 h雨量100～120 mm为高风险水平,风险概率在34%～100%之间,其次为120～150 mm,风险概率为9.7%～34%,而超过175 mm雨量的风险概率不足1%。同时结合天气影响系统统计分析闽清县大暴雨的成因,主要是台风型大暴雨和非台风型大暴雨两种,实际应用中对应不同类型的大暴雨天气用不同的风险概率进行雨量风险预测。

In order to strengthen the rainfall risk prediction capacity of heavy rainstorm and reduce the rainfall prediction error of the conventional statistical methods,information diffusion theory was used for risk prediction on rainfall of heavy rainstorm caused by the typhoon and non-typhoon（including processes and local heavy rainstorm）,based on the rainfall data of 18 heavy rainstorm process from 1971 to 2010 in Minqing County.Results showed that the rainfall risk probability of more than 175 mm in 24 hours was less than 1%,the risk probability that high-risk rainfall level of 100～120 mm was from 34% to 100%,and followed by the risk probability of 120～150 mm of 9.7%～34%.In combination with the weather system heavy rainfall of statistical analysis in Minqing County was mainly caused by typhoon type and non-typhoon factor,corresponding to different types in different risk heavy weather risk forecast probability of rainfall.