广州增城台风降水相关天气型

Synoptic weather typing and typhoon with an application to Zengcheng,Guangzhou

本文利用自动天气分型方法将1990—2012年广州增城每日天气分型并确定与台风降水事件相关的天气型.自动天气分型分为三个步骤,主成分分析,聚类分析以及判别分析,天气分型结果联系台风日降水资料可找出与台风降水相关的天气型.广州市增城区1990—2012年地面小时资料(包括气温,海平面气压,露点温度,南北向风速,东西向风速,总云量六个气象要素)经自动天气分型方法得26种主要天气型,联系台风日降水资料可确定其中四类天气型与台风降水强相关,两类天气型与台风降水部分相关.台风相关降水天气型(包括四类台风降水强相关天气型和两类台风降水部分相关天气型)包含23个台风强降水事件(日降水量≥50mm)中的22个,包含全部7个日降水量≥100mm的台风强降水事件.结果表明利用自动天气分型方法能较好地判定与台风相关的强降水天气事件.该方法可利用未来降尺度大气环流模式(GCM)气候资料估计未来气候变化情境下与台风相关强降水事件变化.

In this study,an automated synoptic weather typing was employed to assign each day of the dataset to a distinctive weather type and then identify the weather types most likely associated with daily typhoon-related rainfall events for Zengcheng,Guangzhou during the period January1990-December 2012.There are three steps for automated synoptic typing：principal components analysis,an average linkage clustering procedure,and discriminant function analysis.The typhoon-related rainfall weather types can be identified using the daily weather types associated with the typhoon track data.26 major weather types were ascertained using hourly surface data and six ty-phoon-related rainfall weather types were identified based on the typhoon best track dataset for Zengcheng, Guangzhou.These six typhoon-related rainfall weather types can capture 22 out of 23 total typhoon-related heavy rainfall events（≥50 mm＆#183;d-1 ）and all seven cases with typhoon-related daily rainfall ≥100 mm during the period January1990-December 2012.This results suggest that automated weather typing can be useful to identify the typhoon-related heavy rainfall events.Moreover,the method has the potential to proj ect future typhoon-related heavy rainfall events using future downscaled general circulation model（GCM）climate data.