武汉市10个主要极端天气气候指数变化趋势分析

Trends of Ten Main Extreme Weather Indices in Wuhan

根据武汉市1951—2007年逐日气温、降水量计算分析了10个极端天气气候指数的变化特征。结果表明:1)4个气温指数中,年及四季高、低温阈值均为上升趋势,并造成最长热浪天数的延长和霜冻日数的减少;低温阈值升速明显快于高温阈值,高温阈值仅在春季变化显著,最长热浪天数仅在冬季变化显著;低温阈值则为极显著上升趋势,尤其是年和冬季,造成"热春"、"暖冬"频繁;暖夜、闷热、傍晚至夜间的强对流等显著增多,暖日、高温热浪增加,霜冻日大幅减少。2)6个极端降水指数以增趋势为主,其中强降水阈值、比例、日数以及最大5日降水量在冬季增趋势最明显,仅夏季强降水阈值、比例略有减小,冬季日降水强度的增大趋势、夏季持续干期的缩短趋势显著性水平分别可达0.1、0.01。3)一些气温指数在1980—1990年代发生突变,而降水指数未现突变。

Based on the daily temperature and precipitation data from 1951 to 2007 in Wuhan, 10 main series of extreme indices on weather and climate were calculated and the change of extreme weather and climate were revealed through trends analysis, abrupt analysis and decadal comparisons. It is found that： 1） among 4 extreme indices concerning temperature, the annual and four seasonal 90th percentiles of maximum temperatures （txq90） and 10th percentiles of minimum temperatures （tnq10） all showed rising trends that led to the extension of the maximum length of heat wave duration （txhw90） and the reduction of the day number of frosts （tnfd）, respectively, but the tnq10 rose quicker than the txq90, and the trends of tnfd and tnq 10 were all extremely significant especially for winter and year, and the trends oftxq90 and txhw90 were significant only in spring and winter, respectively. Those changes led to more frequent ＂ hot spring ＂, ＂ warm winter ＂, ＂ warm night ＂, ＂ sultry weather ＂, and ＂ strong convective phenomenon ＂ in the evening-night and also more ＂warm day ＂, ＂heat wave ＂, but less ＂ frost day ＂. 2） trends of 6 major extreme indices concerning precipitation were mostly rising, among them the 90th percentile of rainday rainfall （pq90） and its percentage （pfl90） and days （pnl90） and also the maximum 5-day total rainfall （px5d） were rising most significantly in winter, but the pq90 and pfl90 were declining slightly in summer. The rising trend of the simple daily intensity （pint） in winter and the reducing trend of the maximum length of consecutive dry days （pxccd） were significant at the 0.1 and 0.01 significance level, respectively. 3） some abrupt points were detected from the late 1980s to early 1990s for some temperature indices.