河套地区1951—2014年极端气温事件特征分析

在模糊聚类法划分子区域的基础上，采用河套地区73个国家基本气象站1951—2014年气象数据，将Kendall-tau非参数检验方法，Sens斜率估计方法以及Savitzky-Golay平滑滤波方法相结合，分析该地区18个气温指数的变化趋势。结果表明：（1）在河套地区的干旱半干旱子区（地处黄河上游），湿润半湿润子区（地处黄河中下游），极端气温事件的变化趋势基本相近，但强度不同，空间分布不一致。（2）在时间尺度上，发现河套地区近60年存在极端气温事件上升趋势，说明河套地区与全球变暖之间有紧密的响应关系。（3）在空间尺度上，各个指数在河套地区分布差异明显。多个指数在干旱子区的变化率大于湿润子区；干旱区各个指数变化显著的气象站点的占比高于湿润子区。有些指数在个别小区域与整个河套地区变化趋势不同，可能存在的原因有城市热岛、山丘地局部小气候，但有待进一步分析验证。（4）气温日较差、平均最高气温、平均最低气温、日最低气温的极低值、日最低气温的极高值、日最高气温的极低值、日最高气温的极高值指数变化趋势都在2000年左右开始变缓，这可能与全球气候变暖减缓有关。

Analysis on extreme temperature events in Hetao Region from 1951 to 2014

On the basis of sub-region dividing with fuzzy cluster method，the meteorological data from 1951~2014 in 73 national basic meteorological observing stations within Hetao Region are adopted for analyzing the variation trends of 18 temperature indexes in the region with the integration of the Kendall-tau nonparametric test method，the Sens slope estimation method and the Savitzky-Golay smoothing filtering method. The result shows that （1） the extreme temperature event variation trends of the arid and semi-arid sub-region （located at the Upper Yellow River），humid and sub-humid sub-region （located at the Middle and Lower Yellow River） are almost the same，but the intensity is different with inconsistent spatial distributions； （2） the upward trend of the extreme temperature events during the recent 60 years is found out on temporal scale，which is indicated that a close response relationship is there between Hetao Region and the global warming； （3） obvious distribution differences are there among all the indexes in Hetao region on spatial scale. The variation rates of several indexes in the arid sub-regions are larger than those in the humid sub-regions，while the proportions of all the meteorological observing stations with significantly changed indexes are higher than those in the humid sub-regions. The variation trends of several indexes in some individual small sub-regions are not same with those in the whole Hetao Region，for which it is possible that urban heat islands and local hill micro-climates are there，but the further analysis and demonstration are to be made； （4） the variation trends of all the diurnal temperature range，the mean maximum temperature，the mean minimum temperature，the extreme minimal value of daily minimum temperature，the extreme maximal value of daily minimum temperature，the extreme minimal value of daily maximum temperature，the extreme maximal value of daily maximum temperature start to be slowed around 2000，which is likely to be related to the global warming hiatus.