基于GPM的新疆降水聚集区域提取及特征分析

Extraction and features of precipitation clustering areas in Xinjiang based on GPM

气候暖湿化背景下,降水聚集特征与旱涝事件密切相关。基于2001-2020年逐日全球降水测量数据,根据空间热点分析方法提取新疆降水聚集区域,利用信息熵和相关系数分析了聚集区域降水集中度和地表特征。结果表明:新疆降水高值聚集区域占21.45%,降水量在130mm以上,主要分布在新疆西北部,低值聚集区域占27.97%,降水量在72mm以下,主要分布在塔里木盆地西南部、吐鲁番盆地和昆仑山中部山区;聚集区域降水量年内呈单峰变化,除显著冷点区域年降水量呈弱下降趋势外,其他区域年降水量呈增加趋势;聚集区域降水集中度呈现从西北向东南呈现出“低-高-低”的空间分布格局,且高值聚集区域降水集中度高于低值聚集区域,与降水量具有较高的相关性;低值聚集区域荒地面积占比最大,高值聚集区域草原和农田面积较大;高值聚集区域海拔、地表粗糙度和地面起伏度与降水量呈正相关,低值聚集区域呈负相关,且聚集性越显著,相关性越大。研究结果可以为防灾减灾、水资源的综合利用提供科学参考。

The clustering of precipitation is closely related to the risk of floods and droughts in arid area with the warming-wetting climate.Understanding of precipitation clustering in Xinjiang play a key role in disaster prevention and ecosystem services.Based on the daily Global Precipitation Measurement(GPM)data from 2001 to 2020,the precipitation clustering areas in Xinjiang is extracted according to the spatial hotspot analysis method,and the precipitation concentration degree(PCD)in the clustering areas are calculated by using information entropy.The surface characteristics of precipitation clustering areas in Xinjiang are systematically analyzed.The results show that the high value precipitation clustering areas in Xinjiang accounts for 21.45%of the total area of Xinjiang,and the precipitation is more than 130mm,which is mainly distributed in the northwest of Xinjiang,while the low value clustering areas accounts for 27.97%of the total area of Xinjiang,and the precipitation is less than 72mm,which is mainly distributed in the southwest of Tarim Basin,Turpan Basin and the central mountainous area of Kunlun Mountain.The precipitation in the clustering areas shows a single peak change in a year.Except that the annual precipitation in the significant cold spot area shows a weak downward trend,the annual precipitation in other areas shows an increasing trend.The PCD in the clustering areas is between 0.05 and 0.88,showing a spatial distribution pattern of"low-high-low"from northwest to southeast.The PCD in the high value clustering areas is higher than that in the low value clustering areas,which has a high correlation with the precipitation.The area of barren in the low value clustering areas accounts for the largest proportion,while the area of grasslands and croplands in the high value clustering areas is large.The altitude,surface roughness and topographic relief of the high value clustering areas are positively correlated with the precipitation,while that of the low value clustering areas is negatively correlated with the precipitation,and the more significant the aggregation is,the greater the correlation is.The research results can provide theoretical support for multi-purpose use of water resources,disaster prevention and reduction in Xinjiang.