基于不同蒸散方法的SPEI在天山南北坡气象干旱的差异性分析

为更好的研究与指导天山南北坡干旱灾害的预警与应对,采用Penman-Monteith、器测法和Thornthwaite方法确定潜在蒸散量(PET),并分别计算相应的标准化降水蒸散指数(SPEI),分别记为SPEI-PM、SPEI-SJ和SPEITH,探讨1960-2016年气象干旱时空变化特征与分析不同蒸散方法下干旱指数的差异性,并评估与历史干旱事件的相符性。结果表明:(1)从趋势变化分析可知,天山南北坡在SPEI-SJ和SPEI-PM指数下呈现湿润化和干旱面积率下降的特征,SPEI-TH指数则相反;(2)空间频率分析显示SPEI-SJ和SPEI-PM指数下研究区年和季节尺度上干旱发生频率一致,即春、夏干旱频发、秋季次之、冬季最少,而SPEI-TH指数下干旱发生频率在各时间变化特征相差不大,SPEI-SJ和SPEI-PM指数下干旱发生频率在空间上保持一致,局部区域存一定的偏差,总体与干旱强度变化趋势一致;(3)基于历次干旱事件可知,SPEI-PM指数对于1986年前的旱灾有较为准确的描述,而SPEI-SJ指数仅针对1990年后对干旱有较准确的描述,SPEI-TH指数对干旱的监测能力较差。综合分析表明SPEI-PM指数对历史干旱事件监测适用性较好,20世纪90年代后,SPEI-SJ指数对干旱的监测有重要的参考价值。

The differences of meteorological drought of SPEI on the north and south slopes of Tianshan Mountains with different evapotranspiration methods

The spatial and temporal variations of meteorological drought on the north and south slopes of Tianshan mountains in Xinjiang from 1960 to 2016 were investigated with three sets of standardized precipitation evapotranspiration index (SPEI),based on three different methods to estimate the potential evapotranspiration (PET).They were Penman Monteith method (SPEI-PM),instrument measurement (SPEI-SJ) and Thornthwaite method (SPEI-TH).The results showed:(1) Both the SPEI-SJ and SPEI-PM indexes showed a wetting trend and a decrease of drought area,but the SPEI-TH index showed an opposite trend;(2) The SPEI-SJ and SPEI-PM indexes had the same occurrence frequency of drought on the annual and seasonal scales,i.e.,drought occurred more frequently in the spring and summer seasons,followed by autumn,and the least in winter season.However,the drought frequency of SPEI-TH did not show significant differences in various time scales.The spatial patterns of drought frequency from the SPEI-SJ and SPEI-PM indices were consistent.There was a difference of drought frequency in local regions,which was the same as drought intensity variation;(3) SPEI-PM had a relatively higher accuracy before 1986,while SPEI-SJ only had a higher accuracy after 1990 in comparison with observed drought events.SPEI-TH was the worst index to reflect drought events.Overall,SPEI-PM was the most suitable index for monitoring historical drought events,and SPEI-SJ could add some values for monitoring drought after 1990s.