SPEI指数在中国东北地区干旱研究中的适用性分析

干旱指数的区域适用性是准确表征区域干旱的重要前提,本文以中国东北地区为典型研究区,探讨标准化降水蒸散指数(SPEI)在该地区应用的有效性。基于研究区90个气象台站的逐日气象资料,计算1961—2014年多时间尺度的SPEI指数。从Kolmogorov-Smirnov(K-S)拟合优度检验、SPEI与典型干旱事件核准、SPEI与农作物受旱灾面积及与土壤湿度相关性分析等方面,验证SPEI指数在东北地区的适用性。分析结果表明:1)东北地区多时间尺度的累积水分亏缺量符合Log-logistic分布,SPEI指数在东北地区的应用具备数学统计理论基础;2)生长季平均SPEI值与黑龙江省、吉林省和辽宁省农作物受旱灾面积比例均呈极显著负相关(P0.01);3)在1、3、6和12个月尺度下,SPEI与土壤湿度呈显著正相关(P0.05)的站点比例分别为90.2%、92.16%、90.2%和88.24%。综上所述,SPEI指数不仅满足数学理论统计的要求,而且与干旱灾情数据和土壤水分监测值均具有极度的关联性,说明其在东北地区干旱预测和定量化研究中具有较好的适用性。

Applicability analysis of SPEI for drought research in Northeast China

Determining the applicability of a drought index to a specific area is a prerequisite for drought quantification and mapping on a regional scale. In this paper, we explored the effectiveness of the Standardized Precipitation Evapotranspiration Index (SPEI) in northeast China. SPEI values at a time scale of 1, 3, 6, and 12 months were calculated using a daily climate dataset that had been collected by 90 meteorological stations in northeast China from 1961 to 2014. The climate variables analyzed included daily temperature, precipitation, atmospheric pressure, wind speed, relative humidity, and sunshine duration. The applicability of the SPEI for drought quantification in northeast China was investigated using Kolmogorov-Smirnov (K-S) test, validation analysis from specific drought events and soil moisture, and correlation analyses between SPEI and the statistics of the area suffering from drought.The results showed that our null hypothesis was rejected by the K-S Test at a 0.05 significance level for each station, and that the Log-logistic distribution matched very well to the water surplus or deficit series for all four time scales at each station. The drought-affected areas showed a significant negative relationship to the SPEI during the growing season (April to September) at these four time scales during 1971-2013. The index of SPEI01 was found to be more suitable for predicting the time, location, and intensity of drought events. There was a strong positive correlation between soil moisture at a depth of 20 cm and the SPEI for most of the climate stations (90.2% of stations for SPEI01, 92.16% of stations for SPEI03, 90.2% of stations for SPEI06 and 88.24% of stations for SPEI12). In addition, the number of climate stations decreased as the time scale increased. Our results indicate that the SPEI is suitable for quantifying droughts, and therefore, should be widely used in scientific drought prediction research in Northeast China.