基于多源数据的干旱监测指数对比研究——以西南地区为例

干旱作为一种时常发生的自然灾害,影响范围广,对农业和粮食安全、人类生活等有深远影响。目前常用的干旱监测指数,都有各自的优缺点,无法适用于所有类型的干旱。论文利用基于气象要素驱动数据集的SPI（Standardized Precipitation Index）、基于MODIS（Moderate Resolution Imaging Spectroradiometer）的ESI（Evaporative Stress Index）、ETI（Evapotranspiration Index）和基于GRACE（Gravity Recovery and Climate Experiment）观测数据的水储量变化TWSC（Terrestrial Water Storage Changes）,对西南地区2005—2014年间的干旱情况进行分析,对比了几种不同数据源下的干旱监测指标的监测效果。结果表明：1）4种干旱监测指标对西南地区的干旱都较为敏感,其中6个月尺度的SPI（即SPI-6）与3个月尺度的ESI（即ESI-3）相关性相对最强（R²=0.431, P<0.01）;2）基于GRACE的水储量变化受全局性大干旱的影响较大,且秋冬比夏天的影响大;3）SPI-6、ESI-3、ETI-3能够较为准确地监测出干旱的空间分布及干旱过程中重心的移动,ETI-3在2009—2010年的干旱中有明显滞后,SPI-6则在干旱末期夸大干旱严重程度。

Evaluation of Drought Monitoring Indices Based onMulti-source Data in Southwest China

Drought is one of the costliest natural hazards and its impacts on economic sectors and people are significant and widespread. However, drought is hard to be monitored appropriately since they are caused by the combination of anomalies in precipitation, temperature and the overall status of surface water and ground water supplies in a region. In this study, remote sensing based data, including MODIS ET, GRACE datasets, and climate datasets are used to calculate multiple drought indicators including SPI (Standardized Precipitation Index), ESI (Evaporative Stress Index), ETI (Evapotranspiration Index) and TWSC (Terrestrial Water Storage Changes). Then these drought indicators were used to track the drought events in Southwest China during 2005-2014. The results showed that all of the indicators can capture the droughts which occurred in the past decade, and the correlation coefficient between the computed SPI and ESI is the highest which is greater than 0.43 (P<0.01), compared with the correlation coefficients between other indicators. In addition, TWSC is more sensitive to widespread severe droughts and is more variable during fall and winter. However, ESI can capture the spatio-tempral distribution of the droughts in 2006, 2009 to 2010 and 2011 more accurately than SPI, ETI and TWSC, even without any input of rainfall data. Here, ESI responds to variability of soil moisture and vegetation water content which are introduced by the variations of precipitation and radiation load and is a useful complement for drought monitoring in regions where rainfall data are spare or unreliable.