《Hydrological Processes》杂志刊登“气候变化对干旱程度-持续时间-频率关系影响的多模型评估”

作者：Joo Heon Lee, Chang Joo Kim

刊物：Hydrological Processes ，2013年，27卷19期，2800–2813页

关键词： 标准化降水指数；干旱频率分析 ；干旱时间；干旱程度-持续时间-频率曲线；气候变化

摘要：该研究利用位于朝鲜半岛的首尔气象站的观测数据对首尔地区过去和未来的干旱发生率进行了分析。分析使用了四种不同类型的大气环流模型（GCMS），即GFDL:CM2_1, CONS:ECHO-G, MRI:CGCM2_3_2 和UKMO:HADGEM1。利用标准化降水指数统计分析了干旱频率持续时间和重现期，绘制出干旱程度-持续时间-频率（SDF）曲线。此外，进行了干旱时间分析，以估计每种干旱分类下的频率和持续时间的变化。分析结果表明，未来中度干旱发生的概率将有所降低，但是重度干旱情况的发生概率将增加。此外，干旱平均持续时间也将增加。把利用观测数据得到的干旱程度-持续时间-频率（SDF）曲线关系与利用大气环流模型得到的关系作对比分析发现，每一重现期的干旱严重程度在降低，这是因为干旱持续时间在不断增加，而且在相同的持续时间和重现期内，由大气环流模型得出的干旱严重程度比由观测数据得出的严重程度要严重的多。此外，在本研究使用的四种类型的大气环流模型中，MRI模型对首尔地区未来可能发生的最严重的干旱进行了预测，并且使用MRI模型导出的SDF曲线干旱的严重强度也比其他大气环流模型严重的多。

A multimodel assessment of the climate change effect on the drought severity–duration–frequency relationship

Authors: Joo Heon Lee, Chang Joo Kim

Journal: Hydrological Processes, Volume 27, Issue 19, pages 2800–2813, 2013

Keywords: SPI; drought frequency analysis; drought spell; SDF curve; climate change

Abstract：In this study, the patterns of past and future drought occurrences in the Seoul region were analysed using observed historical data from the Seoul weather station located in the Korean Peninsula and four different types of general circulation models (GCMs), namely, GFDL:CM2_1, CONS:ECHO-G, MRI:CGCM2_3_2 and UKMO:HADGEM1. To analyse statistical properties such as drought frequency duration and return period, the Standardized Precipitation Index was used to derive the severity–duration–frequency (SDF) curve from the drought frequency analysis. In addition, a drought spell analysis was conducted to estimate the frequency and change of drought duration for each drought classification. The results of the analysis suggested a decrease in the frequency of mild droughts and an increase in the frequency of severe and extreme droughts in the future. Furthermore, the average duration of droughts is expected to increase. A comparison of the SDF relationship derived from the observed data with that derived via the GCMs indicated that the drought severity for each return period was reduced as drought duration increased and that the drought severity derived from the GCMs was severer than the severity obtained using the observed data for the same duration and return period. Furthermore, among the four types of GCMs used in this study, the MRI model predicted the most severe future drought for the Seoul region, and the SDF curve derived using the MRI model also resulted in the highest degree of drought severity compared with the other GCMs.

原文链接：http://onlinelibrary.wiley.com/doi/10.1002/hyp.9390/abstract