摘要
全球气候持续变化下。极端水文事件频发且时空分布规律显著变异,极端水文事件阈值及其时空分布不确定性成为当前水文水资源研究领域的热点问题之一。以珠江流域62个雨量站1959～2009年逐日降水序列为样本,运用固定临界值法、百分位值法和极值分布拟合法等方法,探讨了珠江流域极端降水阈值的时空分布规律和不确定性特征。研究表明,在流域尺度上,利用上述方法分析极端降水阈值时,均存在较多不确定性。(1)固定临界值法无法反映流域降水空间分布不均对极端降水阈值的影响;(2)百分位值法受样本容量、百分位值选取标准等影响较大,流域不同地区同一百分位或同一地区不同百分位下极端降水阈值波动幅度较大,易导致降水高值区阈值偏低,降水低值区阂值偏高;(3)极值分布拟合法受极值分布拟合函数类型、样本容量大小等影响较大,拟合精度不高。
Extreme hydrological events occur with increasing frequency and the spatial and temporal distributions change significantly under the influence of ：onstant change of the global climate. One of the hot issues in the field of hydrology and water resources is the uncertainty of threshold and its spatial and temporal distribution of extreme hydrological events. Based on daily precipitation data of the 62 rainfall stations in the Pearl River Basin from 1959 to 2009, a study on the spatial and temporal distribution and uncertainty of the threshold of extreme precipitation was made systematically based on fixed critical value method, special percentile value method and goodness of extreme value distribution fitting test. This study shows that studying the precise meaning of the threshold of extreme precipitation at the basin-scale by the mentioned methods leads to some uncertain problems. The effects of the unbalance of spatial and temporal distribution on the threshold of extreme precipitation can＇t be reflected by fixed critical value method. Using special percentile value method, the thresholds of extreme precipitation of the same percentile value under different regions or of different percentile values in the same area are significantly diverse, because of the shortage of the standard. The accuracy of the threshold of extreme precipitation calculated by the goodness of extreme value distribution fitting test is limited, for the uncertainty of extreme value distribution and the error of data of extreme value.
出处
《水文》
CSCD
北大核心
2013年第4期59-64,共6页
Journal of China Hydrology
基金
国家自然科学基金项目(50909106)
教育部青年教师培育项目(3161399)
广东省科技计划项目(201180308 00008)
关键词
极端降水
阈值
不确定性
珠江流域
extreme precipitation
threshold
uncertainty
Pearl River Basin
