摘要
基于香港地区1947—2006年雨季(4—9月)、干季(10月—次年3月)、水文年(4月—次年3月)降水序列,利用非参数Mann-Kendall(MK)趋势检验方法对该地区降水趋势及其演变过程进行分析,并利用改进的非参数Mann-Kendall(MMK)趋势检验方法分析序列自相关对趋势显著性的影响.研究表明,各序列总体均呈上升趋势,无论考虑相关性与否,各序列趋势性在0.05显著性水平下均不是统计显著的.相关性对于趋势显著性有明显影响,尤其是对相关性相对强的干季降水序列.基于各序列趋势演变过程,及其与累积离差过程的对比分析,表明研究区降水趋势性不显著.
The precipitation trend and evolution in Hong Kong were investigated by performing Mann-Kendall(MK) tests on the precipitation series during the wet season(Apr.~Sept.),dry season(Oct.~Mar.) and water year(Apr.~Mar.) over the period of 1947 to 2006 separately.The modified nonparametric Mann-Kendall(MMK) tests were employed to analyze the effects of the serial autocorrelation on the significant level of the trend.The results indicate that the precipitation series generally have an increasing trend,and none of them is statistically significant at the significant level of 0.05 considering the serial autocorrelation or not.The effects of the serial autocorrelation on the significant level of the trend are obvious,especially for the dry season series with relatively strong autocorrelation.According to the trend evolution process and based on the cumulative deviation process of each series,it can be inferred that the precipitation trend in Hong Kong is not significant.
出处
《河海大学学报(自然科学版)》
CAS
CSCD
北大核心
2010年第5期505-510,共6页
Journal of Hohai University(Natural Sciences)
基金
香港特别行政区中央政策组
香港特别行政区研究资助局(HKU7022-PPR-2)