日雨量随机解集模式研究

全球气候模式(GCMs)预测的气候变化情景，必须经解集模式得出小尺度上未来气候变化时空分布资料，才能满足评估气候变化对资源、环境和社会经济等影响的需要。本文提出由随机天气生成器和统计参数尺度转换关系组成的随机解集模式，应用17个站32年实测日降雨资料，对随机解集模式进行了分析和验证。首先利用随机天气生成器，通过对站点和GCM尺度面平均降雨系列的模拟，确定模型参数，验证模型模拟历史降雨过程的可靠性。然后，建立模型参数从大尺度向站点转换的关系，并从历史降雨系列中抽出某一日雨量系列，假设为未来气候变化情形，对降雨系列在不同尺度间的转换关系进行了验证。在此基础上，对GCMs预测结果的时空解集方法进行了探讨。

Downscaling of daily precipitation using a stochastic weather generator

A statistical downscaling approach,including a stochastic weather generator and a downscaling technique,is developed to transfer daily rainfall in general circulation model (GCM) grid to local area.To analyze its reasonability,an area with grid of 2.5°in latitude and 3.75° in longitude in Guangdong Province is selected and daily precipitation series of 32 years in 7 stations are used.The statistical characteristics of daily precipitation series generated by the weather generator are compared with the observation data recorded both in local and grid scale.Based on the assumption of the future changed climate precipitation series from one set of history records, verification of the downscaling technique is also made by comparing the statistical characteristics as well as monthly and amount from downscaling way with those from assumed precipitation series.