高级检索
张冬峰, 高学杰, 马洁华. CCSM4.0模式及其驱动下RegCM4.4模式对中国夏季气候的回报分析[J]. 气候与环境研究, 2015, 20(3): 307-318. DOI: 10.3878/j.issn.1006-9585.2015.14158
引用本文: 张冬峰, 高学杰, 马洁华. CCSM4.0模式及其驱动下RegCM4.4模式对中国夏季气候的回报分析[J]. 气候与环境研究, 2015, 20(3): 307-318. DOI: 10.3878/j.issn.1006-9585.2015.14158
ZHANG Dongfeng, GAO Xuejie, MA Jiehua. Analysis of Summer Climate over China from Hindcast Experiments by CCSM4.0 and RegCM4.4 Models[J]. Climatic and Environmental Research, 2015, 20(3): 307-318. DOI: 10.3878/j.issn.1006-9585.2015.14158
Citation: ZHANG Dongfeng, GAO Xuejie, MA Jiehua. Analysis of Summer Climate over China from Hindcast Experiments by CCSM4.0 and RegCM4.4 Models[J]. Climatic and Environmental Research, 2015, 20(3): 307-318. DOI: 10.3878/j.issn.1006-9585.2015.14158

CCSM4.0模式及其驱动下RegCM4.4模式对中国夏季气候的回报分析

Analysis of Summer Climate over China from Hindcast Experiments by CCSM4.0 and RegCM4.4 Models

  • 摘要: 使用区域气候模式RegCM4.4(Regional Climate Model version 4.4)单向嵌套CCSM4.0(Community Climate System Model version 4.0)气候系统模式输出结果, 进行了2001~2010年逐年2月1日至9月1日共10年长度的季节尺度气候预测回报试验, 针对平均气温和降水, 分析了两个模式对中国地区夏季(6~8月)气候的回报能力。首先对气候态的分析表明, RegCM4.4对气温和降水的回报/模拟效果均较CCSM4.0有所改进, 特别是在提供更详细可靠的局地信息方面, 其中降水回报与观测的空间相关系数, 由CCSM4.0的0.39提高到RegCM4.4的0.53, 但同时RegCM4.4对中国东部季风降水的回报表现出类似CCSM4.0北方偏多的偏差。对两个模式2001~2010年逐年气温和降水距平的回报能力, 通过回报与观测空间和时间距平相关系数(ACCs和ACCt)、回报与观测空间和时间距平符号一致率(PCs和PCt)以及趋势异常综合评分(PS)进行了考察, 结果表明两个模式的表现在整体分布上有一定相似的同时, RegCM4.4能够提供更多的空间分布细节, 并对降水的回报结果有一定的改善, 如CCSM4.0和RegCM4.4回报降水的ACCs多年平均分别为0.03和0.10, PS分别为70.4和71.4。同时给出了两个具体年份(2003年和2009年)的个例分析。

     

    Abstract: Hindcast experiments for the 10-year period from 2001 to 2010 are completed using the regional climate model RegCM4.4 (Regional Climate Model version 4.4), driven by the global model CCSM4.0 (Community Climate System Model version 4.0). All experiments are conducted from 1 Feb to 1 Sep. The hindcast simulations from CCSM4.0 and RegCM4.4 are intercompared and evaluated against observations in the summer season (June-July-August, JJA). RegCM4.4 captures the observed temperature and precipitation climatology better than CCSM4.0 and provides more detailed and reliable local information. The spatial correlation coefficient of precipitation improves from 0.39 in CCSM4.0 to 0.53 in RegCM4.4, while similar bias in reproducing monsoonal precipitation, characterized by the overestimation in northern China, is apparent. The anomalies of temperature and precipitation are in general consistent between the two models, as measured by the spatial anomaly correlation coefficient (ACCs) and temporal anomaly correlation coefficient (ACCt), the prediction consistency of the anomaly sign (PCs and PCt), and the prediction score (PS) between hindcasts and observations. Meanwhile, more detailed spatial information and a slightly better performance in precipitation are presented by RegCM4.4; the multi-annual ACCs values for precipitation from CCSM4.0 and RegCM4.4 are 0.03 and 0.10, and the corresponding PS values are 70.4 and 71.4, respectively. The anomalies of precipitation in two cases (2003 and 2009) are also analyzed and discussed.

     

/

返回文章
返回