摘要
基于1951—2013年江西省87个台站逐日降水数据和NCEP/NCAR再分析资料,利用偏相关统计方法选取影响江西省汛期降水的强迫因子,并进行归因分析,最终基于GCM模式预报结果建立江西省汛期降水的降尺度预测模型。结果表明:对江西省汛期降水产生主要影响的两个强迫因子分别是蒙古500 hPa位势高度和前期1—3月黑潮延伸区海表温度,前者反映中高纬的冷空气活动,后者能影响西太平洋副热带高压的位置和强度,两者都将影响江西省汛期降水。利用这两个强迫因子建立降尺度模型能够准确拟合江西省汛期降水,可较好地把握汛期降水的主要趋势,因而可用于江西省汛期降水预测。
Based on the daily precipitation data of 87 stations in Jiangxi province and the NCEP/NCAR reanalysis data during 1951-2013,forcing factors are selected for the precipitation of April-June over Jiangxi province using the partial correlationbased method,and then by attribution analysis,a statistical downscaling model is finally established for the precipitation forecast.The major forcing factors for the precipitation over the region include the 500-hPa geopotential height over Mongolia(M_(gp5)) and the sea surface temperature in the Kuroshio extension(K_(123)).The M_(gp5) may indicate the influence of the cold air from middle-high latitudes,and the K_(123) may reflect the intensity and position of the subtropical high in the western Pacific.The results simulated by the downscaling model based on M_(gp5) and K_(123) can well fit the variation of rainfall in flood season.The independent validation demonstrates the robustness of the downscaling model,which can be hopefully used in the seasonal prediction of the April-June rainfall over the region.
出处
《气象与减灾研究》
2017年第3期171-177,共7页
Meteorology and Disaster Reduction Research
基金
江西省气象局重点项目"动力气候模式产品在月季温度降水预测中的降尺度应用研究"
2016年江西省重点项目"延伸期(16-30天)逐日预报技术研究和业务建立"
江西省科学技术厅指导性计划项目(编号:2013ZBBG70022)
关键词
汛期降水
预测模型
降尺度
强迫因子
偏相关性分析
precipitation in flood season
forecast model
downscaling
forcing factors
partial correlation