基于月动力延伸预报的陕西降水降尺度预测模型  

Downscaling precipitation prediction in Shaanxi Province based on monthly dynamic extended range forecast

作　　者：王娜[1] 方建刚[1] 崔巍[1] 肖科丽[2] 王琦[1] 

机构地区：[1]陕西省气候中心,陕西西安710014 [2]陕西省气象学会,陕西西安710016

出　　处：《西北大学学报（自然科学版）》2013年第5期809-814,共6页Journal of Northwest University（Natural Science Edition）

基　　金：国家重点基础研究发展规划基金资助项目(2010CB833406)

摘　　要：建立陕西站点月降水量的统计降尺度预测模型。利用国家气候中心月动力延伸预报结果、NCEP/NCAR再分析资料和陕西40个站观测资料,通过陕西各月降水量与大气环流相关要素的统计相关分析,采用最优子集回归方法,建立了基于月动力延伸预报的陕西降水降尺度预测方法。预测方法对陕西月降水的预测效果夏季高于冬季,夏秋季预测效果相对稳定,春季预测效果不稳定;区域预测质量陕南效果最好,关中次之,陕北相对较差,陕北、陕南预测效果季节变化比较明显,关中相对稳定;基于月动力延伸预报的陕西降水降尺度预测方法对极端气候事件的异常趋势与实况基本一致,但在异常程度和范围上仍与实况有较大偏差;同时发现基于月动力延伸预报的陕西降水降尺度预测方法对预测初始场有明显的敏感性。预测方法有助于提高陕西气候异常的预报水平和极端事件的预报能力。To establish downscaling monthly precipitation prediction model in Shaanxi Province. By using monthly dynamic extended range forecast （DERF） products, NCEP/NCAR reanalysis data and 40 station data in Shaanxi, downscaling monthly precipitation are predicted using the optimal subset regression from correlation analysis be- tween monthly precipitation and atmospheric circulation elements. It has high prediction quality in summer than in winter and more stable prediction ability in summer and autumn than in spring. The prediction ability is higher in South Shaanxi and Guanzhong, lower in North Shaanxi; meanwhile it has obvious seasonal variation in North Shaanxi and South Shaanxi and more stable prediction ability in Guanzhong. Based on the forecast method, unusual trends of extreme events are consistent with the fact, but there is discrepancy in abnormal degree and range between method result and the fact. Meanwhile the method has apparent sensitivity of the forecast initial field. The method has improved predictive capability of unusual climate and extreme events in Shaanxi Province.

关 键 词：月动力延伸预报 降水 降尺度预测 陕西省 

分 类 号：P456.3[天文地球—大气科学及气象学]