长江中下游地区位于东亚季风区，其夏季降水的水汽部分来源于孟加拉湾的水汽输送。本文利用青藏高原地区全球定位系统（GPS）站点观测到的大气可降水量（PW）资料，采用WRF模式（Weather Research and Forcasting Model）的同化模块（WRFDA），将这支水汽输送带的信息同化进数值模式，并用WRF模式对长江中下游地区的7月份降水预报进行批量试验和个例分析。批量试验和个例分析采用3种方案：无资料同化的控制试验（NoDA），冷启动同化试验（Cold）和循环同化试验（Cycling）。此外，还针对Cycling方案进行延长预报时长的补充试验以探究同化带来正效果最明显的时段。同时为了探究同化正效果的来源，针对Cycling方案进行只同化被主要水汽输送带覆盖的GPS站点的补充试验（Cycling_less_a）以及只同化不被主要水汽输送带覆盖的GPS站点的补充试验（Cycling_less_b）。试验结果表明：同化青藏高原地区的GPS数据能在一定程度上改善长江中下游地区的降水预报，对于48～72小时的降水预报改善效果尤为明显，且Cycling方案在整体上优于Cold方案。对于Cycling方案，在120小时预报时长内，同化正效果最明显时段为48～72小时。当水汽输送带较多地经过同化区域时，降水的TS评分能得到明显改善，而当水汽输送带较少地经过同化区域时，降水的TS评分改善效果不明显。如果只同化被水汽输送带覆盖到的GPS站点的GPSPW数据，仍然可以保留住大部分的同化正效果，因此，针对性地同化GPSPW数据是可行的。
The middle and lower reaches of the Yangtze River are located in the East Asian monsoon region where part of its summer precipitation water vapor comes from the moisture transport of the Bay of Bengal. In this paper, the atmospheric precipitable water (PW) data collected from the global positioning system (GPS) sites in the Tibetan Plateau and the assimilation module (WRFDA) of WRF (Weather Research and Forecasting Model) are used to assimilate the water vapor transport information into the numerical model. At the same time, WRF is also applied to do batch testing and case analysis of the precipitation forecast in the middle and lower reaches of the Yangtze River region in July. Three schemes are adopted during batch experiments and analysis of individual cases: a control experiment with no data assimilation (NoDA), a cold-start assimilation experiment (Cold) and a cycling assimilation experiment (Cycling). In addition, the experiment of extending the forecast time of the cycling scheme is performed to find out the most obvious time period of active effect. For the purpose of investigating the source of active effect, additional experiments of only assimilating the GPS sites that are mainly covered with the water vapor conveyor belt (Cycling_less_a) and that are not mainly covered the with water vapor conveyor belt (Cycling_less_b) are carried out according to the cycling scheme. The results show that the assimilation of the GPSPW data of the Tibetan Plateau area can improve the forecast of precipitation in the middle and lower reaches of the Yangtze River to a certain extent, especially for the 48-72-h period precipitation forecast. The cycling scheme outperforms the cold scheme on the whole. For the cycling scheme, during the forecast time of 120 h, the active effect is most obvious in the 48-72-h period. When the conveyor belt of water vapor goes through the assimilation area, the TS score of precipitation can be obviously improved, while the improvement will not be so obvious when the water vapor does not go through the assimilation area. If we only assimilate the GPSPW data of the GPS sites covered with the water vapor transportation belt, we can still retain most of the positive effect. Therefore, targeted assimilation of the GPSPW data is feasible.
朱丰,徐国强,李莉,郑晓辉,张胜军.同化青藏高原地区GPSPW数据对长江中下游地区降水预报的影响评估.大气科学,2014,38(1):171~189 ZHU Feng, XU Guoqiang, LI Li, ZHeng Xiaohui, Zhang Shengjun. An Assessment of the Impact on Precipitation Prediction in the Middle and Lower Reaches of the Yangtze River Made by Assimilating GPSPW Data in the Tibetan Plateau. Chinese Journal of Atmospheric Sciences (in Chinese),2014,38(1):171~189复制