对流天气系统自动站雨量资料同化对降雨预报的影响

Variational Assimilation of Automatic Weather Stations Rainfall in Convective Systems and its Impact on Rain Forecast

利用GRAPES(Global and Regional Assimilation and Prediction Enhanced System,全球/区域同化预报系统)三维变分同化系统,针对对流天气系统特点,用改进的郭晓岚对流参数化方案作为观测算子,同化广东省自动站记录的对流天气系统的雨量资料,并且与同化探空资料进行了比较。在雨带有明显改进的区域,分别同化这两种资料都可以调整大气低层水汽辐合增加(或辐散),对流层中下层增暖增湿(或变冷变干),从而增加(或减少)降水,表明降水的同化方案对初始场的调整在一定程度上符合探空观测。进一步讨论同时同化这两种资料对暴雨预报的影响,结果表明同化自动站降水资料对暴雨系统短时预报有正面影响,同时同化这两种资料,可以弥补资料的不足。

The lack of data over the tropical regions contributes greatly to the uncertainties in the initial state of numeric weather prediction models, which in turn limits their forecast skill. Because in tropical regions the atmospheric motions are driven largely by diabatic processes, precipitation observations could be a valuable data source for improving initial fields. In Guangdong Province, there are almost 600 automatic weather stations （AWS）, providing hourly rainfall records. Focusing on the convective system, this study use the extension of the KUO cumulus parameter scheme as the observation operator to assimilate the AWS rainfall records in Guangdong Province on the GRAPES （Global and Regional Assimilation and Prediction Enhanced System） 3D variational assimilation system, and compares with the assimilation of sounding data In the case of 1 April 2004, hourly AWS rainfall records are compared with the GOES （Geostationary Operational Environmental Satellite） infrared channel temperature, TRMM （Tropical Rainfall Measuring Mission） rain rate data and flash location observations. Results show that the hourly AWS rain records can properly describe the convective system rain bands. Using WRF（Weather Research and Forecasting Model）as the forecast model, the control test shows that the initial rain bands are located in the northern part of the observation. Three assimilation experiments are designed to assimilate the AWS rain records sounding data and all the data respectively. Results show that in the regions where the initial rain bands are adjusted, assimilating the AWS rainfall and sounding data respectively both can adjust the low level atmosphere moisture convergence （or divergence）, low and middle troposphere temperature and moisture increasing （or decreasing） to enhance （or weaken） the initial rainfall The results mean that the adjustment by AWS rain assimilation scheme is consistent with the sounding data assimilation to some extent. This paper also discusses the influence of AWS rainfall assimilation on the short-range rain forecast. Results show that AWS rainfall assimilation scheme has the positive impact on the convective system short-range rain forecast. Assimilating AWS rainfall and sounding data at the same time can eliminate the deficiency of both the data, improve the rain location and structure forecast.