ATMS卫星资料的同化应用及与AMSUA/MHS的比较研究

STUDY ON THE ASSIMILATION OF ATMS SATELLITE DATA AND COMPARISON WITH AMSUA/MHS

针对数值预报中ATMS卫星资料的同化应用问题,在WRFDA框架下扩展了对ATMS卫星资料的同化功能,选择2012年7月第9号强台风"苏拉"开展初步研究,比较了NOAA18卫星上搭载微波传感器AMSUA/MHS的结果,分析卫星资料特征并试验资料同化应用对区域数值预报的影响。结果表明,ATMS探测质量优于或与AMSUA/MHS相当;同时由于ATMS具备的高空间覆盖率、增多的温度探测扫描点和湿度探测通道,可为资料同化系统提供更丰富的观测信息,有效改善数值预报效果。ATMS卫星资料的降噪处理是资料同化应用的一个初始环节,对于个例的研究结果表明,总体上可以降低噪声和提高资料的使用效果,但在湿度探测段低层通道存在经偏差订正后降噪处理结果误差有所增大的现象,说明针对处理方式还需要调整资料应用的多个环节。偏差订正前,卫星微波资料同化目前普遍使用的ATMS温度探测9通道和湿度探测22通道的卫星观测和模拟间存在较大偏差,是ATMS资料同化应用中需要注意的。此外,ATMS卫星探测与AMSUA/MHS探测通道设置较大的差别之一在于窗区通道,由此主要依据窗区通道探测的云检测方案在ATMS资料同化应用中需要加以调整和试验。

To use the ATMS satellite observation in numerical forecast,the ATMS data assimilation module is extended in the framework of WRFDA.The preliminary application of ATMS data,along with the comparison with the result of microwave sensor AMSUA/MHS onboard NOAA18 satellite,is conducted on the No.9 typhoon case “SAOLA” in July,2012.The satellite data characteristics are analyzed and the effect of data assimilation on the regional numerical forecast is investigated.The results show that the ATMS data is generally of good quality,which is better than or comparable to that of NOAA 18 AMSUA/MHS.At the same time,the use of ATMS data could enrich the observation information for the data assimilation system and improve the numerical weather forecast because the ATMS observation has higher data coverage,more scan points for temperature remote sensing and more channels for humility observation.The process of reducing the noise is an initial step that is of relevance to the assimilation of ATMS data.Generally it decreases the noise and improves the performance of satellite data in our case study.It is found that the biases of reducing the noise in several low-humility channels become large after the bias correction.It is suggested that the associated factors need to be adjusted in data usage that are related with this treatment.ATMS temperature unit channel 9 and humility unit channel 22,which are utilized mainly in currenting data assimilation,have marked large bias between the observation and simulated brightness temperature before the bias correction.It makes an issue on which special attention should be paid in the use of ATMS.In addition,compared to AMSUA/MHS,the ATMS window channel has changed significantly.The cloud examination scheme,which is generally based on the observation of the window channel,must be revised and investigated in the assimilation of ATMS satellite data.