摘要
实测与模拟的微波辐射计亮温存在偏差,导致基于BP神经网络模型的大气温湿廓线反演精度的降低。研究了一种基于资料订正后的BP神经网络反演大气温湿廓线的方法。首先,基于2014年6月南京江宁探空资料,利用MonoRTM模式,模拟中心频率在22.234 GHz^58.800 GHz范围内22通道亮温;对比模拟和实测南京站微波辐射计资料,建立实测微波辐射计资料订正模型。然后,以南京地区2011—2013年探空资料为输入,模拟22通道亮温数据,并基于模拟的22通道亮温数据和当地探空资料,利用BP神经网络算法,建立大气温度、水汽密度、相对湿度廓线反演模型。最后,利用构建的订正模型,对2014年7月试验获取的微波辐射计资料进行订正,并将订正后的微波辐射计资料输入BP神经网络反演模型,反演0~10 km高度58层的大气温度、水汽密度和相对湿度,对比实际探空资料以及微波辐射计二级产品,评估分析反演效果。实验结果表明:所建的反演模型提高了大气温湿廓线反演精度,大气温度、水汽密度和相对湿度均方根误差范围分别为1.0~2.0 K、0.20~1.93 g/m^3和2.5%~18.6%。
There exists deviation between the measured and simulated microwave radiometer sounding data. The bias results in low-accuracy atmospheric temperature and humidity profiles simulated by Back Propagation artifi- cial neural network models. This paper studied a retrieving atmospheric temperature and humidity profiles method by adopting an input data adjustment based Back Propagation artificial neural networks model. Firstly, the sounding data acquired at a Nanjing meteorological site in June 2014 was inputted into the MonoRTM Radiative transfer model to simulate atmospheric downwelling radiance at the 22 spectral channels from 22. 234 GHz to 58.8 GHz, and perform comparison and analysis between the real observed data, results in building up an ad- justment model for the measured microwave radiometer sounding data. Secondly, we simulated the 22 channels＇ sounding data by using the sounding data acquired at Nanjing meteorological site from 2011 to 2013. Based on the simulated rightness temperature data and the sounding data, BP neural network-based models were trained for the retrieval of atmospheric temperature, water vapor density and relative humidity profiles, respectively. Finally, we apply the adjustment model to the microwave radiometer sounding data collected in July 2014, genera- ting the corrected data. After that, we input the corrected data into the BP neural network regression model to predict the atmospheric temperature, vapor density and relative humidity profile at 58 high levels during 0 - 10 km. We evaluated our model＇s effect by comparing the out model＇s output with the real measured data and the microwave radiometer＇s own second-level product. The experiments showed that the inversion model improve atmospheric temperature and humidity profiles retrieval accuracy, the atmospheric temperature RMS error is between 1K and 2. 0K; the water vapor density＇s RMS error is between 0. 2 g/m^3 and 1.93 g/m^3 ; the relative humidity＇s RMS error is between 2. 5% and 18.6%.
出处
《热带气象学报》
CSCD
北大核心
2016年第2期163-171,共9页
Journal of Tropical Meteorology
基金
国家重点基础研究发展计划(973计划)资助项目(2013CB430101)
六大人才高分项目(2015-JY-013)
中国博士后科学基金资助项目(20090461131
201003596)
江苏高校优势学科建设工程资助项目(PAPD)共同资助
关键词
地基微波辐射计
BP神经网络
大气廓线
反演精度
ground-based microwave radiometer
BP Neural Network
atmospheric profiles
regression accuracy
