资源环境科技发展态势分析平台

A numerical correction algorithm (NCA) is proposed to improve the accuracy of atmospheric temperature profile in clear and cloudy days based on the weighting functions of different frequencies. The weighting functions of atmospheric brightness temperatures (BTs) with respect to temperature profiles are analyzed in detail. A performance assessment of the microwave radiometer (MWR) MP-3000A is operated with the observations from March to August 2018 in Harbin China (126.40 degrees E 45.46 degrees N). And a numerical method is used to calculate weighting functions with the downwelling monochromatic radiative transfer model (MonoRTM). According to the weighting functions, a cost function is defined in NCA to minimize the biases between the measured and calculated BTs by correcting the measured temperature profiles. A simulation is designed to examine the NCA based on the methodology and data. The experimental results show that the NCA can adjust the temperature profiles below 5 km. However, there is almost no improvement above 5 km because of the attenuation of radiation. And the statistic results show that the RMSE and MAE of temperature profiles corrected with NCA are reduced by 12.5% and 17.9% when compared with the uncorrected observations. Especially, both of the RMSE and MAE of the temperature profiles below 2 km are reduced by more than 50%.