TAU-OMEGA遥感辐射模型改进与参数化

大尺度地表土壤水分信息的获取对水资源管理、农业生产以及气候变化等相关研究具有重要意义。TAU-OMEGA（$ tau -omega $）模型是利用被动微波遥感技术进行大尺度土壤水分信息提取的常用模型。由于$ tau -omega $模型忽略了植被层的体散射作用，该模型仅适用于L波段，存在C波段适用性差及模型参数无法定量计算的问题。该研究在分析植被冠层对微波散射机制的基础上，通过增加反映冠层多次散射作用的辐射添加项对$ tau -omega $模型进行了改进，成功解决了模型在C波段的适用性问题；通过理论推导，得出了模型$ omega $参数的理论计算方法，基于模拟数据集，实现了基于冠层叶面积指数的模型$ omega $和$ tau $参数的计算，解决了$ tau -omega $模型中$ omega $参数无法根据遥感数据进行定量计算的问题。以玉米冠层为例，在C波段（6.6 GHz），改进的$ tau -omega $模型对地表微波辐射亮温的模拟值与实测数据保持了较好的一致性，模拟误差较改进前有了极大的下降，V极化均方根误差为3.02 K，H极化均方根误差为3.94 K，结果表明了该研究提出的模型改进与参数化方案的合理性，研究结果为联合光学和被动微波遥感数据进行大尺度土壤水分反演奠定了基础。

Improvement and parameterization of the TAU-OMEGA radiation model

Soil moisture information was of great significance for water resources management, agricultural production, and climate change. Passive microwave remote sensing had gradually become one of the most important technical means to obtain surface soil moisture. The TAU-OMEGA model had been successfully applied in global soil moisture inversion using L band passive microwave data but there would be some problems, such as the poor performance of the model and the uncertainty of the model parameter at C-band. Based on the analysis of the mechanism of microwave scattering by vegetation canopy, this study realized the quantitative calculation of multiple scattering effect of the model on canopy by adding the radiation addition term of unobserved radiation of soil scattered by vegetation layer to the observation direction, which greatly improved the simulation accuracy of the model in C-band and successfully solved the applicability problem of the model in C-band. Through theoretical deduction, the theoretical calculation formula of the important parameters of the model was obtained. Based on the simulation data set of the physical model, the quantitative calculation of the parameters of the model was realized by canopy leaf area index (LAI), so that the model parameters could be quantitatively calculated in large areas according to the LAI products obtained from satellite remote sensing data. In this study, a C-band (6.6 GHz) microwave radiometer was used to verify the simulation accuracy of the model before and after the improvement. The data included the observation results of 8 times from seedling stage to flowering stage of maize. The results showed that in the case of using the model parameters based on quantitative calculation of LAI, the microwave radiation brightness temperature of surface would be seriously underestimated by the model before the improvement, while the simulated value of surface microwave radiation brightness temperature of the improved model was in good agreement with the measured data. The simulation error of the model before improvement showed a trend of first increasing and then decreasing from seedling stage to flowering stage, because the body scattering term of canopy on soil radiation was ignored by the Before the improvement of the model, resulting in a larger error of simulation value with the enhancement of canopy scattering effect. With the further growth of the canopy, the attenuation effect of the canopy on the soil microwave radiation exceeded the scattering effect on the soil microwave radiation, and the contribution of the soil microwave radiation to the observed brightness temperature gradually decreased. The average RMSE of the model before improvement in eight growth stages of maize canopy was 7.27 K for V polarization and 7.92 K for H polarization. After the improvement, the average root mean square error (RMSE) of V polarization and H polarization were 3.02 K and 3.94 K respectively. It could be inferred that the soil moisture inversion error caused by the model simulation error could be reduced to less than 1.5%, which greatly improves the potential of the model in soil moisture extraction, according to the simulation errors of the improved model for the brightness temperature of microwave radiation in the C-band V polarization and H polarization of the corn plot. The results also showed that the model improvement and parameterization scheme proposed in this study are reasonable, and the research results laid a foundation for the large-scale soil moisture inversion by combining optical and passive microwave remote sensing data.