中文摘要：

提出了一种基于微波双极化数据的土壤水分反演经验模型，该模型引入了新的综合粗糙度参数Rs=√s^2／L来描述地表粗糙状况，将两个粗糙度参数均方根高度s和相关长度L合二为一，因而模型的未知量仅为Rs与法向菲涅尔反射系数Г^0.基于AIEM模型数值模拟，建立了后向散射系数与Rs、Г0的经验关系，并利用两个极化的微波数据同时反演得到粗糙度参数Rs和Г^0，进而得到地表土壤水分。实测数据表明，该模型反演的土壤水分与地表实测值相关性较高（R^2=0．681，RMS=0．043），在土壤水分反演方面具有较大的潜力。

英文摘要：

This paper proposes an original experiential methodology to retrieve bare surface soil moisture by two-polarized microwave remote sensing data. In the model, we combined the roughness parameters, the root mean square S and correlative length L, and introduced a new synthetic roughness parameter Rs to describe the land surface. So, the unknown parameter in this model reduces to R, and Fresnel reflection coefficient in normal direction Г0. Then, Г0 and Rs can both be retrieved using two- polarized microwave data. In situ measurements from Heihe experiments were used to test the empirical model. Results indicate that there is a strong linear relationship between the estimated soil moisture and the in situ measurements （R2 =0. 681, RMS=0. 043）