根据离散方法建模垄行结构农田表面微波发射率,与地基多频率微波辐射计实测发射率比较发现：二者之间的平均绝对偏差小于0.01 ,证实了利用离散化方法建模农田表面微波发射率的可行性.在给定条件下不同观测方位角农田表面微波发射率与平坦表面的发射率差值在0.02 与0.05 之间,这说明农田结构微波辐射具有各向异性,行结构对发射率的影响在农田电磁波辐射建模过程中不可忽略.本文分析了不同土壤湿度条件下农田垄行结构可能引起的土壤湿度反演误差,结果表明,土壤湿度变化范围是0.02—0.5 cm³/cm³,垄行结构引起的土壤湿度反演误差为0—0.1 cm³/cm³, 此误差超过了土壤湿度反演的容限值,因此在进行农田参数的遥感提取过程中不可忽略周期性垄行结构对表面发射率的影响.

The discrete method is used to simulate emissivity of one dimension of rough surface with row structure. Compared to the emissivity measured by ground radiometer, the bias between measured and simulated emissivity is less than 0.01, and this conf irms that the discrete method is feasibility to simulate emissivity of rough surface with row structure. At a given row structure, the difference of emissivity between row structure and fl at smooth surface changes with observed azimuth angle, and the value falls in the range of 0.02 and 0.05. This indicates that emissivity of row structure surface is anisotropic, and it should be considered in the process of modeling microwave radiation of row structure surface. In addition, the error of soil moisture inversion caused by neglecting the existence of row structure is analyzed across different soil moisture, and results show that the absolute error of soil moisture is between 0 cm³/cm³ and 0.1 cm³/cm³ when soil moisture changes from 0.02 cm³/cm³ to 0.5 cm³/cm³. This error is greater than the error tolerance limit of soil moisture inversion. Therefore, in the process of farmland parameter inversion from passive microwave remote sensing, surface row structure effect cannot be ignored.