Abstract The inversion of soil moisture from Synthetic Aperture Radar (SAR) closure phases is intrinsically plagued by ambiguities that affect the moisture order. This work shows a characterization of the ambiguities and a way to solve for them with the help of interferometric coherence. This allows to properly constrain the inversion and to retrieve the moisture signal. A data set of ALOS-2/PALSAR-2 L-band images is used as an example of successful inversion at the scene level, with sub-kilometer resolution. The results are validated with soil moisture products based on ASCAT and show a high degree of correlation. The raw moisture derived by the algorithm could be immediately used to correct SAR interferometric phases; however, for applications that need absolute moisture levels, a calibration step is likely necessary. Unexpectedly, a good performance was observed over forested areas, which suggests a sensitivity of closure phases to tree moisture; at the same time, over pastures and agricultural fields the closure phase signal was found relatively weak. Additional research is needed to evaluate the applicability of the same measurements principle to shorter wavelengths and exploitation of potential synergies with backscatter and polarimetric information.

中文翻译：

---

SAR 闭合阶段的植被和土壤水分反演：初步实验和结果

摘要 合成孔径雷达 (SAR) 闭合阶段的土壤水分反演本质上受到影响水分顺序的模糊性的困扰。这项工作展示了模糊度的表征以及借助干涉相干性解决它们的方法。这允许适当地约束反演并检索水分信号。ALOS-2/PALSAR-2 L 波段影像数据集被用作场景级成功反演的示例，具有亚公里分辨率。结果用基于 ASCAT 的土壤水分产品进行了验证，并显示出高度的相关性。算法得出的原始水分可立即用于校正 SAR 干涉相位；然而，对于需要绝对湿度水平的应用，校准步骤可能是必要的。不料，在森林地区观察到良好的表现，这表明关闭阶段对树木水分很敏感；同时，在牧场和农田上发现关闭阶段信号相对较弱。需要进一步研究来评估相同测量原理对较短波长的适用性以及利用反向散射和偏振信息的潜在协同作用。