基于AIEM和实地观测数据对GNSS-R反演土壤水分的研究

The study of soil moisture retrieval from GNSS-R signals based on AIEM model and experiment data

介绍了用GPS反射信号反演土壤水分的原理及反演研究的进展,并用改进的积分方程模型(AIEM)和实地观测数据对利用GPS反射信号反演土壤水分的方法进行了分析,分析结果表明,由于单个频率的雷达信号受地表粗糙度、角度和地表类型(裸地和植被类型)的影响比较大,很难提出一个实用的通用物理算法。利用美国2002年土壤水分实验(SMEX02)实测数据对上述的反演算法进行了分析,分析结果表明,采用经验统计算法对单个站点观测比较实用,平均相关系数达到0.85以上。整个分析表明,利用GPS前向散射信号与噪声之比反演土壤水分在单个站点能够取得比较高的精度。

This paper makes a survey of the principle of the soil moisture retrieval from the global navigation satellite system-reflection （GNSS-R） signals and the developments in the retrieval studies, and then makes a thorough analysis of the retrieval algorithm based on the data from the advanced integral equation model （AIEM） and the experimental observation. The result shows that the forward scatter single of the GPS is influenced much by the angle and the roughness of the earth＇s surface, so the retrieval algorithm should consider the influence of them. The GPS satellite just has one band （L1） for civil application, so it is very difficult to develop a general physical algorithm for retrieving soil moisture from GNSS-R signals. The experiment data in the soil moisture experiment 2002 （SMEX02） is utilized to analyze the retrieval algorithm, which indicates that the average correlation coefficient is above 0.85 for a single field site, so the soil moisture can be accurately retrieved by GNSS-R for single site. Finally, the prospect of passive microwave for soil moisture retrieval is discussed with the development of microwave remote sensing technique.