Investigation of the Global Influence of Surface Roughness on Space-Borne GNSS-R Observations

Abstract

Accurately characterizing the impact of vegetation and roughness on CYGNSS observations, which are two main sources of disturbance, is essential for achieving high-quality estimates of soil moisture through this mission. While there are several ancillary data sets that can be employed to address vegetation influence, the lack of a global data set for soil surface roughness motivates us to globally map the contribution of soil roughness to CYGNSS observations. To accomplish this, since separating the contribution of surface roughness and vegetation on reflected signals is often challenging, we initially integrate the vegetation and roughness contributions into a unique variable, denoted as VR. Next, the impacts of vegetation integrated into the CYGNSS-derived VR were separated using Leaf Area Index to map the roughness parameter Hr. The mean value of Hr obtained in this research through CYGNSS observations ranges from 3.2 to 4.6. We observed that the spatial distribution of Hr values is influenced by the predominant vegetation types, with forests exhibiting higher roughness values (Hr = 4.47–4.67), while deserts, shrubs, crops, and bare soils exhibit the smallest Hr values (Hr = 3.25–3.36). Furthermore, we inferred vegetation optical depth (VOD) through CYGNSS observations in conjunction with estimated Hr values. The good agreement observed between the estimated VOD in this study and other vegetation indices, including Vegetation Water Content and tree height, highlights the effectiveness of the introduced Hr global data set in our research and its promising potential in the future GNSS-R studies.