P-band SAR polarimetry for estimating trunk permittivity in forested regions

Abstract

In times of global warming with increasing weather extremes such as droughts and floods [1], the estimation of relevant geophysical variables on global scale become of significant importance. Vegetation moisture is hereby one of the essential parameters within the soil-plant-atmosphere system (SPAS), influencing the water and energy cycles as well as controlling the plant water and carbon uptakes [2]. Thus, its monitoring on large scale is crucial for a better understanding and prediction of the consequences of global climate change. In this study, we propose a new method for estimating the trunk permittivity by using the decomposed dihedral scattering component of polarimetric P-band SAR signals. At strongly vertically orientated landcovers like forests, the dihedral mechanism from P-band data has shown to be the dominant scattering mechanism compared to soil or volume scattering [3-5]. Hence, it can be explored for trunk permittivity estimation. For that, the extended Fresnel model is employed, which computes the Fresnel reflection coefficients of the soil and trunk scattering plane by accounting for potential phase differences between horizontal and vertical polarized backscattering, and for soil-roughness induced depolarization [6]. The method is tested on polarimetric P-band (435 MHz) SAR observations from NASA’s AirMOSS mission, conducted between 2013 and 2015 [5]. The method was validated at in situ measuring stations, covered by boreal, temporal or mixed forests. First results indicate less influence of the surface roughness at P-band compared to L- or C-band. However, the necessity to account for co-polarimetric phase difference is key, since the assumption on negligible propagation differences of individual polarizations in oriented vegetation, can lead to an overestimation of model parameters. As a first check at the site, where the forest was cut at a ~140 m radius around the station, unrealistic trunk permittivity values are estimated due to the fact that the dihedral scattering is negligible. In contrast, at the temporal forest site with the second highest density of evergreen needle-leaf trees, realistic trunk permittivity levels are estimated, corresponding to soil permittivity measurements. Next steps include more detailed analyses of estimated trunk permittivity at larger scales in correlation with soil permittivity and auxiliary parameters (i.e., evapotranspiration, relative water content) in order to evaluate the method. [1] Vereecken, H. et al., 2022. Soil hydrology in the Earth system. Nat Rev Earth Environ 3, 573–587 [2] Jagdhuber, T. et al., 2022. Toward estimation of seasonal water dynamics of winter wheat from ground-based L-band radiometry: a concept study, Biogeosciences, 19, 2273–2294 [3] Fluhrer, A. et al., 2022. Remote Sensing of Complex Permittivity and Penetration Depth of Soils Using P-Band SAR Polarimetry. Remote Sensing 14, 2755 [4] Moghaddam, M. & Saatchi, S., 1995. Analysis of Scattering Mechanisms in SAR Imagery over Boreal Forest: Results from BOREAS ’93. IEEE Trans. Geosci. Remote Sens. 33, 1290–1296 [5] Alemohammad, S.H. et al., 2019. Soil and Vegetation Scattering Contributions in L-Band and P-Band Polarimetric SAR Observations. IEEE Trans. Geosci. Remote Sens. 57, 8417–8429 [6] Jagdhuber, T., 2016. An Approach to Extended Fresnel Scattering for Modeling of Depolarizing Soil-Trunk Double-Bounce Scattering. Remote Sensing, 8(10), 818