Preprints
https://doi.org/10.5194/egusphere-2024-1825
https://doi.org/10.5194/egusphere-2024-1825
24 Jul 2024
 | 24 Jul 2024
Status: this preprint is open for discussion.

Time Series Analysis of C-Band Sentinel-1 SAR Over Mountainous Snow with Physical Models of Volume and Surface Scattering

Firoz Kanti Borah, Jonas-Fredrick Jans, Zhenming Huang, Leung Tsang, Hans Lievens, and Edward Kim

Abstract. In this article, we analyze the time series data collected by the Sentinel-1 C-band synthetic aperture radar over the Alps mountains in Southern France. Both the co-polarized and cross-polarized radar data are analyzed. We study the combined effects of the volume scattering of snow and the rough soil surface scattering below the snow using physical models to show the contributions from both components at C-band. For volume scattering, the bi-continuous DMRT equations are used to obtain backscattering coefficients of the dense media layer. To calculate the rough surface scattering component, 3D Numerical Solution of Maxwell’s Equations (NMM3D) method is used. The bi-continuous DMRT model is based on the dense media radiative transfer where the microstructure of the media is controlled by two parameters: mean grain size <ζ> and aggregation parameter b. Radiative transfer (RT) equations are used to obtain backscattering coefficients of the dense media layer. In the dense media layer, we consider ice grains aggregated into clusters. The NMM3D model is a method of moments (MoM) based electromagnetic solver that calculates backscattering from a rough surface of a given permittivity, rms height and correlation length. For applications to remote sensing of terrestrial snow at C-band, the results of DMRT simulations show that cross-polarization components of snow volume scattering at large snow depth are larger than that of the of soil surface scattering. The time series are analyzed by varying the snow parameters with time yet keeping the rough surface parameters constant with time. There is good agreement between the time series of the modelled and measured backscatter at both co-pol and cross-pol within 2dB. Results are presented for three seasons between 2017 and 2020 and for four different locations in the Alps. Using this physical model of volume and surface scattering, we also discuss the reason for Sentinel-1’s sensitivity in cross pol to snow.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Firoz Kanti Borah, Jonas-Fredrick Jans, Zhenming Huang, Leung Tsang, Hans Lievens, and Edward Kim

Status: open (until 04 Sep 2024)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Firoz Kanti Borah, Jonas-Fredrick Jans, Zhenming Huang, Leung Tsang, Hans Lievens, and Edward Kim
Firoz Kanti Borah, Jonas-Fredrick Jans, Zhenming Huang, Leung Tsang, Hans Lievens, and Edward Kim

Viewed

Total article views: 67 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
39 22 6 67 2 2
  • HTML: 39
  • PDF: 22
  • XML: 6
  • Total: 67
  • BibTeX: 2
  • EndNote: 2
Views and downloads (calculated since 24 Jul 2024)
Cumulative views and downloads (calculated since 24 Jul 2024)

Viewed (geographical distribution)

Total article views: 65 (including HTML, PDF, and XML) Thereof 65 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 26 Jul 2024
Download
Short summary
In this paper, we study radar data collected by Sentinel-1 over mountain regions of Alps. Using physical models of snow and soil surface scattering, we show the reasons for the high sensitivity of cross-polarized observations with snow depth. This accurate modelling for cross-pol using physical models can be then used to retrieve snow depth at for very deep snow at mountain regions using the cross-pol signal.