Monitoring of subarctic shrub vegetation characteristics using TerraSAR-X and RADARSAT-2 data

Due to its sensitivity to the vertical structure of vegetation, SAR remote sensing has the potential to become a useful tool to map changes in shrub vegetation in subarctic environments. In this paper we evaluate the sensitivity of RADARSAT-2 and TerraSAR-X data to subarctic shrub vegetation. Backscattering was compared with field measurements of shrub height and regression analyses were performed to estimate the sensitivity of SAR signal to shrub height. An unsupervised Wishart classification was also performed on the RADARSAT-2 images to evaluate the separability between shrubs and other elements of the environment. RADARSAT-2 proved to be the most sensitive to shrub height with R² values reaching 0.64 for σ⁰_HV for fall acquisitions and losing little sensitivity during the winter. The unsupervised classification provided good separability between shrub dominated areas and other land cover types and was able to distinguish high shrubs from low shrubs.