Feasibility of single-polarized TanDEM-X data for Hyrcanian forest height estimation (Case study: Shast-Kalateh forest)

In this study, the capability of single-polarized TanDEM-X data was assessed for estimating the Hyrcanian forest height. Two approaches of Sinc interferometric coherence model based on Random Volume over Ground (RVoG) and differential interferometry utilizing airborne LiDAR-derived DTM were compared. The study was conducted in part of Dr. Bahram Nia’s forest management plan (Shast-Kalateh). 308 circular sample plots with an area of 0.1 ha were used in evaluating the accuracy of predicted forest height using two different approaches. For this reason, Lorey’s mean tree height weighted by basal area was calculated. The interferometric coherency was involved in the Sinc function for inverting forest height. We also removed topography from the SAR phase calibrated with LiDAR-derived DTM on open areas for estimating forest height based on differential interferometry. After terrain correction, the average predicted canopy height was extracted for each plot. Our results showed higher accuracy of differential interferometry than the Sinc model in forest height estimation when compared with ground reference data (RMSE=2.86 m and rRMSE=12.28%). Although, the accuracy of estimated forest height by the Sinc model was comparable with RMSE=3.41 m and rRMSE=14.46%. We found TanDEM-X data and coherence-based models a relatively promising approach in the Hyrcanian forest even though only 15% of Lorey’s men height change was explained by the Sinc model. Hence, further studies are needed to figure out the effects of species, slope, and image aquisition features on the results.