Spaceborne PolInSAR and ground-based TLS data modeling for characterization of forest structural and biophysical parameters

Mapping of forest biophysical parameters is an effective tool for determination of forest inventories, vegetation modeling, and the global carbon cycle. The present study aims at quantification of the forest biophysical parameters of a sub-tropical Forest of India with Terrestrial Laser Scanner (TLS) and Polarimetric Interferometry SAR (PolInSAR) modeling approaches. TLS data was acquired in single and multiple scans and it was found that the multiple scanning approaches have the capability to provide forest parameters with very high accuracy. The RANSAC algorithm was implemented on TLS point cloud data to derive forest structural parameters. The TLS modeled DBH and stem volume exhibited a coefficient of determination of 0.88 and 0.80 and RMSE of 2.85 cm and 0.4 cu.m respectively. PolInSAR Coherence Amplitude Inversion (CAI) and RVoG techniques were implemented to retrieve forest stand height. Both the PolInSAR inversion based models were employed with the integration of complex coherences occurring in different polarimetric channels. CAI estimated the forest height precisely nearly equivalent to the field measured forest height with a coefficient of determination of 0.52, percent accuracy 85.85% and RMSE of 2.94 m. Application of the RVoG model further improved the statistics of the forest height with a coefficient of determination of 0.57, percent accuracy 89.94% and RMSE of 2.13 m. Measurement of tree height estimation using RVoG showed false vegetation height for dry riverbed and the same feature was characterized by the CAI model with zero vegetation height. The research showed that the forest vegetation height characterization of CAI model is better than the RVoG. This study has successfully investigated the potential of spaceborne PolInSAR and ground-based TLS data for forest structural and biophysical parameters retrieval.