RVoG Modelling for Forest Height Retrieval using L-Band NISAR Data

Accurate estimation of forest canopy height is essential for characterizing forest structure, biomass, carbon stocks, and ecosystem dynamics. The recently launched NASA–ISRO Synthetic Aperture Radar (NISAR) mission provides a new opportunity for large-scale forest height retrieval using L-band Interferometric SAR observations. The present study evaluates the potential of NISAR L-band for forest height estimation over the Malhan forest region using available NISAR Geocoded Single-Look Complex (GSLC) products. Canopy height was retrieved using the Random Volume over Ground (RVoG) model applied to a repeat-pass cross-pol HV InSAR pair acquired on 1 January 2026 and 13 January 2026, with a temporal baseline of 12 days. The retrieved height was validated using canopy height measurements from the Global Ecosystem Dynamics Investigation (GEDI) Level 2A (L2A) product. Statistical analysis with GEDI L2A canopy height yielded an RMSE of 6.77 m, an MAE of 3.94 m, and a bias of −2.34 m. The NISAR-derived height exhibited a significant relationship with GEDI measurements (R² = 0.46, p < 0.001), suggesting 46 % variability in GEDI height was explained by the NISAR-derived estimates. The retrieved height suggested compressed dynamic range, followed by underestimation of taller canopies and overestimation of shorter vegetation. These observed discrepancies may be attributed to the temporal decorrelation, simplified inherent assumptions in the RVoG model, and coherence uncertainty. These findings highlight the capability of NISAR L-band repeat-pass interferometry for forest canopy height estimation in heterogeneous forest environments, and its potential for regional forest structure.