Structural–carbon decoupling and forest structural thinning in degrading forests of Southwestern Nigeria using GEDI LiDAR and multi-sensor data fusion

Oluwafemi David Bejide , Kunle David Emiola, Hezekiah Daramola Olaniran

Accurate monitoring of forest degradation requires indicators that capture both structural condition and carbon dynamics. While canopy height derived from spaceborne LiDAR is widely used as a proxy for forest condition, its ability to represent aboveground biomass (AGB) under ongoing degradation remains uncertain. This study examines the relationship between canopy height and AGB in tropical forests of Southwestern Nigeria between 2020 and 2025 using GEDI LiDAR and multi-sensor data.

Canopy height (RH98) and AGB were modelled independently using machine learning and multi-source predictors. Model performance was moderate (R² = 0.38–0.49 for canopy height; R² = 0.45–0.48 for AGB). Both variables declined over time; however, biomass loss (20–39%) consistently exceeded canopy height reduction (10–23%).

A structural–carbon decoupling index (DI = 1.83) indicates that biomass declined approximately 1.8 times faster than canopy height. Aboveground carbon decreased from 111.6 to 77.7 tCO₂ ha⁻¹, corresponding to a loss of ~33.9 tCO₂ ha⁻¹. Spatial patterns indicate structural thinning rather than complete canopy loss.

These findings demonstrate that canopy height alone may underestimate carbon loss in degrading tropical forests and highlight the importance of integrating multi-sensor data for large-scale forest monitoring.