Allometric modeling for accurate biomass and carbon estimation in shea trees of west african parklands

Accurate estimation of biomass and carbon stocks in agroforestry systems is essential to assess their role in climate mitigation and sustainable land management. This study presents allometric models to estimate above and Biomass below ground and carbon stocks of shea tree, a keystone species in West African parklands. We destructively sampled 30 trees on a size gradient and applied Generalized Additive Models for Location, Scale, and Shape (GAMLSS) to derive predictive models using tree diameter, height, and crown area as predictors. The best models captured structural and functional variation between tree sizes and explained more than 89% of the variance in biomass components. The root-to-shoot ratio decreased with increasing crown area, suggesting a developmental change in biomass allocation. Our findings confirm that GAMLSS provides a flexible and accurate framework for ecological modeling and underscore the significance of species-specific approaches in carbon accounting. These models offer practical tools designed to improve agroforestry management, facilitate national forest inventories, and address the accounting requirements for carbon in shea tree parklands, catering to the needs of the burgeoning carbon market.