Development of aboveground biomass allometric equations for seven agroforestry tree species in northern Uganda
Accurate estimating aboveground biomass (AGB) is essential for quantifying carbon stocks and assessing the climate-mitigation potential of agroforestry systems. However, many existing biomass models in Uganda and other eastern African countries were developed for natural forests, rely on limited sample sizes, or employ generalized equations that may not adequately represent trees grown under smallholder agroforestry setting This study developed species-specific allometric equations to estimate AGB for seven agroforestry tree species in northern Uganda: Acacia polyacantha, Albizia lebbeck, Gmelina arborea, Maesopsis eminii, Markhamia lutea, Melia volkensii, and Senna siamea. Destructive sampling was conducted on 36–44 trees per species across a range of diameter classes and management regimes in Gulu, Amuru, Nwoya, and Oyam districts. Diameter at breast height (D), tree height (H), and crown radius (CR) were fitted as predictors of AGB using nonlinear maximum-likelihood regression models. Model performance was assessed using coefficient of determination (R2), root mean square error (SE), prediction error (PE%), Akaike Information Criterion (AIC), and repeated k-fold cross-validation. The selected models explained most of the variation in AGB (R2 = 0.90–0.99) and PE% below 3.5%. Model structure varied among species: for M. eminii and S. siamea, D alone adequately explained AGB variation, whereas other species required combinations of D with H or CR. Comparison with previously developed models revealed that generalised or pantropical equations produced biased estimates. Therefore, the developed species-specific equations provide reliable tools for estimating AGB and carbon stocks in smallholder agroforestry systems and woodlots in northern Uganda, supporting carbon accounting and restoration monitoring.
Download :
This works is licensed under a Creative Commons Attribution 4.0 International License.
Authors
Mugasha, W.A.,Kimaro, A.A.,Trimarco, J.
Publication year
2026
