Allometric equations to predict aboveground biomass and carbon stock of <i>Faidherbia albida</i> in parkland agroforestry

Maintaining trees on farmland improves the carbon sequestration potential of agricultural lands and contributes to global carbon balance. Samples of 35 trees were harvested for aboveground biomass (AGB) and carbon stock measurements in parkland agroforestry fields of Tigray, Ethiopia. Allometric equations were developed using the regression models that related the total AGB with a diameter at stump height (DSH), diameter at breast height (DBH), and a combination of DSH, DBH, height (H), and crown diameter (CD). The allometric model related to diameter at breast height, Ln (AGB) = -2.85 + 2.4 ln (DBH), was considered the best fit, with the highest adjusted coefficient of determination (0.88) and lowest prediction residuals sum of squares (3.51 kg), bias (-2 kg), root mean square error (0.3 kg), and Akaike information criterion (P &lt; 0.01). The best-fitting model for aboveground biomass estimation was validated using cross-validation. When assessed with the aboveground biomass of <i>F. albida</i> data sets from this work, the species-specific and multi-species previously published models were substantially overestimated by 60% and underestimated by 50%, respectively. <i>F. albida</i> total aboveground biomass was 3372.25 kg ha⁻¹ and 1584.96 kg ha⁻¹ of carbon for a density of 47 trees ha⁻¹ in the parklands. The results showed that the statistical fits of our model were generally good, enabling confidence in its use to predict the aboveground biomass of <i>F. albida</i> stands. Our study identified patterns of carbon distribution in different <i>F. albida</i> tree sizes that may serve as a guide for programs that measure, report, and verify carbon stock and promote a climate-resilient farming system.