Stump diameter is a robust predictor of biomass and carbon stocks in tropical Acacia plantations

Accurate estimation of tree biomass is essential for tropical forest management, climate change mitigation, and REDD+ monitoring. While conventional inventories rely on diameter at breast height (DBH), measurement of DBH is often impractical in dense stands, buttressed stems, or post-harvest contexts. This study evaluates stump diameter (dst, measured at 30 cm above ground) as a complementary proxy for DBH and stem biomass in Acacia auriculiformis , A. mangium , and their hybrid plantations in Bangladesh. We measured a stratified sample of 1,050 trees and destructively sampled 75 individuals for calibration. Fourteen models (linear, ln–ln, polynomial, mixed-effects, and machine learning) were tested using adjusted R 2 , RMSE, AIC, and Durbin–Watson statistics. The parsimonious ln–ln regression performed very well (R 2 > 0.98), with low RMSE, MAE, and high concordance coefficients confirming unbiased predictions under plantation conditions. Biomass models based on stump performed comparably to DBH models, while mixed-effects models reduced bias, particularly for hybrid Acacia. Validation with independent datasets and five-fold cross-validation confirmed unbiased predictions ( p > 0.05) with narrow confidence intervals. Random Forest identified stump diameter as the strongest predictor, while Gradient Boosting showed signs of overfitting. Species-specific carbon stock estimates indicated that hybrid Acacia was the highest-yielding, followed by A. mangium and A. auriculiformis . Carbon estimates are limited to stem biomass only. This study provides a practical tool for rapid biomass assessment and post-harvest carbon accounting. The findings may contribute to climate and biodiversity goals (SDG 13 and SDG 15), though broader policy relevance requires validation across diverse plantation types and ages.