Development of Stem Volume Equation for Urban Trees of Abomey-Calavi in Southern Benin (West Africa)

Abstract

Information on growing stock is important for understanding health assessment, environmental analysis, carbon storage estimation, and economic analysis of urban forest. The stand volume estimation enables the calculation of ecosystemic services value and growth stock of urban forests. However, most of volume models fitted for multiple species in tropical forests may not be suitable for urban trees. This study was conducted to develop generic volume models for urban trees in Abomey-Calavi at the southern Benin. A total of 1608 trees belonging to 80 plant species were measured for their diameter at breast height (DBH), stem height (h) and stem volume using non-destructive sampling methods. Using a nonlinear procedure, six volume models were constructed. Cross validation and Fit statistics like standard error of estimate (SEE), relative absolute error (RAE), root mean square error (RMSE), fit index (FI), Akaike information criterion (AIC) and Willmott’s agreement index (dw) were used to evaluate the efficiency and stability of different models. The six generic volume models developed in this study included both diameter and height. These models exhibited an absence of multicollinearity, with normal and homoscedastic residuals. Furthermore, they show high efficiency (IF > 0.997) and reduce of prediction errors (RMSE: 0.05388–0.06629 m³; RAE: 0.05186–0.06952), which ensuring stability in the estimates. However, the Model II was the best for predicting the stem volume of urban tree according to evaluation statistics and rank analysis. The models developed can provide stem volumes prediction with accurate estimations. Though, stem heights should be systematically measured. These models can contribute to assess the productivity of urban forests in order to pursue their sustainable management and planning.