Estimating stem volume using breast height and stump diameters for Tectona grandis Lin F. in Afe Babalola plantation, Ado Ekiti, Nigeria

Abstract

Forest plantation has been identified as a quick fix to the attendant problems of deforestation. Information on appropriate volume models estimated with Diameter at Breast Height (dbh) and stump diameter of plantation grown species for sustainable management is inadequate. Therefore, this study was carried out to develop volume models using dbh and stump diameter for Tectona grandis in Afe Babalola plantation, Ado Ekiti, Nigeria. Four age series (22, 20, 14 and 11 years old) of Tectona grandis were purposively selected based on availability. A total of 40 temporary sample plots (20 x 20 m) were laid using stratified random sampling, considering each age series as a stratum, all trees with DBH ≥ 10cm in each plot were enumerated. Tree dbh, Diameter at Stump Height (DSH), Total Height (TH) and Merchantable Height (MH) were measured. Basal area (BA, m²/ha) and Total Volume (TV, m³/ha) were estimated. Five models were selected each for DBH and DSH to predict stem volume. The suitability of these models was evaluated and the best selected based on least Root Mean Square Error (RMSE), Akaike Information Criterion (AIC) and highest adjusted R² (R²adj). Data were analyzed using descriptive statistics and regression at α0.05. The dbh observed ranges between 10.0 to 34.0 across all the age series while the mean height varied from 16.15±0.21m to 16.79±1.69m. The basal area ranged from 20.17m²h^-1 to 23.81m²h^-1. The DSH ranges between 19.55±0.32cm and 19.43±0.26cm. The estimated merchantable volume range between 140.0±0.007m³/ha and 167.32±0.005m³/ha for the whole stand. The selected model for volume - DBH relationship was V = 0.0000476D1.666 × H1.296 (AIC = -2693.595, R²adj = 0.8388, RSME = 0.0466) while for volume – DSH relationship, the best model was: V = 0.2447 + 0.00591Dst + 0.0004Dst2 + 0.0134H (AIC = -2642.177, R²adj = 0.8285, RSME = 0.0480). Stand volumes can be estimated using the selected models.