Semi-destructive method to derive allometric aboveground biomass model for village forest of Bangladesh: Comparison of regional and pan-tropical models

Abstract

Pan-tropical biomass models were developed for natural and plantation forests, which cover a wide range of geographical areas and tree species. Tree architecture of a species significantly varies among plantations and natural forests as well as village forests or homestead areas. Tree architecture has a significant influence on biomass estimation in allometric models. Therefore, it was hypothesised that pan-tropical biomass models may not be able to address the desired accuracy in biomass estimation for village forests. The objective of this study was to derive a common allometric above-ground biomass model for a village forest in Bangladesh and to compare the efficiency of the derived model with frequently used pan-tropical models. This study adopted a semi-destructive method, where the biomass of individual sampled trees was derived from stem volume, wood density and biomass expansion factor. Eight linear models [natural logarithm(Ln) transformed] were used to derive the best-fit allometric biomass model. In comparison with the best-fit model, the frequently used pan-tropical models showed significant estimation of total above-ground biomass (TAGB). Therefore, the derived model for the village forest showed higher capacity to reduce uncertainty in biomass estimation compared to pan-tropical models. This finding may restrict the indiscriminate use of pan-tropical models without checking their accuracy towards a particular forest type and species.