Indirect methods of tree biomass estimation and their uncertainties

Abstract

Tree aboveground biomass (AGB) may be determined using direct (weighing or biomass models) and indirect (involving the use of a form factor [FF]) or volume models, basic density [BD] and biomass expansion factor [BEF]) methods. Focusing on three dominant mangrove species (Avicennia marina (Forssk.) Vierh., Sonneratia alba J.Smith and Rhizophora mucronata Lam.) in Tanzania, this study assessed uncertainties of biomass estimation using indirect methods. To achieve this, volume models were developed and FF, BD and BEF were determined. The study covered four sites along the coasts of Tanzania. Volume, FF and BEF were determined for trees with diameter at breast height (dbh) > 15 cm (n = 55). Based on goodness-of-fit statistics, both developed volume models, based on dbh only and dbh and height, fitted well to data. Depending on data availability (dbh only or both dbh and total tree height) either of the models may be applied to generate satisfactory estimates of tree volume needed for planning and decision-making in management of mangrove forests. The study found an overall mean FF value of 0.65 ± 0.03 (SE), 0.56 ± 0.03 (SE) and 0.57 ± 0.03 (SE) for A. marina, S. alba and R. mucronata, respectively. The BEF was 1.37, 1.62 and 1.34 for A. marina, S. alba and R. mucronata, respectively. On average, the BEF values accounted for more than 20% of AGB. Neither FF nor BEF varied significantly between the three species. In general, the use of indirect methods of AGB estimation resulted in uncertainties ranging from small to large. Therefore, volume (volume model or FF), BD and BEF based approaches as alternative means to biomass estimation should be used with caution because they are likely to lead to biased estimates. For accurate estimation of biomass and carbon stocks, biomass models based on data obtained through destructive sampling (direct method) are recommended.

Keywords: accuracy, basic density, biomass expansion factor, mangrove, precision, tree volume models, Tanzania