Allometric models for above- and below-ground biomass of Sonneratia spp

Allometric models are needed to estimate the stored carbon stocks in a mangrove forest ecosystem (blue carbon). Nevertheless, the development of allometric models for above- and below-ground biomass models for Sonneratia spp. species specifically has not yet been carried out either at global, regional and local scale and therefore it is very important to conduct a research on it. This is because this species is a pioneer species in the mangrove forests and a CO2 absorber, has a very important role in reducing wave energy and retain sediments. Therefore this research was conducted to formulate an allometric model for the above- and below-ground biomass of Sonneratia spp. The biomass data of Sonneratia spp. (S. alba and S. caseolaris) were acquired through destructive sampling technique of 30 tree samples that have a stem diameter (diameter breast height/DBH)) within the range of 1.27 -20.06 cm. The allometric model is obtained through non-linear regression analysis (using GNLS function in R), with a combination of the stem diameter, tree heights and wood density as estimated variables. The largest biomass proportion is found on the part of the stem (31.28%), which was followed by the below-ground roots (23.40%), twig (13.35%), branches (12.96%), foliage (12.50%), while the smallest proportion is present on the pneumatophore (6.51%). The biomass comparison of the above- and below-ground is 3: 1 (76.60%: 23.40%). The result of the analysis indicate that the belowground, aboveground and total biomass can be estimated well with the power function. The aboveground and total biomass can be estimated with just the stem diameter variable, while the belowground biomass can be well estimated with a combination of the stem diameter, tree height, and wood density. Overall, the addition of the tree heights or wood density variable in biomass allometric model development can increase the accuracy of the model, however the results are not significantly different than the model that uses only the stem diameter variable. (C) 2018 The Authors. Published by Elsevier B.V.