SPECIES-SPECIFIC BIOMASS EQUATIONS FOR SMALL-SIZE TREE SPECIES IN SECONDARY TROPICAL FORESTS

Abstract

Secondary tropical forests are largely dominated by small-diameter trees. Their contribution to the total stand productivity may be determinant for ecosystem functioning. However, the biomass contained in small-size trees is frequently ignored because they are rarely included in forest inventories, as well as due to the lack of biomass equations for such trees. The aim of this study was to generate specific biomass equations by tree structural components and total-tree biomass for 22 species (diameter at breast height-DBH ≤ 10 cm diameter) structurally important of secondary tropical forests in southeast Yucatán peninsula. The biomass equations were generated using two fitting approaches. First, using the additive approach where the biomass equations by structural tree components (stem, branches and foliage) were simultaneously fitted. Second, the tree-level approach where the total-tree biomass was independently fitted. We used DBH, total height (TH) and wood density (WD) as predictor variables during the fitting process. The performance of each total-tree biomass equation was compared to generalized biomass equations developed in other tropical forests. The variance explained of observed biomass across tree species were stem, 64% to 99%; branch, 24% to 95%; foliage, 35% to 94% for equations by structural tree components; while the explained variance for total tree biomass equations was between 94 to 99 %. The inclusion of TH and WD as predictor variables in the equation structure, significantly improved the goodness of fit statistics. These are the first equations developed for small-size tree species in secondary forests of Yucatán peninsula. We observed that generalized equations developed for other tropical forests, in some cases overestimated up to ~ 44.2% and in other cases underestimated up to ~ 48.4% the mean total-tree biomass. When we applied our equations to forest inventory data, they showed a better predicting performance than generalized equations. Therefore, they are reliable for estimating biomass of small-size tree species in similar forest types. Our equations could be used for biomass-carbon estimation in tropical secondary forests composed by high densities of small-size tree species; therefore, they are relevant within climate change contexts.