A generalized tree component biomass model derived from principles of variable allometry

Accurate estimates of forest biomass stocks are critical for scientists, policymakers and forest managers trying to address an increasing array of demands on forests, to sustain human well-being and a broader diversity of life forms on Earth. Thus, it is important that forest biomass estimates are translatable into both biologically and economically meaningful components. Here, a new variable-form, variable-density tree mass component model is presented. The model decomposes a tree into a system of tree component-specific equations that: (a) reflect variation in scaling relationships between major portions of the tree body that define variation in whole-tree growth form and (b) relate to commercially relevant portions of the tree. When tested using data collected from felled and dissected hardwood trees of different size and species, growing over a range of stand conditions, the variable-form, variable-density models gave superior predictions for all components of tree mass, when compared to standard fixed-form, fixed-density models that predict tree mass components only from stem diameter at breast height (DBH). The results demonstrated why the standard approach of estimating mass components from DBH with a power function is fairly limited, because base-, trunk-, crown- and main stem-DBH relationships are all variable within and between tree species. Species-specific models were generally superior, but a mixed-species model gave equivalent and sometimes better results than equations fitted to each species individually. The results provide a theoretical basis for biologically-meaningful, robust estimation of tree biomass components over a range of species and forest conditions and may offer new flexibility in producing ecologically and economically relevant biomass inventories.