Allometric regressions for improved estimate of secondary forest biomass in the central Amazon

Autor: Batista, G. T., Couto, L. B., Aquino de Souza, S. G., Mesquita, R., Nelson, B. W., Pereira, J. L. G.
Předmět:
Zdroj: Forest Ecology & Management. 5/17/1999, Vol. 117 Issue 1-3, p149. 0p.
Abstrakt: Estimates of the sequestering of carbon by secondary forests - whichoccupy almost half the deforested area of the Brazilian Amazon - will be improved by the use of accurate allometric relationships for non-destructive measurement of standing biomass and by an evaluation of the suitability of existing equations for application in secondary forest. Species-specific and mixed-species regressions for estimating total above-ground dry weight (DW) were therefore developed using eight abundant secondary forest tree species in the central Amazon. Usingonly DBH as the input variable, the species-specific equations estimated DW of individual trees with an average error of 10-15%. For the mixed-species equations, developed using 132 trees from seven of the eight species (excluding Cecropia), average error in estimating DW ofindividual trees was 19.8% using only DBH and 15.0% using DBH plus specific density of the wood (SD). Average SD for each species can be substituted without increasing the error of the estimate. Adding total tree height (H) as an input variable provided only a slight reduction in error to 14.0%. Previously published mixed-species biomass regression models, based on primary and secondary forest trees of the Amazon, were also cross-validated against the trees of this study. Two of these models, based on primary forest plots and using only DBH as an input, overestimated biomass by 10-60% for central Amazonian secondary forest trees in the size range 5-25cm. The overestimate was greatest for the larger trees. Including Cecropia in the test group will make the overestimate even greater. Those published equations using DBH, H and SD as inputs, whether from secondary or primary forest plots, showed better agreement with the sample-derived regressions and lower average errors in estimation of individual tree dry weights. [ABSTRACT FROM AUTHOR]
Databáze: GreenFILE