Estimation of biomass in a Neotropical forest in French Guiana: spatial and temporal variability
Chave J; Riera B; Dubois M
Journal of Tropical Ecology
Regression model relating biomass of a tree to its dbh. ln(AGTB) = alpha + b*ln(D) Standard deviation = 0.27 ln(AGTB) = -2.00 +/- 0.27 + 2.42 ln(D). in kg Above the size of several hectares the statistical variations in biomass estimates for the forest type are averaged out. Trees greater than 30 cm influenced average biomass a lot. They contribute about 80% of the biomass in two forests. The critical size of the sampling area can be related to the number of large trees in the sample. As a rule of thumb an area with more than 100 large trees (>= 70 cm dbh) gives a statistically averaged biomass estimate. Although biomass of lianas is generally believed to be small in disturbed areas such as liana forests small trees and lianas contribute more to the total biomass. Small fluctuations in soil nutrient content mainly affect floristic characteristics (Newbery and Proctor 1984) although they were recently claimed to account for as much as a third of the observed biomass variability (Laurance et al. 1999) in a large-scale experiment of the Brazilian lowland rain forest. Clear correlation between well-drained soils and large trees in various forests. Estimate of biomass based only on a 1-ha sampling unit would have led to an error far greater than the error resulting from the use of an allometric regression. Values suggest a biomass accumulation (NPP) of 2-4 Mg ha-1 y-1 and significantly more for secondary forests.