Above ground biomass (AGB) is increasing in most of the Amazon forests. One hypothesis is that forests are responding to wide-spread and intense human intervention prior to the European conquest (>450 years ago). In this study we confront this hypothesis with changes in AGB over 6.3 years in a large western Amazonian forest plot (>150,000 shrubs and trees and 1100 species with dbh ≥ 10 mm in 25 ha). We examined AGB flux in different habitats and across diameter classes. 

Results/Conclusions

The forest lost small stems (by 4.6% over 6.3 years), gained large trees (2.6%), and gained biomass (0.7%). The change in AGB stock was due entirely to an upward shift in size, leading to more canopy trees and fewer saplings after just 6 years. Across habitats, the biggest increment in biomass was in a secondary forest patch (3.4% y^-1), which we know was cleared about 27 years ago, whereas mature forest on ridges and valleys had small increases (0.10 and 0.09% y^-1, respectively). In both censuses, AGB stocks were >50% higher on the ridge than in the valley, but relative growth and mortality were higher in the valley. Mean wood specific gravity (WSG) decreased with increasing diameter class; WSG changed very slightly and did not contribute to the change in AGB stocks. The forest increased its standing biomass, but far less than the average reported for other Amazonian forests (i.e., 0.31 vs. 0.98 Mg ha y^-1, Baker et al. 2004). We find no evidence to support the notion that the forest is recovering from long-past human intervention. Instead, we believe the forest changed in response to natural fluctuations of the environment (e.g., changes in precipitation, higher CO_²), windstorms, or other recent disturbances. The significant differences in AGB stocks between valley and ridge suggest that the terra firme forests are a mosaic of natural habitats, and that this mosaic is in part responsible for the variation in biomass stocks detected in Amazonian forests.