Spatial distribution of forest biomass in Brazil's state of Roraima, northern Amazonia

Forest biomass is an important variable for calculating carbon stocks and greenhouse gas emissions from deforestation and forest fires in Brazilian Amazonia. Its spatial distribution has caused controversy due to disagreements over the application of different calculation methodologies. Standardized networks of forest surveys provide an alternative to solve this problem. This study models the spatial distribution and original total stock of forest biomass (Aboveground + Belowground + Fine and coarse litter) in Brazil's state of Roraima, taking advantage of data from georeferenced forest surveys in the region. Commercial volume (bole volume) from surveys was expanded to total biomass. Kriging techniques were used to model the spatial distribution of biomass stocks and generate a benchmark map. All results were associated with phytophysiognomic groups, climatic regions and land uses (protected areas; agricultural use). We estimate forest in the state of Roraima to have an original biomass stock of 6.32 x 10(9) Mg. Forest biomasses in areas with shorter dry seasons were higher as compared to forests in regions with longer dry seasons. The original vegetation in protected areas, independent of phytophysiognomic group, has higher biomass compared to areas currently under agricultural use. Protected areas support 65.8% of Roraima's stock of forest biomass, indicating an important potential role in REDD projects for conservation of forest carbon. Information on spatial distribution of biomass stocks at a more refined scale is needed to reduce uncertainties about the regional character of carbon pools in Amazonia. (C) 2016 The Authors. Published by Elsevier B.V.