Abstract. Fires associated with land use and land cover changes release into the atmosphere large amounts of aerosols and trace gases. Although there are several inventories of biomass burning emissions covering Brazil, there are still considerable uncertainties and differences among these. While most fire emissions inventories still utilize the parameters of vegetation fuel load, emission factors and other parameters to estimate the biomass burned and its associated emissions, certain more recent inventories tend to apply an alternative method based on fire radiative power (FRP) observations to estimate the amount of biomass burned and the corresponding emissions of trace gases and aerosols. The Brazilian Biomass Burning Emission Model (3BEM) and Fire Inventory from NCAR (FINN) are examples of the first, while Brazilian Biomass Burning Emission Model with FRP assimilation (3BEM_FRP) and Global Fire Assimilation System (GFAS) are examples of the latter method mentioned. In this paper, the output of four biomass burning emission inventories used during the South American Biomass Burning Analysis (SAMBBA) field campaign are analyzed and intercompared, focusing on eight predefined grids. Aerosol optical thickness derived from measurements made by the MODIS sensor operating onboard the Aqua satellite is applied to assess the inventories consistency. Significant correlation coefficients (r, p>0.05 level, Student t-test) were found between 3BEM and FINN, and between 3BEM_FRP and GFAS, with approximately 0.86 and 0.85, respectively. These results indicate that emissions estimates in this region derived via similar methods tend to agree with one other, but differ more from the estimates derived via the alternative approach. However, correlations in specific grids indicate that 3BEM and FINN typically underestimated the smoke emission loading in the eastern region of Amazon Forest, whilst 3BEM_FRP presents a tendency to overestimate fire emissions in the same area. The relationship between the 3BEM and FINN fire inventories present a correlation coefficient of 0.75-0.92, with a tendency of FINN to overestimate the emission of carbon monoxide by 20-30%. Moreover, 3BEM and GFAS shows a correlation coefficient of between 0.75-0.85, with higher values near the arc of deforestation in Amazon Rainforest. However, GFAS has a tendency to present higher carbon monoxide emissions in this region, and 3BEM_FRP tends to overestimate the emissions in soybean expansion (east of Amazon forest).