Tropical forests have great potential for mitigating global warming due to atmospheric CO₂ emissions. However, this potential is impaired by deforestation and forest degradation, which contributes about one fifth of total anthropogenic CO₂ emissions. Avoiding deforestation, afforesting degraded lands and proper management of forest resources are necessary to tackle problems of climate change and global warming. This requires adopting land use changes to those that are ‘forest friendly’ or stop further disturbance of forest resources. However, developing countries have little economic or policy incentives to adopt land uses that counter emissions. Against this background, a recent meeting at Kyoto devised a mechanism of compensating developing regions for the amount of carbon they sequester. To implement this, knowledge of the extent of carbon pool is necessary. This requires cutting large number of trees and measuring biomass to get quantitative data on carbon storage and establish baseline information for future reference. However, measuring biomass is difficult in the tropics where diverse vegetation species exist. Accessibility and availability of resources also prohibit undertaking extensive survey and measure biomass. As a result, allometric equations are interesting alternatives to estimate aboveground biomass (AGB) on wider geographical scale. We used dataset collected by the Africa Soil Information Service (AfSIS) project to estimate aboveground carbon stock of trees. Key tree attribute data such as diameter at breast height and tree height were measured for over 29000 trees in 15 randomly selected sites of Southern Africa. Ten generalized multispecies and species-specific allometric equations were used to estimate AGB of miombo woodlands. The results show that miombo woodlands have an average 35 kg C tree^-1. Statistical analyses of AGB vis-à-vis different environmental attributes reveal key determinants of AGB in the region to be cultivation, prevalence of erosion, and slope.