Management of tropical soils as sinks or sources of atmospheric carbon

The prevailing paradigm for anticipating changes in soil organic carbon (SOC) with changes in land use postulates reductions in SOC in managed systems (agriculture and tree plantations) relative to mature tropical forests. Variations of this notion are used in carbon models to predict the role of tropical soils in the global carbon cycle. Invariably these models show tropical soils as sources of atmospheric carbon. We present data from a variety of studies that show that SOC in managed systems can be lower, the same as, or greater than mature tropical forests and that SOC can increase rapidly after the abandonment of agricultural fields. History of land use affects the comparison of SOC in managed and natural ecosystems. Our review of the literature also highlights the need for greater precautions when comparing SOC in mature tropical forests with that of managed ecosystems. Information on previous land use, bulk density, and consistency in sampling depth are some of the most common omissions in published studies. From comparable SOC data from a variety of tropical land uses we estimate that tropical soils can accumulate between 168 and 553 Tg C/yr. The greatest potential for carbon sequestration in tropical soils is in the forest fallows which cover some 250 million hectares. Increased attention to SOC by land managers can result in greater rates of carbon sequestration than predicted by current SOC models.