There is considerable interest in reforesting dryland farming systems (rainfall 300-600 mm/year) across Australia to protect soil, water supplies and biodiversity from the onset of salinity. It is estimated that up to 17 Mha of land will be affected by salinity by 2050, so rapid and extensive reforestation is required to tackle the problem. Profitable systems are required, so that these can be applied across large swathes of farmland using private rather than public investment. Carbon sequestration investment in farmland reforestation is seen as a very prospective method of financing this reforestation, however it is necessary to determine how much will occur and how this will vary with soil and site conditions. Similarly, understanding how recharge varies across farms and landscapes will allow the targeting of reforestation to areas where it is most needed. The study was undertaken in the agricultural region of the southwest of Western Australia. Existing soil survey and climatic data at the farm (1:10,000), watershed (1:250,000 ha) and regional (1:2,500,000) scales were combined with pedotransfer functions of tree suitability, yield and recharge reduction. Areas of land compliant with Kyoto Protocol Article 3.3 were identified. The analysis indicated broad areas where reforestation was suitable and those areas where reforestation was either unlikely to succeed or not required on the basis of small amounts of recharge to groundwater. For those areas suitable for reforestation, the potential amount of carbon sequestration was estimated from climatic data, with these data also used to indicate the risk from drought. This approach allows the use of existing data-sets and expert knowledge to make assessments at different scales of the suitability of land for reforestation and carbon sequestration potential. We suggest that the approach can be extended to other regions, where similar data are available, and for other species. Results at the watershed and regional scales are useful as a synoptic planning tool, rather than for making decisions about changes in practice within a particular farm. Often soil data are not available at the farm scale and these will have to be collected, either with traditional techniques of soil survey or with new techniques such as gamma radiometry.