A methodology is presented to construct supply curves and cost–supply curves for carbon plantations based on land-use scenarios from the Integrated Model to Assess the Global Environment (IMAGE 2). A sensitivity analysis for assessing which factors are most important in shaping these curves is also presented. In the IPCC SRES B2 Scenario, the carbon sequestration potential on abandoned agricultural land increases from 60 MtC/year in 2010 to 2,700 MtC/year in 2100 for prices up to 1,000 $/tC, assuming harvest when the mean annual increment decreases and assuming no environmental, economical or political barriers in the implementation-phase. Taking these barriers into consideration would reduce the potential by at least 60%. On the other hand, the potential will increase 55 to 75% if plantations on harvested timberland are considered. Taking into account land and establishment costs, the largest part of the potential up to 2025 can be supplied below 100 $/tC, assuming harvest when the mean annual increment decreases and assuming no environmental, economical or political barriers in the implementation-phase. Taking these barriers into consideration would reduce the potential by at least 60%. On the other hand, the potential will increase 55 to 75% if plantations on harvested timberland are considered. Taking into account land and establishment costs, the largest part of the potential up to 2025 can be supplied below 100 /tC (In this article all dollar values are in US dollars of 1995, unless indicated otherwise.). Beyond 2050, more than 50% of the costs come to over 200 $/tC. Compared to other mitigation options, this is relative cheap. So a large part of the potential will likely be used in an overall mitigation strategy. However, since huge emission reductions are probably needed, the relative contribution of plantations will be low (around 3%). The largest source of uncertainty with respect to both potentials and costs is the growth rate of plantations compared to the natural vegetation.