Storing carbon in natural ecosystems, particularly forests, is key to reducing human impact on global climate. In terrestrial ecosystems, about four times as much carbon is stored in soil, rather than in trees and plants. Although critically important, soil carbon dynamics are much more difficult to understand, assess, and monitor than aboveground dynamics. Despite these obstacles, since soil carbon stocks are likely to be disturbed during forest management, it is important to consider changes in soil carbon in forest carbon offset projects. For this reason, the Climate Action Reserve, with EcoShift’s support, has taken the bold step of including soil carbon dynamics in the latest version of its Forest Project Protocol. Version 3.3 of the Forest Project Protocol, still in draft form, includes lookup tables, based on soil type, management practice, and other factors, to easily calculate changes in soil carbon in forest offset projects. The Reserve is now in the process of finalizing the new version. Given scientific uncertainty and lack of comprehensive research, creating the methodology for including soil carbon was not an easy task, and the process started over two years ago. To initiate the effort, the Reserve contracted EcoShift to examine potential methods for accounting for soil carbon in the Forest Project Protocol. This first stage of the process resulted in a comprehensive white paper identifying the aspects of forest management and site conditions that could result in soil carbon losses. This white paper, Accounting for Carbon in Soils, also sought to develop a comprehensive set of recommendations to minimize the losses of, and in some cases increase, soil carbon in forests managed under CAR’s Forest Project Protocol. In the second stage of the process, EcoShift designed a lookup table and associated methodology to determine which types of management were likely to result in soil carbon losses at what time intervals, and to identify potential risk factors that would suggest significant potential for soil carbon loss. The updated Climate Action Reserve guidance for Soil Carbon Accounting in avoided conversion projects that incorporates EcoShift models can be found here.CAR used this information to create the draft of Version 3.3 of the protocol, which can be found here. The biggest challenge in developing models for changes in soil carbon is the fact that data for soil carbon dynamics in different systems remain sparse, and decisions often have to be made with incomplete information. In all cases, the Reserve and EcoShift used a conservative approach to assigning values in the lookup tables, and this approach ensures that actual carbon stored is likely greater that what is shown on the lookup table, rather than less. As better and more comprehensive research becomes available, we will have a better understanding of soil carbon dynamics, and this knowledge can then be incorporated into future versions of protocol. Although current scientific understanding is far from complete, continuing to ignore soil carbon dynamics in offset calculations carries a risk of gross inaccuracies – in both in additional gains and losses – in the largest terrestrial carbon pool.