Quantification and Decision Support Tools for Greenhouse Gas Mitigation in Agriculture.

For agriculture to participate in greenhouse gas mitigation policies, adoption of mitigation practices need to provide real mitigation benefits that are valued equally to other GHG mitigation activities.  Thus they need robust quantification with known uncertainty and issues such as additionality, permanence, leakage and ensured net GHG reductions need to be dealt with effectively.  For implementing national policies, the science and information need to support effective policy is partially in place, although important research and development areas remain. Combining the attributes of both direct measurement and modeling for estimating soil C stock changes and non-CO2 emissions provides the most flexible and cost-effective alternative for measurement and monitoring.  Combining information from existing research sites, validated models and a coordinated set of on-farm measurements, would enables practice-based monitoring that can utilize low cost monitoring methods (e.g. remote sensing, existing surveys).  Process-based models are capable of providing local and region-specific estimates that consider climate, soil and previous land use history and thus provide more accurate estimates than broad national or regional single factor stock change estimates.  However, sufficient direct measurements are needed to determine model uncertainties and to reduce them to acceptable levels. A critical need is for a coordinated national sampling network where direct measurements of soil C stocks can be taken over time to provide the necessary ground-truth for accurate estimation with low uncertainty.  Such a network should encompass all the major ecozones and management systems in the US and be designed for long-term viability.  Maintenance of strict protocols for sampling and analysis and for permanent archiving of samples is essential.  Data on land use and management practices over time are needed for attribution of stock changes and thus there are major advantages to incorporating a measurement network within the existing landuse inventory systems such as USDAs National Resource Inventory.