Fields To Continents: Issues Of Scale and Uncertainty For Modeling Soil GHG Emissions and Mitigation.

Emissions of GHGs from soils are widely dispersed, spatially and temporally variable and influenced by a host of environmental and management factors – and thus accurate quantification is inherently difficult.  Sources of uncertainty for model-based estimates include uncertainties in process understanding, environmental driving variables, as well as management activity data.

Most ecosystem models used to simulate GHG fluxes are ‘field-scale centric’ in terms of their representation of key processes and drivers.  Improvements in predictions are most limited by fundamental process understanding but also measurement limitations to capturing spatial and temporal variation are field scales are an important issue.  Here we provide some examples of using a diversity of long-term field experiments and static vs continuous flux measurements to access and improve the DayCent simulation model predictions of soil GHG emissions and removals.

Modeling soil GHG emissions at regional and continental scales generally involves ‘upscaling’ of our field-centric models, either as points or grids, by accounting for the spatial and temporal distribution of soil and environmental conditions, and land management practices, across this larger area.  Examples from DayCent modeling of soil C and N2O fluxes for the US national greenhouse gas inventory are used to illustrate some of the major uncertainties and improvements needed for better continental-scale predictions.  Distributed monitoring networks that combine soil measurements with monitoring of land use and management practices over time are a key ingredient to better understanding and reduced uncertainty of continental-scale fluxes.