Back to CDIAC Home Page
Global Patterns of Carbon Dioxide Emissions from Soils on a 0.5 Degree Grid Cell Basis
Semi-mechanistic, empirically-based statistical models were used to predict the spatial (0.5 degree grid cells) and temporal (monthly and annual) patterns of global carbon emissions from terrestrial soils. Emissions include the respiration of both soil organisms and plant roots. Geographically referenced data of mean monthly air temperature and precipitation, soil organic carbon and nitrogen content, soil type and natural vegetation type were used in the model development. It was found that at the global scale the rates of soil CO2 efflux correlate significantly with temperature and precipitation, have a pronounced seasonal pattern in most locations, and contribute to observed wintertime increases in atmospheric CO2.
The DB1015 data are the predicted CO2 emissions of the model based on untransformed precipitation data and the exponential relationship between soil biological activity and temperature. The data at the 0.5° latitude by longitude spatial and monthly temporal resolution represents the best resolved estimate to date of global CO2 fluxes from soils and should facilitate investigations of net exchanges between the atmosphere and terrestrial biosphere.
The data files (~10 MB total) consist of 12 monthly and one annual model output files, one geographic information file and a Read-Me file with simple FORTRAN data access information.
Back to CDIAC Home Page