Atmospheric Constraints on Physiological Processes from CO2 Inversions

        An improved understanding of the biological and biophysical controls on carbon exchange processes is crucial for understanding carbon-climate interactions.  We present the results of a relatively new approach to using atmospheric CO2 data to constrain parameters of carbon exchange processes.  CO2 concentration data collected at remote monitoring stations of the NOAA-CMDL network are combined with a biosphere model and an atmospheric transport model to perform a Bayesian synthesis inversion.  Rather than inferring regional-to-continental-scale net carbon fluxes as is typically done, we solve for the light-use efficiency (LUE) of net photosynthesis for broad ecosystem types (e.g. boreal biomes). Our inversion results suggest higher growing-season LUE values for Eurasian boreal regions than North American boreal regions at similar latitudes (north of 56 N). This is likely due to the greater abundance of deciduous tree cover (relative to coniferous tree cover) in Eurasia than N. America at these latitudes, since deciduous trees generally have higher LUEs than conifers. This longitudinal LUE asymmetry mirrors the results of interannual net flux inversions, which generally infer higher carbon sinks in Eurasian boreal latitudes than North American boreal latitudes. The LUE values returned from boreal and temperate latitudes on both continents are consistent with literature-derived values for major ecosystem types in these latitudes. These results should prove useful for terrestrial carbon cycle models, which often estimate LUE as a function of environmental variables like temperature and precipitation. The spread in these estimates is large, contributing the greatest uncertainty in global models of terrestrial production. Future research will include other tracers, liked13C andd18O, as well as assimilate eddy flux data directly in the inversion process.