Evaluation of constraint provided by current atmospheric monitoring network for quantifying anthropogenic emissions and biospheric carbon fluxes (Invited)

Predicting future changes to the global carbon cycle (and therefore climate), as well as quantifying anthropogenic emissions of carbon dioxide, requires an understanding of carbon sources and sinks, and their variability, across a variety of spatial and temporal scales. Atmospheric observations have historically been used to quantify biospheric and oceanic carbon fluxes under the assumption that anthropogenic emissions are well known. Increasingly, however, questions are being asked about the information content of these atmospheric observations for quantifying anthropogenic emissions and their associated uncertainties. These questions are particularly relevant for climate treaty monitoring. Because the natural components of the carbon cycle still have large uncertainties, the ability to answer these questions ultimately depends on the degree to which atmospheric observations can be used to not only quantify the total CO2 flux (including anthropogenic, biospheric, and oceanic fluxes), but also the extent to which these estimated fluxes can be disaggregated to provide information about the individual flux components. This presentation will use recent work focusing on applying the geostatistical inverse modeling (GIM) approach to constraining the North American carbon cycle and CO2 fluxes to inform these questions. The GIM approach is used to estimate the total CO2 flux at a high spatial and temporal resolution. The inferred relationships between estimated fluxes and ancillary environmental and socioeconomic data are used to disaggregate anthropogenic and biospheric fluxes. The analysis is conducted within a framework that consistently assesses and tracks the uncertainty associated with each element of the analysis. The impact of the expanding North American monitoring network is also assessed.