The goal of the project is to improve understanding of the mechanisms driving carbon exchange processes and associated uncertainties on the regional scale by (1) maintaining and extending a ground-based network of CO2 observation sites in Oregon and aircraft sampling of boundary layer profiles, and (2) continuing development and application of a high-resolution data assimilation system with regional atmospheric inverse modeling. The ultimate objective is to produce regionally representative atmospheric CO2 observations and terrestrial fluxes to understand causes of variability at different temporal scales. The work was was previously undertaken as part of an NACP project, with five CO2 observation sites operated continuously since 2006. Under this project one of the existing sites will be relocated, and add two new sites will be added, extending the network to a total of seven sites. New mountain/ridge-top locations have been selected to generate a combined footprint that covers most of Oregon, and form a �??ring of towers�?? with the Metolius ridge-top site at the center to improve our ability to track the CO2 changes of air masses passing the model domain. One new site is in the convergence/divergence zone between Cape Mendocino, CA and Newport, OR that combined with the Mary�??s Peak site will provide data on incoming marine air. Tower measurements will be supplemented by proposed aircraft campaigns that are linked with existing NOAA aircraft campaigns in Northern California to characterize atmospheric mixing conditions. The atmospheric inverse modeling framework uses the CO2 mole fraction observations to train and validate a terrestrial CO2 flux model. The biophysics-biogeochemistry model CLM-CN in the modeling framework will be used to improve representation of terrestrial carbon fluxes.