Regulatory air quality management as well as scientific understanding of air quality processes and models relies heavily on observation systems. Observations can be used through collective decision metrics, such as population health metrics. In this study, the adjoint of gas-phase CMAQ is implemented to relate ozone (O3) decision metrics (e.g. short-term ozone mortality) to the information value added by the observations at different locations across North America. Furthermore, a spatial interpolation (i.e. Kriging method) is conducted to account for uncertainties in the estimated
concentrations. Finally, maps of information content are produced to illustrate how observations of ozone at various locations can help constrain decision metrics.