Mission CO₂ntrol: A Statistical Scientist's Role in Remote Sensing of Atmospheric Carbon Dioxide

Too much carbon dioxide (CO₂) in the atmosphere is a threat to long-term sustainability of Earth's ecosystem. Atmospheric CO₂ is a leading greenhouse gas that has increased to levels not seen since the middle Pliocene (approximately 3.6 million years ago). One of the US National Aeronautics Space Administration's (NASA) remote sensing missions is the Orbiting Carbon Observatory-2, whose principal science objective is to estimate the global geographic distribution of CO₂ sources and sinks at Earth's surface, through time. This starts with raw radiances (Level 1), moves on to retrievals of the atmospheric state (Level 2), from which maps of gap-filled and de-noised geophysical variables and their uncertainties are made (Level 3). With the aid of a model of transport in the atmosphere, CO₂ fluxes (Level 4) can be obtained from Level 2 data directly or possibly through Level 3. Decisions about how to mitigate or manage CO₂ could be thought of as Level 5. Hierarchical statistical modeling is used to qualify and quantify the uncertainties at each level. Supplementary materials for this article are available online.