Tracers For Evaluating Computational Models Of Atmospheric Transport And Oxidation At Regional To Global Scales

Atmospheric tracers are effective tools for characterizing dispersion and for testing computational models of atmospheric transport. Atmospheric trace gas measurements are now used widely to infer geographical surface flux distributions. However, robust flux estimates critically rely on well-validated knowledge of atmospheric chemistry, loss processes and transport, without which we are not fully realizing the potential of atmospheric measurements collected on the ground, or from aircraft and satellites. This challenge has taken on renewed importance in the shadow of the Paris Agreement that will likely take advantage of atmospheric trace gas measurements to help improve national and global greenhouse gas emission budgets. We describe a wide range of existing and new potential atmospheric tracers for improving our understanding of atmospheric dispersion. We consider the investigation of atmospheric transport over two scales: (1) short-to-medium length scale (on the order of 1-1000 km) to improve our understanding of convection and boundary layer transport processes, and (2) hemisphere-to-global length scale (on the order of 1000-10,000 km), where large-scale mixing, cross hemisphere transport and stratosphere-troposphere exchange are important. Although we note the possibility of using "tracers of opportunity," our primary focus is on deliberate-release tracers, and we explore the use of cyclic perfluorocarbons, hydrofluorocarbons, hydrochlorofluorocarbons, hydrofluoroethers and novel tracers of deuterium-substituted halocarbons. We examine how we might exploit existing instrumentation already deployed at remote global monitoring sites as well as requirements for new instrumentation. To guide the discussion, we provide example scenarios for how experiments might be set up, covering regional to global spatial scales for the evaluation and improvement of atmospheric transport models. However, we stress that appropriate three-dimensional modelling studies and preliminary experiments would need to be carried out to determine the specific details of any real-world experiment.