Skip to main content

A process system approach for addressing climate change uncertainties

Christina
McCluskey
NSF NCAR
Talk
(Invited)
Aerosol-cloud interactions (ACI) are a key uncertainty in future climate projections. In this presentation, we will discuss the wealth of knowledge that can be gained by intentional integration of field observations and numerical modeling to address knowledge gaps in specific ACI process systems. 

Mixed phase ACI occurring over the Southern Ocean (SO) region will be discussed as an illustration of the need for considering the full process system. Efforts to assess the Community Earth System Model (CESM) against field campaign observations revealed a number of important CESM biases that help explain conflicting model findings on the role of ice nucleation in simulated Southern Ocean clouds. Specifically, results highlight that the role of ice nucleation in SO cloud phase and cloud feedbacks is uncertain and strongly dependent on simulated aerosol availability, cloud droplet and cloud condensation nuclei number concentrations, precipitation processes, and ice processes. Improved model physics are needed to determine future SO cloud phase and feedbacks, but rely on process-oriented observations that include Lagrangian sampling of process rates and capture so-called “anchor” metrics for assessing multiple components of the process system.  

Finally, this presentation will include an introduction to a new NSF NCAR strategic initiative to INtegrating Field Observations and Research Models (INFORM), where we aim to expand the current top of atmosphere approach for assessing CESM to include process-scale diagnostics. The outcomes of INFORM will serve a two-way benefit to the modeling and observation communities by 1) levering field campaign data and establishing best practices for assessing models, 2) providing community ready tools for field campaign design and execution, and 3) creating a research infrastructure focused on addressing process system uncertainties for improving Earth system predictability.