Assessing The Impact of SST Anomalies on Polar Climate Using Global Teleconnection Operators from Multiple Models Uncertainties

Abstract:

The predictability of polar climate is limited by uncertainties in the given forcing, the response to this forcing, and the internal variability of the fully coupled climate system. Given these factors, we estimate how anomalous sea surface temperature (SST) patterns can influence polar climate and ultimately impact ice sheets and polar feedbacks. Using different versions of NCAR Community Atmospheric Model (i.e. CAM3.1, CAM3.5, CAM4.0, CAM5.0), we assess the capabilities of multiple atmospheric general circulation models (AGCMs) to respond to SST forced changes by perturbing SST fields that influence polar climate via atmospheric teleconnections. By decomposing uncertainties, we are able to address the impact of structural differences in climate models.

From large-ensembles of model simulations, we estimate the Global Teleconnection Operator (GTO) for each AGCM. The GTO is a linear approximation or empirical Green’s function and can be used to diagnose the sensitivities of polar climate to the boundary condition forcing from anomalous SSTs patterns. Primarily, the GTO identifies the ocean sectors where SST anomalies are effective at forcing polar climate response. To explore predictability issues, the multi-linear model is evaluated by comparing the linearly reconstructed response with both the results from the full non-linear coupled model and observations. We find that the multi-linear model can capture polar climate variability that the Coupled Model Intercomparison Project (CMIP5) simulations produce at seasonal scales for several polar regions in the near future. Overall, this approach provides a tool for exploring polar climate response as a first-order assessment of the climate variability being driven by SST forcings and the internal variability and model uncertainties by using large ensembles to estimate GTO. Furthermore, the uncertainty decomposition can help identify key directions where further research is required to improve predictive skill.