Decadal Prediction and Predictability of Extremes in Ocean Eddy Resolving Coupled Models
Increasingly, high-resolution observations and coupled-model experiments with eddy-resolving oceans indicate that western boundary currents (WBCs) are regions of strong ocean-atmosphere interactions that are critical components of the climatic mean state and variability. The high sea-surface temperatures (SSTs) and strong SST gradients couple with the atmosphere to pump moisture into the marine-boundary layer, accelerate winds, sharpen SST fronts, and introduce significant decadal climate variability that affects the frequency and intensity of extreme events (e.g., heat waves, cold spells, droughts, floods, extreme winds) at remote locations. These extreme events are embedded within sub-seasonal to decadal variability that may be predicted by global models–this project seeks to determine this predictability in global coupled models that capture the essential feedbacks associated with ocean mesoscale features (e.g., eddies and WBC).
The proposed work leverages previous experience in predictability using an ocean eddy resolving version of CCSM4 to perform a suite of initialized retrospective decadal predictions experiments with ocean eddy resolving versions of CESM2 and E3SMv1. Similar retrospective forecasts with eddy parameterized versions will be used for comparison. The proposed research will have two main objectives: (i) we will investigate how the mesoscale features in the ocean feedback onto representation of known modes of decadal variability (e.g., AMO, PDO, among others); and (ii) diagnose how these known modes teleconnect to regional US extremes. The analysis will also include a detailed examination of sub-seasonal (e.g., MJO) variability, seasonal (e.g., ENSO) variability, and how their interactions impact regional US extremes in terms of intensity and occurrence. Essentially, we seek to diagnose how the predictability of decadal variability and the embedded extreme events are affected by oceanic mesoscale features and WBCs.
Specifically, we propose to perform decadal retrospective predictions with ocean eddy resolving version CESM2 (i.e., 0.1 degree for ocean and ice and 0.25 degree for atmosphere and land) and the high-resolution version of E3SMv1. The atmosphere-land-ice-ocean initialization will follow the same procedure that is used for preliminary ocean eddy resolving predictions with CCSM4; that is, scientists will use the historical oceanic and atmospheric reanalysis products (e.g., CFSR) to initialize the retrospective forecasts. The team has already competed a preliminary set of three ensemble member retrospective forecasts (January initial states from CFSR for 1980-2016) and demonstrated that not only does the initialization of CCSM4 work effectively, but the team also found an improved representation of regional imprint of global climate modes, particularly in costal zones. Similar initialization procedures will be implemented for E3SMv1, and we have extensive experience in applying our initialization using disparate reanalysis products to a diverse set of coupled models. The analysis of the forecasts will identify processes that are consistent across the two models and seek to understand the sources of differences between the two models.
This project is in progress.