Oral

IAMAS

JMCP18 - Sub-seasonal to Decadal Prediction (S2S-S2D)

Robust Estimates of Earth System Predictability of the 1st kind using the CESM2 MultiyearPrediction System (CESM2-MP)

Here we present a new seasonal-to-multiyear Earth prediction system (CESM2-MP) based on the Community Earth System Model version 2 (CESM2). A 20-member ensemble which assimilates oceanic temperature and salinity anomalies provides the initial conditions for 5-year predictions from 1960 to 2020. We analyze skills using pairwise ensemble statistics, calculated among individual ensemble members (IM) and compare the results with the more commonly used ensemble mean (EM) approach. This comparison is motivated by the fact that an EM of a nonlinear dynamical system generates, unlike reality, a heavily smoothed trajectory, akin to a slow manifold evolution. However, for most autonomous nonlinear systems, the EM does not even represent a solution of the underlying physical equations, and it should, therefore, not be used as an estimate of the expected trajectory. The IM-based approach is less sensitive to the ensemble size than EM-based skill computations, and its estimates of potential predictability are closer to the actual skill. Using IM-based statistics helps to unravel the physics of predictability patterns in CESM-MP and their relationship to ocean-atmosphere-land interactions and climate modes. Furthermore, the IM-based method emphasizes potential predictability of the 1st kind which is associated with the propagation of the initial conditions. In contrast, the EM-based method is more sensitive to the predictability of the 2nd kind, which is associated with external forcing and time-varying boundary conditions. Calculating IM-based skills for the CESM-MP provides new insights into predictability sources due to ocean initial conditions and helps delineate and quantify forecast limits of internal variability.