Investigation of the Causes for Systematic Breakdown of Numerical Subseasonal Forecasts

“A recent analysis of the THORPEX Interactive Grand Ensemble (TIGGE) showed that while the operational global ensemble forecast systems of the world’s leading numerical weather prediction centers were efficient, in general, in capturing the uncertainty dynamics associated with the high-frequency (synoptic scale) transients, they all predicted the slowly varying large-scale component of the flow with a systematic error whose magnitude increased with the forecast time. Such a systematic error poses a major obstacle to extending skillful forecasting into the subseasonal to seasonal (S2S) forecast range. The fact that the different ensemble forecast systems, which use different models and are also generated differently, all fail in the same general fashion, suggests that there may be one or more important dynamical processes that are not accounted for in the current forecast models. Our goal is to investigate the possibility that ocean mesoscale eddy-atmosphere (OME-A) feedback from the ocean to the atmosphere is such a process. This goal is driven by the hypothesis that oceanic mesoscale eddies, which can persist for months in the western boundary current regimes, may play an important role in S2S predictability through modulating the midlatitude storm tracks by inducing mesoscale SST variability.

We will carry out both deterministic and ensemble forecast experiments with a global atmospheric model (the NCAR CAM model at 1/4° resolution) coupled to a simple thermodynamic ocean model. The focus will be on representing the OME-A feedback (and the uncertainty in the OME-A feedback) from oceanic fronts and eddies in the mid-latitude storm track regions. The effect of the parameterization of the OME-A feedback on the synoptic scale waves in the global model simulations will be validated against the results of previous high (9 km atmospheric and 3 km oceanic) resolution process studies with a regional coupled model. We will analyze the results of the forecast experiments by both detailed computations of the energy conversion processes and ensemble-based diagnostics.

The results of the proposed research project are likely to improve the understanding of predictability of phenomena occurring at the S2S time scales. The advances made are expected to lead to improved prediction and better understanding of such S2S process as the transitions between the two phases of NAO and a blocked flow regime.”