Investigating the Downstream Impacts of TPVs on Predictability Using MPAS/MPAS-DART

The prediction of a particular feature called a Tropopause Polar Vortex (TPV) has downstream implications on larger-scale atmospheric evolution and forecast skill. The TPV is a feature found in the Arctic that can persist for many days before ultimately exerting a major impact on weather forecasts over North America. A forecast model’s ability to represent the dynamical and physical features of a TPV leads to considerable uncertainty in forecasts. The extended-range predictability of weather events can be particularly sensitive to the initial condition representation of TPVs, which derives partially in how a model represents physical processes related to TPVs in short-term forecasts. This talk will investigate the effect of TPVs on downstream predictability using the Model for Prediction Across Scales (MPAS). MPAS is a global, non-hydrostatic atmospheric model that offers smooth grid refinement down to higher resolutions. Since MPAS is a global model, forecast uncertainties from the treatment of lateral boundary conditions are eliminated. The use of data assimilation will be used with MPAS to quantify TPV sensitivities, and will be achieved by coupling MPAS with the Data Assimilation Research Testbed (DART) ensemble Kalman filter (EnKF).

Ensemble data assimilation is used here as a tool to investigate forecast sensitivity of TPV characteristics to an extreme weather event that was poorly forecast. This presentation will investigate the potential effects that multiple TPVs had in a case of a major forecast failure. Results show that forecast spread rapidly increases in association with an equatorward-moving TPV in the Canadian Arctic that interacts with the North Atlantic jet stream. Discussion will be focus on the ensemble sensitivity to interaction of multiple TPVs interacting with the jet stream over the course of a 5-day forecast. MPAS is initialized with analyses from the Global Ensemble Forecasting System (GEFS), creating a 21-member ensemble of MPAS forecasts that are integrated on a uniform 45-km mesh and a 60-15 km variable resolution mesh. The use of MPAS’s grid refinement feature, with higher-resolution over the TPV, allows for a better representation and evolution of TPVs.