Influence of gravity wave drag parametrisations on the stratospheric circulation of seasonal ICON-NWP experiments

The stratospheric polar vortex is highly variable in winter and thus, models often struggle to capture its variability and strength. Yet, the influence of the stratosphere on the tropospheric circulation becomes highly important in Northern Hemisphere winter and is one of the main potential sources for subseasonal to seasonal prediction skill in mid latitudes. Mid-latitude extreme weather patterns in winter are often preceded by sudden stratospheric warmings (SSWs), which are the strongest manifestation of the coupling between stratosphere and troposphere. Misrepresentation of the SSW-frequency and stratospheric biases in models can therefore also cause biases in the troposphere.

In this context this work comprises the analysis of four seasonal ensemble experiments with a high-resolution, nonhydrostatic global atmospheric general circulation model in numerical weather prediction mode (ICON-NWP). The main focus thereby lies on the variability and strength of the stratospheric polar vortex. We identified the gravity wave drag parametrisations as one important factor influencing stratospheric dynamics. As the control experiment with default gravity wave drag settings exhibits an overestimated amount of SSWs and a weak stratospheric polar vortex, three sensitivity experiments with adjusted drag parametrisations were generated. Hence, the parametrisations for the non-orographic gravity wave drag and the subgrid‐scale orographic (SSO) drag were chosen with the goal of strengthening the stratospheric polar vortex. Biases to ERA-Interim are reduced with both adjustments, especially in high latitudes. Whereas the positive effect of the reduced non-orographic gravity wave drag is strongest in the mid-stratosphere in winter, the adjusted SSO-scheme primarily affects the troposphere by reducing mean sea level pressure biases in all months. A fourth experiment using both adjustments exhibits improvements in the troposphere and stratosphere. Although the stratospheric polar vortex in winter is strengthened in all sensitivity experiments, it is still simulated too weak compared to ERA-Interim. Further mechanisms causing this weakness are also investigated in this study.