Investigating Arctic Cyclone-Tropopause Polar Vortex Interactions with Observing System Simulation Experiments

Arctic cyclones (ACs) play a key role in transporting heat and moisture into the Arctic and forcing sea ice extent variability on synoptic to sub-seasonal time scales. Upper-level forcing for AC intensification and maintenance is commonly provided by tropopause polar vortices (TPVs), often long-lived potential vorticity (PV) anomalies in the upper troposphere and lower stratosphere (UTLS), which are common in the Arctic. Although the link between these two features has been established via case studies and modeling experiments, questions remain about the exact role of TPVs and associated PV features in AC development and medium-range predictability. In the current study, we use observing system simulation experiments (OSSEs) with highly idealized simulated dropsondes of a coupled AC-TPV system to explore the physical mechanisms that link the upper and lower-level features. We take the ECMWF Cubic Octahedral grid Nature Run (ECO1280) as the truth and use the Model for Prediction Across Scales (MPAS) coupled with the Data Assimilation Research Testbed (DART) ensemble adjustment Kalman filter to run experiments with several sets of simulated observations. We expect that the system will be highly sensitive to additional moisture observations in the UTLS region and that the modeled development of the AC will be connected to the mesoscale PV structures around the TPV.

In addition to a control, four main experiments are conducted, assimilating observations of (1) temperature, (2) humidity, (3) temperature and humidity, and (4) temperature and humidity over a broader region with sparser coverage. Each of the experiments reduces errors throughout the troposphere and at the surface relative to the control, with the fourth performing the best overall. Additional humidity observations lead to greater PV generation in the lower stratosphere and a slightly deeper surface cyclone. Experiments with additional temperature observations exhibit notable improvements in the representation of TPV structure and surrounding PV features. Those experiments, in turn, produce stronger surface cyclones and skillful forecasts of the TPV for up to 36 hours longer than the control.