Impact of Assimilating WindBorne Observations Following Different Parts of a TPV on the Predictability of an Arctic Cyclone During THINICE

Abstract The impact of assimilating in‐situ observation inside a Tropopause Polar Vortex (TPV) from a novel weather balloon system, the WindBorne, on the predictability of the TPV and the coupled Arctic Cyclone (AC) is investigated using an AC case from the THINICE filed campaign. Two WindBornes continuously sampled inside and near the center of the TPV at various vertical levels for 27 hr before the cyclogenesis. The 27 hr were divided into three phases based on the different vertical levels sampled by WindBornes. The Root Mean Square Error (RMSE) of the forecasted cyclone is reduced from 12‐hr to 36‐hr forecast lead time as more phases of WindBornes are assimilated. This period corresponds to when the surface cyclone becomes superimposed with the TPV and rapidly deepens. Comparing the experiment with three phases of WindBorne observations assimilated versus the baseline, the cyclone expands southwestwards toward the TPV and forms a stronger coupled structure. Two analysis improvements leading to the improved AC predictability are revealed. First, a stronger circulation at and below the tropopause of the TPV is observed, leading to stronger coupling between the TPV and the surface cyclone in the forecast. Second, a mesoscale shortwave embedded in the synoptic trough is better constrained as more phases of WindBorne are assimilated. The north part of the TPV, which evolves from the shortwave trough, is therefore forecasted to be stronger. The sensitivity of forecast performance to horizontal localization parameter of the WindBorne observations is also studied. Larger localization parameters result in larger RMSE reduction compared to smaller values.