The stratosphere at high latitudes exhibits an annual cycle dominated by the evolution of the stratospheric polar vortex. Most general circulation models show a significant bias in the estimation of the date of the vortex breakdown in the Southern Hemisphere, which may be attributed to missing gravity wave drag. We use a four dimensional variational data assimilation technique to estimate the missing gravity wave drag during the stratospheric polar vortex breakdown, using a middle atmosphere model and MERRA reanalysis as observations. The winter-summer transition is improved with data assimilation, and the gravity wave drag estimation is then used a posteriori to check the performance of the gravity wave drag parameterization. Important resolved waves-parameterized waves interactions are found during the vortex breakdown. Finally, a set of optimal parameters for the gravity wave drag parameterization is derived using a genetic algorithm, aiming to obtain gravity wave drag profiles which best resemble the estimated missing gravity wave drag. The estimated parameters largely reduced the biases in the representation of the polar vortex breakdown in the model and give insight on possible deficiencies of the gravity wave parameterization.

---

*email: pulido@unne.edu.ar
*Preference: Oral