Stratospheric ozone has been shown to influence stratospheric variability and subseasonal-to-seasonal (S2S) prediction through its strong radiative effects. This talk presents research that analyses and quantifies the impact of interactive ozone on high-latitude Southern Hemisphere stratospheric variability during the vortex breakdown period. The analysis is applied to two seasonal hindcast ensembles generated with the ECMWF IFS model: one with interactive ozone and one with prescribed ozone. The analysis focuses on wave amplitudes, defined as longitudinal deviations from the zonal mean, and on zonal-mean deviations from the ensemble mean, for both temperature and zonal wind. The effect is quantified as a function of day of year to account for the strong nonstationarity during this period, with particular emphasis on the lower stratosphere, a region important for stratosphere-troposphere coupling. For both wave and zonal-mean variability, interactive ozone provides a positive radiative feedback on variability, increasing the variances of both the waves and the zonal-mean deviations from the ensemble mean.

This study trains three machine learning models with varying complexity—Random Forest, Support Vector Machine, and Neural Network—to predict cyclogenesis at a forecast lead time of 24 hr for given tropical disturbances identified by an optimized Kalman Filter algorithm. The overall performance is competent in terms of f1-scores (∼0.8) compared to previous research of the same kind. An assessment by SHapley Additive exPlanations (SHAP) values reveals that mid-level (500 hPa) vorticity is the most influential factor in deciding if a tropical disturbance is developing or non-developing for all three models. Wind shear and tilting are found to hold a certain level of importance as well. These results encourage further experiments that use physical models to explore the dynamical, mid-level pathway to tropical cyclogenesis. Another usage of SHAP values in this work is to explain how a machine learning model decides if an individual tropical disturbance case will develop, by listing the contribution of each feature to the output genesis probability, illustrated by a case study of Typhoon Halong. This increases the reliability of the machine learning models, and forecasters can take advantage of such information to issue tropical cyclone formation warnings more accurately. Several caveats of the current machine learning application in the studies of tropical cyclogenesis are discussed and can be considered for future research. These can benefit the interpretation and emphasis of certain output fields in the operational dynamical prediction system, which can contribute to more timely cyclogenesis forecasts.