Atmosphere Ocean Science Friday Seminar

---

Using Explainable AI and Transfer Learning to understand and predict the maintenance of Atlantic blocking with limited observational data

Speaker: Huan Zhang

Location: Warren Weaver Hall 1314

Date: Friday, March 8, 2024, 4 p.m.

Synopsis:

Blocking events are an important cause of extreme weather, especially long-lasting blocking events that trap weather systems in place. The duration of blocking events is, however, systematically underestimated in climate models. Explainable Artificial Intelligence (XAI) refers to a class of data analysis methods that can help us identify the physical causes of prolonged blocking events, as well as diagnose model deficiencies. We first demonstrate such an approach on an idealized quasigeostrophic model developed by Marshall and Molteni (1993). We train a convolutional neural network (CNN), and subsequently, build a sparse predictive model for the persistence of Atlantic blocking, conditioned on an initial high-pressure anomaly. Shapley Additive ExPlanation (SHAP) analysis reveals that high-pressure anomalies in the American Southeast and North Atlantic, separated by a trough over Altantic Canada, contribute significantly to the prediction of sustained blocking events in the Atlantic region. This agrees with previous work that identified precursors in the same regions via wave train analysis. When we apply the same CNN to blockings in the ERA5 reanalysis data provided by ECMWF, there is insufficient data to accurately predict persistent blocks. We partially overcome this limitation, however, by pre-training the CNN on the plentiful data of the Marshall-Molteni model, and then using Transfer Learning (TL) to achieve better predictions than direct training. Our work demonstrates the power of machine learning methods to extract meaningful precursors of extreme weather events and achieve better prediction using limited observational data.