Investigating Inductive Biases for Machine Learning Emulation of Sudden Stratospheric Warmings in Idealised Isca Simulations
This work addresses the challenge of exploiting stratospheric predictability anchors for subseasonal-to-seasonal forecasting, but findings are incremental as they confirm known needs for vertical coupling and highlight diagnostic limitations.
The study tested how architectural inductive bias affects machine-learning emulation of sudden stratospheric warming (SSW) dynamics using idealised Isca simulations. Results show that explicit three-dimensional vertical coupling is a key bias for emulating stratospheric dynamics, but low forecast error does not guarantee physically faithful wave-mean-flow interaction.
Machine-learning emulators are increasingly used for weather prediction and have the potential to extend skill on subseasonal-to-seasonal timescales by learning dynamically important sources of predictability. A key challenge is whether the models can exploit predictability anchors, such as stratospheric variability, that influence tropospheric circulation beyond short lead times. We test how architectural inductive bias affects emulation of sudden stratospheric warming (SSW) dynamics using paired idealised Isca simulations that differ only in an imposed wave-2 heating perturbation. Across convolutional, transformer, and graph-based architectures trained for one-step prediction, model differences are modest when the stratosphere is dynamically quiet but widen substantially when SSW-like variability is active. Our results identify explicit three-dimensional vertical coupling as a key inductive bias for machine-learning emulation of stratospheric dynamics. However, Eliassen-Palm flux diagnostics show that low forecast error does not guarantee physically faithful wave-mean-flow interaction, with coherent errors remaining in stratospheric wave-driving structure.