Impacts of tropical forecast errors on two extreme precipitation events: insights from relaxation experiments using machine-learning weather prediction models

Abstract. This study explores the use of relaxation experiments in 2 machine learning-based weather prediction (MLWP) models to identify sources of subseasonal predictability in comparison to a traditional numerical weather prediction (NWP) system. Tropical relaxation involves nudging specific tropical regions of a model toward reanalysis data to isolate their influence on forecast skill. We apply this technique to Pangu-Weather (fully data-driven) and NeuralGCM (hybrid) and compare the experiments to the Unified Forecast System (UFS). The focus is on the week 3–4 forecast of 2 major precipitation events in western North America in winter 2022/2023, both linked to Madden–Julian Oscillation (MJO) activity. For the 2 cases, MLWP models exhibit higher forecast skill than the UFS at subseasonal lead times. Though tropical relaxation improves the skill in all forecast systems, gains are greater for UFS, reflecting the MLWP models' stronger baseline performance. A Rossby wave source (RWS) analysis shows that tropical relaxation consistently improves the large-scale dynamic processes associated with the tropical–extratropical teleconnections leading to both events. These results highlight the potential of relaxation experiments as an effective diagnostic for understanding and improving subseasonal forecasts, especially in emerging MLWP systems.