Toward Skillful Forecasting of Super El Niño Events Using a Diffusion-Based Westerly Wind Burst Parameterization

<p><strong>Citation:</strong> Ji, C., Mu, M., Qin, B., Lian, T., Yuan, S., Feng, J., Song, S., Wei, Y., Dai, G., Wang, J., &amp; Fang, X. (2025). Toward skillful forecasting of super El Niño events using a diffusion-based westerly wind burst parameterization. <em>npj Climate and Atmospheric Science</em> (Published in partnership with CECCR at King Abdulaziz University). https://doi.org/10.1038/s41612-025-01158-x</p><p><strong>Key Takeaways:</strong></p><ul><li><strong>Innovative Generative AI Parameterization:</strong> The study introduces a state-of-the-art <strong>Denoising Diffusion Probabilistic Model (DDPM)</strong> to parameterize westerly wind bursts (WWBs). These wind bursts are critical, episodic atmospheric events that inject wind energy into the Pacific, playing a pivotal role in triggering super El Niños. This new generative AI framework successfully captures the complex, <strong>joint modulation of wind bursts by both slow-varying oceanic states and rapid atmospheric processes</strong>.</li><li><strong>Superior Representation of Wind Burst Physics:</strong> Traditional schemes rely heavily on ocean-state indicators like the warm pool eastern edge, which fails to capture high-frequency atmospheric noise. By incorporating multiple physical conditions—<strong>Sea Surface Temperature Anomalies (SSTA), Outgoing Longwave Radiation Anomalies (OLRA), and Sea Level Pressure Anomalies (SLPA)</strong>—the DDPM-based scheme dramatically improves the simulated frequency, intensity, duration, and spatial distribution of wind bursts compared to observational data.</li><li><strong>Drastic Improvements in Super El Niño Intensity Predictions:</strong> When coupled online with the <strong>Community Earth System Model (CESM)</strong>, the DDPM scheme significantly outperforms both standard control runs and traditional parameterization schemes. It accurately predicts the absolute amplitude of historic super El Niño events—specifically the <strong>1982/83, 1997/98, and 2015/16</strong> events—by correcting the severe underestimations found in baseline climate models.</li><li><strong>Mitigation of Seasonal Phase-Locking Bias:</strong> A persistent challenge in climate modeling is &quot;seasonal phase-locking&quot; prediction bias, where models incorrectly project a double-peak warming cycle (peaking in summer, weakening, then re-intensifying in winter). The DDPM scheme overcomes this issue by generating <strong>stronger and more realistically eastward-shifted wind stress anomalies</strong>, which correctly trigger the positive dynamical feedbacks (such as the Bjerknes feedback) necessary to sustain a steady, natural warming progression toward a single December peak.</li></ul><p><br></p><p><br></p>