Comparing Noise Models for Simulating Westerly Wind Bursts in ENSO
Published:Dec 27, 2025 21:44
•1 min read
•ArXiv
Analysis
This paper investigates different noise models to represent westerly wind bursts (WWBs) within a recharge oscillator model of ENSO. It highlights the limitations of the commonly used Gaussian noise and proposes Conditional Additive and Multiplicative (CAM) noise as a better alternative, particularly for capturing the sporadic nature of WWBs and the asymmetry between El Niño and La Niña events. The paper's significance lies in its potential to improve the accuracy of ENSO models by better representing the influence of WWBs on sea surface temperature (SST) dynamics.
Key Takeaways
- •Gaussian noise, commonly used to represent WWBs, has limitations in capturing their characteristics.
- •CAM noise offers a more realistic representation of WWBs, including their sporadic nature and the asymmetry between El Niño and La Niña.
- •A conditional noise model, combining additive Gaussian and CAM noise, is proposed to better model the full spectrum of warm events.
Reference
“CAM noise leads to an asymmetry between El Niño and La Niña events without the need for deterministic nonlinearities.”