Towards accurate extreme event likelihoods from diffusion model climate emulators

2026-05-05

AI Summary

This work addresses the challenge of efficiently and accurately estimating the probability of climate extremes—such as tropical cyclones—under specified boundary conditions. To this end, we leverage the diffusion model “Climate in a Bottle” to generate atmospheric states consistent with prescribed sea surface temperatures and solar positions, and introduce a guidance mechanism to steer generation toward extreme events in targeted regions. For the first time, we integrate the diffusion model’s probability density estimation with importance sampling to quantify the likelihood of guided samples, substantially improving the accuracy of probability estimates. The method successfully computes the odds ratio of tropical cyclone occurrence under guided versus unguided conditions, dramatically reducing the standard error of Monte Carlo estimates and establishing a novel paradigm for extreme event attribution.

Abstract

ML climate model emulators are useful for scenario planning and adaptation, allowing for cost-efficient experimentation. Recently, the diffusion model Climate in a Bottle (cBottle) has been proposed for generation of atmospheric states compatible with boundary conditions of solar position and sea surface temperatures. Crucially, cBottle can be guided to generate extreme events such as Tropical Cyclones (TCs) over locations of interest. Diffusion models such as cBottle work by approximating the probability density of the training data. Here, we show use cases of the probability density estimates of atmospheric states obtained from this climate emulator. Most importantly, these estimates allow us to calculate likelihoods of extreme events under guidance. When guiding the model towards states including TCs, comparing the probability density under the guided and unguided model enables us to quantify how much more likely the guidance has made the TC. We show how these odds ratios allow us to importance-sample from the TC distribution, reducing the standard error of the probability estimate compared to simple Monte Carlo sampling. Furthermore, we discuss results and limitations of the application of model probability densities to extreme event attribution-like experiments. We present these early but encouraging results hoping they will spur more research into probabilistic information that can be gained from diffusion models of the atmosphere.

Problem

• extreme event likelihood
• diffusion model
• climate emulator
• tropical cyclone
• probability density

Innovation

• diffusion model
• extreme event likelihood
• climate emulator
• importance sampling
• probability density estimation