Physics-Informed Diffusion Model for Generating Synthetic Extreme Rare Weather Events Data

TL;DR

This work proposes a physics-informed diffusion model based on the Context-UNet architecture to generate synthetic, multi-spectral satellite imagery of extreme weather events and addresses the extreme class imbalance in the dataset, demonstrating a scalable framework for enhancing operational weather detection algorithms.

Abstract

Data scarcity is a primary obstacle in developing robust Machine Learning (ML) models for detecting rapidly intensifying tropical cyclones. Traditional data augmentation techniques (rotation, flipping, brightness adjustment) fail to preserve the physical consistency and high-intensity gradients characteristic of rare Category 4-equivalent events, which constitute only 0.14\% of our dataset (202 of 140,514 samples). We propose a physics-informed diffusion model based on the Context-UNet architecture to generate synthetic, multi-spectral satellite imagery of extreme weather events. Our model is conditioned on critical atmospheric parameters such as average wind speed, type of Ocean and stage of development (early, mature, late etc) -- the known drivers of rapid intensification. Using a controlled pre-generated noise sampling strategy and mixed-precision training, we generated $16\times16$ wind-field samples that are cropped from multi-spectral satellite imagery which preserve realistic spatial autocorrelation and physical consistency. Results demonstrate that our model successfully learns discriminative features across ten distinct context classes, effectively mitigating the data bottleneck. Specifically, we address the extreme class imbalance in our dataset, where Class 4 (Ocean 2, early stage with average wind speed 50kn hurricane) contains only 202 samples compared to 79,768 samples in Class 0. This generative framework provides a scalable solution for augmenting training datasets for operational weather detection algorithms. The average Results yield an average Log-Spectral Distance (LSD) of 4.5dB, demonstrating a scalable framework for enhancing operational weather detection algorithms.

Figures (7)

• Figure S1: Figure illustrates the three-phase pipeline for generating synthetic extreme events: Forward Process: $x_0$ (original wind field) is transformed into Gaussian noise $x_t$ using a pre-generated noise strategy. Context-UNet Training: A UNet ($n\_feat=64$) predicts noise by conditioning the noisy input ($x_t$) on sinusoidal time steps ($t$) and physics-based context ($c$), such as one-hot encoded wind shear classes. Reverse Process: The trained model performs 500 iterative denoising steps starting from $x_T$ to produce a synthetic extreme event $x^{\hat{}}_0$.
• Figure S2: Visual and statistical verification of eight random noise samples. Each patch represents a $16 \times 16$ realization of $\epsilon \sim \mathcal{N}(0, 1)$. The consistent Gaussian distribution across samples ensures that rare event detection is not biased by noise artifacts.
• Figure S4: Storm Wind Fields representing physical intensity on a $16 \times 16$ grid. Samples are conditioned on specific atmospheric parameters (Param 0 and 1) representing Rapid Intensification (RI) conditions. The colormap highlights the high-intensity gradients preserved by the physics-informed model.
• Figure S5: Distribution of unique physical parameter labels within the dataset. The extreme scarcity of samples in Class 4 (202 samples) demonstrates the data bottleneck for rare weather events that this research aims to mitigate through synthetic augmentation.
• Figure S6: Context-conditioned generation for labels 0-9. Each image represents a wind field pattern generated conditioned on its corresponding class label using one-hot encoding. The model successfully produces distinct patterns for different context inputs.