Monthly Diffusion v0.9: A Latent Diffusion Model for the First AI-MIP
For climate modeling, it offers a computationally efficient emulator for low-frequency atmospheric variability, though results are preliminary.
This paper introduces Monthly Diffusion v0.9, a latent diffusion model for emulating monthly climate variability at 1.5-degree resolution, achieving efficient forward-stepping with modest computational resources.
Here, we describe Monthly Diffusion at 1.5-degree grid spacing (MD-1.5 version 0.9), a climate emulator that leverages a spherical Fourier neural operator (SFNO)-inspired Conditional Variational Auto-Encoder (CVAE) architecture to model the evolution of low-frequency internal atmospheric variability using latent diffusion. MDv0.9 was designed to forward-step at monthly mean timesteps in a data-sparse regime, using modest computational requirements. This work describes the motivation behind the architecture design, the MDv0.9 training procedure, and initial results.