LaDCast: A Latent Diffusion Model for Medium-Range Ensemble Weather Forecasting

TL;DR

LaDCast introduces a novel latent diffusion framework for medium-range ensemble weather forecasting, achieving high accuracy and efficiency in probabilistic predictions, especially for extreme events, while significantly reducing computational costs.

Contribution

It is the first global latent-diffusion model for medium-range ensemble weather forecasting operating entirely in a learned latent space.

Findings

Performs comparably to ECMWF IFS-ENS in skill metrics.

Excels in tracking rare extreme events like cyclones.

Reduces storage and computation by orders of magnitude.

Abstract

Accurate probabilistic weather forecasting demands both high accuracy and efficient uncertainty quantification, challenges that overburden both ensemble numerical weather prediction (NWP) and recent machine-learning methods. We introduce LaDCast, the first global latent-diffusion framework for medium-range ensemble forecasting, which generates hourly ensemble forecasts entirely in a learned latent space. An autoencoder compresses high-dimensional ERA5 reanalysis fields into a compact representation, and a transformer-based diffusion model produces sequential latent updates with arbitrary hour initialization. The model incorporates Geometric Rotary Position Embedding (GeoRoPE) to account for the Earth's spherical geometry, a dual-stream attention mechanism for efficient conditioning, and sinusoidal temporal embeddings to capture seasonal patterns. LaDCast achieves deterministic and probabilistic skill close to that of the European Centre for Medium-Range Forecast IFS-ENS, without any explicit perturbations. Notably, LaDCast demonstrates superior performance in tracking rare extreme events such as cyclones, capturing their trajectories more accurately than established models. By operating in latent space, LaDCast reduces storage and compute by orders of magnitude, demonstrating a practical path toward forecasting at kilometer-scale resolution in real time. We open-source our code and models and provide the training and evaluation pipelines at: https://github.com/tonyzyl/ladcast.