Apeliotes: A Diffusion-Based Modeling Framework for km-scale Multi-Level Atmospheric Fields

2026-07-18

AI Summary

High-resolution, multivariate atmospheric field data are scarce, and traditional dynamical downscaling methods suffer from high computational costs and limited generalization. This work proposes an end-to-end generative framework that, for the first time, integrates a globally pretrained weather foundation model with a regional diffusion model to directly synthesize kilometer-scale, multivariate, and multi-level atmospheric fields from reanalysis data. The approach enables stochastic high-resolution generation of unobservable variables and achieves exceptional accuracy: vertical wind profile predictions exhibit errors below 3%, while 10-meter wind speed and 2-meter temperature attain correlation coefficients of 0.91 and 0.99, respectively, with normalized root mean square errors (NRMSE) as low as 0.42 and 0.17—significantly outperforming existing methods.

Abstract

High-resolution atmospheric data are required to resolve mesoscale and localized meteorological structures, however such datasets remain limited in many regions of the world. Existing high-resolution weather products are typically produced through dynamical downscaling, which is computationally expensive and difficult to scale across locations, variables, and forecast scenarios. These limitations motivate machine-learning-based downscaling systems that can generate multiple weather variables stochastically while producing new high-resolution fields directly. In this paper we present Apeliotes, a framework for high-resolution weather forecasting. Built on the global re-analysis atmospheric data, a pre-trained global weather foundation model, and a regionally trained generative diffusion model, Apeliotes not only provides accurate kilometer-scale weather variables, but also multi-level atmospheric fields which are not directly available in the existing global atmospheric data. Our comprehensive evaluation demonstrates that Apeliotes achieves highly competitive performance. The model predicts vertical wind profile with less than 3\% error between truth and predicted fields, achieving correlations of 0.91 for 10-m wind speed and 0.99 for 2-m temperature, with NRMSE values of 0.42 and 0.17, respectively.

Problem

• high-resolution atmospheric data
• dynamical downscaling
• multi-level atmospheric fields
• weather forecasting
• data scarcity

Innovation

• diffusion model
• atmospheric downscaling
• multi-level atmospheric fields
• high-resolution weather forecasting
• generative modeling