# Field-Space Autoencoder for Scalable Climate Emulators

arXiv: https://arxiv.org/abs/2601.15102 · HTML: https://mlsanity.com/paper/2601.15102/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 53 references that are in the map.
- First posted on arXiv 21 Jan 2026.

## Builds on (most cited first, 21 of 21)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [GraphCast: Learning skillful medium-range global weather forecasting](https://mlsanity.com/paper/2212.12794/) (2022)
- [FourCastNet: Accelerating Global High-Resolution Weather Forecasting using Adaptive Fourier Neural Operators](https://mlsanity.com/paper/2208.05419/) (2022)
- [GenCast: Diffusion-based ensemble forecasting for medium-range weather](https://mlsanity.com/paper/2312.15796/) (2023)
- [Pangu-Weather: A 3D High-Resolution Model for Fast and Accurate Global Weather Forecast](https://mlsanity.com/paper/2211.02556/) (2022)
- [DYffusion: A Dynamics-informed Diffusion Model for Spatiotemporal Forecasting](https://mlsanity.com/paper/2306.01984/) (2023)
- [Improved Techniques for Maximum Likelihood Estimation for Diffusion ODEs](https://mlsanity.com/paper/2305.03935/) (2023)
- [An Analysis of the t-SNE Algorithm for Data Visualization](https://mlsanity.com/paper/1803.01768/) (2018)
- [Variational Autoencoder with Learned Latent Structure](https://mlsanity.com/paper/2006.10597/) (2020)
- [AtmoRep: A stochastic model of atmosphere dynamics using large scale representation learning](https://mlsanity.com/paper/2308.13280/) (2023)
- [CRA5: Extreme Compression of ERA5 for Portable Global Climate and Weather Research via an Efficient Variational Transformer](https://mlsanity.com/paper/2405.03376/) (2024)
- [Fourier Amplitude and Correlation Loss: Beyond Using L2 Loss for Skillful Precipitation Nowcasting](https://mlsanity.com/paper/2410.23159/) (2024)
- [HiLAB: A Hybrid Inverse-Design Framework](https://mlsanity.com/paper/2505.17491/) (2025)
- [LaDCast: A Latent Diffusion Model for Medium-Range Ensemble Weather Forecasting](https://mlsanity.com/paper/2506.09193/) (2025)
- [Compression and Conditional Emulation of Climate Model Output](https://mlsanity.com/paper/1605.07919/) (2016)
- [Latent Diffusion Model for Generating Ensembles of Climate Simulations](https://mlsanity.com/paper/2407.02070/) (2024)
- [Scalable Climate Data Analysis: Balancing Petascale Fidelity and Computational Cost](https://mlsanity.com/paper/2507.08006/) (2025)
- [Explicit Time Embedding Based Cascade Attention Network for Information Popularity Prediction](https://mlsanity.com/paper/2308.09976/) (2023)
- [Wasserstein GAN-Based Precipitation Downscaling with Optimal Transport for Enhancing Perceptual Realism](https://mlsanity.com/paper/2507.17798/) (2025)
- [Field-Space Attention for Structure-Preserving Earth System Transformers](https://mlsanity.com/paper/2512.20350/) (2025)

## Abstract (arXiv)

Kilometer-scale Earth system models are essential for capturing local climate change. However, these models are computationally expensive and produce petabyte-scale outputs, which limits their utility for applications such as probabilistic risk assessment. Here, we present the Field-Space Autoencoder, a scalable climate emulation framework based on a spherical compression model that overcomes these challenges. By utilizing Field-Space Attention, the model efficiently operates on native climate model output and therefore avoids geometric distortions caused by forcing spherical data onto Euclidean grids. This approach preserves physical structures significantly better than convolutional baselines. By producing a structured compressed field, it serves as a good baseline for downstream generative emulation. In addition, the model can perform zero-shot super-resolution that maps low-resolution large ensembles and scarce high-resolution data into a shared representation. We train a generative diffusion model on these compressed fields. The model can simultaneously learn internal variability from abundant low-resolution data and fine-scale physics from sparse high-resolution data. Our work bridges the gap between the high volume of low-resolution ensemble statistics and the scarcity of high-resolution physical detail.