# Conditional diffusion models for downscaling and bias correction of Earth system model precipitation

arXiv: https://arxiv.org/abs/2404.14416 · HTML: https://mlsanity.com/paper/2404.14416/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Placed in Flow matching, time series and science, one of 14 directions in Diffusion models.
- Builds on 15 of its 47 references that are in the map.
- First posted on arXiv 5 Apr 2024.

## Placed in

- Diffusion models / [Flow matching, time series and science](https://mlsanity.com/diffusion/flow-matching-time-series-and-science/)

## Builds on (most cited first, 15 of 15)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Elucidating the Design Space of Diffusion-Based Generative Models](https://mlsanity.com/paper/2206.00364/) (2022)
- [Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distances](https://mlsanity.com/paper/1306.0895/) (2013)
- [Palette: Image-to-Image Diffusion Models](https://mlsanity.com/paper/2111.05826/) (2021)
- [Cascaded Diffusion Models for High Fidelity Image Generation](https://mlsanity.com/paper/2106.15282/) (2021)
- [Knowledge Distillation in Iterative Generative Models for Improved Sampling Speed](https://mlsanity.com/paper/2101.02388/) (2021)
- [Efficient Diffusion Training via Min-SNR Weighting Strategy](https://mlsanity.com/paper/2303.09556/) (2023)
- [DeepSD: Generating High Resolution Climate Change Projections through Single Image Super-Resolution](https://mlsanity.com/paper/1703.03126/) (2017)
- [Debias Coarsely, Sample Conditionally: Statistical Downscaling through Optimal Transport and Probabilistic Diffusion Models](https://mlsanity.com/paper/2305.15618/) (2023)
- [Physically Constrained Generative Adversarial Networks for Improving Precipitation Fields from Earth System Models](https://mlsanity.com/paper/2209.07568/) (2022)
- [Fast, Scale-Adaptive, and Uncertainty-Aware Downscaling of Earth System Model Fields with Generative Machine Learning](https://mlsanity.com/paper/2403.02774/) (2024)
- [Deep learning for bias-correcting CMIP6-class Earth system models](https://mlsanity.com/paper/2301.01253/) (2022)

## Built on by (most cited first, 17 of 17)

- [Machine learning emulation of precipitation from km-scale UK regional climate simulations using a diffusion model](https://mlsanity.com/paper/2407.14158/) (2024)
- [A Generative Framework for Probabilistic, Spatiotemporally Coherent Downscaling of Climate Simulation](https://mlsanity.com/paper/2412.15361/) (2024)
- [ORBIT-2: Scaling Exascale Vision Foundation Models for Weather and Climate Downscaling](https://mlsanity.com/paper/2505.04802/) (2025)
- [Satellite Observations Guided Diffusion Model for Accurate Meteorological States at Arbitrary Resolution](https://mlsanity.com/paper/2502.07814/) (2025)
- [Stochastic Flow Matching for Resolving Small-Scale Physics](https://mlsanity.com/paper/2410.19814/) (2024)
- [WIND: Weather Inverse Diffusion for Zero-Shot Atmospheric Modeling](https://mlsanity.com/paper/2602.03924/) (2026)
- [Generating realistic global precipitation fields from modelled atmospheric circulation](https://mlsanity.com/paper/2504.00307/) (2025)
- [Reframing Generative Models for Physical Systems using Stochastic Interpolants](https://mlsanity.com/paper/2509.26282/) (2025)
- [Generating time-consistent dynamics with discriminator-guided image diffusion models](https://mlsanity.com/paper/2505.09089/) (2025)
- [Vision Transformers for Multi-Variable Climate Downscaling: Emulating Regional Climate Models with a Shared Encoder and Multi-Decoder Architecture](https://mlsanity.com/paper/2506.22447/) (2025)
- [Physics-constrained generative machine learning-based high-resolution downscaling of Greenland's surface mass balance and surface temperature](https://mlsanity.com/paper/2507.22485/) (2025)
- [How Can Machine Learning Emulators Best Support Climate Science?](https://mlsanity.com/paper/2603.22320/) (2026)
- [Regional Climate Model Emulation with Diffusion Approaches: What is the Added Value of Generative Machine Learning?](https://mlsanity.com/paper/2606.14570/) (2026)
- [Domain-Adaptive Climate Downscaling Under Temporal Distribution Shift](https://mlsanity.com/paper/2607.05645/) (2026)
- [Apeliotes: A Diffusion-Based Modeling Framework for km-scale Multi-Level Atmospheric Fields](https://mlsanity.com/paper/2607.17037/) (2026)
- [Temporal Bridges for Spatial Resolution: Enhancing Climate Data Super-Resolution with Bidirectional Alignment](https://mlsanity.com/paper/2608.05981/) (2026)
- [Generalizable Neural Downscaling of Earth System Model Wind Fields via Continuous Dynamics Modeling](https://mlsanity.com/paper/2610.03757/) (2026)

## Abstract (arXiv)

Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damage, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle to resolve small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We first map observational and ESM data to a shared embedding space, where both are unbiased towards each other, and then train a conditional diffusion model to reverse the mapping. Only observational data is used for the training, so that the diffusion model can be employed to correct and downscale any ESM field without need for retraining. Our approach ensures statistical fidelity and preserves spatial patterns larger than a chosen spatial correction scale. We demonstrate that our approach outperforms existing statistical and deep learning methods especially regarding extreme events.