mlsanityData to 7 Oct 2026

Generative Diffusion-based Downscaling for Climate

arXiv 2404.17752 ↗First posted 27 Apr 2024physics.ao-ph cs.LGcited by 25

Not on the map. Why?

Abstract (arXiv)

Downscaling, or super-resolution, provides decision-makers with detailed, high-resolution information about the potential risks and impacts of climate change, based on climate model output. Machine learning algorithms are proving themselves to be efficient and accurate approaches to downscaling. Here, we show how a generative, diffusion-based approach to downscaling gives accurate downscaled results. We focus on an idealised setting where we recover ERA5 at 0.25degree~resolution from coarse grained version at 2degree~resolution. The diffusion-based method provides superior accuracy compared to a standard U-Net, particularly at the fine scales, as highlighted by a spectral decomposition. Additionally, the generative approach provides users with a probability distribution which can be used for risk assessment. This research highlights the potential of diffusion-based downscaling techniques in providing reliable and detailed climate predictions.

Key facts. Cited by 25 papers in the map (data through 7 Oct 2026). Builds on 44 of its 85 references that are in the map. First posted on arXiv 27 Apr 2024.

Builds on

44 of 85 references.

References · 44

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  5. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  6. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  7. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  8. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  9. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  10. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  11. 2022FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operatorscited by 748
  12. 2023ClimaX: A foundation model for weather and climatecited by 325
32 more
  1. 2019Tackling Climate Change with Machine Learningcited by 264
  2. 2023Text-to-image Diffusion Models in Generative AI: A Surveycited by 264
  3. 2023FuXi: A cascade machine learning forecasting system for 15-day global weather forecastcited by 237
  4. 2023Neural General Circulation Models for Weather and Climatecited by 232
  5. 2023GenCast: Diffusion-based ensemble forecasting for medium-range weathercited by 169
  6. 2022Pangu-Weather: A 3D High-Resolution Model for Fast and Accurate Global Weather Forecastcited by 164
  7. 2023DYffusion: A Dynamics-informed Diffusion Model for Spatiotemporal Forecastingcited by 105
  8. 2017DeepSD: Generating High Resolution Climate Change Projections through Single Image Super-Resolutioncited by 80
  9. 2020Stochastic Super-Resolution for Downscaling Time-Evolving Atmospheric Fields with a Generative Adversarial Networkcited by 68
  10. 2022A Generative Deep Learning Approach to Stochastic Downscaling of Precipitation Forecastscited by 64
  11. 2024DiffDA: a Diffusion Model for Weather-scale Data Assimilationcited by 59
  12. 2019Machine Learning for Precipitation Nowcasting from Radar Imagescited by 56
  13. 2023ACE: A fast, skillful learned global atmospheric model for climate predictioncited by 53
  14. 2024Estimating Epistemic and Aleatoric Uncertainty with a Single Modelcited by 52
  15. 2022Increasing the accuracy and resolution of precipitation forecasts using deep generative modelscited by 35
  16. 2023Debias Coarsely, Sample Conditionally: Statistical Downscaling through Optimal Transport and Probabilistic Diffusion Modelscited by 33
  17. 2022Hard-Constrained Deep Learning for Climate Downscalingcited by 29
  18. 2023SEEDS: Emulation of Weather Forecast Ensembles with Diffusion Modelscited by 26
  19. 2020ClimAlign: Unsupervised statistical downscaling of climate variables via normalizing flowscited by 25
  20. 2017Intercomparison of Machine Learning Methods for Statistical Downscaling: The Case of Daily and Extreme Precipitationcited by 24
  21. 2023Fourier Neural Operators for Arbitrary Resolution Climate Data Downscalingcited by 24
  22. 2023Unpaired Downscaling of Fluid Flows with Diffusion Bridgescited by 21
  23. 2022Machine learning emulation of a local-scale UK climate modelcited by 20
  24. 2020TRU-NET: A Deep Learning Approach to High Resolution Prediction of Rainfallcited by 19
  25. 2020Climbing down Charney's ladder: Machine Learning and the post-Dennard era of computational climate sciencecited by 18
  26. 2020A generative adversarial network approach to (ensemble) weather predictioncited by 18
  27. 2021Fast and accurate learned multiresolution dynamical downscaling for precipitationcited by 18
  28. 2023Forecasting Tropical Cyclones with Cascaded Diffusion Modelscited by 10
  29. 2022Physics-Informed Learning of Aerosol Microphysicscited by 9
  30. 2023DiffESM: Conditional Emulation of Earth System Models with Diffusion Modelscited by 9
  31. 2022Deep generative model super-resolves spatially correlated multiregional climate datacited by 8
  32. 2016End-to-End Image Super-Resolution via Deep and Shallow Convolutional Networkscited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 13; 2026: 10.

25 papers.

Sort:

By area:

  1. 2025FLEX: A Backbone for Diffusion-Based Modeling of Spatio-temporal Physical Systemscited by 8
  2. 2025Multimodal Atmospheric Super-Resolution With Deep Generative Modelscited by 8
  3. 2025RainShift: A Benchmark for Precipitation Downscaling Across Geographiescited by 5
  4. 2025Spatiotemporal Pyramid Flow Matching for Climate Emulationcited by 4
  5. 2025DiffScale: Continuous Downscaling and Bias Correction of Subseasonal Wind Speed Forecasts using Diffusion Modelscited by 3
  6. 2024TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion modelscited by 2
  7. 2025Guided Unconditional and Conditional Generative Models for Super-Resolution and Inference of Quasi-Geostrophic Turbulencecited by 2
  8. 2026PODiff: Latent Diffusion in Proper Orthogonal Decomposition Space for Scientific Super-Resolutioncited by 2
  9. 2025A PDE-Informed Latent Diffusion Model for 2-m Temperature Downscalingcited by 1
  10. 2025China Regional 3km Downscaling Based on Residual Corrective Diffusion Modelcited by 1
  11. 2026Super-Resolving Coarse-Resolution Weather Forecasts With Flow Matchingcited by 1
  12. 2026Patch-PODiff-ViT: Structured Latent Diffusion with Patchwise POD for Super-Resolution and Uncertainty Quantificationcited by 1
13 more
  1. 2024Advancing Marine Heatwave Forecasts: An Integrated Deep Learning Approachcited by 0
  2. 2025Weakly-Constrained 4D Var for Downscaling with Uncertainty using Data-Driven Surrogate Modelscited by 0
  3. 2025Vision Transformers for Multi-Variable Climate Downscaling: Emulating Regional Climate Models with a Shared Encoder and Multi-Decoder Architecturecited by 0
  4. 2025Breaking the Statistical Similarity Trap in Extreme Convection Detectioncited by 0
  5. 2025A Probabilistic U-Net Approach to Downscaling Climate Simulationscited by 0
  6. 2025Spatiotemporal Satellite Image Downscaling with Transfer Encoders and Autoregressive Generative Modelscited by 0
  7. 2026IPSL-AID: Generative Diffusion Models for Climate Downscaling from Global to Regional Scalescited by 0
  8. 2026Domain-Adaptive Climate Downscaling Under Temporal Distribution Shiftcited by 0
  9. 2026Physics-Informed Super-Resolution of Atmospheric Datacited by 0
  10. 2026Multimodal Spatiotemporal Atmospheric Data Assimilation with Latent Video Flow-matchingcited by 0
  11. 2026Temporal Bridges for Spatial Resolution: Enhancing Climate Data Super-Resolution with Bidirectional Alignmentcited by 0
  12. 2026RainAtlas: A Multi-Continental Dataset for Precipitation Downscalingcited by 0
  13. 2026Artificial intelligence pathways from weather to climatecited by 0