Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
166 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Residual Corrective Diffusion Modeling for Km-scale Atmospheric Downscaling (2309.15214v4)

Published 24 Sep 2023 in cs.LG and physics.ao-ph

Abstract: The state of the art for physical hazard prediction from weather and climate requires expensive km-scale numerical simulations driven by coarser resolution global inputs. Here, a generative diffusion architecture is explored for downscaling such global inputs to km-scale, as a cost-effective machine learning alternative. The model is trained to predict 2km data from a regional weather model over Taiwan, conditioned on a 25km global reanalysis. To address the large resolution ratio, different physics involved at different scales and prediction of channels beyond those in the input data, we employ a two-step approach where a UNet predicts the mean and a corrector diffusion (CorrDiff) model predicts the residual. CorrDiff exhibits encouraging skill in bulk MAE and CRPS scores. The predicted spectra and distributions from CorrDiff faithfully recover important power law relationships in the target data. Case studies of coherent weather phenomena show that CorrDiff can help sharpen wind and temperature gradients that co-locate with intense rainfall in cold front, and can help intensify typhoons and synthesize rain band structures. Calibration of model uncertainty remains challenging. The prospect of unifying methods like CorrDiff with coarser resolution global weather models implies a potential for global-to-regional multi-scale machine learning simulation.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (62)
  1. Machine learning emulation of a local-scale uk climate model. arXiv preprint arXiv:2211.16116, 2022.
  2. Tru-net: a deep learning approach to high resolution prediction of rainfall. Machine Learning, 110:2035–2062, 2021.
  3. Configuration and intercomparison of deep learning neural models for statistical downscaling. Geoscientific Model Development, 13(4):2109–2124, 2020.
  4. Usability of best track data in climate statistics in the western north pacific. Monthly Weather Review, 140(9):2818–2830, 2012.
  5. Conditional image generation with score-based diffusion models. arXiv preprint arXiv:2111.13606, 2021.
  6. The rise of data-driven weather forecasting, 2023.
  7. Accurate medium-range global weather forecasting with 3d neural networks. Nature, pages 1–6, 2023.
  8. How well is outer tropical cyclone size represented in the era5 reanalysis dataset? Atmospheric Research, 249:105339, 2021.
  9. Unpaired downscaling of fluid flows with diffusion bridges. arXiv preprint arXiv:2305.01822, 2023.
  10. Introduction to probability. Crc Press, 2019.
  11. Spherical fourier neural operators: Learning stable dynamics on the sphere. arXiv preprint arXiv:2306.03838, 2023.
  12. An operational multi-radar multi-sensor qpe system in taiwan. Bulletin of the American Meteorological Society, 102(3):E555–E577, 2021.
  13. Improving afternoon thunderstorm prediction over taiwan through 3dvar-based radar and surface data assimilation. Weather and Forecasting, 35(6):2603–2620, 2020.
  14. Fuxi: A cascade machine learning forecasting system for 15-day global weather forecast. npj Climate and Atmospheric Science, 6:190, 2023.
  15. Diffusion models in vision: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023.
  16. P. Dhariwal and A. Nichol. Diffusion models beat gans on image synthesis. In Proceedings of NeurIPS, volume 34, pages 8780–8794, 2021.
  17. Regional data assimilation with the ncmrwf unified model (ncum): impact of doppler weather radar radial wind. Pure and Applied Geophysics, 176:4575–4597, 2019.
  18. Why should ensemble spread match the RMSE of the ensemble mean? J. Hydrometeorol., 15(4):1708–1713, August 2014.
  19. Radar super resolution using a deep convolutional neural network. Journal of Atmospheric and Oceanic Technology, 37(12):2197–2207, 2020.
  20. Strictly proper scoring rules, prediction, and estimation. J. Am. Stat. Assoc., 102(477):359–378, March 2007.
  21. Enhancing spatial variability representation of radar nowcasting with generative adversarial networks. Remote Sensing, 15(13):3306, 2023.
  22. The ongoing need for high-resolution regional climate models: Process understanding and stakeholder information. Bulletin of the American Meteorological Society, 101(5):E664–E683, 2020.
  23. A generative deep learning approach to stochastic downscaling of precipitation forecasts. Journal of Advances in Modeling Earth Systems, 14(10):e2022MS003120, 2022.
  24. Diffusion models for high-resolution solar forecasts. arXiv preprint arXiv:2302.00170, 2023.
  25. The era5 global reanalysis. Quarterly Journal of the Royal Meteorological Society, 146(730):1999–2049, 2020.
  26. Denoising diffusion probabilistic models. In Proceedings of NeurIPS, volume 33, pages 6840–6851, 2020.
  27. Icon-sapphire: simulating the components of the earth system and their interactions at kilometer and subkilometer scales. Geoscientific Model Development, 16(2):779–811, 2023.
  28. Elucidating the design space of diffusion-based generative models. arXiv preprint arXiv:2206.00364, 2022.
  29. On convergence and stability of gans. arXiv preprint arXiv:1705.07215, 2017.
  30. Learning skillful medium-range global weather forecasting. Science, 0(0):eadi2336, 2023.
  31. Latent diffusion models for generative precipitation nowcasting with accurate uncertainty quantification. arXiv preprint arXiv:2304.12891, 2023.
  32. Stochastic super-resolution for downscaling time-evolving atmospheric fields with a generative adversarial network. IEEE Transactions on Geoscience and Remote Sensing, 59(9):7211–7223, 2020.
  33. Seeds: Emulation of weather forecast ensembles with diffusion models. arXiv preprint arXiv:2306.14066, 2023.
  34. On distillation of guided diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14297–14306, 2023.
  35. A climate downscaling deep learning model considering the multiscale spatial correlations and chaos of meteorological events. Mathematical Problems in Engineering, 2020:1–17, 2020.
  36. Forecasting tropical cyclones with cascaded diffusion models. arXiv preprint arXiv:2310.01690, 2023.
  37. Ncep gfs 0.25 degree global forecast grids historical archive, 2015.
  38. Comparison of a novel machine learning approach with dynamical downscaling for australian precipitation. Environmental Research Letters, 2023.
  39. Fourcastnet: A global data-driven high-resolution weather model using adaptive fourier neural operators. arXiv preprint arXiv:2202.11214, 2022.
  40. Stephen B Pope. Turbulent flows. Cambridge university press, 2000.
  41. The weather research and forecasting model: Overview, system efforts, and future directions. Bulletin of the American Meteorological Society, 98(8):1717–1737, 2017.
  42. Large-eddy simulation in an anelastic framework with closed water and entropy balances. Journal of Advances in Modeling Earth Systems, 7(3):1425–1456, 2015.
  43. Increasing the accuracy and resolution of precipitation forecasts using deep generative models. In International conference on artificial intelligence and statistics, pages 10555–10571. PMLR, 2022.
  44. Using bayesian model averaging to calibrate forecast ensembles. Mon. Weather Rev., 133(5):1155–1174, May 2005.
  45. High-resolution downscaling with interpretable deep learning: Rainfall extremes over new zealand. Weather and Climate Extremes, 38:100525, 2022.
  46. Skilful precipitation nowcasting using deep generative models of radar. Nature, 597(7878):672–677, 2021.
  47. Deepdownscale: a deep learning strategy for high-resolution weather forecast. arXiv preprint arXiv:1808.05264, 2018.
  48. Improved techniques for training gans. Advances in neural information processing systems, 29, 2016.
  49. Progressive distillation for fast sampling of diffusion models. In International Conference on Learning Representations, 2021.
  50. Machine-learning-based downscaling of hourly era5-land air temperature over mountainous regions. Atmosphere, 14(4):610, 2023.
  51. Upscale error growth in a high-resolution simulation of a summertime weather event over europe. Monthly Weather Review, 143(3):813–827, 2015.
  52. Denoising diffusion implicit models. In Proceedings of the International Conference on Learning Representations (ICLR), 2021.
  53. Generative modeling by estimating gradients of the data distribution. In Advances in Neural Information Processing Systems (NeurIPS), 2019.
  54. Improved techniques for training score-based generative models. Advances in neural information processing systems, 33:12438–12448, 2020.
  55. Score-based generative modeling through stochastic differential equations. In Proceedings of the International Conference on Learning Representations (ICLR), 2021.
  56. Dyamond: the dynamics of the atmospheric general circulation modeled on non-hydrostatic domains. Progress in Earth and Planetary Science, 6(1):1–17, 2019.
  57. Physics-informed deep learning framework to model intense precipitation events at super resolution. Geoscience Letters, 10(1):19, 2023.
  58. Score-based generative modeling in latent space. Advances in Neural Information Processing Systems, 34:11287–11302, 2021.
  59. Deep learning for downscaling tropical cyclone rainfall to hazard-relevant spatial scales. Journal of Geophysical Research: Atmospheres, page e2022JD038163, 2023.
  60. Statistical downscaling of general circulation model output: A comparison of methods. Water resources research, 34(11):2995–3008, 1998.
  61. Tackling the generative learning trilemma with denoising diffusion GANs. In Internatiodnal Conference on Learning Representations (ICLR), 2022.
  62. Fast sampling of diffusion models via operator learning. In International Conference on Machine Learning, pages 42390–42402. PMLR, 2023.
Citations (19)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now