Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
175 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

Distributional Regression U-Nets for the Postprocessing of Precipitation Ensemble Forecasts (2407.02125v1)

Published 2 Jul 2024 in stat.ML, cs.LG, and stat.AP

Abstract: Accurate precipitation forecasts have a high socio-economic value due to their role in decision-making in various fields such as transport networks and farming. We propose a global statistical postprocessing method for grid-based precipitation ensemble forecasts. This U-Net-based distributional regression method predicts marginal distributions in the form of parametric distributions inferred by scoring rule minimization. Distributional regression U-Nets are compared to state-of-the-art postprocessing methods for daily 21-h forecasts of 3-h accumulated precipitation over the South of France. Training data comes from the M\'et\'eo-France weather model AROME-EPS and spans 3 years. A practical challenge appears when consistent data or reforecasts are not available. Distributional regression U-Nets compete favorably with the raw ensemble. In terms of continuous ranked probability score, they reach a performance comparable to quantile regression forests (QRF). However, they are unable to provide calibrated forecasts in areas associated with high climatological precipitation. In terms of predictive power for heavy precipitation events, they outperform both QRF and semi-parametric QRF with tail extensions.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (80)
  1. TensorFlow: Large-scale machine learning on heterogeneous systems, 2015. URL https://www.tensorflow.org/. Software available from tensorflow.org.
  2. Censored and shifted gamma distribution based EMOS model for probabilistic quantitative precipitation forecasting. Environmetrics, 27(5):280–292, May 2016. ISSN 1099-095X. https://doi.org/10.1002/env.2391.
  3. Relational inductive biases, deep learning, and graph networks. June 2018. https://doi.org/10.48550/ARXIV.1806.01261.
  4. The quiet revolution of numerical weather prediction. Nature, 525(7567):47–55, sep 2015. https://doi.org/10.1038/nature14956.
  5. Generation of scenarios from calibrated ensemble forecasts with a dual-ensemble copula-coupling approach. Monthly Weather Review, 144(12):4737–4750, November 2016. ISSN 1520-0493. https://doi.org/10.1175/mwr-d-15-0403.1.
  6. Improving medium-range ensemble weather forecasts with hierarchical ensemble transformers. Artificial Intelligence for the Earth Systems, 3(1), January 2024. ISSN 2769-7525. https://doi.org/10.1175/aies-d-23-0027.1.
  7. Patrick Bénichou. Cartography of statistical pluviometric fields with an automatic allowance for regional topography. In Global Precipitations and Climate Change, pages 187–199. Springer Berlin Heidelberg, 1994. https://doi.org/10.1007/978-3-642-79268-7_11.
  8. Accurate medium-range global weather forecasting with 3d neural networks. Nature, 619(7970):533–538, July 2023. ISSN 1476-4687. https://doi.org/10.1038/s41586-023-06185-3.
  9. Sensitivity of the AROME ensemble to initial and surface perturbations during HyMeX. Quarterly Journal of the Royal Meteorological Society, 142(S1):390–403, sep 2015. https://doi.org/10.1002/qj.2622.
  10. Leo Breiman. Random forests. Machine Learning, 45(1):5–32, 2001. https://doi.org/10.1023/a:1010933404324.
  11. John Bjørnar Bremnes. Ensemble postprocessing using quantile function regression based on neural networks and bernstein polynomials. Monthly Weather Review, 148(1):403–414, jan 2020. ISSN 1520-0493. https://doi.org/10.1175/mwr-d-19-0227.1.
  12. Glenn W. Brier. Verification of forecasts expressed in terms of probability. Monthly Weather Review, 78(1):1–3, 1950. ISSN 1520-0493. https://doi.org/10.1175/1520-0493(1950)078<0001:vofeit>2.0.co;2.
  13. The heavy precipitation event of 14–15 october 2018 in the aude catchment: a meteorological study based on operational numerical weather prediction systems and standard and personal observations. Natural Hazards and Earth System Sciences, 21(3):1135–1157, March 2021. ISSN 1684-9981. https://doi.org/10.5194/nhess-21-1135-2021.
  14. Les mesures de précipitations et l’estimation des lames d’eau à Météo-France : état de l’art et perspectives. La Houille Blanche, 95(5):28–34, October 2009. ISSN 1958-5551. https://doi.org/10.1051/lhb/2009052.
  15. Probabilistic predictions from deterministic atmospheric river forecasts with deep learning. Monthly Weather Review, 150(1):215–234, jan 2022. https://doi.org/10.1175/mwr-d-21-0106.1.
  16. Fengwu: Pushing the skillful global medium-range weather forecast beyond 10 days lead. April 2023. https://doi.org/10.48550/ARXIV.2304.02948.
  17. François Chollet et al. Keras. https://keras.io, 2015. URL https://keras.io.
  18. Deep graphical regression for jointly moderate and extreme australian wildfires. Spatial Statistics, 59:100811, March 2024. ISSN 2211-6753. https://doi.org/10.1016/j.spasta.2024.100811.
  19. The schaake shuffle: A method for reconstructing space–time variability in forecasted precipitation and temperature fields. Journal of Hydrometeorology, 5(1):243–262, feb 2004. https://doi.org/10.1175/1525-7541(2004)005<0243:tssamf>2.0.co;2.
  20. Y. Dai and S. Hemri. Spatially coherent postprocessing of cloud cover ensemble forecasts. Monthly Weather Review, 149(12):3923–3937, dec 2021. https://doi.org/10.1175/mwr-d-21-0046.1.
  21. Probabilistic weather prediction with an analog ensemble. Monthly Weather Review, 141(10):3498–3516, September 2013. ISSN 1520-0493. https://doi.org/10.1175/mwr-d-12-00281.1.
  22. Approximation of the distribution of excesses through a generalized probability-weighted moments method. Journal of Statistical Planning and Inference, 137(3):841–857, mar 2007. https://doi.org/10.1016/j.jspi.2006.06.012.
  23. Elizabeth E. Ebert. Fuzzy verification of high‐resolution gridded forecasts: a review and proposed framework. Meteorological Applications, 15(1):51–64, 2008. https://doi.org/10.1002/met.25.
  24. Grid‐ versus station‐based postprocessing of ensemble temperature forecasts. Geophysical Research Letters, 46(13):7744–7751, July 2019. ISSN 1944-8007. https://doi.org/10.1029/2019gl083189.
  25. C. A. T. Ferro. Fair scores for ensemble forecasts. Quarterly Journal of the Royal Meteorological Society, 140(683):1917–1923, December 2013. ISSN 0035-9009. https://doi.org/10.1002/qj.2270.
  26. Postprocessing for extreme events. In Statistical Postprocessing of Ensemble Forecasts, pages 127–154. Elsevier, 2018. https://doi.org/10.1016/b978-0-12-812372-0.00005-4.
  27. Probabilistic forecasting. Annual Review of Statistics and its Applications, 2014. https://doi.org/10.1146/annurev-statistics-062713-085831.
  28. Calibrated probabilistic forecasting using ensemble model output statistics and minimum crps estimation. Monthly Weather Review, 133(5):1098–1118, May 2005. ISSN 0027-0644. https://doi.org/10.1175/mwr2904.1.
  29. Deep learning for post-processing ensemble weather forecasts. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 379(2194):20200092, 2021. https://doi.org/10.1098/rsta.2020.0092.
  30. Thomas M. Hamill. Practical Aspects of Statistical Postprocessing, pages 187–217. Elsevier, 2018. ISBN 9780128123720. https://doi.org/10.1016/b978-0-12-812372-0.00007-8.
  31. Verification of eta–rsm short-range ensemble forecasts. Monthly Weather Review, 125(6):1312–1327, June 1997. ISSN 1520-0493. https://doi.org/10.1175/1520-0493(1997)125<1312:voersr>2.0.co;2.
  32. Trends in the predictive performance of raw ensemble weather forecasts. Geophysical Research Letters, 41(24):9197–9205, dec 2014. https://doi.org/10.1002/2014gl062472.
  33. Deep learning for post-processing global probabilistic forecasts on sub-seasonal time scales. June 2023.
  34. Deep learning forecast uncertainty for precipitation over the western united states. Monthly Weather Review, 151(6):1367–1385, jun 2023. https://doi.org/10.1175/mwr-d-22-0268.1.
  35. Evaluating rank histograms using decompositions of the chi-square test statistic. Monthly Weather Review, 136(6):2133–2139, jun 2008. https://doi.org/10.1175/2007mwr2219.1.
  36. Evaluating probabilistic forecasts with scoringrules. Journal of Statistical Software, 90(12), 2019. ISSN 1548-7660. https://doi.org/10.18637/jss.v090.i12.
  37. Ryan Keisler. Forecasting global weather with graph neural networks. February 2022. https://doi.org/10.48550/ARXIV.2202.07575.
  38. Graphcast: Learning skillful medium-range global weather forecasting. December 2022. https://doi.org/10.48550/ARXIV.2212.12794.
  39. Convolutional autoencoders for spatially-informed ensemble post-processing. April 2022.
  40. Comparison of non-homogeneous regression models for probabilistic wind speed forecasting. Tellus A: Dynamic Meteorology and Oceanography, 65(1):21206, December 2013. ISSN 1600-0870. https://doi.org/10.3402/tellusa.v65i0.21206.
  41. Forecaster’s dilemma: Extreme events and forecast evaluation. Statistical Science, 32(1), February 2017. ISSN 0883-4237. https://doi.org/10.1214/16-sts588.
  42. Visualizing the loss landscape of neural nets. In S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett, editors, Advances in Neural Information Processing Systems, volume 31. Curran Associates, Inc., 2018. URL https://proceedings.neurips.cc/paper_files/paper/2018/file/a41b3bb3e6b050b6c9067c67f663b915-Paper.pdf.
  43. Convolutional neural network-based statistical post-processing of ensemble precipitation forecasts. Journal of Hydrology, 605:127301, feb 2022. https://doi.org/10.1016/j.jhydrol.2021.127301.
  44. Scoring rules for continuous probability distributions. Management Science, 22, 1976. https://doi.org/10.2307/2629907.
  45. Nicolai Meinshausen. Quantile regression forests. Journal of Machine Learning Research, 7:983–999, 2006.
  46. Nonhomogeneous boosting for predictor selection in ensemble postprocessing. Monthly Weather Review, 145(1):137–147, 2017. https://doi.org/10.1175/mwr-d-16-0088.1.
  47. Robust weather-adaptive post-processing using model output statistics random forests. Nonlinear Processes in Geophysics, 30(4):503–514, November 2023. ISSN 1607-7946. https://doi.org/10.5194/npg-30-503-2023.
  48. Modeling jointly low, moderate, and heavy rainfall intensities without a threshold selection. Water Resources Research, 52(4):2753–2769, apr 2016. https://doi.org/10.1002/2015wr018552.
  49. Evaluation of 33 years of quantitative precipitation forecasting at the NMC. Weather and Forecasting, 10(3):498–511, sep 1995. https://doi.org/10.1175/1520-0434(1995)010<0498:eoyoqp>2.0.co;2.
  50. Neural networks for extreme quantile regression with an application to forecasting of flood risk. 2024. https://doi.org/10.48550/ARXIV.2208.07590.
  51. Fourcastnet: A global data-driven high-resolution weather model using adaptive fourier neural operators. February 2022.
  52. Expected impacts of climate change on extreme climate events. Comptes Rendus Geoscience, 340(9-10):564–574, sep 2008. https://doi.org/10.1016/j.crte.2008.07.009.
  53. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, 2023. URL https://www.R-project.org/.
  54. Neural networks for postprocessing ensemble weather forecasts. Monthly Weather Review, 146(11):3885–3900, 2018. https://doi.org/10.1175/mwr-d-18-0187.1.
  55. A climatology of the mesoscale environment associated with heavily precipitating events over a northwestern mediterranean area. Journal of Applied Meteorology and Climatology, 51(3):468–488, March 2012. ISSN 1558-8432. https://doi.org/10.1175/jamc-d-11-017.1.
  56. U-net: Convolutional networks for biomedical image segmentation. In Lecture Notes in Computer Science, pages 234–241. Springer International Publishing, 2015. https://doi.org/10.1007/978-3-319-24574-4_28.
  57. Ensemble Postprocessing Methods Incorporating Dependence Structures, pages 91–125. Elsevier, 2018. ISBN 9780128123720. https://doi.org/10.1016/b978-0-12-812372-0.00004-2.
  58. Uncertainty quantification in complex simulation models using ensemble copula coupling. Statistical Science, 28(4), nov 2013. https://doi.org/10.1214/13-sts443.
  59. Statistical postprocessing of ensemble precipitation forecasts by fitting censored, shifted gamma distributions*. Monthly Weather Review, 143(11):4578–4596, October 2015. ISSN 1520-0493. https://doi.org/10.1175/mwr-d-15-0061.1.
  60. Using artificial neural networks for generating probabilistic subseasonal precipitation forecasts over california. Monthly Weather Review, 148(8):3489–3506, aug 2020. https://doi.org/10.1175/mwr-d-20-0096.1.
  61. Distributional regression forests for probabilistic precipitation forecasting in complex terrain. The Annals of Applied Statistics, 13(3), sep 2019. https://doi.org/10.1214/19-aoas1247.
  62. Machine learning methods for postprocessing ensemble forecasts of wind gusts: A systematic comparison. Monthly Weather Review, 150(1):235–257, 2022a. https://doi.org/10.1175/mwr-d-21-0150.1.
  63. Aggregating distribution forecasts from deep ensembles. April 2022b. https://doi.org/10.48550/ARXIV.2204.02291.
  64. Very deep convolutional networks for large-scale image recognition. September 2014. https://doi.org/10.48550/ARXIV.1409.1556.
  65. From research to applications – examples of operational ensemble post-processing in france using machine learning. Nonlinear Processes in Geophysics, 27(2):329–347, may 2020. https://doi.org/10.5194/npg-27-329-2020.
  66. Calibrated ensemble forecasts using quantile regression forests and ensemble model output statistics. Monthly Weather Review, 144(6):2375–2393, 2016. https://doi.org/10.1175/mwr-d-15-0260.1.
  67. Forest-based and semiparametric methods for the postprocessing of rainfall ensemble forecasting. Weather and Forecasting, 34(3):617–634, 2019. https://doi.org/10.1175/waf-d-18-0149.1.
  68. Corrigendum. Weather and Forecasting, 37(7):1305, July 2022. ISSN 1520-0434. https://doi.org/10.1175/waf-d-22-0057.1.
  69. Alexander Tsyplakov. Evaluation of probabilistic forecasts: Proper scoring rules and moments. SSRN Electronic Journal, 2013. ISSN 1556-5068. https://doi.org/10.2139/ssrn.2236605.
  70. Alexander Tsyplakov. Evaluation of probabilistic forecasts: Conditional auto-calibration, 2020. URL https://www.sas.upenn.edu/~fdiebold/papers2/Tsyplakov_Auto_calibration_sent_eswc2020.pdf.
  71. Bert Van Schaeybroeck and Stéphane Vannitsem. Ensemble post-processing using member-by-member approaches: theoretical aspects. Quarterly Journal of the Royal Meteorological Society, 141(688):807–818, jun 2014. https://doi.org/10.1002/qj.2397.
  72. Statistical postprocessing and multivariate structuring of high-resolution ensemble precipitation forecasts. Journal of Hydrometeorology, 19(11):1815–1833, nov 2018. https://doi.org/10.1175/jhm-d-18-0105.1.
  73. Statistical postprocessing for weather forecasts: Review, challenges, and avenues in a big data world. Bulletin of the American Meteorological Society, 102(3):E681–E699, mar 2021. https://doi.org/10.1175/bams-d-19-0308.1.
  74. Statistical postprocessing of wind speed forecasts using convolutional neural networks. Monthly Weather Review, 149(4):1141–1152, apr 2021. https://doi.org/10.1175/mwr-d-20-0219.1.
  75. Comparing area probability forecasts of (extreme) local precipitation using parametric and machine learning statistical postprocessing methods. Monthly Weather Review, 146(11):3651–3673, oct 2018. https://doi.org/10.1175/mwr-d-17-0290.1.
  76. A comparison of ensemble post-processing methods for extreme events. Quarterly Journal of the Royal Meteorological Society, 140(680):1112–1120, jul 2013. https://doi.org/10.1002/qj.2198.
  77. Scoring rules and the evaluation of probabilities. Test, 5(1):1–60, June 1996. ISSN 1863-8260. https://doi.org/10.1007/bf02562681.
  78. Robert L. Winkler. Rewarding Expertise in Probability Assessment, pages 127–140. Springer Netherlands, 1977. ISBN 9789401012768. https://doi.org/10.1007/978-94-010-1276-8_10.
  79. ranger: A fast implementation of random forests for high dimensional data in c++ and r. Journal of Statistical Software, 77(1), 2017. ISSN 1548-7660. https://doi.org/10.18637/jss.v077.i01.
  80. Michaël Zamo. Statistical Post-processing of Deterministic and Ensemble Wind Speed Forecasts on a Grid. Theses, Université Paris Saclay (COmUE), December 2016. URL https://theses.hal.science/tel-01598119.

Summary

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

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