Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
169 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 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

Convolutional GRU Network for Seasonal Prediction of the El Niño-Southern Oscillation (2306.10443v1)

Published 18 Jun 2023 in physics.ao-ph, cs.LG, and physics.data-an

Abstract: Predicting sea surface temperature (SST) within the El Ni~no-Southern Oscillation (ENSO) region has been extensively studied due to its significant influence on global temperature and precipitation patterns. Statistical models such as linear inverse model (LIM), analog forecasting (AF), and recurrent neural network (RNN) have been widely used for ENSO prediction, offering flexibility and relatively low computational expense compared to large dynamic models. However, these models have limitations in capturing spatial patterns in SST variability or relying on linear dynamics. Here we present a modified Convolutional Gated Recurrent Unit (ConvGRU) network for the ENSO region spatio-temporal sequence prediction problem, along with the Ni~no 3.4 index prediction as a down stream task. The proposed ConvGRU network, with an encoder-decoder sequence-to-sequence structure, takes historical SST maps of the Pacific region as input and generates future SST maps for subsequent months within the ENSO region. To evaluate the performance of the ConvGRU network, we trained and tested it using data from multiple large climate models. The results demonstrate that the ConvGRU network significantly improves the predictability of the Ni~no 3.4 index compared to LIM, AF, and RNN. This improvement is evidenced by extended useful prediction range, higher Pearson correlation, and lower root-mean-square error. The proposed model holds promise for improving our understanding and predicting capabilities of the ENSO phenomenon and can be broadly applicable to other weather and climate prediction scenarios with spatial patterns and teleconnections.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (54)
  1. ENSO as an integrating concept in earth science. \JournalTitleScience 314, 1740–1745 (2006).
  2. Climate anomalies in Europe associated with ENSO extremes. \JournalTitleInternational Journal of Climatology 12, 25–31 (1992).
  3. On the weakening relationship between the Indian monsoon and ENSO. \JournalTitleScience 284, 2156–2159 (1999).
  4. Glantz, M. H. et al. Currents of change: Impacts of El Niño and La Niña on climate and society (Cambridge University Press, 2001).
  5. The impact of El Niño–Southern Oscillation (ENSO) on winter and early spring US tornado outbreaks. \JournalTitleJournal of Applied Meteorology and Climatology 56, 2455–2478 (2017).
  6. Western North Pacific tropical cyclone intensity and ENSO. \JournalTitleJournal of Climate 18, 2996–3006 (2005).
  7. Squire, D. T. et al. Likelihood of unprecedented drought and fire weather during Australia’s 2019 megafires. \JournalTitleClimate and Atmospheric Science 4, 64 (2021).
  8. Interannual variations in fire weather, fire extent, and synoptic-scale circulation patterns in northern California and Oregon. \JournalTitleTheoretical and Applied Climatology 95, 349–360 (2009).
  9. Variations in tropical sea surface temperature and surface wind fields associated with the Southern Oscillation/El Niño. \JournalTitleMonthly Weather Review 110, 354–384 (1982).
  10. Trenberth, K. E. The definition of El Niño. \JournalTitleBulletin of the American Meteorological Society 78, 2771–2778 (1997).
  11. Indices of El Niño evolution. \JournalTitleJournal of Climate 14, 1697–1701 (2001).
  12. Translating climate forecasts into agricultural terms: Advances and challenges. \JournalTitleClimate Research 33, 27–41 (2006).
  13. Hammer, G. et al. Advances in application of climate prediction in agriculture. \JournalTitleAgricultural Systems 70, 515–553 (2001).
  14. Deterministic skill of ENSO predictions from the North American Multimodel Ensemble. \JournalTitleClimate Dynamics 53, 7215–7234 (2019).
  15. Prediction of Niño 3 sea surface temperatures using linear inverse modeling. \JournalTitleJournal of Climate 6, 1067–1076 (1993).
  16. The optimal growth of tropical sea surface temperature anomalies. \JournalTitleJournal of Climate 8, 1999–2024 (1995).
  17. Extended-range statistical ENSO prediction through operator-theoretic techniques for nonlinear dynamics. \JournalTitleScientific Reports 10, 1–15 (2020).
  18. A vector autoregressive ENSO prediction model. \JournalTitleJournal of Climate 28, 8511–8520 (2015).
  19. Lorenz, E. N. Atmospheric predictability as revealed by naturally occurring analogues. \JournalTitleJournal of Atmospheric Sciences 26, 636–646 (1969).
  20. Analog forecasting with dynamics-adapted kernels. \JournalTitleNonlinearity 29, 2888 (2016).
  21. Kernel analog forecasting: Multiscale test problems. \JournalTitleMultiscale Modeling & Simulation 19, 1011–1040 (2021).
  22. Deep learning for multi-year ENSO forecasts. \JournalTitleNature 573, 568–572 (2019).
  23. Long short-term memory. \JournalTitleNeural Computation 9, 1735–1780 (1997).
  24. Analyzing El Niño–Southern Oscillation predictability using long-short-term-memory models. \JournalTitleEarth and Space Science 6, 212–221 (2019).
  25. Shi, X. et al. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. \JournalTitleAdvances in Neural Information Processing Systems 28 (2015).
  26. Shi, X. et al. Deep learning for precipitation nowcasting: A benchmark and a new model. \JournalTitleAdvances in Neural Information Processing Systems 30 (2017).
  27. Delving deeper into convolutional networks for learning video representations. \JournalTitlearXiv preprint arXiv:1511.06432 (2015).
  28. ENSO-like interdecadal variability: 1900–93. \JournalTitleJournal of Climate 10, 1004–1020 (1997).
  29. El Niño and its impact on fire weather conditions in Alaska. \JournalTitleInternational Journal of Wildland Fire 10, 1–13 (2001).
  30. Recurrent neural networks. \JournalTitleDesign and Applications 5, 64–67 (2001).
  31. On the properties of neural machine translation: Encoder–decoder approaches. In Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation (SSST-8), 103–111 (2014).
  32. Graves, A. Generating sequences with recurrent neural networks. \JournalTitlearXiv preprint arXiv:1308.0850 (2013).
  33. Unsupervised learning for physical interaction through video prediction. \JournalTitleAdvances in Neural Information Processing Systems 29 (2016).
  34. Wang, Y. et al. Predrnn: A recurrent neural network for spatiotemporal predictive learning. \JournalTitleIEEE Transactions on Pattern Analysis and Machine Intelligence 45, 2208–2225 (2022).
  35. Continuous spatiotemporal transformers. \JournalTitlearXiv preprint arXiv:2301.13338 (2023).
  36. TCTN: A 3D-temporal convolutional transformer network for spatiotemporal predictive learning. \JournalTitlearXiv preprint arXiv:2112.01085 (2021).
  37. Keisler, R. Forecasting global weather with graph neural networks. \JournalTitlearXiv preprint arXiv:2202.07575 (2022).
  38. Lam, R. et al. Graphcast: Learning skillful medium-range global weather forecasting. \JournalTitlearXiv preprint arXiv:2212.12794 (2022).
  39. Ravuri, S. et al. Skilful precipitation nowcasting using deep generative models of radar. \JournalTitleNature 597, 672–677 (2021).
  40. Skillful climate forecasts of the tropical Indo-Pacific ocean using model-analogs. \JournalTitleJournal of Climate 31, 5437–5459 (2018).
  41. Operator Theoretic Aspects of Ergodic Theory, vol. 272 (Springer, 2015).
  42. Giannakis, D. Data-driven spectral decomposition and forecasting of ergodic dynamical systems. \JournalTitleApplied and Computational Harmonic Analysis 47, 338–396 (2019).
  43. Empirical evaluation of gated recurrent neural networks on sequence modeling. In NIPS 2014 Workshop on Deep Learning, December 2014 (2014).
  44. Adam: A method for stochastic optimization. \JournalTitleInternational Conference on Learning Representations (2015).
  45. Adaptive subgradient methods for online learning and stochastic optimization. \JournalTitleJournal of Machine Learning Research 12 (2011).
  46. Paszke, A. et al. PyTorch: An imperative style, high-performance deep learning library. \JournalTitleAdvances in Neural Information Processing Systems 32 (2019).
  47. Gent, P. R. et al. The community climate system model version 4. \JournalTitleJournal of Climate 24, 4973–4991 (2011).
  48. DeepAR: Probabilistic forecasting with autoregressive recurrent networks. \JournalTitleInternational Journal of Forecasting 36, 1181–1191 (2020).
  49. Alexandrov, A. et al. GluonTS: Probabilistic and neural time series modeling in Pyhon. \JournalTitleThe Journal of Machine Learning Research 21, 4629–4634 (2020).
  50. Delworth, T. L. et al. SPEAR: The next generation GFDL modeling system for seasonal to multidecadal prediction and projection. \JournalTitleJournal of Advances in Modeling Earth Systems 12 (2020).
  51. Van Vuuren, D. P. et al. A new scenario framework for climate change research: Scenario matrix architecture. \JournalTitleClimatic C 122, 373–386 (2014).
  52. O’Neill, B. C. et al. The scenario model intercomparison project (ScenarioMIP) for CMIP6. \JournalTitleGeoscientific Model Development 9, 3461–3482 (2016).
  53. Compo, G. P. et al. The twentieth century reanalysis project. \JournalTitleQuarterly Journal of the Royal Meteorological Society 137, 1–28 (2011).
  54. Inconsistencies between long-term trends in storminess derived from the 20CR reanalysis and observations. \JournalTitleJournal of Climate 26, 868–874 (2013).
Citations (2)
View on Semantic Scholar

Summary

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

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