Hostname: page-component-76fb5796d-dfsvx Total loading time: 0 Render date: 2024-04-25T14:59:41.599Z Has data issue: false hasContentIssue false
  • English
  • Français

Can Machines Learn How Clouds Work? The Epistemic Implications of Machine Learning Methods in Climate Science

Published online by Cambridge University Press:  01 January 2022

Suzanne Kawamleh
Get access Rights & Permissions [Opens in a new window]

Abstract

Scientists and decision makers rely on climate models for predictions concerning future climate change. Traditionally, physical processes that are key to predicting extreme events are either directly represented (resolved) or indirectly represented (parameterized). Scientists are now replacing physically based parameterizations with neural networks that do not represent physical processes directly or indirectly. I analyze the epistemic implications of this method and argue that it undermines the reliability of model predictions. I attribute the widespread failure in neural network generalizability to the lack of process representation. The representation of climate processes adds significant and irreducible value to the reliability of climate model predictions.

Type
Computer Simulation and Computer Science
Information
Philosophy of Science , Volume 88 , Issue 5 , December 2021 , pp. 1008 - 1020
Copyright
Copyright 2021 by the Philosophy of Science Association. All rights reserved.

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

†

I would like to thank Gary Ebbs, Elisabeth Lloyd, and Greg Lusk for their detailed comments on this article. This article also benefited greatly from feedback from participants of the Data Science in Climate and Climate Impact Research Workshop at ETH Zurich, particularly from Julie Jebeile, Tim Raz, and Vincent Lam.

References

Alvarado, Rafael, and Humphreys, Paul. 2017. “Big Data, Thick Mediation, and Representational Opacity.” New Literary History 48 (4): 729–49.CrossRefGoogle Scholar
Betts, Alan K., and Miller, M. J.. 1984. A New Convective Adjustment Scheme. Technical report no. 43. Reading: European Centre for Medium Range Weather Forecasts.Google Scholar
Bony, Sandrine, et al. 2015. “Clouds, Circulation and Climate Sensitivity.” Nature Geoscience 8, no. 4 (April): 261–68.CrossRefGoogle Scholar
Dueben, Peter D., and Bauer, Peter. 2018. “Challenges and Design Choices for Global Weather and Climate Models Based on Machine Learning.” Geoscientific Model Development 11 (10): 3999–4009.CrossRefGoogle Scholar
Ebert-Uphoff, Imme, and Deng, Yi. 2015. “Identifying Physical Interactions from Climate Data: Challenges and Opportunities.” Computing in Science and Engineering 17 (6): 27–34.CrossRefGoogle Scholar
Ganguly, A. R., Kodra, E. A., Agrawal, Ankit, Banerjee, A., Boriah, S., Chatterjee, S. N., Chatterjee, S. O., Choudhary, A., Das, D., and Faghmous, J.. 2014. “Toward Enhanced Understanding and Projections of Climate Extremes Using Physics-Guided Data Mining Techniques.” Nonlinear Processes in Geophysics 21 (4): 777–95.CrossRefGoogle Scholar
Gray, Jim. 2007. “Data Management: Past, Present, and Future.” ArXiv, Cornell University. https://arxiv.org/abs/cs/0701156.Google Scholar
McGovern, Amy, Lagerquist, Ryan, Gagne, David John, Jergensen, G. Eli, Elmore, Kimberly L., Homeyer, Cameron R., and Smith, Travis. 2019. “Making the Black Box More Transparent: Understanding the Physical Implications of Machine Learning.” Bulletin of the American Meteorological Society 100 (11): 2175–99.CrossRefGoogle Scholar
Monteleoni, Claire, Schmidt, Gavin A., Alexander, Francis J., Niculescu-Mizil, Alexandru, Steinhaeuser, Karsten, Tippett, Michael, Banerjee, Arindam, Blumenthal, M. Benno, Ganguly, Auroop R., Smerdon, Jason E., and Tedesco, M.. 2013. “Climate Informatics.” In Computational Intelligent Data Analysis for Sustainable Development, ed. Yu, T., Chawla, N., and Simoff, S., 81–126. Data Mining and Knowledge Discovery Series. London: Chapman & Hall.Google Scholar
O’Gorman, Paul A., and Dwyer, John G.. 2018. “Using Machine Learning to Parameterize Moist Convection: Potential for Modeling of Climate, Climate Change, and Extreme Events.” Journal of Advances in Modeling Earth Systems 10 (10): 2548–63.Google Scholar
Parker, Wendy S. 2009. “Confirmation and Adequacy-for-Purpose in Climate Modelling.” In Proceedings of the Aristotelian Society, Supplementary Volumes 83:233–49.CrossRefGoogle Scholar
Prein, Andreas F., et al. 2015. “A Review on Regional Convection-Permitting Climate Modeling: Demonstrations, Prospects, and Challenges.” Reviews of Geophysics 53 (2): 323–61.CrossRefGoogle ScholarPubMed
Rasp, Stephan, Pritchard, Michael S., and Gentine, Pierre. 2018. “Deep Learning to Represent Subgrid Processes in Climate Models.” Proceedings of the National Academy of Sciences 115 (39): 9684–89.CrossRefGoogle ScholarPubMed
Scher, Sebastian, and Messori, Gabriele. 2019. “Generalization Properties of Feed-Forward Neural Networks Trained on Lorenz Systems.” Nonlinear Processes in Geophysics 26 (4): 381–99.CrossRefGoogle Scholar
Schmidhuber, Jürgen. 2015. “Deep Learning in Neural Networks: An Overview.” Neural Networks 61:85–117.CrossRefGoogle ScholarPubMed
Steinhaeuser, Karsten, Chawla, Nitesh V., and Ganguly, Auroop R.. 2010. “Complex Networks in Climate Science: Progress, Opportunities and Challenges.” In Proceedings of the 2010 Conference on Intelligent Data Understanding, CIDU 2010, October 5–6, 2010, Mountain View, California, USA, ed. Srivastava, Ashok N., Chawla, Nitesh V., Yu, Philip S., and Melby, Paul, 16–26. Moffett Field, CA: NASA Ames Research Center.Google Scholar
Stocker, T. F., et al. 2001. “Physical Climate Processes and Feedbacks.” In Climate Change 2001: The Scientific Basis, ed. Houghton, J. T. et al., 417–70. Cambridge: Cambridge University Press.Google Scholar
Thompson, Robert M. Jr., Payne, Steven W., Recker, Ernest E., and Reed, Richard J.. 1979. “Structure and Properties of Synoptic-Scale Wave Disturbances in the Intertropical Convergence Zone of the Eastern Atlantic.” Journal of the Atmospheric Sciences 36 (1): 53–72.2.0.CO;2>CrossRefGoogle Scholar
Yuval, Janni, and O’Gorman, Paul A.. 2020. “Stable Machine-Learning Parameterization of Subgrid Processes for Climate Modeling at a Range of Resolutions.” Nature Communications 11 (1): 1–10.CrossRefGoogle Scholar
Zelinka, Mark D., Randall, David A., Webb, Mark J., and Klein, Stephen A.. 2017. “Clearing Clouds of Uncertainty.” Nature Climate Change 7 (10): 674–78.CrossRefGoogle Scholar