Articles | Volume 14, issue 1
https://doi.org/10.5194/gmd-14-107-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-14-107-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Model description paper
| 
08 Jan 2021
Model description paper |
| 08 Jan 2021
| 08 Jan 2021
ClimateNet: an expert-labeled open dataset and deep learning architecture for enabling high-precision analyses of extreme weather
Prabhat
Lawrence Berkeley National Laboratory, Berkeley, CA, USA
Department of Earth and Planetary Science, University of California, Berkeley, CA, USA
Karthik Kashinath
CORRESPONDING AUTHOR
Lawrence Berkeley National Laboratory, Berkeley, CA, USA
Mayur Mudigonda
Terrafuse, Berkeley, CA, USA
Department of Earth and Planetary Science, University of California, Berkeley, CA, USA
Lukas Kapp-Schwoerer
ETH Zurich, Zürich, Switzerland
Andre Graubner
ETH Zurich, Zürich, Switzerland
Ege Karaismailoglu
ETH Zurich, Zürich, Switzerland
Leo von Kleist
ETH Zurich, Zürich, Switzerland
Thorsten Kurth
NVIDIA, Santa Clara, CA, USA
Annette Greiner
Lawrence Berkeley National Laboratory, Berkeley, CA, USA
Ankur Mahesh
Department of Earth and Planetary Science, University of California, Berkeley, CA, USA
Lawrence Berkeley National Laboratory, Berkeley, CA, USA
Kevin Yang
Department of Earth and Planetary Science, University of California, Berkeley, CA, USA
Colby Lewis
Department of Earth and Planetary Science, University of California, Berkeley, CA, USA
Jiayi Chen
Department of Earth and Planetary Science, University of California, Berkeley, CA, USA
Andrew Lou
Department of Earth and Planetary Science, University of California, Berkeley, CA, USA
Sathyavat Chandran
Department of Computer Science, Rice University, Houston, TX, USA
Ben Toms
Department of Atmospheric Science, Colorado State University, Fort Collins, CO, USA
Will Chapman
Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA, USA
Katherine Dagon
National Center for Atmospheric Research, Boulder, CO, USA
Christine A. Shields
National Center for Atmospheric Research, Boulder, CO, USA
Travis O'Brien
Department of Atmospheric Science, Indiana University, Bloomington, IN, USA
Lawrence Berkeley National Laboratory, Berkeley, CA, USA
Michael Wehner
Lawrence Berkeley National Laboratory, Berkeley, CA, USA
William Collins
Lawrence Berkeley National Laboratory, Berkeley, CA, USA
Department of Earth and Planetary Science, University of California, Berkeley, CA, USA
Related authors
Benjamin A. Toms, Karthik Kashinath, Prabhat, and Da Yang
Geosci. Model Dev., 14, 4495–4508, https://doi.org/10.5194/gmd-14-4495-2021, https://doi.org/10.5194/gmd-14-4495-2021, 2021
Short summary
Short summary
We test whether a type of machine learning called neural networks can be used trustfully within the geosciences. We do so by challenging the networks to understand the spatial patterns of a commonly studied geoscientific phenomenon. The neural networks can correctly identify the spatial patterns, which lends confidence that similar networks can be used for more uncertain problems. The results of this study may give geoscientists confidence when using neural networks in their research.
Travis A. O'Brien, Mark D. Risser, Burlen Loring, Abdelrahman A. Elbashandy, Harinarayan Krishnan, Jeffrey Johnson, Christina M. Patricola, John P. O'Brien, Ankur Mahesh, Prabhat, Sarahí Arriaga Ramirez, Alan M. Rhoades, Alexander Charn, Héctor Inda Díaz, and William D. Collins
Geosci. Model Dev., 13, 6131–6148, https://doi.org/10.5194/gmd-13-6131-2020, https://doi.org/10.5194/gmd-13-6131-2020, 2020
Short summary
Short summary
Researchers utilize various algorithms to identify extreme weather features in climate data, and we seek to answer this question: given a
plausibleweather event detector, how does uncertainty in the detector impact scientific results? We generate a suite of statistical models that emulate expert identification of weather features. We find that the connection between El Niño and atmospheric rivers – a specific extreme weather type – depends systematically on the design of the detector.
Ankur Mahesh, William D. Collins, Boris Bonev, Noah Brenowitz, Yair Cohen, Joshua Elms, Peter Harrington, Karthik Kashinath, Thorsten Kurth, Joshua North, Travis O'Brien, Michael Pritchard, David Pruitt, Mark Risser, Shashank Subramanian, and Jared Willard
Geosci. Model Dev., 18, 5575–5603, https://doi.org/10.5194/gmd-18-5575-2025, https://doi.org/10.5194/gmd-18-5575-2025, 2025
Short summary
Short summary
Simulating extreme weather events in a warming world is a challenging task for current weather and climate models. These models' computational cost poses a challenge in studying low-probability extreme weather. We use machine learning to construct a new probabilistic system. We give an in-depth explanation of how we constructed this system. We present a thorough pipeline to validate our method. Our method requires fewer computational resources than existing weather and climate models.
Ankur Mahesh, William D. Collins, Boris Bonev, Noah Brenowitz, Yair Cohen, Peter Harrington, Karthik Kashinath, Thorsten Kurth, Joshua North, Travis A. O'Brien, Michael Pritchard, David Pruitt, Mark Risser, Shashank Subramanian, and Jared Willard
Geosci. Model Dev., 18, 5605–5633, https://doi.org/10.5194/gmd-18-5605-2025, https://doi.org/10.5194/gmd-18-5605-2025, 2025
Short summary
Short summary
We use machine learning emulators to create a massive ensemble of simulated weather extremes. This ensemble provides a large sample size, which is essential to characterize the statistics of extreme weather events and study their physical mechanisms. Also, these ensembles can be beneficial to accurately forecast the probability of low-likelihood extreme weather.
William E. Chapman, Francine Schevenhoven, Judith Berner, Noel Keenlyside, Ingo Bethke, Ping-Gin Chiu, Alok Gupta, and Jesse Nusbaumer
Geosci. Model Dev., 18, 5451–5465, https://doi.org/10.5194/gmd-18-5451-2025, https://doi.org/10.5194/gmd-18-5451-2025, 2025
Short summary
Short summary
We introduce the first state-of-the-art atmosphere-connected supermodel, where two advanced atmospheric models share information in real time to form a new dynamical system. By synchronizing the models, particularly in storm track regions, we achieve better predictions without losing variability. This approach maintains key climate patterns and reduces bias in some variables compared to traditional models, demonstrating a useful technique for improving atmospheric simulations.
Malcolm J. Roberts, Kevin A. Reed, Qing Bao, Joseph J. Barsugli, Suzana J. Camargo, Louis-Philippe Caron, Ping Chang, Cheng-Ta Chen, Hannah M. Christensen, Gokhan Danabasoglu, Ivy Frenger, Neven S. Fučkar, Shabeh ul Hasson, Helene T. Hewitt, Huanping Huang, Daehyun Kim, Chihiro Kodama, Michael Lai, Lai-Yung Ruby Leung, Ryo Mizuta, Paulo Nobre, Pablo Ortega, Dominique Paquin, Christopher D. Roberts, Enrico Scoccimarro, Jon Seddon, Anne Marie Treguier, Chia-Ying Tu, Paul A. Ullrich, Pier Luigi Vidale, Michael F. Wehner, Colin M. Zarzycki, Bosong Zhang, Wei Zhang, and Ming Zhao
Geosci. Model Dev., 18, 1307–1332, https://doi.org/10.5194/gmd-18-1307-2025, https://doi.org/10.5194/gmd-18-1307-2025, 2025
Short summary
Short summary
HighResMIP2 is a model intercomparison project focusing on high-resolution global climate models, that is, those with grid spacings of 25 km or less in the atmosphere and ocean, using simulations of decades to a century in length. We are proposing an update of our simulation protocol to make the models more applicable to key questions for climate variability and hazard in present-day and future projections and to build links with other communities to provide more robust climate information.
Bo Dong, Paul Ullrich, Jiwoo Lee, Peter Gleckler, Kristin Chang, and Travis A. O'Brien
Geosci. Model Dev., 18, 961–976, https://doi.org/10.5194/gmd-18-961-2025, https://doi.org/10.5194/gmd-18-961-2025, 2025
Short summary
Short summary
A metrics package designed for easy analysis of atmospheric river (AR) characteristics and statistics is presented. The tool is efficient for diagnosing systematic AR bias in climate models and useful for evaluating new AR characteristics in model simulations. In climate models, landfalling AR precipitation shows dry biases globally, and AR tracks are farther poleward (equatorward) in the North and South Atlantic (South Pacific and Indian Ocean).
Ryan J. O'Loughlin, Dan Li, Richard Neale, and Travis A. O'Brien
Geosci. Model Dev., 18, 787–802, https://doi.org/10.5194/gmd-18-787-2025, https://doi.org/10.5194/gmd-18-787-2025, 2025
Short summary
Short summary
We draw from traditional climate modeling practices to make recommendations for machine-learning (ML)-driven climate science. Our intended audience is climate modelers who are relatively new to ML. We show how component-level understanding – obtained when scientists can link model behavior to parts within the overall model – should guide the development and evaluation of ML models. Better understanding yields a stronger basis for trust in the models. We highlight several examples to demonstrate.
Arne M. E. Winguth, Mikaela Brown, Pincelli Hull, Elizabeth Griffith, Christine Shields, Ellen Thomas, and Cornelia Winguth
EGUsphere, https://doi.org/10.5194/egusphere-2024-4209, https://doi.org/10.5194/egusphere-2024-4209, 2025
Short summary
Short summary
The Paleocene-Eocene Thermal Maximum (PETM) about 56 million years ago is characterized by a rapid perturbation of the global carbon cycle. Comparison of sedimentary records with results from a comprehensive Earth system model suggest that environmental changes including benthic foraminifera extinction may have caused by a massive carbon input at the PETM and associate collapse of the ocean circulation due to the greenhouse-gas induced warming.
Xiaodong Zhang, Brett J. Tipple, Jiang Zhu, William D. Rush, Christian A. Shields, Joseph B. Novak, and James C. Zachos
Clim. Past, 20, 1615–1626, https://doi.org/10.5194/cp-20-1615-2024, https://doi.org/10.5194/cp-20-1615-2024, 2024
Short summary
Short summary
This study is motivated by the current anthropogenic-warming-forced transition in regional hydroclimate. We use observations and model simulations during the Paleocene–Eocene Thermal Maximum (PETM) to constrain the regional/local hydroclimate response. Our findings, based on multiple observational evidence within the context of model output, suggest a transition toward greater aridity and precipitation extremes in central California during the PETM.
Jiwoo Lee, Peter J. Gleckler, Min-Seop Ahn, Ana Ordonez, Paul A. Ullrich, Kenneth R. Sperber, Karl E. Taylor, Yann Y. Planton, Eric Guilyardi, Paul Durack, Celine Bonfils, Mark D. Zelinka, Li-Wei Chao, Bo Dong, Charles Doutriaux, Chengzhu Zhang, Tom Vo, Jason Boutte, Michael F. Wehner, Angeline G. Pendergrass, Daehyun Kim, Zeyu Xue, Andrew T. Wittenberg, and John Krasting
Geosci. Model Dev., 17, 3919–3948, https://doi.org/10.5194/gmd-17-3919-2024, https://doi.org/10.5194/gmd-17-3919-2024, 2024
Short summary
Short summary
We introduce an open-source software, the PCMDI Metrics Package (PMP), developed for a comprehensive comparison of Earth system models (ESMs) with real-world observations. Using diverse metrics evaluating climatology, variability, and extremes simulated in thousands of simulations from the Coupled Model Intercomparison Project (CMIP), PMP aids in benchmarking model improvements across generations. PMP also enables efficient tracking of performance evolutions during ESM developments.
Ankur Mahesh, Travis A. O'Brien, Burlen Loring, Abdelrahman Elbashandy, William Boos, and William D. Collins
Geosci. Model Dev., 17, 3533–3557, https://doi.org/10.5194/gmd-17-3533-2024, https://doi.org/10.5194/gmd-17-3533-2024, 2024
Short summary
Short summary
Atmospheric rivers (ARs) are extreme weather events that can alleviate drought or cause billions of US dollars in flood damage. We train convolutional neural networks (CNNs) to detect ARs with an estimate of the uncertainty. We present a framework to generalize these CNNs to a variety of datasets of past, present, and future climate. Using a simplified simulation of the Earth's atmosphere, we validate the CNNs. We explore the role of ARs in maintaining energy balance in the Earth system.
John T. Fasullo, Jean-Christophe Golaz, Julie M. Caron, Nan Rosenbloom, Gerald A. Meehl, Warren Strand, Sasha Glanville, Samantha Stevenson, Maria Molina, Christine A. Shields, Chengzhu Zhang, James Benedict, Hailong Wang, and Tony Bartoletti
Earth Syst. Dynam., 15, 367–386, https://doi.org/10.5194/esd-15-367-2024, https://doi.org/10.5194/esd-15-367-2024, 2024
Short summary
Short summary
Climate model large ensembles provide a unique and invaluable means for estimating the climate response to external forcing agents and quantify contrasts in model structure. Here, an overview of the Energy Exascale Earth System Model (E3SM) version 2 large ensemble is given along with comparisons to large ensembles from E3SM version 1 and versions 1 and 2 of the Community Earth System Model. The paper provides broad and important context for users of these ensembles.
Arjun Babu Nellikkattil, Danielle Lemmon, Travis Allen O'Brien, June-Yi Lee, and Jung-Eun Chu
Geosci. Model Dev., 17, 301–320, https://doi.org/10.5194/gmd-17-301-2024, https://doi.org/10.5194/gmd-17-301-2024, 2024
Short summary
Short summary
This study introduces a new computational framework called Scalable Feature Extraction and Tracking (SCAFET), designed to extract and track features in climate data. SCAFET stands out by using innovative shape-based metrics to identify features without relying on preconceived assumptions about the climate model or mean state. This approach allows more accurate comparisons between different models and scenarios.
Michelle L. Maclennan, Jan T. M. Lenaerts, Christine A. Shields, Andrew O. Hoffman, Nander Wever, Megan Thompson-Munson, Andrew C. Winters, Erin C. Pettit, Theodore A. Scambos, and Jonathan D. Wille
The Cryosphere, 17, 865–881, https://doi.org/10.5194/tc-17-865-2023, https://doi.org/10.5194/tc-17-865-2023, 2023
Short summary
Short summary
Atmospheric rivers are air masses that transport large amounts of moisture and heat towards the poles. Here, we use a combination of weather observations and models to quantify the amount of snowfall caused by atmospheric rivers in West Antarctica which is about 10 % of the total snowfall each year. We then examine a unique event that occurred in early February 2020, when three atmospheric rivers made landfall over West Antarctica in rapid succession, leading to heavy snowfall and surface melt.
Sjoukje Y. Philip, Sarah F. Kew, Geert Jan van Oldenborgh, Faron S. Anslow, Sonia I. Seneviratne, Robert Vautard, Dim Coumou, Kristie L. Ebi, Julie Arrighi, Roop Singh, Maarten van Aalst, Carolina Pereira Marghidan, Michael Wehner, Wenchang Yang, Sihan Li, Dominik L. Schumacher, Mathias Hauser, Rémy Bonnet, Linh N. Luu, Flavio Lehner, Nathan Gillett, Jordis S. Tradowsky, Gabriel A. Vecchi, Chris Rodell, Roland B. Stull, Rosie Howard, and Friederike E. L. Otto
Earth Syst. Dynam., 13, 1689–1713, https://doi.org/10.5194/esd-13-1689-2022, https://doi.org/10.5194/esd-13-1689-2022, 2022
Short summary
Short summary
In June 2021, the Pacific Northwest of the US and Canada saw record temperatures far exceeding those previously observed. This attribution study found such a severe heat wave would have been virtually impossible without human-induced climate change. Assuming no nonlinear interactions, such events have become at least 150 times more common, are about 2 °C hotter and will become even more common as warming continues. Therefore, adaptation and mitigation are urgently needed to prepare society.
Mari R. Tye, Katherine Dagon, Maria J. Molina, Jadwiga H. Richter, Daniele Visioni, Ben Kravitz, and Simone Tilmes
Earth Syst. Dynam., 13, 1233–1257, https://doi.org/10.5194/esd-13-1233-2022, https://doi.org/10.5194/esd-13-1233-2022, 2022
Short summary
Short summary
We examined the potential effect of stratospheric aerosol injection (SAI) on extreme temperature and precipitation. SAI may cause daytime temperatures to cool but nighttime to warm. Daytime cooling may occur in all seasons across the globe, with the largest decreases in summer. In contrast, nighttime warming may be greatest at high latitudes in winter. SAI may reduce the frequency and intensity of extreme rainfall. The combined changes may exacerbate drying over parts of the global south.
Ashesh Chattopadhyay, Mustafa Mustafa, Pedram Hassanzadeh, Eviatar Bach, and Karthik Kashinath
Geosci. Model Dev., 15, 2221–2237, https://doi.org/10.5194/gmd-15-2221-2022, https://doi.org/10.5194/gmd-15-2221-2022, 2022
Short summary
Short summary
There is growing interest in data-driven weather forecasting, i.e., to predict the weather by using a deep neural network that learns from the evolution of past atmospheric patterns. Here, we propose three components to add to the current data-driven weather forecast models to improve their performance. These components involve a feature that incorporates physics into the neural network, a method to add data assimilation, and an algorithm to use several different time intervals in the forecast.
Claudia Tebaldi, Kalyn Dorheim, Michael Wehner, and Ruby Leung
Earth Syst. Dynam., 12, 1427–1501, https://doi.org/10.5194/esd-12-1427-2021, https://doi.org/10.5194/esd-12-1427-2021, 2021
Short summary
Short summary
We address the question of how large an initial condition ensemble of climate model simulations should be if we are concerned with accurately projecting future changes in temperature and precipitation extremes. We find that for most cases (and both models considered), an ensemble of 20–25 members is sufficient for many extreme metrics, spatial scales and time horizons. This may leave computational resources to tackle other uncertainties in climate model simulations with our ensembles.
Benjamin A. Toms, Karthik Kashinath, Prabhat, and Da Yang
Geosci. Model Dev., 14, 4495–4508, https://doi.org/10.5194/gmd-14-4495-2021, https://doi.org/10.5194/gmd-14-4495-2021, 2021
Short summary
Short summary
We test whether a type of machine learning called neural networks can be used trustfully within the geosciences. We do so by challenging the networks to understand the spatial patterns of a commonly studied geoscientific phenomenon. The neural networks can correctly identify the spatial patterns, which lends confidence that similar networks can be used for more uncertain problems. The results of this study may give geoscientists confidence when using neural networks in their research.
Katherine Dagon, Benjamin M. Sanderson, Rosie A. Fisher, and David M. Lawrence
Adv. Stat. Clim. Meteorol. Oceanogr., 6, 223–244, https://doi.org/10.5194/ascmo-6-223-2020, https://doi.org/10.5194/ascmo-6-223-2020, 2020
Short summary
Short summary
Uncertainties in land model projections are important to understand in order to build confidence in Earth system modeling. In this paper, we introduce a framework for estimating uncertain land model parameters with machine learning. This method increases the computational efficiency of this process relative to traditional hand tuning approaches and provides objective methods to assess the results. We further identify key processes and parameters that are important for accurate land modeling.
Travis A. O'Brien, Mark D. Risser, Burlen Loring, Abdelrahman A. Elbashandy, Harinarayan Krishnan, Jeffrey Johnson, Christina M. Patricola, John P. O'Brien, Ankur Mahesh, Prabhat, Sarahí Arriaga Ramirez, Alan M. Rhoades, Alexander Charn, Héctor Inda Díaz, and William D. Collins
Geosci. Model Dev., 13, 6131–6148, https://doi.org/10.5194/gmd-13-6131-2020, https://doi.org/10.5194/gmd-13-6131-2020, 2020
Short summary
Short summary
Researchers utilize various algorithms to identify extreme weather features in climate data, and we seek to answer this question: given a
plausibleweather event detector, how does uncertainty in the detector impact scientific results? We generate a suite of statistical models that emulate expert identification of weather features. We find that the connection between El Niño and atmospheric rivers – a specific extreme weather type – depends systematically on the design of the detector.
Mark D. Risser and Michael F. Wehner
Adv. Stat. Clim. Meteorol. Oceanogr., 6, 115–139, https://doi.org/10.5194/ascmo-6-115-2020, https://doi.org/10.5194/ascmo-6-115-2020, 2020
Short summary
Short summary
Evaluation of modern high-resolution global climate models often does not account for the geographic location of the underlying weather station data. In this paper, we quantify the impact of geographic sampling on the relative performance of climate model representations of precipitation extremes over the United States. We find that properly accounting for the geographic sampling of weather stations can significantly change the assessment of model performance.
Cited articles
Allen, M. and Ingram, W.: Constraints on Future Changes in Climate and the
Hydrologic Cycle, Nature, 419, 224–32, https://doi.org/10.1038/nature01092, 2002. a, b
Bonfanti, C., Stewart, J., Maksimovic, S., Hall, D., Govett, M., Trailovic, L., and Jankov, I.: Detecting Extratropical and Tropical Cyclone Regions of
Interest (ROI) in Satellite Data using Deep Learning, available at: https://ui.adsabs.harvard.edu/abs/2018AGUFM.H31H1992B/abstract (last access: 14 December 2020),
2018a. a
Bonfanti, C., Trailovic, L., Stewart, J., and Govett, M.: Machine Learning:
Defining Worldwide Cyclone Labels for Training,
2018 21st International Conference on Information Fusion (FUSION), IEEE, https://doi.org/10.23919/ICIF.2018.8455276, 2018b. a
Brenowitz, N. D. and Bretherton, C. S.: Prognostic validation of a neural
network unified physics parameterization, Geophys. Res. Lett., 45,
6289–6298, 2018. a
Chapman, W., Subramanian, A., Delle Monache, L., Xie, S., and Ralph, F.:
Improving Atmospheric River Forecasts With Machine Learning, Geophys.
Res. Lett., 46, 10627–10635, 2019. a
Chavas, D., Lin, N., and Emanuel, K.: A Model for the Complete Radial Structure
of the Tropical Cyclone Wind Field. Part I: Comparison with Observed
Structure, J. Atmos. Sci., 72, 3647–3662,
https://doi.org/10.1175/JAS-D-15-0014.1, 2015. a
Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., and Adam, H.:
Encoder-Decoder with Atrous Separable Convolution for Semantic Image
Segmentation, arXiv e-prints, arXiv:1802.02611, 2018. a, b
Chollet, F.: Xception: Deep Learning with Depthwise Separable Convolutions,
arXiv e-prints, arXiv:1610.02357, 2016. a
Dettinger, M. D., Ralph, F. M., Das, T., Neiman, P. J., and Cayan, D. R.:
Atmospheric rivers, floods and the water resources of California, Water, 3,
445–478, 2011. a
Gershunov, A., Shulgina, T., Clemesha, R. E., Guirguis, K., Pierce, D. W.,
Dettinger, M. D., Lavers, D. A., Cayan, D. R., Polade, S. D., Kalansky, J.,
and Ralph, F. M.: Precipitation regime change in Western North America: the role of
Atmospheric Rivers, Sci. Rep., 9, 1–11, 2019. a
Hodges, K. I.: Feature Tracking on the Unit Sphere, Mon. Weather Rev.,
123, 3458–3465, 1995. a
Hong, S., Kim, S., Joh, M., and Song, S.-K.: Globenet: Convolutional neural
networks for typhoon eye tracking from remote sensing imagery, arXiv preprint
arXiv:1708.03417, 2017. a
Kapp-Schwoerer, L., Graubner, A., Karaismailoglu, E., von Kleist, L., and
Greiner, A.: ClimateNet dataset and trained deep learning model, available at: https://portal.nersc.gov/project/ClimateNet/, last access: 14 December 2020. a
Kingma, D. P. and Ba, J.: Adam: A Method for Stochastic Optimization,
arXiv e-prints, arXiv:1412.6980, 2014. a
Knutson, T., Camargo, S. J., Chan, J. C. L., Emanuel, K., Ho, C.-H., Kossin,
J., Mohapatra, M., Satoh, M., Sugi, M., Walsh, K., and Wu, L.: Tropical
Cyclones and Climate Change Assessment: Part II. Projected Response to
Anthropogenic Warming, B. Am. Meteorol. Soc., 101,
E303–E322, https://doi.org/10.1175/BAMS-D-18-0194.1, 2019. a, b
Kurth, T., Treichler, S., Romero, J., Mudigonda, M., Luehr, N., Phillips, E.,
Mahesh, A., Matheson, M., Deslippe, J., Fatica, M., Prabhat, and Houston, M.: Exascale deep
learning for climate analytics, in: Proceedings of the International
Conference for High Performance Computing, Networking, Storage, and Analysis,
p. 51, IEEE Press, 2018. a, b, c, d
LeCun, Y., Bengio, Y., and Hinton, G.: Deep learning, Nature, 521, 436–444, 2015. a
Levine, S., Finn, C., Darrell, T., and Abbeel, P.: End-to-end training of deep visuomotor policies, J. Mach. Learn. Res., 17,
1334–1373, 2016. a
Liu, Y., Racah, E., Correa, J., Khosrowshahi, A., Lavers, D., Kunkel, K.,
Wehner, M., and Collins, W.: Application of deep convolutional neural
networks for detecting extreme weather in climate datasets, arXiv preprint
arXiv:1605.01156, 2016. a
Lotter, W., Sorensen, G., and Cox, D.: A multi-scale CNN and curriculum
learning strategy for mammogram classification, in: Deep Learning in Medical
Image Analysis and Multimodal Learning for Clinical Decision Support,
169–177, Springer, 2017. a
Mahesh, A., O'Brien, T., Collins, W., Prabhat, Kashinath, K., and Mudigonda,
M.: Probabilistic Detection of Extreme Weather Using Deep Learning Methods, 99th Annual Meeting of the American Meteorological Society, 6–10 January 2019, available at:
https://ams.confex.com/ams/2019Annual/webprogram/Paper354370.html (last access: 14 December 2020), 2019a. a
Mahesh, A., Evans, M., Jain, G., Castillo, M., Lima, A., Lunghino, B., Gupta,
H., Gaitan, C., Hunt, J. K., Tavasoli, O., Brown, P. T., and Balaji, V.: Forecasting El Niño
with Convolutional and Recurrent Neural Networks, 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada, 8–14 December 2019b. a
McGovern, A., Elmore, K. L., Gagne, D. J., Haupt, S. E., Karstens, C. D.,
Lagerquist, R., Smith, T., and Williams, J. K.: Using artificial intelligence
to improve real-time decision-making for high-impact weather, B.
Am. Meteorol. Soc., 98, 2073–2090, 2017. a
Mitchell, D., AchutaRao, K., Allen, M., Bethke, I., Beyerle, U., Ciavarella, A., Forster, P. M., Fuglestvedt, J., Gillett, N., Haustein, K., Ingram, W., Iversen, T., Kharin, V., Klingaman, N., Massey, N., Fischer, E., Schleussner, C.-F., Scinocca, J., Seland, Ø., Shiogama, H., Shuckburgh, E., Sparrow, S., Stone, D., Uhe, P., Wallom, D., Wehner, M., and Zaaboul, R.: Half a degree additional warming, prognosis and projected impacts (HAPPI): background and experimental design, Geosci. Model Dev., 10, 571–583, https://doi.org/10.5194/gmd-10-571-2017, 2017. a, b
Neu, U., Akperov, M. G., Bellenbaum, N., Benestad, R., Blender, R., Caballero, R., Cocozza, A., Dacre, H. F., Feng, Y., Fraedrich, K., Grieger, J., Gulev, S., Hanley, J., Hewson, T., Inatsu, M., Keay, K., Kew, S. F., Kindem, I., Leckebusch, G. C., Liberato, M. L. R., Lionello, P., Mokhov, I. I., Pinto, J. G., Raible, C. C., Reale, M., Rudeva, I., Schuster, M., Simmonds, I., Sinclair, M., Sprenger, M., Tilinina, N. D., Trigo, I. F., Ulbrich, S., Ulbrich, U., Wang, X. L., and Wernli, H.: IMILAST: A Community Effort to Intercompare Extratropical Cyclone Detection and Tracking Algorithms, B. Am. Meteorol. Soc., 94, 529–547,
https://doi.org/10.1175/BAMS-D-11-00154.1, 2013. a, b, c, d
NOAA: ENSO Indices, available at: https://www.weather.gov/fwd/indices (last access: 14 December 2020), 2019. a
O'Brien, T. A., Risser, M. D., Loring, B., Elbashandy, A. A., Krishnan, H., Johnson, J., Patricola, C. M., O'Brien, J. P., Mahesh, A., Prabhat, Arriaga Ramirez, S., Rhoades, A. M., Charn, A., Inda Díaz, H., and Collins, W. D.: Detection of Atmospheric Rivers with Inline Uncertainty Quantification: TECA-BARD v1.0, Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2020-55, in review, 2020. a
O'Gorman, P. A. and Dwyer, J. G.: Using machine learning to parameterize moist convection: Potential for modeling of climate, climate change, and extreme
events, J. Adv. Model. Earth Sy., 10, 2548–2563, 2018. a
O’Brien, T. A., Collins, W. D., Rauscher, S. A., and Ringler, T. D.: Reducing
the computational cost of the ECF using a nuFFT: A fast and objective
probability density estimation method, Comput. Stat. Data An., 79, 222–234, https://doi.org/10.1016/j.csda.2014.06.002,
2014. a
Pall, P., Allen, M., and Stone, D. A.: Testing the Clausius–Clapeyron
constraint on changes in extreme precipitation under CO2 warming, Clim.
Dynam., 28, 351–363, 2007. a
Patricola, C. and Wehner, M.: Anthropogenic influences on major tropical
cyclone events, Nature, 563, 339–346, https://doi.org/10.1038/s41586-018-0673-2, 2018. a, b
Racah, E., Beckham, C., Maharaj, T., Ebrahimi Kahou, S., Prabhat, M., and Pal, C.: ExtremeWeather: A large-scale climate dataset for semi-supervised
detection, localization, and understanding of extreme weather events, 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA, 3405–3416, 2017. a
Risser, M. D. and Wehner, M. F.: Attributable Human-Induced Changes in the
Likelihood and Magnitude of the Observed Extreme Precipitation during
Hurricane Harvey, Geophys. Res. Lett., 44, 12457–12464,
https://doi.org/10.1002/2017GL075888, 2017. a, b
Russell, B. C., Torralba, A., Murphy, K. P., and Freeman, W. T.: LabelMe: a
database and web-based tool for image annotation, Int. J.
Comput. vision, 77, 157–173, 2008. a
Shields, C. A., Rutz, J. J., Leung, L.-Y., Ralph, F. M., Wehner, M., Kawzenuk, B., Lora, J. M., McClenny, E., Osborne, T., Payne, A. E., Ullrich, P., Gershunov, A., Goldenson, N., Guan, B., Qian, Y., Ramos, A. M., Sarangi, C., Sellars, S., Gorodetskaya, I., Kashinath, K., Kurlin, V., Mahoney, K., Muszynski, G., Pierce, R., Subramanian, A. C., Tome, R., Waliser, D., Walton, D., Wick, G., Wilson, A., Lavers, D., Prabhat, Collow, A., Krishnan, H., Magnusdottir, G., and Nguyen, P.: Atmospheric River Tracking Method Intercomparison Project (ARTMIP): project goals and experimental design, Geosci. Model Dev., 11, 2455–2474, https://doi.org/10.5194/gmd-11-2455-2018, 2018. a, b, c, d
Toms, B. A., Kashinath, K., Prabhat, and Yang, D.: Testing the Reliability of Interpretable Neural Networks in Geoscience Using the Madden-Julian Oscillation, Geosci. Model Dev. Discuss. [preprint], https://doi.org/10.5194/gmd-2020-152, in review, 2020. a
Ullrich, P. A. and Zarzycki, C. M.: TempestExtremes: a framework for scale-insensitive pointwise feature tracking on unstructured grids, Geosci. Model Dev., 10, 1069–1090, https://doi.org/10.5194/gmd-10-1069-2017, 2017. a, b
van Oldenborgh, G. J., van der Wiel, K., Sebastian, A., Singh, R., Arrighi, J., Otto, F., Haustein, K., Li, S., Vecchi, G., and Cullen, H.: Attribution of extreme rainfall from Hurricane Harvey, August 2017, Environ. Res. Lett., 12, 124009, https://doi.org/10.1088/1748-9326/aa9ef2, 2017. a, b
Walsh, K., Lavender, S., Murakami, H., Scoccimarro, E., Caron, L.-P., and
Ghantous, M.: The Tropical Cyclone Climate Model Intercomparison Project,
Springer Netherlands, Dordrecht, 24 pp., https://doi.org/10.1007/978-90-481-9510-7_1,2010. a, b
Wang, S.-Y. S., Zhao, L., Yoon, J.-H., Klotzbach, P., and Gillies, R. R.:
Quantitative attribution of climate effects on Hurricane Harvey's extreme
rainfall in Texas, Environ. Res. Lett., 13, 054014,
https://doi.org/10.1088/1748-9326/aabb85, 018. a, b
Wehner, M. F., Reed, K. A., Li, F., Bacmeister, J., Chen, C.-T., Paciorek, C., Gleckler, P. J., Sperber, K. R., Collins, W. D., Gettelman, A., and Jablonowski, C.: The
effect of horizontal resolution on simulation quality in the Community
Atmospheric Model, CAM5. 1, J. Adv. Model. Earth Sy., 6,
980–997, 2014. a, b
Wehner, M. F., Reed, K. A., Loring, B., Stone, D., and Krishnan, H.: Changes in tropical cyclones under stabilized 1.5 and 2.0 ∘C global warming scenarios as simulated by the Community Atmospheric Model under the HAPPI protocols, Earth Syst. Dynam., 9, 187–195, https://doi.org/10.5194/esd-9-187-2018, 2018. a, b, c, d
Weinshall, D., Cohen, G., and Amir, D.: Curriculum learning by transfer
learning: Theory and experiments with deep networks, arXiv preprint
arXiv:1802.03796, 2018.
a
Zamir, A. R., Sax, A., Shen, W., Guibas, L. J., Malik, J., and Savarese, S.:
Taskonomy: Disentangling task transfer learning, in: Proceedings of the IEEE
Conference on Computer Vision and Pattern Recognition, IEEE, 3712–3722, 2018. a
Short summary
Detecting extreme weather events is a crucial step in understanding how they change due to climate change. Deep learning (DL) is remarkable at pattern recognition; however, it works best only when labeled datasets are available. We create
ClimateNet– an expert-labeled curated dataset – to train a DL model for detecting weather events and predicting changes in extreme precipitation. This work paves the way for DL-based automated, high-fidelity, and highly precise analytics of climate data.
Detecting extreme weather events is a crucial step in understanding how they change due to...
