Deep Learning Atmospheric Models Reliably Simulate Out-of-Sample Land Heat and Cold Wave Frequencies

• URL: http://arxiv.org/abs/2507.03176v1
• Date: Thu, 03 Jul 2025 21:09:27 GMT
• Title: Deep Learning Atmospheric Models Reliably Simulate Out-of-Sample Land Heat and Cold Wave Frequencies
• Authors: Zilu Meng, Gregory J. Hakim, Wenchang Yang, Gabriel A. Vecchi,
• Abstract summary: We evaluate two deep learning (DL)-based general circulation models (GCMs) against a conventional high-resolution land-atmosphere model (HiRAM)<n>Both DL models generalize successfully to unseen climate conditions, broadly reproducing the frequency and spatial patterns of heatwave and cold wave events during 1900-1960.<n>An exception is over portions of North Asia and North America, where all models perform poorly during 1940-1960.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Deep learning (DL)-based general circulation models (GCMs) are emerging as fast simulators, yet their ability to replicate extreme events outside their training range remains unknown. Here, we evaluate two such models -- the hybrid Neural General Circulation Model (NGCM) and purely data-driven Deep Learning Earth System Model (DL\textit{ESy}M) -- against a conventional high-resolution land-atmosphere model (HiRAM) in simulating land heatwaves and coldwaves. All models are forced with observed sea surface temperatures and sea ice over 1900-2020, focusing on the out-of-sample early-20th-century period (1900-1960). Both DL models generalize successfully to unseen climate conditions, broadly reproducing the frequency and spatial patterns of heatwave and cold wave events during 1900-1960 with skill comparable to HiRAM. An exception is over portions of North Asia and North America, where all models perform poorly during 1940-1960. Due to excessive temperature autocorrelation, DL\textit{ESy}M tends to overestimate heatwave and cold wave frequencies, whereas the physics-DL hybrid NGCM exhibits persistence more similar to HiRAM.

Related papers

• Physics-constrained generative machine learning-based high-resolution downscaling of Greenland's surface mass balance and surface temperature [44.99833362998488]
  High-resolution projections of the Greenland ice sheet's surface mass balance (SMB) and surface temperature are essential for understanding future sea-level rise.<n>Here, we introduce a novel physics-constrained generative modeling framework based on a consistency model (CM) to downscale low-resolution SMB and surface temperature fields.<n>Our approach delivers realistic, high-resolution climate forcing for ice-sheet simulations with fast inference.
  arXiv  Detail & Related papers  (2025-07-30T08:43:48Z)
• Pushing the Limits of Extreme Weather: Constructing Extreme Heatwave Storylines with Differentiable Climate Models [49.1574468325115]
  We present a novel framework that leverages a differentiable hybrid climate model, NeuralGCM, to optimize initial conditions and generate physically consistent worst-case heatwave trajectories.<n>Results demonstrate that differentiable climate models can efficiently explore the upper tails of event likelihoods, providing a powerful new approach for constructing targeted storylines of extreme weather under climate change.
  arXiv  Detail & Related papers  (2025-06-12T12:50:38Z)
• Advancing Seasonal Prediction of Tropical Cyclone Activity with a Hybrid AI-Physics Climate Model [2.5701544858386396]
  Machine learning models are successful with weather forecasting and have shown progress in climate simulations.<n>We show this feasibility using Neural General Circulation Model (NeuralGCM), a hybrid ML-physics atmospheric model developed by Google.
  arXiv  Detail & Related papers  (2025-04-30T19:42:16Z)
• Prediction Beyond the Medium Range with an Atmosphere-Ocean Model that Combines Physics-based Modeling and Machine Learning [0.0]
  This paper explores the potential of a hybrid modeling approach that combines machine learning (ML) with conventional physics-based modeling for weather prediction beyond the medium range.<n>The model is based on the low-resolution, simplified parameterization atmospheric general circulation model SPEEDY.<n>The model has skill in predicting the El Nino cycle and its global teleconnections with precipitation for 3-7 months depending on the season.
  arXiv  Detail & Related papers  (2024-05-29T20:56:44Z)
• Residual Corrective Diffusion Modeling for Km-scale Atmospheric Downscaling [58.456404022536425]
  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.
  arXiv  Detail & Related papers  (2023-09-24T19:57:22Z)
• Dynamical Tests of a Deep-Learning Weather Prediction Model [0.0]
  Deep-learning weather prediction models have been shown to produce forecasts that rival those from physics-based models run at operational centers. It is unclear whether these models have encoded atmospheric dynamics, or simply pattern matching that produces the smallest forecast error. Here we subject one such model, Pangu-weather, to a set of four classical dynamical experiments that do not resemble the model training data. We conclude that the model encodes realistic physics in all experiments, and suggest it can be used as a tool for rapidly testing ideas before using expensive physics-based models.
  arXiv  Detail & Related papers  (2023-09-19T18:26:41Z)
• DiffESM: Conditional Emulation of Earth System Models with Diffusion Models [2.1989764549743476]
  A key application of Earth System Models (ESMs) is studying extreme weather events, such as heat waves or dry spells. We show that diffusion models can effectively emulate the trends of ESMs under previously unseen climate scenarios.
  arXiv  Detail & Related papers  (2023-04-23T17:12:33Z)
• Multi-scale Digital Twin: Developing a fast and physics-informed surrogate model for groundwater contamination with uncertain climate models [53.44486283038738]
  Climate change exacerbates the long-term soil management problem of groundwater contamination. We develop a physics-informed machine learning surrogate model using U-Net enhanced Fourier Neural Contaminated (PDENO) In parallel, we develop a convolutional autoencoder combined with climate data to reduce the dimensionality of climatic region similarities across the United States.
  arXiv  Detail & Related papers  (2022-11-20T06:46:35Z)
• Spatiotemporal modeling of European paleoclimate using doubly sparse Gaussian processes [61.31361524229248]
  We build on recent scale sparsetemporal GPs to reduce the computational burden. We successfully employ such a doubly sparse GP to construct a probabilistic model of paleoclimate.
  arXiv  Detail & Related papers  (2022-11-15T14:15:04Z)
• Deep Learning based Extreme Heatwave Forecast [8.975667614727648]
  Using 1000 years of state-of-the-art PlaSim Planete Simulator Climate Model data, it is shown that Convolutional Neural Network-based Deep Learning frameworks, with large-class undersampling and transfer learning achieve significant performance in forecasting the occurrence of extreme heatwaves.
  arXiv  Detail & Related papers  (2021-03-17T16:10:06Z)
• DeepClimGAN: A High-Resolution Climate Data Generator [60.59639064716545]
  Earth system models (ESMs) are often used to generate future projections of climate change scenarios. As a compromise, emulators are substantially less expensive but may not have all of the complexity of an ESM. Here we demonstrate the use of a conditional generative adversarial network (GAN) to act as an ESM emulator.
  arXiv  Detail & Related papers  (2020-11-23T20:13:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.