Prakhar Srivastava, Ruihan Yang, Gavin Kerrigan, Gideon Dresdner, Jeremy McGibbon, Christopher Bretherton, Stephan Mandt
In climate science and meteorology, high-resolution local precipitation (rain and snowfall) predictions are limited by the computational costs of simulation-based methods. Statistical downscaling, or super-resolution, is a common workaround where a low-resolution prediction is improved using statistical approaches. Unlike traditional computer vision tasks, weather and climate applications require capturing the accurate conditional distribution of high-resolution given low-resolution patterns to assure reliable ensemble averages and unbiased estimates of extreme events, such as heavy rain. This work extends recent video diffusion models to precipitation super-resolution, employing a deterministic downscaler followed by a temporally-conditioned diffusion model to capture noise characteristics and high-frequency patterns. We test our approach on FV3GFS output, an established large-scale global atmosphere model, and compare it against six state-of-the-art baselines. Our analysis, capturing CRPS, MSE, precipitation distributions, and qualitative aspects using California and the Himalayas as examples, establishes our method as a new standard for data-driven precipitation downscaling.
James P. C. Duncan, Elynn Wu, Surya Dheeshjith, Adam Subel, Troy Arcomano, Spencer K. Clark, Brian Henn, Anna Kwa, Jeremy McGibbon, W. Andre Perkins, William Gregory, Carlos Fernandez-Granda, Julius Busecke, Oliver Watt-Meyer, William J. Hurlin, Alistair Adcroft, Laure Zanna, Christopher Bretherton
Traditional numerical global climate models simulate the full Earth system by exchanging boundary conditions between separate simulators of the atmosphere, ocean, sea ice, land surface, and other geophysical processes. This paradigm allows for distributed development of individual components within a common framework, unified by a coupler that handles translation between realms via spatial or temporal alignment and flux exchange. Following a similar approach adapted for machine learning-based emulators, we present SamudrACE: a coupled global climate model emulator which produces centuries-long simulations at 1-degree horizontal, 6-hourly atmospheric, and 5-daily oceanic resolution, with 145 2D fields spanning 8 atmospheric and 19 oceanic vertical levels, plus sea ice, surface, and top-of-atmosphere variables. SamudrACE is highly stable and has low climate biases comparable to those of its components with prescribed boundary forcing, with realistic variability in coupled climate phenomena such as ENSO that is not possible to simulate in uncoupled mode.
Clayton Sanford, Anna Kwa, Oliver Watt-Meyer, Spencer Clark, Noah Brenowitz, Jeremy McGibbon, Christopher Bretherton
While previous works have shown that machine learning (ML) can improve the prediction accuracy of coarse-grid climate models, these ML-augmented methods are more vulnerable to irregular inputs than the traditional physics-based models they rely on. Because ML-predicted corrections feed back into the climate model's base physics, the ML-corrected model regularly produces out of sample data, which can cause model instability and frequent crashes. This work shows that adding semi-supervised novelty detection to identify out-of-sample data and disable the ML-correction accordingly stabilizes simulations and sharply improves the quality of predictions. We design an augmented climate model with a one-class support vector machine (OCSVM) novelty detector that provides better temperature and precipitation forecasts in a year-long simulation than either a baseline (no-ML) or a standard ML-corrected run. By improving the accuracy of coarse-grid climate models, this work helps make accurate climate models accessible to researchers without massive computational resources.
Sungduk Yu, Zeyuan Hu, Akshay Subramaniam, Walter Hannah, Liran Peng, Jerry Lin, Mohamed Aziz Bhouri, Ritwik Gupta, Björn Lütjens, Justus C. Will, Gunnar Behrens, Julius J. M. Busecke, Nora Loose, Charles I. Stern, Tom Beucler, Bryce Harrop, Helge Heuer, Benjamin R. Hillman, Andrea Jenney, Nana Liu, Alistair White, Tian Zheng, Zhiming Kuang, Fiaz Ahmed, Elizabeth Barnes, Noah D. Brenowitz, Christopher Bretherton, Veronika Eyring, Savannah Ferretti, Nicholas Lutsko, Pierre Gentine, Stephan Mandt, J. David Neelin, Rose Yu, Laure Zanna, Nathan Urban, Janni Yuval, Ryan Abernathey, Pierre Baldi, Wayne Chuang, Yu Huang, Fernando Iglesias-Suarez, Sanket Jantre, Po-Lun Ma, Sara Shamekh, Guang Zhang, Michael Pritchard
Modern climate projections lack adequate spatial and temporal resolution due to computational constraints, leading to inaccuracies in representing critical processes like thunderstorms that occur on the sub-resolution scale. Hybrid methods combining physics with machine learning (ML) offer faster, higher fidelity climate simulations by outsourcing compute-hungry, high-resolution simulations to ML emulators. However, these hybrid ML-physics simulations require domain-specific data and workflows that have been inaccessible to many ML experts. As an extension of the ClimSim dataset (Yu et al., 2024), we present ClimSim-Online, which also includes an end-to-end workflow for developing hybrid ML-physics simulators. The ClimSim dataset includes 5.7 billion pairs of multivariate input/output vectors, capturing the influence of high-resolution, high-fidelity physics on a host climate simulator's macro-scale state. The dataset is global and spans ten years at a high sampling frequency. We provide a cross-platform, containerized pipeline to integrate ML models into operational climate simulators for hybrid testing. We also implement various ML baselines, alongside a hybrid baseline simulator, to highlight the ML challenges of building stable, skillful emulators. The data (https://huggingface.co/datasets/LEAP/ClimSim_high-res) and code (https://leap-stc.github.io/ClimSim and https://github.com/leap-stc/climsim-online) are publicly released to support the development of hybrid ML-physics and high-fidelity climate simulations.
Salva Rühling Cachay, Brian Henn, Oliver Watt-Meyer, Christopher S. Bretherton, Rose Yu
Data-driven deep learning models are transforming global weather forecasting. It is an open question if this success can extend to climate modeling, where the complexity of the data and long inference rollouts pose significant challenges. Here, we present the first conditional generative model that produces accurate and physically consistent global climate ensemble simulations by emulating a coarse version of the United States' primary operational global forecast model, FV3GFS. Our model integrates the dynamics-informed diffusion framework (DYffusion) with the Spherical Fourier Neural Operator (SFNO) architecture, enabling stable 100-year simulations at 6-hourly timesteps while maintaining low computational overhead compared to single-step deterministic baselines. The model achieves near gold-standard performance for climate model emulation, outperforming existing approaches and demonstrating promising ensemble skill. This work represents a significant advance towards efficient, data-driven climate simulations that can enhance our understanding of the climate system and inform adaptation strategies.
Liran Peng, Peter N. Blossey, Walter M. Hannah, Christopher S. Bretherton, Christopher R. Terai, Andrea M. Jenney, Michael Pritchard
High-Resolution Multi-scale Modeling Frameworks (HR) -- global climate models that embed separate, convection-resolving models with high enough resolution to resolve boundary layer eddies -- have exciting potential for investigating low cloud feedback dynamics due to reduced parameterization and ability for multidecadal throughput on modern computing hardware. However low clouds in past HR have suffered a stubborn problem of over-entrainment due to an uncontrolled source of mixing across the marine subtropical inversion manifesting as stratocumulus dim biases in present-day climate, limiting their scientific utility. We report new results showing that this over-entrainment can be partly offset by using hyperviscosity and cloud droplet sedimentation. Hyperviscosity damps small-scale momentum fluctuations associated with the formulation of the momentum solver of the embedded LES. By considering the sedimentation process adjacent to default one-moment microphysics in HR, condensed phase particles can be removed from the entrainment zone, which further reduces entrainment efficiency. The result is an HR that is able to produce more low clouds with a higher liquid water path and a reduced stratocumulus dim bias. Associated improvements in the explicitly simulated sub-cloud eddy spectrum are observed. We report these sensitivities in multi-week tests and then explore their operational potential alongside microphysical retuning in decadal simulations at operational 1.5 degree exterior resolution. The result is a new HR having desired improvements in the baseline present-day low cloud climatology, and a reduced global mean bias and root mean squared error of absorbed shortwave radiation. We suggest it should be promising for examining low cloud feedbacks with minimal approximation.
Elynn Wu, Finn Rebassoo, Pappu Paul, Cristian Proistosescu, Jacqueline Nugent, Daniel McCoy, Peter Caldwell, Christopher S. Bretherton
Green's functions are a useful technique for interpreting atmospheric state responses to changes in the spatial pattern of sea surface temperature (SST). Here we train version 2 of the Ai2 Climate Emulator (ACE2) on reference historical SST simulations of the US Department of Energy's EAMv3 global atmosphere model. We compare how well the SST Green's functions generated by ACE2 match those of EAMv3, following the protocol of the Green's Function Model Intercomparison Project (GFMIP). The spatial patterns of top-of-atmosphere (TOA) radiative response from the individual GFMIP SST patch simulations are similar for ACE and the EAMv3 reference. The derived sensitivity of global net TOA radiation sensitivity to SST patch location is qualitatively similar in ACE as in EAMv3, but there are statistically significant discrepancies for some SST patches, especially over the subtropical northeast Pacific. These discrepancies may reflect insufficient diversity in the SST patterns sampled over the course of the EAMv3 AMIP simulation used for training ACE. Both ACE and EAMv3 Green's functions reconstruct the historical record of the global annual-mean TOA radiative flux from a reference EAMv3 AMIP simulation reasonably well. Notably, under our configuration and compute resources, ACE achieves these results approximately 100 times faster in wall-clock time compared to EAMv3, highlighting its potential as a powerful and efficient tool for tackling other computationally intensive problems in climate science.
W. Andre Perkins, Anna Kwa, Jeremy McGibbon, Troy Arcomano, Spencer K. Clark, Oliver Watt-Meyer, Christopher S. Bretherton, Lucas M. Harris
Kilometer-scale simulations of the atmosphere are an important tool for assessing local weather extremes and climate impacts, but computational expense limits their use to small regions, short periods, and limited ensembles. Machine learning offers a pathway to efficiently emulate these high-resolution simulations. Here we introduce HiRO-ACE, a two-stage AI modeling framework combining a stochastic version of the Ai2 Climate Emulator (ACE2S) with diffusion-based downscaling (HiRO) to generate 3 km precipitation fields over arbitrary regions of the globe. Both components are trained on data derived from a decade of atmospheric simulation by X-SHiELD, a 3 km global storm-resolving model. HiRO performs a 32x downscaling--generating 3 km 6-hourly precipitation from coarse 100 km inputs by training on paired high-resolution and coarsened X-SHiELD outputs. ACE2S is a $1^\circ \times 1^\circ$ ($\sim$100 km) stochastic autoregressive global atmosphere emulator that maintains grid-scale precipitation variability consistent with coarsened X-SHiELD, enabling its outputs to be ingested by HiRO without additional tuning. HiRO-ACE reproduces the distribution of extreme precipitation rates through the 99.99th percentile, with time-mean precipitation biases below 10% almost everywhere. The framework generates plausible tropical cyclones, fronts, and convective events from poorly resolved coarse inputs. Its computational efficiency allows generation of 6-hourly high-resolution regional precipitation for decades of simulated climate within a single day using one H100 GPU, while the probabilistic design enables ensemble generation for quantifying uncertainty. This establishes an AI-enabled pathway for affordably leveraging the realism of expensive km-scale simulations to support local climate adaptation planning and extreme event risk assessment.
Oliver Watt-Meyer, Brian Henn, Jeremy McGibbon, Spencer K. Clark, Anna Kwa, W. Andre Perkins, Elynn Wu, Lucas Harris, Christopher S. Bretherton
Existing machine learning models of weather variability are not formulated to enable assessment of their response to varying external boundary conditions such as sea surface temperature and greenhouse gases. Here we present ACE2 (Ai2 Climate Emulator version 2) and its application to reproducing atmospheric variability over the past 80 years on timescales from days to decades. ACE2 is a 450M-parameter autoregressive machine learning emulator, operating with 6-hour temporal resolution, 1° horizontal resolution and eight vertical layers. It exactly conserves global dry air mass and moisture and can be stepped forward stably for arbitrarily many steps with a throughput of about 1500 simulated years per wall clock day. ACE2 generates emergent phenomena such as tropical cyclones, the Madden Julian Oscillation, and sudden stratospheric warmings. Furthermore, it accurately reproduces the atmospheric response to El Niño variability and global trends of temperature over the past 80 years. However, its sensitivities to separately changing sea surface temperature and carbon dioxide are not entirely realistic.
Oliver Watt-Meyer, Gideon Dresdner, Jeremy McGibbon, Spencer K. Clark, Brian Henn, James Duncan, Noah D. Brenowitz, Karthik Kashinath, Michael S. Pritchard, Boris Bonev, Matthew E. Peters, Christopher S. Bretherton
Existing ML-based atmospheric models are not suitable for climate prediction, which requires long-term stability and physical consistency. We present ACE (AI2 Climate Emulator), a 200M-parameter, autoregressive machine learning emulator of an existing comprehensive 100-km resolution global atmospheric model. The formulation of ACE allows evaluation of physical laws such as the conservation of mass and moisture. The emulator is stable for 100 years, nearly conserves column moisture without explicit constraints and faithfully reproduces the reference model's climate, outperforming a challenging baseline on over 90% of tracked variables. ACE requires nearly 100x less wall clock time and is 100x more energy efficient than the reference model using typically available resources. Without fine-tuning, ACE can stably generalize to a previously unseen historical sea surface temperature dataset.
Anna Kwa, Spencer K. Clark, Brian Henn, Noah D. Brenowitz, Jeremy McGibbon, W. Andre Perkins, Oliver Watt-Meyer, Lucas Harris, Christopher S. Bretherton
Due to computational constraints, running global climate models (GCMs) for many years requires a lower spatial grid resolution (${\gtrsim}50$ km) than is optimal for accurately resolving important physical processes. Such processes are approximated in GCMs via subgrid parameterizations, which contribute significantly to the uncertainty in GCM predictions. One approach to improving the accuracy of a coarse-grid global climate model is to add machine-learned state-dependent corrections at each simulation timestep, such that the climate model evolves more like a high-resolution global storm-resolving model (GSRM). We train neural networks to learn the state-dependent temperature, humidity, and radiative flux corrections needed to nudge a 200 km coarse-grid climate model to the evolution of a 3~km fine-grid GSRM. When these corrective ML models are coupled to a year-long coarse-grid climate simulation, the time-mean spatial pattern errors are reduced by 6-25% for land surface temperature and 9-25% for land surface precipitation with respect to a no-ML baseline simulation. The ML-corrected simulations develop other biases in climate and circulation that differ from, but have comparable amplitude to, the baseline simulation.
Noah D. Brenowitz, W. Andre Perkins, Jacqueline M. Nugent, Oliver Watt-Meyer, Spencer K. Clark, Anna Kwa, Brian Henn, Jeremy McGibbon, Christopher S. Bretherton
Cloud microphysical parameterizations in atmospheric models describe the formation and evolution of clouds and precipitation, a central weather and climate process. Cloud-associated latent heating is a primary driver of large and small-scale circulations throughout the global atmosphere, and clouds have important interactions with atmospheric radiation. Clouds are ubiquitous, diverse, and can change rapidly. In this work, we build the first emulator of an entire cloud microphysical parameterization, including fast phase changes. The emulator performs well in offline and online (i.e. when coupled to the rest of the atmospheric model) tests, but shows some developing biases in Antarctica. Sensitivity tests demonstrate that these successes require careful modeling of the mixed discrete-continuous output as well as the input-output structure of the underlying code and physical process.
Noah D Brenowitz, Christopher S Bretherton
General circulation models (GCMs) typically have a grid size of 25--200 km. Parametrizations are used to represent diabatic processes such as radiative transfer and cloud microphysics and account for sub-grid-scale motions and variability. Unlike traditional approaches, neural networks (NNs) can readily exploit recent observational datasets and global cloud-system resolving model (CRM) simulations to learn subgrid variability. This article describes an NN parametrization trained by coarse-graining a near-global CRM simulation with a 4~km horizontal grid spacing. The NN predicts the residual heating and moistening averaged over (160 km)^2 grid boxes as a function of the coarse-resolution fields within the same atmospheric column. This NN is coupled to the dynamical core of a GCM with the same 160 km resolution. A recent study described how to train such an NN to be numerically stable when coupled to specified time-evolving advective forcings in a single column model, but feedbacks between NN and GCM components cause spatially-extended simulations to crash within a few days. Analyzing the linearized response of such an NN reveals that it learns to exploit a strong synchrony between precipitation and the atmospheric state above 10 km. Removing these variables from the NN's inputs stabilizes the coupled simulations, which predict the future state more accurately than a coarse-resolution simulation without any parametrizations of sub-grid-scale variability, although the mean state slowly drifts.
Noah D. Brenowitz, Tom Beucler, Michael Pritchard, Christopher S. Bretherton
Neural networks are a promising technique for parameterizing sub-grid-scale physics (e.g. moist atmospheric convection) in coarse-resolution climate models, but their lack of interpretability and reliability prevents widespread adoption. For instance, it is not fully understood why neural network parameterizations often cause dramatic instability when coupled to atmospheric fluid dynamics. This paper introduces tools for interpreting their behavior that are customized to the parameterization task. First, we assess the nonlinear sensitivity of a neural network to lower-tropospheric stability and the mid-tropospheric moisture, two widely-studied controls of moist convection. Second, we couple the linearized response functions of these neural networks to simplified gravity-wave dynamics, and analytically diagnose the corresponding phase speeds, growth rates, wavelengths, and spatial structures. To demonstrate their versatility, these techniques are tested on two sets of neural networks, one trained with a super-parametrized version of the Community Atmosphere Model (SPCAM) and the second with a near-global cloud-resolving model (GCRM). Even though the SPCAM simulation has a warmer climate than the cloud-resolving model, both neural networks predict stronger heating/drying in moist and unstable environments, which is consistent with observations. Moreover, the spectral analysis can predict that instability occurs when GCMs are coupled to networks that support gravity waves that are unstable and have phase speeds larger than 5 m/s. In contrast, standing unstable modes do not cause catastrophic instability. Using these tools, differences between the SPCAM- vs. GCRM- trained neural networks are analyzed, and strategies to incrementally improve both of their coupled online performance unveiled.
Spencer K. Clark, Oliver Watt-Meyer, Anna Kwa, Jeremy McGibbon, Brian Henn, W. Andre Perkins, Elynn Wu, Lucas M. Harris, Christopher S. Bretherton
While autoregressive machine-learning-based emulators have been trained to produce stable and accurate rollouts in the climate of the present-day and recent past, none so far have been trained to emulate the sensitivity of climate to substantial changes in CO$_2$ or other greenhouse gases. As an initial step we couple the Ai2 Climate Emulator version 2 to a slab ocean model (hereafter ACE2-SOM) and train it on output from a collection of equilibrium-climate physics-based reference simulations with varying levels of CO$_2$. We test it in equilibrium and non-equilibrium climate scenarios with CO$_2$ concentrations seen and unseen in training. ACE2-SOM performs well in equilibrium-climate inference with both in-sample and out-of-sample CO$_2$ concentrations, accurately reproducing the emergent time-mean spatial patterns of surface temperature and precipitation change with CO$_2$ doubling, tripling, or quadrupling. In addition, the vertical profile of atmospheric warming and change in extreme precipitation rates up to the 99.9999th percentile closely agree with the reference model. Non-equilibrium-climate inference is more challenging. With CO$_2$ increasing gradually at a rate of 2% year$^{-1}$, ACE2-SOM can accurately emulate the global annual mean trends of surface and lower-to-middle atmosphere fields but produces unphysical jumps in stratospheric fields. With an abrupt quadrupling of CO$_2$, ML-controlled fields transition unrealistically quickly to the 4xCO$_2$ regime. In doing so they violate global energy conservation and exhibit unphysical sensitivities of and surface and top of atmosphere radiative fluxes to instantaneous changes in CO$_2$. Future emulator development needed to address these issues should improve its generalizability to diverse climate change scenarios.
Noah D. Brenowitz, Brian Henn, Jeremy McGibbon, Spencer K. Clark, Anna Kwa, W. Andre Perkins, Oliver Watt-Meyer, Christopher S. Bretherton
Climate models are complicated software systems that approximate atmospheric and oceanic fluid mechanics at a coarse spatial resolution. Typical climate forecasts only explicitly resolve processes larger than 100 km and approximate any process occurring below this scale (e.g. thunderstorms) using so-called parametrizations. Machine learning could improve upon the accuracy of some traditional physical parametrizations by learning from so-called global cloud-resolving models. We compare the performance of two machine learning models, random forests (RF) and neural networks (NNs), at parametrizing the aggregate effect of moist physics in a 3 km resolution global simulation with an atmospheric model. The NN outperforms the RF when evaluated offline on a testing dataset. However, when the ML models are coupled to an atmospheric model run at 200 km resolution, the NN-assisted simulation crashes with 7 days, while the RF-assisted simulations remain stable. Both runs produce more accurate weather forecasts than a baseline configuration, but globally averaged climate variables drift over longer timescales.