ORBIT-2: Scaling Exascale Vision Foundation Models for Weather and Climate Downscaling
Oak Ridge National Laboratory（奥克岭国家实验室）；AMD Research and Advanced Development（AMD研究与高级开发）；RIKEN Center for Computational Science（理化学研究所计算科学中心）；University of Alabama at Birmingham（阿拉巴马大学伯明翰分校）
Xiao Wang, Jong-Youl Choi, Takuya Kurihaya, Isaac Lyngaas, Hong-Jun Yoon, Xi Xiao, David Pugmire, Ming Fan, Nasik M. Nafi, Aristeidis Tsaris, Ashwin M. Aji, Mal… 展开作者
Xiao Wang, Jong-Youl Choi, Takuya Kurihaya, Isaac Lyngaas, Hong-Jun Yoon, Xi Xiao, David Pugmire, Ming Fan, Nasik M. Nafi, Aristeidis Tsaris, Ashwin M. Aji, Maliha Hossain, Mohamed Wahib, Dali Wang, Peter Thornton, Prasanna Balaprakash, Moetasim Ashfaq, Dan Lu
Sparse observations and coarse-resolution climate models limit effective regional decision-making, underscoring the need for robust downscaling. However, existing AI methods struggle with generalization across variables and geographies and are constrained by the quadratic complexity of Vision Transformer (ViT) self-attention. We introduce ORBIT-2, a scalable foundation model for global, hyper-resolution climate downscaling. ORBIT-2 incorporates two key innovations: (1) Residual Slim ViT (Reslim), a lightweight architecture with residual learning and Bayesian regularization for efficient, robust prediction; and (2) TILES, a tile-wise sequence scaling algorithm that reduces self-attention complexity from quadratic to linear, enabling long-sequence processing and massive parallelism. ORBIT-2 scales to 10 billion parameters across 65,536 GPUs, achieving up to 4.1 exaFLOPS sustained throughput and 74--98% strong scaling efficiency. It supports downscaling to 0.9 km global resolution and processes sequences up to 4.2 billion tokens. On 7 km resolution benchmarks, ORBIT-2 achieves high accuracy with $R^2$ scores in the range of 0.98--0.99 against observational data.
Compressing high-resolution data through latent representation encoding for downscaling large-scale AI weather forecast model
Shanghai Academy of Artificial Intelligence for Science（上海科学人工智能研究院）；Shanghai Normal University（上海师范大学）；Jiangsu Meteorological Observatory（江苏省气象台）；Artificial Intelligence Innovation and Incubation Institute, Fudan University（复旦大学人工智能创新与孵化研究院）；Nanjing Meteorological Observatory（南京市气象台）
Hard-Constrained Deep Learning for Climate Downscaling
Fraunhofer ITWM（弗劳恩霍夫信息技术与数学研究所）；Mila Quebec AI Institute（米拉魁北克人工智能研究所）；University of Montreal（蒙特利尔大学）；New York University（纽约大学）；IBM Research（IBM研究院）；McGill University（麦吉尔大学）
China Regional 3km Downscaling Based on Residual Corrective Diffusion Model
State Key Laboratory of Severe Weather Meteorological Science and Technology（severe weather meteorological science and technology state key laboratory）；CMA Earth System Modeling and Prediction Centre；NVIDIA；Tsinghua University
Modeling the risk of extreme weather events in a changing climate is essential for developing effective adaptation and mitigation strategies. Although the available low-resolution climate models capture different scenarios, accurate risk assessment for mitigation and adaption often demands detail that they typically cannot resolve. Here, we develop a dynamic data-driven downscaling (super-resolution) method that incorporates physics and statistics in a generative framework to learn the fine-scale spatial details of rainfall. Our method transforms coarse-resolution ($0.25^{\circ} \times 0.25^{\…
University of Oxford（牛津大学）；European Centre for Medium-Range Weather Forecasts（欧洲中期天气预报中心）
Despite continuous improvements, precipitation forecasts are still not as accurate and reliable as those of other meteorological variables. A major contributing factor to this is that several key processes affecting precipitation distribution and intensity occur below the resolved scale of global weather models. Generative adversarial networks (GANs) have been demonstrated by the computer vision community to be successful at super-resolution problems, i.e., learning to add fine-scale structure to coarse images. Leinonen et al. (2020) previously applied a GAN to produce ensembles of reconstruct…
Generative 3D Gaussian Splatting for Arbitrary-ResolutionAtmospheric Downscaling and Forecasting
Department of Computer Science and Engineering, The Hong Kong University of Science and Technology（香港科技大学计算机科学与工程系）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Department of Computer Science and Engineering, Southern University of Science and Technology（南方科技大学计算机科学与工程系）；School of Computer and Information Sciences, University of Newcastle（纽卡斯尔大学计算机与信息科学学院）；School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University（西北工业大学人工智能、光学与电子学学院（iOPEN））
Downscaling Extreme Precipitation with Wasserstein Regularized Diffusion
Department of Electrical and Computer Engineering, Rice University（电气与计算机工程系， Rice大学）；Department of Civil and Environmental Engineering, Rice University（土木与环境工程系， Rice大学）；Ken Kennedy Institute, Rice University（肯尼迪研究所， Rice大学）；Department of Computer Science, Rice University（计算机科学系， Rice大学）
Global spatio-temporal downscaling of ERA5 precipitation through generative AI
Institute of Meteorology and Climate Research - Atmospheric Environmental Research (IMK-IFU)（气象与气候研究所——大气环境研究（IMK-IFU））；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；Institute of Meteorology and Climate Research - Atmospheric Trace Gases and Remote Sensing (IMK-ASF)（气象与气候研究所——大气痕量气体与遥感（IMK-ASF））；University of Augsburg（奥格斯堡大学）
Jackson School of Geosciences（杰克逊地球科学学院）；University of Texas at Austin（德克萨斯大学奥斯汀分校）；CIRES, University of Colorado Boulder（科罗拉多大学博尔德分校环境科学研究合作研究所）；NOAA/Physical Sciences Laboratory（美国国家海洋和大气管理局物理科学实验室）；Purdue University（普渡大学）；Cockrell School of Engineering（科克雷尔工程学院）
Urban downscaling is a link to transfer the knowledge from coarser climate information to city scale assessments. These high-resolution assessments need multiyear climatology of past data and future projections, which are complex and computationally expensive to generate using traditional numerical weather prediction models. The city of Austin, Texas, USA has seen tremendous growth in the past decade. Systematic planning for the future requires the availability of fine resolution city-scale datasets. In this study, we demonstrate a novel approach generating a general purpose operator using dee…
École polytechnique fédérale de Lausanne（洛桑联邦理工学院）；Federal Office of Meteorology and Climatology MeteoSwiss（联邦气象和气候学办公室 MeteoSwiss）
Generative adversarial networks (GANs) have been recently adopted for super-resolution, an application closely related to what is referred to as "downscaling" in the atmospheric sciences: improving the spatial resolution of low-resolution images. The ability of conditional GANs to generate an ensemble of solutions for a given input lends itself naturally to stochastic downscaling, but the stochastic nature of GANs is not usually considered in super-resolution applications. Here, we introduce a recurrent, stochastic super-resolution GAN that can generate ensembles of time-evolving high-resoluti…
Prithvi WxC: Foundation Model for Weather and Climate
NASA Marshall Space Flight Center（NASA马歇尔太空飞行中心）；IBM Research（IBM研究院）；Earth System Science Center, The University of Alabama in Huntsville（阿拉巴马大学亨茨维尔分校地球系统科学中心）；Department of Earth System Science, Stanford University（斯坦福大学地球系统科学系）；Department of Atmospheric Science, Colorado State University（科罗拉多州立大学大气科学系）；National Center for Computational Sciences / Oak Ridge National Laboratory（橡树岭国家实验室国家计算科学中心）；NASA/Goddard Space Flight Center（NASA戈达德太空飞行中心）；NASA Headquarters（NASA总部）
Deep learning, particularly convolutional neural networks for image recognition, has been recently used in meteorology. One of the promising applications is developing a statistical surrogate model that converts the output images of low-resolution dynamic models to high-resolution images. Our study exhibits a preliminary experiment that evaluates the performance of a model that downscales synoptic temperature fields to mesoscale temperature fields every 6 hours. The deep learning model was trained with operational 22-km gridded global analysis surface winds and temperatures as the input, opera…
MAUNet-Light: A Concise MAUNet Architecture for Bias Correction and Downscaling of Precipitation Estimates
Department of Artificial Intelligence, Indian Institute of Technology Kharagpur（人工智能系，印度理工学院科丁格浦分校）；Sustainability and Data Sciences Laboratory (SDS Lab), Civil and Environmental Engineering, Northeastern University（可持续性与数据科学实验室（SDS实验室），土木与环境工程，东北大学）；AI for Climate and Sustainability (AI4CaS), The Institute for Experiential AI (IEAI), Northeastern University（人工智能用于气候与可持续性（AI4CaS），体验式人工智能研究所（IEAI），东北大学）
High-Resolution Climate Projections Using Diffusion-Based Downscaling of a Lightweight Climate Emulator
Information Sciences and Technology, The Pennsylvania State University（宾夕法尼亚州立大学信息科学与技术系）；Department of Applied Mathematics, University of California, Santa Cruz（加州大学圣克鲁兹分校应用数学系）；Allen Institute for Artificial Intelligence (AI2)（人工智能研究所）；School of Mechanical Engineering, Purdue University（普渡大学机械工程学院）
Artificial intelligence pathways from weather to climate
University of California, Los Angeles（加州大学洛杉矶分校）；Hong Kong University of Science and Technology（香港科技大学）；University of Lausanne（洛桑大学）；Allen Institute for AI（艾伦人工智能研究所）；University of Colorado Boulder（科罗拉多大学博尔德分校）；California Institute of Technology（加州理工学院）；NOAA/Geophysical Fluid Dynamics Laboratory（NOAA地球物理流体动力学实验室）；Google Research（谷歌研究院）；New York University（纽约大学）
