Ocean General Circulation Models require extensive computational resources to reach equilibrium states, while deep learning emulators, despite offering fast predictions, lack the physical interpretability and long-term stability necessary for climate scientists to understand climate sensitivity (to greenhouse gas emissions) and mechanisms of abrupt % variability such as tipping points. We propose to take the best from both worlds by leveraging deep generative models to produce physically consistent oceanic states that can serve as initial conditions for climate projections. We assess the viability of this hybrid approach through both physical metrics and numerical experiments, and highlight the benefits of enforcing physical constraints during generation. Although we train here on ocean variables from idealized numerical simulations, we claim that this hybrid approach, combining the computational efficiency of deep learning with the physical accuracy of numerical models, can effectively reduce the computational burden of running climate models to equilibrium, and reduce uncertainties in climate projections by minimizing drifts in baseline simulations.
A machine learning model for skillful climate system prediction
National Climate Centre, China Meteorological Administration（中国气象局国家气候中心）；Artificial Intelligence Innovation and Incubation Institute, Fudan University（复旦大学人工智能创新与产业研究院）；School of Data Science, Fudan University（复旦大学大数据学院）；Institute for Big Data, Fudan University（复旦大学大数据研究院）；MOE Laboratory for National Development and Intelligent Governance, Fudan University（复旦大学教育部国家发展智能治理实验室）；Xinjiang Climate Center（新疆气候中心）；Nanjing University of Information Science and Technology（南京信息工程大学）；School of Atmospheric Sciences, Nanjing University of Information Science and Technology（南京信息工程大学大气科学学院）；Chinese Academy of Meteorological Sciences（中国气象科学研究院）
Data-driven Global Ocean Modeling for Seasonal to Decadal Prediction
School of Computer Science, Fudan University（复旦大学计算机科学学院）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Potsdam Institute for Climate Impact Research（波茨坦气候影响研究所）；Japan Agency for Marine-Earth Science and Technology（日本海洋研究开发机构）；Institut de Recherche pour le Développement (IRD)（法国发展研究院）；French National Research Institute for Sustained Development (IRD)（法国可持续发展研究院）；Cooperative Institute for Research in Environmental Sciences（合作环境科学研究所）；NOAA Physical Sciences Laboratory（美国国家海洋和大气管理局物理科学实验室）；Nanjing University of Information Science and Technology（南京信息工程大学）；Technical University of Munich（慕尼黑工业大学）；University of Colorado Boulder（科罗拉多大学博尔德分校）
Data-driven global ocean model resolving ocean-atmosphere coupling dynamics
Center for Climate and Carbon Cycle Research, Korea Institute of Science and Technology, Seoul, Republic of Korea（韩国科学技术院气候与碳循环研究中心，首尔，大韩民国）；Department of Environment and Energy, Jeonbuk National University, Jeonju, Republic of Korea（全南国立大学环境与能源系，全州，大韩民国）；School of Earth and Environmental Sciences, Seoul National University, Seoul, Republic of Korea（首尔国立大学地球与环境科学学院，首尔，大韩民国）；Department of Environmental Management, Seoul National University, Seoul, Republic of Korea（首尔国立大学环境管理系，首尔，大韩民国）
Global Climate Model Bias Correction Using Deep Learning
Department of Computational；Data Sciences, Indian Institute of Science, Bengaluru 560012, India. Divecha Center for Climate Change, Indian Institute of Science, Bengaluru 560012, India.
Samudra 2: Scaling Ocean Emulators across Resolutions
Courant Institute School of Mathematics, Computing, and Data Science, New York University（Courant学院数学、计算与数据科学系，纽约大学）；Open Athena AI Foundation, Inc.（开放Athena人工智能基金会）；Program in Atmospheric and Oceanic Sciences, Princeton University（大气与海洋科学项目，普林斯顿大学）
