[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120143-en":3,"doc-seo-120143-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},120143,8796095461610,"Oliver","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Machine Learning for Climate Physics and Simulations - Overview","The document examines emerging advances and opportunities at the intersection of machine learning and climate physics, focusing on supervised learning, unsupervised methods, and equation discovery to speed up climate knowledge discovery and simulation workflows. It distinguishes two complementary directions: machine learning for climate physics and machine learning for climate simulations. It argues that when observations are scarce and future conditions are non-stationary, physics knowledge and interpretability are essential for generalizable models, motivating tighter collaboration across climate physics, ML theory, and numerical analysis.","Machine Learning for Climate Physics and  \narXiv:2404.13227v2 [[physics. ao-ph](physics. ao-ph)] 18 Aug 2024  \nSimulations  \nChing-Yao Lai, 1 Pedram Hassanzadeh,2 Aditi Sheshadri3 , Maike Sonnewald4 , Raffaele Ferrari5 , Venkatramani Balaji6  \n1 Department of Geophysics, Stanford University, Stanford, CA 94305, USA; email: [cyaolai@stanford.edu](cyaolai@stanford.edu)  \n2 Department of Geophysical Sciences and Commitee on Computational and Applied Mathematics, University of Chicago, Chicago, IL 60637, USA  \n3 Department of Earth System Sciences, Stanford University, Stanford, CA 94305, USA  \n4 Department of Computer Science, University of California Davis, Davis, CA 95616, USA  \n5 Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA  \n6 Schmidt Sciences, USA  \nXxxx. Xxx. Xxx. Xxx. YYYY. AA:1–29 This article’s doi:  \n10.1146/((please add article doi))  \nCopyright © YYYY by Annual Reviews. All rights reserved  \nKeywords  \nclimate, machine learning, physics-informed machine learning, equation discovery, parameterization, emulator  \nAbstract  \nWe discuss the emerging advances and opportunities at the intersection of machine learning (ML) and climate physics, highlighting the use of ML techniques, including supervised, unsupervised, and equation discovery, to accelerate climate knowledge discoveries and simulations. We delineate two distinct yet complementary aspects: (1) ML for climate physics and (2) ML for climate simulations. While physics-free ML-based models, such as ML-based weather forecasting, have demonstrated success when data is abundant and stationary, the physics knowledge and interpretability of ML models become crucial in the small-data/non-stationary regime to ensure generalizability. Given the absence of observations, the long-term future climate falls into the small-data regime. Therefore, ML for climate physics holds a critical role in addressing the challenges of ML for climate simulations. We emphasize the need for collaboration among climate physics, ML theory, and numerical analysis to achieve reliable ML-based models for climate applications.  \nModel: A representation of a system to make predictions. Models can be  \nphysics-based, ML-based, or coupled.  \nSimulation:  \nPhysics-based models solved numerically. For example the General Circulation Model is a climate simulation that solves the Navier-Stokes equation...etc.  \nEmulator: In this article emulator refers to a subset of models that fit the data, bypassing  \nsolving physics-based equations.  \nML-based emulator:  \nEmulators that fit the data with ML models (e.g., deep neural networks), bypassing solving physics-based equations.  \nNeural Network (NN): Function approximators parameterized by function operations and parameters γ , that are optimized to minimize specified cost functions L  \nNon-stationarity:  \nSystems with  \ntime-evolving statistical properties, so that a limited time series is not representative of the past or the future.  \n2  \n1. Introduction  \nMachine learning (ML) has led to breakthroughs in various areas, from playing Go to text generated with large language models (LLM), and more recently, to weather forecasting (1, 2, 3, 4, 5) . Different from Go and LLM, the language scientists use to understand and simulate weather and climate has been equations rooted in fundamental physics. Physicsbased equations, often differential equations, are essential to simulate systems where direct observations are limited and noisy. Even more so for projections of future climates where data are altogether unavailable. (Fig. 1) . Recently, ML has emerged as an alternative tool for predictive modeling as well as improving the understanding of climate physics (Fig. 2) . For instance, physics-free ML models such as neural networks (NNs), which are universal approximators of functions (6) and operators (7), trained on data from observations or physics-based simulations have demonstrated remar","cbCailzneOlSzlEH","https://ap.wps.com/l/cbCailzneOlSzlEH","pdf",4295108,1,29,"English","en",105,"# Introduction\n## Weather versus climate: non-stationarity\n## ML approaches across timescales\n# Concepts and definitions\n## Model\n## Simulation\n## Emulator\n## ML-based emulator\n## Non-stationarity","[{\"question\":\"What are the two complementary areas discussed for machine learning in climate applications?\",\"answer\":\"The document distinguishes machine learning for climate physics and machine learning for climate simulations, presenting them as distinct yet complementary aspects.\"},{\"question\":\"Why are physics knowledge and interpretability especially important in the small-data, non-stationary regime?\",\"answer\":\"In regimes with limited observations and time-evolving statistics, relying on physics-free patterns can fail to generalize, so interpretability and physical grounding help ensure robustness.\"},{\"question\":\"How does the article define an emulator in climate modeling?\",\"answer\":\"An emulator is described as a subset of models that fit data while bypassing the numerical solving of physics-based equations; an ML-based emulator uses machine learning models such as neural networks to perform this fitting.\"}]","Machine Learning for Climate Physics and Simulations - Overview | PDF",1785728418,73,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-for-climate-physics-and-simulations-overview","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-for-climate-physics-and-simulations-overview/120143/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-03",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What are the two complementary areas discussed for machine learning in climate applications?","Question",{"text":76,"@type":77},"The document distinguishes machine learning for climate physics and machine learning for climate simulations, presenting them as distinct yet complementary aspects.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Why are physics knowledge and interpretability especially important in the small-data, non-stationary regime?",{"text":81,"@type":77},"In regimes with limited observations and time-evolving statistics, relying on physics-free patterns can fail to generalize, so interpretability and physical grounding help ensure robustness.",{"name":83,"@type":74,"acceptedAnswer":84},"How does the article define an emulator in climate modeling?",{"text":85,"@type":77},"An emulator is described as a subset of models that fit data while bypassing the numerical solving of physics-based equations; an ML-based emulator uses machine learning models such as neural networks to perform this fitting.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]