[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118175-en":3,"doc-seo-118175-105":30,"detail-sidebar-cat-0-en-105":91},{"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":4,"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},118175,8796095462418,"Noah","https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780",8,"Research & Report","Interpretable Machine Learning for Weather and Climate Prediction: A Survey","Advanced machine learning models deliver high predictive accuracy for weather and climate forecasting, yet their complexity often limits inherent transparency and interpretability, turning them into “black boxes” that reduce user trust and slow model improvement. This survey reviews interpretable machine learning approaches for meteorological prediction, grouping methods into post-hoc attribution of pre-trained models and inherently interpretable architectures. It also outlines research challenges, including deeper mechanistic explanations tied to physical principles, standardized evaluation benchmarks, interpretability within iterative development, and explainability for large foundation models.","arXiv:2403.18864v1 [[physics. ao-ph](physics. ao-ph)] 24 Mar 2024  \nInterpretable Machine Learning for Weather and Climate Prediction: A Survey  \n[Ruyi Yang](Ruyi Yang1 yangruyi853@gmail. com)[1](Ruyi Yang1 yangruyi853@gmail. com)[ yangruyi853@gmail. com](Ruyi Yang1 yangruyi853@gmail. com)  \n[Jingyu Hu](Jingyu Hu2 ym21669@bristol. ac.uk)[2](Jingyu Hu2 ym21669@bristol. ac.uk)[ ym21669@bristol. ac.uk](Jingyu Hu2 ym21669@bristol. ac.uk)  \nZihao [Li](Li3 lizihao9885@gmail. com)[3](Li3 lizihao9885@gmail. com)[ lizihao9885@gmail. com](Li3 lizihao9885@gmail. com)  \n[Jianli Mu](Jianli Mu1 mujl668@sina. com)[1](Jianli Mu1 mujl668@sina. com)[ mujl668@sina. com](Jianli Mu1 mujl668@sina. com)  \n[Tingzhao Yu](Tingzhao Yu1 tsingzao@hotmail. com)[1](Tingzhao Yu1 tsingzao@hotmail. com)[ tsingzao@hotmail. com](Tingzhao Yu1 tsingzao@hotmail. com)  \n[Jiangjiang Xia](Jiangjiang Xia4 xiajj@tea. ac. cn)[4](Jiangjiang Xia4 xiajj@tea. ac. cn)[ xiajj@tea. ac. cn](Jiangjiang Xia4 xiajj@tea. ac. cn)  \n[Xuhong Li](Xuhong Li5 jacqueslixuhong@gmail. com)[5](Xuhong Li5 jacqueslixuhong@gmail. com)[ jacqueslixuhong@gmail. com](Xuhong Li5 jacqueslixuhong@gmail. com)  \n[Aritra Dasgupta](Aritra Dasgupta6 aritra. dasgupta@njit. edu)[6](Aritra Dasgupta6 aritra. dasgupta@njit. edu)[ aritra. dasgupta@njit. edu](Aritra Dasgupta6 aritra. dasgupta@njit. edu)  \n[Haoyi Xiong](Haoyi Xiong5 haoyi.xiong.fr@ieee. org)[5](Haoyi Xiong5 haoyi.xiong.fr@ieee. org)[ haoyi.xiong.fr@ieee. org](Haoyi Xiong5 haoyi.xiong.fr@ieee. org)  \n1 Public Meteorological Service Center, China Meteorological Administration 2 University of Bristol  \n3 Zhejiang University 4 Institute of Atmospheric Physics, Chinese Academy of Science 5 Baidu Inc.  \n6 New Jersey Institute of Technology  \nAbstract  \nAdvanced machine learning models have recently achieved high predictive accuracy for weather and climate prediction. However, these complex models often lack inherent transparency and interpretability 1 , acting as \"black boxes\" that impede user trust and hinder further model improvements. As such, interpretable machine learning techniques have become crucial in enhancing the credibility and utility of weather and climate modeling. In this survey, we review current interpretable machine learning approaches applied to meteorological predictions. We categorize methods into two major paradigms: 1) Post-hoc interpretability techniques that explain pre-trained models, such as perturbation-based, game theory based, and gradient-based attribution methods. 2) Designing inherently interpretable models from scratch using architectures like tree ensembles and explainable neural networks.  \nWe summarize how each technique provides insights into the predictions, uncovering novel meteorological relationships captured by machine learning. Lastly, we discuss research challenges around achieving deeper mechanistic interpretations aligned with physical principles, developing standardized evaluation benchmarks, integrating interpretability into iterative model development workflows, and providing explainability for large foundation models.  \n1 Introduction  \nWeather and climate change have a significant impact on social, economic, and environmental systems around the world. Therefore, accurate weather forecasting and climate prediction are crucial to hazard preparation, resource management, and understanding long-term climate change. Traditionally, these predictions have relied heavily on complex numerical models that solve fundamental physics equations influencing atmospheric dynamics (Richardson, 1922), such as Numerical Weather Prediction (NWP) models and General Circulation Models (GCMs) . However, these physics-based numerical predictions have some major limitations, including uncertainties in initial conditions, incomplete representations of sub-grid processes, and constraints on spatial resolution and computing power. In recent years, machine learning (ML) techniques, particularly deep learning models, have achieved d","cbCaidLk1s1OAagw","https://ap.wps.com/l/cbCaidLk1s1OAagw","pdf",10743457,1,26,"English","en",105,"# Introduction\n## Motivation and limitations of numerical models\n## Machine learning progress in meteorology\n## The interpretability challenge","[{\"question\":\"Why is interpretability important for weather and climate prediction models?\",\"answer\":\"Many high-performing ML models act as “black boxes,” reducing trust from domain experts and limiting developers’ ability to diagnose errors or understand captured relationships, which slows refinement and obscures atmospheric processes.\"},{\"question\":\"How does the survey categorize interpretable machine learning methods?\",\"answer\":\"It groups approaches into two paradigms: post-hoc interpretability that explains pre-trained models using attribution methods, and inherently interpretable models built from architectures such as tree ensembles and explainable neural networks.\"},{\"question\":\"What research challenges does the survey highlight for future work?\",\"answer\":\"It discusses the need for deeper mechanistic interpretations aligned with physical principles, standardized evaluation benchmarks, integrating interpretability into iterative model development workflows, and providing explainability for large foundation models.\"}]","Interpretable Machine Learning for Weather and Climate Prediction: A Survey | PDF",1785682002,66,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":28},"interpretable-machine-learning-for-weather-and-climate-prediction-a-survey","",{"@graph":36,"@context":85},[37,54,68],{"@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/interpretable-machine-learning-for-weather-and-climate-prediction-a-survey/118175/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is interpretability important for weather and climate prediction models?","Question",{"text":75,"@type":76},"Many high-performing ML models act as “black boxes,” reducing trust from domain experts and limiting developers’ ability to diagnose errors or understand captured relationships, which slows refinement and obscures atmospheric processes.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the survey categorize interpretable machine learning methods?",{"text":80,"@type":76},"It groups approaches into two paradigms: post-hoc interpretability that explains pre-trained models using attribution methods, and inherently interpretable models built from architectures such as tree ensembles and explainable neural networks.",{"name":82,"@type":73,"acceptedAnswer":83},"What research challenges does the survey highlight for future work?",{"text":84,"@type":76},"It discusses the need for deeper mechanistic interpretations aligned with physical principles, standardized evaluation benchmarks, integrating interpretability into iterative model development workflows, and providing explainability for large foundation models.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]