[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122120-en":3,"doc-seo-122120-105":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},122120,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Machine Learning Methods for Seasonal Forecasts of Climate Variables","This thesis presents an overview of leading machine learning approaches for seasonal climate forecasting, explaining core implementation methods and demonstrating model performance with illustrative examples. The study details the internal workings of LSTM and transformer architectures, including their strengths and limitations. It describes data preprocessing using EOF decomposition of key climate fields, emphasizes practical training techniques, and evaluates the resulting example LSTM and transformer models on forecast outcomes.","Department of Physics and Astronomy Department of Biological, Geological and Environmental Sciences  \nMaster Degree in Science of Climate  \nMACHINE LEARNING METHODS FOR SEASONAL FORECASTS OF CLIMATE VARIABLES  \nSupervisor:  \nProf. Antonio Navarra  \nSubmitted by: Nicol`o Landi  \nAcademic year 2023/2024  \nAbstract  \nThe purpose of this thesis work is to give an overview of the most prominent and successful machine learning models in the field of climate forecasts, to exhibit some of the most important methods used in their implementation, and to showcase their performance through the use of a few simple example models. We first go through the inner workings of the LSTM and transformer models, highlighting their strengths and shortcomings. We then go on to the data preprocessing phase, which in our case included the EOF decomposition of our input fields, particularly the tropical sea surface temperature and surface air temperature. We also list some practical methods useful during the training process. Finally, we present the performance of our example LSTM and transformer models.  \nContents  \nIntroduction 5  \n1 Data and Methods 7  \n1.1 LSTM Architecture .............................. 7  \n1.1.1 Recurrent Neural Networks ..................... 7  \n1.1.2 LSTM ................................. 8  \n1.2 Transformer Architecture ........................... 11  \n1.2.1 Attention Mechanism ......................... 11  \n1.2.2 Decoder-Encoder Mechanism .................... 14  \n1.2.3 Positional Embedding ........................ 14  \n1.3 Data Preprocessing .............................. 16  \n1.3.1 ERA5 Dataset ............................. 16  \n1.3.2 EOF Analysis ............................. 16  \n1.3.3 Data Splitting and Sequencing .................... 18  \n1.4 Training Process ............................... 19  \n1.4.1 Optimizer ............................... 20  \n1.4.2 Scheduler and Gradient Clipping .................. 21  \n1.4.3 Dropout ................................ 22  \n1.4.4 Validation Phase ........................... 24  \n2 Results 27  \n2.1 Obtaining the Forecast ............................ 28  \n2.1.1 Greedy Inference ........................... 28  \n2.1.2 Recomposition and Testing ...................... 28  \n2.2 LSTM ..................................... 29  \n2.2.1 Tropical SST ............................. 29  \n2.2.2 Pacific SST .............................. 30  \n2.2.3 Tropical T2M ............................. 30  \n2.2.4 T2M over Sea ............................. 30  \n2.3 Transformer .................................. 31  \n3 Discussion 45  \n3.1 LSTM ..................................... 45  \n3.1.1 Sea Surface Temperature ....................... 45  \n3.1.2 Surface Air Temperature ....................... 49  \n3.2 Transformer .................................. 51  \nSummary and Conclusions 55  \nAppendix 63  \nIntroduction  \nWhen dealing with the climate system and in particular its evolution, we are mostly dealing with physical laws that can be described by partial differential equations, like the fundamental Navier-Stokes equations. Since these equations cannot be solved analytically, one needs to use approximations to solve the forecasting problem. From the first onset of the computer, weather prediction models started to obtain valid solutions with numerical methods, such as the finite differences method [1] . From those earliest years much development has occurred, with the increase of computational power allowing for extremely complex models, with a myriad of equations representing all kinds of physical, biological, and even social processes that occur on Earth. Nevertheless, these models are still fundamentally based on the same numerical methods used in the first forecasts.  \nIn recent years, with the advent of big data, efficient supercomputers with Graphics Processing Units (GPU), and scientific interest in emerging new methods [2], a new way of solving the partial differential equations governing the climate system has","cbCailNd9Ey4iUlP","https://ap.wps.com/l/cbCailNd9Ey4iUlP","pdf",3472592,1,68,"English","en",105,"# Introduction\n# Data and Methods\n## LSTM Architecture\n## Transformer Architecture\n## Data Preprocessing\n## Training Process\n# Results\n## Obtaining the Forecast\n## LSTM\n## Transformer\n# Discussion\n## LSTM\n## Transformer\n# Summary and Conclusions\n# Appendix","[{\"question\":\"What machine learning models are examined for seasonal climate forecasting?\",\"answer\":\"The thesis focuses on LSTM and transformer models, outlining their internal mechanisms and how they are applied to climate prediction.\"},{\"question\":\"How is the input climate data preprocessed before training?\",\"answer\":\"Data preprocessing includes EOF decomposition of input fields, with emphasis on tropical sea surface temperature and surface air temperature.\"},{\"question\":\"What is evaluated in the results section?\",\"answer\":\"The work evaluates forecast generation methods and reports the performance of example LSTM and transformer models, including comparisons across specified climate variables.\"}]","Machine Learning Methods for Seasonal Forecasts of Climate Variables | PDF",1785808912,171,{"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},"machine-learning-methods-for-seasonal-forecasts-of-climate-variables","",{"@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/machine-learning-methods-for-seasonal-forecasts-of-climate-variables/122120/",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","2026-08-04",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},"What machine learning models are examined for seasonal climate forecasting?","Question",{"text":75,"@type":76},"The thesis focuses on LSTM and transformer models, outlining their internal mechanisms and how they are applied to climate prediction.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the input climate data preprocessed before training?",{"text":80,"@type":76},"Data preprocessing includes EOF decomposition of input fields, with emphasis on tropical sea surface temperature and surface air temperature.",{"name":82,"@type":73,"acceptedAnswer":83},"What is evaluated in the results section?",{"text":84,"@type":76},"The work evaluates forecast generation methods and reports the performance of example LSTM and transformer models, including comparisons across specified climate variables.","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"]