[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120752-en":3,"doc-seo-120752-105":29,"detail-sidebar-cat-0-en-105":93},{"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":11},120752,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",8,"Research & Report","Integrating Empirical Orthogonal Functions (EOFs) into Machine Learning Model - Research Computing Days 2023","Integrating empirical orthogonal function (EOF) analysis into machine learning offers an approach to represent soil moisture for Earth System Models such as CESM while addressing the high computational cost of increasingly detailed land-process parameterizations. The study constructs soil-moisture data from Climate Land Model output, decomposes hydrological variables using EOFs, and uses the resulting EOF-based modes as input features for neural network emulators. Model effectiveness is assessed via RMSE, MAE, and Nash–Sutcliffe efficiency, comparing the EOF-based method with traditional models using raw features.","Boise State University  \nScholarWorks  \n\n| Research Computing Days 2023 | Research Computing Days |\n| --- | --- |\n\n3-28-2023  \nIntegrating Empirical Orthogonal Functions (EOFs) into Machine Learning Model  \nKachinga Silwimba Boise State University  \nIntegrating Empirical Orthogonal Functions (EOFs) into Machine Learning Model  \nAbstract  \nThe representation of soil moisture in Earth System Models, like the Community Earth System Model (CESM), is an essential facet in modeling the response of the Earth System to climate change. Since their inception, land models have grown to represent critical processes like carbon cycling, ecosystem dynamics, terrestrial hydrology, and agriculture. They serve as a lower boundary condition for atmospheric general circulation models. With increasing process representation, they are computationally expensive. Hydrologists and modelers use several parameterization schemes to describe the water and energy balance. However, this is regarded as computationally expensive. Alternative tools called emulators (e.g., machine learning and artificial intelligence) incorporated with the empirical orthogonal function analysis can represent soil moisture.  \nThis student presentation is available at ScholarWorks: [https://scholarworks.boisestate.edu/rcd_2023/14](https://scholarworks.boisestate.edu/rcd_2023/14)  \nIntegrating Empirical Orthogonal Functions (EOFs) into Machine Learning Model  \nKachinga Silwimba  \nDepartment of Geoscience, Boise State University  \nIntroduction Integration of EOF into Machine Learning Model Constructed Soil Moisture Data  \nEOFs are a popular tool for analyzing large datasets in various fields, such as climate science, oceanography, and geology. They identify dominant spatial and temporal patterns in data and reduce dimensionality while preserving essential information . Recently, there has been growing interest in integrating EOF analysis into machine learning models to enhance their performance and interpretability by extracting key features and relationships between variables . This study explores the effectiveness of incorporating EOF analysis into machine learning models using Climate Land Model output to predict and investigate the performance of the EOF-based model.  \nEOF Analysis  \n• EOF analysis finds a structure (or pattern) that conveys as much information as the original dataset without redundancy.  \n• EOF analysis provides means of filtering and compression of the data for better understanding.  \nX (t, s) =  ck (t)uk (s)  \nFig.2 . Schematic workflow diagram of the machine learning model and decomposition of the hydrological soil variable using the EOF analysis and the integration of EOF into the machine learning model to predict soil moisture .  \nNeural Network Model  \nX-gridded climate dataset, ck (t) -principal component, uk (s) - empirical orthogonal functions, M - number of modes in a field  \nMathematical Description  \n• The singular value Decomposition (SVD) technique is used to compute the EOF for climate dataset A. This avoids having to compute the covariance matrix directly and is optimal for datasets with large spatial dimensions . The decomposition of A is  \nFig.3. Schematic workflow of the machine learning model to predict soil moisture .  \nEOF Analysis Modes  \nFig.5 . comparing the constructed soil moisture with the actual soil moisture dataset from the Climate Land Model. The constructed soil moisture is used in the output layer of the neural network to be predicted using the US.  \nPredicted Soil Moisture  \nFig.6 . Scatter and distribution plot Comparing soil moisture prediction using the ML model with the actual CLM output over the US compared ton the EOF-based method.  \nFig.7 shows the Neural Network's Performance in predicting soil moisture in green color compared to the actual soil moisture output from the Climate Land Model in black color.  \nConclusion  \nA = URVT U ∈ ℝm×m, V ∈ ℝn×n, R ∈ ℝn×n  \n• The singular values along the R diagonal are customarily ","cbCaicLcO0V0YThi","https://ap.wps.com/l/cbCaicLcO0V0YThi","pdf",2818092,1,3,"English","en",105,"# Introduction\n## EOF Analysis\n## Neural Network Model\n## Mathematical Description\n## EOF Analysis Modes\n## Predicted Soil Moisture\n## Conclusion\n## References","[{\"question\":\"Why integrate EOFs into a machine learning model for soil moisture?\",\"answer\":\"Land models become computationally expensive as they include more processes. EOF-based emulators combine EOF decomposition with machine learning to represent soil moisture efficiently while preserving essential information.\"},{\"question\":\"How are EOFs computed in the presented method?\",\"answer\":\"The singular value decomposition (SVD) technique is used to compute EOFs from the climate dataset, avoiding direct covariance-matrix computation and working well for large spatial dimensions.\"},{\"question\":\"What metrics are used to evaluate prediction performance?\",\"answer\":\"Performance is evaluated using statistical metrics including root mean square error (RMSE), mean absolute error (MAE), and Nash–Sutcliffe efficiency (NSE).\"},{\"question\":\"What do the results indicate about EOF-based models versus traditional approaches?\",\"answer\":\"EOF-based machine learning models outperform traditional machine learning models that use raw data features as inputs, improving prediction accuracy and agreement with climate land model outputs.\"}]","Integrating Empirical Orthogonal Functions (EOFs) into Machine Learning Model - Research Computing Days 2023 | PDF",1785731843,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":28},"integrating-empirical-orthogonal-functions-eofs-into-machine-learning-model-research-computing-days-2023","",{"@graph":35,"@context":87},[36,52,66],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,49],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":21},"https://docshare.wps.com/document/research-report/",{"item":50,"name":13,"@type":42,"position":51},"https://docshare.wps.com/document/integrating-empirical-orthogonal-functions-eofs-into-machine-learning-model-research-computing-days-2023/120752/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":23,"description":14,"dateModified":60,"datePublished":60,"encodingFormat":59,"isAccessibleForFree":61,"interactionStatistic":62},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":63,"interactionType":64,"userInteractionCount":20},"InteractionCounter",{"@type":65},"ViewAction",{"@type":67,"mainEntity":68},"FAQPage",[69,75,79,83],{"name":70,"@type":71,"acceptedAnswer":72},"Why integrate EOFs into a machine learning model for soil moisture?","Question",{"text":73,"@type":74},"Land models become computationally expensive as they include more processes. EOF-based emulators combine EOF decomposition with machine learning to represent soil moisture efficiently while preserving essential information.","Answer",{"name":76,"@type":71,"acceptedAnswer":77},"How are EOFs computed in the presented method?",{"text":78,"@type":74},"The singular value decomposition (SVD) technique is used to compute EOFs from the climate dataset, avoiding direct covariance-matrix computation and working well for large spatial dimensions.",{"name":80,"@type":71,"acceptedAnswer":81},"What metrics are used to evaluate prediction performance?",{"text":82,"@type":74},"Performance is evaluated using statistical metrics including root mean square error (RMSE), mean absolute error (MAE), and Nash–Sutcliffe efficiency (NSE).",{"name":84,"@type":71,"acceptedAnswer":85},"What do the results indicate about EOF-based models versus traditional approaches?",{"text":86,"@type":74},"EOF-based machine learning models outperform traditional machine learning models that use raw data features as inputs, improving prediction accuracy and agreement with climate land model outputs.","https://schema.org",{"og:url":50,"og:type":89,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":91,"canonical":50},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":94},[95,99,103,107,112,117,122,125,130,133,137],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":45,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":45,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":113,"doc_module":4,"doc_module_name":45,"category_name":114,"show_sort_weight":115,"slug":116},6,"Technology",50,"technology",{"id":118,"doc_module":4,"doc_module_name":45,"category_name":119,"show_sort_weight":120,"slug":121},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":123,"slug":124},30,"research-report",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":128,"slug":129},9,"Religion & Spirituality",20,"religion-spirituality",{"id":128,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":128,"slug":132},"World Cup","world-cup",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":134,"slug":136},10,"Lifestyle","lifestyle",{"id":138,"doc_module":4,"doc_module_name":45,"category_name":139,"show_sort_weight":108,"slug":140},19,"General","general"]