{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/101011"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/101011","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Analyzing el nino southern oscillation predictions from long-short-merm-memory models","abstract":"El Nino Southern Oscillation (ENSO) can have global impacts across the world. Because of its prevalence, scientists run models to forecast its next move. Here, long-short-term-memory models (LSTM) were compared to linear regression models (LR) as first steps to explore the potential benefits of simple deep neural networks for predicting ENSO. Each model’s prediction capabilities were tested with sea surface temperatures (SST), warm water volumes, and zonal winds as predictors, individually and in combinations, utilizing both monthly and daily resolution data, across a total of 11 leads. By utilizing these three variables, we examine different forms of climate variability within the coupled system (SST), the subsurface ocean (warm water volume), and the atmosphere (zonal winds), and we quantify the relative importance of each of these processes for ENSO predictability through two statistical modeling approaches: LSTM and LR. Results show that when using monthly data as predictors, predictions from LSTM were similar to predictions from LR. However, with daily data, LSTM exhibited some advantage over LR in terms of the correlation coefficient, especially with daily resolution SST as a predictor and at longer leads. This can be appealing because once the computationally expensive training of LSTM is complete, the predictions employing the trained model can be relatively cheap to perform thereafter.","abstract_html":"El Nino Southern Oscillation (ENSO) can have global impacts across the world. Because of its prevalence, scientists run models to forecast its next move. Here, long-short-term-memory models (LSTM) were compared to linear regression models (LR) as first steps to explore the potential benefits of simple deep neural networks for predicting ENSO. Each model’s prediction capabilities were tested with sea surface temperatures (SST), warm water volumes, and zonal winds as predictors, individually and in combinations, utilizing both monthly and daily resolution data, across a total of 11 leads. By utilizing these three variables, we examine different forms of climate variability within the coupled system (SST), the subsurface ocean (warm water volume), and the atmosphere (zonal winds), and we quantify the relative importance of each of these processes for ENSO predictability through two statistical modeling approaches: LSTM and LR. Results show that when using monthly data as predictors, predictions from LSTM were similar to predictions from LR. However, with daily data, LSTM exhibited some advantage over LR in terms of the correlation coefficient, especially with daily resolution SST as a predictor and at longer leads. This can be appealing because once the computationally expensive training of LSTM is complete, the predictions employing the trained model can be relatively cheap to perform thereafter.","abstract_has_math":false,"creators":["Huang, Andrew K."],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Atmospheric Sciences","degree_department":null,"school":null,"contributors":["Sriver, Ryan L."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2018,"date_issued":"2018-09-04T20:27:16Z","date_published":"2018-09-04T20:27:16Z","updated_at":"2026-07-22T22:24:38Z","subjects":["ENSO","LSTM","LR","modeling","deep learning","forecast"],"languages":["en"],"rights":["Copyright 2018 Andrew Huang"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/101011","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Sriver, Ryan L."]},{"key":"dc:creator","label":"Author","values":["Huang, Andrew K."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2018-09-04T20:27:16Z","2018-04-18","2018-05"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Atmospheric Sciences"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["ENSO","LSTM","LR","modeling","deep learning","forecast"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2018 Andrew Huang"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/101011"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["El Nino Southern Oscillation (ENSO) can have global impacts across the world. Because of its prevalence, scientists run models to forecast its next move. Here, long-short-term-memory models (LSTM) were compared to linear regression models (LR) as first steps to explore the potential benefits of simple deep neural networks for predicting ENSO. Each model’s prediction capabilities were tested with sea surface temperatures (SST), warm water volumes, and zonal winds as predictors, individually and in combinations, utilizing both monthly and daily resolution data, across a total of 11 leads. By utilizing these three variables, we examine different forms of climate variability within the coupled system (SST), the subsurface ocean (warm water volume), and the atmosphere (zonal winds), and we quantify the relative importance of each of these processes for ENSO predictability through two statistical modeling approaches: LSTM and LR. Results show that when using monthly data as predictors, predictions from LSTM were similar to predictions from LR. However, with daily data, LSTM exhibited some advantage over LR in terms of the correlation coefficient, especially with daily resolution SST as a predictor and at longer leads. This can be appealing because once the computationally expensive training of LSTM is complete, the predictions employing the trained model can be relatively cheap to perform thereafter.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2018-08-31 without embargo terms","The student, Andrew Huang, accepted the attached license on 2018-04-18 at 10:06.","The student, Andrew Huang, submitted this Thesis for approval on 2018-04-18 at 10:21.","This Thesis was approved for publication on 2018-04-18 at 10:56.","DSpace SAF Submission Ingestion Package generated from Vireo submission #12337 on 2018-08-31 at 17:13:30","Made available in DSpace on 2018-09-04T20:27:16Z (GMT). No. of bitstreams: 2 HUANG-THESIS-2018.pdf: 889980 bytes, checksum: cf73f97a3eb1bf9cf99a5dfc7202d078 (MD5) LICENSE.txt: 4209 bytes, checksum: 1f6876e245f959b603c7885c86dbce75 (MD5) Previous issue date: 2018-04-18"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Analyzing el nino southern oscillation predictions from long-short-merm-memory models"]}]}],"canonical_facts":{"dc:contributor":["Sriver, Ryan L."],"dc:creator":["Huang, Andrew K."],"dc:date":["2018-09-04T20:27:16Z","2018-04-18","2018-05"],"dc:description":["El Nino Southern Oscillation (ENSO) can have global impacts across the world. Because of its prevalence, scientists run models to forecast its next move. Here, long-short-term-memory models (LSTM) were compared to linear regression models (LR) as first steps to explore the potential benefits of simple deep neural networks for predicting ENSO. Each model’s prediction capabilities were tested with sea surface temperatures (SST), warm water volumes, and zonal winds as predictors, individually and in combinations, utilizing both monthly and daily resolution data, across a total of 11 leads. By utilizing these three variables, we examine different forms of climate variability within the coupled system (SST), the subsurface ocean (warm water volume), and the atmosphere (zonal winds), and we quantify the relative importance of each of these processes for ENSO predictability through two statistical modeling approaches: LSTM and LR. Results show that when using monthly data as predictors, predictions from LSTM were similar to predictions from LR. However, with daily data, LSTM exhibited some advantage over LR in terms of the correlation coefficient, especially with daily resolution SST as a predictor and at longer leads. This can be appealing because once the computationally expensive training of LSTM is complete, the predictions employing the trained model can be relatively cheap to perform thereafter.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2018-08-31 without embargo terms","The student, Andrew Huang, accepted the attached license on 2018-04-18 at 10:06.","The student, Andrew Huang, submitted this Thesis for approval on 2018-04-18 at 10:21.","This Thesis was approved for publication on 2018-04-18 at 10:56.","DSpace SAF Submission Ingestion Package generated from Vireo submission #12337 on 2018-08-31 at 17:13:30","Made available in DSpace on 2018-09-04T20:27:16Z (GMT). No. of bitstreams: 2 HUANG-THESIS-2018.pdf: 889980 bytes, checksum: cf73f97a3eb1bf9cf99a5dfc7202d078 (MD5) LICENSE.txt: 4209 bytes, checksum: 1f6876e245f959b603c7885c86dbce75 (MD5) Previous issue date: 2018-04-18"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/101011"],"dc:language":["en"],"dc:rights":["Copyright 2018 Andrew Huang"],"dc:subject":["ENSO","LSTM","LR","modeling","deep learning","forecast"],"dc:title":["Analyzing el nino southern oscillation predictions from long-short-merm-memory models"],"dc:type":["text"],"thesis:degree_discipline":["Atmospheric Sciences"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:24:38Z"}