Publicação

Using analog ensembles with alternative metrics for hindcasting with multistations

Ver documento

Detalhes bibliográficos
Resumo:This study concerns making weather predictions for a location where no data is available, using meteorological datasets from nearby stations. The hindcast with multiple stations is performed with different variants of the Analog Ensemble (AnEn) method. In addition to the traditional Monache metric used to identify analogs in datasets from one or two stations, several new metrics are explored, namely cosine similarity, normalization, and k-means clustering. These were analyzed and benchmarked to find the ones that bring improvements. The best results were obtained with the k-means metric, yielding between 3% and 30% of lower quadratic error when compared against the Monache metric. Also, by making the predictors to include two stations, the performance of the hindcast improved, decreasing the error up to 16%, depending on the correlation between the predictor stations.
Autores principais:Balsa, Carlos
Outros Autores:Rodrigues, Carlos Veiga; Lopes, Isabel Maria; Rufino, José
Assunto:Analog ensembles Metrics Hindcasting Time series Meteorological data
Ano:2020
País:Portugal
Tipo de documento:artigo
Tipo de acesso:acesso aberto
Instituição associada:Instituto Politécnico de Bragança
Idioma:inglês
Origem:Biblioteca Digital do IPB
_version_ 1867173244967059456
author Balsa, Carlos
author2 Rodrigues, Carlos Veiga
Lopes, Isabel Maria
Rufino, José
author2_role author
author
author
author_facet Balsa, Carlos
Rodrigues, Carlos Veiga
Lopes, Isabel Maria
Rufino, José
author_role author
contributor_name_str_mv Biblioteca Digital do IPB
country_str PT
creators_json_txt [{\"Person.name\":\"Balsa, Carlos\",\"Person.identifier.orcid\":\"0000-0003-2431-8665\"},{\"Person.name\":\"Rodrigues, Carlos Veiga\"},{\"Person.name\":\"Lopes, Isabel Maria\",\"Person.identifier.orcid\":\"0000-0002-5614-3516\"},{\"Person.name\":\"Rufino, José\",\"Person.identifier.orcid\":\"0000-0002-1344-8264\"}]
datacite.contributors.contributor.contributorName.fl_str_mv Biblioteca Digital do IPB
datacite.creators.creator.creatorName.fl_str_mv Balsa, Carlos
Rodrigues, Carlos Veiga
Lopes, Isabel Maria
Rufino, José
datacite.date.Accepted.fl_str_mv 2020-01-01T00:00:00Z
datacite.date.available.fl_str_mv 2023-02-16T10:03:53Z
datacite.date.embargoed.fl_str_mv 2023-02-16T10:03:53Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_abf2
datacite.subjects.subject.fl_str_mv Analog ensembles
Metrics
Hindcasting
Time series
Meteorological data
datacite.titles.title.fl_str_mv Using analog ensembles with alternative metrics for hindcasting with multistations
dc.contributor.none.fl_str_mv Biblioteca Digital do IPB
dc.creator.none.fl_str_mv Balsa, Carlos
Rodrigues, Carlos Veiga
Lopes, Isabel Maria
Rufino, José
dc.date.Accepted.fl_str_mv 2020-01-01T00:00:00Z
dc.date.available.fl_str_mv 2023-02-16T10:03:53Z
dc.date.embargoed.fl_str_mv 2023-02-16T10:03:53Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10198/26978
dc.language.none.fl_str_mv eng
dc.rights.cclincense.fl_str_mv http://creativecommons.org/licenses/by/4.0/
dc.rights.none.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.subject.none.fl_str_mv Analog ensembles
Metrics
Hindcasting
Time series
Meteorological data
dc.title.fl_str_mv Using analog ensembles with alternative metrics for hindcasting with multistations
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_6501
description This study concerns making weather predictions for a location where no data is available, using meteorological datasets from nearby stations. The hindcast with multiple stations is performed with different variants of the Analog Ensemble (AnEn) method. In addition to the traditional Monache metric used to identify analogs in datasets from one or two stations, several new metrics are explored, namely cosine similarity, normalization, and k-means clustering. These were analyzed and benchmarked to find the ones that bring improvements. The best results were obtained with the k-means metric, yielding between 3% and 30% of lower quadratic error when compared against the Monache metric. Also, by making the predictors to include two stations, the performance of the hindcast improved, decreasing the error up to 16%, depending on the correlation between the predictor stations.
dirty 0
eu_rights_str_mv openAccess
format article
fulltext.url.fl_str_mv https://bibliotecadigital.ipb.pt/bitstreams/feaf52fa-66cf-4ffe-85f8-b25b7c14d78c/download
id ipb_dfcf75333f07b00a7c9e4cfe5e2e5088
identifier.url.fl_str_mv http://hdl.handle.net/10198/26978
instacron_str ipb
institution Instituto Politécnico de Bragança
instname_str Instituto Politécnico de Bragança
language eng
network_acronym_str ipb
network_name_str Biblioteca Digital do IPB
oai_identifier_str oai:bibliotecadigital.ipb.pt:10198/26978
organization_str_mv urn:organizationAcronym:ipb
person_str_mv Balsa, Carlos
Balsa, Carlos
https://www.ciencia-id.pt/DE1E-2F7A-AAB1
DE1E-2F7A-AAB1
http://orcid.org/0000-0003-2431-8665
0000-0003-2431-8665
Rodrigues, Carlos Veiga
Lopes, Isabel Maria
Lopes, Isabel Maria
https://www.ciencia-id.pt/8812-AE1C-A316
8812-AE1C-A316
http://orcid.org/0000-0002-5614-3516
0000-0002-5614-3516
Rufino, José
Rufino, José
https://www.ciencia-id.pt/C414-F47F-6323
C414-F47F-6323
http://orcid.org/0000-0002-1344-8264
0000-0002-1344-8264
publishDate 2020
reponame_str Biblioteca Digital do IPB
repository_id_str urn:repositoryAcronym:ipb
service_str_mv urn:repositoryAcronym:ipb
spelling engpt_PTThis study concerns making weather predictions for a location where no data is available, using meteorological datasets from nearby stations. The hindcast with multiple stations is performed with different variants of the Analog Ensemble (AnEn) method. In addition to the traditional Monache metric used to identify analogs in datasets from one or two stations, several new metrics are explored, namely cosine similarity, normalization, and k-means clustering. These were analyzed and benchmarked to find the ones that bring improvements. The best results were obtained with the k-means metric, yielding between 3% and 30% of lower quadratic error when compared against the Monache metric. Also, by making the predictors to include two stations, the performance of the hindcast improved, decreasing the error up to 16%, depending on the correlation between the predictor stations.application/pdfpt_PTUsing analog ensembles with alternative metrics for hindcasting with multistationsPersonalBalsa, CarlosDSpacehttp://dspace.org/items/d0e5ccff-9696-4f4f-9567-8d698a6bf17dDSpacehttp://dspace.org/items/d0e5ccff-9696-4f4f-9567-8d698a6bf17dBalsaCarlosCiência IDhttps://www.ciencia-id.ptDE1E-2F7A-AAB1ORCIDhttp://orcid.org0000-0003-2431-8665Researcher IDhttps://www.researcherid.comM-8735-2013Scopus Author IDhttps://www.scopus.com23391719100Rodrigues, Carlos VeigaPersonalLopes, Isabel MariaDSpacehttp://dspace.org/items/111716db-94a0-4c24-b739-330dc2ae79fcDSpacehttp://dspace.org/items/111716db-94a0-4c24-b739-330dc2ae79fcLopesIsabel MariaCiência IDhttps://www.ciencia-id.pt8812-AE1C-A316ORCIDhttp://orcid.org0000-0002-5614-3516Researcher IDhttps://www.researcherid.comA-1728-2014Scopus Author IDhttps://www.scopus.com55211017300Scopus Author IDhttps://www.scopus.com57190212117Scopus Author IDhttps://www.scopus.com57207843433PersonalRufino, JoséDSpacehttp://dspace.org/items/1e24d2ce-a354-442a-bef8-eebadd94b385DSpacehttp://dspace.org/items/1e24d2ce-a354-442a-bef8-eebadd94b385RufinoJoséCiência IDhttps://www.ciencia-id.ptC414-F47F-6323ORCIDhttp://orcid.org0000-0002-1344-8264Scopus Author IDhttps://www.scopus.com55947199100Scopus Author IDhttps://www.scopus.com57188967176HostingInstitutionOrganizationalBiblioteca Digital do IPBe-mailmailto:dspace@ipb.ptdspace@ipb.ptDOIIsPartOfDOI: 10.55969/paradigmplus.v1n2a12023-02-16T10:03:53Z20202020-01-01T00:00:00ZHandlehttp://hdl.handle.net/10198/26978http://purl.org/coar/access_right/c_abf2open accessAnalog ensemblesMetricsHindcastingTime seriesMeteorological data790162 bytesliteraturehttp://purl.org/coar/resource_type/c_6501journal article2020http://creativecommons.org/licenses/by/4.0/http://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://bibliotecadigital.ipb.pt/bitstreams/feaf52fa-66cf-4ffe-85f8-b25b7c14d78c/downloadJournal ParadigmPlus12117
spellingShingle Using analog ensembles with alternative metrics for hindcasting with multistations
Balsa, Carlos
Analog ensembles
Metrics
Hindcasting
Time series
Meteorological data
status SINGLETON
subject.fl_str_mv Analog ensembles
Metrics
Hindcasting
Time series
Meteorological data
title Using analog ensembles with alternative metrics for hindcasting with multistations
title_full Using analog ensembles with alternative metrics for hindcasting with multistations
title_fullStr Using analog ensembles with alternative metrics for hindcasting with multistations
title_full_unstemmed Using analog ensembles with alternative metrics for hindcasting with multistations
title_short Using analog ensembles with alternative metrics for hindcasting with multistations
title_sort Using analog ensembles with alternative metrics for hindcasting with multistations
topic Analog ensembles
Metrics
Hindcasting
Time series
Meteorological data
topic_facet Analog ensembles
Metrics
Hindcasting
Time series
Meteorological data
url http://hdl.handle.net/10198/26978
visible 1