North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago

Detalhes bibliográficos
Autor(a) principal: Monteiro, Nuno Manuel Rocha
Data de Publicação: 2021
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10773/32005
Resumo: This work focuses on implementing an unstructured grid wind-wave model for the North Atlantic with high resolution around the Azores archipelago. For that purpose, the OceanMesh2D software was implemented to generate the computational mesh, and the WAVEWATCH III (WW3) was used to model the wind-waves. Two HPC clusters were used to run WW3: ARGUS (from the University of Aveiro) and BOB (from the Minho Advanced Computing Center). Two different numerical schemes (explicit and implicit) and two different CPU parallelization algorithms (card-deck and domain decomposition) were used and tested in WW3 for two-week hindcast simulations in ARGUS and BOB. The explicit scheme was less efficient than the implicit scheme in BOB for simulations with 32 to 500 CPUs. Implementing the explicit scheme in ARGUS was slightly more efficient than the implicit scheme implementation in BOB for a lower number of CPU cores. However, in BOB, the implicit scheme with domain decomposition has proven to be scalable for up to 1000 CPU cores, while the explicit scheme with card-deck parallelization was scalable up to 500 cores. WW3 was validated using almost two years of data recorded by four nearshore wave buoys located in the different islands of the archipelago. Error metrics and wave roses for the four buoys were estimated. High energetic wind-wave events induced by extreme atmospheric phenomena passing by the Azores are studied in detail. Moreover, the Azores islands’ wind-wave shadow effect in storm conditions is analyzed.
id RCAP_7c91bb9d9c548f8f2af87e0de652716d
oai_identifier_str oai:ria.ua.pt:10773/32005
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling North Atlantic unstructured grid wind-wave model with high resolution around Azores ArchipelagoWAVEWATCH IIIOceanMesh2DNumerical simulationHPCExtreme eventsThis work focuses on implementing an unstructured grid wind-wave model for the North Atlantic with high resolution around the Azores archipelago. For that purpose, the OceanMesh2D software was implemented to generate the computational mesh, and the WAVEWATCH III (WW3) was used to model the wind-waves. Two HPC clusters were used to run WW3: ARGUS (from the University of Aveiro) and BOB (from the Minho Advanced Computing Center). Two different numerical schemes (explicit and implicit) and two different CPU parallelization algorithms (card-deck and domain decomposition) were used and tested in WW3 for two-week hindcast simulations in ARGUS and BOB. The explicit scheme was less efficient than the implicit scheme in BOB for simulations with 32 to 500 CPUs. Implementing the explicit scheme in ARGUS was slightly more efficient than the implicit scheme implementation in BOB for a lower number of CPU cores. However, in BOB, the implicit scheme with domain decomposition has proven to be scalable for up to 1000 CPU cores, while the explicit scheme with card-deck parallelization was scalable up to 500 cores. WW3 was validated using almost two years of data recorded by four nearshore wave buoys located in the different islands of the archipelago. Error metrics and wave roses for the four buoys were estimated. High energetic wind-wave events induced by extreme atmospheric phenomena passing by the Azores are studied in detail. Moreover, the Azores islands’ wind-wave shadow effect in storm conditions is analyzed.O foco deste trabalho esteve na implementação de um modelo de agitação marítima não-estruturado para o Atlântico Norte com alta resolução em torno do arquipélago dos Açores. Para isso, o software OceanMesh2D foi implementado para gerar a malha e o WAVEWATCH III (WW3) foi utilizado para modelar a agitação marítima. Dois clusters HPC foram utilizados para executar o WW3: ARGUS (da Universidade de Aveiro) e o BOB (do Minho Advanced Computing Center). Dois esquemas numéricos distintos (explícito e implícito) e dois algoritmos de paralelização CPU (card deck e decomposição do domínio) foram utilizados e testados no WW3. As performances dessas configurações em termos do tempo elapsado foram comparadas para uma simulação hindcast de duas semanas, no ARGUS e no BOB. O esquema explícito foi menos eficiente que o esquema implícito nas simulações no BOB para 32 a 500 CPUs. Na implementação do esquema explícito do ARGUS, com card deck, foi ligeiramente mais eficiente do que o esquema implícito no BOB para um número mais baixo de CPUs. Contudo, no BOB, o esquema implícito com decomposição do domínio provou ser escalável até 1000 cores de CPU, enquanto que o esquema explícito com paralelização card-deck provou ser escalável até 500 cores. O WW3 foi validado com dados de quase dois anos registados por bóias costeiras localizadas em diferentes ilhas do arquipélago. Métricas de erro e rosas de vento de agitação marítima foram estimados para as bóias. Eventos de agitação marítima de elevada energia induzidos por fenómenos atmosféricos extremos que passaram pelos Açores são estudados em detalhe. Além disso, o efeito de sombra das ilhas dos Açores em condições de tempestade é analisado.2026-07-23T00:00:00Z2021-07-16T00:00:00Z2021-07-16info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10773/32005engMonteiro, Nuno Manuel Rochainfo:eu-repo/semantics/embargoedAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2024-02-22T12:01:53Zoai:ria.ua.pt:10773/32005Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-20T03:03:50.227324Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago
title North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago
spellingShingle North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago
Monteiro, Nuno Manuel Rocha
WAVEWATCH III
OceanMesh2D
Numerical simulation
HPC
Extreme events
title_short North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago
title_full North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago
title_fullStr North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago
title_full_unstemmed North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago
title_sort North Atlantic unstructured grid wind-wave model with high resolution around Azores Archipelago
author Monteiro, Nuno Manuel Rocha
author_facet Monteiro, Nuno Manuel Rocha
author_role author
dc.contributor.author.fl_str_mv Monteiro, Nuno Manuel Rocha
dc.subject.por.fl_str_mv WAVEWATCH III
OceanMesh2D
Numerical simulation
HPC
Extreme events
topic WAVEWATCH III
OceanMesh2D
Numerical simulation
HPC
Extreme events
description This work focuses on implementing an unstructured grid wind-wave model for the North Atlantic with high resolution around the Azores archipelago. For that purpose, the OceanMesh2D software was implemented to generate the computational mesh, and the WAVEWATCH III (WW3) was used to model the wind-waves. Two HPC clusters were used to run WW3: ARGUS (from the University of Aveiro) and BOB (from the Minho Advanced Computing Center). Two different numerical schemes (explicit and implicit) and two different CPU parallelization algorithms (card-deck and domain decomposition) were used and tested in WW3 for two-week hindcast simulations in ARGUS and BOB. The explicit scheme was less efficient than the implicit scheme in BOB for simulations with 32 to 500 CPUs. Implementing the explicit scheme in ARGUS was slightly more efficient than the implicit scheme implementation in BOB for a lower number of CPU cores. However, in BOB, the implicit scheme with domain decomposition has proven to be scalable for up to 1000 CPU cores, while the explicit scheme with card-deck parallelization was scalable up to 500 cores. WW3 was validated using almost two years of data recorded by four nearshore wave buoys located in the different islands of the archipelago. Error metrics and wave roses for the four buoys were estimated. High energetic wind-wave events induced by extreme atmospheric phenomena passing by the Azores are studied in detail. Moreover, the Azores islands’ wind-wave shadow effect in storm conditions is analyzed.
publishDate 2021
dc.date.none.fl_str_mv 2021-07-16T00:00:00Z
2021-07-16
2026-07-23T00:00:00Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10773/32005
url http://hdl.handle.net/10773/32005
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/embargoedAccess
eu_rights_str_mv embargoedAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799137693482876928