Developing an Explainable Variational Autoencoder (VAE) Framework for Accurate Representation of Local Circulation in Taiwan

This study develops an explainable variational autoencoder (VAE) framework to efficiently generate high‐fidelity local circulation patterns in Taiwan, ensuring an accurate representation of the physical relationship between generated local circulation and upstream synoptic flow regimes. Large ensemb...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Journal of geophysical research. Atmospheres Jg. 129; H. 12
Hauptverfasser:	Hsieh, Min‐Ken, Wu, Chien‐Ming
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Washington Blackwell Publishing Ltd 28.06.2024
Schlagworte:	Accuracy Circulation Circulation patterns Climate and weather Climate change Climate change scenarios Climate prediction Climatic conditions deep generative model deep learning explainable artificial intelligence Flow distribution Flow pattern Fluid flow large eddy simulation local circulation Local flow Neural networks Representations Simulation Synoptic weather conditions Training Uncertainty Vortices Weather Weather conditions Weather patterns Wind speed Taiwan
ISSN:	2169-897X, 2169-8996
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Abstract	This study develops an explainable variational autoencoder (VAE) framework to efficiently generate high‐fidelity local circulation patterns in Taiwan, ensuring an accurate representation of the physical relationship between generated local circulation and upstream synoptic flow regimes. Large ensemble semi‐realistic simulations were conducted using a high‐resolution (2 km) model, TaiwanVVM, where critical characteristics of various synoptic flow regimes were carefully selected to focus on the effects of local circulation variations. The VAE was constructed to capture essential representations of local circulation scenarios associated with the lee vortices by training on the ensemble data set. The VAE's latent space effectively captures the synoptic flow regimes as controlling factors, aligning with the physical understanding of Taiwan's local circulation dynamics. The critical transition of flow regimes under the influence of southeasterly synoptic flow regimes is also well represented in the VAE's latent space. This indicates that the VAE can learn the nonlinear characteristics of the multiscale interactions involving the lee vortex. The latent space within VAE can serve as a reduced‐order model for predicting local circulation using synoptic wind speed and direction. This explainable VAE binds the physical reasoning to the predictions of the local circulation that ensures the physical examination of the uncertainty in accelerating the local weather assessments under various climate change scenarios. Plain Language Summary This research introduces an advanced neural network framework for generating high‐fidelity local flow patterns in Taiwan. This framework, known as an explainable variational autoencoder, can accurately simulate how wind patterns of synoptic weather conditions interact in this region. We used detailed simulations to train the variational autoencoder, ensuring it captures the complex relationships between local flow and larger‐scale weather patterns. By training on the detailed simulations, the variational autoencoder learned and represented these large‐scale weather patterns in a way that helps maintain the physical relationship between local flow prediction and the large‐scale weather patterns. One of the key outcomes of this study is the development of a reduced‐order model. This simplified model takes advantage of what we have learned about complex weather interactions and can quickly predict local weather under different conditions. This approach provides opportunities for physical examination of uncertainty in local circulation predictions using a neural network model under complex situations involving changing climate conditions. Key Points An explainable variational autoencoder is constructed to capture Taiwan's local circulation using TaiwanVVM ensemble simulations The representation of local circulation in the latent space of the VAE can be formulated as synoptic wind speed and direction This framework can effectively generate accurate local circulation in Taiwan for fast climate response assessment
AbstractList	This study develops an explainable variational autoencoder (VAE) framework to efficiently generate high‐fidelity local circulation patterns in Taiwan, ensuring an accurate representation of the physical relationship between generated local circulation and upstream synoptic flow regimes. Large ensemble semi‐realistic simulations were conducted using a high‐resolution (2 km) model, TaiwanVVM, where critical characteristics of various synoptic flow regimes were carefully selected to focus on the effects of local circulation variations. The VAE was constructed to capture essential representations of local circulation scenarios associated with the lee vortices by training on the ensemble data set. The VAE's latent space effectively captures the synoptic flow regimes as controlling factors, aligning with the physical understanding of Taiwan's local circulation dynamics. The critical transition of flow regimes under the influence of southeasterly synoptic flow regimes is also well represented in the VAE's latent space. This indicates that the VAE can learn the nonlinear characteristics of the multiscale interactions involving the lee vortex. The latent space within VAE can serve as a reduced‐order model for predicting local circulation using synoptic wind speed and direction. This explainable VAE binds the physical reasoning to the predictions of the local circulation that ensures the physical examination of the uncertainty in accelerating the local weather assessments under various climate change scenarios. This research introduces an advanced neural network framework for generating high‐fidelity local flow patterns in Taiwan. This framework, known as an explainable variational autoencoder, can accurately simulate how wind patterns of synoptic weather conditions interact in this region. We used detailed simulations to train the variational autoencoder, ensuring it captures the complex relationships between local flow and larger‐scale weather patterns. By training on the detailed simulations, the variational autoencoder learned and represented these large‐scale weather patterns in a way that helps maintain the physical relationship between local flow prediction and the large‐scale weather patterns. One of the key outcomes of this study is the development of a reduced‐order model. This simplified model takes advantage of what we have learned about complex weather interactions and can quickly predict local weather under different conditions. This approach provides opportunities for physical examination of uncertainty in local circulation predictions using a neural network model under complex situations involving changing climate conditions. An explainable variational autoencoder is constructed to capture Taiwan's local circulation using TaiwanVVM ensemble simulations The representation of local circulation in the latent space of the VAE can be formulated as synoptic wind speed and direction This framework can effectively generate accurate local circulation in Taiwan for fast climate response assessment This study develops an explainable variational autoencoder (VAE) framework to efficiently generate high‐fidelity local circulation patterns in Taiwan, ensuring an accurate representation of the physical relationship between generated local circulation and upstream synoptic flow regimes. Large ensemble semi‐realistic simulations were conducted using a high‐resolution (2 km) model, TaiwanVVM, where critical characteristics of various synoptic flow regimes were carefully selected to focus on the effects of local circulation variations. The VAE was constructed to capture essential representations of local circulation scenarios associated with the lee vortices by training on the ensemble data set. The VAE's latent space effectively captures the synoptic flow regimes as controlling factors, aligning with the physical understanding of Taiwan's local circulation dynamics. The critical transition of flow regimes under the influence of southeasterly synoptic flow regimes is also well represented in the VAE's latent space. This indicates that the VAE can learn the nonlinear characteristics of the multiscale interactions involving the lee vortex. The latent space within VAE can serve as a reduced‐order model for predicting local circulation using synoptic wind speed and direction. This explainable VAE binds the physical reasoning to the predictions of the local circulation that ensures the physical examination of the uncertainty in accelerating the local weather assessments under various climate change scenarios. Plain Language Summary This research introduces an advanced neural network framework for generating high‐fidelity local flow patterns in Taiwan. This framework, known as an explainable variational autoencoder, can accurately simulate how wind patterns of synoptic weather conditions interact in this region. We used detailed simulations to train the variational autoencoder, ensuring it captures the complex relationships between local flow and larger‐scale weather patterns. By training on the detailed simulations, the variational autoencoder learned and represented these large‐scale weather patterns in a way that helps maintain the physical relationship between local flow prediction and the large‐scale weather patterns. One of the key outcomes of this study is the development of a reduced‐order model. This simplified model takes advantage of what we have learned about complex weather interactions and can quickly predict local weather under different conditions. This approach provides opportunities for physical examination of uncertainty in local circulation predictions using a neural network model under complex situations involving changing climate conditions. Key Points An explainable variational autoencoder is constructed to capture Taiwan's local circulation using TaiwanVVM ensemble simulations The representation of local circulation in the latent space of the VAE can be formulated as synoptic wind speed and direction This framework can effectively generate accurate local circulation in Taiwan for fast climate response assessment This study develops an explainable variational autoencoder (VAE) framework to efficiently generate high‐fidelity local circulation patterns in Taiwan, ensuring an accurate representation of the physical relationship between generated local circulation and upstream synoptic flow regimes. Large ensemble semi‐realistic simulations were conducted using a high‐resolution (2 km) model, TaiwanVVM, where critical characteristics of various synoptic flow regimes were carefully selected to focus on the effects of local circulation variations. The VAE was constructed to capture essential representations of local circulation scenarios associated with the lee vortices by training on the ensemble data set. The VAE's latent space effectively captures the synoptic flow regimes as controlling factors, aligning with the physical understanding of Taiwan's local circulation dynamics. The critical transition of flow regimes under the influence of southeasterly synoptic flow regimes is also well represented in the VAE's latent space. This indicates that the VAE can learn the nonlinear characteristics of the multiscale interactions involving the lee vortex. The latent space within VAE can serve as a reduced‐order model for predicting local circulation using synoptic wind speed and direction. This explainable VAE binds the physical reasoning to the predictions of the local circulation that ensures the physical examination of the uncertainty in accelerating the local weather assessments under various climate change scenarios.
Author	Hsieh, Min‐Ken Wu, Chien‐Ming
Author_xml	– sequence: 1 givenname: Min‐Ken orcidid: 0000-0002-4622-8524 surname: Hsieh fullname: Hsieh, Min‐Ken organization: National Taiwan University – sequence: 2 givenname: Chien‐Ming orcidid: 0000-0001-9295-7181 surname: Wu fullname: Wu, Chien‐Ming email: mog@as.ntu.edu.tw organization: National Taiwan University
BookMark	eNp9kE1LxDAQhoOsoK578wcEvChYTZM2TY7L7vrFgiCreCtpOpFoN6lp6yr4461WxJNzmWF45oV59tDIeQcIHcTkNCZUnlFCk-s5SeKYZ1tol8ZcRkJKPvqds4cdNGmaJ9KXICxJk130MYdXqHxt3SNWDi_e6kpZp4oK8L0KVrXWO1Xhadd6cNqXEPDR_XRxjM-DWsPGh2dsfMBTrbugWsC3UAdowLXfl9gbvPS6D5jZoLtqWFqHV8pulNtH20ZVDUx--hjdnS9Ws8toeXNxNZsuI015IiIl0gxSZjTPCqG41EZmpU5JIoUhjJNS6qIAZXgiE0bKDGIqCqZFLKg02gg2RodDbh38SwdNmz_5LvR_NTkjGaXsy0VPnQyUDr5pApi8Dnatwnsek_xLcf5XcY-zAd_YCt7_ZfPri9t5KjkT7BMTfH9U
Cites_doi	10.2151/jmsj.2022‐028 10.48550/arXiv.2112.08440 10.1038/s41598‐023‐49455‐w 10.1175/JAMC‐D‐22‐0102.1 10.1029/2020GL092032 10.48550/arXiv.2104.12469 10.48550/arXiv.2309.15214 10.1002/wea.543 10.1126/science.1127647 10.1561/2200000056 10.48550/arXiv.2212.12794 10.5194/gmd‐13‐3887‐2020 10.48550/arXiv.2012.01233 10.17605/OSF.IO/4ZUTJ 10.48550/arXiv.1406.2661 10.1175/2007MWR2095.1 10.1002/asl.1150 10.48550/arXiv.2102.04534 10.48550/arXiv.2202.11214 10.1029/2021MS002631 10.1029/2011MS000061 10.23919/MVA51890.2021.9511404 10.1002/essoar.10512517.1 10.1029/2020EA001340 10.2151/jmsj.2019‐031 10.5194/gmd‐9‐1937‐2016 10.5194/acp‐21‐16893‐2021 10.1038/s41586‐023‐06185‐3 10.1029/2022MS003130 10.1016/j.atmosenv.2020.117418 10.1109/TKDE.2017.2720168 10.48550/arXiv.2108.00048 10.48550/arXiv.1710.11431 10.5194/acp‐21‐16709‐2021 10.5194/gmd‐16‐6433‐2023 10.1073/pnas.1918964117 10.1109/MCSE.2007.55 10.48550/arXiv.2009.08454 10.1002/aic.690370209 10.1002/2015MS000514 10.48550/arXiv.2006.11239 10.1007/s13143‐019‐00116‐x
ContentType	Journal Article
Copyright	2024. American Geophysical Union. All Rights Reserved.
Copyright_xml	– notice: 2024. American Geophysical Union. All Rights Reserved.
DBID	AAYXX CITATION 7TG 7UA 8FD C1K F1W FR3 H8D H96 KL. KR7 L.G L7M
DOI	10.1029/2024JD041167
DatabaseName	CrossRef Meteorological & Geoastrophysical Abstracts Water Resources Abstracts Technology Research Database Environmental Sciences and Pollution Management ASFA: Aquatic Sciences and Fisheries Abstracts Engineering Research Database Aerospace Database Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources Meteorological & Geoastrophysical Abstracts - Academic Civil Engineering Abstracts Aquatic Science & Fisheries Abstracts (ASFA) Professional Advanced Technologies Database with Aerospace
DatabaseTitle	CrossRef Aerospace Database Civil Engineering Abstracts Aquatic Science & Fisheries Abstracts (ASFA) Professional Meteorological & Geoastrophysical Abstracts Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources Technology Research Database ASFA: Aquatic Sciences and Fisheries Abstracts Engineering Research Database Advanced Technologies Database with Aerospace Meteorological & Geoastrophysical Abstracts - Academic Water Resources Abstracts Environmental Sciences and Pollution Management
DatabaseTitleList	CrossRef Aerospace Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Meteorology & Climatology
EISSN	2169-8996
EndPage	n/a
ExternalDocumentID	10_1029_2024JD041167 JGRD59638
Genre	researchArticle
GeographicLocations	Taiwan
GeographicLocations_xml	– name: Taiwan
GrantInformation_xml	– fundername: National Science and Technology Council funderid: 112‐2111‐M‐002‐015‐ – fundername: National Taiwan University funderid: NTU112L7832
GroupedDBID	05W 0R~ 1OC 24P 33P 50Y 52M 5VS 702 8-1 A00 AAESR AAHHS AAHQN AAIHA AAMNL AANLZ AAXRX AAYCA AAZKR ABCUV ABJNI ACAHQ ACCFJ ACCZN ACGFS ACIWK ACPOU ACXBN ACXQS ADBBV ADEOM ADKYN ADMGS ADOZA ADXAS ADZMN AEEZP AEIGN AEQDE AEUYR AFBPY AFFPM AFGKR AFPWT AFRAH AFWVQ AHBTC AITYG AIURR AIWBW AJBDE ALMA_UNASSIGNED_HOLDINGS ALUQN ALVPJ AMYDB AZFZN AZVAB BFHJK BMXJE BRXPI DPXWK DRFUL DRSTM EBS G-S HGLYW HZ~ LATKE LEEKS LITHE LOXES LUTES LYRES MEWTI MSFUL MSSTM MXFUL MXSTM MY~ O9- P2W R.K RNS ROL SUPJJ WBKPD WIN WXSBR WYJ ~OA AAMMB AAYXX AEFGJ AEYWJ AGHNM AGXDD AGYGG AIDQK AIDYY CITATION 7TG 7UA 8FD C1K F1W FR3 H8D H96 KL. KR7 L.G L7M
ID	FETCH-LOGICAL-c2648-a857e53fc67b8a69cf97dc50498f0360d9cbbeaf649430d7e128b3c81829fcf83
IEDL.DBID	DRFUL
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001252823500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2169-897X
IngestDate	Sat Nov 01 14:46:32 EDT 2025 Sat Nov 29 05:39:26 EST 2025 Wed Jan 22 17:18:14 EST 2025
IsPeerReviewed	true
IsScholarly	true
Issue	12
Language	English
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c2648-a857e53fc67b8a69cf97dc50498f0360d9cbbeaf649430d7e128b3c81829fcf83
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-4622-8524 0000-0001-9295-7181
PQID	3072233454
PQPubID	54657
PageCount	20
ParticipantIDs	proquest_journals_3072233454 crossref_primary_10_1029_2024JD041167 wiley_primary_10_1029_2024JD041167_JGRD59638
PublicationCentury	2000
PublicationDate	28 June 2024
PublicationDateYYYYMMDD	2024-06-28
PublicationDate_xml	– month: 06 year: 2024 text: 28 June 2024 day: 28
PublicationDecade	2020
PublicationPlace	Washington
PublicationPlace_xml	– name: Washington
PublicationTitle	Journal of geophysical research. Atmospheres
PublicationYear	2024
Publisher	Blackwell Publishing Ltd
Publisher_xml	– name: Blackwell Publishing Ltd
References	2021; 8 2021; 48 2021; 21 2023; 13 1991; 37 2019; 97 2019; 55 2023; 16 2019; 12 2020; 227 2020; 13 2017; 29 2006; 313 2011; 3 2022; 100 2021; 35 2023; 62 2010; 65 2023; 24 2023 2022 2021 2020 2007; 9 2022; 14 2020; 117 2017 2014 2008; 136 2013 2023; 619 2016; 8 2016; 9 e_1_2_7_6_1 e_1_2_7_5_1 e_1_2_7_4_1 e_1_2_7_3_1 e_1_2_7_9_1 e_1_2_7_8_1 e_1_2_7_7_1 e_1_2_7_19_1 e_1_2_7_18_1 e_1_2_7_17_1 e_1_2_7_16_1 e_1_2_7_40_1 e_1_2_7_2_1 e_1_2_7_15_1 e_1_2_7_41_1 e_1_2_7_14_1 e_1_2_7_42_1 e_1_2_7_13_1 e_1_2_7_43_1 e_1_2_7_12_1 e_1_2_7_44_1 e_1_2_7_11_1 e_1_2_7_10_1 e_1_2_7_26_1 e_1_2_7_27_1 e_1_2_7_28_1 e_1_2_7_29_1 Kingma D. P. (e_1_2_7_21_1) 2013 e_1_2_7_30_1 e_1_2_7_25_1 e_1_2_7_31_1 e_1_2_7_24_1 e_1_2_7_32_1 e_1_2_7_23_1 e_1_2_7_33_1 e_1_2_7_22_1 e_1_2_7_34_1 e_1_2_7_35_1 e_1_2_7_20_1 e_1_2_7_36_1 e_1_2_7_37_1 e_1_2_7_38_1 e_1_2_7_39_1
References_xml	– year: 2020 article-title: Denoising diffusion probabilistic models publication-title: arXiv – year: 2021 article-title: Generative modeling of spatio‐temporal weather patterns with extreme event conditioning publication-title: arXiv – year: 2021 article-title: A modular framework for extreme weather generation publication-title: arXiv – volume: 117 start-page: 16805 issue: 29 year: 2020 end-page: 16815 article-title: Adversarial super‐resolution of climatological wind and solar data publication-title: Proceedings of the National Academy of Sciences of the United States of America – volume: 65 start-page: 180 issue: 7 year: 2010 end-page: 185 article-title: Robust adaptation to climate change publication-title: Weather – year: 2013 article-title: Auto‐encoding variational Bayes publication-title: arXiv – volume: 313 start-page: 504 issue: 5786 year: 2006 end-page: 507 article-title: Reducing the dimensionality of data with neural networks publication-title: Science – year: 2022 article-title: FourCastNet: A global data‐driven high‐resolution weather model using adaptive Fourier neural operators publication-title: arXiv – volume: 16 start-page: 6433 issue: 22 year: 2023 end-page: 6477 article-title: Machine learning for numerical weather and climate modelling: A review publication-title: Geoscientific Model Development – volume: 619 start-page: 533 issue: 7970 year: 2023 end-page: 538 article-title: Accurate medium‐range global weather forecasting with 3D neural networks publication-title: Nature – volume: 21 start-page: 16709 issue: 22 year: 2021 end-page: 16725 article-title: Tracking the influence of cloud condensation nuclei on summer diurnal precipitating systems over complex topography in Taiwan publication-title: Atmospheric Chemistry and Physics – volume: 9 start-page: 1937 issue: 5 year: 2016 end-page: 1958 article-title: Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization publication-title: Geoscientific Model Development – year: 2021 article-title: Climate‐invariant machine learning publication-title: arXiv – volume: 48 issue: 5 year: 2021 article-title: PrecipGAN: Merging microwave and infrared data for satellite precipitation estimation using generative adversarial network publication-title: Geophysical Research Letters – volume: 24 issue: 5 year: 2023 article-title: A deep learning framework for analyzing cloud characteristics of aggregated convection using cloud‐resolving model simulations publication-title: Atmospheric Science Letters – volume: 3 issue: 2 year: 2011 article-title: Inclusion of surface topography into the vector vorticity equation model (VVM): Inclusion of surface topography into the VVM publication-title: Journal of Advances in Modeling Earth Systems – volume: 9 start-page: 90 issue: 3 year: 2007 end-page: 95 – volume: 14 issue: 8 year: 2022 article-title: Non‐linear dimensionality reduction with a variational encoder decoder to understand convective processes in climate models publication-title: Journal of Advances in Modeling Earth Systems – year: 2022 article-title: Implicit learning of convective organization explains precipitation stochasticity publication-title: Atmospheric Sciences – volume: 8 start-page: 212 issue: 1 year: 2016 end-page: 223 article-title: Representation of topography by partial steps using the immersed boundary method in a vector vorticity equation model (VVM): VVM partial step publication-title: Journal of Advances in Modeling Earth Systems – start-page: 1 year: 2021 end-page: 5 – volume: 14 issue: 3 year: 2022 article-title: A library of large‐eddy simulations forced by global climate models publication-title: Journal of Advances in Modeling Earth Systems – year: 2017 article-title: Physics‐guided neural networks (PGNN): An application in lake temperature modeling publication-title: arXiv – volume: 13 issue: 1 year: 2023 article-title: Comparing storm resolving models and climates via unsupervised machine learning publication-title: Scientific Reports – volume: 55 start-page: 701 issue: 4 year: 2019 end-page: 717 article-title: Implementation of the land surface processes into a vector vorticity equation model (VVM) to study its impact on afternoon thunderstorms over complex topography in Taiwan publication-title: Asia‐Pacific Journal of Atmospheric Sciences – volume: 13 start-page: 3887 issue: 9 year: 2020 end-page: 3904 article-title: Taiwan Earth system model version 1: Description and evaluation of mean state publication-title: Geoscientific Model Development – year: 2022 – year: 2014 article-title: Generative adversarial networks publication-title: arXiv – volume: 227 year: 2020 article-title: Characteristics of the upstream flow patterns during PM2.5 pollution events over a complex island topography publication-title: Atmospheric Environment – volume: 29 start-page: 2318 issue: 10 year: 2017 end-page: 2331 article-title: Theory‐guided data science: A new paradigm for scientific discovery from data publication-title: IEEE Transactions on Knowledge and Data Engineering – volume: 12 start-page: 307 issue: 4 year: 2019 end-page: 392 article-title: An introduction to variational autoencoders publication-title: Foundations and Trends® in Machine Learning – volume: 35 start-page: 6750 issue: 8 year: 2021 end-page: 6758 article-title: ExGAN: Adversarial generation of extreme samples publication-title: arXiv – volume: 100 start-page: 555 issue: 3 year: 2022 end-page: 573 article-title: The roles of local circulation and boundary layer development in tracer transport over complex topography in Central Taiwan publication-title: Journal of the Meteorological Society of Japan. Series II – year: 2020 article-title: Investigating two super‐resolution methods for downscaling precipitation: ESRGAN and CAR publication-title: arXiv – volume: 97 start-page: 501 issue: 2 year: 2019 end-page: 517 article-title: The precipitation hotspots of afternoon thunderstorms over the Taipei Basin: Idealized numerical simulations publication-title: Journal of the Meteorological Society of Japan. Series II – volume: 62 start-page: 427 issue: 3 year: 2023 end-page: 439 article-title: The observation‐based index to investigate the role of the lee vortex in enhancing air pollution over northwestern Taiwan publication-title: Journal of Applied Meteorology and Climatology – volume: 136 start-page: 276 issue: 1 year: 2008 end-page: 294 article-title: A three‐dimensional anelastic model based on the vorticity equation publication-title: Monthly Weather Review – year: 2022 article-title: GraphCast: Learning skillful medium‐range global weather forecasting publication-title: arXiv – volume: 8 issue: 3 year: 2021 article-title: A deep learning approach to radar‐based QPE publication-title: Earth and Space Science – year: 2023 article-title: Residual diffusion modeling for Km‐scale atmospheric downscaling publication-title: arXiv – year: 2021 article-title: Controlling weather field synthesis using variational autoencoders publication-title: arXiv – volume: 37 start-page: 233 issue: 2 year: 1991 end-page: 243 article-title: Nonlinear principal component analysis using autoassociative neural networks publication-title: AIChE Journal – volume: 21 start-page: 16893 issue: 22 year: 2021 end-page: 16910 article-title: Air quality deterioration episode associated with a typhoon over the complex topographic environment in central Taiwan publication-title: Atmospheric Chemistry and Physics – ident: e_1_2_7_16_1 doi: 10.2151/jmsj.2022‐028 – ident: e_1_2_7_3_1 doi: 10.48550/arXiv.2112.08440 – ident: e_1_2_7_31_1 doi: 10.1038/s41598‐023‐49455‐w – ident: e_1_2_7_17_1 doi: 10.1175/JAMC‐D‐22‐0102.1 – ident: e_1_2_7_38_1 doi: 10.1029/2020GL092032 – ident: e_1_2_7_23_1 doi: 10.48550/arXiv.2104.12469 – ident: e_1_2_7_30_1 doi: 10.48550/arXiv.2309.15214 – ident: e_1_2_7_40_1 doi: 10.1002/wea.543 – ident: e_1_2_7_14_1 doi: 10.1126/science.1127647 – ident: e_1_2_7_22_1 doi: 10.1561/2200000056 – ident: e_1_2_7_27_1 doi: 10.48550/arXiv.2212.12794 – ident: e_1_2_7_28_1 doi: 10.5194/gmd‐13‐3887‐2020 – ident: e_1_2_7_39_1 doi: 10.48550/arXiv.2012.01233 – ident: e_1_2_7_37_1 doi: 10.17605/OSF.IO/4ZUTJ – ident: e_1_2_7_12_1 doi: 10.48550/arXiv.1406.2661 – ident: e_1_2_7_19_1 doi: 10.1175/2007MWR2095.1 – ident: e_1_2_7_7_1 doi: 10.1002/asl.1150 – ident: e_1_2_7_44_1 doi: 10.48550/arXiv.2102.04534 – ident: e_1_2_7_33_1 doi: 10.48550/arXiv.2202.11214 – ident: e_1_2_7_35_1 doi: 10.1029/2021MS002631 – ident: e_1_2_7_41_1 doi: 10.1029/2011MS000061 – ident: e_1_2_7_13_1 doi: 10.23919/MVA51890.2021.9511404 – ident: e_1_2_7_34_1 doi: 10.1002/essoar.10512517.1 – ident: e_1_2_7_43_1 doi: 10.1029/2020EA001340 – ident: e_1_2_7_25_1 doi: 10.2151/jmsj.2019‐031 – ident: e_1_2_7_11_1 doi: 10.5194/gmd‐9‐1937‐2016 – ident: e_1_2_7_29_1 doi: 10.5194/acp‐21‐16893‐2021 – ident: e_1_2_7_5_1 doi: 10.1038/s41586‐023‐06185‐3 – ident: e_1_2_7_2_1 doi: 10.1029/2022MS003130 – year: 2013 ident: e_1_2_7_21_1 article-title: Auto‐encoding variational Bayes publication-title: arXiv – ident: e_1_2_7_26_1 doi: 10.1016/j.atmosenv.2020.117418 – ident: e_1_2_7_20_1 doi: 10.1109/TKDE.2017.2720168 – ident: e_1_2_7_32_1 doi: 10.48550/arXiv.2108.00048 – ident: e_1_2_7_9_1 doi: 10.48550/arXiv.1710.11431 – ident: e_1_2_7_6_1 doi: 10.5194/acp‐21‐16709‐2021 – ident: e_1_2_7_10_1 doi: 10.5194/gmd‐16‐6433‐2023 – ident: e_1_2_7_36_1 doi: 10.1073/pnas.1918964117 – ident: e_1_2_7_18_1 doi: 10.1109/MCSE.2007.55 – ident: e_1_2_7_4_1 doi: 10.48550/arXiv.2009.08454 – ident: e_1_2_7_24_1 doi: 10.1002/aic.690370209 – ident: e_1_2_7_8_1 doi: 10.1002/2015MS000514 – ident: e_1_2_7_15_1 doi: 10.48550/arXiv.2006.11239 – ident: e_1_2_7_42_1 doi: 10.1007/s13143‐019‐00116‐x
SSID	ssj0000803454
Score	2.260416
Snippet	This study develops an explainable variational autoencoder (VAE) framework to efficiently generate high‐fidelity local circulation patterns in Taiwan, ensuring...
SourceID	proquest crossref wiley
SourceType	Aggregation Database Index Database Publisher
SubjectTerms	Accuracy Circulation Circulation patterns Climate and weather Climate change Climate change scenarios Climate prediction Climatic conditions deep generative model deep learning explainable artificial intelligence Flow distribution Flow pattern Fluid flow large eddy simulation local circulation Local flow Neural networks Representations Simulation Synoptic weather conditions Training Uncertainty Vortices Weather Weather conditions Weather patterns Wind speed
Title	Developing an Explainable Variational Autoencoder (VAE) Framework for Accurate Representation of Local Circulation in Taiwan
URI	https://onlinelibrary.wiley.com/doi/abs/10.1029%2F2024JD041167 https://www.proquest.com/docview/3072233454
Volume	129
WOSCitedRecordID	wos001252823500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVWIB databaseName: Wiley Online Library Free Content customDbUrl: eissn: 2169-8996 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000803454 issn: 2169-897X databaseCode: WIN dateStart: 20130101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell – providerCode: PRVWIB databaseName: Wiley Online Library Full Collection 2020 customDbUrl: eissn: 2169-8996 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000803454 issn: 2169-897X databaseCode: DRFUL dateStart: 20130101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3di9NAEB-8Ox988Vuueh7zoKJgMM3HZvextFf1qEXKXe1b2Ex2ISCppO3di3-8s9ttrS-C-BIICUvIfP1mdua3AK-UrIW2tY5IxDrKKI0jWRNfhO0bKVVSeQa--aSYTuViob6GgpubhdnyQ-wLbs4yvL92Bq6rVSAbcByZnLVnl6M4cxsJR3Di5qo4-ToZzcbXk32VhfFQmvmj0JK-UJFUxSJ0v_MiHw6X-DMu_Qabh5DVx5zxg__92odwP6BNHGzV4xHcMe1j6H1hoLzsfD0d3-Dwe8Oo1d89gZ-j_RAV6hZdh14Yr8I5Z9WhcoiDzXrpGDBr0-Hb-eDiHY53TV7IKBgHRBvHQYEz32gb5ptaXFqcuOCJw6ajcHAYNi1e6eZWt0_henxxNfwUhQMaInKNcZGWeWHy1JIoKqmFIquKmnJOOqTlyBjXiqrKaCsyR_JeF4aDYZUSY4REWbIyfQbH7bI1p4AppYUqDCtHrjkjqnWsq76OJaVK5iRED17vxFP-2PJwlH7_PFHl4b_twdlOdmWwxlXJfoxRkFOEHrz3UvrrGuXlx9kod57p-b-9_gLuuQeukyyRZ3C87jbmJdylm3Wz6s6Dbp7D0bfP018V0uTU
linkProvider	Wiley-Blackwell
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3da9swED-6dLC9bN1HWbauu4dtbDBTx7L18RiSpl3nhhHSkDcjyxIYilPcpHvZHz9JUbLsZVD6YjAIYXynu9-d7n4H8FHwikpTyUjRWEapInHEK2Uf1PQ05yIpPQPfLGfjMZ_Pxc8w59T1wqz5IbYJN3cyvL12B9wlpAPbgCPJtGF7ejGMU3eT8Aj2U0oY78D-cDK6yrdpFguISOpnoSU9KiIu2DyUv9tNTna3-Ncx_UWbu5jVO53R8wd_7gE8C3gT-2sFeQF7unkJ3UsLlRetz6jjZxxc1xa3-rdX8Hu4baNC2aCr0QsNVjizcXXIHWJ_tVw4DsxKt_hl1j_9iqNNmRdaHIx9pVaOhQInvtQ2dDg1uDCYO_eJg7pVYXQY1g1OZf1LNq_hanQ6HZxHYURDpFxpXCR5xnRGjKKs5JIKZQSrVGbDDm6sb4wrocpSS0NTR_NeMW3dYUmURQmJMMpwcgidZtHoN4BEESaYtuqRSRsTVTKWZU_GXBHBM0VpFz5t5FPcrJk4Cn-Dnohi99924WgjvCKcx9vCWjKLg5wmdOGbF9N_9yguzibDzNmmt_db_gGenE8v8yL_Pv7xDp66Ra6uLOFH0Fm2K_0eHqu7ZX3bHgdF_QNu4ee1
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1NaxsxEBVtWkovTfoR4iZN55CWFrpkvR9a6WjsuE3qmBBS17dFO5JgoazDxm4v-fEZybLrXAqhl4UFIZbVSPNm9OYNY0dSaK6sVhHyWEUZpnEkNNKD264RQiaVV-CbjIrxWEyn8iL0OXW1MEt9iHXCze0Mf167DW6utQ1qA04kk8L27GwQZ-4m4TF7khEUd5yun6fjdZKF4FCa-U5oSZfLSMhiGsjvNMXx5gT33dJfrLmJWL3LGW7_98fusBcBbUJvaR4v2SPTvGKdcwLKs9bn0-Ej9H_VhFr922t2O1gXUYFqwDH0QnkVTCiqDplD6C3mM6eAqU0Lnya9k88wXJG8gFAw9BAXToMCLj3RNtQ3NTCzMHLOE_p1i6FxGNQNXKn6j2resB_Dk6v-tyg0aIjQEeMiJfLC5KlFXlRCcYlWFhpzCjqEJc8Ya4lVZZTlmRN514UhZ1ilSBghkRatSHfZVjNrzB6DFNNCFoaMI1cUEWkVq6qrYoGpFDly3mEfVutTXi91OEp_f57IcvPfdtjBavHKsBtvSjrHCAU5S-iwL36Z_jlHefb1cpC7k-ntw4a_Z88uBsNydDr-vs-euzGOVJaIA7Y1bxfmHXuKv-f1TXvorfQOfznmCw
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Developing+an+Explainable+Variational+Autoencoder+%28VAE%29+Framework+for+Accurate+Representation+of+Local+Circulation+in+Taiwan&rft.jtitle=Journal+of+geophysical+research.+Atmospheres&rft.au=Hsieh%2C+Min%E2%80%90Ken&rft.au=Wu%2C+Chien%E2%80%90Ming&rft.date=2024-06-28&rft.issn=2169-897X&rft.eissn=2169-8996&rft.volume=129&rft.issue=12&rft.epage=n%2Fa&rft_id=info:doi/10.1029%2F2024JD041167&rft.externalDBID=10.1029%252F2024JD041167&rft.externalDocID=JGRD59638
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2169-897X&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2169-897X&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2169-897X&client=summon