Non‐Linear Dimensionality Reduction With a Variational Encoder Decoder to Understand Convective Processes in Climate Models

Deep learning can accurately represent sub‐grid‐scale convective processes in climate models, learning from high resolution simulations. However, deep learning methods usually lack interpretability due to large internal dimensionality, resulting in reduced trustworthiness in these methods. Here, we...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Journal of advances in modeling earth systems Ročník 14; číslo 8; s. e2022MS003130 - n/a
Hlavní autoři:	Behrens, Gunnar, Beucler, Tom, Gentine, Pierre, Iglesias‐Suarez, Fernando, Pritchard, Michael, Eyring, Veronika
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	United States John Wiley & Sons, Inc 01.08.2022 American Geophysical Union (AGU)
Témata:	Atmosphere Climate Climate models Clouds Convection Cumulus clouds Deep learning Diabatic heating dimensionality reduction Dimensions ENVIRONMENTAL SCIENCES explainable artificial intelligence generative deep learning Long wave radiation Machine learning Modelling Moisture content parameterization Precipitation Radiation Simulation Topography convection explainable artificial intelligence parameterization dimensionality reduction generative deep learning machine learning
ISSN:	1942-2466, 1942-2466
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	Deep learning can accurately represent sub‐grid‐scale convective processes in climate models, learning from high resolution simulations. However, deep learning methods usually lack interpretability due to large internal dimensionality, resulting in reduced trustworthiness in these methods. Here, we use Variational Encoder Decoder structures (VED), a non‐linear dimensionality reduction technique, to learn and understand convective processes in an aquaplanet superparameterized climate model simulation, where deep convective processes are simulated explicitly. We show that similar to previous deep learning studies based on feed‐forward neural nets, the VED is capable of learning and accurately reproducing convective processes. In contrast to past work, we show this can be achieved by compressing the original information into only five latent nodes. As a result, the VED can be used to understand convective processes and delineate modes of convection through the exploration of its latent dimensions. A close investigation of the latent space enables the identification of different convective regimes: (a) stable conditions are clearly distinguished from deep convection with low outgoing longwave radiation and strong precipitation; (b) high optically thin cirrus‐like clouds are separated from low optically thick cumulus clouds; and (c) shallow convective processes are associated with large‐scale moisture content and surface diabatic heating. Our results demonstrate that VEDs can accurately represent convective processes in climate models, while enabling interpretability and better understanding of sub‐grid‐scale physical processes, paving the way to increasingly interpretable machine learning parameterizations with promising generative properties. Plain Language Summary Deep neural nets are hard to interpret due to their hundred thousand or million trainable parameters without further postprocessing. We demonstrate in this paper the usefulness of a network type that is designed to drastically reduce this high dimensional information in a lower‐dimensional space to enhance the interpretability of predictions compared to regular deep neural nets. Our approach is, on the one hand, able to reproduce small‐scale cloud related processes in the atmosphere learned from a physical model that simulates these processes skillfully. On the other hand, our network allows us to identify key features of different cloud types in the lower‐dimensional space. Additionally, the lower‐order manifold separates tropical samples from polar ones with a remarkable skill. Overall, our approach has the potential to boost our understanding of various complex processes in Earth System science. Key Points A Variational Encoder Decoder (VED) can predict sub‐grid‐scale thermodynamics from the coarse‐scale climate state The VED's latent space can distinguish convective regimes, including shallow/deep/no convection The VED's latent space reveals the main sources of convective predictability at different latitudes
AbstractList	Deep learning can accurately represent sub‐grid‐scale convective processes in climate models, learning from high resolution simulations. However, deep learning methods usually lack interpretability due to large internal dimensionality, resulting in reduced trustworthiness in these methods. Here, we use Variational Encoder Decoder structures (VED), a non‐linear dimensionality reduction technique, to learn and understand convective processes in an aquaplanet superparameterized climate model simulation, where deep convective processes are simulated explicitly. We show that similar to previous deep learning studies based on feed‐forward neural nets, the VED is capable of learning and accurately reproducing convective processes. In contrast to past work, we show this can be achieved by compressing the original information into only five latent nodes. As a result, the VED can be used to understand convective processes and delineate modes of convection through the exploration of its latent dimensions. A close investigation of the latent space enables the identification of different convective regimes: (a) stable conditions are clearly distinguished from deep convection with low outgoing longwave radiation and strong precipitation; (b) high optically thin cirrus‐like clouds are separated from low optically thick cumulus clouds; and (c) shallow convective processes are associated with large‐scale moisture content and surface diabatic heating. Our results demonstrate that VEDs can accurately represent convective processes in climate models, while enabling interpretability and better understanding of sub‐grid‐scale physical processes, paving the way to increasingly interpretable machine learning parameterizations with promising generative properties. Deep learning can accurately represent sub‐grid‐scale convective processes in climate models, learning from high resolution simulations. However, deep learning methods usually lack interpretability due to large internal dimensionality, resulting in reduced trustworthiness in these methods. Here, we use Variational Encoder Decoder structures (VED), a non‐linear dimensionality reduction technique, to learn and understand convective processes in an aquaplanet superparameterized climate model simulation, where deep convective processes are simulated explicitly. We show that similar to previous deep learning studies based on feed‐forward neural nets, the VED is capable of learning and accurately reproducing convective processes. In contrast to past work, we show this can be achieved by compressing the original information into only five latent nodes. As a result, the VED can be used to understand convective processes and delineate modes of convection through the exploration of its latent dimensions. A close investigation of the latent space enables the identification of different convective regimes: (a) stable conditions are clearly distinguished from deep convection with low outgoing longwave radiation and strong precipitation; (b) high optically thin cirrus‐like clouds are separated from low optically thick cumulus clouds; and (c) shallow convective processes are associated with large‐scale moisture content and surface diabatic heating. Our results demonstrate that VEDs can accurately represent convective processes in climate models, while enabling interpretability and better understanding of sub‐grid‐scale physical processes, paving the way to increasingly interpretable machine learning parameterizations with promising generative properties. Plain Language Summary Deep neural nets are hard to interpret due to their hundred thousand or million trainable parameters without further postprocessing. We demonstrate in this paper the usefulness of a network type that is designed to drastically reduce this high dimensional information in a lower‐dimensional space to enhance the interpretability of predictions compared to regular deep neural nets. Our approach is, on the one hand, able to reproduce small‐scale cloud related processes in the atmosphere learned from a physical model that simulates these processes skillfully. On the other hand, our network allows us to identify key features of different cloud types in the lower‐dimensional space. Additionally, the lower‐order manifold separates tropical samples from polar ones with a remarkable skill. Overall, our approach has the potential to boost our understanding of various complex processes in Earth System science. Key Points A Variational Encoder Decoder (VED) can predict sub‐grid‐scale thermodynamics from the coarse‐scale climate state The VED's latent space can distinguish convective regimes, including shallow/deep/no convection The VED's latent space reveals the main sources of convective predictability at different latitudes Abstract Deep learning can accurately represent sub‐grid‐scale convective processes in climate models, learning from high resolution simulations. However, deep learning methods usually lack interpretability due to large internal dimensionality, resulting in reduced trustworthiness in these methods. Here, we use Variational Encoder Decoder structures (VED), a non‐linear dimensionality reduction technique, to learn and understand convective processes in an aquaplanet superparameterized climate model simulation, where deep convective processes are simulated explicitly. We show that similar to previous deep learning studies based on feed‐forward neural nets, the VED is capable of learning and accurately reproducing convective processes. In contrast to past work, we show this can be achieved by compressing the original information into only five latent nodes. As a result, the VED can be used to understand convective processes and delineate modes of convection through the exploration of its latent dimensions. A close investigation of the latent space enables the identification of different convective regimes: (a) stable conditions are clearly distinguished from deep convection with low outgoing longwave radiation and strong precipitation; (b) high optically thin cirrus‐like clouds are separated from low optically thick cumulus clouds; and (c) shallow convective processes are associated with large‐scale moisture content and surface diabatic heating. Our results demonstrate that VEDs can accurately represent convective processes in climate models, while enabling interpretability and better understanding of sub‐grid‐scale physical processes, paving the way to increasingly interpretable machine learning parameterizations with promising generative properties. Deep learning can accurately represent sub‐grid‐scale convective processes in climate models, learning from high resolution simulations. However, deep learning methods usually lack interpretability due to large internal dimensionality, resulting in reduced trustworthiness in these methods. Here, we use Variational Encoder Decoder structures (VED), a non‐linear dimensionality reduction technique, to learn and understand convective processes in an aquaplanet superparameterized climate model simulation, where deep convective processes are simulated explicitly. We show that similar to previous deep learning studies based on feed‐forward neural nets, the VED is capable of learning and accurately reproducing convective processes. In contrast to past work, we show this can be achieved by compressing the original information into only five latent nodes. As a result, the VED can be used to understand convective processes and delineate modes of convection through the exploration of its latent dimensions. A close investigation of the latent space enables the identification of different convective regimes: (a) stable conditions are clearly distinguished from deep convection with low outgoing longwave radiation and strong precipitation; (b) high optically thin cirrus‐like clouds are separated from low optically thick cumulus clouds; and (c) shallow convective processes are associated with large‐scale moisture content and surface diabatic heating. Our results demonstrate that VEDs can accurately represent convective processes in climate models, while enabling interpretability and better understanding of sub‐grid‐scale physical processes, paving the way to increasingly interpretable machine learning parameterizations with promising generative properties. Deep neural nets are hard to interpret due to their hundred thousand or million trainable parameters without further postprocessing. We demonstrate in this paper the usefulness of a network type that is designed to drastically reduce this high dimensional information in a lower‐dimensional space to enhance the interpretability of predictions compared to regular deep neural nets. Our approach is, on the one hand, able to reproduce small‐scale cloud related processes in the atmosphere learned from a physical model that simulates these processes skillfully. On the other hand, our network allows us to identify key features of different cloud types in the lower‐dimensional space. Additionally, the lower‐order manifold separates tropical samples from polar ones with a remarkable skill. Overall, our approach has the potential to boost our understanding of various complex processes in Earth System science. A Variational Encoder Decoder (VED) can predict sub‐grid‐scale thermodynamics from the coarse‐scale climate state The VED's latent space can distinguish convective regimes, including shallow/deep/no convection The VED's latent space reveals the main sources of convective predictability at different latitudes Deep learning can accurately represent sub-grid-scale convective processes in climate models, learning from high resolution simulations. However, deep learning methods usually lack interpretability due to large internal dimensionality, resulting in reduced trustworthiness in these methods. Here, we use Variational Encoder Decoder structures (VED), a non-linear dimensionality reduction technique, to learn and understand convective processes in an aquaplanet superparameterized climate model simulation, where deep convective processes are simulated explicitly. We show that similar to previous deep learning studies based on feed-forward neural nets, the VED is capable of learning and accurately reproducing convective processes. In contrast to past work, we show this can be achieved by compressing the original information into only five latent nodes. As a result, the VED can be used to understand convective processes and delineate modes of convection through the exploration of its latent dimensions. A close investigation of the latent space enables the identification of different convective regimes: (a) stable conditions are clearly distinguished from deep convection with low outgoing longwave radiation and strong precipitation; (b) high optically thin cirrus-like clouds are separated from low optically thick cumulus clouds; and (c) shallow convective processes are associated with large-scale moisture content and surface diabatic heating. Our results demonstrate that VEDs can accurately represent convective processes in climate models, while enabling interpretability and better understanding of sub-grid-scale physical processes, paving the way to increasingly interpretable machine learning parameterizations with promising generative properties.Deep learning can accurately represent sub-grid-scale convective processes in climate models, learning from high resolution simulations. However, deep learning methods usually lack interpretability due to large internal dimensionality, resulting in reduced trustworthiness in these methods. Here, we use Variational Encoder Decoder structures (VED), a non-linear dimensionality reduction technique, to learn and understand convective processes in an aquaplanet superparameterized climate model simulation, where deep convective processes are simulated explicitly. We show that similar to previous deep learning studies based on feed-forward neural nets, the VED is capable of learning and accurately reproducing convective processes. In contrast to past work, we show this can be achieved by compressing the original information into only five latent nodes. As a result, the VED can be used to understand convective processes and delineate modes of convection through the exploration of its latent dimensions. A close investigation of the latent space enables the identification of different convective regimes: (a) stable conditions are clearly distinguished from deep convection with low outgoing longwave radiation and strong precipitation; (b) high optically thin cirrus-like clouds are separated from low optically thick cumulus clouds; and (c) shallow convective processes are associated with large-scale moisture content and surface diabatic heating. Our results demonstrate that VEDs can accurately represent convective processes in climate models, while enabling interpretability and better understanding of sub-grid-scale physical processes, paving the way to increasingly interpretable machine learning parameterizations with promising generative properties.
Author	Beucler, Tom Gentine, Pierre Pritchard, Michael Iglesias‐Suarez, Fernando Eyring, Veronika Behrens, Gunnar
Author_xml	– sequence: 1 givenname: Gunnar orcidid: 0000-0002-5921-5327 surname: Behrens fullname: Behrens, Gunnar email: Gunnar.Behrens@dlr.de organization: Columbia University – sequence: 2 givenname: Tom orcidid: 0000-0002-5731-1040 surname: Beucler fullname: Beucler, Tom organization: University of Lausanne – sequence: 3 givenname: Pierre orcidid: 0000-0002-0845-8345 surname: Gentine fullname: Gentine, Pierre organization: Columbia University – sequence: 4 givenname: Fernando orcidid: 0000-0003-3403-8245 surname: Iglesias‐Suarez fullname: Iglesias‐Suarez, Fernando organization: Institut für Physik der Atmosphäre – sequence: 5 givenname: Michael orcidid: 0000-0002-0340-6327 surname: Pritchard fullname: Pritchard, Michael organization: University of California Irvine – sequence: 6 givenname: Veronika orcidid: 0000-0002-6887-4885 surname: Eyring fullname: Eyring, Veronika organization: Institute of Environmental Physics (IUP)
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/36245669$$D View this record in MEDLINE/PubMed https://www.osti.gov/biblio/1881374$$D View this record in Osti.gov
BookMark	eNp9kU1vFDEMhiNURD_gxhlFcOHQLfmYZDLHatkC1S4goHCMMhmPmmo2KUmm1R6Q-hP4jfwS0k4rVZXgZFt-Xst-vYu2fPCA0HNKDihhzRtGGFt9JYRTTh6hHdpUbMYqKbfu5dtoN6UzQqSUTDxB21yySkjZ7KBfH4P_c_V76TyYiN-6NfjkgjeDyxv8BbrR5lLiHy6fYoO_m-hMvunjhbehg6KBKeaAT3xJUja-w_PgL6BoLwB_jsFCSpCw83g-uLXJgFdFM6Sn6HFvhgTPbuMeOjlafJu_ny0_vfswP1zObMUUmUHNGqpEyxvb075hyghCoOVV21nV963oDLAWKIUGhBS16lrDOBDoueW1AL6HXk5zQ8pOJ-sy2FMbvC8raqoU5XVVoNcTdB7DzxFS1muXLAyD8RDGpFnNRMWVYk1BXz1Az8IYiyvXFKmFqiRRhXpxS43tGjp9HsvtcaPv3C8AmwAbQ0oRel02u_E3R-MGTYm-frG-_-Ii2n8gupv7D5xP-KUbYPNfVh8frhaMSkn4X2wfthQ
CitedBy_id	crossref_primary_10_1017_eds_2025_17 crossref_primary_10_1029_2024MS004272 crossref_primary_10_1007_s13253_023_00539_0 crossref_primary_10_1029_2024MS004398 crossref_primary_10_1038_s41370_024_00712_8 crossref_primary_10_1029_2024MS004401 crossref_primary_10_1038_s41561_024_01527_w crossref_primary_10_3390_app14198884 crossref_primary_10_1016_j_buildenv_2024_112325 crossref_primary_10_1017_eds_2025_10010 crossref_primary_10_1029_2024JD041167 crossref_primary_10_1007_s13143_023_00330_8 crossref_primary_10_5194_amt_16_911_2023 crossref_primary_10_1029_2024JD041954
Cites_doi	10.3894/james.2009.1.12 10.1029/2020ms002076 10.1175/jas3453.1 10.1029/2019JD032321 10.1145/3429309.3429324 10.1175/1520-0469(2001)058<0978:ccpwtl>2.0.co;2 10.1016/j.dynatmoce.2015.05.003 10.1017/CBO9781139087513 10.1073/pnas.1810286115 10.1109/CVPR.2019.01269 10.1029/2001GL013552 10.1038/nclimate3190 10.1038/ngeo2398 10.1175/1520-0469(2000)057<1741:aqetcm>2.0.co;2 10.1175/BAMS-84-11-1547 10.2151/jmsj.2020-021 10.1002/2014ms000340 10.1175/jas-d-13-0119.1 10.1175/JAS-D-11-0148.1 10.1175/JCLI-D-17-0590.1 10.1256/qj.03.130 10.1007/s00382-013-1678-z 10.1002/2013GL057998 10.1029/2018GL078202 10.1175/JCLI3762.1 10.1093/oso/9780195066302.001.0001 10.1029/2020ms002385 10.1175/jas3677.1 10.1029/2004RG000158 10.1175/1520-0469(2003)060<0607:crmota>2.0.co;2 10.1038/s41467-020-17142-3 10.1175/JCLI-D-11-00168.1
ContentType	Journal Article
Copyright	2022 The Authors. Journal of Advances in Modeling Earth Systems published by Wiley Periodicals LLC on behalf of American Geophysical Union. 2022. This work is published under http://creativecommons.org/licenses/by-nc/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Copyright_xml	– notice: 2022 The Authors. Journal of Advances in Modeling Earth Systems published by Wiley Periodicals LLC on behalf of American Geophysical Union. – notice: 2022. This work is published under http://creativecommons.org/licenses/by-nc/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
CorporateAuthor	Columbia Univ., New York, NY (United States)
CorporateAuthor_xml	– name: Columbia Univ., New York, NY (United States)
DBID	24P AAYXX CITATION NPM 7TG ABUWG AEUYN AFKRA AZQEC BENPR BHPHI BKSAR CCPQU DWQXO F1W H96 HCIFZ KL. L.G PCBAR PHGZM PHGZT PIMPY PKEHL PQEST PQQKQ PQUKI PRINS 7X8 OTOTI
DOI	10.1029/2022MS003130
DatabaseName	Open Access: Wiley-Blackwell Open Access Journals CrossRef PubMed Meteorological & Geoastrophysical Abstracts ProQuest Central (Alumni) ProQuest One Sustainability ProQuest Central UK/Ireland ProQuest Central Essentials AUTh Library subscriptions: ProQuest Central Natural Science Collection Earth, Atmospheric & Aquatic Science Collection ProQuest One Community College ProQuest Central Korea ASFA: Aquatic Sciences and Fisheries Abstracts Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources SciTech Premium Collection Meteorological & Geoastrophysical Abstracts - Academic Aquatic Science & Fisheries Abstracts (ASFA) Professional Earth, Atmospheric & Aquatic Science Database ProQuest Central Premium ProQuest One Academic Publicly Available Content Database ProQuest One Academic Middle East (New) ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic (retired) ProQuest One Academic UKI Edition ProQuest Central China MEDLINE - Academic OSTI.GOV
DatabaseTitle	CrossRef PubMed Publicly Available Content Database Aquatic Science & Fisheries Abstracts (ASFA) Professional ProQuest One Academic Middle East (New) ProQuest Central Essentials ProQuest One Academic Eastern Edition Earth, Atmospheric & Aquatic Science Database ProQuest Central (Alumni Edition) SciTech Premium Collection ProQuest One Community College ProQuest Central China Earth, Atmospheric & Aquatic Science Collection ProQuest Central ProQuest One Sustainability Meteorological & Geoastrophysical Abstracts Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources ProQuest One Academic UKI Edition Natural Science Collection ProQuest Central Korea ASFA: Aquatic Sciences and Fisheries Abstracts ProQuest Central (New) ProQuest One Academic Meteorological & Geoastrophysical Abstracts - Academic ProQuest One Academic (New) MEDLINE - Academic
DatabaseTitleList	Publicly Available Content Database CrossRef MEDLINE - Academic PubMed
Database_xml	– sequence: 1 dbid: 24P name: Wiley Online Library Open Access url: https://authorservices.wiley.com/open-science/open-access/browse-journals.html sourceTypes: Publisher – sequence: 2 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 3 dbid: PIMPY name: Publicly Available Content Database url: http://search.proquest.com/publiccontent sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Meteorology & Climatology Environmental Sciences
EISSN	1942-2466
EndPage	n/a
ExternalDocumentID	1881374 36245669 10_1029_2022MS003130 JAME21660
Genre	article Journal Article
GrantInformation_xml	– fundername: EC ERC HORIZON EUROPE European Research Council funderid: 855187 – fundername: Advanced Research Projects Agency - Energy funderid: DE‐SC0022331 – fundername: National Science Foundation Science and Technology Center Learning the Earth with Artificial intelligence and Physics funderid: 2019625 – fundername: Columbia sub‐award 1 funderid: PG010560‐01 – fundername: Deutsches Klimarechenzentrum funderid: 1179
GroupedDBID	0R~ 1OC 24P 29J 31~ 5VS 8-1 8FE 8FH AAHHS AAZKR ABDBF ACCFJ ACCMX ACUHS ACXQS ADBBV ADKYN ADZMN ADZOD AEEZP AEGXH AENEX AEQDE AEUYN AFKRA AIWBW AJBDE ALAGY ALMA_UNASSIGNED_HOLDINGS ALUQN AVUZU AZFZN BCNDV BENPR BHPHI BKSAR CCPQU D1K EAD EAP EAS EBS EJD EPL ESX GODZA GROUPED_DOAJ HCIFZ IAO IGS IPNFZ ITC K6- KQ8 LK5 M7R M~E O9- OK1 P2P PCBAR PIMPY PROAC RIG RNS TUS WIN ~OA AAMMB AAYXX AEFGJ AFFHD AGXDD AIDQK AIDYY CITATION HZ~ PHGZM PHGZT NPM 7TG ABUWG AZQEC DWQXO F1W H96 KL. L.G PKEHL PQEST PQQKQ PQUKI PRINS 7X8 PUEGO ABPTK AEUQT OTOTI
ID	FETCH-LOGICAL-c4280-e729185b39cf1f928a500eb34bdc8ffb5dae2be11e9e56578dba23e0ef3c375e3
IEDL.DBID	BENPR
ISICitedReferencesCount	18
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000842506900001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1942-2466
IngestDate	Mon Aug 28 20:02:17 EDT 2023 Thu Sep 04 17:36:19 EDT 2025 Mon Oct 06 17:29:51 EDT 2025 Mon Jul 21 05:57:47 EDT 2025 Tue Nov 18 22:36:14 EST 2025 Sat Nov 29 02:09:16 EST 2025 Wed Jan 22 16:23:21 EST 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	8
Keywords	convection explainable artificial intelligence parameterization dimensionality reduction generative deep learning machine learning
Language	English
License	Attribution-NonCommercial 2022 The Authors. Journal of Advances in Modeling Earth Systems published by Wiley Periodicals LLC on behalf of American Geophysical Union.
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c4280-e729185b39cf1f928a500eb34bdc8ffb5dae2be11e9e56578dba23e0ef3c375e3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 National Science Foundation Science and Technology Center USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) Columbia University European Research Council (ERC) SC0022331; 855187; PG010560-01; 2019625
ORCID	0000-0002-0845-8345 0000-0002-5731-1040 0000-0002-0340-6327 0000-0002-5921-5327 0000-0003-3403-8245 0000-0002-6887-4885 0000000268874885 0000000203406327 0000000259215327 0000000208458345 0000000257311040 0000000334038245
OpenAccessLink	https://www.proquest.com/docview/2707584608?pq-origsite=%requestingapplication%
PMID	36245669
PQID	2707584608
PQPubID	616667
PageCount	0
ParticipantIDs	osti_scitechconnect_1881374 proquest_miscellaneous_2725438829 proquest_journals_2707584608 pubmed_primary_36245669 crossref_citationtrail_10_1029_2022MS003130 crossref_primary_10_1029_2022MS003130 wiley_primary_10_1029_2022MS003130_JAME21660
PublicationCentury	2000
PublicationDate	August 2022
PublicationDateYYYYMMDD	2022-08-01
PublicationDate_xml	– month: 08 year: 2022 text: August 2022
PublicationDecade	2020
PublicationPlace	United States
PublicationPlace_xml	– name: United States – name: Washington
PublicationTitle	Journal of advances in modeling earth systems
PublicationTitleAlternate	J Adv Model Earth Syst
PublicationYear	2022
Publisher	John Wiley & Sons, Inc American Geophysical Union (AGU)
Publisher_xml	– name: John Wiley & Sons, Inc – name: American Geophysical Union (AGU)
References	2017; 7 2015; 71 2013; 40 2013; 41 2005; 62 2006; 19 2005; 43 1994 2020; 12 2001; 28 2020; 11 2020; 125 2020; 98 2018; 45 2015; 8 2021; 13 2019; 2426 2006; 63 2022 2021 2020 2000; 57 2004; 130 2018; 115 2019 2018 2016 1996; 1 2014 2012; 25 2012; 69 2003; 60 2009; 2 2003; 84 2001; 58 2018; 31 2014; 71 2014; 6 e_1_2_8_28_1 e_1_2_8_29_1 e_1_2_8_24_1 Alemi A. (e_1_2_8_3_1) 2018 e_1_2_8_27_1 Mamalakis A. (e_1_2_8_26_1) 2021 e_1_2_8_5_1 e_1_2_8_4_1 e_1_2_8_7_1 e_1_2_8_6_1 e_1_2_8_8_1 Krinitskiy M. A. (e_1_2_8_23_1) 2019; 2426 e_1_2_8_20_1 e_1_2_8_42_1 Emanuel K. (e_1_2_8_9_1) 1994 e_1_2_8_41_1 e_1_2_8_17_1 e_1_2_8_18_1 e_1_2_8_19_1 Alberdi X.‐A. T. (e_1_2_8_2_1) 2018 e_1_2_8_13_1 e_1_2_8_36_1 e_1_2_8_14_1 e_1_2_8_35_1 e_1_2_8_15_1 e_1_2_8_38_1 e_1_2_8_16_1 e_1_2_8_37_1 Wang P. (e_1_2_8_39_1) 2022 Kingma D. P. (e_1_2_8_22_1) 2014 Wang X. (e_1_2_8_40_1) 2021 Kingma D. P. (e_1_2_8_21_1) 2014 Lorenz E. N. (e_1_2_8_25_1) 1996; 1 e_1_2_8_32_1 e_1_2_8_10_1 e_1_2_8_31_1 e_1_2_8_11_1 e_1_2_8_34_1 e_1_2_8_12_1 e_1_2_8_33_1 e_1_2_8_30_1
References_xml	– volume: 58 start-page: 978 issue: 9 year: 2001 end-page: 997 article-title: Coupling cloud processes with the large‐scale dynamics using the clouds‐resolving convection parameterization (CRCP) publication-title: Journal of the Atmospheric Sciences – year: 2020 article-title: Generative modeling of atmospheric convection publication-title: arXiv – volume: 71 start-page: 800 issue: 2 year: 2014 end-page: 815 article-title: Causal evidence that rotational moisture advection is critical to the superparameterized madden–julian oscillation publication-title: Journal of the Atmospheric Sciences – volume: 8 start-page: 261 issue: 4 year: 2015 end-page: 268 article-title: Clouds, circulation and climate sensitivity publication-title: Nature Geoscience – volume: 43 issue: 2 year: 2005 article-title: Madden‐julian oscillation publication-title: Reviews of Geophysics – volume: 125 start-page: 1 issue: 21 year: 2020 end-page: 28 article-title: Quantifying progress across different CMIP phases with the ESMValTool publication-title: Journal of Geophysical Research: Atmospheres – volume: 115 start-page: 9684 issue: 39 year: 2018 end-page: 9689 article-title: Deep learning to represent subgrid processes in climate models publication-title: Proceedings of the National Academy of Sciences of the United States of America – start-page: 159 year: 2018 end-page: 168 – volume: 13 issue: 5 year: 2021 article-title: Assessing the potential of deep learning for emulating cloud superparameterization in climate models with real‐geography boundary conditions publication-title: Journal of Advances in Modeling Earth Systems – volume: 11 start-page: 1 issue: 1 year: 2020 end-page: 10 article-title: Stable machine‐learning parameterization of subgrid processes for climate modeling at a range of resolutions publication-title: Nature Communications – year: 2021 article-title: Neural network attribution methods for problems in geoscience: A novel synthetic benchmark dataset publication-title: arXiv preprint arXiv:2103.10005 – volume: 41 start-page: 1527 issue: 5–6 year: 2013 end-page: 1551 article-title: Stochastic and deterministic multicloud parameterizations for tropical convection publication-title: Climate Dynamics – volume: 2426 start-page: 52 year: 2019 end-page: 61 article-title: Clustering of polar vortex states using convolutional autoencoders publication-title: CEUR Workshop Proceedings – start-page: 1 year: 2021 end-page: 35 article-title: Stable climate simulations using a realistic gcm with neural network parameterizations for atmospheric moist physics and radiation processes publication-title: Geoscientific Model Development Discussions – year: 2016 – volume: 57 start-page: 1741 issue: 11 year: 2000 end-page: 1766 article-title: A quasi‐equilibrium tropical circulation model—Formulation publication-title: Journal of the Atmospheric Sciences – year: 2018 – year: 1994 – year: 2014 – volume: 63 start-page: 1308 issue: 4 year: 2006 end-page: 1323 article-title: A simple multicloud parameterization for convectively coupled tropical waves. part i: Linear analysis publication-title: Journal of the Atmospheric Sciences – volume: 12 issue: 9 year: 2020 article-title: A moist physics parameterization based on deep learning publication-title: Journal of Advances in Modeling Earth Systems – volume: 6 start-page: 723 issue: 3 year: 2014 end-page: 739 article-title: Restricting 32–128 km horizontal scales hardly affects the mjo in the superparameterized community atmosphere model v. 3.0 but the number of cloud‐resolving grid columns constrains vertical mixing publication-title: Journal of Advances in Modeling Earth Systems – volume: 2 year: 2009 article-title: Assessing the diurnal cycle of precipitation in a multi‐scale climate model publication-title: Journal of Advances in Modeling Earth Systems – volume: 71 start-page: 35 year: 2015 end-page: 55 article-title: Cloud‐radiation feedback and atmosphere‐ocean coupling in a stochastic multicloud model publication-title: Dynamics of Atmospheres and Oceans – year: 2014 article-title: Adam: A method for stochastic optimization publication-title: arXiv preprint arXiv:1412.6980 – volume: 1 year: 1996 article-title: Predictability: A problem partly solved publication-title: Proc. seminar on predictability – volume: 45 start-page: 5742 issue: 11 year: 2018 end-page: 5751 article-title: Could machine learning break the convection parameterization deadlock? publication-title: Geophysical Research Letters – start-page: 12406 year: 2019 end-page: 12415 – volume: 7 start-page: 3 issue: 1 year: 2017 end-page: 5 article-title: Climate goals and computing the future of clouds publication-title: Nature Climate Change – volume: 40 start-page: 5970 issue: 22 year: 2013 end-page: 5976 article-title: Beyond deadlock publication-title: Geophysical Research Letters – volume: 28 start-page: 3617 issue: 18 year: 2001 end-page: 3620 article-title: A cloud resolving model as a cloud parameterization in the NCAR community climate system model: Preliminary results publication-title: Geophysical Research Letters – volume: 62 start-page: 2136 issue: 7 year: 2005 end-page: 2154 article-title: Simulations of the atmospheric general circulation using a cloud‐resolving model as a superparameterization of physical processes publication-title: Journal of the Atmospheric Sciences – volume: 25 start-page: 2782 issue: 8 year: 2012 end-page: 2804 article-title: Moist static energy budget of MJO‐like disturbances in the atmosphere of a zonally symmetric aquaplanet publication-title: Journal of Climate – volume: 84 start-page: 1547 issue: 11 year: 2003 end-page: 1564 article-title: Breaking the cloud parameterization deadlock publication-title: Bulletin of the American Meteorological Society – volume: 31 start-page: 2563 issue: 7 year: 2018 end-page: 2577 article-title: Assessing the vertical latent heating structure of the east Pacific itcz using the cloudsat cpr and trmm pr publication-title: Journal of Climate – volume: 60 start-page: 607 issue: 4 year: 2003 end-page: 625 article-title: Cloud resolving modeling of the arm summer 1997 iop: Model formulation, results, uncertainties, and sensitivities publication-title: Journal of the Atmospheric Sciences – volume: 130 start-page: 3055 issue: 604 year: 2004 end-page: 3079 article-title: Sensitivity of moist convection to environmental humidity publication-title: Quarterly Journal of the Royal Meteorological Society – year: 2022 article-title: Non‐local parameterization of atmospheric subgrid processes with neural networks publication-title: arXiv preprint arXiv:2201.00417 – volume: 98 start-page: 395 issue: 2 year: 2020 end-page: 435 article-title: The added value of large‐eddy and storm‐resolving models for simulating clouds and precipitation publication-title: Journal of the Meteorological Society of Japan – volume: 69 start-page: 1080 issue: 3 year: 2012 end-page: 1105 article-title: Using the stochastic multicloud model to improve tropical convective parameterization: A paradigm example publication-title: Journal of the Atmospheric Sciences – volume: 19 start-page: 2162 issue: 11 year: 2006 end-page: 2183 article-title: The dynamical simulation of the Community Atmosphere Model version 3 (CAM3) publication-title: Journal of Climate – ident: e_1_2_8_32_1 doi: 10.3894/james.2009.1.12 – ident: e_1_2_8_15_1 doi: 10.1029/2020ms002076 – ident: e_1_2_8_17_1 doi: 10.1175/jas3453.1 – ident: e_1_2_8_5_1 doi: 10.1029/2019JD032321 – ident: e_1_2_8_28_1 doi: 10.1145/3429309.3429324 – ident: e_1_2_8_14_1 doi: 10.1175/1520-0469(2001)058<0978:ccpwtl>2.0.co;2 – year: 2014 ident: e_1_2_8_21_1 article-title: Adam: A method for stochastic optimization publication-title: arXiv preprint arXiv:1412.6980 – ident: e_1_2_8_12_1 doi: 10.1016/j.dynatmoce.2015.05.003 – ident: e_1_2_8_24_1 doi: 10.1017/CBO9781139087513 – ident: e_1_2_8_35_1 doi: 10.1073/pnas.1810286115 – ident: e_1_2_8_36_1 doi: 10.1109/CVPR.2019.01269 – ident: e_1_2_8_18_1 doi: 10.1029/2001GL013552 – ident: e_1_2_8_37_1 doi: 10.1038/nclimate3190 – ident: e_1_2_8_6_1 doi: 10.1038/ngeo2398 – ident: e_1_2_8_29_1 doi: 10.1175/1520-0469(2000)057<1741:aqetcm>2.0.co;2 – ident: e_1_2_8_34_1 doi: 10.1175/BAMS-84-11-1547 – volume-title: 8th international workshop on climate informatics year: 2018 ident: e_1_2_8_2_1 – volume: 1 year: 1996 ident: e_1_2_8_25_1 article-title: Predictability: A problem partly solved publication-title: Proc. seminar on predictability – ident: e_1_2_8_38_1 doi: 10.2151/jmsj.2020-021 – volume-title: 2nd international conference on learning representations. ICLR 2014–Conference track proceedings (Ml) year: 2014 ident: e_1_2_8_22_1 – ident: e_1_2_8_31_1 doi: 10.1002/2014ms000340 – ident: e_1_2_8_30_1 doi: 10.1175/jas-d-13-0119.1 – start-page: 159 volume-title: International conference on machine learning year: 2018 ident: e_1_2_8_3_1 – volume: 2426 start-page: 52 year: 2019 ident: e_1_2_8_23_1 article-title: Clustering of polar vortex states using convolutional autoencoders publication-title: CEUR Workshop Proceedings – ident: e_1_2_8_10_1 doi: 10.1175/JAS-D-11-0148.1 – ident: e_1_2_8_16_1 doi: 10.1175/JCLI-D-17-0590.1 – ident: e_1_2_8_8_1 doi: 10.1256/qj.03.130 – ident: e_1_2_8_11_1 doi: 10.1007/s00382-013-1678-z – ident: e_1_2_8_33_1 doi: 10.1002/2013GL057998 – ident: e_1_2_8_13_1 doi: 10.1029/2018GL078202 – year: 2021 ident: e_1_2_8_26_1 article-title: Neural network attribution methods for problems in geoscience: A novel synthetic benchmark dataset publication-title: arXiv preprint arXiv:2103.10005 – ident: e_1_2_8_7_1 doi: 10.1175/JCLI3762.1 – volume-title: Atmospheric convection year: 1994 ident: e_1_2_8_9_1 doi: 10.1093/oso/9780195066302.001.0001 – ident: e_1_2_8_27_1 doi: 10.1029/2020ms002385 – ident: e_1_2_8_20_1 doi: 10.1175/jas3677.1 – year: 2022 ident: e_1_2_8_39_1 article-title: Non‐local parameterization of atmospheric subgrid processes with neural networks publication-title: arXiv preprint arXiv:2201.00417 – ident: e_1_2_8_42_1 doi: 10.1029/2004RG000158 – ident: e_1_2_8_19_1 doi: 10.1175/1520-0469(2003)060<0607:crmota>2.0.co;2 – ident: e_1_2_8_41_1 doi: 10.1038/s41467-020-17142-3 – start-page: 1 year: 2021 ident: e_1_2_8_40_1 article-title: Stable climate simulations using a realistic gcm with neural network parameterizations for atmospheric moist physics and radiation processes publication-title: Geoscientific Model Development Discussions – ident: e_1_2_8_4_1 doi: 10.1175/JCLI-D-11-00168.1
SSID	ssj0066625
Score	2.425156
Snippet	Deep learning can accurately represent sub‐grid‐scale convective processes in climate models, learning from high resolution simulations. However, deep learning... Deep learning can accurately represent sub-grid-scale convective processes in climate models, learning from high resolution simulations. However, deep learning... Abstract Deep learning can accurately represent sub‐grid‐scale convective processes in climate models, learning from high resolution simulations. However, deep...
SourceID	osti proquest pubmed crossref wiley
SourceType	Open Access Repository Aggregation Database Index Database Enrichment Source Publisher
StartPage	e2022MS003130
SubjectTerms	Atmosphere Climate Climate models Clouds Convection Cumulus clouds Deep learning Diabatic heating dimensionality reduction Dimensions ENVIRONMENTAL SCIENCES explainable artificial intelligence generative deep learning Long wave radiation Machine learning Modelling Moisture content parameterization Precipitation Radiation Simulation Topography
SummonAdditionalLinks	– databaseName: Open Access: Wiley-Blackwell Open Access Journals dbid: 24P link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3LatwwFBVhmkU3bdOnk7Qo0HTTmloP2_KyTCZ0kyHk1eyMJMtkYLDL2OkukE_IN_ZLeq-sMQm0hdKVMZKwsO6VzrmSziXkvRaJVRWvgeToNJbKgM_Zuoil4NrItLJWDMkm8vlcXV4WxyHghndhBn2IMeCGnuHna3RwbbogNoAamcDa-dGp1x4Eyv6IMaEwdQOXx-uZGJC5T7oKPJ3HXGZZOPgO7T_fb_1gSZq04Fq_g5sP0atffg6f_m_Hn5EnAXjSL4OlbJEN1zwn0RFg5nblQ-v0A50uFwBg_dsLcjNvm5-3d0BWwRnoAWYBGBQ8ALfTE1R8xTGl3xb9FdX0Ajh3iCvSWYMX5aGNG559S8_HOzR0iufc_SxLwy0F19FFEz7uKGZnW3Yvyfnh7Gz6NQ7JGmILDCaJHaB0WPuNKGzN6oIrTLUATF2ayqq6NmmlHTeOMVc43GpVldFcuMTVwoo8deIVmTRt494QKhOTgoFXKre51JapXNQpzDOJTVgleR6Rj-vxKm1QMseEGsvS76jzorz_iyOyP9b-Pih4_KHeDg59CcgD5XMtnjOyfcmUYiKXEdldW0QZvLwreQ6ACwBcoiKyNxaDf-Kmi25ce411UG4AeEwRkdeDJY3dAPAA-DWDkk_eYP7avxJW4RlnWZZs_1v1HfIYC4ZTi7tk0q-u3VuyaX_0i271znvMLz40FJ0 priority: 102 providerName: Wiley-Blackwell
Title	Non‐Linear Dimensionality Reduction With a Variational Encoder Decoder to Understand Convective Processes in Climate Models
URI	https://onlinelibrary.wiley.com/doi/abs/10.1029%2F2022MS003130 https://www.ncbi.nlm.nih.gov/pubmed/36245669 https://www.proquest.com/docview/2707584608 https://www.proquest.com/docview/2725438829 https://www.osti.gov/biblio/1881374
Volume	14
WOSCitedRecordID	wos000842506900001&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: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 1942-2466 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0066625 issn: 1942-2466 databaseCode: DOA dateStart: 20090101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources customDbUrl: eissn: 1942-2466 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0066625 issn: 1942-2466 databaseCode: M~E dateStart: 20090101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre – providerCode: PRVPQU databaseName: Earth, Atmospheric & Aquatic Science Database customDbUrl: eissn: 1942-2466 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0066625 issn: 1942-2466 databaseCode: PCBAR dateStart: 20090201 isFulltext: true titleUrlDefault: https://search.proquest.com/eaasdb providerName: ProQuest – providerCode: PRVPQU databaseName: ProQuest Central customDbUrl: eissn: 1942-2466 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0066625 issn: 1942-2466 databaseCode: BENPR dateStart: 20090201 isFulltext: true titleUrlDefault: https://www.proquest.com/central providerName: ProQuest – providerCode: PRVPQU databaseName: Publicly Available Content Database customDbUrl: eissn: 1942-2466 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0066625 issn: 1942-2466 databaseCode: PIMPY dateStart: 20090201 isFulltext: true titleUrlDefault: http://search.proquest.com/publiccontent providerName: ProQuest – providerCode: PRVWIB databaseName: Wiley Online Library Free Content customDbUrl: eissn: 1942-2466 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0066625 issn: 1942-2466 databaseCode: WIN dateStart: 20090101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell – providerCode: PRVWIB databaseName: Wiley Online Library Open Access customDbUrl: eissn: 1942-2466 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0066625 issn: 1942-2466 databaseCode: 24P dateStart: 20090101 isFulltext: true titleUrlDefault: https://authorservices.wiley.com/open-science/open-access/browse-journals.html providerName: Wiley-Blackwell
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3NjtMwEB6xXQ5c-P8Ju1RGAi4QrWM7jXNCULJiD42ihWXLKUocR1SqkqXpcgPxCDwjT8JM4hZWgr1wSZXGUSyNZ-ab8fgbgCeF5EZXosYgpwh9pUvUOVPHvpKiKFVYGSOHZhNRmur5PM5cwq1zZZUbm9gb6qo1lCM_EBE6N3SWXL88--xT1yjaXXUtNHZgl5jK1Ah2XydpdryxxYjNRejK3bmIKdIXs3c9XyG_4IhGLSrU30DmRczaO53DG_873Ztw3cFN9mpYH7fgim1ugzdDpNyu-oQ6e8amywXC1v7uDnxN2-bn9x8YoqIKsDfE_T_wdiBaZ8fE80qSZKeL9SdWsA8YabtsIksaOh6P79jhd92yk-3JGTal6vbetjJ3NsF2bNG4j1tGPdmW3V04OUzeT9_6rkWDbzBu4b5FbI4ev5SxqYM6FpoaLGB8rsrK6Louw6qworRBYGNLG6y6KgshLbe1NDIKrbwHo6Zt7ANgipchLutKRyZShQl0JOsQrQs3PKiUiDx4vpFXbhx_ObXRWOb9PrqI8z-l68HT7eizgbfjH-P2SPQ54g0izTVUXWTWeaB1ICPlwf5GurnT7S7_LVoPHm8fo1bSVkvR2PacxhDJAEYvsQf3h5W0nQZCBkStE3zyol9al84vR9-biGAy4Q8vn8seXKP3htrEfRitV-f2EVw1X9aLbjWGHaEyvEZzPXY6Mu7TD3idfUvwv-xoln3Eu9Oj9BdhIx0Y
linkProvider	ProQuest
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V3NbtQwEB6VLRJcKL8lbQEjUS4Q1bGTjXNACG236qrd1QpaWk4hcRwRaZW0mxTUAxKPwJPwUDwJM_lZqAS99cApiuI4I-fzzDf2eAbgWSS5VolI0cmJPNtVMc45nQa2K0UUu16itWyKTfiTiTo-DqZL8KM7C0NhlZ1OrBV1UmhaI98SPho3NJZcvT45talqFO2udiU0GljsmfMv6LKVr0bb-H83hdgZHgx27baqgK2RanPbIJ1EIxXLQKdOGghFNQHQpXTjRKs0jb0kMiI2jmMCQ3uCKokjIQ03qdTS94zEfq_BsotgVz1Yno7G0w-d7kdfQHhteD0XAa0siPG7Oj8iv2D4egVO4L-R2oscuTZyOyv_2_DchlstnWZvGvzfgSWT3wVrjJ5AMa83DNhzNphlSMvru3vwdVLkP799RxcchWXbVNugyUuC3gh7S3lsCansKKs-sYi9j-ZZu1rKhjkd_8d3THOtCna4OBnEBhS9X9sO1p69MCXL8vbjhlHNuVl5Hw6vZDQeQC8vcvMQmMtjD6dtonztu5F2lC9TD7Un19xJXOFb8KLDR6jb_OxUJmQW1nECIgj_RJMFm4vWJ01ekn-0WyeohcinKCmwpugpXYWOUo70XQs2OjSFre4qw99QsuDp4jFqHdpKinJTnFEbSqKA3llgwWqD3IUYSImQlffxycsaypfKFyK3GAqn3-drl8vyBG7sHoz3w_3RZG8dblIfTRzmBvSq-Zl5BNf15yor54_bOcng41Xj_BfNanVd
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V3NbtQwEB6VLUJc-P8JLWAkygWiOnaycQ4Iof0Rq7KrFVBaTiFxHBFplZTNFsQBiUfgeXgcnoSZxFmoBL31wCmK4jiW883MN_Z4BuBhIrlWmcjRyUkC11cpypzOI9eXIkn9INNatsUmwtlMHR5G8w340Z2FobDKTic2ijqrNK2R74oQjRsaS652cxsWMR-Onx19dKmCFO20duU0WojsmS-f0X2rn06G-K93hBiP3gxeuLbCgKuRdnPXILVEg5XKSOdeHglF9QHQvfTTTKs8T4MsMSI1nmciQ_uDKksTIQ03udQyDIzEfs_BJlJyH2Vscz6Zzt91dgD9AhHYUHsuIlplENPXTa5EfsII9ioU5r8R3JN8uTF448v_81RdgUuWZrPnrVxchQ1TXgNnih5CtWw2EtgjNlgUSNebu-vwdVaVP799R9ccB8uGVPOgzVeCXgp7RfltCcHsoFh9YAl7mywLu4rKRiWlBcB3THtdVWx_fWKIDSiqv7EpzJ7JMDUrSvtxw6gW3aK-AftnMhs3oVdWpbkNzOdpgOKcqVCHfqI9Fco8QK3KNfcyX4QOPO6wEmubt53KhyziJn5ARPGfyHJgZ936qM1X8o92WwS7GHkWJQvWFFWlV7GnlCdD34HtDlmx1Wl1_BtWDjxYP0ZtRFtMSWmqY2pDyRXQa4scuNWieD0MpErI1vv45EkD61PHFyPnGAmv3-d3Th_LfbiA4I5fTmZ7W3CRumjDM7eht1oem7twXn9aFfXynhVPBu_PGua_AFEPfh0
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=Non%E2%80%90Linear+Dimensionality+Reduction+With+a+Variational+Encoder+Decoder+to+Understand+Convective+Processes+in+Climate+Models&rft.jtitle=Journal+of+advances+in+modeling+earth+systems&rft.au=Behrens%2C+Gunnar&rft.au=Beucler%2C+Tom&rft.au=Gentine%2C+Pierre&rft.au=Iglesias%E2%80%90Suarez%2C+Fernando&rft.date=2022-08-01&rft.pub=American+Geophysical+Union+%28AGU%29&rft.issn=1942-2466&rft.eissn=1942-2466&rft.volume=14&rft.issue=8&rft_id=info:doi/10.1029%2F2022MS003130&rft.externalDocID=1881374
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1942-2466&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1942-2466&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1942-2466&client=summon