Using Machine Learning to Predict Urban Canopy Flows for Land Surface Modeling

Simple item page
dc.contributor.authorLu, Yanleen
dc.contributor.authorZhou, Xu-Huien
dc.contributor.authorXiao, Hengen
dc.contributor.authorLi, Qien
dc.date.accessioned2023-03-29T13:25:24Zen
dc.date.available2023-03-29T13:25:24Zen
dc.date.issued2023-01-16en
dc.description.abstractDeveloping urban land surface models for modeling cities at high resolutions needs to better account for the city-specific multi-scale land surface heterogeneities at a reasonable computational cost. We propose using an encoder-decoder convolutional neural network to develop a computationally efficient model for predicting the mean velocity field directly from urban geometries. The network is trained using the geometry-resolving large eddy simulation results. Systematic testing on urban structures with increasing deviations from the training geometries shows the prediction error plateaus at 15%, compared to errors sharply increasing up to 35% in the null models. This is explained by the trained model successfully capturing the effects of pressure drag, especially for tall buildings. The prediction error of the aerodynamic drag coefficient is reduced by 32% compared with the default parameterization implemented in mesoscale modeling. This study highlights the potential of combining computational fluid dynamics modeling and machine learning to develop city-specific parameterizations.en
dc.description.notesQL acknowledge support from the US National Science Foundation (NSF-CAREER-2143664, NSF-AGS-2028633, NSF-CBET-2028842) and computational resources from the National Center for Atmospheric Research (UCOR-0049).en
dc.description.sponsorshipUS National Science Foundation [NSF-CAREER-2143664, NSF-AGS-2028633, NSF-CBET-2028842]; National Center for Atmospheric Research [UCOR-0049]en
dc.description.versionPublished versionen
dc.format.mimetypeapplication/pdfen
dc.identifier.doihttps://doi.org/10.1029/2022GL102313en
dc.identifier.eissn1944-8007en
dc.identifier.issue1en
dc.identifier.urihttp://hdl.handle.net/10919/114220en
dc.identifier.volume50en
dc.language.isoenen
dc.publisherAmerican Geophysical Unionen
dc.rightsCreative Commons Attribution-NonCommercial-NoDerivatives 4.0 Internationalen
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/en
dc.subjectmachine learningen
dc.subjectlarge-eddy simulationen
dc.subjecturban canopy flowen
dc.subjecturban canopy modelen
dc.titleUsing Machine Learning to Predict Urban Canopy Flows for Land Surface Modelingen
dc.title.serialGeophysical Research Lettersen
dc.typeArticle - Refereeden
dc.type.dcmitypeTexten

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
LuUsing2023.pdf
Size:
2.03 MB
Format:
Adobe Portable Document Format
Description:
Published version
Download

Collections

Scholarly Works, Aerospace and Ocean Engineering