Machine Learning Emulation of Urban Land Surface Processes
Abstract Can we improve the modeling of urban land surface processes with machine learning (ML)? A prior comparison of urban land surface models (ULSMs) found that no single model is “best” at predicting all common surface fluxes. Here, we develop an urban neural network (UNN) trained on the mean pr...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
American Geophysical Union (AGU)
2022-03-01
|
| Series: | Journal of Advances in Modeling Earth Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2021MS002744 |
| _version_ | 1827958794495197184 |
|---|---|
| author | David Meyer Sue Grimmond Peter Dueben Robin Hogan Maarten van Reeuwijk |
| author_facet | David Meyer Sue Grimmond Peter Dueben Robin Hogan Maarten van Reeuwijk |
| author_sort | David Meyer |
| collection | DOAJ |
| description | Abstract Can we improve the modeling of urban land surface processes with machine learning (ML)? A prior comparison of urban land surface models (ULSMs) found that no single model is “best” at predicting all common surface fluxes. Here, we develop an urban neural network (UNN) trained on the mean predicted fluxes from 22 ULSMs at one site. The UNN emulates the mean output of ULSMs accurately. When compared to a reference ULSM (Town Energy Balance; TEB), the UNN has greater accuracy relative to flux observations, less computational cost, and requires fewer input parameters. When coupled to the Weather Research Forecasting (WRF) model using TensorFlow bindings, WRF‐UNN is stable and more accurate than the reference WRF‐TEB. Although the application is currently constrained by the training data (1 site), we show a novel approach to improve the modeling of surface fluxes by combining the strengths of several ULSMs into one using ML. |
| first_indexed | 2024-04-09T15:41:48Z |
| format | Article |
| id | doaj.art-4552a66260c14b4eb98316c9359af5a8 |
| institution | Directory Open Access Journal |
| issn | 1942-2466 |
| language | English |
| last_indexed | 2024-04-09T15:41:48Z |
| publishDate | 2022-03-01 |
| publisher | American Geophysical Union (AGU) |
| record_format | Article |
| series | Journal of Advances in Modeling Earth Systems |
| spelling | doaj.art-4552a66260c14b4eb98316c9359af5a82023-04-27T07:53:09ZengAmerican Geophysical Union (AGU)Journal of Advances in Modeling Earth Systems1942-24662022-03-01143n/an/a10.1029/2021MS002744Machine Learning Emulation of Urban Land Surface ProcessesDavid Meyer0Sue Grimmond1Peter Dueben2Robin Hogan3Maarten van Reeuwijk4Department of Meteorology University of Reading Reading UKDepartment of Meteorology University of Reading Reading UKEuropean Centre for Medium‐Range Weather Forecasts Reading UKDepartment of Meteorology University of Reading Reading UKDepartment of Civil and Environmental Engineering Imperial College London London UKAbstract Can we improve the modeling of urban land surface processes with machine learning (ML)? A prior comparison of urban land surface models (ULSMs) found that no single model is “best” at predicting all common surface fluxes. Here, we develop an urban neural network (UNN) trained on the mean predicted fluxes from 22 ULSMs at one site. The UNN emulates the mean output of ULSMs accurately. When compared to a reference ULSM (Town Energy Balance; TEB), the UNN has greater accuracy relative to flux observations, less computational cost, and requires fewer input parameters. When coupled to the Weather Research Forecasting (WRF) model using TensorFlow bindings, WRF‐UNN is stable and more accurate than the reference WRF‐TEB. Although the application is currently constrained by the training data (1 site), we show a novel approach to improve the modeling of surface fluxes by combining the strengths of several ULSMs into one using ML.https://doi.org/10.1029/2021MS002744machine learningneural networkWeather Research Forecasting (WRF)numerical weather prediction (NWP)couplingurban land surface |
| spellingShingle | David Meyer Sue Grimmond Peter Dueben Robin Hogan Maarten van Reeuwijk Machine Learning Emulation of Urban Land Surface Processes Journal of Advances in Modeling Earth Systems machine learning neural network Weather Research Forecasting (WRF) numerical weather prediction (NWP) coupling urban land surface |
| title | Machine Learning Emulation of Urban Land Surface Processes |
| title_full | Machine Learning Emulation of Urban Land Surface Processes |
| title_fullStr | Machine Learning Emulation of Urban Land Surface Processes |
| title_full_unstemmed | Machine Learning Emulation of Urban Land Surface Processes |
| title_short | Machine Learning Emulation of Urban Land Surface Processes |
| title_sort | machine learning emulation of urban land surface processes |
| topic | machine learning neural network Weather Research Forecasting (WRF) numerical weather prediction (NWP) coupling urban land surface |
| url | https://doi.org/10.1029/2021MS002744 |
| work_keys_str_mv | AT davidmeyer machinelearningemulationofurbanlandsurfaceprocesses AT suegrimmond machinelearningemulationofurbanlandsurfaceprocesses AT peterdueben machinelearningemulationofurbanlandsurfaceprocesses AT robinhogan machinelearningemulationofurbanlandsurfaceprocesses AT maartenvanreeuwijk machinelearningemulationofurbanlandsurfaceprocesses |