TerraWind: A Deep Learning‐Based Near‐Surface Winds Downscaling Model for Complex Terrain Region

Abstract Wind downscaling is crucial for refining coarse‐scale wind estimates, improving local‐scale predictions, and supporting various applications like risk assessment and planning. Dynamic downscaling models demand extensive computational resources and time, leading to a shift toward more effici...

Full description

Saved in:
Bibliographic Details
Main Authors: Jie Lian, Sirong Huang, Jiahao Shao, Peiyan Chen, Shengming Tang, Yi Lu, Hui Yu
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:Geophysical Research Letters
Online Access:https://doi.org/10.1029/2024GL112124
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850126653485744128
author Jie Lian
Sirong Huang
Jiahao Shao
Peiyan Chen
Shengming Tang
Yi Lu
Hui Yu
author_facet Jie Lian
Sirong Huang
Jiahao Shao
Peiyan Chen
Shengming Tang
Yi Lu
Hui Yu
author_sort Jie Lian
collection DOAJ
description Abstract Wind downscaling is crucial for refining coarse‐scale wind estimates, improving local‐scale predictions, and supporting various applications like risk assessment and planning. Dynamic downscaling models demand extensive computational resources and time, leading to a shift toward more efficient statistical downscaling, whereas it often overlooks inter‐variable and inter‐station spatial correlations. Addressing this, we propose TerraWind, a deep learning‐based downscaling method for complex terrain regions. TerraWind enhances accuracy by incorporating topographic factors and inter‐station linkages, capturing wind field interactions with terrain at multiple scales. Experimental results in Eastern China demonstrate that TerraWind reduces wind speed Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) by an average of 42.6% and 33.3%, respectively, compared to three interpolation methods (bicubic, bilinear, and Inverse Distance Weighting). Furthermore, TerraWind achieves an average reduction of 35.3% in wind speed MAE and 25.6% in wind speed RMSE compared to four deep learning models (Wind‐Topo, DeepCAMS, RCM‐emulator, and Uformer).
format Article
id doaj-art-e1db8867007244b0bc7c80cdc23ce87e
institution OA Journals
issn 0094-8276
1944-8007
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series Geophysical Research Letters
spelling doaj-art-e1db8867007244b0bc7c80cdc23ce87e2025-08-20T02:33:52ZengWileyGeophysical Research Letters0094-82761944-80072024-12-015123n/an/a10.1029/2024GL112124TerraWind: A Deep Learning‐Based Near‐Surface Winds Downscaling Model for Complex Terrain RegionJie Lian0Sirong Huang1Jiahao Shao2Peiyan Chen3Shengming Tang4Yi Lu5Hui Yu6Shanghai Engineering Research Center of Intelligent Education and Bigdata Shanghai Normal University Shanghai ChinaShanghai Engineering Research Center of Intelligent Education and Bigdata Shanghai Normal University Shanghai ChinaShanghai Engineering Research Center of Intelligent Education and Bigdata Shanghai Normal University Shanghai ChinaShanghai Typhoon Institute China Meteorological Administration Shanghai ChinaShanghai Typhoon Institute China Meteorological Administration Shanghai ChinaShanghai Typhoon Institute China Meteorological Administration Shanghai ChinaShanghai Typhoon Institute China Meteorological Administration Shanghai ChinaAbstract Wind downscaling is crucial for refining coarse‐scale wind estimates, improving local‐scale predictions, and supporting various applications like risk assessment and planning. Dynamic downscaling models demand extensive computational resources and time, leading to a shift toward more efficient statistical downscaling, whereas it often overlooks inter‐variable and inter‐station spatial correlations. Addressing this, we propose TerraWind, a deep learning‐based downscaling method for complex terrain regions. TerraWind enhances accuracy by incorporating topographic factors and inter‐station linkages, capturing wind field interactions with terrain at multiple scales. Experimental results in Eastern China demonstrate that TerraWind reduces wind speed Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) by an average of 42.6% and 33.3%, respectively, compared to three interpolation methods (bicubic, bilinear, and Inverse Distance Weighting). Furthermore, TerraWind achieves an average reduction of 35.3% in wind speed MAE and 25.6% in wind speed RMSE compared to four deep learning models (Wind‐Topo, DeepCAMS, RCM‐emulator, and Uformer).https://doi.org/10.1029/2024GL112124
spellingShingle Jie Lian
Sirong Huang
Jiahao Shao
Peiyan Chen
Shengming Tang
Yi Lu
Hui Yu
TerraWind: A Deep Learning‐Based Near‐Surface Winds Downscaling Model for Complex Terrain Region
Geophysical Research Letters
title TerraWind: A Deep Learning‐Based Near‐Surface Winds Downscaling Model for Complex Terrain Region
title_full TerraWind: A Deep Learning‐Based Near‐Surface Winds Downscaling Model for Complex Terrain Region
title_fullStr TerraWind: A Deep Learning‐Based Near‐Surface Winds Downscaling Model for Complex Terrain Region
title_full_unstemmed TerraWind: A Deep Learning‐Based Near‐Surface Winds Downscaling Model for Complex Terrain Region
title_short TerraWind: A Deep Learning‐Based Near‐Surface Winds Downscaling Model for Complex Terrain Region
title_sort terrawind a deep learning based near surface winds downscaling model for complex terrain region
url https://doi.org/10.1029/2024GL112124
work_keys_str_mv AT jielian terrawindadeeplearningbasednearsurfacewindsdownscalingmodelforcomplexterrainregion
AT sironghuang terrawindadeeplearningbasednearsurfacewindsdownscalingmodelforcomplexterrainregion
AT jiahaoshao terrawindadeeplearningbasednearsurfacewindsdownscalingmodelforcomplexterrainregion
AT peiyanchen terrawindadeeplearningbasednearsurfacewindsdownscalingmodelforcomplexterrainregion
AT shengmingtang terrawindadeeplearningbasednearsurfacewindsdownscalingmodelforcomplexterrainregion
AT yilu terrawindadeeplearningbasednearsurfacewindsdownscalingmodelforcomplexterrainregion
AT huiyu terrawindadeeplearningbasednearsurfacewindsdownscalingmodelforcomplexterrainregion