Preprints
https://doi.org/10.5194/npg-2023-10
https://doi.org/10.5194/npg-2023-10
10 May 2023
 | 10 May 2023
Status: a revised version of this preprint was accepted for the journal NPG and is expected to appear here in due course.

A two-folds deep learning strategy to correct and downscale winds over mountains

Louis Le Toumelin, Isabelle Gouttevin, Clovis Galiez, and Nora Helbig

Abstract. Assessing wind fields at a local scale in mountainous terrain has long been a scientific challenge partly because of the complex interaction between large-scale flows and local topography. Traditionally, the operational applications that require high resolution wind forcings rely on downscaled outputs of numerical weather predictions systems. Downscaling models either proceed from a function that links large scale wind fields to local observations (hence including a corrective step), or use operations that account for local scale processes, through statistics or dynamical simulations, and without prior knowledge of large scale modeling errors. This work presents a strategy to first correct and then downscale the wind fields of the numerical weather prediction model AROME operating at 1300 m grid spacing, by using a modular architecture composed of two artificial neural networks and the DEVINE downscaling model. We show that our method is able to first correct the wind direction and speed from the large scale model (1300 m), and then accurately downscale it to a local scale (30 m) by using the DEVINE downscaling model. The innovative aspect of our method lies in its optimization scheme that accounts for the downscaling step in the computations of the corrections of the coarse scale wind fields. This modular architecture yields competitive results without suppressing the versatility of the downscaling model DEVINE, which remains unbounded to any wind observations.

Louis Le Toumelin et al.

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on npg-2023-10', Anonymous Referee #1, 05 Jul 2023
    • AC1: 'Reply on RC1', Louis Le Toumelin, 15 Oct 2023
  • RC2: 'Comment on npg-2023-10', Anonymous Referee #2, 18 Aug 2023
    • AC2: 'Reply on RC2', Louis Le Toumelin, 15 Oct 2023

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on npg-2023-10', Anonymous Referee #1, 05 Jul 2023
    • AC1: 'Reply on RC1', Louis Le Toumelin, 15 Oct 2023
  • RC2: 'Comment on npg-2023-10', Anonymous Referee #2, 18 Aug 2023
    • AC2: 'Reply on RC2', Louis Le Toumelin, 15 Oct 2023

Louis Le Toumelin et al.

Louis Le Toumelin et al.

Viewed

Total article views: 971 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
814 137 20 971 32 9 6
  • HTML: 814
  • PDF: 137
  • XML: 20
  • Total: 971
  • Supplement: 32
  • BibTeX: 9
  • EndNote: 6
Views and downloads (calculated since 10 May 2023)
Cumulative views and downloads (calculated since 10 May 2023)

Viewed (geographical distribution)

Total article views: 942 (including HTML, PDF, and XML) Thereof 942 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 24 Nov 2023
Download
Short summary
Forecasting wind fields over mountains is of high importance for several applications and particularly to understand how wind erodes and deposes snow. Forecasters rely on operational wind forecasts over mountains, which are currently only available on kilometric scales. These forecasts can also be affected by errors of diverse origins. Here we introduce a new strategy based on artificial intelligence to correct large scale wind forecasts in mountains and increase their spatial resolution.