Geoscientific Model Development
Geoscientific Model Development
GMD
Articles | Volume 19, issue 18
Articles | Volume 19, issue 18
https://doi.org/10.5194/gmd-19-8895-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-19-8895-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
Articles | Volume 19, issue 18
Development and technical paper
 | 
21 Sep 2026
Development and technical paper |  | 21 Sep 2026

A deep learning framework for gridding daily climate variables from a sparse station network

Alexandru Dumitrescu

Viewed

Total article views: 1,432 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
948 383 101 1,432 252 105 80
  • HTML: 948
  • PDF: 383
  • XML: 101
  • Total: 1,432
  • Supplement: 252
  • BibTeX: 105
  • EndNote: 80
Views and downloads (calculated since 04 Mar 2026)
Cumulative views and downloads (calculated since 04 Mar 2026)

Viewed (geographical distribution)

Total article views: 1,432 (including HTML, PDF, and XML) Thereof 1,412 with geography defined and 20 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 21 Sep 2026
Download
Short summary
Accurate daily climate maps are essential for environmental monitoring, yet many regions lack dense weather-station networks. We developed a deep-learning method that converts sparse station measurements into high-resolution gridded temperature and precipitation fields over complex terrain. Tested over Romania, it outperforms traditional geostatistical interpolation by learning how topography shapes local climate, while providing reliable uncertainty estimates for each prediction.
Read more
Share