GLIDE-SOL: a GPU-accelerated global lightweight infrastructure for diagnostic environmental modeling with SOLWEIG

Zonato, Andrea; Kamath, Harsh G.; Sudharsan, Naveen; Monaco, Luca; Kittner, Jonas; Wolf, Luise; Demuzere, Matthias; Middel, Ariane; Bechtel, Benjamin; Milelli, Massimo

doi:10.5194/gmd-19-7389-2026

Articles | Volume 19, issue 15

https://doi.org/10.5194/gmd-19-7389-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-19-7389-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 19, issue 15

Model description paper

|

11 Aug 2026

Model description paper |

| 11 Aug 2026

GLIDE-SOL: a GPU-accelerated global lightweight infrastructure for diagnostic environmental modeling with SOLWEIG

Andrea Zonato, Harsh G. Kamath, Naveen Sudharsan, Luca Monaco, Jonas Kittner, Luise Wolf, Matthias Demuzere, Ariane Middel, Benjamin Bechtel, and Massimo Milelli

Data sets

Data2Resilience - A Biometeorological Weather Station Network in Dortmund (Version 1.0.0) C. Hüser et al. https://doi.org/10.5281/zenodo.18221203

Model code and software

GLIDE-SOL: A GPU-accelerated Global Lightweight Infrastructure for Diagnostic Environmental Modeling with SOLWEIG A. Zonato et al. https://doi.org/10.5281/zenodo.18671813

Short summary

Cities need fast, reliable heat-stress maps to plan cooling measures and protect people. We built an automated workflow that gathers global public data, runs an outdoor comfort model much faster on graphics processing units, and adds simple corrections for wind and night-time warming. Tested in Dortmund against many sensors, errors fell from about ten to under three °C.