Paper · 2026 · Open access

Multi-domain: A dynamic way of training across domains and resolutions

Aram Farhad Shafiq Salihi, Sophie Buurman, Even Norhagen, Mario Santa Cruz, Michiel Van Ginderachter, Thomas N. Nipen

Published 22 June 2026 · DOI 10.5194/ems2026-582

Open full textCite this paper

In brief

“Multi-domain: A dynamic way of training across domains and resolutions” is a 2026 paper by Aram Farhad Shafiq Salihi and 5 co-authors. OpenAlex counts 0 citations as of 21 November 2026. It is open access under a CC BY licence: the full text is free to read.

Citations
0
Authors
6

Abstract

The domain of weather forecasting is currently undergoing a significant transformation driven by advances in machine learning. State of the art data-driven weather models have demonstrated performance that surpasses the state of the art traditional numerical weather prediction (NWP) models, while operating at a fraction of the computational cost (Bouallegue et al., 2024). While global models like AIFS (Lang et al.,2024), GraphCast, and Pangu have gained a lot of attention, different flavours of high-resolution regional modeling have emerged developed by different meteorological institutes. Stretched-grid is a global model with an increased spatial resolution and dynamics over a region of interest, a novel approach to regional modelling. This capability is demonstrated to be highly competitive and even surpass state of the art regional NWP for certain variables (Nipen et al., 2025; Nordhagen et al., 2025).Building on the concept of stretched-grid and generalizing the idea to incorporate more high-resolution data, while avoiding intermediate fine-tuning and transfer learning steps, one can achieve high resolution predictions for any domain in Europe. In this work, we propose a probabilistic multi-domain model, introducing a new way of training across domains and resolutions, by utilizing the concept of dynamical graphs. This is done by alternating between different global and regional data and its corresponding graph across different spatial regions, grid-types and resolutions. This capability is enabled through a flexible encoder, processor and decoder architecture. The idea is to make the model less biased towards certain grid types, terrain and dynamics induced by the training data, enabling the model to generalize across resolutions and climate zones. The kilometre scale model has been trained on analyses from four high resolution NWP models covering various parts of Europe, including AROME-Artic (2.5km), the MetCoOp Ensemble Prediction System (MEPS, 2.5km), UWC-West (2km) and the Austrian Reanalysis (ARA, 2.5km). We evaluate the performance of this model on another domain that was left out of the training, and show improved generalizability when compared to a model trained on fewer domains. With one year of verification, we show improvements across parameters such as 2m temperature, mean sea level pressure, 10m wind speed and total precipitation.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Salihi, A. F. S., Buurman, S., Norhagen, E., Cruz, M. S., Van Ginderachter, M., & Nipen, T. N. (2026). Multi-domain: A dynamic way of training across domains and resolutions. https://doi.org/10.5194/ems2026-582

MLA 9

Salihi, Aram Farhad Shafiq, et al. “Multi-domain: A dynamic way of training across domains and resolutions.” 2026. https://doi.org/10.5194/ems2026-582.

Chicago

Salihi, Aram Farhad Shafiq, Sophie Buurman, Even Norhagen, Mario Santa Cruz, Michiel Van Ginderachter, and Thomas N. Nipen. 2026. “Multi-domain: A dynamic way of training across domains and resolutions.” https://doi.org/10.5194/ems2026-582.

Harvard

Salihi, A. F. S. et al. (2026) ‘Multi-domain: A dynamic way of training across domains and resolutions’. doi:10.5194/ems2026-582.

IEEE

A. F. S. Salihi, S. Buurman, E. Norhagen, M. S. Cruz, M. Van Ginderachter, and T. N. Nipen, “Multi-domain: A dynamic way of training across domains and resolutions,” 2026, doi: 10.5194/ems2026-582.

BibTeX

@article{salihi2026multi,
  title     = {{Multi-domain: A dynamic way of training across domains and resolutions}},
  author    = {Salihi, Aram Farhad Shafiq and Buurman, Sophie and Norhagen, Even and Cruz, Mario Santa and Van Ginderachter, Michiel and Nipen, Thomas N.},
  year      = {2026},
  doi       = {10.5194/ems2026-582},
  url       = {https://doi.org/10.5194/ems2026-582}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Meteorological Phenomena and Simulations

  • multi-domain training
  • probabilistic weather forecasting
  • numerical weather prediction

Work with this paper

Most-cited papers in Earth and Planetary Sciences

All highly cited Earth and Planetary Sciences papers

Questions about this paper

How do I cite this paper in APA style?

Salihi, A. F. S., Buurman, S., Norhagen, E., Cruz, M. S., Van Ginderachter, M., & Nipen, T. N. (2026). Multi-domain: A dynamic way of training across domains and resolutions. https://doi.org/10.5194/ems2026-582

How many times has this paper been cited?

0 times, according to OpenAlex on 21 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access: sign in and open the full text from this page.

Who wrote this paper?

Aram Farhad Shafiq Salihi, Sophie Buurman, Even Norhagen, Mario Santa Cruz, Michiel Van Ginderachter, Thomas N. Nipen.

What is the DOI of this paper?

10.5194/ems2026-582, permanently at https://doi.org/10.5194/ems2026-582. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 21 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.