On the sensitivity of machine-learned probabilistic weather forecast models to scale-aware scoring rules

Ethics

Major revisionpanel verdict ยท 2026-09-01

Ethics & Compliance Reviewer

SCORE: N/A CONFIDENCE: 5

Assessment

This manuscript presents no ethics or compliance issues requiring review. The work is a computational study using publicly available ERA5 reanalysis data (1979โ€“2020) to train and evaluate machine-learned weather forecast models. No human subjects, identifiable data, clinical applications, animal research, dual-use risks, or restricted-consent materials are involved. The computational experiments are performed on public supercomputing facilities and involve no data governance concerns beyond standard academic attribution.

Funding and competing interests are not explicitly stated in the manuscript, but this is a minor administrative matter for the editor rather than an ethics trigger โ€” the work involves no category that would require such disclosure to be substantive to the review.

The manuscript is in scope for In Silico and raises no ethics or compliance barriers to publication.

โ† All documents in this review