Temporally Validated Ordinal and Multi-Threshold Models for Retrospective Wildfire Escalation Risk Stratification: A 22-Year Mediterranean Incident Study
Rattachement africain : tr, se. Niveau de preuve : code pays fourni par la source.
Le résumé fourni par la source
Reproducibility package for “Temporally Validated Ordinal and Multi-Threshold Models for Retrospective Wildfire Escalation Risk Stratification: A 22-Year Mediterranean Incident Study”. The archive contains the anonymized modelling table, self-contained Python analysis code, locked analysis protocol, event-level predictions, full-precision result tables, and publication figures for a retrospective analysis of 1,898 outcome-complete wildfire incidents recorded during 2004–2025. The institutional wildfire archive was supplied by the Mersin Regional Directorate of Forestry under research permission. Model development was restricted to 2004–2020, while the 2021–2025 period was retained as a locked temporal evaluation cohort. The archive supports reproduction of the multi-threshold, ordinal, calibration, capacity-constrained ranking, and spatio-temporal robustness analyses reported in the associated manuscript.
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.
Le contrôle bibliographique ouvert
Les institutions déclarées
Une affiliation ne permet pas de déduire la nationalité d’un auteur.