When Data Substitution Changes the Scientific Question: Claim-Level Evidence Admissibility in Wildfire Robustness Analysis — Reproducibility Archive

Abstract

Reproducibility materials supporting the manuscript "When Data Substitution Changes the Scientific Question: Claim-Level Evidence Admissibility in Wildfire Robustness Analysis." This archive supports the claim-specific continuity test through the benchmark and release-contract registries, registered semantic coding, SOCR schema and examples, verification code, conformance and metamorphic tests, rule-to-record checks, and the blinded semantic-coding packet. Deterministic verification is conditional on the registered semantic coding. Independent semantic-coding reliability was not evaluated. The archive does not establish external validation. SOCR is an implementation and serialization format, not a separate contribution; the method is not a generic ranking, scoring, MCDA, fuzzy, or decision framework. Reproduce with "bash run_all.sh" using Python 3. The preserved checks report 15/15 conformance/metamorphic tests and 5/5 rule-to-record derivations PASS. The archive describes the optional jsonschema dependency and its structural fallback. The deposited Reproducibility_Archive.zip is the unchanged canonical archive supplied with the manuscript. SHA-256: 341dfa3f08aa72e1f2716837986217ace6b174de73c1828aa6ed5136853bee16. The public GitHub mirror is https://github.com/shaikhamalkawi-ux/14-FIRE-READY-Claim-Specific-Continuity-Test-Reproducibility-Archive. Its verified scientific-material snapshot is commit 48d8bec070b47829b5ea8a85b35e87f02e67cb7f. JSON formatting differences in the mirror do not alter the registered contents. This is a direct Zenodo deposit; no GitHub release or tag is used for this deposit. Licenses apply by authorship and file scope, as specified in LICENSES_AND_RIGHTS.txt: MIT for the authors' original software; CC BY 4.0 only for the authors' original documentation and data contributions for which they own the rights. No third-party material, external-source excerpt, underlying source dataset, or publication is relicensed. Such material retains its original rights and source terms. Source-reported numbers, citations, identifiers and factual metadata are provided for traceability; no ownership of the underlying sources is claimed. The deposit contains no publisher PDFs or restricted third-party binaries. Author affiliations:Ghassan Malkawi - Faculty of Computer Information Science (CIS), Higher Colleges of Technology, Al Ain, Abu Dhabi, United Arab Emirates.Ahmed Abdelaziz Elsayed - Department of Computer Engineering and Computational Sciences, Canadian University Dubai, Dubai, United Arab Emirates.