FORM-TRACE: a reproducible workflow for converting forest-management documents into auditable transition indicators

This dataset provides the reproducibility package for FORM-TRACE, a workflow developed to convert heterogeneous forest-management documents into auditable transition indicators. It supports the MethodsX method article “FORM-TRACE: a reproducible workflow for converting forest-management documents into auditable transition indicators.” The package includes the transition definition, corpus manifest, extraction log, keyword-domain matrix, document-level domain scores, year-level and period-level score tables, run report, QA checklist, generated figures, and reproducibility documentation for the Harvard Forest / New England worked example covering 1908–2026. The workflow inspected 257 PDF files, identified 215 analytical forest-management documents, successfully extracted and scored 201 documents, flagged 14 failed or OCR-needed documents, and separated 42 methods-only references. FORM-TRACE uses deterministic keyword-domain scoring to generate normalized document and period-level signals for production and silviculture, ecological structure, carbon and ecosystem function, disturbance and climate risk, and governance implementation. AI-assisted tools were used only to support organization, methods-reference review, and manuscript preparation; final scoring was generated by deterministic scripts. Source PDFs and copyrighted source documents are not redistributed in this dataset. The manifest, metadata, logs, scoring outputs, and documentation are provided to support reproducibility with legally accessible source documents.