Submerged karst cave systems in Mexico’s Yucatán Peninsula represent repositories of cultural heritage preserving Pleistocene human and animal remains, ochre mines, Maya artifacts, and paleoenvironmental records documenting thousands of years of interaction between people, animals, and landscape. For more than 25 years, the Centro Investigador del Sistema Acuífero de Quintana Roo (CINDAQ) has conducted systematic exploration and documentation of these underwater systems, generating extensive datasets through the work of numerous divers with varied backgrounds, languages, and levels of scientific specialization. Data were collected using heterogeneous methods, including manual and semi-automated survey, sketch mapping, comments, photography and videography, photogrammetry, and Geographic Information Systems (GIS).

We present a structured approach to transforming long-term, heterogeneous legacy datasets into usable and reproducible scientific resources. The workflow is organized along four axes: (1) curating legacy data through deduplication, indexing, and the establishment of Single Sources Of Truth; (2) defining standards for legacy translation and new data acquisition, such as controlled vocabularies for survey comments and image tagging; (3) developing tools integrating old and new data while enforcing standards, including image search engines, GIS analyses, automated validation scripts, and photogrammetry pipelines; and (4) documenting paradata through knowledge bases and open, software-agnostic formats ensuring long-term usability.

Examples cover the reuse of standardized observations to model halocline depth and assess how environmental dynamics influence cave accessibility and preservation ; the normalization of decades of legacy comments related to ochre, calcite, and charcoal in the study of submerged prehistoric ochre mines ; and the classification of photographic archives to relocate animal and human remains sites for archaeological and paleontological documentation.

This work shows how systematic standardization of heterogeneous datasets can enable new analyses while improving quality and reproducibility, and how these methods can be transferred to other fields beyond cave-related studies to any discipline managing complex, long-term field data.

First Author: Julien Fortin

Co-Authors: Samuel Meacham, Andreas Rosland, Dominique Rissolo, Loren Clark, Wetherbee Dorshow

Keywords:
Legacy data
Data standardization
Underwater cultural heritage
Digital documentation
Geographic Information Systems (GIS)