Temporal Decay and Volunteer Bias in Citizen Science Biodiversity Databases: Epistemological Anchoring as a Structural

Abstract Citizen science platforms such as iNaturalist and national biodiversity information facilities including Thailand's TH-BIF serve as primary references for biodiversity research and policy. However, these systems share a common structural vulnerability: the validity of their records degrades over time as taxonomy evolves, while the records themselves remain static. This paper identifies and characterises three failure modes — Volunteer Confirmation Bias, Temporal Decay, and the Scale Problem — and demonstrates their practical impact through a live validation test of ten Thai species. We show that 80% of tested species carry outdated or incorrect nomenclature in commonly referenced databases. We propose Epistemological Anchoring as a structural remedy: binding every species record to a Digital Object Identifier (DOI) and synchronising against an authoritative taxonomic backbone on a quarterly basis. We describe the Dynamic Periodic Anchor methodology implemented in Master Harmonix™ and present its application in the context of biodiversity finance and the Kunming-Montreal Global Biodiversity Framework. Keywords: biodiversity informatics, taxonomic synonymy, citizen science, temporal decay, DOI anchoring, biodiversity credits, iNaturalist, TH-BIF, Master Harmonix™