219-12 Reassessing Rodent Diversity in the Oligo-Miocene of the Pacific Northwest Using Museum Dark Data

Session: Paleontology, Diversity, Extinction, Origination (Posters)

Presenting Author:

Authors:

Abstract:

The Oligo-Miocene rodent fossil record of the Pacific Northwest is critical to understanding how North American small mammals responded to major climatic, tectonic, and ecological shifts. The development of occurrence databases, such as the Paleobiology Database (PBDB), has revolutionized the analysis of diversity dynamics and ecological structure through time. However, this digital revolution has depended on the published record of diversity, which is biased by the priorities of systematic studies and leaves museum catalog occurrences, so-called “dark data,” unexamined. These dark data offer important information not included in research databases.

Here, we reassess the taxonomic diversity of more than 70 localities with rodent occurrences in the PBDB, spanning the Arikareean through Hemphillian NALMAs, by compiling and analyzing collections catalog data from four museums that house most of the fossil material from our study area. After identifying equivalent museum localities, synonymies, and repeat occurrences across the PBDB and museum records, we find that museum catalogs can provide ~40% more genus- and species-level occurrences than the PBDB alone. The problems posed by museum dark data are exacerbated by disparities in attention, with most new occurrences coming from smaller, less-visited museum collections. After incorporating these new occurrences, we find that γ-diversity increases by more than 15% compared to the PBDB-only dataset, and α-diversity increases significantly in time bins with a high number of localities (Arikareean, Hemphillian).

Faunal dissimilarity metrics (e.g., β-diversity) may be less affected by the addition of new occurrences, especially at shallower timescales. However, museum occurrence data may play a key role in increasing confidence in estimates of β-diversity. They are also of vital importance for assessing the certainty of the biostratigraphic first and last appearance data (FAD/LAD) of many taxa. The new occurrences substantially increase the proportion of well-sampled localities (≥10 unique species or genera), which is essential for accurately analyzing aspects of community ecology such as functional diversity or relative abundance. This work highlights the value of museum collections catalogs for facilitating the reconstruction of more complete paleocommunities. Future efforts to evaluate the accuracy of specimen identifications in these catalogs will further clarify their utility for diversity analyses.

Geological Society of America Abstracts with Program. Vol. 57, No. 6, 2025

doi: 10.1130/abs/2025AM-9250

© Copyright 2025 The Geological Society of America (GSA), all rights reserved.