189 Paleontology, Paleoecology/Taphonomy (Posters)

Tuesday, 21 October 2025
8:00 AM - 5:30 PM
HBGCC, Hall 1

Assessing the Impact of Dataset Bias in Understanding Small Mammal Communities: A Comparison of Trapping and Owl Pellet Data From the Koanaka Hills, Botswana

Paper number: 189-24
Program: Technical Sessions
Time: 8:00 AM – 8:00 AM

Authors

Abstract

Small mammal fossil assemblages are frequently used to infer paleoenvironments because some species associate with specific habitats. However, these assemblages may be affected by taphonomic processes, collector biases, and taxonomic misidentification. Many small mammal assemblages were accumulated by owls, introducing predator bias. Nevertheless, such assemblages are often used to reconstruct paleoenvironments, with fossil taxa often compared directly with trapping datasets to interpret community composition and environmental shifts. Trapping data may also be biased, including species misidentification; behavior, age, sex, and activity levels; application, duration, and timing of trapping efforts; trap type effectiveness; and baits used. To assess the results of dataset type-driven biases, we compared two genus-level datasets from the Koanaka Hills bushveld, Botswana: (1) genetically-identified small mammals trapped using Sherman and pitfall traps in 2008 (n=185) and 2009 (n=348), and (2) modern Tyto alba pellets collected from caves in 2008 (n=127) and 2009 (n=50). Minimum numbers of individuals for pellets were calculated using the maximum number of non-repeating mandibles or crania per genus. Shannon-Wiener diversity indices (H') and Jaccard and Bray-Curtis dissimilarities assessed taxonomic and abundance differences. Fourteen small mammal genera appeared across datasets. Trapping yielded higher H' than pellets, reflecting broader, more even community sampling. Presence/absence comparisons showed overlap across years within each dataset type, indicating consistent type-specific biases; however, four taxa appeared in only one trapping year, suggesting additional within-dataset bias. For example, Steatomys was captured once in 2009 but was absent in 2008, yet accounted for 57% of the 2008 pellet assemblage, suggesting trap underrepresentation. Trapping and pellet datasets showed dissimilarity. Five genera trapped, including three abundant taxa (Aethomys, Micaelamys, Elephantulus), were absent from pellets, indicating predation biases that excluded common species. Abundance was consistent between trapping years but varied more in pellets, potentially reflecting owl dietary variation. Furthermore, field identifications had ~50% error before genetic confirmation, implying lower reliability of pellet identifications based solely on craniodental remains. In summary, modern small mammal communities are challenging to characterize fully, and single-year trapping data, especially without genetic confirmation, should not be considered definitive. Pellet biases suggest fossil assemblages may be distorted before taphonomic changes, and any paleoenvironmental reconstruction warrants caution.

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

doi: 10.1130/abs/2025AM-412934

© Copyright 2025 The Geological Society of America (GSA), all rights reserved. Permission is hereby granted to the author(s) of this abstract to reproduce and distribute it freely, for noncommercial purposes. Permission is hereby granted to any individual scientist to download a single copy of this electronic file and reproduce up to 20 paper copies for noncommercial purposes advancing science and education, including classroom use, providing all reproductions include the complete content shown here, including the author information. All other forms of reproduction and/or transmittal are prohibited without written permission from GSA Copyright Permissions.