INVESTIGATING ENVIRONMENTAL HETEROGENEITY IN THE COMMUNITY STRUCTURES OF CONTEMPORARY PALEONTOLOGICAL ASSEMBLAGES

Environmental heterogeneity can occur at fine scales in modern ecosystems, and new studies illustrate the importance of considering small-scale environmental differences and micro-habitats in ecological studies. Yet spatially defining a paleontological assemblage often makes assumptions which can mask sub-regional variation in community structure. John Day Fossil Beds National Monument (JODA) has substantial paleontological collections with concisely dated and correlated stratigraphy across an extensive region, presenting a well-constrained paleontological context to test the effects of these assumptions on observed spatial diversity patterns.

We investigated whether sub-regional environmental heterogeneity could be identified at JODA through two paleoecological contexts. First, we compared community structure between six contemporaneous palaeobotanical localities comprising the Bridge Creek Flora (BCF) known from the three traditional “facies” of the Big Basin Member (~33 Ma) of the John Day Formation (JDF), covering ~1500 sq km and including 5032 fossils. Second, we compared contemporaneous mammalian communities through time from three major localities (Foree, Blue Basin, and Sheep Rock) of the Turtle Cove (TC) Member (~30-25 Ma) of the JDF, covering ~20 sq km and including 1911 fossils. For each context, we designated genus-level assemblages binned spatially by locality. The TC mammal assemblages were further binned temporally by geologic subunit, creating a series of locality-time paired assemblages. We applied Chord-distance analysis as a similarity index to compare the community structures of the assemblages in each context.

The BCF demonstrated disparity both between individual assemblages (Chord dist. up to 1.09) and within the traditional “facies” (Chord dist. up to 0.84), indicating spatial habitat variation. We also found evidence of heterogeneity between contemporary TC mammal assemblages (Chord dist. up to 0.59), which may reflect differences in habitat openness based on the key taxa associated with each faunal assemblage. Our results illustrate how different localities within this region can display variation in community structure and may not uniformly follow diversity trends, suggesting caution when evaluating analyses with significant regional binning.