CHANGES IN SMALL MAMMAL ABUNDANCE DISTRIBUTIONS FOLLOWING THE LOSS OF LARGE MAMMAL ECOSYSTEM ENGINEERS AT THE TERMINAL PLEISTOCENE

Much attention has been paid to the decline of large-bodied mammals because of their importance as ecosystem engineers and their elevated extinction rates in the Pleistocene. Less is known about the coinciding shifts in ecological structure among small mammals. Here, we evaluate the response of small mammals at Halls Cave, TX to the terminal Pleistocene extinction using species abundance distributions (SADs). Differences in SADs can distinguish communities with little to no resource partitioning (saturated model) or where niche partitioning is a “zero sum” game (e.g., Geometric, Log-Series or Zero-Sum Multinomial models) from communities with complex guild structure (e.g., Lognormal, Zipf or Zipf-Mandelbrot models). We divided the small mammal record at Hall’s Cave over the last 21,000 years into 16 intervals, 8 pre-, and 8 post- extinction. We fit each interval with the aforementioned models. Model likelihoods reflect the probability of observing X species with 1...n specimens given N total specimens. AICc weights were used to contrast the models. We find that although most intervals pre-extinction were best fit by a Lognormal, only one post-extinction interval was. Instead, post-extinction, SADs were best fit by saturated or Geometric models. Our results demonstrate that the Halls Cave small-mammal community was greatly altered after the terminal Pleistocene extinction, with underlying ecological structure severely disrupted or simplified. Moreover, our results suggest a loss of a more complex structure that never re-appeared. Future body size downgrading caused by the loss of currently at-risk large mammals is likely to result in similar simplification of surviving mammal communities.