AMS 14C analysis of Late Pleistocene non-analog faunal components from 21 cave deposits in southeastern North America

Abstract

Late Wisconsinan micromammal faunas collected from caves in southeastern North America are characterized by nearly twice the number of species than are present in the region today. It has been proposed that this richness was achieved by paleoenvironments that permitted both the immigration and intermingling of boreal, steppe, and sub-tropical taxa with the present day deciduous forest micromammal community. Contemporaneity of component taxa cannot be assumed without 14C dating because taphonomic processes can produce similar configurations. If ‘non-analog’ specimens are contemporaneous, they represent communities with no modern analogs. If the dates on component species are disjunct, they are an artifact of time averaging and the species have a different paleoecological significance. The question remains: What is a significant interval of time to assert a community versus a time-averaging interpretation for Pleistocene faunal configurations? Analysis of 132 AMS 14C dates from 21 caves in southeastern North America (Fig. 1) on 16 presently allopatric or marginally sympatric (distantly) micromammal taxa demonstrate that these species are contemporaneous in many deposits. While other sites produced non-analog associations because of taphonomic processes, dates on many component specimens are contemporaneous regionally. The regional, as well as local, non-analog associations documented here demonstrate that a widespread, non-analog micromammal community occupied southeastern North America during the last glacial. These species-rich, non-analog communities, which contrast sharply to the species-depauperate Holocene faunas, probably were supported by unique but chronologically and geographically variable last-glacial environments. Although the AMS dates are not evenly distributed chronologically and radiocarbon plateaus occasionally exert influence, late Wisconsinan, non-analog communities of variable composition were present and likely influenced by Heinrich events. Other factors, some not yet identified, undoubtedly contributed to community structures.