Future climates are projected to warm substantially and rapidly, requiring correspondingly fast responses if species are to remain viable.  Climates have changed similarly rapidly at some periods in the recent past, most notably during the initiation of the Bølling-Allerød approximately 14.5 thousand years ago (ka) and the initiation and termination of the Younger Dryas (12.9-11.5 ka).  Many high-resolution fossil pollen records are available to investigate vegetational responses to these rapid climatic changes, but synthesis has been hindered by poor accuracy and precision in the underlying radiocarbon dates of many sites.  The accuracy of some radiocarbon dates has been demonstrated to be off by thousands of years.  However, a new generation of fossil pollen records, with high-precision radiocarbon dates on carefully chosen organic carbon, has been collected over the last decade.  A systematic revision of many chronologies is needed to provide a firm temporal foundation with which to reconstruct the spatial responses of species and communities to rapid climatic change.

We investigated ecological responses to rapid climatic change in eastern North America using the Neotoma paleoecological database (www.neotomadb.org) and data contributed by individual investigators.  We first developed criteria to identify highly precise and accurate chronologies (‘golden spike’ chronologies) and searched for pollen records that met these criteria.  We then used the golden spikes to reassess the regional timing of important vegetational events over the last 21,000 years, and investigated in more detail the ecological responses of key taxa to rapid paleoclimatic change.

Results/Conclusions

Very few existing pollen cores can be characterized as golden spikes, and the few golden spike sites that exist have a non-uniform spatiotemporal distribution.  For example, over 400 pollen cores span the crucial Bølling-Allerød and Younger Dryas periods, yet less than 25 met our criteria for ‘golden spike’ chronologies.  The majority of golden spike sites were located in eastern North America and the upper Midwest; only two were in the southeastern US.  Using the golden spike sites, we reassessed and mapped the regional timing of major vegetation events such as Picea decline, the rise and decline in Fraxinus and Ulmus, and the expansion of Pinus, Quercus, and other taxa.  The responses of vegetation to abrupt climate change are spatially variable, both within and across regions, and are often time-transgressive.  Finally, we investigated range shifts of several taxa during periods of abrupt climate change and explored the relationship between abundance change and range shifts.