Large-Scale Approaches in the Quaternary: New Approaches, New Insights
The Quaternary offers multiple advantages for biogeographers and global change ecologists interested in understanding the climatic and biotic processes governing species distributions, community assembly, and species persistence (or extinction) during periods of large environmental change. Climate changes were large, repeated, and abrupt, with regional rates of change similar to those projected for the 21st century. Shifts in insolation and temperature seasonality produced climates with no modern analog. There is a long tradition of data-model comparisons in paleoclimatology, with paleoclimatic proxies used as benchmarks for earth system models, to put recent climate changes into context, and to understand the natural and anthropogenic drivers of climate variations. A new wave of data-model comparisons has emerged over the past decade, primarily focused on testing biogeographic theory and assessing and improving the predictive ability of ecological forecasting models. This new wave is driven by three main factors: 1) The development of paleoclimate model simulations and independent paleoclimatic proxies, which allows paleoecological data to serve as signals of ecological response rather than as direct paleoclimate proxies, which has traditionally been the case. 2) The on-going assembly of paleoecological data into well-organized and community-curated data resources such as the Neotoma Paleoecology Database (www.neotomadb.org), enabling continental-scale multi-proxy syntheses. 3) The increasing urgency to assess the exposure and sensitivity of species to 21st-century climate change and develop robust science-based adaptation strategies. Examples of continental-scale syntheses of fossil pollen data and their application to test key biogeographic hypotheses include understanding the velocity of species distribution shifts during rapid climate change, the stability of species-climate relationships, and whether we can substitute space for time when building biodiversity models. The development of Bayesian statistical models that incorporate multiple sources of temporal and proxy uncertainty are enabling more comprehensive and more robust inference. This in turn is opening the door to new kinds of data-model assimilation and collaborations with terrestrial ecosystem modelers, in which paleoecological data can be used to initialize, validate, and improve ecosystem models. Additionally, the growth and growing availability of archaeological databases offer the opportunity for new integrated assessments of the interactions among climate, terrestrial ecosystems, and human land use and land cover change.
