ABSTRACT

This project links functional anatomical traits in mammals and reptiles to temperature, precipitation, and vegetation cover using a Bayesian framework that allows us to explicitly test these vertebrate environmental proxies against independent pedogenic and paleobotanical evidence in order to reconstruct paleoclimates of the North American interior during the Miocene. We build estimates of paleoenvironment from entire faunas rather than individual species in order to combine the predictive strengths of ectotherms (reptiles) and endotherms (mammals). Our objectives are: (1) assemble a database of functional traits from Miocene vertebrate fossils at the local assemblage and biocenosis scale in the central Great Plains based on the extensive fossil collections of the Nebraska State Museum and biomolecular proxy data from select fossil localities; (2) develop Bayesian probabilistic climate spaces from sampled traits in modern faunas; (3) use the trait-climate spaces to estimate Miocene paleoenvironment and test for consilience between vertebrate estimates and independent proxies; (4) use the trait-based proxy results to test competing hypotheses of mid-continental floral composition and to determine the extent of coupling between global climate change and local environments during episodes of warming and cooling. This study is the most taxonomically and anatomically comprehensive analysis of faunal traits ever performed on a continuous terrestrial vertebrate fossil record, and our method can be extended to other proxies (stable isotopes, biomolecules, paleoflora) as a general method for paleoenvironmental estimation. We will generate paleoclimate profiles through the Miocene that can be used to test General Circulation Models and to forecast regional impacts of future anthropogenic global climate change.

Non-technical description: This project uses the relationship between the physiology and ecology of living mammals and reptiles to their modern environments to reconstruct the past climates and environments of the interior of North America during the Miocene Epoch (5.3 to 23.0 million years ago) using the fossil record. We will use the functional properties of different living and fossil species that are represented by their skeletal anatomy, such as method of locomotion, dietary specializations, and physiological regulation of body temperature, to reconstruct temperature, precipitation, and floral composition during known histories of global climate change. This approach provides paleontologists with a sophisticated way of understanding the geographic complexity of environmental change using fossils and it provides a record of what changes actually occurred in the Miocene. We will develop a new statistical framework for conducting such research that allows us to determine the relative probabilities of different past climates. For example, environmental temperature and precipitation are correlated to body size and tail length in modern snakes and correspond to the relative proportions of ankle bones in modern carnivorous mammals. By measuring these traits in fossil assemblages, we can statistically estimate past climates based on the relative probabilities of different temperature and precipitation values from both snake and mammals. Our method allows us to better assess the complex relationship between organisms and changing climate at both local and global scales, and will provide precise past climate values that can be used to test General Circulation Models and refine the ability to forecast future climate change.
In addition to increasing our understanding about these important issues, the funds spent on this project will provide training in geology, anatomy, computer programing, and mathematics to university undergraduate and graduate students. Additionally, it will contribute to scientific conferences and museum exhibits for the general public.

Please report errors in award information by writing to: awardsearch@nsf.gov.