2009 Portland GSA Annual Meeting (18-21 October 2009)

Paper No. 10
Presentation Time: 10:30 AM

GEOLOGICAL TO EVOLUTIONARY SCALES OF TIME: CASE STUDIES OF CHANGE AND RATES OF CHANGE IN EOCENE MAMMALS


GINGERICH, Philip D., Department of Geological Sciences and Museum of Paleontology, The Univ of Michigan, Ann Arbor, MI 48109-1079, gingeric@umich.edu

North American mammalian biostratigraphy has traditionally relied on land-mammal ages of 1 to 5 million-year duration. In the late Paleocene and early Eocene of the Western Interior we have been able to develop more refined biochronologies yielding biozones on the order of 0.5 million years. An effort to develop an orbitally-tuned Paleocene-Eocene time scale is under way, which, if successful, will enable precession-scale (ca. 20 kyr) resolution in favorable stratigraphic settings. None of these come close to the ca. 0.1-20 year generational time scale of interest for mammalian evolutionary studies.

However, cumulative evolutionary change is systematically less than expected on long time scales and greater than expected on short time scales compared to change observed on average time scales. Rates behave the same way. Change and rates observed on a range of observable time scales can be used to extrapolate to other time scales of interest. The precision of the extrapolation is dependent on the variance of sampling for each observed time scale, and on the spacing of observed time scales relative to each other and relative to the extrapolated scale of interest.

In the case of mammals, change and rates of change extrapolated from empirical Eocene time scales 50 million years back in time predict change and rates of change observed from one living generation to the next today. Thus use of a range of relatively long time scales can sometimes provide answers for questions involving change and rate of change on shorter time scales. In the examples developed here these time scales are far shorter than we will ever achieve at the limits of geochronometry. For some questions observations on a range of time scales have a power greater than observations on the shortest achievable time scale.