Southeastern Section - 66th Annual Meeting - 2017

Paper No. 5-14
Presentation Time: 8:00 AM-12:00 PM

EARLY MIOCENE PALEOLANDSCAPE OF THE LOPEROT PRIMATE FOSSIL SITE (KENYA) DETERMINED BY RHIZOLITH SIZE VARIANCE AND STABLE ISOTOPIC SIGNATURE


LIUTKUS-PIERCE, Cynthia M.1, WRIGHT, James D.2 and BEANE, Luke1, (1)Dept. of Geology, Appalachian State University, PO Box 32067, Boone, NC 28608, (2)Department of Earth and Planetary Sciences, Rutgers University, 610 Taylor Road, Piscataway, NJ 08854, beanela@appstate.edu

Since 2012, the Loperot Miocene Project has analyzed a highly fossiliferous site in western Turkana that contains a unique assemblage of Early Miocene apes and monkeys never before recorded. Our work indicates that the ~28 m of section surrounding the fossils at Loperot represent deposition on a meandering river (identified by fluvial sands) floodplain (noted by paleosols with abundant rhizoliths). Our goal is to determine how the landscape may have affected the faunal assemblage. We analyzed the size variation and stable isotopic signatures of rhizoliths from 3 distinct stratigraphic units that contained important fossils. Comparison of the length vs. width data of rhizoliths collected from each unit (AS = ape site, MS = monkey site, and CS = croc skull site) indicates similar ranges in lengths, but distinct average widths (5.2mm for AS [n=58], 7.9mm for MS [n=56], and 10.2mm for CS samples [n=33]). The range in widths for CS samples (1-σ st dev) is higher (3.8) than for AS (2.4) and MS (2.4) samples, as is the average rhizolith width. This may indicate a greater variety of plant types on the surface at that site, or could indicate that plants in the CS unit had a longer amount of time to grow. It is unlikely that the range is due only to variation in plant type, since δ13C values of CS samples differ by only ~1‰ (-11.2 to -10.1‰, C3 plants). Covariance between δ18O and δ13C values from microsamples across individual rhizoliths is consistent regardless of stratigraphic unit. In general, higher δ18O and δ13C values are found on the outsides of the rhizoliths and decrease towards the center. Optical and cathodoluminescence microscopy confirm that the rhizoliths are not diagenetically overprinted. The maximum decrease in δ18O (from edge to center) is ~2‰, and is unlikely due only to temperature change (2‰ decrease = increase of ~10˚C). Instead, calcite on the edges precipitated first around plants on the floodplain in evaporatively-concentrated water. As plants died and decayed away (releasing low δ13C from C3 vegetation), the resulting holes infilled during the next flood resulting in lower δ18O and δ13C values in the rhizoliths’ interiors. The rhizolith stable isotope values thus record short-term environmental changes on the floodplain (weeks to months?) and provide critical information about rapid, localized landscape variability.