GSA Annual Meeting in Phoenix, Arizona, USA - 2019

Paper No. 15-10
Presentation Time: 10:45 AM


ANTHONY, Kait1, LIUTKUS-PIERCE, Cynthia M.2, ARMENGOLT, Oscar2 and TAKASHITA-BYNUM, Kevin K.3, (1)Southwestern Community College, Sylva, NC 28779, (2)Department of Geological and Environmental Sciences, Appalachian State University, 572 Rivers St., Boone, NC 28608, (3)Watershed Studies Institute, Murray State University, Murray, KY 42071

Previous work reconstructs the early Miocene paleolandscape of the primate-fossil-bearing site of Loperot, Kenya as a large perennial river system bordered by a riparian forest under semi-arid climate. With exposures of ~30m, we question whether landscape (and/or climate) changed through time. δ13C results from Loperot rhizolith samples suggest that vegetation at the site fluctuated from pure C3 (Units 26, 17) to water-stressed C3/CAM (Unit 22), but δ18O values show little variation, indicating little change in precipitation regime. Thus, we test the hypothesis that vegetation shifts between individual units are due to proximity to a meandering river rather than climate. In order to do this, we examined the soil chroma (Munsell) and sediment texture (grain size, sorting, roundness, % clay) of 15 paleosol units from the exposed strata at Loperot. We hypothesize that river proximity (i.e., energy) should be a first order control on sediment maturity; paleosols that formed close to the river channel (dominated by C3 riparian vegetation) should be coarser grained, more angular, poorly sorted, and contain less clay (low textural maturity), and those forming farther from the channel (in more open shrubland) should be more texturally mature. We report that grain size and roundness do not change significantly, that units reportedly forming close to the river channel (26, 17) are better sorted than those forming farther away (22), and that the highest clay content is found in samples proposed to be on the distal floodplain (22). Without repeated inundation from river flooding, the distal floodplain samples retain a high clay content and poor sorting, whereas proximal samples experienced frequent elluviation of clay. Upsection, clay content and soil chroma values of paleosols decrease above Unit 17; lower soil chroma is a function of wetter climate. Thus, we note 2 important conclusions: (1) differences in sedimentological properties of individual units can be used to determine river proximity and explain vegetation distribution, and (2) the permanent shift in clay content and soil chroma above Unit 17 is due to climate change (increased humidity). Paleosol geochemical data support this interpretation as the ratio of mobile cations to aluminum decreases, and CIA-K increases, permanently above Unit 17.