Paper No. 277-8
Presentation Time: 3:25 PM
LATE HOLOCENE PALEOFLOODS AND GRAIN-SIZE VARIABILITY ALONG THE MIDDLE TENNESSEE RIVER IN NORTHERN ALABAMA
STEWART, C. Lance1, STINCHCOMB, Gary E.2, FORMAN, Steven L.3, DAVIS, Lisa4, LOMBARDI, Rachel4, LEIGH, David S.5, BLACKABY, Emily6 and HOCKADAY, William C.7, (1)Watershed Studies Institute and Earth and Environmental Sciences, Murray State University, 334 Blackburn Hall, Murray, KY 42071; Watershed Studies Institute and Earth and Environmental Sciences, Murray State University, Murray, KY 42071, (2)Watershed Studies Institute and Earth and Environmental Sciences, Murray State University, Murray, KY 42071, (3)Institute of Archaeology, Baylor University, Waco, TX 76798, (4)Geography, University of Alabama, Tuscaloosa, AL 35487-0322, (5)Department of Geography, The University of Georgia, Geog.-Geol. Building, 210 Field St., Room 204, Athens, GA 30602, (6)Department of Geosciences, Baylor University, 1 Bear Place, mail stop 97354, Waco, TX 76798, (7)Geology, Baylor University, One Bear Place #97354, Waco, TX 76798
Gauged and observed peak streamflow measurements spanning at most the past 80 years are often the only metric to forecast the likelihood of future floods. The time-series of flood events at decadal and sub-decadal resolution can be extended 1000s of years by incorporating paleoflood data collected from floodplains and terraces along river valley bottoms. Grain-size proxies can be used to estimate the magnitude of unobserved floods; however, landform proximity to the channel and its potential effect on grain-size distributions preserved in paleoflood deposits needs to be better defined. This study examines the late Holocene flood chronology of the middle Tennessee River, Alabama (USA), using clay-free relative fine sand (RFS) content and quantifies grain-size variation across fluvial landforms. Age control is provided by 30 optically stimulated luminescence (OSL) ages on quartz grains, ranging from 3120 ± 220 BP to 1970 ± 30 CE. Grain-size analysis was performed on sediment from flood deposits and buried soils formed on four distinct landforms including, the inside and outside bends of a paleomeander, a terrace, and a relict levee. Normalized RFS shows two large historic flood deposits that date to 75 ± 10 and 40 ± 10 years BP, which may reflect the 1867 CE flood of record and the 1917 CE flood. Seven large flood deposits with RFS magnitude equal to or greater than the 1867 CE flood date to 40 ± 10, 120 ± 15, 490 ± 30, 1370 ± 110, 1785 ± 125, 1830 ± 135 and 2190 ± 140 years BP.
Particle size (n=157) showed an ~normally distributed fine silt for the terrace, bi-modal distributions with distinct medium silt peaks for the inside and outside paleomeander bends and a large fine sand peak, positively skewed for the relict levee. The relict levee site has the clearest record of flood activity over the late Holocene interval with evidence of five weakly-developed buried soils separated by a minimum of five flood deposits > 20 cm thick. Despite weak soil development in the buried soils for all landforms, micromorphology shows evidence of clay translocation, reaffirming the need to examine depositional trends on a clay-free basis. Given that grain-size appears to vary as a function of alluvial landform and soil development, normalization and clay-free calculations are a necessary step when reconstructing paleofloods with a grain-size proxy.
![[Visit Client Website]](/img/gsa/banner.jpg)