Cody Mason, Jourdan Speessen, Brian Romans , Daniel F. Stockli, Michael D Blum, Jack Heltzer, G. Paul Miles, Andrea Fildani

The Pleistocene Mississippi sediment routing system has experienced significant drainage reorganizations in response to Northern Hemisphere glaciation. Geologically recent (<25 ka) and large (~90 km) river avulsions have occurred in the Mississippi’s lower alluvial valley, yet whether these punctuated autogenic phenomena influenced down system records of sedimentary provenance is unknown. We present U–Pb detrital zircon (DZ) geochronology from near last glacial maximum (LGM; ~25 ka) and early deglacial (~14 ka) aged fluvial deposits of the Mississippi Valley, United States, integrated with published DZ samples from the Gulf of Mexico (Mississippi fan), to show a high similarity of DZs from late Pleistocene fluvial deposits and those from time correlative deep-marine deposits. These results imply the last glacial Mississippi River efficiently transferred DZs from river to deep-sea with little modification to the riverine DZ signature. Stratigraphically deeper DZ samples in the Mississippi fan contain distinct DZ age spectra, which likely represent the primary DZ signature of the Mississippi between isotope (MIS) stages 5 – 3. Mixture modeling of DZ samples recovered from late Pleistocene (MIS-2) deposits in the Western Lowlands and Eastern Lowlands of the Mississippi Valley suggests ~80% of the DZs in a LGM-to-deglacial Mississippi River were sourced from the glacial Missouri River, with lesser but appreciable contributions from the glacial Upper Mississippi River (13 – 20%). Only minute contributions of DZs are interpreted from the glacial Arkansas (1 – 5%) and Ohio Rivers (0 – 2%). Given the interpreted lack of Ohio river derived DZs in the Western Lowlands at ca. 25 ka, we suggest a glacial Ohio river did not join a combined Missouri-Mississippi River at the sampled location in the Western Lowlands at ~25, and may or may not have joined the Missouri-Mississippi in the Eastern Lowlands at ~15 ka. Our geochronological analyses included several grain size fractions for each sample location; a lack of Gondwanan-Appalachian aged DZs in one sub sample grain size fraction illustrates the need for mineral separation methods and U-Pb age measurement methods that include all available size fractions, or risks serious error in provenance interpretations.