USING PAIRED OPTICALLY STIMULATED LUMINESCENCE AND COSMOGENIC NUCLIDE 10BE DATING TO UNDERSTAND CHANGES IN EROSION RATE WITHIN A FILL-CUT TERRACE SEQUENCE: WHITEWATER RIVER, SOUTHEASTERN MINNESOTA, USA

As river catchments evolve in response to changing climate and base level, they experience varying water supplies, erosion rates, and sediment storage and release within their valley networks. Here we combine optically stimulated luminescence (OSL) and ¹⁰Be cosmogenic nuclide analyses to track the geomorphic evolution of a fill–cut terrace sequence within a river catchment and link it to paleoerosion rates from Marine Isotope Stage 3 (~30–40 ka BP) to present. To do so, we sampled two terraces cut within aggradational fill in the lower Whitewater River basin, part of the upper Mississippi watershed. The Upper Trout Creek Terrace, so named because of its position at the confluence of Trout Creek with the larger Whitewater River, lies 23 m above the modern Whitewater River floodplain. Using hand augers and a Geoprobe, we paired logarithmically spaced ¹⁰Be sampling with OSL samples taken from 1.5, 2.5, and 6 meters depth. We repeated this (with the hand auger only) on the Low Trout Creek Terrace, 8 m above the modern floodplain. Preliminary analysis of the ¹⁰Be depth profiles gives an age of ~13.5 ka for the high surface and ~9.5 ka for the low surface. These preliminary ages correlate with Mississippi River incision during Lake-Agassiz-sourced flooding (high) and the end of meltwater reoccupation of the southern Lake Agassiz outlet (low), which impacted the mainstem Mississippi and therefore base level for the Whitewater River and Trout Creek. Paleoerosion rates calculated from the ¹⁰Be inheritance were 53 mm/kyr (13.5 ka terrace) and 40 mm/kyr (9.5 ka terrace); we augmented this with a modern catchment-averaged erosion rate sample from the mouth of the Whitewater River (45 mm/kyr) and a sample from the aggradational phase (>20 ka; 56 mm/kyr). Characteristic averaging times for these erosion rates range from 13–19 kyr, meaning that the data smooth the true variability in erosion rates. Nonetheless, they suggest higher glacial stage erosion rates, perhaps due to periglacial processes. At the time of writing, single-grain OSL dating has not yet been completed, but this should provide additional geochronological constraints on surface ages and permit us to bring nuclide concentrations and calculated paleoerosion rates from the deep ¹⁰Be samples into chronological context.