Evidence for bioturbation of luminescence signals in eolian sand on upland ridgetops, southeastern Minnesota, USA

This study sought to identify the age of eolian sand on narrow ridges in the Root River valley of southeastern Minnesota. The ridges in the region are cored by Ordovician dolomite and Cambrian sandstone, but the ridge tops are typically covered with Late Wisconsin Peoria and older loess units. In some locations Peoria loess is absent and the ridgetops are covered with up to 3 m of eolian sand that was likely sourced from local river valleys and transported to the uplands by sand ramps. We studied nine ridgetop soil profiles and collected seventeen OSL samples from eolian sand at depths ranging from 0.3 to 2.6 m below the present ground surface. All OSL samples were analysed using small aliquots where 90–150 μm quartz grains were applied to the inner 2 mm of 10 mm aluminum disks. The OSL ages ranged from 12.3 to 1.5 ka indicating a significant amount of age variability, and potentially suggesting nearly continual eolian deposition throughout the Holocene. However, several key differences were identified between those samples taken from within 1 m of the ground surface compared with samples that were more deeply buried. Those samples collected from depths of greater than 1 m yielded ages that were tightly clustered between 12.3 and 10.3 ka, while samples taken from depths of less than 1 m showed ages with much higher spread that ranged from 1.5 to 10.1 ka. The samples collected from within 1 m of the present ground surface also commonly showed higher spread in their equivalent dose distributions and higher overdispersion values relative to the samples that were more deeply buried. This suggests the luminescence signals from the upper portions of these deposits were reset after burial, most likely by bioturbation, and that the OSL ages are not depositional ages. This interpretation is supported by evidence from a core sample collected from one of our sites that shows primary eolian lamina are preserved below 1.4 m depth but not above this depth. Presumably, the bioturbation agents were effective at both resetting the luminescence signal and disturbing the primary bedding to depths of at least 1 m. If bioturbation occurred below this threshold it apparently did not impact either of these indicators. Our findings suggest the upper 1 m of these profiles were impacted by bioturbation and that all of the eolian sand on these ridgetops was likely deposited prior to 10.1 ka.