12 Non-Glacial Quaternary Research in the Great Lakes Region

Tuesday, 19 May 2015
1:30 PM - 4:50 PM
Hall of Ideas F

GEOCHEMISTRY AND MINERALOGY OF LOESS IN WISCONSIN INDICATES GREAT PLAINS SOURCE AREA

Paper number: 12-5
Program: Theme Sessions
Time: 2:35 PM – 2:55 PM

Authors

Abstract

Loess associated with transport and accumulation during the last glaciation occurs in several landscape regions in Wisconsin. Jacobs et al. (2011 Quat. Res. 75: 574-583) proposed that loess on the southern Green Bay Lobe land surface was sourced in the Upper Mississippi Valley region and was delivered to the deglaciated land surface by repeated entrainment across an eolian sandsheet that acted as a surface of transport. The loess eventually accumulated on topographically high landscapes above the migrating sand. Jacobs et al. demonstrated that the mineralogy of 8-63 μm silt was not useful for determining the provenance of the loess, because all possible source units had statistically similar concentration of major minerals. However, because silt and clay travel together in loess sedimentary systems, they concluded that the smectitic clay mineralogy distinctly indicated a western source for the loess. Here we present geochemistry of the silt minerals data set.

Lab methods include an HF digestion of the 8-63 μm silt fraction samples and ICP-OES analysis for trace elements and a mix of light (LREE) and heavy rare earth (HREE) elements. Elements near the detection limit or with overlapping spectra were typically excluded; the data set includes 24 elements, including 9 trace elements, 8 LREE, and 7HREE.

Results tend to clearly group the loess from the Green Bay Lobe, the Driftless Area, and the various Western source units together in nearly all plots. The glaciogenic Glacial Lake Oshkosh sediments often plot together and are commonly separated from the cluster of points for the loess and Western source units. The silt from Glacial Lake Wisconsin often groups with the loess, but in some plots is intermediate between the clusters described above.

Our results support the contention of Jacobs et al. that loess mantling the Green Bay Lobe land surface is similar to loess in the Upper Mississippi Valley region. Furthermore, we hypothesize that Upper Mississippi Valley loess deposits have a strong provenance signal from rocks in the Great Plains, in part because both Late Wisconsin and Pre-Illinoian ice sheet lobes carried eroded material from the eastern Great Plains into the Upper Mississippi basin, but also possibly because of far-traveled dust transport from the Plains.