RELIEF PRODUCED BY LOESS DEPOSITION IN THE CENTRAL GREAT PLAINS: A "NATURAL EXPERIMENT" IN GEOMORPHOLOGY

Relief production drives landscape evolution, both in classical conceptual models and in more recent numerical studies. In this study, I took advantage of an unusual form of relief production, to evaluate the relationships between local relief and drainage density predicted by process-based models. In parts of the central Great Plains, rapid deposition of thick late Pleistocene Peoria Loess increased local relief by up to 50 m, because loess accumulated on gently sloping upland summits but not on the active channel belts of larger streams. Across a study area of 77,500 km2, the systematic downwind decrease in Peoria Loess thickness explains much of the variance in maximum local relief (R2 = 0.73). This results in a regional gradient of maximum local relief, with greatest relief where the loess is thickest. Across much of this gradient, slopes are underlain primarily by Peoria or older loess units; Hortonian overland flow is common in this subhumid to semi-arid landscape, although piping may also be significant.

I investigated the response of drainage density to this spatial pattern of local relief, using standard deviation of plan curvature (SDPC) as a readily computed proxy for drainage density (SDPC is correlated with drainage density in 15 small test areas; R2 = 0.87). SDPC increases systematically with local relief in the study area, and high SDPC is associated with the belt of thickest Peoria Loess. This is consistent with a shear stress threshold for channel formation by Hortonian overland flow (Montgomery and Dietrich, 1991; 1992), and more broadly with dependence of channel erosion on local slope and drainage area: Where relief is greatest, and slopes are generally steepest, less drainage area is required for channel formation and the landscape is more dissected. A semi-arid bedrock-controlled landscape adjacent to the study area displayed a very similar relationship between SDPC and relief, but with an offset consistent with greater resistance to channel erosion. Within the limited climatic range of the study area (mean annual precipitation is 45-75 cm), the relationship between relief and drainage density appears to be stronger than any effects of climate.