Excess sediment affects many streams and rivers throughout the world and has a negative impact on the ecological health of surface waters. The unifying objective of this dissertation is to better understand sediment dynamics and how sediment is measured in large river systems, particularly where excess sediment is related to anthropogenic landscape modifications. Lake Pepin, the principal focus of this study, is a natural riverine impoundment that traps sediment eroding from three major sub-watersheds of the upper Mississippi River (UMR), the St. Croix, headwater Mississippi, and Minnesota rivers. Sediment accumulation in Lake Pepin has increased nearly 10-fold since the onset of Euro-American settlement c. 1830. This dissertation explores: (1) the geological evolution of the UMR and Lake Pepin, (2) the most recent sediment-accumulation trends in Lake Pepin, (3) the deconstruction of magnetic susceptibility profiles in a Lake Pepin core, and (4) an evaluation of direct atmospheric deposition of <super>210</super>Pb to the river surface and its consequences for geochemical fingerprinting. The four study elements are summarized as follows: (Chapter 1) Lake Pepin currently occupies 34 km of the Mississippi River valley, but it extended nearly 81 km further up-river at its inception approximately 10,000 years ago. Sediment fans deposited by several tributaries (the Chippewa, Cannon, and Vermillion rivers) entering the valley segmented this early Lake Pepin into several smaller lake sections. Later, accretion of the Chippewa fan surpassed that of the tributary fans to create a single large lake. Since that time, the length of Lake Pepin has decreased progressively as its delta prograded downstream from present-day St. Paul. Based on geomorphic evidence, the delta is estimated to have migrated past Hastings c. 6.0 ka and Red Wing c. 1.4 ka.
(Chapter 2) A new method was developed to measure recent sediment accumulation rates in Lake Pepin by aligning new cores to a previous set of cores with an established chronology. This method relies on prominent magnetic susceptibility peaks ubiquitous throughout the lake to align the new cores with the older core set. The amount of recent (1996-2008) bulk sediment, total phosphorus, and mercury at each core site was extrapolated to the depositional area of the lake to estimate whole-basin loads. These core-based estimates of recent sediment loads compared well (<3%) with load estimates from monitored inflow data, validating the alignment procedure and multiple-core approach. Recent (1996-2008) bulk sediment accumulation remains high (772,000 t/yr), and almost an order of magnitude greater than pre-settlement accumulation rates. Total phosphorus deposition remains constant and mercury continues to decline following peaks in the 1960s. (Chapter 3) Three techniques for measuring magnetic susceptibility were compared using a single sediment core from Lake Pepin: (1) loop-sensor logging on wet sediment of the intact core; (2) point-sensor logging on wet sediment of a split (lengthwise) core; and (3) discrete measurements of dried subsamples (2-cm intervals) using a susceptibility bridge. Overall trends agree reasonably well among the three techniques. Profiles from the loop-sensor and susceptibility bridge were most similar, while the point sensor profile provided higher down-core temporal resolution. All three techniques captured two distinct magnetic susceptibility excursions c. 1900 and 1940. Concentrations of ferrimagnetic components were modeled to further explore the magnetic sources of these susceptibility peaks. Model results indicate that the two prominent excursions were a result of higher superparamagnetic particle concentrations (<30 nm). A later peak (1980) was present in the point-sensor and susceptibility bridge logs, but was not obvious in the loop-sensor log. This 1980 peak corresponded to higher concentrations of interacting single domain (<0.1 μm) and multi-domain (>1-10 μm) ferrimagnetic particles, but not superparamagnetic particles, as did the two earlier excursions. These results highlight the likely causes for discrepancies among magnetic susceptibility techniques on sediment cores. (Chapter 4) Geochemical fingerprinting using atmospheric radioisotopes (ARI) such as <super>210</super>Pb is an important tool for quantifying riverine sediment sources. Although sediment fingerprinting methods are well established, they rely on the assumption that direct deposition of ARI to the river surface is negligible relative to that derived indirectly from watershed erosion. In this study a strategy was developed for quantifying the amount of <super>210</super>Pb deposited directly to the river surface (i.e., <super>210</super>Pb not entering on eroded particles) by exploiting the initial disequilibrium between newly deposited <super>210</super>Pb and its granddaughter <super>210</super>Po. Direct <super>210</super>Pb deposition (<italic>Pb<sub>w</sub></italic>) was estimated by modeling <super>210</super>Po ingrowth based on repeated measurements of the same sample over the course of a year. A significant amount of <italic>Pb<sub>w</sub></italic> was detected at each site, ranging from 20-90% of the total unsupported annual <super>210</super>Pb load, and generally increasing with watershed size. This method provides an important correction to atmospheric radioisotope fingerprinting, particularly in larger watersheds. This dissertation offers several key findings that help to better understand sediment movement in the upper Mississippi River system and provide innovative and novel approaches to more accurately measure sediment dynamics in general. There is now a new understanding of the early history of Lake Pepin and the Mississippi River valley it occupies. There is also a new method for determining recent trends in sediment and contaminant accumulation by building on previously analyzed cores, and results show that despite efforts to remediate high sediment loads in the UMR, Lake Pepin is still filling in at an alarming rate. There is now a better understanding of the ubiquitous magnetic susceptibility peaks in Lake Pepin cores that have been important stratigraphic markers in prior studies. It is now understood the two most prominent peaks (1900 and 1940) are associated with a greater influx of superparamagnetic particles, and the most recent (1980) peak is heavily influenced by greater concentrations of interacting single domain and multi-domain ferrimagnetic particles. Finally, efforts to determine sediment sources with atmospheric radioisotopes have neglected the influence of isotopes deposited directly to the river surface, largely because there was not a viable method to estimate this component. There is now a new method to measure atmospherically deposited <super>210</super>Pb, which provides an important correction for these types of studies.
University of Minnesota Ph.D. dissertation. August 2013. Major: Geology. Advisors: computer file (PDF); vii, 126 pages. Daniel R. Engstrom and Herbert E. Wright, Jr. 1 computer file (PDF); ix, 112 pages.
Reconstructing the erosional history of the Upper Mississippi River from magnetic, isotopic, and geomorphic evidence.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.