Upper Mississippi River Restoration Program Long Term Resource Monitoring

Introduction

Sedimentation is a major concern of resource managers of the Upper Mississippi River (UMR). Land use changes (i.e., agricultural practices and urban development) in the basin have contributed to high sediment delivery rates. The problem has been exacerbated by river projects for navigation and flood control (i.e., dams, training structures, and levees) that have altered the natural sediment transport process within the river. Even in a natural state, the river’s complex floodplain was conducive to deposition of sediments in transport. Accumulation of sediment may result in loss of backwater volume and encroachment of terrestrial vegetation into formerly aquatic areas. Such accumulation will largely determine the fate of off-channel areas in the UMR (Fremling and Claflin 1984; Nielsen et al. 1984), and the fate of the biota that rely on those habitat conditions. Studies to determine sedimentation rates and processes were among the recommendations developed by the Sediment Transport and Geomorphology Work Group convened to define informational needs for the UMR (Gaugush and Wilcox 2002).

The UMR extends over 1,365 km from Minneapolis-St. Paul, Minnesota, to Cairo, Illinois, with floodplain complexity that differs greatly longitudinally. The 523-km reach of the UMR from Pool 2 to Pool 13 is particularly complex, having a mosaic of channel and off-channel environments that are still in the active floodplain. Off-channel areas, generally grouped into impounded areas and backwater lakes, are plentiful in this reach. Impounded areas are large expanses of open water formed just upstream of the dam and typically have flow even at low river discharge. In contrast, backwater lakes have little or no flow during low river discharge periods, thus, periodically providing a lentic environment in a lotic system. Most backwater lakes are typically shallow (<1 m) during low discharge conditions. The degradation of these highly productive backwater lakes because of sedimentation has been a long-standing issue for river managers (GREAT 1980).

Using a variety of methods, several researchers have estimated sedimentation rates in off-channel areas of the UMR. Using Cesium-137 sediment dating techniques to estimate net sedimentation in mostly impounded areas, McHenry et al. (1984) found rates between 1954 and 1975 ranging from 1 to 4 cmּyr^­1 in Pools 4 through 10. Korschgen et al. (1987) reported lower rates (0.2 cmּyr^­1) for a period of 50 years in an impounded backwater area in Pool 7 using bathymetric maps. In a large backwater lake in Pool 9, an annual sedimentation rate of 1.69 cmּyr^­1 was reported between 1964 and 1974 by Eckblad et al. (1977) using the Cesium-137 dating technique. Using deep coring to parent material, Rogala et al. (1997) observed sedimentation rates for the 58-yr period since impoundment ranged from 0 to 1.5 cmּyr^­1 in backwater lakes of Pool 8. For a 6-yr period ending in 1996, Rogala and Boma (1996) reported average rates of 0.29, 0.12, and 0.80 cmּyr^­1 in Pools 4, 8, and 13, respectively, from bed elevation surveys conducted along a nonrandom set of transects.

There are several shortcomings regarding these past studies, particularly for predicting future conditions of backwater lakes in the UMR. First, many of the studies were conducted in impounded areas, where sedimentation rates may differ greatly from rates in backwater lakes. Second, most of these studies probably overestimated sedimentation rates by sampling only in areas known to have accumulated sediments, or provided an opportunity for bias by selecting sample sites nonrandomly. Third, most of these studies did not provide estimates of sedimentation for a recent period. The historical sedimentation rates may be inappropriate for predicting future conditions of the river because impounded systems are expected to trap less sediment as the system ages (McHenry et al. 1984; Church 1995). Finally, past studies provided little information on spatial variability in sedimentation among and within pools, which is critical for predicting future conditions.

To address the shortcomings of these earlier studies, we redesigned the short-term monitoring study conducted by Rogala and Boma (1996). One objective of our study was to compare sedimentation rates we obtained for a present 5-yr period with past estimates. A second objective was to determine variability in rates along transects and among years.  Finally, we examined spatial and temporal variability to identify variables of interest for future predictive modeling efforts. Overall, we addressed spatial and temporal variability in sedimentation rates at finer scales than has previously been investigated.