Home


Habitat Changes in the Upper Mississippi River Floodplain


by
Joseph Wlosinski
National Biological Service
Douglas A. Olsen
Carol Lowenberg
Thomas W. Owens
Jim Rogala
Mark Laustrup
National Biological Service
The U.S. Congress recognized the Upper Mississippi River (UMR) as a nationally significant ecosystem in 1986. The UMR extends northward from the confluence of the Mississippi and Ohio rivers to the Twin Cities, Minnesota, a distance of more than 1,360 km (850 mi). The floodplain (area between the bluffs) of the UMR includes 854,000 ha (2,110,000 acres) of land and water. The Mississippi River is a major migration corridor for waterfowl and provides habitat for more than 150 fish and 40 freshwater mussel species.
Fig. 1. The Upper Mississippi River. Numbers indicate reservoirs formed by navigation dams and known locally as pools.
Since 1824 the federal government has implemented numerous changes on the UMR. The river was first modified by removing snags and then sandbars, with changes progressing to rock excavation, elimination of rapids, closing of side channels, and the construction of hundreds of wing dams, 27 navigation dams, and hundreds of kilometers of levees. Reservoirs formed by the navigation dams are known locally as pools (Fig. 1), which are numbered from north to south. Construction of the dams (mostly during the 1930's) significantly altered the northern 1,040 km (650 m) of the UMR (north of St. Louis, Missouri) by increasing the amount of open water and marsh areas. Wing dams and levees have altered aquatic habitats south of St. Louis (the open river) by reducing open-water habitats and isolating the river from much of the floodplain. Most of the changes to the river ecosystem were either designed for navigational improvements or to control the movement of river water. Here we investigate some of the habitat changes at various levels of resolution.
Spatial data were analyzed by using a geographic information system (GIS). Floodplain areas (bluff to bluff) and systemic land-cover/land-use data were obtained from Landsat Thematic Mapper data collected in 1989. Land-cover/land-use data from 1891 were created from ground surveys conducted by the Mississippi River Commission. High-resolution land-cover/land-use data were created from 1:15,000 (scale) color infrared aerial photographs taken in 1989. Data for 1891 and 1989 were compared for Pools 4, 5, 8, 13, 26, and for a 64-km (40-mi) stretch of river, near Cape Girardeau, Missouri, which is not affected by navigation dams. Historical aerial photographs from 1939, 1954, 1967, and 1989 were used to measure island loss in an area just upriver of the dam in Pool 8.
Long-term daily data at three stations on the open-river portion of the UMR were analyzed to evaluate changes in the relationship between discharges and water-surface elevations.

Status and Trends
Comparison of the land-cover/land-use data between 1891 and 1989 in the dammed portion of the UMR showed that open water and marsh habitats generally increased, mostly at the expense of grass/forb, woody terrestrial, and agricultural classes. For example, the combined classes of open water and marsh in Pool 8 have increased from 3,600 ha (8,900 acres) in 1891 to 9,500 ha (23,430 acres) in 1989 (Figs. 2a, b). Similar increases in these two classes were found at Pools 5 and 13. In Pools 4 and 26 increases were less significant.

Fig. 2. Land-cover/land-use comparisons for a portion of Pool 8 in (a) 1891 and (b) 1989, and the open river near Cape Girardeau, Missouri, in (c) 1891 and (d) 1989.
In many pools inundation created an impounded area with a mosaic of islands, open water, and marsh, which, in general, increased aquatic habitat for fish and wildlife. Although dam construction has benefited aquatic habitat in many pools, the reservoir aging process has reduced these benefits, especially in areas just upriver of dams. For example, island areas have been steadily eroding upriver of the dam in Pool 8 (Fig. 3). The dam that forms Pool 8 began operating in 1937, and photographs taken 2 years later showed 253 ha (624 acres) of islands. By 1989 the island area in the same location was reduced by 79% to 52 ha (129 acres).

Fig. 3. Island loss that has occurred in Pool 8, in the area just upriver of the dam, since construction of the lock and dam system.
Sedimentation is also a major concern on the UMR; rates of 1 to 3 cm/yr (0.4-1.2 in/yr) have been measured (McHenry et al. 1984). Erosion and sedimentation were both detected in comparisons between present elevation data and surveys before dam construction. Erosion was more prevalent in shallow areas and sedimentation more prevalent at greater depths. Erosion and sedimentation converge at depths of between 0.9 and 1.5 m (3 to 5 ft). This has resulted in a more homogeneous distribution of depth, which is dominated by areas 0.9 to 1.5 m (3 to 5 ft) in depth. Similar frequency distributions of water depth were observed for lower portions of Pools 8 and 13. Comparison of historical and present bottom geometry revealed the loss of elevational diversity.
In areas of the UMR unaffected by navigation dams (the 40-mi stretch of river near Cape Girardeau), there was a 28% reduction in open water and a 38% reduction in woody and terrestrial habitat between 1891 and 1989 (Figs. 2c, d). Agricultural areas increased by 6,360 ha (15,700 acres). The 1,900-ha (4,710-acre) reduction of open water can be explained by the construction of levees and wing dams (also known as pile dikes). One large side channel that existed in 1891 was cut off by construction of a levee, reducing the area of water by 550 ha (1,350 acres). In all, nearly 2,000 km (1,240 mi) of levees now isolate more than 400,000 ha (988,000 acres) from the river during all but the highest discharge rates.
Wing dams and levees, along with other changes to the watershed, have also had a major effect on habitats by changing the relationship between discharge and water-surface elevations. Wing dams have narrowed and deepened the main channel so that water elevations at low discharges are now lower than they were historically. Levees restrict flows and result in higher water elevations during high discharges. Water-surface elevations at relatively low discharges (60,000 cfs) have dropped about 2.4 m (8.0 ft) over the record 133-year period at St. Louis, Missouri, 0.5 m (1.5 ft) over the 52-year record at Chester, Illinois, and 1.5 m (5.0 ft) over the 60-year record at Thebes, Illinois. Water-surface elevations at relatively high discharges (780,000 cfs), however, have risen about 2.7 m (9 ft) over the record period at St. Louis, 1.5 m (5.0 ft) at Chester, and 1.1 m (3.6 ft) at Thebes.
For further information:
Joseph Wlosinski
National Biological Service
Environmental Management Technical Center
575 Lester Ave.
Onalaska, WI 54650
References
McHenry, J.R., J.C. Ritchie, C.M. Cooper, and J. Verdon. 1984. Recent rates of sedimentation in the Upper Mississippi River. Pages 99-118 in J.G. Wiener, R.V. Anderson, and D.R. McConville, eds. Contaminants in the Upper Mississippi River. Butterworth Publishers, Boston, MA.


Home