Upper Midwest Environmental Sciences Center
LTRMP Reports
Summary
of Monitoring Findings for Fiscal Year 2004
Monitoring Activities and Highlights
Daily discharge data were collected at Winona, Minnesota; Keokuk, Iowa;
and St. Louis, Missouri, for the Upper Mississippi River and at Kingston
Mines, Illinois, for the Illinois River (Table).
The period of record for the four stations ranged from 65 to 127 years.
Watershed areas for the four stations vary from about 2% to 99% of the
total watershed of the Upper Mississippi River, including the Missouri
River watershed (Table). Other information
concerning the discharge database are provided in a procedures manual
(Wlosinski
et al. 1995). Mean daily discharge (m3/sec) was calculated
for each station for each water year (defined as the period from October
through the following September). Compared to historical means, mean discharge
for the 2004 water year was below normal for all of the stations except
for Keokuk which was above normal. The St. Louis hydrograph for the period
of record and for 2004 is presented in Figure
2. Water surface elevations (in feet above mean sea level) for the
LTRMP study areas are shown in Figures 3–8.
The UMRS is a dynamic ecosystem that has experienced dramatic change
over the past 150 years. Its transition from a free-flowing, untrained
river to its current impoundment by levees, locks, and dams can be documented
through the generation of digital land cover/land use data sets using
archival maps and aerial photography. Knowing where and what change has
occurred is critical to assessing habitat status and trends (Figure
9). To assist in this effort, UMESC provides detailed systemic land
cover/land use data for the 1890s, 1975, 1989, and 2000.
In 2004, base and over-target funds were used to complete the development
of the 2000 systemic land cover/land use data set. The bluff-to-bluff
2000 data set used 1:24,000-scale color-infrared aerial photography acquired
in late summer 2000 and the LTRMP 31-class vegetation classification system
to map floodplain vegetation. This data set complements the LTRMP baseline
land cover/land use developed using 1:15,000-scale aerial photography
collected in late summer 1989. All 2000 year data are referenced in Universal
Transverse Mercator Zones 15 and 16 (NAD27 and NAD83) and served on UMESC’s
Web site as Arc Export files and ArcView shapefiles (http://www.umesc.usgs.gov/data_library/land_cover_use/land_cover_use_data.html).
Water quality monitoring within the LTRMP focused on the UMRS and its major tributaries and included variables, such as physicochemical features, suspended sediment, and major plant nutrients likely to affect aquatic habitats. The sampling design combined quarterly stratified random sampling (SRS) episodes with fixed-site sampling conducted at 2- and 4-week intervals. Stratified random sampling (about 130 sites per study reach quarterly) provided unbiased, seasonal information on water quality across broad areas, such as entire navigation pools, whereas fixed-site sampling (9 to 14 sites per study reach) provided more continuous information at specific locations. The stratified random sampling was not conducted in 2003 and resumed in 2004 following the original 1993 design as modified in 2000 (a 30% reduction in sampling effort was implemented in January 2000).
The transport of sediment and major plant nutrients within the UMRS is an important management concern. Total suspended solids (TSS; mg/L) in spring random sampling episodes declined in the upper reaches (e.g., Pool 8) between 1994 and 2004 (Figure 10). Spring TSS concentrations in the lower reaches (Pool 26 and Open River Reach) varied substantially over the same period. Spring TSS concentrations peaked in 1999 in the Open River Reach when the river was above flood stage; this was followed in 2000 by lower concentrations that coincided with river stages that were about 10 feet below average. The TSS concentrations were low in 2004 in the lower reaches (approximately 50–80 mg/L), similar to 2000 (approximately 50–65 mg/L), and much lower than the high concentrations observed in 2002 (approximately 300–330 mg/L; Figure 10).
Water temperature (oC) affects a wide range of physical, chemical, and biological phenomena. Because it affects the presence or absence of species, growth rates, oxygen solubility, and chemical equilibria, even subtle changes in temperature have the potential to produce many associated changes. Median temperatures in both the lower and upper study reaches were greater during 2002 than any of the previously monitored winters, but were notably lower in 2004. In the upper reaches, the range of temperatures in 2004 was still broader than previous years (Figure 11a). In the lower reaches, 2004 temperatures were similar to 2000 and 2001 (Figure 11b).
Oxygen concentrations (percent dissolved oxygen saturation) in the waters of the Mississippi River in winter are a function of many factors such as water temperature, ice cover, primary production, and oxygen demand. Historically, winter oxygen saturation has been slightly lower in the northern study areas than in the southern study areas, probably because of greater ice and snow cover. Saturation values below 60% are not uncommon in winter in the northern study areas (Figure 12a), but are rare in the southern study areas (Figure 12b). Most years, median winter oxygen saturation in the southern study areas was typically about 100% and was often supersaturated. Since 1996, saturation values have generally increased in both the upper and lower Mississippi River (Figure 12). Median values recorded in 2002 exceeded those recorded in the previous eight winters, possibly because of less extensive areas of ice and snow cover (hence greater atmospheric exchange) and higher levels of primary productivity as evidenced by chlorophyll-a concentrations (µg/L; Figure 13). However, dissolved oxygen saturation was slightly lower in 2004 than in 2002.
Phytoplankton play a vital ecological role in the UMRS in that they provide food for filter-feeders, process nutrients, and generate (and consume) oxygen. Algal blooms, often noted in summer, also occur in winter when light penetration is excessively reduced by snow cover. During a period of little snow cover in 1995, a relatively large bloom occurred in the northern study reaches (Figure 13a) and was probably responsible for the higher oxygen saturation that also occurred at the same time (Figure 12a). The highest median winter concentrations observed by the monitoring program occurred in 1995 and similar concentrations were seen in 2002. However, 2004 median chlorophyll-a concentrations in the upper and lower reaches were much lower than was observed in 2002.
Annual changes in limnological data are strongly influenced by both long-term
and episodic changes in weather and hydrology, and limnological response
to prevailing conditions are now evident in the LTRMP data. For example,
results of recent spring sampling events demonstrate the relation between
TSS concentrations and river discharge, much the same as increases in
oxygen saturation and chlorophyll-a may reflect recent mild winters.
The importance of the LTRMP limnological record in documenting changes
and relations such as these cannot be overstated, in as much as these
data will be critical in the interpretation of associated changes in the
biological communities of the Mississippi River and other large rivers.
The break that occurred in the sampling record beginning in 2002 was ill-timed
given the extended period of drought that prevailed over some areas in
2003. There are a number of differences between 2002 and 2004 including
a strong decline in chlorophyll-a concentrations and water temperature.
The lack of 2003 complicates the interpretation of these changes. Unexpected
events (e.g., drought, floods, etc.) contribute significantly to our understanding
of UMRS functioning, but their effects will only be recorded with a commitment
to long-term, uninterrupted data collection.
Fish are an important component of the UMRS. Collectively, the river's fish fauna perform several important ecological functions. First, fish are the principal conveyers of matter and energy upstream. As terminal predators in the system, fish also tend to integrate the effects of perturbations to the system, making them excellent indicators of ecosystem integrity. In addition to responding to changes in the river system, fish themselves can affect ecological functions in the river by locally changing water quality characteristics (e.g., bioturbation) and mediating the transport of other biological life forms (e.g., freshwater mussels). Fish also support socially and economically valuable recreational and commercial fisheries within the UMRS.
Fish monitoring within the LTRMP occurs within six study areas (Pools 4, 8, 13, and 26 and an Open River Reach of the Mississippi River and La Grange Pool of the Illinois River) within the Upper Mississippi River System (Figure 1). Monitoring activities focus on single species and community status and trends using a multiple sampling gear approach (Gutreuter et al. 1995). Sampling in 2004 followed the stratified random sampling design established in 1993 (Gutreuter et al. 1995), as modified in 2002 (annual effort was reduced by about 33% through gear reductions; Ickes and Burkhardt 2002).
In 2004, the number of samples collected ranged from 218 in the Open River Reach to 360 in La Grange Pool, total annual catch ranged from 5,420 fish in the Open River Reach to 80,059 fish in La Grange Pool, and annual species counts ranged from 58 in the Open River Reach to 68 in Pool 8.
Twelve species of concern in one or more UMRS states were collected. Asian carp (Hypopthalmichthys spp.) continued to be collected in Pool 26, the Open River Reach, and La Grange Pool, but no range expansion into the upper pools was observed by the LTRMP in 2004. Nine nonnative species were observed in La Grange Pool, the LTRMP study reach with the highest number of nonnative species perennially. The round goby (Neogobius melanostomus), a species introduced into the Great Lakes a decade ago, was observed for the first time in La Grange Pool in 2004. In 2004, weed shiner (Notropis texanus) were observed for the first time since stratified random sampling began in 1993 in Pool 13 and the Open River Reach.
Aquatic vegetation in the UMRS is desired because of its many values, most notably as food for migratory waterfowl (Korschgen et al. 1988) and habitat for fish. The construction of a series of locks and dams in the 1930s created vast shallow areas where aquatic plants proliferated for nearly 3 decades before symptoms of deterioration associated with permanent impoundment became apparent (Green 1984). A widespread and sudden decline in the distribution and abundance of wild celery (Vallisneria americana) from Pools 5 through 19 in 1987–89 elevated concern as to whether or not the UMRS was on the verge of a drastic degradation in the aquatic vegetation resources (Rogers and Theiling 1999).
Long-term monitoring of aquatic vegetation was initiated in 1991 with the primary objective of determining trends in the presence and distribution of submersed and rooted floating-leaf vegetation in the UMRS (Rogers and Theiling 1999). Between 1991 and 2000, submersed aquatic vegetation was sampled along transects in selected backwaters in Pools 4, 8, 13, and 26 of the Mississippi River and La Grange Pool of the Illinois River twice a year, in spring and summer, using a standard protocol (Rogers and Owens 1995). In 1998, a stratified random sampling protocol was initiated (Yin et al. 2000) to allow for estimation of poolwide means.
Within-pool distribution patterns of submersed aquatic vegetation in the five key pools (4, 8, 13, and 26 and La Grange Pool) have remained relatively stable since 1998 when stratified random sampling was initiated (Figure 14). A few patterns are worth noting based as percent frequency of occurrence (% frequency of occurrence = # of occurrences/total # rake samples), submersed aquatic vegetation in upper Pool 4 has declined from 1998 to 2004 (21.8 to 7.1, respectively; Figure 14). Over the period of record, percent frequency of occurrence for submersed aquatic vegetation in Pool 13 was the highest recorded since stratified random sampling began in 1998 (Figure 14). Percent frequency of SAV in the impounded strata of Pool 8 has increased from 2001 (37%) to 2004 (56%; Figure 15). This may be due in part to the building of islands in lower Pool 8 that provide the vegetation protection from wind and wave action.
Macroinvertebrate monitoring by the LTRMP is intended to provide a better understanding of the conditions needed to support viable macroinvertebrate populations at levels adequate for sustaining native fish and migrating waterfowl.
Mayflies (Ephemeridae), fingernail clams (Sphaeriidae), midges (Chironomidae), the nonnative Asiatic clam (Corbicula spp.), and the exotic zebra mussel (Dreissena polymorpha) are monitored for their ecological significance in the food web or because they are recent nonnative invaders to the UMRS. Waterfowl, shorebirds, and wading birds can consume large numbers of invertebrates (Thompson 1973; Kushlan 1978; Eldridge 1988). A number of fish such as crappies (Pomoxis spp.), shovelnose sturgeon (Scaphirhynchus platorynchus), walleye (Stizostedion vitreum), bluegill (Lepomis macrochirus), freshwater drum (Aplodinotus grunniens), and yellow perch (Perca flavescens) use the target organisms (Hoopes 1960; Jude 1968; Ranthum 1969; Tyson and Knight 2001). Macroinvertebrate patterns and abundances are distinct among pools and continue to show few trends from year to year. Macroinvertebrates are contingent on a number of biotic and abiotic factors; therefore, large annual variations are expected.
Because of budget constraints, only Pools 8 and 13 were sampled in 2003; therefore, comparisons to previous year mean densities are not possible for Pools 4 and 26 and La Grange Pool. However, the poolwide estimated mean densities of mayflies, fingernail clams, and midges were all within the range of variation previously observed in all study areas over all years (Figures 16, 17, and 18).
Pool 8 recorded the highest poolwide estimated mean densities of fingernail clams and midges seen in the pool over the period of record. Pool 8 fingernail clam densities in 2004 continue the high levels seen beginning in 1999 (Figure 19). The mean density from 1992 to 1998 was 13 m-2 whereas from 1999 to 2004 was 347 m-2. This represents a 25-fold increase in density. The reason for this substantial and persistent increase is unknown; but could be because of a decrease in inorganic suspended solids (Gray et al. 2005).
This report provides only the major highlights from the Long Term Resource Monitoring Program for 2004. For more detailed information, we highly recommend the reader review the individual annual updates for 2004 or use the LTRMP Fish and Vegetation Graphical Browsers.
Page Last Modified: April 17, 2018