Freshwater mussels in the Clinton River, southeastern Michigan: an assessment of community status.
Freshwater mussels (unionids) constitute an important and diverse element of freshwater ecosystems in the United States (Cummings and Mayer 1992). No other region of the world equals North America in unionid diversity. While all of Europe supports only 12 species, the nearly 300 species in the U.S. live primarily within the Mississippi River drainage and to the east (Stein et al. 2000; Graf and Cummings 2007; Bogan 2008). More than 70% of North American unionids are categorized as endangered, threatened, or of special concern, a percentage that leads all other species of freshwater-dependent animals. This includes more than one in 10 freshwater mussel species that have become extinct (Stein et al. 2000).
The Clinton River drains most of Macomb County, the eastern half of Oakland County, and small areas of southern Lapeer and St. Clair counties, eventually ending in Lake St. Clair. The Clinton River watershed covers 760 square miles and includes over 1,000 miles of streams in addition to the 80-mile-long main branch (Zorn and Seelbach 1992). Along the Clinton River there are several lakes that keep the streams unusually clear, and fluctuations in temperatures, water chemistry, and current speed are generally very small (Strayer 1983).
The watershed's geology and certain components of the biota are discussed in earlier studies of the river and these are still largely applicable to the Clinton River today (Strayer 1980; Zorn and Seelbach 1992; USACE 2005; Sinha and Creech 2006; Francis and Haas 2006). In terms of the mussel fauna, the river was unusually species-rich historically, and Strayer considered the original 31 species to be exceptional for a stream in the Great Lakes drainage (Strayer 1980). In that report he indicated that in urban areas, "Man's activities have had a devastating effect" whereas in other more rural areas there were still relatively healthy mussel communities in terms of density and diversity (Strayer 1980). Strayer concluded that "the factor determining the present state of mussel fauna is the degree to which urban pollution has affected a stream stretch" (Strayer 1980, p. 148). His survey revealed a loss of species from 31 to 26 and a loss of rare and endangered species from seven to four in 1978 with the caveat that his methods were not designed to discover rare species. Well after Strayer's survey, two exotic bivalves, the zebra mussel (Dreissena polymorpha) and the Asian clam (Corbicula fluminea), were discovered in the watershed in the early 1990s (Janech and Hunter 1995; Hunter 1996; Hunter et al. 1997).
Due to the regulation of virtually all point source discharges, water quality in the lower Clinton River has improved over the last 30 years (Francis and Haas 2006). Industries now discharge into municipal sewers and are required to meet NPDES (National Pollution Discharge Elimination System) effluent limits. While live fish couldn't be found from Pontiac to the mouth of the Clinton in the 1960's, a large, varied fishery (dependent on stocking) exists today consisting of 100 species as of 2005 (Zorn and Seelbach 1992; Francis and Haas 2006).
As indicated by Strayer, the tributaries of the St. Clair corridor and western Lake Erie had a rich freshwater mussel community that was affected by water quality problems arising from human activities in this rapidly growing area (Strayer 1980). Given the available historical data, we have a basis of comparison for a re-examination of the mussel community of the Clinton River. This new survey allows us to determine what, if any, further changes have occurred and to make inferences about the potential causes given the continuing development of this region over the past 50 years that has included some pollution abatement as well as continuing geomorphologic instability in the watershed.
Seventy-six sites located in Oakland, Macomb, and St. Clair Counties were sampled in May through September of 2004. These sites were distributed in seven separate regions of the Clinton watershed corresponding to major tributaries (Figure 1). Overall density of unionids was determined from the estimated area of river bottom that was sampled (20,982 [m.sup.2]) and this was expressed as number of individuals/[m.sup.2] The area of river bottom searched was measured using a 100 m Keeson fiberglass tape. Water temperature was recorded and characteristics of each site (clarity of water, relative current speed, and river bottom type) were noted as well as the man-hours spent at each location.
[FIGURE 1 OMITTED]
Area searched and man-hours spent were matched as closely as possible to the time and area searched by David Strayer in the original 1978 survey (Strayer 1980). In most cases, time and area were identified in Strayer's field notes, but in a few cases these was no mention of either of these. In the latter cases we made an estimate of what was likely based on site characteristics and rate of return of specimens found. On average, searching the bottom was entirely visual and lasted for about 30 min involving a minimum of two people (= one person-hour). The authors are aware that this cursory level of sampling undoubtedly underestimates density and misses some rare species; however the highest priority was to duplicate Strayer's methods in order to maintain comparability of results (Metcalfe-Smith et al. 2000).
Mussels were collected at each location using glass-bottomed buckets where necessary. Net handles were used to probe the bottom and nets were used to lift the unionids. Identification was done using standard keys and manuals (Cummings and Mayer 1992; Burch 1975; Strayer and Jirka 1997; Mathiak 1979; Clarke 1981). External features were used to identify live specimens whereas both external and internal shell characteristics were utilized for spent shells. Spent shells (shells of dead individuals) were also identified to species. Following identification, live unionids were reinserted unharmed into the river bottom. Permits for the handling of unionids, both endangered and non-endangered species, were obtained from the Wildlife Division of the Michigan DNR.
Density. Table 1 gives overall density and relative density by species in the Clinton River in 1978 and 2004, ranked by overall density in 1978. Raw data for density, with site location and number collected by species at each site, are available at: http://hdl.handle.net/10323/163.
Overall density was calculated by dividing total live individuals found of each species by the total area searched in the entire study (= 20,982 [m.sup.2] Relative density is overall density of a species as a percentage of the sum of the overall densities of all species. Fourteen species of live freshwater mussels were found in 2004. Table 1 also includes six additional species considered likely to still be present based on recent sightings by others (mostly D. Sweet and RDH), three species for which spent shells were all that was found in 2004, and four species that were not observed in 2004 at any of the 76 sites. In 1978, density ranged from 0.0004/[m.sup.2] for Potamilis alatus and Quadrula pustulosa to 0.01902/[m.sup.2] for Elliptio dilatata. In 2004, unionid density ranged from 0.0005/[m.sup.2] for Amblema plicata and Lampsilis ovata to 0.00658/[m.sup.2] for E. dilatata. Hence, E. dilatata was the most abundant species in both survey years. In the case of Toxolasma parvus (proposed Michigan endangered) it is likely to still be present at very low density in a tributary of Plum Brook at Joseph Delia Park near site 53. Joseph Delia Park is a site not previously sampled by Strayer or by the authors; it was independently surveyed by D. Sweet in 2002. Likewise, Pleurobema sintoxia and Alasmidonta viridis were observed by D. Sweet at Wolcott Mills Metropark (site 28; D. Sweet, pers. comm.).
TABLE 1. Overall density (X1000) and relative density of freshwater mussels in the Clinton River (SE Michigan) in 1978 and 2004 based on living mussel data only. Total area surveyed = 20,982 [m.sup.2]. Relative density is overall density of a species as a percentage of the sum of the overall densities of all species. Data for 1978 are based on Strayer (1980). Raw data for density, with site location and number collected by species at each site, are available at http://hdl.handle.net/10323/163. 1978 2004 Number of Density Density live No./[m.sup.2] No./[m.sup.2] individuals Species X 1,000 X 1,000 2004 Elliptic dilatata (1) 19.02 6.58 138 Strophitus undulatus (1) 8.64 1.05 22 Pyganodon grandis (1) 8.33 0.62 13 Anodontoides 7.40 1.05 22 ferrussacianus (1) Lasmigona complanata (1) 6.74 0.10 2 Ptychobranchus 6.55 1.53 32 fasciolaris (1) Lasmigona compressa (1) 5.15 0.38 8 Pleurobema sintoxia (2) 4-45 0.00 0 Lampsi siliquoidea (2) 3.60 0.38 8 Actinonaias ligamentina (4) 2.87 0.00 0 Villosa iris (1) 2.87 1.67 35 Amblema plicata (1) 2.29 0.05 1 Utterbackia imbecillis (2) 2.21 0.00 0 Fusconaia flava (1) 2.01 3.00 63 Alasmidonta viridis (2) 1.36 0.00 0 Lasmigona costata (1) 1.08 0.00 0 Toxolasma lividus (2) 0.93 0.00 0 Lampsilis ovata (1) 0.58 0.05 1 Epioblasma triquetra (1) 0.46 0.29 6 Lampsilis fasciola (1) 0.31 0.10 2 Alasmidonta marginata (3) 0.27 0.00 0 Quadrula quadrula (4) 0.27 0.00 0 Potamilis alatus (4) 0.04 0.00 0 Quadrual pustulosa (4) 0.04 0.00 0 Toxolasma parvus (2) 0.00 0.00 0.00 Ligumia nasuta (3) 0.00 0.00 0 Villosa fabalis (2) 0.00 0.00 0 Totals 87.47 16.83 353 1978 Relative Relative Density 2004 (%) Species Density (%) Elliptic dilatata (1) 21.74 39.09 Strophitus undulatus (1) 9.88 6.23 Pyganodon grandis (1) 9.52 3.68 Anodontoides ferrussacianus (1) 8.46 6.23 Lasmigona complanata (1) 7.71 0.57 Ptychobranchus fasciolaris (1) 7.48 9.07 Lasmigona compressa (1) 5.89 2.27 Pleurobema sintoxia (2) 5.09 0.00 Lampsi siliquoidea (2) 4.12 2.27 Actinonaias ligamentina (4) 3.28 0.00 Villosa iris (1) 3.28 9.92 Amblema plicata (1) 2.61 0.28 Utterbackia imbecillis (2) 2.52 0.00 Fusconaia flava (1) 2.30 17.85 Alasmidonta viridis (2) 1.55 0.00 Lasmigona costata (1) 1.24 0.00 Toxolasma lividus (2) 1.06 0.00 Lampsilis ovata (1) 0.66 0.28 Epioblasma triquetra (1) 0.53 1.70 Lampsilis fasciola (1) 0.35 0.57 Alasmidonta marginata (3) 0.31 0.00 Quadrula quadrula (4) 0.31 0.00 Potamilis alatus (4) 0.04 0.00 Quadrual pustulosa (4) 0.04 0.00 Toxolasma parvus (2) 0.00 0.00 Ligumia nasuta (3) 0.00 0.00 Villosa fabalis (2) 0.00 0.00 Totals (1.) Observed live in 2004. (2.) Not observed live in 2004 but likely to exist based on recent sightings by D. Sweet or R. D. Hunter. (3.) Observed in 2004 as spent shells only. (4.) Not observed in 2004 either live or as spent shells.
Villosa fabalis was observed by one of the authors (RDH) in 1995 and 1996 at site number 11 (Cooley Lk. Rd. Bridge). Like T parvus, it is a small unionid and is very difficult to locate by the methods we used hence its density is generally underestimated. It is likely to still be present at site 11 so it is listed in Table 1.
Overall, relative density has decreased for every species except Elliptio dilatata, Epioblasma triquetra, Fusconaia flava and Villosa iris. In 1978, E. dilatata had the greatest relative density of all species found (21.74%) while Potamilis alatus and Quadrula pustulosa had the least (0.04%). In 2004 E. dilatata also had the greatest relative density (39.09%) while Amblema plicata and Lampsilis ovata had the least (0.28%). Ten species reported in 1978 were not found live in 2004: Actinonaias ligamentina, Alasmidonta viridis, Alasmidonta marginata, Lasmigona costata, Pleurobema sintoxia, Potamilis alatus, Quadrula pustulosa, Quadruda quadrula, Toxolasma lividus and Utterbackia imbecillis. Spent shells were found for L. costata, P. sintoxia, T. lividus and U. imbecillis in 2004. The continuing presence of T lividus (Michigan endangered) is supported by studies in 1995 and 1997 in which that species was observed over three consecutive years at site 9 (RDH). This finding is supported by the discovery of T. lividus shells at site 9 in 2004; however, disturbances at this site make the population's future doubtful.
A multiple analysis of variance (MANOVA; [alpha] = 0.05) was performed on the mean site density data by region that was square root transformed after adding 0.0001 to all values. Univariate comparisons indicated significant regional differences between 1978 and 2004 (Figure 2). At least one of the six regions was different from at least one other region. There was no significant interaction between year and region. Lacking a significant interaction, an ANOVA was performed on all regions, with each region treated as a normal variable. A comparison of the first six regions in 1978 and the last six in 2004 indicated significant differences in mean site densities (Figure 2; F = 3.954; P = 0.0001). A Tukey/Kramer multiple comparison test ([alpha] = 0.05) indicated that region 5 was significantly different in 1978 from regions 1 and 2, and region 6 was significantly different from all other regions in both 1978 and 2004.
[FIGURE 2 OMITTED]
When changes in region density are viewed on a percentage basis, the total density for live mussels has declined the most in the middle branch (100%) where nine species were found in 1978 but no live mussels were found in 2004. The remaining percentage change in density ranged from-70.2% in the upper mainstem to-93.7% at Coon Creek.
Frequency. Similarly, the overall frequency has also declined in nearly all species from 1978 to 2004 (Table 2). Frequency is defined here as the number of sites, for both live and spent shells, where each species was found. In 1978, the species that occurred at the most sites were Pyganodon grandis (29), Strophitus undulatus (27), Lasmigona compressa (26) and Villosa iris (26). In 2004, the species that occurred at the most sites were V. iris (21), Anodontoides ferrussacianus (18) and P. grandis (17).
Two species showed a small increase in frequency from 1978 to 2004: Fusconaia flava (from 12 to 14) and Ligumia nasuta (from 0 to 1). All the remaining 21 species showed a decline in frequency from 1978 to 2004. Most of these were relatively rare in 1978 (i.e. found at one site only). Perhaps most noteworthy are those species that have undergone a drastic reduction in frequency. These include Alasmidonta viridis (23 to 0) and Pleurobema sintoxia (20 to 2). Two other species went from modest frequency to zero: Actinonaias ligamentina and Alasmidonta marginata.
TABLE 2. Species frequency (number of sites out of 76 total where each species was found) for living plus spent shell data in the Clinton River (SE Michigan) in 1978 and 2004. Rare species status is indicated by superscripts based on data from the Michigan Natural Features Inventory (http://web4.msue.msu.edu/mnfi/). Data for 1978 are based on Strayer (1980). 1978 % of Scientific Name Common Name Frequency Total Pyganodon grandis Giant floater 29 38.16 Strophitus undulatus Squawfoot 27 35.53 Lasmigona compressa Creek heclsplitfer 26 34.21 Villosa iris (3) Rainbow 26 34.21 Anodontoides ferrussacianus Cylindrical papershell 24 31.58 Elliptic dilatata Spike 24 31.58 Alasmidonta viridis (3) Slippershell mussel 23 30.26 Pleurobema sinioxia (3) Round pigtoe 20 26.32 Lampsilis siliquoidea Fatmucket 19 25.00 Lasmigona complanaia While heelsplitter 14 18.42 Fusconaia flava Wabash pigtoe 12 15.79 Ptychobranchus fasciolaris Kidneyshell 11 14.47 Actinonaias Ugatnen tina Mucket 7 9.21 Lampsilis fascioki (2) Wavy-rayed lamp mussel 7 9.21 Lampsilis ovata Pocketbook 7 9.21 Alasmidonta marginata (3) Elktoe 6 7.89 Amblema plicata Three-ridge 6 7.89 Epioblasma iriquetra (1) Snuffbox 6 7.89 Utterbackia imbecillis Paper pondshell 6 7.89 Lasmigona costata Fluted shell 5 6.58 Toxolasma lividus (1) Purple lilliput 2 2.63 Potamilis alatus Pink heelsplitter 1 1.32 Quadrula pustulosa Pimpleback 1 1.32 Quadrula quadrula Mapleleaf 1 1.32 Toxolasma parvus Lilliput 1 1.32 Villosa fabalis (1) Rayed bean 1 1.32 Ligumia nasuta Eastern pondmussel 0 0.00 Total 312 2004 % of Scientific Name Common Name Frequency Total Pyganodon grandis Giant floater 17 22.37 Strophitus undulatus Squawfoot 14 18.42 Lasmigona compressa Creek heclsplitfer 11 14.47 Villosa iris (3) Rainbow 21 27.63 Anodontoides ferrussacianus Cylindrical papershell 18 23.68 Elliptic dilatata Spike 15 19.74 Alasmidonta viridis (3) Slippershell mussel 0 0.00 Pleurobema sinioxia (3) Round pigtoe 2 2.63 Lampsilis siliquoidea Fatmucket 10 13.16 Lasmigona complanaia While heelsplitter 4 5.26 Fusconaia flava Wabash pigtoe 14 18.42 Ptychobranchus fasciolaris Kidneyshell 6 7.89 Actinonaias Ugatnen tina Mucket 0 0.00 Lampsilis fascioki (2) Wavy-rayed lamp mussel 3 3.95 Lampsilis ovata Pocketbook 2 2.63 Alasmidonta marginata (3) Elktoe 0 0.00 Amblema plicata Three-ridge 2 2.63 Epioblasma iriquetra (1) Snuffbox 6 7.89 Utterbackia imbecillis Paper pondshell 1 1.32 Lasmigona costata Fluted shell 1 1.32 Toxolasma lividus (1) Purple lilliput 2 2.63 Potamilis alatus Pink heelsplitter 0 0.00 Quadrula pustulosa Pimpleback 0 0.00 Quadrula quadrula Mapleleaf 0 0.00 Toxolasma parvus Lilliput 0 0.00 Villosa fabalis (1) Rayed bean 1 1.32 Ligumia nasuta Eastern pondmussel 1 1.32 Total 151 (1.) Endangered, Michigan list. (2.) Threatened, Michigan list. (3.) Special concern, Michigan list.
Species Richness. Table 3 presents species richness data by region for 1978 and 2004 for live mussels and for live mussels plus spent shells (= inclusive richness) ranked by 1978 richness. All regions showed a decline in live species from 1978 to 2004 except for the lower mainstem for which no live individuals were found in either survey. These declines ranged from a complete disappearance (9 to 0) in the middle branch to relatively small declines in Paint Creek (4 to 3) and in the upper mainstem (13 to 9). The north branch remains the region with the most species but it has also declined from 21 to 11. Adding spent shell data has little effect on the regional numbers tor live species with the exception of the lower mainstem where 11 species of spent shells were reported in 1978 but none were found in 2004.
TABLE 3. Species richness (live only) and inclusive richness (live plus spent shells) for unionids in the Clinton River (SE Michigan) in 1978 and 2004, Data for 1978 are based on Strayer (1980). Species Species Inclusive Inclusive Richness Richness Richness Richness Region 1978 2004 1978 2004 North Branch 21 11 22 12 Upper Mainstem 13 9 16 17 Stony Creek 10 4 10 8 Middle Branch 9 0 9 5 Coon Creek 8 1 12 9 Paint Creek 4 5 8 4 Lower Mainstem 0 0 11 0 Entire River 24 14 26 20
There are significant regional differences in species richness between 1978 and 2004. The original data were square root transformed after adding 0.0001 to all values. A MANOVA. ([alpha] = 0.05) indicated at least one of the six regions was different with no significant interaction between year and region. An ANOVA was then performed on all regions with each region treated as a normal variable. A comparison of the first six regions in 1978 and the last six in 2004 indicated significant differences in mean site richness (F = 4.296; P < 0.0001). A Tukey/Kramer rest ([alpha] = 0.05) indicated that region six (middle branch) was significantly different from all other regions in both 1978 and 2004.
Exotic Bivalves (Mussels and Clams)
Two exotic bivalves were found in this survey that were not present in 1978; the Asian clam (Corbicula fluminea) and the zebra mussel (Dreissena polymorpha). Sites where these exotics were found are indicated in Figure 1. A crude visual estimate was made for Asian clams based on the relative number of shells visible on the river bottom. Live individuals are invariably buried in the sediment and no attempt was made to sample these. Zebra mussel density is also a crude field estimate except in cases where counts or weights per unionid were measured. In the majority of cases where zebra mussels were found, the site was located downstream from a lake with a known zebra mussel infestation. Sites 13 and 18 had the highest zebra mussel loads on unionids. Because the bottom at these sites was mostly sand, unionids were the only stable sub-stratum to which zebra mussels could attach. Zebra mussels occurred in four of the seven regions of the Clinton River (western tributaries only), while Asian clams were found in all regions except Paint Creek (Figure 1). Zebra mussels were most prevalent at sites located in the upper mainstem (10 sites) with Stony Creek next (4 sites). Zebra mussels were found at only one site in the lower mainstem, a site that is immediately downstream from the Pontiac sewage treatment plant. The upper mainstem had the largest number of sites with both zebra mussels and Asian clams. There were no zebra mussels found in the north branch, middle branch, or Coon Creek. Asian clams were found at only one site each in the north branch, middle branch, Coon Creek, and Stony Creek regions.
Endangered and Threatened Species
Epioblasma triquetra, Toxolasma lividus and Villosa fabalis are all on the Michigan endangered species list. E. triquetra was only found in the upper mainstem in our 2004 survey. A relatively large population for that species at the Cooley Lake Road Bridge site has been known to local malacologists for several years (site 11). Live T lividus was not found in 2004 but spent shells were located at site 9 (Dawson's millpond outlet) indicating the possibility of live individuals being present or recently present. Only shells were found of V. fabalis in 2004 at site 11 in the upper mainstem.
Lampsilis fasciola has threatened status on the Michigan rare species list. In the 1978 survey there were only a few sites where this species was located, yielding eight individuals. In 2004 we found spent L. fasciola in the upper mainstem and north branch regions, however the only area where live individuals were observed was in the north branch region (site 27). Live L. fasciola was observed at site 15 (upper mainstem) in 1994-1997 by RDH.
Of the four remaining listed species (Alasmidonta marginata, Alasmidonta viridis, Pleurobema sintoxia, and Villosa iris) only V iris (special concern) was found in 2004 but it was surprisingly common and widespread having been found at 21 of the 76 sites. In relative density it ranked third in 2004 at nearly 10% and was highest in frequency at 21.
Many of the land use changes occurring in the Clinton River watershed today are a continuation of those that began before Strayer's 1978 freshwater mussel survey (Strayer 1980). These changes are largely an outgrowth of the extensive development (human population growth, residential and commercial development, road construction, etc.) that has occurred in the watershed (Zorn and Seelbach 1992). The human population living in the watershed has increased from about 300,000 in the 1930's to over 1.6 million in 2005. The land cover, which was primarily forests in the 1800's, became croplands by 1900 to present day suburban sprawl. It has become increasingly more industrialized with both commercial and information technology business complexes surrounding many areas of the river (USACE 2005).
In Strayer's 1978 survey he reported that the mussel fauna had been destroyed in several areas by pollution, but most of the watershed still contained a healthy mussel community. Of the original 31 species that were found in the Clinton River from 1870 through 1933, 26 species still remained in 1978 (Strayer 1980). Strayer identified human activities as having a devastating effect on the mussels. The north branch, Stony Creek, and the upper mainstem still had a healthy mussel community in terms of diversity and density. He also found that the mussels were present in about the same proportions as before 1933. His conclusion was that all streams subjected to urban pollution had lost their mussel faunas, whereas those that were not subjected to pollution still enjoyed a rich mussel community (Strayer 1980).
The Clinton River: Land Use and Watershed Changes
From 1990 to 2000 the amount of developed land in the Clinton watershed increased. In Macomb County from 1990 to 2000 there was a 17.6% increase in the amount of land used for single-families and a 28.8% increase in multiple-family dwellings with other categories of development also increasing. Overall, there was a 15.5% increase in those 10 years in the amount of developed land with a decrease of 26.9% in open land (SEMCOG 2003).
In Oakland County there was a 17.9% increase from 1990 to 2000 in the amount of land used for single-families and a 29.1% increase in multiple-family dwellings. Overall there was a 17.9% increase in these 10 years in the amount of developed land. The amount of open land in Oakland County decreased 33.5% between 1990 and 2000 (SEMCOG 2003).
Like many North American rivers, the Clinton River has experienced a variety of watershed and water quality degradation problems over the past several decades. Historically, the primary causes were industrial and municipal discharges contaminating both water and sediments. In recent years, the majority of point discharges were eliminated or treated and at present the ongoing contamination is nearly all non-point source in origin, largely in the form of urban runoff (Sinha and Creech 2006). Currently there are no major discharges to the river or its tributaries, with most municipalities having pretreatment programs (USEPA 2008). Treatment is now so extensive that at the point where the river enters Lake St. Clair, half of the river's flow is treated wastewater originating from six treatment plants. Stormwater runoff is therefore likely to be the greatest source of water quality degradation.
Another major category of environmental degradation due to urbanization is an increase in impervious surface area within the watershed. This has resulted in increased geomorphologic instability in the river with increased soil erosion, bottom scouring, stream bank erosion, and flooding (Sinha and Creech 2006). Land use data for a typical sub-region of the watershed (middle branch of the Clinton River upstream of the Romeo Plank Rd. crossing) illustrates this change. In 1978 the percentage imperviousness was 10.5%; in 2001 it was 19.7% (Sinha and Creech 2006).
A recent analysis of the hydrology and geomorphology of the Clinton River watershed indicated that there have been substantial changes in stream flow trends over the past 40 years. Based on data from 16 USGS gauges, the average peak storm flow increased 31%, annual mean flow increased 54%, and bankfull flow increased 13% in that period. These increases were most pronounced where the greatest watershed development had occurred (Sinha and Creech 2006). The Clinton River has a number of impoundments in the watershed that are maintained by 75 dams, some of which control lakes at statutory levels. Reduction of flow due to maintenance of lake levels, especially during drought periods, is a contributing factor in reducing flow rates and is presumably an added stressor for unionid populations. In 2002-2003, drought resulted in discharges below 5 f(t.sup.3]/sec for periods of over six months at Drayton Plains, MI (near site 16; USGS 2003). The authors have observed complete cessation of flow at one such dam in the upper mainstem that reduces the river to a series of stagnant pools. Unfortunately, this dam is immediately upstream from a small population of state endangered Toxolasma lividus (site 9), only one of two known for the state. In 1978, Strayer reported finding 135 live unionids at site 9 including 24 T lividus (formerly Carunculina glans, Strayer 1980). In 2004 we found no live unionids at this site and only spent shells of T lividus and three other species, an indication that this site is one of the most heavily impacted among those of the Clinton River, very likely a consequence of lake level regulation by this dam during drought periods.
The suitability of rivers for supporting unionids is a combination of biogeography, physical and chemical characteristics, sediment composition, geology and impacts from watershed land use and discharges (Strayer 1983; Cummings and Mayer 1992; Box and Mossa 1999). Intense agricultural land use may increase the levels of pesticides that end up in the river, ultimately impacting the viability of the unionid populations (Badra and Goforth 2003). Land disturbance from construction or agricultural use may also increase sediment load in streams. The Michigan DEQ completed a survey of freshwater mussels in eight rivers (58 sites) finding a total of 32 species (Badra and Goforth 2003). They identified elevated input of fine particles as a contributing factor to the decline in unionids. They suggested that a relatively small amount of the variation in species richness and abundance is caused by the diversity in substrate composition. Changes in sediment composition and quantity can interfere with the host fish-mussel relationship. Some conglutinates (packets of larval mussels) that mimic aquatic insects may not be able to attach to silt-covered substrata. However, a contrasting view indicated there is no conclusive evidence that supports a decline in mussel populations with an increase in stream sediments (Box and Mossa 1999).
Clinton River Fish Community
Fish are especially relevant, to the maintenance of freshwater mussel populations due to their unique parasitic dependence on a host fish for maturation and dispersal of larvae. There are over 100 species of fish that have historically been found in the Clinton River (Francis and Haas 2006). A 2006 report of the fish community in the Clinton River watershed indicates that the fish populations comprising this community have not declined in the Clinton River. Most areas received an acceptable to excellent rating for river quality. The density and diversity of the fish community has actually improved in the last 30 years (Francis and Haas 2006). Since many of the fish species that were found are known hosts to the native mussels, this suggests that host species reduction or elimination is not likely to be a reason for the decline in density and richness of the unionids.
Endangered Unionids and Exotic Bivalves in the Clinton River
Forty-five unionid species have been documented in Michigan rivers and lakes and over one-third of them (19) species are state listed as threatened or endangered (Badra and Goforth 2003). Eight species of unionids found in the Clinton River in the 1978 and/or 2004 surveys are listed as endangered, threatened, or of special concern (Table 2). Of the three species that are presently listed as endangered in Michigan (Epioblasma triquetra, Toxolasma lividus and Villosa fabalis), E. triquetra was only found in the upper mainstem in both the 1978 and 2004 surveys except for one shell found by Strayer at site 23 located in the north branch. Neither of the above surveys found Obovaria subrotunda, which was known from museum specimens in the 1930s (Strayer 1980). Live T lividus was found only at site 9 in the upper mainstem in the 1978 survey. Strayer found 24 live individuals at site 9 (Orchard Lake Rd.), but in 2004 only spent shells were located there and due to occasional interruptions in the volume of water passing over the dam at Dawson's Millpond, this population appears to be imperiled. An intensive search for T lividus at site 9 in 2002 by Doug Sweet produced a density estimate of 1.42/[m.sup.2], however this site is very small (approximately 340 [m.sup.2] and has since been subjected to repeated flow stoppage; hence the population of this species is unlikely to be stable. Only shells were found of Villosa fabalis in both 1978 and 2004 at site 11 in the upper mainstem. As far as can be determined from Strayer's and our surveys, of the three endangered unionids in the Clinton River, E. triquetra has the densest and most stable population. This is in spite of the nearby Cass Lake population having been completely decimated by zebra mussels in the late 1990s (Hunter 1996).
Lampsilis fasciola is listed as threatened. In the 1978 survey there were only a few sites in the upper mainstem and north branch regions where this species was located, yielding eight individuals. By 2004 the same two regions contained L. fasciola, but the only area where live individuals were found was in the north branch region. L. fasciola was found by one of the authors (RDH) at site 15 (upper mainstem) in 1994-1997.
Epioblasma rangiana is on the federal endangered list. This species was not found in the Clinton system, either live or as spent shells, in the 1978 survey or the 2004 survey. It is only known from museum specimens dating from the 1930s (Strayer 1980).
During Strayer's 1978 survey neither zebra mussels nor Asian clams had been introduced in the Clinton River system. Since that time both exotic species have spread throughout many lakes of southeastern Michigan. Although we did not find high densities of zebra mussels in the river, they have infested many lakes through which the Clinton River passes and they have decimated the native mussel populations living in those lakes (Hunter et al. 1997). Hence, there have been clear negative effects of zebra mussels on unionids in the Clinton system but those effects are primarily observed in lakes. As in most other studies, there have been no observed negative effects of Asian clams on unionid populations in the Clinton River despite there being 10 locations where that exotic species is in evidence (Kraemer 1979).
Our study indicates that the freshwater mussel community of the Clinton River has continued to decline compared to historical levels. Since the last survey in 1978, two community7 attributes, number of species (24 to 14) and average density (0.087/[m.sup.2] to 0.017/[m.sup.2], are substantially lower throughout the entire river system. Although determining reasons for this decline was not the focus of this study, we can note that some potentially causative factors have improved while others have worsened. Point-source pollution, previously suggested as a probable cause, has abated and the host fish community is regarded as intact and stable, suggesting that host fish elimination is an unlikely contributor (Strayer 1980; Francis and Haas 2006). Two factors that have clearly worsened are the introduction of exotic bivalves, most notably the zebra mussel, and increased urban development of the watershed resulting in flashy hydrodynamics and geomorphic instability of the riverbed (Janech and Hunter 1995; Hunter et al. 1997; Sinha and Creech 2006). Zebra mussels have devastated unionids in lakes through which the Clinton River passes, but as far as we can determine have had little effect on unionids in the river itself (Hunter et al. 1997). Since watershed urbanization has increased substantially since Strayer's 1980 report, it is likely to be the major factor that has led to unionid species loss and density reduction. One inference of our results is that the negative effects of human activities on the freshwater mussels of the Clinton River have continued although they differ somewhat today from those suggested by Strayer in 1980. A continuing need exists for studies to examine the specific mechanisms by which hydrodynamic instability of rivers negatively impacts freshwater mussel faunas.
The authors gratefully acknowledge the following people: David Strayer (Cary Institute of Ecosystem Studies) for providing us with field notes from his 1978 survey; Jim Francis and Bob Haas (Michigan DNR, Fisheries Div.) for sharing a preliminary copy of their Clinton River Report; and Doug Sweet for his purple and common lilliput reports, done for the MDNR Wildlife Division.
BADRA, P J., AND R. R. GOFORTH. 2003. Freshwater Mussel Surveys of Great Lakes Tributary Rivers in Michigan. Michigan Department of Environmental Quality Coastal Zone Management Unit. MNFI Report Number 2003-15.
BURCH, J. B. 1975. Freshwater Unionacean Clams (Mollusca: Pelecypoda) of North America. Hamburg, MI: Malacological Publications.
BOGAN, A. E. 2008. Global diversity of freshwater mussels (Mollusca: Bi-valvia) in freshwater. Hydrobiologia 595:139-147.
BOX, J. B., AND J. MOSSA. 1999. Sediment, land use, and freshwater mussels: Prospects and problems. Journal of the North American Benthological Society 18:99-177.
CLARKE, A. H. 1981. The Freshwater Molluscs of Canada. Ottawa: National Museum of Canada.
CUMMINGS, K. S., AND C. A. MAYER. 1992. Field Guide to Freshwater Mussels of the Midwest. Manual 5. Champaign: Illinois Natural History Survey.
FRANCIS, J. T., AND R. C. HAAS. 2006. Clinton River Assessment. Michigan Department of Natural Resources Fisheries Division, SR .39.
GRAF, D. L., AND K. S. CUMMINGS. 2007. Review of the systematics and global diversity of freshwater mussel species (Bivalvia:Unionoida). Journal of Molluscan Studies 73:291-314.
HUNTER, R. D. 1996. Impact of Zebra Mussels on Endangered and Non-endangered Unionids in the Upper Clinton River: 1996. Report for the Michigan DNR, Wildlife Division, Natural Heritage Small Grants Program.
HUNTER, R. D., S. A. TOCZYLOWSKI, AND M. G. JANECH. 1997. Zebra mussels in a small river: Impact on unionids. In Zebra Mussels and Aquatic Nuisance Species, ed. F. D'Itri, 161-186. Chelsea, Michigan: Ann Arbor Press, Inc.
JANECH, M. G., AND R. D. HUNTER. 1995. Corbicula fluminea in a Michigan river: Implications for low temperature tolerance. Malacological Review 28:125-130.
KRAEMER, L. R. 1979. Corbicula (Bivalvia:Sphaeriacea) vs indigenous mussels (Bivalvia Unionacea) in U.S. rivers: A hard case for interspecific competition? American Zoologist 199:1085-1096.
MATH1AK, H. A. 1979. A River Survey of the Unionid Mussels of Wisconsin 1973-1977. Horicon, WI: Sand Shell Press.
METCALFE-SMITH, J. L., J. DI MAIO, S. K. STATON, And G. L. MACKIE. 2000. Effect of sampling effort on the efficiency of the timed search method for sampling freshwater mussel communities. Journal of the North American Benthological Society 19:725-732.
SEMCOG. 2003. Land Use Change In Southeast Michigan: Causes and Consequences. Southeast Michigan Council of Governments. http://www.semcog.org/Data/Apps/landuse.cfm?mcd=8999http://www.semcog.org/WebSearch.aspx?searchterm=land%20use#.
SlNHA, S. K., AND C. T. CREECH. 2006. Hydrologic and Geomorphic Analysis of the Clinton River Watershed: Final Report. Environmental Consulting and Technology, Inc., Ann Arbor, MI. (email@example.com).
STHN, B. A., L. S. JUTNER, AND J.S. Adams, eds. 2000. Precious Heritage--The Status of Biodiversity in the United States. New York: Oxford University Press.
STRAYER, D. L. 1980. The freshwater mussels (Bivalvia: Unionidae) of the Clinton River, Michigan, with comments on man's impact on the fauna, 1870-1978. The Nautilus 94:142-149.
STRAYER, D.L. 1983. The effects of surface geology and stream size on freshwater mussel (Bivalvia, Unionidae) distribution in southeastern Michigan, U.S.A. Freshwater Biology 13:253-264.
STRAYER, D.L, And K.J.JlRKA. 1997. The Pearly Mussels of New York State. New York State Education Department.
USACE. 2005. Clinton River Sediment Transport Modeling Study. Report by W.K Baird & Associates. U.S. Army Corps of Engineers, Detroit District. http://www.glc.org/tributary/pubs/documents/ClintonRiverModelingStudy--main1.pdf.
USEPA. 2008. Great Lakes Area of Concern: Clinton River AoC, Beneficial Use Impairments, http://www.epa.gov/greatlakes/aoc/clintriv.html.
USGS. 2003. Annual Water Data Reports, Clinton River near Drayton Plains, Ml. http://waterdata.usgs.gov/mi/nwis/uv/?site_no=041609006&agency--cd=USGS.
ZORN, T.G., AND P W. SEELBACH. 1992. A Historical Perspective of the Clinton River, Its Watershed and Fish Communities. Michigan Department of Natural Resources Fisheries Division. Fisheries Technical Report No. 92-10.
DEBBIE MOROWSKI, LUKE J. JAMES, AND R. DOUGLAS HUNTER
|Printer friendly Cite/link Email Feedback|
|Author:||Morowski, Debbie; James, Luke J.; Hunter, R. Douglas|
|Date:||Sep 22, 2009|
|Previous Article:||Inter- and intraspecific patterns of genetic variation in three species of West Michigan Ranid frogs: a preliminary report.|
|Next Article:||Freshwater mussel (Unionidae) distribution and demographics in relation to microhabitat in a first-order Michigan stream.|