Proceedings of the First Freshwater Mollusk Conservation Society Symposium, 1999, pages 261-274

**** **** ********** ******

Freshwater mussels and water quality:

A review of the effects of hydrologic and instream habitat alterations

G. Thomas Watters1

Ohio Biological Survey and Aquatic Ecology Laboratory; 1315 Kinnear Road,

The Ohio State University, Columbus, OH 43212

ABSTRACT: Hydraulic impacts represent a suite of habitat alterations that, although having different causes, often have similar methods

of affecting the mussel fauna. For instance, logging and channelization are very different disturbances, but both generate sediments.

These hydraulic impacts thus overlap each other to one degree or another. I have attempted to break them down into categories based

on the type of disturbance, but what applies to 1 impact often may apply to others. By far, there is more published information on the

effects of impoundments than on all other hydrologic impacts combined, and this review is dominated by that subject. Other subjects are

not covered in any detail because they are too infrequent or ancillary to North American mussel conservation. For example, log runs in

Finland are known to damage mussel populations (Valovirta 1990), but this is probably not a widespread problem.

Keywords: freshwater mussels, water quality, impoundments, hydrology, habitat

Impoundments impoundment on the Mississippi River. In adjacent

Lake Pepin, a naturally-formed pool, 30 species were

Perhaps mankind s earliest attempt to manipulate free- encountered.

flowing water was the dam. Dams could be used to

divert water to mills and turbines, where its seemingly The general impact of impoundments on existing

limitless power ground grain, cut lumber, and later aquatic habitats was reviewed by many authors.

generated electricity, generally freeing humanity to Yeager s (1993, 1994) reviews are particularly

toil elsewhere. Dams could be used to divert water to thorough. Ellis (1942) gave an early review of the

irrigate ground that would not otherwise support biotic and abiotic effects of impounding a river. He

crops. They could, in theory, alleviate flooding if the noted such deleterious consequences as silt accumula-

amount of water passing through the dam could be tion, loss of shallow water habitat, stagnation,

regulated. Dams could make a shallow river deep, accumulation of pollutants, and nutrient-poor water.

allowing watercraft to operate. Impounded rivers He concluded that the initial period of high

could act as reservoirs for holding water to support the productivity may be very short in some reservoirs, and

populace. Pristine natural areas could be turned into a longer in others but the decline will inevitably come

recreational goldmine through impoundment. Real, unless man makes some adjustments. Baxter (1977)

perceived, or pork barrel, there were many reasons to gave an excellent review drawing from examples

dam rivers. worldwide. The author characterized impoundments

as a distinct type of ecosystem, characterized by

It must be stressed that an artificial impoundment is complicated flow patterns that may involve hypolim-

not analogous to a naturally occurring pool within a nion discharge, long periods of flooding, and heavy

river. Impoundments typically become deeper toward sediment loads. Downstream areas may be affected as

their downstream end, until they abut the dam. In well, particularly by flow regime. Baxter sagely noted

contrast, natural pools are deepest toward their middle, that since not all the hydraulic head of the world s rivers

then becoming shallow, forming runs and riffles. This has yet been utilized, it seems likely that more remain to

results in a very different water flow pattern through be built. Neel (1963) gave a somewhat lopsided review

the pool/impoundment, and subsequently down- of reservoirs, where they were largely viewed as

stream. These hydrologic differences result in faunal desirable features. However, he romantically noted

differences. For example, van der Schalie (1938) that fish migration was blocked, with frustrated piscine

found 15 mussel species in Lake Cooper, a man-made migrants, monotonously working their way, time after

time, toward the overpowering jets. Although the

benthic fauna was adversely affected, he speculated

For correspondence contact G.T. Watters

3

(in error) that no certainty exists that missing species

(Email: abqk2k@r.postjobfree.com)

261

have become extinct or cannot return. Despite Neel s extirpated, and several are now extinct. Five species,

(and others) claims, the evidence that impoundment is mainly soft-substrate tolerant taxa, have invaded the

detrimental to aquatic life, and most mussels in reservoir. Four original species that survived the

particular, is overwhelming and indisputable. impoundment also have increased in abundance.

Even below impoundments benthic diversity in Mussel diversity has declined from 64 species to 30 in

general is reduced (Yeager 1993, 1994). We know the upper Chickamauga Reservoir of the Tennessee

that perhaps several dozen mussel species, and River (Ahlstedt and McDonough 1994). Although

numerous more freshwater snails, were driven to relict populations of four federally endangered

extinction wholly or in large part by the construction mussels still occurred there, based on pre-impound-

of dams (Layzer et al. 1993, Lydeard and Mayden ment shell middens, two of these were once the most

1995, Stansbery 1973). Almost without exception, abundant species in the river reach.

rivers that have been impounded, have lost or changed

their mussel faunas. In the Tennessee River portion of Wheeler Reservoir,

19 of the historical 38 species were either absent or

Examples present only as relicts (non-reproducing individuals)

A. Tennessee River: Isom (1969) reported that the (Ahlstedt and McDonough, in press). Thick-shelled

Tennessee River s mussel diversity had decreased species may have been severely affected during a 6-y

from 100 species to 44, largely because of changes drought in the 1980 s. Overall abundance had

associated with dams. Although caused in part by declined from an estimated 39 million mussels in

over-harvesting, the author attributed most of the 1960, to 14 million in 1991.

decline to changes in habitat associated with the dam.

Water flow in the impoundment had decreased to the B. Cumberland River: At least 37 of the 60 pre-

point where silt could accumulate on the river bed, impoundment species of the Caney Fork River of the

smothering mussels. Coincidentally, silt-tolerant Cumberland River have been extirpated (Layzer et al.

species were expanding their ranges in the river. 1993). Two are extinct. The authors attributed this, in

large part, to the presence of the Center Hill Dam. No

In the Fort Loudoun Reservoir on the Tennessee living mussels were found for 12 km below the dam,

River, Isom (1971) found a drastic decline in mussel which they attributed to the discharge of cold

diversity post-impoundment. A survey in 1970 of the hypolimnetic water, periodic scouring, and dewater-

reservoir found four mussel species. Prior to its ing. Near-anoxic conditions occasionally occurred in

impoundment, Ortmann (1918) reported 64 species the metalimnion and hypolimnion of the impound-

from the same general area. In addition, the 1970 ment. This acted as a barrier to fish movement across

survey found mussels primarily on flooded pre- the impoundment, isolating mussel and fish popula-

impoundment land (overbank), not in the original tions in tributaries. Sedimentation of the impound-

river channel habitat. This probably was due to ment also was taking place.

critically depleted oxygen levels in the channel.

Schmidt (1986), in a separate survey, found 36 mussel

Elsewhere, during construction of the Nickajack Dam, species in the Caney Fork River prior to impound-

the Tennessee River was dewatered at the construction ment. By this estimate, the construction of the Center

site (Isom 1972). Seventeen species were found. Hill hydroelectric dam resulted in the loss of 78% of

Ortmann (1925) had found 25, for a cumulative total of those species. The author concluded that the

33 species reported for this reach. Missing in the latter extirpated species could not adapt to the widely

survey were many endangered and rare species. Whether fluctuating daily flows, cold water discharges, and

this was due to the presence of the Hales Bar Dam, 6.4 loss of nutrients.

miles upstream, or other causes was not known.

In Lake Barkley on the Cumberland River, 64% of

It may be argued that mussel faunal composition the pre-impoundment mussel fauna was lost as a

changes over time whether dams are built or not. consequence of impoundment (Blalock and Sickel

However, Parmalee et al. (1982) documented a fauna 1996). Prior to impoundment, 25 mussel species were

that remained essentially unchanged for several reported from this reach. This study found only nine

millennia until impounded by a dam. The species 29 years after completion of the reservoir.

Chickamauga Reservoir of the Tennessee River However, 15 of the original species still occurred

supported 46 species for perhaps 2,000 years prior to below the dam. Species had either invaded the

impoundment. After impoundment, 28 species were impoundment, or once insignificant ones had became

262

dominant, including anodontines and species of Mussel habitat in impoundments

Quadrula. The authors attributed the overall decline, A free-flowing river has great habitat heterogeneity:

in part, to the accumulation of sediments, anoxic riffles, runs, pools, shoals, water-willow stands, and

conditions in the impoundment, and loss of fish hosts. meanders, often with considerable tree canopy. After

impoundment, 3 - 4 habitats remain, usually with no

In the lowest reach of the Cumberland River, only 25 tree canopy. These are either new habitats or highly

of the 45 pre-impoundment mussel species remained modified existing ones (Bates 1962, Blalock and

(Sickel and Chandler 1996). Several new species had Sickel 1996). These remaining habitats are discussed

invaded, probably from the impounded Ohio River. below.

C. Little Tennessee River: Only 6 of the original 50 Original channel

mussel species at Tellico Lake of the Little Tennessee The first is the original channel, which remains intact

River remained after impoundment (Parmalee and but under deeper water. Studies demonstrate that

Hughes 1993). Although diversity had declined mussels, in general, are most abundant in shallow

before completion of the dam, additional species were water, with relatively few species able to tolerate

lost when they were impounded. By 1972, this impoundment depths (Haukioja and Hakala 1974,

diversity had dropped to 18 species. Twelve years Lewandowski and Stanczykowska 1975). Haag and

after impoundment and the formation of Tellico Lake, Thorpe (1991) examined the relationship between

only 6 of the original species still lived there. depth and substrate type on benthic invertebrates at a

However, another 8 species were encountered for the site approximately 1.5 km downstream of Kentucky

first time. These species were apparently able to Lock and Dam on the Tennessee River. Although

colonize the extensive sand/mud/silt substrate of the mussel abundance was not correlated with substrate

new embayments and overbanks. type, abundance decreased with depth.

D. Kaskaskia River: Forty mussel species were As water velocity decreases, water loses its ability to

recorded from the Kaskaskia River prior to carry sediment. The old river channel effectively

impoundment. Approximately 8 years after impound- becomes a sediment trap, eventually smothering

ment, only 24 species were found (Suloway et al. mussels that cannot adapt to soft substrates (Isom

1981). Some sites no longer supported any mussels, 1969). This is particularly true near the upstream side

and overall density had declined. Once clean sand- of the dam, where the substrate may be composed of

gravel substrates were overlain with silt and debris mud mixed with debris and rubbish (Clark and Gillette

caused by bank erosion, substrate instability, and 1911). Ellis (1936) summarized the effects of silt and

runoff. Changes in host abundance also may have sediment on the aquatic habitat in general, and on

occurred. mussels in particular. Silt resulted in the loss of light

penetration, causing diminished algal abundance, an

E. Tombigbee River: The once highly diverse mussel important food of mussels. Thermal changes also

fauna of the Tombigbee River was nearly obliterated occurred, specifically the creation of an annual lag in

by the construction of the Tenn-Tom Waterway and the cooling and warming of the hypolimnion. Silt

associated dams (Williams et al. 1992). Comparisons caused organic material to be retained on the bottom,

with unimpounded and pre-impounded areas revealed leading to oxygen depletion. And finally, silt

that perhaps two-thirds of the fauna were eliminated in smothered the benthic fauna. Ellis (1936) demon-

impounded areas by loss of necessary habitat due to strated this using over 2,000 mussels of 18 species in

decreased water flow, increased depth, and sedimenta- an artificial stream to which silt was added. A silt

tion. accumulation of 0.6 2.5 cm depth resulted in

mortality approaching 90%.

These and other case studies reveal clear-cut patterns

and trends in the change associated with converting a Even before extirpation by smothering occurs,

free-flowing riverine habitat into an impoundment. recruitment may be diminished or stopped, and growth

Clearly, impoundments interfere with the basic rates reduced. Bates (1962), for example, found no

ecological processes of free-flowing systems (Sparks evidence of mussel recruitment in the channel of

1995). Most indigenous mussel species are extirpated Kentucky Lake, although adult mussels were present.

from the impounded region. These may be replaced by Semenova et al. (1992) compared growth rates of

soft-substrate adapted species, such as anodontines Margaritifera margaritifera (Linnaeus, 1758) be-

and heelsplitters. The reason for this lies in the nature tween rivers having different habitats, and found that

of the habitat modifications caused by impoundment. rivers having high levels of suspended solids inhibited

263

mussel growth. Bauer et al. (1980) determined that M. or conservation importance. Although there is

margaritifera juveniles required substrate with low conflicting evidence as to what degree mussels are

organic content, conditions not met in submerged habitat specific, the change from a sand/gravel/cobble

channels. Buddensiek et al. (1993) showed that bottom to one overlain with silt entails at least a

juvenile mussels, which live completely buried for change in relative abundance from the original fauna,

several years, required a substrate where water could if not the loss of original species (Burkhead et al.

be freely exchanged between the overlying water 1992, Layzer et al. 1993, Williams et al. 1992).

column and the interstitial water. These interstices are Harman (1972) believed that freshwater mollusks had

clogged by silt in impoundment conditions. substrate preferences. He showed that species

occurring in clean grave-cobble did not occur in silt,

The hypolimnion of the channel also may become and vice versa. In another study, Ghent et al. (1978)

excessively cold, near-anoxic, and nutrient-poor. examined the distributions of Elliptio, a silt-intolerant

Cold water ( 25 m), and control). In some areas, water levels may be become

any new juvenile mussels that may been deposited shallow enough that thermal buffering is lost,

there earlier in the year, now exposed. allowing extreme temperatures to occur. Blinn et al.

(1995) reported that substrate subjected to 2-12 h

Dams and impoundments as barriers to mussels exposures to air required more than 4 mo. to regain a

and hosts biomass similar to unexposed habitat. Federally

It was recognized early that dams acted as barriers to endangered mussel species were reported by Neck and

fish movement (Coker 1914). Coker (1914) believed Howells (1994) as casualties of scheduled dewatering

impoundments had both good and bad qualities for processes. Riggs and Webb (1956) reported that

mussels, and envisioned the tremendously increased several thousand mussels died in the tailwaters of Lake

surface area of flooded dry land as potential mussel Texoma, an impoundment of the Red River formed by

and fish habitat. But he realized that dams would form Denison Dam, when water levels dropped, allowing the

effective barriers to fish host dispersal. Hubbs and water to become excessively warm (>26 C). This area

Pigg (1976) showed that the decline in fish diversity in was exposed for at least 20 days before being again

Oklahoma was due, at least in part, to these barriers. inundated. Exposure to cold air may be equally lethal.

Migrating fishes, in particular, would be affected Nagel (1987) believed mussels were more sensitive to

(Coker 1914, Branson 1974, Unkenholz 1986, cold water during frosts than warm water during

Alexander 1987). Some fishes are now rare, due in temporary droughts, and Blinn et al. (1995) showed that

large part to these migration obstacles (Stuebner a single overnight exposure to subzero temperatures

1993). In some cases, such as the Hells Canyon resulted in at least a 90% loss of invertebrate mass.

Complex in Idaho, migration has become impossible Valovirta (1990) reported that mussels were killed when

for some species of fishes (Collier et al. 1996). For water froze to the river bottom.

freshwater mussels, this meant that hosts may not

be accessible to their glochidial parasites. Other- Harman (1974) found few mussels within the first 17-

wise healthy mussel populations would simply 34 km below impoundments in the Delaware River

grow old and die without recruiting. Such a loss of system due to cold water discharges. Low

hosts has been implicated in the decline of mussels temperatures associated with an impoundment

in several areas (Burkhead e t a l . 1992, Jones 1991, hypolimnion were thought to result in slow growth

Suloway e t a l . 1981). Watters (1996) showed that and inhibited reproduction. This was dramatically

dams as low as 1 m in height restricted the shown by Heinricher and Layzer (1999), who

distribution of some mussel species. Because the transplanted non-reproducing mussels out of a cold

dams caused minimal habitat disturbance, this water discharge to warmer water where the mussels

effect was probably due to interference with host began to reproduce.

movements and migrations. Even if not absolutely

trapped by dams, the route a fish must take to Associated with fluctuating water levels are dramatic

surmount a dam are often circuitous and, at best, changes in water velocity. Neck and Howells (1994)

unlikely. Clark and Gillete (1911) noted that as the noted that high-volume water discharges and abrupt

result of Sullivan s Dam on the Little River, a fish stoppages resulted in a river bed composed of large

trying to move upstream would have to backtrack rocks and shifting sand, habitat inhospitable to most

downstream to the Big Arkansas River, enter mussels. In the River Schwalm in Germany, Nagel

Chisholm Creek and move upstream until it (1992) compared growth rates of three mussel species

encountered a cut-off ditch that communicated back to (Unio pictorum, Pseudanodonta complanata, Anodonta

265

piscinalis) along the length of the river. Little change (1993). In otherwise habitat-homogeneous impounded

in growth rate was associated with either cool- or rivers, these structures may offer a small measure of

warm-water discharges. More important were habitat diversity for some macroinvertebrates (Beckett

differences in water flow velocity. A similar result et al. 1983, Burress et al. 1982). However, dikes and

was found for the declining mussel population in the levees often trap fine-grained sediments. These areas

Licking River of Kentucky (McMurray et al., 1999). would be colonized by soft-substrate adapted mussels.

The authors believed that spikes of cold hypolimnionic The rip-rap often used with these structures would not

discharge from the Cave Run Lake dam was more be suitable for colonization by mussels (Libois and

important to the mussel fauna than were changes in Hallet-Libois 1987).

long-term discharges. Scour immediately below dams

also may preclude colonization by mussels (Miller and Upland reservoirs and changes in the water table

Payne 1992). These authors characterized a large There is little information on the effects of upland

mussel bed below a dam by density, recruitment rates, reservoirs on mussel communities. Hauck and Edson

and demography. Compared with earlier surveys, (1976) discussed some consequences on fishes of

there was little indication that this bed had changed as upland storage ponds used for hydroelectric generation.

the result of the presence of the dam, except the area Some of these effects influence mussels, directly or

closest to the dam. That area experienced greater indirectly. Pumping cycles produce fluctuating water

scour and erosion than downstream and had less levels that adversely affect mussels. Hosts also may

mussel diversity. However, in other rivers this be entrained incidentally by water intakes.

unstable zone may be extensive. In Texas, the Possum

Kingdom Reservoir on the Brazos River exhibited Land use practices

unstable substrate for 150 km below the dam (Yeager Logging, mining, construction, farming, livestock,

1993). These areas of erosion may move downstream and a host of other land uses often adversely affect

at a rate of up to tens of km yr-1 (Fedorov 1969) until mussel populations, generally by releasing runoff of

water velocity falls below the threshold necessary to sediments, salt, and other pollutants into the stream, as

transport sediments (Yeager 1993). well as an increased volume of water (Allan and

Flecker 1993, Osborne and Kovacic 1993, Patric and

Beyond the area of scour sedimentation below dams Aubertin 1977). Houp (1993) documented that an

may be extreme. The Rio Grande at El Paso, below the otherwise pristine Wild River segment of the Red

Elephant Butte Dam, accumulated nearly 4 m of River in eastern Kentucky, was experiencing a decline

sediment in 26 y (Reinhardt 1937). This sedimenta- in mussels due to sedimentation from mining, stream-

tion reduced fish diversity and abundance by altering relocation, logging, and farming, although these

habitat necessary for spawning and overwintering occurred miles upstream of the effected area. The

(Petts 1984, Holland and Huston 1985, Nelson et al. result was a changing mussel community composition

1987). Changes in fish faunal composition as the favoring more tolerant and less habitat-specific

result of impoundments have been demonstrated for species. A similar condition was reported for the

the Clinch River in Tennessee (Fitz 1968), the Barren Buttahatchee River, Mississippi (Jones 1991), where

River in Kentucky (Swink and Jacobs 1983), and the mussel surveys in 1990 demonstrated pronounced

Guadeloupe River in Texas (Edwards 1978). declines in several areas since 1977. These declines

were attributed to impoundment associated with the

Clearly, silt deposited in the tailwaters may smother Tenn-Tom Waterway, increased turbidity from an

mussels. But Chutter (1969) showed that even sublethal abandoned kaolin strip mine, runoff from logging

amounts of silt and sand rendered the tailwaters areas, and sand and gravel mines. The substrate was

inhospitable. In case studies on aquatic invertebrates in unstable at several sites, perhaps due to gravel mining.

South Africa, he found that increases in the amount of Bank erosion had resulted from the removal of the

sand and silt below dams caused instability of the river riparian corridor through logging or agriculture.

substrate. Therefore, it was not necessary to smother the Mussel populations in the Powell River, Virginia

substrate before deleterious changes in habitat occurred. based on early surveys of the 1900 s, were degraded

Neck and Howells (1994) proposed this as one cause (of by land mismanagement (Wolcott and Neves 1994).

several) for the decline of the imperiled Potamilus Runoff from coal mines, abandoned mine lands, and

amphichaenus (Frierson, 1898). wastewater treatment plant effluents all had contributed

to the decline. Although mussels still lived at most

Dikes and levees sites, virtually no recruitment was found. Woodward

The overall biotic and abiotic impacts of dikes and (1990) noted several changes in land use in Scotland

levees were summarized by Pennington and Shields that harmed existing mussel populations: strip mines

266

and peat extraction contributed to sediment and acid bottom sediments, and increase the amount of

water runoff; fish farms discharged waste; and suspended solids for undetermined amounts of time.

improvements in the road system allowing easier Muddy plumes behind tugs are not uncommon,

access to streams not only to pearl fishers, but to indicating that bottom sediments have been forcibly

tourists, skiers, divers, and hikers, which were removed from the channel and suspended. (Apocryphal

counter-productive due to their increasing presence reports exist of mussels being lifted from the bottom

through ready access resulting in the destruction of the into tug propellers.) Wake impinging on the shoreline

natural environment which drew them to the area in may escalate erosion and introduce additional

the first place. Proximity of streams to roads may be suspended solids into the water. It is believed that

detrimental by increasing the amount of salt, heavy these factors interfere with mussel respiration and/or

metals, petroleum products, and other pollutants feeding, and may result in diminished health in

washed into the system (Van Hassel et al. 1980, mussels in these areas.

Winger et al. 1985). With runoff, changes in host

composition may occur as well (Boschung and O Neil Aldridge et al. (1987) studied the effect of propeller-

1981, Berkman and Rabeni 1987, Houp 1993, induced turbulence in controlled laboratory conditions

Rutherford et al. 1992). Finally, sediments from by exposing mussels to varying levels of suspended

channelization and other sources of runoff erode solids and periods of turbulence and measuring

mussel shells rendering them susceptible to shell- physiological rates. They concluded that frequent

dissolving acids (Harmon 1974). turbulence (once every 0.5 h) and high concentrations

of suspended solids (600-750 mg L-1) changed the

Where recognized, erosional problems often require physiological energetics of the mussels by significantly

bank stabilization practices. Libois and Hallet-Libois decreasing food clearance rates and oxygen uptake,

(1987) examined four mussel species in the River and nitrogen elimination. However, Miller and Payne

Meuse in Belgium during a maintenance dewatering. (1995), in an in situ study, investigated the possible

The banks of the river varied from several types of effects of navigation traffic on mussel beds in the Ohio

natural substrate to man-made reinforcements: riprap, River. They concluded that although there were

gabions, concrete, and open stone pitching. Mussel changes in water velocity associated with passage, the

density was highest in the natural habitats of mud, duration was too short and the velocity change too

sand, and fine gravel, and lowest by several orders of small to affect mussels. Clearly, more work needs to

magnitude on man-made reinforcements. They be done on this important aspect of mussel

concluded that mussel density was significantly lower conservation and management.

in the stabilized areas than in unaltered habitat.

Trampling

Land mismanagement need not be confined to There is very little literature addressing the issue of

physical destruction of the riparian corridor. In an age trampling on mussels. By this I mean the physical

when everyone wants stream-front property with a destruction to individual mussels caused by being

green lawn to the water s edge, obvious habitat crushed by people, animals, and machinery. As early

alterations become apparent. Morris and Corkum as the 1700 s, Scottish and German pearl fishers

(1996) studied differences in mussel community realized that mussels downstream of fords often had

composition between a forested riparian corridor and more pearls than in other places (Kunz and Stevenson

one composed of grass. They found greater 1908). This was due, no doubt, to the increased

temperature fluctuations and higher ammonia and sediments (that became the pearl nuclei) generated by

nitrogen concentrations at the grassy site. Although horses and wagons crossing the fords. In areas where

the two types of sites had the same mussel diversity, livestock have access to a waterway, crushed mussels,

the grassy site was dominated by an anodontine broken shells, and deformed living individuals are not

species. The authors suggested that species had a uncommon. Similarly, fords across streams are often

higher tolerance to the fluctuating temperature and rendered devoid of mussels. In areas where waterways

different chemical composition found there. have a firm substrate, the practice of driving offroad

vehicles in the river may be a chronic problem.

Watercraft However these usually represent fairly insignificant

Watercraft, particularly large vessels such as tugs nuisances that act on small areas.

pushing barges, cause changes in habitat, and may

affect mussels in and adjacent to navigation channels. More serious is the problem posed by canoeists and

Water turbulence increases as vessels pass by, naturalists, who pose a peculiar paradox. Although

creating short-lived surges. These surges stir-up these people are often supportive and appreciative of

267

natural areas, the portage of canoes and people across tests. Besides smothering, sediments resulting from

riffles and runs must have a deleterious impact on channelization may resuspend contaminants (Engler

mussels, particularly juveniles. Juveniles and adult 1979), increase concentrations of inorganic plant

mussels live buried in these riffles. Hosts (e.g., nutrients, lower photosynthesis (Loar et al. 1980), and

darters, sculpins) must frequent these riffles to be increase BOD (Ebert 1993). Sediments generated by

parasitized. I have witnessed up to 40 canoes an hour channelization may eventually erode shells, rendering

being dragged across riffles supporting federally them more susceptible to shell-dissolving acids

endangered species in Big Darby Creek, Ohio. This (Harman 1974).

scenario, played repeatedly at every riffle, may have a

substantial effect on mussel populations. There is no Full-scale dredging is not necessary to disrupt mussel

doubt that the presence of hikers, divers, skiers, and populations. Valovirta (1990) noted that simply

tourists may be detrimental to a natural area removing large rocks from the channels to facilitate

(Woodward 1990). Denying use of this recreation, water flow caused the substrate to become unstable,

however, could have negative political and social resulting in mussel mortality. In southern Europe, M.

repercussions. margaritifera was found in only 25% of its original

range, and many populations were not reproducing

Channelization, dredging and snagging (Bauer 1986). Bauer attributed this decline largely to

Channelization, dredging, and snagging, like eutrophication through pollution, but several cases

impoundment, reduce habitat heterogeneity and were given of channelization adversely affecting

aquatic diversity (Nelson 1993). Meanders are mussel populations. Although there is not a large

removed (Simpson et al. 1982) and the riffle-run-pool body of literature on the impact of channelization to

sequence is disrupted (Keller 1978, Wesche 1985) mussels, channelization has been implicated in mussel

reducing available fish and mussel habitat. Circulation declines (Grace and Buchanan 1981, Hartfield 1993,

patterns and substrate composition are altered (Loar et Schuster et al. 1989, Valovirta 1990, Yokley and

al. 1980). Macroinvertebrate assemblages and trapped Gooch 1976).

organic matter that form integral parts of the trophic

web are eliminated (Cummins et al. 1973, Ebert As the forces of nature attempt to reclaim these altered

1993). Potential mussel hosts may be lost as fish habitats, it becomes necessary to perform maintenance

faunal composition changes. Examples of fish faunal to preserve the existing modifications. Channels

changes associated with channelization and dredging dredged for navigation or flood control quickly begin

were documented in Mississippi (Arner et al . 1976), to refill, requiring a periodic re-dredging to ensure a

Ohio (Trautman and Gartman 1974), Illinois (Smith minimum depth. Impoundments are dredged to

1968, 1971), and Pennsylvania (Lee 1973). maintain storage capacity. Dredge spoil and

associated contaminants disposed of in upland areas

The process of dredging and channelization may be may inevitably re-enter the river through surface

catastrophic. Most obviously, mussels caught in the runoff, biological uptake and cycling, and leaching

dredge path are destroyed. But the effects of into groundwater (Gambrell et al. 1978). The

channelization are more far-ranging than just the problems associated with channelization thus become

immediate dredge area. Silt and other suspended chronic, resulting in long-term ecological changes.

solids generated by channelization may travel

downstream and smother, or otherwise adversely Channelization may result in an increase in soft-

affect mussels. Instream dredging for minerals has substrate adapted mussels. Increasing the depth of

similar effects. Marking (1979) buried 3 mussel channels may lower water velocity, causing

species to depths of 25 cm to determine how much sedimentation to take place (Hubbard et al. 1993),

smothering they could survive. Fifty percent of allowing these mussels to colonize the new habitat.

Lampsilis cardium (Rafinesque, 1820) and Lampsilis On one occasion this increased the population of an

radiata (Gmelin, 1791) individuals were able to endangered mussel species. The federally endangered

extricate themselves from up to 17.5 cm in depth. Fifty Potamilus capax (Green, 1832) had colonized

percent of Fusconaia flava (Rafinesque, 1820) portions of the severely channelized St. Francis River

individuals were able to extricate themselves from up flood-control channels in Arkansas (Ahlstedt and

to 10 cm in depth. But Ellis (1942), working with over Jenkinson 1987). The modification had resulted in an

2,000 mussels of 18 species, reported that 90% apparently optimum habitat for that soft-substrate

mortality occurred with 25 mm of siltation. Granted tolerant species. Many of these mussels were

that dredge spoil and silt may be two different things, relocated when re-channelization became necessary.

but the disparity of these results beg for additional The Hocking River in Ohio, channelized for flood

268

control, was colonized by several otherwise uncommon Nevertheless, once connected, we know that fishes

species, including a then state endangered species and mussels populated these canals and certainly must

(Watters 1988). These were all soft-substrate adapted have moved between systems. What effect this had on

species. It must be emphasized that these are unusual mussel and host distributions cannot be ascertained;

cases and that despite their rarity, these new mussel the event is too distant in time with sketchy baseline

faunas flourished at the expense of the original faunas. data for comparison. While it might be assumed that

Furthermore, most dredged areas are not quickly these canals were sediment laden, supporting only

colonized by mussels of any kind. For example, Grace anodontines and other soft-substrate tolerant species,

and Buchanan (1981) found no mussels in an area this was not always the case. Higgins (1858)

dredged 15 y earlier. described the Columbus [Ohio] feeder canal of the

Ohio Canal: Many species have traversed the whole

Snagging is a common practice to alleviate perceived length of the canal, and many species there thrive and

flooding ( i.e., preventing a river from inundating its become abundant which are quite rare in the adjacent

natural floodplain). Fallen trees and debris are rivers. Records at the Carnegie Museum of Natural

removed by dragging them from the stream bed. This History, Pittsburgh, also document now endangered

action inevitably reduces the available habitat by species living in these canals.

creating a more homogeneous environment (Marzolf

1978). Habitat heterogeneity is important to fish Most of these canals are now abandoned and no longer

diversity, and therefore mussel diversity. Snagging maintain connections between rivers. However, in

increases bank erosion (Hubbard et al. 1993) and Texas, connections recently have been made between

creates unstable substrates as the stream recovers the Red and Trinity Rivers. Plans also are being

(Cobb and Kaufman 1993). considered to move water from the Neches or Sabine

River, across the Trinity, Brazos, and Colorado River

Channelization and snagging may actually increase systems, to the Lavaca-Navidad, to be removed

flood heights (Belt 1975). This is due, in part, to a downstream for transfer to Corpus Christi (Howells in

reduction of stream length and increased gradient lit. 1997).

(Hubbard et al. 1993). (One stretch of the Rio Grande,

for instance, was shortened from 155 miles to 88 miles

(Mueller 1975)). These amplified flood events create Summary

additional runoff, and additional remediation.

Hydraulic impacts on freshwater mussel habitats are

often catastrophic, both immediately and over time.

Headcutting

Most such impacts involve complicated interrelated

Headcuts are regions of disturbance moving upstream,

actions; rarely is there a single causative agent for

in a zipper-like fashion, as the result of the upper

mussel declines. For example, although impoundments

boundary of the modification collapsing. Headcuts

may lead to the immediate smothering of some

may move miles upstream, destroying habitat and

mussels, many effects may take years to become

mussels as they pass. Headcutting is a form of channel

apparent: changes in seasonal temperatures within the

modification that has operated undetected on many

impoundment, isolation from necessary hosts, changes

rivers and streams (Hartfield 1993). Instream

in component fauna, etc.

modifications such as dredging obviously affect the

mussels in the immediate area, as well as those

The data clearly support the fact that impoundments,

downstream. But sometimes, mussels above the

dredging, snagging, channelization, and other

modification become extirpated as well.

improvements once taken for granted may have long-

term detrimental effects on freshwater mussels. In

Canals

many cases these changes appear to be irreversible.

To our ancestors, traveling on rivers was much easier

Dozens of freshwater mussels and snails have become

than traveling by land, as long as the river went where

extinct within the past 200 years as the result of these

you wanted it to go. It quickly became apparent that

practices. Although our knowledge of these animals

divine provenance had not placed rivers where they

has dramatically increased in the past decades, it is

should be, an oversight humanity attempted to rectify.

now apparent that their basic biology is much more

By digging canals between rivers and lakes, watercraft

complex than ever imagined. Our future use and

could move great distances without laborious portage.

alteration of their habitat must be carefully planned

No thought was given to the possibility that

with this knowledge in mind to prevent any further

connecting two different river systems possessing

irrevocable loss of biodiversity.

different faunas might not be a desirable goal.

269

Literature Cited Baxter, R.M. 1977. Environmental effects of dams

and impoundments. Annual Review of Ecology

Ahlstedt, S.A. and J.J. Jenkinson. 1987. Distribution and Systematics 8: 255-283.

and abundance of Potamilus capax and other Beckett, D.C., Bingham, C.R., Sanders, L.G.,

freshwater mussels in the St. Francis River Mathais, D.B., and E.M. McLemore. 1983.

system, Arkansas and Missouri. U.S. Army Corps Benthic macroinvertebrates of selected aquatic

of Engineers, Memphis District, Tennessee. habitats on the lower Mississippi River.

Contract No. PD-86-C052. 164 pp. Unpublished Report E-83-10, US Army Corps of

Ahlstedt, S.A. and T. McDonough. 1994. Summary Engineers Waterways Experiment Station,

of preoperational monitoring of the mussel fauna Vicksburg, Mississippi.

in upper Chickamauga Reservoir in the vicinity of Belt, C.B. 1975. The 1973 flood and man s

the Watts Bar Nuclear Plant 1983-1993. constriction of the Mississippi River. Science

Tennessee valley Authority, Water Management 189: 681-684.

Environmental Compliance, Norris, TN. 29 pp. Berkman, H.E. and C.F. Rabeni. 1987. Effect of

Ahlstedt, S.A. and T. McDonough. 1993. siltation on stream fish communities.

Quantitative evaluation of commercial mussel Environmental Biology of Fishes 18: 285-294.

populations in the Tennessee River portion of Blalock, H.N. and J.B. Sickel. 1996. Changes in

Wheeler Reservoir, Alabama. Pages 38-49. I n K. mussel (Bivalvia:Unionidae) fauna within the

S. Cummings, A. C. Buchanan, and L. M. Koch Kentucky portion of Lake Barkley since

(editors) Conservation and Management of impoundment of the lower Cumberland River.

Freshwater Mussels. Proceedings of a UMRCC American Malacological Bulletin 13: 111-116.

Symposium, October 1992, St. Louis, Missouri. Blinn, D.W., Shannon, J.P., Stevens, L.E., and

Upper Mississippi River Conservation Committee, J.P. Carder. 1995. C onsequences of

Rock Island, Illinois. fluctuating discharge for lotic communities.

Aldridge, D.W., B.S. Payne, and A.C. Miller. 1987. Journal of the North American Benthological

The effects of intermittent exposure to suspended Society 1 4: 233-248.

solids and turbulence on three species of freshwater Boschung, H.T. and P. O Neil. 1981. The effects of

mussels. Environmental Pollution 45: 17-28. forest clear-cutting on fishes and

Alexander, C.M. 1987. Fishery status assessment of macroinvertebrates in an Alabama stream. Pp.

Watts Bar Reservoir with management 200-217 I n L.A. Krumholz, C.F. Bryan, G.D.

recommendations. Unpublished report, Tennessee Pardue, and G.A. Hall (editors), Proceedings of

Valley Authority TVA/ONR/WRF-83/4(a), the Warmwater Stream Symposium, Southern

Knoxville. Division of the American Fisheries Society,

Allan, J.D. and A.S. Flecker. 1993. Biodiver

Contact this candidate