Freshwater Biology - 2006 - SPOONER - Context Dependent Effects of Freshwater Mussels On Stream Benthic Communities

Freshwater Biology (2006) 51, 1016–1024 doi:10.1111/j.1365-2427.2006.01547.
Context-dependent effects of freshwater mussels on

stream benthic communities
DANIEL E. SPOONER AND CARYN C. VAUGHN
Oklahoma Biological Survey and Department of Zoology, University of Oklahoma, Norman, OK, U.S.A.
SU M M A R Y
1. We asked whether unionid mussels influence the distribution and abundance of
co-occurring benthic algae and invertebrates. In a yearlong field enclosure experiment in a
south-central U.S. river, we examined the effects of living mussels versus sham mussels
(shells filled with sand) on periphyton and invertebrates in both the surrounding sediment
and on mussel shells. We also examined differences between two common unionid
species, Actinonaias ligamentina (Lamarck 1819) and Amblema plicata (Say 1817).
2. Organic matter concentrations and invertebrate densities in the sediment surrounding
mussels were significantly higher in treatments with live mussels than treatments with
sham mussels or sediment alone. Organic matter was significantly higher in the sediment
surrounding Actinonaias than that surrounding Amblema. Actinonaias was more active than
Amblema and may have increased benthic organic matter through bioturbation.
3. Living mussels increased the abundance of periphyton on shells and the abundance and
richness of invertebrates on shells, whereas effects of sham mussels were similar to
sediment alone. Differences in the amount of periphyton growing on the shells of the two
mussel species reflected differences in mussel activity and shell morphology.
4. Differences between living and sham mussel treatments indicate that biological activities
of mussels provide ecosystem services to the benthic community beyond the physical
habitat provided by shells alone. In treatments containing live mussels we found
significant correlations between organic matter and chlorophyll a concentrations in the
sediment, organic matter concentrations and invertebrate abundance in the sediment and
the amount of chlorophyll a on the sediment and invertebrate abundance. There were no
significant correlations among these response variables in control treatments. Thus, in
addition to providing biogenic structure as habitat, mussels likely facilitate benthic
invertebrates by altering the availability of resources (algae and organic matter) through
nutrient excretion and biodeposition.
5. Effects of mussels on sediment and shell periphyton concentrations, organic matter
concentrations and invertebrate abundance, varied seasonally, and were strongest in late
summer during periods of low water volume, low flow, and high water temperature.
6. Our study demonstrates that freshwater mussels can strongly influence the co-occurring
benthic community, but that effects of mussels are context-dependent and may vary
among species.
Keywords: context dependent, ecosystem engineer, ecosystem function, macroinvertebrate, periphy-

ton, Unionidae
Correspondence: Daniel E. Spooner, Oklahoma Biological Survey and Department of Zoology, University of Oklahoma, Norman, OK
73019, U.S.A. E-mail: [email protected]
1016 2006 The Authors, Journal compilation 2006 Blackwell Publishing Ltd
13652427, 2006, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2427.2006.01547.x by University Of Jyväskylä Library, Wiley Online Library on [24/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Mussel effects on stream benthic communities 1017
with which they are most closely associated, the co-
Introduction
occurring benthic community; however, the effects of
In many shallow-water marine systems, filter-feeding unionids on the rest of the benthic community have not
bivalves dominate the benthic biomass and serve as a been investigated.
link between pelagic and benthic compartments by Here we describe the results of a year-long field
filtering large quantities of phytoplankton and fine experiment that examined the effects of unionid
particulate organic matter (FPOM) from the water mussels on other benthic invertebrates and periphy-
column and biodepositing organic matter to the ton occurring in both the sediment and on mussel
sediment (Dame, 1996; Gutierrez et al., 2003). Marine shells. Our experiment addresses three fundamental
bivalve aggregations increase secondary space and questions: (i) do mussels influence the composition
thus habitat for colonising epifauna, and bivalve and abundance of the benthic community and if so,
activities such as nutrient cycling and transformation, are effects the results of processes performed by living
biodeposition, and bioturbation enhance infaunal mussels or merely the results of habitat provided by
communities (Dame, 1996; Peterson & Heck, 1999). shells?, (ii) do different mussel species have different
Comparative studies of the ecological roles of fresh- effects on the benthic community? and (iii) do effects
water bivalves have been primarily limited to the of mussels on the benthic community vary with
epifaunal zebra mussel (Dreissena polymorpha) and environmental context (season)?
infaunal Asian clam (Corbicula fluminea), both invasive
species in North America. These species can control
both phytoplankton dynamics and benthic inverte- Methods
brate community structure through a combination of
Study site
their biological activities (i.e. filter-feeding, nutrient
cycling, biodeposition) and the physical habitat We performed the experiment in the Kiamichi River, a
provided by their shells (Stewart, Miner & Lowe, medium-sized tributary (watershed area 4560 km2) of
1998; Strayer et al., 1999; Hakenkamp et al., 2001). the Red River in the Ouachita Mountains of south-
Freshwater mussels (Unionacea) are a guild of eastern Oklahoma, U.S. The Kiamichi is a relatively
benthic, burrowing, filter-feeding bivalves. In rivers, pristine river known for its high aquatic biodiversity
the biomass of healthy unionid assemblages can exceed (Master, Flack & Stein, 1998). The river harbours
the biomass of all other benthic organisms by an order approximately 30 species of mussels and there are no
of magnitude (Negus, 1966; Vaughn & Hakenkamp, documented extirpations of mussel species within the
2001) and production by mussels (range from 1 to 20 g last century (Vaughn & Pyron, 1995). Mussel beds
dry mass m)2 year)1) can equal that by all other typically occur in reaches several hundred metres long
macrobenthos (Strayer, 1994). Mussels filter phyto- with densities as high as 64 individuals m)2. Seasonal
plankton and other suspended material from the water discharge variation is high, ranging from an average
column, excrete nutrients back to the water column and 200 cm3 s)1 in August to 6000 cm3 s)1 in February
biodeposit organic material to the sediment as faeces (Fig. 1). Our study site, a shallow, 200 m stream reach
and pseudofaeces. By burrowing in the sediment they with a gravel/cobble streambed and homogenous
increase sediment water and oxygen content and depth and flow, was chosen to minimise any effects of
release nutrients from the sediment to the water habitat heterogeneity. The site was located between a
column. Finally, the physical presence of both living large upstream pool and a downstream riffle sequence
mussels and their spent shells stabilises sediment and and could only be conveniently accessed from private
likely creates habitat for other benthic organisms land, protecting the experiment from tampering.
(Vaughn & Hakenkamp, 2001; Strayer et al. 2004).
Given the processes they perform and their high
Experimental design
biomass, unionid mussels have the potential to have
strong effects in rivers where they are abundant, by We were interested in differentiating effects on the
modifying habitat and controlling the availability of benthic community of processes performed by live
resources to other organisms. We would expect them to mussels versus the mere physical presence of mussel
have particularly strong influences on the organisms shells and in determining differences between unionid
2006 The Authors, Journal compilation 2006 Blackwell Publishing Ltd, Freshwater Biology, 51, 1016–1024
1018 D.E. Spooner and C.C. Vaughn
7000 30 year average 2 m from shore and approximately 1.5 m apart (to
Monthly discharge 2000–2001
6000 minimise downstream cage-effects). The 1-month
Mean discharge (cm3 s–1)
5000 block was furthest downstream and the 12-month

Month 3
4000 block was furthest upstream. This design allowed us
3000 to sample from and remove enclosures at the end of a
2000 Month 1 time period, without disturbing enclosures for subse-
Month 12
1000
quent time treatments.
0
Enclosures were buried 15 cm into the streambed
and filled with homogenised sediment (see below),
August S O N D J F M A M J J August
2000 2001
so that the sediment in the enclosures was level with
the streambed and the upper 15 cm of the cage
Fig. 1 Mean monthly discharge (2000–2001 and 30 year average; extended into the water column. This design allowed
USGS data) near the study site. Arrows indicate sampling
movement of invertebrates in and out of enclosures
periods.
through both the sediment and water column, but
prevented immigration/emigration of mussels so
species. Our design consisted of four mussel treat- that we could maintain constant mussel densities
ments [live Actinonaias ligamentina (Lamarck 1819), live over time.
Amblema plicata (Say 1817), ‘sham’ mussels (see below) Prior to the experiment, sediment was extracted
and a mussel-free control (sediment alone)] and three from the riverbed and mixed in 246 L plastic trash
time treatments (1, 3 and 12 months), with each mussel cans to homogenise the distribution of invertebrates,
by time treatment combination replicated five times. organic matter and algae among treatments. Mussels
This design allowed us to use planned orthogonal were removed prior to homogenisation. We define
contrasts, to compare effects of live mussels sediment to encompass both the organic (invertebrate
(Actinonaias + Amblema) to sham mussels as well as and other organic matter) and inorganic fractions
examine differences between species. typically found in the benthic substrate. Live Actinon-
Actinonaias ligamentina and Amblema plicata are aias [mean wet weight (mg) 7.79 ± 0.39 SE] and
typical of the Interior Basin mussel fauna (Parmalee Amblema [mean wet weight (mg) 4.02 ± 0.16 SE] were
& Bogan, 1998), and together represent over 70% of collected at the site. Prior to placing mussels in
mussel biomass in the Kiamichi River (Vaughn & enclosures, periphyton and other biofilm were
Pyron, 1995). The species differ in phylogeny, mor- removed from their shells by scrubbing with a plastic
phology and behaviour, characteristics that might brush. Each mussel-treatment enclosure (Actinonaias,
influence their ecological role. Actinonaias (subfamily Amblema and sham) was stocked with 10 individuals,
Lampsilinae) has a smooth shell and is more active a density representative of the local assemblage
than Amblema (subfamily Ambleminae), which has a (Vaughn & Pyron, 1995). Four glass microscope slides
ridged shell and tends to be sedentary (Vaughn, Gido were placed on the streambed surface in each
& Spooner, 2004). Sham mussels were created by enclosure to allow measurement of benthic organic
filling clean, relict Actinonaias and Amblema shells with matter and periphyton.
sand, then gluing the shells together with non-toxic
epoxy.
Response variables
The experiment was performed using 60 enclosures
(50 cm · 50 cm · 30 cm deep) constructed with a Enclosures were placed in the river and stocked with
polyvinyl chloride (PVC; 3.3 cm schedule 40) pipe mussels in August 2000. Twenty enclosures were
frame and the sides and bottom encased in 2.5 cm removed and response variables measured after one
diameter wire poultry netting. To control for depth, (September 2000), three (November 2000) and 12
current velocity and substrate type, enclosures were (August 2001) months. The experiment was moni-
placed within one stream reach (200 m). Enclosures tored frequently to ensure that leaf packs and other
were placed in the stream reach within three blocks, debris did not significantly influence water velocities
with one block of 20 enclosures for each time in the cages. The following procedures were followed
treatment. Within each block, enclosures were located for each time-period. For each enclosure except
controls, five mussels were randomly selected, Shell response variables (invertebrate abundance,
removed and individually placed in a plastic bag chlorophyll a and organic matter) across treatments
with 500 mL of distilled water. Each mussel was (Actinonaias, Amblema, sham mussels and time-period)
scrubbed with a plastic brush for 2 min, creating a were compared using an A N C O V A with biomass as a
water-biofilm slurry that was subsampled (50 mL) covariate. Bonferonni multiple comparison proce-
and stored on ice for chlorophyll a determination. dures were performed to control for type I error.
Two sediment cores (10 cm wide by 8 cm deep) were We used Pearson product-moment correlation to
taken from each enclosure and preserved in 5% examine associations between response variables that
formalin. Glass slides were placed in a jar with might otherwise go unnoticed in an A N O V A design.
125 mL distilled water and stored on ice. We examined associations between organic matter,
In the laboratory, core samples were elutriated, chlorophyll a and invertebrate abundance in both the
passed through a 210 lm sieve, and invertebrates sediment and on shells.
identified and counted. Glass slides were scraped
with a razor blade into 125 mL of water and two
50 mL aliquots were removed. To determine organic Results
content, one of these aliquots was filtered through a
Sediment
47 mm, 0.45 lm glass fibre filter, dried at 105 C for
12 h, and ashed at 550 C for 1 h. The other aliquot Mean discharge at the study site was minimal in late
was analysed for chlorophyll a. Chlorophyll a samples summer when we began and ended the experiment,
of both the mussel shell slurry and glass slide slurry but increased considerably in late autumn and
were filtered as above. Chlorophyll a was extracted peaked in the winter (Fig. 1). Some glass slides and
with acetone and measured spectrophotometrically sham mussels were lost to high flow during the
with a correction for pheophytin. winter, thus these data were unavailable for sedi-
We recorded the shell length of all mussels used ment chlorophyll a and organic matter estimates in
in the experiment, and the tissue dry weight of a August 2001. Therefore, sediment organic matter for
subsample of these mussels. We used a dry weight- month 12 was estimated from a 125 mL sub-sample
shell length regression (y ¼ 1.06)14.14x, R ¼ 0.88) to of the sediment cores collected for invertebrate
estimate biomass of non-sacrificed mussels and enumeration. Although these different sampling
sham mussels. Response variables were standard- methods (glass slides versus cores) hindered tem-
ised to tissue dry mass to factor out potential poral comparisons in sediment organic matter, we
confounding effects of mussel size differences were able to test for differences between mussel
among treatments. treatments.
Sediment organic matter was significantly higher in
treatments with live mussels (biologically active) than
Data analyses
treatments with sham mussels or sediment alone
Sediment response variables (invertebrate abundance (non-biologically active). This effect varied seasonally,
in the core samples, chlorophyll a and organic matter with organic matter significantly higher in late sum-
on the glass slides) were compared among treatments mer (September 2000 F1,16 ¼ 6.637, P < 0.05 and
using planned orthogonal contrasts for each time- August 2001 F1,14 ¼ 5.897, P < 0.05), but not in
period (Toothaker 1993). Our a priori hypothesis was autumn (November 2000 F1,13 ¼ 1.256, P > 0.05;
that that living mussels should have stronger effects Fig. 2a). Organic matter was significantly higher in
than shells or sediment alone. We tested this hypo- the sediment surrounding Actinonaias than Amblema,
thesis by comparing ‘biologically active treatments’ even after correcting for biomass differences (F1,28 ¼
(Actinonaias + A. plicata) to ‘non-biologically active 11.019, P ¼ 0.003; Fig. 2b). Sediment chlorophyll a did
treatments’ (sham mussels + sediment). To examine not differ significantly in treatments with live mussels
differences in sediment response variables between versus sham mussels or sediment (September 2000
the two live mussel species treatments we used F1,16 ¼ 3.444, P > 0.05, November 2000 F1,13 ¼ 0.281,
A N C O V A with biomass as a covariate and time as a P > 0.05; Fig. 2c) or between mussel species (F1,18 ¼
separate factor. 0.145, P > 0.05; Fig. 2d).
(a) (b)
Chlorophyll a (µg cm–2) Organic matter (g cm–2)
0.010 0.010
Sham mussel / sediment treatments Actinonaias
Live mussel treatment * Amblema *
0.008 0.008
0.006 0.006
0.004 0.004
0.002 * 0.002 *
0.000 0.000
(c) 0.04 (d)

0.04
0.03
0.03
0.02 0.02
0.01 0.01
0.00 0.00
(e) * * (f) 675

450
600
375 525
invertebrates
Fig. 2 Mean + 1 SE for sediment response

Number of
300 450
375 variables. (a), (c) and (e) are A N O V A
225 300 comparisons of live mussel versus sham
150 225 mussel/sediment treatments. (b), (d) and
150 (f) are A N C O V A comparisons of the two
75
75
0 0 species. Asterisks represent significant
Sep Nov Aug Sep Nov Aug differences from multiple comparison
2000 2000 2001 2000 2000 2001 procedures.
Overall, there were more invertebrates in the September 2000 but not November 2000 (Fig. 3a; sham
sediment of live mussel treatments than in sham mussels were not available for August 2001). When
mussel and sediment treatments. These differences standardised for biomass, Amblema had higher chlo-
were significant for summer months (September 2000 rophyll a concentrations on their shells than Actinon-
F1,16 ¼ 6.12, P < 0.05; August 2001 F1,12 ¼ 10.119, aias (Fig. 3a).
P < 0.05), but not in November (F1,14 ¼ 0.587, Overall, total invertebrate abundance was not sig-
P > 0.05; Fig. 2e). We found no significant differences nificantly different on the shells of live and sham
in sediment invertebrate abundance between Actinon- mussels (F2,31 ¼ 1.319, P > 0.05). There were seasonal
aias and Amblema treatments (F1,27 ¼ 1.761; Fig. 2f). differences, with higher invertebrate abundance on
In live mussel treatments we found significant both live and sham mussel shells in September 2000
correlations between the amount of organic matter (F2,31 ¼ 32.740, P < 0.001; Fig. 3b). At a finer taxo-
and chlorophyll a in the sediment (r ¼ )0.649, P ¼ nomic scale, there were significantly more tardigrades
0.003), organic matter and invertebrate abundance in (F2,31 ¼ 4.081, P ¼ 0.031) and mites (F2,31 ¼ 4.712,
the sediment (r ¼ 0.491, P ¼ 0.008), and the amount P ¼ 0.02) on the shells of Amblema and Actinonaias
of chlorophyll a on the sediment and invertebrate than on sham mussel shells.
abundance (r ¼ )0.452, P ¼ 0.052). There were no Mussel biomass (size) was correlated with inver-
significant correlations among sediment response tebrate abundance in the live mussel treatments (r ¼
variables in the sham mussel or sediment treatments. 0.51, P ¼ 0.026), but not in the sham mussel treat-
ments. Despite a marginally significant correlation
between shell invertebrate abundance and chloro-
Shells
phyll a (r ¼ 0.49, P ¼ 0.067), there were no other
Chlorophyll a was significantly higher on the shells of significant correlations among shell response varia-
living than sham mussels (F2,31 ¼ 3.684, P ¼ 0.037) in bles in either the living or sham mussel treatments.
(a) 0.16 nutrient content and was a function of increased

* mussel density, with pore water ammonia and phos-
µg chlorophyll a g–1 mussel
0.14 *
0.12 Actinonaias phate concentration four times higher in the densest
0.10
Amblema mussel beds. Radziejewska (1986) documented a dis-
dry mass
Sham tinct meiofauna community of higher abundance in the

0.08
0.06 sediments of marine mussel beds compared with

0.04 adjacent areas of sediment. Meiofauna appeared to be
0.02 responding to higher levels of organic material from
0.00 biodeposits. Mussels and other invertebrates may
interact synergistically to increase organic matter in
(b) 350 the sediment (Hakenkamp & Morin, 2000). In addition,
Total number of invertebrates
300 mussels stabilise the substrate (Strayer, 1999), which

250 would increase the retention time of organic matter in
200
the sediment (Hakenkamp & Morin, 2000).
150
Oligochaetes are benthic worms that feed on sedi-
ment organic matter (Brinkhurst & Gelder, 2000).
100
Oligochaete abundance in the sediment was higher in
50
live than sham mussel or sediment treatments. Living
0
Sep Nov Aug mussels may facilitate oligochaetes by biodepositing
2000 2000 2001
food in the form of faeces and pseudofaeces and by
Fig. 3 Mean + 1 SE for shell response variables. Results shown bioturbating the sediment, which decreases compac-
are for A N C O V A comparisons of Actinonaias ligamentina and tion and provides higher quality habitat. Ephemerop-
Amblema plicata with sham mussels. Asterisks represent signifi- terans in the sediment were significantly higher in live
cant differences from multiple comparison procedures.
than sham mussel or sediment treatments and likely
responding to increases in organic matter in the
Discussion
sediment (Merritt & Cummins, 1996).
We found that live unionid mussels influenced the Periphyton abundance, as represented by chloro-
distribution and abundance of other benthic organ- phyll a concentration, was higher on live than sham
isms, periphyton and invertebrates, in both the sur- mussel shells. Previous work at the study site suggests
rounding sediment and on mussel shells. Differences that the Kiamichi River may undergo periods of both
between living and sham mussel treatments indicate phosphorous and nitrogen limitation (C.C. Vaughn
that biological activities of mussels provide ecosystem and D.E. Spooner, unpubl. data). Thus, higher
services to the benthic community beyond the phys- periphyton abundance on living mussels likely rep-
ical habitat provided by shells alone. Mechanisms by resents a response of nutrient-limited algae to local
which mussels likely provide these ecosystem services nutrient excretion by mussels. Algal abundance on
include biodeposition of faeces and pseudofaeces, glass slides placed on the sediment did not differ
excretion of nutrients, and bioturbation of sediments among treatments. However, we think the glass slide
(Vaughn & Hakenkamp, 2001; Vaughn et al., 2004). methodology underestimated local effects of mussel
Both organic matter and invertebrate abundance excretion. Slides were placed near the corners of each
were higher in live than sham mussel or sediment enclosure, but mussels were rarely in enclosure
treatments. These results suggest that colonising inver- corners. Assuming excretion of nutrients causes local
tebrates are responding to higher levels of biodepos- increases in periphyton, slides may have been located
ited organic matter and excreted nutrients in live too far from mussels for periphyton on the slides to
mussel treatments. Our field observations of both respond to nutrient excretion. Nutrients directly
Actinonaias and Amblema support this conclusion; excreted from the exhalant siphon of mussels may
discrete piles of faeces/pseudofaeces are often be taken up by periphyton on mussel shells, whereas
observed beside mussel exhalent siphons in the nutrients would likely be sequestered before reaching
streambed. Peterson & Heck (1999) showed that glass slides (Blumenshine et al., 1997). Water flow
biodeposits from marine mussels increased pore water likely magnified this effect by diluting nutrients
excreted by mussels or by transporting nutrients effects in late summer (September 2000 and August
downstream, away from the enclosures. 2001) and minimal to no detectable effects in mid
Invertebrate and periphyton abundance were posi- autumn (November 2000). These differences can be
tively correlated on the shells of live but not sham attributed to seasonal differences in depth, discharge,
mussels, indicating that invertebrates are responding to and temperature in the river. At the beginning and
increased algal abundance on living mussel shells as end of the experiment water levels were quite low
food and/or shelter. Invertebrate abundance and mus- (30 cm depth), flow was minimal (Fig. 1), and water
sel size were positively correlated for shells of live but temperatures were high (as high as 41 C). Under
not sham mussels, and invertebrate richness was higher these shallow, low-flow conditions, materials excreted
on larger mussels. Higher abundance and richness of and biodeposited by mussels would remain in the
invertebrates on larger mussels might be a reflection of mussel bed where they could be used by the rest of
both increased habitat availability (Beckett, Green & the benthic community (Vaughn et al., 2004), as
Thomas, 1996), as well as increased local ecosystem our data demonstrate. Higher temperatures would
services. For example, larger mussels excrete more increase mussel metabolic rates, which would in-
nutrients than smaller mussels (Vaughn et al., 2004). crease nutrient excretion and biodeposition rates
Total invertebrate abundance on shells did not (Aldridge, Payne & Miller, 1995; McMahon & Bogan,
differ between treatments, but there were some 2001). In contrast in October and November our study
differences among taxonomic groups. Significantly site experienced high flows (Fig. 1), with large increa-
more mites and tardigrades were found on the shells ses in both current velocity and water volume. High
of live mussels than sham mussels. Tardigrades flows likely scoured organic matter, algae, and inver-
typically live on plants or the sediment and feed on tebrates from both sediment and shells. Flow remained
algae (Nelson & Marley, 2000), thus they may be relatively high throughout the winter and spring,
responding to local increases in algae on live mussel decreasing in June (Fig. 1) and allowing mussel effects
shells as both food and habitat, whereas adult mites to again predominate. These results are corroborated
are likely responding to increased habitat availability by other studies demonstrating that organic matter
(Di Sabatino, Gerecke & Martin, 2000). dynamics in streams are governed by seasonal flow
After correcting for biomass, Amblema had more conditions (Brennan, McLachlan & Wotton, 1978;
algae on their shells than Actinonaias. We believe this Palmer et al., 1997) and that the ability of freshwater
reflects both morphological and behavioural differ- bivalves to influence ecosystem processes decreases
ences between the species. Throughout the experiment with increases in flow and water volume (Strayer
Actinonaias was more active; individuals moved about et al., 1999; Vaughn et al., 2004).
enclosures and burrowed up and down in the sedi- This study demonstrates that riverine, unionid
ment. These activities would tend to inhibit algal mussels influence the co-occurring benthic commu-
colonisation and slough off attached algae. In contrast, nity, but that effects of mussels are context-dependent
Amblema individuals were sedentary and generally and stronger during periods of low water volume,
stayed in one location, with part of the shell always low flow and higher water temperatures. Our results
exposed to the water column and thus sunlight, are based on standing crop estimates of production
encouraging algal growth. In addition, shells of Amble- (chl a and organic matter) and community structure
ma are composed of multiple ridges while Actinonaias (benthic invertebrates in the sediment and on shells).
shells are smooth (McMahon & Bogan, 2001). Ridges While these data provide an important first step, our
are thought to help mussels maintain position during understanding of how unionids influence the rest of
high flow events (Watters, 1994) and might present the benthic community will be greatly strengthened
algae a refuge from grazing and/or flow providing by studies that track processing rates of energy and
increased habitat heterogeneity and area. Amblema had nutrients. While some comparative, field estimates of
more mites and tardigrades on their shells than unionid processing rates have been made (Nichols
Actinonaias, which probably is a result of the increased and Garling 2000, Raikow and Hamilton 2001, Chris-
algal food and habitat resources on Amblema shells. tian et al. 2004), it is now imperative to examine these
Effects of mussels on the benthic community varied rates experimentally and under different environmen-
greatly with season. Overall, mussels had strong tal contexts.
Species whose abundance and/or biomass domin- for statistical advice. Comments from G. Wellborn,
ates an ecosystem often have corresponding strong W.J. Matthews, H.S. Galbraith, B. Downes and J.D.
effects on ecosystem function by causing physical Ackerman improved the study design and/or manu-
state changes in biotic or abiotic materials and script. Funding was provided by the Department of
controlling the availability of resources to other Zoology and Graduate College at the University of
organisms (Levinton, 1995; Gutierrez et al., 2003; Lill Oklahoma and by NSF grants DEB-9870092 and
& Marquis, 2003; Statzner, Peltret & Tomanova, 2003). DEB-0211010 to Vaughn.
Examples of such ‘ecosystem engineers’ include
detrital feeding fish (Flecker, 1996), beavers (Wright,
References
Jones & Flecker, 2002), suspension-feeding blackfly
larvae (Wotton et al., 1998), marine bivalves (Dame, Ackerman J.D., Loewen M.R. & Hamblin P.F. (2001)
1996) and zebra mussels (Strayer et al., 1999; Benthic-pelagic coupling over a zebra mussel reef in
Ackerman, Loewen & Hamblin, 2001). Based on our western Lake Erie. Limnology and Oceanography, 46,
results and other recent studies (Vaughn et al., 2004) 892–904.
Aldridge D.W., Payne B.S. & Miller A.C. (1995) Oxygen
we think that unionid mussels potentially act as
consumption, nitrogenous excretion, and filtration
‘ecosystem engineers’ in rivers where they are abun-
rates of Dreissena polymorpha at acclimation tempera-
dant because they both modify the habitat and control
tures between 20 and 32 C. Canadian Journal of
availability of resources to other organisms. Further Fisheries and Aquatic Sciences, 52, 1761–1767.
research in this area is needed. Beckett D.C., Green B.W. & Thomas S.A. (1996) Epizoic
Given the recent body of research on the ecological invertebrate communities on upper Mississippi River
role of invasive freshwater bivalves (i.e. zebra mussels unionid bivalves. American Midland Naturalist, 135,
and Asian clams), it is tempting to make comparisons 102–114.
between the effects of the invasive and native fauna. Blumenshine S.C., Vadeboncoeur Y., Lodge D.M.,
However, inferences drawn from studies of invasive Cottingham K.L. & Knight S.E. (1997) Benthic-pelagic
species that downplay the role of native mussels links: responses of benthos to water-column nutrient
should be regarded with skepticism. Native mussels enrichment. Journal of the North American Benthological
Society, 16, 466–479.
occupy different niches (e.g. zebra mussels are epifa-
Bogan A.E. (1993) Freshwater bivalve extinctions (Mol-
unal), are longer lived and thus provide ecosystem
lusca: Unionidae): a search for causes. American Zoolo-
services on a much longer temporal scale, and are an
gist, 33, 599–609.
integral component of many freshwater systems. This Brennan A., McLachlan A.J. & Wotton R.S. (1978)
study represents the first empirical test and confir- Particulate material and midge larvae (Chironomidae,
mation of ecological benefits provided by native Diptera) in an upland river. Hydrobiologia, 59, 67–74.
freshwater mussels. Brinkhurst R. & Gelder S.E. (2000) Annelida: Oligochaeta
Freshwater mussels are threatened and declining and Branchiobdellida. In: Ecology and Classification of
globally (Bogan, 1993). Historically, mussels domin- North American Freshwater Invertebrates (Eds J.H. Thorp
ated the biomass of rivers in eastern North America & A.P. Covich), pp. 431–433. Academic Press, New
(Parmalee & Bogan, 1998), but in recent years York, NY.
populations of both rare and common species have Christian A.D., Smith B.N., Berg D.J., Smoot J.C. &
Findley R.H. (2004) Trophic position and potential
undergone catastrophic declines (Vaughn & Taylor,
food sources of 2 species of unionid bivalves (Mollus-
1999). Our results, and results of other recent studies
ca: Unionidae) in 2 small Ohio streams. Journal of North
of ecosystem services performed by riverine bivalves
American Benthological Society, 23, 101–113.
(Strayer et al., 1999; Vaughn et al., 2004), indicate that Dame R.F. (1996) Ecology of Marine Bivalves: an Ecosystem
this catastrophic loss of mussel biomass may lead to Approach. CRC Press, New York.
changes in the functioning of river ecosystems. Di Sabatino A., Gerecke R. & Martin P. (2000) The
biology and ecology of lotic water mites (Hydrachni-
dia). Freshwater Biology, 44, 47–62.
Acknowledgments
Flecker A.S. (1996) Ecosystem engineering by a dominant
We thank R. Cothran, J. Johnson and J. Golan for field detritivore in a diverse tropical stream. Ecology, 77,
assistance, K. Roberts for river access, and M. Patten 1845–1854.
Gutierrez J.L., Jones C.G., Strayer D.L. & Iribarne O.O. perspectives on pearly mussels, North America’s most
(2003) Mollusks as ecosystem engineers: the role of imperiled animals. BioScience, 54, 429–439.
shell production in aquatic habitats. Oikos, 101, 79–90. Statzner B., Peltret O. & Tomanova S. (2003) Crayfish as
Hakenkamp C.C. & Morin A. (2000) The importance of geomorphic agents and ecosystem engineers: effect of
meiofauna to lotic ecosystem functioning. Freshwater a biomass gradient on baseflow and flood-induced
Biology, 44, 165–175. transport of gravel and sand in experimental streams.
Hakenkamp C.C., Ribblett S.G., Palmer M.A., Swan C.M., Freshwater Biology, 48, 147–163.
Reid J.W. & Goodison M.R. (2001) The impact of an Stewart T.W., Miner J.G. & Lowe R.L. (1998) Quantifying
introduced bivalve (Corbicula fluminea) on the benthos mechanisms for zebra mussel effects on benthic
of a sandy stream. Freshwater Biology, 46, 491–501. macroinvertebrates: organic matter production and
Levinton J.S. (1995) Bioturbators as ecosystem engineers: shell-generated habitat. Journal of the North American
control of the sediment fabric, inter-individual inter- Benthological Society, 17, 81–94.
actions, and material fluxes. In: Linking Species and Strayer D.L. (1994) Body size and abundance of benthic
Ecosystems (Eds C.G. Jones & J.H. Lawton), pp. 29–38. animals in Mirror Lake, New Hampshire. Freshwater
Chapman and Hall, New York. Biology, 32, 83–90.
Lill J. & Marquis R. (2003) Ecosystem engineering by Strayer D.L. (1999) Use of flow refuges by unionid
caterpillars increases insect herbivore diversity on mussels in rivers. Journal of the North American Bentho-
white oak. Ecology, 84, 682–690. logical Society, 18, 468–476.
Master L.M., Flack S.R. & Stein B.A. (1998) Rivers of Life: Strayer D.L., Caraco N.F., Cole J.J., Findley S. & Pace M.L.
Critical Watersheds for Protecting Freshwater Diversity. (1999) Transformation of freshwater ecosystems by
The Nature Conservancy, Arlington, VA. bivalves. BioScience, 49, 19–27.
McMahon R.F. & Bogan A.E. (2001) Mollusca: Bivalvia. Toothaker L.E. (1993) Multiple Comparison Procedures.
In: Ecology and Classification of North American Fresh- Sage Publications, Newbury Park, California. pp. 1–95.
water Invertebrates (Eds J.H. Thorp & A.P. Covich), pp. Vaughn C.C. & Hakenkamp C.C. (2001) The functional
230–250. Academic Press, New York, NY. role of burrowing bivalves in freshwater ecosystems.
Merritt R.W. & Cummins K.W. (1996) An Introduction to Freshwater Biology, 46, 1431–1446.
the Aquatic Insects of North America, 3rd edn. Kendall- Vaughn C.C. & Pyron M. (1995) Population ecology of
Hunt, Dubuque, IA. the endangered Ouachita Rock Pocketbook mussel,
Negus C.L. (1966) A quantitative study of growth and Arkansia wheeleri (Bivalvia: Unionidae), in the Kiamichi
production of unionid mussels in the river Thames at River, Oklahoma. American Malacological Bulletin, 11,
Reading. Journal of Animal Ecology, 35, 513–532. 145–151.
Nelson D.R. & Marley N.J. (2000) The biology and Vaughn C.C. & Taylor C.M. (1999) Impoundments and
ecology of lotic Tardigrada. Freshwater Biology, 44, the decline of freshwater mussels: a case study of an
93–108. extinction gradient. Conservation Biology, 13, 912–
Nichols S. & Garling D. (2000) Food-web dynamics and 920.
trophic-level interactions in a multispecies community Vaughn C.C., Gido K.B. & Spooner D.E. (2004) Ecosystem
of freshwater unionids. Canadian Journal of Zoology, 78, processes performed by unionid mussels in stream
871–882. mesocosms: species roles and effects of abundance.
Palmer M.A., Covich A.P., Finlay B.J. et al. (1997) Hydrobiologia, 527, 35–47.
Biodiversity and ecosystem processes in freshwater Watters G.T. (1994) Form and function of unionoidean
sediments. Ambio, 26, 571–577. shell sculpture and shape (Bivalvia). American Malac-
Parmalee P.W. & Bogan A.E. (1998) The Freshwater ological Bulletin, 11, 1–20.
Mussels of Tennessee. University of Tennessee, Knox- Wotton R.S., Malmqvist B., Muotka T. & Larsson K.
ville. (1998) Fecal pellets from a dense aggregation of
Peterson B.J. & Heck K.L. (1999) The potential for suspension-feeders in a stream: An example of
suspension feeding bivalves to increase seagrass ecosystem engineering. Limnology and Oceanography,
productivity. Journal of Experimental Marine Biology 43, 719–725.
and Ecology, 240, 37–52. Wright J.P., Jones C.G. & Flecker A.S. (2002) An
Raikow D.F. & Hamilton S.K. (2001) Bivalve diets in a ecosystem engineer, the beaver, increases species
Midwestern U.S. stream. A stable isotope enrichment richness at the landscape scale. Oecologia, 132, 96–101.
study. Limnology and Oceanography, 46, 514–522.
Strayer D.L., Downing J.A., Haag W.R., King T.L., Layzer (Manuscript accepted 28 February 2006)
J.B., Newton T.J. & Nichols S.J. (2004) Changing

Freshwater Biology - 2006 - SPOONER - Context Dependent Effects of Freshwater Mussels On Stream Benthic Communities

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Freshwater Biology - 2006 - SPOONER - Context Dependent Effects of Freshwater Mussels On Stream Benthic Communities

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Freshwater Biology - 2006 - SPOONER - Context Dependent Effects of Freshwater Mussels On Stream Benthic Communities

Uploaded by

Copyright:

Available Formats

Freshwater Biology (2006) 51, 1016–1024 doi:10.1111/j.1365-2427.2006.01547.

Context-dependent effects of freshwater mussels on

Keywords: context dependent, ecosystem engineer, ecosystem function, macroinvertebrate, periphy-

5000 block was furthest downstream and the 12-month

(c) 0.04 (d)

(e) * * (f) 675

Fig. 2 Mean + 1 SE for sediment response

(a) 0.16 nutrient content and was a function of increased

Sham tinct meiofauna community of higher abundance in the

0.06 sediments of marine mussel beds compared with

300 mussels stabilise the substrate (Strayer, 1999), which

You might also like