The benthification of freshwater lakes: Exotic
mussels turning ecosystems upside down
Collaborators
Patty Armenio**
Thomas Bridgeman
Kristen DeVanna**
University of Toledo
Lake Erie Center
Lars Rudstam
Ed Mills
Cornell University
Bin Zhu**
University of Hartford
Kim Schulz
Rebecca Cecala**
Peibing Qin**
Xinli Ji
SUNY ESF/Syracuse University
Funding sources
NOAA (NY) Sea Grant; Lake Erie Protection Fund
Dreissenid Mussels & Benthification
 High population density
 High filtration rate
Increased water clarity
 Form clusters
Increased benthic structure
Dreissena are Ecosystem Engineers
Organisms that directly or indirectly modulate the
availability of resources to other species, by causing
physical state changes in biotic or abiotic materials.
Jones et al. 1994
Ecology, Ricklefs & Miller, 2000
Population Interactions
Resources and Consumers
Competition Theory
Competition in Nature
Predation
Herbivory and Parasitism
Coevolution and Mutualism
Suite of expected changes with benthification
eutrophic
light penetration
benthified
light penetration
 Limited benthic production
Extensive benthic production
 Low benthic complexity
 High benthic complexity
 Low foraging efficiency by
benthic fish (low benthic to
pelagic flux)
 High foraging efficiency by
benthic fish (high benthic to
pelagic flux)
Dreissena affect lakes at multiple spatial scales
and across trophic levels
System level
Increased light penetration
Local level
Resource importation
Structural complexity
•Benthic PP
•Benthic grazers
•Visual foragers
Benthic Processes Affected by Dreissena
 Primary production
Macrophytes: System-wide
Algae: System-wide & Local
 Benthic populations
Microbes: Local
Macroinvertbrates: System-wide & Local
 Visual foragers: System-wide & Local
Benthic Processes Affected by Dreissena
 Primary production
Macrophytes: System-wide
Algae: System-wide & Local
 Benthic populations
Microbes: Local
Macroinvertbrates: System-wide & Local
 Visual foragers: System-wide & Local
Oneida Lake
W. basin
you are here
Maximum plant depth (m)
SAV maximum depth increased after Dreissena
Species evenness increased
Myriophylum only spp to decrease
7
Diver survey
Hydroacoustic survey
6
5
4
R2 = 0.74
p=0.001
3
2
1
0
1975
1980
1985
1990
1995
2000
2005
Zhu et al. 2006 Ecosystems 9:1-12
Benthic Processes Affected by Dreissena
 Primary production
Macrophytes: System-wide
Algae: System-wide & Local
Oneida Lake
Experiments
Local mechanisms
 Benthic populations
Microbes: Local
Macroinvertbrates: System-wide & Local
 Visual foragers: System-wide & Local
Response of benthic algal primary production
to increased clarity in Oneida Lake
o Whole lake GPP 2003 & 2004
oLight response curves (N ~ 200) &
modified Fee model to estimate
production
o Back casting using long-term
clarity data
Substrate and depth
map of Oneida Lake
Legend
<all other values>
S_D_NAME
H(0-6f)
H(7-12f)
H(13-18f)
H(19-24f)
H(25-30f)
H(31-35f)
H(36-40f)
SD(0-6f)
SD(7-12f)
SD(13-18f)
SD(19-24f)
SD(25-30f)
SD(31-35f)
SD(36-40f)
SF(0-6f)
SF(7-12f)
SF(13-18f)
SF(19-24f)
SF(25-30f)
SF(31-35f)
SF(36-40f)
SF(41-45f)
SF(46-50f)
SF(51-55f)
Becky Johnson-Cecala; Cecala et al. 2008. J. Integrated Plant Biology 50:1452-1466
Diver collecting surface core to measure
primary production on soft sediment
52,000
51,000
50,000
49,000
48,000
Benthic GPP
2003
1999
1995
1991
1987
1983
1979
47,000
1975
Average benthic production ( Kg C/day)
Whole-lake summer benthic GPP has increased and become
less variable
~4% increase, was net change & included areas of
reduced production due to photoinhibiton
51,500
Average kgC/day
51,000
50,500
50,000
49,500
49,000
48,500
48,000
Pre
Post
52,000
1
51,000
0.8
50,000
0.7
0.6
49,000
0.5
48,000
Benthic GPP
Light attenuation
0.4
0.3
2003
1999
1995
1991
1987
1983
1979
47,000
Avg July attenuation (k)
0.9
1975
Average benthic production ( Kg C/day)
Benthic GPP has become less variable following changes in
attenuation
Benthic Processes Affected by Dreissena
 Primary production
Macrophytes: System-wide
Algae: System-wide & Local
Oneida Lake
Experiments
Local mechanisms
 Benthic populations
Microbes: Local
Macroinvertbrates: System-wide & Local
 Visual foragers: System-wide & Local
Experimental Approach: light x Dreissena x P x other grazers
w/ Kim Schulz, Peibing Qin, Xinli Xi
Fluorometric
measure of
photosynthesis
Electron Transport
Rate (ETR) =
proportional to
photosynthesis
 System wide and localized effects of Dreissena
on Cladophora-dominated algal community
High light
High P
Among treatments
Within treatments
Low light
Low P
Systemwide
effects
Localized
effects
 Light (but not P) strongly affected both NPP and ETR
3
25
2.5
2
1.5
1
0.5
0
low
high
NPP (mgO2/m2/h)
ETR (µmol e-m-2s-1)
Non-colonized rocks (system wide effect)
Light Level
20
15
10
5
0
-5
low
high
 Dreissena sequestered P in experimental mesocosms,
mimicking near shore shunt
TP in water (µg L-1)
90
Light
Dreissena
P
Other grazers
80
70
60
50
40
30
20
10
0
high
low
yes
no
high
treatment
low
high
low
Localized Effects
ETR higher with Dreissena at both high & low light
Statistical model accounts for other- treatment variance
ETR (µmol e-m-2s-1)
Difference small compared to light effect
3
2.5
2
1.5
1
0.5
0
no ZM
with ZM
High light
no ZM
with ZM
Low light
Benthic Processes Affected by Dreissena
 Primary production
macrophytes: System-wide
algae: System-wide & Local
Oneida Lake
Experiments
Local mechanisms
 Benthic populations
microbes: Local
macroinvertbrates: System-wide & Local
 Visual foragers: System-wide & Local
Do Dreissena contribute nutrients to promote algal blooms?
Manipulative experiment with Lyngbya wollei
Patricia Armenio; MS student U. Toledo
Live Dreissena (N=10)
Dreissena shells (N=10)
Pottery shards (N=10)
Sand (N=10)
6
400
5
Carbon (mg/g)
Phosphorus (mg/g)
No mass change, but Dreissena contributed some
nutrients to Lyngbya
4
3
2
350
300
250
200
150
100
1
50
0
0
4
40
Potassium (mg/g)
Nitrogen (mg/g)
45
35
30
25
20
15
10
3.5
3
2.5
2
1.5
1
5
0.5
0
0
live
mussels
mussel
shells
pottery
shards
substrate type
sand
live
mussels
mussel
shells
pottery
shards
substrate type
sand
However…
160
120
100
80
60
40
20
5
0
4.5
live
mussel
mussels shells
pottery
shards
substrate type
sand
Chl a: dry weight ratio
Calcium (mg/g)
140
4
3.5
3
2.5
2
1.5
1
0.5
0
live
mussels
mussel
shells
pottery
shards
substrate type
sand
Lyngbya density in western Lake Erie
Conclusions
1. Increased water clarity, hence bottom light promotes
increased benthic PP, both plants and algae
2. Dreissena also increase benthic algal photosynthesis
at local scale
3. Transfer of nutrients is a possible mechanism for
local-scale effects– work in progress
4. When ZM aggregations are large, they may elevate
water column P and other nutrients and thereby
increase benthic algal photosynthesis, Near Shore
Shunt
Benthic Processes Affected by Dreissena
 Primary production
Macrophytes: System-wide
Algae: System-wide & Local
Oneida Lake
Experiments
Local mechanisms
 Benthic populations
Microbes: Local
Macroinvertbrates: System-wide & Local
Hard substrate-Oneida Lake
Soft substrate-western Lake Erie
 Visual foragers: System-wide & Local
Rocky site
0.8
background
experimental
0.4
0.0
2.0
Mud site
3.35
1.5
1.0
Hydra
Planariidae
Oligochaeta
Isopoda
Hirudinidae
Sphaeriidae
Chir
0.0
Gastropoda
0.5
Amphipoda
Invertebrate density (no/cm2)
Dreissena attached to hard substrate increase invertebrate
density, especially on soft background
Mayer et al. 2002 JNABS
Combined lake-wide and local effects of Dreissena may
favor grazers and predators
detritivores
+; local, structure
-; soft sediment
dwellers; loss of
detritus
predators
grazers
+ lake-wide and
local; refuge
from predators
and higher prey
+ local;
structure
+ lake-wide;
higher benthic
primary
production
periphyton
Water clarity and benthic GPP have increased in Oneida Lake
Mean annual Secchi Depth (m)
5.0
4.0
3.0
2.0
1.0
0.0
1975
1980
1985
1990
Year
1995
2000
2005
% of total at shallow (3.5m) station
As water clears, relatively more amphipods, but not
chironomids found at shallow station
90
y = 13.641x + 20.821
R2 = 0.28
p=0.005
80
70
60
y = 0.0234x + 50.847
R2 <0.00
p=0.995
50
40
30
20
chironomids
amphipods
10
0
1.5
2.0
2.5
3.0
3.5
4.0
Mean annual Secchi depth (m)
4.5
5.0
Periodic intensive survey of embayment shows predatory
taxa higher in Dreissena-colonized habitats
16000
14000
density (no*m-2)
no zm
12000
with zm
10000
8000
6000
4000
2000
0
Predators
Total
1995
Predators
Total
2000
Percent predatory taxa consistently higher with Dreissena
18.00
14000
16.00
12000
no zm
with zm
14.00
% predators
density (no*m-2)
16000
10000
8000
6000
12.00
10.00
8.00
6.00
4000
4.00
2000
2.00
0
0.00
Predators
1995
Total
Predators
2000
Total
1995
2000
Combined lake-wide and local effects of Dreissena may
favor grazers and predators
detritivores
+; local, structure
-; soft sediment
dwellers; loss of
detritus
predators
grazers
+ lake-wide and
local; refuge
from predators
and higher prey
+ local;
structure
+ lake-wide;
higher benthic
primary
production
periphyton
Combined lake-wide and local effects of Dreissena may
favor grazers and predators
Underestimating secondary
production?
detritivores
+; local, structure
-; soft sediment
dwellers; loss of
detritus
predators
grazers
+ lake-wide and
local; refuge
from predators
and higher prey
Distributional
shift observed
+ local;
structure
periphyton
Benthic Processes Affected by Dreissena
 Primary production
Macrophytes: System-wide
Algae: System-wide & Local
Oneida Lake
Experiments
Local mechanisms
 Benthic populations
Microbes: Local
Macroinvertbrates: System-wide & Local
Hard substrate-Oneida Lake
Soft substrate-western Lake Erie
 Visual foragers: System-wide & Local
Kristen DeVanna; PhD student U. Toledo
Habitat Type Selection
Objective
Dreissena clusters on hard substrate repeatedly shown to
elevate localized invertebrate density
How do Dreissena affect soft sediment
invertebrates such as Hexagenia?
?
Hexagenia habitat selection experiments
5 mayfly densities; 100 -1200/m2
Mayflies allowed to chose habitat
Removed after 48 hr
Bare
sediment
Artificial
clusters
Live
Dreissena
Hexagenia prefer live Dreissena clusters
Suggests resource importation
Only go to bare sediment at high density
% mayflies in habitat
100%
80%
60%
live mussels
artificial clusters
40%
bare sediment
20%
0%
100
200
400
800
1200
Mayfly density (number/m2)
DeVanna et al. in review
% Hexagenia in habitat
75% of Hexagenia inhabited highdensity Dreissena habitats
50%
B
B
40%
30%
A
20%
A
10%
0%
0%
25%
50%
100%
% Dreissena coverage
DeVanna et al. in review
Do Hexagenia always prefer Dreissena
habitat: hypoxia ?
H1
Hexagenia avoid Dreissena clusters during hypoxic
conditions
H2
Hexagenia leave burrows more often in hypoxic
conditions compared to normoxic conditions
O2
?
Does predation threat change Hexagenia
behavior under different O2 conditions?
H1
Hexagenia select for Dreissena clusters and leave
burrows less when fish predation threat is present
H2
Hexagenia leave burrows under low O2 even when
predation threat is present
O2 &
?
Methods
• Hexagenia Behavioral
Arenas
• 2 habitat types:
Dreissena and sediment
• Treatments
• Fish presence (N=5) vs. no fish (N=5)
• High oxygen vs. hypoxia (imposed at 24 hr to both
treatments)
Methods
• Day 1
• Introduced 6 mayflies
• Observed for 15
minutes (Initial)
X
• Day 2
• Initial Observations
(Pre-hypoxia)
• Lowered oxygen
(<30% saturation)
• Observed for 15
minutes (post-hypoxia)
• Observed after 3
hours of hypoxia
?
6
Initial Selection for Dreissenacovered sediment
4.5
Number of Hexagenia
4
3.5
3
2.5
No Fish
2
Yellow Perch
1.5
1
0.5
0
Bare Sediment
Dreissena
More Hexagenia were exposed in the
structured habitat during hypoxia
% Hexagenia fully exposed
No Fish Present
50
Bare Sediment
Dreissena
40
30
20
10
0
Pre-hypoxia
15 min of hypoxia
3 hours of hypoxia
Hexagenia waited longer to expose
themselves when fish were present
% Hexagenia fully exposed
Yellow Perch Present
Bare Sediment
50
Dreissena
40
30
20
10
0
Pre-hypoxia
15 min of hypoxia
3 hours of hypoxia
Hexagenia left their burrows during
hypoxia, but many were still sheltered
beneath cluster
Yellow Perch Present
% Hexagenia fully exposed
35
below cluster
in cluster
above cluster
30
25
20
15
10
5
0
Pre-hypoxia
15 min of hypoxia
3 hours of hypoxia
Conclusions:
Hexagenia select for habitat with Dreissena
May be exploiting microbial, algal or other
resources in clusters
Ongoing experiments to determine factors
controlling habitat choice, e.g. fish predation, hypoxia
Preliminary results show that
hypoxia results in neutral habitat
selection and Hexagenia exit
burrows and expose heads and
bodies
Trophic Cascade
Effects
transmitted
among trophic
levels; up, down,
or middle out
Ecosystem Engineering Cascade
Light
Direct effects on
consumers and
primary
producers