Reference: Biol. Bull. 213: 141–151. (October 2007)
© 2007 Marine Biological Laboratory
Morphological and Ecological Determinants of Body
Temperature of Geukensia demissa, the Atlantic
Ribbed Mussel, and Their Effects On Mussel
Mortality
JENNIFER JOST* AND BRIAN HELMUTH
Department of Biological Sciences and Belle W. Baruch Institute, University of South Carolina,
Columbia, South Carolina 29208
Abstract. Measurements of body temperatures in the field
have shown that spatial and temporal patterns are often far
more complex than previously anticipated, particularly in
intertidal regions, where temperatures are driven by both
marine and terrestrial climates. We examined the effects of
body size, body position within the sediment, and microhabitat (presence or absence of Spartina alterniflora) on the
body temperature of the mussel Geukensia demissa. We
then used these data to develop a laboratory study exposing
mussels to an artificial “stressful” day, mimicking field
conditions as closely as possible. Results suggested that G.
demissa mortality increases greatly at average daily peak
temperatures of 45 °C and higher. When these temperatures
were compared to field data collected in South Carolina in
the summer of 2004, our data indicated that mussels likely
experienced mortality due to high-temperature stress at this
site during this period. Our results also showed that body
position in the mud is the most important environmental
modifier of body temperature. This experiment suggested
that the presence of marsh grass leads to increases in body
temperature by reducing convection, overwhelming the effects of shading. These data add to a growing body of
evidence showing that small-scale thermal variability can
surpass large-scale gradients.
abundance and biodiversity is an important and increasingly
pressing issue. Evidence of the impacts of climate change
can be seen in a variety of environments, through influences
on population abundances, species ranges, and interactions
within ecological communities (Walther et al., 2002). Current models suggest that the rate of climate change is
accelerating (IPCC, 2001; Stainforth et al., 2005), and
changes in air, sea, and surface temperature are likely to
have great consequences for the planet’s ecosystems, especially in intertidal zones (Ray et al., 1992; Suchanek 1994)
where many organisms are thought to live close to their
thermal tolerance (Doty, 1946; Connell, 1972; Somero,
2002; Davenport and Davenport, 2005). Body temperature
has been shown to control local and latitudinal distributions
of many intertidal species (Wethey, 1983; Fields et al.,
1993; Bertness et al., 1999), and small changes in global air,
surface, and water temperatures are likely to cause shifts in
patterns of species distribution over a range of spatial scales
(e.g., Southward et al., 1995; Sagarin et al., 1999; Gilman et
al., 2006; Helmuth et al., 2006a,b). Therefore, intertidal
invertebrate species are potentially very good indicators of
the effects of global climate change on species distribution
patterns (Barry et al., 1995; Southward et al., 1995; Sagarin
et al., 1999; Harley et al., 2006, Helmuth et al., 2006a).
In many cases, however, we do not have sufficient information on the body temperatures of intertidal organisms in
the field to gauge how closely intertidal animals are living to
their thermal limits (Denny et al., 2006, Helmuth et al.,
2006a). Moreover, for many intertidal invertebrates, our
knowledge of lethal limits comes from studies conducted in
water, despite the fact that recent studies have suggested
Introduction
Quantitative prediction of the effects of global climate
change on species ranges and on patterns of organismal
Received 22 September 2006; accepted 9 May 2007.
* To whom correspondence should be addressed. E-mail: jostja@
unc.edu
141
142
J. JOST AND B. HELMUTH
that both aquatic and aerial temperatures may be important
to physiological performance and survival (Stillman, 2003;
Tomanek and Sanford, 2003; Somero, 2005; Denny et al.,
2006; Fields et al., 2006). During submersion at high tide,
organisms are likely to have body temperatures similar to
that of the surrounding water (Helmuth, 1998). In contrast,
at low tide, the organism must cope with the changing
conditions of the terrestrial environment. During aerial exposure, the body temperature of an intertidal invertebrate is
driven by multiple climatic factors such as wind speed, solar
radiation, cloud cover, and air and surface temperature, and
the animal’s temperature is often considerably warmer or
cooler than the surrounding air (Helmuth, 2002; Denny and
Harley, 2006). Significantly, body temperatures experienced during low tide have been shown to cause physiological damage (e.g., Hofmann and Somero, 1995; Stillman,
2003; Somero, 2002, 2005), suggesting that an understanding of geographic patterns in aerial body temperature is
crucial for estimating the effects of climate change on
intertidal species (Gilman et al., 2006; Helmuth et al.,
2006b). Moreover, we often have an incomplete understanding of what aspect of body temperature (maximum,
minimum, or time history) most strongly affects survival,
growth, and reproduction. Understanding and predicting the
effects of climate change on species distribution thus requires that we have not only a detailed understanding of
each species’ physiological responses to body temperature,
but also a concomitantly detailed knowledge of what the
patterns of body temperature are in nature, in both aquatic
and aerial thermal regimes.
Recent studies have suggested that patterns of intertidal
body temperature can be far more complex than previously
recognized. Helmuth and Hofmann (2001) showed that
body temperatures of rocky intertidal mussels varied by
more than 10 °C owing to the effects of substratum angle,
and Denny et al. (2006) found that these variations in
substratum angle could account for variability in risk of
mortality. Also, since the physical characteristics of an
organism (e.g., factors such as shape, color, and mass) affect
the likelihood that it will experience extremes in body
temperature, patterns in high- and low-temperature stress
are likely to vary between species (Denny and Harley, 2006)
and with body size (Helmuth, 2002). Thus, two species of
organisms, or two organisms of the same species but of
varying size, can experience very different body temperatures when exposed to identical physical environments
(Helmuth, 2002; Denny and Harley, 2006). While variability in responses to past climate change has been shown to
occur with body size (Roy et al., 2001), the underlying
biophysical and physiological determinants of this size effect are poorly understood for most intertidal organisms. For
rocky intertidal organisms, larger mussels are thought to be
more thermally stable than smaller animals because they
have greater thermal inertia (Lent, 1968), and small mussels
may experience more extreme body temperature during
brief exposures to extreme climatic conditions (Helmuth,
1998, Helmuth, 1999). However, over long exposures, large
mussels are predicted to reach a more extreme body temperature than small mussels (Helmuth, 1998, Helmuth,
2002).
This effect has yet to be explored for intertidal salt marsh
organisms. Thus, the failure to consider patterns of aerial
temperature and the interactions of morphological factors
such as body size on body temperature neglects the potential
to detect responses to climate change that may operate on
smaller spatial scales, and which may be important determinants of physiological stress. By mechanistically identifying the environmental and morphological determinants of
body temperature, we will be better prepared to understand
how climate sets local zonation patterns and levels of competitive ability (Wethey, 1983). In addition, the added factor
of vertical body position (the proportion of the organism
exposed above the substrate) in soft-sediment habitats has
not been examined.
Here we quantitatively examine the primary determinants
of within-site variability in body temperature of the salt
marsh mussel Geukensia demissa. We explicitly examine
the effects of body size and small-scale microhabitat (position within the mud surface) on body temperature, and
quantify differences in body temperature between sites with
and without marsh grass cover. By examining the complex
patterns of body temperature and how these temperatures
vary with microhabitat, body size, and body position, our
goal is to understand where and when the maximum body
temperatures are most likely to occur in the field, and to
assess how important each of these morphological (size),
behavioral (position), and environmental (grass cover) factors are to body temperature. Using these field data to create
a laboratory experiment that closely mimics actual field
body temperatures, our goal is to determine at what daily
maximum body temperature mussel mortality occurs. We
are then able to compare laboratory mortality rates to field
data to determine whether mortality due to heat stress is
likely to be occurring in the field, and if so, where and when
it is likely to take place.
Study system
Geukensia demissa (Dillwyn, 1817; formerly Modiolus
demissus) is an abundant salt marsh species that plays an
ecologically significant role in marsh dynamics by alleviating nutrient deficiencies through deposition of nitrogenous
wastes on the sediment surface (Bertness, 1984). In addition, the interaction between Spartina alterniflora and this
mussel species prevents marsh disturbance and erosion.
Therefore, this species is ideal for an examination of the
possible effects of climate change on salt marsh ecosystems
in the western Atlantic. G. demissa has a wide geographical
GEUKENSIA DEMISSA BODY TEMPERATURE
range along the eastern coast of North America, from the
Gulf of St. Lawrence to Northern Florida (Abbott, 1954),
and is found in a range of intertidal microhabitats that can
potentially vary greatly in thermal regime.
In salt marsh environments, mussels are buried in the
substrate, often up to the level of the siphon, and they have
a greatly reduced surface area exposed to air during low
tide. However, there can be high variability in mussel body
position; some mussels may be completely buried, while
others may be almost entirely exposed above the sediment
surface. This variation is likely to have a great effect on
mussel body temperature. Other sources of microhabitat
variation include the amount of standing water in the substrate during low tide (as a function of sediment grain size)
and the intensity of solar radiation and wind speed at the
level of the sediment surface (as a function of amount of
vegetation present). These three variables (body size, body
position, and amount of vegetation) thus represent the most
likely sources of within-site thermal variability, and we
tested their relative importance in modifying the body temperature of G. demissa.
Materials and Methods
Estimating mussel body temperature using physical
models
To estimate Geukensia demissa body temperature in the
field, we modified Helmuth and Hofmann’s (2001) design
for a thermally matched (biomimetic) temperature logger.
The logger is a physical mimic composed of an empty G.
demissa shell, 100% clear silicone caulking, and a small,
battery-operated temperature logger (iButton, Maxim Integrated Products, Dallas Semiconductor) that measures estimated mussel body temperature at 20-min intervals in the
field. Continuous body temperature data were collected
every 20 min from 20 May through 31 August 2004. All
data were collected at the Belle W. Baruch Institute in the
North Inlet Estuary of South Carolina (33⬘21.0⬘⬘N,
79⬘10.8⬘⬘W).
To determine mussel mimic accuracy, body temperatures
of living mussel were compared to those of mussel mimics
at a variety of wind speeds in a wind tunnel in the laboratory. For each trial, 6 living mussels (3 allowed to gape, and
3 rubber-banded shut to prevent gaping) and 2 mussel
mimics were arranged in a circle under a halogen heat lamp,
and all mussels were buried in a layer of sediment with 3 cm
of the animal exposed above the level of substrate. To
eliminate potential effects of mussel size, experimental
mussels were between 7 and 8 cm in shell length. All
mussels were oriented with valves facing into the wind, and
all mussels (including mimics) were equidistant from the
halogen heat lamp. A ramp was designed using plastic foam
(Styrofoam) insulation from the bottom of the wind tunnel
143
to the top of the sediment layer to reduce disturbances in
wind flow across the mussels.
We used a metal file to make a small hole in the mussel
shell on the posterior end of the shell and inserted copper/
constantan thermocouples into the mantle tissue to measure
living mussel body temperature every minute. Body temperature data were recorded using a Campbell datalogger
(Campbell Scientific, Logan, UT). Estimated mussel body
temperatures were measured every minute using iButtons
inserted into the mussel mimics. Trials were run under a
halogen heat lamp in a wind tunnel for 3 h, with mussels
slowly heating under the heat lamp for 90 min and cooling
with the lamp off for 90 min. Prior to the experiments, two
1-l flasks of boiling tapwater were added to the wind tunnel
downwind of the mussels to increase the humidity level to
at least 80% relative humidity. These humidity levels were
maintained throughout the experiments. Three replicate trials were conducted for each of three wind speeds: 0.5, 1,
and 1.5 m/s.
Effects of microhabitat, body position, and body size
Three within-site microhabitats were selected on the basis
of vegetation (Spartina alterniflora) and sediment type:
Microsite 1 (Dry) was not vegetated and had well-drained
sediment at low tide; microsite 2 (Wet) had no vegetation
and had standing water at low tide; microsite 3 (Grass) had
Spartina alterniflora present and had well-drained sediment. All sites were located within 50 m of one another and
were at the same tidal height with an average of 67% aerial
exposure time per tidal cycle.
Three body sizes were examined: small (4 – 6 cm in shell
length), medium (6 – 8 cm), and large (8 –10 cm). In addition, three body positions of mussels within the sediment
(measured in terms of length of shell exposed above the
sediment surface) were examined, and all are biologically
relevant to field conditions: 0 cm (entirely buried within
sediment), 3 cm, and 6 cm (Fig. 1). Mussel models were
attached to wooden dowels (3 loggers per dowel, one at
each vertical body position; 3 body sizes ⫻ 2 replicates ⫽
6 dowels per microsite) using marine epoxy (Evercoat Zspar), and positioned in the sediment to ensure proper positioning.
Temperature analysis
Because of the large number of samples collected by the
loggers, body temperatures were summarized on a monthly
basis, using average daily maxima (Fitzhenry et al., 2004).
For each day of the month, the daily maximum body temperature was calculated for each logger. The daily maxima
were then averaged over the entire month. This measurement is useful because it is an indicator of “chronic” hightemperature stress, and it integrates body temperatures experienced during both aerial and aquatic conditions
144
J. JOST AND B. HELMUTH
Figure 1. Diagram of mussel mimics as they are deployed in the field.
Mussels are attached to wooden dowels, and dowels are buried to allow the
vertical body positions to correspond to 0, 3, or 6 cm above the sediment
surface.
(Helmuth and Hofmann, 2001). Statistical analyses were
conducted using an ANOVA and a post hoc Tukey’s test
(SAS, ver. 8.1). Average daily maximum body temperatures
were analyzed separately for each variable: body size, body
position, and microhabitat.
Laboratory study
Thermal regime. Using estimated body temperatures collected in the field, we set up a laboratory experiment to
examine the effects of daily maximum temperatures on the
survival and growth of G. demissa. Estimated mussel body
temperatures collected using mussel thermal mimics during
summer 2004 were analyzed to estimate a typical “hot” day
in the field for these mussels. Based on past research examining G. demissa heat tolerance (Lent, 1968), days with
a body temperature of 45 °C and higher were considered
“hot.” We deployed 54 physical loggers in the field during
summer 2004 and recorded a total of 141 instances in which
the body temperature exceeded 45 °C. For each of these
events, we determined (1) the time of the daily minimum
body temperature, (2) the time of the daily maximum body
temperature, (3) the rate of heating during the day, and (4)
the timing and duration of the high tide. On average, we
determined that the mussels experience a nighttime high
tide for about 4 h, and that the daily minimum temperature
generally occurred around 0600. There was then a period of
heating (at an average rate of 2.66 °C per hour) until 1500
when the maximum body temperature was reached. There
was then a cooling period during low tide until the tide came
in later in the evening.
Experimental tank set-up. The laboratory experiment was
set up to mimic mussel body temperatures under field conditions as closely as possible (Fig. 2), with field conditions
based on mussel body temperature data estimated using
thermal mimics. In a screened laboratory, we set up six
temperature treatments (varying in daily maximum temperature: 30, 35, 40, 45, 50, or 55 °C). For each treatment,
halogen heat lamps (on timers) hanging from the ceiling
(varying in intensity and distance from the sediment surface) were used to regulate mussel body temperature from
0600 to 1500. The rate of heating for each treatment was set
to fall within the range seen in the field. In addition, there
was an artificial tidal cycle (created using a solenoid valve
on a timer), with a high tide coming in every day at 1800
and lasting for 4 h. The mussels were fed unfiltered seawater
pumped directly from the marsh at the Belle W. Baruch
Marine Institute in Georgetown, South Carolina. The seawater used in the experiment was identical in both temperature and salinity to the seawater from the marsh field sites
so that any changes in these parameters during the experiment mimicked changes that would also have been experienced by mussels in the field.
For each of the six treatments (20 individuals ⫻ 6 treatments ⫻ 2 replicates ⫽ 240 mussels total), we used a small
plastic bin to set up artificial marsh conditions. The bins
were layered with a screen mesh along the bottom, a 3-cm
layer of natural aquarium gravel (no coloring added), another layer of mesh screen, and a 3-cm layer of sediment
collected from the same marsh site as the mussels as the top
layer. Five small drainage holes (4 cm in diameter) were
punched out of the bottom panel and the side panel (above
the level of sediment) of each bin. Unfiltered seawater
entered into each tub through PVC pipe coming from the
main seawater line, and the water was allowed to flow along
the bottom of each tub before draining back out to the
Figure 2. Comparison of the estimated mussel body temperatures
recorded in the field and the laboratory set-up for a 24-h period.
145
GEUKENSIA DEMISSA BODY TEMPERATURE
marsh. Therefore, there was a constant supply of water to
the base of each bin to ensure that the sediment stayed moist
even during low tide, mimicking the average marsh conditions as closely as possible. During high tide, water would
slowly fill each bin from the bottom. The water level would
then reach the holes in the side panel and slowly drain out
of the bin. Therefore, this system allowed the water to flood
the bins at high tide and also maintained sediment moisture
during low tide. In addition, this system allowed seawater to
be constantly moving through the system, preventing stagnant conditions and giving a constant food supply to the
mussels during high tide.
Specimen collection and experimental exposure. Live mussels (ranging in shell length from 45 to 70 mm) were
collected from the high intertidal marsh at the Belle W.
Baruch Marine Institute on 20 May 2005. The mussels were
rinsed in saltwater to clean the shells of byssus threads and
other debris and then transferred to a holding tank with
continuously running unfiltered seawater in the screened
laboratory. The mussels were held under these conditions
for 2 weeks to remove any effects of thermal history in the
field prior to collection. On 4 June 2005, mussels were
removed from the holding tanks and individually tagged
with a number (a strip of paper, glued to the shell with
cyanoacrylate adhesive and coated with a layer of clear nail
polish to waterproof the paper), and labeled with a dot of
colored nail polish corresponding to the temperature treatment. Mussels were oriented in growth position with the
posterior end of the organism perpendicular to the sediment
surface. Within 24 h, each mussel had attached to the layer
of screen mesh at the bottom of the sediment layer with new
byssus threads, and a natural mussel body position was
maintained actively throughout the experiment.
The heating regime began on 4 June 2005 and ran for 12
weeks. Estimated body temperature of each treatment was
measured every 5 min using a mussel mimic (described
above) buried to the same level as the living mussels. Once
per week, each bin was examined for dead individuals.
Dead mussels were removed, and the date and mussel
numbers were recorded. In addition, the estimated body
temperature was collected from the mimics each week to
ensure that the heating regime was accurate. To maintain
accuracy, the heat lamps were slightly repositioned on the
basis of the temperatures of the previous week.
Results
Logger accuracy
Comparison of mussel mimic body temperatures and
living mussel body temperatures showed that mimics provide a close estimate of living tissue temperatures and that
mimic accuracy increases with increasing wind speed (Table 1). Mimics tended to warm more quickly and cool more
Table 1
Average (⫾ standard error) and maximum temperature differences (°C)
between average mussel mimic estimated body temperature and living
mussel body temperatures for three wind speeds
Wind speed
(m/s)
Average
difference (°C)
Standard
Error
Maximum
difference (°C)
0.5
1
1.5
1.50
1.11
1.15
0.0605
0.0856
0.0798
4.87
4.05
4.72
Data are averaged over three 3-h replicate trials per wind speed.
quickly than living mussels, but the general patterns of body
temperature remained the same (Fig. 3). Additionally, due
to the potential effects of cooling through evaporation on
mussel body temperature, some mussels were banded shut
to prevent gaping. If cooling was occurring via evaporation,
we expected banded mussels to have higher body temperatures. However, results showed that the body temperature of
banded mussels closely resembled that of mussels allowed
to gape, and at times it tended to be cooler (Fig. 3).
Temperature analysis
On several occasions, loggers were lost or damaged. In
these situations, temperature data have been omitted, causing a few gaps in the summer temperature data. Body
temperatures were similar for all three body sizes, regardless of body position, microhabitat, or month (Fig. 4). There
was no significant effect of body size and no significant
interactions between body size and either body position or
microhabitat (Table 2). Therefore, the analysis was conducted a second time, eliminating the variable of body size
from the ANOVA. The results of the two-way ANOVAs by
month (Table 3) showed a significant effect of body position
on mussel body temperature across all summer months. In
general, the 0-cm body position was warmer at night than
the others and cooler in the day, suggesting a less extreme
environment (Fig. 4). Mussels in the 6-cm body position
experienced both the hottest and coldest temperature (in
most cases) throughout the summer (Fig. 4). At one point, a
mussel in the 6-cm position was up to 25 °C hotter than
mussels in the 0-cm position (compared within one body
size and one microhabitat), the largest difference seen
among all of the variables examined.
Body temperature differences based on microhabitat were
more difficult to interpret (Fig. 4). When microhabitat differences within one body size and body position were
compared there was a very large effect of marsh location in
some cases (Fig. 4). In each of these cases, Grass was the
hottest site, Dry was intermediate, and Wet was the coolest.
There was a significant effect of microhabitat in every
month except August, suggesting that there was a differen-
146
J. JOST AND B. HELMUTH
significantly distinct across microhabitats. However, at the
6-cm body position, microhabitat had a significant effect,
with Grass being warmer than Dry, which was warmer than
Wet. When comparing average daily maximum mussel
body temperature by month, May was the hottest summer
month for all sites and body positions (Fig. 5). In addition,
there was a significant interaction between microsite and
vertical position during June (Table 3). At the Dry and Wet
microsites, mussels at the 3-cm body position more closely
resembled those at the 0-cm body position early in summer,
but more closely resembled those at the 6-cm body position
in August (Fig. 5). On the other hand, at the Grass microsite,
the 0-cm and 3-cm positions were more closely matched,
while the 6-cm one was distinctly warmer (Fig. 5).
Figure 3. Average estimated mussel body temperature (from mussel
mimic loggers) and average living mussel body temperature (for mussels
rubber-banded shut or allowed to gape freely) over time for a 3-h wind
tunnel experiment at a wind speed of 1m/s. Mussels and mimics experienced 90 min of heating using a halogen heat lamp and 90 min of cooling.
tial rate of heating in early and late summer months between
sites, but that sites were consistently hot in August, with no
statistical difference in mussel body temperature between
sites (Table 3). Mussels at the 0-cm body position were not
Mussel survival
Owing to daily fluctuations in air temperature in the
screened seawater laboratory, the distinct body temperature
treatments were not maintained as expected. Therefore,
laboratory data are presented as a linear regression based on
actual average daily maxima experienced by the mussels.
The mussel survival rate decreased sharply at average daily
maximum body temperatures above 45 °C, and mortality
was 100% above 50 °C (Fig. 6). In addition, an increase in
Figure 4. Estimated mussel body temperatures over one 24-h period for mussel mimics for (from top left
to bottom right): three body positions for the same microhabitat and same body size, three microhabitats with
the same body size at the 0-cm body position, three body sizes from the same microhabitat held at the same body
position, and three microhabitats with the same body size at the 6-cm body position.
GEUKENSIA DEMISSA BODY TEMPERATURE
147
Table 2
Fixed effects ANOVA by month for the average daily maximum mussel
body temperature taken from a South Carolina estuary for three body
sizes during the summer of 2004 show no significant effects
Month
DF
Type III SS
Mean Square
F value
P value
June
July
August
2
2
2
0.3003135
5.3729286
11.65681
0.1501567
2.6864643
5.828405
0.24
2.7
1.34
0.7917
0.0865
0.2793
the total time spent above 45°C (measured in degree minutes) resulted in an increase in mortality rates (Fig. 7).
Using the laboratory data, we determined the number of
days in summer 2004 in which field conditions corresponded to a potential mortality event for G. demissa (Table
4A). The highest potential mortality risk occured in the
vegetated microhabitat and for mussels exposed at the 6-cm
vertical body position. We also determined the number of
days in summer 2004 that would have resulted in a potential
mortality event based on an increase of either 2 °C (Table
4B) or 3 °C (Table 4C) in mussel body temperature. If body
temperatures increase even a few degrees, the potential for
mortality will be greater in all microhabitats and depths;
with an increase of 3 °C in body temperature, there is a
potential for mortality to occur across the entire population,
regardless of microhabitat or vertical body position.
Discussion
Temperature patterns
The results of this study suggest that patterns of mussel
body temperature in soft-sediment habitats, and specifically
the effects of body size on body temperature, can be drastically different from those seen in the rocky intertidal zone
(Denny and Harley, 2006). For example, the lower thermal
Table 3
Fixed effects ANOVA by month for the average daily maximum mussel
body temperature taken from a South Carolina estuary for three
microsites, three vertical body positions, and possible interactions
during the summer of 2004
Month
June
July
August
Source
DF
Type III
SS
Mean
Square
F
value
P
value
site
depth
site ⫻ depth
site
depth
site ⫻ depth
site
depth
site ⫻ depth
2
2
4
2
2
4
2
2
4
9.47
211.38
17.47
7.64
249.20
5.35
2.18
179.98
16.09
4.74
105.69
4.37
3.82
124.60
1.34
1.09
89.99
4.02
7.66
170.88
7.06
3.38
110.14
1.18
0.30
24.66
1.10
0.0015
⬍0.0001
0.0002
0.0435
⬍0.0001
0.3323
0.7438
⬍0.0001
0.3679
Figure 5. Average daily maximum temperature with standard errors
for three microsites and three vertical body positions for May through
August 2004. Note: May data begins on 20 May 2004, and therefore is only
a partial representative of the month.
inertia of small rocky intertidal mussels allows them to heat
more quickly than large mussels, leading to differences in
body temperature across a range of body sizes in this
environment (Helmuth, 1998). Conversely, no pattern was
seen across body size and body temperature in the salt
marsh habitat. It is likely that this difference is due in part
to the mussels’ large contact area with the substrate. The
mud acts as one large heat sink, heating slowly and retaining
heat once warm. With an increase in contact between the
organism and the substrate, the organism body temperature
is more closely related to the substrate temperature, as has
been observed for rocky intertidal barnacles (Wethey,
2002). This mechanism works well to describe the lack of
difference in body size for mussels held at the 0-cm body
position. However, mussels at the 6-cm body position have
very little contact with the soft-sediment substrate, and still
148
J. JOST AND B. HELMUTH
Table 4
Number of days in 2004 when mussel body temperatures were above 45
°C in the South Carolina salt marsh based on microhabitat and vertical
body position for three conditions: (A) summer 2004 estimated mussel
body temperatures, (B) a body temperature increase of 2 °C, and (C) a
body temperature increase of 3 °C
Condition
Site
0 cm
3 cm
6 cm
A
SC1
SC2
SC3
SC1
SC2
SC3
SC1
SC2
SC3
1
0
3
4
0
9
5
2
10
11
3
2
17
6
5
23
10
8
14
5
29
31
16
44
35
23
55
B
C
Figure 6. Percent survival as a function of mussel body temperature
(°C, average, average daily maximum and maximum) for groups of 20
mussels during a 12-week laboratory experiment.
body size does not play a significant role in determining
body temperature. The reasons why body size does not
appear to affect body temperature in salt marsh mussels
therefore remains enigmatic.
We found that the effect of microhabitat is similar to that
seen in other habitats, including the rocky intertidal zone.
The mussels are likely to experience body temperature
differences over very small scales (1–10 m) within the same
salt marsh, due to presence or absence of vegetation. However, past studies have suggested that the presence of Spartina alterniflora in the habitat would act as a cooling agent
by providing shade to the substrate and the mussels (Kuenzler, 1961). Our results suggest that the presence of Spartina
may have quite the opposite effect for G. demissa, as the
vegetated site was the hottest of three microhabitats. One
possible explanation for this trend is a decrease in convec-
Figure 7. Percent survival as a function of degree minutes above 45° C
for groups of 20 mussels during a 12-week laboratory experiment.
tion due to the cordgrass. With an increase in vegetation
close to the sediment surface, wind speed decreases, reducing cooling through evaporation and convection. In general,
evaporative cooling is already low in this species due to the
high humidity levels (80%–90% relative humidity) close to
the sediment substrate. Additionally, comparison of banded
and gaping mussels in this study suggested that evaporative
cooling does not play a role in the body temperature of this
species. However, the reduced wind speed is likely to
greatly decrease convective cooling. In most cases, the
sediment heats to a higher temperature than the air above
the substrate. With reduced air flow, the ability of the
substrate and the organisms embedded in it to lose heat to
the environment decreases. Also, although the vegetation is
likely to increase shading to the substrate in some situations,
we have found that mussels are most likely to experience
maximum body temperatures early in the afternoon, when
the sun is directly overhead, reducing any effect shading
would have on body temperature.
Although we found that microhabitat plays a role in
determining body temperature, the main factor that appears
to affect body temperature is vertical body position. A
vertical difference of 6 cm results in up to a 25 °C difference
in body temperature. These results suggest that small-scale
differences in vertical position (cm) are more important in
determining potential heat stress than are meters of horizontal position (microhabitat) or several hundred kilometers in
latitude (climate).
We suggest that mussels in the 6-cm body position are at
a strong thermal disadvantage to those at more buried body
positions. The question remaining is, can mussels actively
regulate their vertical position in the sediment? As larvae
mussels settle out of the water column onto the substrate,
there is a preferential settlement on or near the shells of
conspecifics (Bertness and Grosholz, 1985; Nielson and
Franz, 1995). As a result of this behavior, mussels are often
found in large clumps, which may result in crowding that
GEUKENSIA DEMISSA BODY TEMPERATURE
causes them to be pushed farther above the sediment surface
as surrounding mussels grow in size. However, it is not rare
to find a solitary mussel in the salt marsh habitat, especially
at the higher tidal heights where density is lower. If exposed
body positions are negative, one would expect solitary mussels to be completely buried, which is not always the case.
Often solitary mussels are found at vertical body positions of 3– 6 cm above the sediment surface, suggesting
either an advantage to this position or the inability to regulate body position above a certain body size. It is possible
that an increase in flow rate during high tide allows for
better food consumption as the individual moves away from
the substrate. In addition, a mussel may have to use less
energy to remove sediment from the water column during
feeding events when it is farther from the mud surface.
Another possible explanation for the large numbers of mussels found at the 3– 6-cm position in the field may be that
increased body temperatures are beneficial to a point. For
example, it is possible that at these positions, mussels experience increased rates of metabolism, digestion, and
growth, while experiencing hot, but sublethal, temperatures
most of the summer. Even if these mussels run some risk of
being exposed to lethal body temperatures throughout the
summer months, thermal benefits at the 3- and 6-cm body
positions may outweigh these risks.
Mussel survival
We showed that mussels can die when exposed to a
heating regime similar to that seen regularly during the
summer months. Even a few moments of exposure at about
50 °C is enough to result in 100% mortality. At the same
time, we recorded temperatures above 50 °C in the field
during summer 2004. In addition, we found a strong decrease in survival at daily maximum body temperatures
about 45 °C, conditions that are seen in the field fairly
regularly.
A previous study has suggested that, thanks to the cooling
from evaporation from the surrounding mud. mussels in the
salt marsh habitat are not likely to reach their thermal limits
(Lent, 1969). Although most previous studies have examined the thermal limits of submerged mussels (e.g., Hilbish,
1987), some references to heating during aerial exposure
can be found (Read and Cumming, 1967; Lent, 1968).
However, these studies concluded that the upper tolerance
limit lies between 36 and 40 °C. In this study, technological
advances allowed us to measure mussel body temperature in
the field and discover that mussel temperature often exceeds
that of air temperature, contrary to past suggestions (Lent,
1969). In addition, we have shown that mussel body temperature often exceeds both the previously determined limits (36 – 40 °C) and the current thermal limits discovered
here (⬃45 °C). On the basis of these results, we suggest that
mortality occurs in South Carolina estuaries during the
149
summer months at a variety of microhabitats within the salt
marsh.
Implications for climate change
The Intergovernmental Panel on Climate Change 2001
Synthesis report (IPCC, 2001) outlines several several scenarios of climate change, in particular with expected increases in air temperature. For two of these scenarios, A2
and B2, air temperature in coastal South Carolina is expected to increase by 2 to 3 °C, respectively, by the year
2100. Given that mortality due to heat stress should already
be occurring in South Carolina salt marshes, it is likely that
heat stress will continue and possibly become a greater
concern over time. More specifically, the majority of heatrelated mortalities in this mussel population are predicted to
be currently occurring at the 6-cm position and to a lesser
extent at the 3-cm position. While air temperatures may not
translate directly into changes in body temperature (see
below) they may give some indication of what changes lie
ahead. With a 3 °C increase in mussel body temperature,
mortality due to heat stress may be likely to occur at all
microhabitats and at all vertical body positions, making the
entire mussel population vulnerable to heat mortality.
However, the current climate-change scenarios predict
temperature changes in terms of air and water temperatures
only. Although these predictions may allow for some rough
estimates of future conditions for Geukensia demissa populations in South Carolina, these predictions do not suggest
what future trends in mussel body temperature will be. As
shown by Gilman et al. (2006), climate-change predictions
for a species require the following three items: (1) current
and future climate predictions, (2) the relationship between
climate and body temperature for that species, and (3) the
physiological tolerances of that species. In this study, we
have successfully measured animal body temperature in the
field and begun exploring the physiological tolerance of G.
demissa. However, we still do not understand the exact
relationship between air temperature and body temperature,
and therefore we cannot accurately predict how mussel
body temperature will change with the current climatechange scenarios. It is likely that a 1 °C increase in air
temperature will not result in an equal increase in mussel
body temperature (Gilman et al., 2006).
Assuming that an increase in air temperature will result in
at least a small increase in mussel body temperature, more
and more mussels may become prone to heat stress, and G.
demissa populations may suffer. This species has been
shown to play a key role in the salt marsh ecosystem,
specifically—through its interaction with Spartina alterniflora—in terms of the nitrogen cycle (Jordan and Valiela,
1982) and the prevention of marsh erosion (Bertness, 1984).
Therefore, it is possible that a strong decrease in the abundance of G. demissa may lead to a die-off of Spartina
150
J. JOST AND B. HELMUTH
alterniflora in some areas. With a loss of vegetation, the salt
marshes will be more prone to erosion. Therefore, the
ability to monitor the body temperature of this species and
the potential effects of this temperature on population may
have larger implications if climate change influences the
levels of heat stress being experienced by this mussel population.
Conclusions
Unlike past studies examining patterns in the body temperature of rocky intertidal invertebrates (e.g., Helmuth and
Hofmann, 2001), our results, obtained by directly measuring body temperature patterns in the field, show that body
size is of little importance in determining body temperature
in the soft sediment habitats. Second, in agreement with
other studies in the rocky intertidal zone (e.g., Helmuth,
1998), we found that small horizontal scales are important
in the soft-sediment habitat. However, the dominant determinant of body temperature is vertical position in the sediment, and a few centimeters can result in temperature
differences of more than 20 °C. Therefore, the ability of
mussels to regulate vertical position and the causes for
mussel positioning in the field have become of greater
interest than initially expected. In the laboratory, we found
it is possible to recreate the typical “stressful” day in the
intertidal, showing that mussel mortality is occurring at
temperatures of 45 °C and higher. What is of greatest
interest in this study is the potential for mussel mortality to
be occurring in the field on a regular basis. On the basis of
field and laboratory data, we have concluded that mussel
mortality due to heat stress should be occurring at a variety
of microhabitats within South Carolina estuaries.
Acknowledgments
The authors thank S. Forehand for assistance with laboratory set-up, and the Belle W. Baruch Marine Institute for
access to field sites and for use of the running seawater laboratory. In addition, we thank K. A. Smith, L. Szathmary, C.
Purvis, L. Burnett, J. Hilbish, D. Wethey, and S. Woodin for
help in editing earlier versions of the manuscript. This
research was funded by the National Science Foundation
(OCE-0323364) and by NOAA NA04NOS4780264.
Literature Cited
Abbott, R. T. 1954. American Seashells. D. van Nostrand, Princeton, NJ.
Barry, J. P., C. H. Baxter, R. D. Sagarin, and S. E. Gilman. 1995.
Climate-related, long-term faunal changes in a California rocky-intertidal community. Science 267: 672– 675.
Bertness, M. D. 1984. Ribbed mussels and Spartina alterniflora production in a New England salt marsh. Ecology 65: 1794 –1807.
Bertness, M. D., and E. Grosholz. 1985. Population dynamics of the
ribbed mussel, Geukensia demissa: the costs and benefits of an aggregated distribution. Oecologia 67: 192–204.
Bertness, M. D., G. H. Leonard, J. M. Levine, and J. F. Bruno. 1999.
Climate-driven interactions among rocky intertidal organisms caught
between a rock and a hot place. Oecologia 120: 446 – 450.
Connell, J. H. 1972. Community interactions on marine rocky intertidal
shores. Annu. Rev. Ecol. Syst. 3: 169 –192.
Davenport, J., and J. L. Davenport. 2005. Effects of shore height,
wave exposure and geographical distance on thermal niche width of
intertidal fauna. Mar. Ecol. Prog. Ser. 292: 41–50.
Denny, M. W., and C. D. G. Harley. 2006. Hot limpets: predicting body
temperature in a conductance-mediated thermal system. J. Exp. Biol.
209: 2409 –2419.
Denny, M. W., L. P. Miller, and C. D. G. Harley. 2006. Thermal stress
on intertidal limpets: long-term hindcasts and lethal limits. J. Exp. Biol.
209: 2420 –2431.
Doty, M. S. 1946. Critical tide factors that are correlated with the vertical
distribution of marine algae and other organisms along the Pacific
coast. Ecology 27: 315–328.
Fields, P. A., J. B. Graham, R. H. Rosenblatt, and G. N. Somero. 1993.
Effects of expected global climate change on marine faunas. Trends
Ecol. Evol. 8: 361–367.
Fields, P. A., E. L. Rudomin, and G. N. Somero. 2006. Temperature
sensitivities of cytosolic malate dehydrogenases from native and invasive species of marine mussels (genus Mytilus): sequences-function
linkages and correlations with biogeographic distribution. J. Exp. Biol.
209: 656 – 667.
Fitzhenry, T., P. M. Halpin, and B. Helmuth. 2004. Testing the effects
of wave exposure, site, and behavior on intertidal mussel body temperatures: applications and limits of temperature logger design. Mar.
Biol. 145: 339 –349.
Gilman, S. E., D. S. Wethey, and B. Helmuth. 2006. Variation in the
sensitivity of organismal body temperature to climate change over local
and geographical scales. Proc. Natl. Acad. Sci. 103: 9560 –9565.
Harley, C. D. G., A. R. Hughes, K. Hultgren, B. G. Miner, C. J. B.
Sorte, C. S. Thornber, L. F. Rodriguez, L. Tomanek, and S. L.
Williams. 2006. The impacts of climate change in coastal marine
systems. Ecol. Lett. 9: 228 –241.
Helmuth, B. 1998. Intertidal mussel microclimates: predicting the body
temperature of a sessile invertebrate. Ecol. Monogr. 68: 51–74.
Helmuth, B. 1999. Thermal biology of rocky intertidal mussels: quantifying body temperatures using climatological data. Ecology 80: 15–
34.
Helmuth, B. 2002. How do we measure the environment? Linking
intertidal thermal physiology and ecology through biophysics. Integr.
Comp. Biol. 42: 837– 845.
Helmuth, B., and G. E. Hofmann. 2001. Microhabitats, thermal heterogeneity, and patterns of physiological stress in the rocky intertidal
zone. Biol. Bull. 201: 374 –384.
Helmuth, B., B. N. Mieszkowska, P. Moore, and S. J. Hawkins. 2006a.
Living on the edge of two changing worlds: forecasting the responses
of rocky intertidal ecosystems to climate change. Annu. Rev. Ecol.
Evol. Syst. 37: 373– 404.
Helmuth, B., B. R. Broitman, C. A. Blanchette, S. Gilman, P. Halpin,
C. D. G. Harley, M. J. O’Donnell, G. E. Hofmann, B. Menge, and
D. Strickland. 2006b. Mosaic patterns of thermal stress in the rocky
intertidal zone: implications for climate change. Ecol. Monogr. 76:
461– 479.
Hilbish, T. J. 1987. Response of aquatic and aerial metabolic rates in the
ribbed mussel Geukensia demissa (Dillwyn) to acute and prolonged
changes in temperature. J. Exp. Mar. Biol. Ecol. 105: 207–218.
Hofmann, G. E., and G. N. Somero. 1995. Evidence for protein damage
at environmental temperatures: seasonal changes in levels of ubiquitin
conjugates and HSP70 in the intertidal mussel Mytilus trossulus. J. Exp.
Biol. 7: 1509 –1518.
IPCC (Intergovernmental Panel on Climate Change). 2001. Climate
change 2001: the scientific basis. [Online]. Available http://www.
GEUKENSIA DEMISSA BODY TEMPERATURE
grida.no/climate/ipcc_tar/wg1/031.htm#figTechSum20. Accessed 19
September 2007.
Jordan, T. E., and I. Valiela. 1982. A nitrogen budget of the ribbed
mussel, Geukensia demissa, and its significance in nitrogen flow in a
New England salt marsh. Limnol. Oceanogr. 27: 75–90.
Kuenzler, E. J. 1961. Structure and energy flow of a mussel population
in a Georgia salt marsh. Limnol. Oceanogr. 6: 191–204.
Lent, C. M. 1968. Air-gaping by the ribbed mussel, Modiolus demissus
(Dillwyn): effects and adaptive significance. Biol. Bull. 134: 60 –73.
Lent, C. M. 1969. Adaptations of the ribbed mussel, Modiolus demissus
(Dillwyn), to the intertidal habitat. Am. Zool. 9: 283–292.
Nielson, K. J., and D. R. Franz. 1995. The influence of adult conspecifics and shore level on recruitment of the ribbed mussel, Geukensia
demissa (Dillwyn). J. Exp. Mar. Biol. Ecol. 188: 89 –98.
Ray, G. C., B. P. Hayden, A. J. Bulger, Jr., and M. G. McCormickRay. 1992. Effects of global warming on the biodiversity of coastalmarine zones. Pp. 91–104 in Global Warming and Biological Diversity,
R. L. Peters and T. E. Lovejoy, eds. Yale University Press, New Haven,
CT.
Read, K. R. H., and K. B. Cumming. 1967. Thermal tolerance of the
bivalve molluscs Modiolus modiolus L., Mytilus edulis L. and Brachidontes demissus Dillwyn. Comp. Biochem. Physiol. 22: 149 –155.
Roy, K., D. Jablonski, and J. W. Valentine. 2001. Climate change,
species range limits, and body size in marine bivalves. Ecol. Lett. 4:
366 –370.
Sagarin, R. D., J. P. Barry, S. E. Gilman, and C. H. Baxter. 1999.
Climate-related change in an intertidal community over short and long
time scales. Ecol. Monogr. 69: 465– 490.
151
Somero, G. N. 2002. Thermal limits to life: underlying mechanisms and
adaptive plasticity. Integr. Comp. Biol. 42: 1316.
Somero, G. N. 2005. Linking biogeography to physiology: evolutionary
and acclamatory adjustments of thermal limits. Front. Zool. 2: 1.
Southward, A. J., S. J. Hawkins, and M. T. Burrows. 1995. Seventy
years’ observations of changes in distribution and abundance of zooplankton and intertidal organisms in the western English channel in
relation to rising sea temperature. J. Therm. Biol. 20: 127–155.
Stainforth, D. A., T. Aina, C. Christensen, M. Collins, N. Faull, D. J.
Frame, J. A. Kettleborough, S. Knight, A. Martin, J. M. Murphy,
C. Piani, D. Sexton, L. A. Smith, R. A. Spicer, A. J. Thorpe, and
M. R. Allen. 2005. Uncertainty in predictions of the climate response
to rising levels of greenhouse gases. Nature 433: 403– 406.
Stillman, J. 2003. Acclimation capacity underlies susceptibility to climate change. Science 301: 65.
Suchanek, T. H. 1994. Temperate coastal marine communities: biodiversity and threats. Am. Zool. 34: 100 –114.
Tomanek, L., and E. Sanford. 2003. Heat-shock protein 70 (Hsp70) as a
biochemical stress indicator: an experimental field test in two congeneric intertidal gastropods (genus: Tegula) Biol. Bull. 205: 276 –284.
Walther, G., E. Post, P. Convey, A. Menzel, C. Parmesan, T. Beebee,
J. Fromentin, O. Hoegh Guldberg, and F. Bairlein. 2002. Ecological responses to recent climate change. Nature 416: 389 –395.
Wethey, D. S. 1983. Geographic limits and local zonation: the barnacles
Semibalanus (Balanus) and Chthamalus in New England. Biol. Bull.
165: 330 –341.
Wethey, D. S. 2002. Biogeography, competition, and microclimate: the
barnacle Chthamalus fragilis in New England. Integr. Comp. Biol. 42:
872– 880.