Kelp in hot water: effects of warming seawater temperature on kelp quality as a food source and settlement substrate

Abstract

Predicting the effect of climate change on communities requires an understanding of the effects of environmental conditions on species and their interactions. We investigated the potential for warming seawater temperature to modify the interactions of the gastropod mesograzer Lacuna vincta and the invasive bryozoan Membranipora membranacea with kelps in Nova Scotia. The nutritional content (C/N) of the kelps Saccharina latissima, Laminaria digitata and Agarum clathratum were unaffected by temperature (11, 18 and 21°C), and chemical defenses (phlorotannins) were reduced only in A. clathratum after 1 wk exposure to 21°C. C/N and phlorotannin content increased over the season in S. latissima collected monthly in summer 2013 and 2014. The effect of temperature-induced changes in kelp on the grazing of L. vincta was assessed using feeding experiments with S. latissima pretreated at 11 or 21°C. Snails consumed more kelp pretreated at 21°C only when grazing rate was high. The quality of S. latissima as a food source for L. vincta was not affected by temperature, as diets of kelp pretreated at 11 and 21°C supported similar growth, reproduction, and survival of snails. Temperature also did not affect the quality of kelp as a substrate for M. membranacea, since settlement rates were not different between S. latissima pretreated at ambient temperature (9 to 14) and 21°C. The absence of temperatureinduced changes in kelp quality suggests that the effects of L. vincta and M. membranacea will act additively with the direct effects of temperature and cause increased biomass loss from kelp beds.

Full text

MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 53 7: 105 –119, 2015
doi: 10.3354/meps11421 Published October 14
INTRODUCTION
Effects of climate change on community structure
and function will integrate environmental impacts on
individual species with changes in interactions
among species (Harley et al. 2006). Alteration of bio -
tic interactions can have significant effects on com-
munity composition and ecosystem function that can
enhance or diminish the effects of climate change
(Zarnetske et al. 2012, HilleRisLambers et al. 2013).
Climate-driven changes in biotic interactions can
occur through introduction or loss of interacting spe-
cies due to changes in distributional range or pheno -
logy, or through changes in the strength of interac-
tions due to alteration of behavior or physiology
(Zarneste etl al. 2012, HilleRisLambers et al. 2013,
Vergés et al. 2014). Changing climate also can facili-
tate the invasion of non-native species or exacerbate
their effects on native communities (Cockrell & Sorte
2013, Floerl et al. 2013). Alteration of biotic inter -
actions can combine with the direct effects of climate
change on a species, resulting in impacts that may
not have been expected when considering the effect
of climate alone (HilleRisLambers et al. 2013).
© Inter-Research 2015 · www.int-res.com*Corresponding author: erika.simonson@dal.ca
Kelp in hot water: II. Effects of warming seawater
temperature on kelp quality as a food source
and settlement substrate
E. J. Simonson1,*, A. Metaxas2, R. E. Scheibling1
1Department of Biology, Dalhousie University, Halifax, Nova Scotia B3H 4R2, Canada
2Department of Oceanography, Dalhousie University, Halifax, Nova Scotia B3H 4R2, Canada
ABSTRACT: Predicting the effect of climate change on communities requires an understanding of
the effects of environmental conditions on species and their interactions. We investigated the
potential for warming seawater temperature to modify the interactions of the gastropod meso-
grazer Lacuna vincta and the invasive bryozoan Membranipora membranacea with kelps in Nova
Scotia. The nutritional content (C/N) of the kelps Saccharina latissima, Laminaria digitata and
Agarum clathratum were unaffected by temperature (11, 18 and 21°C), and chemical defenses
(phlorotannins) were reduced only in A. clathratum after 1 wk exposure to 21°C. C/N and phloro-
tannin content increased over the season in S. latissima collected monthly in summer 2013 and
2014. The effect of temperature-induced changes in kelp on the grazing of L. vincta was assessed
using feeding experiments with S. latissima pretreated at 11 or 21°C. Snails consumed more kelp
pretreated at 21°C only when grazing rate was high. The quality of S. latissima as a food source
for L.vincta was not affected by temperature, as diets of kelp pretreated at 11 and 21°C supported
similar growth, reproduction, and survival of snails. Temperature also did not affect the quality of
kelp as a substrate for M. membranacea, since settlement rates were not different between
S. latissima pretreated at ambient temperature (9 to 14) and 21°C. The absence of temperature-
induced changes in kelp quality suggests that the effects of L. vincta and M. membranacea will act
additively with the direct effects of temperature and cause increased biomass loss from kelp beds.
KEY WORDS: Kelp · Climate change · Temperature · Saccharina latissima · Lacuna vincta ·
Herbivory · Membranipora membranacea · Bryozoans
Resale or republication not permitted without written consent of the publisher
This authors' personal copy may not be publicly or systematically copied or distributed, or posted on the Open Web,
except with written permission of the copyright holder(s). It may be distributed to interested individuals on request.

Mar Ecol Prog Ser 537: 105–119, 2015
Globally, increasing seawater temperature has
been linked to range contractions and declines in kelp
populations (Andersen et al. 2011, Fernández 2011,
Tuya et al. 2012, Wernberg et al. 2013). Kelp (large
brown algae of the order Laminariales) are im portant
foundation species on temperate rocky reefs, support-
ing high community productivity and biodiversity
(Dayton 1985, Steneck et al. 2002). Direct effects of
increasing temperature on kelp include reduced
growth (Bolton & Lüning 1982, Gerard & Du Bois 1988,
Davison 1991, Andersen et al. 2013), weak en ing and
loss of kelp tissue (Simonson et al. 2015, this volume),
and mortality when physiological limits are exceeded
(Wernberg et al. 2013). Temperature-mediated
changes in kelp bed communities also may occur indi-
rectly through the alteration of interspecific inter -
actions (Harley et al. 2012, Vergés et al. 2014). The
interaction between herbivores and their algal food
sources can regulate kelp communities through the
consumption of algal biomass (Lubchenco & Gaines
1981), and may be altered by temperature stress due
to changes in either algal growth rate or consumption
(O’Connor 2009), or changes in algal palatability
(Harley et al. 2012). For example, warming seawater
temperatures in southern Japan have both impacted
kelp directly and enhanced feeding rates of herbi -
vorous fish, triggering shifts from beds of Ecklonia
cava to barrens (Vergés et al. 2014).
Changes in the palatability of kelp, and thus its
vulnerability to herbivores, could be driven by tem-
perature-induced changes to the morphological or
chemical characteristics of the kelp tissue. Herbi-
vores alter both rate of consumption and feeding
preference based on the palatability of their food
sources, selecting foods that have fewer mechanical
or chemical defenses, or greater nutritional content
(Steneck & Watling 1982, Duffy & Paul 1992, Krauf -
velin et al. 2006, Pansch et al. 2008). Warmer temper-
atures decreased palatability of Ecklonia radiata by
increasing the C/N ratio (Staehr & Wernberg 2009),
as herbivores tend to select food sources enriched in
nitrogen (Duffy & Paul 1992). In contrast, increases in
temperature prevented the induction of anti-herbi-
vore defenses in the brown alga Fucus vesiculosus,
increasing consumption (Weinberger et al. 2011).
Kelp (and other brown algae) produce phlorotan-
nins (polymers of phloroglucinol) that can act as a
chemical defense against herbivory (Steinberg 1984,
Targett & Arnold 1998). However, the effectiveness
of phlorotannins as a deterrent may depend on their
molecular size or structural characteristics (Boettcher
& Targett 1993, Van Altena & Steinberg 1992), as
well as herbivore tolerance (Kubanek et al. 2004).
Phlorotannins likely have multiple ecological roles
(Amsler & Fairhead 2006), as they also act as antioxi-
dant compounds and can be induced by UV stress
(Cruces et al. 2012). Kelp may defend against cellular
damage caused by thermal stress by upregulating
antioxidant compounds, such as phlorotannins, to
combat the effects of reactive oxygen species. How-
ever, in the 3 species examined to date, thermal
stress was not observed to induce phlorotannins, and
at high levels may even inhibit phlorotannin produc-
tion (Cruces et al. 2012, 2013). Phlorotannins have
also been implicated as chemical defenses against
epibionts (Wikström & Pavia 2004, Iken et al. 2009),
suggesting that temperature-induced changes in
phlorotannin content could affect the vulnerability of
kelp to both grazing and fouling.
In Nova Scotia kelp beds, the gastropod mesograzer
Lacuna vincta is a dominant herbivore that feeds by
creating surface excavations and perforations on kelp
blades (Johnson & Mann 1986). Grazing by L. vincta
re moves kelp tissue not only directly, but also indi-
rectly, by increasing erosion and breakage of blades
(Johnson & Mann 1986, Duggins et al. 2001, Krum -
hansl & Scheibling 2011, Krumhansl et al. 2011),
which can result in large-scale loss of kelp biomass
(Fralick et al. 1974, O’Brien et al. 2015). Al though a
generalist herbivore, L. vincta exhibits dietary selec-
tivity both among algal species and among tissue
types within species, with a strong preference for
Saccharina latissima and avoidance of tissues high in
phlorotannins (Johnson & Mann 1986, Chavanich &
Harris 2002, Toth & Pavia 2002, Dubois & Iken 2012).
Temperature-mediated changes in defensive chemi-
cals, tissue toughness, or nutritional value of S. latis-
sima could affect feeding rate and selectivity of L.
vincta and the impact of these snails on kelp beds.
The invasive bryozoan Membranipora membrana -
cea also causes extensive biomass loss from Nova
Scotian kelp beds, by encrusting kelp blades, weak-
ening the tissue and leading to blade breakage
(Krum hansl et al. 2011). Outbreaks of M. membrana -
cea, combined with large wave events, result in the
large-scale defoliation of kelp beds (Saunders &
Metaxas 2008, Scheibling & Gagnon 2009). Since the
introduction of M. membranacea to the region in the
early 1990s, the extent of bryozoan outbreaks and
consequent damage to kelp beds have been closely
tied to warming temperatures (Scheibling & Gagnon
2009, Saunders et al. 2010), which result in earlier
settlement (Saunders & Metaxas 2007) and increased
colony growth of the bryozoan (Saunders & Metaxas
2009). The damage inflicted on kelp beds by M.
membranacea is an indirect effect of increasing tem-
106
Author copy

Simonson et al.: Warming seawater effects on kelp quality
perature, which may act synergistically with the
direct effects of increased temperature on kelp.
Increased temperature also may limit population
growth of M. membranacea on kelp blades if it
causes a reduction in tissue quality that limits larval
settlement or the persistence of newly established
colonies. In the laboratory, larvae of M. mem-
branacea exhibit settlement preferences, both for
algal species and for specific areas of kelp blade
(Matson et al. 2010), suggesting that they can detect
differences in substrate quality. Increases in temper-
ature that degrade the structural integrity of blade
tissue, or increase phlorotannin content, could pre-
vent settling larvae from attaching, resulting in an
antagonism between the direct effects of tempera-
ture on the kelp and the effects of temperature on the
interaction between kelp and M. membranacea.
We examined the potential for synergistic or anta -
go nistic effects of L. vincta, M. membranacea and
warming seawater temperature on biomass loss from
Nova Scotia kelp beds. Specifically, we investigated
the impact of warming temperatures on the phloro -
tannnin content and C/N ratio of tissue of Saccharina
latissima, Laminaria digitata and Agarum clathratum
in a laboratory experiment, as changes in these chem-
ical properties could potentially alter interactions of
the kelp with L. vincta or M. membranacea. We used
the feeding experiments to determine whether tem-
perature-induced changes in kelp tissue quality al-
tered the feeding rate of L.vincta.Wepredictedthat
changes in the kelp tissue at warmer temperatures
would make kelp easier to consume or more palatable,
leading toincreased feeding rate.Increases in feed ing
rate would indicate a synergism where warming tem-
perature both directly and indirectly (through herbi -
vory) increases kelp tissue loss. We also examined the
effect of temperature-induced changes in kelp tissue
quality on settlement of M. membranacea,predicting
that settlement would be reduced on degraded kelp
exposed to warmer temperatures. Reduced settlement
rate would indicate an antagonism, whereby increas-
ing temperature directly would result in increased loss
of kelp biomass, but at the same time mitigate loss
caused through encrustation by the bryozoan.
MATERIALS AND METHODS
Chemical composition of kelp tissue
Experimental design. Mature individuals of Sac -
cha rina latissima (1 to 1.5 m blade length), Laminaria
digitata and Agarum clathratum (0.5 to 1.0 m), were
collected by SCUBA from 12 m depth at Splitnose
Point (44° 28’ 38.45”N, 63° 32’ 48.21”W) in June/July
2013. Kelp were transported to the laboratory in cool-
ers, and stored upon arrival in a 3000 l circular tank
(1.87 m diameter, 1.08 m height) with continuously
flowing ambient seawater (flow rate: 430 l h−1) by fix-
ing the kelp holdfasts with elastic bands to plastic
racks suspended within the tank. Within 24 h of col-
lection, 9 individuals of each species were suspended
in experimental tanks of similar size and flow rate
(total: 27 individuals per tank) and exposed to 1 of 3
temperature treatments: 11, 18 or 21°C. These tem-
peratures represent a growth optimum for S. latis-
sima and L. digitata (11°C), a typical maximum aver-
age temperature experienced over 1 to 2 weeks
(18°C) (Scheibling et al. 2013), and an anticipated
maximum temperature based on climate change pre-
dictions (21°C) (Müller et al. 2009). Any herbivores
present on the surface of the kelp were removed
before placement in the temperature treatments.
Tissue samples (5.5 cm diameter) were collected
for chemical analysis from 3 individuals of each spe-
cies within 24 h of field collection (before tempera-
ture treatment), from 3 individuals of each species
after 1 wk exposure to 11 and 21°C, and from another
3 individuals of each species after 2 wk exposure to
11 and 18°C (for phlorotannin content, % dry wt) or
3wk exposure to 11 and 18°C (for C/N ratio) (Fig. 1).
Exposure time at which phlorotannins and C/N were
assessed differed because low survival after 3 wk at
18°C did not provide sufficient replicates for phloro-
tannin quantification. Tissue samples were excised
30 cm from the blade−stipe interface in S. latissima,
and 15 cm from the blade−stipe interface in L. digi-
tata and A. clathratum, to control for any variation in
chemical properties along the length of a blade.
Three trials of the experiment were conducted in
June/ July 2013 and all tissue samples were stored at
−80°C until further processing. Tissue samples for
determination of C/N were collected only in the first
trial (n = 3 individuals) and were dried at 60°C for
48 h until constant weight. Tissue samples for phloro-
tannin content were collected in all 3 trials (n = 9
individuals). During the trials, temperature was re -
corded in each tank with data loggers (Maxim Inte-
grated Thermochron iButtons) as 11.6 ± 0.9, 18.2 ±
1.6, and 20.8 ± 1.1°C (mean ±1 SD, n = 58).
To track changes in the chemical composition of
kelp blades over the summer, 10 to 15 mature S. latis-
sima (1.0 to 1.5 m blade length) were collected
monthly from 8 m depth at Splitnose Point, from 25
June to 27 August 2013 and from 30 June to 25 Sep-
tember 2014. Kelp were transported to the laboratory
107
Author copy

Mar Ecol Prog Ser 537: 105–119, 2015
in coolers, and, immediately on arrival, 5.5 cm dia -
meter tissue samples were excised 30 cm from the
blade− stipe interface. Tissue samples were stored
at –80°C until further processing. Temperature at the
collection site was continuously monitored over the
collection periods using data loggers (Onset HOBO
Pendant) anchored to the seabed.
C/N analysis. Dried algal tissue was ground to a
fine homogeneous powder using a mortar and pestle
and packaged in known weights into
tin capsules. C and N content were
determined using a Costec ECS 4010
CHNSO analyzer with acetanilide as a
standard (detection limit = 0.001 mg).
Phlorotannin analysis. To extract
phlorotannins, tissue samples were
first freeze-dried and ground to a fine
powder with a mortar and pestle. For
each sample, 100 mg dried tissue was
placed in 5 ml 70% acetone and ex -
tracted overnight with continuous
shaking (Koivikko et al. 2005). Sam-
ples were then centrifuged for 10 min
at 3200 × gand 0.05 ml supernatant
was withdrawn for phlorotannin
quan tification. Phlorotannin content
was de termined using the Folin-Cio-
calteu assay (Van Alstyne 1995) with
phloro glucinol (1,3,5-trihodroxyben-
zene, Sigma-Aldrich) as a standard.
The 0.05 ml aliquot was mixed with
1.0 ml distilled water and 1.0 ml 40%
Folin-Ciocalteu reagent. After stand-
ing 5 min, 1.0 ml NaCO3was added
and samples were incubated for
30 min at 50°C. Absorbance was read
at 765 nm using a Cary WinUV 4000
spectro photometer (Agilent Technolo-
gies, detection limit = 0.23 mg).
Statistical analyses. The effect of
temperature treat ment after 1 wk (11
and 21°C) or after 3 wk (11 and 18°C)
on C/N ratio was analyzed using 2-
tailed independent samples t-tests, for
each kelp species. Variance was het-
erogeneous for C/N ratio data for A.
clathratum and L. digitata after 1 wk,
so the Welch-Satterthwaite modifica-
tion of degrees of freedom was used.
The effect of temperature treatment
(1 wk at 11 or 21°C, or 2 wk at 11 or
18°C; fixed factors) and trial (random
factor) on phlorotannin content was ex -
amined using 2-way ANOVA. If the trial × treatment
interaction was high ly non-significant (p > 0.20), the
interaction was removed and the main effects of tem-
perature and trial were tested against the pooled trial
× treatment MS and residual MS (Winer et al. 1991).
Seasonal differences in phlorotannin content or C/N
ratio among sampling dates were examined using 1-
way ANOVA. Post hoc comparisons of sampling
dates were performed using Tukey’s HSD test.
108
11°C 18°C 21°C
C/N & Phlorotannins
A. clathratum, L. digitata, S. latissima
A. clathratum, L. digitata, S. latissima
S. latissima &
L. vincta
No-Choice Expt
Settlement Expts
S. latissima &
M. membranacea
Ambient
A. clathratum, L. digitata, S. latissima
Phlorotannins
C/N
Feeding Expts
Choice Expt
Snails No snails
ChoiceNo choice
1 wk
10 d
2 wk
3 wk
Snails No snails
Growth,
Reproduction &
Survival Expts
S. latissima &
L. vincta
Snails
No kelp 11° kelp 21˚ kelp
Temperature pretreatment
Fig. 1. Schematic of temperature pretreatments and experimental design for
quantification of C/N and phlorotannin in Agarum clathratum, Laminaria dig-
itata and Saccharina latissima, feeding experiments and growth, reproduction
and survival experiments (Lacuna vincta on Saccharina latissima), and settle-
ment experiments (Membranipora membranacea on S. latissima). In feeding,
growth, reproduction and survival, and settlement experiments, temperature
pretreatment of disks or half-disks is indicated by colour. Pairs of disks or half-
disks sharing the same texture (hatched or open) come from one individual of
S. latissima. Pairs with one hatched and one open disk or half-disk are made
up of 2 individuals. See text for full description of experimental design
Author copy

Simonson et al.: Warming seawater effects on kelp quality
Temper atur e-induc ed change s in ke lp as a
food source
Material collection and preparation.Lacuna vinc -
ta were collected along with associated S. latissima
from 8 to 12 m depth at Splitnose Point in June/July
2013. Snails and kelp were transported to the labo-
ratory in coolers. Within 24 h of collection, L. vincta
were removed from the kelp blades and placed
within fine mesh bags in a tank with flow-through
ambient seawater. Once cleaned of snails, thalli of
S.latissima were suspended in the temperature
treatment tanks (as in temperature experiment) and
pretreated for 10 d at either 11 or 21°C before use in
feeding or survival, growth and reproduction exper-
iments. S. latissima were placed in temperature pre-
treatments within 24 h of collection for use in feed-
ing experiments, and within 2 wk of collection for
the survival, growth and reproduction experiment.
Be fore the start of the feeding experiments, L.
vincta were fed kelp ad libitum for 7 d, and then
starved for 3 d.
Feeding experiments. The effect of temperature-
induced changes in kelp tissue quality on L. vincta
feeding rate and preference were examined by
conducting choice and no-choice feeding experi-
ments with 2 diet treatments: S. latissima pretreated
at 11 or 21°C. Two disks of kelp (5.5 cm diameter)
were ex cised 30 cm from the blade−stipe interface
from each of 8 individuals of S. latissima (main-
tained for 10 d in the experimental temperatures),
and then cut in half. In the choice experiment, the
initial blotted wet weight of each half-disk was
recorded, and half-disks from the 2 diet treatments
were paired by weight (Fig. 1). Each set of paired
half-disks was placed in a flow-through cylindrical
container (10 cm diameter, 8 cm height, with
0.2 cm holes) with a mesh top (n = 8). Four of the
containers were randomly designated as autogenic
controls to record changes in kelp mass in the
absence of snails. Groups of 8 snails (shell length 4
to 7 mm) were individually measured (0.1 mm pre-
cision) and added to each of the other 4 containers.
In the no-choice experiment, half-disks from the
same individual of S. latissima were paired,
weighed and placed together in 16 flow-through
containers. Eight containers (4 with S. latissima
pre treated at 11°C, 4 with S. latissima pretreated at
21°C) were randomly designated as autogenic con-
trols and the remaining 4 containers for each diet
treatment received groups of 8 snails from the
same population and size range as those in the
choice experiment (Fig. 1).
The containers were weighted and submerged,
and randomly interspersed in the same flow-through
seawater table, where they were maintained for 5 d.
The mass of algal tissue grazed in each container was
calculated as the change in blotted wet weight
(0.001 g precision) of the half-disk(s) from the begin-
ning to the end of the 5 d period. For both the choice
and no-choice experiments, 3 trials with 3 to 4 repli-
cate containers were conducted in June and July
2013. Water temperature during each trial was (°C,
mean ± SD, n = 6): trial 1, 7.89 ± 1.27; trial 2, 10.17 ±
0.68; trial 3, 8.96 ± 1.58. Because of a recording fail-
ure within the Aquatron facility at Dalhousie Univer-
sity, temperature during each trial was obtained from
average daily seawater temperature at 8 m depth at
Splitnose Point (~15 km from the Aquatron seawater
intake, and at the same depth). Temperatures in
Aquatron during August 2013 were generally within
1°C of those at Splitnose Point (mean ± SD difference
in temperature: 0.11 ± 0.86°C; n = 28)
Survival, growth and egg production. The effect
of temperature-induced changes to kelp tissue qual-
ity on the survival, growth and reproduction of L.
vincta was determined in an 8 wk laboratory experi-
ment (13 July to 5 September 2013). Groups of 8
snails were placed in flow-through containers (as in
feeding experiments) and randomly assigned 1 of 3
diet treatments: S. latissima pretreated at 11°C, S.
latissima pretreated at 21°C, or starved controls (n =
15 containers for each treatment). All containers
were randomly placed in a flow-through seawater
table. Fed treatments were provided 5.5 cm diameter
disks of kelp pretreated for 10 d in the 11 or 21°C
temperature treatments, as in the feeding experi-
ments. Kelp disks were replaced weekly. Snails were
marked with nail polish and their individual growth
rates recorded by measuring changes in shell length
from digital photographs, taken initially and then at
biweekly intervals for 8 wk. Shell length was meas-
ured using image analysis software (ImageJ). Initial
shell lengths were 3 to 8 mm, and did not differ
among treatments (1-way ANOVA: F2, 357 = 0.33, p =
0.72). Growth was calculated as the change in shell
length over each 2 wk sampling interval. Dead snails
and egg masses were counted and removed at each
measurement interval. Reproductive output per snail
for each container was calculated as the number of
egg masses produced divided by the number of
snails. Upon termination of the experiment, a subset
of surviving snails (from 30 containers) was sexed.
Sex ratio in the sampled containers did not differ
among treatments (1-way ANOVA, F2, 27 = 0.11, p =
0.90) and the mean (± SD) proportion of females was
109
Author copy

Mar Ecol Prog Ser 537: 105–119, 2015
0.51 ± 0.29 (n = 123). The sex ratio in each container
was therefore assumed to be 1:1.
Statistical analyses. Analysis of feeding experi-
ments incorporated the controls for autogenic
changes in kelp mass as described in Peterson &
Renaud (1989). In the no-choice experiment, mass
changes (mg d−1) of half-disks were analyzed using
3-way ANOVA with diet treatment (pretreatment at
11 or 21°C) and herbivore (presence, absence) as
fixed factors, and trial as a random factor. There was
a significant trial × herbivore × diet interaction (F2,32
= 11.42, p < 0.001), so trials were analyzed separately
with 2-way ANOVA. Data in trial 3 were log trans-
formed to meet the assumption of homogeneity of
variance (Levene’s test, p > 0.05). In the choice exper-
iment, the difference in mass change between diet
treatments was calculated for each replicate con-
tainer, and the differences for containers with and
without snails were compared using a 2-way
ANOVA with herbivore (presence, absence) as a
fixed factor and trial as a random factor. Although the
interaction between trial and herbivore was not sig-
nificant (F2, 16 = 13.03, p = 0.11), the trials were ana-
lyzed separately using 1-way ANOVA for consis-
tency with the no-choice feeding experiment. Data in
trial 3 did not meet the assumption of homogeneity of
variance even after transformation, and results from
untransformed data are presented.
Differences in growth rate of L. vincta between diet
treatments (S. latissima pretreated at 11 or 21°C;
fixed factor) and among weeks (random factor), with
repeated measures within weeks, were examined
using repeated-measures ANOVA on the average
length change (mm) of snails within each container
over each 2 wk interval. Similarly, differences in the
effect of diet (kelp pretreated at 11 or 21°C; fixed fac-
tor) on the number of egg masses snail−1and propor-
tion of snails surviving over each 2 wk interval (ran-
dom factor) in each container were analyzed using
repeated measures ANOVA, with repeated meas-
ures within weeks. The proportion of snails surviving
violated the assumption of sphericity (Mauchly’s test,
p < 0.05), and the Greenhouse-Geisser adjustment
was used for p-values.
Temper atur e-induc ed change s in ke lp as a
substrate
Material collection and preparation. The effect of
the temperature-induced changes in kelp tissue on
settlement of Membranipora membranacea was
examined with a settlement choice experiment. Lar-
vae of M. membranacea were isolated from plankton
samples collected from St. Margaret’s Bay (~35 km
west of Splitnose Point) in October/November 2013
and July/August 2014. S. latissima was collected over
the same periods from 8 to 12 m depth at Splitnose
Point, and pretreated for 10 d in experimental tanks
with either ambient (see below) or 21°C seawater
(tank set up as in temperature experiment). Tissue
samples (trials 1 and 2: 11 cm2half-disks; trials 3 to 5:
8cm
2disks) were excised from pretreated kelp,
30 cm from the blade-stipe interface, paired and
placed in 250 ml beakers with filtered seawater. Pairs
consisted either of 2 samples pretreated at ambient
temperature, 2 samples pretreated at 21°C, or 1 sam-
ple pretreated at ambient temperature and 1 sample
at 21°C (Fig. 1). Groups of 30 competent larvae of M.
membranacea were visually identified and isolated
from the plankton samples using a dissecting micro-
scope (20× magnification), and then added to each
beaker and allowed 72 h to settle on the kelp sub-
strates. During this time, beakers were placed in a
continuous-flow seawater table to maintain the
beakers at ambient seawater temperature. After 72 h,
the kelp was examined microscopically and the num-
ber of larvae that settled on each sample was
counted. 3 trials were conducted in October to
November 2013 (mean ± SD ambient water tempera-
ture: 14.4 ± 2.2°C, n = 30) and 2 trials in July to
August 2014 (9.3 ± 2.1°C, n = 20). Ambient water
temperature was recorded in the pretreatment tank
at 1 h intervals with data loggers (Maxim Integrated
Thermochron iButtons). A total of 28 larvae were
scored, and all were observed to settle on one of the
2 kelp substrates. Although area of tissue varied
among trials, total settlement within beakers did not
differ with total area available for settlement (1-way
ANOVA, F1, 18 = 0.02, p = 0.40). The number of set-
tlers on each sample was divided by the area of the
sample, giving settlers cm−2.
Statistical analyses.The effect of the temperature-
mediated quality of the kelp tissue on settlement of M.
membranacea was determined by comparing the
number of larvae that settled on a substrate (kelp pre-
treated at ambient temperature or 21°C) when larvae
were given a choice of substrates to the number of lar-
vae that settled on that substrate when there was no
choice. Beakers in which no settlement oc curred were
not included in the analyses. For the comparisons us-
ing kelp pretreated at ambient temperature, all 5
trials were included in the analysis. However, for the
comparisons using kelp pretreated at 21°C, only 3 tri-
als (those with settlement in both choice and no-
choice beakers) were included. In the no-choice
110
Author copy

Simonson et al.: Warming seawater effects on kelp quality
beakers, settlement did not differ between the 2 tissue
samples within each beaker (paired t-tests, ambient
beakers: t5= 0.31, p = 0.77; 21°C beakers: t4= −0.93,
p= 0.41); to maintain independence of replicates, 1
sample was randomly selected from each no-choice
beaker for the analysis. The effect of treatment
(choice, no-choice; fixed factor) and trial (random fac-
tor) on the number of settlers cm−2was examined us-
ing 2-way ANOVA for each temperature pretreatment
(ambient, 21°C). Because there was no treatment by
trial interaction (p > 0.20) at either temperature, the
interaction was removed and treatment and trial were
tested against pooled treatment × trial MS and resid-
ual MS (Winer et al. 1991).
RESULTS
Chemical composition of kelp tissue
There was no effect of temperature
on C/N ratio for any of the 3 kelp spe-
cies after 1 wk exposure be tween 11
and 21°C, or after 3 wk exposure be -
tween 11 and 18°C (Fig. 2, Table 1).
There was a trend of in creasing C/N
ratio over the summer in both 2013
and 2014 (Fig. 3), although differ-
ences among sampling dates were
not significant (Table 2).
Phlorotannin content of Agarum
clathratum was significantly lower
after 1 wk exposure to 21°C com-
pared to 11°C (Fig. 2, Table 3), but
there was no difference after 2 wk ex -
posure between 11 and 18°C (Fig. 2,
Table 3). Phlorotannin content in
Laminaria digitata and Saccharina
latissima was approximately 10% and
25% that of A. clathratum, respec-
tively (Fig. 2). Phlorotannin content of
L. digitata was slightly lower (by
~0.2% dry wt) at 18°C than 11°C,
after 2 wk exposure. There was no
difference in phlorotannin content for
S. latissima species after 1 wk expo-
sure between 11 and 21°C, or after
2wk exposure between 11 and 18°C
(Fig. 2, Table 3). Phlorotannin content
of S. latissima collected from Split-
nose Point was higher in 2013 than
in 2014 across all sampling dates
(Fig. 3). During 2013, phlorotannin
content increased throughout the summer (Fig. 3),
although there were no significant differences
among sampling dates (Table 2). In 2014, phlo ro -
tannin content of S. latissima increased slightly from
June to August and then decreased significantly by
0.6% dry wt in September (Fig. 3, Table 2).
Temper atur e-induc ed change s in ke lp as a
food source
In the no-choice feeding experiment, kelp loss was
greater in the presence of Lacuna vincta, and auto-
111
Fig. 2. Mean (+ 1 SE) C/N ratio (n = 3) and phlorotannin content (% dry wt; n =
9) of Agarum clathratum, Laminaria digitata and Saccharina latissima immedi-
ately after collection from the field (initial) and after 1 wk exposure to 11 and
21°C, and 2 wk or 3 wk exposure to 11 and 18°C in the laboratory
Species Exposure Temperature t df p
(wk) (°C)
Agarum clathratum 1 11 vs. 21 1.24 2.1a 0.34
3 11 vs. 18 1.37 4 0.24
Laminaria digitata 1 11 vs. 21 1.45 2.0a 0.28
3 11 vs. 18 −0.73 4 0.51
Saccharina latissima 1 11 vs. 21 −3.76 4 0.73
3 11 vs. 18 1.17 4 0.30
aWelch-Satterthwaite adjustment due to unequal variance (F-test, p < 0.05)
Table 1. Results of independent samples t-test to examine differences in C/N
ratio between 11 and 21°C temperature treatments after 1 wk exposure, or be-
tween 11 and 18°C treatments after 3 wk exposure, for each kelp species
Author copy

Mar Ecol Prog Ser 537: 105–119, 2015
genic loss of kelp pretreated at 21°C was greater
than for kelp pretreated at 11°C (Fig. 4, Table 4). In
trial 1, a significant diet by herbivore interaction indi-
cated that grazing rates of L. vincta were greater on
kelp pretreated at 21°C than at 11°C (Fig. 4, Table 4).
There was no difference in grazing rates between
diets in trials 2 and 3, as indicated by the non-signif-
icant interaction term (Fig. 4, Table 4). In the choice
experiment, the difference in mass change between
half-disks of kelp pretreated at 11 and 21°C was sig-
nificantly greater in treatments with L. vincta than
without L.vincta in trial 1, indicating that snails fed
preferentially on kelp pretreated at 21°C (Fig. 4,
Table 5). This was not the case in trials 2 and 3,
although the direction of differences
was consistent (Fig. 4, Table 5). As in
the no-choice experiment, autogenic
mass loss was greater for kelp pre-
treated at 21°C in the choice experi-
ment (Fig. 4).
At the end of the 8 wk growth ex -
tperiment, mean growth (change in
shell length) in containers of snails
fed either of the 2 kelp diets was an
order of magnitude higher than the
mean growth of starved controls
(Fig. 5a). There was no effect of kelp
diet (S. latissima pretreated at 11 or
21°C) on the growth of snails over
the 8 wk experiment, and no interac-
tion be tween kelp diet and sampling
week, although growth did differ
among sampling weeks (Table 6).
There was no production of egg
masses in the starved controls after
week 2, and egg production snail−1
increased with sampling week in
containers fed either kelp diet
(Fig. 5b). There was no difference in
the number of egg masses produced snail−1be tween
containers fed kelp pretreated at 11°C and those fed
kelp pretreated at 21°C (Table 6). Egg production did
differ among weeks, but there was no interaction
between kelp diet and sampling week (Table 6).
Mean survival of snails in the experiment was high,
exceeding 90% over the first 6 wk in all diet treat-
ments (Fig. 5c). After 8 wk, mean survival of the
starved controls declined to 75% (Fig. 5c). There was
no difference in the survival between the 2 kelp
diets, and no interaction between diet and sampling
week, but there was a difference in survival among
weeks (Table 6).
Temper atur e-induc ed change s in ke lp as a
substrate
There was no difference in the number of larvae of
Membranipora membranacea cm−2that settled on S.
latissima pretreated at 21°C or at ambient tempera-
ture when it was offered as a choice, compared to
when it was offered without a choice (Fig. 6, Table 7).
The lack of a difference between choice and no-
choice treatments indicates that larvae of M. mem-
branacea have no preference for, or avoidance of,
kelp pretreated at 21°C compared to kelp pretreated
at ambient temperature.
112
Year Variable F df p Tukey HSD
2013 C/N 3.70 3, 8 0.06
Phl 2.61 4, 31 0.09
2014 C/N 1.46 3, 8 0.30
Phl 4.08 3, 24 0.001 Sep < Aug = Jul
Table 2. Results of ANOVA comparing phlorotannin content
(Phl) or C/N ratio of blade tissue of Saccharina latissima in
the field, among sampling dates (2013: 18 and 25 June, 15
July (Phl only), 28 July, 27 August; 2014: 30 June, 31 July, 27
August, 25 September). Significant results are shown in bold
Fig. 3. Mean (± 1 SE) C/N ratio (n = 3) and phlorotannin content (% dry wt; n =
10) of blade tissue of Saccharina latissima collected monthly in summer 2013
and 2014 from 8 m depth at Splitnose Point (black squares), and mean daily
temperature at the collection site
Author copy

Simonson et al.: Warming seawater effects on kelp quality
DISCUSSION
Effects of temperature on chemical composition ofÊ
kelp tissue
The increase in C/N ratio observed in Saccharina
latissimaover the summer follows the expected pat-
tern of seasonal variation in this species (Gevaert etÊ
al. 2001, Nielsen et al. 2014). This pattern resultsÊ
from both a decrease in N and an increase in C overÊ
the summer (see Fig. S1 in the Supplement at www.Ê
int-res. com/articlesÉ suppl/ m537 p105_ supp. pdf,Ê
Gevaert et al. 2001, Nielsen et al. 2014). Over winter,Ê
when growth is light-limited, N is stored and thenÊ
used to support growth in spring and summerÊ
(Nielsen et al. 2014). During N-limited growth in
113
Exposure Source of variation df MS F p
(wk)
Agarum clathratum
1Temp. (11 vs. 21) 1 4.27 6.37 0.03
Trial 2 0.43 0.64 0.54
Temp. × Trial 2 0.38 0.51 0.61
Residual 10 0.73
2Temp. (11 vs. 18) 1 0.30 0.22 0.64
Trial 2 2.16 1.56 0.24
Temp. × Trial 2 0.47 0.31 0.74
Residual 12 1.53
Laminaria digitata
1Temp. (11 vs. 21) 1 0.27 2.64 0.13
Trial 2 0.14 1.38 0.28
Temp. × Trial 2 0.05 0.50 0.62
Residual 12 0.11
2Temp. (11 vs. 18) 1 0.15 5.28 0.04
Trial 2 0.06 2.20 0.15
Temp. × Trial 2 0.04 1.65 0.23
Residual 11 0.03
Saccharina latissima
1Temp. (11 vs. 21) 1 0.43 0.69 0.42
Trial 2 1.02 1.65 0.23
Temp. × Trial 2 0.46 0.71 0.51
Residual 12 0.65
2Temp. (11 vs. 18) 1 0.09 0.21 0.65
Trial 2 0.67 1.51 0.25
Temp. × Trial 2 0.31 0.66 0.54
Residual 12 0.46
Table 3. Results of 2-way ANOVA to examine differences in
phlorotannin content (% dry wt) between 11 and 21°C tem-
perature treatments (fixed factor) and trials (random factor)
after 1 wk exposure, or between 11 and 18°C treatments after
2 wk exposure, for each kelp species. Temperature and Trial
were tested against pooled Temp. × Trial MS and residual
MS. Significant results are shown in bold
Fig. 4. Mean (+ 1 SE) rates of mass loss of Saccharina latissima
pretreated for 10 d at 11°C or 21°C in the presence of Lacuna
vincta, or in controls without L. vincta in 3 trials of choice and
no-choice feeding experiments (trial 1: n = 3; trials 2 & 3: n = 4)
Trial Source df Fp
1Diet 1, 8 45.81 <0.001
Herbivore 1, 8 30.55 <0.001
Diet × Herbivore 1, 8 23.53 0.001
2Diet 1, 12 13.25 0.003
Herbivore 1, 12 6.08 0.03
Diet × Herbivore 1, 12 0.00 0.99
3a Diet 1, 12 33.89 <0.001
Herbivore 1, 12 18.91 <0.001
Diet × Herbivore 1, 12 3.87 0.073
aMass change log-transformed to meet assumption of
homoscedasticity
Table 4. Results of 2-way ANOVA examining effects of diet
(Saccharina latissima pretreated for 10 d at 11 or 21°C) and
herbivore (presence or absence of Lacuna vincta) on rate of
kelp mass loss (mg d−1) in 3 trials of the no-choice feeding
experiment. Significant results are given in bold
Trial df F p
11, 4 102.6 <0.001
21, 6 0.016 0.91
3a1, 6 0.605 0.47
aData heteroscedastic even after transformation.
Untrans formed data are presented
Table 5. Results of 1-way ANOVA comparing difference in
rate of mass change (mg d−1) of Saccharina latissima pre-
treated for 10 d at 11 vs. 21°C between herbivore treatments
(presence or absence of Lacuna vincta) in 3 trials of the choice
feeding experiment. Significant results are given in bold
Author copy

Mar Ecol Prog Ser 537: 105–119, 2015
summer, C is stored as carbohydrates to allow for
continued growth through the winter (Nielsen et al.
2014). It has been suggested that changes in temper-
ature also may contribute to this pattern. A decrease
in N content with increasing temperature has been
observed for S. latissima from Helgoland (Olis-
chläger et al. 2014) as well as in the kelps Eklonia
radiata (Staehr & Wernberg 2009) and Eisenia
arborea (Matson & Edwards 2007). We found that
neither 1 wk exposure to 21°C nor 3 wk exposure to
18°C changed the C/N ratio in any of the species we
examined, relative to 11°C. N content (% dry wt) of
field-collected kelps at the start our experiment
(mean ± SE: Agarum clathratum, 1.29 ± 0.02; Lami-
naria digitata, 0.92 ± 0.05; S. latissima, 0.67 ± 0.08)
was less than the critical value of 1.8%, below which
the growth of S. latissima is N-limited (Chapman et
al. 1978). This suggests that upon collection any
stores of N had already been consumed and the kelps
were N limited throughout the experiment.
Phlorotannin content in Agarum clathratum and
L. digitata was reduced after 1 wk exposure to 21°C
and 2 wk exposure to 18°C, respectively. Phlorotan-
nins are a known antioxidant defense in brown
algae, induced by exposure to ultraviolet radiation
(Gómez & Huovinen 2010, Cruces et al. 2012, 2013).
In these studies, phlorotannins were not induced
under thermal stress for periods of up to 72 h,
despite increased lipid peroxidation indicative of a
rise in activity of reactive oxygen species (Cruces et
al. 2012, 2013). Furthermore, temperature stress (20
and 28°C) was found to inhibit induction of phloro-
tannin by ultraviolet radiation (Cruces et al. 2012,
2013), suggesting that high temperatures may hin-
der the ability of kelp to produce phlorotannins,
possibly by damaging the membranes of the Golgi-
ER complex, where phlorotannins are produced
(Schoenwaelder & Clayton 2000).
Temperature did not effect phlorotannin content of
S. latissima and had only a small effect on phlorotan-
nin content of L. digitata. The observed ~0.2% dry wt
decrease in phlorotannin content in L. digitata, is
unlikely to be ecologically significant, as concentra-
tions of phlorotannins below 1% dry wt have not
been reported to deter herbivores (Targett & Arnold
1998). Phlorotannin content of S. lattisima and L. dig-
itata were low, although similar to previously re -
ported levels in these species: 0.9 − 2.5 % dry wt and
0.15 − 0.3 % dry wt, respectively (Johnson & Mann
1986, Connan et al. 2006, Dubois & Iken 2012),
reducing our ability to detect any effect of tempera-
ture. Measurement of non-phlorotannin compounds
may have further hindered our ability to detect any
effect of temperature (Van Alstyne 1995).
In the field, there was a pattern of increasing phlo -
ro tannin content over the summer in Saccharina
latissima, followed by a decrease in September in
2014. Although phlorotannin content was similar in
summer and winter in S. latissima, and unrelated to
irradiance or nutrient availability (Dubois & Iken
2012), in other brown algae (primarily the order
Fucales) phlorotannin content peaks in the spring or
summer (Steinberg 1995, Stiger et al. 2004, Kamiya
et al. 2010). Higher phlorotannin content in summer
has been attributed to increased grazer density,
which can induce phlorotannin production (Van Al -
styne 1988), or greater energy availability for phlo ro -
tannin production during periods of growth limita-
tion (Steinberg 1995, Stiger et al. 2004). Phlorotannin
114
Fig. 5. (a) Mean (± 1 SE) growth (length change) of Lacuna
vincta relative to initial shell length when fed 3 diets: Sac-
charina latissima pretreated 10 d at 11°C, S. latissima pre-
treated 10 d at 21°C, or no kelp. SE represents the variation
among 15 containers. (b) Mean (+ 1 SE) egg masses snail−1
produced by L. vincta when fed 3 diets (as above; n = 15 con-
tainers). (c) Mean (± 1 SE) proportion of L. vincta surviving
(relative to initial number of individuals) when fed 3 diets (as
above; n = 15 containers)
Author copy

Simonson et al.: Warming seawater effects on kelp quality
content of S. latissima was higher in 2013
than in 2014, by ~1% dry wt. Interannual
variation in phloro tannin content can
reflect changes in environmental condi-
tions, such as grazer densities, irradiance
or nutrient availability (Van Alstyne
1988, Pavia & Toth 2000).
Effects of temperature on kelp as a
food source
Lacuna vincta consumed more kelp
pre treated at 21 than at 11°C in trial 1 of
both the choice and no-choice feeding
experiments, but not in the other 2 trials.
The rate of consumption of kelp in trial 1
was more than twice that in trial 2 or 3 in
the choice experiment. The rate of con-
sumption of kelp pretreated at 21°C in
trial 1 was at least 2.5-fold greater than
that in trials 2 or 3 in the no-choice ex -
periment. Differences in ambient tem-
perature among trials could have caused
these differences in feeding rate in
response to metabolic demand; however,
temperature in trial 1 was lower than that
in trial 2 or 3.
The greater feeding rate of L. vincta on
kelp pretreated at 21°C in trial 1 could
reflect a temperature-induced change in
the palatability of kelp tissue that is apparent only at
high feeding rates. Given that the chemical quality of
S. latissima was unaffected by temperature, temper-
ature-induced changes to the mechanical properties
of kelp may account for the observed increased feed-
115
Variable Source df F p
Proportion survivinga Between Subjects
Diet 1 0.0003 0.98
Container(Diet) 28
Within Subjects
Week 3 13.73 <0.001
Week × Diet 3 0.90 0.42
Week × Container(Diet) 84
Growth Between Subjects
Diet 1 0.52 0.48
Container(Diet) 28
Within Subjects
Week 3 11.28 <0.001
Week × Diet 3 0.71 0.55
Week × Container(Diet) 84
Egg production Between Subjects
Diet 1 0.21 0.65
Container(Diet) 28
Within Subjects
Week 3 8.25 <0.001
Week × Diet 3 1.05 0.37
Week × Container(Diet) 84
aAssumption of sphericity violated (Mauchly’s test p < 0.001), and
Greenhouse-Geisser adjustment applied
Table 6. Results of repeated-measures ANOVA comparing proportion of
Lacuna vincta surviving, average growth (length change) and egg produc-
tion (egg masses snail−1) between 2 diets (Saccharina latissima pretreated
at 11 or 21°C; fixed factor). Repeated measures were taken at 2 wk intervals
(2, 4, 6 and 8 wk). Significant results are given in bold
Fig. 6. Mean (+ 1 SE) number of settlers of Membranipora
membranacea cm−2on Saccharina latissima pretreated for
10 d at ambient temperature, or S. latissima pretreated for
10 d at 21°C, when a choice of substrates was offered and in
a control with no choice of substrates (data from 5 trials
pooled, n = 5 to 9). Ambient temperature (mean ± SD): 14.4 ±
2.2°C for 3 trials in Oct/Nov 2013; 9.3 ± 2.1°C for 2 trials in
Jul/Aug 2014
Temp erature Source of variation df MS F p
Ambient Treatment 1 0.0019 0.31 0.59
Trial 4 0.0048 0.76 0.57
Treatment × Trial 4 0.0088 2.05 0.23
Residual 5 0.0043
21°C Treatment 1 0.0006 0.12 0.74
Trial 2 0.0031 0.60 0.57
Treatment × Trial 2 0.0054 1.04 0.42
Residual 5 0.0052
Table 7. Results of 2-way ANOVA examining the effect of
treatment (choice of substrates or no-choice; fixed factor)
and trial (random factor) on the settlement of larvae of Mem-
branipora membranacea in 2 temperature treatments, ambi-
ent (9.3−14.4°C) and 21°C. Treatment and Trial were tested
against pooled Treatment × Trial MS and residual MS.
Author copy

Mar Ecol Prog Ser 537: 105–119, 2015
ing rate. Higher grazing rates on kelp pretreated at
21°C in both the choice and no-choice experiments
indicate that L. vincta did not have an active prefer-
ence for this tissue, but rather that kelp pretreated at
21°C is easier to consume. L. vincta has a taenioglos-
san radula that is more efficient on softer tissues (Ste-
neck & Watling 1982), and prefers young tissue, even
in the absence of any difference in C/N ratio or phlo-
rotannin content, likely due to differences in tissue
toughness (Toth & Pavia 2002, Chenelot & Konar
2007). Increasing temperature damages and weak-
ens kelp tissue (Table 8), and this weaker tissue may
be easier for L. vincta to consume.
The greater consumption of kelp pretreated at
21°C in trial 1 indicates that the effect of temperature
on the palatability of kelp tissue may depend on the
feeding rate of L. vincta, and that the direct effects of
temperature on kelp tissue loss and temperature-
induced changes in herbivory of L. vincta can be syn-
ergistic when feeding rates are high. However, when
feeding rates are lower (as in trials 2 and 3) there is
no evidence for temperature-induced changes in
palatability of S. latissima, and the impacts of tem-
perature and L.vincta on kelp tissue loss will likely be
additive.
S. latissima is a nutritious food source for L. vincta
that supports greater growth rates than other algal
diets (Chavanich & Harris 2002). In our study, sur-
vival, growth and egg production of snails were unaf-
fected by temperature-induced changes in kelp tis-
sue, indicating that the nutritional quality of kelp did
not change, as evidenced by the lack of variation in
C/N between temperature pretreatments. Mean
growth rate of fed snails in our experiment (0.04 to
0.23 mm wk−1) were similar to those previously
recorded in the laboratory on a diet of S. latissima
(0.04 to 0.12 mm wk−1; Chavanich & Harris 2002).
The similarity in the success of L. vincta (similar
growth rates, survival, and reproductive output) fed
kelp diets pretreated at 11 and 21°C suggests that
populations of L. vincta will likely remain stable with
increasing temperature, as long as kelp is available:
i.e. there are no indirect effects of temperature that
affect growth, survival or reproduction of snails.
Effects of temperature on kelp as a substrate
Temperature pretreatment of kelp tissue did not
affect larval settlement of Membranipora membrana -
cea. Abundance of settlers observed on kelp pre-
treated at ambient temperatures and that pretreated
at 21°C (mean: 0.05 to 0.07 settlers cm−2) were com-
parable to the abundance of settlers on S. latissima in
the field during peak settlement (mean: 0.03 to 0.19
settlers cm−2; Saunders & Metaxas 2007).
Larvae of M. membranacea demonstrate settle-
ment preferences both among kelp species and
among locations on the kelp thallus (Brumbaugh et
al. 1994, Matson et al. 2010). They also exhibit fine-
scale searching behaviour along the kelp substrate,
which suggests they are able to detect differences in
substrate quality (Matson et al. 2010). Across kelp
species, larvae of M. membranacea show a prefer-
ence for settlement on young tissue proximal to the
blade meristem (Brumbaugh et al. 1994, Denley et al.
2014). The cues attracting larvae to
young tissue are unknown, but the
persistence of the preference when
flow is reversed suggests that larvae
use a physical or chemical character-
istic of the substrate to cue settlement
(Brumbaugh et al. 1994). Phloro -
tannins have been suggested as a
chemical cue preventing the settle-
ment of fouling organisms (Wikström
& Pavia 2004). Because the phloro-
tannin content of S. latissima was
unaffected by temperature, the kelp
substrates in our experiment (pre-
treated at 21°C and pretreated at
ambient temperature, 9.3−14.4°C)
had similar levels of chemical deter-
rents. Brumbaugh et al. (1994) found
that damage to the blade reduced
settlement and postulated that lar-
116
Property Species Temperature Reference
C/N A. clathratum No effect This study
L. digitata
S. latissima
Phlorotannin A. clathratum 21°C This study
L. digitata No effect
S. latissima
Strength A. clathratum No effect Simonson et al. (2015)
L. digitata 18, 21°C
S. latissima 14, 18, 21°C
Extensibility A. clathratum No effect Simonson et al. (2015)
L. digitata 21°C
S. latissima 18, 21°C
Table 8. Effects of temperature treatments (11, 14, 18 and 21°C) on chemical
and mechanical metrics of kelp tissue quality (C/N, phlorotannin content [%
dry wt], blade tissue strength [MPa] and blade tissue extensibility [% length
change]) for Agarum clathratum, Laminaria digitata and Saccharina latis-
sima. Quality metrics were significantly decreased by exposure to the listed
temperatures
Author copy

Simonson et al.: Warming seawater effects on kelp quality
vae may avoid older sections of the blade due to
greater levels of physical damage there. Tempera-
ture stress of 21°C also damages and weakens kelp
tissue (Table 8), however we saw no effect of this
damage on settlement preference.
The lack of any temperature-induced changes in
kelp quality as a substrate for settlement of M. mem-
branacea suggests that increased population growth
rate due to warming seawater temperatures (Saun-
ders & Metaxas 2007, 2009) will not be retarded by
lower settlement rates. The direct effects of tempera-
ture on kelp combined with temperature-mediated
increases in M. membranacea populations are likely
to cause large-scale biomass loss from kelp beds in
Nova Scotia. However, this biomass loss could be
miti gated if temperature-induced changes in sub-
strate quality affect post-settlement mortality. Tem-
perature-induced damage to the meristoderm could
cause it break and peel away with the associated
bryozoan colonies. Peeling of the outer layer of cells
has been documented as a mechanical defense
against fouling in several algal species (Sieburth &
Tootle 1981, Nylund & Pavia 2005). Even in the
absence of changes to post- settlement mortality, the
loss of kelp biomass may in turn limit further out-
breaks of M. membranacea by limiting the availabil-
ity of preferred substrate.
CONCLUSIONS
An understanding of the simple and cumulative
effects of environmental conditions on species inter-
actions is imperative for predicting the effects of cli-
mate change on community function. We expected
that warming seawater temperature would alter kelp
tissue quality, which would in turn affect interactions
between kelp and the mesograzer L. vincta and en -
crusting bryozoan M. membranacea. However,
higher temperature did not affect the quality of S.
latissima as a substrate for M. membranacea, and
we observed temperature-induced changes in the
palatability of kelp only when grazing rates of L.
vincta were high. Nutrient content and chemical
defense of both S. latissima and Laminaria digitata
were not strongly affected by temperature, suggest-
ing that the quality of these species as a food and a
substrate would be similarly unaffected by increases
in temperature. A temperature-induced reduction in
the chemical defenses of Agarum clathratum at 21°C
suggests that this species might become more vul-
nerable to damage by L. vincta or M. membranacea
at high temperatures.
We predict that the direct effects of temperature on
kelp tissue, herbivory by L. vincta and encrustation
by M. membranacea will act additively to increase
biomass lost from kelp beds as seawater temperature
increases. Warmer temperatures are expected to in -
crease both outbreaks of M. membranacea (Scheib -
ling & Gagnon 2009, Saunders & Metaxas 2007,
2009) and metabolic rates, and therefore feeding, of
herbivores (O’Connor 2009). Encrustation by bryo -
zoans, herbivory by L. vincta, and temperature-
in duced damage all weakened kelp tissue and in -
creased vulnerability to wave forces (Krumhansl et
al. 2001, Table 8). No antagonistic effects were man-
ifested through the inhibition of feeding by snails or
settlement by M. membrana cea that would mitigate
other temperature effects on kelp. Using future cli-
matic conditions (warming temperatures and larger
waves), models of kelp detrital production have also
predicted a loss of kelp biomass in Nova Scotia
(Krumhansl et al. 2014). Reductions in standing kelp
biomass could impact both habitat availability and
community productivity (Dayton 1985, Steneck et al.
2002), while changes in the production and export of
kelp detritus will impact adjacent coastal and deep-
water ecosystems (Krumhansl & Scheibling 2012).
Acknowledgements. We thank J. Lindley, K. Filbee-Dexter,
C. Feehan, J. O’Brien, D. Denley, and F. Francis for assis-
tance with field work, and M. Harris, D. Durette Morin and J.
O’Brien for their help in the laboratory. This research was
funded by Discovery Grants to A.M. and R.E.S from the Nat-
ural Sciences and Engineering Research Council (NSERC) of
Canada. E.J.S. was supported by an NSERC Canada Gradu-
ate Scholarship and a Dalhousie University Scholarship.
LITERATURE CITED
Amsler CD, Fairhead VA (2006) Defensive and sensory
chemical ecology of brown algae. Adv Bot Res 43: 1−91
Andersen GS, Steen H, Christie H, Fredriksen S, Moy FE
(2011) Seasonal patterns of sporophyte growth, fertility,
fouling, and mortality of Saccharina latissima in Skager-
rak, Norway: implications for forest recovery. J Mar Biol
2011: 690375
Andersen GS, Pedersen MF, Nielsen SL (2013) Temperature
acclimation and heat tolerance of photosynthesis in Nor-
wegian Saccharina latissima (Laminariales, Phaeo-
phyceae). J Phycol 49: 689−700
Boettcher AA, Targett NM (1993) Role of polyphenolic
molecular size in reduction of assimilation efficiency in
Xiphister mocusus. Ecology 74: 891−903
Bolton JJ, Lüning K (1982) Optimal growth and maximal
survival temperatures of Atlantic Laminaria species
(Phaeo phyta) in culture. Mar Biol 66: 89−94
Brumbaugh DR, West JM, Hintz JL, Anderson FE (1994)
Determinants of recruitment by an epiphytic marine
bryo zoan: field manipulations of flow and host quality.
117
Author copy

Mar Ecol Prog Ser 537: 105–119, 2015
In: Wilson WH, Stricker SA , Shinn GL ( eds) Reproduct ion
and development of marine invertebrates. Johns Hop-
kins University Press, Baltimore, p 287−301
Chapman ARO, Markham JW, Lüning K (1978) Effect of
nitrate concentration on the growth and physiology of
Laminaria saccharina (Phaeophyta) in culture. J Phycol
14: 195−198
Chavanich S, Harris LG (2002) The influence of macroalgae
on seasonal abundance and feeding preference of a sub-
tidal snail, Lacuna vincta (Montagu) (Littorinidae) in the
Gulf of Maine. J Mollusc Stud 68: 73−78
Chenelot H, Konar B (2007) Lacuna vincta (Mollusca, Neo-
taenioglossa) herbivory on juvenile and adult Nereocys-
tic luetkeana (Heterokontophyta, Laminariales). Hydro-
biologia 583: 107−118
Cockrell ML, Sorte CJB (2013) Predicting climate-induced
changes in population dynamics of invasive species in a
marine epibenthic community. J Exp Mar Biol Ecol 440:
42−48
Connan S, Delisle F, Deslandes E, Gall EA (2006) Intral-
thallus phlorotannin content and antioxident activity in
Phaeophycea of temperate waters. Bot Mar 49: 39−46
Cruces E, Huovinen P, Gómez I (2012) Phlorotannin and
antioxidant responses upon short-term exposure to UV
radiation and elevated temperature in three South
Pacific kelps. Photochem Photobiol 88: 58−66
Cruces E, Huovinen P, Gómez I (2013) Interactive effects of
UV radiation and enhanced temperature on photosyn-
thesis, phlorotannin induction and antioxident activities
of two sub-Antarctic brown algae. Mar Biol 160: 1−13
Davison IR (1991) Environmental effects on algal photosyn-
thesis: temperature. J Phycol 27: 2−8
Dayton PK (1985) Ecology of kelp communities. Annu Rev
Ecol Syst 16: 215−245
Denley D, Metaxas A, Short J (2014) Selective settlement by
larvae of Membranipora membranacea and Electra
pilosa (Ectocarpa) along kelp blades in Nova Scotia,
Canada. Aquat Biol 21: 47−56
Dubois A, Iken K (2012) Seasonal variation in kelp phloro -
tannins in relation to grazer abundance and environmen-
tal variables in the Alaskan sublittoral zone. Algae 27:
9−19
Duffy JE, Paul VJ (1992) Prey nutritional quality and the
effectiveness of chemical defenses against tropical reef
fishes. Oecologia 90: 333−339
Duggins D, Eckman JE, Siddon CE, Klinger T(2001) Interac-
tive roles of mesograzers and current flow in survival of
kelps. Mar Ecol Prog Ser 223: 143−155
Fernández C (2011) The retreat of large brown seaweeds on
the north coast of Spain: the case of Saccorhiza poly-
schides. Eur J Phycol 46: 352−360
Floerl O, Rickard G, Inglis G, Roulston H (2013) Predicted
effects of climate change on potential sources of non-
indigenous marine species. Divers Distrib 19: 257−267
Fralick RA, Turgeon KW, Mathieson AC (1974) Destruction
of kelp populations by Lacuna vincta (Montagu). Nau-
tilus 88: 112−114
Gerard VA, Du Bois KR (1988) Temperature ecotypes near
the southern boundary of the kelp Laminaria saccharina.
Mar Biol 97: 575−580
Gevaert F, Davoult D, Creach A, Kling R, Janquin MA, Seu-
ront L, Lemoine Y (2001) Carbon and nitrogen content of
Laminaria saccharina in the eastern English Channel:
biometrics and seasonal variations. J Mar Biol Assoc UK
81: 727−734
Gómez I, Huovinen P (2010) Induction of phlorotannins dur-
ing UV exposure mitigates inhibition of photosynthesis
and DNA damage in the kelp Lessonia nigrescens. Pho-
tochem Photobiol 86: 1056−1063
Harley CDG, Hughes AR, Hultgren KM, Miner BG and oth-
ers (2006) The impacts of climate change in coastal mar-
ine systems. Ecol Lett 9: 228−241
Harley CDG, Anderson KM, Demes KW, Jorve JP, Kordas
RL, Coyle TA, Graham MH (2012) Effects of climate
change on global seaweed communities. J Phycol 48:
1064−1078
Hille Ris Lambers J, Harsch MA, Ettinger AK, Ford KR,
Theobald EJ (2013) How will biotis interactions influence
climate change-induced range shifts? Ann N Y Acad Sci
1297: 112−125
Iken K, Amsler CD, Amsler MO, McClintock JB, Baker BJ
(2009) Field studies on deterrent properties of phlorotan-
nins in Antarctic brown algae. Bot Mar 52: 547−557
Johnson CR, Mann KH (1986) The importance of plant
defense abilities to the structure of subtidal seaweed com-
munities: the kelp Laminaria longicruris de la Pylaie sur-
vives grazing by the snail Lacuna vincta (Montagu) at high
population densities. J Exp Mar Biol Ecol 97: 231−267
Kamiya M, Nishio T, Yokoyama A, Yatsuya K, Nishigaki T,
Yoshikawa S, Ohki K (2010) Seasonal variation of phloro-
tannin in sargassacean species from the coast of the Sea
of Japan. Phycol Res 58: 53−61
Koivikko R, Loponen J, Honkanen T, Jormalainen V (2005)
Contents of soluble, cell-wall-bound and exuded phloro -
tannins in the brown alga Fucus vesiculosus, with impli-
cations on their ecological functions. J Chem Ecol 31:
195−212
Kraufvelin P, Salovius S, Christie H, Moy FE, Karez R, Ped-
ersen MF (2006) Eutrophication-induced changes in
benthic algae affect the behaviour and fitness of the mar-
ine amphipod Gammarus locusta. Aquat Bot 84: 199−209
Krumhansl KA, Scheibling RE (2011) Detrital production in
Nova Scotian kelp beds: patterns and processes. Mar
Ecol Prog Ser 421: 67−82
Krumhansl KA, Scheibling RE (2012) Production and fate of
kelp detritus. Mar Ecol Prog Ser 467: 281−302
Krumhansl KA, Lee MJ, Scheibling RE (2011) Grazing dam-
age and encrustation by an invasive bryozoan reduce the
ability of kelps to withstand breakage by waves. J Exp
Mar Biol Ecol 407: 12−18
Krumhansl KA, Lauzon-Guay JS, Scheibling RE (2014)
Modeling effect of climate change and phase shifts on
detrital production of a kelp bed. Ecology 95: 763−774
Kubanek J, Lester SE, Fenical W, Hay ME (2004) Ambiguous
role of phlorotannins as chemical defenses in the brown
alga Fucus vesiculosus. Mar Ecol Prog Ser 277: 79−93
Lubchenco J, Gaines SD (1981) A unified approach to mar-
ine plant−herbivore interactions. I. Populations and com-
munities. Annu Rev Ecol Syst 12: 405−437
Matson PG, Edwards MS (2007) Effects of ocean tempera-
ture on the southern range limits of two understory kelps,
Pterygophora californica and Eisenia arborea, at multi-
ple life-stages. Mar Biol 151: 1941−1949
Matson PG, Steffen BT, Allen RM (2010) Settlement behav-
ior of cyphonautes larvae of the bryozoan Membranipora
membranacea in response to two algal substrata. Inver-
tebr Biol 129: 277−283
Müller R, Laepple T, Bartsch I, Wiencke C (2009) Impact of
oceanic warming on the distribution of seaweeds in polar
and cold-temperate waters. Bot Mar 52: 617−638
118
Author copy

Simonson et al.: Warming seawater effects on kelp quality
Nielsen MM, Krause-Jensen D, Olesen B, Thinggaard R,
Christensen PB, Bruhn A (2014) Growth dynamics of
Saccharina latissima (Laminariales, Phaeophyceae) in
Aarhus Bay, Denmark, and along the species’ distribu-
tion range. Mar Biol 161: 2011−2022
Nylund GM, Pavia H (2005) Chemical versus mechanical
inhibition of fouling in the red alga Dilsea carnosa. Mar
Ecol Prog Ser 299: 111−121
O’Brien JM, Scheibling RE, Krumhansl KA (2015) Positive
feedback between large-scale disturbance and density-
dependent grazing decreases resilience of a kelp bed
ecosystem. Mar Ecol Prog Ser 522: 1−13
O’Connor MI (2009) Warming strengthens an herbivore−
plant interaction. Ecology 90: 388−398
Olischläger M, Iñiguez C, Gordillo FJL, Wiencke C (2014)
Biochemical composition of temperate and Arctic popula-
tions of Saccharina latissima after exposure to increased
pCO2and temperature reveals ecotypic variation. Planta
240: 1213−1224
Pansch C, Gómez I, Rothäusler E, Veliz K, Thiel M (2008)
Species-specific defense strategies of vegetative versus
reproductive blades of the Pacific kelps Lessonia nigre -
scens and Macrocystis integrifolia. Mar Biol 155: 51−62
Pavia H, Toth GB (2000) Influence of light and nitrogen on
the phlorotannin content of the brown seaweeds Asco-
phyllum nodosum and Fucus vesiculosus. Hydrobiologia
440: 299−305
Peterson CH, Renaud PE (1989) Analysis of feeding prefer-
ence experiments. Oecologia 80: 82−86
Saunders M, Metaxas A (2007) Temperature explains settle-
ment patterns of the introduced bryozoan Membranipora
membranacea in Nova Scotia, Canada. Mar Ecol Prog
Ser 344: 95−106
Saunders M, Metaxas A (2008) High recruitment of the
introduced bryozoan Membranipora membranacea is
associated with kelp bed defoliation in Nova Scotia, Can-
ada. Mar Ecol Prog Ser 369: 139−151
Saunders M, Metaxas A (2009) Effects of temperature, size,
and food on the growth of Membranipora membranacea
in laboratory and field studies. Mar Biol 156: 2267−2276
Saunders M, Metaxas A, Filgueira R (2010) Implication of
warming temperatures for population outbreaks of a
nonindigenous species (Membranipora membranacea,
Bryozoa) in rocky subtidal ecosystems. Limnol Oceanogr
55: 1627−1642
Scheibling RE, Gagnon P (2009) Temperature-mediated out-
break dynamics of the invasive bryozoan Membranipora
membranacea in Nova Scotian kelp beds. Mar Ecol Prog
Ser 390: 1−13
Scheibling RE, Feehan CJ, Lauzon-Guay JS (2013) Climate
change, disease and the dynamics of a kep-bed ecosys-
tem in Nova Scotia. In: Fernández-Palacios JM, de
Nascimento L, Hernández JC, Clemente S, González A,
Díaz-González JP (eds) Climate change perspectives
from the Atlantic: past, present and future. Servicio de
Publicaciones, Universidad de La Laguna, p 361−387
Schoenwaelder MEA, Clayton MN (2000) Physode forma-
tion in embryos of Phyllospora comosa and Hormosira
banksii (Phaeophyceae). Phycologia 39: 1−9
Sieburth J, Tootle JL (1981) Seasonality of microbial fouling
on Ascophyllum nodosum (L.) LeJol., Fucus vesiculosus
L., Polysiphonia lanosa (L.) Tandy and Chondrus crispus
Stackh. J Phycol 17: 57−64
Simonson EJ, Scheibling RE, Metaxas A (2015) Kelp in hot
water: I. Warming seawater temperature induces weak-
ening and loss of kelp tissue. Mar Ecol Prog Ser 537:
89–104
Staehr PA, Wernberg T (2009) Physiological responses of
Ecklonia radiata (Laminariales) to a latitudinal gradient
in ocean temperature. J Phycol 45: 91−99
Steinberg PD (1984) Algal chemical defense against herbi-
vores: allocation of phenolic compounds in the kelp
Alaria marginata. Science 223: 405−407
Steinberg PD (1995) Seasonal variation in the relationship
between growth rate and phlorotannin production in the
kelp Ecklonia radiata. Oecologia 102:
169−173
Steneck RS, Watling L (1982) Feeding capabilities and limi-
tation of herbivorous molluscs: a functional group
approach. Mar Biol 68: 299−319
Steneck RS, Graham MH, Bourque BJ, Corbett D, Erlandson
JM, Estes JA, Tegner MJ (2002) Kelp forest ecosystems:
biodiversity, stability, resilience and future. Environ
Conserv 29: 436−459
Stiger V, Deslandes E, Payri CE (2004) Phenolic contents of
two brown algae, Turb inar ia or nata and Sargassum man-
garevenses on Tahiti (French Polynesia): interspecific,
ontogenic and spatio−temporal variations. Bot Mar 47:
402−409
Targett NM, Arnold TM (1998) Predicting the effects of
brown algal phlorotannins on marine herbivores in trop-
ical and temperate oceans. J Phycol 34: 195−205
Toth G, Pavia H (2002) Intraplant habitat and feeding pref-
erence of two gastropod herbivores inhabiting the kelp
Laminaria hyperborea. J Mar Biol Assoc UK 82: 243−247
Tuya F, Cacabelos E, Duarte P, Jaci nto D and o thers (20 12)
Patterns of landscape and assemblage structure along a
latitudinal gradient in ocean climate. Mar Ecol Prog Ser
466: 9−19
Van Alstyne KL (1988) Herbivore grazing increases poly -
phenolic defenses in the intertidal brown algae Fucus
distichus.Ecology69: 655−663
Van Alstyne KL (1995) Comparison of three methods for
quantifying brown algal polyphenolic compounds.
J Chem Ecol 21: 45−58
Van Altena IA, Steinberg PD (1992) Are differences in the
response between North American and Australasian
marine herbivore to phlorotannins due to differences in
phlorotannin structure? Biochem Syst Ecol 20: 493−499
Vergés A, Steinberg PD, Hay ME, Poore AGB and others
(2014) The tropicalization of temperate marine ecosys-
tems: climate-mediated changes in herbivory and com-
munity phase shifts. Proc R Soc B 281: 20140846
Weinberger F, Rohde S, Oschmann Y, Shahnaz L, Dobretsov
S, Wahl M (2011) Effects of limitation stress and of dis-
ruptive stress on induced antigrazing defense in the
bladder wrack Fucus vesiculosus. Mar Ecol Prog Ser 427:
83−94
Wernberg T, Smale DA, Tuya F, Thomsen MS and others
(2013) An extreme climatic event alters marine ecosys-
tem structure in a global biodiversity hotspot. Nat Clim
Change 3: 78−82
Wikström SA, Pavia H (2004) Chemical settlement inhibition
versus post-settlement mortality as an explanation for
differential fouling of two congeneric seaweeds. Oecolo-
gia 138: 223−230
Winer BJ, Brown DR, Michels KM (1991) Statistical princi-
ples in experimental design, 3rd edn. McGraw-Hill, New
York, NY
Zarnetske PL, Skelly DK, Urban MC (2012) Biotic multipliers
of climate change. Science 336: 1516−1518
119
Editorial responsibility: Morten Pedersen,
Roskilde, Denmark
Submitted: March 20, 2015; Accepted: July 10, 2015
Proofs received from author(s): September 8, 2015
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
Author copy