Ecology, 96(1), 2015, pp. 241–251
Ó 2015 by the Ecological Society of America
Kelp canopy facilitates understory algal assemblage via competitive
release during early stages of secondary succession
KYLLA M. BENES1
ROBERT C. CARPENTER
AND
Department of Biology, California State University, 18111 Nordhoff Street, Northridge, California 91330-8303 USA
Abstract. Kelps are conspicuous foundation species in marine ecosystems that alter the
composition of understory algal assemblages. While this may be due to changes in the
competitive interactions between algal species, how kelp canopies mediate propagule supply
and establishment success of understory algae is not well known. In Southern California,
USA, Eisenia arborea forms dense kelp canopies in shallow subtidal environments and is
associated with an understory dominated by red algal species. In canopy-free areas, however,
the algal assemblage is comprised of mostly brown algal species. We used a combination of
mensurative and manipulative experiments to test whether Eisenia facilitates the understory
assemblage by reducing competition between these different types of algae by changes in biotic
interactions and/or recruitment. Our results show Eisenia facilitates a red algal assemblage via
inhibition of brown algal settlement into the canopy zone, allowing recruitment to occur by
vegetative means rather than establishment of new individuals. In the canopy-free zone,
however, high settlement and recruitment rates suggest competitive interactions shape the
community there. These results demonstrate that foundation species alter the distribution and
abundance of associated organisms by affecting not only interspecific interactions but also
propagule supply and recruitment limitation.
Key words: brown algae; disturbance; Eisenia arborea; facilitation; foundation species; indirect
interaction; recruitment; red algae; settlement; shading; understory.
INTRODUCTION
Canopy-forming primary producers act as foundation
species (sensu Dayton 1972) by providing habitat and
modifying key resources such as light and nutrients (see
Dayton [1985], Ellison et al. [2005] for reviews).
Understory community composition is often regarded
as a subset of local species that have a physiological
tolerance/preference for understory environments, or as
a consequence of a shift in competitive dominants (e.g.,
Beatty 1984, Naumburg and DeWald 1999, Clark et al.
2004, Toohey et al. 2004). However, canopies may alter
recruitment processes by limiting propagule dispersal or
propagule success (e.g., Matlack 1994, Beckage et al.
2000, Dalling et al. 2001, Fayolle et al. 2009), limiting
the local range of species and reducing local diversity
and abundance (e.g., Primack and Miao 1992, Ehrlén
and Eriksson 2000, Foster and Tilman 2003). Alternatively, mediation of the abiotic and biotic environment
by canopies may alter the outcome or nature of
interspecific interactions (e.g., Liancourt et al. 2005),
thereby shaping the distribution and abundance of
associated organisms. Disturbance to the canopy and/
Manuscript received 13 January 2014; revised 11 April 2014;
accepted 6 June 2014; final version received 7 July 2014.
Corresponding Editor: M. H. Graham.
1 Present address: Northeastern University Marine Science
Center, 430 Nahant Road, Nahant, Massachusetts 01908
USA. E-mail: [email protected]
or understory can change the distribution of resources
(i.e., space, light, nutrients, etc.) and allow for succession and changes in community structure, depending on
the resistance and resilience of disturbed patches (e.g.,
Dayton et al. 1984). The resulting assemblage is likely a
result of variation in propagule supply and interspecific
interactions, as both can contribute to the distribution
and abundance of organisms (Tilman 1997).
Kelps are conspicuous foundation species in temperate marine habitats. Canopies create dynamic and
patchy light environments (e.g., Gerard 1984), which
can indirectly affect understory macroalgae (Dayton
1985, Irving and Connell 2006a), since competition for
light is important in shaping algal species assemblages
(Carpenter 1990). Additionally, shorter kelps (e.g.,
Laminaria, Egregia, etc.) may directly interact with
understory algae via abrasion by whiplash from kelp
blades (e.g., Velimirov and Griffiths 1979, Hughes
2010). Both natural and manipulative experiments have
demonstrated the influence of kelp canopies on understory algal assemblages (e.g., Reed and Foster 1984,
Kennelly 1987a, Arkema et al. 2009). In these studies,
disturbances to canopies (thinning or complete removal)
result in significant changes in understory community
structure, usually resulting in an increase in macroalgae
cover and/or a shift in macroalgae and sessile invertebrate composition. Loss of kelp and changes to the
understory algal assemblage could lead to changes in
primary productivity and community-level diversity and
241
242
KYLLA M. BENES AND ROBERT C. CARPENTER
function (e.g., Graham 2004, Miller et al. 2011). While
previous studies have focused on recruitment processes
of macroscopic understory algae, canopy mediation of
flow and light conditions (Gerard 1984, Eckman et al.
1989) may alter the abundance and viability of
propagules (e.g., Graham 2003, Roleda et al. 2004).
Changes in the kelp forest understory could therefore be
due to dispersal or settlement patterns of understory
macroalgal species (Kennelly and Larkum 1983).
Eisenia arborea Areschoug (hereafter referred to as
Eisenia) is a stipitate kelp that occurs from Vancouver
Island, British Columbia, Canada to Bahı́a Magdalena,
Baja California, Mexico in the eastern North Pacific
(Abbott and Hollenberg 1992). In Southern California,
Eisenia forms dense monospecific stands in shallow (,3
m mean lower low water [MLLW]), wave-swept areas
(Dayton et al. 1984, Murray and Bray 1993). This is
likely due to its ability to better withstand sedimentation
and/or scour compared to larger species of kelps (i.e.,
Macrocystis pyrifera), which can be competitively
dominant over Eisenia by shading in deeper areas or in
areas with less wave intensity (Dayton 1985). During
summer, peak densities create canopies that cover up to
95% of the substratum (Kastendiek 1982) and reduce
light to as little as 3% of levels above the canopy (;20 6
4 lmol photonss1m2). During winter, however,
storm swells disturb the canopy, reducing Eisenia
densities and increasing light to approximately 30% of
levels above the canopy (;159 6 13 lmol photonss1m2; R. C. Carpenter, unpublished data collected from 2003 to 2005). Due to its shallow nature and
high densities, Eisenia may interact both directly and
indirectly with the understory community by scouring
the substratum and significantly reducing light.
We used observations and an Eisenia canopy manipulation to identify at what stage (settlement or postsettlement) of community succession kelps facilitate
foliose algal understory assemblages. In particular, we
asked the following questions. What is the distribution
and abundance of macroalgae in the canopy and
canopy-free zones? Do settlement patterns of macroalgae differ between the canopy and canopy-free zones?
How does Eisenia affect the recruitment of foliose algae
following a simulated disturbance? Since winter storms
alter the canopy density (and associated abiotic conditions), observations and experiments were conducted for
nearly two years to capture seasonal variation in
understory assemblages. For the purposes of this study,
settlement is defined as the appearance of microscopic
recruits estimated from a short sampling period and
then quantified in the lab, whereas recruitment is defined
as the appearance of macroscopic individuals identifiable in the field.
METHODS
Study site
This study was carried out at the Wrigley Marine
Science Center (WMSC), Santa Catalina Island, Cal-
Ecology, Vol. 96, No. 1
ifornia. Surveys and field experiments were conducted at
the western end of Bird Rock, a small islet just off the
leeward coast of Santa Catalina Island (33827 0 N,
118829 0 W). This region of Southern California is
characterized by strong, storm-driven, northwesterly
swells during the winter and spring followed by a period
of relatively calm southwesterly swells in the summer
and fall (NOAA buoy, station 46069; June 2005–
February 2006; data obtained from the National Data
Buoy Center, Stennis Space Center, Mississippi, USA).
The western end of Bird Rock faces the predominant
swell and receives the most intense wave action at this
site (Robles 1997); here, the Eisenia canopy occupies an
approximately 600-m2 area from the low intertidal to 3
m depth. Also at this site, an area devoid of a kelp
canopy (canopy-free zone) exists from approximately 3
to 5 m depth, and an area dominated by a Macrocystis
pyrifera canopy occurs .5 m depth (for further
description of the subtidal zonation, see Morrow and
Carpenter [2008a]). All portions of this study were
conducted in the Eisenia canopy (at ;2–3-m) and
canopy-free (at ;3–4-m) zones.
Eisenia density and macroalgal community structure
Community surveys were conducted to determine if
the distribution and abundance of macroalgae varied
between canopy habitats. Winter storms alter kelp
density, therefore surveys were conducted during winter
and summer seasons from 2004 to 2005. For both
canopy density and algal percent cover, quadrats were
placed haphazardly along depth contours, due to the
difficulty of using metered transects in the canopy area.
To estimate Eisenia density, all adult individuals within
a 0.25-m2 quadrat were counted. Data were transformed
to log(x þ 1) to achieve normality and homoscedasticity,
then analyzed using a two-way, mixed-model ANOVA
to test for the effects of season (fixed) and year
(random). All statistical analyses were conducted using
SYSTAT v. 9.0 (Systat Software, San Jose, California,
USA), unless otherwise noted.
To estimate percent cover of non-kelp macroalgae, a
0.0625-m2 quadrat was divided into 25 squares, each
square representing 4% cover, and the dominant species
was recorded for each square. Using this method,
changes ,4% cover (i.e., species present in only a few
quadrats generating a large number of 0% cover
observations) were unreliable. However, we did not
use these values to make quantitative predictions, and
were concerned with relative patterns. Analyses were
conducted at the macroalgae group level (i.e., red,
brown, articulated coralline, and crustose coralline
algae), not species level (Appendix A), and the majority
of cover estimates were .4% for each group, sufficient
for assessing the relative influence of the canopy on
macroalgae. Percent cover data were arcsine-transformed to achieve normality and homoscedasticity
before statistical analyses. Algal percentage data were
analyzed using a three-way, mixed-model ANOVA to
January 2015
EISENIA FACILITATES ALGAL ASSEMBLAGE
243
FIG. 1. (A) Experimental design (split plot) used to test the effects of Eisenia arborea canopy removal and disturbance to the
macroalgal community (no removal [NR], upright removal [UR], and complete removal [CR]) on the recruitment of understory
algal species. (B) Canopy removal treatment within the Eisenia bed at Bird Rock, Santa Catalina Island, California, USA.
test for the effects of canopy (fixed), season (fixed), and
year (random) on macroalgal cover for each macroalgal
group separately.
Settlement
To compare settlement densities of macroalgae
between canopy and canopy-free zones, settlement tiles
were places in the two zones for short sampling periods.
To quantify seasonal effects on settlement due to canopy
density or timing of reproduction, tiles were placed in
the field at three randomly chosen sampling periods
during each season over three seasons (fall 2004, spring
2005, and summer 2005; the winter season could not be
sampled due to severe weather). Settlement tiles (38.5
cm2) were made by molding Sea Goin’ epoxy putty
(Permalite Plastics, Costa Mesa, California) into Silastic
J silicon molds (Dow Corning, distributed by KR
Anderson, Santa Ana, California), using coarse sandpaper and table salt to create surface heterogeneity for
spore settlement (Brawley and Johnson 1991). These
tiles could be placed in small areas imposed by the high
density of kelp holdfasts in the canopy zone, and
minimized loss of replicate tiles, as larger tiles were often
dislodged during winter storms (K. M. Benes, personal
observation). Tiles were attached to the substratum
under the Eisenia canopy and in the canopy-free zone
for 6-d periods using permanent one-quarter-inch (;0.6
cm) stainless steel bolts, washers, and nuts. Although the
same number of tiles were deployed for each sampling
period (n ¼ 10 per habitat), some replicate tiles were lost
due to severe weather during some sampling periods (see
Results: Settlement for actual replication).
At the end of each sampling period, tiles were
collected and placed in running seawater tables (12:12
h light : dark cycle) and monitored until spores could be
identified to the lowest taxonomic level possible. Control
tiles were placed in the tables, and monitored with tiles
from the field to estimate spore input from the seawater
system. No algal spores were observed on control tiles.
To account for differential mortality/growth rates of
spores in the lab, all tiles were counted every 14 days (for
a maximum of 84 days), and the peak density observed
for each taxon was used as the settlement value. The
total number of spores per tile, as well as the settlement
of each macroalgal group, were transformed to log(x þ
1) to achieve normality. Data were analyzed using a
partially nested, three-way ANOVA to test the effects of
canopy (fixed), season (fixed), sampling period within
season (random), and their interactions on settlement.
Recruitment
We used a manipulative experiment to examine the
effects of canopy and understory disturbance on the
recruitment of macroalgae. The experiment was conducted in the canopy zone and followed a split-plot
design (Fig. 1). In each randomly placed block (n ¼ 5), a
canopy-present treatment and a canopy-removal treatment were established over an area of approximately 7
m2. Canopy clearings were checked every other month
for Eisenia recruits; any individuals found were removed
to keep these areas canopy free. Within each canopy
treatment, three types of understory algal disturbance
plots were established in 0.0625-m2 plots; no removal
(NR, served as a control), upright removal (UR, algal
244
KYLLA M. BENES AND ROBERT C. CARPENTER
thalli were cut just above their holdfasts), and complete
removal (CR, entire algal individuals were removed and
the substratum was abraded). Each disturbance plot was
replicated twice and randomly established in each
canopy treatment (n ¼ 10 plots per canopy treatment
and disturbance plot combination; Fig. 1). The understory macroalgal disturbance plots, in combination with
the canopy treatments, mimicked varying levels of
disturbance (none [NR with canopy present] to severe
[CR with canopy removed]), such that recruitment of
macroalgae could be estimated and compared with the
canopy present (i.e., canopy reestablishment after winter
disturbances) and the canopy absent (i.e., no canopy
reestablishment after winter disturbances).
Natural disturbances to the canopy and understory
assemblage vary in size from ,1 m2 to ;10 m2
(Kastendiek 1982; K. M. Benes personal observation),
likely due to the severity of disturbance imposed by
winter storms. Our canopy clearing size is representative
of a large and severe disturbance. Individual Eisenia can
be up to 1 m tall and interact with the substratum within
a ;3-m2 area; therefore, the clearing size (7 m2) and
small understory plots were necessary to limit contact
between Eisenia and the understory plots in the canopy
removal treatments.
The experiment was established in May 2004 to
coincide with the end of the natural disturbance period.
Percent cover of algal species was initially recorded 40
days following the start of the experiment and every 30–
50 days afterward up to 505 days following the
disturbance (see Methods: Community structure for
estimation of percent cover of macroalgae). To test for
the effects of block (random), canopy (fixed), and
treatment (fixed) on the recruitment of understory algae,
a three-way repeated-measures ANOVA was used on
arcsine-transformed data. Due to storm swells, tags
distinguishing the various plots in an entire block and
two additional plots in different blocks were dislodged
resulting in a loss of replicates (see Results: Recruitment
for actual replication used in analyses).
RESULTS
Eisenia density and macroalgal community structure
Over the course of this study (2004–2005), Eisenia
densities at the study site were significantly higher in
summer (up to 17 individuals/m2) compared to winter
seasons (up to 11 individuals/m2 [two-way, Model III
ANOVA, F1,1 ¼ 191.500, P ¼ 0.046; Fig. 2A]). This
pattern did not vary across years (ANOVA, F1, 186 ¼
1.684, P ¼ 0.196) or between seasons and years
(ANOVA, F1, 186 ¼ 0.015, P ¼ 0.903).
Foliose algal community structure differed with
respect to canopy, season, and year (Fig. 2B, C); see
Appendix B for complete ANOVA results for each
macroalgal group. Foliose red algae was highest in
percent cover in the canopy (12–16%) compared to the
canopy-free (2–7%) zone for all of 2004 and summer
2005, but had a similar percent cover between the two
Ecology, Vol. 96, No. 1
habitats in winter 2005 (5–7% [three-way Model III
ANOVA, canopy 3 year; P , 0.001; Fig. 2B]). In
contrast, brown algal cover was up to six times higher in
the canopy-free compared to the canopy zone across all
seasons and years sampled (canopy; P ¼ 0.019; Fig. 2C).
Articulated coralline cover was overall higher in winter
in 2005 (38–40%) than in summer (21–25% [season 3
year; P ¼ 0.03]). However, in 2004, articulated coralline
cover was higher in the canopy zone each season
(canopy 3 year; P ¼ 0.028; Fig. 2D). Crustose corallines
were nearly 2.53 greater in abundance in the canopy-free
compared to the canopy zone in winter of 2004, but
showed similar abundances in both habitats in winter
2005. In the summer seasons, however, percent cover
was higher under the Eisenia canopy (39–40%) compared to the canopy-free zone (25–26%) in both 2004
and 2005 (canopy 3 season 3 year; P ¼ 0.004; Fig. 2E).
These results indicate that although both habitats
have high abundances of articulated and crustose
coralline algae, they differ in foliose algal species. The
understory assemblage is dominated by foliose red algae
with extremely low abundances of brown algae. In
contrast, the assemblage in the canopy-free zone is
dominated by brown macroalgae (Table 1, Appendix A;
see Plate 1).
Settlement
Total settlement was up to sixfold higher in the
canopy-free zone (nested ANOVA, F1,6 ¼ 13.349, P ¼
0.011; Fig. 3A). Significant differences in total settlement
were found among sampling periods within each season
(F6, 144 ¼ 18.265, P , 0.001), and in the interaction
between habitat and the weeks within season (F6, 144 ¼
6.147, P , 0.001), suggesting high temporal variability
in propagule supply.
Settlement patterns of separate macroalgal groups
showed striking differences (Fig. 3B–E, Appendix C). In
most analyses, the effect of sampling period within
season was significant; these data are not presented,
however, since they are highly variable among seasons
and algal groups and do not show any striking patterns
with respect to the canopy. Foliose red algal settlement
was higher overall during spring and summer compared
to fall, but did not differ between zones (Fig. 3B). Brown
algal settlement was up to 203 higher in the canopy-free
zones (Fig. 3C), and the difference in settlement of
brown algae between the two habitats was lowest in
spring. Spore density of articulated coralline algae
compared to other macroalgal groups was relatively
low across habitats and was highest in spring (Fig. 3D).
Crustose coralline settlement was relatively similar
between canopy habitats in the fall and spring but
higher in the canopy-free zone during the summer (Fig.
3E).
Even though spore densities of red algae were similar
among habitats, settlement by conspicuous species
(species that make up the assemblage studied) was
nearly zero for both habitats and all sampling periods.
January 2015
EISENIA FACILITATES ALGAL ASSEMBLAGE
245
FIG. 2. Patterns of (A) Eisenia density (mean þ SE) and macroalgal composition (mean þ SE) in the canopy and canopy-free
zones for (B) red algae, (C) brown algae, (D) articulated corallines, and (E) crustose corallines. Data are for winter and summer
seasons from 2004 to 2005. Replication for each sampling period is shown above the bar in parentheses. In panel (B), replication is
shown for canopy/canopy-free zones.
Almost all (.99%) of the red algal settlers were small,
filamentous species such as Polysiphonia and Ceramium
spp. This is in contrast to settlement of brown algal
species, in which most (.80%) of the spores were
macroalgal species that were members of the study
assemblage (brown, turf-building species included Ectocarpus sp.).
Recruitment
Recruitment of the major macroalgal groups varied
over time both between canopy and canopy-removal
treatments and among disturbance treatments (Fig. 4,
Appendix D).
Foliose red algae recruitment varied significantly over
time following the disturbance and in the interactions
between canopy and plot types and time (Appendix D).
With the canopy present, foliose red algal cover was
highest in the NR and UR plots and nearly doubled in
cover over time, while cover in the CR plots remained
low (0–1%). In contrast, with the canopy removed, the
percent cover of red algae was initially higher in the NR
plots, but by the end of the experiment was similar in all
plot types (Fig. 4A, B).
For the percent cover of brown algae, a significant
three-way interaction between time, canopy treatment,
and plot type was found (Appendix D). We observed
246
KYLLA M. BENES AND ROBERT C. CARPENTER
Ecology, Vol. 96, No. 1
TABLE 1. Foliose macroalgae observed at the study site in the canopy and canopy-free zones.
Relative abundance of foliose algae (%)
Under Eisenia canopy
Canopy-free zone
1.00
1.00
1.00
1.00
0.98
0.75
0.20
0.79
1.00
0
4
5
19
20
,1
12
33
1
1
0
0
0
,1
1
8
2
0
1.00
0.98
0.82
0.57
1.00
0.94
0
1
2
,1
0
3
2
27
7
,1
12
40
Species and abbreviation
CI
Articulated corallines
Crustose corallines
Foliose red algae
Asparagopsis taxiformis (AT)
Carpopeltis bushiae (CB)
Chondracanthus spinosus (CS)
Gelidium purpurascens (GP)
G. robustum (GR)
Laurencia pacifica (LP)
Plocamium cartilagineum (PC)
Pterocladiella capillacea (PtC)
Rhodymenia californica (RC)
0.09
0.01
Foliose brown algae
Colpomenia sinuosa (CoS)
Dictyopteris undulate (DU)
Dictyota binghamiae (DB)
D. coriacea (DC)
Sargassum palmeri (SP)
Zonaria farlowii (ZF)
Notes: The relative abundance of foliose macroalgal species within each habitat is presented;
species are listed in alphabetical order (foliose red algae; Rhodophyta, foliose brown algae;
Ochrophyta). The index of canopy association (CI) of each species and articulated and crustose
coralline groups are also given. CI ranges from 1.00 to 1.00, where a value of 1.00 indicates a
strong association to the canopy zone (see Appendix A). Percent cover is not shown for articulated
and crustose corallines, which were not included in a Bray-Curtis multidimensional scaling
(Appendix A). Please see Abbott and Hollenberg (1992) for species authorities.
Species with stoloniferous (vegetative) growth capability.
virtually no recruitment of brown algae with the canopy
present (0–2%), but variable recruitment with the
canopy removed (Fig. 4C, D). With the canopy
removed, abundance in the UR and CR plots was
initially high (35–43%), followed by a decrease over 40–
160 days to as little as 4%. This was due to rapid
recruitment and disappearance of the opportunistic and
ephemeral Colpomenia sinuosa (Appendix E). At the end
of the study, percent cover of brown algae increased in
all plot types with the highest cover in the CR plots
compared to the NR and UR plots (which were similar
in percent cover; Fig. 4C, D).
For both articulated and crustose coralline algae,
recruitment into treatment plots varied temporally over
the first 160 days, but converged by the end of the
experiment (Fig. 4E–H). Percent cover of articulated
coralline algae was two times higher with the canopy
removed compared to under the canopy (Fig. 4E, F). In
contrast, percent cover of crustose coralline algae was
similar in both canopy treatments (Fig. 4G, H).
DISCUSSION
Eisenia arborea dominates wave-exposed shallow
subtidal sites around Santa Catalina Island and
throughout the Southern California Bight (Murray
and Bray 1993, Benes 2006). Canopy and canopy-free
zones have a high abundance of articulated and crustose
coralline algae, but they differ in amounts of foliose red
and brown macroalgae, a pattern common over time
and seasons at our study site. Our results show the
potential for direct and indirect effects of Eisenia on
understory algae, thereby facilitating a foliose red algal
assemblage via competitive release and modification of
secondary succession.
Eisenia influenced the distribution and abundance of
algae by direct and indirect interactions (i.e., abrasion
by kelp blades and reduced light levels, respectively)
with the understory assemblage. For foliose macroalgae,
we often observed higher percent cover during the
summer in the community surveys (Fig. 2) and the
recruitment experiment (Fig. 4, winter observations at
;160 days). Additionally, the kelp canopy affected the
rate of recruitment. In areas with the canopy removed,
we observed rapid recruitment and/or growth within 40–
130 days of the start of the experiment. However, plots
under the canopy did not diverge until late in the
experiment (foliose red algae) or remained relatively
similar over the course of the experiment. These
observations may reflect greater growth and primary
productivity of macroscopic foliose algae in spring/
summer, and increased abrasion by Eisenia blades and/
or limiting growth conditions in the winter.
Disturbance by Eisenia blades could account for the
overall low algal spore settlement under the canopy. Yet
the difference in settlement density between canopy and
canopy-free zones is notably less in spring, when canopy
density is low (i.e., higher light levels), but disturbance
from storm swells is high. This indicates a negative
indirect effect of the canopy on algal spores or early
recruits. Patterns of brown algal settlement agree with
January 2015
EISENIA FACILITATES ALGAL ASSEMBLAGE
247
FIG. 3. (A) Total number of spores per tile (mean þ SE) settling in canopy and canopy-free zones. Data also are presented for
each algal group; (B) foliose red algae, (C) brown algae, (D) articulated corallines, and (E) crustose corallines, during the fall 2004,
spring 2005, and summer 2005 seasons. The number of replicates tiles for each sampling period are in parentheses above the bar in
panel (A).
this hypothesis, but the lack of red algal settlement by
conspicuous species makes it difficult to know how the
canopy affects this macroalgal group. Near zero
settlement of conspicuous red algal species suggests
propagule supply or survival is very low and/or
filamentous red algae is occupying space and competing
with large red algal species. Our data does not allow us
to distinguish among these hypotheses, but additional
experiments could help determine whether primary
successional species (i.e., filamentous algae) can further
modify the patterns and processes seen in our current
study.
These results suggest that Eisenia, via modification of
the light environment, has an important indirect effect on
the survival of new algal settlers and/or very young
recruits. Although we cannot distinguish between propagule dispersal vs. survival, our results demonstrate that
Eisenia alters the establishment of foliose algae very early
in secondary succession. Reduced settlement and slow
recovery under the canopy indicate community structure
in this habitat is driven by propagule and recruitment
limitation. Indeed, reduced light levels can affect the
success of sporelings (e.g., Swanson and Druehl 2000,
Roleda et al. 2004), and algal spores can resist
dislodgement from extremely high forces (Charters et
al. 1971). Given reduced settlement into the canopy
environment, vegetative growth and colonization (i.e.,
stoloniferous growth) in some taxa (Table 1) are likely
important modes of colonization for the understory
assemblage (e.g., Santelices and Varela 1994, Downing
1995; K. M. Benes, personal observation). Direct abrasion
and removal of understory macroalgae by the stipitate
canopy (Irving and Connell 2006b) and/or indirect
suppression of growth due to reduced light levels (e.g.,
Kain 1987) could further limit the rate of recruitment
into the understory. Reduction in abundance of red algae
by more than half in winter 2005 may have been due to
increased abrasion from the canopy because of recordbreaking storms at that time. Slow recovery rates, as seen
in our recruitment experiment, could explain the smaller
difference in red algal cover between canopy habitats in
summer 2005 compared to 2004.
248
KYLLA M. BENES AND ROBERT C. CARPENTER
Ecology, Vol. 96, No. 1
FIG. 4. Percent cover (mean 6 SE) of (A, B) foliose red algae, (C, D) brown algae, (E, F) articulated corallines, and (G, H)
crustose corallines over the experiment. Data are for the no removal (NR), upright removal (UR), and complete removal (CR)
treatments of macroalgae with the presence (left column) and removal (right column) of the Eisenia canopy. For reference, stars
indicate the percent cover of macroalgal groups from community surveys in the canopy zone conducted ;1.5 months prior to the
start of the experiment. Note the varying scales of the y-axes.
In contrast, the community outside the canopy is likely
a result of interspecific competition, since abundance of
propagules was high and recruitment into disturbed
patches was relatively fast. Changes in percent cover of
algae in canopy-removal areas were due mostly to the
establishment of new individuals. This was evidenced by
the change in species composition over the course of the
experiment with the canopy removed, but the relative no
change with the canopy present (Appendix E). Although
Eisenia could negatively affect red foliose algae physiologically, the canopy has a positive indirect effect via its
suppression of dominant brown foliose algae, thereby
maintaining a red algal assemblage through indirect
facilitation among potential competitors (Levine 1999).
The effects of Eisenia on understory macroalgae may
depend on both canopy density and disturbance size.
Surveys of kelp beds along 20 km of Santa Catalina
Island coastline (Morrow 2008a) revealed a minimum
density of ;10 Eisenia individuals/m2 is required for a
red-algal-dominated understory assemblage (Benes
2006). Kastendiek (1982) found that Pterocladiella
capillacea (as Pterocladia), a dominant understory
species (Table 1), could resist invasion by other macroalgal species even when Eisenia density was reduced to
;20 individuals/m2. Additionally, in our study, brown
algal species typically found in the canopy-free zone
recruited into the no-disturbance plots (NR), and this
recruitment even outpaced potential regrowth in the
January 2015
EISENIA FACILITATES ALGAL ASSEMBLAGE
249
PLATE 1. (A) Eisenia arborea canopy, (B) example of understory algal assemblage with the foliose red alga, Pterocladiella
capillacea, and (C) example of canopy-free algal assemblage with several fleshy brown algal species, including Sargassum palmeri
and Zonaria farlowii. Photo credits: (A) Jonathon P. Williams; (B and C) Kathleen M. Morrow.
upright-removal plots (UR) when the canopy was
removed (Appendix E). This suggests in canopy gaps,
even undisturbed patches of the understory community
cannot resist invasion. The persistence and recovery of a
foliose red algal assemblage is therefore dependent on
the presence and stability of Eisenia canopy patches. The
importance of long-term stability of a canopy patch for
the presence of the associated understory assemblage
has been demonstrated in both terrestrial and marine
systems (e.g., Dayton et al. 1984, Beckage et al. 2000).
Since disturbance by swells decreases with depth,
zonation of the different habitats is naturally confounded with depth. We attempted to limit this confounding
factor by restricting the depths over which our study
took place. However, the differences in foliose algae
composition and abundance could be driven by depth
rather than canopy, and the low brown algal cover in the
canopy zone may be due to limits imposed by species
dispersal distances. But foliose brown algae can be
found in high abundance at shallow depths (i.e., depth
of the Eisenia bed) where Eisenia is sparse or absent
(Benes 2006), and can easily colonize large disturbed
patches within the canopy area (as was found in this
study). Recruitment into canopy-cleared areas, where
foliose brown algae was not initially present, indicates
that brown algae recruitment into the canopy area is not
limited by dispersal distance. Settlement was measured
in the canopy and canopy-free zones (not the canopy
clearings in the recruitment experiment), therefore the
actual density of spores arriving to the clearings may be
250
KYLLA M. BENES AND ROBERT C. CARPENTER
lower due to dilution as spores travel away from the
source population.
Large disturbances that remove the Eisenia canopy
shift the composition of the understory macroalgal
assemblage, leading to potential cascading effects on the
associated community. Removal of the Eisenia canopy
led to an increase in percent cover (a proxy for biomass)
of macroalgae, indicating higher net primary production
of the benthic assemblage (Miller at al. 2011). However,
the change in composition of the foliose assemblage
might have negative effects on the entire community.
The reduced abundance and distinct morphology of
understory red algae increase sessile invertebrate abundance (Morrow and Carpenter 2008a) and prey capture
efficiency (Morrow and Carpenter 2008b), thereby
enhancing overall community richness and function.
Additionally, brown foliose algae can support high
densities of micro-grazers such as amphipods (e.g.,
Christie et al. 2009), which collectively reduce brown
and ephemeral macroalgae (Duffy and Hay 2000) and
kelp abundance (Sala and Graham 2002). The dense
foliose algae in canopy gaps can prevent recruitment of
kelps, inhibiting the recovery of the canopy following a
disturbance via competitive exclusion (Dayton et al.
1984, Kennelly 1987b) and possibly the supported
herbivore assemblage. The potential for increased storm
frequency and severity with climate change (Easterling
et al. 2000) therefore may generate long-term changes in
shallow kelp forest communities.
Eisenia arborea facilitates an understory assemblage
via modification of secondary succession, releasing red
foliose algae from competition with brown foliose algae.
By mediating abiotic and biotic factors, Eisenia limits
the arrival and/or settlement of propagules into the
understory environment causing slow recruitment and
low abundance of foliose red algae. The results
presented suggest that the development and maintenance of the understory community is dependent upon
the Eisenia canopy even though physiological trade-offs
due to low light levels may occur during seasons of peak
canopy density. We demonstrate that kelps, as foundation species, do not affect interspecific interactions (i.e.,
competition) alone, but also alter community structure
and development via mediation of propagule supply
and/or success. The Eisenia canopy is a small portion of
the larger kelp forest ecosystem (canopies were never
observed greater than 900 m2), therefore also demonstrating the importance of foundation species at smaller
spatial scales (meters) in determining the distribution
and abundance of organisms within a larger ecosystem.
ACKNOWLEDGMENTS
Fieldwork was made possible (and much more pleasant) by
the assistance of R. Elahi, A. Hettinger, S. Kruger, K. Morrow,
G. Smith, and S. Talmage. A special thanks to K. A. Miller for
help with algal spore identification. Thanks to the staff at the
University of Southern California, Wrigley Institute for
Environmental Studies, who made working at Santa Catalina
Island possible. Advice and comments from L. Allen, S.
Dudgeon, and P. Wilson enhanced the project. We also thank
Ecology, Vol. 96, No. 1
C. Newton and two anonymous reviewers for comments that
greatly improved the manuscript. This research was supported
by grants from the University Corporation and Graduate
Research and International Studies Program offices at California State University, Northridge (to K. M. Benes) and NIHSCORE 506GM048680-11 (to R. C. Carpenter). This is
contribution no. 217 of the CSUN marine biology program.
LITERATURE CITED
Abbott, I. A., and G. J. Hollenberg. 1992. Marine algae of
California. Stanford University Press, Stanford, California,
USA.
Arkema, K. K., D. C. Reed, and S. S. Schroeter. 2009. Direct
and indirect effects of giant kelp determine benthic community structure and dynamics. Ecology 90:3126–3137.
Beatty, S. W. 1984. Influence of microtopography and canopy
species on spatial patterns of forest understory plants.
Ecology 65:1406–1419.
Beckage, B., J. S. Clark, B. D. Clinton, and B. L. Haines. 2000.
A long-term study of tree seedling recruitment in southern
Appalachian forests: the effects of canopy gaps and shrub
understories. Canadian Journal of Forest Research 30:1617–
1631.
Benes, K. M. 2006. Effects of Eisenia arborea Areschoug on
understory algal species: implications for populations and
communities. Thesis. California State University, Northridge, Northridge, California, USA.
Brawley, S. H., and L. E. Johnson. 1991. Survival of fucoid
embryos in the intertidal zone depends upon developmental
stage and microhabitat. Journal of Phycology 27:179–186.
Carpenter, R. C. 1990. Competition among marine macroalgae:
a physiological perspective. Journal of Phycology 26:6–12.
Charters, A. C., M. Neushul, and D. Coon. 1971. The effect of
water motion on algal spore adhesion. Limnology and
Oceanography 18:884–896.
Christie, H., K. M. Norderhaug, and S. Fredriksen. 2009.
Macrophytes as habitat for fauna. Marine Ecology Progress
Series 396:221–233.
Clark, R. P., M. S. Edwards, and M. S. Foster. 2004. Effects of
shade from multiple kelp canopies on an understory algal
assemblage. Marine Ecology Progress Series 267:107–119.
Dalling, J. W., S. P. Hubbell, and K. Silvera. 2001. Seed
dispersal, seedling establishment and gap partitioning among
tropical pioneer trees. Journal of Ecology 86:674–689.
Dayton, P. K. 1972. Toward an understanding of community
resilience and the potential effects of enrichments to the
benthos at McMurdo Sound, Antarctica. Pages 81–95 in
B. C. Parker, editor. Proceedings of the colloquium on
conservation problems in Antarctica. Allen Press, Lawrence,
Kansas, USA.
Dayton, P. K. 1985. Ecology of kelp communities. Annual
Review of Ecology and Systematics 16:215–245.
Dayton, P. K., V. Currie, T. Gerrodette, D. D. Keller, R.
Rosenthal, and D. Ven Tresca. 1984. Patch dynamics and
stability of some California kelp communities. Ecological
Monographs 54:253–289.
Downing, J. W. 1995. The effects of vegetative reproduction on
the recruitment and small-scale distribution of the red alga
Plocamium cartilagineum. Thesis. San Jose State University,
San Jose, California, USA.
Duffy, J. E., and M. E. Hay. 2000. Strong impacts of grazing
amphipods on the organization of a benthic community.
Ecological Monographs 70:237–263.
Easterling, D. R., G. A. Meehl, C. Parmesan, S. A. Changnon,
T. R. Karl, and L. O. Mearns. 2000. Climate extremes:
observations, modeling, and impacts. Science 289:2068–2074.
Eckman, J. E., D. O. Duggins, and A. T. Sewell. 1989. Ecology
of understory kelp environments. I. Effects of kelps on flow
and particle transport near the bottom. Journal of Experimental Marine Biology and Ecology 129:173–187.
January 2015
EISENIA FACILITATES ALGAL ASSEMBLAGE
Ehrlén, J., and O. Eriksson. 2000. Dispersal limitation and
patch occupancy in forest herbs. Ecology 81:1667–1674.
Ellison, A. M., et al. 2005. Loss of foundation species:
consequences for the structure and dynamics of forested
ecosystems. Frontiers in Ecology and the Environment 3:
479–486.
Fayolle, A., C. Violle, and M. Navas. 2009. Differential impacts
of plant interactions on herbaceous species recruitment:
disentangling factors controlling emergence, survival, and
growth of seedlings. Oecologia 159:817–825.
Foster, B. L., and D. Tilman. 2003. Seed limitation and the
regulation of community structure in oak savanna grassland.
Journal of Ecology 91:999–1007.
Gerard, V. A. 1984. The light environment in a giant kelp
forest: influence of Macrocystis pyrifera on spatial and
temporal variability. Marine Biology 84:189–195.
Graham, M. H. 2003. Coupling propagule output to supply at
the edge and interior of a giant kelp forest. Ecology 84:1250–
1264.
Graham, M. H. 2004. Effects of local deforestation on the
diversity and structure of Southern California giant kelp
forest food webs. Ecosystems 7:341–357.
Hughes, B. B. 2010. Variable effects of a kelp foundation
species on rocky intertidal diversity and species interactions
in central California. Journal of Experimental Marine
Biology and Ecology 393:90–99.
Irving, A. D., and S. D. Connell. 2006a. Predicting understory
structure from the presence and composition of canopies: an
assembly rule for marine algae. Oecologia 148:491–502.
Irving, A. D., and S. D. Connell. 2006b. Physical disturbance by
kelp abrades erect algae from the understorey. Marine
Ecological Progress Series 324:127–137.
Kain, J. M. 1987. Seasonal growth and photoinhibition in
Plocamium cartilagineum. Phycologia 26:88–99.
Kastendiek, J. 1982. Competitor-mediated coexistence: interactions among three species of benthic macroalgae. Journal of
Experimental Marine Biology and Ecology 62:201–210.
Kennelly, S. J. 1987a. Physical disturbances in an Australian
kelp community. II. Effects on understory species due to
differences in kelp cover. Marine Ecological Progress Series
40:155–165.
Kennelly, S. J. 1987b. Inhibition of kelp recruitment by turfing
algae and consequences for an Australian kelp community.
Journal of Experimental Marine Biology and Ecology 1112:
49–60.
Kennelly, S. J., and A. W. D. Larkum. 1983. A preliminary
study of temporal variation in the colonization of subtidal
algae in an Ecklonia radiata community. Aquatic Botany 17:
275–282.
Levine, J. M. 1999. Indirect facilitation: evidence and predictions from a riparian community. Ecology 80:1762–1769.
Liancourt, P., R. M. Callaway, and R. Michalet. 2005. Stress
tolerance and competitive-response ability determine the
outcome of biotic interactions. Ecology 86:1611–1618.
Matlack, G. R. 1994. Plant species migration in a mixed-history
forest landscape in eastern North America. Ecology 75:1491–
1502.
251
Miller, R. J., D. C. Reed, and M. A. Brzezinski. 2011.
Partitioning of primary production among giant kelp
(Macrocystis pyrifera), understory macroalgae, and phytoplankton on temperate reef. Limnology and Oceanography
56:119–132.
Morrow, K., and R. Carpenter. 2008a. Shallow kelp canopies
mediate macroalgal composition: effects on the distribution
and abundance of Corynactis californica (Corallimorpharia).
Marine Ecological Progress Series 361:119–127.
Morrow, K. M., and R. C. Carpenter. 2008b. Macroalgal
morphology mediates particle capture by the corallimorpharian Corynactis californica. Marine Biology 155:273–280.
Murray, S. N., and R. N. Bray. 1993. Benthic macrophytes.
Pages 304–368 in M. D. Dailey, D. J. Reish, and J. W.
Anderson, editors. Ecology of the Southern California Bight:
a synthesis and interpretation. University of California Press,
Berkley, California, USA.
Naumburg, E., and L. E. DeWald. 1999. Relationships between
Pinus ponderosa forest structure, light characteristics, and
understory graminoid species presence and abundance.
Forest Ecology and Management 124:205–215.
Primack, R. B., and S. L. Miao. 1992. Dispersal can limit local
plant distribution. Conservation Biology 6:513–519.
Reed, D. C., and M. S. Foster. 1984. The effects of canopy
shading on recruitment and growth in a giant kelp forest.
Ecology 65:937– 948.
Robles, C. D. 1997. Changing recruitment in constant species
assemblages: implications for predation theory in intertidal
communities. Ecology 78:1400–1414.
Roleda, M. Y., W. H. van de Poll, D. Hanelt, and C. Wiencke.
2004. PAR and UVBR effects on photosynthesis, viability,
growth and DNA in differential life stages of two coexisting
Gigartinales: implications for recruitment and zonation
pattern. Marine Ecology Progress Series 281:37–50.
Sala, E., and M. H. Graham. 2002. Community-wide distribution of predator–prey interaction strength in kelp forests.
Proceedings of the National Academy of Sciences USA 99:
3678–3683.
Santelices, B., and D. Varela. 1994. Abiotic control of
reattachment in Gelidium chilense (Montagne) Santelices &
Montalva (Gelidium; Rhodophyta). Journal of Experimental
Marine Biology and Ecology 177:145–155.
Swanson, A., and L. Druehl. 2000. Differential meiospore size
and tolerance of ultraviolet light stress within and among
kelp species along a depth gradient. Marine Biology 136:657–
664.
Tilman, D. 1997. Community invasibility, recruitment limitation, and grassland diversity. Ecology 78:81–92.
Toohey, B., G. A. Kendrick, T. Wernberg, J. C. Phillips, S.
Malkin, and J. Price. 2004. The effects of light and thallus
scour from Ecklonia radiata canopy on an associated foliose
algal assemblage: the importance of photoacclimation.
Marine Biology 144:1019–1027.
Velimirov, B., and C. L. Griffiths. 1979. Wave-induced kelp
movement and its importance for community structure.
Botanica Marina 22:169–172.
SUPPLEMENTAL MATERIAL
Ecological Archives
Appendices A–E are available online: http://dx.doi.org/10.1890/14-0076.1.sm