Download Direct and indirect effects of nutrients on

J. Exp. Mar. Biol. Ecol., 151 (1991) 139-153
@ 1991 Elsevier Science Publishers B.V. All rights reserved 0022-0981/91/$03.50
139
JEMBE 01637
Direct and indirect effects of nutrients on intertidal
community structure: variable consequences of seabird
guano
J. Timothy Wootton
Department of Zoology, and Friday Harbor Laboratories, University of Washington, Seattle, Washington,
USA
(Received 18 December 1990; revision received 22 March 1991; accepted 16 April 1991)
Abstract: The structure of communities on cliff sections with and without overhead seabird colonies was
compared to assess whether local nutrient enrichment by birds affected the abundances of rocky-shore
species in the San Juan archipelago of Washington state. In the presence of guano, the vertical distribution
of orange lichens (Caloplaca marina Wedd. and Xanthoria elegans (Link.) Th. Fr.) was elevated and gray
lichens (Physcia (Schreb.) Michx.) were eliminated in the supralittora! fringe. In the splash and upper
intertidal zones, guano appeared to directly enhance the abundance of the green alga Prasiola meridionalis
S. & G. and fleshy crustose algae (Mastocarpus papillatus (C. Ag.) Kutzing, Ralfsia pacifica Hollenb., and
Hiidenbrandia Nardo spp.), but to directly reduce the amount of area covered by Fucus distichus L., and Ulva
L. spp. By directly affecting the abundance of these species, guano indirectly affected other species:
Verrucaria mucosa Wahl. declined because of enhanced competition with Prasiola in the splash zone, and
Balanus glandula Darwin declined because of increased desiccation in areas with reduced Fucus canopy.
Overall, guano positively influenced only four of the eighteen taxa examined in the study, suggesting that,
with the exception of species that live in unusual situations with low nutrient transport, macronutrients
probably do not limit the abundance of species in this community.
Key words: Alga; Barnacle; Indirect effect; Intertidal community; Nutrient; Seabird guano
INTRODUCTION
Field studies assessing the effects of nutrients on community structure and of specific
species on ecosystem properties provide valuable links between the subdisciplines of
community and ecosystem ecology. Aside from investigating human impacts (e.g.,
Edmondson, 1970; Borowitzka, 1972; Littler & Murray, 1975; Brown et al., 1990), few
studies have addressed how one group of species may affect other species by altering
ecosystem properties such as the nutrient regime (Vitousek, 1986). Additionally, in
contrast to terrestrial and aquatic systems (e.g., Schindler, 1977; Tilman et al., 1982;
Chapin & Shaver, 1985; Tilman, 1987; Goldberg & Miller, 1990; Pringle, 1990), the
impact of nutrient regime on community composition in rocky intertidal marine systems
Correspondence address: J.T. Wootton, Department of Integrative Biology, Mulford Hall, University of
California, Berkeley, CA 94720, USA.
140
J.T. WOOTTON
has not been considered, perhaps because relatively high nutrient levels and extensive
water exchange via currents and wave action make local nutrient depletion less likely
(Mann, 1973; Leigh et al., 1987; but see Hanisak, 1983).
Studying the effects of seabirds on coastal marine communities provides an opportunity to probe the effects of nutrients on intertidal community composition and
examine the effect of one group of species on other species via the alteration of
ecosystem properties. Seabirds forage over a wide area of ocean, then return to roosting
sites or nesting colonies where they deliver food to their young. By collecting, concentrating, and then partially releasing nutrients to a local area through the excretion of
guano, birds may significantly alter the local nutrient flow of coastal ecosystems
(Hutchinson, 1950; Golovkin, 1967; Leentvaar, 1967; Golovkin & Garkavaya, 1975;
McColl & Burger, 1976; Hayes & Caslick, 1984; Bosman & Hockey, 1986; but see
Bedard et al., 1980, for a different perspective at a large scale).
Guano is rich in nitrogen, phosphorus, potassium and some salts (Hutchinson, 1950;
Gillham, 1956; Blakemore & Gibbs, 1968; Ganning & Wulff, 1969; Lindeboom, 1984;
Bosman & Hockey, 1986). Therefore, where plants are nutrient limited, the deposition
of guano might increase plant and animal biomass, and, because plants respond differentially to nutrients (Tilman et al., 1982; Chapin & Shaver, 1985; Tilman, 1987), change
species composition. In South Africa, intertidal communities on offshore islands that
support dense seabird colonies exhibit higher algal productivity and altered species
composition compared to mainland sites that have no colonies (Hockey & Branch,
1984; Bosman & Hockey, 1986, 1988). In an unreplicated experiment, Bosman et al.
(1986) found that algal productivity increased at an experimental site with guano
dripped on the rock relative to control sites with and without water drip. Likewise
adding guano to laboratory cultures of marine algae enhances productivity (Golovkin,
1967), and on the Russian coast, phytoplankton abundance is sometimes higher in areas
near bird colonies than is typical of an area without bird colonies (Golovkin, 1967;
Golovkin & Garkavaya, 1975). Onuf et al. (1977) found higher mangrove growth and
herbivorous insect production on an island containing a seabird colony than on one
without a colony, and Powell et al. (1989) reported enhanced productivity of seagrass
where guano deposition was locally elevated with experimental bird perches.
Although guano adds nutrients to the system, guano may not necessarily enhance
plant biomass or productivity. Benthic plant biomass may be limited by factors other
than nutrients, such as herbivores, attachment space, light availability, or physical
conditions (see reviews in Lewis, 1964; ConneU, 1972; Paine, 1977). Also, nutrients in
very high concentrations can reduce plant productivity (Bradshaw et al., 1964;
Golovkin, 1967). Indeed, high guano concentration inhibits many terrestrial (Gillham,
1956, 1960; Weseloh & Brown, 1971; Lindeboom, 1984) and some marine plants
(Golovkin, 1967). Therefore assessing the effects of guano requires examination on a
species-specific basis.
Here I examine the response of an intertidal community to guano input from seabird
colonies in order to examine the consequences of an ecosystem alteration by one group
GUANO EFFECTS ON INTERTIDAL COMMUNITIES
141
of species on the abundance of associated members of the community. This study
provides an assessment of the degree to which intertidal species are nutrient-limited, and
permits an evaluation of the importance of birds affecting intertidal species from the
bottom (via nutrient regime) and from the top of the food web (via predation, see
Wootton, 1990).
STUDY SITE AND METHODS
Intertidal communities in the vicinity of Lopez Island, in the San Juan Islands of
Washington state, USA, were studied. I examined vertical cliff faces of Willow Island
(48 ° 32' N, 122 ° 49' W), Flower Island (48 ° 32' N, 122 ° 51' W), and two sites on Lopez
Island: Humphrey Head ( 4 8 ° 3 3 ' N , 122°52'W), and Davis Bay ( 4 8 ° 2 7 ' N ,
122 ° 55' W, Fig. 1). These cliffs provided a particularly powerful "natural experiment"
because each cliff face contained small colonies (3-20 pairs) of pigeon guillemots
Cepphus columba Pallas, pelagic cormorants Phalacrocorax pelagicus Pallas, and
glaucous-winged gulls Larus glaucescens Naumann interspersed with sections lacking
colonies overhead (described in Speich & Wahl, 1989). Because the same cliff face with
and without standing guano was compared, I minimized the probability of confounding
factors such - differences in current, wave wash, and sunlight affecting the conclusions.
The location of guano on a particular part of the cliff face was determined by small-scale
topography that permitted suitable nest or perch sites, thus being relatively independent
of features characterizing the intertidal zone below. Because these birds cannot feed on
vertical walls, examining cliff sites additionally eliminated possible confounding effects
ofbird predation (see Wootton, 1990). Limited data (see below) were also taken at an
additional cliff site (Upright Head, 48 ° 33' N, 122 ° 53' W) and at several offshore island
sites, three with birds (northeast and southwest sides of Whale Rocks [48 26' N,
122 ° 56' W]; Hall Island [48 ° 26' N, 122 ° 54' W]) and three without birds (northwest
and southeast sides ofa subisland of Long Island [48 ° 26' N, 122 ° 55' W]; Secar Rock
[48 ° 26' N, 122 ° 54' W], Fig. 1). Sites in the analysis differed in wave exposure. The
Davis Bay, Whale Rocks, Hall Island, Long Island, and Secar Rock sites receive
heavier wave wash than the other sites because they face both Puget Sound to the south,
and the Strait of Juan de Fuca to the west, and therefore are exposed to bodies of water
with relatively large fetch to develop waves during storms. In contrast, the Willow
Island, Flower Island, and Humphrey Head sites are surrounded by larger islands of
the San Juan archipelago which shelter them from large waves.
To assess the effects of guano deposition on lichens in the supralittoral fringe (areas
receiving only salt spray, see Lewis, 1964), I measured from the upper limit of the acorn
barnacle Balanus glandula Darwin band to the upper and lower limits of the orange
lichen band (comprised of Caloplaca marina Wedd. and Xanthoria elegans (Link.) Th.
Fr.). Because the upper limit of Balanus did not change on adjacent portions of the cliff
face in the presence and absence of guano (pers. obs.), these tidal heights are consistent
among all sites within a particular cliff. This analysis included the Uptight Head cliff
142
J.T. WOOTTON
#
/......
0
LI
HI
SR~
5 km
Fig. 1. Study sites within the San Juan Islands, San Juan County, Washington State. Site abbreviations:
WI, Willow Island; FI, Flower Island; HH, Humphrey Head; UH, Upright Head; WR, Whale Rocks; DB,
Davis Bay cliffs; LI, Long Island subisland; HI, Hall Island; SR, Secar Rock. FHL, Location of Friday
Harbor Laboratories on San Juan Island.
site (Fig. 1). Upright Head was unsuitable for analysis of lower tidal heights because
fallen boulders produced a heterogeneous cliff face.
I censused 33 x 33-cm quadrats for the percent cover of space-holding plant and
animal species at three relative tidal heights (hereafter referred to as the splash, high,
and mid zones, respectively): 0.5 m above, 0.5 m below and 1.5 m below the upper limit
GUANO EFFECTS ON INTERTIDAL COMMUNITIES
143
of the acorn barnacle Balanus glandula band. For the high and mid zones, quadrats were
taken only on the four main cliff sites, but splash zone quadrats were taken at all sites
except Upright Head. Island sites were deemed appropriate to include in the splash zone
analysis because potentially confounding effects of bird predation were unlikely given
the absence of invertebrate prey species at this tide height. Six replicate pairs ofquadrats
were taken at each cliff site, four replicates at each island site. To avoid possible
i confounding effects of freshwater on intertidal organisms, no quadrats were taken under
areas that experience high run-off during rains.
Overall differences in species abundances between guano and non-guano sites were
generally tested with Wiicoxon paired ranks tests because of failure in normality
assumptions. Mid and high quadrats were pooled for the analysis because species
compositions were similar. Mann-Whitney U tests were conducted for comparisons
where the data structure did not allow pairing (e.g., between sites with different wave
exposure). I also performed correlations between species showing a response to guano
and other members of the community to look for significant patterns of association
among the dominant species. Significant correlations would be consistent with a hypothesis of interactions among species, which might imply that important indirect effects
of guano could occur. When significant correlations were present, I performed multiple
regression on percent cover of a given species, using guano, relative tidal height, and
all correlated species whose abundance varied with guano as independent variables in
an effort to probe whether any of the differences found between sites with and without
guano resulted from direct effects of guano or indirect effects of guano on other
interacting species. Because multiple regression estimates the direct linear effects on the
dependent variable of an independent variable with the other independent variables
included in the model held constant (Sokal & Rohlf, 1981), if the guano effect remained
in this analysis, changes in abundance were deemed likely to be, at least in part, a direct
effect of guano. However, if the apparent effect of guano in the regression differed from
that detected in simple paired comparisons between sites with and without guano, the
regression was analyzed further to determine likely indirect effects of guano via an
interaction with other species. All data were examined for linearity prior to the regression analysis.
RESULTS
SUPRALITTORAL ZONE
The height of the Caloplaca/Xanthoria band above the upper limit of Balanus increased
in the presence of guano. Normally the orange lichen band occurs above the black lichen
(Verrucaria mucosa Wahl.) band in the zone of wave-splash, and below the terrestrial
gray lichens (Physcia (Schreb.) Michx. sp.). At the site of a seabird colony, the bottom
of the gray lichen band and all of the orange lichen band occurred above the top
of the guano area (Table I). The bottom of the orange lichen band at a bird colony
144
J.T. WOOTTON
TABLE I
Height (m) above the upper limit of Balanus glandula of upper and lower limits of the orange lichen band
in areas with and without seabird colonies on the cliff face. N = 13.
No guano
Guano
Top (SD)
Bottom (SD)
3.6 (2.4)
17.3 (6.1)
1.2 (0.8)
11.6 (5.4)
averaged 8.0 + 4.9 m above the top of the orange lichen band in areas without bird
colonies (Table I, Wilcoxon test, P < 0.001). Similarly, the boundary of the orange and
gray lichen zones averaged 13.7 + 5.3 m higher on shorelines with bird colonies
(Table I, Wilcoxon test, P < 0.001).
SPLASH ZONE
The two dominant taxa in the splash zone, the black lichen Verrucaria mucosa and
the green alga Prasiola meridionalis S. & G. responded differently to guano. Prasiola
cover across all sites increased > 60-fold in areas with guano compared to areas without
guano (Wilcoxon test, P < 0.001, Fig. 2a). Wave exposure influenced the response of
Prasiola to guano; in areas ofhigh wave exposure, Prasiola became the dominant species
(Fig. 2a, Wilcoxon test, P < 0.015), but in sheltered areas, it showed little response to
guano (Fig. 2a, paired t test, P > 0.1).
Unlike Prasiola, Verrucaria cover across all sites declined by 53 ~o in the presence of
guano (Wilcoxon test, P < 0.0001, Fig. 2b). In the absence of Prasiola, Verrucaria cover
did not respond significantly to guano (Fig. 2b, Wilcoxon test, P > 0.05), but declined
by 79% at guano sites with Prasiola present (paired t test, P < 0.001, Fig. 2b). In quadrat
pairs without Prasiola, Verrucaria cover did not differ among wave exposure sites either
with (37.7 + 36.6% sheltered, 38.8 + 19.3% exposed, Mann-Whitney U test, P > 0.9)
or without guano (53.8 + 32.6% sheltered, 47.5 + 42.1Y/o exposed, Mann-Whitney
U test, P > 0.1).
HIGH AND MIDDLE INTERTIDAL ZONE
The species composition in the upper intertidal zone shifted in the presence of guano
(Fig. 3). The percent cover of the brown alga Fucus distichus L., when exposed to guano,
declined by 93 Yo in high quadrats, and by 85 % in mid quadrats, relative to Fucus cover
in areas without guano (Wilcoxon test, P < 0.0001). Similarly, the area covered by the
acorn barnacle Balanus glandula declined by 2 8 ~ in high quadrats and by 45 ~o in mid
quadrats when exposed to guano (Wilcoxon test, P < 0.03). In contrast, the cover of
fleshy crustose red algae (mostly the "Petrocelis" stage of Mastocarpus papillatus
(C. Ag.) Kutzing, Ra!fsia pacifica Hollenb., and Hildenbrandia Nardo spp., see Dethier,
1987) increased by a factor of 2 in high quadrats and by a factor of 2.4 in mid quadrats
GUANO EFFECTS ON INTERTIDAL COMMUNITIES
145
Prasiola
70
60
50
L_
o
0
40
30
20
10
0
.....
i.
k
no guano
guano
Sheltered
_
,v
.
.
.
.
no guano
guano
Exposed
Verrucaria
80
70
60
=>
50
o
L)
30
20
10
0
no guano
guano
No Prasiola
no guano
guano
Prasiola
Fig. 2. Effect of guano on the green alga Prasiola meridionalis and the black lichen Verrucaria in the splash
zone. Response of Prasiola to guano at sites exposed to (N = 24) and sheltered (N = 12) from heavy wave
wash (see text), response of Verrucaria to guano in replicate pairs where Prasiola was (N = 15) or was not
(N = 21) present.
in the presence of guano (Wilcoxon test, P < 0.0001). Ulva L. spp. responded to guano
in a pattern related to wave exposure. At the low wave exposure sites (Willow Island,
Flower Island, and Humphrey Head), Ulva cover declined by 59~o in the presence of
guano (Fig. 4, Wilcoxon test, P < 0.035). In contrast, at the high exposure site (Davis
Bay), Ulva exhibited a near-significant doubling in cover in areas with guano (Fig. 4,
Wilcoxon test, 0.1 > P > 0.05). No significant effect of guano was observed in the cover
of the acorn barnacle Chthamalus dalli Pilsbry, and seven taxa of algae (Acrosiphonia
coalita (Rupr.) Scagel, Garbary, Golden & Hawkes, the foliose stage of Mastocarpus
papillatus, Endocladia muricata (Post. & Rupr.) J. Ag., Microcladia borealis Rupr.,
Porphyra C. Ag. spp., Rhodomela larix (Turn.) C. Ag., and diatoms; Table II, Wilcoxon
test, all P > 0.1).
146
J.T. WOOTTON
FUCUS
50
40
>
o
O
---F--
30
20
lO
0
•
rJ ..................
high guano
high no guano
mid guano
mid no guano
mid guano
mid no guano
mid guano
mid no guano
Balanus
40
-r-
30
>
o
(3
20
10
0
high guano
high no guano
Fleshy Crustose Algae
50
40
i._
(I)
>
o
(3
30
20
10
0
high guano
high no guano
Fig. 3. Response of Fucus distichus, Balanus glandula, and fleshy crustose algae to guano and relative tide
height. High transects are 0.5 m below the upper limit of Balanus, mid transects are 1.5 m below the upper
limit of Balanus. N = 24.
16
14
!
!~
[]
Guano
No Guano
12
10
O)
8
0
6
~)
4
/
2
0
high sheltered
mid sheltered
high exposed
mid exposed
Fig. 4. Response of Ulva to guano at two tidal heights in sites sheltered from (N = 18) and exposed (N 6)
to heavy wave wash. In sheltered areas, Ulva declines significantly in response to guano, but in exposed
areas, Ulva tends to increase.
GUANO EFFECTS ON INTERTIDAL COMMUNITIES
147
TABLE II
Percent cover (SD) of sessile species at high and mid-intertidal quadrats that showed no significant
differences (Wilcoxon paired ranks tests, P > 0.1) between guano and nonguano areas. N = 24.
Species
Chthamalus dalli
Endocladia muricata
Porphyra spp.
Mastocarpus papillatus
Rhodomela larix
Acrosiphonia coalita
Microcladia borealis
Diatoms
High intertidal
Mid intertidal
Guano
No guano
Guano
No guano
30.92
(30.12)
1.44
(5.14)
0.88
(3.18)
0.25
(0.72)
0.21
25.00
(25.88)
4.63
(8.00)
0.08
(0.41)
0.33
(0.76)
0.49
25.21
(25.09)
2.13
(7.64)
1.67
(4.50)
1.13
(2.31)
1.88
25.83
(29.33)
0.63
(1.53)
1.46
(3.65)
1.67
(5.03)
2.04
(0.59)
(1.14)
(6.34)
(5.78)
0
(0)
0
(0)
0.83
(4.08)
0
(0)
0.04
(0.20)
0
(0)
0.58
(1.44)
1.83
(7.10)
3.67
(7.41)
0.29
(1.04)
0.67
(2.10)
4.75
(14.32)
Abundances of species that exhibited differences between guano and nonguano sites
correlated significantly with other species in some cases, raising the possibility that
interactions among species might have affected the patterns observed among sites with
and without guano. Therefore the response of one species to guano might have counteracted the direct effect of guano, or indirectly caused differences in abundance of other
species. Fucus cover correlated positively with Balanus (r = 0.329, P < 0.01), and negatively with Chthamalus cover (r = -0.296, P < 0.01). Balanus cover correlated negatively with Chthamalus (r = - 0.351, P < 0.01). Fleshy crusts correlated negatively with
Chthamalus ( r = - 0 . 0 2 2 0 , P < 0 . 0 5 ) and positively with Acrosiphonia ( r = 0.216,
P < 0.05). Ulva correlated negatively with Chthamalus (r = -0.258, P < 0.02), and
positively with Porphyra spp. (r = 0.535, P < 0.01), and Acrosiphonia (r = 0.209,
P < 0.05).
Multiple regression analysis suggested a direct effect of guano on Fucus (Table III).
Fucus cover remained negatively correlated with guano when potential interacting
species were included in the regression. Indirect effects of species interactions appeared
important in the patterns of acorn barnacle abundance. When Fucus was included in
the regression model, the effect of guano on Balanus disappeared (P > 0.5), whereas a
strong positive relationship with Fucus remained (P < 0.0001, Table III). Chthamalus
showed no response to guano even after accounting for apparent interactions with other
species. Multiple regression suggested strong negative effects of Fucus, B. glandula, and
fleshy crustose algae on Chthamalus (Table III), as well as a reduction in abundance
148
J.T. WOOTTON
TABLE III
Results of multiple regressions estimating the direct effects of guano and potentially interacting species o n
focal species. Dependent variables included guano, tide height, and any species exhibiting differences
between guano and nonguano sites that also correlated with the focal species. Wave exposure was also
included in the model if it correlated significantly with abundance of the focal species in pairwise
correlations.
Species
Fucus
(r 2 = 47 %)
Dependent variable
Guano
Tide height
Balanus
Balanus
(r 2 = 27 % )
Guano
Tide height
Fucus
Chthamalus
(r 2 =
71%)
Guano
Tide height
Exposure
Fucus
Balanus
Porphyra
(r 2 = 30%)
Fleshy crusts
UIva
Guano
Tide height
Ulva
Acrosiphonia
(r 2 = 16%)
Guano
Tide height
Fleshy crusts
Ulva
Slope
sE
P
- 24.86
- 15.36
0~38
3.19
17.75
0.41
1.31
- 2.14
- 41.11
- 0.21
- 0.56
- 0.57
- 0.25
0.44
- 0.57
0.29
0.02
- 0.26
0.007
0.03
3.74
3.87
0.09
4.72
3.94
0.10
4.20
4.09
4.19
0.09
0.09
0.09
0.30
0.58
0.58
0.05
0.19
0.20
0.005
0.03
< 0.0001"
< 0.0002*
< 0.0001"
> 0.5
< 0.0001"
< 0.0001"
> 0.75
>0.6
< 0.0001"
< 0.025*
< 0.0001"
< 0.0001"
> 0.4
> 0.4
> 0.3
< 0.0001"
> 0.8
> 0.15
> 0. I
< 0.05
with increased wave exposure (Chthamalus did not occur at Davis Bay). Multiple
regression analysis did not indicate that any indirect interactions counteracted the
effects of guano on any of the six less-common algae species examined (Table III).
DISCUSSION
Guano altered intertidal community structure at several tidal heights, however in
most instances, species cover did not increase with guano addition. Guano directly
reduced the cover of some species, whereas changes in species abundance by guano
appeared to indirectly affect other interacting species. Many species showed no
response to guano.
In the supralittoral, lichen zonation shifted in the presence of guano. At seabird
colonies, the orange lichens Caloplaca and Xanthoria lived at much higher levels than
in areas without guano. The shift in the level of the orange lichen band could have
reflected a changing response to differing guano concentrations between the edges and
center of the colony: orange lichen could respond positively at moderate concentrations
and negatively at high concentrations. The shift in lichen zonation at the colony
GUANO EFFECTS ON INTERTIDAL COMMUNITIES
149
periphery could correspond to lower intensities of bird use, and therefore lower guano
deposition that subsequently washes away annually, or to guano transport via winds
or rain splash to a localized zone around the colony. Alternatively, the change in lichen
distributions could result from decreased competition for free space or other resources
caused by the death of the grey lichen Pkyscia above the orange lichen band. Physcia,
a leafy form of lichen, frequently appears to overgrow other adherent lichens, notably
Caloplaca (pers. obs.). The orange lichens may be more tolerant than the grey lichens
to several conditions caused by the guano. First, lichens live in nutrient-poor habitats,
and may be poorly adapted to deal with high nutrient concentrations. Second, seabird
guano also contains high sea salt concentrations (Blakemore & Gibbs, 1968). The
"terrestrial Physcia may be more sensitive to salt conditions than the orange lichens,
which could explain the zonation patterns in both the presence and absence of guano.
Elevated salt concentrations result from guano around bird colonies and wave spray
in other areas. Gillham (1956, 1960)noted that the composition of terrestrial plants in
seabird colonies shifted toward halophytes and other species generally found in areas
of high salt spray, although she did not specifically invoke salt content of the guano as
a causal mechanism.
In the splash zone, the area covered by Prasiola increased in the presence of guano,
but only in areas of higher wave exposure. This observation suggests that the concentration of guano might have been important in determining the response of algae to
guano. Where wave wash is greater, guano becomes more diluted and Prasiola growth
is enhanced. At more sheltered sites, guano concentration remains high, and Prasiola
growth is depressed. Alternatively, high desiccation stress in the sheltered sites could
prohibit the growth of Prasiola independent of guano.
Guano appeared to indirectly inhibit the abundance of Vermcaria in the splash zone
by enhancing Prasiola. At sites without Prasiola, Verrucaria cover did not change with
differences in guano, but declined in the presence of guano at sites that contained
Prasiola. Prasiola is a faster-growing upright species which may easily shade out the
slowly growing Vermcaria (Dethier, 1987), thus the decrease of Verrucaria in the
presence of guano appears to have been an indirect consequence of competition.
In the intertidal zone, guano directly enhanced the cover of fleshy crustose algae
independently of its effects on other species. The other dominant plant in this intertidal
community, Fucus, declined in the presence of guano, even after accounting for possible
positive interactions with acorn barnacles as a result of grazer interference (see Farrell,
1991). Therefore guano appears to inhibit Fucus in some manner. Guano also affected
UIva, but in a habitat-specific manner. Like Prasiola, the response of Ulva to guano
depended on the degree of wave exposure. In sites with low wave exposure, guano
reduced Ulva cover, but in more wave-washed areas, Ulva tended to cover more area
in the presence of guano.
Guano indirectly influenced the abundance of Balanus by reducing a facilitating
species. Fucus and Balanus cover correlated positively, and Fucus responded adversely
to guano. Statistically accounting for the effects of Fucus eliminated the differences in
150
J.T. WOOTTON
Balanus cover in areas with and without guano. These interpretations are consistent
with the results of other studies on these species. Dayton (1971) found that experimental
removal of Fucus caused decreased survivorship of juvenile Balanus, presumably
because the Fucus canopy reduced desiccation stress, and Farrell (1989) found that
Balanus cover increased with increasing algal cover in gaps undergoing succession.
The cover of the second acorn barnacle in the system, Chthamalus, did not differ in
areas with and without guano, despite the apparent negative effects it apparently
experiences from several species influenced by guano. Chthamalus correlated negatively
with Fucus, Balanus, and fleshy crustose algae, suggesting inhibition by all three species.
Balanus beats Chthamalus in competition for space (Connell, 1961a,b; Dayton, 1971;
Farrell, 1991), and barnacles do not successfully attach and grow on the smooth surface
of fleshy crusts (Paine et al., 1979; pers. obs.). Because guano affects the three apparent
competitors in different ways (increasing fleshy crusts, decreasing Fucus and Balanus),
their effects on Chthamalus cancel. Where one of these species is absent, guano would
be expected to have an indirect effect on Chthamalus abundance.
Overall, guano appeared to directly enhance only four of the eighteen taxa of sessile
organisms investigated. Given the high nutrient content of guano (Hutchinson, 1950;
Gillham, 1956; Blakemore & Gibbs, 1968; Ganning & Wulff, 1969; Lindeboom, 1984;
Bosman & Hockey, 1986), the failure to find many increases in the presence of guano
does not lend much support to the hypothesis that the biomass of most species in this
intertidal system is limited by inorganic nutrients. Without gaining benefits of additional
nutrients in the system, some species reveal probable toxic effects of enhanced guano
input. The mechanism ofbiomass reduction with the addition of guano is unknown, but
for marine-adapted forms, may relate to enhanced ammonium rather than increased salt
input (discussed above). With moisture, guano breaks down primarily to ammonia
(Hutchinson, 1950, Blakemore & Gibbs, 1968), which is toxic to most organisms in high
concentrations.
The two groups of species that clearly increased in biomass with guano input, Prasiola
and fleshy crustose algae, exist in situations in which they may be particularly likely to
benefit from enhanced nutrients. Prasiola typically lives at a tidal height where it is
generally not immersed in water, hence nutrient delivery from the surrounding waters
is limited. The addition of guano may thus provide a critical supplemental nutrient
source when provided in moderate concentrations.
Because of their morphology, fleshy crustose algae may take up nutrients less
effectively than other algae, a factor which may cause their relatively slow growth (Paine
et al., 1979; Dethier, 1987). Crustose algae are completely adherent to the rock, thus
they have approximately half the surface area exposed to the water for nutrient uptake
that they would have as free-floating fronds. Furthermore, the hydrodynamic conditions
these species face may be less favorable for nutrient delivery than species with fronds.
Nutrient delivery to aquatic plants is enhanced with water motion, which reduces the
boundary layer across which nutrients must diffuse (Neushul, 1972; Wheeler, 1980;
Gerard, 1982; Denny, 1988; Koehl & Alberte, 1988). Crustose algae reside in the
GUANO EFFECTS ON INTERTIDAL COMMUNITIES
151
boundary layer of the rock substrate, and because of their smooth surface, may
experience reduced turbulence on a micro-scale. In contrast, species with fronds extend
beyond the rock boundary layer and experience increased turbulent flow as their fronds
flap in moving water (Koehl & Alberte, 1988), potentially enhancing nutrient exchange.
Thus, although the crustose morph of Mastocarpus papillatus responds positively to
guano, the foliose morph does not. This hypothesis requires further study.
In summary, by concentrating guano on or along the margins of sea cliffs, birds alter
the structure of Washington shore communities at several tidal levels. By importing
nutrients into the system, birds enhance the growth of fleshy crustose algae, and, at
lower guano concentrations, Prasiola and perhaps Ulva. However in most cases,
increasing local nutrient concentrations did not enhance biomass. In many instances
guano had no direct effects, and in several cases apparently had toxic effects. The varied
direct responses exhibited by species to guano in this community thus caution against
generalizations characterizing an entire community in terms of nutrient limitation.
Finally, some differences seen between sites with and without guano were not the direct
result of guano input, but instead appeared to be the consequences of interactions with
affected species. Such indirect effects should be considered more often in studies of
community response to alteration in nutrient regime.
ACKNOWLEDGMENTS
I thank P. Spatig, F. Thomas, and B. Wootton for field assistance, the Friday Harbor
Laboratories for logistical support and transportation, the U S Fish and Wildlife Service
for permitting access to study sites in the San Juan Islands National Wildlife Refuge,
and K. Banse, M. Dethier, D. Duggins, D. Harvell, P. Karieva, J. Kenagy, A. Kohn,
G. Orians, R. Paine, C. Pfister, W. Sousa, and K. VanAlstyne for providing helpful
comments on the manuscript and during the study. This study was supported in part
by an NSF predoctoral fellowship and NSF grants OCE-8415707 and OCE-8614463
to Robert T. Paine.
REFERENCES
Bedard, J., J. C. Therriault & J. Berube, 1980. Assessment of the importance of nutrient recycling by seabirds
in the St. Lawrence Estuary. Can. J. Fish. Aquat. Sci., Vol. 37, pp. 583-588.
Blakemore, L.C. & H.S. Gibbs, 1968. Effects of gannets on soil at Cape Kidnappers, Hawke's Bay. N. Z.
J. Sci., Vol. 11, pp. 54-62.
Borowitzka, M.A., 1972. Intertidal algal species diversity and the effects of pollution. Aust. J. Mar.
Freshwater Res., Vol. 23, pp. 73-84.
Bosman, A. L., J.T. Du Toit, P. A. R. Hockey & G.M. Branch, 1986. A field experiment demonstrating the
influence of seabird guano on intertidal primary production. Estuarine Costal Shelf Sci., Vol. 23,
pp. 283-294.
Bosman, A.L. & P.A.R. Hockey, 1986. Seabird guano as a determinant of rocky intertidal community
structure. Mar. Ecol. Progr. Ser., Vol. 32, pp. 247-257.
Bosman, A. L. & P.A.R. Hockey, 1988. The influence of seabird guane on the biolog~ca! structure of rocky
152
J.T. WOOTTON
intertidal communities on islands off the west coast of southern Africa. S. Aft. J. Mar. Sci., Vol. 7,
pp. 61-68.
Bradshaw, A.D., M.J. Chadwick, D. Jowett & R.W. Snaydon, 1964. Experimental investigation into the
mineral nutrition of several grass species IV. Nitrogen level. J. Ecol., Vol. 52, pp. 665-676.
Brown, V.B., S.A. Davies & R.N. Synnot, 1990. Long-term monitoring of the effects of treated sewage
effluent on the intertidal macroalgal community near Cape Schanck, Victoria, Australia. Bot. Mar.,
Vol. 33, pp. 85-98.
Chapin, F. S. & G. R. Shaver, 1985. Individualistic growth response of tundra plant species to environmental
manipulation in the field. Ecology, Vol. 66, pp. 564-567.
ConneU, J.H., 1961a. Effects of competition, predation by Thais lapillus, and other factors on natural
population of the barnacle Balanus balanoides. Ecol. Monogr., Vol. 31, pp. 61-104.
Connell, J. H., 1961b. The influence ofinterspecific competition and other factors on the distribution ofthe
barnacle Chthamalus stellatus. Ecology, Vol. 42, pp. 710-723.
ConneU, J.H., 1972. Community interactions on marine rocky intertidal shores. Annu. Rev. Ecol. Syst.,
Vol. 3, pp. 169-192.
Dayton, P.K., 1971. Competition, disturbance, and community organization: the provision and subsequent
utilization of space in a rocky intertidal community. Ecol. Monogr., Vol. 41, pp. 351-389.
Denny, M.W., 1988. Biology and the mechanics of the wave-swept environment. Princeton University Press,
Princeton, New Jersey, 329 pp.
Dethier, M.N., 1987. The distri~otion and reproductive phenology of intertidal fleshy crustose algae in
Washington. Can. J. Bot., Vol. 65, pp. 1838-1850.
Edmondson, W.T., 1970. Phosphorus, nitrogen and algae in Lake Washington al~er diversion of sew:~ge.
Science, Vol. 169, pp. 690-691.
Farrell, T.M., 1989. Succession in a rocky intertidal community: the importance of disturbance size and
position within a disturbed patch. J. Exp. Mar. Biol. EcoL, Vol. 128, pp. 57-73.
Farrell, T. M., 1991. Models and mechanisms of succession: an example from a rocky intertidal community.
Ecol. Monogr., Vol. 61, pp. 1838-1850.
Gunning, B. & F. Wulff, 1969. The effects ofbird droppings on chemical and biological dynamics in brackish
water rockpools. Oikos, Vol. 20, pp. 274-286.
Gerard, V.A., 1982. In situ water motion and nutrient uptake by the giant kelp Macrocystus pyrifera. Mar.
Biol., Vol. 69, pp. 51-54.
Gillham, M.E., 1956. Ecology of the Pembrokeshire Islands V. Manuring by the colonial seabirds and
mammals, with a note on seed distribution by gulls. J. Ecol., Vol. 44, pp. 429-454.
Giilham, M. E., 1960. Destruction of indigenous heath vegetation in Victorian sea-bird colonies. Aust. J. Bot.,
Voi. 8, pp. 277-317.
NA
Goldberg, D. E. & T. E. Miller, 19~,v.
Effects of different resource additions on species u"~:
, v.....
, ~ , o ,:,.,
, s :..
~,, ~,,
n., annual
plant community. Ecology, Vol. 71, pp. 213-225.
Golovkin, A.N., 1967. The effect of colonial sea birds on development of the phytoplankton. Oceanology,
Vol. 7, pp. 521-529.
Golovkin, A.N. & G.P. Garkavaya, 1975. Fertilization of waters of the Murmansk coast by bird excreta
near various types of colonies. Soy. J. Mar. Sci., Vol. l, pp. 345-351.
Hanisak, M.D., 1983. Nitrogen relationships of marine macroalgae. In, Nitrogen in the marine environment,
edited by E.J. Carpenter & D.G. Capone, Academic Press, New York, pp. 699-730.
Hayes, J. P. & J.W. Caslick, 1984. Nutrient deposition in cattail stands by communally roosting blackbirds
and starlings. Am. Mid/. Nat., Vol. ll2, pp. 320-331.
Hockey, P. A. R. & G. M. Branch, 1984. Oystercatchers and limpets: impact and implications. A preliminary
assessment. Ardea, Vol. 72, pp. 199-206.
Hutchinson, G. E., 1950. Survey of contemporary knowledge of biogeochemistry 3. The biogeochemistry of
vertebrate excretion. Bull. Am. Mus. Nat. Hist., Vol. 96, pp. 1-554.
Koehl, M.A.R. & R.S. Alberte, 1988. Flow, flapping, and photosynthesis of Nereocystis luetkeana: a
-"~functional comparison of undulate and fiat blade morphologies. Mar. Biol., Vol. 99, pp. 435-444.
L e e n ~ r , P., 1967. Observations in guanotrophie environments. Hydrobiologia, Vol. 29, pp. 441-489.
Leigh, E. G..,~. T.~._F.___Qu_inn & T. H. Suchanek, 1987. Wave energy and intertidal productivity. Proc.
Natl. Acad. Sci. U.S.A., Vol. 84, p p . ~ - i 4 ~
.......
Lewis, J.R., 1964. The ecology ofrockv, shores. English Universities Press, London, 323 pp.
Lindeboom, H.J., 1984. The nitrogen pathway in a penguin rookery. Ecology, Vol. 65, pp. 269-277.
GUANO EFFECTS ON INTERTIDAL COMMUNITIES
153
Littler, M.M. & S.M. Murray, 1975. Impact of sewage on the distribution, abundance and community
structure of rocky intertidal macro-organisms. Mar. Biol., Vol. 30, pp. 277-291.
Mann, K.H., 1973. Seaweeds: their productivity and strategy for growth. Science, Vol. 182, pp. 975-981.
McColl, J. G. & J. Burger, 1976. Chemical inputs by a colony of Franklin's gulls nesting in cattails. Am. Midl.
Nat., Vol. 96, pp. 270-280.
Neushul, M., 1972. Functional interpretation of benthic marine algal morphology. In, Contributions to the
systematics of benthic marine algae of the North Pacific, edited by I.A. Abbot & M. Kurogi, Japanese Society
of Phycology, Kobe, Japan, pp. 47-74.
Onuf, C. P., J. M. Teal & I. Valiela, 1977. Interactions of nutrients, plant growth and herbivory in a mangrove
ecosystem. Ecology, Vol. 58, pp. 514-526.
Paine, R.T., 1977. Controlled manipulations in the marine intertidal zone and their contributions to
ecological theory. Acad. Natl. Sci. Philadelphia Spec. Publ., Vol. 12, pp. 245-270.
Paine, R.T., C.J. Slocum & D.O. Duggins, 1979. Growth and longevity in the crustose red algae Petrocelis
middendorffii. Mar. Biol., Vol. 51, pp. 185-192.
Poweli, G. V. N., W.J. Kenworthy & J.W. Fourqurean, 1989. Experimental evidence for nutrient limitation
of seagrass growth in a tropical estuary with restricted circulation. Bull. Mar. Sci., Vol. 44, pp. 324-340.
Pringle, C.M., 1990. Nutrient spatial heterogeneity: effects on community structure, physiognomy, and
diversity of stream algae. Ecology, Vol. 71, pp. 905-920.
Schindler, D.W., 1977. Natural compensation for deficiencies of nitrogen and carbon by eutrophied lake
ecosystems: why phosphorus control works. Science, Vol. 195, pp. 260-262.
Sokal, R.R. & F.J. Rohlf, 1981. Biometry, W.H. Freeman and Co., San Francisco, California, second
edition, 859 pp.
Speich, S.M. & T. R. Wahl, 1989. Catalog of Washington seabird colonies. U.S. Fish. Wildl. Serv. Biol. Rep.,
Vol. 88 (6), pp. 1-510.
Tilman, D., 1987. Secondary succession and the pattern of plant dominance along an experimental nutrient
gradient. Ecol. Monogr., Vol. 57, pp. 189-214.
Tilman, D., S.S. Kilham & D. Kilham, 1982. Phytoplankton community ecology: the role of limiting
nutrients. Annu. Rev. Ecol. Syst., Vol. 13, pp. 349-372.
Vitousek, P.M., 1986. Biological invasions and ecosystem properties: can species make a difference? In,
Ecology of biological invasions of North America and Hawaii, edited by H.A. Mooney & J.A. Drake,
Springer-Verlag, Berlin, FRG, pp. 163-176.
Weseloh, D.V. & R.T. Brown, 1971. Plant distribution within a heron rookery. Am. Midl. Nat., Vol. 86,
pp. 57-64.
Wheeler, W.N., 1980. Effect of boundary layer transport on the fixation of carbon by the giant kelp
Macrocystis pyrifera. Mar. BioL, Vol. 56, pp. 103-110.
Wootton, J. T., 1990. Direct and indirect effects ofbirdpredation and excretion on the spatial and tempora!patt~,'ns
of intertidal species. Ph.D. thesis, University of Washington, Seattle, 297 pp.