Behavioral Ecology Vol. 11 No. 5: 565-571
© 2000 International Society for Behavioral Ecology
Integrating individual behavior and population ecology: the potential for habitat-dependent population regulation in a reef fish
a Northwest Fisheries Science Center, National Marine Fisheries Service, 2725 Montlake Boulevard E, Seattle, WA 98112, USA b University of Auckland, Leigh Marine Laboratory, PO Box 349, Warkworth, New Zealand c Department of Biology, Northeastern University, Boston, MA 02115 d Department of Biology, University of Windsor, Windsor, Ontario N9B 3P4, Canada
Address correspondence to P. S. Levin. E-mail: phil.levin{at}noaa.gov .
Received 23 August 1999; revised 3 April 2000; accepted 5 April 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
We used the predictions of the ideal free and ideal despotic distributions (IFD and IDD, respectively) as a basis to evaluate the link between spatial heterogeneity, behavior, and population dynamics in a Caribbean coral reef fish. Juvenile three-spot damselfish (Stegastes planifrons) were more closely aggregated in patch reef habitat than on continuous back reef. Agonistic interactions were more frequent but feeding rates were lower in the patch versus the continuous reef habitat. Growth rates were lower in patch reef habitat than on the continuous reef, but mortality rates did not differ. A separate experiment using standard habitat units demonstrated that the patterns observed in natural habitat were the result of the spatial distribution of the habitat patches rather than resource differences between habitats. Our results do not follow the predictions of simple IFD or IDD models. This deviation from IFD and IDD predictions may be the result of a number of factors, including lack of perfect information about habitat patches, high movement costs, and higher encounter rates of dispersed patches. Our results demonstrate that behavioral interactions are an integral part of population dynamics and that it is necessary to consider the spatial organization of the habitat in both behavioral and ecological investigations.
Key words: damselfish, habitat structure, ideal free distribution, recruitment, Stegastes planifrons.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Behavioral interactions among individuals both influence and are influenced by demographic processes. Many of the ecological processes that determine population size or dynamics are explicitly behavioral, and fully understanding ecological phenomena such as predator-prey interactions, some forms of interference competition, dispersal, and patterns of habitat use often requires knowledge of specific behavioral interactions occurring among individuals (Fryxell and Lundberg, 1998
; Hassell and May,
1985; Mangel and Clark,
1988; Smith and Sibly,
1985; Sutherland,
1996). Behavior can also profoundly affect the dynamics of
populations by adding or modifying density dependence
(Chesson and Rosenzweig,
1991). Although there is a compelling body of theory linking
behavioral and population ecology (e.g., Abrams,
1989,
1992;
Fryxell and Lundberg, 1998;
Holt, 1983;
Houston and McNamara; 1997;
Ives and Dobson, 1987;
Sih, 1984;
Sutherland, 1996), empirical
work exploring this linkage is sparse
(Anholt, 1997;
Dunbar, 1985;
Fryxell and Lundberg, 1998;
Gilliam, 1987).
The effects of spatial heterogeneity of resources on patch use and movement
of animals has been a mainstay of behavioral ecology. Fretwell
(1972;
Fretwell and Lucas, 1970)
deduced that individuals should distribute themselves among patches that vary
in resources and conspecific density such that the average reward is the same
for individuals in different patches. Thus, individuals will distribute
themselves such that the ratio of density between any two patches will equal
the ratio of the resource levels of those patches
(Hugie and Grand, 1998). This
model assumes that individuals have perfect information about the relative
availability of resources and are free to move among patches. Following these
assumptions, it is referred to as the ideal free distribution (IFD). Although
the IFD model was originally conceived in terms of birds choosing nesting
habitats, it has been widely applied to a diversity of animal taxa
(Kacelnik et al., 1992;
Milinski and Parker, 1991;
Tregenza, 1994) and recently
to plants (Gersani et al.,
1998). Because IFD models directly address the behavioral
mechanisms producing density-dependent growth, fecundity, and mortality, they
have clear implications for population dynamics
(Fryxell and Lundberg, 1998;
Sutherland, 1996).
Fretwell (1972) also
acknowledged that frequency-dependent habitat selection need not necessarily
result in IFDs. In particular, Fretwell proposed the ideal despotic
distribution (IDD) as an alternative to IFD when territorial behavior prevents
free entry to or movement among patches. In IDD models, the suitability of a
patch for an individual declines with the order of settling, but the presence
of new individuals does not decrease the suitability for those that have
already settled (Sutherland,
1996). The IDD differs from IFD in that the average reward differs
among patches, even though at any time, settling individuals obtain the same
reward in both places (Kramer et al.,
1997).
In addition to resource levels, the structure of a habitat can affect local
densities of animals as well as the frequency or intensity of behavioral
interactions. For example, animals in structurally complex habitats may
encounter each other at lower frequencies than in structurally simple habitats
(Chesson and Rosenzweig,
1991). As a consequence, aggressive interactions will be reduced
in complex habitats, and thus, the strength of interference competition will
also be reduced (Anholt, 1990;
Dye, 1984). The pattern of
dispersion of a population and thus local densities often reflects responses
of individuals to structural attributes of the habitat
(Brown, 1975;
Wilson, 1975). For instance,
the degree of aggregation of endophytaphagous insects and reef fishes can be
closely linked to the dispersion pattern of the vegetation they use as
foraging and refuge habitat (e.g., Levin,
1993; Morris et al.,
1992). In sites where vegetation is clumped, individual animals
will experience a higher local density relative to sites in which plants are
randomly or regularly dispersed. If increased density influences the
occurrence of aggressive interactions
(Boccia et al., 1988;
Ens and Goss-Custard, 1984;
Goss-Custard and Sutherland,
1997; Jones, 1984;
Sale, 1972), the degree of
patchiness of a habitat will affect the frequency of behavioral
interactions.
Foraging rates of animals are affected by at least two conflicting
parameters: the density of prey and the density of conspecific and
heterospecific competitors and thus levels of interference
(Sutherland, 1996). IFD models
describe a solution to this conflict in which consumers adjust their densities
in relation to prey levels such that foraging rates are equal in all prey
patches. For animals in which local densities and thus levels of interference
are determined, at least in part, by the structure of the habitat, attributes
of the habitat must also be included in assessment of foraging behavior. Two
patches that vary in structural complexity, for example, may have the same
prey levels, but for given consumer densities, levels of interference will be
lower in structurally complex than in simple habitat
(Anholt, 1990;
Dye, 1984). As a result, the
effect of habitat structure on encounter rates will increase the strength of
density dependence even though both prey and consumer densities are identical.
The increased interference can influence population size or dynamics by
reducing rates of growth, survivorship, or reproduction
(Sutherland, 1996).
Although theory and individual empirical studies suggest that habitat
structure, behavioral interactions, and demography should be linked, few
studies have explicitly explored these links. This is especially true in
marine habitats in general and coral reefs in particular, where simple
linkages between habitat and demography have only recently received attention
(Booth and Wellington, 1998).
In the present study, we extended previous work linking habitat and demography
or behavior by exploring linkages between habitat structure, behavioral
interactions, and juvenile demography in the three-spot damselfish
Stegastes planifrons. Specifically, we asked: (1) Does the spatial
pattern of juvenile fish vary among habitats? (2) Are foraging rate or the
frequency of aggressive interactions affected by the structure of the habitat?
and (3) Do habitat-dependent differences in behavior result in changes in
growth or mortality rates?
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study species
Three-spot damselfish are common members of Caribbean reef communities (Emery, 1973
). Adults
vigorously defend small territories year-round on the corals Acropora
palmata or A. cervicornis against numerous fish and
invertebrates (Itzkowitz,
1977; Myrberg and Thresher,
1974; Thresher,
1976; Williams,
1978). Within territories, three-spot damselfish cultivate algal
lawns as a food resource and guard demersal eggs
(Itzkowitz, 1977;
Kaufman, 1977;
Williams, 1978). Both adults
and juveniles feed on benthic algae as well as on benthic and planktonic
crustacea (Emery, 1973;
Randall, 1967). Three-spot
damselfish spawn throughout the year on a lunar cycle and produce demersal
eggs that are defended by males (Williams,
1978). After hatching, larvae enter an approximately 26-day
dispersive planktonic stage before settling primarily to habitat dominated by
the mounding coral Montastrea annularis
(Tolimieri, 1995). Settlement
occurs episodically around new moons
(Robertson, 1992), and fish
average 10.5 mm (SL) at settlement.
Spatial patterns of fish recruitment
We quantified nearest neighbor distances of juvenile three-spot damselfish
(< 45 mm SL) in two reef habitats within Tague Bay on the northeast shore
of St. Croix in the United States Virgin Islands in 1991 and again in 1997. A
well-developed bank barrier reef separates Tague Bay from open water. The reef
surface is composed mostly of dead Acropora palmata rubble. Heads of
Montastrea annularis and Porites porites and massive corals
such as Siderastea sideria are scattered throughout the back reef
area. Small patch reefs (frequently <1 m3) composed mostly of
M. annularis as well as massive corals were scattered just landward
of the backreef. Tolimieri
(1995,
1998) provides a more detailed
description of this site. We sampled along the continuous back reef and patch
reefs 20-25 m shoreward of the main reef.
To measure nearest neighbor distances, we haphazardly selected individual
fish and noted the point of first observation. We then searched for other
three-spot damselfish in the vicinity and measured the distance between the
original fish and its nearest neighbor. No fish was sampled twice (i.e., once
as the focal animal and again as a nearest neighbor). We used a two-factor
ANOVA to compare nearest neighbor distances between continuous reef and patch
reef habitat with habitat type and sampling year as the main effects. Before
analysis we tested for homogeneity of variances using Levene's test
(Wilkinson et al., 1996).
Variances of log-transformed data were homogeneous.
Behavioral observations
We conducted focal animal observations
(Altman, 1974) on juvenile
three-spot damselfish in both continuous and patch reef habitat in 1991 and
1997. In 1991, we conducted 5-min focal observations, and in 1997 we performed
10-min observations. Fish were observed from a distance of at least 2 m, which
was sufficient to allow fish to behave normally. When fish were not in view
for the entire observation period, sampling was abandoned and restarted when
the fish reappeared. This only rarely occurred, and there was no systematic
difference between back and patch reefs in our ability to observe fish. Thus,
all data reported here are from individuals that were in full view during the
entire observation period. All observations were conducted between 1100 and
1530 h.
During each observation, we quantified the number of bites at food items
and the number of agonistic interactions in which juvenile damselfish in each
habitat were involved. In July 1991, we considered all agonistic behavior
including lateral displays, chases, and nips both directed at other fish and
received from other fish. In July 1997, we quantified only chases directed at
other fish (some of which concluded with nips at other fish) because this was
the most prevalent agonistic behavior we observed. Because foraging rate
(i.e., the number of bites at food) and the number of agonistic interactions
were recorded on individual fish during the same observation period, these
behaviors may not be independent of each other. Consequently, we refer to
these two parameters collectively as "behavior" and used
multivariate analysis of variance (MANOVA) to compare the behavior of
three-spot damselfish on continuous versus patch reefs. MANOVA is used to deal
with multiple dependent variables that may be correlated. Such variables
cannot be tested separately because it is unclear what the correct 
would be given the correlation between the two variables. Thus, we first used
MANOVA to test the hypothesis that habitat affects behavior, and then tested
each behavior (foraging and aggression) individually. We performed two
separate MANOVAs on data collected in 1991 and 1997 because the duration of
the observational period varied. Before analysis we tested for homogeneity of
variances using Levene's test (Wilkinson
et al., 1996). Variances of log-transformed data were
homogeneous.
Because continuous and patch reefs differ from each other in both the
degree of patchiness and physical location, we performed an experiment on
standard habitat units (SHU) in a single location to remove this potentially
confounding effect from our observations. SHUs consisted of four PVC pipes
(10.2 cm long, 7.6 cm diam) bolted in a cross pattern to a 0.25-m2
Plexiglas base (Itzkowitz and Makie,
1986). Coral rubble was placed in the center of each unit. SHUs
were arrayed in two spatial configurations on a barren sand flat in Discovery
Bay,Jamaica (18°30' N, 77°24' W) in February 1997. In the
first treatment we positioned four SHUs in the corners of a square such that
the center of adjacent SHUs were 1 m apart. In our second treatment SHUs were
arrayed in an identical fashion, except the centers of adjacent SHUs were
separated by 0.5 m. These distances were selected to reflect average nearest
neighbor distances between fish (see Results). Five replicates of each
treatment were arrayed in a complete randomized block design such that
treatments were separated from each other by 10 m and blocks were separated
from each other and natural habitat by >15 m.
Juvenile three-spot damselfish (<45 mm SL) were captured on adjacent natural habitat. We then measured fish (SL) and transplanted one fish to each SHU. Fish were assigned haphazardly to experimental locations. Thus, each experimental plot of four SHUs received four fish. All transplanted fish remained on SHUs. Three days after we transplanted fish, we commenced 5-min focal observations. We recorded both foraging bites and agonistic behavior as we did in our 1997 observations in natural habitat. Data were analyzed with MANOVA as described above. Where MANOVA detected significant differences, we tested each dependent variable (feeding rate, aggression rate) with t tests.
Effects of habitat on growth and mortality
Although demography typically includes age-specific rates of movement,
mortality, growth, and reproduction, we limited our demographic analyses to
rates of mortality and growth. We first compared the change in standard length
of fish on natural continuous and patch reefs to determine whether differences
in nearest neighbor distance and behavior might affect fish growth. From 26
May to 10 June 1997, we searched both continuous and patch reef habitat in
Tague Bay for newly recruited three-spot damselfish. Fish were collected after
being anesthetized with quinaldine and were placed in plastic bags and
measured to the nearest 0.5 mm SL. We then gave each fish a unique mark using
florescent elastomer (Malone et al.,
1999) and returned fish to their original location on continuous
or patch reefs. We censused marked fish at approximately weekly intervals
until we terminated the sampling between 26 and 31 July 1997 by collecting and
remeasuring all marked fish. During each census and upon termination of the
sampling, we searched the surrounding habitat for marked individuals that were
not at their initial capture/release point. We did not find any marked
individuals that were classified as missing during the regular censuses.
Because three-spot damselfish have extremely small home ranges
(Itzkowitz, 1977;
Myrberg and Thresher, 1974;
Thresher, 1976;
Williams, 1978), losses of
marked fish were ostensibly due to mortality rather than migration.
We used analysis of covariance (ANCOVA) to compare change in SL per day between fishes on continuous and patch reef. Initial SL was included as the covariate to control for differences in the starting size of damselfish. Before analysis, we checked data for normality, homogeneity of variance, and homogeneity of slopes. To determine whether mortality rates differed between habitats, we used Fisher's Exact test to compare the number of fish missing on continuous and patch reefs. We included only those fish that survived to their first census to control for mortality that might be attributable to handling effects. We have thus defined mortality rate as the presence of fish from the first census to the final collection.
We also examined growth rates of fish on our SHUs. After fish were captured and measured, individuals on sets of four SHUs were individually marked using fin clips. After 15 days, fish were recaptured, placed in plastic bags, and measured to the nearest 0.5 mm SL.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Spatial patterns of fish recruitment
In both habitats, nearest neighbor distances were never >2.5 m. However, nearest neighbor distances were consistently shorter in patch than in continuous reef habitat (F1,167 = 4.12, p <.001; Figure 1). In both 1991 and 1997, nearest neighbor distances on patch reefs averaged about 47 cm, while on continuous reefs nearest neighbor distances averaged about 105 cm (Figure 1). We were unable to detect a difference in nearest neighbor distance between years (F1,167 = 0.01, p =.98), nor was there a significant interaction between habitat type and year (F1,167 = 0.04; p =.85).
|
Behavioral observations
Natural habitat
Our focal observations of three-spot damselfish revealed clear differences
in behavior between continuous and patch reef habitats in both 1991 (MANOVA,
Pillai trace = 0.64, df = 1,50, p <.001) and 1997 (MANOVA, Pillai
trace = 0.56, df = 1,61, p <.001). In 1991, fish in continuous
habitat took significantly more bites at food than fish on patch reefs
(Figure 2; t = 6.70,
df = 52, p <.001). Likewise, in 1997, on continuous reefs, we
observed three-spot damselfish foraging at more than double the rate seen on
patch reefs (Figure 2;
t = 7.50, df = 61, p <.001).
|
The number of agonistic interactions in which three-spots were involved was lower on continuous and than on patch reef habitat (Figure 2). In 1991, damselfish were involved in an average of fourfold more agonistic interactions on patch than continuous reefs (t = 6.028, df = 49, p <.001). Similarly, in 1997 we observed three-spot damselfish in more than twice the number of agonistic interactions on patch than on continuous reefs (t = 4.88, df = 61, p <.001).
Standard habitat units
SHUs were arrayed such that they were 0.5 m or 1 m apart, and this closely
mimicked nearest neighbor distances in patch and continuous reefs,
respectively (Figure 1). Our
behavioral observations of transplanted three-spot damselfish on SHUs also
resembled our observations in natural habitat with clear differences in
behavior between fish on SHUs separated by 0.5 m and 1 m (MANOVA, Pillai trace
= 0.24, F = 6.623, df = 2,42, p =.003). Fish tended to
forage at a higher rate on SHUs separated by 1 m than on SHUs 0.5 m apart,
although this was not significant at an of 0.05 (T = 1.92,
p =.06; Figure 3).
Damselfish on SHUs separated by 0.5 m were involved in nearly double the
agonistic interactions as those on SHUs 1.0 m apart (one-tailed t
test, df = 43, T = 2.94, p <.001).
|
Effects of habitat on growth and mortality
In natural habitat we observed a strong effect of habitat type on growth
rate (Figure 4; ANCOVA,
F1,70 = 47.499, p <.001), with fish on
continuous reef growing more quickly than those on patch reefs. Initial size
of the fish also affected the change in SL per day
(Figure 4; ANCOVA,
F1,70 = 166.82, p <.001). On SHUs, we were
able to unambiguously identify eight fish on SHUs separated by 0.5 m and nine
fish on SHUs separated by 1.0 m. Daily percent growth of fish on SHUs
separated by 1.0 m averaged 0.60%, and this was significantly greater than the
0.03% daily growth we observed on SHUs separated by 0.5 m
(F1,15 = 4.45, p =.05).
|
We did not detect a difference in mortality rates between continuous and patch reefs (Fisher's Exact test, p =.816). Fifty-two fish survived on patch reefs to the first census, while 46 fish survived on continuous reefs. Of these survivors, 76% were recovered on patch reefs and 73.9% on continuous habitats at the end of the experiment.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Species interactions have been the traditional domain of population and community ecologists. However, it is clear that aspects of individual behavior underlie such interactions and may thus have important ecological consequences (Real, 1992
). In this study,
we examined how habitat structure affected patterns of fish dispersion, how
fish behavior was influenced by habitat-induced differences in aggregation,
and how growth and mortality rates differed as a function of these behavioral
differences. Juvenile three-spot damselfish occupy home ranges <1
m2 on coral and do not stray far from the core of their home range
(Itzkowitz, 1977). As a
result, the dispersion of juvenile three-spot damselfish is closely tied to
the patch structure of the reef, and density of juvenile damselfish was higher
on patch reef than on continuous reef habitat (nearest neighbor distances were
shorter on patch than on continuous reef). The higher aggregation of fish on
patch versus continuous reef appears to have resulted in a greater number of
agonistic interactions on patch reef than on continuous habitat. The increased
time spent on aggression on patch reefs apparently occurred at the expense of
foraging, as fish on patch reefs spent less time foraging than their
counterparts on continuous reefs. The elevated rates of aggression and reduced
rates of foraging on patch versus continuous reef habitat were associated with
a decrease in growth rates of fish on patch reefs, although survivorship did
not differ between the two habitats.
Reduced growth rates have a strong potential to feed back to population
dynamics in fishes. When food or access to food is limited, growth rates of
fish tend to be low, and mortality rates tend to be high and negatively size
selective (reviewed by Sogard,
1997). Houde
(1987) termed this the stage
duration hypothesis. Reduction in juvenile growth rate is coupled with
increased mortality rates because fish remain in vulnerable size classes for
longer periods of time. Although we were unable to detect differences in
mortality rates between patch and continuous reef habitats with short-term
monitoring, such differences may have become apparent had we sampled longer.
Indeed, such effects are common in fishes (e.g.,
Blom et al., 1994;
Holtby et al., 1990;
Levin et al., 1997;
Post and Prankevicius,
1987).
Patch and continuous reef habitats vary in many aspects other than the dispersion of coral habitat for juvenile three-spot damselfish. At our study site, patch reefs occurred in deeper water (4-5 m) than the continuous back reef (1.5-2 m). Thus, light levels, flow regime, flux of planktonic food, species and densities of potential competitors, species of corals, and other factors may have differed between the two habitats. However, when we experimentally created standard habitats in a single location that varied only in the distance between habitat patches (which reflected the nearest neighbor distance in the two natural habitats), our results were similar to those from natural habitat. On SHUs separated by 0.5 m (average nearest neighbor distance on patch reefs), the number of agonistic interactions was greater, while foraging and growth rates were lower than on SHUs separated by 1.0 m (average nearest neighbor distance on continuous reef). Thus, it is likely that differences in behavior and growth between the two natural habitats were largely the result of differences in patch structure. Resource levels clearly affect foraging rates; however, the similarity of SHU results that were conducted in Jamaica to those of St. Croix strengthens the argument that the spatial patterning of the habitat is important.
IFD and IDD models provide a framework for linking behavior to population
processes. They are clearly useful when evaluating reef fishes, but there are
problems with the application of both models to such species. Both
distributions assume that animals can correctly assess the suitability of a
patch. The IFD also assumes that new settlers are free to enter a patch
(Fretwell, 1972). The
territoriality and high rates of aggression displayed by three-spot damselfish
suggest that new settlers may not be free to enter a patch, indicating that an
IDD may be more appropriate. However, even though three-spot damselfish are
highly territorial and aggressive, the IDD may not be appropriate for
settlement of juvenile fishes. Because settlement generally occurs at night
when diurnally active residents, including three-spot damselfish, are
inactive, residents may not have the opportunity to prevent settlement into
the patch (e.g., Doherty,
1983; Jones, 1984;
Tolimieri, 1995; but see
Sweatman, 1985) Moreover, most
workers have failed to detect postsettlement movement for coral reef fish,
especially in damselfish (Doherty,
1983; Forrester,
1990,
1995; Jones,
1987a,
b,
1988,
1990;
Tolimieri, 1995), suggesting
that these new settlers may not be forced out by residents after settlement
has occurred.
IFD models predict that fish should leave patch reefs where interference is
high for other habitats where the density of potential competitors relative to
available food resources is lower. However, juvenile three-spot damselfish
remained on patch reefs despite experiencing lower growth rates than those on
continuous reef. Higher than predicted use of poor patches has been observed
in number of different taxa (Kohlmann and
Risenhoover, 1997; Messier et
al., 1990; Tregenza et al.,
1996). There are several reasons to explain why juvenile
three-spots overused patch reef habitat, and these fall into a number of
traditionally proposed categories: (1) inability to correctly assess patches,
(2) high cost of movement among patches, (3) factors related to perceptual
ability or encounter rate, and (4) unquantified costs or benefits, which in
our study would most likely be early, postsettlement mortality or survivorship
over a longer period than the duration of our experiments.
A likely explanation for the patterns we observed combines points one and
two. Settling individuals may not be capable of correctly (or completely)
assessing patch quality in terms of conspecific density, and the costs of
postsettlement movement may be high enough to prevent redistribution of
juveniles after settlement. Many reef fishes respond to the presence of
conspecifics presumably because the presence of conspecifics indicates
high-quality habitat (Booth,
1992; Sweatman,
1985; Tolimieri,
1998); however, settling three-spot damselfish do not appear to
respond to conspecifics (Tolimieri,
1995). Even when fish respond to conspecifics, settlers may still
have only partial information on the quality of the patch. Settlement occurs
at night when most diurnal residents have retreated into the reef and are
inactive. Although settlers may be able to use olfactory cues to detect
conspecifics (Sweatman, 1988),
this may only provide them with information on presence or absence of
conspecifics, not density. Once fish settle, patch movement, especially in
damselfish, is minimal (Doherty,
1983; Forrester,
1990,
1995; Jones,
1987a,
b,
1988,
1990; Tolimieri,
1995). In the present study,
we recorded no cases of juvenile three-spots moving among patches, although
they are known to move to different substrata upon reaching maturity
(Williams, 1978). Lack of
movement after settlement alone should not exclude IFD because fish are free
to choose among patches during settlement. However, the inability to
completely assess patch quality at the time of settlement in concert with lack
of movement among patches after settlement may produce the deviation from
predictions of both IFD and IDD models that we observed.
The ability of individuals to perceive patches may also affect
distribution. Small, dispersed patches will be encountered more often than
less numerous large ones (e.g., Levin,
1993). Many ecological studies have found that rates of patch
colonization are higher in small than in larger patches
(Bell et al., 1987; Eggleston
et al., 1998,
1999;
Keough, 1984;
McNeil and Fairweather, 1993;
Paine and Levin, 1981;
Sogard, 1989;
Sousa, 1984). Thus, three-spot
damselfish may settle in high densities on patch reefs simply because these
habitats are more easily detected. In an experiment similar to the one here,
Levin (1993) found higher
settlement of a temperate wrasse to patchily distributed artificial habitat
units than to clumped ones. His results are relevant here because mortality
rates of these recently settled fish differed such that mortality was higher
on the preferred settlement patch.
The action of factors that we were unable to measure could affect both our
choice between the IFD and IDD and the fit of our data to the appropriate
model. For example early postsettlement predation may have been important.
Many IFD and IDD models, as well as empirical studies examining the
predictions of these models, ignore habitat differences in predation
ostensibly because predation on adult birds, the subject of many of these
studies, is rare (Houston and McNamara,
1997). However, even when predation is rare, it can have
significant effects on individual behavior
(Abrams, 1993;
Houston et al., 1993;
Houston and MacNamara, 1997).
Although differences in predation rates among varying habitats are well known
in fishes (Beukers and Jones,
1997; Hixon, 1991;
Werner and Gilliam, 1984), we
did not detect a difference in mortality rates between patch and continuous
reef habitats. Juvenile three-spot damselfish recruit to reefs when they are
about 10.5 mm, and the fish we observed ranged from 15 to 45 mm. Because
predation on newly settled fish is often size dependent
(Levin et al., 1997;
Sogard, 1997), it is possible
that predators affected fish smaller than those we used in this study.
Although predation rates are often higher at habitat margins, continuous back
reefs, with their diversity of habitats, may harbor a more diverse and thus
more effective suite of a predators than patch reefs
(Hixon and Beets, 1993;
Hixon and Carr, 1997). For
example, Connell (1996)
reported that mortality rates of Acanthochromis polycanthus were
higher on continuous reef than on patch reef, and this difference was
associated with the distribution of large predators. Consequently, the
apparent preference for lower quality patches by three-spot damselfish may be
incorrect, and they may actually follow IFD or IDD predictions.
Regardless of why the patterns we observed deviated from basic IFD and IDD predictions, our results suggest strong linkages between habitat, behavior, and growth rate. Patches used by juvenile three-spot damselfish varied in their spatial distribution and dispersion. This habitat-level pattern was associated with higher levels of aggression and lower feeding rates in patchy habitat, and this was correlated with a decrease in growth. These results illustrate that behavioral interactions are an integral part of population dynamics and that it is necessary to consider the spatial organization of the environment in both behavioral and ecological investigations.
|
ACKNOWLEDGEMENTS |
|---|
We thank K. Gestring, M. Ganger and students from Northeastern University's East/West program for diving assistance. R. Petrik, F. Michelli, D. Westneat, and R. Zabel provided critical comments on various drafts of this paper. Funding was provided by National Science Foundation (NSF) grants OCE-9018724 to P.F.S. and OCE-9102027 to P.S.L. and P.F.S. Supplemental support was provided by NSF grant DEB-9610353 to P.S.L. and J. A. Coyer. This is contribution number 17 from PISCO, the Partnership for Interdisciplinary Studies of Coastal Oceans: A Long-Term Ecological Consortium, funded by the David and Lucile Packard Foundation.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abrams PA, 1989. Decreasing functional responses as a result of adaptive consumer behavior. Evol Ecol 3: 95-114.
Abrams PA, 1992. Ecological effects of the interaction of adaptive foraging behaviors in food webs. Am Nat 140: 573-600.
Abrams PA, 1993. Optimal traits when there are several
costs: the interaction of mortality and energy costs in determining foraging
behavior. Behav Ecol 4:
246-253.
Altman J, 1974. Observational study of behavior: sampling method. Behaviour 49: 227-267.[Medline]
Anholt BR, 1990. An experimental separation of interference and exploitative competition in a larval damselfly. Ecology 76: 2235-2239.
Anholt BR, 1997. How should we test for the role of behaviour in population dynamics. Evol Ecol 11: 633-640.
Bell JD, Westoby M, Steffe AS, 1987. Fish larvae settling in seagrass beds of different leaf density? J Exp Mar Biol Ecol 111: 133-144.
Beukers JS, Jones JP, 1997. Habitat complexity modifies the impact of piscivores on a coral reef fish population. Oecologia 114: 50-59.
Blom GT, Sv
sand T, Jøstad
KE, Otter H, Paulson OI, Holm JC,
1994. Comparative survival and growth of two strains of Atlantic
cod (Gadus morhua) through the early life stages in a marine pond.
Can J Fish Aquat Sci 51:
1012-1023.
Boccia ML, Landenslager M, Reite M, 1988. Food distribution, dominance and aggressive behavior in bonnet macaques. Am J Primatol 16: 123-130.
Booth DJ, 1992. Larval settlement patterns and preferences by domino damselfish Dascyllus albisella Gill. J Exp Mar Biol Ecol 155: 85-104.
Booth DJ, Wellington G, 1998. Settlement preferences in coral-reef fishes: effects on patterns of adult and juvenile distributions, individual fitness and population structure. Aust J Ecol 23: 274-279.
Brown JL, 1975. The evolution of behavior. Toronto: Norton.
Chesson P, Rosenzweig M, 1991. Behavior, heterogeneity, and the dynamics of interacting species. Ecology 72: 1187-1195.
Connell SD, 1996. Variations in mortality of a coral-reef fish: links with predator abundance. Mar Biol 126: 347-352.
Doherty PJ, 1983. Tropical territorial damselfishes: is density limited by aggression or recruitment? Ecology 64: 176-190.
Dunbar RIM, 1985. Population consequence of social structure. In: Behavioural ecology: ecological consequences of adaptive behaviour (Sibly RM, Smith RH, eds). Oxford: Blackwell; 507-520.
Dye C, 1984. Competition among larval Aedes aegypti: the role of interference. Ecol Entomol 9: 355-357.
Eggleston DB, Elis WE, Etherington LL, Dahlgren CP, Posey MH, 1999. Organism responses to habitat fragmentation and diversity: habitat colonization by estuarine macrofauna. J Exp Mar Biol Ecol 236: 107-132.
Eggleston DB, Etherington LL, Elis WE, 1998. Organism response to habitat patchiness: species and habitat-dependent recruitment of decapod crustaceans. J Exp Mar Biol Ecol 223: 111-132.
Emery AR, 1973. Comparative ecology and functional osteology of fourteen species of damselfish (Pices: Pomacentridae) at Alligator Reef, Florida Keys. Bull Mar Sci 23: 649-770.
Ens BJ, Goss-Custard JD, 1984. Interference among oystercatchers, Haematopus ostralegus, feeding on mussels, Mytilus edulis, on the Exe estuary. J Anim Ecol 53: 217-231.
Forrester GF, 1990. Factors influencing the juvenile demography of a coral reef fish. Ecology 71: 1666-1681.
Forrester GE, 1995. Strong density-dependent survival and recruitment regulate the abundance of a coral reef fish. Oecologia 103: 275-282.
Fretwell SD, 1972. Populations in a seasonal environment. Princeton, New Jersey: Princeton University Press.
Fretwell SD, Lucas HL, 1970. On territorial behavior and other factors influencing habitat distribution in birds. Acta Biotheor 19: 16-19.
Fryxell JM, Lundberg P, 1998. Individual behavior and community dynamics. London: Chapman and Hall.
Gersani M, Abramsky Z, Falik O, 1998. Density-dependent habitat selection in plants. Evol Ecol 12: 223-234.
Gilliam JF, 1987. Individual behavior and population dynamics (book review). Ecology 68: 456-457.
Goss-Custard JD, Sutherland WJ, 1997. Individual behaviour, populations and conservation. In: Behavioural ecology: an evolutionary approach (Krebs JR, Davies NB, eds). London: Blackwell Science; 373-398.
Hassell MP, May RM, 1985. From individual behaviour to population dynamics. In: Behavioural ecology (Sibly RM, Smith RH, eds). Oxford: Blackwell; 3-32.
Hixon MA, 1991. Predation as a process structuring coral reef fish communities. In: The ecology of fishes on coral reefs (Sale PF, ed). San Diego, California: Academic Press; 475-508.
Hixon MA, Beets JP, 1993. Predation, prey refuges, and the structure of coral-reef fish communities. Ecol Monogr 63: 77-101.
Hixon MA, Carr MH, 1997. Synergistic predation,
density dependence, and population regulation in marine fish.
Science 277:
946-949.
Holt RD, 1983. Optimal foraging and the form of the predator isocline. Am Nat 122: 521-541.
Holtby LN, Andersen BC, Kadowaku RK, 1990. Importance of smolt size and early ocean growth to interannual variability in marine survival of coho salmon (Oncorhynchus kisutch). Can J Fish Aquat Sci 47: 2181-2194.
Houde ED, 1987. Fish early life dynamics and recruitment variability. Am Fish Soc Symp 2: 17-29.
Houston AI, McNamara JM, 1997. Patch choice and population size. Evol Ecol 11: 703-722.
Houston AI, McNamara JM, Hutchinson JMC, 1993. General results concerning the trade-off between gaining energy and avoiding predation. Phil Trans R Soc Lond B 341: 375-397.
Hugie DM, Grand TC, 1998. Movement between patches, unequal competitors and the ideal free distribution. Evol Ecol 12: 1-19.
Itzkowitz M, 1977. Spatial organization of the Jamaican damselfish community. J Exp Mar Biol Ecol 28: 217-241.
Itzkowitz M, Makie D, 1986. Habitat structure and reproductive success in the beaugregory damselfish. J Exp Mar Biol Ecol 97: 305-312.
Ives AR, Dobson AP, 1987. Antipredator behavior and the population dynamics of simple predator-prey systems. Am Nat 130: 431-447.
Jones GP, 1984. The influence of habitat and behavioural interactions on the local distribution of the wrasse, Pseudolabrus celidotus. Anim. Behav. 10: 43-58.
Jones GP, 1987a. Competitive interactions among adults and juveniles in a coral reef fish. Ecology 68: 1534-1547.
Jones GP, 1987b. Some interactions between residents and recruits in two coral reef fishes. J Exp Mar Biol Ecol 114: 169-182.
Jones GP, 1988. Experimental evaluation of the effects of habitat structure and competitive interactions on the juveniles of two coral reef fishes. J Exp Mar Biol Ecol 123: 115-126.
Jones GP, 1990. The importance of recruitment to the dynamics of a coral reef fish population. Ecology 71: 1691-1698.
Kacelnik A, Krebs JR, Bernstein C, 1992. The ideal free distribution and predator-prey populations. Trends Ecol Evol 7: 50-55.
Kaufman LS, 1977. The threespot damselfish: effects on benthic biota of Caribbean coral reefs. Proc Int Coral Reef Symp 3(1): 559-564.
Keough MJ, 1984. Effects of patch size on the abundance of sessile marine invertebrates. Ecology 65: 423-437.
Kohlmann SG, Risenhoover KL, 1997. White-tailed deer in a patchy environment: a test of the ideal-free-distribution theory. J Mammal 78: 1261-1272.
Kramer DL, Rangeley RW, Chapman LJ, 1997. Habitat selection: patterns of spatial distribution from behavioural decisions. In: Behavioural ecology of teleost fishes (Godin JJ, ed). Oxford: Oxford University Press; 37-80.
Levin PS, 1993. Habitat structure, conspecific presence and spatial variation in the recruitment of a temperate reef fish. Oecologia 94: 176-185.
Levin P, Petrik R, Malone J, 1997. Interactive effects of habitat selection, food supply and predation on recruitment of an estuarine fish. Oecologia 112: 55-63.
Malone JC, Forrester GE, Steele MA, 1999. Effects of subcutaneous microtags on the growth, survival, and vulnerability to predation of small reef fishes. J Exp Mar Biol Ecol 237: 243-253.
Mangel M, Clark CW, 1988. Dynamic modeling in behavioral ecology. Princeton, New Jersey: Princeton University Press.
McNeil SE, Fairweather PG, 1993. Single large or several small marine reserves? An experimental approach with seagrass fauna. J Biogeogr 20: 429-440.
Messier F, Virgil JA, Marinelli L, 1990. Density-dependent habitat selection in muskrats: a test of the ideal free distribution model. Oecologia 84: 380-385.
Milinski M, Parker GA, 1991. Competition for resources. In: Behavioural ecology (Krebs, JR, Davies NB, eds). Oxford: Blackwell; 137-168.
Morris WF, Wiser SD, Klepetka B, 1992. Causes and consequences of spatial aggregation in the phytophagous beetle Altica tombacina. J Anim Ecol 61: 49-58.
Myrberg AA, Thresher RE, 1974. Interspecific aggression and its relevance to the concept of territoriality in reef fishes. Am Zool 14: 81-96.
Paine RT, Levin SA, 1981. Intertidal lanscapes: disturbances and the dynamics of pattern. Ecol Monogr 51: 145-178.
Post JR, Prankevicius AB, 1987. Size-selective mortality in young-of-the-year yellow perch (Perca flavescens:) evidence from otolith microstructure. Can J Fish Aquat Sci 44: 1840-1847.
Randall JE, 1967. Food habits of the fishes of the West Indies. Stud Trop Oceanogr 5: 665-874.
Real LA, 1992. Introduction to the symposium. Am Nat 140: S1-S4.
Robertson DR, 1992. Patterns of lunar settlement and early recruitment in Caribbean reef fishes at settlement. Mar Biol 114: 527-537.
Sale PF, 1972. Influence of corals in the dispersion of the pomacentrid fish, Dascyllus aruanus. Ecology 53: 741-744.
Sih A, 1984. The behavioral race between predators and prey. Am Nat 123: 143-150.
Smith RH, Sibly R, 1985. Behavioural ecology and population dynamics: towards a synthesis. In: Behavioural ecology: ecological consequences of adaptive behaviour (Sibly RM, Smith RH, eds). Oxford: Blackwell; 577-592.
Sogard SM, 1989. Colonization of artificial seagrass by fishes and decapod crustaceans—importance of proximity to natural eelgrass. J Exp Mar Biol Ecol 133: 15-37.
Sogard CM, 1997. Size-selective mortality in the juvenile stage of teleost fishes: a review. Bull Mar Sci 60: 1129-1157.
Sousa WP, 1984. Intertidal mosaics: patch size, propagule availability, and spatially variable patterns in succession. Ecology 65: 1918-1935.
Sutherland WJ, 1996. From individual behaviour to population ecology. Oxford: Blackwell.
Sweatman H, 1988. Field evidence that settling coral reef fish larvae detect resident fishes using dissolved chemical cues. J Exp Mar Biol Ecol 124: 163-174.
Sweatman HPA, 1985. The influence of adults of some coral reef fishes on larval recruitment. Ecol Monogr 55: 469-485.
Thresher RE, 1976. Field analysis of the territoriality of the threespot damselfish, Eupomacentrus planifrons (Pomacentridae). Copeia 1976: 266-276.
Tolimieri N, 1995. Effects of microhabitat characteristics on the settlement and recruitment of a coral reef fish at two spatial scales. Oecologia 102: 52-63.
Tolimieri N, 1998. Effects of substrata, resident conspecifics and damselfish on the settlement and recruitment of the stoplight parrotfish, Sparisoma viride. Environ Biol Fish 53: 393-404.
Tregenza T, 1994. Common misconceptions in applying the ideal free distribution. Anim Behav 47: 485-487.
Tregenza T, Thompson DJ, Parker GA, 1996. Interference
and the ideal free distribution: oviposition in a parasitoid wasp.
Behav Ecol 7:
387-394.
Werner EE, Gilliam JF, 1984. The ontogenetic niche and species interactions in size-structured populations. Annu Rev Ecol Syst 15: 393-425.[Web of Science]
Wilkinson L, Blank G, Gruber C, 1996. Desktop data analysis with SYSTAT. Upper Saddle River, New Jersey: Prentice Hall.
Williams AH, 1978. Ecology of threespot damselfish: social organization, age structure, and population stability. J Exp Mar Biol Ecol 34: 197-213.
Wilson EO, 1975. Sociobiology. Cambridge: Harvard University Press.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Wall and J. Herler Postsettlement movement patterns and homing in a coral-associated fish Behav. Ecol., January 1, 2009; 20(1): 87 - 95. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Zimmerman, W. S. LaHaye, and R. J. Gutierrez Empirical support for a despotic distribution in a California spotted owl population Behav. Ecol., May 1, 2003; 14(3): 433 - 437. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||