Mar Biol (2008) 153:755­768
DOI 10.1007/s00227-007-0844-4
R E S E A R C H A R T I C L E
Long-term changes in coral colony size distributions on Kenyan
reefs under different management regimes and across the 1998
bleaching event
T. R. McClanahan Æ M. Ateweberhan Æ
J. Omukoto
Received: 20 June 2007 / Accepted: 12 October 2007 / Published online: 15 November 2007
Ó Springer-Verlag 2007
Abstract
Colony size is an important life-history char-
distributions were significantly skewed and had right-tailed
acteristic of corals and changes in colony size will have
distributions. After 1998, the right-tailed distributions of
significant effects on coral populations. This study sum-
Acropora, Hydnophora, and Montipora were significantly
marizes *21,000 haphazard colony size measurements of
reduced. Most taxa had peaky distributions and only
26 common coral taxa (mostly coral genera) collected
Acropora experienced a statistically significant change
annually between 1992 and 2006 in seven Kenyan reef
from peaky to flat. The mean sizes of taxa were not related
lagoons. There was a major coral bleaching and mortality
to their mortality across 1998, which indicates that the size
event in early 1998 and all seven reefs were affected. The
effect was within rather than between taxa. Astreopora and
seven
locations
include
two
long-protected
Marine
Platygyra were well-sampled taxa that did not show an
National Parks (Malindi and Watamu), one relatively
effect of management, but had reduced median sizes across
recently established park (Mombasa), and four unprotected
1998. Consequently, no taxa were tolerant of both fishing
locations (Vipingo, Kanamai, Ras Iwatine, and Diani).
and bleaching disturbances and the combined effect was to
They span about 150 km and represent three distinct fish-
reduce the size of all corals.
ery management regimes: old protected (OP), newly
protected (NP), and unprotected (UP). Seventeen taxa had
statistically significant different sizes for comparisons of
the management regimes, with only one genus, Pavona,
Introduction
having larger sizes in the unprotected reefs. The size of
eight coral genera showed a significant time and manage-
Body size and, in the case of corals and other colonial
ment interaction, and size frequency differences that
invertebrates, colony size is an important life-history trait
existed in management areas prior to 1998 were further
that can reflect taxonomic adaptations and responses to
increased after the bleaching event. Time alone was a
evolutionary and environmental stress (LaBarbera 1989;
significant factor for eleven genera, and in all cases colo-
Brown 1995; Karlson 1986, 1988; Belgrano and Brown
nies were smaller after 1998. For most taxa, colony size
2002). Size distributions of corals are often used for
describing population processes and environmental effects
because coral growth is often weakly related to size and
age (Hughes 1984; Hughes and Connell 1987; Bak and
Communicated by J.P. Grassle.
Meesters 1998; Ruesink 1997; Fong and Glynn 1998,
2001). The number of sexually mature polyps in coral
T. R. McClanahan (&)
colonies determines their fecundity, and human-induced or
Wildlife Conservation Society, Marine Programs,
Bronx, NY 10460, USA
natural disturbances that affect colony size are expected to
e-mail: tmcclanahan@wcs.org
have consequences for their reproduction and population
dynamics (Hall and Hughes 1996). Corals experience
M. Ateweberhan Á J. Omukoto
partial mortality due to both environmental and human
Coral Reef Conservation Project, P.O. Box 99470,
80107 Mombasa, Kenya
disturbances (Meesters et al. 1996, 1997a; Nugues and
123
756
Mar Biol (2008) 153:755­768
Roberts 2003a, b; Wielgus et al. 2004). The colonial life
coral reefs of different fishing levels. The corals were
form and partial mortality pattern may provide corals with
studied over a high-mortality-bleaching event in 1998, and
the ability to respond rapidly to environmental and human
low-mortality events in 1994, 2002, and 2005 (McClana-
disturbances, which is in turn reflected in size distribution
han et al. 2001, 2007a). In comparison to global surveys,
patterns.
coral colonies on Kenyan lagoonal reefs are generally
Size distributions in corals mostly reflect differences
small to medium-sized (*10 to 50 cm in one linear
among taxa but within taxa there are also size patterns that
dimension) (McClanahan and Mutere 1994; McClanahan
are expected to result from growth differences in relation to
et al. 2001) and with moderate recruitment rates (*50 m-2
community structure and different disturbance histories
recruits with linear dimensions of 0.5­5 cm, Glassom et al.
(Meesters et al. 2001). For example, a comparison of
2004; McClanahan et al. 2005). We tested the hypotheses
urbanized versus control sites in the Netherlands Antilles
that the disturbances changed distributions towards domi-
found that environments with degraded water quality had
nance by larger colonies and more normal skew and
fewer small colonies, lower variance, and more centralized
kurtosis. Variations and responses in colony size occur
distributions (Meesters et al. 2001) and this led to a prop-
both below and above the species level and we focused
osition that climate change would increasingly produce
mainly on genus level differences. Size based studies,
coral colonies and communities with similar properties
irrespective of taxonomic resolution are becoming com-
(Bak and Meesters 1999). The implicated causative factors
mon in ecosystem level approaches, particularly where
in these Caribbean studies were increased sedimentation,
species-level taxonomy is difficult or unresolved, which is
turbidity, and elevated nutrient concentrations, which
the case for corals in east Africa.
appeared to reduce coral recruitment and the abundance of
small corals, and produced low variance and more cen-
tralized size distributions. These environmental conditions
Materials and methods
are expected to increase with climate change (McClanahan
2002) and the responses may be generic to reef degrada-
Study sites
tion. Caribbean reefs are, however, unusual in having low
recruitment rates compared to Indian Ocean and other reefs
Coral size data were collected over 15 years, from 1992 to
(Glassom et al. 2004), and this could result in different
2006, as part of annual monitoring program of seven
responses to climate change--reefs with high recruitment
Kenyan reef lagoons in three management categories: two
having smaller and reefs with low recruitment having lar-
old protected (Malindi and Watamu Marine National Parks,
ger colony sizes. Nevertheless, it also remains to be tested
legally established in 1968 but closed to fishing in the early
whether or not these size-distribution patterns are specific
1970s), one newly protected (Mombasa Marine National
to environmental disturbances or regions, and if this will
Park, legally established in 1987 but closed to fishing in
produce different responses.
1991) and four unprotected sites, open to unregulated
Coral reefs face many threats from multiple sources of
fishing (Vipingo, Kanamai, Ras Iwatine, and Diani)
disturbance (Wilson et al. 2006). In the Indo-Pacific, the
(Fig. 1). The distance spanned by these locations (Malindi
two major and frequently cited human influences on coral
to Diani) is *150 km and Mombasa and Ras Iwatine are
reefs are fishing and coral bleaching (Donohue et al. 2001;
separated by *5 km. A total of three sites were monitored
McClanahan 2002; Cros and McClanahan 2003; Rodgers
in the two old protected, two in the newly protected, and
and Cox 2003). Fishing may affect coral size through direct
seven in the four unprotected reefs. The closed areas cover
physical destruction or by influencing competitive pro-
*5% of Kenya's near shore areas and the individual parks
cesses that may reduce coral growth, such as competitive
of Mombasa and Watamu parks cover similar areas
interactions with algae (McCook et al. 2001). Similarly,
(*6 km2) and the Malindi Park is *10 km2. The unpro-
bleaching can cause both partial and total mortality and
tected reefs have fishing pressures of [4 fishers km-2
there is evidence for size-dependent responses to bleaching
(McClanahan and Mangi 2001) and a considerably reduced
where recruits (Mumby 1999) and small coral colonies
abundance and biomass of fish (McClanahan et al. 2007b).
survive better than larger colonies (Loya et al. 2001;
The sites are all located in shallow back reef lagoons
Nakamura and van Woesik 2001; Bena and van Woesik
(\3 m at low tide). The tidal range at spring tide is 4 m.
2002; Shenkar et al. 2005). Consequently, shifts in reefs
Stony and soft corals dominate lagoons, and these and
frequently disturbed by fishing and bleaching may be
rubble are interspersed with sand and seagrass. The seven
towards smaller taxa and colony sizes.
study locations are all within one zoogeographic region
Here, we examine the changes in the size distribution of
(Veron 2000) and have similar community structure
26 common coral taxa over a 15-year period on Kenyan
(McClanahan et al. 2007a).
123

Mar Biol (2008) 153:755­768
757
by their different growth forms. Surveys of coral species
40o E
in Kenya, such as in the Malindi­Watamu marine reserve,
found 45 genera and 113 species, which suggests a fairly
Malindi
small species/genus ratio (Lemmens 1993). The most
speciose genus, Acropora, contained 14 species and
Watamu
Pavona (8 species), Montipora (6 species), and Porites (6
species) were the other speciose genera. There is often
a single dominant species per genus in each study site
(McClanahan and Mutere 1994). Acropora spp. are often
difficult to distinguish in the field and the genus-level
Vipingo
resolution used here for this and other speciose genera
Kanamai
4oS
constitutes only a rough estimate of species-level patterns.
Transects were laid haphazardly and draped to follow
Mombasa
the bottom contour, hence, the length measurements used
Ras lwatine
to estimate size was as likely to over- as under-estimate
colony size for any particular size, substratum aspect, or
dimension. When the distance between living coral tissues
Diani
of the same colony exceeded 3 cm, a new measurement
4o30'
was made and recorded as a new coral size measurement.
For distances [3 cm it was not possible to objectively
distinguish living coral that might have been separated by
partial mortality or from larval recruitment and growth on
dead coral. Corals [3 cm are larger than the stage at which
INDIAN
0CEAN
they are usually considered recruits (Mumby et al. 2007).
Consequently, the size measurements presented here are
5oS
estimates of haphazardly measured lengths of continuous
Ecological study site
live post-recruitment coral tissue cover. They are not an
Marine park
estimate of a total colony size connected by a shared
skeleton, which would require making efforts to include
40o
patchy remnants of living tissue into a single colony size
Fig. 1 Map of seven study locations in southern Kenya, including
measurement. A total of 20,855 coral sizes were made
three National Marine Parks
during the study, of which 9,270 were made before and
11,585 after the 1998 bleaching event (Table 1).
Data collection
Statistical analysis
Size data for all coral taxa were collected using the
line-intercept transect method (McClanahan and Shafir
All coral size data were tested for normality using a
1990). Within each reef 9­18, 10-m long line transects
Shapiro­Wilk W test and for homogeneity of variance by
were surveyed each year in January/February. A major
Levene's test before statistical analyses were done. Like-
bleaching event occurred in early 1998. Data were col-
lihood ratio (L-R) tests were done to test the significance of
lected just before the bleaching occurred (November
taxon, management, time, interaction between management
1997­January 1998) and in August 1998, 5­6 months
and time, and interaction between management and taxon
after the bleaching event, when cumulative coral mortality
on coral sizes (with all the variables entered simulta-
had leveled off (McClanahan et al. 2001). All hard corals
neously). This method is more conservative and makes no
[3 cm intercepting the line transect were identified to
assumptions about the distribution of the data, hence we
genus level and measured along the transect line to the
present the results based on this model. Mombasa is pre-
nearest cm with a flexible tape measure. Corals with
sented as a separate management category because the
branches that shared a base were measured as a single
analysis found significant differences at the three levels of
colony. One genus Porites, was separated into three
management and between Mombasa and the older parks
structural categories, namely branching, massive, and
[L-R v2 (Chi-square) = 1055.8, P \ 0.0001]. Mombasa
Synarea (=Porites rus) due to ease in distinguishing these
sampling was based on two sites in the park and could not
123
758
Mar Biol (2008) 153:755­768
Table 1 Summary of the coral
Taxa
Management and bleaching
Total
length sample sizes by
management categories
OP-Before
OP-After
NP-Before
NP-After
UP-Before
UP-After
Acanthastrea
7
21
7
12
0
5
52
Acropora
868
17
45
31
25
28
1,014
Alveopora
1
3
11
3
9
95
122
Astreopora
46
36
6
17
6
36
147
Cyphastrea
8
15
30
50
16
39
158
Echinopora
144
288
21
30
2
18
503
Favia
38
54
151
205
52
209
709
Favites
52
34
140
213
30
54
523
Fungia
3
20
1
2
10
33
69
Galaxea
275
342
150
209
53
283
1,312
Goniastrea
63
67
10
18
8
8
174
Goniopora
19
24
32
43
43
28
189
Hydnophora
36
27
21
39
11
32
166
Leptastrea
11
4
4
6
4
7
36
Leptoria
13
12
4
19
2
6
56
Millepora
76
36
18
14
200
333
677
Montipora
365
94
163
50
19
5
696
Pavona
23
52
40
114
322
1,166
1,717
Platygyra
50
52
38
92
33
44
309
Pocillopora
129
220
80
64
76
286
855
Porites branching
165
261
746
41
2,049
2,927
6,189
Porites massive
270
408
516
976
764
1,204
4,138
Stylophora
11
20
20
0
505
192
748
OP Old protected = 3 study
Synarea
0
0
16
140
0
3
159
sites, NP newly protected = 2
sites, UP unprotected = 7 sites
Tubipora
1
0
7
3
68
33
112
and bleaching (before and
Turbinaria
0
1
6
11
6
1
25
after); coral genera are listed
Total
2,674
2,108
2,283
2,402
4,313
7,075
20,855
alphabetically
be replicated as a management category as no other new
r
ffiffiffiffiffi!
24
parks were successfully established in Kenya during the
kurtosis ¼ 2 Â
n
study period.
Coral size distributions were dominated by smaller
and
colonies (positively skewed) in all taxa, and we therefore
rffiffiffi!
analyzed for significant differences in size among the
6
skewness ¼ 2 Â
three management categories using the non-parametric
n
Kruskal­Wallis test (Sokal and Rohlf 1995). Distribution
parameters, such as the mean size, standard deviation,
Differences in the kurtosis, skewness, and median for all
median, kurtosis, and skewness were computed for each
genera before and after the early 1998 bleaching event
taxon. Taxa size frequency distributions in each man-
were explored using univariate ANOVA, after tests of
agement category and before and after bleaching years
homogeneity of variance and normality and, where nec-
were compared using the Kolmogorov­Smirnov (KS) two
essary, transformations were done. A regression test for
sample comparison, which is sensitive to differences in
relationships between mean coral size and mortality of
most distribution parameters. Deviations in kurtosis and
each taxon across the 1998-bleaching event was under-
skewness from the normal distribution curve were tested
taken using mortality data from before and 6 months after
(Tabachnick and Fidell 1996). Kurtosis and skewness
the 1998 bleaching event (McClanahan et al. 2001). All
greater than two times the standard error (of the kurtosis/
statistical analyses were done using JMP 5.1 for Mac (Sall
skewness statistic) were considered significantly different
et al. 2001) and SPSS 10.0 for Windows following Sokal
from normal.
and Rohlf (1995).
123
Mar Biol (2008) 153:755­768
759
Results
taxon had significant influence on total coral size (Likeli-
hood ratio tests: taxon: v2 = 969.0, P \ 0.0001; time:
Number of colonies
v2 = 586.9, P \ 0.0001; Time 9 management interaction:
v2 = 139.0, P \ 0.0001; Management 9 taxon interaction:
Most taxa exhibited changes in the number of sampled
v2 = 878.3, P \ 0.0001). A total of 17 taxa showed sig-
colonies across 1998 (Table 1). In terms of relative change,
nificant differences in their colony sizes among the three
increases were observed for Fungia in the old protected,
management categories (Kruskal­Wallis test, P \ 0.05;
Synerea in the newly protected, and for Alveopora, Ast-
Table 2). With the exception of Pavona, all of the statis-
reopora, and Echinopora in the unprotected reefs. Declines
tically significant coral sizes were larger in the protected
were observed for Acropora, Leptastrea, Millepora, Mon-
than the unprotected reefs. In the old protected reefs, the
tipora, and Tubipora in the old protected, Alveopora,
smallest was Alveopora (11.3 ± 0.8 cm) and largest size
Montipora, branching Porites, Stylophora, and Tubipora in
recorded was for Montipora (42.7 ± 2.5 cm). Three gen-
the newly protected, and Montipora, Stylophora, Tubipora,
era, Cyphastrea, Goniopora, and branching Porites were
and Turbinaria in the unprotected reefs.
largest in the newly protected reef. In the newly protected
reef, Pavona was smallest (8.8 ± 0.9 cm) while massive
Colony size
Porites was largest (29.4 ± 0.7 cm). In the unprotected
reefs, Acanthastrea was the smallest (7.6 ± 1.3 cm) and
Taxon, time, and the interactions between time and man-
Acropora was the largest (22.7 ± 2.1 cm). Eight genera
agement and the interaction between management and
were significantly influenced by the interaction between
Table 2 Comparison of colony size for the major coral taxa recorded in old protected (Malindi and Watamu), newly protected (Mombasa) and
unprotected sites (Diani, Ras Iwatine, Kanamai and Vipingo)
Taxa
Old protected
Newly protected
Unprotected sites
Kruskal­Wallis test
Mean
SD
n
Mean
SD
n
Mean
SD
n
Chi-square
P value
Montipora
42.69
53.42
459
21.73
24.11
213
18.79
20.99
24
55.82
\0.0001
Leptastrea
38.47
36.09
15
25.70
29.15
10
11.18
5.74
11
8.08
0.0176
Echinopora
37.47
32.07
432
16.61
12.74
51
18.55
9.13
20
47.90
\0.0001
Hydnophora
36.68
33.18
63
11.25
6.81
60
15.35
10.76
43
57.07
\0.0001
Porites massive
31.11
30.30
678
29.44
25.20
1,492
18.29
17.33
1,968
298.84
\0.0001
Leptoria
29.76
27.39
25
35.13
29.89
23
26.13
14.77
8
0.94
NS
Acropora
29.64
27.81
885
16.80
12.14
76
22.77
15.43
53
37.80
\0.0001
Millepora
27.35
27.78
112
20.53
16.84
32
13.43
9.96
533
43.73
\0.0001
Astreopora
24.45
22.73
82
16.04
11.39
23
20.71
13.02
42
3.59
NS
Platygyra
21.04
14.83
102
16.83
11.14
130
17.52
10.81
77
3.39
NS
Galaxea
20.41
26.41
617
12.86
8.46
359
11.97
7.05
336
63.26
\0.0001
Acanthastrea
19.46
12.43
28
9.32
4.66
19
7.60
2.88
5
12.59
0.0018
Goniastrea
19.45
20.53
130
17.86
13.69
28
13.38
9.96
16
3.05
NS
Cyphastrea
18.78
10.81
23
10.91
6.68
80
18.16
13.26
55
19.67
\0.0001
Pocillopora
17.95
12.39
349
14.82
11.31
144
11.82
7.10
362
57.94
\0.0001
Favia
16.65
14.90
92
9.55
7.94
356
11.95
7.72
261
36.83
\0.0001
Favites
16.53
12.39
86
11.58
9.03
353
12.69
7.85
84
12.52
0.0019
Goniopora
15.07
13.59
43
26.12
22.66
75
10.37
11.04
71
41.55
\0.0001
Stylophora
15.00
8.06
31
22.60
16.81
20
18.71
11.70
697
2.91
NS
Porites branching
14.09
15.99
426
26.34
22.08
787
14.10
12.11
4,976
401.58
\0.0001
Turbinaria
14.00
1
18.71
14.03
17
11.57
3.10
7
0.29
NS
Pavona
12.07
10.67
75
8.76
10.67
154
15.39
12.99
1488
81.29
\0.0001
Alveopora
11.25
1.50
4
12.57
12.11
14
7.62
8.16
104
9.83
0.0073
Fungia
10.13
5.90
23
7.67
4.73
3
9.53
4.04
43
0.61
NS
Tubipora
5.00
1
10.40
6.47
10
13.44
11.90
101
3.19
NS
Synarea
28.78
29.12
156
29.33
11.68
3
0.69
NS
Coral taxa are listed in descending order of their mean lengths in the old parks
123
760
Mar Biol (2008) 153:755­768
time and management and colonies were smaller after
in unprotected reefs. The size frequency distribution of all
1998, with the exception of Millepora in the old protected
colonies combined was significantly different among
and Cyphastrea in the unprotected reefs. Eleven genera
management regimes for comparisons across the 1998
were influenced by time alone, with no effect of manage-
bleaching event (KS comparison P \ 0.0001).
ment, and in all cases colonies were smaller after 1998.
Eight taxa showed significant changes in size structure
in the old protected, 14 in newly protected, and 3 in
unprotected reefs before versus after 1998 (Fig. 2; KS
Colony size frequency distributions
comparison P \ 0.05). Astreopora, Goniopora, and Hydno-
phora showed change in old protected reefs, Acropora,
Both before and after 1998, size frequency distributions
Alveopora, Cyphastrea, Favia, Favites, Montipora, Pocillo-
were dominated by small- to intermediate-sized colonies in
pora, and Synarea in the newly protected; and Stylophora
all taxa with the highest relative proportion of the largest
was the only taxon that changed in the unprotected loca-
size classes in old protected, than newly protected reefs,
tions. Galaxea, Pavona, Platygyra and massive Porites
and least in unprotected reefs (Fig. 2). In each taxon and
changed in both old and newly protected reefs, where as
management category, the largest decrease across 1998
Millepora changed in both newly and unprotected reefs.
occurred mainly in medium-sized colonies and this decline
Branching Porites changed in all three management
resulted in further dominance by the smaller size classes.
categories across 1998. In all the above taxa size structure
The differences in size structure across 1998 were largest
shifted towards smaller size classes across the bleaching
in old protected, followed by newly protected, and smallest
event.
Fig. 2 Cumulative frequency
plots for coral taxa with
significant differences in
distributions as determined by
Kolmogorov­Simirnov (KS)
comparison of distributions.
OP Old protected, NP newly
protected, UP unprotected.
Before and after 1998 indicated
123
Mar Biol (2008) 153:755­768
761
Fig. 2 continued
Before and after 1998, comparisons of size structures
before 1998. For most taxa that had difference both before
between the management categories showed significant
and after 1998, the differences were larger after 1998.
differences in 11 taxa each between old protected-unpro-
Kurtosis was significantly positive in 20 taxa before and
tected and newly protected- unprotected and 12 taxa
in 18 taxa after the bleaching event (Appendix), indicating
between old protected-newly protected reefs (Table 3).
that most of the taxa had peaky or leptokurtic size distri-
More taxa showed differences in distributions after than
butions. Only Acropora showed a statistically significant
123
762
Mar Biol (2008) 153:755­768
Table 3 Non-parametric Kolmogorov­Smirnov (KS) two-sample test comparison of the coral size distribution among the three management
categories (OP old protected, NP newly protected, UP unprotected) before and after bleaching
OP vs. NP
OP vs. UP
NP vs. UP
Taxon
Z
P
Taxon
Z
P
Taxon
Z
P
Before 1998
Acropora
1.926
0.001
Goniopora
1.6
0.012
Millepora
1.36
0.066
Porites branching
2.782
\0.0001
Montipora*
1.7
0.006
Pocillopora
1.43
0.034
Total
3.451
\0.0001
Total
10.5
\0.0001
Porites branching
6.73
\0.0001
Total
6.26
\0.0001
After 1998
Cyphastrea*
1.61
0.011
Alveopora*
1.47
0.026
Acropora*
1.69
0.007
Millepora*
1.83
0.003
Echinopora*
1.72
0.006
Cyphastrea
2.26
\0.0001
Pocillopora
2.25
\0.0001
Hydnopora*
1.98
0.001
Echinopora
1.6
0.012
Total
2.93
\0.0001
Pavona
1.98
0.001
Favia
2.83
\0.0001
Total
5.74
\0.0001
Favites
1.35
0.053
Total
4.15
\0.0001
Before
After
Before
After
Before
After
Z
P
Z
P
Z
P
Z
P
Z
P
Z
P
Both before and after 1998
Acanthastrea 1.34
0.056
1.51
0.02
Galaxea
1.71
0.006
2.1
\0.0001 Goniopora
2.52 \0.0001 2.09 \0.0001
Echinopora
1.46
0.028
2.99 \0.0001 Millepora
1.64
0.01
2.81 \0.0001 Pavona
1.52 0.019
4.29 \0.0001
Favia
1.67
0.008
2.91 \0.0001 Pocillopora
1.74
0.005
2.58 \0.0001 Porites massive 5.89 \0.0001 4.19 \0.0001
Galaxea
1.99
0.001
2.16 \0.0001 Porites branching 1.35
0.05
3.1
\0.0001
Hydnophora 2.49 \0.0001 2.29 \0.0001 Porites massive
4
\0.0001 3.08 \0.0001
Montipora
2.3
\0.0001 2.47 \0.0001
Pavona
1.37
0.048
1.93
0.001
* Sample size \30
change in kurtosis, with a reduction in kurtosis across 1998
(Fig. 3). Although comparisons of mean kurtosis before
and after 1998 were not statistically different for other
genera, kurtosis did change from positive to not signifi-
cantly different from normal for Acanthastrea, Acropora,
Goniastrea, Hydnophora and Leptastrea across 1998.
Alveopora, Leptoria and Synarea were non-significantly
different from normal but became significantly positive
after 1998.
Comparison of skewness indicated that most of the coral
genera were positively skewed or had right-tailed distri-
butions both before (22 taxa) and after (20 taxa) 1998.
Fig. 3 Variation in kurtosis of Acropora, the only taxon that showed
Acanthastrea, Leptastrea, Tubipora, and Turbinaria had
statistically significant difference over the 15-year study
skewnesses that were significantly different from normal
before 1998. Alveopora and Synarea had skewnesses that
were significantly different from normal only after 1998.
displayed a statistically significant increase in skewness,
After 1998, the skewness of many genera was reduced,
with a peak immediately after 1998 followed by some
but Acropora, Hydnophora, and Montipora were the only
return to the pre-1998 condition by 2002 (Fig. 4). Acropora
ones where the reductions were statistically significant
was the only taxon that showed a significant difference for
(ANOVA, P \ 0.05; Appendix; Fig. 4). Montipora showed
both kurtosis and skewness across 1998 (ANOVA,
some recovery in its skewness after 1998 but that was lost
P \ 0.05). It showed little evidence for a recovery in its
again after 2002­2004. Only one taxon, massive Porites,
distribution after 1998. Alveopora, Favites, Galaxea,
123
Mar Biol (2008) 153:755­768
763
Fig. 4 Variation of skewness
over time for taxa that had
statistically significant changes
before and after bleaching
(a, b and c had a decrease while
d had an increase in skewness
after the 1998 bleaching event)
Hydnophora, Pavona, Platygyra, branching Porites and
Stylophora had significant reductions in their median sizes
across 1998 (ANOVA, P \ 0.05; Appendix). The rela-
tionship between the mean coral size and mortality across
1998 was not statistically significant (Fig. 5; P [ 0.05).
Discussion
Our results agree with the general pattern that coral pop-
ulation structure is skewed towards small-sized colonies
(Bak and Meesters 1999; Meesters et al. 2001) but, for
most taxa, was not supportive of the hypothesized increase
in coral size or normality of distributions with fishing and
bleaching disturbances. In most cases in Kenya, mean coral
size and size distributions responded to both fishing pres-
Fig. 5 Scatter-plot of the relationship between mean coral size and
mortality across the 1998-bleaching event based on line transect
sure and bleaching factors by a reduced size and, in fewer
measurements before and after the event (mortality data from
cases, skewness, and in the case of Acropora, kurtosis of
McClanahan et al. 2001)
the coral populations. For a number of taxa, bleaching and
associated mortality appeared to largely compound the
effects of fishing. A reduction in size appears to hold for
their submassive growth form or high skeletal density. The
the majority of affected taxa with only a few exceptions.
regain of skewness in Montipora after 1998 and loss again
Nine taxa had non-significant differences in mean size due
after 2002 was probably due to a disease that devastated
to fishing. In Fungia, Goniastrea, Leptoria, Tubipora,
populations of this genus in 2002 (McClanahan et al.
Turbinaria, Stylophora, and Synarea, the absence of sig-
2004a).
nificance was likely due to small sample size in at least one
In most of the taxa, bleaching resulted in further dom-
of the management categories. Consequently, although the
inance by small-sized colonies. The larger reduction in size
results indicate no effect of these factors, this could be a
in the protected reefs after 1998 resulted in size structures
consequence of low statistical power. Among the well-
that were either similar to or had greater dominance by
sampled taxa, mean sizes of Astreopora and Platygyra
small colonies than in the unprotected reefs. In the case of
were not affected by fishing. Astreopora and Platygyra
Pavona and Pocillopora the relative proportion of small
may have some resistance to fishing disturbances due to
colonies was, however, higher in new than old protected
123
764
Mar Biol (2008) 153:755­768
reefs. In the case of Acropora, the 1998 event equalized the
reefs before the bleaching and low sample size may have
size structures between old and new protected reefs
influenced this result. These observations indicate that
because of the greater mortality in the former.
other factors including species interactions and the local
Pavona and branching Porites were exceptions to the
environment can also influence coral size distributions.
general pattern. Pavona had larger colonies in unprotected
Despite the observation that taxa were smaller in
reefs and displayed only a small reduction in size following
unprotected locations and following a bleaching event,
1998. A study of predation on corals in Kenya found that
there was no relationship between mean colony size and
Pavona was highly susceptible to predation (McClanahan
mortality. There were consistent patterns of bleaching
et al. 2005). Predators such as parrotfish and triggerfish
between taxa in different locations across bleaching events
quickly consumed Pavona transplanted from unprotected
(McClanahan et al. 2004b), but there is a great deal of
to protected reefs, and these fish have been greatly reduced
scatter in the relationship and other aspects of the taxa, such
by fishing in unprotected reefs (McClanahan et al. 2005,
as their growth rate (Gates and Edmunds 1999), morphol-
2007b). Pavona is also one of the taxa most tolerant to
ogy, and other life-history traits (Loya et al. 2001;
warm-water anomalies (McClanahan 2004). Consequently,
McClanahan 2004; McClanahan et al. 2004b) are suspected
it is expected to do relatively well in environments with
to be more important than mean colony size. This suggests
reduced predation and warm-water anomalies that induce
that disturbances are breaking up or eliminating larger
high bleaching mortality in other corals.
colonies irrespective of taxa. Under such conditions, taxa
The combination of management and the 1998 bleach-
that tolerate bleaching or partial mortality and fragmenta-
ing was most pronounced in branching Porites. It was the
tion (Meesters et al. 1997b) are likely to benefit from
only taxon that showed change in size structure in all
disturbances due to human resource use and climate change.
management categories. It was also the only taxon with a
Both bleaching and fishing were observed to cause
higher abundance of smaller colonies in old protected reefs
partial as well as whole-colony mortality (Loya et al. 2001;
that showed a stronger shift towards small colonies after
Baird and Marshall 2002; Rodgers and Cox 2003). Con-
1998. Similarly, it was initially very abundant in the newly
sequently, because of the colonial nature of most corals, the
protected category but the 1998 event equalized the size
effect may often be to break large colonies into smaller
structure between the old and newly protected reefs. Sim-
ones, which is expected to result in reduced size and
ilar to Pavona, branching Porites is also susceptible to
skewness among taxa that experience heavy fishing and
predation (McClanahan et al. 2005). Given that it was
frequent bleaching. The mean size and size distribution of
abundant prior to 1998 in the newly protected reefs, it may
the single polyp, mobile coral Fungia were not affected by
be more tolerant of predation than Pavona. Branching
management regime or bleaching and Fungia actually
Porites had one of the fastest bleaching responses and
increased in number after 1998. Fungia has been reported
highest mortalities in 1998 (McClanahan et al. 2001) and,
to be common in reefs influenced by fishing disturbances;
in combination with high susceptibility to predation, it is
including dynamite fishing (Yap and Gomez 1988;
likely to become rare in protected reefs with frequent
McClanahan et al. 1999) and the life history of this taxon
bleaching-induced mortality.
may provide it with some resilience to human disturbances.
Another difference in the overall trend of decreased size
The observed changes in mean colony size and size
with protection and bleaching was the higher proportion of
distributions differ from previous reports from the Carib-
larger colonies of Acropora, Cyphastrea, Echinopora, and
bean (Bak and Meesters 1998, 1999; Meesters et al. 2001)
Favia in the unprotected than in the newly protected reefs.
and may result from different ecological processes in the
The size structure of Acropora in the newly protected reef
two regions. An alternative explanation could, however, be
remained the same across 1998 and was probably influ-
the different methods, definitions, and measurements of the
enced by unmeasured factors. The difference in Cyphastrea
colonies. Bak and Meesters (1998) reported that their
was caused by a real increase in the unprotected reefs, as its
studied corals shifted towards larger colonies in impacted
size structure remained the same in the newly protected
sites and this was attributable to reduced recruitment and
reefs across 1998. The difference in Favia was caused by
lower survival of small colonies in sites with poor water
the larger reduction in size in the newly protected com-
quality. It is possible those responses to bleaching and
pared to the unprotected reefs across 1998. The difference
fishing were different from those due to elevated nutrients,
in Echinopora was caused by both a reduction in size in the
turbidity and sedimentation, or that differences were due to
newly protected and an increase in the unprotected reefs.
lower rates of coral recruitment in the Caribbean compared
Echinopora had a very low presence in the unprotected
to the Indian and Pacific Oceans (Glassom et al. 2004).
123
Mar Biol (2008) 153:755­768
765
Reduced water quality associated with elevated nutrients
consequences for population dynamics. Partial mortality of
(Wielgus et al. 2004) and sediments (Nugues and Roberts
adult colonies results in reduced size and fragmented col-
2003a) have, however, also caused partial mortality and
onies that will have fewer fertile polyps. Translocation of
might be expected to reduce coral size and shift colonies to
energy resources critical for reproductive processes is
smaller sizes. Conversely, increased nutrients, turbidity,
probably more limited in fragmented and discontinuous
and sedimentation is expected to reduce recruitment of
living tissue. Coral recruitment can decrease or completely
corals and the frequency of small-sized corals due to burial
fail during and after severe ENSO events, due to the direct
beneath sediments or competition with algae (Hughes
effects of extreme temperatures (Glynn et al. 2000). It is
1989; Hughes and Tanner 2000; McCook et al. 2001;
less likely that our observation of positive skewness was
Nugues and Roberts 2003b). Fishing, bleaching, and mor-
caused by increased recruitment but rather by partial
tality of corals have also been associated with the
mortality of larger colonies, as suggested above. Partial
succession and recovery of algae on dead substratum that is
mortality, according to Meesters et al. (2001), results in
expected to retard coral recruitment (Diaz-Pulido and
higher proportion of medium-sized colonies. Our results
McCook 2002; Bellwood et al. 2004; Aronson and Precht
indicate that high proportions of positive skewness and
2006; Mumby et al. 2007). Given that bleaching mortality,
kurtosis was generally maintained and even increased after
fishing, and water quality have all been associated with
bleaching. Exceptions were taxa that were uncommon and,
increased algal succession and cover, there is not strong
therefore, not well sampled.
evidence to suggest that any of these factors would inhibit
coral recruitment more than another.
One difficulty in comparing our results with those of
Conclusions
Bak and Meesters (1998) is that these authors estimated
the size of whole coral colonies, which often include a
This study suggests that fishing and bleaching effects are
mix of living tissue and dead skeleton. In our study, we
expected to reduce the size of living coral colonies. Some
did not attempt to determine colony size independent of
of this reduction is probably due to partial mortality,
the size of continuous live tissue, as it was difficult to
which breaks larger colonies into smaller live coral sur-
determine if there was a previous connection between
faces. Coral reefs with no fishing also have partial
separated living tissues from examination of what are
mortality factors such as predation and tourism damage
often eroded skeletons. This difference in definition of
but, in the case of Kenya, these appear to be small and
what constitutes a colony, a continuous skeleton or live
largely undetectable from the design of this monitoring
tissue, and efforts to measure it could partly explain dif-
program. Bleaching, in addition to reducing coral cover
ferences between the two studies. We suggest that most
and changing taxonomic composition (Loya et al. 2001;
common disturbances on coral reefs will reduce the area
McClanahan et al. 2001) is expected to produce smaller
of continuous live coral tissue, often a result of partial
coral colonies but not necessarily to favor taxa with
mortality. If recruitment also occurs, smaller mean colony
smaller mean colony sizes. Although a few taxa appeared
size will be the result. Continuous disturbances and
to tolerate one of the disturbances, none of the studied
reduced size of corals is expected to reduce reproduction
coral taxa appeared to tolerate or benefit from both fishing
and recruitment from larval stages (Hall and Hughes
and bleaching.
1996). Consequently, in the most disturbed environments,
such as the Caribbean (McClanahan 2002), low recruit-
Acknowledgments
Research was supported by the Wildlife Con-
servation Society through a number of sources; USAID's Program in
ment and preferential survival of small corals (Mumby
Science in Technology, Eppley, McBean, Perretti, Stebbins Founda-
1999) could eventually result in shifts towards large sizes
tions, the Pew Charitable Trust, and the World Bank Coral Reef
of the most disturbance-tolerant taxa. Alternatively, corals
Targeted Research Program on coral bleaching. Fieldwork was
with larger skeletons may appear to persist in disturbed
assisted by a number of collaborators including H. Machano Ali, B.
Kaunda-Arara, R. Arthur, A.T. Kamukuru, R. Moothien-Pillay, S.
environments (Hughes 1989), but could be an artifact of
Mangi, N.A. Muthiga, H. Peters, and M.J. Rodriques. N.A. Muthiga
their slower decay and disappearance, or the difficulty of
read and improved the manuscript. Permission to undertake research
recognizing small and recently dead corals against a
was granted by Kenya's Office of Science and Technology and Kenya
background of coral rubble.
Wildlife Services.
Coral colony size is an important life-history trait and
Appendix
disturbances that affect size are expected to have
123
766
Mar Biol (2008) 153:755­768
and
value
0.0001
P
NS
NS
NS
0.02
NS
NS
NS
\
NS
0.0003
NS
NS
0.003
NS
NS
NS
0.03
0.03
NS
0.003
NS
0.004
NS
NS
kurtosis
value
ANOVA
F
0.76
4.08
0.42
7.48
1.41
4.24
3.56
29.63
0.08
23.61
0.73
0.24
12.49
na
0.61
0.3
0.59
5.62
5.58
4.23
17.68
1.27
12.06
na
1.46
0.02
5.07
6.16
3.28
4.61
2.58
6.54
1.32
1.54
4.05
1.20
5.08
2.81
3.52
0.75
5.54
2.70
0.88
2.22
3.42
3.11
SD
16.81
17.77
26.11
11.42
14.97
15.35
nonsignificant
6.88
8.61
8.50
9.89
8.50
9.72
9.44
9.67
After
Mean
10.81
15.22
14.39
11.56
22.89
10.72
14.11
18.33
12.33
25.79
19.06
16.00
13.50
10.13
15.78
12.00
21.44
14.25
indicate
.
3.94
4.56
1.15
9.24
7.43
4.57
1.63
1.27
2.59
1.18
3.59
8.94
4.56
2.75
4.51
1.80
4.60
2.41
2.19
4.54
2.14
3.92
1.77
SD
39.47
34.82
NS
8.00
9.30
Median
Before
Mean
13.00
20.86
24.17
15.07
29.25
10.00
12.43
13.64
16.17
15.00
18.86
40.63
38.08
14.29
19.21
11.29
19.64
12.86
13.14
17.71
17.14
47.50
12.07
13.25
skewness,
S
S
S
S
After
Sig
N
+
+
+
+
+
+
+
NS
+
+
+
+
N
NS
+
+
+
+
+
+
+
+
+
N
N
and
Before
Sig
+
+
NS
+
+
+
+
+
NS
+
+
+
+
+
NS
+
+
+
+
+
+
+
+
NS
+
+
kurtosis
value
P
0.00
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
0.01
NS
NS
0.04
NS
NS
NS
NS
0.04
NS
NS
positive
value
ANOVA
F
na
27.60
3.49
0.33
0.76
0.02
0.47
1.30
0.03
0.34
0.12
0.60
9.73
na
0.02
0.03
5.58
0.59
0.32
0.89
1.28
5.21
1.10
na
0.35
na
significant
bleaching
SD
1.31
0.74
1.04
0.75
0.78
0.93
1.68
0.68
0.95
1.71
0.70
0.87
0.42
0.33
1.29
1.76
0.96
0.27
0.83
0.31
1.68
1.06
0.47
0.99
0.41
0.29
after
indicate
+
After
Mean
0.28
0.64
1.92
0.71
1.55
1.69
2.98
1.53
0.57
3.54
1.14
1.33
0.84
1.14
0.69
2.80
1.45
2.06
0.97
1.34
3.50
2.92
1.01
1.60
0.54
0.86
and
size,
SD
0.51
0.97
1.08
0.85
0.64
0.66
0.99
1.70
0.91
1.90
0.70
1.40
0.49
1.06
0.66
1.28
0.68
0.47
0.90
0.91
0.76
0.21
0.31
NC
1.43
0.19
before
sample
Skewness
Before
Mean
1.84
2.95
0.53
0.96
1.18
1.64
2.46
2.43
0.45
4.09
1.27
1.82
1.59
1.66
0.82
2.36
2.54
1.92
1.23
1.65
2.67
1.90
1.24
0.28
0.93
1.46
genera
low
S
S
S
S
S
to
coral
After
Sig
N
N
+
NS
+
+
+
+
NS
+
N
+
N
N
+
+
+
+
+
+
+
+
+
+
+
NS
due
all
for
Before
Sig
+
+
NS
NS
+
+
+
+
NS
+
+
+
+
+
NS
+
+
+
+
+
+
+
+
NS
+
NS
applicable
median
value
P
0.00
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
not
and
significant
of
value
ANOVA
F
na
15.27
1.58
0.02
0.32
0.26
0.84
0.13
0.06
0.02
0.99
1.18
3.97
na
1.60
0.51
3.58
1.31
0.66
0.04
1.43
3.38
0.47
na
1.57
na
analysis
test
skewness
3.35
1.82
5.08
1.08
3.11
5.74
2.94
2.55
2.58
2.44
2.00
0.20
2.25
3.78
1.87
2.43
1.18
2.40
4.56
1.07
1.43
SD
14.21
20.28
15.50
25.63
11.82
statistical
Fidell's
kurtosis,
1.13
0.02
4.30
0.53
2.51
4.11
2.99
0.68
0.80
1.99
0.23
1.17
1.89
2.90
5.09
0.94
1.93
0.81
3.56
1.37
1.46
and
After
Mean
-
13.89
-
18.96
13.08
23.17
13.27
-
-
mean
ANOVA
of
SD
NC
7.99
2.52
3.27
2.16
2.56
6.60
11.50
1.59
18.13
2.46
5.94
1.87
NC
1.84
7.77
4.72
2.05
3.30
5.98
7.33
1.53
1.11
NC
5.27
1.61
na
Tabachnick
4.86
0.57
0.70
1.52
2.83
8.15
8.95
1.18
2.23
4.65
2.33
6.80
0.14
8.15
7.36
3.92
2.14
3.67
4.24
1.54
0.75
1.67
1.30
Kurtosis
Before
Mean
12.70
-
-
22.31
-
10.21
-
Comparison
computed,
following
not
SD
branching
massive
Appendix
Taxa
Acanthastrea
Acropora
Alveopora
Astreopora
Cyphastrea
Echinopora
Favia
Favites
Fungia
Galaxea
Goniastrea
Goniopora
Hydnophora
Leptastrea
Leptoria
Millepora
Montipora
Pavona
Platygyra
Pocillopora
Porites
Porites
Stylophora
Synarea
Tubipora
Turbinaria
NC
skewness
123
Mar Biol (2008) 153:755­768
767
References
Karlson RH (1988) Size-dependent growth in two zoanthid species: a
contrast in clonal strategies. Ecology 69:1219­1232
Aronson RB, Precht WF (2006) Conservation, precaution, and
LaBarbera M (1989) Analyzing body size as a factor in ecology and
Caribbean reefs. Coral Reefs 25:441­450
evolution. Ann Rev Ecol Syst 20:97­117
Baird AH, Marshall PA (2002) Mortality, growth and reproduction in
Lemmens JWTJ (1993) Reef-building corals (Cnidaria: Scleractinia)
scleractinian corals following bleaching on the Great Barrier
from the Watamu marine national reserve, Kenya: an annotated
Reef. Mar Ecol Prog Ser 237:133­141
species list. Zool Med Leiden 67:453­465
Bak RPM, Meesters EH (1998) Coral population structure: the hidden
Loya Y, Sakai K, Yamazato K, Nakano Y, Sambali H, van Woesik R
information of colony size­frequency distributions. Mar Ecol
(2001) Coral bleaching: the winners and the losers. Ecol Lett
Prog Ser 162:301­306
4:122­131
Bak RPM, Meesters EH (1999) Population structure as a response of
McClanahan TR (2002) The near future of coral reefs. Environ
coral communities to global change. Amer Zool 39:56­65
Conserv 29:460­483
Belgrano A, Brown JH (2002) Oceans under the macroscope. Nature
McClanahan TR (2004) The relationship between bleaching and
419:128­129
mortality of common corals. Mar Biol 144:1239­1245
Bellwood DR, Hughes TP, Folke C, Nystrom M (2004) Confronting
McClanahan TR, Mutere JC (1994) Coral and sea urchin assemblage
the coral reef crisis. Nature 429:827­832
structure and interrelationships in Kenyan reef lagoons. Hydro-
Bena C, van Woesik R (2002) The impact of two bleaching events on
biologia 286:109­124
the survival of small coral colonies (Okinawa, Japan). Bull Mar
McClanahan TR, Mangi S (2001) The effect of closed area and beach
Sci 75:115­125
seine exclusion on coral reef fish catches. Fish Manage Ecol
Brown JH (1995) Macroecology. The University of Chicago Press,
8:107­121
Chicago
McClanahan TR, Shafir SH (1990) Causes and consequences of sea
Cros A, McClanahan TR (2003) Coral transplant damage under
urchin abundance and diversity in Kenyan coral reef lagoons.
various management conditions in the Mombasa Marine
Oecologia 83:362­370
National Park, Kenya. West Ind Ocean J Mar Sci 2:127­136
McClanahan TR, Muthiga NA, Kamukuru AT, Machano H, Kiambo
Diaz-Pulido G, McCook LJ (2002) The fate of bleached corals:
RW (1999) The effects of marine parks and fishing on coral reefs
patterns and dynamics of algal recruitment. Mar Ecol Prog Ser
of northern Tanzania. Biol Cons 89:161­182
232:115­128
McClanahan TR, Muthiga NA, Mangi S (2001) Coral and algal
Donohue MJ, Boland RC, Sramek CM, Antonelis GA (2001) Derelict
changes after the 1998 coral bleaching: Interaction with reef
fishing gear in the Northwestern Hawaiian Islands: diving
management and herbivores on Kenyan reefs. Coral Reefs
surveys and debris removal in 1999 confirm threat to coral reef
19:380­391
ecosystems. Mar Pollut Bull 42:1301­1312
McClanahan TR McLaughlin SM, Davy JE, Wilson WH, Peters EC,
Fong P, Glynn PW (1998) A dynamic size-structured population
Price KL, Maina J (2004a) Observations of a new source of coral
model: does disturbance control size structure of a population of
mortality along the Kenyan coast. Hydrobiologia 530/531:469­
the massive coral Gardineroseris panulata in the Eastern
479
Pacific? Mar Biol 130:663­674
McClanahan TR, Baird AH, Marshall PA, Toscano MA (2004b)
Fong P, Glynn PW (2001) Population abundance and size­structure of
Comparing bleaching and mortality responses of hard corals
an eastern tropical Pacific reef coral after the 1997­98 ENSO: a
between southern Kenya and the Great Barrier Reef, Australia.
simulation model predicts field measures. Bull Mar Sci 69:187­
Mar Pollut Bull 48:327­335
202
McClanahan TR, Maina J, Herron-Perez P, Dusek E (2005) Detri-
Gates RD, Edmunds PJ (1999) The physiological mechanisms of
ments to post-bleaching recovery of corals. Coral Reefs 24:230­
acclimatization in tropical reef corals. Am Zool 39:30­43
246
Glassom D, Zakai D, Chadwick-Furman NE (2004) Coral recruit-
McClanahan TR, Ateweberhan M, Sebastian CR, Graham NAJ,
ment: a spatio-temporal analysis along the coastline of Eilat,
Wilson SK, Guillaume MMM, Bruggemann JH (2007a) Western
northern Red Sea. Mar Biol 144:641­651
Indian Ocean coral communities: bleaching responses, and
Glynn PW, Colley SB, Ting JH, Mate JL, Guzman HM (2000) Reef
susceptibility to extinction. Mar Ecol Prog Ser 337:1­13
coral reproduction in the eastern Pacific: Costa Rica, Panama and
McClanahan TR, Graham NAJ, Calnan JM, MacNeil MA (2007b)
Galapagos Islands (Ecuador). IV. Agaricidae, recruitment and
Towards pristine biomass: reef fish recovery in coral reef marine
recovery of Pavona varians and Pavona sp. A. Mar Biol
protected areas in Kenya. Ecol Appl 17:1055­1067
136:785­805
McCook LJ, Jompa J, Diaz-Pulido G (2001) Competition between
Hall VR, Hughes TP (1996) Reproductive strategies of modular
corals and algae on coral reefs: a review of evidence and
organisms: comparative studies of reef-building corals. Ecology
mechanisms. Coral Reefs 19:400­417
77:950­963
Meesters EH, Wesseling I, Bak RPM (1996) Partial mortality in three
Hughes TP (1984) Population dynamics based on individual size
species of reef-building corals and the relation with colony
rather than age: a general model with a reef coral example. Am
morphology. Bull Mar Sci 58:838­852
Nat 123:778­795
Meesters EH, Pauchli W, Bak RPM (1997b) Predicting regener-
Hughes TP (1989) Community structure and diversity of coral reefs:
ation of physical damage on a reef-building coral by
the role of history. Ecology 70:275­279
regeneration capacity and lesion shape. Mar Ecol Prog Ser
Hughes TP, Connell JH (1987) Population dynamics based on size or
146:91­99
age? A reef-coral analysis. Am Nat 129:818­829
Meesters EH, Wesseling I, Bak RPM (1997a) Coral colony tissue
Hughes TP, Tanner JE (2000) Recruitment failure, life histories,
damage in six species of reef-building corals: partial mortality in
and long-term decline of Caribbean corals. Ecology 81:2250­
relation with depth and surface area. J Sea Res 37:131­144
2263
Meesters EH, Hilterman M, Kardinaal E, Keetman M, de Vries M,
Karlson RH (1986) Disturbance, colonial fragmentation, and size-
Bak RPM (2001) Colony size frequency distributions of
dependent life history variation in two coral reef cnidarians. Mar
scleractinian coral populations: spatial and interspecific varia-
Ecol Prog Ser 28:245­249
tion. Mar Ecol Prog Ser 209:43­54
123
768
Mar Biol (2008) 153:755­768
Mumby PJ (1999) Bleaching and hurricane disturbances to popula-
Shenkar N, Fine M, Loya Y (2005) Size matters: bleaching dynamics
tions of coral recruits in Belize. Mar Ecol Prog Ser 190:27­35
of the coral Oculina patagonica. Mar Ecol Prog Ser 294:181­
Mumby PJ, Harborne AR, Williams J, Kappel CV, Brumbaugh DR,
188
Micheli F, Holmes KE, Dahlgren CP, Paris CB, Blackwell PG
Sokal PR, Rohlf FJ (1995) Biometry, 3rd edn. Freeman and Co., San
(2007) Trophic cascade facilitates coral recruitment in a marine
Francisco
reserve. Proc Natl Acad Sci USA 104:8362­8367
Tabachnick BG, Fidell LS (1996) Using multivariate statistics, 3rd
Nakamura T, van Woesik R (2001) Differential survival of corals
edn. Harper Collins, New York
during the 1998 bleaching event is partially explained by water-
Veron J (2000) Corals of the World, 1st edn. Australian Institute of
flow rates and passive diffusion. Mar Ecol Prog Ser 212:301­304
Marine Science, Townsville
Nugues MM, Roberts CM (2003a) Partial mortality in massive reef
Wielgus J, Chadwick-Furman NE, Dubinsky Z (2004) Coral cover
corals as an indicator of sediment stress on coral reefs. Mar Poll
and partial mortality on anthropogenically impacted coral reefs
Bull 46:314­323
at Eilat, northern Red Sea. Mar Pollut Bull 48:248­253
Nugues MM, Roberts CM (2003b) Coral mortality and interaction
Wilson SK, Graham NAJ, Pratchett MS, Jones GP, Polunin NVC
with algae in relation to sedimentation. Coral Reefs 22:507­516
(2006) Multiple disturbances and the global degradation of coral
Rodgers KS, Cox EF (2003) The effects of trampling on Hawaiian
reefs: are reef fishes at risk or resilient? Global Change Biol
corals along a gradient of human use. Biol Conserv 112:383­389
12:2220­2234
Ruesink JL (1997) Coral injury and recovery: matrix models link
Yap HT, Gomez ED (1988) Aspects of benthic recruitment on a
process to pattern. J Exp Mar Biol Ecol 210:187­208
northern Philippine reef. Proc 6th Int Coral Reef Symp 2:279­
Sall J, Lehmaan A, Creighton L (2001) JMP start statistics, 2nd edn.
283
Thomson Learning, Duxbury
123

Document Outline