Partial mortality in massive reef corals as an indicator of sediment stress on coral reefs

Marine Pollution Bulletin 46 (2003) 314–323 www.elsevier.com/locate/marpolbul

Partial mortality in massive reef corals as an indicator of sediment stress on coral reefs Maggy M. Nugues b

a,b,*

, Callum M. Roberts

a

a Environment Department, University of York, York YO10 5DD, UK Netherlands Institute for Sea Research (NIOZ), P.O. Box 59, AB Den Burg, Texel 1790, The Netherlands

Accepted 15 October 2002

Abstract Partial mortality and fission on colonies of four common massive coral species were examined at sites differing in their exposure to river sediments in St. Lucia, West Indies. Rates of partial mortality were higher close to the river mouths, where more sediments were deposited, than away from the rivers in two coral species. Frequency of fission showed no significant trend. The percent change in coral cover on reefs from 1995 to 1998 was negatively related to the rate of partial mortality estimated in 1998 in all species. This suggests that partial mortality rates could reflect longer-term temporal changes in coral communities. Similar conclusions could also be reached using a less precise measure and simply recording partial mortality on colonies as <50% and P 50% dead tissue. We conclude that partial mortality in some species of massive reef corals, expressed as the amount of dead tissue per colony, could provide a rapid and effective means of detecting sediment stress on coral reefs. Ó 2002 Elsevier Science Ltd. All rights reserved. Keywords: Bioindicators; Caribbean; Colony size; Coral reefs; Partial mortality; Sediment pollution

1. Introduction There is serious concern that coral reefs are being rapidly degraded by a wide range of human activities (Wilkinson, 1993; Ginsburg, 1994; Birkeland, 1996; Bryant et al., 1998). This perception has stimulated the need to develop methods for assessing reef condition that can be applied easily and rapidly over wide areas and with limited resources (e.g., Alcolado et al., 1994; Dustan, 1994; Risk et al., 1994; Ginsburg et al., 1996; Hodgson, 1999). Coral cover and colony number have traditionally been considered as an essential part of long-term monitoring programs. However, their use as indicators of reef condition in ÔsnapshotÕ surveys has been limited by the erroneous assumption that ÔhealthyÕ reefs always have high coral cover and colony density (Thomason and Roberts, 1992). To date, the possibility of using other aspects of the population biology of corals as indicators of environmental stress has hardly *

Corresponding author. Tel.: +31-222-369530; fax: +31-222319674. E-mail address: [email protected] (M.M. Nugues).

been explored, namely the characteristics of partial mortality and fission in coral colonies. Corals are modular colonial organisms and consequently they have the ability to survive the death of parts of their living tissue (Hughes and Jackson, 1980, 1985; Sebens, 1987; Babcock, 1991). This characteristic is known as partial mortality. When a colony surface is damaged, the surrounding living tissue may regrow and recover the wound through the regeneration of tissue and skeleton (Bak and Steward-van Es, 1980; Bak, 1983; Meesters et al., 1992; Meesters and Bak, 1993). However, if this fails, the trace of partial mortality becomes permanent. Additionally, through partial mortality, coral colonies often subdivide into physiologically separate but genetically identical ÔdaughterÕ colonies, a phenomenon known as fission (Highsmith, 1982; Hughes and Jackson, 1980, 1985). Hughes and Jackson (1985) showed that partial mortality was even greater than whole colony mortality in terms of loss of living tissue by coral populations. A decade later, Meesters et al. (1996, 1997) described partial mortality in various coral species and demonstrated a strong influence of coral colony size, morphology, depth

0025-326X/03/$ - see front matter Ó 2002 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0025-326X(02)00402-2

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323

and species on partial mortality patterns. Colony fission has also been described by earlier workers as being advantageous in many species of corals since it increases chances of long-term colony survival and plays an important role in reproduction and colonisation of space in some species (Hughes and Jackson, 1980, 1985; Highsmith, 1982; Babcock, 1991). Despite their seemingly important role in the structure and dynamics of coral population, few studies (e.g., Guzman et al., 1991; Bythell et al., 1993; Lewis, 1997; Epstein et al., 1999) have quantified partial mortality and fission at reef locations subject to different types and levels of stress. The purpose of this study was to examine how partial mortality and fission in reef corals with massive growth forms vary with sedimentation, and ultimately to assess whether they could be used as indicators of sediment stress on coral reefs. Massive corals are long-lived and highly persistent. Unlike branching corals, they can withstand the impact of recurrent storms common on reefs (Stoddart, 1974; Highsmith, 1982; Woodley et al., 1981). Therefore, we hypothesised that their longevity and their resistance to natural disturbances could make them good subjects with which to detect slow, chronic perturbations of the reef such as those caused by sediment pollution. We investigated two natural gradients of sediment input to coral reefs extending from close to river mouths to nearby headlands (1–2 km away) on the Caribbean island of St. Lucia. Beside the input of river sediments, water circulation and flushing may vary considerably

315

from the inner regions of bays to headlands. Therefore, we also established two sets of control sites at varying distances from the heads of bays, but unexposed to river sediments, to account for natural variability along the gradients (Glasby and Underwood, 1998). The specific objectives of this study were: (1) to describe sedimentation rate and particle size-distribution of the surface sediment along the sediment gradients and control sites (thereafter referred to as Ôsediment-exposedÕ and ÔcontrolÕ reefs, respectively), (2) to examine partial mortality and fission on colonies of four common massive coral species at each locality, and (3) to compare partial mortality and fission with changes in live coral cover over a four-year period to test whether they could reflect temporal changes in total coral cover.

2. Materials and methods 2.1. Study sites This study was conducted on the fringing reefs of the Caribbean island of St. Lucia. We chose four reefs, each extending from bay head to headland (Fig. 1). Two reefs, termed Ôsediment-exposed reefsÕ, were exposed to river sediments from the Soufri
ere and the Anse La Raye/Anse Galet rivers. Both river catchments were settled and farmed. The Soufri
ere river contains 44.1 hm3 (1 hm ¼ 0.1 km) of water while Anse Galet and Anse La Raye rivers are smaller with 12.7 and 37.2 hm3 ,

Fig. 1. Location of St. Lucia (a) and the study sites. The enlargements (b) and (c) show study sites (black arrows), river systems and towns. The two Ôsediment-exposed reefsÕ (Anse La Raye and Soufri
ere) are indicated by continuous lines and the two Ôcontrol reefsÕ (Petit Piton and Gros Piton) by dotted lines. There were three locations in each reef: Ô1Õ at the bay head or ÔnearÕ location, Ô2Õ approximately half way between the bay head and the headland or ÔmidÕ location (0.5–1 km away), and Ô3Õ at the headland or ÔfarÕ location (>1 km away). Site Ô3bÕ is an additional site. Sediments discharged by the Soufri
ere river flow across the northern part of the bay due to predominant southwest currents. Thus the Petit Piton control reef is mostly kept free of river sediment discharge. Asterisks (*) denote sites where live coral cover was not surveyed. Grey areas represent beaches.

316

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323

respectively (Sladek Nowlis et al., 1997). Sediment deposition decreased moving from the heads of bays (close to the river mouths) to the nearby headlands, providing two natural gradients of sediment input. The two other reefs, termed Ôcontrol reefsÕ, were located on the north sides of Petit Piton and Gros Piton (Fig. 1). They had no river at their bay heads and thus were exposed to very low levels of sediment inputs from bay head to headland. Study sites were established at three locations along each of the four reefs: one at the bay head (ÔnearÕ location), one approximately half way between the bay head and the nearby headland (ÔmidÕ location) and one at the headland (ÔfarÕ location). There was an additional site (3b in Fig. 1) along the Soufri
ere reef, making a total of 13 sites. 2.2. Sediment analyses Sedimentation rate was estimated at each site at a depth of 15 m using two sediment traps cemented vertically to the reef and placed less than 2 m apart. Traps were 4 cm diameter PVC cylinders with a height to width ratio of 4, as recommended by Blomqvist and Kofoed (1981) for estimating vertical fluxes of sediment. The mouths of the traps lay approximately 25 cm above the sea bed. Traps were deployed on 11 occasions between August 1997 and December 1998 for periods of 14 days. The trap contents were filtered onto 0.45 lm filter papers and dried to constant weight. Particle size of sediment deposits on the reef was estimated at each site, also at a depth of 15 m in August 1997. Within each site, four samples of the surface sediment were collected in randomly selected reef pockets and sediment-filled channels using a cylindrical PVC corer (100 mm high  40 mm diameter). The samples were dried at 60 °C to constant weight. A 100 g subsample was removed and soaked overnight in 800 ml of fresh water and 40 ml of sodium hexametaphosphate solution (6.2 g l1 ) to disaggregate the particles (Buchanan and Kain, 1971). Samples were then washed to remove the silt, redried and sieved through 850, 500, 250, 125 and 63 lm Endicott screens and a collecting container using an electrically driven reciprocating shaker. After weighing the resulting fractions, the proportion of each fraction was determined. Here only data on the percent of particles less than 125 lm in size (including silt) are presented since this was the fraction differing the most among sites and fine particles are the most harmful to corals (Bak and Elgershuizen, 1976; Fisk, 1981). 2.3. Coral population surveys Partial mortality and fission were surveyed in colonies of Porites astreoides, Siderastrea siderea, Colpophyllia

natans and two sibling species of Montastraea annularis: M. faveolata and M. annularis (Knowlton et al., 1992). All these species form massive skeletons and are abundant on the St. Lucian reefs. They represent 14%, 5%, 19% and 19% (M. faveolata and M. annularis combined) of live coral cover, respectively, averaged across sites (MN, unpublished data). In the analysis, we grouped M. faveolata and M. annularis together since sites tended to be dominated either by one or the other rendering it difficult to obtain an adequate sample size for each species at each site. Coral colonies were surveyed at each site at a depth of 15 m from January to February 1998 using 2 or 4 m wide belt transects varying in length from 5 to 10 m, depending on the density of colonies. Transects were laid parallel to the shore. Computer-generated random numbers were used to determine the distances between each transect; however, major sand channels or monospecific stands of Madracis mirabilis were avoided. Colonies with P 50% of their surface area lying outside of the transect were not recorded. Sufficient transects were laid to obtain a minimum sample of 30 colonies per species at each site. For the purpose of this study, a colony was defined as an autonomous mass of skeleton with living tissue. As such, a colony divided by partial mortality or morphological characteristics (e.g. M. annularis) into separate patches of living tissue, but located on the same mass of skeleton, was considered to be one colony. The projected total surface area of each colony was calculated using the formula for a two-dimensional ellipse p  ðd1 =2Þ  ðd2 =2Þ where d1 was defined as the maximum diameter and d2 the diameter at right angles to the maximum diameter, all measured to the nearest 1 cm. Partial mortality was defined as areas of bare and algal-covered skeleton present on the colony surface (Fig. 2). Every colony was assigned visually into one of ten partial mortality categories (0, 1–9, 10–19, 20– 29; . . . ; 90–99%). Average partial mortality was then estimated from the midpoint of the mortality category for each of the colonies sampled. Whole colony death was not recorded on the basis that it is more frequent in small colonies (Hughes and Jackson, 1985; Babcock, 1991) and completely dead colonies, especially if small, are difficult to detect in the field. From a practical viewpoint, the presence of remaining living tissue also facilitates field identification of individual colonies and species. In addition, the percent of damaged colonies (>0% partial mortality) and the percent of colonies with P 50% partial mortality were calculated to determine whether they could provide a simple and rapid indicator of sediment stress. Frequency of fission was recorded by counting the number of patches of living tissue (daughter colonies) on a parent colony (Fig. 2). Two patches of living tissue on one colony were recorded as one fission, three patches as

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323

317

plots fixed factor. The size classes were <60, 60–<180 and P 180 cm2 for P. astreoides, <150, 150–<500 and P 500cm2 for S. siderea, <1000, 1000–<3500 and P 3500 cm2 for C. natans, and <800 cm2 , 800–<2700 cm2 and P 2700 cm2 for M. faveolata-annularis. These categories were chosen to maintain at least 5 colonies per site in each size class for each species. Site 3b was excluded from the ANOVAs to even up sample sizes (Fig. 1). Because of the low level of replication, marginally significant results with a significance level between 0.05 and 0.1 were noted in all ANOVAs. Statistical assumptions of normality and homoscedasticity were tested and data were transformed when necessary.

2.4. Statistical analyses The sampling design was analogue to a split-plot experiment and the appropriate ANOVA procedure was applied (Cochran and Cox, 1957). The four reefs were treated as plots. Sediment exposure (sediment-exposed reefs vs. control reefs) was the between-plots fixed factor and location (near, mid and far) was the within-plots fixed factor. The effect of sediment exposure (1 df) was tested using reefs within sediment exposure (2 df) as the error term. For the effect of location (2 df) and the interaction between sediment exposure and location (2 df), location by reefs within sediment exposure (4 df) was used as the error term. Partial mortality and fission were closely related to colony size. Therefore, to exclude variations due to differences in colony size, colonies of each species were assigned to three size classes and results were also examined by adding size class as a within-

Patterns of variation in the two sediment analyses between sediment exposure and among locations confirmed the greater influence of terrigenous sediments on the sediment-exposed reefs close to the river mouths. Both analyses were significantly, or marginally significantly, higher on the sediment-exposed reefs and increased with proximity to the bay heads (Fig. 3; Table 1). This rate of increase was greater on the sedimentexposed reefs than on the control reefs for the percent of particles <125 lm in size, as indicated by a marginally significant interaction between sediment exposure and location (Fig. 3b; Table 1). However, this interaction was not significant for sedimentation rate. When adding time (11 levels) as a within-plots factor in the ANOVA, the analysis showed a small, but marginally significant effect of time (F10;20 ¼ 1:98, p ¼ 0:09) and no significant interactions (p > 0:1). The same general patterns emerged for partial mortality in two coral species: S. siderea and M. faveolataannularis (Fig. 4b and d). Both species showed a significant interaction between sediment exposure and

5 4 3 2 1 0 n m f n m f C S a)

Sediments <125µm (%)

two fissions, etc. On M. annularis and M. faveolata, only different patches present on the same column, sheet or plate were considered as fission. To compare partial mortality and fission with change over time in living coral cover, we used data on percent live coral cover available for 11 of the 13 sites studied (Fig. 1). These data were taken at a depth of 15 m in 1995 (December 1994–February 1995) and 1998 (October– November 1998). Live coral cover was visually estimated using randomly placed 1 m2 quadrats (n P 10 per site). The 1998 survey was conducted by MN while the 1995 survey was conducted by Sladek Nowlis. Both observers used the same method and were trained by the same person (CR) and therefore should show minimal differences in coral identification and percent estimates.

3. Results

Sedimentation rate -2 -1 (mg.cm .d )

Fig. 2. In situ photograph of a colony of Colpophyllia natans showing partial mortality at the ÔnearÕ location on the Soufri
ere reef (Fig. 1). The white line had been added to the photograph and represents the interface between living and dead coral. This colony was estimated to have suffered 75% partial mortality. The two separated patches of living tissue (or ÔdaughterÕ colonies) represent one fission event. Quadrat size: 60  40 cm.

25 20 15 10 5 0 b)

n m f C

n m f S

Fig. 3. Patterns of variation in the two sediment measures (means  S:E:) along the two control (C) reefs and the two sedimentexposed (S) reefs in relation to location (n, near; m, mid; f, far). Sedimentation rate values are means of 22 replicates pooled across 11 sampling periods. Particle size data are means of 4 core samples.

318

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323

Table 1 Analysis of variance of the effects of sediment exposure (sediment or S) and location (L) on sedimentation rate and percentage of particles <125 lm in size Source of variation

df

Sedimentation rate

% Sediments <125 lm

MS

F

p

MSa

F

P

Between reefs Sediment Reef (S)

1 2

8.24 0.46

18.14

*

1.78 0.04

46.58

**

Within reefs Location LS L  reef ðSÞ

2 2 4

0.74 0.33 0.15

5.02 2.24

* ns

2.12 1.26 0.22

9.60 5.71

** *

Sedimentation rate was loge transformed and percent of particles <125 lm in size was arcsin square root transformed. Significance levels are indicated by:  < 0:1;  < 0:05; ns, non-significant. a Values have been multiplied by 102 .

a) P. astreoides 0.5 0.4 0.3 0.2 0.1 0.0 b) S. siderea 60 50 40 30 20 10 0 60 50 40 30 20 10 0 60 50 40 30 20 10 0

c) C. natans

Frequency of fission (no. fissions per colonies)

Partial colony mortality (% colony surface area)

location (Table 2). Their rates of partial mortality increased with proximity to the bay heads more strongly on the sediment-exposed reefs than on the control reefs, suggesting that the input of river sediments induced greater tissue mortality in these two species. Rates of partial mortality in S. siderea close to river mouths (ÔnearÕ location on sediment-exposed reefs) were 46  5% on Soufri
ere reef and 52  5% on Anse la Raye/Anse Galet reef, compared with 13  3% and 18  3%, respectively, on the same reefs at the ÔfarÕ location, or maxima of 36  4% at any location on the control reefs (Fig. 4b). In M. faveolata-annularis, rates of partial mortality were 48  5% on Soufri
ere reef and 37  4% on Anse la Raye/Anse Galet reef at the ÔnearÕ location, compared with 23  4% and 7  2%, respectively, on the same reefs at the ÔfarÕ location, or maxima of 18  2% at any location on the control reefs (Fig. 4d). Partial mortality in P. astreoides was not significantly affected by sediment exposure or location (Fig. 4a; Table 2). In C. natans, partial mortality differed marginally among location, but there was no significant interaction between sediment exposure and location (Table 2). This was partially due to large differences between the two sediment-exposed reefs. Only one (Anse la Raye/Anse Galet reef) showed a strong increase in rates of partial mortality in C. natans close to the river mouth (Fig. 4c). Fission showed no significant effect of sediment exposure or location in any of the species studied (Fig. 4; Table 2). When colony size class was added in the ANOVAs (Table 3), partial mortality varied significantly among size class in all species but there was no major change in the significance level of the factors sediment exposure and location and their interaction (only C. natans showed no more significant location effect). Therefore, previous variations in partial mortality were not simply due to differences in colony size. There was no significant 3-way interaction between size class, sediment exposure and location in any of the species studied, thus differences in partial mortality among each sediment exposure

60 50 40 30 20 10 0

4 3 2 1 0 0.8 0.6 0.4 0.2 0.0

d) M. faveolata-annularis 2.5 2.0 1.5 1.0 0.5 0.0 n m f C

n m f S

n m f n m f C S

Fig. 4. Patterns of variation in partial colony mortality and frequency of fission (means  S:E:) for the four species studied along the two control (C) reefs and the two sediment-exposed (S) reefs in relation to location (n, near; m, mid; f, far). Data are means of a minimum of 30 colonies per reef per location.

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323

319

Table 2 Analysis of variance of the effects of sediment exposure (sediment or S) and location (L) on rates of partial colony mortality and frequency of fission for the four studied coral species Parameter

Source of variation

Partial mortality

Between reefs Sediment Reef (S) Within reefs Location LS L  reef ðSÞ

Fission

Between reefs Sediment Reef (S) Within reefs Location LS L  reef ðSÞ

df

P. astreoides

S. siderea

MS

F

p

MS

1 2

0.03 0.12

0.24

ns

0.13 0.40

2 2 4

0.52 0.40 0.43

1.20 0.91

ns ns

1 2

2.08 0.33

6.38

2 2 4

1.20 0.63 0.78

1.54 0.80

C. natans

M. faveolata-annularis

p

MS

F

p

0.34

ns

1.02 0.28

3.67

ns

5.18 5.11 0.32

16.42 16.21

** **

2.35 0.18 0.51

4.64 0.36

ns

1.69 22.20

0.08

ns

0.21 0.28

ns ns

26.80 38.30 41.90

0.64 0.91

ns ns

0.62 0.65 1.10

F

MS

F

p

6.33 5.00

1.27

ns

* ns

3.18 5.16 0.68

4.70 7.64

* **

0.75

ns

85.00 33.00

2.57

ns

0.56 0.59

ns ns

21.70 49.40 14.10

1.54 3.51

ns ns

Partial colony mortality was arcsin square root transformed. Significance levels as in Table 1. MS values have been multiplied by 102 .

Table 3 Analysis of variance of the effects of sediment exposure (sediment or S), location (L) and size class (SC) on rates of partial colony mortality and frequency of fission for the four studied coral species Source of variation

df

P. astreoides MS

Between reefs Sediment Reef (S) Within reefs Location LS L  reef ðSÞ

F

S. siderea p

MS

C. natans

F

p

MS

M. faveolata-annularis F

p

MS

F

p

1 2

0.52 0.27

1.93

ns

0.15 0.30

0.49

ns

5.93 2.93

2.02

ns

15.10 15.50

0.98

ns

2 2 4

1.22 0.94 1.27

0.96 0.74

ns ns

17.90 10.00 1.29

13.88 7.76

** **

10.30 2.66 3.40

3.03 0.78

ns ns

9.94 11.90 1.68

5.91 7.05

* **

Size class SC  S SC  reef ðSÞ

2 2 4

18.00 1.08 0.24

74.91 4.50

** *

20.00 2.09 1.61

12.44 1.30

** ns

20.40 0.81 1.37

14.92 0.59

** ns

4. 78 0.20 0.93

5.16 0.21

* ns

L  SC L  SC  S L  SC  reef ðSÞ

4 4 8

0.68 0.35 0.13

5.33 2.71

** ns

7.05 1.97 2.09

3.37 0.94

* ns

3.24 0.93 1.09

2.97 0.86

* ns

0.68 1.10 0.84

0.81 1.32

ns ns

Partial colony mortality was arcsin square root transformed. Significance levels as in Table 1. MS values have been multiplied by 102 .

and location combination were not significantly higher in larger colony size classes. Three species had a significant, or marginally significant, interaction between colony size and location. This was due to partial mortality being greater at the ÔnearÕ location in the larger size class. SpearmanÕs rank correlations between partial mortality and fission with the sediment measures indicated that the percent of particles <125 lm in size of the surface sediment was generally the best correlate of both rate of partial mortality and frequency of fission in all species (Table 4a and b). There was no significant correlation between partial mortality and sedimentation rates measured in traps. Fig. 5a and b show the relationship between the percent of particles <125 lm in size of the surface sediment and rate of partial mortality for S. siderea and M. faveolata-annularis.

The percent change in live coral cover from 1995 to 1998 was negatively related to partial mortality in all species (Table 4c). This relationship was significant for M. faveolata-annularis. Again, fission gave less significant results (Table 4c). Fig. 5c and d depict the relationship between percent change in live coral cover from 1995 to 1998 and rate of partial mortality for S. siderea and M. faveolata-annularis. Live coral cover declined in all sites with average rates of partial mortality >40% for S. siderea and >30% for M. faveolataannularis. Note that all these sites also had >10% of particles <125 lm in size in their surface sediment (Fig. 5a and b). Finally, the percent of damaged colonies did not show any consistent variation between sediment exposure and location in any of the species studied; whereas,

320

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323

Table 4 SpearmanÕs rank-order correlation matrix between the two sediment measures and (a) rate of partial colony mortality and (b) frequency of fission for each coral species and between (c) percent change in live coral cover from 1995 to 1998 and both rate of partial colony mortality and frequency of fission for each coral species Variables

P. astreoides

S. siderea

C. natans 0.313 0.550

(a) Partial mortality

Sedimentation rate % Sediment <125 lm

0.396 0.790

0.214 0.769

(b) Fission

Sedimentation rate % Sediment <125lm

0.361 0.458

0.316 0.586

(c) Change in coral cover

Partial mortality Fission

)0.455 )0.370

)0.445 )0.100

M. faveolata-annularis 0.077 0.489

)0.295 0.490

0.511 0.324

)0.291 )0.169

)0.745 )0.691

Correlations are based on mean values for each site. n ¼ 13 sites except for (c) where n ¼ 11 sites. * < 0:05. ** < 0:01. *** < 0:001.

Sediments < 125 µm (%)

S. siderea 25 20

M. faveolata-annularis 25

rs = 0.77**

20

15

15

10

10

5

5

0

0 0 10 20 30 40 50 60

Change in live coral cover from 1995 to 1998 (%)

a)

0 10 20 30 40 50 60 b)

40

40

20

20

0

0

-20

-20

-40

-40

-60

rs = -0.45

-80

rs = -0.75**

-60 -80

0 10 20 30 40 50 60 c)

rs = 0.49

Partial mortality (% colony surface area)

0 10 20 30 40 50 60 d)

Partial mortality (% colony surface area)

Fig. 5. Percent of the 125 lm particle size-fraction in sediment deposits, and percent change in live coral cover from 1995 to 1998, each as a function of partial colony mortality for S. siderea ((a) and (c), respectively) and M. faveolata-annularis (b) and (d). Error bars represent 1 S.E. Partial mortality means are based on a minimum of 30 colonies per site. Particle size data are means of four core samples. n ¼ 13 sites for a and b, and n ¼ 11 sites for c and d. rs ¼ Spearman rank-order correlation coefficient. Significance levels are indicated by:  < 0:05;  < 0:01.

the percent of colonies with P 50% partial mortality strongly increased close to river mouths on both sediment-exposed reefs in S. siderea and M. faveolata-annularis (Fig. 6). This suggests that this simple measure on these two species could reliably indicate sediment stress on coral reefs.

4. Discussion Results from this study support the use of partial mortality in some species of massive corals such as S. siderea and M. faveolata-annularis as an indicator of sediment stress. Rates of partial mortality in these species

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323

a) P. astreoides 25

80

20

60

15

40

10

20

5

0

0 b) S. siderea

100 Damaged colonies (% of colonies)

80 60 40 20 0 100

c) C. natans

80 60 40 20 0

Colonies > 50 % partial mortality (% of colonies)

100

80 60 40 20 0 50 40 30 20 10 0

100

d) M. faveolata-annularis 80

80

60

60

40

40

20

20 0

0 n m f n m f C S

n m f n m f C S

Fig. 6. Patterns of variation in percent of damaged colonies and percent of colonies with P 50% partial mortality (means  S:E:) along the two control (C) reefs and the two sediment-exposed (S) reefs in relation to location (n, near; m, mid; f, far). Values were calculated over a minimum of 30 colonies per reef per location.

were higher close to the river mouths where more sediments were deposited than away from the rivers. Furthermore, our data suggest that partial mortality could reflect temporal changes in coral communities. Therefore, it could reveal areas of serious decline when baseline data are missing (e.g., Ôpost-impactÕ studies). Conclusive results could also be obtained by simply recording partial mortality on colonies as <50% or P 50%, suggesting that this method could be applied easily and rapidly. Although we could have expected large-sized colonies to be better indicators of sediment stress, colony size did not greatly influence the effect of sediment exposure, site location and their interaction on partial mortality.

321

Colony fission was a less sensitive indicator. Fission mostly occurs at high rates of partial mortality. Therefore, in species with generally low rates of partial mortality (e.g., P. astreoides and C. natans), or at low levels of sedimentation, partial mortality is likely to be a more sensitive indicator. At high rates of partial mortality, fission may take place. However, total colony mortality is known to increase sharply with decreasing colony size (Hughes and Jackson, 1985; Babcock, 1991). Therefore, the probability of complete mortality of the surviving patches is greatly increased, reducing the opportunity to record fission events even further. S. siderea and M. faveolata-annularis were particularly good indicator species. They suffered much greater damage on reefs exposed to river sediments than C. natans and P. astreoides. These differences may be interpreted by the differing abilities in these species to reject fine particles. S. siderea and M. annularis had a lower ability to reject silt compared to P. astreoides and C. natans in Bak and ElgershuizenÕs, 1976 laboratory experiments. Their results showed that it took less than 3 h for the two latter species and between 5 to 10 h for the two former species to remove 0.75 g of carborundum powder (a simulator of silt). In the present study, this explanation is further supported by the positive correlations between partial mortality and percent of particles <125 lm in size of the surface sediment. In addition, partial mortality originates when patches of coral die and are not regenerated. Therefore, a possibly related factor may be a lower capability in these species to regenerate lesions. On the fringing reefs of Curacßao, Meesters et al. (1992) found a decrease in regeneration capability in M. annularis at a site exposed to a short-term increase in sedimentation compared to a relatively unaffected upcurrent site. They did not find such a decrease in S. siderea colonies exposed to prolonged sediment stress. However, this species was generally extremely poor at regenerating lesions. In this study, damage and partial mortality in S. siderea was substantial and occurred across all sites, including those unaffected by river sediments (Fig. 4b). Other studies have also reported high levels of lesions in this species (e.g., Guzman et al., 1991; Meesters et al., 1992, 1997; Ginsburg et al., 1996; Lewis, 1997; Ruesink, 1997; Debrot et al., 1998). Although sedimentation has long been known to affect corals, the ways in which it does so are complex. Sediments can cause coral mortality directly by increasing energy expenditure from removing sediment particles from coral tissue. However, they can also affect corals indirectly by decreasing the light available to photosynthesising symbiotic algae, or stimulating the growth of competitors of corals (reviewed by Rogers, 1990). In this study, there were no significant correlations between partial mortality and sedimentation rates measured in traps. Reefs at the ÔfarÕ location on the

322

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323

sediment-exposed reefs were subjected to considerable levels of sediment resuspension from swells, particularly in the dry season, but corals in these sites had low rates of partial mortality (Fig. 4). Instead, damage was localised to places where riverine sediment plumes were deposited and where sediments were fine. This suggests that the composition of bottom sediments in term of particle size may be more critical to sediment effects on corals than sedimentation rates. Other studies demonstrated the importance of particle size on corals (Hubbard and Pocock, 1972; Bak and Elgershuizen, 1976; Lasker, 1980; Fisk, 1981; Stafford-Smith and Ormond, 1992). The aforementioned differences among species in relation to their ability to reject fine particles lend further support to this view. Finally, several studies showed the importance of turbidity from suspended sediments, as opposed to sediment deposition rates in affecting corals (Bak, 1978; Rogers, 1979; Tomascik and Sanders, 1985, 1987). Although this parameter was not measured in this study, it may also play an important role in determining partial mortality patterns. Partial mortality in corals is a natural phenomenon. It can be caused by a wide variety of disturbances, such as hurricanes, predation, competition, diseases or human impacts (Pearson, 1981; Brown and Howard, 1985; Bythell et al., 1993) to which species respond differently (e.g., Bythell et al., 1993; this study). Babcock (1991) also noted that partial mortality could be part of a natural process of ÔsenescenceÕ in corals. Therefore, much remains to be learned about the significance and causes of partial mortality. Results from this study suggest that partial mortality in some species of massive reef corals could be used as a rapid and effective means of detecting sediment stress on coral reefs. Where the reef habitat is affected by riverine sediment inputs, this should translate into an increase in partial mortality in species having a poor ability to reject fine sediments (e.g., species of Montastraea and Siderastrea).

Acknowledgements We thank the St. Lucia Department of Fisheries, Soufri
ere Marine Management Area, Soufri
ere Foundation and Scuba St. Lucia for their continuing support of our work. We much appreciate the help and involvement of staff from these organisations including Horace Walters, Sarah George, Dawn Pierre-Nathaniel, Susannah Scott, Kai Wulf, Jean-Thierry Winstel, and Karyn and Michael Allard. We thank Joshua Sladek Nowlis for undertaking the 1995 survey of benthic reef communities. We are also grateful to Roger Malone and Ronnie Nicholas for their invaluable help in the field. Field assistance in sediment trap collection was ably provided by Tessa McGarry, Chris Schelten, Annalie

Morris, Rachel Miller and Douane Joseph. Allan Smith provided technical assistance and John Byrne and Jaap van der Meer helped with the statistical analyses. This study was supported by grants from the UK Darwin Initiative and NERC secured by CR.

References Alcolado, P.M., Herrera-Moreno, A., Martinez-Estalella, N., 1994. Sessile communities as environmental bio-monitors in Cuban coral reefs. In: Ginsburg, R.N. (Ed.), Proceedings of the Colloquium on Global Aspects of Coral Reefs, University of Miami, Miami, pp. 27–33. Babcock, R.C., 1991. Comparative demography of three species of scleractinian corals using age- and size-dependent classifications. Ecological Monographs 61, 225–244. Bak, R.P.M., 1978. Lethal and sublethal effects of dredging on reef corals. Marine Pollution Bulletin 9, 14–16. Bak, R.P.M., 1983. Neoplasia, regeneration and growth in the reefbuilding coral Acropora palmata. Marine Biology 77, 221–227. Bak, R.P.M., Elgershuizen, J.H.B.W., 1976. Patterns of oil-sediment rejection in corals. Marine Biology 37, 105–113. Bak, R.P.M., Steward-van Es, Y., 1980. Regeneration of superficial damage in the scleractinian corals Agaricia agaricites f. purpurea and Porites astreoides. Bulletin Marine Science 30, 883–887. Birkeland, C., 1996. Life and Death of Coral Reefs. Chapman & Hall, New York, 536pp. Blomqvist, S., Kofoed, C., 1981. Sediment trapping––a subaquatic in situ experiment. Limnology and Oceanography 26, 585–590. Brown, B.E., Howard, L.S., 1985. Assessing the effects of ‘‘stress’’ on reef corals. Advances in Marine Biology 22, 1–63. Bryant, D., Burke, L., McManus, J., Spalding, M., 1998. Reefs at risk––a map-based indicator of threats to the worldÕs coral reefs. Joint publication by World Resources Institute, International Center for Living Aquatic Resources Management, World Conservation Monitoring Centre, United Nations Environment Programme. Buchanan, J.B., Kain, J.M., 1971. Measurement of the physical and chemical environment. In: Holme, N.A., McIntyre, A.D. (Eds.), Methods for the Study of Marine Benthos. In: IBP Handbook No. 16. Blackwell Scientific Publications, Oxford and Edinburgh, pp. 30–58. Bythell, J.C., Gladfelter, E.H., Bythell, M., 1993. Chronic and catastrophic natural mortality of three common Caribbean reef corals. Coral Reefs 12, 143–152. Cochran, W.G., Cox, G.M., 1957. Experimental designs, second ed. John Wiley and Sons, New York, 611pp. Debrot, A.O., Kuenen, M.M.C.E., Dekker, K., 1998. Recent declines in the coral fauna of the Spaanse Water, Curacao, Netherlands Antilles. Bulletin Marine Science 63, 571–580. Dustan, P., 1994. Developing methods for assessing coral reef vitality: a tale of two scales. In: Ginsburg, R.N. (Ed.), Proceedings of the Colloquium on Global Aspects of Coral Reefs, University of Miami, Miami, pp. 38–45. Epstein, N., Bak, R.P.M., Rinkevich, B., 1999. Implementation of a small-scale ‘‘no-use zone’’ policy in a reef ecosystem: EilatÕs reef lagoon six years later. Coral Reefs 18, 327–332. Fisk, D.A., 1981. Sediment shedding and particulate feeding in two free-living, sediment-dwelling corals (Heteropsammia cochlea and Heterocyathus aequicstatus) at Wistari Reef Great Barrier Reef. In: Proceedings of the 4th International Coral Reef Symposium 2, 21–25. Ginsburg, R.N., 1994. In: Proceedings of the Colloquium on Global Aspects of Coral Reefs, University of Miami, Miami, 420pp.

M.M. Nugues, C.M. Roberts / Marine Pollution Bulletin 46 (2003) 314–323 Ginsburg, R.N., Bak, R.P.M., Kiene, W.E., Gischler, E., Kosmynin, V., 1996. Rapid assessment of reef condition using coral vitality. Reef Encounter 19, 12–14. Glasby, T.M., Underwood, A.J., 1998. Determining positions for control locations in environmental studies of estuarine marinas. Marine Ecology Progress Series 171, 1–14. Guzman, H.M., Jackson, J.B.C., Weil, E., 1991. Short-term ecological consequences of a major oil spill on Panamanian subtidal coral reefs. Coral Reefs 10, 1–12. Highsmith, R.C., 1982. Reproduction by fragmentation in corals. Marine Ecology Progress Series 7, 207–226. Hodgson, G., 1999. A global assessment of human effects on coral reefs. Marine Pollution Bulletin 38, 345–355. Hubbard, J.A.E.B., Pocock, Y.P., 1972. Sediment rejection by recent scleractinian corals: a key to paleo-environmental reconstruction. Geologische Rundschau 61, 598–626. Hughes, T.P., Jackson, J.B.C., 1980. Do corals lie about their age? Some demographic consequences of partial mortality, fission, and fusion. Science 209, 713–715. Hughes, T.P., Jackson, J.B.C., 1985. Population dynamics and life histories of folacious corals. Ecological Monographs 55, 141–166. Knowlton, N., Weil, E., Weight, L.A., Guzman, H.M., 1992. Sibling species in Montastrea annularis, coral bleaching, and the coral climate record. Science 255, 330–333. Lasker, H.R., 1980. Sediment rejection by reef corals: the roles of behavior and morphology in Montastrea cavernosa. Journal of Experimental Marine Biology and Ecology 47, 1158–1159. Lewis, J.B., 1997. Abundance, distribution and partial mortality of the Massive Coral Siderastrea siderea on degrading coral reefs at Barbados, West Indies. Marine Pollution Bulletin 34, 622–627. Meesters, E.H., Bak, R.P.M., 1993. Effects of coral bleaching on tissue regeneration potential and colony survival. Marine Ecology Progress Series 96, 189–198. Meesters, E.H., Bos, A., Gast, G.J., 1992. Effects of sedimentation and lesion position on coral tissue regeneration. In: Proceedings of the 7th International Coral Reef Symposium 2, 681–688. Meesters, E.H., Wesseling, I., Bak, R.P.M., 1996. Partial mortality in three species of reef-building corals and the relation with coral morphology. Bulletin Marine Science 58, 838–852. Meesters, E.H., Wesseling, I., Bak, R.P.M., 1997. Coral colony tissue damage in six species of reef-building corals: partial mortality in relation with depth and surface area. Journal of Sea Research 37, 131–144.

323

Pearson, R.G., 1981. Recovery and recolonisation of coral reefs. Marine Ecology Progress Series 4, 105–122. Risk, M.J., Dunn, J.J., Allison, W.R., Horrill, C., 1994. Reef monitoring in Maldives and Zanzibar: low-tech and high-tech science. In: Ginsburg, R.N. (Ed.), Proceedings of the Colloquium on Global Aspects of Coral Reefs, University of Miami, Miami pp. 66–72. Rogers, C.S., 1979. The effect of shading on coral reef structure and function. Journal of Experimental Marine Biology and Ecology 41, 269–288. Rogers, C.S., 1990. Responses to coral reefs and reef organisms to sedimentation. Marine Ecology Progress Series 62, 185–202. Ruesink, J.L., 1997. Coral injury and recovery: matrix models link process to pattern. Journal of Experimental Marine Biology and Ecology 210, 187–208. Sebens, K.P., 1987. The ecology of indeterminate growth in animals. Annual Review of Ecological Systems 18, 371–407. Sladek Nowlis, J., Roberts, C.M., Smith, A.H., Siirila, E., 1997. Human-enhanced impacts of a tropical storm on nearshore coral reefs. Ambio 26, 515–521. Stafford-Smith, M.G., Ormond, R.F.G., 1992. Sediment-rejection mechanisms of 42 species of Australia Scleractinian corals. Australian Journal of Marine and Freshwater Research 43, 683– 705. Stoddart, D.R., 1974. Posthurricane changes on the British Honduras reefs: Resurvey of 1972. In: Proceedings of the 2nd International Coral Reef Symposium 2, 473–483. Thomason, J., Roberts, C.M., 1992. What is a healthy reef? Reef Encounter 11, 8–9. Tomascik, T., Sanders, F., 1985. Effects of eutrophication on reefbuilding corals. I. Growth rate of the reef-building coral Montastrea annularis. Marine Biology 87, 143–155. Tomascik, T., Sanders, F., 1987. Effects of eutrophication on reefbuilding corals. II Structure of scleractinian coral communities on fringing reefs, Barbados, West Indies. Marine Biology 94, 53– 75. Wilkinson, C.R., 1993. Coral reefs of the world are facing widespread devastation: can we prevent this through sustainable management practices? In: Proceedings of the 7th International Coral Reef Symposium 1, 11–21. Woodley, J., Chornesky, E., Clifford, P., Jackson, J., Kaufman, L., Knowlton, N., et al., 1981. Hurricane AllenÕs impact on Jamaican coral reefs. Science 214, 749–755.