Aquatic Botany 81 (2005) 225–243 www.elsevier.com/locate/aquabot Natural settlement dynamics of a young population of Turbinaria ornata and phenological comparisons with older populations Valérie Stiger *, Claude E. Payri Laboratoire d’Ecologie Marine, Université de la Polynésie Française, B.P. 6570 FAAA/Aéroport, Tahiti, French Polynesia Received 2 July 2003; received in revised form 27 September 2004; accepted 6 December 2004 Abstract Our study was done during 16 months on two sites at Moruroa to (i) compare Moruroan populations (young) with Tahitian populations (old) of Turbinaria ornata and (ii) follow the natural dynamic of a young population at Moruroa. Moruroan and Tahitian populations have maximal density and maximal reproductive characteristics during the cool and hot seasons, respectively. Recruits are present throughout the year at all sites. Within the reference field site, individuals were initially distributed regularly and then became clumped through time, recruits appeared around adults. Recruits mortality followed a similar pattern, whereas mortality of adults and juveniles was even throughout the monitoring period. Populations initially had similar proportions of recruits, juveniles and adults (28, 40 and 32%, respectively), but at the end, recruits represent 92% of the population. A linear relationship between the number of individuals present within the station and the colonized substrate area was determined (R2 = 0.96), suggesting a limiting effect of the substrate against the settlement of T. ornata with time. Moreover, the growth of recruits in the close vicinity of adults was arrested in the young populations: recruitment may therefore, significantly buffer high rates of adult mortality. It may contribute to the invasion and, more interesting to the persistence of T. ornata in new colonized area. # 2005 Elsevier B.V. All rights reserved. Keywords: Coral reef algae; Proliferation; Substrate limitation; Recruitment; Turbinaria ornata * Corresponding author. Present address: LEBHAM-IUEM, Université de Bretagne Occidentale, Technopôle Brest-Iroise, Place Nicolas Copernic, 29280 Plouzane, France. Tel.: +33 298 498792; fax: +33 298 498772. E-mail address: stiger@univ-brest.fr (V. Stiger). 0304-3770/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.aquabot.2004.12.005 226 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 1. Introduction French Polynesia has two reef ecosystems, the high volcanic islands and the low carbonate atolls where the abundance of fleshy macroalgae decreases markedly from the former to the latter (Payri and Denizot, 1993). The marine flora of these volcanic islands, such as the Society Archipelago, is characterized by large beds of brown fleshy algae (Dictyotales, Scytosiphonales, Fucales), whereas the atoll reef system is covered with turfs, green algal mats (Caulerpa, Halimeda and Microdictyon) and coralline algae pavements. In the 1980s, a spectacular benthic macroalgal proliferation started on reefs of the high islands (Payri and Naim, 1982; Done et al., 1991) and is still persisting (Augustin et al., 1997; Stiger and Payri, 1999a). The macroalgae involved are mostly the perennial frondose brown algae of the orders Fucales and Dictyotales, but with lesser amounts of the green algae Boodlea kaeneana, the red alga Acanthophora spicifera and Cyanobacteria assemblages. Identification of the factors responsible for this macroalgal proliferation is difficult without long-term monitoring and small-scale manipulations. The widely held view that macroalgal proliferation is caused by enhanced level of nutrients has recently been challenged (Schaffelke, 2001; Smith et al., 2001; Larkum and Koop, 1997; Atkinson et al., 1995). Moreover, the control of the macroalgae by herbivores is questionable when considering the large algae beds often observed in areas accessible to grazers. An alternative hypothesis is that shift from coral to algal dominance results from cascading changes in the variables that control community structure. The replacement of corals by algae may often indicate coral mortality due to external disturbances rather than competitive overgrowth by macroalgae (Williams and Polunin, 2001; Mc Cook et al., 2001). On the volcanic islands, the fucalean Turbinaria ornata biomass is high and populations are well developed (Payri, 1982; Payri, 1984; Stiger and Payri, 1999a, 1999b). Its proliferation on Tahiti Island implies important spatial variations in biological characteristics, i.e. dwarfism on the algal ridge and boosting of reproductive abilities under high hydrodynamic conditions and around the fringing reefs (FR) (Stiger and Payri, 1999a). This species is highly fecund; the released germlings disperse passively and settle within 90-cm of parent thalli (Stiger and Payri, 1999b). Moreover, when thalli are broken off, their floating vesicles allow drifting. It seems, therefore, that T. ornata can spread and proliferate in new areas because of its reproductive potential together with its ability to drift and rapidly create a new population through short-range dispersal. In French Polynesia, observed in large biomass on volcanic islands (Andréfouët et al., 2004), T. ornata has spread towards low-lying atolls of the Tuamotu Archipelago where it was not found before (Montaggioni et al., 1985; Payri and N’Yeurt, 1997). In 1985, T. ornata was first observed in North islands of this archipelago. Moreover, as early as 1990, T. ornata was observed at Moruroa, an atoll in the southern Tuamotu Archipelago, located about 1300 km to the southeast of Tahiti (Tartinville et al., 1997). Boat shipping between Tahiti and Tuamotu islands was first hypothesized as the introduction vector of T. ornata in this archipelago; but this hypothesis was rejected by observation of water-ballasts (Stiger, personal observation). The arrival of T. ornata by drifting thalli coming from high islands (Society Archipelago) has been V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 227 suggested as an explanation for its presence in atolls of the Tuamotu Archipelago (Stiger and Payri, 1999b; Payri and Stiger, 2001). To gain a better insight into how a novel population settles in a new environment, the monitoring of T. ornata around the high islands of French Polynesia conducted over the past decades was extended to the atoll of Moruroa. It showed two types of populations (considered as young): a large and dense one in the north of the atoll and a small and dispersed one in the south of the atoll. The biological behavior of the former was compared with that of known Tahitian populations (considered as old) from three sites: fringing reef (FR), inner barrier reef (IBR) and outer barrier reef (OBR) studied from November 1994 to February 1996. On the other hand, Southern population provided us with the opportunity both to document natural colonization and follow the persistence of T. ornata in a new environment. The results presented here are from a 16-month survey with the aim to determine first the biological characteristics of this species for self-maintenance in this new environment. For this, we followed seasonal variations in density, maturity index, fertility and measured the length at first maturity. Secondly, the colonization dynamics of a new habitat was studied by following recruitment and mortality of individuals classified in three life-history stages. The specific questions addressed in this study were: (1) Do biological characteristics, i.e. density, maturity, sizes, fertility, vary among young and old populations? (2) How the settlement of a young population take place? (3) How is the distribution of individuals in a young population? (4) Is there a limitant factor in the settlement of a young population, as substrate for example? Since the colonization of new substrata is a fundamental process in benthic organisms, our study should result in a better understanding of the invasive ability of T. ornata, and may contribute to management efforts against its proliferation. 2. Materials and methods 2.1. Field sites The study was done at two field sites of Moruroa atoll (Tuamotu Archipelago, French Polynesia) (218500 S, 1388550 W). In the north-east of the atoll, site A was characterized by dispersed coral colonies near the pass with large and dense populations of T. ornata attached to two coral colonies (circumference  100 m) dominated by some Pocillopora; at low tide Pocillopora and Turbinaria emerged. This site was chosen not only to get biological data but also for comparison to literature data recorded in T. ornata on the high Island of Tahiti (178340 S, 1498340 W) (Stiger and Payri, 1999a). In the south of the atoll, site B had a few sparse individuals of T. ornata at the beginning of the study. It was chosen because the number of mature thalli was great enough to get a sufficient progeny over the period of study. In addition, isolated mature thalli allowed recruits to be counted; grid layout as well as diving for data collection was made easier by the near-flat substratum surface and shallow depth (about 2 m) of this area. No replicates were done since this area was the only one to present few individuals of T. ornata. This site was always underwater. The reef flat had a 5-cm-high turf of Coelothrix irregularis (Rhodymeniales), and frondose species such as Dictyota, Codium, Padina and Caulerpa. 228 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 2.2. Data collection and analysis at site A The algal population was analyzed by SCUBA diving using 0.25 m2 quadrats. The density, maturity and mean fertility of females were determined as described by Stiger and Payri (1999a). Following a pre-sampling design and in accordance with individuals within quadrats, the number of collected quadrats was fixed at two. At each collection date, all the thalli were removed within each haphazardly quadrat, which permitted us (i) to determine density and maturity index, i.e. number of mature thalli divided by the total number of thalli in the sample, and (ii) to also follow the recolonization of the area after clearing. The length between the holdfast and the apex of T. ornata was measured for each plant and placed into 17 size classes from class 0 (0–2 cm) to class 16 (32–34 cm) according with Scherrer (1984) for calculation. The maturity index in each size class was determined. A regression between maturity index (multiplied by 100) evolution and size was made to find the good relation between these two parameters. The curve coefficients were estimated using the Macintosh software MacCurveFit; this relation permitted us to determine the length at first maturity: it corresponds to the length at which 50% of individuals in the population are mature. The fertility, i.e. number of oogonia produced by the female population, was determined according to Stiger and Payri (1999a). We refer to female because of their dominance in all the studied populations (Stiger, 1997). To satisfy the criteria of normality and homoscedasticity for parametric tests, transformation was applied to maturity data before processing them with the Super Anova software for Macintosh to perform one-way ANOVA tests with time as factor. Then, Student–Newman–Keuls tests (S–N–K) were carried out to constitute homogenous groups (Zar, 1984). The whole data set was compared to the ones from Tahitian populations. For all the field sites, the variability of the studied parameters was compared using Bartlett’s test (for homogeneity of variances) at a 5% significance level. 2.3. Data collection and analysis at site B To illustrate the spatio-temporal pattern of recruitment of T. ornata, its abundance within a given area was mapped as follows: first, all the individuals living in the area were recorded. Then, it was delimited with a squared stainless steel frame (6 m  6 m) maintained in place throughout the period of study. This frame was divided into four quadrats with stainless steel bars. Ropes were placed on them at precise locations to split the whole framed area into 144 quadrats of 0.25 m2 each. A previous study had demonstrated that the rate of mortality in fucoid recruits was higher on the edges than at the center of experimental areas (Emmerson and Collings, 1998). Then, we discard the outer 44 quadrats to free from substrate-induced effect on the recruits. In the end, 100 quadrats were really used for sampling analysis. The three-step data collection procedure was the following one: (i) recording the existing thalli, (ii) noting down the new macro-recruits visible to the unaided eye and (iii) determining the fate of each developmental stage. Developmental stages were investigated by spotting in situ each individual or group of individuals. Two rigid rulers were placed in the upper left and right corners of the examined quadrats. For each quadrat, the distances (in cm) between each of the two rulers and the point where they intersected with each individual were measured using X and Y coordinates 229 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 as previously described by Kendrick and Walker (1994). When groups were observed, we took into account the coordinates of the center of the group and counted the number of individuals at each developmental stage. In order to realistically depict the life-history of T. ornata, three ontogenic stages were defined as follows: (i) macro-recruits: size < 1 cm and soft texture, (ii) juveniles: length within 1 and 10 cm together with tough texture, but no receptacles and (iii) mature adults: thalli bearing mature receptacles. Within the study area, the algal density in each 0.25 m2 quadrat was determined; then, the area was stratified at a given sampling time (T0, T1, T2, T3 and T4) on considering the number of individuals as a stratifying variable. Within each stratum, a simple random sampling was done by determining the number of quadrats to be sampled from Elliot’s relation (Plante and Le Loeuff, 1982) with p = 20% (Table 1). Within each stratum, the mean numbers of individuals, recruits, juveniles and adults as well as their maturity index were determined at each collection date as already described in Section 2.2. Table 1 Number of 0.25-m2 quadrats randomly collected in each stratum within the permanent 36-m2 station at site B (Moruroa, French Polynesia) Sampling months Stratum A Stratum B Stratum C February 1995 June 1995 October 1995 February 1996 June 1996 4 4 11 12 15 4 3 4 3 5 3 8 3 5 6 Using the software ‘Statgraphics’, two-way ANOVA tests were performed on the number of recruits, juveniles and adults and on the maturity index to examine the effects of the factors ‘month’ and ‘stratum’. Multiple-range test comparisons were carried out in order to constitute homogeneous groups. To satisfy the criteria of normality and homoscedasticity for parametric tests, the data were log-transformed, except for the maturity index for which an angular transformation was performed (Underwood, 1999). The spatial pattern of this population was studied using the modified Clark–Evans (CE) method, which is a simple univariate nearest neighbor method. For this, the sum (T) and the expected sum (E(T)) of the nearest neighbor distances, variance of T (varT) and the modified Clark–Evans statistic were calculated as described in Creed (1995). It indicates whether individuals within a population are clumped (CE < 0) or display a regular pattern (CE > 0) (Creed, 1995). A random distribution is demonstrated by equivalence of mean and variance in compliance with the Poisson distribution (Scherrer, 1984). This CE statistic was followed also for recruitment through the observation of new recruits and for mortality (disappearance) at each stage of development within the reference field site. We investigated the mortality of recruits, juveniles and adults in order to make tentative conclusions about the role of intraspecific competition in the population of concern. Further to the death of corals, we also wondered about seaweed recruitment limitation by free substrate. By using WinCurveFit software, a relationship between the substrate area and the number of individuals followed in relation with time was determined. 230 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 This study was conducted over 16 months from February 1995 to June 1996. Algal populations were sampled every 4 months, during the hot (February 1995, October 1995 and February 1996) and cool (June 1995 and 1996) season. A 4-month period between two successive samplings may sound too long, but it was imposed by the very long distance (1200 km) between Tahiti, where our laboratory precincts are, and Moruroa; on the other hand, it allowed visible changes within the monitored populations. 3. Results 3.1. Biological characteristics at field site A The density varied over time ( p = 0.0349). It was maximal in June 1995 and 1996, with respectively 1248  148 and 1850  114 individuals m2 (Fig. 1A). In both years, recruitment occurred in June, but it was greater in 1996 than in 1995. The minimal density of 732  128 individuals m2 was observed in October 1995. T. ornata can reproduce yearround and its maturity index was always greater than 0 (Fig. 1B). Mature adults were seen throughout the period of study, but their number was reduced over the coldest months (June 1995 and 1996, weak maturity index), presumably due to their growth from one size to the next. The maturity of T. ornata seemed to be inversely related to density and varied over time ( p = 0.0249). In February 1995, 18% of the population was mature. Later, its maturity decreased to a level considered as stable (S–N–K test) with a maturity index within 0.08 and 0.06 in October 1995 and June 1996, respectively. One should, however, note variability in data in October 1995 (Fig. 1B). The histogram on Fig. 1C shows the dominance of small thalli (classes 0–3: length < 8 cm) within the population. The length at first maturity estimated from the logistic curve of growth was 10.98 cm (a = 0.010; b = 0.589; c = 0.690; R2 = 0.966). The female thalli of T. ornata produce oogonia throughout the year. Their fertility, however, varied versus time ( p = 0.0349) (Fig. 1D). Maximal fertility was observed in February 1995 with a production of 50.3  105  8.5  105 oogonia. The S–N–K test showed no significant variation in the female fertility rate between June 1995 and June 1996 with values within 8.7  105  3.6  105 and 19  105  8.32  105 oogonia. 3.2. Comparison of biological characteristics between Tahitian and Moruroan populations The population from Moruroa exhibited relatively high density compared to those from Tahiti (Table 2); noticeable variations are observed between the different field sites. Moruroa and IBR populations display an alike maturity index, which is lower than those of the two others (Table 2). The mature individuals of Moruroa displayed the smallest mean size: 10.95  4.2 cm. About their length at first maturity (Lfm), the different populations were ranked as follows: OBR (9.10 cm), FR (10.25 cm), Moruroa (10.98 cm) and IBR (15.44 cm). One should note that FR and Moruroa populations display similar Lfm. Fertility expressed in number of oogonia among female thalli allowed us to classify them as follows: first OBR (4.2  106  3.1  106), then FR (3.2  106  3.0  106) which was V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 231 Fig. 1. T. ornata at site A (Moruroa, French Polynesia). (A) Time variation of the density (individuals m2) of the whole population. (B) Variation in maturity index vs. sampling time. (C) Estimation of the length at first maturity using frequency of individuals in each size class and maturity index. (D) Evolution of the number of oogonia produced by the female population vs. time. Data in Fig. 1A, 1B and 1D represent mean and standard deviation (S.D.); values below the bars were homogenous according to the Student–Newman–Keuls test at the 5% significance level. 232 Variables Density (individuals m2) Maturity index Mean size of mature individuals (cm) Lfm (cm) Fertility (number of oogonia per female population) Tahiti Moruroa Statistical analysis FR IBR OBR Site A Bartlett’s test SNK test (for all populations) 287 (105) 0.233 (0.02) 16.2 (5.1) 10.25 3.2  106 (3.0  106) 608 (223) 0.122 (0.005) 19.6 (4.3) 15.44 1.6  106 (1.3  106) 717 (197) 0.226 (0.018) 13.1 (3.7) 9.10 4.2  106 (3.1  106) 1127 (460) 0.09 (0.008) 10.95 (4.2) 10.98 2.38  106 (1.8  106) Bobserved = 5.77, x2 = 7.8 FR < IBR < OBR < A Bobserved = 3.43, x2a = 7.8 IBR = A < FR = OBR Bobserved = 1.58, x2a = 7.8 A < OBR < FR < IBR – Bobserved = 8.30, x2a = 7.8 No SNK test performed IBR < A < FR < OBR FR refers to fringing reef, IBR to inner barrier reef and OBR to outer barrier reef. V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 Table 2 Comparisons of density, maturity index, mean size, length at first maturity and fertility data between Tahitian and Moruroan populations of T. ornata using the Student– Newman–Keuls and Bartlett’s tests at the 5% significance level V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 233 twice the IBR one (1.6  106  1.3  106). Oogonia production by the population of Moruroa felt within the two values (2.38  106  1.8  106). Barlett’s testing of the density, maturity index and mean size data versus time for Moruroa and Tahiti populations highlighted homogeneity in temporal variability (Bobserved = 5.77; Bobserved = 3.43; Bobserved = 1.58 and x2a = 7.8, respectively, Table 2). The location and environment had no significant effect on T. ornata seasonal behavior: all of them were characterized with high density and large size over the cool months and a high maturity index over the hottest months. Conversely, environmental conditions affected the fertility of female T. ornata (Bobserved = 8.30 and x2a = 7.8, Table 2). Four months after each clearing, recruits were observed within each quadrat. But no noticeable growth of recruits was observed over time: they constituted a crust-like mat covering the quadrat area. 3.3. Temporal pattern of density and maturity at field site B The variation of the total number of individuals versus time indicated three-phase colonization within the reference area (Fig. 2A). Two-way ANOVA test on the total number of individuals with respect to months ( p < 0.0001) and stratum ( p = 0.0001) showed that these factors significantly affected the population (Table 3). The mean densities varied significantly from one year (1995) to the other (1996) with maximal values on the second year. Statistic testing of mean density evidenced that values were stable over the first year of monitoring (1995) and ranged from 88  28 individuals in February to 104  36 individuals in June (Fig. 2A, Table 3). Four months later, the mean density increased to 754  71 individuals; the maximal density, i.e. 3190  957 individuals, was reached in February 1996 and remained stable at 3011  1523 individuals in June 1996. The maturity index being always above 0 (Fig. 2D for the number of mature individuals), T. ornata within the reference station can reproduce all over the year. The factor ‘month’ had a significant effect ( p < 0.0001) on the maturity index (Table 3); the number of mature thalli decreased along the period of study (Fig. 2D). The evolution of the three chosen stages of development, i.e. recruits, juveniles and adults, differed and varied versus time (Fig. 2B–D). The factors ‘month’ and ‘stratum’ significantly affected the mean density of total recruits within the station ( p < 0.0001 and p = 0.0014, respectively; Table 3). A continuous production of recruits was observed throughout the period of study (Fig. 2B). The highest recruitment occurred between October 1995 and February 1996 (Fig. 2B, Table 3). On the other hand, the mean number of juveniles in the field station was affected only by ‘month’ ( p = 0.0001, Table 3). The lowest number of juveniles was observed in June and October 1995, whereas the highest one was recorded in February 1996 (Fig. 2C). Adults were found all over the study (Fig. 2D); their number was significantly affected by the factor ‘month’ ( p = 0.0085), but not by stratum ( p = 0.2535) (Table 3): they were, indeed, the most numerous over the hottest months, i.e. February 1995 and 1996 (31.5  10.6 and 36.00  43.2, respectively) and minimal values during the coldest months. Throughout the study juveniles and adults were far less numerous than recruits (Fig. 2). 234 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 Fig. 2. Time variation of T. ornata mean number m2 at the permanent reference station: (A) for all the individuals, (B) recruits, (C) juveniles and (D) adults. Transformed data depict mean and S.D. (only the maximal value is represented); values below the bars were homogenous according to the Student–Newman–Keuls test at the 5% significance level. 235 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 Table 3 Analysis of months and stratum effects on the number of T. ornata (total individuals, recruits, juveniles and adults) and maturity Total number of Source of variation p Individuals Month Stratum 0.0000 0.0001 Feb-95 = Jun-95 < Oct-95 > Feb-96 = Jun-96 Recruits Month Stratum 0.0000 0.0014 Feb-95 = Jun-95 < Oct-95 > Feb-96 = Jun-96 Juveniles Month Stratum 0.0001 0.0634 Feb-95 > Jun-95 = Oct-95 < Feb-96 > Jun-96 Mature adults Month Stratum 0.0085 0.2535 Feb-95 > Jun-95 > Oct-95 < Feb-96 = Jun-96 Mature within the whole population Month Stratum Multiple range test 95% LSD 0.0000 0.7163 Feb-95 = Jun-95 > Oct-95 = Feb-96 = Jun-96 The multiple range test was used for multiple comparisons of month effects within the reference station. 3.4. Spatial pattern of individuals at field site B At the start of the monitoring, February 1995, the population consisted of some regularly (CE significantly > 0) distributed individuals (Fig. 3A and B, Table 4). Four months later, the negative results of CE statistic indicated a change in this distribution pattern and suggest a clumping of individuals with time (Fig. 3D and E, Table 4). The smallest nearest neighbor distance was 12 cm in February 1995 and fell down to 8 cm in June 1995, then to 2 cm at the next collection dates. Then, after some time, aggregation of germlings to the parent thalli was observed (Fig. 3). 3.5. Spatial pattern of individuals’ recruitment and mortality at field site B Recruitment and mortality of recruits were followed all over the study. At first, recruits appeared and disappeared regularly within the station (Table 5). Between June and October 1995, a sudden change was noticed in this pattern with the occurrence of clumped recruits; this phenomenon was visible till the end of the monitoring (Table 5). The mortality of recruits followed the same pattern with a 4-month delay; aggregation of recruits likely explains why their mortality followed a clumped distribution between October 1995 and February 1996. On the other hand, the mortality of juveniles and adults was regular Table 4 Modified Clark–Evans (CE) statistics (in meters) for the population of T. ornata determined for each collection date from February 1995 to June 1996 within the reference area at site B (Moruroa, French Polynesia) Sampling months CE statistic (m) n Interpretation February 1995 June 1995 October 1995 February 1996 June 1996 0.13 0.11 0.05 0.12 0.37 88 104 754 3190 3011 Regular distribution Regular distribution Clumped distribution Clumped distribution Clumped distribution 236 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 Fig. 3. Spatio-temporal distribution of individuals of T. ornata within the reference area split into 100 quadrats of 0.25 m2: time evolution of T. ornata mean number at each developmental stage: recruits, juveniles and adults at Moruroa: (A) in February 1995, (B) in June 1995, (C) in October 1995, (D) in February 1996 and (E) in June 1996. Each point corresponds to one individual or one group of individuals. R refers to Recruits, J to Juveniles and A to adults. throughout the study except between October 1995 and February 1996 when their survival was high (Table 5). 3.6. Temporal composition of the population The T. ornata population within the field station was always composed of recruits, juveniles and mature individuals, but the relative proportion between these stages varied 237 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 Table 5 Modified Clark–Evans (CE) statistics (in meter) for the appearance of recruits and disappearance of recruits, juveniles and adults of T. ornata determined from February 1995 to June 1996 within the reference area at site B (Moruroa, French Polynesia) Period Stage of development Event CE statistic Interpretation Feb-95 to Jun-95 Recruits Recruits Juveniles Adults Recruitment Mortality Mortality Mortality 0.137 0.065 0.064 0.066 Jun-95 to Oct-95 Recruits Recruits Juveniles Adults Recruitment Mortality Mortality Mortality 0.059 0.038 0.023 0.025 Clumped distribution Regular distribution Regular distribution Regular distribution Oct-95 to Feb-96 Recruits Recruits Juveniles Adults Recruitment Mortality – – 0.151 0.025 – – Clumped distribution Clumped distribution No disappearance No disappearance Feb-96 to Jun-96 Recruits Recruits Juveniles Adults Recruitment Mortality Mortality Mortality 0.227 0.557 0.160 0.010 Clumped distribution Clumped distribution Regular distribution Regular distribution Regular Regular Regular Regular distribution distribution distribution distribution over the monitoring period. At the beginning of the survey, the population consisted of recruits (28%), juveniles (40%) and mature individuals (32%). Four months later, i.e. June 1995, with the reproduction of mature thalli new recruits accounted for 50% of the population while juveniles and adults were 24 and 26%, respectively. By the end of the study, further discrepancy between the various stages was obvious. Recruits were exactly 92% of the population, juveniles 6% and mature thalli only 2%; these data demonstrate a large production of recruits all along the study. 3.7. Relation between the number of individuals and the substrate area A linear relationship was established between the number of individuals present within the station (N) and the colonized substrate area (S): log N = 0.116S + 1.42 with SSE = 0.09306 and R2 = 0.96 (significant at p = 0.001). It shows that, with the death of coral colonies, the size of T. ornata population is related to an increase in the substrate area available for settlement and suggests a limitant effect of the substrate against the settlement of T. ornata with time. 4. Discussion At Moruroa, two populations of T. ornata were observed: a large and dense at site A and a small and sparse at site B. These populations were considered as young while Tahitian ones were considered as old populations. 238 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 4.1. Comparison of young and old populations of T. ornata: is there a limitant factor in the settlement of a young population? Population at site A (Moruroa) highlighted spatio-temporal variations in terms of density and size, as already demonstrated in old populations at Tahiti (Stiger and Payri, 1999a). Seasonality of reproductive characteristics is less obvious due to high variability; it nevertheless suggests an increase of these parameters during the hottest months in agreement with the various literature data discussed in Stiger and Payri (1999a). The high values of the maturity index and the fertility in February 1995 remain unexplained. One should note the decrease of atmospheric pressure and higher precipitations in February 1995 compared to February 1996 (Stiger, 1997). In February 1995, we can suggest a boosting of reproductive abilities in T. ornata in response of environmental stress, as already demonstrated in older populations at Tahiti (Stiger and Payri, 1999a). T. ornata at Moruroa is a monospecific population found on the top of huge coral colonies covered by dead corals; its development is limited by low availability of substrate in the colonization area (linear relationship in this study) as already suspected (Payri, 1987; Venier and Fahrig, 1996). At Tahiti, in the fringing reef, T. ornata covered small pieces of rock, whereas it shares its ecological niche with Sargassum species in the inner and outer barrier reef: the latter has a substrate area alike the one in Moruroa, whereas the former consists of small and isolated coral colonies; all these observations partly explain why density was higher at Moruroa than at Tahiti and similar to the one at the OBR. Another cause is likely hydrodynamism. Indeed, the abundance of macroalgae at Moruroa is favored by violent wave action and emersion of the top of the coral colony at each low tide. Both factors limit the coral settlement and survival. However, some living colonies of Pocillopora remain, and compete for space with T. ornata. The inability of algal propagules to settle on healthy coral tissue has been demonstrated (Mc Cook et al., 2001; Diaz-Pulido and Mc Cook, 2002). Nevertheless, the current-driven T. ornata fronds cause necrosis of corals by scraping of coral recruits as already mentioned by Gleason (1996). Then, this dominance of algae on corals, plus the high rate of fertility and the quasi-continuous production of algal propagules all year long represent a major risk for corals whose reproduction is usually restricted, especially in French Polynesia (Jardin, 1994). One of our results, though preliminary, clearly suggests that substrate availability is the main limitation for T. ornata. Nowadays, the novel populations of T. ornata in the atolls of the Tuamotu Archipelago are limited in space. But, the hypothesis of an overgrowth at larger scale cannot be rejected, because of a possible increase in substrate availability further to natural and human degradations as suggested by McClanahan et al. (2001) and Connell et al. (1997). 4.2. How the settlement of a young population of T. ornata take place? This study evidenced a progressive establishment of T. ornata in the reference field site; it suggested a recruitment process all over the monitoring period. Over the 16-month monitoring, the increase in the number of individuals was, first, slow then quicker after the cold season. At the end, the number of individuals seemed to be stable. Since recruits dispersed within a restricted perimeter around parent thalli, the substrate quickly became V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 239 limiting (see Section 4.1), which may explain the stabilization of density observed at the end of the monitoring. Density of young recruits was elevated due to a continuous production of oogonia throughout the year. This observation was corroborated by the fertility of the population from site A (north of the atoll): its seasonal recruitment reflects its seasonal release of gametes as already demonstrated at site A (Section 3.1) and in an old population at Tahiti (Stiger and Payri, 1999a, 1999b). One should, however, note that only few recruits reach the adult stage because of high mortality rate. This phenomenon has been previously evidenced in other Fucales such as Sargassum (Ang, 1985; Ang and De Wreede, 1990; Kendrick and Walker, 1994), Fucus (Mc Cook and Chapman, 1992) and Ascophyllum (Vadas et al., 1990; Viejo et al., 1999). About Sargassum species, Kendrick and Walker (1994) observed that only 15–18 individuals became mature within 9 months and Ang (1985) noticed a 6% survival rate between recruitment and mature stages. In both studies, the mature stage was characterized by a high mortality rate. During colonization, the growth of recruits was slowed down. This phenomenon was also observed in the large and dense population at site A where most of coral colonies were covered with a mat of recruits whose size did not vary over several field trips (this study; Stiger, 1997). In populations from sites A and B, individuals were likely delayed in a ‘‘seed bank’’ (or ‘‘recruit bank’’) year round. This pool of recruits may buffer stochastic events that eliminate macroscopic individuals from the population. Algal propagule banks analogous to seed banks of terrestrial plants (Santelices et al., 1991) have been recently described in a wide range of genera: Codium bursa (Vidondo and Duarte, 1998), Ecklonia radiata (Kirkman, 1981), Fucus distichus (Ang, 1991), Fucus serratus (Creed et al., 1996) and Sargassum sp. (Kendrick and Walker, 1991). The decrease in growth may result from adult thalli-induced inhibition and/or overcrowding of recruits. 4.2.1. Effect of adult thalli Such an inhibition of recruit growth may be produced through a chemical or physical action, i.e. allelopathy or shading, by adult thalli in agreement with previous studies on shading (Reed and Foster, 1984; Kennelly, 1987; Ang, 1991; Brawley and Johnson, 1991; Kendrick, 1994). The death of adult thalli releases inhibition, and then allows the re-start of recruits growth as already observed in several species of Cystoseira (Benedetti-Cecchi and Cinelli, 1992). 4.2.2. Overcrowding of recruits In the present study, recruits were spatially distributed in the close vicinity of mature thalli. According to Stiger and Payri (1999b), the dispersal distance for T. ornata on Tahiti Island was below 0.9 m. Consequently, near the parent thalli, settlement rate is high and the new germlings are clumped. Schiel (1985) noted that the size of Sargassum sinclairii individuals was reduced when recruit densities were high. Arrested development of juveniles has also been reported in F. distichus (Ang, 1991). In this study, the recruits isolated from parent thalli were distinguished from those close to parent thalli. No investigation about the factors responsible for mortality in T. ornata was done, but field observations permitted us to conclude that grazers, i.e. sea urchins and fishes, are present in the vicinity of thalli at site B (personal observation). Within this field station, at the beginning of the study, regular death of isolated recruits was noticed and may 240 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 come from grazing on the turf as already shown by Stimson et al. (2001). Indeed, these authors observed that, by grazing heavily on their preferred species of algae, herbivorous fishes facilitate the access to more substrate for the less palatable species. In the present study, mortality was not caused by the rubbing of ropes and steel bars on the bottom since we considered only the central quadrats. 4.3. How is the distribution of individuals in a young population of T. ornata? The spatial pattern of individuals within the station evolved versus time: it was regular at the beginning of the study and became contagious or clumped at the end. The mortality of recruits followed a similar pattern (regular and then clumped), whereas that of adults and juveniles was regular all over the study. The spatial pattern of algae was discussed in Creed (1995). According to this author, the distribution of any algal population is expected to become more regular over time: the stronger the clumping of individuals is, the stronger the competition is within the clump and the greater the mortality is because of self-thinning. A regular, rather than random or clumped, spatial pattern should be interpreted as strong evidence of intraspecific competition (Creed, 1995). In our study, the regular distribution of recruits can be explained by a competition among recruits that starts at the time of settlement. Over time, the competition-induced mortality is likely compensated by the numerous parent-thalli-produced recruits together with the slowing down in recruit growth. Mortality of the recruits settled around the parent thalli was low. Indeed, visual observation at the reference station demonstrated a clumped distribution of thalli: adults were at the center of the clump with recruits and juveniles just around them. The persistence of this distribution suggests that immature individuals are protected by parent thalli against grazing and overgrowth of other algae species. The parents may produce repulsive chemical substances able to limit turf growth and push back grazers. Considering that phenolic compounds, whose major role is chemical defense (Hay, 1997) and marine antibioticity, are known to play a role in chemical communication in the marine environment, we wonder about their possible action in the growth inhibition of T. ornata recruits. 4.4. Progressive colonization of T. ornata around the atoll of Moruroa Since its first observation at Moruroa, in 1985, T. ornata has progressively colonized the entire reef. At first, its distribution was confined to the northwest zone with high densities on three coral colonies. Nowadays, T. ornata is found on the internal reef flat surrounding the atoll. The T. ornata-made sea-drift rafts observed on a small island (Motu) in the southwest area do not exclude colonization from drifting thalli. The long-range transport (remote dispersal, Farnham, 1980) of fertile thalli of T. ornata from the high volcanic islands, e.g. Tahiti and Moorea, is likely a vector in the dispersal of the species to the different atolls (Stiger and Payri, 1999b; Payri and Stiger, 2001). We hypothesized that the arrival of propagules in the new ecosystem is indirectly related to the degradation of the coral reef and to the subsequent algal blooms in the high islands at the origin of the numerous drifting thalli displayed on SPOT satellite images (Belsher et al., 1990). The development of the population in the new environment is not correlated to water quality or V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 241 low grazing, but to recruitment capacity and substrate availability. The presence of T. ornata in atolls can be interpreted as a consequence of reef degradation. The colonization of new environment may result from long-distance dispersal of drifting thalli-produced neopropagules, whereas the expansion of a population is locally controlled by shortdistance dispersal of germlings released by fixed thalli. The ecological consequences of the T. ornata proliferation in French Polynesia are still unknown. Further investigations about the origin of T. ornata in the Tuamotu Archipelago and its impact on coral are needed to complete our understanding on this species and its effects on coral reefs. Acknowledgements This research was part of the first author PhD carried out within the Laboratory of Marine Ecology of the Université de la Polynésie Française (French Polynesia) with the financial support of the Ministère Français de l’Environnement (Direction de la Recherche et des Affaires Etrangères et Internationales). The authors thank the Service Mixte de Surveillance Radiologique et Biologique (SMSRB) for financial and logistical help on Moruroa atoll as well as B. Gout, who contacted Prof. C.E. Payri for this study and provided support. The authors are particular grateful to the scuba divers of SMSRB for their kind and efficient assistance in the field. They also sincerely thank M.P. Friocourt for fruitful help in the English writing of this manuscript and anonymous reviewers for the improving of the manuscript. Contribution No. 939 of the IUEM, European Institute for Marine Studies (Brest, France). References Andréfouët, S., Zubia, M., Payri, C., 2004. Mapping and biomass estimation of the invasive brown algae Turbinaria ornata (Turner) J. Agardh and Sargassum mangarevense (Grunow) Setchell on heterogeneous Tahitian coral reefs using 4-meter resolution IKONOS satellite data. Coral Reefs 23, 26–38. Ang Jr., P.O., 1991. Age- and size-dependent growth and mortality in a population of Fucus distichus. Mar. Ecol. Prog. Ser. 78, 173–187. Ang Jr., P.O., De Wreede, R.E., 1990. Matrix models for algal life history stages. Mar. Ecol. Prog. Ser. 59, 171– 181. Ang Jr., P.O., 1985. Studies on the recruitment of Sargassum spp. (Fucales, Phaeophyta) in Balibago, Calatagan, Philippines. J. Exp. Mar. Biol. Ecol. 91, 293–301. Atkinson, M.J., Carlson, B., Crow, G.L., 1995. Coral growth in high-nutrient, low-pH seawater: a case study of corals cultured at the Waikiki Aquarium, Honolulu, Hawaii. Coral Reefs 14, 215–223. Augustin, D., Galzin, R., Legendre, P., Salvat, B., 1997. Variation interannuelle des peuplements récifaux du récifbarrière de Tiahura (ı̂le de Moorea, Polynésie française). Oceanol. Acta 20 (5), 743–756. Belsher, T., Meinesz, A., Payri, C., Ben Moussa, H., 1990. Apport du satellite SPOT à la connaissance des écosystèmes récifaux coralliens. La végétation marine de l’ı̂le de Moorea, Polynésie française. Oceanol. Acta 13, 513–524. Benedetti-Cecchi, L., Cinelli, F., 1992. Canopy removal experiments in Cystoseira-dominated rockpools from western coast of the Mediterranean (Ligurian Sea). J. Exp. Mar. Biol. Ecol. 155, 69–83. Brawley, S.H., Johnson, L.E., 1991. Survival of fucoid embryos in the intertidal zone depends upon developmental stage and microhabitat. J. Phycol. 27, 179–186. Connell, J.H., Hughes, T.P., Wallace, C.C., 1997. A 30-year study of coral abundance, recruitment and disturbance at several scales in space and time. Ecol. Monogr. 67 (4), 461–488. 242 V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 Creed, J.C., Norton, T.A., Harding, S.P., 1996. The development of size structure in a young Fucus serratus population. Eur. J. Phycol. 31, 203–209. Creed, J.C., 1995. Spatial dynamics of a Himanthalia elongata (Fucales, Phaeophyta) population. J. Phycol. 31, 851–859. Diaz-Pulido, G., Mc Cook, L., 2002. The fate of bleached corals: patterns and dynamics of algal recruitment. Mar. Ecol. Prog. Ser. 232, 115–128. Done, T.J., Dayton, P.K., Dayton, A.E., Steger, R., 1991. Regional and local variability in recovery of shallow coral communities: Moorea, French Polynesia and Central Great Barrier Reef. Coral Reefs 9, 183–192. Emmerson, L.M., Collings, G.J., 1998. Macroalgal recruitment in artificially disturbed areas: interactive effects of temporal and spatial scale. Marine Freshwater Res. 49 (6), 541–546. Farnham, W.F., 1980. Studies on aliens in the marine flora of southern England. In: Price, J.H., Irvine, D.E.G., Farnham, W.F. (Eds.), The Shore Environment. Academic Press, London, Systematics Association Special Volume, 17b, pp. 875–914. Gleason, M.G., 1996. Coral recruitment in Moorea, French Polynesia: the importance of patch type and temporal variation. J. Exp. Mar. Biol. Ecol. 207, 79–101. Hay, M.E., 1997. The ecology and evolution of seaweed–herbivore interactions on coral reefs. Proc. 8th Coral Reef Symp. 1, 23–32. Jardin, C., 1994. Biologie de la reproduction de quelques coraux scléractinaires en Polynésie française: application à la restauration d’un site lagonaire degrade. Thèse de Doctorat, Univ. Fr. du Pacifique. Kendrick, G.A., 1994. Effects of propagule settlement density and adult canopy on survival of recruits of Sargassum spp. (Sargassaceae, Phaeophyta). Mar. Ecol. Prog. Ser. 103, 129–140. Kendrick, G.A., Walker, D.I., 1994. Role of recruitment in structuring beds of Sargassum spp. (Phaeophyta) at Rottnest island, Western Australia. J. Phycol. 30, 200–208. Kendrick, G.A., Walker, D.I., 1991. Dispersal distances for propagules of Sargassum spinuligerum (Sargassaceae, Phaeophyta) measured directly by vital staining and venturi suction sampling. Mar. Ecol. Prog. Ser. 79, 133–138. Kennelly, S.J., 1987. Inhibition of kelp recruitment by turfing algae and consequences for an Australian kelp community. J. Exp. Mar. Biol. Ecol. 112, 49–60. Kirkman, H., 1981. The first year in the life history and the survival of the juvenile marine macrophyte, Ecklonia radiata (Turn.) J. Agardh. J. Exp. Mar. Biol. Ecol. 55, 243–254. Larkum, A.W.D., Koop, K., 1997. ENCORE, algal productivity and possible paradigm shifts. Proc. 8th Int. Coral Reef Symp. 1, 881–884. McClanahan, T.R., Muthiga, N.A., Mangi, S., 2001. Coral and algal changes after the 1998 coral bleaching: interaction with reef management and herbivores on Kenyan reefs. Coral Reefs 19, 380–391. Mc Cook, L., Jompa, J., Diaz-Pulido, G., 2001. Competition between corals and algae on coral reefs: a review of evidence and mechanisms. Coral Reefs 19 (4), 400–417. Mc Cook, L., Chapman, A.R.O., 1992. Vegetative regeneration of Fucus rockweed canopy as a mechanism of secondary succession on an exposed rocky shore. Bot. Mar. 35, 35–46. Montaggioni, L., Richard, G., Bourrouilh-Le Jan Gabrié, C., Humbert, L., Monteforte, M., Naı̈m, O., Payri, C., Salvat, B., 1985. Geology and marine biology of Makatea, an uplifted atoll, Tuamotu archipelago, French Polynesia. J. Coastal Res. 1 (2), 165–171. Payri, C.E., Stiger, V., 2001. Macroalgal community changes on French Polynesian reefs, 1980–2000. IPC7, Thessaloniki. Phycologia 40 (4 Suppl.), 111. Payri, C.E., N’Yeurt, A.D.R., 1997. A revised checklist of Polynesian benthic Marine algae. Aust. Syst. Bot. 10, 867–910. Payri, C.E., Denizot, M., 1993. Les peuplements d’algues. In: Atlas de la Polynésie française pl.49. Editions de l’ORSTOM, Paris. Payri, C.E., 1987. Zonation and seasonal variation of the commonest algae on Tiahura reef (Moorea Island, French Polynesia). Bot. Mar. 30, 141–149. Payri, C.E., 1984. Variations biologiques et morphologiques en fonction du milieu chez Turbinaria ornata (Turner) J. Agardh (Phéophycées) du récif de Tiahura–Ile de Moorea–Polynésie Française. Bot. Mar. 27, 327– 333. V. Stiger, C.E. Payri / Aquatic Botany 81 (2005) 225–243 243 Payri, C.E., 1982. Les macrophytes du lagon de Tiahura (ı̂le de Moorea, Polynésie française). Inventaire floristique–Répartition–Biomasse–Variations saisonnières–Dynamique des populations de Turbinaria ornata (Phéophycées, Fucales). Thèse d’Etat, Univ. Sci. Techniques du Languedoc. Payri, C.E., Naim, O., 1982. Variations entre 1971 et 1980 de la biomasse et de la composition des populations de macroalgues sur le récif corallien de Tiahura (ı̂le de Moorea, Polynésie française). Cryptogamie: Algol. 3, 229–240. Plante, R., Le Loeuff, P., 1982. Le benthos marin et les stratégies d’échantillonnage. In: Frontier, S. (Ed.), Stratégies d’échantillonnage en écologie. Masson, Paris, collection d’écologie 17, 325–340. Reed, D.C., Foster, M.S., 1984. The effects of canopy shading on algal recruitment and growth in a giant kelp forest. Ecology 65, 937–948. Santelices, B., Hoffman, A., Otaiza, R., 1991. Banks of microscopic forms. I. Ecological roles in the establishment of the local vegetation. J. Phycol. 27 (Suppl.), 64. Schaffelke, B., 2001. Surface alkaline phosphatase activities of macroalgae on coral reefs of the central Great Barrier Reef, Australia. Coral Reefs 19, 310–317. Scherrer, B., 1984. Biostatistiques. Gaetan Morin. Schiel, D.R., 1985. Growth, survival and reproduction of two species of marine algae at different densities in natural stands. J. Ecol. 73, 199–217. Smith, J.E., Smith, C.M., Hunter, C.L., 2001. An experimental analysis of the effects of herbivory and nutrient enrichment on benthic community dynamics on a Hawaiian reef. Coral Reefs 19, 332–342. Stiger, V., 1997. Contribution à l’étude de la biologie des populations de deux grandes algues brunes Turbinaria ornata (Turner) J. Agardh et Sargassum mangarevense (Grunow) Setchell, proliférant sur les récifs de Polynésie française. Thèse de Doctorat, Univ. Fr. du Pacifique. Stiger, V., Payri, C., 1999a. Spatial and seasonal variations in the biological characteristics of two invasive brown algae, Turbinaria ornata (Turner) J. Agardh and Sargassum mangarevense (Grunow) Setchell (Sargassaceae, Fucales) spreading on the reefs of Tahiti (French Polynesia). Bot. Mar. 42, 295–306. Stiger, V., Payri, C., 1999b. Spatial and temporal patterns of settlement of the brown macroalgae Turbinaria ornata and Sargassum mangarevense in a coral reef on Tahiti. Mar. Ecol. Prog. Ser. 191, 91–100. Stimson, J., Larned, S.T., Conklin, E., 2001. Effects of herbivory, nutrient levels, and introduced algae on the distribution and abundance of the invasive macroalga Dictyosphaeria cavernosa in Kaneohe Bay, Hawaii. Coral Reefs 19, 343–357. Tartinville, B., Deleersnijder, E., Rancher, J., 1997. The water residence time in the Mururoa atoll lagoon: sensitivity analysis of a three-dimensional model. Coral Reefs 16, 193–203. Underwood, A.J., 1999. Experiments in Ecology: Their Logical Design and Interpretation using Analysis of Variance. Cambridge University Press. Vadas Sr., R.L., Wright, W.A., Miller, S.L., 1990. Recruitment of Ascophyllum nodosum: wave action as a source of mortality. Mar. Ecol. Prog. Ser. 61, 263–272. Venier, L.A., Fahrig, L., 1996. Habitat availability causes the species abundance–distribution relationship. Oikos 76, 564–570. Vidondo, B., Duarte, C.M., 1998. Population structure, dynamics, and production of the Mediterranean macroalga Codium bursa (Chlorophyceae). J. Phycol. 34, 918–924. Viejo, R.M., Aberg, P., Cervin, G., Lindegarth, M., 1999. The interactive effects of adult canopy, germling density and grazing on germling survival of the rockweed Ascophyllum nodosum. Mar. Ecol. Prog. Ser. 187, 113–120. Williams, I.D., Polunin, N.V.C., 2001. Large-scale associations between macroalgal cover and grazer biomass on mid-depth reefs in the Carribean. Coral Reefs 19, 358–366. Zar, J.H., 1984. Biostatistical Analysis, second ed. Prentice-Hall, Englewood Cliffs, NY.