Skip to main content
NCBI home page
Zoological Studies logoLink to Zoological Studies
. 2016 Oct 26;55:e45. doi: 10.6620/ZS.2016.55-45

Spatial Patterns and Environmental Settings of Non-reefal Coral Communities Across the Tropic of Cancer in the Penghu Archipelago (Pescadores), Taiwan

Hernyi Justin Hsieh 1, Colin Kou-Chang Wen 2,3, Yuan-Chao Huang 4, Kao-Sung Chen 5, Chang-Feng Dai 6,*, Chaolun Allen Chen 2,6,7,*
PMCID: PMC6511904  PMID: 31966190

Abstract

Hernyi Justin Hsieh, Colin Kou-Chang Wen, Yuan-Chao Huang, Kao-Sung Chen, Chang-Feng Dai, and Chaolun Allen Chen (2016) Non-reefal coral communities occurring at the edges of scleractinian coral distribution ranges normally show a sharp gradient in the composition of coral species. Environmental and biological factors such as sea surface temperature (SST), competition with other benthos, and human disturbances might play roles in shaping the structure of coral communities. The Penghu Archipelago is located on the east side of Taiwan Strait straddling the Tropic of Cancer and hosts non-reefal coral communities. In this study, benthic surveys throughout the Penghu Archipelago were conducted and potential environmental and biological factors that shape coral species distributions were inferred by multivariate analyses. A total of 103 species representing 28 genera of scleractinian corals were recorded. Three major ecological sectors (northeast, south, and inner) were defined based on a canonical analysis of the principal coordinates of scleractinian species composition. Correlation analyses showed that scleractinians in the south and northeast sectors were strongly influenced by SST-related variables. In contrast, the coral communities in inner sector were mainly affect by turbidity or nutrition, which supposed come from human activities. Distance-based redundancy analysis showed that benthos, except soft corals, hardly interacted with scleractinian coral distributions. Our study demonstrated a distinct coral species assemblage among different islands across the Tropic of Cancer in the Penghu Archipelago. Natural and human-derived environmental factors both showed a strong correlation with coral species distribution. It’s clear that either natural and human-derived factors influenced coral composition in Penghu archipelago.

Keywords: Non-reefal coral community, Penghu Archipelago, Environmental settings, Tropic of Cancer, Spatial pattern

BACKGROUND

Scleractinian corals are normally associated with warm shallow waters ranging more or less between the Tropics of Cancer (23.5°N) and Capricorn (23.5°S), where coral reef formation is limited by environmental factors like temperature, light, and nutrients (reviewed in Veron 1995; Kleypas et al. 1999). Environmental settings within this geographic range have constantly warm sea surface temperatures (SST ~30°C), high irradiation, and low nutrient levels. Toward the Tropics of Cancer and Capricorn (i.e., the Subtropics), fluctuations in SST and nutrients increases and irradiation decreases, creating dynamic subtropical environments compared to equatorial regions. Within the tropics, coral species richness is relatively high in general, and there is a trend in decreasing species diversity toward the borders of the subtropics (Veron 1995, 2000; Chen 1999; Jones et al. 2011; Dalton and Roff 2013). Beyond the subtropics, the development of coral reefs is significantly reduced due to lower SSTs (< 18°C) and other environmental and/or biological correlates, resulting in marginal “non- reefal” coral communities founded on rocky substrata and characterized by low coral diversity containing a combination of tropical, subtropical, and temperate species, with exceptions like Japan, Australia, and the Red Sea (Veron 1995; Kleypas et al. 1999, 2001; Denis et al. 2013). Coral distribution and geography from Philippines to Japan were clustered into tropical reefs, non- reefal communities and outlying populations, which are closely related to latitudinal and oceanographical gradients (Veron 1995). The definitions of “coral reef”, “coral reef community” and “coral community” have also been discussed. “Coral reef” is a sedimentary structure produced by a “coral reef community”. A “coral reef community” has the potential for reef construction, while a “coral community” does not have that potential (Buddemeier and Hopley 1988).

Coral community structures and environ- mental settings that modulate coral spatial patterns at the local scale across the tropics and the subtropics have been examined over the last few decades. For example, subtropical coral communities in eastern Australia show wide variations in range and latitude in species diversity and living coral cover at scales from 10 to 100 km (Harriott et al. 1994, 1995, 2002). Acropora corals generally are dominant at the Tropic of Capricorn (23.1-23.5°N; e.g., Keppel Islands and Heron Reef) and then decrease in coverage toward higher latitudes (e.g., Solitary Islands; Dalton and Roff 2013). Many environmental factors, such as water temperature, light availability, aragonite saturation, currents, larval dispersal, competition between corals and macroalgae, growth rate, and reproduction operate synergistically to create this latitudinal variation (reviewed in Harriott and Banks 2002). Therefore, corals in subtropical regions with histories of naturally fluctuating environmental factors (e.g., temperature) are suspected to have greater resistance to elevated temperatures compared to tropical counterparts (Chen and Keshavmurthy 2009; Oliver and Palumbi 2011). In addition, marginal high-latitude regions might provide refuges for tropical corals in response to rising seawater temperatures (Hughes et al. 2010; Beger et al. 2011, Baird et al. 2012), although other environmental factors like low pH and high nutrient levels might confine this potential (Huang et al 2011). However, it is difficult to anticipate the possible shift of environmentally disturbed subtropical marginal corals without understanding the interactions between scleractinian coral composition and environmental settings.

The Penghu Archipelago (also known as the Pescadores), located at the eastern side of Taiwan Strait between Taiwan and China, is composed of 90 islands and islets extending 70 km in a north-south direction across the Tropic of Cancer (Fig. 1). The marine environment of the Penghu Archipelago is influenced by the seasonal changes in ocean currents around the western Pacific Ocean (reviewed in Chen 1999; Chen and Keshavmurthy 2009). In summer, the South China Sea Surface Current (SCSSC), driven by the warm southeastern monsoon, flows northward into Taiwan Strait. In winter, while southern Penghu is affected by a weak but warm branch of the Kuroshio Current, northern Penghu is influenced by cold, fresh China Coastal Water (CCW) driven by the northeastern monsoon (Wang and Chern 1998, 1992; reviewed in Chen and Keshavmurthy 2009). Qualitative surveys of coral distributions around Taiwan and Penghu show different compositions of Acropora and Faviidae corals between the southern and other parts of Penghu (Chen 1999). Most of corals in Penghu were also found attached to the bare substrates and were distinguished from coral reefs which are able to accumulate a build-up of calcium carbonate structure (Buddemeier and Smith 1999). These corals in Penghu were considered as non- reefal coral communities and can be interpreted as incoherent growths composed of only a single generation of potential framework builders that do not produce an interlocking framework (Halfar et al. 2005). In addition, this coral community is strongly influenced by human activities including oyster farming, marine cages, and sewage run-off from nearby urban areas, causing high sedimentation and eutrophication in the bay (Huang et al. 2011, 2012). Whether the impacts of regional environmental settings and localized human disturbances affect coral species composition across different islands remains unexplored. Thus, full-scale surveys of corals/benthos and environmental settings were necessary in order to elucidate spatial variations in non-reefal coral communities and potential environmental drivers in the Penghu Archipelago.

Fig. 1.

Fig. 1.

Open in a new tab

Fig. 1. Location of Penghu Archipelago (Pescadores) and 12 sampling sites. 1, Aimen; 2, Chudra; 3, Chinwan; 4, Chungkwang; 5, Chimei; 6, Dawen; 7, Gibei; 8, Gupo; 9, Mudo; 10, Ponpon; 11, Watung; 12, Yupin. Dotted line: Tropic of Cancer; solid lines delineate the four sampling site sectors.

In order to know the distribution of coral communities in Penghu, preliminary surveys were carried out before the sites selection. Besides, possible environment factors that might result in the community structure variations such as temperature, turbidity, sedimentation, water quality (organic content, Chl. a concentration), and biological interactions were also noticed, together with literature surveys, as possible quantitative environmental variables in this study. According to the results of preliminary surveys, a nested hierarchical design was adapted and quantitative surveys of benthic communities at 12 selected sites in 4 geographic sectors (3 sites in each sector) in the Penghu Archipelago were conducted. In addition, environmental data, including seawater temperature, turbidity, chlorophyll-a, and organic matter, as well as other benthic data, were collected and correlated with benthic composition to infer the factors that might drive scleractinian coral community patterns across the Tropic of Cancer in the Penghu Archipelago.

MATERIALS AND METHODS

Sampling sites

Surveys of benthic communities around the Penghu Archipelago were conducted in the summer of 2004. Three random sites were selected to represent four geographic sectors (North, East, South and Inner; Fig. 1) based on qualitative surveys using the manta-tow method from 1998 to 2003. The western sector was excluded due to the unsuitable pebble habitat for corals from pilot surveys. Coral communities of the Penghu Archipelago are limited to the upper 10 m of the water column at most of the studied sites due to its non-reefal characteristics (Hsieh 2008). Thus, six 20 m replicated transects were established to quantify community structure at two depth zones, 3-5 m (upper zone), 6-8 m (lower zone) at each site.

Benthic composition processing

An underwater digital video recorder (Sony TRV-950) was operated by scuba divers to record the images of benthic organisms within a distance of 0.5 m of each transect line. A total of 24 frames (25 cm × 25 cm) were randomly selected from the footages of each transect with iMovie software. Images were analyzed with CPCe (Coral Point Count with Excel extensions) for proportional coverage of benthos (Kohler and Gill 2006). Thirty points were randomly spread onto frames and codes were assigned that corresponded to objects (i.e. all live organisms and substrate types) identified in each frame. Scleractinian corals were identified to species and other benthic organisms to genus if possible. Scleractinian taxonomy followed Veron and Stafford-Smith (2000), Dai and Horng (2009a; 2009b), Budd et al. (2012), and Huang et al. (2014). A total of 144 transects were surveyed and 3456 frames were analyzed. Proportional coverage of scleractinian coral species and other benthic organisms were expressed as spreadsheet data files.

Environmental parameters

Loggers were set at a depth of 5 m at the studied sites to collect seawater temperature in one-hour intervals throughout the year. No significant vertical variations in temperature occur at these locations (Hsieh 2008). Several temperature parameters were generated to examine the environmental variables of the survey sites, including mean temperature (meanT), maximum/minimum daily temperature (maxDT/ minDT), maximum/minimum monthly temperature (maxMT/minMT), and daily/monthly temperature differences (ΔDT and ΔMT). Three replicated water samples (1000 ml) from two depths were collected with metal-free Niskin bottles before conducting underwater surveys. Sampling equipment and processes followed Taiwan EPA quality assurance and quality control protocols (Huang et al. 2012). Water samples were stored at 4°C during transportation to the laboratory for analyses. Turbidity, particulate organic matter (POM), and chlorophyll-a were measured as proxies of human activity. Turbidity was measured with a nephelometer and then filtered through pre- weighted GF/C glass microfiber filters (0.45 μm; Whatman). Half of the collected solids were desalinated by washing with distilled deionized water then dried at 70°C in an oven to constant weights. Microfiber filters were ignited in a furnace at 450°C for 4 h to combust the organic matter, the residual compound representing POM. Planktonic chlorophyll-a was extracted from the remaining half of the filtered solids in 90% aqueous acetone for 24 hrs at 4°C, then measured with a spectrophotometer (Huang et al. 2012).

Statistical analysis

Coral community

Coral similarity among different geographical areas was examined by multivariate analysis under a three-factors design. Geographic factors - sector and zone were selected as fixed factors and site (nested in sector) was selected as a random factor. Present/absent data of scleractinian coral was reassembled according to taxonomic dissimilarity (Theta+; genus and species) prior to multivariate analysis (Clarke et al. 2006). Taxonomic dissi- milarity was considered in this study due to these sampling sites across a large geographical region, and not many same species might be found among sites. In addition, the similarity of species from same genus or family might show similar environment preference and using this dissimilarity index to distinguish these sites around Penghu might be clearly differentiated from human or environmental influences. The 12 sites have been examined by different environmental processes and the difference can not easily be seen by unconstrained ordination method (i.e. MDS; Hsieh 2008). Therefore, a constrained ordination - canonical analysis of principal coordinates (CAP) with a prior hypothesis (factor- sector) was used to separate the sites based on geographical distribution of coral species around Penghu (Anderson and Willis 2003). A three- term model [sector, site (sector), and zone] was used to examine the significance levels of coral composition by permutational multivariate analysis of variance (Anderson 2001). The homogeneity of multivariate variances were verified for all model- terms [sector, site (sector), and zone] by using PERMDISP (permutational analysis of multivariate dispersions, p > 0.05; Anderson 2001). Type III (partial) sums of squares were used because it is more suitable for ecological data with unbalanced design and generate a probability distribution to calculate a p value for each factor. Pairwise tests were also conducted to determine the significant differences among each sector for both zones (upper and lower) to seek if there is an alternative sector based on coral composition besides our arbitrarily group based on geographical location. The scleractinian coral data matrix was analyzed and modeled by DistLM (Distance based Linear Model) for both zones separately to build a relationship model between coral distribution and environmental/benthos predictor variables (Anderson 2004). Environmental and benthos variables have been standardized transformation in prior to DistLM analysis. The significance level of each predictor variable was generated by a marginal test in DistLM. Akaike Information Criterion corrected (AICc) was used to chose the most parsimonious models from all the possible combinations of variables with BEST selection procedure (Anderson 2004). The parsimonious fitted model with only significant and correlated predictors (correlation > 0.2) was visualized by dbRDA (Distance-based redundancy analysis; McArdle and Anderson 2001). All multivariate analyses were conducted in Primer-e (v. 7; Anderson et al. 2008).

Model selection for coral distribution

GLMM (generalized linear mixed model) and GLM were used to examine the responses of each coral genus to geographic (sector, site, and zone) and environmental (abiota and biota) factors to understand which ones determine coral distribution. For geographic factors, sector and zone were selected as fixed factors and site (nested in sector) was selected as a random factor in GLMM. For GLM environmental factors, several temperature-related factors were generalized from temperature data prior to being analyzed with other environmental parameters. Such as mean temperature (meanT), maximum/minimum daily temperature (maxDT/minDT), maximum/minimum monthly temperature (maxMT/minMT), and daily/ monthly temperature differences (ΔDT and ΔMT) as well as water quality parameters (turbidity, organic matter, and chlorophyll-a) were the variable ocean- physical parameters. Four different data distribution models (Poisson, negative binomial, zero-inflated Poisson, and zero-inflated negative binomial) were used to determine the best goodness-of-fit of corals with all possible combination of factors. Four different models were tested due to the possible over-dispersion (high proportion of zeros) that is typical of ecological data and makes zero-inflated models more suitable. Poisson and negative binomial distribution components were chosen because of the possibility for larger variation in our data (Joseph et al. 2009). Factors were examined in a framework of four models using maximum likelihood to estimate parameters and a log-link function was chosen to approximate the above mentioned data distribution models. This approach was taken as no prior assumption of homogeneity was necessary (Adams et al. 2011). The null model (no factor) and alternative models (combination of different factors) were compared with Akaike’s information criterion (AIC; Symonds and Moussalli 2011). Corrected AIC (AICc) was used in this study due to the small number of sample replicates. The model with lowest AICc was selected to represent the best goodness of fit and Akaike weight of models were also calculated for the likelihood (Wagenmakers and Farrell 2004). If there are more than two models with the same AICc value, the parsimony model should be selected due to the extra parameters being uninformative (Arnold 2010). GLMM and GLM modelling and AIC analyses were conducted in R (v. 3.2.2; R Core Team 2015).

RESULTS

A total of 103 species representing 28 genera of scleractinian corals were recorded in our study (Table 1). Coral species richness around Penghu ranged from 7 (upper zone of Watung (11)) to 37 species per site (upper zone of Chudra(2) (Fig. 2)). Acroporidae corals, including Acropora and Montipora were dominant at most sampling sites around Penghu, except Watung (11) which was dominant by Goniopora, Coelastrea, Tubastrea, and Porites (Table 1). The lower zones of Gupo (8) and Chudra (2) were dominated by non- Acroporidae genera like Goniopora, Coelastrea, Turbinaria, Porites, and Echinopora (Table 1). The south sector had a higher number of coral species (45) than East (29) and North sectors (34; Table 1). The Inner sector had the lowest species number (15) and medium live coral cover (LCC; 33%). The lowest LCC was found at the lower zone in Chungkwang (4)(15.84 ± 3.36%), whereas the lower zone in Chimei (5) had the highest LCC (75.53 ± 3.41%). On average, the South sector had a relatively higher LCC (59%) compared to the Inner sector (33%), East sector (31%), and North sector (32%).

Table 1. Coral cover and coral species number of each site around Penghu Archipelago.

Site Region based on geographic locality* Depth zone No. of coral species Live coral cover (%) Top 2 dominant coral species (coverage %) Region based on coral species composition
Watung Inner U 7 19.63 ± 1.50 Goniopora djiboutiensis (18.45%) Goniopora favulus (0.51%) Inner§
Inner L 13 28.76 ± 4.45 Turbinaria frondens (7.10%) Porites lutea (5.91%) Inner§
Chungkwang Inner U 14 47.66 ± 2.59 Acropora muricata (18.68%) Montipora cactus (13.62%) Inner
Inner L 10 15.84 ± 3.36 Montipora cactus (3.28%) Acropora muricata (2.32%) Inner
Chinwan Inner U 15 53.40 ± 2.96 Acropora muricata (20.42%) Montipora cactus (10.07%) Inner
Inner L 8 32.48 ± 4.32 Acropora humilis (15.63%) Montipora cactus (6.50%) Inner
Gibei North U 17 37.82 ± 6.40 Acropora muricata (25.79%) Pocillopora damicornis (2.75%) Northeast
North L 18 17.75 ± 3.89 Acropora muricata (8.22%) Acropora hyacinthus (2.29%) Northeast
Gupo North U 34 61.97 ± 1.99 Acropora muricata (17.31%) Montipora efflorescens (11.39%) Northeast
North L 30 35.90 ± 2.45 Turbinaria mesenterina (5.73%) Acropora solitaryensis (5.01%) Northeast
Mudo North U 20 26.99 ± 3.35 Acropora muricata (10.74%) Acropora monticulosa (6.57%) Northeast
North L 18 40.66 ± 3.40 Acropora muricata (11.89%) Acropora hyacinthus (8.45%) Northeast
Aimen East U 18 48.77 ± 4.44 Acropora muricata (32.29%) Acropora hyacinthus (11.48%) Northeast
East L 26 22.48 ± 2.78 Turbinaria mesenterina (6.37%) Porites solida (2.94%) Northeast
Chudra East U 37 19.86 ± 4.98 Acropora hyacinthus (7.64%) Acropora muricata (2.99%) Northeast
East L 29 16.27 ± 1.94 Echinopora lamellosa (2.48%) Turbinaria mesenterina (2.20%) Northeast
Ponpon East U 19 58.43 ± 2.39 Acropora muricata (27.50%) Acropora hyacinthus (23.75%) Northeast
East L 8 22.80 ± 4.06 Acropora muricata (7.73%) Acropora valida (4.06%) Northeast
Chimei South U 14 75.39 ± 3.09 Acropora intermedia (21.99%) Acrpora muricata (18.84%) South
South L 25 75.53 ± 3.41 Montipora efflorescens (16.62%) Echinopora gemmacea (15.05%) South
Dawen South U 11 62.07 ± 10.17 Acropora muricata (25.03%) Acropora aspera (10.95%) South
South L 22 41.83 ± 5.63 Acropora intermedia (14.03%) Echinopora lamellosa (9.14%) South
Yupin South U 17 52.25 ± 5.66 Acropora intermedia (17.06%) Acropora muricata (16.71%) South¶
South L 22 47.92 ± 8.52 Acropora hyacinthus (16.57%) Acropora muricata (12.22%) South¶
Open in a new tab

Fig. 2.

Fig. 2.

Open in a new tab

Fig. 2. Underwater images of the (A) north sector (Mudo), (B) east sector (Aimen), (C) inner sector (Chinwan), and (D) south sector (Chimei).

Scleractinian coral communities

Scleractinian species composition was significantly different among sectors and sites (within sector), between depth zones, and interactions between zone and site (within sector) (P = 0.001), but there were no significant interactions between zone and sector (P = 0.16, PERMANOVA). Canonical analysis of principal coordinates (CAP) showed that significant differences between sectors (P = 0.001, Fig. 3). Three sites in the South sector were grouped into one cluster, three sites from the Inner sector were scattered, and sites from the North and East sectors generally overlapped (Fig. 3). CAP results were consistent with post hoc pairwise tests in that no significant differences were found between North and East sectors at both depth zones (Table 2). Thereafter, North and East sectors were grouped as a single ecological sector, the Northeast sector, for the analyses. Nevertheless, there was no consistent pattern in coral compo- sition differences between depth zones among other sectors. We separated scleractinian coral composition into upper and lower zones in order to reveal which corals drove the differences in depth zones among sites and sectors. Fewer coral species were driving coral composition differences in the upper zone compared to the lower zone in the dbRDA analysis (Fig. 4). In the upper zone (Fig. 4A), Goniopora dijiboutiensis was dominant at Watung (11) and Chiwan (3) (inner sector), while Acropora intermedia, Echinopora gemmacea, and Montipora stellata were common in most South sector transects. Acropora valida and Acropora hyacinthus were the main contributors to the scleractinian coral community in the Northeast sector. In contrast, the composition of lower-zone corals were more scattered and not consistent with ecological sectors (Fig. 4B). The lower zone of the Inner sector had more Acropora humilis, Montipora cactus, Lithophyllon undulatum, and Cyphastrea chalcidicum. The South sector and Gibei (7) and Mudo (9) of the Northeast sector had abundant Montipora aequituberculata, Acropora hyacinthus, and Echinopora lamellosa. Many other scleractinian coral species, such as Turbinaria peltata, Tubinaria mesenterina, Platygyra sinensis, Porites lutea, Dipsastraea maxima, and Coelastrea aspera, lead the different groups in the Northeast sector into different directions.

Fig. 3.

Fig. 3.

Open in a new tab

Fig. 3. Canonical analysis of principal coordinates (CAP) with a prior hypothesis (four different sectors: inner, north, east, and south) for coral communities among Penghu (Eigenvalues of correlation: 1 = 0.9284, 2 = 0.7247; Permutation test: P = 0.001). The dash circles indicated the significant differences from PERMANOVA pairwise tests.

Table 2.Post hoc pairwise PERMANOVA test and average similarity distances results of coral composition at both upper zone (lower-left triangle) and lower zone (upper-right triangle) among sectors.

East Inner South North
East 63.917n.s. 62.934* 52.916n.s.
Inner 65.346* 74.746* 65.558n.s.
South 58.961* 72.443* 57.269*
North 49.988n.s. 65.976* 53.38n.s.
Open in a new tab

*: p < 0.05, n.s.: not significant.

Fig. 4.

Fig. 4.

Open in a new tab

Fig. 4. Distance-based redundancy analysis (dbRDA) of coral communities with coral species as variables at (A) upper and (B) lower zones among three different sectors at Penghu. Only significantly related coral species with correlations > 0.2 are included on the plot.

Scleractinian coral composition correlated with environmental parameters

Environmental parameters from four geographic sectors around Penghu are listed in Table S1. The mean temperature around Penghu is from 23.12°C in the Inner sector to 23.82°C in the East and South sectors. Significance and correlations of coral composition and environmental factors among sites were examined based on the Distance-based Linear Model (DistLM; Table 3).

Table 3. Results of variables from marginal test of DistLM on (a) environmental parameters and (b) benthos (biota and abiota) of both upper and lower zones.

(a).
Upper zone Lower zone
Variable Pseudo-F p Pseudo-F p
MeanT 6.1974 0.001* 8.8863 0.001*
minDT 1.867 0.102 3.6444 0.003*
MAXDT 4.8569 0.002* 5.2645 0.001*
minMT 13.117 0.001* 14.534 0.001*
MAXMT 7.2105 0.001* 3.2051 0.002*
ΔDT 8.5159 0.001* 6.3117 0.001*
ΔMT 10.989 0.001* 8.0842 0.001*
Turbidity 4.5505 0.003* 13.342 0.001*
Organic matter 1.4266 0.178 2.8982 0.016*
Chlorophyll a 18.651 0.001* 17.662 0.001*
Open in a new tab

*symbol indicates the significant correlation of factors among sampling sites, p < 0.05.

(b).
Upper zone Lower zone
Variable Pseudo-F p Pseudo-F p
Soft corals 4.0614 0.005* 2.9484 0.008*
Sponges 1.2298 0.252 1.5485 0.166
Zoanthids 1.3335 0.182 2.033 0.054
Macroalgae 0.7591 0.523 1.0681 0.391
Other benthos 1.4544 0.183 2.2112 0.033*
Dead coral (with algae) 9.1123 0.001* 5.84 0.001*
Coralline algae 2.3246 0.039* 0.51965 0.880
Diseased corals No test 0.77924 0.675
Abiota (sand, pavement, rubble) 1.4306 0.187 1.5178 0.159
Open in a new tab

Scleractinian composition in upper zones significantly correlated with temperature-related variables, including MeanT, MAXDT, minMT, MAXMT, ΔDT, ΔMT and water quality such as turbidity and chlorophyll a (Table 3a). ΔMT was higher with corals of inner sector minMT was higher with most corals in the Northeast sector except Ponpon (10) and Gupo (8) (Fig. 5A). Other temperature factors such as MeanT, MAXDT, MAXMT, ΔDT were all higher with corals of south sector which is also low latitude. Higher chlorophyll-a was with corals in the Inner sector and higher turbidity with the Northeast sector, especially at Ponpon (10). In lower zone of reefs, all the temperature and water environmental factors were significant with coral distribution (Table 3a). The temperature correlations were similar to those in upper zones, except that MeanT were higher in Aimen (1) and Chudra (2) in the Northeast sector (Fig. 5B). Water quality factors - chlorophyll-a and turbidity were higher at most sites in the Inner sector. Overall, temperature parameters were consistent with the latitudinal distribution of sites in the archipelago, except that temperature variations at some sites in the Inner and Northeast sectors were higher. Meanwhile, turbidity and chlorophyll-a (nutrients) were consistently higher in the Inner sector and higher turbidity in upper zone or northwest sites (Fig. 5B).

Fig. 5.

Fig. 5.

Open in a new tab

Fig. 5. dbRDA of coral communities with environmental variables at (A) upper and (B) lower zones among three different sectors at Penghu. Only significantly related environmental variables with correlations > 0.2 are included on the plot.

Correlation of scleractinians with other benthos

Only a couple benthos and abiota, such as soft corals and dead coral with algae, at both zones were significantly correlated with corals around the Penghu archipelago (Table 3b). In addition, coralline algae coverage were significantly higher with the corals in the upper zone, other benthos including polychaetes, bryozoans, echinoderms, bivalves, anemones, and gastropods, were significantly more abundant in the lower zones (Table 3b). The dbRDA showed similar pattern that soft corals had higher coverage significantly in the south sector and dead coral were more abundant in other sectors in upper zone (Fig 6A). Higher coralline algae coverage was more abundant in few places only, such as Ponpon (10), Mudo (9), and Yupin (12). Contrarily, soft corals were more abundant in lower zone of Chudra (2), Mudo (9), and Gibei (7) in the Northeast sector (Fig. 6B). Other benthos were more abundant in South sector, while dead corals were consistently abundant in both zones of the Inner and in upper zone of Northeast sector.

Fig. 6.

Fig. 6.

Open in a new tab

Fig. 6. dbRDA of coral communities with benthos/abiota variables at (A) upper and (B) lower zones among three different sectors at Penghu. Only significantly related benthos/abiota with correlations > 0.2 are included on the plot.

Coral genus distribution based on abiotic/ biotic factors

Each genus was modeled with geographic factors as well as coverage of other biotic and abiotic data. The distribution of all coral genera seemed to be affected by three factors: site (nested in sector), sector, and depth, except Euphyllia and Stylocoeniella (Table 4). In addition, 16 of 25 genera in our survey showed a preference of either the upper zone or bottom zone of the reefs at Penghu. Few genera, such as Acanthastrea, Echinopora, Dipsastraea, Favites, Hydnophora, Pocillopora, Stylophora were equally abundant in both depth zones with some preference of certain collecting site around Penghu.

Table 4. Summary of GLMM/GLM selection for coral genera distribution with geographic factors. The best goodness-of-fit models were selected from lowest Akaike Information Criterion correction(AICc) among all the geographic factors combination (Site(nested in Sector), Zone and Sector). Akaike weight indicates the likelihood of given model as best goodness-of-fit model. Total coral cover of each genus showed the contribution of each genus in Penghu coral reefs. #indicates the best model also has factor “sector” but with same AICc and Akaike weight.

Genera Model structure AICc Akaike weight (%) of total coral cover
Acanthastrea Coral cover ~Site(Sector)# 365.11 0.385 0.88
Acropora Coral cover ~Site(Sector)+Zone# 1611.93 0.500 56.37
Coscinaraea Coral cover ~Site(Sector)+Zone# 62.26 0.476 0.03
Cyphastrea Coral cover ~Site(Sector)+Zone 411.74 0.489 0.79
Echinophyllia Coral cover ~Site(Sector)+Zone# 348.01 0.303 0.78
Echinopora Coral cover ~Site(Sector) # 571.71 0.386 4.98
Euphyllia Coral cover ~NULL 143.65 1.000 0.28
Dipsastraea Coral cover ~Site(Sector) # 448.43 0.382 0.88
Favites Coral cover ~Site(Sector) # 667.95 0.386 2.19
Galaxea Coral cover ~Site(Sector)+Zone# 316.23 0.312 0.49
Goniastrea Coral cover ~Site(Sector)+Zone# 583.93 0.296 1.48
Goniopora Coral cover ~Site(Sector)+Zone# 320.98 0.500 2.79
Hydnophora Coral cover ~Sector 196.60 0.397 0.43
Lithophyllon Coral cover ~Site(Sector)+Zone# 240.09 0.477 0.68
Phymastrea Coral cover ~Zone 109.73 0.698 0.13
Montipora Coral cover ~Site(Sector)+Zone# 1078.83 0.256 13.50
Mycedium Coral cover ~Site(Sector)+Zone# 293.50 0.500 0.69
Pavona Coral cover ~Site(Sector)+Zone# 415.80 0.415 0.95
Platygyra Coral cover ~Site(Sector)+Zone# 466.93 0.268 0.97
Pocillopora Coral cover ~Site(Sector)# 789.90 0.387 2.71
Porites Coral cover ~Zone 799.42 0.365 3.47
Psammocora Coral cover ~Zone 70.08 0.716 0.89
Stylocoeniella Coral cover ~NULL 44.61 0.998 0.02
Stylophora Coral cover ~Sector 246.48 0.462 0.86
Turbinaria Coral cover ~Site(Sector)+Zone# 55.99 0.499 3.53
Open in a new tab

Most coral genera showed a negative trend (negative coefficient in the model) with most benthos in the GLM model (Table 5). No responses in abiota and diseased corals were detected relative to all scleractinian genera. Several scleractinian genera, such as Pocillopora, Galaxea, Euphyllia, and Coscinaraea, had neither positive nor negative responses to all benthic substrata. Dead coral with algae showed half negative and half positive responses to many coral genera (15/28). There were only a very few positive responses occurring for Acropora vs. coralline algae, Goniopora vs. macroalgae, and Stylocoeniella vs. other benthos.

Table 5. Summary of GLM for coral genera distribution with abiotic and biotic benthos. Symbols – and + indicates the best fitting model structure with given biota or abiota factors and either with negative coefficient and positive coefficient in the selected model. X in NULL column indicated no factor was detected in the best goodness-of-fit model. Akaike weight indicates the likelihood of given model as best goodness-of-fit model.

Genera Soft coral Sponge Zoanthids Macroalgae Other benthos Dead coral with algae Coralline algae
Acropora − +
Montipora − − −
Echinopora −
Turbinaria − + −
Porites −
Goniopora + − −
Pocillopora
Favites − − −
Goniastrea −
Platygyra − − −
Pavona + −
Psammocora -
Acanthastrea − − − −
Favia − −
Stylophora −
Cyphastrea − − − +
Echinophyllia − −
Mycedium − +
Lithophyllon − +
Galaxea
Hydnophora −
Euphyllia
Montastrea +
Coscinaraea
Stylocoeniella +
Genera Diseased coral Abiota NULL AICc Akaike weight (%) of total coral cover
Acropora 1145.77 0.195 56.37
Montipora 718.12 0.251 13.50
Echinopora 403.35 0.206 4.98
Turbinaria 359.27 0.106 3.53
Porites 426.08 0.093 3.47
Goniopora 232.75 0.070 2.79
Pocillopora X 408.20 0.063 2.71
Favites 357.13 0.148 2.19
Goniastrea 596.99 0.079 1.48
Platygyra 230.89 0.120 0.97
Pavona 227.64 0.055 0.95
Psammocora 35.60 0.097 0.89
Acanthastrea 207.35 0.081 0.88
Favia 215.16 0.078 0.88
Stylophora 164.62 0.154 0.86
Cyphastrea 195.07 0.095 0.79
Echinophyllia 184.09 0.060 0.78
Mycedium 152.08 0.103 0.69
Lithophyllon 136.34 0.063 0.68
Galaxea X 145.35 0.080 0.49
Hydnophora 106.96 0.142 0.43
Euphyllia X 82.61 0.079 0.28
Montastrea 58.68 0.037 0.13
Coscinaraea X 13.97 0.075 0.03
Stylocoeniella 12.97 0.087 0.02
Open in a new tab

DISCUSSION

Two main biogeographic distribution patterns were found in corals at Penghu Archipelago. One was a latitudinal distribution pattern from the south sector to the east sector to the north sector, which was related to seawater temperature and sea currents connecting Penghu to nearby island groups. Another distribution pattern was the differences among the Inner and other sectors, which correlated with natural/human disturbances and water circulation within the Inner sector. In addition, we did not detect a strong interaction between corals and other benthic invertebrates and substrata from our data. Only a few benthos were consistent with the biogeographic distributions of corals at Penghu.

Latitudinal distribution of corals and environmental settings

Coral coverage and coral species number around Penghu were generally consistent with latitude. Coral coverage and species number decreased from tropical (south sector) to subtropical areas (north and east sectors) at Penghu. Compared to other places in Taiwan, numbers of coral species within 70 km of Penghu (103 species) were higher than in subtropical northern Taiwan (~60 species) but lower than in tropical southern Taiwan (~300 species; Dai 1991; Hsieh 2008). Compared to other islands around the northwestern Pacific Ocean, Penghu has lower richness than the Yaeyama Islands (368 species) and Miyako Islands (249 species) of the Okinawa Archipelago (Kayanne et al. 2012). Similar latitudes in the southern hemisphere and in the southern Great Barrier Reef of Australia (400 km) have about 144 coral species in inshore reefs (De Vantier et al. 2006) and 244 species in whole region. However, the Great Keppel Island group (23°S) at the edge of the tropical zone has only 40+ coral species (Sweatman et al. 2007). The latitudinal decrease in coral species number from southern Taiwan, Penghu, and northern Taiwan is comparable to eastern Australia and the southwestern Pacific Ocean (Harriott and Banks 2002), even though the distance between north and south Taiwan is significantly shorter than that of east Australia. Oceanographical variations such as convergence of Kuroshio Current and China coastal current as well as seasonal monsoon are the possible major factors (Jan et al. 2002) which result in a sharp temperature changes and coral distributions in Penghu. In addition, coral coverage around Penghu is compatible to marginal corals of Australia (30-60%) suggested the possible consistency of marginal coral communities at western Pacific region. (Harriott and Banks 2002; De Vantier et al. 2006; Sweatman et al. 2007).

The coral community in the south sector of Penghu is separated from the northeast sector because corals in the south sector are normally associated with warm, clear water and less temperature fluctuation. For example, Stylophora pistillata, Pocillopora eydouxi, P. meandrina, and Acropora intermedia, which are typical tropical species strongly associated with clear and warm waters, were only found in the south sector. Conversely, the coral community of the upper northeast sector was dominated by subtropical species such as Turbinaria mesenterina, T. frondens, Mycedium elephantotus, and Echinopora lamellosa, which are cup-shaped or encrusting morphs that have the advantages of increased sediment removal efficiency and increased surface area for light absorption (Riegl et al. 1996). Other dominant species in the Northeast region have relatively robust structures (Acropora solitaryensis and A. monticulosa) or massive and encrusting growth forms (Montipora efflorscens, Favites halicora) for protection against the physical impacts of seasonal monsoons. In addition, fast- growing species like A. muricata, A. hyacinthus, and A. valida can recruit to available spaces between disturbances. The rapidly changing environments between the northeast and south sectors are related to geological location, ocean- physical conditions, and topography. The Penghu Archipelago is located at the intersection of three sea currents, which are the CCW, SCSSC, and a branch of the Kuroshio Current (Chen 1999). CCW brings cold water to the northeastern region of Penghu, while the SCSSC and the Kuroshio branch bring warm waters from tropical areas. These currents probably also connect the coral communities of Penghu with Japan, South China, and Philippines as well as Taiwan (reviewed Chen and Keshavmurthy 2009). Further investigations of genetic connectivity across island groups will help to reveal the patterns in this area.

Human disturbances to corals in the Inner sector

The Inner sector had the lowest coral species number and living coral coverage compared to other sectors (South and Northeast). Marginal tests with the Distance-based Linear Model suggested that the highest temperature and high variations in temperature, turbidity, and nutrients might affect coral species composition in the Inner sector, which is surrounded by three major islands and characterized by shallow water (depth mostly < 15 m) and limited circulation with outside bay waters (Fig. 1). Coastal urbanization plus industrial scale oyster farming and marine cage culture stretching from the intertidal zone to deeper waters in the Inner sector has resulted in the discharge of sewage, modification of benthic fauna, and eutrophication derived from feed debris and feces of cultivated fish (Huang et al. 2011). These may account for the high levels of turbidity, sedimentation, and nutrients in the seawater of the Inner sector (Huang et al. 2011; 2012). Consequently, only coral genera like Galaxea, Goniopora, Cyphastrea, and Montipora that can tolerate these stressors colonize the limited substructure of the Inner sector. These genera can either remove the sediments efficiently by secreting mucus (Brown and Bythell 2005) and polyp sweeping (Stafford-Smith 1993), or may benefit from the dissolved or suspended inorganic/ organic nutrient loads (Bongiorni et al. 2003).

Relationship between corals and other marine organisms on coral reefs

We did not see many benthic invertebrates or abiota substrata relating to coral distribution around Penghu except for soft corals, which were consistently correlated with hard corals in both zones. Soft corals (alcyonacean) are quite abundant in tropical/subtropical areas around Taiwan (Dai 1991). Interactions and competition between soft and hard corals are diverse and depend on the species involved (Dai 1990; Maida et al. 2001). The differing competitive hierarchy of each hard and soft coral species might also determine the distribution of both coral types around Penghu (Dai 1993). However, the limited ability to identify soft corals at Penghu (but see Benayahu et al. 2012), especially from photo images, makes it difficult to understand the possible interactions of hard and soft corals in this area. Macroalgae is quite common as a benthos driver for coral distribution in most tropical and subtropical areas (McCook 1999; Hoey et al. 2011), especially eutrophic environments (McCook et al. 2001) like the Inner sector in the present study. Surprisingly, we did not see macroalgae as a significant driver of coral distribution around Penghu, which might be due to fish herbivory. Macroalgae plays an important role in the competition between coral and algae (Ceccarelli et al. 2005; Wismer et al. 2009). It is necessary to include reef-associated fish surveys in the future to understand the interactions among corals, macroalgae, and herbivorous fishes.

In conclusion, spatial variation in coral species diversity and assemblages was detected between the Penghu Archipelago and other islands distributed along a 70 km transect oriented north-to-south across the Tropic of Cancer in Taiwan Strait. Environmental factors such as sea surface temperature, turbidity, and nutrients showed significant correlations with coral species distribution in different sectors, suggesting that these environmental factors might account for the formation of coral assemblage patterns at Penghu. Considering the increasing threats from local disturbances (Hsieh et al. 2011) and rising global sea surface temperatures, our study provides an important baseline for the management and conservation of the unique non-reefal coral com- munities of Taiwan.

Acknowledgments

Acknowledgement: This article was based on the data in PhD dissertation of Hsieh (2008). Coral and environmental data can be acquired from first and corresponding authors. Special thanks to members of the Coral Reef Evolutionary Ecology and Genetics Laboratory (CREEG), Biodiversity Research Center, Academia Sinica (BRCAS), for assistance during countless field trips. Many thanks to colleagues at Penghu Marine Biology Research Center (PMBRC) for full administrative support. This study was funded by research grants from the National Science Council, Taiwan, and BRCAS to CAC and PMBRC to HJH.

References

  1. Adams VM, Mills M, Jupiter SD, Pressey RL. 2011. Improving social acceptability of marine protected area networks: A method for estimating opportunity costs to multiple gear types in both fished and currently unfished areas. Biol Conserv 144:350-361. doi:10.1016/j.biocon.2010.09.012.
  2. Anderson MJ. 2001. A new method for non parametric multivariate analysis of variance. Austral Ecology 26:32- 46. doi:10.1111/j.1442-9993.2001.01070.pp.x.
  3. Anderson MJ. 2004. DISTLM v. 5: a FORTRAN computer program to calculate a distance-based multivariate analysis for a linear model. Department of Statistics, University of Auckland, New Zealand 10.
  4. Anderson MJ, Gorley R, Clarke K. 2008. PERMANOVA+ for PRIMER: guide to software and statistical methods. Primer-e, Plymouth, UK 237.
  5. Anderson MJ, Willis TJ. 2003. Canonical analysis of principal coordinates: a useful method of constrained ordination for ecology. Ecology 84(2):511-525. doi:10.1890/0012- 9658(2003)084[0511:CAOPCA]2.0.CO;2.
  6. Arnold TW. 2010. Uninformative parameters and model selection using Akaike’s information criterion. J Wildlife Manage 74:1175-1178. doi:http://dx.doi.org/10.2193/2009- 367.
  7. Baird AH, Sommer B, Madin JS. 2012. Poleward range expansion of Acropora spp. along the east coast of Australia. Coral Reefs 31:1063. doi:10.1007/s00338-012- 0928-6.
  8. Beger M, Babcock R, Booth DJ, Bucher D, Condie SA, Creese B, Cvitanovic C, Dalton SJ, Harrison P, Hoey A, Jordan A, Loder J, Malcolm H, Purcell W, Roelfsma C, Sachs P, Smith SDA, Sommer B, Stuart-Smith R, Thomson D, Wallace CC, Zann M, Pandolfi JM. 2011. Research challenges to improve the management and conservation of subtropical reefs to tackle climate change threats. Ecol Manage Restor 12:e7-e10. doi:10.1111/j.1442- 8903.2011.00573.x.
  9. Benayahu Y, van Ofwegen LP, Dai C-f, Jeng M-S, Soong K, Shlagman A, Hsieh HJ, McFadden CS. 2012. Diversity, Distribution, and Molecular Systematics of Octocorals (Coelenterata: Anthozoa) of the Penghu Archipelago, Taiwan. Zool Stud 51:1529-1548.
  10. Bongiorni L, Shafir S, Angel D, Rinkevich B. 2003. Survival, growth and gonad development of two hermatypic corals subjected to in situ fish-farm nutrient enrichment. Mar Ecol-Prog Ser 253:137-144. doi:10.3354/meps253137.
  11. Brown B, Bythell J. 2005. Perspectives on mucus secretion in reef corals. Mar Ecol-Prog Ser 296:291-309. doi:10.3354/ meps296291.
  12. Budd AF, Fukami H, Smith ND, Knowlton N. 2012. Taxonomic classification of the reef coral family Mussidae (Cnidaria: Anthozoa: Scleractinia). Zool J Linn Soc 166:465-529. doi:10.1111/j.1096-3642.2012.00855.x.
  13. Buddemeier RW, Hopley D. 1988. Turn-ons and turn-offs: Causes and mechanisms of the initiation and termination of coral reef growth. Proc. 6th Int. Coral Reef Symp 2:485-490.
  14. Buddemeier RW, Smith SV. 1999. Coral adaptation and accli- matization: A most ingenious paradox. Am Zool 39:1-9. doi:http://dx.doi.org/10.1093/icb/39.1.1 1-9.
  15. Ceccarelli D, Jones G, McCook L. 2005. Effects of territorial damselfish on an algal-dominated coastal coral reef. Coral Reefs 24:606-620. doi:10.1007/s00338-005-0035-z.
  16. Chen CA. 1999. Analysis of scleractinian distribution in Taiwan indicating a pattern congruent with sea surface temperatures and currents: examples from Acropora and Faviidae corals. Zool Stud 38:119-129.
  17. Chen CA, Keshavmurthy S. 2009. Taiwan as a connective stepping-stone in the Kuroshio Triangle and the conservation of coral ecosystems under the impact of climate change. Kuroshio Science 3:15-22.
  18. Clarke KR, Somerfield PJ, Chapman MG. 2006. On resemblance measures for ecological studies, including taxonomic dissimilarities and a zero-adjusted Bray–Curtis coefficient for denuded assemblages. J Exp Mar Biol Ecol 330:55-80. doi:10.1016/j.jembe.2005.12.017.
  19. Dai C-F. 1990. Interspecific competition in Taiwanese corals with special reference to interactions between alcyonaceans and scleractinians. Mar Ecol-Prog Ser Oldendorf 60:291- 297.
  20. Dai C-F. 1991. Distribution and adaptive strategies of alcyonacean corals in Nanwan Bay, Taiwan. Hydrobiologia 216:241-246.
  21. Dai C-F. 1993. Patterns of Coral Distribution and Benthic Space Partitioning on the Fringing Reefs of Southern Taiwan. Mar Ecol 14:185-204. doi:10.1111/j.1439-0485.1993. tb00479.x.
  22. Dai C, Horng S. 2009a. Scleractinia fauna of Taiwan. I. The robust group. Taipei, Taiwan. National Taiwan University, 162 pp.
  23. Dai C, Horng S. 2009b. Scleractinia fauna of Taiwan. II. The complex group. Taipei, Taiwan. National Taiwan University, 172 pp.
  24. Dalton SJ, Roff G. 2013. Spatial and temporal patterns of eastern Australia subtropical coral communities. PLoS ONE 8(9):e75873. doi:10.1371/journal.pone.0075873. [DOI] [PMC free article] [PubMed]
  25. Denis V, Chen CA, Song JI, Woo S. 2013. Alveopora japonica beds thriving under kelp. Coral Reefs 32:503. doi:10.1007/ s00338-013-1019-z.
  26. De Vantier L, De’Ath G, Turak E, Done T, Fabricius K. 2006. Species richness and community structure of reef-building corals on the nearshore Great Barrier Reef. Coral Reefs 25:329-340. doi:10.1007/s00338-006-0115-8.
  27. Halfar J, Godinez-Orta L, Riegl B, Valdez-Holguin JE, Borges JM. 2005. Living on the edge: high-latitude Porites carbonate production under temperate eutrophic conditions. Coral Reefs 24(4):582-592. doi:10.1007/ s00338-005-0029-x.
  28. Harriott V, Banks S. 2002. Latitudinal variation in coral communities in eastern Australia: a qualitative biophysical model of factors regulating coral reefs. Coral Reefs 21:83- 94. doi:10.1007/s00338-001-0201-x.
  29. Harriott VJ, Harrison PL, Banks SA. 1995. The coral communities of Lord Howe island. Mar Freshw Res 46:457-465. doi:10.1071/MF9950457.
  30. Harriott VJ, Smith SDA, Harrison PL. 1994. Patterns of coral community structure of subtropical reefs in the Solitary Islands marine reserve, eastern Australia. Mar Ecol-Prog Ser 109:67-76.
  31. Hoey AS, Pratchett MS, Cvitanovic C. 2011. High macroalgal cover and low coral recruitment undermines the potential resilience of the world’s southernmost coral reef assemblages. PLoS One 6:e25824. doi:http://dx.doi. org/10.1371/journal.pone.0025824. [DOI] [PMC free article] [PubMed]
  32. Hsieh HJ. 2008. Studies on subtropical scleractinian coral communities in Penghu Archipelago. PhD dissertation, National Taiwan Univ., Taipei, Taiwan, 206 pp.
  33. Hsieh HJ, Chen KS, Lin YI, Huang YA, Baird AH, Tsai WS, Su WC, Chen CA. 2011. A No-take Area (NTA) is no guarantee for preserving coral communities in the Chinwan Inner Bay, Pescadores, Taiwan. Zool Stud 50:443-453.
  34. Hsieh HJ, Wei NV, Lu YL, Jeng MS, Tsai WS, Chen CA. 2001. Unexpectedly high coral coverage in Chinwan Inner Bay, Pescadores: a proposed site for a marine protection area. Coral Reefs 20:316-317. doi:10.1007/s003380100169.
  35. Huang D, Benzoni F, Fukami H, Knowlton N, Smith ND, Budd AF. 2014. Taxonomic classification of the reef coralfamilies Merulinidae, Montastraeidae and Diploastraeidae (Cnidaria: Anthozoa: Scleractinia). Zool J Linn Soc 171:277-355. doi:10.1111/zoj.12140.
  36. Huang Y-C, Hsieh HJ, Huang S-C, Meng P-J, Chen Y-S, Keshavmurthy S, Nozawa Y, Chen CA. 2011. Nutrient enrichment caused by marine cage culture and its influence on subtropical coral communities in turbid waters. Mar Ecol-Prog Ser 423:83-93. doi:10.3354/ meps08944.
  37. Huang Y-C A, Huang S-C, Hsieh HJ, Meng P-J, Chen CA. 2012. Changes in sedimentation, sediment characteristics, and benthic macrofaunal assemblages around marine cage culture under seasonal monsoon scales in a shallow- water bay in Taiwan. J Exp Mar Biol Ecol 422-423:55-63. doi:10.1016/j.jembe.2012.04.008.
  38. Hughes TP, Graham NAJ, Jackson JBC, Mumby PJ, Steneck RS. 2010. Rising to the challenge of sustaining coral reef resilience. Trends Ecol Evol 25:633-642. doi:10.1016/ j.tree.2010.07.011. [DOI] [PubMed]
  39. Jan S, Wang J, Chern CS, Chao SY. 2002. Seasonal variation of the circulation in the Taiwan Strait. J Mar Syst 35(3):249-268. doi:10.1016/S0924-7963(02)00130-6.
  40. Jones AM, Berkelmans R, Houston W. 2011. Species richness and community structure on a high latitude reef: implications for conservation and management. Diversity 3:329-355. doi:10.3390/d3030329.
  41. Joseph LN, Elkin C, Martin TG, Possingham HP. 2009. Modeling abundance using N-mixture models: the importance of considering ecological mechanisms. Ecological Applications 19(3):631-642. doi:10.1890/07- 2107.1. [DOI] [PubMed]
  42. Kayanne H, Hongo C, Okaji K, Ide Y, Hayashibara T, Yamamoto H, Mikami N, Onodera K, Ootsubo T, Takano H, Tonegawa M, Maruyama S. 2012. Low species diversity of hermatypic corals on an isolated reef, Okinotorishima, in the northwestern Pacific. Galaxea, Journal of Coral Reef Studies 14:73-95.
  43. Kleypas JA, McManus JW, Menez LAB. 1999. Environmental limits to coral reef development: where do we draw the line? Am Zool 39:146-159.
  44. Kleypas JA, Buddemeier RW, Gattuso J-P. 2001. The future of coral reefs in an age of global change. Int J Earth Sci 90:426-437.
  45. Kohler KE, Gill SM. 2006. Coral Point Count with Excel extensions (CPCe): A Visual Basic program for the determination of coral and substrate coverage using random point count methodology. Comput Geosci 32:1259-1269.
  46. Maida M, Sammarco PW, Coll JC. 2001. Effects of Soft Corals on Scleractinian Coral Recruitment. II: Allelopathy, Spat Survivorship and Reef Community Structure. Mar Ecol 22:397-414. doi:10.1046/j.1439-0485.2001.01709.x.
  47. McArdle BH, Anderson MJ. 2001. Fitting multivariate models to community data: a comment on distance- based redundancy analysis. Ecology 82:290-297. doi:10.2307/2680104.
  48. McCook L. 1999. Macroalgae, nutrients and phase shifts on coral reefs: scientific issues and management consequences for the Great Barrier Reef. Coral Reefs 18:357-367. doi:10.1007/s003380050213.
  49. McCook L, Jompa J, Diaz-Pulido G. 2001. Competition between corals and algae on coral reefs: a review of evidence and mechanisms. Coral Reefs 19:400-417. doi:10.1007/ s003380000129.
  50. Oliver TA and Palumbi SR. 2011. Do fluctuating temperature environments elevate coral thermal tolerance? Coral Reefs 30:429-440. doi:10.1007/s00338-011-0721-y.
  51. R Core Team. 2015. R: A language and environment for statistical computing. R Foundation for Statistical Computing Vienna Austria.
  52. Riegl B, Heine C, Branch G. 1996. Function of funnel-shaped coral growth in a high-sedimentation environment. Mar Ecol-Pro Ser 145:87-93. doi:10.3354/meps145087.
  53. Smith SV, Buddemeier RW. 1992. Global change and coral reef ecosystems. Annu Rev Ecol Syst 23:89-118.
  54. Stafford-Smith MG. 1993. Sediment-rejection efficiency of 22 species of Australian scleractinian corals. Mar Biol 115:229-243. doi:10.1007/BF00346340.
  55. Sweatman HH, Thompson AA, Delean SS, Davidson JJ, Neale SS. 2007. Status of near-shore reefs of the Great Barrier Reef 2004. Marine and Tropical Sciences Research Facility Research Report Series. Reef and Rainforest Research Centre Limited, Cairns (169 pp.).
  56. Symonds MRE, Moussalli A. 2011. A brief guide to model selection, multimodel inference and model averaging in behavioural ecology using Akaike’s information criterion. Behav Ecol Sociobiol 65:13-21. doi:10.1007/s00265-010- 1037-6.
  57. Veron JEN. 1995. Corals In space and time: The biogeography and evolution of the scleractinia. Cornell University Press pp. 321.
  58. Veron JEN, Stafford-Smith M. 2000. Corals of the World. Australian Institute of Marine Science, Townsville.
  59. Wagenmakers EJ, Farrell S. 2004. AIC model selection using Akaike weights. Psychon B Rev 11:192-196. doi:10.3758/ BF03206482. [DOI] [PubMed]
  60. Wang J, Chern CS. 1988. On the Kuroshio branch in the Taiwan Strait during wintertime. Progressive Oceanography 21:469-491. doi:10.1016/0079-6611(88)90022-5.
  61. Wang J, Chern CS. 1992. On the distribution of bottom cold waters in Taiwan Strait during summertime. La mer 30:213-221.
  62. Wismer S, Hoey AS, Bellwood DR. 2009. Cross-shelf benthic community structure on the Great Barrier Reef: relationships between macroalgal cover and herbivore biomass. Mar Ecol-Pro Ser 376:45-54. doi:10.3354/ meps07790.

Articles from Zoological Studies are provided here courtesy of Biodiversity Research Center, Academia Sinica

RESOURCES