Density-associated recruitment in octocoral communities in St. John, US Virgin Islands

Journal of Experimental Marine Biology and Ecology 473 (2015) 103–109

Contents lists available at ScienceDirect

Journal of Experimental Marine Biology and Ecology journal homepage: www.elsevier.com/locate/jembe

Density-associated recruitment in octocoral communities in St. John, US Virgin Islands Kristin Privitera-Johnson a, Elizabeth A. Lenz b,c, Peter J. Edmunds b,⁎ a b c

Department of Biology, California State University, 1250 Bellflower Blvd, Long Beach, CA 90840, USA Department of Biology, California State University, 18111 Nordhoff Street, Northridge, CA 91303-8303, USA Hawai'i Institute of Marine Biology, University of Hawai'i, PO Box 1346, Kāne'ohe, HI 96744, USA

a r t i c l e

i n f o

Article history: Received 2 April 2015 Received in revised form 16 August 2015 Accepted 17 August 2015 Available online 1 September 2015 Keywords: Caribbean Coral reefs Soft corals

a b s t r a c t To evaluate the possibility that density-associated effects modulate octocoral abundance on a Caribbean coral reef, we tested the hypothesis that the density of octocoral recruits (colonies ≤ 4 cm tall) and adult colonies are positively associated on shallow reefs (≤14 m depth) in St. John, US Virgin Islands. Both life stages were censused for density at 8–10 sites along 7 km of shore in 2013 and 2014, and a correlative approach was used to evaluate the extent to which the densities were associated using sites as replicates. For 8 sites censused in both years, mean densities of adults (pooled among taxa) varied from 2.95 ± 1.16 colonies m−2 to 20.60 ± 2.62 colonies m−2 in 2013, and from 3.20 ± 0.75 colonies m−2 to 13.00 ± 1.04 colonies m−2 in 2014; for recruits, mean densities varied from 1.05 ± 0.34 colonies m−2 to 4.25 ± 0.81 colonies m−2 in 2013, and from 0.60 ± 0.31 colonies m−2 to 1.44 ± 0.40 colonies m−2 in 2014 (all ± SE). The most common taxa in both years among all sites were Antillogorgia spp., Gorgonia spp., and plexaurids. Density of recruits was significantly and positively correlated with population density of adult octocorals (pooled among taxa) and plexaurids in both years, and for Gorgonia spp. in 2013 (with a similar trend in 2014). Densities of recruits and adults of Antillogorgia spp. were not associated in either year. Together, these data suggest that densities of adult octocorals positively influence the density of co-occurring octocoral recruits, thereby potentially promoting population growth. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Understanding the factors regulating population size is a central theme of ecology (Hixon et al., 2002; Sale and Tolimieri, 2000) that has interested scholars for centuries (Malthus, 1798). This preoccupation originates from the desire to understand why populations increase in size, remain stable, decline, or terminate with extinction (Green, 2003; Holling, 1973). For more than a century, anthropogenic disturbances have been responsible for large fluctuations in population sizes of animals and plants in every major ecosystem (Lande, 1998; Vitousek et al., 1997), and these trends underscore the contemporary importance of fundamental ecological principles involved in the regulation of population size. Surprisingly, however, understanding of the factors regulating population size remains incomplete, particularly for sessile marine organisms in open populations (Hixon et al., 2002, 2012; Sibly and Hone, 2002). A population is considered regulated if it displays persistence, boundedness, and recovery (Hixon et al., 2012), with compensatorydensity feedback (sensu Herrando-Pérez et al., 2012) serving as a regulatory mechanism. Feedback controls population size in ways that vary ⁎ Corresponding author. E-mail address: [email protected] (P.J. Edmunds).

http://dx.doi.org/10.1016/j.jembe.2015.08.006 0022-0981/© 2015 Elsevier B.V. All rights reserved.

with the density of organisms, with classic examples provided by interand intra-specific resource competition (Herrando-Pérez et al., 2012). One interesting case of population size being affected by densityassociated processes is provided by arborescent taxa like some plants, macroalgae, and animals (Carr, 1994; Reed and Foster, 1984), for which high population densities create a canopy that modulates conspecific recruitment beneath (Callaway, 1992). Where such organisms are photosynthetic, the attenuation of light as it passes through the canopy can play an important role in depressing conspecific recruitment below (Callaway, 1992). Population regulation has been extensively studied on coral reefs (Hughes, 1984; Connell et al., 1997), where large changes in population sizes of key taxa can be found (Aronson and Precht, 2001; De'ath and Moran, 1998; Lessios et al., 1984). Within the last few decades, the population sizes of scleractinians and macroalgae have dramatically changed in an inverse pattern throughout the Caribbean and parts of the Indo-Pacific (Edmunds, 2013; Jackson et al., 2014; Schutte et al., 2010), and these changes are central components of the phase shifts affecting many reefs (Done, 1992; Hughes, 1994; Ostrander et al., 2000). A phase shift to macroalgal-dominance is however, only one phase into which coral reefs can transition, and in a few cases ascidians, sponges, and octocorals have become the dominant taxa on disturbed coral reefs (Loh and Pawlik, 2014; Norström et al., 2009; Ruzicka et al., 2013).

104

K. Privitera-Johnson et al. / Journal of Experimental Marine Biology and Ecology 473 (2015) 103–109

One change on coral reefs that has recently received attention is an increase in abundance of octocorals that has been recorded in the Florida Keys (Ruzicka et al., 2013), and in St. John, US Virgin Islands (Lenz et al., 2015). In the Florida Keys, between 1999 and 2009, the percentage cover of octocorals increased 0.97% y− 1, and as rapidly as 1.43% y−1 in shallow fore reef habitats (Ruzicka et al., 2013). Between 1992 and 2012 in St. John, densities of octocorals increased 2–5 fold at four of the six sites (they remained unchanged at the other two) (Lenz et al., 2015). Throughout the Caribbean, increased densities of octocorals creates an assemblage of taxa forming mostly arborescent colonies, which at high densities can create a canopy analogous to dense growths of trees or a kelp forest (Rossi, 2013). The formation of a flexible forest, which contrasts with the rigid forest built by scleractinians (Alvarez-Filip et al., 2011), has potentially important implications for the way in which flowing seawater and the particles it contains interacts with the benthos beneath (Denny and Gaylord, 2010; Gili and Coma, 1998). In this study, we quantified the association between the abundance of octocoral recruits and the population density of their adults to explore the ecological implications of varying densities of octocorals that appear to be increasing in at least two locations (Lenz et al., 2015; Ruzicka et al., 2013). We tested the hypothesis that octocoral recruitment and adult population density are positively associated on shallow reefs in St. John, US Virgin Islands. Multiple sites on shallow reefs were examined in 2013 (10 sites) and 2014 (8 sites), and we used sites with different densities of adult octocorals to accomplish our goal. The outcome of an analysis of recruitment is strongly affected by the way in which recruits are defined, and here we follow Caley et al. (1996) and define a recruit as the “initial sighting of a recently settled juvenile in the adult habitat”. Our preliminary surveys revealed that the smallest octocoral colonies we could detect were 1–4 cm tall, and therefore we defined recruits as colonies ≤ 4 cm tall. This definition included sufficient numbers of colonies for their densities to be normally distributed and amenable to parametric statistics, and moreover, it is similar to definitions of octocoral recruits used in previous studies (Jamison and Lasker, 2008; Lasker, 1990, 2013). Given the lively debate associated with the role of densityassociated processes in mediating population size (Caley et al., 1996; Sale and Tolimieri, 2000), it is important to note that a correlative study such as the present analysis cannot establish cause-and-effect relationships. Density-association between recruitment and population size can arise from multiple mechanisms with implications differing between open and closed populations (Sale and Tolimieri, 2000), and critically, does not necessarily correspond to density-dependent population regulation (Sale and Tolimieri, 2000). Evidence of density-association of early life stages does, however, motivate process-oriented studies to test alternative hypotheses regarding the mechanisms causing the association. While evidence of an association between recruitment and population size for octocorals is not new (Lasker, 2013; Yoshioka, 1996), the paucity of studies of such effects, as well as the shifting community ecology of Caribbean coral reefs (Jackson et al., 2014; Ruzicka et al., 2013), makes this a timely point to revisit the association and explore the role it might play in promoting increasing abundances of octocorals (Lenz et al., 2015; Ruzicka et al., 2013).

64° 43.452’ N

St. John 1 km

18° 19.002’ W

RS9 Booby Rock

RS11 RS6

White Point

East Cabritte RS15 Yawzi Point

RS5 Tektite

RS2

Cabritte Horn

Fig. 1. Map of study sites in St. John, US Virgin Islands. Yawzi Point (Y, 9 m depth) and Tektite (T, 14 m depth) were selected in 1987, and six more sites at 7–9-m depth were selected in 1992 (RS2, RS5, RS6, RS9, RS11, RS15) (see Edmunds, 2013); these legacy sites were censused for octocorals in 2013 and 2014. In 2013, east Cabritte (EC, N18° 18.613′, W64°, 43.127′) and Booby Rock (BR, N18° 18.159′, W64° 42.598′) at 9-m depth were sampled in order to capture sites with high densities of octocorals.

algae, bare space, and other invertebrates (Colvard and Edmunds, 2011; Idjadi and Edmunds, 2006; Edmunds, 2013). Eight sites were censused in 2013 and 2014, and are annually censused for analysis of coral reef community structure (Edmunds, 2002, 2013) that now includes octocorals (Lenz et al., 2015). In 2013, two additional sites (Booby Rock and east Cabritte) were censused to sample sites that local knowledge suggested had high densities of octocorals. Species-level resolution for octocorals often involves inspection of sclerites, which requires sampling of tissue that is not feasible at a large scale in a U.S. National Park and Biosphere Reserve. Morphological taxonomy (Sanchez and Wirshing, 2005, http://www.nova.edu/ncri/ sofla_octocoral_guide/), voucher specimens, and expert identification by H.R. Lasker was used to identify octocorals, and we designed an analysis with genus- or family-level resolution. Preliminary surveys in 2012 identified 10 genera (Antillogorgia spp., Briareum spp. [erect form], Eunicea spp., Gorgonia spp., Muricea spp., Muriceopsis spp., Plexaura spp., Plexaurella spp., Pseudoplexaura spp., and Pterogorgia spp.), of which 4 (Plexaura spp., Plexaurella spp., Pseudoplexaura spp., and Eunicea spp.) were challenging to distinguish when small and were pooled and scored as plexaurids (Lenz et al., 2015). Surveys focused on taxa forming arborescent colonies that dominate octocoral communities on the shallow reefs of St. John (Lenz et al., 2015). This focus excluded Erythropodium caribaeorum and the encrusting form of Briarium asbestinum, both of which occur on shallow reefs in St. John, although only E. caribaeorum was common, and only then in select habitats (see Witman, 1992).

2. Methods 2.1. Octocoral surveys in 2013 Field surveys in 2013 and 2014 were completed along ~7 km of the south shore of St. John to test for an association between densities of octocoral recruits and adult octocorals (Fig. 1). Coral reef community structure in this location is well known (Edmunds, 2002, 2013; Rogers et al., 2008), and in shallow water (b 14 m depth) is represented by two habitats. One is dominated by Orbicella annularis (where coral cover was as high as 45% in 1987), and one by igneous boulders and cliffs where mean coral cover has remained b5% since 1992, and hard surfaces are dominated by macroalgae, algal turf, crustose coralline

In 2013, surveys were conducted in July and August at ten sites at 7–14 m depth (Fig. 1). Octocoral recruits were defined as colonies ≤ 4 cm tall, which includes colonies ≤ 1–2 y old assuming linear growth of ca. 2.0–4.0 cm y−1 (Goffredo and Lasker, 2006; Lasker et al., 2003; Yoshioka and Yoshioka, 1991), and they were counted in surveys separate from those used to census all octocorals. At two sites that have been surveyed for coral reef community structure since 1987 when they were dominated by O. annularis (Yawzi Point at 9 m depth and Tektite at 14 m

K. Privitera-Johnson et al. / Journal of Experimental Marine Biology and Ecology 473 (2015) 103–109

depth [Edmunds, 2013]), the octocoral community (all colony sizes) was censused in quadrats (1 × 1 m) positioned contiguously along two 10–12 m transects that were parallel (10 quadrats transect−1) to one another (see Edmunds, 2013). Octocoral recruits were censused along the same transects within a few days of the surveys of the octocoral community, and quadrats (1 × 1 m, 10 quadrats transect−1) for this purpose again were placed contiguously along the transects. All octocorals (including recruits) were censused at Booby Rock (n = 20 quadrats) and east Cabritte (n = 10 quadrats) using quadrats (1 × 1 m) positioned randomly along a single 40-m transect placed along a 7–9 m depth contour; sample sizes differed due to logistical constraints. Octocorals also were surveyed at an additional six sites that have been studied since 1992 ((“random sites” [RS]), and are located on shallow reefs at 7–9 m depth) where the cover of scleractinians has remained b 5% (see Edmunds, 2013). At these sites, small quadrats (0.5 × 0.5 m, n = 20 site−1) were used to census the octocoral community, and the quadrat size matched the size used to quantify the overall community structure (Edmunds, 2013). The twenty small quadrats were not effective at quantifying the density of octocoral recruits, and therefore, larger quadrats (1 × 1 m, n = 20 site−1) were used for this purpose. Quadrats of both sizes were placed at the same random positions along the transect lines as described below (see Table 1 for an overview of survey methods). Octocoral recruits were detected by inspecting the substratum and parting macroalgal thalli where necessary (macroalgae were not removed). To estimate densities of adult octocorals, densities of juveniles were subtracted from densities of all octocoral colonies (i.e., regardless of size). Densities of both life stages were normalized to a square meter, which required linear interpolation from densities recorded in 0.5 × 0.5 m quadrats.

2.2. Octocoral surveys in 2014 In August 2014, surveys were repeated at Yawzi Point, Tektite, and the six random sites, and the methodology was matched to that used for scleractinians (Edmunds, 2013), and therefore differed slightly from that used for octocoral surveys in 2013. As in 2013, sampling focused on arborescent octocorals and excluded E. caribaeorum and encrusting B. asbestinum. At Yawzi Point and Tektite, octocorals were censused in ten, 1 × 1 m quadrats placed contiguously along the permanent transects, but the survey effort was increased to include all three of the permanent transects. At the six random sites, octocorals were censused in 40, 0.5 × 0.5 m quadrats placed at random positions along a 40 m transect line. In all cases, every arborescent octocoral was counted and measured for height, which allowed separation of recruits (≤4 cm height) from pooled juvenile and adult octocoral (N4 cm height) colonies.

105

2.3. Statistical analyses Total (adults and recruits) and recruit densities for octocorals were compared among sites using one-way ANOVA to first test whether densities differed among sites. It was reasoned that our ability to interpret an association between densities of adults and recruits would be strengthened in a case where the sites statistically differed in population density of octocorals. A correlative approach was then used to test for associations between densities of adult octocorals and their recruits using sites as replicates and mean densities as dependent variables. These analyses were completed separately for all octocorals (pooled among taxa), Gorgonia spp., Antillogorgia spp., and plexaurids. Briareum spp., Muricea spp., Muriceopsis spp., and Pterogorgia spp. occurred at such low densities that it was not meaningful to analyze these genera separately. Linear–linear and linear–log associations between the variables were tested using Pearson correlation, and where significant, a functional relationship was described using Model II regression (Sokal and Rohlf, 2011). Statistical analyses were conducted using Systat 13. All data are available on line (http://www.bco-dmo.org). 3. Results 3.1. Density of octocorals in 2013 Octocoral communities consisted of 10 common genera, of which 18% was Antillogorgia spp., 22% was Gorgonia spp., and 62% was plexaurids (n = 376 colonies in 120, 0.25 m2 quadrats, and n = 607 in 70, 1 m2 quadrats). Mean (± SE) densities of octocorals (pooled among taxa, adults plus recruits) ranged from 3.60 ± 1.36 colonies m−2 at RS9 (n = 20) to 23.60 ± 2.54 colonies m−2 at RS2 (n = 20) and differed among sites (Table 2A). Where high population densities were found (e.g., Booby Rock), all quadrats contained at least one octocoral, but where densities were low, up to half of the quadrats at each site contained no octocorals (e.g., RS9). Densities of Antillogorgia spp., Gorgonia spp. and plexuarids also differed among sites (Table 2A). The density of Antillogorgia spp. ranged from 0.60 ± 0.44 colonies m−2 (RS9) to 4.15 ± 0.42 colonies m−2 (Booby Rock) (±SE, n = 20); the density of Gorgonia spp. ranged from 0.80 ± 0.37 colonies m−2 (RS15) to 6.20 ± 1.55 colonies m−2 (RS6) (±SE, n = 20); and the density of plexaurids ranged from 1.10 ± 0.30 colonies m− 2 (Tektite) (± SE, n = 36) to 9.10 ± 0.97 colonies m− 2 (Booby Rock) (± SE, n = 20). Octocoral recruits were found at all sites and were represented by at least 9 genera. Overall, recruits represented 21% of the entire octocoral fauna, but by site, they accounted for as few as 13% (at RS2) or as many as 38% (at RS5) of the total colonies. For the most common taxa, recruits accounted for 12% of all Antillogorgia spp., 9% of all Gorgonia spp., and 31% of all plexaurids. Overall, the density of octocoral recruits (pooled among taxa) differed among sites (Table 2A), with mean

Table 1 Description of the sampling design used to census octocoral densities (adults and recruits) in 2013 and 2014 in St. John for random sites (RS), Tektite, Yawzi Point, Booby Rock, and east Cabritte. All sites were censused in 2013 and 2014, except for Booby Rock and east Cabritte that were censused only in 2013. Transects = length and number of transects employed (n, the same for both years except Tektite and Yawzi Point as shown); quadrat size = area of quadrat used to census adults and recruits; quadrats 2013/2014 = number of replicate quadrats censused in each year; depth (m) = depth of study site; and adult density category. Sites

RS2 RS5 RS6 RS9 RS11 RS15 Tektite Yawzi Point Booby Rock East Cabritte

Transect

40 m (n = 1) 40 m (n = 1) 40 m (n = 1) 40 m (n = 1) 40 m (n = 1) 40 m (n = 1) 10 m (2013, n = 2; 2014, n = 3) 10 m (2013, n = 2; 2014, n = 3) 40 m (n = 1) 40 m (n = 1)

Quadrat size (m2)

Quadrats

Transects

Adults

Recruits

2013/2014

Fixed/random

0.25 0.25 0.25 0.25 0.25 0.25 1 1 1 1

1 1 1 1 1 1 1 1 1 1

20/40 20/40 20/40 20/40 20/40 20/40 20/30 20/30 20 10

Fixed Fixed Fixed Fixed Fixed Fixed Fixed Fixed Random Random

Depth (m)

9 9 9 7 9 9 14 9 9 9

106

K. Privitera-Johnson et al. / Journal of Experimental Marine Biology and Ecology 473 (2015) 103–109

Table 2 Results of one-way ANOVAs comparing densities of recruits and total octocorals among sites for pooled octocorals, Antillogorgia, Gorgonia, and plexaurids in (A) 2013 and (B) 2014. Recruits

A 2013 Pooled octocorals Antillogorgia Gorgonia Plexaurids B 2014 Pooled octocorals Antillogorgia Gorgonia Plexaurids

Total

df

F

P

F

P

9,180 9,180 9,180 9,180

5.003 1.834 3.364 5.138

b0.001 0.065 0.001 b0.001

18.905 4.722 6.831 13.376

b0.001 b0.001 b0.001 b0.001

7,291 7,291 7,291 7,291

1.196 2.540 4.302 1.103

0.031 0.015 b0.001 0.361

11.425 2.089 13.117 11.939

b0.001 0.045 b0.001 b0.001

values varying from 0.75 ± 0.27 colonies m− 2 at Tektite to 4.30 ± 1.42 colonies m− 2 at east Cabritte (± SE, n = 20). Similar patterns were observed for recruits of the most abundant taxa: the density of Gorgonia spp. recruits differed among sites (Table 2A), with mean values ranging from 0 colonies m− 2 (at RS5) to 0.98 ± 0.15 colonies m− 2 (at RS6) (± SE, n = 20); the density of plexaurids recruits differed among sites (Table 2A) with mean values ranging from 0.05 ± 0.17 colonies m − 2 (Tektite and Yawzi Point) to 3.20 ± 0.95 colonies m− 2 (Booby Rock) (± SE, n = 20). The density of Antillogorgia spp. recruits did not differ among sites (Table 2A) with mean values ranging from 0 colonies m − 2 (east Cabritte) to 0.50 ± 0.17 colonies m− 2 (at RS5).

3.2. Density of octocorals in 2014 In 2014 the octocoral community again was represented by 10 common genera, of which 17% was Antillogorgia spp., 26% was Gorgonia spp., and 53% was plexaurids (n = 670). Mean densities (pooled among taxa, adults plus juveniles) ranged from 2.67 ± 0.42 colonies m−2 at Tektite (n = 30) to 14.10 ± 1.11 at RS2 (n = 40), and differed among sites (Table 2B). Likewise, the density of Antillogorgia spp., Gorgonia spp. and plexaurids all differed among sites (Table 2B). For Antillogorgia spp., densities and ranged from 0.27 ± 0.11 colonies m− 2 at Tektite (n = 30) to 2.56 ± 0.45 colonies m− 2 at RS5 (n = 40); density of Gorgonia spp. ranged from 0.40 ± 0.19 colonies m−2 at RS15 (n = 40) to 3.18 ± 0.68 colonies m−2 at RS6 (n = 39); and the density of plexaurids ranged from 0.73 ± 0.22 colonies m−2 at Tektite (n = 30) to 10.20 ± 0.89 m−2 at RS2 (n = 40) (all mean ± SE). Octocoral recruits represented 12% of colonies (n = 670), but by site they accounted for as few as 8% (RS2, n = 141) or as many as 19% (RS11, n = 68) of colonies. For the most common taxa, recruits accounted for 5% of Antillogorgia spp. (n = 112), 17% of Gorgonia spp. (n = 173), and 12% of plexaurids (n = 353). Overall, the density of octocoral recruits (pooled among taxa) differed among sites (Table 2B), with mean values varying from 0.40 ± 0.13 colonies m−2 at Tektite (±SE, n = 30) to 1.44 ± 0.40 colonies m−2 at RS6 (±SE, n = 39). Similar patterns were observed for Gorgonia spp., with densities of recruits differing among sites (Table 2B) and ranging from 0.10 ± 0.10 colonies m−2 at RS5, RS9 and RS15 (n = 40) to 0.41 ± 0.20 colonies m−2 (RS6) (n = 39), and for Antillogorgia spp., with densities of recruit differing among sites (Table 2B), and ranging from 0 colonies m− 2 (Yawzi, RS2, RS5, RS11 and RS15) to 0.40 ± 0.19 colonies m− 2 (RS9) (n = 40) (all means ± SE). Densities of plexaurid recruits did not vary among sites (Table 2B), but mean values ranged from 0.03 ± 0.03 colonies m− 2 (Yawzi Point [n = 30] and RS9 [n = 50]) to 0.90 ± 0.39 colonies m−2 (RS11) (n = 40) (±SE).

3.3. Associations between density of recruits and density of adults in 2013 and 2014 The density of octocoral recruits increased with density of adults for three taxa in 2013, and 2 taxa in 2014, with another taxon showing a trend for a positive association (Fig. 2). The density of Gorgonia spp. recruits was positively correlated with the density of adult Gorgonia spp. in 2013 (r = 0.644, df = 8, P = 0.044), with the Model II regression showing an increase of 0.13 recruit adult−1, and in 2014, there was a trend for a similar relationship (r = 0.637, df = 6, P = 0.089), with the Model II regression showing that density increased at 0.41 recruit adult−1. For all octocorals (pooled among taxa) and plexaurids, the density of recruits was correlated with the log of the density of adults in 2013 (r = 0.706, df = 8, P = 0.022, and r = 0.767, df = 8, P = 0.010, respectively) and 2014 (r = 0.804, df = 6, P = 0.016, and r = 0.874, df = 6, P = 0.005, respectively), showing that the density of recruits was associated positively with adult density at low adult densities, with the effect ameliorated at higher densities. In contrast, the density of Antillogorgia spp. recruits was unrelated either to the density or the log of the density of adult Antillogorgia spp. in 2013 (r b 0.261, df = 8, P N 0.467) and 2014 (r b |0.151|, df = 6, P N 0.050). 4. Discussion The large changes that have taken place on Caribbean coral reefs over the last 50 y (Jackson et al., 2014; Schutte et al., 2010) make it likely that the function of these ecosystems in the future will be influenced by organisms other than scleractinians and macroalgae (Norström et al., 2009). This possibility underscores the values of early detection of increasing abundance of other benthic organisms, and of exploring the mechanisms promoting increases in their population size. Aside from evidence that some scleractinians might function as ecological winners on future reefs (Edmunds et al., 2014; Loya et al., 2001), there already are signs that the community structure of some coral reefs are becoming more heavily influenced by invertebrates other than scleractinians (Norström et al., 2009), such as octocorals (Ruzicka et al., 2013) and sponges (Loh and Pawlik, 2014). The focus of the present study was on one possible example of this phenomenon from St. John, where octocorals have at least maintained abundances (or become more common) since 1992 (Lenz et al., 2015), and explored associations that have value in understanding the mechanisms mediating changes in population sizes of octocorals. Our tests for associations between densities of octocoral recruits and adults reveal positive relationships for octocorals pooled among taxa, for two of three functional groups, and in two consecutive years. Our analyses of recruit–adult relationships for four groups of octocorals reveal three patterns of association: (1) a log–linear relationship in which densities of recruits initially rises rapidly with the density of adults, but show signs of stabilizing at a maximum value (i.e., all octocorals and plexaurids), (2) a linear relationship with no amelioration of the density effect within the range of adult densities encountered (≤5.85 colonies m−2 [e.g., Gorgonia spp.]), and (3) independence between densities of the two life stages (Antillogorgia spp.). Our study is not the first to report a positive association between octocoral recruitment and the density of adults in the Caribbean. Yoshioka (1996) reported a positive relationship between these life stages for two reefs at 7–11 m depth off the south coast of Puerto Rico 19 y ago. Further, between 2004 and 2007 on the south edge of the Little Bahama Bank, Lasker (2013) found that recruitment of Antillogorgia elisabethae was highest at sites where adult A. elisabethae were most abundant. At the scale of the reefs investigated, Yoshioka (1996) and Lasker (2013) considered the positive association between recruitment and adult density as examples of stock–recruitment relationships, in which a larger adult population supported greater local recruitment. Stock–recruitment relationships are well known for fish populations (Caley et al., 1996), but the extent to which they represent mechanisms underlying

K. Privitera-Johnson et al. / Journal of Experimental Marine Biology and Ecology 473 (2015) 103–109

107

Fig. 2. Relationships between population density of adult octocorals (N4 cm) and their recruits (≤4 cm) on reefs along the south shore of St. John in 2013 (top row) and 2014 (bottom row). (A, B) Octocorals, pooled among taxa, (C, D) Antillogorgia spp., (E, F) Gorgonia spp., and (G, H) plexaurids (Plexaura spp., Plexaurella spp., Pseudoplexaura spp., and Eunicea spp.). Values show mean ± SE (2013: n = 20 quadrats site−1, except east Cabritte [n = 10]; 2014: ~40 quadrats site−1 except Yawzi and Tektite [n ~ 30]) for 8–10 sites (site codes in Fig. 1). There is a significant positive relationship between the density of recruits and the population density for: Octocorals, Gorgonia spp. in 2013 (with a trend in 2014), and plexaurids, but not for Antillogorgia spp. Lines represent Model II regressions. Note axes differ among panels.

associations between recruit and adult density depends on whether the populations are open or closed (Caley et al., 1996; Sale and Tolimieri, 2000). On the shallow reefs of St. John, most octocorals formed flexible colonies extending as high as ~1–2 m above the benthos, and at the highest population densities (e.g., ~ 22 colonies m−2 at Booby Rock) formed dense stands of swaying branches. In 2013 and 2014, mean densities of adult colonies and recruits differed among 8–10 sites, but within a site, some quadrats contained colonies of both life stages, while others had none (e.g., up to 35% of quadrats at RS9 in 2013). It was reasoned that the differences among sites would facilitate a mensurative test (sensu Hurlbert, 1984) for an association between recruit and adult densities, with the assumption that the mechanism(s) underlying such effects would operate at a scale bigger than a quadrat but constrained within the scale of each site. The detection of similar relationships in two years strengthens the inferences that can be made from our results, which likely included different cohorts of recruits. Since recruits were defined as ≤ 4 cm tall, and growth rates of Caribbean octocorals in shallow water are in the range of ~ 2–4 cm y− 1, with Antillogorgia spp., Gorgonia spp., and plexaurids growing at 2 cm y−1 (Lasker et al., 2003), 6.1 cm y− 1 (Cary, 1914), and 1.9 cm y− 1 (Yoshioka and Yoshioka, 1991), respectively, many of the octocoral recruits in 2013 probably would have grown out of this size class 12 months later. Therefore, our surveys conducted in 2014 evaluated the relationships between adult and recruitment largely with a new cohort of octocoral recruits. Again, this analysis revealed positive associations between the two. Our study sites are ≤2.6 km apart, which represents an open system where pelagic larvae are likely to come from (and go to) distant locations. The potential extent of these effects are revealed by measurements of surface flow speeds in Great Lameshur Bay (1–13 cm s− 1 [Horst and Edmunds, 2010]), and the pelagic larval durations for octocorals (up to 3 weeks [Lasker and Kim, 1996; Zaslow and Benayahu, 1996]). As our study sites are likely open populations, the mechanism underlying the associations between recruit and adult densities may differ from that proposed by Lasker (2013) and Yoshioka (1996) (i.e., a stock–recruitment relationship), by involving larvae imported from locations more distant than the immediate octocoral populations (in addition to those produced locally), and a

role of the octocoral canopy both in retaining and possibly entraining octocoral larvae. Our study are consistent with the hypothesis that aggregates of octocoral colonies can facilitate population growth in general (i.e., for octocorals pooled among taxa), and separately for Gorgonia spp., and plexaurids, with these effects potentially contributing to recent increased abundance of this taxon in parts of the Caribbean (Lenz et al., 2015; Ruzicka et al., 2013). Manipulative experiments (with several possibilities described below) are required to evaluate whether this is a cause-and-effect relationship and to resolve the mechanism(s) upon which it is based. In Puerto Rico 28 y ago, Yoshioka (1996) carried out experiments relevant to this issue by clearing octocorals from small plots (0.5 m2) and measuring the effect on octocoral recruitment. Yoshioka's experiment was designed to test for inhibition of recruitment at high population density, and while this was not supported by the depressed recruitment he recorded, the outcome is consistent with the opposite effect, that adult colonies promote recruitment (Yoshioka and Yoshioka, 1989). Two possible explanations for the present results are that: (1) spatial variation in the supply of octocoral larvae causes some sites to always have higher rates of recruitment than other sites, and (2) different reproductive strategies of octocoral (i.e., brooders versus broadcast spawners) influence recruitment in dissimilar ways within octocoral canopies varying in adult densities. In the case of the first possibility, this explanation does not provide a good match to our data, because spatial variation in octocoral recruitment is inconsistent with spatial variation in recruitment of scleractinians, at least as evaluated on settlement tiles placed at ~5 m depth at five sites close to sites in the present analysis (see Green and Edmunds, 2011). Over three years (2011–2013) the mean densities of scleractinian recruits ranked these sites Yawzi Point N Cabritte Horn N Tektite N West Little Lameshur Bay N White Point (P.J. Edmunds unpublished data), which does not match with nearby sites ranked by mean densities of octocoral recruits (Fig. 2). Therefore, if larvae of octocorals and scleractinians are present in similar densities and are moved in similar ways by seawater flowing along the coast of St. John, then high densities of octocoral recruits do not occur in locations where their larvae are likely to occur at elevated densities. As for the second potential explanation, it is possible that the mode of sexual reproduction of octocorals influences the rate of recruitment beneath their canopy (see Jamison and Lasker, 2008). This effect

108

K. Privitera-Johnson et al. / Journal of Experimental Marine Biology and Ecology 473 (2015) 103–109

Fig. 3. Line drawing illustrating how an animal forest composed of octocorals could promote stock-recruitment within the canopy of an octocoral community. We hypothesize that octocoral larvae are concentrated beneath the canopy, both by entrainment from the ambient flow of seawater, and by retention of larvae from the adult colonies constructing the animal forest. These effects are associated positively with octocoral density, and high concentrations of larvae translate to high densities of recruits. The tallest Gorgonia ventalina represents 1 m, while the particles are not to scale.

might occur, for example, because brooded larvae are well developed when released and capable of settlement almost immediately, whereas spawned gametes require time for fertilization and development to competence before settlement is possible (Lasker et al., 1996, 1998; Lasker and Kim, 1996). While it was not possible to explore the relationship between recruitment and mode of sexual reproduction in the present study, it is interesting that recruitment and adult densities were unrelated for Antillogorgia spp., which consists mostly of brooding species (Coelho and Lasker, 2014; Gutiérrez-Rodríguez and Lasker, 2004). In contrast, Gorgonia ventalina showed a positive linear relationship between recruitment and adult density, and is presumed to reproduce by mass spawning of gametes (Andras et al., 2013). Plexaurids and pooled octocorals had a positive threshold relationship between recruitment and adult density, and both spawning and brooding reproductive strategies are found among their species (Brazeau and Lasker, 1989; Gutiérrez-Rodríguez and Lasker, 2004; Kahng et al., 2011; Kapela and Lasker, 1989). Further studies are needed to tease apart the association between recruitment patterns observed inside canopies of octocoral communities consisting of varied reproductive modes. When our correlative results are interpreted against the backdrop of the changing benthic communities of Caribbean reefs (Jackson et al., 2014), and the recent evidence that octocorals are becoming more abundant (Ruzicka et al., 2013) or at least maintaining population sizes (Lenz et al., 2015), the importance to determine the cause-andeffect relationships driving these patterns is apparent. It is beyond the scope of the present study to speculate on the experimental designs that would be valuable for this purpose, but it might be valuable to target such experiments on testing at least four mechanisms: (1) selfrecruitment of octocoral larvae (Gutiérrez-Rodríguez and Lasker, 2004) retained within the octocoral canopy (Fig. 3), (2) entrainment of octocoral larvae from ambient seawater above the canopy (Fig. 3) in a mechanism resembling that affecting recruitment of sessile taxa and mobile benthic invertebrates in New England (Osman and Whitlatch, 1998; see Sponaugle et al. (2002) for an overview of this topic), (3) enhanced fertilization success within dense populations (Brazeau and Lasker, 1992; Coma and Lasker, 1997), or (4) enhanced post-settlement success beneath the octocoral canopy, for example, because recruits are sheltered from large predatory fishes like some monocanthids (filefishes) and ostraciontids (trunkfishes) that consume octocorals (Randall, 1967).

Acknowledgments This research was funded by the US National Science Foundation (DEB 08 41441, DEB 13 50146, and OCE 13-32915) and an REU supplement (to KPJ). Fieldwork was completed under permit VIIS-2013-SCI0025 from the Virgin Islands National Park. We thank A. Yarid and C. Didden for assistance in the field, H.R. Lasker for critical feedback that improved an earlier draft of this manuscript, S. Prosterman and V. Powell for on-site logistics, the staff of the Virgin Islands Environmental Resource Station for making our visits to St. John productive and enjoyable, and M. Lenz for preparing Fig. 3. This is contribution number 231 of the marine biology program of California State University, Northridge. [SS]

References Alvarez-Filip, L., Cote, I.M., Gill, J.A., Watkinson, A.R., Dulvy, N.K., 2011. Region-wide temporal and spatial variation in Caribbean reef architecture: is coral cover the whole story? Glob. Chang. Biol. 17, 2470–2477. Andras, J.P., Rypien, K.L., Harvell, C., 2013. Range-wide population genetic structure of the Caribbean sea fan coral, Gorgonia ventalina. Mol. Ecol. 22, 56–73. Aronson, R.B., Precht, W.F., 2001. White-band disease and the changing face of Carribean coral reefs. Hydrobiologia 460, 25–38. Brazeau, D.A., Lasker, H.R., 1989. The reproductive cycle and spawning in a Caribbean gorgonian. Biol. Bull. 176, 1–7. Brazeau, D.A., Lasker, H.R., 1992. Reproductive success in the Caribbean octocoral Briareum asbestinum. Mar. Biol. 114, 157–163. Caley, M.J., Carr, M.H., Hixon, M.A., Hughes, T.P., Jones, G.P., Menge, B.A., 1996. Recruitment and the local dynamics of open marine populations. Annu. Rev. Ecol. Syst. 27, 477–500. Callaway, R.M., 1992. Effect of shrubs on recruitment of Quercuz douglash and Quercus lobata in California. Ecology 73, 2118–2128. Carr, M.H., 1994. Effects of macroalgal dynamics on recruitment of a temperate reef fish. Ecology 75, 120–1333. Cary, L.R., 1914. Observation on the growth rate and oecology of gorgonians. Carnegie Inst Washington Publ 413, 79–90. Coelho, M.A.G., Lasker, H.R., 2014. Reproductive biology of the Caribbean brooding octocoral Antillogorgia hystrix. Invertebr. Biol. 133, 299–313. Colvard, N.B., Edmunds, P.J., 2011. Decadal-scale changes in abundance of nonscleractinian invertebrates on a Caribbean coral reef. J. Exp. Mar. Biol. 397, 153–160. Coma, R., Lasker, H.R., 1997. Small-scale heterogeneity of fertilization success in a broadcast spawning octocoral. J. Exp. Mar. Biol. Ecol. 214, 107–120. 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. 7, 461–488. De'ath, G., Moran, P.J., 1998. Factors affecting the behavior of crown-of-thorns starfish (Acanthaster planci L.) on the Great Barrier Reef: 1: patterns of activity. J. Exp. Mar. Biol. 220, 107–126. Denny, M.W., Gaylord, B., 2010. Marine ecomechanics. Ann. Rev. Mar. Sci. 2, 89–114.

K. Privitera-Johnson et al. / Journal of Experimental Marine Biology and Ecology 473 (2015) 103–109 Done, T.J., 1992. Phase changes in coral reefs and their ecological significance. Hydrobiologia 247, 121–132. Edmunds, P.J., 2002. Long-term dynamics of coral reefs in St. John, US Virgin Islands. Coral Reefs 21, 357–367. Edmunds, P.J., 2013. Decadal-scale changes in the community structure of coral reefs of St. John, US Virgin Islands. Mar. Ecol. Prog. Ser. 489, 107–123. Edmunds, P.J., Adjeroud, M., Baskett, M.L., Baums, I.B., Budd, A.F., Carpenter, R.C., Fabina, N.S., Fan, T., Franklin, F.C., Gross, K., Han, X., Jacobson, L., Klaus, J.S., McClanahan, T.R., O'Leary, J.K., van Oppen, M.J.H., Pochon, X., Putnam, H.M., Smith, T.B., Tyler, B., Stat, M., Sweatman, H., van Woesik, R., Gates, R.D., 2014. Persistence and change in community composition of reef corals through present, past, and future climates. PLoS ONE 9 (10), e107525. Gili, J.M., Coma, R., 1998. Benthic suspension feeders: their paramount role in littoral marine food webs. Trends Ecol. Evol. 13, 316–321. Goffredo, S., Lasker, H.R., 2006. Modular growth of a gorgonian coral can generate predictable patterns of colony growth. J. Exp. Mar. Biol. Ecol. 336, 221–229. Green, D.M., 2003. The ecology of extinction: population fluctuation and decline in amphibians. Biol. Conserv. 111, 331–343. Green, D.H., Edmunds, P.J., 2011. Spatio-temporal variability of coral recruitment on shallow reefs in St. John, US Virgin Islands. 397, 220–229. Gutiérrez-Rodríguez, C., Lasker, H.R., 2004. Reproductive biology, development, and planula behavior in the Caribbean gorgonian Pseudopterogorgia elisabethae. Invertebr. Biol. 123, 54–67. Herrando-Pérez, S., Delean, S., Brook, B.W., Bradshaw, C.J.A., 2012. Density dependence: an ecological Tower of Babel. Oecologia 170, 585–603. Hixon, M.A., Pacala, S.W., Sandin, S.A., 2002. Population regulation: historical context and contemporary challenges of open vs. closed systems. Ecology 83, 1490–1508. Hixon, M.A., Anderson, T.W., Buch, K.L., Johnson, D.W., McLeod, J.B., Stallings, C.D., 2012. Density dependence and population regulation in marine fish: a large-scale, longerterm field manipulation. Ecol. Monogr. 82, 467–489. Holling, C.S., 1973. Resilience and stability of ecological systems. Annu. Rev. Ecol. Syst. 4, 1–23. Horst, G., Edmunds, P.J., 2010. Spatio-temporal variation in seawater characteristics in a semi-enclosed bay in St. John, US Virgin Islands. Caribb. J. Sci. 46, 1–10. Hughes, T.P., 1984. Population dynamics based on individual size rather than age: a general model with a reef coral example. Am. Nat. 123, 778–795. Hughes, T.P., 1994. Catastrophes, phase shifts, and large-scale degradation of a Caribbean coral reef. Science 265, 1547–1551. Hurlbert, S.H., 1984. Pseudoreplication and the design of ecological field experiments. Ecol. Monogr. 54, 187–211. Idjadi, J.A., Edmunds, P.J., 2006. Scleractinian corals as facilitators for other invertebrates on a Caribbean reef. Mar. Ecol. Prog. Ser. 319, 117–127. Jackson, J.B.C., Donovan, M.K., Cramer, K., Lam, V. (Eds.), 2014. Status and Trends of Caribbean Coral Reefs: 1970–2012. Global Coral Reef Monitoring Network, IUCN, Gland, Switzerland. Jamison, J.L., Lasker, H.R., 2008. Identification of octocoral recruits using microsatellite primers: relationships between recruitment and adult distribution of Pseudopterogorgia spp. Limnol. Oceanogr. 53, 1963–1972. Kahng, S.E., Benayahu, Y., Lasker, H.R., 2011. Sexual reproduction in octocorals. Mar. Ecol. Prog. Ser. 443, 265–283. Kapela, W., Lasker, H.R., 1989. Size-dependent reproduction in the Caribbean gorgonian Pseudoplexaura porosa. Mar. Biol. 135, 107–114. Lande, R., 1998. Anthropogenic, ecological and genetic factors in extinction and conservation. Res. Popul. Ecol. 40, 259–269. Lasker, H.R., 1990. Clonal propagation and population dynamics of a gorgonian coral. Ecology 71, 1578–1589. Lasker, H.R., 2013. Recruitment and resilience of a harvested Caribbean octocoral. PLoS ONE 8 (9), e74587. Lasker, H.R., Kim, K., 1996. Larval development and settlement behavior of the gorgonian coral Plexaura kuna. J. Exp. Mar. Biol. Ecol. 207, 161–175. Lasker, H.R., Brazea, D.A., Calderon, J., Coffroth, M.A., Coma, R., Kim, K., 1996. In situ rates of fertilization among broadcast spawning gorgonian corals. Biol. Bull. 190, 45–55.

109

Lasker, H.R., Kim, K., Coffroth, M.A., 1998. Production, settlement, and survival of plexaurid gorgonian recruits. Mar. Ecol. Prog. Ser. 162, 111–123. Lasker, H.R., Boller, M.L., Castanaro, J., Sánchez, J.A., 2003. Determinate growth and modularity in a gorgonian octocoral. Biol. Bull. 205, 319–330. Lenz, E.A., Lasker, H.R., Bramanti, L., Edmunds, P.J., 2015. Long-term variation of octocoral populations in St. John, US Virgin Islands. Coral Reefs. http://dx.doi.org/10.1007/ s00338-015-1315-x. Lessios, H.A., Robertson, D.R., Cubit, J.D., 1984. Spread of Diadema mass mortality through the Carribean. Science 226, 335–337. Loh, T., Pawlik, J.R., 2014. Chemical defenses and resource trade-offs structure sponge communities on Caribbean coral reefs. Proc. Natl. Acad. Sci. U. S. A. 111, 4151–4156. Loya, Y., Sakai, K., Yamazato, K., Nakano, Y., Sambali, H., van Woesik, R., 2001. Coral bleaching: the winners and the losers. Ecol. Lett. 4, 122–131. Malthus, T.R., 1798. An essay on the principle of population. Printed for J. Johnson in St. Paul's Church-Yard, London. Norström, A.V., Nystrom, M., Lokrantz, J., Folke, C., 2009. Alternative states on coral reefs: beyond coral–macroalgal phase shifts. Mar. Ecol. Prog. Ser. 376, 295–306. Osman, R.W., Whitlatch, R.B., 1998. Local control of recruitment in an epifaunal community and the consequences to colonization processes. Hydrobiologia 374, 113–123. Ostrander, G.K., Armstrong, K.M., Knobbe, E.T., Gerance, D., Scully, E.P., 2000. Rapid transition in the structure of a coral reef community: the effects of coral bleaching and physical disturbance. Proc. Natl. Acad. Sci. U. S. A. 97, 5297–5302. Randall, J.E., 1967. Food habitats of reef fishes of the West Indies. Stud. Trop. Oceanogr. 5, 665–847. Reed, D.C., Foster, M.S., 1984. The effects of canopy shadings on algal recruitment and growth in a giant kelp forest. Ecology 65, 937–948. Rogers, C.S., Miller, J., Muller, E.M., Edmunds, P.J., et al., 2008. Ecology of coral reefs in the US Virgin Islands. In: Riegl, B.M., Dodge, R.E. (Eds.), Coral Reefs of the USA. Springer, Dordrecht, pp. 303–373. Rossi, S., 2013. The destruction of the ‘animal forests’ in the oceans: towards an oversimplification of the benthic ecosystems. Ocean Coast. Manag. 84, 77–85. Ruzicka, R.R., Colella, M.A., Porter, J.W., Morrison, J.M., Kidney, J.A., Brinkhuis, V., Lunz, K.S., Macaulay, K.A., Barlett, L.A., Meyers, M.K., Colee, J., 2013. Temporal changes in benthic assemblages on Florida Keys reefs 11 years after the 1997/1998 El Nino. Mar. Ecol. Prog. Ser. 489, 125–141. Sale, P.F., Tolimieri, N., 2000. Density dependence at some time and place? Oecologia 124, 166–171. Sanchez, J.A., Wirshing, H.H., 2005. A field key to the identification of tropical western Atlantic zooxanthellate octocorals (Octocorallia: Cnidaria). Caribb. J. Sci. 41, 508–522. Schutte, V.G.W., Selig, E.R.S., Bruno, J.F., 2010. Regional spatio-temporal trends in Caribbean coral reef benthic communities. Mar. Ecol. Prog. Ser. 402, 115–122. Sibly, R.M., Hone, J., 2002. Population growth rate and its determinants: an overview. Philos. Trans. R. Soc. Lond. B 357, 1153–1170. Sokal, R.R., Rohlf, J.F., 2011. Biometry: The Principles and Practice of Statistics in Biological Research. W.H. Freeman, New York. Sponaugle, S., Cowen, R.K., Shanks, A., Morgan, S.G., Leis, J.M., Pineda, J.M., Boehlert, G.W., Kingsford, M.J., Lindeman, K.C., Grimes, C., Munro, J.L., 2002. Predicting selfrecruitment in marine populations: biophysical correlates and mechanisms. Bull. Mar. Sci. 35, 341–375. Vitousek, P.M., Mooney, H.A., Lubchenco, J., Melillo, J.M., 1997. Human domination on Earth's ecosystems. Science 277, 494–499. Witman, J.D., 1992. Physical disturbance and community structure of exposed and protected reefs: a case study from St. John, U.S. Virgin Islands. Am. Zool. 32, 641–654. Yoshioka, P.M., 1996. Variable recruitment and its effects on the population and community structure of shallow-water gorgonians. Bull. Mar. Sci. 59, 433–443. Yoshioka, P.M., Yoshioka, B.B., 1989. Effects of wave energy, topographic relief and sediment transport on the distribution of shallow-water gorgonians Puerto Rico. Coral Reefs 8, 145–152. Yoshioka, P.M., Yoshioka, B.B., 1991. A comparison of the survivorship and growth of shallow-water gorgonian species of Puerto-Rico. Mar. Ecol. Prog. Ser. 69, 253–260. Zaslow, R., Benayahu, Y., 1996. Longevity, competence and energetic content in planulae of the soft coral Heteroxenia fuscescens. J. Exp. Mar. Biol. Ecol. 206, 55–68.