Introduction

Corals are the primary structural architects of coral reef ecosystems, and coral reproduction is one of the fundamental processes that sustain the maintenance of the structure and functions of coral reef ecosystems (Baird et al. 2015). Most scleractinian corals reproduce by broadcast spawning, which involves the release of gamete bundles from different coral colonies into the water column, where fertilization, larval development, and dispersal take place (Harrison et al. 1984). Coral spawning within populations is often synchronized within short periods of a few nights of a few consecutive months, manifested as mass spawning, depending on the assemblage and geographical location (Harrison and Wallace 1990; Baird et al. 2009a, b; Gilmour et al. 2016). When multispecies synchronous mass spawning occurs in reefs with high coral cover, buoyant gametes may aggregate in the surface water and produce a coral spawn slick (Butler 1980; Harrison et al. 1984; Babcock et al. 1986; Oliver and Willis 1987; Heyward et al. 2002; Harrison 2008). Synchrony in coral spawning is of ecological importance and has management implications as this process ensures efficient fertilization of coral gametes (Guest 2008), which is essential for successful coral larval dispersal and recruitment unto nearby degraded reefs. Information on spawn slick formation is relevant for coral larval rearing and reseeding (Oliver and Willis 1987; Heyward et al. 2002; Edwards et al. 2015).

The discovery of multispecies synchronous mass spawning events in the Great Barrier Reef in the 1980s (Harrison et al. 1984) caused a substantial increase in research on coral reproduction (Harrison 2011). Several studies reported spawning observations from a wide range of localities in Western Australia (Simpson 1985), Japan (Hayashibara et al. 1993), and the Caribbean (e.g. Hagman et al. 1998). These early observations appeared to support the view that synchrony in coral spawning was restricted to high-latitude regions because of larger fluctuations in sea-surface temperatures (SSTs) and tidal levels (Oliver et al. 1988). Recent studies, however, have reported observations of multispecies synchronous coral spawning in areas with high coral cover in lower latitudes (Guest et al. 2002, 2005; Bouwmeester et al. 2015; Chelliah et al. 2015). These findings suggest that the occurrence of synchronous spawning is a feature of all speciose coral assemblages at various latitudes (Baird and Guest 2008).

Considering the advancements in research on coral reproduction, few studies have been done at the equatorial regions and in the Coral Triangle, which includes the Philippines (Veron 2000; Hoeksema 2007; Chelliah et al. 2015). The Philippines has the second largest coral reef area (27,000 km2) in Southeast Asia (Burke et al. 2002; Wilkinson 2008; Carpenter et al. 2011). Most studies on broadcast spawning corals here were conducted in the Bolinao-Anda Reef Complex (BARC), northwestern Philippines (Vicentuan et al. 2008; Vicentuan 2009; Bouwmeester et al. 2015; Maboloc et al. 2015; Abecia et al. 2016; Keith et al. 2016). In situ spawning of 36 coral species (Vicentuan et al. 2008) and hatchery-based spawning of Favites species (Guest et al. 2014; Maboloc et al. 2015) have been documented in BARC. The present study reports on spawning timing and extent in synchrony among various Acropora species, and also on slick formation for this region in the Coral Triangle.

Materials and methods

Study site

The study was carried out in Magsaysay reef (16° 18’ 38.3” N, 120° 01’ 46.8” E), which has the highest coral cover in the Bolinao-Anda Reef Complex (Cabaitan, unpublished data) in Lingayen Gulf, northwestern Philippines (Fig. 1). Magsaysay reef has a depth range of 2–6 m. Unlike the other reefs in BARC, Magsaysay reef was previously not considered for coral spawning studies as this reef was impacted by a crown-of-thorns sea star (Acanthaster planci) outbreak in 2007 (Vergara et al. 2010). The deeper area of Magsaysay reef did not fully recover from the infestation, while the shallower portions (2–3 m depths) exhibited a recovery in 2013, which resulted in a high coral cover dominated by tabular species, Acropora hyacinthus and A. cytherea (Villanueva, personal observation).

Fig. 1
figure 1

Lingayen Gulf in northwestern Philippines showing the location of Magsaysay reef and the extent of coral spawn slicks observed on 5 March 2016 and coral spawn slick remnants on 6 March 2016

Full size image

Rapid reproductive surveys

Reproductive maturity of selected Acropora species in Magsaysay reef was examined through biweekly rapid sampling, from 29 January to 8 April in 2015 and from 2 February to 28 March in 2016, which was conducted by four observers and lasted about an hour each. Sampling was done by breaking one or two fragments from each coral colony using a chisel to check for pigmented (mature) oocytes, white (immature) oocytes, or no visible oocytes (Baird et al. 2002). Acropora colonies were noted as gravid (with mature eggs) when oocyte coloration ranged from shades of orange to pink. In 2015, approximately 60 colonies of various Acropora species were sampled from two established 15 × 15-m2 plots. At least 50 additional Acropora colonies were sampled from a 50 × 50-m2 plot in 2016.

Spawning observations

Following observations of gravid colonies, night-time surveys of coral spawning were performed. In 2015, only one night-time spawning survey was conducted (16 March) because of logistical constraints. In 2016, night-time surveys of coral spawning were conducted in eight consecutive nights from 28 February to 6 March. These dates were predetermined, based on the timing of spawning noted in 2015, which occurred 10 days after full moon. Coral colonies inside the established plots were visually checked for gamete-setting every 30 min starting from 1800 h until spawning. Coral species identity (Wallace 1999), times of gamete-setting and spawning were recorded.

Observation and tracking of coral spawn slicks

Visual surveys were conducted to locate the aggregation and extent of coral spawn slicks. Coordinates along the spawn slicks were marked using Garmin Montana 680 Handheld GPS. Computer software, Manifold System 8.0 Universal Edition, was used to visualize the extent of spawn slicks. Water samples were collected during night and morning surveys for visual analysis of fertilized coral gametes using a MOTIC Images Plus 2.0 stereomicroscope.

Results and discussion

The month of spawning can be predicted by examining the reproductive condition of corals. Acropora colonies sampled on the first samplings, 29 January 2015 and 2 February 2016, contained immature eggs (35 and 41%, respectively of the sampled colonies) (Online Resources 1 and 2). Mature eggs were present in colonies in February and March, with most colonies with mature eggs observed on 12 March 2015 and on 29 Feb 2016 (100 and 73%, respectively). Following these sampling dates, colonies with mature eggs decreased in numbers on 26 March 2015 and on 28 March 2016 (17% and 13% respectively). There were no visible eggs in colonies sampled in April 2015. Sampling was not done in April 2016.

In total, five selected Acropora species were sampled in 2015 and nine in 2016 to assess their reproductive maturity. A high level of synchrony in gamete development among species was observed. All colonies of the five species were mature on 12 March 2015, 6 days after a full moon (Online Resource 1). In 2016, all species had at least one mature colony on 29 February, 6 days after a full moon (Online Resource 2). Populations of A. digitifera, A. hyacinthus, A. humilis, A. latistella, A. millepora, and an unidentified Acropora were quite synchronous in terms of gamete development as 73–100% of the colonies sampled had mature eggs. Other species, such as A. florida and A. tenuis, had moderate levels of population synchrony, where 40 and 33%, respectively, of the colonies sampled had mature eggs, suggesting that these species still may spawn at other times. Extended spawning is common among Indo-Pacific corals (Bermas et al. 1992; Bermas 1996; Baird et al. 2009b; Vicentuan 2009; Bouwmeester et al. 2015; Chelliah et al. 2015), specifically in highly fecund Acropora corals (Mangubhai and Harrison 2008).

Spawning of at least eight species of Acropora (A. hyacinthus, A. cytherea, A. millepora, A. digitifera, unknown Acropora sp., A. latistella, A. florida, and A. humilis) was observed at Magsaysay reef between 9 to 11 days after full moon in March of 2015 and 2016 (Table 1). Acropora spp. exhibited different times of gamete-setting in 2015 and 2016, which took place between 1857 and 2029 h. In 2015, spawning began as early as 1940 h by the dominant species A. hyacinthus and synchronized spawning of the different Acropora species was observed at around 2000 h. Aggregation of gamete bundles was seen in the water column at around 2030 h and spawning ended at around 2100 h. In 2016, the highest number of colonies that spawned was noted on 5 March, which occurred 11 days earlier than the spawning observed in 2015. Inter-annual variability in spawning pattern is a usual phenomenon (Simpson 1985; Willis et al. 1985), which follows a reproductive cycle that depends on the annual changes in lunar months across succeeding years (Baird et al. 2009a; Gilmour et al. 2016).

Table 1 Acropora species that spawned in 2015 (10 days after a full moon) and 2016 (9–11 days after a full moon) with respective times (hours) of gamete-bundle setting and spawning
Full size table

On the nights with the highest number of spawning colonies, coral-spawn slicks were observed (Fig. 2). Slight water turbulence in 2015 may have lead to the dispersal of coral gametes in various locations and eventual disappearance of coral-spawn slicks. The surface-water condition during the mass spawning in 2016 was calmer, which may have allowed the formation of the slicks, which were approximately 5– to 10-m wide and extended eastward up to at least 4 km (Fig. 1). Visual surveys conducted in the morning following the major spawning revealed that remnants of spawn slicks extended up to at least 8 km. Developing coral larvae that ranged from 0.5 to 1 μm in diameter were observed in coral spawn slick remnants.

Fig. 2
figure 2

Coral spawn slick observations in Magsaysay reef, northwestern Philippines showing (a) the spawning of Acropora spp., (b) coral spawn slick accumulation on the night of 5 March 2016, and (c) dispersed remnants of the coral spawn slicks on the morning of 6 March 2016

Full size image

This study expands the limited knowledge on coral reproduction in the Coral Triangle, specifically in the Philippines, by describing the seasonality and timing of coral spawning. Seasonality of spawning may be influenced by a number of factors acting at both an ultimate and proximate level such as temperature, insolation, moonlight, tides (Willis et al. 1985; Simpson et al. 1993; Penland et al. 2004; Boch et al. 2011; Kaniewska et al. 2015), but evidence suggests that the strongest proximate driver of spawning (at least for Indo-Pacific Acropora) is the rate of increase in sea temperature prior to spawning (Keith et al. 2016). In the Philippines, the rise of sea surface temperature occurs from March until May (Online Resource 3), which coincides with the major spawning in that period (Vicentuan et al. 2008; Bouwmeester et al. 2015; Keith et al. 2016). Our results corroborate with the existing suggestions that the major spawning for Acropora in the BARC occurs early in the season, from March to April (Vicentuan et al. 2008). The time of the year of spawning in the Philippines also coincides with the spawning timing in the Java Sea and Palau, which were hypothesized to be driven by peaks of insolation (Penland et al. 2004; Permata et al. 2012).

Mass spawning of Acropora assemblages in BARC were observed from 9 to 11 days after full moon, which are 4–6 days later than the spawning reported by Vicentuan et al. (2008). The two studies reported that coral spawning occurred alongside a neap tide, when the tidal amplitude is at its minimum (Babcock et al. 1986). During the spawning observations, the tides were high around 1724 h (Online Resource 4). A low tidal range would cause a reduced water flow, which may decrease the chances of gamete dilution and increase fertilization success (Mendes and Woodley 2002; Guest et al. 2005). This condition also allows the formation of spawn slicks (Oliver and Willis 1987; Heyward et al. 2002). In areas where both mass spawning and slicks were observed, Acropora species constitutes the majority of the spawners (Willis et al. 1985; Heyward et al. 2002; Harrison 2008; Guest et al. 2010; Hanafy et al. 2010; Permata et al. 2012; Chelliah et al. 2015).

Coral spawning in Magsaysay reef occurred during an inter-monsoon, when there is no prevalent strong water movement between the east and west sides of the Lingayen Gulf (Ashikawa et al. 2013). Hence, depending on the current strength generated by the ebbing tides and winds, coral larval retention is likely to happen (Harrison 2008). As local anthropogenic stressors and unprecedented environmental changes are threatening coral reef ecosystems, knowledge on the reproductive timing and synchrony of corals is crucial in understanding the persistence of coral communities, especially when supported with studies on population genetics and oceanographic modeling for the dispersal potential and connectivity patterns.