Coral reefs are complex formations created by the accumulation of calcium carbonate deposited by animals and algae. They are considered to be the marine ecosystems with the widest biological diversity on the planet (Bellwood and Hughes, 2001; Hughes et al., 2017). However, because they need strict conditions for their formation (Kleypas et al., 1999; Muir et al., 2015) they are restricted to less than 0.1% of the ocean’s surface (Spalding and Grenfell, 1997). Despite this, they provide important ecological and economic services (Bell et al., 2013; Cooper et al., 2014; Mitra and Zaman, 2015; Spalding et al., 2017). Hermatypic corals are primarily responsible for the formation of coral reefs (Cairns et al., 2009). For Caribbean reefs, Acropora spp., specifically Acropora palmata is of great importance due to their ability to withstand high energy conditions, high growth rates, ability to produce fragments and regenerate as well as their three-dimensional architecture, which provides habitat for many species (Lirman, 2002).

Acropora palmata, like most hermatypic corals, lives in a symbiotic relationship with a dinoflagellate alga called Symbiodinium that contributes to its host coral products that are derived from photosynthesis, in exchange for protection from external conditions and a constant supply of inorganic nutrients from the efficient recycling of the coral’s metabolic waste (Davy et al., 2012; Pernice et al., 2012). The contribution of energy by the algae is more than the coral’s daily requirement, so the surplus carbon, mainly glucose, is transformed into triglycerides and waxes to form large energy reserves (Oku et al., 2003). These energy reserves allow the coral to maintain its high productivity and sustain metabolic processes such as growth and sexual reproduction (Edmunds and Davies, 1986; Grottoli et al., 2004).

Lipids are key energetic biomolecules for hermatypic corals in particular, representing up to 40% of their dry weight (Harland et al., 1993; Oku et al., 2003; Imbs, 2013), although this percentage can vary in terms of environmental conditions such as light intensity and water temperature, as well as the coral’s nutritional and health status (Harland et al., 1992; Oku et al., 2002). Under normal circumstances there is a balance between the energy that is produced and that which is consumed (Grottoli et al., 2004), however, under stressful conditions this balance can be broken, because the energy reserves, in addition to supporting growth and reproduction, are also the source of energy to combat the consequences of the stress (Seemann et al., 2013). Lipids represent up to 40% of coral tissue dry weight and, as such, are the primary energy reserve, therefore decreases or increases in the concentrations of total lipids are useful indicators of the use or accumulation, respectively, of these reserves (Stimson, 1987; Ward, 1995; Oku et al., 2002, 2003; Grottoli et al., 2004; Imbs, 2013; Lin et al., 2013; Conlan et al., 2017, 2018). For this reason, lipid concentrations are considered to be a good indicator of coral stress.

Corals can suffer injuries or lesions that are caused by events such as non-lethal diseases, predation, high or prolonged sedimentation or as a result of physical damage caused by various types of disturbances such as hurricanes or tropical storms (Lewis, 1997; Meesters et al., 1997; Wesseling et al., 2001; Cantin and Lough, 2014). Coral bleaching, whereby the symbiotic algae leave or are expelled from the coral tissue, can also result in lesions on the colony, especially if the bleaching event is prolonged (Gleason, 1993; Jones, 2008). Corals have mechanisms that respond immediately to an injury, which initiate the regeneration of the skeleton and tissue in the damaged area. The regeneration process consumes large amounts of energy, involving all the adjacent polyps up to 15 cm around the lesion (Oren et al., 2001; Klingenberg, 2008, 2014; Cocito and Sgorbini, 2014). On occasion, the regeneration process cannot cover the entire lesion, probably due to various factors such as the origin, shape, and size of the lesion relative to colony size, environmental conditions, and the high energy cost involved (Bak and Steward-Van Es, 1980; Oren et al., 1997, 2001; Denis et al., 2013). Consequently, the area becomes covered by colonizing organisms or sediment and the lesions become permanent, becoming what is known as partial mortality.

Corals are modular organisms and can survive partial tissue loss (Oren et al., 2001). However, because recovery from partial mortality represents a major expenditure of energy for the coral, in an effort to maintain colony integrity, it may interrupt other vital processes that require energy reserves such as reproduction and growth (Oren et al., 1997, 2001). Many studies have determined the effects of partial mortality on coral energy reserves focusing on reproduction (Kojis and Quinn, 1985; Rinkevich and Loya, 1989; Van Veghel and Bak, 1994; Oren et al., 2001; Graham and van Woesik, 2013; Riegl and Purkis, 2015), and growth (Bak, 1983; Fang et al., 1989; Guzmán et al., 1994; Meesters et al., 1994; Lirman, 2000; Alvarado and Acosta, 2009) while regeneration is active. To our knowledge, no studies have documented the effects, if any, once regeneration has stopped and partial mortality is permanent. This is relevant to the coral’s energy budget if these lesions represent an area of high energy demand, i.e., the amount of lipids near the lesion is less than that in areas at a distance from the lesion. In this case, areas of partial mortality could continue to be a drain on the coral’s energy and negatively affect other processes, such as growth and reproduction. Therefore, the objective of this contribution is to evaluate if there is a prolonged effect of partial mortality on the energy budget and its possible implications on branch growth and egg size, as well as embryonic development in the coral A. palmata, by comparing the growing edge of apparently healthy colonies to the borders of the lesions of colonies with partial mortality.

Partial mortality in A. palmata colonies has a negative effect on their energy budget as shown by the results obtained in this study. The negative effects were detected in colonies with old partial mortality, where the lesions were completely colonized by other species or covered in sediment. Certain aspects of the reproduction biology of A. palmata were affected by the presence of old partial mortality, whereas colony branch growth appeared to be unaffected.

All studies to date, on the effects of partial mortality on reproduction and growth have only focused on the first stages of the injury or lesion when regeneration tends to be accelerated. Reported results differ widely ranging from a reduction in reproductive activity in Stylophora pistillata (Rinkevich and Loya, 1989) and massive Porites spp. (Welsh et al., 2015) and fecundity in Goniastrea favulus (Kojis and Quinn, 1985), to cases where the effects are localized and only affected up to a 15 cm radius around injured areas in Favia favus (Oren et al., 2001). Negative effects also vary depending on the morphotype of colonies of Montastrea annularis (Van Veghel and Bak, 1994). For A. palmata it has been reported that injured colonies completely stop gamete production (Lirman, 2000).

The area occupied by partial mortality in individuals of A. palmata did not change during the period of study, designed to coincide with the months leading up to the summer spawning season. No significant changes were found between the months of March–April and June–July (Figures 5A–C). This is likely explained by the fact that regeneration is a gradual process that decreases with time reaching the point where it stops, even when the injury has not completely regenerated. The most accepted explanation for this fact is that regeneration is stopped in order to not continue with the excessive energy expenditure caused by this process (Oren et al., 1997, 2001). However, the lipid results obtained in this study contradict this explanation because they show that even after the regeneration process was stopped, the lesions continue to represent areas of high energy demand, as reflected by the significantly lower lipid concentrations at the borders of the lesions (Figure 8).

Furthermore, our results show that the amount of lipids at the borders of the lesions does not differ from the lipid concentrations at the edges of maximum growth of the A. palmata branches. This would indicate that large amounts of energy are being consumed near the borders of the partial mortality, which is similar to the expense that supports the accelerated growth in the tips of the branches in A. palmata. Despite the possibility that there are variations in physico-chemical properties between the three sites, the edges of the zones with partial mortality consistently exhibited lower lipid concentrations (Figure 8) even when the data are separated by site.

Based on our results, we infer that the explanation for the arrest or reduction of regeneration and the low amount of lipids at the edges of the lesions (Figure 8) is produced by a possible competitive relationship induced by the organisms that colonize the injured areas, rather than a mechanism to stop energy expenditure. The space exposed by the lesions, as analyzed in this study, was dominated by filamentous algae (38%) and fleshy algae (20%), which are organisms that are characterized by engaging in competitive relationships with coral species by activating chemical mechanisms that allow them to maintain the space they have gained and, if possible, to overgrow the coral. Their very presence represents a physical obstacle that prevents the adequate regeneration of injured areas (Titlyanov et al., 2005, 2006) and promotes the accumulation of organic and inorganic sediment that in turn provides nutrients and support for algal development (Nugues and Roberts, 2003; Titlyanov and Titlyanova, 2008). This coincides with our results, because sediment, in addition to the algae, is the substrate that dominates the lesions with a median of around 20% (Figure 3). Therefore, the coral needs to activate defensive mechanisms, which require significant amounts of energy, and which would explain the results of the low amount of lipids at the borders of the lesions.

On the other hand, Oren et al. (1997, 2001) also mentions that excessive energy expenditure such as that which occurs when regeneration is active can affect other processes such as reproduction and growth since they depend on the same reserves. This energy expenditure has been estimated at about 33% of the total available energy in the coral Porites porites (Edmunds and Davies, 1986). Such a sizeable energy expenditure would affect processes such as growth and reproduction. Therefore, taking into account our results where it is shown that old lesions continue to be large energy sinks, one would expect that these negative effects would continue even after the regeneration stopped. However, our results showed that growth is not affected in corals with partial mortality in any of the three sites sampled (Figure 4), which is unusual and differs from previous studies. For example, Meesters et al. (1994) found that lesions in Montastrea annularis reduced the growth capacity of the coral even after regeneration was stopped. Bak (1983) found that branches of A. palmata that presented lesions that were within 25 cm of the growth edge of the branch had considerably lower linear extension rates. However, if the lesion was located at more than 25 cm from the growth edge, there was no negative effect on its growth (Bak, 1983). This is a plausible explanation for the results obtained in this work since the position of the lesions analyzed were greater than 25 cm from the growth edge.

As seen from the results presented here, the presence of old partial mortality did affect some aspects of the reproductive process in A. palmata colonies. These colonies were capable of producing and releasing gametes. However, the volume of the eggs released in gamete bundles by corals suffering from partial mortality was significantly lower than eggs released by corals that were apparently healthy (p = 8.52e-15). The eggs were almost 20% lower in terms of volume. This result could be due to the energy imbalance produced by partial mortality on the coral such that the eggs from colonies suffering from old injuries have lower lipid contents. These eggs with lower energy reserves would be at a disadvantage during embryonic development and long-distance dispersal in comparison to eggs produced by apparently healthy colonies with larger reserves. Decreased energy reserves have also been found in adult corals that have previously bleached (Jones and Berkelmans, 2011). Bleached colonies produce fewer eggs with lower lipid levels compared to healthy colonies (Ward et al., 2000; Sudek et al., 2012).

The difference in egg size did not affect the ability to be fertilized (Figure 6A). In fact, fertilization percentage was on average higher and much less variable in colonies with signs of partial mortality than the fertilization percentage from apparently healthy colonies. Another process that was not affected by the volume of the eggs was abnormalities in the developing embryos (Figure 6B). This would indicate then, that while smaller, the eggs produced by colonies suffering from partial mortality were of equivalent quality to the eggs produced by apparently healthy colonies. However, small egg size may have post-developmental repercussions such as reduced dispersal distances, ability to settle and colony growth.

With the results of this work we can conclude that partial mortality does affect the coral A. palmata and the negative effect is observed mainly as a constant energy expenditure, due to the process of competition against the colonizing organisms occupying the injured area, which in turn causes a decrease in the common energy reserves for other processes. Although the results show no negative effects on growth, we did detect a significant effect on the reproductive capacity of corals, specifically in the volume of the eggs, however, due to the low sample size used here and the large variability of this process more studies are needed to confirm these results. Finally, it is important to bear in mind that partial mortality is a condition that remains over time, consuming large amounts of energy and possibly weakening the coral, so that it is very likely that its resistance will decrease in the face of future stress events.

VP-G and AB developed the concept for this study and wrote the manuscript. VP-G conducted the field work, collected the data, and undertook statistical analyses.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.