Coral reefs are biodiverse ecosystems that provide coastal protection, food and other economic benefits to coastal and island communities. Scleractinia corals form the structural framework of this habitat. Key requirements for the success and longevity of reefs is for larval corals to settle onto reefs, grow into larger colonies and regularly reproduce. Unfortunately, numbers of juvenile corals have recently decreased on many reefs (Edmunds et al., 2015; Hughes et al., 2019), suggesting that settlement and recruitment processes may be declining. Due to the small size of coral recruits and many possible stressors, specific factors attributing to this decline are challenging to document on reefs.

Ideal settlement substrates are one component contributing to successful coral recruitment. Some coral larvae preferentially settle on reef bedrock covered with crustose coralline algae (CCA; Rhodophyta family Corallinaceae). This relationship is related to chemical signals released by the CCA and/or associated microorganisms (Heyward and Negri, 1999; Negri et al., 2001; Sneed et al., 2014; Tebben et al., 2015). The presence of CCA is generally associated with high reef health or quality (Ritson-Williams et al., 2009). However, CCA has been replaced in some shallow reefs by encrusting red-brown colored algae of the Peyssonneliacean family, including Ramicrusta spp. (Pueschel and Saunders, 2009; Eckrich et al., 2011; Ballantine et al., 2016; Edmunds et al., 2019; Nieder et al., 2019). While Ramicrusta is known to have similar form and growth patterns as CCA, it behaves differently in that it frequently overgrows mature corals (Eckrich and Engel, 2013; Nieder et al., 2019). There is limited evidence of juvenile coral recruitment on Ramicrusta, with studies finding very limited (Edmunds et al., 2019) to no coral recruits on Ramicrusta coated surfaces (Eckrich and Engel, 2013). There are also differences in the microbial biofilms harbored by CCA and Ramicrusta which could additionally impact settlement (Wilson et al., 2020). Ramicrusta is considered invasive to shallow Caribbean reefs and studies suggest that the rapid spread of Ramicrusta may leave little suitable settlement substrate available for coral larvae (Eckrich and Engel, 2013; Edmunds et al., 2019).

While there is evidence that Ramicrusta is a threat to adult corals, little is known about the effects of Ramicrusta on coral larvae and the settlement process. Here, we conducted a laboratory study using larvae from the Caribbean corals Porites astreoides and Favia fragum to investigate the impact of Ramicrusta and CCA solutions on larval mortality/survivorship and settlement. We show that Ramicrusta exposures cause high larval mortality compared to CCA.

For both species of coral, larval mortality was the highest in the Ramicrusta treatments and there was a significant effect of treatment on larval mortality (H test, p < 0.003, Supplementary Table 1 ). For P. astreoides, mean larval mortality was 96% (±8.9% standard deviation, SD) for both the 1 and 2 mg ml^-1 Ramicrusta treatments ( Figure 2A ). Mortality in the 1 mg ml^-1 CCA treatment was 4% (±8.9% SD) and there was no mortality in the 2 mg ml^-1 CCA or sterile seawater treatments. Significantly more P. astreoides larvae survived in the control and 2 mg ml^-1 CCA compared to the Ramicrusta treatments (p = 0.031, H test with Tukey comparison), but the other post-hoc comparisons were either not significant (p > 0.05) or non-testable due to the lack of significant difference between the rank sums ( Supplemementary Table 1 ). There was 100% mortality in the F. fragum treatments with Ramicrusta and 10% (±20% SD) and 5% (±10% SD) mortality in the 0.5 and 1 mg ml^-1 CCA treatments, respectively with complete survival of larvae in the sterile seawater ( Figure 2B ). Post-hoc statistical comparisons were either not significant (p > 0.05) or non-testable due to no significant difference between the rank sums ( Supplemementary Table 1 ).

There was a significant effect of treatment on larval settlement and survivorship for both species of coral (H-test, p < 0.005, Supplementary Table 1 ) and the highest settlement and surface interaction occurred in the CCA treatments. Mean P. astreoides settlement/surface interaction was 68% (±23% SD) and 40% (±37% SD) for the 1 and 2 mg ml^-1 CCA treatments, respectively ( Figure 2C ). One P. astreoides larva survived in each 1 mg ml^-1 and 2 mg ml^-1 Ramicrusta treatment. While both larvae showed evidence of settlement/surface interaction, the settlement/surface interactions in the Ramicrusta treatments were significantly lower than in the 1 mg ml^-1 CCA treatment ( Figure 2C , p = 0.019). Mean F. fragum settlement/surface interaction was 65% (±19.1% SD) and 75% (±10% SD) for the 0.5 and 1 mg ml^-1 CCA treatments, respectively ( Figure 2D ). Significantly more larvae interacted with the surface in the CCA 1 mg ml^-1 compared to each Ramicrusta treatments (p = 0.021), which is partially attributed to the complete mortality of F. fragum larvae exposed to Ramicrusta.

Ramicrusta is a rapidly expanding encrusting algae that is especially prolific on shallow Caribbean reefs. Here we used exposure experiments to show that Ramicrusta causes mortality in larvae from two species of prominent brooding Caribbean corals, suggesting that Ramicrusta may be toxic to the larvae. We observed limited larval mortality (<10%) in our CCA exposure experiments, and 40–80% of these CCA-exposed larvae exhibited settlement or surface interactions, thus suggesting that the experimental conditions and solution concentrations were appropriate for the larvae. Notably, one P. astreoides in each Ramicrusta treatment (1 and 2 mg ml^-1) survived and showed evidence of settlement/substrate interaction, signifying that select coral larva may tolerate Ramicrusta, which could signify metamorphosis under stress (Kitchen et al., 2022).

Our data suggest that Ramicrusta may contain or produce metabolites toxic to coral larvae. A recent study showed that Ramicrusta textilis produced caffeine, among other metabolites (Weber et al., 2022). Caffeine is a common deterrent of herbivores and pathogens by land plants (Kim et al., 2010), and it has adverse growth impacts on insect larvae (Srinivasan and Kesavan, 1977; Maguire et al., 2017). The impact of caffeine, as well as other Ramicrusta producing metabolites, on coral larvae is an area for further investigation.

Our study exposed coral larvae to Ramicrusta that was dissolved in solution at specific concentrations. While our CCA exposure experiments at similar doses resulting in settlement suggests that these concentrations may be in-line with natural exposure and other larval studies (Pollock et al., 2017), future studies could expose larva to surfaces with encrusted with Ramicrusta (instead of solutions) to verify the impact of typical surface-type exposures on larval survival and settlement. Here we used solutions to achieve homogeneity in all replicates, which may be more inconsistent with Ramicrusta surfaces due to differential biofilms on the surfaces. Additionally, we only evaluated two concentrations of Ramicrusta solutions and response could be evaluated with lesser dilutions of Ramicrusta. However, our results are in line with prior observations of no to limited coral settlement on Ramicrusta coated surfaces (Eckrich and Engel, 2013; Edmunds et al., 2019). Our experiments used replicate wells within a plate for each treatment which may introduce a plate-type effect in our study. Future experiments should randomize treatments within each plate and/or use separate plates for each replicate within a treatment to strengthen the experimental design and resulting conclusions. It should also be noted that only two species of coral larvae were used in this experiment, and expansion to additional species are warranted, as well as molecular genotyping to verify the species of Ramicrusta. Here we observed slightly elevated settlement in F. fragum compared to P. astreoides larvae which signifies species-related differences. Indeed, F. fragum typically settles within the first 24 hours (Goodbody-Gringley, 2010), and P. astreoides can maintain longer larval durations (Harrison and Wallace, 1990).

Overall, our results add to the growing knowledge that Ramicrusta is harmful to corals and is a potential threat to Caribbean reefs. Research that explores herbivory controls as well as other intervention mechanisms to remove or inhibit Ramicrusta spread on reefs is urgently needed to prevent further reef decline.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions include study conception, design and funding (AA, TM), experimentation (KC, NA, SF), data analysis (KC, AA), writing (KC, AA), and editing (all authors). All authors contributed to the article and approved the submitted version.