Considerations for maximizing the adaptive potential of restored coral populations in the western Atlantic

Abstract

Active coral restoration typically involves two interventions: crossing gametes to facilitate sexual larval propagation; and fragmenting, growing, and outplanting adult colonies to enhance asexual propagation. From an evolutionary perspective, the goal of these efforts is to establish self-sustaining, sexually reproducing coral populations that have sufficient genetic and phenotypic variation to adapt to changing environments. Here, we provide concrete guidelines to help restoration practitioners meet this goal for most Caribbean species of interest. To enable the persistence of coral populations exposed to severe selection pressure from many stressors, a mixed provenance strategy is suggested: genetically unique colonies (genets) should be sourced both locally as well as from more distant, environmentally distinct sites. Sourcing three to four genets per reef along environmental gradients should be sufficient to capture a majority of intraspecies genetic diversity. It is best for practitioners to propagate genets with one or more phenotypic traits that are predicted to be valuable in the future, such as low partial mortality, high wound healing rate, high skeletal growth rate, bleaching resilience, infectious disease resilience, and high sexual reproductive output. Some effort should also be reserved for underperforming genets because colonies that grow poorly in nurseries sometimes thrive once returned to the reef and may harbor genetic variants with as yet unrecognized value. Outplants should be clustered in groups of four to six genets to enable successful fertilization upon maturation. Current evidence indicates that translocating genets among distant reefs is unlikely to be problematic from a population genetic perspective but will likely provide substantial adaptive benefits. Similarly, inbreeding depression is not a concern given that current practices only raise first-generation offspring. Thus, proceeding with the proposed management strategies even in the absence of a detailed population genetic analysis of the focal species at sites targeted for restoration is the best course of action. These basic guidelines should help maximize the adaptive potential of reef-building corals facing a rapidly changing environment.

Keywords: adaptive potential; assisted gene flow; biomarkers; coral restoration; genetic diversity; inbreeding; outbreeding; phenotypic resilience; population enhancement; species selection; unintended selection.

Figure 1

Overview of basic restoration guidelines to maximize the adaptive potential of reef‐building corals facing a rapidly changing environment. A genet is defined as a genetically unique colony or collection of colonies (ramets) that can trace their ancestry back to the same sexual reproductive event (i.e., they stem from the same settler and, hence, share the same genome).

Figure 2

Coral reproduction. (A) Most reef‐building species in the Caribbean are self‐incompatible hermaphroditic broadcast spawners. Adult colonies release egg–sperm bundles that float to the surface where they break apart and mix with gametes from other colonies of the same species. After fertilization and development, larvae settle onto the reef, metamorphose into primary polyps, and grow into new genets. (B) Over time, genets may fragment. These fragments can reattach to form new ramets of the same genet. A genet is defined as a genetically unique colony or collection of colonies (ramets) that can trace their ancestry back to the same sexual reproductive event (i.e., they stem from the same settler and, hence, share the same genome). Adopted from Devlin‐Durante et al. (2016).

Figure 3

Capturing allelic diversity for coral conservation. Sampling of relatively few genets from a population is sufficient to capture most common alleles. (A) Proportion of alleles observed when sampling a certain number of genets (N _ind) from a population, depending on the allele frequency. For an allele of frequency P, the probability of not observing it among N diploid genets is (1 − P)^2N. Hence, the probability of observing such an allele, which is the same as the proportion of such alleles across the whole genome that are observed, is 1 − (1 − P)^2N. (B) The proportion of common alleles (allele frequency P > 0.05) depends on their allele frequency in three species of reef‐building corals and can be calculated based on single nucleotide polymorphism (SNP) data. The three species (Acropora millepora, Orbicella faveolata, and Acropora cervicornus) have broadly similar distributions of their common alleles, despite substantial phylogenetic and geographic separation. (C) Proportion of common SNP alleles represented within a certain number of sampled genets. A sample of just three genets captures >50% of all SNP alleles found in more than 5% of all genomes in the population, and 12 genets capture 90% of them. (D) Proportion of common alleles represented calculated based on microsatellite loci for Acropora palmata. Note that due to higher mutation rate of microsatellites they require four rather than three genets to represent >50% of alleles.