Introduction
Within an ecosystem, sympatric species can co-occur largely due to ecological adaptations that reduce competition for resources and habitat predation risk (Grether et al. 2017, Lankau 2011, Perri and Randall 1999, Siepielski and McPeek 2010, Zhao et al. 2020). The resulting community structure, where a collection of species fills a variety of ecological niches, is one that is fundamental to ecosystem functioning (Genung et al. 2020, Petchey et al. 1999, Spaak et al. 2017). However, shifts in environmental conditions can cause changes in these dynamics and disrupt the overall ecosystem structure, potentially altering key environmental functions or services. Generally, those niches with lower functional redundancy are less resilient to change (Bellwood et al. 2003, reviewed by Biggs et al. 2020, Micheli and Halpern 2005). For example, in African savannahs, the exclusion of even one of the primary consumers of invasive and problematic shrub (Solanum campylacanthum ) led to increases in its abundance (Pringle et al. 2014). S. campylacanhum is toxic to livestock and has negative impacts on grazing species as well as vegetation (Pringle et al. 2014). Even though three herbivorous species are known to forage in the understory where the shrub grows, they function as complementary rather than redundant consumers.
Ecological specialists differ from generalists in their overall niche width (Clavel et al. 2010, Futuyma and Moreno 1988). That is, specialists occupy a narrower niche, typically with regard to some combination of habitat choice and food preference (Futuyma and Moreno 1988). Specialists have generally evolved to outcompete generalists, maximizing the exploitation of the selected resource. They thrive when environmental conditions are stable but can struggle to adapt in variable conditions (Scheiner 2002). Examples of the vulnerability of specialists to environmental perturbations exist throughout the fossil record (McKinney 1997) and modern-day specialists are similarly susceptible (Fourcade et al. 2021, Norden et al. 2013). The flexibility of generalists makes them more resilient to change, but the effectiveness of specialists makes them more difficult to replace. However, with enough functional redundancy, it is possible for multiple species of generalists to maintain an ecosystem service. For example, Memmott et al. (2007) showed that while specialist pollinators were more vulnerable to extinction, the network of resilient generalist pollinators is sufficient to sustain plant populations.
Grooming behaviors are common in terrestrial organisms, having been reported for mammals (Heine et al. 2017, Mooring et al. 2004), birds (Cox 2012, Goodman et al. 2020), and arthropods (El-Awami and Dent 1995, Invernizzi et al. 2015). These behaviors typically present in the form of autogrooming (grooming oneself), or social grooming (typically between conspecifics), and have two potential functions. First, the removal of parasites from the body of the client reduces the direct impacts of the parasites on the host and may also reduce the overall abundance of the parasites. Grooming can have a secondary effect of maintaining overall hygiene of external body surfaces, facilitating social interaction, and/or reducing stress (Soares et al. 2011).
Grooming of heterospecifics is rare in terrestrial systems and is limited primarily to generalists such as arthropod-feeding birds that will opportunistically eat insects and arachnids from the bodies of other animals (Samish and Rehacek 1999). The only apparent interspecific-grooming specialists in terrestrial systems are the two species in the family Buphagidae (oxpeckers, Bezuidenhout and Stutterheim 1980). In contrast to terrestrial species, intraspecific grooming is uncommon in marine organisms, having been reported for a small number of fish species (Clague et al. 2011, Sikkel 1986, Sikkel and Smit 2018). However, interspecific grooming, commonly referred to as “cleaning”, is well documented and is especially common in shallow reef systems.
In interspecific cleaning, certain fish or invertebrate species will obtain most of their dietary requirements by gleaning ectoparasites off other species of fishes. These interspecific mutualisms have evolved more than once in small, “picker-type”, fishes and in some shrimps (Vaughan et al. 2016). Cleaning interactions are of sufficient importance to individual “client” or “host” fishes that such fishes will interrupt other, biologically important activities (such as spawning) to get cleaned (Cheney and Côté 2003, Poulin and Grutter 1996, Sikkel et al. 2005). Within the many fish species that have evolved cleaning behaviors, most are “facultative” cleaners that clean occasionally or during specific life history stages but are overall not heavily dependent on consumption of external parasites. Examples of these include Thalassoma bifasciatum in the Caribbean (Dunkley et al. 2018, Feddern 1965, Itzkowitz 1979, Losey 1974), Coris julisin the Mediterranean (Zander and Sotje 2002), and Johnrandallia nigrirostris, in the Pacific (Quimbayo et al. 2017). Unique to coral reef systems, there are also obligate (or “dedicated”) cleaning specialists that rely primarily on the ectoparasites of client fishes as their main food source. (Vaughn et al. 2016). This group includes several species of Elacatinus gobies on coral reefs in the Caribbean region (Arnal and Côté 2000, Côté et al. 1998, Côté and Soares 2011) and Labroides wrasses on coral reefs in the Indo-Pacific (Côté and Brandl 2021, Grutter 1997a, Grutter 1997b, Losey 1972). Whereas wrasses tend to wander (even if associated with a “cleaning station”), cleaner gobies are often tightly associated with large boulder corals where they sit and await clients (Losey 1974, Whittey et al. 2021). While facultative cleaners tend to interact with a narrower diversity of client species compared to dedicated cleaners, they can at times eat a similar number of parasites (Grutter and Feeney 2016, Narvaez et al. 2015).
Cleaners provide numerous benefits to both individual clients and the overall community. Client-level effects of cleaners include facilitating wound healing (Foster 1985), reduced stress (Bshary et al. 2007, Soares et al. 2011), a decreased reliance on immune function (Ros et al. 2011), and increased client growth rates (Waldie et al. 2011). At the community level, the presence of cleaners impacts coral reef communities by locally increasing species richness and abundance (Bshary 2003, Grutter et al. 2003, Waldie et al. 2011). Among the myriad of services they provide, cleaners are best known for feeding on, and thus, removing the ectoparasites of client fishes. Although they consume several types of ectoparasites, the primary diet of cleaner fishes consists of gnathiid isopods (Arnal and Côté 2000, Grutter 1997a, Grutter and Feeney 2016, Losey 1974, Sikkel et al. 2004, Vaughan et al. 2016) and cleaning activity can significantly reduce gnathiid populations (Grutter et al. 2018).
Gnathiid isopods (“gnathiids”) are hematophagous arthropods that spend the majority of their lives in the benthic substrate but emerge to feed on a fish host (Sikkel and Welicky 2019, Smit and Davies 2004). As the primary “broker” of cleaner-client interactions they can impact hosts in multiple ways. These include influencing host behaviors such as spawning (Sikkel et al. 2005), and migration (Sikkel et al. 2017), as well as interactions between clients and cleaners (Grutter 2001, Sikkel et al. 2004), reducing hematocrit (Jones and Grutter 2005), increasing stress hormones (Triki et al. 2016), reducing juvenile fitness (Allan et al. 2021, Sellers et al. 2019), and killing the host (Artim et al. 2015, Hayes et al. 2011, Penfold et al. 2008, Sellers et al. 2019). Consequently, gnathiids can negatively affect host populations (Hayes et al. 2011, Penfold et al. 2008) and thus impact community dynamics (Coile and Sikkel 2013, Grutter et al. 2018, Sikkel and Welicky 2019).
Gnathiids are most active during crepuscular and nocturnal periods (Grutter 1999, Santos and Sikkel 2017, Sikkel et al. 2006) and will parasitize a wide range of fish species (Coile and Sikkel 2013, Hendrick et al. 2019, Jones et al. 2007, Nagel and Grutter 2007, Santos and Sikkel 2017, reviewed by Sikkel and Welicky 2019). They attach to hosts only temporarily (up to several hours for most species) to feed, but once engorged with blood/body fluids, they return to the benthos to digest and molt. Thus, gnathiids spend most of their life free-living (Smit and Davies 2004, Tanaka 2007). Because of this unusual lifecycle, they have been referred to variously as temporary ectoparasites, protelean parasites, and micropredators.
There are several methods by which external parasites are consumed that do not involve grooming but still impact the parasite burden of hosts by reducing parasite population densities (reviewed by Johnson et al. 2010). This includes concomitant predation as well as predation of free-living stages (Artim et al. 2017, Kaplan et al. 2009, Thieltges et al. 2013). Because gnathiids spend most of their lives not associated with a host, they are susceptible to consumption by many fish species, including the up to 70% of species that consume invertebrates (Kramer et al. 2015). However, only one study, on a few nocturnal carnivores, has examined consumption by non-cleaners. (Artim et al. 2017).
Tropical western Atlantic coral reefs possess little functional redundancy of cleaners, and dedicated cleaner fish provide an apparently unique and critical ecosystem service that may not be compensated for were they to suffer declines in abundance. Even dedicated invertebrate cleaners are unlikely to provide sufficient compensation in the absence of dedicated cleaner fish. For example, Ancylomenes pedersoni(Pederson’s cleaner shrimp), are dependent on the short-lived,Bartholomea annulata (corkscrew anemone, Huebner and Chadwick 2012, O’Reilly et al. 2018, Titus et al. 2017) that are patchy in their distribution.
In contrast to the mostly nocturnal gnathiids, the dedicated cleaners that are considered their primary consumers are diurnal, consuming them only off the hosts’ body, and most active at dawn (Côté and Molloy 2003, Grutter 1999, Pierera et al. 2022, Sazima et al. 2000, Sikkel et al. 2004). When determining functional redundancy, we must therefore widen the scope beyond dedicated cleaners to consider the degree to which facultative cleaner species and non-cleaners might provide additional sources of gnathiid consumption. The main goal of this study was, therefore, to test predictions of the hypothesis that coral reefs possess a sufficiently high level of functional redundancy of gnathiid consumption through some combination of facultative cleaners and additional functional groups of non-cleaner fishes that consume free-living gnathiids. Specifically, we tested whether: 1) other fish functional groups consume significant numbers of free-living stages of gnathiids and 2) whether consumption by facultative cleaners is comparable to dedicated cleaners on Caribbean coral reefs. We accomplished this by sampling gut contents from 61 reef fish species from 16 families, representing multiple feeding guilds, collected across three sites in the northeastern Caribbean.