Coral Reefs (2018) 37:827–840 https://doi.org/10.1007/s00338-018-1709-7 REPORT Variation in coral-associated cryptofaunal communities across spatial scales and environmental gradients 1 1 2 Chelsie W. W. Counsell • Megan J. Donahue • Kyle F. Edwards • 1 3 Erik C. Franklin • Mark A. Hixon Received: 2 March 2018 / Accepted: 20 June 2018 / Published online: 5 July 2018 Ó Springer-Verlag GmbH Germany, part of Springer Nature 2018 Abstract Most of the diversity on coral reefs is in the wave height, and chlorophyll-a were significant drivers of cryptofauna, the hidden organisms that inhabit the inter- occurrence. Depth and percent live coral tissue were also stitial spaces of corals and other habitat-forming benthos. identified as important correlates for community compo- However, little is known about the patterns and drivers of sition with distinct responses across taxa. Analyzing spe- diversity in cryptofauna. We investigated how the crypto- cies-specific responses to environmental gradients faunal community associated with the branching coral documented a unique pattern for the guard crab Trapezia Pocillopora meandrina varies across spatial scales and intermedia, which had a higher probability of occurring on environmental gradients. We performed nondestructive smaller colonies (in contrast to 18 other common taxa). visual surveys of the cryptofaunal community on 751 P. The results of a principal coordinates analysis on com- meandrina colonies around the island of O‘ahu (30–73 munity composition and a co-occurrence analysis further colonies per site, 3–6 sites per region, five regions). We supported T. intermedia as having a unique distribution identified 91 species, including 48 fishes and 43 inverte- across colonies, even in comparison with four other Tra- brates. Most of these species were observed rarely, with pezia species. Overall, these patterns emphasize the only 19 species occurring on greater than 5% of surveyed importance of host coral characteristics (i.e., colony size colonies. Variation in community abundance and species and percent live tissue) and physical characteristics of the richness was greatest at the scale of the coral colony and surrounding habitat (i.e., wave energy, chlorophyll-a, and lowest at the site scale. Abundance and species richness depth) in structuring cryptofaunal communities and char- increased with increasing colony size and maximum wave acterize species-specific responses to environmental height, and decreased with increasing surface chlorophyll- gradients. a. In an analysis of species-specific responses, colony size, Keywords Community ecology Á Cryptofauna Á Hawaiian Topic Editor Morgan S. Pratchett Islands Á Pocillopora meandrina Á Environmental gradients Á Spatial scales Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00338-018-1709-7) contains supple- mentary material, which is available to authorized users. Introduction & Chelsie W. W. Counsell [email protected] Most of the diversity on coral reefs is in the cryptofauna, the hidden species that inhabit the branches, crevices, and 1 School of Ocean and Earth Science Technology, Hawai‘i Institute of Marine Biology, University of Hawai‘i at Ma¯noa, interstitial spaces of corals and other habitat-forming ses- Ka¯ne‘ohe, HI 96744, USA sile organisms (Reaka-Kudla 1997; Plaisance et al. 2011). 2 Department of Oceanography, University of Hawai‘i at Reef-associated cryptofauna constitute 91% of the known Ma¯noa, Honolulu, HI 96822, USA species on coral reefs (Stella et al. 2010) and are a critical 3 Department of Biology, University of Hawai‘i at Ma¯noa, component of coral reef trophic webs. Cryptofauna capture Honolulu, HI 96822, USA and recycle nutrients by consuming very small prey items 123 828 Coral Reefs (2018) 37:827–840 (e.g., plankton, detritus, coral mucus), and they are a pri- one of two categories: large visual surveys focused on a mary food source for many reef fishes, including squir- few key species (e.g., Sin and Lee 2000; Holbrook et al. relfishes, wrasses, triggerfishes, snappers, and groupers 2008; Stier and Leray 2014), or thorough, albeit destruc- (Randall 1967; Enochs 2012; Leray et al. 2015). Despite tive, sampling limited to a small number of colonies (e.g., their abundance and importance, reef cryptofauna are Austin et al. 1980; Coles 1980; Black and Prince 1983; under-represented in traditional reef surveys and, as a Gotelli and Abele 1983; Britayev et al. 2017;Lo´pez-Pe´rez result, relatively little is known about the composition of et al. 2017). Previous surveys have identified host colony these communities and the ecological processes that size as a strong correlate with the number of individuals structure them. and diversity of species in the associated community One of the most diverse coral-associated cryptofaunal (Abele and Patton 1976; Stella et al. 2010; Holbrook et al. communities is associated with living and dead corals in 2011). In addition, many studies have suggested that a the family Pocilloporidae (Stella et al. 2010). Pocilloporids decline in host coral health due to tissue bleaching or are structurally complex, reef-building corals that are mortality shifts the composition of the decapod community common, especially on exposed reefs, and widespread from a few obligate species to a more diverse group of throughout much of the Indo-Pacific. The relatively small facultative species (Coles 1980; Stewart et al. 2006; size and spatial isolation of individual Pocillopora colonies Enochs and Hockensmith 2008; Plaisance et al. 2009; facilitates studies of discrete, replicate communities. Given Stella et al. 2010, 2011; Enochs and Manzello 2012; Leray the tractable nature of these communities, Pocillopora-as- et al. 2012), although some obligate species, including sociated assemblages have been the focus of research on Trapezia crabs, have been observed on dead coral colonies species interactions and cryptofaunal distribution since the (Preston and Doherty 1990; Stella et al. 2011; Head et al. 1960s. Previous research on Pocillopora-associated com- 2015). munities has identified cryptofauna-coral host mutualisms While some research has suggested that Pocillopora- for trapeziid crabs, alpheid shrimps, and damselfishes. associated communities vary over space as a result of Crabs in the genus Trapezia and shrimps in the genus environmental drivers (Abele 1976; Austin et al. 1980; Alpheus help protect their host corals from corallivores, Black and Prince 1983;Lo´pez-Pe´rez et al. 2017), previous including the predatory gastropod Drupella cornus studies were limited in spatial extent (two to four sites) and (McKeon and Moore 2014), the cushion star Culcita did not directly investigate the correlation between specific novaeguineae (McKeon et al. 2012; McKeon and Moore environmental factors and community composition. Envi- 2014), and the crown-of-thorns seastar Acanthaster planci ronmental factors including depth, reef zone, and wave (Pratchett 2001; McKeon et al. 2012: McKeon and Moore energy can drive shifts in community composition for 2014; Rouze´ et al. 2014). These mutualistic decapods also corals (Franklin et al. 2013; Gove et al. 2015), non-cryptic increase the growth and survival of host corals by remov- reef fishes (Nunes et al. 2013; Jankowski et al. 2015; ing sediments from the coral tissue (Stewart et al. 2006; Darling et al. 2017), and cryptofaunal communities not Stier et al. 2012; Stewart et al. 2013; Rouze´ et al. 2014) and directly associated with coral hosts (Klumpp et al. 1988; reducing negative interactions with vermetid snails (Stier Depczynski and Bellwood 2005). et al. 2010). Damselfishes associated with Pocillopora Here, we examined how environmental factors influence colonies promote coral growth and survival through a the composition of Pocillopora-associated cryptofaunal variety of mechanisms, including territorial defense that communities, providing a broader ecological context for minimizes predation from other reef fishes (Gochfeld 2009; the existing experimental work on these communities. We Chase et al. 2014), sleep-swimming behavior that circu- included measures of colony size and percent live tissue, lates water between coral branches at night (Goldshmid factors highlighted in previous studies, to reflect the habitat et al. 2004), and excretion-based nutrient enrichment quality of the host coral. We hypothesized that species (Holbrook et al. 2008). In addition to describing coral exhibit taxon-specific responses across gradients of depth mutualisms, several studies have reported that some spe- and wave energy, resulting in unique community compo- cies in the coral-associated community can affect the sitions. We also considered the effect of host density, occurrence and survival of other species through predation hypothesizing that the abundance of obligate species on and territorial behavior (Schmitt et al. 2009; Holbrook et al. each colony will decrease with increasing availability of 2011; Stier et al. 2013; Stier and Leray 2014; Leray et al. adjacent host colonies. Further, we hypothesized that 2015). increasing benthic complexity at the site scale would Given the focus of Pocillopora research on species decrease the species richness of the Pocillopora-associated interactions and the considerable effort required to community due to increased habitat structure available for exhaustively sample cryptofaunal communities, most sur- facultative species. Finally, we investigated whether veys of Pocillopora-associated communities can be fit into abundance