Checklist of Fishes from Madagascar Reef, Campeche Bank, México

Abstract This study presents the first list of fish species from Madagascar Reef, Campeche Bank, Gulf of México. Field surveys and literature review identified 54 species belonging to 8 orders, 30 families and 43 genera, comprising both conspicuous and cryptic fishes. Species richness was lower at this reef site compared to reefs in the Mexican Caribbean, Veracruz or Tuxpan, but was similar to other reefs in the same region. Species composition was a mixture of species present in all the reef systems of the Mexican Atlantic. Hypoplectrus ecosur was recorded here for the first time in the Gulf of Mexico, Mycteroperca microlepis, Equetus lanceolatus and Chaetodipterus faber were new records for the reefs of the Campeche Bank, Elacatinus xanthiprora was recorded for the second time in Mexico and expanded its known distribution westwards from Alacranes Reef and Sanopus reticulatus, endemic of the Yucatan state, was recorded here for the first time on a reef.


Introduction
Coral reefs are important centres of fish biodiversity. About 4000 fish species are associated with coral reefs around the world (Allen 2008), representing about 25% of all the species of marine fishes known today (Nelson 2006). Fishes are one of the most diverse groups of organisms in this ecosystem (Sale 2002) and occupy all consumer trophic levels and reef habitats (Holmlund and Hammer 1999). The absence or presence of certain guilds regulate the abundance of other reef organisms, such as corals or macroalgae, and can cause drastic changes in ecosystem states (Bellwood et al. 2006).
The Mexican Atlantic is characterized by several reef ecosystems with diverse fish communities, containing about 40% of all the reef species in the Western Atlantic (Floeter et al. 2008). Mexican Caribbean reefs host approximately 393 species (Schmitter-Soto et al. 2000), while the Gulf of Mexico (GoMx), which can be divided into the Tuxpan Reef System (TRS), Veracruz Reef System (VRS) and Campeche Bank (CB), has 376 species recorded . Although the species richness of these reef systems is alike, their similarity in species composition differs as a function of their environmental conditions and their connectivity (Chávez-Hidalgo et al. 2008). The Mexican Caribbean reefs are located in a tropical environment, whilst the TRS and VRS regions are more temperate. The CB reefs lie in between, receiving waters from the Caribbean through the Yucatan Channel that then travel to the inner areas of the GoMx, reaching the VRS and TRS. Thus, the reefs in the CB could act as stepping stones between the Mexican Caribbean reefs and the TRS/VRS (Jordán-Dahlgren 2002, Villegas-Sánchez et al. 2014).
The ichthyofauna of the TRS, VRS and Mexican Caribbean has been studied extensively, while most of the CB reefs lack information. Detailed lists of species for the TRS, VRS and Mexican Caribbean have been generated and updated (Schmitter-Soto et al. 2000, Del Moral Flores et al. 2013, González-Gándara et al. 2013). In contrast, icthyological studies in the CB are scarce and limited to just a few reefs: Cayo Arcas (Garduño and Chávez 2000), Cayo Arenas (Chávez 1966, Garduño andChávez 2000), Triangulos Oeste (Chávez 1966, Garduño andChávez 2000) and Alacranes reef (González-Gandara and Arias-González 2001). At least nine other recognized reefs within this system do not have information about their fish communities , and a further indefinite number of reefs remain to be described even at the most basic level (Zarco-Perelló et al. 2013).
Given their proximity to the Caribbean and their closeness to the shore (Chávez 1994, Zarco-Perelló et al. 2013, these reefs are potentially important centres of biodiversity and sources of fishery products for human communities living on the coast of the Yucatan state. Thus far, only three reefs known as Sisal Reefs have being researched, and only regarding some aspects of their benthic communities (Duarte et al. 2014, González-Muñoz et al. 2013, Ortigosa et al. 2013, Santana-Moreno et al. 2013, Zarco-Perelló et al. 2013. In this study, we improve the information known about one of these reefs by providing a list of cryptic, benthic and pelagic fish species associated with Madagascar Reef. The richness and composition found is then compared with other reefs systems of the Mexican Atlantic.

Study Site
Madagascar Reef is part of a cluster of three reefs named Sisal Reefs. These reefs are located in the Campeche Bank, a large carbonate platform of gentle slope (Fig. 1) that presents a low topographic complexity that is only increased sporadically by the presence of scattered reefs. Madagascar Reef is located 40 km from the fishing port of Sisal, whose inhabitants exploit its marine resources; however, fishermen from nearby ports, such as Celestun and Progreso, also visit the reef. Its morphology is peculiarly elongated, extending 2.5 km East-West and 130 m North-South in its widest point. The reef rises from sandy plains at 14m of depth to the reef crest at 4 m depth, where the illumination is high and the water current stronger (Zarco-Perelló et al. 2013).

Data collection
Fish species occurrence was registered during two different surveys. During the first campaign (2007), seven transects of 50 m were deployed and all fishes on sight were photographed along the way for later identification (Fig. 2). The second campaign (2010) focused on cryptic species, which historically have received less attention (MacNeil et al. 2008). In this campaign, six transects of 50 m were deployed and species were counted visually and/or collected using clove oil (eugenol) diluted to 10% with 70% ethanol and sea water (Cunha and Rosa 2006). Collected fishes were preserved in 70% ethanol for later identification in the laboratory. The surveys included areas from the shallow reef crest (5 m) to the deep sandy plains around the reefs (25 m). The taxonomic identification of the species was based on Robins and Ray (1999), Böhlke and Chaplin (1993), McEachran and Fechhelm (1998), McEachran and Fechhelm (2006), Humann and DeLoach (2008). The nomenclature was revised using the Catalog of Fishes (Eschmeyer and Fricke 2013). The order of the species in the list was arranged following Nelson (2006) for the suprageneric categories, while the genera and species were ordered alphabetically. Information about the general geographic distribution of the species was based on Felder and Camp (2009), Carpenter (2002a), Carpenter (2002b and Colin (2010) for the species of the genus Elacatinus. All the species collected and photographed were registered in the Ichthyology Collection (YUC-PEC-239-01-11) of the Unidad Multidiciplinaria de Docencia e Investigacion Sisal (UMDI-Sisal), a research station of the National Autonomous University of Mexico (UNAM). The list of species found in our field work was complemented with species documented in the M.Sc. thesis work of Martínez de la Portilla (2008) who surveyed Madagascar Reef in 2005.

Species richness estimation
To evaluate if more sampling effort is needed to register all the fish species present on Madagascar Reef we calculated a species accumulation curve using the software EstimateS v.9. which utilizes a novel method developed by Colwell et al. (2004) and Colwell et al. (2012) that links rarefaction and extrapolation for presence/absence data samples. The species richness at Madagascar Reef is similar to other reefs in the CB, but lower than richness values reported for reefs in the VRS and TRS (Fig. 7). Of the four reefs in the CB with ichthyological information, three present lower species richness than Madagascar Reef. Cayos Arcas and Arenas have 37 species registered, whereas Triangulos Oeste has 52 ). Only Alacranes reef surpasses these numbers with 294 species (González-Gandara and Arias-González 2001). All the reefs in the TRS present higher richness. For example, Medio Reef has the lowest with 83 and Lobos Reef has the highest with 248 species, while each of the remaining reefs have approximately 100 species (González-Gándara et al. 2013). Similarly, Enmedio Reef in the SAV has 145 species ). The estimation of the species richness through the species accumulation curve clearly shows that the list of species presented here is not an exhaustive compendium of the species inhabiting this reef and that more species remain undiscovered (Fig. 8). With only three sampling campaigns comprising about 15 surveys (including Martínez de la Portilla  (2008)) Madagascar Reef presented about 1/7 of the reef fish species in the Gulf of Mexico (376 species) . This numbers are lower than other better studied reefs but still quite significant given the low coral cover in the reef (Zarco-Perelló et al. 2013). The implementation of more studies at Madagascar Reef and other reefs in the CB not only would probably increase its species richness, reflecting the fish biodiversity of nearby hotspots such as Alacranes Reef (González-Gandara and Arias-González 2001) and the Caribbean (Schmitter-Soto et al. 2000) but also has the potential to extend distribution ranges of species from the Gulf of Mexico and Caribbean by finding new species records in this region, as was the case in the present study. Fish species accumulation curve for Madagascar Reef. Calculated using sample-based incidence data with the software EstimateS v9 (http://purl.oclc.org/estimates) using Eq. 5 in Colwell et al. (2004) and Eq. 17 in Colwell et al. (2012) for rarefaction and Eq. 18 in Colwell et al. 2012Colwell et al. (2012 for extrapolation. Bars represent 95% confidence intervals calculated using Eq. 6 in Colwell et al. (2004) for rarefaction and Eq. 19 in Colwell et al. (2012) for extrapolation. For a full output of the analysis see Suppl. material 3.

Author contributions
S.Z.P. participated in the design of the study, in the collection of samples, in the identification of species, interpretation of the data, and in the writing of the manuscript; R.M.M. participated in the design of the study, in the collection of samples and identification of species; N.S. participated in the design of the study, collection of samples and assisted in the writing of the manuscript.