451 Abstract—The smooth hammerhead (Sphyrna zygaena) is the third most Trophic ecology of the smooth hammerhead shark captured shark species in Peru, a nation with one of the largest shark (Sphyrna zygaena) off the coast of northern Peru fisheries in the Pacific Ocean. We sought to better understand the tro- Adriana Gonzalez-Pestana (contact author)1,2,3 phic ecology of this shark in north- 1,4 ern Peru by analyzing stomach Nicolas Acuña-Perales contents. From 2013 through 2015, Javier Coasaca-Cespedes1,2 we collected 485 samples of gut con- Francisco Cordova-Zavaleta1,5 tents from sharks measuring 53–294 Joanna Alfaro-Shigueto1,2,6 cm in total length. Our results show 1,6 that the smooth hammerhead is a Jeffrey C. Mangel top predator with a diet dominated Pepe Espinoza2,7 by jumbo flying squid (Dosidicus gi- gas) and the Patagonian squid (Do- Email address for contact author: [email protected] ryteuthis (Amerigo) gahi). Smooth hammerheads displayed different di- 1 ProDelphinus 5 Universidad Nacional Agraria La Molina ets with different body size, and this Calle Jose Galvez 780-e Avenida La Molina s/n finding indicates that sharks change Lima 18, Peru La Molina, Lima 12, Peru their distribution and habitat during 2 development. This study represents Universidad Científica del Sur 6 Center for Ecology and Conservation the most comprehensive investiga- Panamericana Sur Km 19 College of Life and Environmental Sciences tion to date of the trophic ecology Villa El Salvador, Lima 42, Peru University of Exeter. Penryn Campus of smooth hammerhead in waters 3 James Cook University Penryn, Cornwall TR10 9FE, United Kingdom off Peru. We propose that these wa- 1 James Cook Dr. 7 Instituto del Mar del Perú ters represent an important eastern Townsville City, Queensland 4811, Australia Esquina Gamarra y General Valle s/n Pacific Ocean feeding ground for 4 Universidad Ricardo Palma Chucuito, Callao 05, Peru smooth hammerhead. Because this Avenida Alfredo Benavides 5440 species is commercially important Santiago de Surco, Lima 33, Peru and also preys upon other commer- cial species, these findings could contribute to the design and imple- mentation of plans for ecosystem- based fisheries management. The smooth hammerhead (Sphyrna cies has been included in Appendix zygaena) is distributed from Califor- II of the Convention on Internation- nia to Chile within the eastern Pa- al Trade in Endangered Species of cific Ocean. This species reaches a Wild Fauna and Flora (https://cites. maximum size of 370–400 cm in total org/eng/app/appendices.php). Smooth length (TL), and is one of the largest hammerhead fins are prized in Asian fishes in the southeast Pacific Ocean markets and sharks are increasingly (Fowler et al., 2005). At birth, neo- targeted in some areas. Peru is one nates measure 50–61 cm-TL (Fowler of the top 12 countries that supply et al., 2005). Neonate and juvenile shark fins to Asian markets (Cheung smooth hammerheads use shallow and Chang, 2011). The smooth ham- coastal waters as nursery grounds merhead is commonly caught in the and as they grow larger they move to southeast Pacific Ocean but the im- Manuscript submitted 14 October 2016. more offshore oceanic waters (Smale, pact of these fisheries on its popula- Manuscript accepted 7 June 2017. 1991; Diemer et al., 2011; Francis, tion is unknown (Fowler et al., 2005). Fish. Bull. 115:451–459 (2017). 2016). The distribution of adults Peru also reports the highest accu- Online publication date: 31 July 2017. worldwide, however, is still unknown mulated historical shark landings in doi: 10.7755/FB.115.2 (Francis, 2016). the Pacific Ocean and smooth ham- Abundance of this species is de- merhead is identified as the third The views and opinions expressed or implied in this article are those of the creasing and the species is classified most captured shark species by the author (or authors) and do not necessarily as vulnerable in the IUCN Red List fisheries of Peru and the most fre- reflect the position of the National of Threatened Species (Casper et al., quently captured shark species off Marine Fisheries Service, NOAA. 2005). As of September 2014, the spe- northern Peru (Gonzalez-Pestana et 452 Fishery Bulletin 115(4) al., 2016). However, the fisheries lack robust W monitoring and management, and species bi- Zorritos ology and ecology remain poorly understood, Acapulco both locally and worldwide (Fowler et al., Cancas Ecuador 2005; Cortés et al., 2010). Mancora There are limited studies from Mexico, Ecuador, and Peru on the diet of smooth hammerhead in the Pacific Ocean. Research Peru shows that the diet of this shark in waters Yacila off Mexico is composed of fishes and cephalo- pods (e.g., California needlefish Strongylura[ exilis]; common clubhook squid [Onychoteu- TEP-MP n=230 this banksia]) (Galvan-Magaña et al., 1989; Ochoa-Díaz, 2009; Galvan-Magaña et al., WTSP-MP n=255 2013). In waters off Ecuador, information suggests that the diet is composed mainly of cephalopods (e.g., jumbo flying squid Dosidi[ - San Jose cus gigas]; purpleback flying squid Stheno[ - Pacific Ocean teuthis oualaniensis]; whip-lash squid [Mas- tigoteuthis dentata]; and sharpear enope squid [Ancistrocheirus lesueurii]) (Castañe- da and Sandoval, 2004; Estupiñan-Montaño and Cedeño-Figueroa, 2005; Bolaño Mar- Salaverry tínez, 2009). In one study in Peru, the diet S of smooth hammerhead was analyzed and smooth hammerheads were found to feed pri- Figure 1 marily on fishes (Pacific sardineSardinops [ Map of the study area and the 7 landing sites where smooth sagax]; Peruvian hake [Merluccius gayi pe- hammerhead (Sphyrna zygaena) were collected off northern Peru ruanus]; and Peruvian anchoveta [Engraulis from December 2012 through June 2015. The black line defines ringens]), as well as on cephalopods (Loligo the division between the Tropical East Pacific Marine Province spp., and jumbo squids) (Castañeda1). Al- (TEP–MP) and the Warm Temperate Southeastern Pacific Marine though this study in Peru had an adequate Province (WTSP–MP), n=number of stomachs collected from each sample size, time series, and size distribu- marine province. This map was created with Seaturtle.org Map- tion for the smooth hammerhead, it is more tool (Seaturtle.org Inc. website, accessed January 2017). descriptive than analytical and is limited to a seasonal comparison. Analysis of stomach contents We sought to better understand the trophic ecology of smooth hammerhead off the coast of northern Peru Prey items from stomach contents were analyzed at the by analyzing stomach contents. We assessed diet vari- Laboratorio de Ecología Trófica of the Instituto de Mar ability by sex, body size, location, season, year, and en- del Perú,2 and identified to the lowest possible taxon, vironmental conditions. counted, and weighed (wet weight). For identification of fishes and cephalopods, and their hard parts (oto- liths and beaks), the following identification guides Materials and methods were used: Iverson and Pinkas (1971); Wolff (1982, 1984); Clarke (1986); Chirichigno and Cornejo (2001); Collections, storage, and analysis of samples Garcia-Godos (2001); Lu and Ickeringill (2002); and Samples were collected from a small-scale driftnet fish- Xavier and Cherel (2009). Cephalopod beaks were used ery from December 2012 through June 2015 at 7 land- to reconstruct total mass at ingestion, by using regres- ing sites along the coast of northern Peru: Zorritos, sion equations (Lu and Ickeringill, 2002). Values for Acapulco, Cancas, Mancora, Yacila, San Jose, and Sa- stage of digestion were allocated to each prey item and laverry (Fig. 1). Nets in this fishery are typically set at ranged from 1 (little or no digestion) to 4 (advanced the time of sunset and retrieved the following morning state of digestion) (Bolaño Martínez, 2009). for an average set length of ca. 14 h (Alfaro-Shigueto Diet was quantified by using percentage of prey, by et al., 2010). Sharks were measured (total length) and number (%N), weight (%W), and frequency of occurrence sex was determined. Stomachs were extracted and pre- (%O) (Hyslop, 1980). The index of relative importance served in 10% formalin solution. (IRI) was calculated as IRI=%O (%N + %W). It was then divided by the total IRI for all items to express the 1 Castañeda, J. 2001. Biología y pesquería del “tiburón mar- tillo” (Sphyrna zygaena) en Lambayeque, 1991–2000. Inst. 2 and at the Laboratorio de Biologia Marina of the Universi- Mar Perú Inf. Prog. 139:17–32. [Available from website.] dad Cientifica del Sur. Gonzalez-Pestana et al.: Trophic ecology of Sphyrna zygaena off northern Peru 453 IRI as a percentage (%IRI; Cortés, 1997). Items rarely cause of biogeographic characteristics of the Tropical found in stomachs (e.g., rocks, snails) and parasites (e.g., East Pacific and Warm Temperate Southeastern Pacific isopoda) were not included in the analysis. marine provinces where the collection sites were locat- ed (Spalding et al., 2007) (Fig. 1). The division of the Statistical analysis seasons was based upon the seasonality of chlorophyll- a concentration and primary production; for which the Cumulative prey curves were constructed to deter- highest levels occurred during the austral summer and mine whether an adequate number of stomachs had fall (Pennington et al., 2006). Therefore, we divided the been collected to accurately describe the diet of smooth data into 2 seasons: season 1 (austral summer and fall) hammerhead (Jimenez and Hortal, 2003). The order in and season 2 (austral winter and spring). which stomachs were analyzed was randomized 1000 Nonmetric dimensional scaling (nMDS) ordinations times to eliminate bias. The number of stomachs an- generated from a Bray–Curtis similarity matrix on nu- alyzed is considered sufficient in describing the diet meric abundance of prey (%N) was used to determine when a cumulative prey curve reaches an asymptote. whether sex, body size, capture location, season, year, Therefore, a slope value less than 0.1 indicates a good or environmental conditions exerted the greatest over- representation of diet (Soberón and Llorente, 1993).