310 Abstract—The blue shark (Prionace Food habits of the blue shark, Prionace glauca glauca) is the most landed shark species in Peruvian shark fisheries, (Linnaeus, 1758), in waters off northern Peru representing 42% of total landings. Despite its importance for fisheries, 1,2 the ecological role of the blue shark Francisco Córdova-Zavaleta (contact author) in Peruvian waters remains poorly Jaime Mendo2 understood. Therefore, in this study, Sergio A. Briones-Hernández3 we aimed to assess the food habits 1 of blue sharks off northern Peru Nicolas Acuña-Perales through stomach content analysis. Adriana Gonzalez-Pestana1,4 Between February and December of Joanna Alfaro-Shigueto1,4,5 2015, 143 stomachs were collected Jeffrey C. Mangel1,5 and 28 found to be were empty. In- dividuals measured from 110.0 to 299.4 cm in total length. The results Email address for contact author: [email protected] indicate that blue sharks have epi- pelagic and mesobathypelagic feed- 1 ProDelphinus 4 Facultad de Biología Marina ing habits, preying upon a wide Calle Jose Galvez 780-e Universidad Científica del Sur number of prey taxa (42 species) 15074 Lima, Peru Panamericana Sur Km 19 but with a diet dominated by cepha- 2 Facultad de Pesquería 15067 Lima, Peru lopods, especially Argonauta spp. Universidad Nacional Agraria la Molina 5 Center for Ecology and Conservation and Ancistrocheirus lesueurii. Diets Avenida La Molina s/n College of Life and Environmental Sciences differed by size class and location, 15024 Lima, Peru University of Exeter, Penryn Campus suggesting that longitudinal move- 3 Centro Interdisciplinario de Ciencias Marinas Penryn, Cornwall TR10 9FE, United Kingdom ments are related to increments in Instituto Politécnico Nacional body size. In addition, we propose Avenida Instituto Politécnico Nacional s/n that blue sharks scavenge for food Colonia Playa Palo de Santa Rita Apdo. on the basis of finding the cyprid 23096 La Paz, Baja California Sur, Mexico larval stage and juvenile cosmopoli- tan duck barnacle (Lepas anatifera) associated with the prey item ‘un- identified cephalopods.’ The present study contributes new information on the diet of blue sharks. In the southeastern Pacific Ocean, the al.,1 2016). Seventy percent of blue area off northern Peru is an important shark landings come from the small- area of convergence between temper- scale longline fishery operating along ate (Humboldt Current) and tropical the coast (Doherty et al., 2014), es- (South Equatorial Current) waters pecially off southern Peru where (Spalding et al., 2007). This particular cold to temperate waters support area hosts a diversity of marine spe- higher abundances of pelagic sharks cies (Chirichigno and Cornejo, 2001), (Adams et al., 2016). Likewise, it is including a large number of sharks one of the most frequently species (Cornejo et al., 2015). One of these caught in Peruvian small-scale drift- is the blue shark (Prionace glauca), net shark fisheries (Alfaro-Shigueto which is considered a key cosmopoli- et al., 2010). However, despite its tan pelagic species that exerts top- importance for fisheries, biological down control on community structure studies describing the ecological role Manuscript submitted 14 December 2017. (Stevens et al., 2000). of blue sharks in Peruvian waters Manuscript accepted 16 August 2018. On a global scale, the blue shark are few. Fish. Bull. 116:310–322 (2018). is the most landed shark species (as There have been several studies Online publication date: 11 September 2018. target and nontarget catch) and the regarding the diet of blue sharks in doi: 10.7755/FB.116.3-4.9 major component of the international shark fin trade (Clarke et al., 2006; The views and opinions expressed or 1 Gonzalez-Pestana, A., C. Kouri, and X. implied in this article are those of the Nakano and Stevens, 2008). It is also Velez-Zuazo. In review. Shark fisher- author (or authors) and do not necessarily the most landed species by Peruvian ies in the southeast Pacific: a 61-year reflect the position of the National shark fisheries, representing 42% of analysis from Peru. F1000Research Marine Fisheries Service, NOAA. total landings (Gonzalez-Pestana et 3:164. [Available at website.] Córdova-Zavaleta et al.: Food habits of Prionace glauca in waters off northern Peru 311 sagax), Peruvian hake (Merluccius gayi peruanus), and eggs from flying fish (Exocoetidae) and Chilean Ecuador silverside (Odontesthes regia) (Hoyos et al., 1991; Elliot et al., 1995, 1996). Although these previous studies provide valuable insights into prey species of blue sharks off the coast of Peru, their results have been limited by sample size, time series, and size classes. Therefore, to determine the ecological role of blue sharks in the Peruvian marine ecosys- tem, we focused on providing broader information on blue shark food habits off northern Peru by ana- lyzing stomach contents and assessing diet variabil- ity by sex, size class, season, and fishing ground. Peru Materials and methods Sampling and stomach content analysis Pacific Ocean Stomach contents of blue sharks were collected between February and December 2015 by onboard observers during small-scale driftnet and longline fishing trips that landed in Mancora and Sala- verry ports in northern Peru (Fig. 1). Sex of each shark was determined and total length (TL) was measured to the nearest centimeter (Bigelow and Schroeder, 1948). Stomachs were extracted and pre- served in 10% formalin for analysis (Galván-Maga- ña et al., 1989). Stomach samples were examined at the Laboratorio de Recursos Hidrobiológicos of the Universidad Nacional Agraria la Molina, where Figure 1 prey items were identified to the lowest possible Map of locations where blue sharks (Prionace glauca) were taxon and weighed (wet weight) to the nearest 0.01 collected in northern Peru between February and December g. Fish items (entire specimens and skeletons) were 2015. The blue vertical gradient defines the position of the identified according to Chirichigno and Velez (1998) Peru–Chile Trench (with an average distance from shore of and Clothier (1950), and fish otoliths were identi- 130 km or 70 nautical miles). fied according to Rivaton and Bourret (1999) and García-Godos (2001). Cephalopods were identified the eastern Pacific Ocean. For example, Tricas (1979) by examination of beaks (Wolff, 1982, 1984; Clarke, and Harvey (1989) reported that blue sharks consume 1986; Xavier and Cherel, 2009). Crustaceans and ma- euphausiids (mostly Thysanoessa spinifera), small fish rine mammals were identified according to Moscoso (e.g., the northern anchovy, Engraulis mordax), and (2013) and Jefferson et al. (2015), respectively. Recon- a variety of cephalopods (e.g., Loliginidae and Histio- struction of cephalopod weights by regression equa- teuthidae). Studies from the Mexican Pacific Ocean tions was based on measurements of the lower rostral have revealed a high consumption of the pelagic red length of cephalopod beaks (Wolff, 1982, 1984; Clarke, crab (Pleuroncodes planipes), as well as a variety of 1986; Smale et al., 1993; Xavier and Cherel, 2009), squid, including Histioteuthis heteropsis, Gonatus cali- whereas reconstruction of fish weights was based on forniensis, and Ancistrocheirus lesueurii (Markaida and the relationship of length to weight (Fernández, 1987; Sosa-Nishizaki, 2010; Hernández-Aguilar et al., 2016). IMARPE2) and the relationship of otolith radius to to- Loor-Andrade et al. (2017) showed that A. lesueurii, tal length (Goicochea and Arrieta, 2008) when possible. Stigmatoteuthis hoylei, and the jumbo squid (Dosidi- To assess whether the sample size collected was ap- cus gigas) were the most consumed species off Ecuador. propriate to describe the diet of blue sharks, we con- In Chilean waters, prey species included a variety of structed a cumulative prey curve at the family level fishes: mackerels Trachurus( murphyi and the Pacific from the total number of prey species identified, ex- chub mackerel, Scomber japonicus) and tunas (Thun- cluding ‘unidentified remains’ (i.e., cephalopod, fish, nus spp.), as well as jumbo squid (Lopez et al., 2010; and crustacean) prey items by using the software Es- Klarian et al., 2018). In Peru, three studies of blue shark diet have re- 2 IMARPE (Instituto del Mar del Peru). 2012. Crucero 1202- vealed that they feed on jumbo squid, Peruvian ancho- 04 de “Evaluación hidroacústica de los recursos pelágicos.” veta (Engraulis ringens), Pacific sardine Sardinops( Inst. Mar Peru, Inf. Ejec., 49 p. [Available from website.] 312 Fishery Bulletin 116(3–4) n timateS, vers. 9.1.0 (Colwell, 2013). The order of sam- TP = 1 + (Σ j=1 DCij ) × (TPj ), (2) ples was permuted 1000 times to reduce bias. Follow- where DC = the composition of the prey j in the diet of ing the method of Jiménez-Valverde and Hortal (2003), ij the predator I; we plotted the results to analyze the suitability of the TP = to the trophic level of prey j; and sample size. When the curve approaches the asymp- j n = the number of prey species in the diet of tote (slope<0.1), the number of samples is deemed to predator j. be sufficient to describe the diet (Soberón and Llorente, 1993). Furthermore, we used the calculation method Values of trophic position for fish prey were obtained 3 proposed by Bizzarro et al. (2007) to reinforce the sam- from Froese and Pauly and Espinoza (2014); and pling assessment. This method states that the slope for cephalopod prey from Cortés (1999) and Espinoza of the line generated from the curve endpoints (mean (2014). cumulative number of prey taxa generated for the final To identify possible differences in diet, individual 4 stomach samples) should be compared to a line of 0 sharks were analyzed according to 5 factors: sex, sea- slope to establish whether a cumulative prey curve has son, size class, latitude of fishing ground, and longitude reached an asymptote. Slopes are compared by using of fishing ground. The analysis by sex was performed Student’s t-test, where slopes that are not significant to clarify whether the composition of diets of females (P>0.05) indicate that the curve has reached an asymp- and males is related to the reported spatial segregation tote (Bizzarro et al., 2007).