The artisanal elasmobranch fishery of the Pacific coast of Baja California Sur, Mexico, management implications ; Pesquería artesanal de elasmobranquios en la costa Pacifico de Baja California Sur, México, implicaciones para su gestión

1 Centro Interdisciplinario de Ciencias Marinas, Instituto Politécnico Nacional, Av. IPN s/n Col. Playa Palo de Santa Rita, CP 23096, La Paz, Baja California Sur, Mexico. E-mail: galvan.felipe@gmail.com 2 Marine Biology Research Division, Scripps Institution of Oceanography, University of California, San Diego, 9500 Gilman Dr., La Jolla, CA 92093-0204, USA. 3 Universidad Autonoma de Baja California Sur, Km. 5.5 Carretera al Sur, CP 23080, La Paz, Baja California Sur, Mexico. 4 Laboratorio de Ecologia Pesquera, Departamento de Ecologia, CICESE, Ensenada, Baja California, Mexico.


INTRODUCTION
Artisanal fisheries are generally characterized as small-scale traditional fisheries using small amounts of capital and energy and small fishing vessels, which make short fishing trips close to shore, primarily for local consumption. These fisheries often take place in remote areas of developing countries. With the widespread adoption of motorization, artisanal fisheries have grown significantly over the past two decades, contributing more than 25% of the world catch and accounting for half of the fish used for direct human consumption. The rapid expansion of artisanal fishing capacity under open access regimes has begun to exert overfishing pressures on coastal fisheries resources (Mathew 2001).
Elasmobranchs (i.e. sharks and rays) are currently one of the resources of greatest concern in artisanal fisheries. Worldwide overfishing of several species has caused population declines of their populations, mainly because they are k-selected fishes with slow growth and low reproductive rates (Bonfil 1994, Camhi 1998, Walker 1998, Musick 1999.
In Mexico, artisanal fisheries account for approximately 40% of the total national catch and comprise up to 80% of the elasmobranch fishing effort (Arreguin-Sanchez et al. 2004). Mexico has been one of the most important elasmobranch fishing nations in the world; in 2007 it had the sixth largest catch of elasmobranchs, representing 4.3% of total world catch, with approximately 34638 t , FAO 2009. In 2008, 102807 vessels were recorded in Mexican artisanal fisheries, exploiting mainly coastal finfish, sharks, crustaceans, mollusks and echinoderms (Ramirez-Rodriguez 2011). This fishery represents an important source of employment, providing both sustenance and income for some of the poorest sectors of Mexican society (Arreguin-Sanchez et al. 2004, Ponce-Diaz et al. 2009, Cartamil et al. 2011.
The management of sustainable elasmobranch fisheries in Mexico has been hampered by a lack of reliable data. For example, official records of elasmobranchs recognize only three categories: sharks (sharks larger than 150 cm total length [TL]), cazones (sharks smaller than 150 cm TL) and rays (all batoids) (Bonfil 1997, Castillo-Geniz et al. 1998, Galvan-Magaña 2009. Detailed quantitative information on the specific composition of the catch is a basic requirement to determine possible effects of fishing on populations of target species and to establish baselines for comparison of biological diversity (Bonfil 1997, Marquez-Farias 2002. In recent years the artisanal elasmobranch fisheries have been described in several regions of northwestern Mexico, including Sonora (Bizzarro et al. 2009a), the east coast of Baja California (BC) (Smith et al. 2009), the east coast of Baja California Sur (BCS) (Bizzarro et al. 2009b), Sinaloa (Bizzarro et al. 2009c) and the Pacific coast of BC (Cartamil et al. 2011).  BCS was the state with the second largest catch of elasmobranchs, representing 17% (4004 t) of the total catch on the Mexican Pacific coast (SAGARPA 2009). However, there have been no studies describing elasmobranch artisanal fisheries on the highly productive Pacific coast of BCS.
The objectives of the present study were to identify and describe the Pacific coast BCS artisanal elasmobranch fishery, to determine elasmobranch species composition and catch rates, and to provide biological information (size, sex, seasonality) for the most abundant species captured in BCS.

Study area
BCS has an area of 73677 km 2 , representing 3.8% of the land mass of Mexico and comprising the southern half of the Baja California Peninsula. It has 2220 km of coastline: 1400 km of which corresponds to the Pacific coast (between 28°16' and 22°33'N) (Fig.1). In 2010, the population of BCS was 637027, indicating a relatively low population density (Cortes-Ortiz et al. 2006, INEGI 2011.
The Pacific coast of BCS is characterized by a narrow continental shelf that is generally less than 37 km wide with a peak width of approximately 68.5 km between Laguna San Ignacio and the area north of the Bahia Magdalena lagoon complex (Fig. 1). This region is affected by the California Current System, which dominates during the cool part of the year, and by the northward intrusion of a branch of the tropical North Equatorial Current during the warm part. It has a high primary productivity driven by coastal upwelling (Alvarez-Borrego 1983, Espinoza-Carreon et al. 2001, Zaytsev et al. 2003.

Artisanal fishery survey
The artisanal elasmobranch fishery on the Pacific coast of BCS was surveyed during the period 2000-2010. The location of each camp was determined with a global positioning system (GPS) unit. Sites supporting artisanal fishing activities were characterized based on the level of infrastructure (Bizzarro et al. 2009a): A, little to no infrastructure; B, moderate infrastructure; and C, significant infrastructure. We determined whether camps were permanent (P) or temporary (T), and the number of active artisanal fishing vessels.
Elasmobranch sampling was conducted at 16 artisanal camps. Elasmobranchs landed at the camps were identified to the lowest possible taxon and enumerated. Sharks of the genus Mustelus were sometimes grouped into the category Mustelus spp. due to the difficulty of identifying them to species level. In addition, species-specific size and sex composition data were collected. Standard measurements used were total length (TL; using the natural extension of the caudal fin) and disc width (DW), recorded to the nearest 0.5 cm (Compagno 2001).
Total elasmobranch landings by season were assessed according to species composition. Catch per unit effort (CPUE) was defined as the number of individuals per vessel per trip, and was calculated for each species within each major gear type and for each season surveyed. Measured specimens were used to determine size composition and sex ratio of landings. Sex-specific size composition data were evaluated for normality (Kolmogorov-Smirnov test) and homoscedasticity (F test) and compared using a t-test or Mann-Whitney U test, as appropriate. As a result, differences in the size composition between the sexes were compared using raw size data in all cases. The sex ratio among landings was evaluated using chi-square analysis with Yates correction for continuity for specimens directly examined (Zar 1996). Histograms of size and sex composition of each species with ≥50 measured individuals were computed and compared with size at maturity (taken from several scientific literature sources: , Villavicencio-Garyazar 1995, 1996a,b, Compagno 2001, Ebert 2003, Villavicencio-Garayzar and Bizzarro 2004, Perez-Jimenez et al. 2005, Shoou-Jeng and Hua-Hsun 2005, Castillo-Geniz et al. 2007, Bizzarro et al. 2007a, Bejarano-Alvarez et al. 2010, Carrera-Fernandez et al. 2010, Hoyos-Padilla et al. 2012. Sample size-sufficiency for each fishing gear and each seasonal catch composition was verified using cumulative taxon curves (Gotelli and Cowell 2001). Seasons were defined as follows: spring (22 March-21 June), summer (22 June-21 September ), autumn (22 September-21 December) and winter . To determine whether the sampled landings were adequate to describe the catch compo-sition, the average curve of accumulated number of elasmobranch taxa present in each vessel was plotted against the number of vessels grouped at random (Ferry and Cailliet 1996). Catch composition of 1000 randomly selected vessels was re-sampled (EstimateS 8.2.0.), and used to calculate a mean and standard deviation estimate for each sample (Colwell 2009). Linear regression was used to determine quantitatively whether the curve reached an asymptote, signifying an adequate number of samples (Bizzarro et al. 2007b).
The Shannon-Wiener index (H') was used to evaluate the levels of elasmobranch diversity within the study area, using the equation: where S is the number of species in the sample, p i is the proportion that the ith species contributes to the total abundance of the sample (p i =N i /N), N i the number of individuals of the ith species, and N the number of individuals in the sample. Nonparametric permutation and bootstrap statistical methods (Manly 2007) were applied to estimate the mean and variance of the diversity index of the Pacific coast of BCS.
Elasmobranch dominance within the study area was further tested using the Simpson index (D) with the equation where n i is the number of organisms of a particular species and n the total number of organisms of all species. Zoogeographic affinity analysis was done according to the biogeographical divisions suggested for the Pacific coast of BC Peninsula by several authors (Briggs 1974, Walker 1960, Hastings 2000, Horn et al. 2006, Spalding et al. 2007, Robertson and Cramer 2009. The following divisions were considered: Californian Province (corresponding to temperate waters) from Point Conception, California to Bahia Magdalena, BCS, and the Cortez Province (corresponding to tropical waters), which extends south from Bahia Magdalena to the Gulf of California (Fig. 1). Based on these studies we divided the study area into north and south zones (relative to Bahia Magdalena). We calculated the diversity index for each zone and made comparisons between zones using a t-test. The Jaccard index (Jac) was calculated as a measure of similarity of species composition for each zone. This presence-absence index takes into account the relationship between the number of common species and total species found in the samples being compared, and is calculated as Jac a a b c ij = + + where a is number of shared species (present in both zones, i and j), b is the number of species present only in zone i and c is the number of species present only in zone j.  Table  1. Biogeographical zones: CP, California province; COP, Cortez province.

Fishing sites and global fishery characteristics
A total of 60 artisanal fishing camps were documented along the Pacific coast of BCS ( Fig. 1). High variability was observed in fishing camp size, from small temporary camps to permanent camps with significant infrastructure. Table 1 details the comparative characteristics of the documented camps. Tables 2 and 3 indicate the number of days and trips sampled for each season and fishing gear. Table 1. -Characteristics of the surveyed artisanal camps. "Type" refers to infrastructure: A, little to no infrastructure; B, moderate infrastructure; C, significant infrastructure. "Perm" refers to whether a camp is permanent (P) or temporary (T). "Target" refers to the fishery typology : EG, elasmobranch gillnet; EL, elasmobranch longline; T, teleosts; I, invertebrates (A, abalone; L, lobster; C, crab; G, gastropod; B, bivalve; H, sea cucumber; O, octopus; S, shrimp). FV = number of fishing vessels present during the surveys (or minimum/maximum in case of seasonal fluctuations in effort). Asterisks denote artisanal camps where elasmobranch biological data were sampled. Elasmobranchs were targeted at 75% (n=45; Table  1) of the artisanal camps, as either the primary (n=6) or the secondary target (n=39). Most camps contained moderate infrastructure (57%, n=34). The artisanal elasmobranch fishery was conducted with small vessels of less than 10.5 m length, locally called "pangas". These fishing vessels were open-hulled fiberglass boats with outboard motors. The number of vessels varied among camps, ranging from two (e.g. La Freidera camp) to 113 (e.g. Adolfo Lopez Mateos camp) ( Table 1).
Bottom set gillnets were the most common fishing gear used, and were deployed on the continental shelf (<100 m depth). These nets were monofilament with lengths of 200-800 m, a drop of up to 6 m, and highly variable mesh sizes, ranging from 7.6 to 25.4 cm. On average, two gillnets were used per vessel.
Pelagic sharks were targeted mainly with longlines, which ranged from 1.5 to 3 km in length. The number of hooks per longline was highly variable (range 250-400), and set depths ranged from 5 to 10 m. The hook type commonly used is the "J-style hook" with a 6-8 cm length. On average, two longlines were used per vessel, and fishing time ranged from 10 to 15 hours.
In most artisanal camps, elasmobranchs were dressed, iced and sold fresh to local buyer or cooperatives. However, in some camps elasmobranchs were filleted, dried and salted; this was observed mainly in remote camps with more difficult access. Prices for elasmobranchs varied between seasons by species, size and buyers, but typically ranged between MX$7.00 and $14.00 per kilogram.

Seasonality
CPUE for sharks and rays differed considerably between seasons (Table 5). In the shark group, the blue shark was captured in all seasons, while among batoids, R. productus was the dominant species, though its capture decreased significantly in winter. Overall, the CPUEs were greatest in spring because of high catch rates of R. productus (8.34±1.54), P. glauca (8.19±1.68) and M. californica (5.18±1.95) ( Table 5). Echinorhinus cookei, Negaprion brevirostris, Squalus acanthias and Torpedo californica were recorded only in summer.
In autumn the CPUEs were lower than in spring and summer (

Biological information
For gillnet-captured elasmobranchs, size distribution differed significantly between males and females for C. ventriosum, G. galeus, H. francisci, I. oxyrhinchus, M. californicus, M. henlei, P. glauca, S. californica, G. marmorata, M. californica, R. glaucostigmata, R. productus and Z. exasperata. In addition, the male-to-female sex ratio differed significantly from the expected 1:1 ratio for the following gillnet-captured elasmobranch species: C. falciformis, C. ventriosum, G. galeus, H. francisci, I. oxyrhinchus, M. henlei, S. zygaena, S. californica, G. marmorata, M. californica, R. glaucostigmata, R. productus and Z. exasperata. For longline-captured elasmobranchs, size distribution differed significantly between males and females for C. falciformis, P. glauca, and S. zygaena, while sex ratio differed significantly from the expected 1:1 ratio for I. oxyrhinchus, P. glauca and S. zygaena. Information detailing the size and sex composition of elasmobranchs captured by gillnet and longline is summarized in Figures 2, 3, and 4. Table 6 provides detailed information in sex ratios of all captured elasmobranchs, and Table 7 provides information on the size structure of infrequently captured species.

Diversity index and biogeography
The species richness and diversity recorded in BCS was high with 52 elasmobranch species documented, the Shannon-Wiener index value (H') of 2.34±0.01 SE and the Simpson index (D) of 0.86. There was a higher species richness in the north zone (46 species), resulting in a high diversity index (H'=2.32±0.02 SE) and dominance index (D=0.84), while the south zone (26 species) had lower richness; the diversity index was 2.19 (±0.14), but the dominance index was high (D=0.83). Significant differences (t=14.93; p<0.001) were found in the diversity indices estimated between north and south zones. Also, the Jaccard index between the two zones was low (J=0.42), indicating that there is a difference in the species composition of elasmobranchs from the sites studied.

Survey of artisanal camps
In conjunction with previous studies (Bizzarro et al. 2009a, Smith et al. 2009, Bizzarro et al. 2009b, Bizzarro et al. 2009cand Cartamil et al. 2011, the results of the present study allow for the first time a comprehensive characterization of artisanal elasmobranch fisheries throughout the Mexican northwest. In this region, the Pacific coast of BCS represents one of the largest catch regions of elasmobranch artisanal fisheries. A total of 60 artisanal camps were recorded and the area was surpassed in number only by the 83 camps on the east coast of BCS (Bizzarro et al. 2009b).
Elasmobranchs were the primary target in relatively few camps (6 camps). This may be due to the economics of this fishery, including high operating costs, fuel, oil, equipment, gear, poor infrastructure, limited marketing channels and consequently lower prices for products (Ponce-Diaz et al. 2009). On the Pacific coast of BCS the artisanal elasmobranch fishery was conducted by licensees (concessionaires), free fishers and cooperatives.
Elasmobranchs were often caught incidentally in teleost fisheries, such as the California halibut (Paralichthys californicus) gillnet fishery, in which elasmobranch bycatch such as R. productus and M. henlei was recorded. In addition, fishermen in some camps were opportunistic and focused catch on abundant local fauna, which varied according to the season, due to the low price of elasmobranchs (7-14 pesos kg -1 ) compared with other fishery resources in the study area [e.g., California halibut (20-40 pesos kg -1 ), white seabass (Atractoscion nobilis; 40-60 pesos kg -1 ), abalone (300-500 pesos kg −1 ) and lobster (150-250 pesos kg −1 )]. Landings in the present study were dominated by sharks (61%). Compared with the shark fishery that began in 1930, batoid fisheries in Mexico are a relatively new activity. In the late 1980s, demand increased as shrimp vessels began to commercialize their batoid bycatch. In the early 1990s, due to a drop in the production of Pacific sharpnose shark (Rhizoprionodon longurio) and growing market demand, artisanal fishermen began to engage formally in targeting batoids (Cudney-Bueno and Turk-Boyer 1998). Currently the artisanal batoid fishery is increasing, and official data show that in BCS from 2008 (983 t) to 2009 (1504 t) there was an increase of over 50% (Marquez-Farias andBlanco-Parra 2006, SAGARPA 2009).

Biological data
The prevalence of juveniles of several elasmobranch species (e.g. G. galeus, C. ventriosum and M. californica) throughout the year and the frequency of occurrence of small size classes of several shark species (e.g. P. glauca and I. oxyrinchus) within landings suggests that considerable fishing effort may be opportunistically directed on breeding or nursery areas. Along the Pacific coast of BCS there are three major coastal lagoon systems (Guerrero Negro-Ojo de Liebre, Laguna San Ignacio and Bahia Magdalena; Fig. 1), which represent highly productive and diverse environments (Ibarra-Obando et al. 2001). It has been documented that coastal elasmobranch species use bays, estuaries and lagoons as foraging, resting, mating and nursery grounds (Pratt and Carrier 2001, Heupel and Simpfendorfer 2005, Farrugia et al. 2011.
Sexual segregation has been reported in elasmobranchs (Sims 2005) and in the present study several species had a tendency to be caught in same sex grouping, such as R. productus, M. californica, Z. exasperata and species of the Mustelus genus. This is a behavior that can occur due to differences in sex, body size, behavior, nutritional requirements and/or habitat selection (Klimley 1987, Magurran and Macias-Garcia 2000, Sims 2005, Wearmouth and Sims 2008). R. productus and P. glauca, the two most abundant elasmobranch species recorded, were also the most important species in landings along the Pacific coast of BC (Cartamil et al. 2011). Shovelnose guitarfish was also the species most captured in the upper Gulf of California (Marquez-Farias 2002, Bizzarro et al. 2009a, Smith et al. 2009. High fishing pressure recorded in this species could cause population declines, so this species should be monitored closely.

Seasonality
Seasonal variation in targeted elasmobranch species was recorded along the Pacific coast of BCS. Several species recorded in the landings could be involved in seasonal migrations (e.g. R. productus, M. californica) (Bizzarro et al. 2007a). Temperature plays an important role in the seasonal migrations of several elasmobranchs species in nearshore waters (Talent 1985, Wallman and Bennett 2006, Wiley and Simpfendorfer 2008. Fishing effort was greater during the spring and summer seasons than in autumn and winter. This lateyear decrease was caused primarily by adverse oceanographic conditions due to the presence of northwesterly winds (Jaramillo et al. 2004).

Diversity and biogeography
Estimates of diversity levels in fishery studies can be useful because changes can be detected in the structure of commercially exploited populations (Tavares and Arocha 2008). Further, it has been suggested that managers can use the diversity of species in an area and the presence of foundation species as indicators of marine ecosystem functioning (Bracken et al. 2007). The application of diversity indices in elasmobranch fisheries biology is relatively recent, and few studies using this technique have been reported (e.g. Worm et al. 2003, Tavares andArocha 2008).
The Shannon-Wiener index estimated for the study area suggests high levels of elasmobranch diversity. The number of elasmobranch species (52) recorded on the Pacific coast of BCS is higher than that reported in any other region of northwestern Mexico. This high richness and diversity included species of different affinities and ecological requirements; this finding can be attributed to the convergence of currents found in the study area, as well as topography and bathymetry. Currents converging on the Pacific coast of BCS are the cold-temperate California current, and the North Equatorial Current with warm tropical characteristics (Hickey 1979, Lynn andSimpson 1987). These currents influence elasmobranch species composition (e.g. tropical species such as R. glaucostigma and N. brevirostris and temperate species such as T. californica and Alopias vulpinus) (Hubbs 1960, Walker 1960, Lluch-Belda et al. 2003, Dawson et al. 2006. We observed a significant change in the abundance of some species to the south and north of Punta San Lazaro, Bahia Magdalena (Table 8). Bahia Magdalena is located at the confluence of several water masses and is therefore considered a transition zone (Brinton and Reid 1986). Our results also suggest that from Laguna San Ignacio to Bahia Magdalena there exists a transi- Table 8. -Arrangement of species according to marine province divisions. The first column represents species distributed along the entire Pacific coast of BCS. The "North of Bahia Magdalena" column represents species restricted to the California Province. The "South of Bahia Magdalena" represents species restricted to the Cortez Province. Species are arranged from highest to lowest relative abundance.
tion zone between the Californian Province and Cortez Province; this is corroborated by the mixed tropicaltemperate species found in this study. This transition zone is a possible boundary for the distribution of some elasmobranchs (e.g. Rhinoptera steindachneri, Rhinobatos glaucostigma and Myliobatis longirostris).

Management implications
Despite the importance of elasmobranch fishing in Mexico, few regulations have been instituted for this fishery. The first step for management was taken by the Mexican National Institute of Fisheries, which recommended a moratorium on issuing new shark-fishing permits beginning in 1993 (Castillo-Geniz et al. 1998), which was carried out in 1998 . This was followed by the development of a National Action Plan for the Conservation and Management of sharks, rays and related species in Mexico in 2004 by the Comision Nacional de Agricultura y Pesca (CONAPESCA-INP 2004). Finally, a regulation known as NOM-029- PESCA-2006(DOF 2007 instituted in 2007 aimed to protect and ensure proper fisheries management and conservation for elasmobranchs (DOF 2007).
Our results will serve as a baseline for determining future changes in the artisanal fishery. For proper management in Mexico it is necessary to improve fishery statistics and there is a need for a continued monitoring programme to obtain specific information about captures by sites and seasons. It is also important to incorporate an ecosystem approach to fisheries management, which represents a substantial change in perspective and poses equally substantial challenges (Bracken et al. 2007). In addition, our survey indicates that considerable fishing pressure is occurring upon juvenile elasmobranchs, likely within their nursery habitat. Further research into the determination and protection of these nursery areas is critical for the sustainability of elasmobranch populations (Branstetter 1990).
After decades of elasmobranch exploitation in waters of the Pacific coast of BCS, there have likely been decreases in populations and changes in size structure among less fecund species (Stevens et al. 2000). A specific case that was observed in the present study is that of the angel shark, which was exploited significantly in Laguna San Ignacio and Bahia Magdalena many years ago (Villavicencio-Garayzar and Abitia-Cardenas 1994). In the present study this species was not recorded in either zone; its catch was restricted only to the zone north of Laguna San Ignacio. This suggests that fishing pressure has caused the decline of its populations. Genetic studies suggest that the angel shark captured on the Pacific coast of the BC Peninsula is going through a "bottleneck" process caused by overfishing (Ramirez-Amaro et al. 2011), so particular attention should be paid to the protection of this species.