Biology of the razor clam (Ensis magnus Schumacher, 1817), in the ría de Pontevedra (NW Spain)application to fishery managemen

Résumé

Small-scale or artisanal fisheries produce nearly half of the world¿s catch and employ 50 million of the 51 million people engaged in fisheries (FAO 2008). However, most scientific studies have focused on the analysis of industrial fisheries, and little attention has been given to small-scale fisheries. Bivalves are among the most important resources in S-fisheries (Small-Scale and Spatially Structured fisheries targeting Sedentary resources with artisanal gears, Orensanz et al., 2005a) worldwide (landings around 3 million t, FAO 2013) and some species have high commercial value in international markets, represent substantial food resources and play important economic and social roles. Razor clams (Bivalvia, Solenoidea) are distributed throughout the world, in the Pacific, the Atlantic and the Mediterranean Sea ranging from tropical to temperate areas inhabiting fine sand, silt or sandy-mud ocean floors (Hayward and Ryland, 1998). These species represent a substantial food resource that has experienced a considerable economic valuation development during the last decade. The species most commonly commercialized are Ensis magnus Schumacher, 1817 [syn. E. arcuatus (Jeffreys, 1865)], E. siliqua (Linnaeus, 1758) and E. directus (Conrad, 1843). The sword razor E. magnus is the razor clam with the highest commercial value in European markets (according to the Eurostat information database) and the most important commercial species of razor clam in Spain. Within Spain E. magnus is mainly exploited in Galicia (NW Spain). The increasing demand and the new capture methods introduced in some regions during the last decades (hydraulic and suction dredging or electrical fishing) leaded to sudden increases of landings and subsequently several razor clams beds showed signs of overexploitation (Hall et al., 1990; Fahy and Carroll, 2007; Hauton et al., 2007; Espinoza et al., 2010, FAO 2015). This situation is enhanced by the recruitment variability (del Piero and Dacaprile, 1998; Szarzi et al., 1995) and the slow turnover of the species (Fahy and Carroll, 2007). Although the Galician razor clams beds are harvested manually by diving with a very selective method, they have also been more heavily exploited in the last years, being their maximum sustainable fishing rate still unknown though. Moreover, in winter, coinciding with reproductive season (Darriba et al., 2004) up to 40% of sword razor specimens break at the foot when harvested, and must be discarded, as they are not marketable. To avoid the sword razor discards, the harvest is totally closed for one-two months during winter. This market shortage is overcome by the importation of razor clams mainly from Argentina and Chile, which are less than a third of the price of the local product. Together with the Spanish economic crisis, this has generated a 50% fall in the market price of the local product. Knowledge of marine resources biology is essential for fisheries management. Although a great effort has been conducted to study E. magnus biology during the last two decades (e.g. Robinson and Richardson, 1998; Fahy et al., 2002; Darriba et al., 2004, 2005a, 2010; da Costa et al., 2008, 2011, 2013; Varela et al., 2009; Ruiz et al., 2011, 2012), some important biological information for fisheries management such as growth parameters or size at first maturity has been little studied. Indeed, none of these features were previously described for the species in the Iberian Peninsula. Moreover, razor clams, which like many other benthic organisms are sedentary, inhabit spatially heterogeneous environments and are distributed in dispersed beds, which generate a spatial variability in their population dynamics between near areas (eg. Robinson and Richardson, 1998; Rabaoui et al., 2007). This variability could even happen within a small spatial range like the Galician rias. This fact complicates the assessment, as reported for other S-fisheries (Cochrane, 1999; Walters and Pearse, 1996; Wilson et al., 2010). Thus, biological studies of the population dynamics need to take into account the spatial variability of this species. Many authors have pointed out that simple assessment and management procedures are needed for small-scale fisheries management. These procedures should be based on indicators determined directly by the catch/landings data or by simple surveys by actively involving fishers (e.g. Orensanz et al., 2005a,b; Prince, 2010). Moreover, S-fisheries need simple local tools to enable adaptive and flexible management that takes into account biological and social aspects (Parada et al., 2012; Prince, 2010; Macho et al., 2013). In Galicia, S-fisheries operate under a co-management system, based on territorial user rights for fishing (TURFs) between cofradías (local fishers¿ guilds) and the fisheries administration (Molares and Freire 2003, Macho et al 2013). These entities cooperate for developing and implementing the annual management plans with the help of a Technical Assistant, an on-site fisheries advisor who design the fisheries management plans according to the status of the resources (Macho et al, 2013). The adaptive character of the management plans allows including the aforementioned simple tools and procedures in daily harvesting strategies (Molares and Freire, 2003; Macho et al., 2013; Parada and Molares, 2013). Of the 26 razor clam harvesting plans in Galicia, which involve a total of 35 cofradías, we have selected as case of study the one developed for E. magnus in the Ría de Pontevedra as a result of the fishers and Technical Assistant demand for biological methods and simple management tools that contributes to the fishery sustainability. Moreover, this razor clam harvesting plan is one of the most productive in Galicia, with landings that accounted for 78 t in 2014, worth ~ 0.52 million € (www.pescadegalicia.com). The Ría de Pontevedra is a partially mixed estuary with a mesotidal and semidiurnal regime located NW of the Iberian Peninsula, at the northern limit of the North Atlantic upwelling system (Wooster et al., 1976). It is a large (25 km long) V-shaped indentation in the coast that gradually widen from the innermost area to the open sea (4 - 12 km wide). According to many studies (e.g. Dale and Prego, 2002) the ria is divided into three well defined zones determined by the presence of the Lérez River, water stratification and topography: the innermost zone, strongly influenced by the Lérez River; the outer zone which has a strong oceanic influence, and the intermediate zone. The sword razor fishery of the Ría de Pontevedra is managed as a joint harvesting plan agreed between seven cofradías of the ria. Exploitation is organized through a rotational harvesting strategy in which six areas are exploited, with a close season of two months during the peak reproductive season (February-March, Darriba et al., 2004) (www.pescadegalicia.com). We selected our sampling sites taking into account the oceanographic divisions of the ria and the exploitation areas of the management plan. In this context, this PhD dissertation aims to support the E. magnus fishery management by studying the spatial heterogeneity in reproduction and growth of the species in the Ría de Pontevedra in order to promote a biologically informed harvesting strategy. We have focused in the main biological aspects that determine the dynamic of the species and that represent the most important features for fishery management: 1) reproduction, essential for the establishment of the close season or the commercial size, and 2) growth, necessary to know the age at which individuals become part of the exploited biomass or to determine how populations respond to exploitation. Both biological features are also fundamental for stablishing the harvesting rotational scheme, a key factor in the management of this fishery. This work was done in close collaboration with the TA and the fishers involved in the management plan. The main objective of the reproductive cycle study is to adapt the rotational harvesting strategy and the close season to the differences of the gonadal cycle development along the ria. This would avoid the foot breakage if its relationship with gonad maturation is confirmed. The study also aims to investigate the relation of the reproductive cycle to the main environmental variables and to estimate the size at first maturity (defined as the length at which 50% of the population is mature, L50). Gametogenic development of E. magnus was studied in six shellfish beds of the Ría de Pontevedra (one in each exploitation area) during two years. Study sites were distributed from the innermost to the outermost part of the ria being separated from each other no more than 15 km. To determine L50, additional razor clams were collected at the innermost part of the ria, where the smallest individuals were found. Results showed that the sword razor started to mature between 60 and 70 mm, reaching L50 at 79.5 mm and L95 at 110 mm, although the observed data confirmed that 100% of the sampled individuals were mature between 100 and 105 mm. This result is in accordance with that obtained in Scotland (Muir and Moore, 2003). The reproductive cycle was characterized by a resting stage during summer and early autumn, initiation of gametogenesis in autumn and a period of successive spawning interspersed with gonad recovery during winter and spring, as previously observed in the Ría de Vigo (Darriba et al., 2004). However, a 15-day to one-month delay in advanced stages of gametogenesis and maturation was observed between the inner and the outermost site of the ria, as well as an extended spawning period in the outermost area. Lower bottom seawater temperatures at the outermost sites appeared to delay maturation and to prolong the spawning periods, whereas salinity fluctuations at the innermost sites appeared to reduce the length of the cycle. Both results are in agreement with the previously reported for E. magnus (Darriba et al., 2004, 2005a) and other species of the Superfamily Solenoidea (Cross et al., 2014; López et al., 2005). Finally, the highest foot breakage was observed throughout winter, during maturity, postspawning and gonad recovery stages, confirming the relationship between foot breakage and gonadal development. In view of these results, and after explaining them to the sector, the razor clam fishery in the Ría de Pontevedra adapted the rotation harvesting scheme and the close period to the gonadal cycle. Nevertheless, the gametogenic cycle may differ from year to year influenced by environmental factors, making impossible to apply the same schemes every year. In order to provide a simple tool that enables adaptation of the exploitation to the gonadal stage of the resource, we develop two rapid, simple and inexpensive methods that provide accurate information about the gonadal stage of E. magnus: percentage coverage of the gonad on and over the digestive gland (gonad coverage hereinafter) and examination of the gonad smear. Although similar gonad coverage scales were established before for other razor clams (Aracena et al., 2003; Remacha-Triviño and Anadón, 2006), no gonad smear scale was previously developed for razor clams. Both methods were validated with most accurate and precise histological and gonadal index methods, to construct a correspondence table between all methods. The correspondence table constitutes a simple tool that can be used in any razor clam fishery to monitor the gonad development stage in order to minimize the discards by foot breakage, caused by extension of the gonad into the foot during maturity and the spawning period, and to allow higher numbers of spawning individuals in the bed. This table can be also used to shorten the close season in the fishery by adapting the rotational scheme, thus helping to avoid market shortages and reduce the need to import razor clams. Modelling growth is required in many ecological studies and stock assessment applications (Quinn and Deriso, 1999). The age estimation method and the growth model used should be chosen with caution since uncertainties in age estimates lead to uncertainties in management. From the different methods that can be used to estimate the age of razor clams (measurement of growth during culture, length-frequency distribution analyses, examination of shell cross sections by light microscopy, counting surface growth rings, analysis of the internal growth rings by acetate peels, etc.) we selected the culture method to follow the growth of E. magnus early stages (from larvae to juveniles of ~1.5 years old) and the analysis of internal growth rings by the acetate peel method (Richardson et al. 1979) to determine the age and growth of E. magnus population, as it has been described as the most suitable method to ageing razor clams (e.g. Gaspar et al. 1994). We also used the simplest one (counting of surface growth rings by the naked eye and under binocular microscope) to test the possibility of using it as an easy management tool in other exploitation areas. Most marine bivalves exhibit different growth patterns depending on the stage of their life history (Urban, 2002) and consequently different mathematical models must be used. To our knowledge, larval and postlarval growth was never modelled for any razor clam whereas adult growth was always modelled assuming annual growth rings. Besides, usually bivalve growth studies do not evaluate different growth models but they use the von Bertalanffy growth model as default without testing their fitness, which has long been criticized (e.g. Cailliet et al., 2006; Katsanevakis, 2007). Even more, these studies neither consider the growth seasonal variation nor incorporate the influence of environmental variables on growth. The objectives of the growth study in the present PhD dissertation are: to describe with precision the early stages of E. magnus growth cultured on its optimal habitat; to provide sword razor growth parameters; to evaluate the growth variability at mesoscale and the influence of environmental variables on growth and, finally, to test the fitness of different growth models (Logistic, Gompertz, von Bertalanffy and Richards) for each developmental stage. Larval and postlarval culture was conducted in the Estación de Ciencias Mariñas de Toralla (ECIMAT, Universidade de Vigo) while juvenile growth were carried out in a subtidal bed of E. magnus in the Ría de Vigo (SW Galicia) because of the proximity to the ECIMAT. The growth and age study was conducted in three fishing beds of the Ría de Pontevedra in each oceanographic zone of the ria. E. magnus larval and postlarval growth results agreed with those reported by da Costa et al. (2008, 2011). Thus, larvae grew linearly at 10 µm day-1 reaching 250 µm 20 days after fertilisation, when settlement started, while postlarvae grew slowly (length increment, ¿L = 0.04 mm day-1) during the first 10 days after settlement, fast during the next month (¿L = 0.34 µm day-1) and finally declined (¿L = 0.21 µm day-1) in the last month of culture, reaching 750 µm after one month of fertilisation, 10.5 mm at two months old and 25 mm at three months old. Larval growth followed a linear function while postlarval growth was sigmoidal and thereby adequately represented by the Richards model. By contrast, E. magnus juveniles seeded on subtidal bed of the Ría de Vigo showed a higher growth rate than other razor clam culture experiences which used intertidal or suspended culture methods (da Costa and Martínez-Patiño, 2009; da Costa et al., 2011, 2013). Thus, E. magnus was 50 mm at six months old, 67 mm at 1 year old and 82 mm when the experiment finished, 1.5 years after fertilisation. During winter, a growth cessation period was observed probably caused by a combination of downwelling conditions and low seawater temperature values. To model juvenile growth and incorporate the seasonal oscillation we used temperature-dependents growth models previously described on literature (Otterlei et al., 1999; Kielbasa et al., 2010) and developed two upwelling-dependent growth models. Juvenile growth was better described when using growth models that incorporate environmental variables in their functions, especially the von Bertalanffy growth model that included temperature, as the estimated curves reflected better the winter growth cessation period. Optimum temperature for growth was estimated on 15ºC. These results provide important information for aquaculture techniques and confirm that each growth ring is laid down annually, which is crucial to apply indirect ageing methods such as the ones used in the adult growth study. Examination of adult individuals¿ internal shell microgrowth patterns (acetate peel method) proved to be the most suitable method for growth estimate, as observed in other studies (e.g. Gaspar et al. 1994). Similarly to other bivalves, the growth of E. magnus is faster during the first three years of life, declines at about four¿six years old and almost ceases in subsequent years, with the organisms entering into an asymptotic phase around the age of eight-nine years. The mean population length is attained at four-five years. Within this general pattern, there was significant variation among study sites. Sword razor growth was slower in the intermediate zone (L¿ = 140.4, k = 0.40) followed by the innermost zone (L¿ = 151.91, k = 0.40) and the outermost zone (L¿ = 172.7, k = 0.33). As a result, razor clams reached commercial size in 1.7, 2.3 and 2.8 years in the outer, innermost and intermediate zones of the ria, respectively. The growth pattern obtained for the sword razor population in the Ría de Pontevedra is consistent with the one of the sword razor individuals reared in the Ría de Vigo and the growth parameters are similar to those obtained for other Ensis spp. populations by the same ageing technique (e.g. Gaspar et al., 1994; Henderson and Richardson, 1994; Robinson and Richardson, 1998). The influence of environmental factors on the growth of E. magnus in the Ría de Pontevedra is difficult to determine because of the synergistic interactions among environmental factors in highly dynamic systems such as the rias. However, our results indicate that growth is negatively correlated with mud content at all of the sites under study. In addition, we suggest that growth is positively correlated with current speed and negatively correlated with salinity oscillations. The present PhD dissertation provides the first estimation of the size and age at which E. magnus reaches sexual maturity in the Iberian Peninsula (L50 = 79.5 mm, attained between the first and second year of life) and confirmed that the European legal commercial size for E. magnus (>100 mm) is consistent with the reproductive biology of the species. It is also the first work that determines the mesoscale variation in gonadal development of any species of the superfamily Solenoidea and the first that estimates growth parameters for the species in the Iberian Peninsula. The findings of the reproductive cycle study are being applied since 2011 in the rotation scheme of the fishery management plan of the Ría de Pontevedra, leading to a more efficient and sustainable exploitation, and have been also employed to determine the size at first maturity of other sword razor populations. Besides, the proposed management tool enables consideration of gonadal cycle differences between beds and years in the daily fishery management of any razor clam fishery. Finally, the growth rates estimated for each site enable to establish a rotational harvest strategy of subareas of the same exploited bed to allow individuals to grow to a certain size or orient fishing effort to areas of highest growth rate or productivity. The results highlight the importance of carrying out mesoscale studies of the biology in coastal fisheries resources and lay the foundation for further research in other Galician shellfisheries.