Ecological risk assessment is a useful methodology for assisting the management of fisheries from an ecosystem perspective. Atlantic tuna fisheries, managed by the International Commission for the Conservation of Atlantic Tunas (ICCAT), are economically important and interact with several bycatch species. In spite of these interactions, no comprehensive ecological risk assessment has been conducted for bycatch species caught in ICCAT fisheries. In this paper, we followed a two stage approach with the objective of assessing the relative risk of species being negatively impacted by Atlantic tuna fisheries. An analysis of the ICCAT bycatch species list (which includes all species reported to have interacted with different tuna fishing gears operating in the Atlantic) revealed that most of these species are caught in longline fisheries, followed by gillnets and purse seines. According to the IUCN red list, 7 species of the ICCAT bycatch list (3 coastal sharks, 3 sea turtles and one seabird) are categorized as critically endangered. In our study, and based on their life history characteristics, marine mammals and coastal sharks caught in ICCAT fisheries showed the highest intrinsic vulnerability values. A productivity susceptibility analysis for the European Union (EU) tropical tuna purse seine fleet and the United States (US) pelagic longline fleet revealed two groups with high relative risk scores. The first one included pelagic and coastal sharks, characterized by relatively low productivities, and the second one included teleosts, characterized by higher productivities but high susceptibility to purse seine and longline gears. Some alternative approaches to conduct productivity susceptibility analyses in the context of ecological risk assessments are discussed.

Spatial distribution and species diversity of Cephalopoda caught in ARSA bottom trawlsurveys were studied out during two different seasons (autumn and spring) in the Gulf ofCadiz (Spain) from 2000 to 2007. Species composition of cephalopod assemblages wasanalysed, using both clustering analyses and non-metric multidimensional scaling analysis(nMDS). Spatial distribution of the assemblages identified and abundance of the mainspecies were mapped with Surfer 8.0 software, using kriging as the geostatistical griddingmethod. A total of 35 cephalopod species belonging to 6 families were found at depthsbetween 20 and 700 m. In each season, Sepiolidae was the most abundant family in terms ofnumber of species, followed by Octopodidae. The species richness increased up to 100–120 mdepth, where it reached the maximum value. From 120 m, the species richness decreasedprogressively with depth. In spring, the species with the highest occurrence wasEledone moschata (34%) and in autumn it was Alloteuthis media(70%). In both seasons, the most abundant species in terms of weight wasOctopus vulgaris, while Alloteuthis media showed thehighest yields in terms of numbers. Most species showed wide bathymetric ranges,especially in autumn. Three different assemblages were found in both seasons during thetime period analysed (although 2003 and 2006 were not included in the cluster analysis):shelf assemblage (20–160 m), deep shelf/upper slope assemblage (100–350 m) and middleslope assemblage (320–700 m). The specific composition of these three assemblages wassimilar between spring and autumn and an overlap could be observed between them, mainly inthe two continental shelf groups: shelf and deep shelf/upper slope assemblages.Alloteuthis media and Alloteuthis subulata were themost abundant species in the shelf assemblage as well as in the deep shelf/upper slopeassemblage. In the middle slope assemblage, Illex coindetii was the mostabundant species. The assemblages and their spatial distributions could be largely relatedto a combination of physical and biological factors and their interactions.

In French Polynesia, the black-lip pearl oyster Pinctada margaritifera has been farmed to produce pearls since the 1980s, forming the basis of a major industry. The sustainability of this activity relies on spat collection in the lagoons. However, pearl oyster spat can be difficult to identify for the evaluation of stock variations. It is especially hard to distinguish Pinctada spp. larvae at a very early stage of development. In the present study, a whole-mount in situ hybridisation (ISH) technique was developed to allow the discrimination of larvae of closely-related pearl oyster species found in the French Polynesian atolls. Using specific ribosomal 16S-DNA sequence data, we were able to successfully differentiate between Pinctada margaritifera and Pinctada maculata larvae from 5 to 13 days old. This is the first description of a non-destructive method allowing bivalve larvae discrimination between species within this genus. The method allowed us to successfully identify P. margaritifera larvae in natural plankton samples. This result is a key step needed to develop monitoring of P. margaritifera larval distribution in French Polynesian lagoons, a procedure which will increase spat collection efficiency and ensure sustainable development of pearl oyster farming.

The Pacific oyster (Crassostrea gigas) is an important commercial species in the US Pacific Northwest with a history of production initially based on wild-caught seed imported directly from the Miyagi region of northern Japan (1920s–1970s) followed by an extended period of seed collection from a small number of naturalized, self-recruiting populations in US estuaries (early 1970s-present) and more recently through large-scale hatchery production of seed oysters (mid 1970s-present). I studied the genetic level consequences of each of these major transformations of the oyster industry by examining the patterns of private alleles (bands unique to one sample), the number of polymorphic loci, expected heterozygosity, genetic distance between populations and genetic divergence between individuals within and among three native Pacific oyster populations in Japan (Hiroshima, Miyagi, Midori River), five naturalized populations in North America (Pipestem Inlet BC, Nootka Inlet BC, Dabob Bay WA, Willapa Bay, WA, and Tillamook Bay, OR), two in New Zealand (Chance Bay and Kaipara Harbor), and seven domesticated and selectively bred cohorts from an ongoing genetic improvement program on the West Coast of the United States using amplified fragment length polymorphism (AFLP) markers. All but one of the naturalized populations in both the USA and New Zealand are genetically more similar to native populations from the Ariake Sea (Midori River) than to the Miyagi region of their origin, but all domesticated stocks more closely resemble the wild Miyagi population. According to local oyster producers, the one exceptional naturalized population (Tillamook) is a very recent colonization derived from farmed oysters. Such consistency is unexpected under random genetic drift, and I speculate that both natural and artificial selection may have altered AFLP allele frequencies in this species in the course of naturalization and domestication.

Successful commercial aquaculture production of spiny lobster is, for the most part, still constrained by an array challenges including the development of nutritionally complete and cost-effective feeds, control of disease vectors, and the design of larval mass culture tanks. Culture tank designs for larval production are a critical step to facilitating the most favorable combinations of water flow, food contact, and larval survivorship over the course of development. The evolution of new plankton-kreisels that are used by aquariums to culture and exhibit gelatinous zooplankton (e.g., jellyfish) provide a unique opportunity for testing the feasibility for spiny lobster larval culture, particularly with tropical species such as Caribbean spiny lobster (Panulirus argus) whose larval duration, although complex, is comparatively shorter than other spiny lobsters. Here, we report on the feasibility of culturing P. argus larvae (i.e., phyllosoma) from hatch to Stage VI using large (180 L) modified acrylic plankton-kreisels. We compared overall growth and survival of phyllosoma at starting densities of 5000 (~27.8 larvae L-1) and 2500 individuals (~13.8 larvae L-1) and found no significant difference with respect to survival or mortality through to 65 days (χ2 = 1.595; df = 1; p = 0.2066) resulting in mean survival rates of 60.7% (s.e. = ±3.7) and 54.5% (s.e. = ±3.2), respectively. Comparable growth was also achieved between both densities to Stages V and VI (mean body lengths of 7.5 and 10.2 mm, respectively) at 25.1 ± 0.41 °C and pH = 8.1. Phyllosoma utilized the entire tank volume and displayed minimal entanglement. The application of such tank designs for larval spiny lobster culture not only contributes to future designs for aquaculture production, but also provides a useful platform for conducting behavioral studies for this complex larval phase.

Among the strategies that can be used to improve fish welfare in a rearing environment,domestication and/or selective breeding was proposed to minimize fish responsiveness tohusbandry practices. To verify this hypothesis on a recently domesticated species, the seabass Dicentrarchus labrax, two experiments were realized, each using twopopulations differing according to their level of domestication or selection. For thefirst experiment, we used one population produced from wild parents (Wild;initial body mass: 106  ±  3 g), and one population from parents selected forgrowth for one generation (Selected 1; initial body mass: 129  ±  4 g).For the second experiment, we used one population produced from parents domesticated fortwo generations (Domesticated; initial body mass: 72  ±  3 g), and oneproduced from parents selected for growth for two generations (Selected 2;initial body mass: 89  ±  4 g). The first experiment was carried out over 112days with 240 fish (60 fish per tank, 120 fish per population), and the second one over 84days with 200 fish (50 fish per tank, 100 fish per population). Two variables,self-feeding behavior and growth performance, were measured over the time of theexperiments. After a control period, the fish were submitted twice, at three-weekintervals, to an acute stress treatment consisting of draining the tank and leaving thefish out of water for one minute. Both self-feeding behavior and growth performance werealtered by the acute stress treatment. During the first post-stress period, theDomesticated and Selected (1 and2) groups showed more pronounced post-stress exposure responses thanthe Wild fish: they modified their feeding rhythm, their feed intake, andtheir growth rate. During the second post-stress period, feeding rhythm was still affected(being more diurnal with a well defined peak), but the feed intake and growth rate resultsshowed that the Domesticated and Wild groups seemed lessaffected than the Selected (1 and 2)populations, which continued to express a high post-stress response.

According to these results, it can be concluded that: (1) an application of two acutestress treatments, at three-week intervals, modified fish feeding behavior and growthperformance; (2) the domestication process seemed to improve fish adaptation abilities tothis kind of stress; and (3) the process of selection for growth led to a final, bettergrowth, but did not seem to improve fish acute stress tolerance.

Water quality assessment is a key factor in the environmental management of freshwater networks, especially those including fish farms, which need cost-effective operational tools to monitor and control their waste output. In France, current legislation specifies limits in concentrations of dissolved compounds and suspended solids at fish-farm outlets. Despite the development of mass-balance modelling tools, chemical analysis of water (hydrological method) remains the most widely used method. To understand better the environmental impact of trout farms on aquatic ecosystems and to compare waste assessment methods, we monitored 20 commercial flow-through trout farms for 24 h, and we compared data obtained with the two methods (hydrological method and mass balance modelling) by linear regression. For total nitrogen and total phosphorus, the correlation between the two methods was high; thus, considering the uncertainty of both methods, this study was not able to determine which was more accurate. The high correlation between observed ammonia concentrations and predicted total nitrogen emissions provides a coefficient for estimating ammonia emissions at the farm level. The same approach is proposed for the evaluation of phosphate emissions. In conclusion, this study confirms the utility of simulation modelling for assessing nutrient release from fish farms.

Water Framework Directive requires reliable and effective monitoring tools, and hydroacoustics has a potential to be one of them. The effect of pulse duration on in situ acoustical estimates of fish density and their size distribution was investigated. Measurements were performed in the oligo-mesotrophic Lake Hancza (Poland) using a SIMRAD EK60 split-beam echo-sounder at 70 kHz frequency. During the survey, two similar transducers pinged alternatively through the multiplexer using 4 different pulse lengths, from short to long ones. The results show that the volume backscattering coefficient (Sv) values, equivalent of the fish biomass, were not influenced by the pulse length. However, the number of the detected fish, the mean target strength (TS), and consequently the fish density, differed significantly for the long pulse duration data. This was especially noticeable in the layer above the thermocline with dense fish populations. In this upper layer, for the long pulse the Sawada index frequently exceeded value of 0.1 leading to overestimation of the mean TS and underestimation of the fish density.

Dramatic declines in American eel (Anguilla rostrata) recruitment haveresulted in strong conservation measures being implemented. Recovery actions in Ontariohave included stocking of glass eels. Given the financial costs and imperative toundertake effective recovery actions, post-stocking monitoring is essential. In thisstudy, point- and transect electrofishing sampling were compared in terms of eeldetection, catch rates, size-selectivity, and power to detect changes in abundance.Transect sampling was more likely to detect eels and captured over twice the number ofeels than point-sampling. Differences in catch rates and statistical power were dependenton the catch-per-unit-effort measure (i.e. sampling unit vs. time). Results support atransect-based sampling program for stocked eels in Lake Ontario tributaries.

The pout, Trisopterus luscus, is one of the most important gadoid fishcaptured by northern Portuguese traditional fisheries. In spite of a substantial decreasein fish catches, little data are available either on the population structure or on themanagement of the species. In this study, chemical analysis with inductively coupledplasma mass spectrometry of whole otoliths of 90 pouts (age group 2, total length:17.9–25.0 cm) provided location-specific elemental signatures. Sampling took place inshallow waters along the coastline in three fishing grounds off the Portuguese north coast(Viana do Castelo, Matosinhos and Aveiro) between February and March 2010. Otolithfingerprint analysis detected the presence of several informative trace elements. Molarconcentrations for each site were analysed through uni- and multivariate statisticaltests. Strontium, barium, magnesium and lithium (Sr, Ba, Mg and Li) differed significantlyamong locations, while no significant differences were found for manganese (Mn) and nickel(Ni). Canonical analysis allowed us to discriminate the tested individuals with respect totheir sampling origin with a mean classification accuracy of 69%. The observedsite-specific elemental differences in pout otoliths suggest a high level of site-fidelityin relation to their growing/feeding areas. Pouts from these locations can be regarded asa single, although not necessary homogenous, stock. Furthermore, this study also suggeststhat the populations of juvenile fish mix partially and, therefore, cannot be assumed tobe separate units for fisheries management purposes.

Juveniles of the mussel Mytilus chilensis were exposed to a dietcontaining paralytic shellfish poisoning (PSP) toxins produced by the dinoflagellateAlexandrium catenella (strain ACC02). The feeding behaviour and thedynamics of intoxication and detoxification were evaluated over an intoxication period ofnine days, followed by a detoxification period of eight days. A significant reduction inthe feeding activity was measured during the first days of exposure to the PSP toxins(days 0 and 2), followed by a period of recovery observed on days 5 and 9, when theclearance rate of the contaminated mussels significantly increased. During thedetoxification period, the contaminated bivalves showed a total recovery of clearancerate, and no significant differences were observed between contaminated and controlgroups. The intoxication dynamic was characterised by a rapid and sustained increase inthe amount of PSP toxins accumulated in the tissues of the bivalves. Due to this increase,the level of PSP toxins critical for human consumption was reached on the first day, andat the end of the intoxication period, the mussels had accumulated 1601μg STX eq. 100 g-1 tissue. During the detoxification period, adecrease of PSP toxins was observed, down to 319 μg STX eq. 100g-1 tissue. The lower clearance rates observed over the first days ofexposure would produce a decrease in the energy intake and could affect the rate of growthof juveniles. Despite this initial effect, the rapid intoxication capacity of M.chilensis corroborates that this species is a good indicator for the earlydetection of harmful algal blooms.