Search

Abstract: We assessed the linkages between environmental variables and the spatio-temporal distribution patterns of larval abundance of coastal fish assemblages within and outside the Medes Islands marine protected area (MPA; NW Mediterranean) to explore possible fisheries effects such as larvae export of the latter. We analyzed small-scale fish larvae distribution of 28 representative taxa from the rocky fish assemblage and combined the former in different groups with the help of cluster analysis. Further we assessed the influence of selected abiotic variables on group and species distributions using Generalized Additive Models (GAM’s). We found a high level of variability in the response of larvae groups and species to environmental variables over the two studied periods (spring and summer); depth and habitat of the adults being the most important factors. In addition, distance to the MPA was found as an important variable in defining the location of larvae of strictly coastal species. Our results provided evidence of larval export of three commercial species affected by the MPA (Epinephelus marginatus, Pagellus erythrinus and Scorpaena porcus). Thus our study contributes to the few empirical assessments of larvae export from MPAs and adds to the understanding of the functioning of MPAs as fisheries management tools. Nevertheless, in the future integrated assessments of fisheries effects of MPAs are required that measure the effect of larvae and biomass export on increased fishing yield while taking into account fishing effort dynamics. [Copyright &y& Elsevier]

Spatio-temporal variability in assemblages of pelagic fish in Chesapeake Bay from 1995 to 2000 was documented based on midwater trawl surveys conducted three times annually (April, July, and October). Numerically dominant pelagic and benthopelagic fishes, 30–256 mm TL, were included in the analysis. Species assemblages, diversity indices, biomass distributions, and a correspondence analysis revealed that physical forcing, primarily driven by freshwater input, shaped the structure of fish communities, both annually and regionally. But, seasonal succession occurred, with adults dominating in the spring and recruited juveniles dominating in the fall. The dominant species were bay anchovy (Anchoa mitchilli), Atlantic croaker (Micropogonias undulatus), white perch (Morone americana), spot (Leiostomus xanthurus), weakfish (Cynoscion regalis), and Atlantic menhaden (Brevoortia tyrannus). Young-of-the-year (YOY) and age-1+ fish were concentrated in different regions and different seasons. A correspondence analysis showed that salinity was the most important environmental factor explaining annual differences in the upbay–downbay gradients in fish community structure. In 1997–2000, following a surge of freshwater input and nutrient loadings in 1996, October baywide fish biomass increased by 120% and mean baywide dissolved oxygen (DO) concentration below the pycnocline in summer decreased by 30%. These interannual variabilities in baywide fish biomass and mean DO suggest a relationship between bay plankton productivity and fish production. Species composition shifted markedly in 1996 when YOY and age-1+ anadromous fishes became dominant, but returned progressively in subsequent years to the usual structure in which bay anchovy and other polyhaline species were dominant. [Copyright &y& Elsevier]

The distribution of epibenthic invertebrate biomass in relation to environmental factors was examined in the St. Lawrence system. Biomass estimates for epibenthos sampled yearly for 9 years on 102 suspended collectors (navigation buoys), were related to environmental data from the literature (surface water temperature, water salinity, water transparency, current velocity, chlorophyll a and primary production) using a weighted multiple linear regression analysis. Regression models were generated for total biomass and the biomass of the single dominant sessile species: Mytilus edulis, Semibalanus balanoides, Balanus crenatus, Obelia longissima and Hiatella arctica. Water temperature and water transparency, as well as some biogeographic groups of buoys represented by dummy variables, collectively explained 90.6% of the variance in total biomass. Water temperature, water transparency, biogeographic groups and, to a lesser degree, primary production, were the variables having a significant influence on the biomass of individual species. The lognormal weighted multiple regression model explained up to 84.5% of the variance in M. edulis biomass data and 67.9, 70.0, 71.6 and 38.9%, respectively, of the variance in S. balanoides, O. longissima, B. crenatus and H. arctica biomass data. The need to consider simultaneous biological and environmental sampling at the relevant temporal and spatial scales to model large marine coastal ecosystems is discussed. [Copyright &y& Elsevier]