Changes in the overexploited demersal fish assemblages in the Northwest Atlantic: the Southern Grand Banks and the Flemish Cap

Abstract

An ecosystem approach to fisheries (EAF) is widely recognized as desirable for fisheries management. To progress towards the implementation of an EAF, a first step is to identify which species co-occur by quantifying assemblage structure in these habitats. A second step is the selection and development of ecological indicators to evaluate and to quantify the ecological state of exploited marine ecosystems. I examined groundfish assemblage structure in relation to depth, in two exploited marine ecosytems, the Southern Grand Banks of Newfoundland and the Flemish Cap, using multivariate analysis. Then I evaluated the status of the two ecosystem using a suit of ecological indicators. Finally, I tested different trajectories at two different depths to examine the population structure of the three species of redfish using Multivariate Autoregressive State Space Models (MARSS). I also included covariates to analyze their effect on population size. I used data obtained from two multispecies bottom trawl surveys performed by the Spanish Administration in the Southern Grand Banks (3NO survey) from 2002 to 2014, and the European Union (EU) in the Flemish Cap (3M survey) from 1991 to 2015, in the Northwest Atlantic Fisheries Organization Regulatory Area (NAFO RA). A total of 1526 hauls were made between 2002 and 2014 in the 3NO survey, and 3633 hauls between 1991 and 2015 in the 3M survey. Total catch of the surveys for those periods was 1.482.116 kg in 3NO and 1.268.660 in 3M. Three major main fish assemblages were indentified (38–300, 301–600, and 601–1460 m depth in 3NO; 129–250, 251–600, and 601–1460 m in 3M). Despite dramatic changes in biomass and abundance of the species in both areas, the overall pattern of demersal fish assemblages remained similar over time. Indices responded to different exploitation patterns, management, and environmental regimes in each assemblage. MARSS analysis identified separate trajectories for each species in each depth but one overall population growth rate with commercial catch as a covariate. Thus, while biomass trends for each species in each depth showed some temporal independence, all showed the same long-term changes. Thus there was data support to confirm that species could be treated together as a single management unit and to suggest that fishing is an important driver. This study revealed important spatial structure in the fish assemblages in the two fishing grounds and highlights the importance of calculating indicators in each assemblage in order to compare Summary areas and periods under different fishing pressure. This thesis represents a contribution towards the management through a multispecies approach in NAFO