The Gulf of Mexico commercial reef fish fishery has experienced significant management changes and disturbance in recent years, including transitioning two major fisheries from a traditional open access system into a limited entry individual fishing quota (IFQ) system in 2007 and 2010. Also in 2010, the Deepwater Horizon oil spill (DWH) released an estimated 4.9 million barrels of oil into the Gulf (~206 million U.S. gallons), and is still the largest U.S. environmental disaster to date. Emergency fishing closures initiated shortly after the oil spill began were successful in keeping tainted seafood from reaching markets. However, effects of DWH closures on fisher decision making, fishery productivity, and distribution of fishing effort all remain poorly understood. Understanding the range and magnitude of fishers’ responses to perturbations — including regulatory change and human-induced environmental disasters — is critical for designing effective management and disaster response policies that can meet biological, ecological, economic, social, and sustainability objectives. This work characterized the spatial and temporal patterns of productivity and fishing effort for the Gulf of Mexico (GoM) commercial reef fish fishery. Patterns of productivity and effort distribution were used to examine the response of fishers to management change and large-scale disturbance, namely the DWH fishing closures. Fisheries-dependent logbook trip reports were used to quantify revenue and catch-per-unit-effort (CPUE) patterns from 2000-2014. Novel to fisheries work in the GoM, complementary vessel monitoring systems (VMS) satellite tracking data were used to quantify high-resolution spatial distribution patterns over time, relative to the DWH fishing closures. A general linear modeling (GLM) approach was also used to examine which variables may have contributed to resilience of fishers after DWH closures. Results suggested that this fishery was largely resilient to the DWH fishing closures in 2010, although exact outcomes varied by region. Overall fleet-level productivity steadily increased over time, but regional patterns were based on major species in catch. Productivity in the western GoM was consistently highest over time, and trips in the west and central GoM were dominated by Red snapper (Lutjanus campechanus) and Vermilion snapper (Rhomboplites aurorubens). Trips in the east were dominated by Red grouper (Epinephelus morio) and Gag grouper (Mycteroperca microlepis). Shifts in spatial distribution to new productive fishing grounds or reduced competition via fewer vessels or trips may explain the increases in productivity observed over the study period. Consolidation in the fleet was apparent, with fewer individual vessels and fewer total trips over time. However, the rate of vessel drop out after DWH (5%) was far below the annual background attrition rate of ~14-20%. Relative productivity patterns inside vs. outside the boundaries of fishing closures did not change over time, and there were even some increases in productivity observed during and after DWH in the eastern GoM. Yet, vessels that dropped out after DWH were concentrated in the north-central and eastern GoM. Distribution of fishing grounds before and after DWH were highly similar, and there were increases in effort along the outer West Florida Shelf. Variability in revenue and CPUE, CPUE magnitude, and magnitude of grouper landings were significant predictors of dropping out of the fishery in the GLMs. Synergies with the Red snapper or Grouper-Tilefish IFQs may have “primed” the fishery for resilience by eliminating inconsistent or marginal fishers before the oil spill, and may further explain some of the spatially varying patterns of productivity and attrition after 2010. Resilience was likely also enhanced by the more than $2 billion in emergency compensation payments made to captains, crew, and vessel owners for lost fishing income and assistance with oil remediation efforts. This work stands to make a significant contribution to our understanding of how the DWH oil spill impacted fisheries and communities in the GoM. The results add to a growing body of literature suggesting that the acute population- and ecosystem-level impacts of the DWH oil spill were not as strong or severe as initially anticipated. This work also stands to make contributions to the broader understanding of how this fishery has performed in the wake of recent management change and major environmental disturbance.

Actuellement, la distribution spatiale des thons tropicaux pêchés par les senneurs français dans l'Océan Indien est représentée à partir des données de captures et d'effort issues des livres de bord. Or ces données ne comportent qu'une position par jour et le nombre de calées réalisées pendant cette journée : l'ensemble des zones prospectées sans action de pêche, synonyme d'absence d'agrégations de thon n'est pas connu. Ces zones seraient pourtant utiles à la cartographie des abondances. Cette flottille étant équipée de systèmes VMS (Vessel Monitoring Systems), les trajectoires données par les VMS sont utilisées pour déduire du comportement des navires, les zones de présence (pêche) et d'absence (non pêche) d'agrégations de thonidés. L'objectif est donc d'estimer l'activité d'un navire à partir de sa trajectoire. L'application d'un modèle markovien permet de distinguer les activités de route, de recherche, et d'arrêt, et de pêche. Pour la majorité des sorties en mer (exceptées celles où un observateur est à bord), l'activité réelle du navire n'est pas connue (variable latente), ce qui justifie l'approche par chaînes de Markov à états cachés. Le modèle a été développé dans un contexte bayésien incluant des priors sur certains des paramètres (vitesses, angles de changement de cap et matrice de transition). Il a été calibré et validé grâce aux données collectées par des observateurs embarqués sur 10% de la flottille. Le taux d'erreur est de 10% pour des positions VMS enregistrées toutes les heures. Une fois estimée, l'activité des senneurs en chaque point de leur trajectoire a permis (1) d'obtenir des indicateurs spatiaux et temporels de l'effort de la flottille de senneurs français, et (2) d'étudier la structure spatiale des zones de présence de bancs de thons (variogrammes d'indicatrices) et d'en déduire, par cokrigeage d'indicatrices, des cartes de présence/présence potentielle/absence de thon capturable, pouvant être interprétées comme indices d'abondance de thon.