Date Awarded

Winter 2016

Document Type


Degree Name

Master of Science (M.Sc.)


Virginia Institute of Marine Science


David M Kaplan

Committee Member

Richard W Brill

Committee Member

John E Graves

Committee Member

Michael A Unger


Thousands of floating objects, known as drifting fish aggregating devices (dFADs), are released every year by commercial tropical tuna purse seine vessels in the three equatorial oceans to aggregate tuna and increase catch. The escalation in the number of dFADs deployed over the last three decades has caused changes in fishing effort that are poorly reflected in traditional indices of purse seine effort and catch per unit of effort (CPUE). In addition, concerns have been raised regarding the impacts of such high numbers of dFADs being deployed on both catch and bycatch species. I studied two aspects of dFAD deployments in order to clarify how dFADs are used by purse seiners and how they affect the magnitude of bycatch. My specific goals were to determine how often purse seine vessels fish on the dFADs they deploy and how regional dFAD density affects the magnitude of bycatch and catch. I analyzed commercial data, independent observer data, satellite buoy trajectories, and estimated floating object densities from the French tropical tuna purse seine fisheries in the Atlantic and Indian Oceans to examine these relationships. My results indicate that only 2.7-20.6% of dFAD fishing sets were on the dFADs that French purse seiners deployed over the period 2007-2013. Although this percentage increased over time, such a low percentage suggests that French vessels do not primarily increase CPUE by using the dFADs they deploy for directed fishing. If French purse seiners are not mainly using their own dFADs in this manner, then using a metric of nominal effort based on individual vessel activity is unlikely to produce reliable CPUE estimates for the fishery. While information about how often purse seiners fish on their own dFADs could be incorporated into indices of fishing effort, it may be just as important, if not more so, to quantify the collective component of dFAD releases for overall fishing effort. The results for the second objective showed that average dFAD density had a relatively weak relationship with total bycatch biomass. There was no statistically significant effect on mean total bycatch biomass, but dFAD density was related to heteroscedasticity in bycatch biomass. In particular, there was a significant negative relationship between dFAD density and the upper quantiles of the bycatch biomass distribution, suggesting that dFAD density determines the upper bound for bycatch biomass caught per set. These relationships were not constant across species, however, as the five most prevalent bycatch species (i.e., those caught in the greatest number of sets) showed a mix of positive, negative, or no relationships to dFAD density. The total biomass of catch increased significantly as average dFAD density increased both at the mean and throughout the majority of the distribution, although this effect only explained a small fraction of the total variance in catch biomass. at this time, the low amount of variance explained and the inconsistent impact of average dFAD density on bycatch and target species indicate that it would be ineffective to mitigate bycatch levels in these fisheries by attempting to control regional dFAD densities.




© The Author