Faculty of Fisheries, Kagoshima University, Kagoshima 890-0056, Japan
Two small experimental fish aggregation devices (FAD) of the same design were deployed separately from each other on a sandy seabed. One FAD was tangled with a bottom gillnet and the other left intact. Fishes enmeshed on the tangled gillnet and those aggregated around two FADs were monitored for the period of 1,149 days since entanglement of the gillnet. The observable days for enmeshed fish of each species were investigated independently. A model was created according to the binomial distribution based on the probability of observing one enmeshed fish. With this model, the ghost-fishing mortality was estimated from the number of observed fish enmeshed on the tangled gillnet. Mainly crimson sea bream, Evynnis japonica, and threadsail filefish, Stephanolepis cirrhifer, were enmeshed. There was no clear declining trend in ghost fishing during the experimental period. The ghost-fishing mortality per year by the tangled gillnet was estimated as 191 fish. The number of all fish aggregated around the gillnet-tangled FAD was greater than that around the control FAD, however, the numbers of aggregated individuals of the ghost-fished species were not significantly different between the two FADs. It was hypothesized that the lack of a difference was due to ghost-fishing of those fishes aggregated by the gillnet-tangled FAD.