Abstract
From an experiment with ordinary dart tags, we have found evidence of the effect of fish-aggregating devices (FADs) and of islands on the movements of skipjack tuna (Katsuwonus pelamis) around the Solomon Islands. By fitting a fish movement model to the tag data, we were able to estimate mortality and movement parameters (including diffusivity), parameters of a function that models FAD attraction, and a separate parameter of island attraction. Diffusivity was high enough to effectively distribute fish throughout the island archipelago (approximately 150 000 km2) within a few months. Estimates of FAD parameters indicate that the presence of up to four or five FADs in an area approximately 50 × 50 km can reduce the propensity for skipjack to leave that area by approximately 50%, but that deploying additional FADs in such an area does not significantly increase their effectiveness in holding skipjack. Estimates of the island attraction parameter imply that the propensity of skipjack for movement away from the archipelago is less than half the propensity for movement within it.