A Poisson and negative binomial regression model of sea turtle interactions in Hawaii's longline fishery

Abstract

Sea turtle interactions with the longline fishery in Hawaii have become a serious concern in recent years. Various measures, including a swordfish harvest ban, have been adopted to protect sea turtles. This study explores the factors and degree of sea turtle interactions with Hawaii's pelagic longline fishery, i.e. the production of undesirable outputs in the pursuit of an economic activity, in an analytical framework of rare events using the count data models for the period 1994–2003. The analysis was based on the type of trip, such as those targeting tuna or swordfish. The fishing technologies associated with the choice of trip or target species, season, and turtle population explained turtle interactions with the longline fishery. The impact of a fishing vessel's previous history of turtle interactions had a negligible impact on subsequent interactions. In the absence of new longline fishing technologies to dramatically avert sea turtle interactions, there are about 6% and 55% chance that at least one turtle per trip may be encountered in tuna- and swordfish-targeted fishing trips, respectively. This study confirms that more turtle interactions are associated with the swordfish-targeted trips. Several of the factors that contributed substantially to turtle interactions can be feasibly regulated. Furthermore, the conservation and management of sea turtles require increased policy dialogues and cooperation among the coastal nations vis-à-vis the adaptation to “turtle-friendly” fishing technologies.