Escaping the tyranny of the grid: a more realistic way of defining fishing opportunities

Abstract

A large part of fishing behavior is choosing where to fish. Trawl skippers usually choose between known fishing opportunities, which are observed as groups of trawls that are conducted in the same portion of a fishing ground, or go exploratory fishing. We outline a simple clustering method based on Euclidean distances between trawls that offers a more realistic way of defining fishing opportunities than grid cells or statistical areas. The resulting cluster tree of trawls is divided into individual groups of trawls (fishing opportunities) using a recommended cut point. Our method correctly classified simulated trawls into fishing opportunities. Fishing opportunities were obtained for vessels in the British Columbia groundfish trawl fishery; each vessel usually fished at a wide variety (mean 26, standard deviation 16, range 2–69) of fishing opportunities. Within each fishing opportunity, trawls generally caught similar species. In the Argentina scallop fishery, our method was able to divide exploratory from regular fishing trawls, with obvious applications for catch-per-unit-effort calculations. Our method could also be used to detect positional errors in data from these fisheries. Fishing opportunities could provide indications of how fishermen might react to marine protected areas and to the imposition of quotas on multispecies fisheries.