Testing CPUE-derived spatial occupancy as an indicator for stock abundance: application to deep-sea stocks

¹ Ifremer, rue de l’île d’Yeu, BP 21105, 44311 Nantes Cedex 3, France
² CEFAS Lowestoft Laboratory, Pakefield Road, Lowestoft, NR33 0HT, UK
³ Division of Biology, Imperial College London, Silwood Park, Ascot, SL5 7PY, UK

^a Corresponding author: verena.trenkel@ifremer.fr

Received: 12 February 2013
Accepted: 22 July 2013

Abstract

The status of an exploited population is ideally determined by monitoring changes in abundance and distributional range and pattern over time. Area of occupancy is a measure of the current distribution. Unfortunately, for many populations, scientific abundance and distribution information is not readily available. To evaluate the reliability of commercial fishing data for deriving occupancy indicators that could serve as proxies for stock abundance, we investigated four questions: 1) Occupancy changes with stock biomass, but is this change strong enough to make occupancy a sensitive indicator of population biomass? 2) Fishing boats follow fish, but when does such activity alter the positive macroecological relationship between occupancy and abundance? 3) When does the activity of pursuing fish adversely affect occupancy estimates derived from catch and effort data? 4) How does uncertainty in fishing effort data affect occupancy estimates? Spatial simulations mimicking the dynamics of four deep-water fish species showed that biomass-occupancy relationships can be weak. Fishers following fish can modify the spatial distribution of target species, even reversing the sign of the biomass-occupancy relationship in certain cases, and can affect the reliability of occupancy indicators, which can also be impaired by error in effort data. Using commercial catch and effort data and abundance indices for deep-sea fish populations to the west of the British Isles it was found that only for roundnose grenadier might occupancy provide insights into biomass changes. In conclusion, care should be taken when using occupancy for evaluating range changes in cases where fishing might have modified spatial distributions, uncertain commercial data are used or when the abundance-occupancy relationship is too flat.