Fisheries Research 129–130 (2012) 110–118 Contents lists available at SciVerse ScienceDirect Fisheries Research jou rn al h om epa ge: www.elsevier.com/locate/fishres Effect of habitat patchiness on the catch rates of a Mediterranean coastal bottom long-line fishery Ben Stobart a,b,∗ , Diego Alvarez-Barastegui a , Raquel Go˜ ni a a Centro Oceanográfico de Baleares – IEO – Muelle de Poniente s/n, 07015 Palma de Mallorca, Spain b South Australian Research and Development Institute, PO Box 120, Henley Beach, SA 2022, Australia a r t i c l e i n f o Article history: Received 20 February 2012 Received in revised form 19 June 2012 Accepted 20 June 2012 Keywords: Catch per unit effort Artisanal fishery Long-line Habitat patchiness a b s t r a c t Use of catch per unit effort (CPUE) as an index of abundance relies on a range of assumptions that are often not met. Variation in habitat is likely to affect catch rates of species and consequently should be taken into account when estimating abundance from CPUE data, yet this is not generally possible due to a lack of spatially detailed habitat information. Mediterranean artisanal fisheries often operate in coastal areas where habitat is patchy, and consequently catches may not be related to effort if a portion of the gear has been set over unsuitable habitats for the target species. We investigate the effect of habitat patchiness on CPUE using bottom long-line by including detailed habitat and bathymetric information in the calculation of CPUE for eight commonly caught species. Comparison of CPUE estimated with and without taking into account habitat information showed exclusion of hooks in “unfavourable” habitats generally increases CPUE estimates for species of restricted mobility and strict habitat associations, while estimates for more mobile species, or those with lower habitat specificity, differ less. We also compare two methods for calculating habitat fished by each hook, one that used the habitat at the hook location, and another that included habitats within a 200 m radius. The former method had the highest effect on CPUE. Results demonstrate habitat information can be usefully incorporated into CPUE estimates, while the effect on CPUE will vary depending on how specific species–habitat preferences are. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Catch per unit effort (CPUE) is arguably the most common mea- sure of abundance used to assess populations of commercially exploited marine species (Hilborn and Walters, 1992; Maunder et al., 2006). However, the use of CPUE as an index of abundance relies on the assumption that catch is proportional to fishing effort, density of fish and a catchability coefficient, which is often treated as a constant (Hilborn and Walters, 1992). Although this assump- tion does not necessarily hold, as many factors may affect catch rates and thus bias their relationship with abundance (Beverton and Holt, 1957; Harley et al., 2001; Kleiber and Maundner, 2008), CPUE remains a key fishery statistic because it is easily calculated from fishery records which are generally readily available, making it con- siderably more cost-effective than other methods for determining measures of abundance (Maunder and Punt, 2004). Habitat preference is an important aspect of species behaviour that can affect CPUE, as it dictates where a particular species is likely ∗ Corresponding author. Present address: South Australian Research and Devel- opment Institute (SARDI), PO Box 1511, Port Lincoln, SA 5606, Australia. Tel.: +61 8 86832567; fax: +61 8 86832520. E-mail address: ben.stobart@sa.gov.au (B. Stobart). to be caught (e.g. demersal species – Demestre et al., 2000; Wieland et al., 2009; Lauria et al., 2011; pelagic species – Babcock, 2006; Bigelow and Maunder, 2007), and if specific enough can potentially be utilised as a proxy for mapping seafloor habitat distributions (Auster et al., 2001). The influence of fine scale habitat variation on CPUE estimates is particularly evident in some forms of arti- sanal fishing which take place on patchy littoral habitats where the characteristics of patches of substrate are likely to influence species presence and abundance (Forcada et al., 2009). Here fisher knowledge of habitat and their fishing strategies, which can vary considerably from one fisherman to another, may heavily influence catch rates. For example, in these fisheries fishers may knowingly choose to deploy one long fishing set that crosses ‘barren’ patches for the target species rather than deploying several, more accurate, shorter sets. In this case fisher decisions will greatly affect the pro- portion of the set laid on suitable habitat, which is likely to lead to significant differences in CPUE among fishers and may in some circumstances break the relationship between catch and effort. The objective of this study is to assess the effect of habitat patch- iness on the estimates of CPUE for species with different movement patterns and strength of habitat preference. To achieve this we related bottom long-line fishing effort to habitat type and depth using geo-referenced habitat maps, and investigated the effect of habitat types on CPUE estimates by comparing a traditional CPUE 0165-7836/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.fishres.2012.06.014