Fish farming and its appeal to common bottlenose dolphins: modelling habitat use in a Mediterranean embayment SILVIA BONIZZONI a,b,c,d, * , NATHAN B. FUREY d,e , ENRICO PIROTTA c,f , VASILIS D. VALAVANIS g , BERND WÜRSIG d and GIOVANNI BEARZI a,b,d a Dolphin Biology & Conservation, Piegaro PG, Italy b OceanCare, CH-8820Wädenswil, Switzerland c Tethys Research Institute, Milan, Italy d Texas A&M University at Galveston, Galveston, TX, USA e Pacific Salmon Ecology and Conservation Laboratory, Department of Forest and Conservation Sciences, University of British Columbia, Vancouver, BC, Canada f Institute of Biological & Environmental Sciences, University of Aberdeen, Aberdeen, U.K. g Marine Geographic Information Systems Lab, Hellenic Center for Marine Research, Iraklion Crete, Greece ABSTRACT 1. Common bottlenose dolphins Tursiops truncatus interact with fish farms in the Mediterranean Sea. These interactions were investigated in a Greek bay by incorporating multiple geographic, bathymetric, oceanographic, and anthropogenic variables. 2. Generalized additive models (GAMs) and generalized estimation equations (GEEs) were used to describe dolphin presence. Visual surveys were conducted over 2909 km under favourable viewing conditions that included 54 dolphin group follows for 457 km. Sea surface temperature (SST) and chlorophyll-a (Chl-a) data were obtained from remote sensing imagery, and distances to sources of human influences including fish farms, a ferro- nickel plant, and a slag disposal area were calculated within a geographic information system (GIS). 3. Bottlenose dolphins were encountered mainly in the south-eastern portion of the study area, and occurrence was not clearly related to SST and Chl-a, nor the ferro-nickel plant or nearby slag disposal area. 4. Dolphin occurrence generally increased within 20 km of fish farms, with four farms and dolphins displaying a positive relationship, seven no clear relationship, and two a negative one. 5. While it is likely that uneaten food and other detritus attract dolphin prey, individual farms (or clusters of farms) clearly had a different appeal. The proximity of the ferro-nickel plant and slag disposal area to ‘attractive’ fish farms could compromise dolphin health, but physiological data are unavailable. 6. The modelling of multiple variables allowed for a description of dolphin habitat use and attraction to some fish farms. More such data analysed in similar manner would be instructive for other areas where marine mammals and fish farms co-occur. Copyright # 2013 John Wiley & Sons, Ltd. Received 9 January 2013; Revised 10 August 2013; Accepted 18 August 2013 KEY WORDS: mammals; coastal; distribution; habitat management; aquaculture; modelling *Correspondence to: S. Bonizzoni. Dolphin Biology & Conservation, via Cupa 40, 06066 Piegaro PG, Italy. E-mail: silvia.bonizzoni@gmail.com Copyright # 2013 John Wiley & Sons, Ltd. AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 696–711 (2014) Published online 30 September 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/aqc.2401