Research article Assessment of groundwater vulnerability to nitrates from agricultural sources using a GIS-compatible logic multicriteria model Boris Rebolledo a, * , Antonia Gil a , Xavier Flotats b , Jos  e  Angel S  anchez c a Center of Research for Energy Resources and Consumption (CIRCE), C/Mariano Esquillor G omez 15, 50018, Zaragoza, Spain b GIRO Joint Research Unit IRTA-UPC, Department of Agrifood Engineering and Biotechnology, Universitat Politecnica de Catalunya, BarcelonaTECH, Parc Mediterrani de la Tecnología, Building D4, E-08860, Castelldefels, Barcelona, Spain c Departamento de Ciencias de la Tierra,  Area de Geodin amica, Edificio de Geol ogicas Pedro Cerbuna, 12, 50009, Zaragoza, Spain article info Article history: Received 5 October 2015 Received in revised form 26 January 2016 Accepted 30 January 2016 Available online xxx Keywords: Groundwater vulnerability Risk mapping Nitrate pollution Logic Scoring of Preferences (LSP) Aragon (Spain) abstract In the present study an overlay method to assess groundwater vulnerability is proposed. This new method based on multicriteria decision analysis (MCDA) was developed and validated using an appro- priate case study in Aragon area (NE Spain). The Vulnerability Index to Nitrates from Agricultural Sources (VINAS) incorporates a novel Logic Scoring of Preferences (LSP) approach, and it has been developed using public geographic information from the European Union. VINAS-LSP identifies areas with five categories of vulnerability, taking into account the hydrogeological and environmental characteristics of the territory as a whole. The resulting LSP map is a regional screening tool that can provide guidance on the potential risk of nitrate pollution, as well as highlight areas where specific research and farming planning policies are required. © 2016 Elsevier Ltd. All rights reserved. 1. Introduction The increasing international concern about nutrient overload into the environment has resulted in the introduction of strict regulations for the protection of water resources. Within this context, groundwater contamination by nitrates (NO 3 - ) from agri- cultural sources is one of the most widespread threats worldwide (Addiscott and Benjamin, 2004; Karr et al., 2001; Weyer et al., 2001). Due to this threat the EU drew up the Nitrate Directive 91/ 676/EC concerning the protection of waters against nitrate from agricultural sources. Since the EU Nitrate Directive was adopted, important differ- ences have been observed in the methods and approaches used to identify Nitrate Vulnerable Zones (NVZs) (European Commission, 2013). Although criteria for identifying the NVZs were established in the Nitrate Directive, the specific procedure for the delimitation of these vulnerable areas is still unclear. Furthermore, recent research has shown that an inadequate designation of NVZs can generate unsatisfactory results in the contamination reduction of affected water bodies (Arauzo and Martínez-Bastida, 2015; Arauzo and Valladolid, 2013; Worrall et al., 2009). In Spain, the regional administrations are responsible for iden- tifying NVZs from agricultural practices. In general, analysis of water quality data from networks of monitoring stations has been used to designate vulnerable zones, and administrative boundaries and groundwater bodies have been used to delineate the shape of these areas. Furthermore, the emphasis on the evidence of envi- ronmental damage, rather than on a proactive planning, can hinder successful conservation of water resources. Therefore, it is neces- sary to develop a more rational, rigorous and systematic approach. Until now, several methods for groundwater vulnerability and risk mapping have been proposed. They range from complex deterministic models of the physical, biological and chemical ni- trate leaching processes occurring in vadose zone and saturated zone (De Paz and Ramos, 2004; Lasserre et al., 1999; Ledoux et al., 2007; Srinivasan and Arnold, 1994), to methods that are based on overlay and index techniques to obtain a final vulnerability score. Index methods are based on combining rated maps of various physiographic factors (e.g., depth to water table, aquifer type, soil organic carbon content) of the region by assigning a subjective numerical score to each factor. Models of index methods include DRASTIC (Aller et al., 1987); GOD (Foster, 1987); AVI (Van Stempvoort et al., 1993); EPIK (Doerfliger et al., 1999); SINTACS * Corresponding author. E-mail address: brebolle@unizar.es (B. Rebolledo). Contents lists available at ScienceDirect Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman http://dx.doi.org/10.1016/j.jenvman.2016.01.041 0301-4797/© 2016 Elsevier Ltd. All rights reserved. Journal of Environmental Management 171 (2016) 70e80