Introduction In this article a new computer tool is applied: Predictive Fuzzy Rules Generator (PreFuRGe) (Aroba 2003) that allows qualitative interpretation of data recorded in a database relative to the chemistry of water. Specifically, the authors intend to contrast the model proposed by Grande et al. (1996), which applies classical statistic techniques, such as factorial analysis, to a data mass resulting from the sampling and analysis of a network of 54 wells distributed across the system’s recharge zone. By studying this model, the existence of a close depen- dency relationship between nitrate ion concentrations in the saturated zone of the study area and the presence of strawberry crops in the medium was established. General setting: location, characteristics and present problems of the aquifer system of Ayamonte-Huelva The study area is on the southwest border of the depression of the Guadalquivir in the Huelva province (Spain) between the Guadiana and Piedras Rivers (Fig. 1). Aquifer system 25 (Ayamonte-Huelva) provides groundwater to a population of 150,000 in various municipalities, and an irrigated land area of approxi- mately 9,000 ha (ITGE 1989). The structure and hydrogeological functioning of the system are typical of a multilayer aquifer in a semicon- fined complex system (Grande 1993). It is characterized by alternating gravels and sands, semipermeable sandy marls, and with ages between the Andalusian period and Holocene period, limited by an impermeable substratum formed by blue marls (Andalusian) and shales (Grande 1993). The permeable materials present a monoclinal geometry from north to south (Grande et al. 1991) with transgressive overlap and thickness increasing from 8 m in the northern sector to 40 m in the southern most part. The depth from groundwater surface ranges between 10 and 28 m. Data (Grande et al. 1996) establish the natural average recharge to the aquifer to be about 100 hm 3 / year. However, this estimate is probably too high J.M. Andujar J. Aroba M.L. la deTorre J.A. Grande Contrast of evolution models for agricultural contaminants in ground waters by means of fuzzy logic and data mining Received: 20 May 2005 Accepted: 13 September 2005 Published online: 20 October 2005 Ó Springer-Verlag 2005 Abstract This work aims at con- trasting, by means of a set of fuzzy logic- and data mining-based algo- rithms, the functioning model of a detritic aquifer undergoing overex- ploitation and nitrate excess input coming from strawberry and citrus intensive crops in its recharge zone. To provide researchers unskilled in data mining techniques with an easy and intuitive interpretation, the au- thors have developed a computer tool based on fuzzy logic that allows immediate qualitative analysis of the data contained in a data mass from the water chemical analyses, and serves as a contrast to functioning models previously proposed with classical statistics. Keywords Detritic aquifer system Æ Pollution Æ Nitrate Æ Aquifer Æ Fuzzy logic Æ Data mining Environ Geol (2006) 49: 458–466 DOI 10.1007/s00254-005-0103-2 ORIGINAL ARTICLE M.L. de la Torre and J.A. Grande belongs to Water Resources and Quality Research Group. J.M. Andujar and J. Aroba belongs to Control and Robotics Research Group J.M. Andujar Æ J. Aroba M.L. la deTorre (&) Æ J.A. Grande Escuela Polite´cnica Superior, Universidad de Huelva, Ctra Palos de la Frontera, s/n., 21819 Palos de la Frontera, Huelva, Spain E-mail: mltorre@uhu.es Tel.: +34-959-217345 Fax: +34-959-217304