Study of variation in groundwater quality in a coastal aquifer in north-eastern Tunisia using multivariate factor analysis Sihem Charfi a, * , Kamel Zouari a , Saber Feki b , Ezeddine Mami c a University of Sfax, National School of Engineers, Laboratory of Radio-Analyses and Environment, B.P1173-3038 Sfax, Tunisia b University of Science and Technology (KAUST), Thuwal, Jeddah 23955, Saudi Arabia c Regional Commissary for Agricultural Development of Nabeul, Nabeul, Tunisia article info Article history: Available online 3 November 2012 abstract This work focuses on the Grombalia aquifer which constitutes the main water resource in Northeast Tunisia, Cap Bon Peninsula. The recharge of this aquifer is ensured mainly by direct infiltration of rainwater through permeable layers. Under semi-arid climatic conditions and increasing water demand for irrigation, about 80% of the Grombalia aquifer system shows different vulnerabilities to anthropogenic activities. The total dissolved solids values range from 0.75 to 5.6 g/l. Isotopic characterization with stable isotopes (d 2 H and d 18 O) of Grombalia aquifer system identified geochemistry processes that control water chemistry. In addition, the multivariate statistical technique (Principal Component Analysis) was used to identify the origin, the recharge mode and geochemical processes controlling groundwater quality. The principal reactions responsible for the hydrochemical evolution in the Grombalia groundwater fall into three categories: (1) denitrification process; (2) dissolution of salts; and (3) irrigation return flow process. Tritium data in groundwater from the study area suggest the existence of pre1950 and post1960 recharge. Ó 2012 Elsevier Ltd and INQUA. All rights reserved. 1. Introduction Groundwater quality could be influenced by many factors including climate, aquifer lithology, surface water recharge (Pulido Leboeuf, 2004), and waste water pollution. Principal component analysis (PCA) is a mathematical technique that is successfully applied to investigate problems in geological investigation, such as geochemistry (Chen et al., 2007), and environmental science (Liu et al., 2003). This statistical method has many advantages over classical graphical approaches as it takes into consideration a huge amount of spatial and temporal data of water quality. The main advantage of this method is the reduction of variables dimension by providing the correlation between measured chemical variables and their multivariate patterns based upon the correlation or covariance matrix (Helena et al., 2000). So, it is mainly used to group together different geochemical variables according to their degree of co-variation among samples. The geochemical interpre- tation of determined factors gives insights into the dominant processes influence groundwater chemistry (Chen et al., 2007). The aim of this study was the specification of regional groundwater and the mechanisms controlling the chemistry in the Grombalia area, using geochemical and isotopic tools. 2. Study area The study area lies in the Cap Bon Peninsula of Tunisia and extends between 36  29 0 00 00 e36  42 0 00 00 N, 10  27 0 00 00 e10  47 0 00 00 E, about 719 km 2 . It is located between the Bouchoucha and Halloufa Jebal in the west, Abderrahman Jebal in the east, the plains of Hammamet in the south, and the Gulf of Tunis in the north (Fig. 2). Grombalia plain is characterized by a semi-arid to sub-humid Mediterranean climate. This type of climate is the result of the convergence of several climatic parameters: rainfall is variable in time and space, with an annual average value of 500 mm. The mean annual temperature is around 18.7  C. The potential evapotrans- piration, about 1200 mm/year, is higher in summer as a result of the increase in temperature and the decrease in the air moisture. 3. Geology and hydrogeology The Cap Bon region contains a series of Quaternary terraces, marine and continental. The survey based on the geophysical data * Corresponding author. E-mail address: charfisihem1@yahoo.fr (S. Charfi). Contents lists available at SciVerse ScienceDirect Quaternary International journal homepage: www.elsevier.com/locate/quaint 1040-6182/$ e see front matter Ó 2012 Elsevier Ltd and INQUA. All rights reserved. http://dx.doi.org/10.1016/j.quaint.2012.11.002 Quaternary International 302 (2013) 199e209