Hydrochemical effects of saltwater intrusion in a limestone and dolomitic limestone aquifer in Lebanon

Highlights

• Hydrochemical effects of saltwater intrusion are assessed via a quad-fold approach.

• Salinization stimulates different hydrochemistry in limestone/dolomitic limestone.

• Part of observed variations in TEs concentrations in groundwater is explained.

• Magnesium distinct behavior has direct effects on major hydrogeochemical processes.

• Carbonate rocks are weak geogenic sources of TEs with and without salinization.

Abstract

This study demonstrates groundwater quality differences between a limestone and a dolomitic limestone, (sub)oxic coastal aquifer in the Eastern Mediterranean (Lebanon), with and without ongoing moderate salinization since the last decades. For this purpose, 8 major and 50 trace elements (TEs) were analyzed in 80 water and 65 rock samples, and interpreted with a quad-fold approach utilizing: (1) nonparametric statistical tests, (2) concentration deviations from ideal conservative freshwater–seawater mixing lines, (3) a new parameter called Mixing Enrichment Factor to assess the mobility of chemical constituents under salinizing conditions, and (4) 1-D dual porosity flow path modeling with PHREEQC. Dissolution/precipitation of Ca_xMg_ySr_zCO₃ and cation exchange were the main disclosed hydrogeochemical processes besides minor signs of organic matter oxidation. In the dolomitic limestone aquifer, less carbonate dissolved as compared to the limestone aquifer, partly because of lower pCO₂ in addition to seawater inflow triggering Mg-calcite precipitation by cation exchange. Saltwater intrusion led to mobilization of As, Ba, Cu, Ni, Rb, Sr and U in both aquifers, sometimes likely by cation exchange (e.g. Ba and Sr). Some of these TEs (notably Cu and Ni) recorded higher concentrations in the dolomitic limestone regardless of salinization. Other elements like Al, Be, Ce, Cr, Nb, Pb, V, Y and Zr revealed no or a low mobilization tendency. The concentration of all TEs in groundwater remained below drinking water limits notwithstanding moderate salinization. This classifies carbonate rocks as a weak geogenic source of TEs, whereas encroaching seawater appears to be a more important source.

Introduction

Coastal aquifers are known for their complex hydrochemical nature due to (1) different inputs from precipitation, infiltrating rivers, intruding seawater, irrigation return flow, and wastewater infiltration (Stuyfzand, 1999, Post, 2002); (2) additional or intensified processes in the saltwater-freshwater mixing zone such as dissolution/precipitation, ion exchange, and oxidation/reduction (Herman et al., 1985, Magaritz and Luzier, 1985, Stuyfzand, 1993a); (3) changes in permeability and porosity due to enhanced karstification or dolomitization of carbonate aquifers (Wicks and Herman, 1995, Emblanch et al., 2005) or clay mobilization in clastic aquifers (Bradford and Kim, 2010); and (4) urbanization of coastal zones (Porter et al., 1996). Approaches to interpret the hydrochemical evolution of groundwater bodies can be subdivided into 3 main groups: (1) mapping (including the use of water typologies and quality indices (e.g. Matthess, 1982, Stuyfzand, 1999), (2) mass balancing (Plummer and Back, 1980, Plummer et al., 1994, Dai et al., 2006, Stuyfzand, 2011, Parkhurst and Appelo, 2013), and (3) 1-D (Appelo, 1994), 3-D (Prommer et al., 2003) reactive transport or variable-density groundwater flow and solute transport modeling (e.g. Sanford and Konikow, 1989, Simmons et al., 2010).

Studying the behavior of trace elements (TEs) in coastal groundwater flow systems is important because they need to be tested against maximum permissible concentrations (MPCs) of drinking (e.g. Mondal et al., 2010, Al Kuisi et al., 2015, Fiket et al., 2015) or irrigation water (e.g. Shi et al., 2013), and some can be utilized as tracers of either infiltration water, pollution or geochemical processes (e.g. Stuyfzand, 1993b, Gonneea et al., 2014, O'Connor et al., 2015, Stuyfzand, 2015a, Sun et al., 2015). The behavior of TEs is initially affected by the geochemistry of hosting rocks, but saltwater intrusion and mixing with freshwater may trigger further hydrogeochemical processes, notably dissolution/precipitation and adsorption/desorption. For instance, a carbonate-philic element like Sr is significantly mobilized upon limestone dissolution (Shand and Edmunds, 2008, Lin et al., 2013, Gonneea et al., 2014). Other elements such as, U and Zn have less bonding affinity to carbonates (Shand and Edmunds, 2008, Gonneea et al., 2014), but they may be mobilized by increased O₂ concentrations (for instance due to leakage of oxygenated water via multi-aquifer wells) in limestone and clastic (sandstone, shale, and alluvium) aquifers (Ayotte et al., 2011). Several studies in salinizing sandy aquifers confirm the affinity of TEs to become mobilized, e.g. Ba (Stuyfzand, 1993b), Pb and Hg (Sun et al., 2015), or immobilized, e.g. F, Mo, Rb, V and U (Stuyfzand, 1993b). In carbonate coastal aquifers with varying Ca/Mg host rock composition, the behavior of TEs has not yet been thoroughly tackled, except for a study of the Italian Dolomite Mountains (Frondini et al., 2014); however, salinization was not an issue there.

This paper focuses on water quality differences between limestone and dolomitic limestone aquifers with and without salinization. A (sub)oxic coastal system in the Eastern Mediterranean (Lebanon) is chosen for this purpose. The selected system has been subject to excessive pumping since 1991. This has intensified saltwater encroachment. It is yet at a moderate salinizing stage with maximum seawater fraction <20%. Such carbonate aquifers are inherently complex due to their karstic nature. Many authors have addressed the hydrochemistry of Mediterranean coastal aquifers, e.g. Price and Herman, 1991, Pulido-Leboeuf, 2004, Tulipano et al., 2005, MED-EUWI, 2007; de Montety et al., 2008, Sola et al., 2013, Zghibi et al., 2014, Ben Ammar et al., 2016. In Lebanon, however, hydrochemical studies have applied classical approaches only, like (a) Piper, Schoeller, Durov, and Radial plots (Arkadan, 1999, Khadra, 2003, Korfali and Jurdi, 2007, Korfali and Jurdi, 2009, Saadeh, 2008), or (b) simple water quality indices (Khadra, 2003, Saadeh, 2008, El-Fadel et al., 2014). Recently, Khadra and Stuyfzand (2014) presented a detailed hydrochemical study of the (dolomitic) limestone Damour coastal aquifer system to the south of Beirut. It included a discrimination scheme to define different groundwater bodies (hydrosomes) and their interbedded hydrochemical zones (facies) in addition to the derivation of baseline quality for main constituents, stable isotopes, and many TEs.

The emphasis in this manuscript is given to the hydrochemical effects induced by saltwater intrusion (SWI) on various major constituents and trace elements in conjunction with potential geochemical differences between limestone and dolomitic limestone units. Four lines of research are followed for this purpose, in logical order of increasing complexity: (1) statistics on 4 water groups (limestone vs. Dolomitic limestone with and without salinization); (2) shifts in concentrations of major ions from ideal seawater and freshwater mixing lines; (3) use of Mixing Enrichment Factor (MEF), which is introduced here as a new parameter to assess mobility of chemical constituents when mixing (including salinization) occurs; and (4) 1-D flow path PHREEQC modeling with dual porosity formulation. Lines 1 and 3 analyze many species including TEs, whereas lines 2 and 4 are limited to Na, K, Ca, Mg, SO₄, TIC (or alkalinity), and occasionally Sr. 50 TE (including metals, metalloids, lanthanoids, and actinoids) are considered. Their total amount in groundwater is <0.3% of the total dissolved solids for all analyzed samples. 15 elements provide meaningful indications, whereas the others have concentrations constantly below their minimum detection limit (MDL).