1. Introduction
The region of the Loess Plateau in China comprises an inland area with limited freshwater resources and a considerable amount of brackish water resources. The brackish water resource is valuable to the irrigated agricultural in this region. However, the formation mechanism of this water resource remains unclear, which is constraining the rational utilization of brackish water. (Luo, Chen, & Han, 2010). Numerous studies have been conducted over the past several decades to assess the origin of brackish groundwater on a regional scale (Barth, 1998; Brenot, Négrel, Petelet-Giraud, Millot, & Malcuit, 2015; Cartwright, 2004; Cary et al., 2015; Jørgensen, Andersen, & Engesgaard, 2008; Werner et al., 2013; Sahib, Marandi, & Schüth, 2016). Most of these previous studies were focused on the coastal areas, where the formation of saline groundwater was usually related to seawater intrusion. With the increasing water demand, there is increasing study and focus on the formation of brackish water in inland areas, where the geological contexts and hydrological conditions are more complex than coastal areas (Farid, Zouari, Rigane, & Beji, 2015; Gamboa, Godfrey, Herrera, Custodio, & Soler, 2019; Gil-Márquez, Barberá, Andreo, & Mudarra, 2017).The sustainable management of inland brackish water resources is a global issue and requires a thorough assessment of the formation mechanism of brackish water (Alcalá & Custodio, 2008; Cartwright, 2004; Cary et al., 2015; Ghassemi, Jakeman, & Nix, 1995; Gil-Márquez et al., 2017; Monjerezi, Vogt, Aagaard, Gebru, & Saka, 2011).
The Zuli River is a first-order tributary of the Yellow River, which is located in the western Loess Plateau. Except in the headwater, the groundwater and surface water in this catchment are both brackish, with a characteristic of “bitterness”. Bitterness is an important factor affecting the quality and usability of water resources, particularly in inland areas (Gil-Márquez et al., 2017). However, few studies have been conducted to investigate the formation mechanism of bitterness in water.
Our previous study showed that groundwater recharge in the Zuli River catchment is produced by the infiltration of precipitation in the headwaters (Liu, Tan, Shi, Xu, & I.Elenga, 2019). During this recharging process, the upstream groundwater undergoes rapid salinization. Several hypotheses have been advanced in previous studies as to the source of salinity in brackish water. Current knowledge typically attributes the salinization of groundwater in most inland areas to the mixing of old groundwater in deep strata (Herrera et al., 2018), which usually means the salinity of groundwater originated from long periods of water-rock interactions (Petrides, Cartwright, & R.Weaver, 2006; Skrzypek, Dogramaci, & F.Grierson, 2013). However, our recent study showed that even under the rapid renewal of groundwater, the fresh groundwater from the headwaters still evolves into brackish water with rapid salinization (Liu et al., 2019). Several hypotheses have been advanced in previous studies as to the source of salinity in brackish water. The large amounts of soluble salts in loess were identified as a potential source of salinity (Luo et al., 2010). The contribution of other hydrogeochemical processes still needs to be considered comprehensively, especially the origin of the bitterness.
The shortage of freshwater resources in many areas has led to the use of brackish water resources for agriculture in the Zuli River catchment. The long-term utilization of this poor-quality water resources has severely restricted local economic development. In recent years, the government launched several agricultural water conservancy projects, such as ”The Taohe River Diversion Project ” and ” The Yellow River Diversion Project ” that divert freshwater flow into the Zuli River for the dilution of saline river water, but they had minimal impact. How to optimize water resources in a cost-effective and sustainable manner is a critical question for this area.
The present contribution uses an integrated approach of geochemical and isotopic analysis to identifying the origins of salinity and evaluating salinization processes. The general water chemical methods are difficult to discriminate the source of solutes in groundwater precisely. Research increasingly suggests that a multi-isotope approach is necessary to resolve the outstanding source of solutes in water (Cary et al., 2015; Gamboa et al., 2019; Jørgensen et al., 2008). Such an approach has been shown to be more effective in characterizing hydrogeochemical processes (Bullen, Krabbenhoft, & Kendall, 1996; Cartwright, Weaver, Cendón, & Swane, 2010; Cary et al., 2015; Harrington & Herczeg, 2003). The study uses strontium isotopes and boron isotopes to investigate the possible processes responsible for the formation of brackish water. As a relatively stable element, strontium isotopes are minimally fractionated in chemical and biochemical processes. Groundwater in equilibrium with Sr-bearing minerals attains87Sr/86Sr values that reflect the isotopic ratio of the minerals, leading to variation of the strontium isotopic ratio in different settings (Harrington & Herczeg, 2003; Monjerezi et al.,2011; Sahib et al., 2016). Therefore, the87Sr/86Sr ratio is classically used to provide information on the water-rock interaction under different geological contexts (Bullen et al., 1996; Brenot et al., 2015; Cartwright, 2004; Cartwright et al., 2010; Monjerezi et al., 2011; Palmer & Edmond, 1992; Pingitore & Eastman, 1986). On the contrary, boron is a relatively reactive element and mainly enriched in various types of rocks and water. Due to the isotopic distinctions, boron isotopes are prone to fractionate in many geochemical processes because of the larger relative difference in mass between its primary isotopes (Palmer, Spivack, & Edmond, 1987; Vengosh, Heumann, Juraske, & Kasher, 1994). Boron isotopes have been gradually applied to distinguish the original solute sources in groundwater, such as marine sources with enriched δ11B values and continental sources with significantly depleted δ11B values (Barth, 1998; Palmer & Swihart, 1996; Vengosh, 2005). Therefore, boron isotope is also an effective tracer for understanding the geochemical evolution processes on a regional scale (Barth, 1993; Cary et al., 2015; Morell et al., 2008; Palmer et al., 1987; Vengosh et al., 1994). Thus, the combination of boron and strontium isotopes can be used more precisely to understand the hydro-geochemical evolution processes in the groundwater system on a regional scale. To our knowledge, this study is the first to report on the strontium isotopes of brackish water as well as the boron isotopes of local precipitation, river water, and groundwater from the study area. It is anticipated that the results obtained herein will fill in that gap in the isotope study of the Loess Plateau.