Effects of River Partial Penetration on the Occurrence of Riparian Freshwater Lenses: Experimental Investigation

Abstract

Freshwater lenses have been observed within riparian aquifers where freshwater rivers traverse and connect to saline groundwater (e.g., Cendón et al., 2010;Laattoe et al., 2017). Such settings are encountered in arid and semi-arid regions where saline groundwater often occurs because of significant evapo-concentration effects (Alaghmand et al., 2013;Bauer et al., 2006). The fresh groundwater found in riparian zones (i.e., buoyant, lenticular-shaped freshwater bodies adjacent to rivers; "riparian lenses" hereafter) play an important role in sustaining fragile riparian and floodplain ecosystems (e.g., the Chowilla floodplain, South Australia; Holland et al., 2009) in arid and semi-arid regions, particularly during low-flow periods (e.g., Holland

Transience of Riparian Freshwater Lenses

“…The cross‐sectional SEAWAT model domain representing the sand tank experiments was discretized into 81,440 cells with uniform size of 2 mm × 2 mm (i.e., the cell size in x and z directions, ∆ x and ∆ z [L], respectively), with unit length in the y ‐direction (i.e., perpendicular to the cross‐section). The adopted mesh resolution is consistent with previous riparian lens studies (e.g., Jazayeri et al., 2021; Werner et al., 2016). This maintained the accuracy of the results while allowing for reasonable model run times (up to 3 hr on a quadcore Intel® Core™ i5‐7500 processor).…”

“…(2016) and Jazayeri et al. (2021) showed that the steady‐state riparian lens theory of Werner and Laattoe (2016) was unable to accurately reproduce the lens thickness in the near‐river region, as ascertained by comparison to numerical models and sand tank experiments. They attributed the near‐river mismatch in the lens thickness to the assumption of saltwater hydrostatic conditions in Werner and Laattoe's (2016) solution, which neglects vertical flow effects.…”

“…A possible range of 0.01-1 was assumed for κ in the calibration processes (Section 2.3) to account for uncertainties involved in the estimation of this parameter. The same method was used in previous riparian lens experiments conducted by and Jazayeri et al (2021) in accounting for the screen resistance to flow.…”

“…Laboratory‐scale sand tanks are commonly employed to investigate variable‐density flow problems, which are usually challenging to characterize under field conditions. For example, laboratory experiments have been utilized in previous studies of riparian lenses (e.g., Jazayeri et al., 2021; Werner et al., 2016; Wu et al., 2020), freshwater lenses in oceanic islands (e.g., Lu et al., 2019; Stoeckl et al., 2016) and the transience of seawater intrusion and retreat in coastal aquifers (e.g., Abdoulhalik & Ahmed, 2018; Badaruddin et al., 2015; Goswami & Clement, 2007). In addition, physical experiments can provide benchmark data to assess the performance of numerical models, particularly where buoyancy forces and dispersion lead to complex transport processes (e.g., Goswami & Clement, 2007; Oswald & Kinzelbach, 2004; Werner et al., 2009).…”

“…Jazayeri et al (2020) combined the riparian lens theory of with existing theory for partially penetrating rivers developed by Morel-Seytoux (2009), Miracapillo andMorel-Seytoux et al (2014) to create a methodology for the prediction of saltwater discharge and stable riparian lens characteristics adjacent to partially penetrating, gaining rivers. The analytical solution of Jazayeri et al (2020) was subsequently validated using laboratory experiments of partially penetrating, gaining rivers by Jazayeri et al (2021), who also showed that the saltwater head boundary, freshwater and saltwater densities, and the aquifer depth below the riverbed (in descending order of sensitivity) are the most important factors in controlling riparian lens occurrence and saltwater discharge. America et al (2020) conducted a numerical study of the influence of evaporation and salt accumulation on riparian lenses.…”