Transience of Riparian Freshwater Lenses

Abstract

The current theory for defining the occurrence of riparian lenses (i.e., buoyant, lenticular‐shaped fresh groundwater bodies overlying saline groundwater in riparian zones) is largely based on steady‐state analyses, which neglect the transient dynamics expected in real‐world settings. In this study, the transience of riparian lens movement is investigated for the first time, considering a fully penetrating, gaining river (i.e., the river receives groundwater influxes), and rapid variations in hydraulic boundary conditions. Controlled laboratory experiments and numerical modeling are used to determine the timescales associated with lens movements. Both numerical and experimental timescales show asymmetric behavior in the growth and decline of lenses, whereby growing lenses involve longer timescales than those of shrinking lenses. Inverse relationships between timescales and final hydraulic gradients are observed, such that higher final hydraulic gradients involved shorter timescales. Results also revealed a time lag in the movement of the lens tip (i.e., where the lens and water table intersect) relative to the lens toe (i.e., intersection of the lens with the riverbank). Riparian lenses have near‐identical timescales regardless of whether the river water or aquifer hydraulic head varies, and rather, timescales are more sensitive to the final hydraulic gradient. Application of typical parameter ranges for the River Murray (South Australia) floodplains reveals timescales of years to more than a decade to expand the riparian lens by tens of meters. Insights into riparian lenses transience attained from the current study build on the understanding of transience obtained for other mixed‐convective hydrogeological processes, such as seawater intrusion and retreat.

Riparian Freshwater Lens Response to Flooding

“…The quantification of lens responses to a single river flood event, within laboratory experiments and numerical models, was based on the temporal changes in x L and V L . In numerical simulations, the 0.5 relative salinity isochlor was tracked as an approximation of the freshwater‐saltwater interface (e.g., Jazayeri et al., 2022). In addition, to examine the salt load to the adjacent river, temporal changes in q sr [L 2 T −1 ] (defined as the saltwater flux per unit width discharged into the river) were quantified and analyzed.…”

“…The Block Centered Flow (BCF) package was applied to control the cell wet‐dry conversions that arose during water table variations, similar to the approach of Jazayeri et al. (2022). Further details of the rewetting algorithms in the BCF package are given by Harbaugh et al.…”

“…The equivalent freshwater hydraulic conductivity of the clogging layer (i.e., the plastic screens), Κ r , can be calculated by κΚ , where κ [−] is the reduction ratio of the screen, that is, the ratio of the area of the holes in the screen to the cross‐sectional area of the sand tank (e.g., Werner et al., 2016). Here, κ was calculated as 0.44, with a possible range of 0.01–1 to account for uncertainty in estimating this parameter (e.g., Jazayeri et al., 2022). The normal and high river levels were 0.289 ± 0.001 m and 0.305 ± 0.001 m, respectively.…”

“…Solute conditions at the CHD and specified-head boundaries were represented by the Sink and Source Mixing (SSM) package of MT3DMS (Zheng & Wang, 1999) that allows inflows at a specified salinity (i.e., saltwater and freshwater with relative salt concentrations of 1 and 0, respectively) and outflows at the ambient groundwater salinity. The Block Centered Flow (BCF) package was applied to control the cell wet-dry conversions that arose during water table variations, similar to the approach of Jazayeri et al (2022). Further details of the rewetting algorithms in the BCF package are given by Harbaugh et al (2000).…”

“…Laboratory-scale physical models are commonly applied to various forms of lenses, including riparian lenses (e.g., Jazayeri, Werner, & Cartwright, 2021;Jazayeri, Werner, Wu, et al, 2021;Wu et al, 2020), as well as the lenses within coastal zones and oceanic islands (e.g., Lu et al, 2019;Stoeckl & Houben, 2012;Tang et al, 2021;Yan et al, 2021). More recently, Jazayeri et al (2022) conducted laboratory experiments to study the transient response of lenses (adjacent to fully penetrating, gaining rivers) to instantaneous boundary head changes due to variations in the river water level or saltwater head at the inland boundary. They found that the lens response timescales are negatively correlated with the final riverward hydraulic gradient.…”