Abstract Freshwater lenses within riparian zones of some arid and semiarid settings assist in maintaining the health of riparian ecosystems. We propose an approach for expanding freshwater lenses in saline aquifers adjacent to gaining rivers through the addition of a vertical barrier of low‐hydraulic‐conductivity (low‐ K ) parallel to the river bank. Sharp‐interface analytical solutions for the lens shape and water table distribution are developed to examine the effectiveness of the proposed method and are verified using sand tank experiments and numerical simulations. The sensitivity analysis is used to apply the method to parameters typical of the Lower River Murray (South Australia) and its floodplain aquifers. The results show that the barrier can create significant freshwater lenses in head‐controlled systems, whereas the barrier may lead to lens shrinkage in flux‐controlled systems due to saline water table rise. That is, the effectiveness of the barrier is highly dependent on the inland boundary condition. The analytical solution presented herein can be used to efficiently predict the riparian freshwater lens extent in response to engineered barriers, adding to existing techniques for studying and modifying riparian freshwater lenses.
Abstract This study examines the occurrence of riparian lenses adjacent to partially penetrating, gaining rivers under the controlled conditions of a laboratory sand tank. Laboratory experiments and numerical modeling of the freshwater lens extent are used to provide physical verification (in light of limited examples of well‐characterized field cases) of the analytical methodology, thereby evaluating the underlying assumptions. Parameter calibration and uncertainty analysis are applied to assess both the experimental conditions and the benefit of lens observations in applying the analytical approach. The experimental freshwater lens was reproduced by both analytical and numerical models, with the exception of small mismatches (between analytical results and measured data) in the lens thickness in the near‐river region. These are most likely due to vertical flow effects that arise from the partial river penetration and saltwater inflows to the river bottom, and that are only partly accounted for in the analytical approach. Uncertainty analysis highlighted that accurate lens predictions based on the analytical method requires calibration to direct lens measurements; a similar finding from earlier studies of fully penetrating river conditions. Sensitivity analysis highlighted 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 freshwater lens occurrence and saltwater discharge. The results provide the first physical verification of the occurrence of stable riparian lenses adjacent to partially penetrating, gaining rivers, and verify a recent analytical solution for lens extent and saltwater discharge.
Abstract Unconfined coastal aquifers with a sloping aquifer bed are ubiquitous. However, most analytical solutions previously developed to estimate pumping‐induced seawater intrusion considered a horizontal aquifer bed. In this study, we first developed a steady‐state analytical solution to quantify the maximum safe pumping rate in sloping unconfined coastal aquifers with a fixed‐flux inland boundary condition, using the single potential approach. The analytical solution corrected by an empirical factor is validated against the numerical simulation, which reproduces the corresponding numerical results well. It is found that coastal aquifers with a higher aquifer bed toward inland (i.e., a positive sloping angle) yield a higher maximum safe pumping rate than that of coastal aquifers with a horizontal aquifer base, since the elevated freshwater head prevents seawater intrusion and enhances groundwater pumping. Specifically, for the aquifer bases with the sloping angle of 0.01 and −0.01, the maximum safe pumping rate is increased by 42.6% and decreased by 48.4%, respectively, in comparison to the case of a horizontal aquifer base, suggesting that neglecting the aquifer base slop can result in a significant error in estimating the maximum safe pumping rate. Moreover, the sensitivity analysis reveals that the maximum safe pumping rate increases with a lower hydraulic conductivity and a larger dispersivity. Our results provide valuable insights into the effect of the aquifer base slope on the maximum safe pumping in coastal aquifers, and analytical solutions developed can serve as a powerful tool for quick assessment of pumping‐induced seawater intrusion in sloping unconfined coastal aquifers.
Abstract Previous studies of freshwater lenses in saline aquifers adjoining gaining rivers (“riparian lenses”) have so far considered only rivers that fully penetrate the aquifer, whereas in most cases, rivers are only partially penetrating. This paper presents a new methodology for obtaining the saltwater discharge and the shape of a steady‐state, non‐dispersive riparian lens, where the river is partially penetrating, combining two previous analytical solutions. The resulting analytical solution is compared to numerical modeling results to assess assumptions and the methodology adopted to approximate the “turning effect”, which is the change in groundwater flow direction (horizontal to vertical) near the partially penetrating river. Model parameters were taken from previous studies, representing simplified conditions in the River Murray floodplains (Australia). Consistency between analytical and numerical results and field observations highlights the capability of the proposed analytical solution to predict the riparian lens geometry and saltwater discharge into partially penetrating rivers. Sensitivity analysis indicates that larger riparian lenses are produced adjacent to the deeper and wider rivers, as expected. The change in width or depth of the river has more influence on the saltwater discharge and the horizontal extent of the riparian lens (and less effect on the vertical extent of the lens adjacent to the river) for shallower and narrower rivers. This research highlights the utility of the new method and demonstrates that the assumption of a fully penetrating river likely leads to significant overestimation of the saltwater discharge to the river and the riparian lens horizontal extent and vertical depth.
Abstract The freshwater lenses (“lenses” hereafter) within saline floodplain aquifers are sensitive to river flooding events. However, the effects of extensive floodplain inundation on saline aquifers are rarely considered and have not been examined previously under controlled laboratory conditions. We conducted laboratory experiments within a two‐dimensional (cross‐section) sand tank (i.e., representing a saline floodplain aquifer) and built both laboratory‐ and field‐scale numerical models to examine lens responses to flood events. Three sets of experiments were performed to explore different lateral extents of floodplain inundation. The temporal behavior of experimental lenses was quantified and compared to variable‐density numerical models that adopted calibrated laboratory parameters, showing good agreement. Results show that more extensive floodplain inundation leads to larger lenses (as expected). The sensitivity analysis was performed based on field‐scale numerical models, demonstrating that the floodplain inundation extent, hydraulic conductivity, and dispersivity are key factors controlling the post‐flood recession in lens extent and volume. In field‐scale simulations of floodplain inundation, the entire lens was significantly salinized during flood recession due to enhanced dispersion accompanying higher groundwater velocities, which may further split into several isolated freshwater bodies before eventually returning to steady‐state conditions. Importantly, field‐scale numerical results indicated that the salt load to the adjacent river increased immediately following the flood event, consistent with reporting of the River Murray (South Australia). These results provide critical new insights into relationships between flood events and the behavior of lenses, highlighting the significance of flooding events on both intermediate and long‐term conditions of saline floodplains.
Abstract Subsurface physical barriers have been recognized as effective in mitigating seawater intrusion in coastal aquifers, although mainly 2D (cross‐sectional) barrier effects have been considered. In this study, impermeable barriers with finite shore‐parallel lengths are investigated through 3D numerical simulation, thereby extending previous analyses. Two scenarios are considered: (a) barrier‐only and (b) barrier‐well systems; and three available barrier types are analyzed and compared: (1) subsurface dam, (2) cutoff wall, and (3) fully penetrating barrier. Barrier location, length, and height are investigated, and barrier effectiveness is evaluated from seawater volumes, seawater wedge toe positions, and maximum safe pumping rates. In the barrier‐only system, a better performance in preventing seawater intrusion was achieved by cutoff walls rather than subsurface dams. Finite‐length subsurface dams may slightly enhance seawater extent along parts of the coastline that are beyond the dam's length. Cutoff walls performed best when located at relatively small distances from the coast in the barrier‐only system, whereas with a well at 450 m from the shoreline, the subsurface dam located at a critical distance from the sea (i.e., 300 m in the current study) performed optimally (from the tested cases) and was superior to cutoff walls in terms of the maximum safe pumping rate. A fully penetrating barrier outperformed cutoff walls and subsurface dams, as expected. Our investigation indicates that subsurface barrier design should consider the effect of the shore‐parallel length, because barrier benefits may otherwise be significantly overestimated.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
'/%3e%3c/defs%3e%3cpath fill='%23012169' d='M0 0h10080v5040H0z'/%3e%3cpath stroke='%23fff' stroke-width='.6' d='m0 0 6 3m0-3L0 3' clip-path='url(%23b)' transform='scale(840)'/%3e%3cpath stroke='%23e4002b' stroke-width='.4' d='m0 0 6 3m0-3L0 3' clip-path='url(%23c)' transform='scale(840)'/%3e%3cpath stroke='%23fff' stroke-width='840' d='M2520 0v2520M0 1260h5040'/%3e%3cpath stroke='%23e4002b' stroke-width='504' d='M2520 0v2520M0 1260h5040'/%3e%3cg fill='%23fff'%3e%3cuse xlink:href='%23d' x='2520' y='3780'/%3e%3cuse xlink:href='%23a' x='7560' y='4200'/%3e%3cuse xlink:href='%23a' x='6300' y='2205'/%3e%3cuse xlink:href='%23a' x='7560' y='840'/%3e%3cuse xlink:href='%23a' x='8680' y='1869'/%3e%3cuse xlink:href='%23e' x='8064' y='2730'/%3e%3c/g%3e%3c/svg%3e)