About eighty percent of the world’s population live close to the coastline and withdraw large amounts of freshwater from shallow coastal aquifers. Consequently, the management of coastal groundwater aquifers is one of the most challenging environmental management problems confronted by the water industry, worldwide. Under normal conditions, the ambient groundwater flux discharging towards the ocean will prevent the dense saltwater from encroaching into a coastal aquifer. Typically, there is a natural transition boundary between the salt and fresh water zones, which is located well below the saturated groundwater level. Over exploitation of coastal aquifers, however, has resulted in the lowering of groundwater levels; this has led to the occurrence of saltwater intrusion in several metropolitan areas. Furthermore, large-scale catastrophic events such as tsunamis and hurricanes may inject saltwater into local aquifers and contaminate large volumes of freshwater reserves. Therefore, understanding the mixing dynamics of saltwater within freshwater aquifer systems is an important fundamental research problem. The solution to this problem requires simulation of density coupled transport under both stable (e.g., saltwater intrusion) and unstable (e.g., tsunami invasion) interface conditions. In this work, we will present experimental datasets to illustrate the dynamics of density coupled flow in saturated groundwater systems. These datasets are useful for visualizing the saltwater flow and mixing patterns within a saturated formation; in addition, they are also useful for benchmarking density-coupled transport codes.
Numerical models are commonly used
as a tool to study large-scale saltwater intrusion management problems. The
performance of these codes is often validated by solving a set of benchmark
problems involving both stable and unstable interfaces. The most widely used
benchmark problem for testing saltwater intrusion codes is the Henry problem,
which is a stable interface problem. This problem is based on an analytical
solution derived by Herald Henry, as a part of his PhD dissertation at Columbia
University (Henry, 1960). Henry considered salt wedge transport in a rectangular,
saturated, two-dimensional, confined porous media domain. He adapted a mathematical
solution developed for modeling heat transfer processes to solve his steady-state,
saltwater intrusion problem. Over the years, the original Henry problem has
undergone several revisions, because none of the numerical models were able
to recreate the exact analytical solution for a variety of reasons (Simpson
and Clement, 2004). Segol (1994) reinvestigated Henry’s analytical expression
and suggested corrections to the mistakes originally made by Henry. These corrections
finally allowed investigators to exactly match their numerical results against
the revised Henry solution (Simpson and Clement, 2004). However, to date, no
one has made careful measurements of fluxes and transport patterns in a Henry-type
problem using a laboratory-scale physical model. In this paper, we present an
experimental dataset that can be used for visualizing the transport patterns
occurring within a Henry-type problem. Furthermore, the dataset can also be
used for benchmarking both transient and steady-state saltwater intrusion models.
In addition, we also used the physical model to observe the freshwater flow
patterns near a sloping saltwater discharge boundary (similar to a natural beach
boundary). Flow and solute transport near a groundwater-surface water boundary
is controlled by the presence of a seepage face (Simpson et al. 2003; Clement
et al. 1993). Our laboratory observations show that the presence of a saline
surface-water body causes interesting upward flow patterns beneath the saltwater
boundary, which were similar to the patterns observed in a recently published
field study (Westbrook et al. 2005).
While testing numerical models it is important to assess the performance of the model under a wide range of flow and transport conditions. Simpson and Clement (2003) developed a novel coupled/uncoupled solution strategy to demonstrate the importance of testing density-coupled numerical models by solving problems involving both stable (saltwater in the bottom) and unstable (saltwater on top) interfaces. In this study, we will also present the results of several unstable saltwater flow experiments that can potentially be used as benchmark problems for testing unstable density coupled flow systems. The focus of all our unstable flow experiments was to simulate various types of saltwater contamination scenarios that would occur under tsunami-type saltwater flooding condition. The experiments were specifically designed to replicate the tsunami event that occurred along the east coast of Sri Lanka (Illangasekare et al. 2006). The experimental data and some preliminary modeling results for a tsunami invasion scenario will be presented.