Groundwater Engineering

The largest source of human drinking water is stored and flows in the subsurface. Geological formations saturated in mobile groundwater that can be exploited for human use are called aquifers. This chapter introduces basic notions that set the ground for the understanding and description of subsurface water flow. First, the main properties of water are illustrated, with a particular focus on the forces it establishes with the solid matrix of a porous medium and on how these affect its mobility. Then, broad aquifer classifications are provided, based on their geographical location, their permeability characteristics as a function of the type of porosity (i.e., intergranular, fracture or karst), and their degree of confinement. The latter, which categorizes aquifers as unconfined, leaky or confined, has crucial implications on both their storage capacity and hydrodynamic behavior. The key parameters that characterize an aquifer’s storage capacity are porosity and storativity. While the former is indicative of the total amount of water that can be stored within a porous medium, the latter indicates the fraction that can be released. Both these notions apply to any aquifer type although the mechanism of water release is distinct in unconfined and confined aquifers: in the former, water is released under the effect of gravity alone, and storativity is called specific yield; in the latter, water is released as a result of water expansion that follows a pressure drop. Subsurface water transport, instead, is driven by the existence of a hydraulic gradient (i.e., a drop in hydraulic head, or piezometric level). Under specific hypotheses, groundwater flow can be described by Darcy’s law, which establishes a proportionality relationship between flow rate and hydraulic gradient, and can be used to map an aquifer’s flow field. The relation defined by Darcy’s law is measured by an aquifer-specific parameter called hydraulic conductivity. This parameter is crucial not only in the description of the transport capacity of a porous medium, but also in the calculation of its productivity, which is a function of the hydraulic conductivity and the thickness of an aquifer.

In this chapter, the differential groundwater flow equation that governs the distribution of the flow directions and rates in an aquifer is derived. The problem is examined at a macroscopic scale, neglecting an analysis of detailed solid–liquid interface distribution, which would entail excessive analytical and computational complexity, without contributing useful information from an operational standpoint. The equation is thus determined as a combination of the equations that express the law of mass conservation (whose terms are described for a representative elementary volume), Darcy’s law, and the storage variation due to changes in hydraulic head. Unsteady state groundwater flow in each aquifer type is described by a different equation, each defining the Laplacian of the hydraulic head as a function of the aquifer’s storage and transport capacity, and of the hydraulic head’s partial derivative with respect to time. In particular, in the case of confined aquifers, flow is a function of specific yield and transmissivity, as is the case even for leaky aquifers, whose hydrodynamic behavior is, however, also affected by the leakage between aquifers (quantified by the leakage factor). A rigorous description of flow in unconfined aquifers would require a nonlinear and nonhomogeneous differential equation due to the inclination of the water table with flow; however, under simplifying conditions an approximate description, analogous to the confined aquifer equation, can be defined as a function of specific yield and transmissivity.

The differential groundwater flow equation derived in Chap. 2 can be solved analytically in various geometries, provided that certain hypotheses are satisfied. In this chapter, a polar coordinate system with radial geometry, describing the radial groundwater flow towards a well, is considered. The hypotheses underlying the analytical solutions concern the aquifer’s geometry (constant thickness, homogeneity and isotropy, unlimited horizontal extension, initially horizontal potentiometric surface) and the pumping well (fully penetrating, infinitesimal radius, negligible storage, laminar flow and constant pumping rate). Steady state and transient analytical solutions, respectively describing the drawdown as a function of the distance from the well (r), or of r and time, are provided for confined, leaky and unconfined aquifers. Theis’ (and Cooper and Jacob’s approximation) and Thiem’s equations describe, respectively, the transient and steady state solutions of the groundwater flow equation for confined aquifers. Hantush and Jacob, instead, derived the transient analytical solution for leaky aquifers, while De Glee formalized the steady state solution. In the case of unconfined aquifers, the steady state solution formally coincides, except for an adjustment to the drawdown, to Thiem’s solution. The transient solution was, instead, derived by Neuman, under specific simplifying hypotheses, given that a fully rigorous description of flow in unconfined aquifers would entail the use of a nonlinear and nonhomogeneous differential equation due to the inclination of the water table with pumping and the generation of a vertical component of flow velocity.

Aquifer tests are the most appropriate method to determine the hydraulic behavior of an aquifer and the distribution of the hydrodynamic parameters that govern such behavior. This chapter illustrates the different type of aquifer tests (i.e., pumping, recovery and slug tests) and how to plan, execute and interpret them. Pumping tests consist in measuring the drawdown induced by the extraction of water from a well at a constant discharge rate in one or more observation points. They allow to first identify the hydraulic behavior of the aquifer, and thus to classify it as confined, leaky or unconfined, and then to determine, via a type curve matching method, the aquifer’s horizontal hydraulic conductivity, transmissivity and storativity. In the case of leaky aquifers, also the leakage factor can be calculated; and in the case of unconfined aquifers, the effective porosity and the vertical hydraulic conductivity can also be derived. Clearly, this interpretation relies on a number of ideal hypotheses being satisfied; this chapter also illustrates how to interpret pumping tests in the presence of factors that cause a deviation from the ideal behavior (e.g., finite, as opposed to infinitesimal, well radius and volume; partially penetrating well; presence of recharging or impermeable boundaries; inconsistent pumping rate; permeability damage close to the well). During recovery tests, residual drawdown measurements are carried out following the interruption of the pump at the end of a constant discharge pumping test. Theis’ recovery method, based on the superposition principle and normally used for the interpretation of the test, allows to determine the transmissivity of an aquifer. The last type of aquifer test, i.e., the slug test, consists in inducing an instantaneous variation of the static water level in a well or piezometer, and subsequently measuring the recovery over time of the undisturbed level in the same well. This method is used to determine the hydraulic conductivity of the aquifer in proximity of the well. In this chapter, the most common interpretation methods are presented, as well as the strategies to overcome limitations due to the existence of factors that cause a deviation from the ideal behavior. Finally, a suite of correlation-based, laboratory, and field methods available for the determination of an aquifer’s hydrodynamic parameters in alternative to aquifer tests are presented, and the applicability to different aquifer types and situations of each method, as well as their reliability is discussed.

In order to effectively and sustainably exploit an aquifer, having determined its hydrodynamic characteristics and behavior (through aquifer tests, as described in Chap. 4) is not sufficient. It is, in fact, necessary to conduct a well test to derive information on the production and efficiency of a water supply system. Operationally, they are step-drawdown tests, consisting in a succession of at least three pumping stages, each with an increased discharge and lasting until a pseudo-steady state drawdown is achieved. The aim of these tests is to correlate each stabilized drawdown value in the well to its corresponding pumping rate, and to discern between head losses attributable to the aquifer system and those deriving from the water supply system (e.g., well design and construction characteristics, formation damage close to the well, non-darcyan flow close to the well and within the screen slots). Step-drawdown test interpretation is based on Rorabaugh’s empirical equation. According to this method, the stabilized drawdown measured in a well as a result of pumping a constant discharge is the sum of a linear term, representing the total head losses resulting from the laminar component of flow, and an exponential term, representing the head losses due to inertial flow in the proximity of the well and to turbulent flow through the screen slots. This chapter provides the means to estimate such terms and how to use them to determine the productivity (quantified by its specific capacity) and efficiency of a water supply well.

In the previous chapter, the methods available for assessing the productivity and efficiency of a well were illustrated. However, a water supply system extends beyond the aquifer-well system, and is composed also of water transmission, treatment, storage and distribution elements. In this chapter, methods for the estimation of head losses occurring from the pumping well along the pipe network that transfers water to the treatment plant, before distribution to consumers, are illustrated. Along this network, there are distributed head losses, in straight pipes, and local head losses, in valves, bends and outlets into reservoirs. Commonly used empirical equations for the calculation of these losses and a method for identifying the optimal operating pumping rate are provided, based on a comparison between pump-related and system head losses. Furthermore, other aspects that need to be kept in mind in the design and long term maintenance of a highly efficient water supply system are highlighted.

Water supply systems must be designed in such a way to ensure groundwater extraction sustainability. In addition, the quality of pumped water must also be guaranteed, and this entails protecting the groundwater source from contamination. To do so, it is necessary to identify the physical and hydraulic characteristics of the soil, the unsaturated medium and the aquifer itself that influence the migration of contaminants spilled at the surface towards the aquifer, and hence potentially towards sensitive targets (i.e., drinking water pumping wells). The susceptibility of an aquifer to become polluted following a contaminant spill is called vulnerability, and its assessment is the focus of this chapter. Of the four categories of vulnerability assessment methods, i.e., overlay, index and statistical methods, and process-based simulation models, this chapter presents examples of the former two, which are of easier implementation and are widely used. Overlay methods define aquifer vulnerability on the basis of groundwater circulation and rely on the superposition of maps of the hydrogeologic, structural and morphologic setting. Index methods, instead, are based on the assignment of scores (sometimes weighed) to sets of parameters that are likely to affect the degree of vulnerability. Specific methods of these two categories described in detail in this chapter are the one developed by the Bureau de Recherches Géologiques et Minières in France, the Italian CNR-GNDCI and SINTACS methods, the US-EPA DRASTIC method and the British GOD method. The suitability of different methods is discussed, and how vulnerability assessment can be used to determine the risk of contamination is presented. On this basis, an example of contamination risk reduction strategies is illustrated.

In the previous chapter, methods for the assessment of the vulnerability of aquifers to contamination are illustrated as a first step towards the protection of groundwater resources. Here, static and dynamic protection measures of sources of water for human consumption are presented. Static protection entails the definition of areas of land around the water source that must be subjected to specific safeguard measures and land use limitations. Such areas can be defined via geometric methods (i.e., defining the area arbitrarily, such as by drawing a circle of set radius around the pumping well) or via the time of travel approach. The latter method takes into account the aquifer type and its hydrodynamic parameters, in particular using the groundwater flow velocity to delineate protection areas defined by the time it takes a contaminant to reach the drinking water extraction well. Dynamic protection entails the establishment of a monitoring network along the perimeter of previously defined protection areas. Practical guidance is provided for appropriately designing this network in terms of monitoring-well positioning and sampling frequency.

Aquifer contamination occurs following a release of chemical compounds in groundwater exploited for human consumption which poses a health risk to the consumers. There is a variety of anthropogenic causes of contamination, spanning from discharge of wastewater to the ground, to industrial or mining activities, from accidental spills to agricultural activities. The wide range of sources of contamination is reflected on the extremely broad and diverse set of contaminants, including biological, chemical and radioactive constituents. This chapter is dedicated to the chemical, physical and toxicological classification and characterization of chemical contaminants. Chemically, compounds can be broadly categorized as inorganic (e.g., metals, certain anions and cations, metalloids) or organic (i.e., containing at least one organic carbon). The main organic groups are described, including hydrocarbons, halogenated hydrocarbons, phenols, chlorobenzenes, nitroaromatic compounds, and a class of recently identified hazardous compounds, named emerging organic contaminants, is presented. A physical characterization of contaminants is essential for the prediction of their behavior once they are released to the ground and migrate either across the unsaturated zone towards the saturated medium, or directly in the aquifer. The most important physical characteristics affecting contaminant migration and illustrated in this chapter are physical state, miscibility with water, mass density, solubility in water and volatility. Finally, a toxicological classification of contaminants is provided, which categorizes them as threshold or non-threshold compounds, depending on whether their health effects are manifested only above a certain concentration or are independent of the exposure level (i.e., they induce genetic mutations which lead to cancer development). This classification lays the foundations for the definition of threshold concentration values in drinking water prescribed by national and international health agencies and regulatory authorities. A comparison of the guideline or regulatory values defined by the WHO, the US-EPA, the EU and the Italian law is provided.

This chapter focuses on the mechanisms that govern the propagation of contaminants in aquifers. A qualitative and analytical description of the main hydrological, physico-chemical and biological process is provided. The hydrological mechanisms responsible for the transport and spreading of contaminants derive from the presence and movement of groundwater. The first of such processes is advection, according to which a compound is transported along the main direction of flow at seepage velocity. Molecular diffusion, instead, is responsible for the migration of the contaminant from high to low concentration areas, as a result of thermal agitation of water molecules. The last hydrological process is mechanical dispersion, which is a consequence of microscale heterogeneities present in the porous medium and results in a non-uniform velocity distribution relative to seepage velocity and the emergence of a transverse velocity component. During their migration within an aquifer, chemical compounds can also undergo chemical reactions that can lead to their transformation or degradation. The main reaction models and the most common types of reactions that are likely to occur in groundwater (i.e., acid-base reactions, complexation, hydrolysis, dissolution and precipitation, radioactive decay) are illustrated. Contaminant transformation and degradation can also be biologically mediated, primarily through microbial activity; such reactions are often described through first-order reaction kinetic models or Monod’s model. Finally, contaminant concentration in groundwater is also affected by sorption, a process by which compounds are removed from solution and transferred to the solid phase through a partitioning process typically characterized through isotherms. All these processes are described individually in this chapter, although in reality they occur simultaneously. The chapters that follow describe contaminant transport accounting for the concomitance of these processes.

This chapter derives the differential equations of mass transport, distinguishing between conservative solutes (i.e., exclusively undergo hydrological processes) and reactive solutes (i.e., also undergo physico-chemical and/or biological processes). The differential equations are derived by imposing the mass balance of a generic solute in a representative elementary volume during a certain time interval.

In this chapter, analytical solutions to the differential equation of mass transport for conservative solutes are illustrated. Their derivation relies on a number of simplifying hypotheses, including that: the medium is saturated, homogeneous and isotropic; water has constant density and viscosity, regardless of solute concentration; Darcy’s law is valid; flow directions and rates are uniform; transport parameters are constant within the domain; boundary conditions are constant in time. Solutions for one-, two- and three-dimensional geometries are presented, the former being mainly used for the interpretation of laboratory experiments, the latter two being more relevant for practical applications. Pulse and continuous solute release are considered. Notably, in a three-dimensional geometry a pulse input from a point source and a continuous input from a plane source are illustrated. A solution of the differential equation of mass transfer for the former contamination scenario was derived by Baetslé, while Domenico and Robbins proposed a model for the latter.

Having provided a few analytical solutions for conservative solutes in the previous chapter, here the focus is on reactive solutes. The underlying hypotheses considered to be verified in order to obtain such solutions are the same as in the case of conservative solutes, with the additional requirement that: natural degradation can be described by first-order kinetics and the sorption isotherm is linear. Solutions for continuous and pulse releases are provided for one-, two-, and three-dimensional geometries, with line sources being considered in 2D geometries, and point and plain sources being hypothesized in 3D geometries.

The previous Chapters described the transport of miscible compounds in aquifers. Here, mechanisms of immiscible compound (also called non-aqueous phase liquid, NAPL) transport and propagation is presented. To do so, relevant properties of a multi-phase system (i.e., wettability, interfacial tension and capillary pressure, effective and relative permeability, drainage and imbibition) are illustrated. Migration of NAPLs in the subsurface is then qualitatively described, separately discussing the behavior of light- and dense-NAPLs. Light-NAPLs spilled at the surface that reach the saturated zone tend to accumulate and float on the water table and spread horizontally, progressively releasing their soluble fraction into the groundwater. Dense-NAPLs are denser than water, so if they reach the saturated zone they tend to displace groundwater from the pores, penetrating vertically in the aquifer. Depending on the volume of the release, Dense-NAPL contamination may affect the entire saturated thickness, and the compound may move along base of the aquifer. As for light-NAPL, also dense NAPLs present as a pure phase act as a continuous source of contaminant by releasing their soluble fraction into the groundwater. Finally, a quantitative approach to immiscible compound transport is provided, allowing for the estimation of mass distribution in the different phases in the saturated and unsaturated zones.

This chapter provides a methodological approach for the characterization of a contamination event. This includes an examination of both the unsaturated (i.e., soil, soil gas and pore water) and the saturated media (i.e., soil and groundwater), and is structured around three main phases, i.e., collection and organization of existing data, development of a conceptual model, verification of the hypotheses made in the conceptual model through targeted investigations and sampling. After illustrating different strategies available for defining the sampling design, sampling techniques for the different phases of the unsaturated and saturated media are described. In the unsaturated medium, soil sampling can be carried out through rotary or direct push techniques; active and passive sampling methods are available for the collection of soil gas samples; lysimeters or filter-tip samplers can be used for sampling pore water. Sampling of the saturated medium should allow to obtain a three-dimensional reconstruction of the contaminated areas. Hence, recommendations on the spatial distribution of monitoring wells, on the available options for vertical sampling and on well-purging prior to sampling are provided. These aspects are fundamental for ensuring the collection of representative samples. Subsequently, the most important aspects that need to be kept into account when planning a sampling campaign are illustrated, in particular as regards sampling rate, sample collection method, sampling devices (e.g., bailers, pumps). On site measurement of water quality parameters is also considered, and the possibility of filtering samples during collection is discussed. Quality assurance and control protocols aimed at ensuring accuracy, precision and defensibility of acquired data are then illustrated. Finally, a brief overview of sample storage, blank collection and sampling materials is provided.

The focus of this chapter is human health risk assessment, which quantifies the human or environmental toxicological effects deriving from the release of a contaminant at a source and its migration towards exposed receptors. Essentially, this entails a quantitative description of the relations in the system “source—pathway—receptor”. The procedure of risk assessment consists in a sequence of steps, starting from site assessment investigations, through the definition of a conceptual model (i.e., identification of potential receptors and migration and exposure pathways, selection of constituents of concern), the determination of concentrations at the point of exposure, actual risk calculation, to a risk management decision making stage (i.e., uncertainty assessment, risk acceptability evaluation, determination of the maximum acceptable concentration levels at the source and the selection of appropriate interventions). The risk assessment itself can be carried out at an increasing degree of detail, through a tiered approach, illustrated in the chapter. A relevant focus of this chapter is the calculation of the concentration at the point of exposure via the determination of the natural attenuation factor. This factor is the cumulative result of the contaminant concentration attenuation in the course of its migration from the source to the point of exposure (e.g., partitioning between environmental components, attenuation in the unsaturated medium, dilution in the aquifer or in rivers, volatilization). Having determined the concentration at the point of exposure, the calculation of the rate of exposure is presented. With these two parameters it is then possible to calculate the risk deriving the exposure to carcinogenic or threshold compounds, following a contamination event. The carcinogenic risk is quantified by the incremental lifetime cancer risk, which is a function of the slope factor (defined in Chap. 9); the non-carcinogenic risk, instead, is quantified by the hazard quotient, which is a function of the reference dose (also defined in Chap. 9). Once the risk has been calculated, its acceptability can be evaluated according to the local legislation, and measures to manage it can be put into place.

In order to decrease the health risk deriving from a contamination event, a number of cleanup and corrective actions, collectively called remediation, can be implemented. Remediation can be applied directly at the site of contamination (in situ) or off site (ex situ), in which case the contaminated environmental component is physically extracted and treated in dedicated facilities at the surface. There are three main remedial approaches, generally categorized as: containment, which aims at preventing the migration of the contamination and hence the exposure of sensitive targets; active restoration, which entails removing or treating the contamination; and natural attenuation, which relies on naturally occurring biological, chemical and physical degradation or transformation processes that convert contaminants into harmless compounds. This Chapter reviews the main containment and remedial strategies available for the management of a groundwater contamination event, and provides valuable information to support the choice of the most suitable approach. The presented strategies include: free product recovery for light non-aqueous phase liquid removal; vacuum enhanced extraction; subsurface containment; pump and treat; air- and bio-sparging; permeable reactive barriers; in situ flushing; in situ oxidation; in situ bioremediation. Applicability, design options and operating conditions, as well as advantages and drawbacks of the presented methods are illustrated.