Aquifer Characterization Techniques

Aquifer characterization is broadly defined as processes by which the three-dimensional structure, hydraulic and transport properties, and chemistry of aquifers are evaluated. Aquifer characterization provides the foundation for groundwater modeling, which is ubiquitously used to evaluate sedimentary aquifers. Detailed aquifer characterization is particularly important where solute transport is a concern, as aquifer heterogeneity has a much greater impact on groundwater flow direction and rates than it does on aquifer heads. An introduction to aquifer hydraulic and transport parameters and basic aquifer heterogeneity concepts is provided. Aquifer heterogeneity includes layered, lateral, and multiple-porosity systems. Aquifer characterization starts with an initial conceptual geological model development, followed by evaluations of the type and scale of aquifer heterogeneity and the values of petrophysical and hydraulic parameters, and finally, data analysis and synthesis, and groundwater flow and solute-transport modeling.

The three-dimensional distribution of bodies of rock and sediments with different sedimentological properties and associated hydraulic properties is controlled to varying degrees by the depositional history of the strata of interest. Primary (depositional) variations in sediment textures and fabrics are modified by diagenetic processes, such as compaction, dissolution, and cement precipitation. A facies is a body of sedimentary rock with specified characteristics, which may include lithology (lithofacies), fossils (biofacies), and hydraulic properties (hydrofacies). Sedimentary facies analysis is based on the concept that facies transitions occur more commonly than would be expected if sedimentation processes were random. A facies model (or type model) is an idealized sequence of facies defined as a general summary of a specific sedimentary environment. Sequence stratigraphy is based on the concept that the sedimentary rock record can be divided into unconformity-bounded sequences, which reflect the sedimentological response to sea level changes, subsidence, and sediment supply. The value of facies analysis and sequence stratigraphy is that they can provide some predictability to the facies distribution between data points (i.e., wells). Where there is an underlying sedimentological control on the distribution of the hydraulic properties in aquifer systems, facies analysis can be used to better incorporate the underlying sedimentological fabric into groundwater models.

Siliciclastic aquifers are composed of sediment and rock that are dominated by silicate minerals, particularly quartz, feldspar, and clays. Siliciclastic aquifer properties are controlled by the grain size, sorting, and diagenesis of the sediments. Well-sorted sand and gravel facies deposited by flowing water and air tend to have the highest hydraulic conductivities and form aquifers, whereas low-energy clay-rich facies form confining and semiconfining strata. Facies models are provided for fluvial, alluvial fan, delta, eolian, glacial, and linear terrigenous shoreline (beach and barrier) depositional systems. Very large (multiple orders of magnitude) variations in hydraulic conductivity occur on multiple scales. A key issue for aquifer characterization is the connectivity and orientation of both clean sandy aquifer strata and clay-rich confining strata, which varies between depositional facies.

Carbonate aquifers consist of rocks composed mainly of the minerals calcite and dolomite. Carbonate minerals are generally much more chemically reactive under near surface geochemical conditions and thus undergo a much greater degree of chemical and physical alteration (diagenesis) than siliciclastic deposits. The textures and fabrics of carbonate sediments are strongly controlled by physical, chemical, and biological conditions in their depositional environment. The petrophysical properties of relatively young (Cenozoic) carbonates often still reflect depositional heterogeneities. In most older (Mesozoic and Paleozoic) carbonates, much of the depositional porosity and permeability has been lost or profoundly modified by physical and chemical diagenesis. Groundwater flow in older carbonates is largely controlled by secondary porosity, particularly fractures and solution conduits.

Aquifer characterization programs need to be designed to provide the specific data required for groundwater resources projects, which commonly involves the development of conceptual and numerical groundwater models. Available characterization techniques are compiled in terms of the type of information provided and scale (investigated volume or radius of influence). Important considerations in the selection and implementation of characterization techniques are the investigated volumes and resolution of the technique, scale of aquifer heterogeneity, scale at which data will be used (e.g., model grid cell size), and whether or not solute transport is of concern. Selection of techniques should also consider the scale dependence of hydraulic conductivity values. Aquifer characterization techniques have underlying assumptions and limitations, and field conditions requirements that constrain their successful implementation.

Well construction and drilling techniques are reviewed focusing on the appropriateness of different methods for various aquifer characterization scenarios. The drilling method and well construction selected for an aquifer characterization program should be based on consideration of the type of testing to be performed, borehole conditions required for testing, formation conditions (borehole stability), formation sample requirements, well depth, water quality, planned use of the well after the completion of the program, local driller capabilities and expertise, drilling fluid disposal, local regulations, and cost. A key factor dictating drilling and formation sampling methods is whether or not a borehole is stable (i.e., penetrated strata are lithified or sufficiently cohesive). Wells should be designed, constructed, and developed to maximize well efficiency, particularly if they are to be used as production wells or for single-well hydraulic testing (e.g., slug testing).

Aquifer pumping tests are an integral component of aquifer characterization because they provide quantitative data on large-scale aquifer hydraulic properties such as aquifer transmissivity, storativity, and the vertical hydraulic conductivity (leakance) of confining strata. Aquifer performance tests (APTs), also referred to as aquifer pumping tests, involve the pumping of an aquifer at a known rate and the measurement of the corresponding changes in water levels in the pumped and observation wells. Aquifer hydraulic testing may also be performed by injecting water into a well. APTs can provide important information on well yields, well efficiency, and the stability of water quality. Successful aquifer hydraulic testing requires attention to detail and careful consideration of the underlying assumptions of the various methods used to interpret the data.

Slug tests are a commonly used method to determine the hydraulic conductivity of strata near a borehole. The tests involve recording the water level (pressure) response in a well to an instantaneous increase or lowering of water level. Slug tests have the advantage of being quick, inexpensive to perform, and do not generate water that requires disposal, which is an important consideration at contaminated sites. The quality of data obtained from slug tests is strongly dependent on well and borehole conditions, particularly skin effects. Multiple-level slug tests performed on a single well or borehole are used to obtain hydraulic conductivity-versus-depth profiles. Straddle-packer and single-packer tests allow for the evaluation of hydraulic properties and collection of water samples from discrete intervals. Pressure transient testing is an important tool in oil and gas industry that has applications in groundwater investigations for evaluation of aquifer properties and wellbore conditions.

Small-volume methods are, in essence, point measurements of the petrophysical and hydraulic properties and lithology of aquifer and confining strata. Small-volume methods have the advantage that they are often relatively simple and inexpensive to perform, but have the limitation that the results of individual analyses are usually not representative of the tested hydrogeological unit as a whole. They can provide valuable information on small-scale heterogeneity, which is important for understanding and predicting solute-transport. Hydraulic conductivity is determined by core analyses, minipermeameter testing, and using grain size data. Methods used to evaluate the lithology and mineralogy of aquifers include core and cutting descriptions, thin-section petrography, and x-ray diffractometry. Thin-section petrography, scanning electron microscopy, and mercury-injection porosimetry are used to evaluate pore type and size distribution.

Borehole geophysical logging is a fundamental element of aquifer characterization because it can provide essentially continuous in situ measurements of the petrophysical properties, lithology, location and types of secondary porosity, and pore-water quality (salinity) of the logged strata.

Surface and airborne geophysical methods can be a valuable element of aquifer characterization programs because they typically are less expensive and can be performed quicker than methods that require the drilling of boreholes, which allow a larger number of measurements and thus greater spatial coverage.

Direct-push technology (DPT) is a widely adopted, low-cost method for collecting groundwater samples from unconsolidated or semi-consolidated shallow aquifers without a need for permanent monitoring wells. DPT methods are also used to install permanent small-diameter monitoring wells. DPT has the great advantage of generating minimal investigation-derived waste, which may require expensive disposal at a regulated waste facility. Direct-push technologies are increasingly being used as part of aquifer characterization programs for the installation of observation wells used in aquifer pumping tests, slug testing, electrical conductivity (resistivity) profiling, and aquifer hydraulic conductivity profiling. Hydostratigraphic profiling procedures have been developed that efficiently incorporate a series of direct-push technologies.

Groundwater tracer tests involve the use of existing or introduced variations in water chemistry or properties to obtain information about groundwater flow rates and directions, aquifer hydraulic and transport properties, and fluid–rock interactions. Tracer tests vary greatly in their objectives and complexity. Natural or existing anthropogenic variations in water quality may be cost-effectively taken advantage of to provide information on groundwater flow direction and rates. Key elements of tracer tests are determination of type test (e.g., natural versus forced gradient; qualitative versus quantitative), selection of tracer(s), development of a monitoring program, and data analysis. Forward numerical modeling of tests is strongly recommended to evaluate if testing objective are feasible and the optimal testing program for meeting project goals, based on plausible site hydrogeological conditions.

Data on aquifer storage properties (storativity and specific yield) are required for transient groundwater models. Storativity is usually determined from aquifer pumping tests using the Theis method or variations thereof. Quantification of specific yield is much more challenging because of the long time required (especially in fine-grained sediments) for gravity drainage to occur to completion. Evaluation of the properties of aquitards (semi-confining units) may also be a key element of aquifer characterization and modeling investigations. Heterogeneity, particularly a strong scale effect, and very slow groundwater flow rates are the main challenges associated with aquitard characterization. Multiple methods should be employed to evaluate aquifer storage and aquitard properties with the values subject to adjustment during the model calibration process.

A wide variety of methods have been proposed or applied to groundwater investigations that are not yet widely utilized either because of limited applications, costs, or perhaps lack of widespread knowledge of their value and limitations. This chapter is a somewhat disjointed review of some specialized aquifer characterization techniques that did not fit well into the other chapters, but should still be considered in aquifer characterization investigations. Stoneley wave borehole geophysical analysis can provide information on the location of hydraulically active fractures. Advanced oil-field technologies, such as cross-well seismic tomography and wireline formation testers have potential specialized applications in groundwater investigations. Multilevel testing and monitoring systems are available that are less expensive alternatives to multiple-monitoring wells or multiple-zone wells. Remote sensing techniques have applications where aspects of subsurface hydrogeology are manifested at land surface.

Hydraulic conductivity is directly measured using Darcy’s law-based methods that induce flow through a formation or sample. Indirect measures predict hydraulic conductivity from other sediment or rock properties. Methods are reviewed for obtaining profiles of permeability (hydraulic conductivity) from borehole geophysical logs, which include use of core porosity- versus-permeability transforms, porosity pore-size-permeability relationships, multivariate methods using multiple logs, and artificial neural networks. Obtaining hydraulic conductivity from geophysical logs is more complex for carbonates because of the presence of multiple pore types and the often dominance of flow by secondary porosity. Model grid cells are often one or more orders of magnitude greater than the volume of investigation of geophysical logs and other small-scale aquifer characterization methods. Upscaling is the process of assigning single equivalent values for each aquifer parameters (e.g., hydraulic conductivity) in model grid cells that results in the same modeled flow and solute transport as the original finer-scale heterogeneous values.

Fractured sedimentary rock contains two domains, the fractures and adjoining rock matrix. Fractures often provide most of the aquifer transmissivity, whereas the bulk of water and solute storage may occur in the matrix. The concentration of flow in fractures, which constitutes a very minor part of the total volume of the strata, results in greater flow velocities and travel distances than would occur in single-porosity systems. Characterization of fractured aquifers typically involves a multiple-method approach that includes identification of fractures, determination of whether or not identified fractures are hydraulically active, and determination of the hydraulic properties of the fractures and fractured zones. Fractured rock aquifers may be modeled using either a single-continuum (equivalent porous media), dual-continuum, or discrete fracture network approach.

Karst aquifer systems are characterized by often extreme heterogeneity as flow is dominated by secondary porosity, which includes fractures and solution conduits of multiple scales. Karst aquifers cannot be fully characterized using conventional hydrogeological methods alone, such as potentiometric surface mapping and well pump testing. A basic limitation of borehole-based methods in karst aquifers is that boreholes usually do not intersect flow-dominating conduits. Greater emphasis needs to be placed on identification of subsurface flow paths, hydrologic boundaries, recharge sources, and distribution and properties of flow conduits. Karst systems are studied using techniques that have large volumes of investigation, such as tracer tests, rainfall-runoff relationships (e.g., spring hydrograph analysis), water balance analysis, and input and output chemical data analyses, in addition to general (non-karstic) aquifer characterization techniques.

Aquifer characterization programs are usually performed with the objective of obtaining the data required to develop numerical groundwater models. Groundwater modeling starts with the development of a conceptual model, which is followed by the selection of a modeling code and model discretization. Initial values for the hydraulic and transport properties are then assigned to each model cell or element, which are subject to adjustment during the model calibration process. Predictive simulations are performed to evaluate the response of the aquifer to various stresses (e.g., groundwater pumping scenarios). A deterministic approach has been taken for most groundwater models, in which the goal is to obtain a single solution that represents a ‘best’ estimate of future conditions. The alternative stochastic approach involves running a large number of simulations in a probabilistic framework to explore the range of possible future conditions. The basic premise of stochastic modeling is that due to an incomplete knowledge of the spatial variability of parameters, the decision is made to analyze all (or least numerous) plausible representations of the aquifer. Stochastic modeling has high data requirements and is not a substitute for a robust aquifer characterization program.

Geostatistical analysis techniques are used to predict the values of parameters between data points. Geostatistical methods are only valid for spatially dependent (i.e., nonrandom) data. The basic method is to first identify and quantify the spatial structure of the variables of concern and then to interpolate or estimate the values of variables from neighboring values taking into account their spatial structure. Conditioning is the incorporation of hard or soft data into a model to reduce uncertainty. Hard data, by definition, has negligible uncertainty (e.g., direct measurements of property of interest), whereas soft data (inferred properties) have significant uncertainty. Geostatistical methods have been used to obtain realizations of sedimentary facies distributions, which typically require upscaling to a groundwater model grid, and assignment of hydraulic parameters values. A promising approach is hybrid methodologies that combine facies models and other soft geological information with geostatistical methods. Geostatistical techniques, when properly applied, are data intensive and are not a substitute for detailed field investigations and hydrogeological knowledge.