U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the Technical Meeting, Colorado Springs, Colorado, September 20-24, 1993, Water-Resources Investigations Report 94-4015

Overview of Research on Use of Hydrologic, Geophysical, and Geochemical Methods to Characterize Flow and Chemical Transport in Fractured Rock at the Mirror Lake Site, New Hampshire

CONTENTS

• ABSTRACT
• INTRODUCTION
• SITE DESCRIPTION
• CURRENT RESEARCH
  • Local-Scale Investigations
  • Regional-Scale Investigations
• SUMMARY
• REFERENCES

Abstract

Research efforts at the Mirror Lake Toxic-Substances Hydrology Program research site in Grafton County, New Hampshire have focused on the development of equipment, field testing procedures, and interdisciplinary interpretive approaches of characterizing ground-water flow and chemical transport in fractured rock over distances that range from tens of meters to kilometers. This paper summarizes the results of current research in bedrock hydrogeology in local- and regional-scale investigations at the Mirror Lake site. Local-scale investigations are conducted over distances of tens of meters and focus on identification of (1) fractures and fracture properties on exposed surfaces, (2) fractures in the subsurface using borehole and surface geophysics, and (3) hydraulic and transport properties of fractures by means of hydrologic testing. In regional-scale investigations, controlled hydrologic testing cannot be conducted to identify hydraulic and transport properties of the bedrock. Regional-scale investigations have focused on (1) methods of collecting hydrologic and geochemical information in heterogeneous bedrock environments, and (2) interdisciplinary methods of synthesizing these data by ground-water-flow and chemical-transport modeling to infer hydraulic and transport properties of the bedrock.

INTRODUCTION

The U.S. Geological Survey (USGS) Mirror Lake Toxic-Substances Hydrology Program research site is located in Grafton County, at the southern extent of the White Mountains in central New Hampshire (fig. 1). The Mirror Lake site was selected for investigations in fractured rock as a part of a nationwide program to describe the fate of contaminants in subsurface environments. Although there are no known contaminants in ground water at the Mirror Lake site, the equipment, field techniques and interdisciplinary interpretive methods that are being developed in this study area are directly transferable to other fractured-rock sites where ground-water has been contaminated. Research at the Mirror Lake site is focused on characterizing ground-water flow and chemical transport in fractured rock over distances of tens of meters to kilometers. The purpose of this paper is to review the current research being conducted at the Mirror Lake site. An earlier summary of bedrock research activities at the Mirror Lake site is in Shapiro and Hsieh (1991).

SITE DESCRIPTION

Mirror Lake is located at the lower end of the Hubbard Brook valley in the southern part of the White Mountains of central New Hampshire (fig. 1). Hubbard Brook, which drains the valley, flows into the Pemigewasset River. The topography of the area generally is steep along mountain sides and flat along the Pemigewasset River. Although ground-water flow extends beyond the surface-water drainage basin associated with Mirror Lake, regional field investigations have concentrated on the Mirror Lake drainage basin. The Mirror Lake drainage basin occupies 85 ha. Land-surface elevations range from 213 m above sea level at the lake surface to 481 m at the top of the drainage divide. Three perennial streams flow into Mirror Lake, and Mirror Lake drains through an outlet stream into Hubbard Brook.

Figure 1. Location of the Mirror Lake watershed in New Hampshire and the location of bedrock wells in the vicinity of Mirror Lake. (33k)

Glacial drift overlies the bedrock in much of the Mirror Lake area. The drift thickness ranges from 0 to 55 m and consists mainly of silty and sandy till, with numerous cobbles and boulders. Kame terraces, consisting of fine- to coarse-grained sandy ice-contact deposits, are found at several levels on the mountain sides. In the area between Mirror Lake and Hubbard Brook, the silty sand and gravel form what is believed to be a delta-type deposit laid down when Hubbard Brook adjusted its grade to the Pemigewasset River during deglaciation. Mirror Lake is believed to be a kettle lake overlying a bedrock saddle, where bedrock valleys descend to the north and south. The ridge north of the lake is a moraine overlying the northern bedrock valley, where drift thickness is as much as 55 m.

The bedrock in the Mirror Lake area is a Silurian sillimanite-grade schist that has been extensively intruded by late Devonian granite (C.C. Barton, U.S. Geological Survey, written commun., 1993). The schist and granite are cut by dikes of pegmatite of unknown age, possibly a residual differentiate of the granitic intrusions. All three rocks are cut by lamprophyre dikes, whose age is believed to be middle Jurassic and early Cretaceous. At a road cut along Interstate Highway 93, directly east of Mirror Lake, the exposed bedrock consists of a complex suite of rock types. The granitic intrusions are present as dikes, irregular pods, and anastomosing fingers that range in width from centimeters to tens of meters. Pegmatite typically is in the form of dikes, centimeters to meters wide, that cross-cut both the schist and the granite. The lamprophyre dikes, having widths of centimeters to meters, are less common than the other three rock types in the Mirror Lake area.

All components of the hydrologic system in the Mirror Lake drainage basin are being monitored. Flow of all streams leading to Mirror Lake are measured with flumes having continuous recorders; precipitation is measured at two locations in the drainage basin; climatological information for energy budget and evaporation studies have been collected since 1983; water levels in bedrock wells, drift piezometers, and water-table wells are continuously recorded (Winter, 1984, 1985). Bedrock wells have been drilled in two clusters, referred to as the FSE and CO well fields (fig. 1). Bedrock wells also have been areally distributed in the vicinity of Mirror Lake. In early 1993, there were 13 bedrock wells in the FSE well field, 4 bedrock wells in the CO well field and 12 bedrock wells areally distributed in the Mirror Lake area. Depths of bedrock wells range from 60 to 230 m. At most bedrock-well sites, piezometers are installed at one or more locations in the saturated drift above the bedrock surface. Water-table wells, whose screens intersect the water table, have been installed throughout the watershed and at most bedrock-well sites.

Ground-water flow in the Mirror Lake area is typical of mountain-valley terranes of the New England uplands. Water from precipitation and snowmelt infiltrates into the subsurface, flows through the glacial drift and fractured bedrock, and discharges into streams, lakes and rivers. Estimates of recharge to ground water vary from 0.2 m/yr (meters per year) to 0.45 m/yr (T. C. Winter, U.S. Geological Survey, oral commun., 1993). The major features controlling ground-water flow in the Mirror Lake area are (1) the steep slope of the mountain sides, (2) the low hydraulic conductivity of the drift and the bedrock, and (3) the presence of surface-water bodies such as streams, Mirror Lake and the Pemigewasset River. In most of the Mirror Lake watershed, the water table is close to land surface, typically less than 10 m below land surface. The only exception is the moraine ridge north of Mirror Lake where depth to water table exceeds 20 m. The presence of a shallow water table suggests that local topography affects the local pattern of ground-water flow.

Hydraulic conductivities of the glacial drift and bedrock in the Mirror Lake area have been estimated from results of hydraulic tests. Based on slug tests in piezometers, the hydraulic conductivity of the till is estimated to range between 10^-6 m/s (meters per second) and 10^-5 m/s (Wilson, 1991). Hydraulic conductivity of the sands in the kame terraces is approximately 10 times higher than the till. The results of single-hole, straddle-packer tests using 5-m long test intervals indicate that the bedrock hydraulic conductivity is highly heterogeneous, varying over many orders of magnitude (Hsieh and Shapiro, 1996). The average hydraulic conductivity over the upper 100 m of the bedrock is 3x10^-7 m/s. Single-hole tests, however, sample the rock in a small region around the bedrock well and do not necessarily represent the hydraulic conductivity controlling regional flow through the bedrock.

CURRENT RESEARCH

Current research at the Mirror Lake site is focused on local- and regional-scale investigations of ground-water flow and chemical transport in fractured rock. The local-scale investigations are conducted over distances of tens of meters and focus on identifying (1) fractures and fracture properties on exposed surfaces, (2) fractures in the subsurface using borehole and surface geophysics, and (3) hydraulic and transport properties of fractures through hydrologic testing (Hsieh and others, 1993). When applied individually, these methods generally yield limited information on fracture characteristics. However, when applied together, these methods become powerful tools that enable an interdisciplinary approach to the study of ground-water flow and chemical transport in fractured rock. In regional-scale investigations, controlled hydrologic testing cannot be conducted to identify hydraulic and transport properties of the bedrock. Hydraulic and transport properties must be inferred from hydrologic, geologic and geochemical information. Regional-scale investigations at the Mirror Lake site have focused on (1) methods of collecting hydrologic and geochemical information in heterogeneous bedrock terranes, and (2) the synthesis of this information by means of ground-water-flow and chemical-transport modeling.

Local-Scale Investigations

The principle problem encountered in characterizing ground-water flow and chemical transport in fractured rock is the physical identification and mathematical characterization of the heterogeneous hydraulic properties of the rock. Fractures are the conduits for fluid movement in the subsurface; therefore, identification of their location, interconnectivity and hydraulic properties is paramount in the characterization of ground-water flow and chemical transport over distances of tens of meters. Hence, a major research effort at the Mirror Lake site is the investigation of geometrical properties of fractures. Understanding the geometric properties of fractures provides a basis for making inferences and predictions of flow and transport in the subsurface. Barton (1996) conducted an extensive investigation of fracture properties and rock types on four vertical highway road cuts and one horizontal glacial pavement. Rock types were delineated, and fractures with trace lengths greater than 1 m were mapped for orientation, surface roughness, aperture, trace length, connectivity and signs of mineralization. From this investigation, the granite was shown to be more heavily fractured and to have shorter and more planar fractures than schist, whereas fractures in schist have greater fracture roughness than those in the granite. In addition, the connectivity of fractures in the granite and schist was poor in comparison to studies of fracture connectivity in other areas of the country. This investigation indicates that fluid moves through highly tortuous paths in the bedrock.

Fracture mapping is conducted only on exposed surfaces. Research on characterizing fractures in the subsurface is being conducted by means of surface- and borehole-geophysical surveys. Geophysical surveys generally yield an inferred distribution of a specific rock-mass property from which the presence of fractures must be inferred. Anomalies in the rock-mass properties attributed to fractures, however, also could be the result of variations in lithology; therefore, the correlation of several geophysical techniques as well as geologic information and hydrologic testing must be used to characterize the geometry of fractures in the subsurface.

Lieblich and others (1991) and Haeni and others (1993, 1996) discuss results of surface-geophysical surveys using azimuthal-seismic refraction and azimuthal-DC resistivity to detect the orientation of steeply dipping fractures. These methods infer the strike of a dominant, saturated, steeply dipping set of fractures from the azimuthal variation in the seismic velocity and the apparent resistivity of the rock, respectively. Azimuthal seismic-refraction data collected near the CO well field indicate that the principle strike of fractures is approximately N. 22° E.--an orientation similar to that of the principle orientation of fractures mapped on the road cut east of Mirror Lake. The principle strike of fractures on the road cut is approximately N. 30° E. Results from azimuthal-DC resistivity using a square array were conducted at the same site and yielded a similar orientation for the principle strike of a steeply dipping set of fractures. An azimuthal-DC resistivity survey using a linear array also was conducted at the same site. This survey indicated a principle fracture strike of N. 352° E. The linear array requires a large current-electrode spacing that may have emphasized environmental interferences such as a buried telephone cable. Also, voltage-electrode spacing should be large relative to the fracture spacing--a condition that may not have been satisfied at this site. The interpretation of these results indicates that multiple geophysical methods must be used to obtain an overall interpretation of fracture orientation (Lieblich and others, 1991).

The results of surface-exposure mapping indicated the importance of identifying distributions of subsurface rock types and fractures in characterizing ground-water flow. Techniques for identifying fractures and rock types in bedrock wells for hydrologic investigations are discussed by Johnson (1996) and Paillet (1996). Johnson (1996) describes the use of a borehole-color television camera to identify rock types, fractures and other conditions in the borehole, including mineralization in and near fracture surfaces. The borehole camera reduces the need to core bedrock wells to identify the distribution of rock type. Paillet (1996) discusses a systematic approach to the characterization of fractures in boreholes using a combination of standard geophysical logging tools that identify the general structure and lithology of the rock, and advanced borehole geophysical techniques that identify fractures and their potential for conducting fluid. The acoustic televiewer log shows the location and orientation of fractures intersecting boreholes. A sensitive heat-pulse flowmeter measures the fluid velocity in the borehole. When used under ambient hydraulic conditions in the borehole, permeable fractures having different hydraulic heads in the borehole can be identified. When used in conjuction with pumping in the borehole, the location of the most permeable fractures can be identified.

Results from the heat-pulse flowmeter can also be used to infer fractures forming hydraulic connections between adjacent boreholes (Paillet, 1996). By pumping one bedrock well and measuring the fluid velocity in an adjacent bedrock well, the location of fractures responding to the hydraulic stress can be identified. Paillet and others (1992) interpreted the transient velocity response to identify hydraulic properties of fractures and propose conceptual models of fracture connectivity between boreholes

The acoustic televiewer log and borehole television camera can identify fractures at the face of the borehole wall including their strike and dip. The distance these fractures extend into the rock and their interconnectivity with other fractures cannot be identified by these borehole methods. A major research effort at the Mirror Lake site has been the evaluation of geophysical methods that image fractures away from boreholes. Three methods of characterizing fractures in the rock mass have been applied at the FSE and CO well fields at the Mirror Lake site: single-hole, directional electromagnetic (EM) imagining; EM tomography; and seismic tomography.

Haeni and others (1993) discuss results of using a single-hole, directional EM tool that measures reflections of electromagnetic wave energy radiated from the transmitter in the tool. Water-filled fractures in the vicinity of a borehole, and fractures intersecting the borehole can be inferred from the reflected signal. The results of using this investigative tool in the CO well field showed that the same fracture zone could be detected from different boreholes, and the lateral extent and orientation of fractures could be inferred.

Seismic and EM tomography also infer the location of fractures in the rock mass away from borehole walls. Unlike the single-hole directional tool, where the transmitter and receiver are located in the same borehole, EM tomography is conducted by radiating electromagnetic energy from one borehole and monitoring its reception at an adjacent borehole (Wright and others, 1996a,b). The transmitted energy is delayed by the propagation distance and by the different dielectric properties of the water and rock matrix. By collecting data of sufficient density at a variety of angles (between the source and receiver) over the length of the boreholes, the spatial distribution of dielectric properties in the rock volume between the boreholes can be inferred by an inversion algorithm. Wright and others (1996a) also discuss the results of seismic-tomography surveys conducted between pairs of bedrock wells in the FSE well field. Seismic tomography is based on principles similar to those of EM tomography; however, instead of using electromagnetic energy, an acoustic signal is generated in one borehole, and the first-arrival of the pressure wave is monitored in an adjacent borehole. A\x11spatial distribution of the seismic velocity in the volume of rock between the two wells can be inferred using an inversion algorithm; areas of low-seismic velocity imply zones of fracturing, and zones of high-seismic velocity imply intact rock matrix.

Wright and others (1996b) compare the seismic and EM tomograms with each other and with fractures identified from acoustic televiewer logs. Results of the seismic- and EM-tomography surveys identified the same general zones of fracturing in the rock mass in wells separated by 10 m in the FSE well field. However, the orientation of the fractures intersecting the borehole (as identified from the acoustic televiewer log) was not the same as that of the fracture zones inferred from the seismic and EM tomograms. Results of hydrologic testing confirm the orientation of the fracture zones as identified by the tomograms, implying that the orientation of permeable fractures intersecting boreholes is not necessarily the orientation of permeable zones in the rock mass.

Seismic and electromagnetic signals respond to both fractures and lithology; therefore, dielectric permittivity and seismic velocity may not correlate with hydraulic properties of the rock. Thus, tomography must be conducted in conjunction with hydrologic testing to identify permeable structures in the bedrock. The use of hydrologic testing to characterize fracture locations and their hydraulic and transport properties is another major research effort being conducted at the Mirror Lake site. Hydrologic testing, consisting of hydraulic and tracer testing, is the most direct method of investigating flow and transport properties in the subsurface. Hydrologic tests generally involve artificially inducing a perturbation (a change in fluid pressure or chemical composition) in the subsurface and measuring the resulting response. The flow metering during pumping, as discussed by Paillet and others (1992) and Paillet (1996), are examples of hydraulic testing. Hsieh (1994), Hsieh and Shapiro (1996) and Shapiro (1996) discuss hydraulic and tracer tests conducted in the FSE well field to quantify hydraulic and transport properties of the bedrock.

Hsieh (1996) and Hsieh and Shapiro (1996) discuss hydraulic tests conducted by installing inflatable packers in all bedrock wells in the FSE well field to eliminate the artificial vertical permeability induced by open boreholes. An hydraulically isolated interval in one bedrock well was pumped continuously, and fluid pressure responses were monitored in hydraulically isolated intervals in other bedrock wells. The results of these tests indicated the bedrock does not respond as an equivalent homogeneous porous medium. Instead, there are several clusters of highly permeable fractures of limited areal extent. These highly permeable fracture clusters are only connected through less transmissive fractures.

Shapiro (1996) discusses radially converging tracer tests conducted in one of the highly permeable fracture clusters identified in the FSE well field. New equipment and field techniques of conducting tracer tests in fractured rock were developed for these tracer tests (Shapiro and Hsieh, 1996), the results of which showed that there is significant variability in rock properties that can affect chemical transport, even within a highly permeable fracture zone.

Other interpretive methods of identifying hydraulic and transport properties of the rock were developed to take advantage of environmental tracers. Goode and others (1993) used the natural variability in the radon content of water sampled during an aquifer test to estimate the effective porosity of the bedrock and the vertical leakage from the overburden. The temporal variability in radon concentration of water measured at the pumped well was attributed to the mixing of water sources; water in the glacial drift was assumed to have a different radon concentration than water in the bedrock.

Regional-Scale Investigations

On the scale of kilometers, hydrologic tests in fractured rocks become ineffective because, over such distances, changes in pressure caused by fluid injection or pumping are generally too small to measure, and transport times become too long for tracer tests to be practical. In order to investigate phenomena affecting flow and transport on the regional scale, the response of the ground-water system to natural perturbations and long-term human disturbances must be monitored. Monitoring hydrologic and geochemical phenomena in bedrock terranes requires the design of equipment and techniques that account for the heterogeneous hydraulic properties of bedrock.

Hsieh and others (1996) discuss methods of measuring hydraulic heads in multiple intervals in a single bedrock well. The monitoring equipment converts an open borehole to the equivalent of a multilevel piezometer. The equipment is removable, thereby allowing access to the bedrock well for hydrologic and geophysical testing when desired. The long-term installation of this equipment in bedrock wells also prevents the mixing of waters of differing chemistry because inflatable packers are used to hydraulically isolate hydraulically discrete intervals in the borehole.

Shapiro and others (1996) briefly discuss methods of collecting water samples for geochemical and isotopic analysis from bedrock wells. Because of the large variability of the hydraulic conductivity of fractures, water samples for geochemical and isotopic analyses cannot be collected by pumping an open borehole. A sample from an open borehole would yield water mixed from all permeable fractures intersecting the borehole. Instead, water samples are collected after isolating a single fracture or fracture zone in the borehole using inflatable packers.

A major research effort at the Mirror Lake site is the interpretation of geologic, hydrologic, and geochemical information to infer regional properties of ground-water flow and chemical transport in the bedrock. Rosenberry and Winter (1993) used 10 years of streamflow, lake-discharge, and climatological data to estimate the components of bedrock recharge and discharge in the water budget of Mirror Lake. Water from bedrock is estimated to account for up to 4 percent of all inflows to the lake and up to 1 percent of all outflows from the lake. These estimates are approximately 10 percent of ground-water inflows to and outflows from the lake; however, they are subject to great uncertainty because they are small relative to the major components of the water budget for the lake. These estimates of bedrock fluxes to and from the lake are being used to calibrate ground-water-flow models in the Mirror Lake drainage basin and its vicinity.

Harte and Winter (1996) used a numerical, cross-sectional-flow model of the drift and bedrock to investigate processes affecting bedrock recharge from the overlying glacial drift. Results of model simulations were compared with the distribution of the hydraulic head along an hypothesized path of ground-water flow running through the northwestern catchment area of the Mirror Lake drainage basin. The locations of bedrock recharge in the Mirror Lake area are attributed to lateral trends in the bedrock horizontal hydraulic conductivity, which is hypothesized to decrease with increasing elevations in the drainage basin to reproduce measured head gradients between the drift and bedrock.

Shapiro (1993) conducted a numerical simulation of areal (two-dimensional) regional ground-water flow in a heterogeneous bedrock terrane, where the heterogeneity was assumed to be the result of a random distribution of constant transmissivity blocks. The random distribution of transmissivity resulted in the poor connectivity of highly transmissive zones in the bedrock. The conceptual model was based on results of hydraulic testing in the FSE well field that illustrated the poor hydraulic connection between highly permeable fracture zones (Hsieh, 1996; Hsieh and Shapiro, 1996). The numerical simulations showed that hydraulic heads increased with increased elevation of upgradient locations in a manner similar to that observed in the Mirror Lake drainage basin. Thus, it is hypothesized that the poor connectivity of highly permeable zones in the bedrock could also play a role in the regional distribution of hydraulic heads.

A major research effort at the Mirror Lake site has been the analysis of ground-water chemistry and environmental isotopes to identify transport properties of the bedrock and processes affecting chemical transport in bedrock terranes over distances of kilometers. As water flows in the subsurface, its chemical composition evolves as the water reacts with the rock along the flow path. Understanding the chemical changes in ground water can help develop conceptual models of regional ground-water flow and chemical transport. In addition, environmental isotopes, such as tritium, and chlorofluorocarbons (CFC's) in ground water can be used to estimate the residence times of shallow ground-water.

Water samples were collected from discrete intervals in bedrock wells and analyzed for a variety of dissolved solids, dissolved gases, and a variety of stable and radioactive isotopes. Busenberg and Plummer (1996) measured the concentrations of CFC's in ground-water samples. By using historical records of CFC concentrations in the atmosphere, the date of ground-water recharge can be estimated. Drenkard and others (1996) estimated ground-water ages on the basis of the ratio of the parent-daughter isotopes tritium and helium. Many ground-water samples contained anomalously high helium concentrations; however, plausible corrections could be made for other sources of helium, and ground-water ages were estimated. The ages estimated from the tritium-helium and CFC analyses were similar. However, these ages do not correspond well with the presumed date of the maximum atmospheric tritium concentration. Both the tritium-helium and CFC ages appear to be approximately 6 years younger than ground-water ages estimated by using the maximum tritium concentration as the date of the maximum atmospheric tritium input. This difference is still being investigated and may be the result of ground-water flow through the unsaturated zone where the air phase can allow further equilibration with atmospheric conditions, or it may be a result of the mixing of waters of various ages, either naturally or through the collection of water samples.

Ground-water ages indicate the residence time of the water since it was recharged; however, the distance the water has traveled must also be identified to estimate the regional flow velocity. In heterogeneous bedrock terranes, the distance the water has traveled is not readily identifiable. Shapiro and others (1996) proposed a simple conceptual model to use changes in the concentration of the bicarbonate ion as a second measure of the ground-water-residence time in the bedrock, and as a means of estimating the average ground-water velocity in the bedrock. The model considers transport through a single fracture and diffusion of bicarbonate ions into the fracture from the rock matrix. When used in conjunction with the ground-water ages estimated from CFC or tritium-helium concentrations, an average ground-water velocity can be estimated to reproduce the measured residence time and bicarbonate concentration at sampling locations.

The basis of the conceptual model proposed by Shapiro and others (1996) stems from discussions by Wood and others (1996) regarding the significance of diffusion in defining the ground-water chemistry. Wood and others (1996) noted that residence times of less than 50 years are not sufficient for dissolution of minerals in the granite and schist along fracture faces to produce the concentration of dissolved solids in the water samples collected in the bedrock. Instead, it is hypothesized that water in the void space of the unfractured granite and schist is in chemical equilibrium with the associated minerals in the granite and schist. In particular, small amounts of calcite (1 percent by weight) have been found in samples of the granite. Thus, the equilibrium bicarbonate concentration in the rock matrix is significantly higher than the concentration in the water recharged through the overburden and into the bedrock. The concentration gradient between water in the rock matrix and water in the fractures results in the mass flux of bicarbonate ions into fractures.

Wood and others (1996) conducted laboratory experiments to identify the effective diffusion coefficient and porosity of the intact rock matrix to verify the role of diffusion from the rock matrix. Porosity of granite samples was approximately 1.5 percent, and the effective diffusion coefficient of ¹³⁷Cs in granite (including the effects of retardation, tortuosity and porosity) was approximately 6 X 10^-13 m²/s. In addition, iron-staining in the rock matrix adjacent to fractures has been noted on outcrops and cores, and this is a further indicator of the effects of diffusion in the ground-water chemistry (Barton, 1996; Wood and others, 1996). Preliminary calculations of precipitation rates for ferric hydroxide using the laboratory-estimated porosity and diffusion coefficients indicate that ferric hydroxide bands can form over tens of years as the result of the diffusion of dissolved oxygen into the rock matrix. These timeframes are consistent with laboratory diffusion experiments and the variability of bicarbonate concentration in the ground water in the bedrock, which is assumed to be the result of diffusion from the rock matrix.

SUMMARY

Research at the Mirror Lake Toxic-Substances Hydrology Research site has focused on characterizing the physical properties that can define ground-water flow and chemical transport in fractured rock over distances that range from meters to kilometers. The characterization of ground-water flow and chemical transport over tens of meters has focused on identifying the geometrical and hydraulic properties of fractures. Mapping of surface exposures, use of surface and borehole geophysical techniques, and hydrologic testing have been used to identify properties of fractures that influence ground-water flow and chemical transport. Over distances associated with regional ground-water flow, processes affecting ground-water movement and chemical transport are inferred through the synthesis of hydrologic and geochemical data by means of ground-water and chemical-transport modeling.

REFERENCES

Barton, C.C., 1996,

Characterizing bed rock fractures in outcrop for ground-water hydrology studies-An example from Mirror Lake, Grafton County, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Busenberg, E., and Plummer, L.N., 1996,

Concentrations of chlorofluorocarbons and other gases in ground waters at Mirror Lake, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Drenkard, S., Torgersen, T., Weppernig, R., Farley, K., Schlosser, P., Michel, R., L., Shapiro, A.M., and Wood, W.W., 1996,

Helium isotope analysis and tritium-helium age dating in the Mirror Lake basin, Grafton County, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Goode, D.J., Hsieh, P.A., Shapiro, A.M., Wood, W.W., and Kraemer, T.F., 1993,

Concentration history during pumping from a leaky aquifer with stratified initial concentration, in Shen, H.W., ed., Hydraulic Engineering, Proceedings of 1993 National Conference, July 25-30, 1993, San Francisco: New York, American Society of Civil Engineers.

Haeni, F.P., Lane, J.W., Lieblich, D.A., 1993,

Use of surface-geophysical and borehole-radar methods to detect fractures in crystalline rocks, Mirror Lake area, Grafton County, New Hampshire, in Hydrogeology of Hard Rocks, 24th Congress of the International Association of Hydrogeologists, June 28 - July 2, 1993, Oslo, Norway, p. 577-587.

Haeni, F.P., Lane, J.W., Lieblich, D.L., and Barton, C.C., 1996,

Using surface geophysical and bedrock outcrop data to identify fractures near Mirror Lake, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Harte, P.T., and Winter, T.C., 1996,

Factors affecting recharge to crystalline rock in the Mirror Lake are, Grafton County, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Hsieh, P.A., 1996,

An overview of field investigations of fluid flow in fractured crystalline rocks on the scale of hundreds of meters, in Stevens, P.R., and Nicholson, T.J., eds., Joint U.S. Geological Survey, Nuclear Regulatory Commission Workshop on Research Related to Low-Level Radioactive Waste Disposal, May 4-6, 1993, National Center, Reston, Virginia, Proceedings: U.S. Geological Survey Water- Resources Investigations Report 95-4015.

Hsieh, P.A., Perkins, R.L., and Rosenberry, D.O., 1996,

Field instrumentation for multi-level monitoring of hydraulic head in fractured bedrock at the Mirror Lake site, Grafton County, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Hsieh, P.A., and Shapiro, A.M., 1996,

Hydraulic characteristics of fractured bedrock underlying the FSE well field at the Mirror lake site, Grafton County, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program-Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Hsieh, P.A., Shapiro, A.M., Barton, C.C., Haeni, F.P., Johnson, C.D., Martin, W.W., Paillet, F.L., Winter, T.C., Wright, D.L., 1993,

Methods of characterizing fluid movement and chemical transport in fractured rock, in Cheney, J.T., and Hepburn, J.C., Field Trip Guidebook for the Northeastern United States; 1993 Boston GSA, Contribution No. 19, Department of Geology and Geography, University of Massachusetts, Amherst, Mass., p. R1-30.

Johnson, C.D., 1996,

Use of a color borehole video camera to identify lithologies, fractures and borehole conditions in bedrock wells in the Mirror Lake area, Grafton County, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Lieblich, D.A., Lane, J.W., Haeni, F.P., 1991,

Results of integrated surface-geophysical studies for shallow subsurface fracture detection at three new Hampshire sites, in Expanded Abstracts with Biographies, 61st Annual International Meeting, November 10-14, 1991: Society of Exploration Geophysicists, Houston, Texas, p. 553-556.

Paillet, F.L., 1996,

Using well logs to prepare the way for packer strings and tracer tests-- Lessons from the Mirror Lake study, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Paillet, F.L., Novakowski, K., and Lapcevic, P., 1992,

Analysis of transient flows in boreholes during pumping in fractured formations, in Society of Professional Well Log Analysts Annual Logging Symposium, 33rd, Oklahoma City, Oklahoma, p. S1-21.

Rosenberry, D.O., and Winter, T.C., 1993,

The significance of fracture flow to the water balance of a lake in fractured crystalline rock terrain, in Hydrogeology of Hard Rocks, 24th Congress of the International Association of Hydrogeologists, June 28 - July 2, 1993, Oslo, Norway, p. 967-977.

Shapiro, A.M., 1993,

The influence of heterogeneity in regional hydraulic properties of crystalline rock, in Hydrogeology of Hard Rocks, 24th Congress of the International Association of Hydrogeologists, June 28-July 2, 1993, Oslo, Norway, p. 125-136.

Shapiro, A.M., 1996,

Estimating effective porosity in fractured crystalline rock using controlled-tracer tests, in Steven, P.R., and Nicholson, T.J., eds., Joint U.S. Geological Survey, Nuclear Regulatory Commission Workshop on Research Related to Low-Level Radioactive Waste Disposal, May 4-6, 1993, National Center, Reston, Virginia, Proceedings: U.S. Geological Survey Water- Resources Investigations Report 95-4015.

Shapiro, A.M., and Hsieh, P.A., 1991,

Research in fractured rock hydrogeology-Characterizing fluid movement and chemical transport in fractured rock at the Mirror Lake drainage basin, New Hampshire, in Mallard, G.E., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the Technical Meeting, Monterey, California, March 11-15, 1991: U.S. Geological Survey Water-Resources Investigations Report 91-4034, p. 155-161.

Shapiro, A.M., and Hsieh, P.A., 1996,

A new method of performing controlled injection of traced fluid in fractured crystalline rock, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Shapiro, A.M., Wood, W.W., Busenberg, E., Drenkard, S., Plummer, L.N., Torgersen, T., and Schlosser, P., 1996,

A conceptual model for estimating regional ground-water velocity in bedrock of the Mirror Lake area, Grafton County, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program-Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Wilson, A., 1991,

Distribution of hydraulic conductivity in the glacial drift at Hubbard Brook Experimental Forest, West Thornton, New Hampshire, unpublished Senior Thesis, Department of Earth Sciences, Dartmouth College, Hanover, New Hampshire, 56 p.

Winter, T.C., 1984,

Geohydrologic setting of Mirror Lake, West Thornton, New Hampshire: U.S. Geological Survey Water-Resources Investigations Report 84-4266, 61 p.

Winter, T.C., 1985,

Physiographic setting and geologic origin of Mirror Lake, in Likens, G.E., ed., An Ecosystem Approach to Aquatic Ecology: New York, Springer-Verlag, p. 40-53.

Wood, W.W., Shapiro, A.M., and Hsieh, P.A., 1996,

Observational, experimental and inferred evidence of solute diffusion in granite rocks: Examples from the Mirror Lake watershed, Grafton County, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Wright, D.L., Olhoeft, G.R., and Grover, T.P., 1996a,

Velocity, amplitude and dispersion EM tomography in fractured rock, 1993, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.

Wright, D.L., Olhoeft, G.R., Hsieh, P.A., Majer, E.L., Paillet, F.L., and Lane, J.W., 1996b,

Electromagnetic and seismic tomography compared to borehole acoustic televiewer and flowmeter logs for subsurface fracture mapping at the Mirror Lake site, New Hampshire, in Morganwalp, D.W., and Aronson, D.A., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the technical meeting, Colorado Springs, Colo., September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations Report 94-4015.