Geophysical Methods Used To Guide Hydrogeological Investigation At An UMTRA Site Near Grand Junction, Colorado

Reference: Wightman, W. E., Martinek, B. C., and Hammermeister, D., "Geophysical Methods Used to Guide Hydrogeological Investigations at an Umtra Site Near Grand Junction, Colorado," Current Practices in Ground Water and Vadose Zone Investigations, ASTM STP 1118, David M. Nielden and Marin N. Sara, Eds,. American Society for testing and Materials, Philadelphia, 1992.

Abstract


Surface geophysical surveys played a key role in identifying a suitable hydrogeologic location for a Department of Energy Uranium Mill Tailings Remedial Action (UMTRA) disposal cell near Grand Junction, Colorado. Drilling and excavation work conducted as part of a first phase of site characterization activities had determined that a previously identified 35 hectare (90 acre) disposal cell location was underlain by shallow ground water (6 to 12 meters deep) suitable for domestic use. This shallow ground water was found to be confined to a complex system of alluvial-colluvial filled paleochannels eroded into the upper surface of the underlying Mancos shale bedrock. Geophysical methods were then used to cost-effectively direct a second phase of drilling and excavation activities which delineated an adjacent disposal cell location that was free of shallow ground water. Since this new disposal cell location was within the original designated site area of approximately 240 hectare (600 acres), considerable cost and time savings were realized by this surface geophysics directed field characterization program.

KEYWORDS: surface geophysics, electromagnetic conductivity, uranium mill tailings

Introduction

The Uranium Mill Tailings Remedial Action (UMTRA) program is a federal program administered by the Department of Energy that is designed to relocate uranium mill tailings in several western states to engineered, environmentally stable containment cells. These cells are to be constructed at a number of carefully selected sites in these states. The selected sites are required to meet specific geologic and hydrogeologic criteria that, together with a property designed containment cell, insure that the environmental impact of the tailings will be minimized. One of these criteria is that hydrogeologic conditions at the site are such that the containment cell can be constructed, and the tailings placed, above the water table.

The subject of this study is the 240 hectare (600 acre) Cheney Reservoir (Cheney) UMTRA site at which a containment cell is to be constructed for uranium mill tailings from the Grand Junction,

Colorado area. The Cheney site is located approximately 24 kilometers (km) south of Grand Junction (Figure 1). Design plans for this site called for the construction of a 35 hectare (90 acre) containment cell at the location shown on Figure 1 (original cell location) which would extend to a depth of 5 to 10 meters. Once constructed, the cell would be filled with uranium tailings and subsequently capped. Initial site characterization, which consisted of drilling and trenching at this original cell location, indicated that ground water was not present to depths of 15 meters and hence engineering designs for the cell were prepared. However, during the drilling conducted as part of engineering design, shallow ground water was unexpectedly encountered in some parts of this original cell location at depths of 5 to 10 meters.

The discovery of shallow ground water within the cell area could, depending on its extent, preclude the cell, or the entire Cheney site, from meeting Department of Energy siting or design criteria. The cost-related implications of this discovery were significant and ranged from the requirement for cell design modifications to abandoning the Cheney site altogether. The costs of these options ran from several hundred thousand to several million dollars. In addition, the tight schedule for building the cell, and existing commitments for contractors and equipment, meant that delays would also be costly. It was therefore imperative to quickly, yet confidently, identify the extent of the occurrence of the shallow ground water both in the original cell location, and over the remainder of the Cheney site if the original cell location proved unsuitable. It was also apparent that this could not be done over such a large area quickly enough and in sufficient detail solely by drilling.

Based on these considerations and favorable site conditions, geophysical methods were selected to evaluate the extent of the shallow ground water in the original cell location. Later, the survey was expanded over the remainder of the 240 hectare site in an attempt to identify a more suitable location for the construction of the cell. The geology of the Cheney site, the geophysical method selection, and the results and interpretation of these two surveys are discussed below.
  Fig. 1.  Site Location Map

Physical Setting and Geology


The Cheney site is located at an elevation of about 1,550 meters and is about 10 km west of the Grand Mesa, which is at an elevation of about 3,400 meters. The site generally slopes gently to the west with a uniform gradient. Its northern boundary is marked by a gentle valley and the southern boundary by a sharp valley with 5 to 10 meters of relief. Both of these valleys contain stream beds that are dry for much of the year. Rainfall in the area is less than 25 centimeters (cm) per year and vegetation comprises desert grasses and shrubs.

The geology of the site consists of up to about 15 meters of alluvium and colluvium resting on the Mancos shale. The alluvium and colluvium consists mostly of mixtures of silty gravel with cobbles and boulders derived from the basalt rock that caps Grand Mesa several kilometers cast of the site [1]. Silty surface soils, derived from the erosion of the Mancos and Mesa Verde formations, have developed on the gently sloping surface pediment. Underlying the surface soils are a chaotic mixture of particle sizes ranging from graded gravel and sand to clay. Basalt boulders as large as 1 meter in diameter were encountered in test pits and trenches. Locally, there are cemented zones of caliche directly overlying perched, or formerly perched, water-bearing zones. Thick gypsiferous deposits also overlie some of these same shallow perched or formerly perched zones in the middle and upper portions of the overburden.

The Mancos shale bedrock consists of finely bedded shale to depths of approximately 15 to 25 meters [1]. Below these depths, the shaley bedding appears only intermittently, with most of the formation appearing as massive claystone. The degree of weathering varies from severely to slightly weathered in the upper 15 meters of bedrock. Stress relief fracturing and surfical erosion are determining factors in the depth of the weathered zone. In the weathered rock, paleochannels have been incised from 1.5 to 7.5 meters below adjacent bedrock surfaces. Debris flows have filled most of these paleochannels. Fractures in the weathered zone are commonly filled with gypsum, and iron and manganese stains are prominent on partings. In the underlying unweathered bedrock, most fractures arc unfilled and unstained, although some joints may have clay or calcite filling.

Ground water is typically encountered at a depth of about 215 meters in the area, in the Dakota Sandstone Formation underlying the Mancos shale. The shallow perched ground water encountered during the engineering design phase drilling was observed at depths of as little as 5 meters. Information on the occurrence of this ground water was, prior to the commencement of geophysical surveys at the site, limited to that obtained from about 35 drillholes and three trenches that had been excavated to the top of the shale. The trenches were about 150 meters long and 10 meters wide. The perched water was observed directly above the Mancos shale in alluvial and colluvial overburden that filled topographic lows on the surface of the shale. Thus, the occurrence of the water appeared to be controlled by the surface configuration of the Mancos shale, probably occurring in channels on top of the shale. Apart from these localized occurrences of water, the overburden and shale were visually dry.

Geophysical Method Selection

Preliminary tests at the site were conducted using seismic techniques along with surface conductivity measurements and in situ measurements of the resistivity of the alluvium and the unsaturated and saturated shale. Based on these results, along with site conditions, electromagnetic conductivity methods using a EM34 instrument (EM34) were tentatively selected to delineate areas of shallow ground water at the site. Since the electrical conductivity of a material is strongly influenced by the presence of water, the selection of this method was based on the hypothesis that the conductivity of wet shale/alluvium would be higher than dry shale/alluvium and, moreover, that there would be sufficient conductivity contrast between the wet and dry areas to discriminate between them in the conductivity data.

Electromagnetic conductivity instruments, such as the EM34, measure the conductivity of a volume of the earth using electromagnetic waves [31,[41. The conductivity measured is influenced by the composition and porosity of the soil and the conductivity of the fluids within the soil pores [2] as well as the degree of saturation. The EM34 consists of separate hand-held transmitter and receiver coils and power source. During the operation of the EM34, the transmitter coil is energized by a low frequency alternating current that radiates an electromagnetic field into the earth. This primary field induces eddy currents in the ground below the instrument. The receiver coil on the EM34 detects both the primary field and the secondary magnetic field resulting from the eddy currents. The ratio between the primary field and the quadrature (out-of-phase with the primary field) component of the secondary field is converted to conductivity which is displayed by the EM34. This reading is a bulk measurement of the conductivity beneath the instrument and is made up of the cumulative response to subsurface conditions extending from the surface to the effective depth of penetration of the instrument. More specific details of the instrument response can be found in Technical Note TN-6 [3].

To test the hypothesis that shallow ground water is a cause of an increase in conductivity that can be detected from surface measurements, conductivity readings were taken along a traverse beside one of the trenches excavated during the engineering design phase work. Tests were conducted using both the horizontal and vertical dipole modes and with spacings between the transmitter and receiver coils of 10, 20, and 40 meters (see 'Data Collection'). Ground water occurred in this trench within a topographic low on the surface of the shale. The depth to the wet shale, which occurred near the center of the trench, was approximately 12 meters. To either side of this, the shale became shallower and dryer. The water in the trench was about 1 meter deep.

Contrary to what was expected, there was not an increase in conductivity over the wet shale. Instead, there were low conductivity values where the shale was relatively deep regardless of whether it was wet or dry, and higher conductivity values where the shale was shallow. These results indicated that the shale, even where visually dry, was significantly more conductive than the overburden and that the occurrence of the shallow ground water had an insignificant effect on the bulk conductivity. This relationship is shown schematically on Figure 2. The conductivity values shown in Figure 2 are representative of those found at the site. The reason for the unexpected results is due to the relatively high conductivity of the shale, presumed to result from the significant amounts of gypsum which could be observed. Under these conditions, only very small amounts of moisture in the rock pores are needed to produce high conductivity. Very little change in conductivity occurs as the rocks become completely saturated. With additional surface geophysical measurements, it was determined that this relationship was consistently observed across the site. Thus, while conductivity readings could not be used to detect the areas of shallow ground water directly, they could be used to map the topography of the shale.

Data Collection

The primary target for the geophysical survey was the location of channels on the surface of the shale in the region of the original cell location. Once these had been defined, and proved by drilling, the survey was extended over the remainder of the site to determine if a more suitable cell location could be identified.

Fig. 2.  Hypothetical relationship of conductivity to depth 
to top of shale

As part of the final design of the survey, the instrument configuration and line and station spacings were selected. The instrument configuration variables consist of coil orientation and spacing. Conductivity readings are made with the EM34 coils coplanar and oriented either with their planes parallel to the ground surface (vertical dipole mode) or at right angles to it (horizontal dipole mode). In addition, readings can be taken with three different spacings between the transmitter and receiver coils. Available coil spacings are 10, 20, and 40 meters. The frequencies of the transmitted electromagnetic waves at these coil spacings, which are automatically selected by the instrument, are 6.4 kilohertz (Khz), 1.6 Khz, and 0.4 Khz, respectively. Together, the variable orientation of the coils and the transmitter-receiver spacings result in a total of six different reading configurations. These coil configurations control both the depth, and volume, of the earth investigated and the response of the EM34 to lateral variations in conductivity [31, [4].

The optimum coil separation and operating mode for the field survey of the Cheney site were selected based on the results of several test lines. These test lines were run using the EM34 at both 10 and 20 meter coil spacing using both the horizontal and vertical dipole modes in areas with drillhole or trench control. The horizontal dipole mode provided much smoother data, and gave a larger amplitude response over the shale undulations, than the vertical dipole mode. Of the two coil spacings, the 20 meter spacing was chosen because the maximum exploration depth required over the site was uncertain and the greatest depth penetration could be achieved using the 20 meter spacing.

Using the horizontal dipole mode and 20 meter coil spacing, conductivity readings were taken at nearly 5,000 stations across the site along lines spaced about 60 meters apart (Figure 3). In the eastern area of the site, in the vicinity of the original cell location and where rapid changes in the conductivity were observed, the station spacing was 6 meters, while in the western area a 12 meters station spacing was sufficient.

Data Interpretation

General

A contour map of the conductivity values obtained over the whole site together with the original cell location is shown on Figure 4. A small regional trend of increasing conductivity values to the east has been removed from the contoured data using a first order trend surface. The relationship between conductivity and the depth to the Mancos shale that was observed during the preliminary testing was the basis for interpreting the contoured conductivity data. Areas of high conductivity occur where the shale is shallow and the overburden thin; areas of low conductivity occur where the shale is deep and the overburden thick. The arrows on the map show the major paths taken by surface water across the site, which usually occurs during summer thunderstorms and during spring snow melt. The higher conductivity in these channels results from topographic effects caused by stream valleys cutting into the overburden. Since the Mancos shale is nearer to the surface in these valleys, the measured conductivity is higher. The valley in the south of the area is the most deeply incised and hence has the highest conductivity values.


Fig. 3. Geophysical Survey Lines and Stations At Cheney Site

Paleochannel in Cell Area
The geophysical results in the cell area (Figure 4) show a conductivity high in the center of the cell and a southwesterly trending, linear, low conductivity feature to the west of this high. This conductivity low extends southwestward, beyond the western margin of the cell, until it is eclipsed by the high conductivity values in the southern strcam valley. The pattern of low conductivity values was interpreted to indicate a relatively large depth to the top of the shale reflecting a paleochannel in this area.

This interpretation is consistent with the results of the engineering design phase drilling which indicated that Mancos shale was at depths as great as 14 meters just south of the center of the western edge of the original containment cell. Additional drilling undertaken based on this geophysical interpretation confirmed that this is a topographic depression on the Mancos shale, and that this depression contains ground water to the east of the western boundary of the original cell. A section through the feature is shown on Figure 5 (the location of the section line, marked A A', is shown on Figure 4 and is approximately topographically flat). The conductivity high in the center of the cell results from the shale being at a depth of only about 3 meters, while in the depression the depth to the shale is as great as 15 meters. It was also confirmed with subsequent drilling and trenching over the entire cell area that this was the main feature controlling the occurrence and flow of ground water. This feature, along with the interpreted ground-water flow, is shown on Figure 6.

Conditions at the Remainder of the Site
As the geophysical survey was expanded to the remainder of the site following the confirmatory drilling described above, it was observed that the variability of the conductivity values to the west of the original cell location (Figure 4) were, on average, much lower than those in the cell area. This is shown on Figure 7 in which histograms of the conductivity values from the two regions called the West Area and East Area arc compared. A wide range of conductivity values, with two predominant peaks at about 25 millisiemens per meter (mS/m) and 55 mS/m is shown in the histogram for the East Area, whereas, as can be seen in the plot for the West Area, the conductivity values are predominantly concentrated between 25 and 30 mS/m with a much smaller peak at between 35 and 40 mS/m.

These differences in the conductivity values between the two areas are interpreted to result from two factors. These are the depth of the shale, the variability of which controls the amplitude variations in the EM34 conductivity measurements, and the shale conductivity, which is primarily a function of moisture content. Both of these factors influence the magnitude of the measured conductivity. Overburden conductivity variations were not found to be important since its conductivity is much lower than that of the shale (15 mS/m versus 150 mS/m). The lower average conductivity values in the West Area, along with their smaller amplitude variations, suggest that either the shale is deep or, if shallow, it has a relatively low conductivity and is comparatively dry. Both of these factors are favorable for the location of the containment cell.

In order to test these interpretations, a number of preliminary test pits and boreholes were excavated in the West Area to identify depth and moisture conditions for the shale. These test pits and borcholes indicated that the depth to the shale was relatively shallow and that the shale was comparatively dry. Based on these observations and the geophysical data, it was therefore concluded that the shale was comparatively dry and shallow throughout most of the West Area. The subsequent excavation of additional confirmatory test pits found no shallow ground water or significant palcochannels in the West Area. As a result of this, the location of the containment cell was changed to a new location on the dry shale areas, shown on Figure 6.


Fig. 4 Contoured EM34 Conductivity Data


Fig. 5 Measured EM34 Conductivity Data (a) and Depth To Top of Mancos Shale (b) Along Profile A-A' (see fig. 4 for line of section).


Fig.6 Contoured conductivity Data Showing Interpreted Ground-Water Flow Path Through Original Cell, and Proposed New Cell Location.


Fig. 7 Histograms of measured EM34 Conductivity Values in West and East Areas of the Cheney Site.


Conclusions

With the use of the geophysical survey, in conjunction with geologic studies and confirmatory drilling, it was possible to cost effectively define the occurrence of shallow ground water at the Cheney site and identify the location of an acceptable area for a tailings containment cell on-site. Hundreds of thousands of dollars to millions of dollars were savcd that might have otherwise been spent selecting an entirely new site or drastically modifying the design of the cell if constructed at the original location. These savings far outweighed the cost of the survey, which was approximately $50,000 USD.

Key components in the success of the project are both technical and organizational. Careful presurvey testing was necessary to develop the correct understanding of how the geophysical information could be used to investigate shallow ground water conditions at the site. Through the close interaction between hydrogeologists and geophysicists involved in the project it was possible for a geological-geophysical hypothesis to be established prior to conducting routine measurements. In addition, project managers were convinced of the importance of allocating time and money for the pre-survey testing. By integrating the geophysical survey with confirmatory drilling work, the confidence and accuracy of the geophysical interpretation was enhanced while unnecessary drilling was minimized. The resulting site characterization achieved by the conductivity survey was more comprehensive and detailed than could be achieved by drilling alone, and more accurate than could be obtained solely through the collection of geophysical data.

References

(1) United States Department of Energy, Remedial Action Plan and Site Design for Stabiliazation of hte Inactive Uranium Mill Tailings Site at Grand Junction, Colorado, Attachment 2: Geology Report, August 1990.

(2) Keller, G.V. and Frank C. Frischknecht, Electrical Methods in Geophysical Prospocting, International Series of Monographs in Electromagnetic Waves, Volume 10, Pergamon Press.

(3) McNeill, J.D., Electromagnetic Terrain Conductivity Measurements at Low Induction Numbers, Geonics Ltd., Technical Note TN-6, 1980, 15 pp.

(4) McNeill, J.D., Electrical Conductivity of Soils and Rocks, Geonics Ltd., Technical Note TN-5, 1980, 22 pp.