December 2007

Helicopter survey aids remediation of produced water-contaminated aquifer

Airborne surveys mapped conductivity to help USGS researchers find areas of high salinity in and near East Poplar oil field in northeastern Montana.

Vol. 228 No. 12  


Helicopter survey aids remediation of produced water-contaminated aquifer

Airborne surveys mapped conductivity to help USGS researchers find areas of high salinity in and near East Poplar oil field in northeastern Montana.

David Michael Cohen, Production Engineering Editor

Beginning in 1976, the US Geological Survey (USGS) has assessed and mapped groundwater resources at the Fort Peck Indian Reservation. Results of these studies have demonstrated that groundwater quality has been adversely affected by various land-use practices in some areas of the reservation.1 Groundwater plumes of saline water were identified in 1997 based on water well sampling, ground geophysical surveys and borehole logs.2 USGS determined that these plumes were related to handling and disposal of brine produced with oil from East Poplar Field. In August 2004, a Helicopter ElectroMagnetic (HEM) survey was conducted over the oil field to better define possible subsurface plumes. The airborne EM measurements greatly expanded data collected from previous ground EM surveys and have helped USGS to understand the sources of anomalies and to define possible flow paths.


The first use of airborne electromagnetic survey to map contamination of shallow (less than 100 m deep) groundwater from saline produced water disposal was at oil fields near Brookhaven, Miss., in 1987.3 The survey successfully mapped pockets of shallow saline waters associated with produced water disposal ponds, some of which had not been documented before the airborne surveys. The EM data was used in one of the first conductivity-depth imaging algorithms-developed by Peter Sengpiel-which greatly enhanced the mapping of conductivity anomalies that were not readily seen in the apparent resistivity maps plotted at individual frequencies. A repeat survey in 1997 showed changes in salinity caused by new point sources, some of which were documented by new monitoring wells drilled in the area.4 The success of the 1987 survey led to other surveys in Texas funded by the Texas Railroad Commission and other state agencies to map saline waters in oil fields.5 These surveys successfully identified point sources of shallow groundwater contamination.

Though most produced waters associated with energy production are saline, water co-produced from CoalBed Methane (CBM) can have lower Total Dissolved Solids (TDS) than local groundwater does. The rapid rise in gas production from coalbeds has raised environmental issues concerning proper disposal methods for the produced water. US Department of Energy and USGS studies of produced waters from CBM development in Wyoming’s Powder River Basin have included HEM surveys of selected areas.6 In some of the areas, the airborne surveys have helped to document how disposal of CBM waters in ponds has led to dissolution of salts nearby. The result can be development of very high-TDS plumes.

By 2004, when the USGS contracted an HEM survey of part of Montana’s East Poplar Field, the analog system in the earlier surveys had been replaced by a broader-band digital system.


In the area of East Poplar Field, ground-based electromagnetic methods were first used during the early 1990s to delineate more than 12 sq mi of saline water contamination. An airborne EM survey was conducted during August 2004 in a 106-sq-mi area. The electromagnetic equipment consisted of six different coil-pair orientations that measured resistivity at separate frequencies from about 400 to 140,000 Hz. The electromagnetic resistivity data were converted to six electrical conductivity grids, each representing different approximate depths of investigation. The range of subsurface investigation is comparable to the depth of shallow aquifers.

Electrical induction conductivity and natural gamma logging was done during 1993, 2004 and 2005 in selected boreholes to aid in interpretation of the airborne geophysical survey and to characterize electrical parameters of the lithology and groundwater. Water quality samples were collected from wells during 2003-2005 to correlate geophysical measurements with the chemical composition of water from shallow aquifers. The airborne, ground and borehole conductivity data were used to delineate subsurface areas of high conductivity and were correlated with hydrologic data to indicate areas of contamination.

The USGS determined that handling and disposal of produced water in East Poplar Field has resulted in contamination of not only the shallow aquifers, but also the Poplar River. In the 10 years since the first delineation, the quality of water from wells completed in the field’s shallow aquifers changed markedly. The current extent of saline water plumes differs from that delineated in the early 1990s. The geophysical-hydrologic study is being used in groundwater resource planning studies for the area, to understand the hydrology and the contaminant’s extent and direction of movement.

Study area. The study area includes the city of Poplar, East Poplar Field and most of Northwest Poplar Field, Fig. 1. The Poplar River flows generally south, and shallow Quaternary deposits (up to 30 m thick) directly overlie the relatively thick (about 300 m) Upper Cretaceous Bearpaw Shale throughout most of the study area. These Quaternary deposits are the sole developed source of groundwater for local residents. Land uses in the study area include dry land farming, livestock ranching, oil production and residential development. Previous investigations on geologic structure, stratigraphy and hydrogeology in the study area are summarized by Thamke and Craigg.2

Fig. 1

Fig. 1. Location of the East Poplar Field study area. Courtesy of USGS.

Field hydrology and water quality. Water quality in the Quaternary deposits is highly variable and is dependent on location relative to sources of saline water. Saline water plumes were delineated using data collected during the early 1990s.1,2

Four principal water types in the study area were described by Thamke and Craigg, Table 1. The dissolved solids and chloride concentration ranges of these water types were updated by Thamke and Midtlyng using data collected between September 1993 and September 2000.1 The interpreted plume locations are based on hydrologic and ground geophysical data acquired in 1991-1992, Fig. 2.1,2

TABLE 1. Characteristics of water types in the East Poplar oil field study area1,2
Table 1


Fig. 2

Fig. 2. Interpreted location of saline water plumes in the study area, based on hydrologic and ground geophysical data acquired in 1991-1992. Courtesy of USGS.

Temporal changes in water quality-and thus, movement of saline water plumes in the study area-can be indicated by comparing early with more recent concentrations of dissolved solids and chloride in well water. In 2003, a consulting firm documented plume movement for the southern part of East Poplar Field, although a rate of groundwater movement estimated by the same company differs substantially from a rate estimated by the Montana Department of Environmental Quality.7,8 Ranges in groundwater quality can be substantial (Table 1), depending on the chemical characteristics and quantity of the saline water, proximity to the source, and groundwater flow characteristics.

Ground electromagnetic surveys. Thamke and Craigg determined that water-quality data from wells and surface water sites sampled from 1979 to 1997 confirm the presence of saline water contamination in the Quaternary deposits of the Poplar River drainage basin.2 The wells that serve as possible sample sites are sparsely located throughout the study area and do not provide sufficient information to define the lateral extent of plumes. Consequently, they conducted ground surveys using an EM meter to map variations in subsurface electrical conductivity. The entire survey area was covered using both vertical and horizontal coils. The northern part of the survey area and selected other portions used 10-m and 20-m separations. The southern part of the survey area and selected other portions used 20-m and 40-m separations. Generally the survey was done with 160-m and 320-m (0.1- and 0.2-mi) spaced lines and stations every 160-320 m, Fig. 3.

Fig. 3

Fig. 3. Apparent conductivity for ground surveys using an EM meter with 20-m loop spacing. The heavy black line is the Poplar River. Black dots indicate measurement stations for the 1991-1992 USGS survey. The boundary of the base map is the approximate extent of the 2004 airborne geophysical survey.2 Courtesy of USGS.

Another EM survey was conducted with a tractor-cart mobile system to extend the USGS survey to the south, Fig. 3. The lines were surveyed at 80-m spacing. Sampling rate and speed of the cart system resulted in samples about every 8 m along the lines. Data from the cart system constituted about 20,000 measurements, in contrast with the USGS survey, which measured fewer than 1,000 points.9 Also, the USGS survey was conducted over two summer field seasons, while the cart survey took about two weeks, including an additional metal detection survey not discussed here.

In the simplest case of saline water in coarse glacial sediments, variations in apparent conductivities can be directly related to the variations in the groundwater’s dissolved solids. However, silts and clays in the glacial sediments and numerous cultural effects (power lines and pipelines) are also sources of high conductivities, particularly in an oilfield environment.4,10 In East Poplar Field, the Bearpaw Shale underlies the glacial and alluvial sediments and is also a source of high conductivity response. Consequently, association of high conductivity with plumes requires integrated interpretation of both geologic mapping and hydrologic well data. This procedure was used in the compilation of Fig. 2.

The ground geophysical surveys provide more detail than the hydrologic data from wells, Figs. 2 and 3. However, high conductivity areas cannot by themselves be categorically attributed to high TDS or high chloride concentrations in the shallow groundwater. Each coil separation (and corresponding frequency) relates to a particular depth of investigation. The water and monitor wells sampled also reflect sampling from different depths (screened intervals). Water samples, however, represent groundwater at different depths depending on the screened intervals in the wells. Clay layers in the glacial deposits can increase the likelihood of perched water tables at different depths. A further complication in the subsurface distribution of electrically conductive, soluble salt (including chloride) is that the conductivity can be high even without saturated conditions. All of these factors need to be considered when interpreting the hydrologic and geologic data.


Fugro Airborne Surveys flew an HEM survey over part of the study area in early August 2004.11 The main objective was to extend the areal apparent conductivity mapping from the previous ground geophysical surveys. The main part of the survey area was covered with north-south flight lines flown with 200-m separation, Fig. 4. Fill-in lines were flown in two selected areas, giving an effective flight-line spacing of 100 m. HEM system sampling rate and flight speed yielded a sample about every 3 m along the flight lines. The nominal EM sensor flight height was 33 m.

Fig. 4

Fig. 4. Location of the helicopter electromagnetic survey of East Poplar Field. The hatched box is the main survey area, which was flown with 200-m-spaced north-south flight lines. Inset boxes show where 100-m fill-in lines were flown. Courtesy of USGS.

Additional instrumentation included a total field magnetometer mounted in the EM sensor, GPS navigation systems on both the helicopter and EM sensor system, and a laser altimeter on the sensor system.

Data processing included leveling of the EM flight-line measurements and reduction to apparent conductivities.11 In addition, differential depths and conductivities were computed.12 Grids were made of apparent conductivity for each EM frequency, total field magnetics and digital elevations, Table 2.

TABLE 2. Helicopter electromagnetic system specifications
Table 2


The highest actual frequency used in the HEM survey (132,640 Hz) has an average differential depth of about 1 m. Consequently, the apparent conductivity reflects the near-surface geology and hydrology, Fig. 5. Correlation of the near-surface geologic features mapped by Colton (not shown here)13,14 with the HEM maps is critical to interpreted separation of the hydrologic and geologic (lithologic) sources for conductivity variations. Much of the geomorphology and surface geology indicates the boundary of the last continental glaciation, which terminated just south of the survey area and influenced the present location of the Missouri River.

Fig. 5

Fig. 5. Apparent conductivity measurements from the HEM survey for nominal frequency of 140,000 Hz. Light black lines outline features shown in Fig. 2. Feature A is a paleo-stream meander, B is a shallow conductive area, C is a pipeline, D is an incised drainage exposing Bearpaw Shale, and E is an area of salinization. Courtesy of USGS.

In general, the terrain southeast of the Poplar River is composed of Pliocene sediments and Pleistocene glacial deposits that are composed of silts (lake sediments), sands and gravels having a total thickness up to 30 m. Low-conductivity areas are associated with Pleistocene alluvial gravels and sands. Low conductivity near feature A in Fig. 5 is likely related to a channel meander that has filled with sand and gravel. Slightly higher-conductivity areas (B, Fig. 5) can be due to either clays or higher-TDS groundwater within the unconsolidated sediments. The trend and the extent of high-TDS shallow groundwater in the area follow the trend of the HEM conductivity high. One of the surprising aspects of early studies was the extent of high chloride contamination in the central part of the field.15

Monitoring wells had to be used to supplement EM survey data in the center of the study area because a cathodically protected pipeline (the NW-SE-trending linear feature near feature C in Fig. 5) interfered with EM conductivity measurements. At the highest frequency (132,640 Hz), most cultural infrastructure in the field does not have as much effect as this pipeline is.

The area northwest of the Poplar drainage (such as feature D, Fig. 5) has thinner alluvial and colluvial deposits with more exposure of the underlying Bearpaw Shale. Here, the shale is exposed in a southeast-trending stream channel that has incised the alluvial deposits. The shale is another source of conductivity that makes it difficult to determine the position and extent of saline plumes. Logs were run in old test wells to confirm the bedrock.

Groundwater movement. One of the important interpretations of these USGS studies of the area’s saline waters is groundwater movement. This is particularly important since the shallow water is the sole source aquifer for many local residents as well as for the city of Poplar’s municipal water. Density controls saline water contamination movement more than freshwater movement. Thus, the subsurface topography of the contact between the shallow alluvial deposits and the shale may be expected to control flow or concentration of saline water. Recent, unpublished USGS work demonstrates the relationship between plume migration and the subsurface topography of the Bearpaw Shale-i.e., that the plumes migrate into lower areas.16

Drill logs from local wells have been used to contour the top of the Bearpaw Shale;7,15 HEM data provides greater resolution. Interpreting the HEM data includes complexities such as cultural noise, shallow conductors and data calibration. Specific calibration procedures required in the USGS contract were conducted by Fugro Airborne and are being used to refine leveling of the electrical data. Also, correlation with borehole induction electrical logs is being used to calibrate the airborne interpretation.

Effect of depth of investigation. The apparent conductivity at 1,800 Hz HEM (Fig. 6) has about the same depth of investigation as the ground EM meter used by Thamke and Craigg with 20-m loop separation and frequency of about 1,600 Hz.2 The average computed depth for this frequency from the differential calculation is 24 m.12 Saline water plumes outlined in Fig. 2 and apparent conductivity shown in Fig. 6 have about the same depth of investigation and can be compared. Many of the plumes delineated in the early 1990s (Fig. 2) are in areas that still had high apparent conductivity values at 1,800 Hz during 2004 (Fig. 6), indicating that there might be lasting effects from the saline water plumes. High apparent conductivity values at 140,000 Hz (Fig. 5) in these areas of delineated plumes indicate possible near-surface conductive sources.

Fig. 6

Fig. 6. Apparent conductivity measurements from the HEM survey for nominal frequency of 1,800 Hz. Courtesy of USGS.

One obvious difference in the ground and airborne surveys (in addition to detail and areal coverage) is feature A (Figs 2 and 5), which has a different apparent conductivity expression. It is likely that the HEM has a larger depth of investigation in this part of the study area and is influenced by the conductive Bearpaw Shale underlying the less-conductive alluvium. This area is topographically lower and is expressed in the topographic map as an embayment in the Poplar River valley.

Potential salinity sources. Several sources of shallow saline waters associated with the oilfield development can be postulated. Shallow sources can be attributed to failures in infrastructure, such as tanks and pipeline breaks. Also, older disposal practices of placing brine in evaporation or infiltration ponds can contribute to long-lasting high TDS and electrical conductivity anomalies. Geochemical and geophysical characteristics of such long-lasting features are described in papers on two Oklahoma oilfield sites.17

One source of produced water contamination was a failure in the Biere-1-22 producing well just southeast of point B in Fig. 6. The well had developed a casing leak at about 1,000 ft and was plugged and abandoned by operator Mesa Energy in 1986. After Pioneer acquired Mesa in 1997, the EPA informed the company that the old Biere well it had acquired was a potential source of salinity contamination that had been detected in several landowners’ water wells. In 2000 Pioneer conducted ground temperature surveys that confirmed the leak, and engaged in a $1 million project to plug the well and drill monitoring wells to ensure that no new contamination entered the aquifer.18

The 2004 HEM survey assisted Pioneer in its remediation efforts by delineating the saline plume that had originated at the Biere well. The plume was found to move south from the area of B, toward water wells supplying the city of Poplar.16 Additional geological and hydrological studies on the Beire plume in mid-2006 showed that it is discrete from other regional contamination. Based on these studies, Pioneer determined that it could drill recovery wells to mitigate the salinity contamination. After drilling additional monitor wells and conducting aquifer tests, groundwater modeling was done to determine the capture zones of potential recovery wells; the models were favorable for both containing and capturing the plume. As of press time, Pioneer was slated to drill its first of five recovery wells in late November 2007, to be followed by a deep disposal well in January 2008 for the removed contaminated water.

However, earlier groundwater quality studies identified several possible shallow sources in the area, so there are likely multiple causes of conductivity anomalies.15 For example, nearby agricultural practices can lead to shallow salinization, so not all saline waters can be attributed to energy development.19 In area E in the northwest part of the survey area (Fig. 6), both agricultural and energy land uses are potential sources for the observed shallow saline waters in water wells.1 A borehole induction electrical conductivity log (Fig. 7) for a well near E (Fig. 6) shows both a shallow and a deep region of high conductivity. Chemical analyses from groundwater at this well are not definitive for either oilfield brine or natural salinization.

Fig. 7

Fig. 7. Borehole geophysical logs for monitor well MSCA-4, near feature E, show both a shallow and a deep region of high conductivity. The large red dots on the conductivity log are computed differential depth and resistivity12 from the HEM survey line that is located above the well. Courtesy of USGS.

However, the well log and airborne survey show both shallow and deep sources for high electrical conductivity. The deep source could be due solely to the Bearpaw Shale, which was reported from the drillers’ log to be at 6 m (18 ft). The shale is reported as weathered, so the conductivity profile below 6 m could reflect a decrease in weathering. Conversely, the water level in the well is about 12 m (35 ft), so the increase in conductivity could be due to saline water in the weathered shale. More geochemical work could shed light on the possible source of the salinity. Thus, the high apparent conductivity anomaly in this area could be separated as shallow and deep sources using data from the HEM survey. As part of its remediation effort, Pioneer has put in additional drill holes and conducted additional well logs and geochemistry testing, which has helped to map the source area.16

Another plume near feature C (Fig. 6) was confirmed from monitor wells, but the source has not yet been confirmed and potential impacts have not been fully evaluated. Like the plume near B, the produced water contamination is moving south toward the residential and municipal water-well system. At both plumes (near B and near C, Fig. 6) ground EM surveys had indicated high conductivity, but the surveys did not have sufficient depth to clearly map the plume. The helicopter EM survey provided the depth and resolution necessary to do this mapping.


The airborne geophysical survey of East Poplar Field identified areas of high conductivity that can be attributed to both hydrologic and geologic sources. Airborne EM measurements provided a greater areal coverage and greater subsurface information than the ground EM surveys. Ongoing studies are directed both to understanding the sources of anomalies and to defining possible flow paths. Understanding the topography of the buried Bearpaw Shale has helped to predict constraints on groundwater flow. WO 


This article draws heavily from documents produced by Bruce D. Smith of the US Geological Survey and others. Specifically, much of the information is taken from Smith, B. D., Thamke, J. N and C. Tyrrell, “Geophysical and hydrologic studies of shallow aquifer contamination, East Poplar oil field area, northeastern Montana,” presented at the Symposium on Engineering and Environmental Geophysics, in Seattle, Wash., 2006; and Smith, B. D. et al., “Airborne geophysics for ground water resources: A decade of study by the US Geological Survey Central Region,” poster presented at Exploration07, in Toronto, Ont., Canada, Sept. 9-12, 2007.


1 Thamke, J. N. and K. S. Midtlyng, “Groundwater quality for two areas in the Fort Peck Indian Reservation, northeastern Montana, 1993-2000,” US Geological Survey Water-Resources Investigations Report 03-4214, 2003.
2 Thamke, J. N. and S. D. Craigg, “Saline-water contamination in Quaternary deposits and the Poplar River, East Poplar oil field, northeastern Montana,” US Geological Survey Water-Resources Investigations Report 97-4000, 1997.
3 Smith, B. D. et al., “Helicopter geophysical survey to detect brine, Brookhaven Oil Field,” Expanded Abstracts with Biographies, 1989 Annual Meeting Society of Exploration Geophysicists Technical Program, 1989.
4 Smith, B. D., Bisdorf, R., Slack, L. J. and A. Mazzella, “Evaluation of electromagnetic mapping methods to delineate subsurface saline waters in the Brookhaven oil field, Mississippi,” Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems, 1997, pp. 685-693.
5 Paine, J. G. and B. R. S. Minty, “Airborne hydrogeophysics,” in Singh, V. P., Rubin, Y. and S. S. Hubbard, Eds., Hydrogeophysics, Springerverlag, 2005, pp. 333-357.
6 Sams, J. I., Smith, B. D., Lipinski, B. and W. Harbert, “Applications of airborne electromagnetic surveys to improve management of produced water in the Powder River Basin,” presented at the Symposium on Environmental and Engineering Geophysics, Seattle, 2006.
7 Land and Water Consulting, Inc., “Public water supply well threat study, East Poplar oil field, Roosevelt County, Montana: Missoula, Mont.,” prepared for Murphy Exploration and Petroleum Company, Pioneer Natural Resources, Marathon Ashland Petroleum and Samson, 2003.
8 Montana Department of Environmental Quality, “Source water delineation and assessment report, city of Poplar: Helena, Mont.,” PWSID 00310, 2002.
9 Echotech Geophysical Inc., written communication to Smith, B. D., 2002.
10 Smith, B. D., Thamke, J. N. and J. G. Paine, “Electrical conductivity geophysical methods applied to subsurface mapping of produced waters,” presented at the American Association of Petroleum Geologists Annual Meeting, Dallas, 2004.
11 Fugro Airborne, RESOLVE survey for the US Geological Survey, East Poplar Oil Fields, Montana, 2005.
12 Huang, H. and D. C. Frasier, “The differential parameter method for multifrequency airborne resistivity mapping,” Geophysics, 55, 1996, pp. 1327-1337.
13 Colton, R. B., “Geologic map of the Poplar quadrangle, Roosevelt, Richland and McCone Counties, Montana,” US Geological Survey Miscellaneous Geologic Investigations, Map I-367, 1963.
14 Colton, R. B., “Geologic map of the Hay Creek quadrangle, Roosevelt County, Montana,” US Geological Survey Miscellaneous Geologic Investigations, Map I-365, 1963.
15 Thamke, J. N., Craigg, S. D. and T. M. Mendes, “Hydrologic data for the East Poplar oil field, Fort Peck Indian Reservation, northeastern Montana,” US Geological Survey Open-File Report 95-749, 1996.
16 Smith, B. D., US Geological Survey, interview by author, Nov. 6, 2007.
17 Kharaka, Y. K. and J. K. Otton, “Environmental impacts of petroleum production: Initial results From the Osage-Skiatook petroleum environmental research sites, Osage County, Oklahoma,” US Geological Survey Water-Resources Investigations Report 03-4360, 2003.
18 Jacobs, M., senior environmental specialist, Pioneer Natural Resources, interview by author, Nov. 19, 2007.
19 Miller, M. R. and R. N. Bergantino, “Distribution of saline seeps in Montana,” Montana Bureau of Mines and Geology Hydrogeologic Map 7, 1983.


Related Articles
Connect with World Oil
Connect with World Oil, the upstream industry's most trusted source of forecast data, industry trends, and insights into operational and technological advances.