APPLYING AIRBORNE ELECTROMAGNETIC INDUCTION IN GROUND-WATER SALINIZATION AND RESOURCE STUDIES, WEST TEXAS

Jeffrey G. Paine and Edward W. Collins, Bureau of Economic Geology, Austin, TX

Abstract

In 2001, two high-resolution airborne geophysical surveys were fl own in West Texas using Fugro’s MEGATEM II system to acquire time-domain EM (TDEM) and magnetic fi eld data. One survey, fl own at 100-m line spacing over a 162-km2 area on the Edwards Plateau, identifi ed and assessed groundwater salinization in an oil-fi eld area. The second survey, fl own at 400-m line spacing over a 372-km2 area in a West Texas basin, identifi ed favorable areas for groundwater exploration.

In the salinization study, the magnetic fi eld data accurately identifi ed most of the more than 400 oil and gas wells. Commonly, magnetic anomalies more accurately located wells than did agency records. Some wells located equidistant from adjacent fl ight lines were undetected. Where many wells are clustered, the airborne magnetometer identifi ed a single anomaly for a group of wells. Apparent conductivities calculated from the airborne geophysical data at 10-m intervals between depths of 10 and more than 200 m below the ground surface are generally low. Low conductivities are consistent with the good water quality reported in most of the shallow wells, where water is fresh to slightly saline. Local areas of elevated ground conductivity are associated with oil fi elds where saline water had been discharged into now-abandoned disposal pits.

In the resource study, ground-based TDEM soundings established the achievable exploration depths and demonstrated that the relatively deep groundwater (80 to 120 m) signifi cantly infl uenced the transient signal and was within the MEGATEM exploration depth. Airborne EM and magnetic fi eld data identifi ed intrabasinal faults that infl uence basin-fi ll deposition. Conductivity-depth slices constructed from airborne TDEM data allowed lateral variations in water quality and lithology to be mapped that are necessary to predict groundwater resource quality within these fault-bounded basins fi lled with alluvial, lacustrine, and eolian sediments.

Introduction

We conducted airborne and supporting ground-based geophysical studies in Sterling County, Texas, to examine the intensity of groundwater salinization (Paine, 2002) and in eastern El Paso County to delineate potentially favorable groundwater resources (Paine and Collins, 2002) (Figure 1). The principal geophysical method employed in both studies was time-domain electromagnetic induction (TDEM), which we used to measure the apparent electrical conductivity of the ground to depths of 200 m or more. The key to the potential success of groundwater salinization and resource studies is that the electrical conductivity of the ground is infl uenced both by sediment type (clayey sediments are more conductive than sandy sediments) and by water quality (saline water is more conductive than fresh water). All else being equal, salinized areas will have relatively high electrical conductivity, whereas nonsalinized areas (including fresh groundwater resources in coarse host deposits) will have relatively low conductivity.

Dense spatial data obtained by fl ying the instruments at low altitude on a fi ne grid should allow subsurface changes in water quality and sediment type to be interpreted. The airborne geophysical data were verifi ed by (1) acquiring ground-based geophysical data at representative locations and comparing results, (2) comparing water well data with conductivity patterns evident in the airborne data set, and (3) incorporating existing geological and hydrological information into the interpretation of the airborne geophysical data.

722

Please cite this document as follows: Paine, J. G., and Collins, E. W., 2003, Applying airborne electromagnetic induction in groundwatersalinization and resource studies, West Texas, in Proceedings, Symposium on the Application of Geophysics to Engineering andEnvironmental Problems: Environmental and Engineering Geophysical Society, p. 722-738 (CD-ROM)

One advantage of airborne TDEM is its ability to examine lateral changes in apparent conductivity at critical depths that can be related to lithology, water saturation, and water quality. High-resolution coverage allows the context of individual soundings to be considered and helps identify geological and hydrological features that would be diffi cult to recognize from sparse well data or ground-based geophysical studies. By combining data from wells, ground-based surveys, and airborne surveys, we can reduce the interpretational uncertainty associated with many types of geophysical data and can better interpolate between typically sparse wells and ground data points.

Methods

We employed airborne and ground-based geophysical methods to rapidly and noninvasively delineate salinization and explore for favorable groundwater resources by measuring changes in electrical conductivity

Figure 1: (a) Location of the airborne survey areas in West Texas. (b) Groundwater salinization study area in Sterling County. (c) Groundwater resource assessment area in El Paso County.

Rio Grande

El P

aso

Co.

N

ElPaso

MEXICO TEXAS

Hud

spet

h C

o.

IH 10

Airborne surveyUS 54

Juarez

0

0

20 mi

20 km

(a)

Sterling Co.

TexasTexasT

0

0

200 mi

200 km

Gulf ofMexico

NewMexico

Airbornesurvey

163

163

158

Lacy Ck N. Concho R.

N. Concho R.

U.S. 87U.S. 87

Sterling Co.Mitchell Co.

TomGreen

Co.

CokeCo.

0

0

10 mi

10 km

N

Irion Co.

SterlingCity

Reagan Co.

(b)

(c)

723

with depth. The principal geophysical method is electromagnetic induction (Parasnis, 1973; Frischknecht and others, 1991; West and Macnae, 1991). This method employs a changing primary magnetic fi eld that is created around a current-carrying transmitter wire to induce a current to fl ow within the ground, which in turn creates a secondary magnetic fi eld that is sensed by a receiver coil. In general, the strength of the secondary fi eld is proportional to the ground conductivity.

The TDEM method (Kaufman and Keller, 1983; Spies and Frischknecht, 1991) used in both airborne surveys and in the El Paso ground-based survey measures the decay of a transient, secondary magnetic fi eld produced by the termination of an alternating primary electric current in the transmitter loop. The secondary fi eld, generated by current induced to fl ow in the ground, is measured by the receiving coil following transmitter current shutoff. Secondary fi eld, or transient, strength at an early time gives information on conductivity in the shallow subsurface; transient strength at later times is infl uenced by conductivity at depth.

Airborne GeophysicsFugro Airborne Surveys fl ew the Sterling and El Paso study areas in August and September 2001

using its MEGATEM II TDEM system and a cesium magnetometer towed behind a four-engine aircraft (Figure 2). The Sterling survey covered 162 km2 west of Sterling City (Figure 1b). The principal fl ight lines were oriented north-south and spaced at 100 m. The El Paso study covered a 372-km2 area in eastern El Paso County (Figure 1c). The principal north-south fl ight lines were spaced at 400 m. Flight height was 120 m; the three-axis EM receiver was towed 131 m behind the transmitter at a height of 75 m above the ground. The primary EM fi eld was generated by a four-turn wire loop fi xed to the aircraft carrying a 30-Hz, discontinuous sinusoidal current of 1,330 amperes (A). The dipole moment was 2.1 × 106 A-m2,

Figure 2: Fugro’s MEGATEM II system acquiring data over El Paso County, Texas, August 2001.

724

more than an order of magnitude larger than the 80,000 A-m2 moment calculated for the ground-based measurements. Transients were recorded during the 11.6-ms window following termination of the 4-ms transmitter pulse.

EM diffusion depth is commonly used as a proxy for exploration depth. It is calculated using the equation d = k ( t r )0.5, where d = diffusion depth (in m), k = 503.3 (m/ohm-s)0.5, t = latest time measured, and r = resistivity (in ohm-m) (Parasnis, 1986). Assuming an average resistivity of 20 ohm-m estimated from ground-based data and a latest measurement time of 11.6 ms, exploration depth for the Sterling survey is about 240 m. Assuming a resistivity range of 13 to 50 ohm-m for the El Paso survey estimated from ground TDEM measurements, expected exploration depth is 190 to 380 m.

Measurement locations were determined using differential GPS. Locational accuracy is 10 m or better. At the 30-Hz transmitter frequency (60-Hz sample frequency) and a nominal airspeed of 250 km/hr, transients were acquired every 1.2 m along the fl ight line. Recording stacked transients at 4 Hz resulted in a sample spacing of about 17 m.

Fugro performed conductivity-depth transforms to produce conductivity profi les for each stacked transient to depths of 300 m or more (Hefford, 2001a, b). We produced horizontal images of subsurface conductivity by (1) extracting conductivity values at 10-m depth intervals, (2) gridding the values using ERMapper, and (3) importing the images into GIS databases. Coverages used to analyze the relationship between the geophysical data and geological and hydrological characteristics included water-well locations and depths, water-quality analyses, roads, and streams. For the Sterling survey, we also imported oil and gas wells and pipeline locations provided by the Railroad Commission of Texas (RRC).

The aircraft towed a cesium magnetometer at a height of 73 m above the ground to measure changes in the magnetic fi eld strength caused by natural effects and local features such as pipelines and oil and gas wells. Magnetometer data were acquired at 10 Hz, yielding a 7-m sample spacing.

Ground GeophysicsIn the Sterling survey, we acquired conductivity data using a ground conductivity meter (Geonics

EM34). We collected data at either the 20- or 40-m coil separation (1,600 or 400 Hz) along nine lines for comparison with airborne survey results. In the El Paso survey, we examined the transients measured at nine representative sites using a TDEM instrument (Geonics Protem 57), used those to verify the feasibility of the method, and compared the conductivity models derived from both data sets to validate the airborne data. Ground-based systems have fewer problems with EM noise and it is easier to keep the transmitter and receiver geometry constant, but data are more diffi cult to acquire. Airborne systems make it possible to produce high-resolution images of subsurface conductivity over large areas that cannot be practically surveyed on the ground.

Sterling County Salinization Study

The Sterling survey covers 162 km2 in a moderate-relief area composed of Cretaceous Edwards Group limestones forming a plateau that has been dissected by Concho River tributaries, such as Lacy Creek, which fl ows from west to east across the survey area (Figure 1b). The area is twice the size of the salinization study conducted in nearby Runnels County using the helicopter-based DIGHEM system (Paine and others, 1999). The TDEM instrument used in the Sterling survey explores to greater depths and produces more detailed vertical conductivity profi les. Groundwater quality in the Sterling survey area is generally good, both from an alluvial aquifer along Lacy Creek and from the Cretaceous Antlers Sand aquifer. Local groundwater salinization related to oil and gas exploration and production has been documented in the Parochial Bade Oil Field (Ashworth and Coker, 1993; Renfro, 1993) in the southwestern part of the

725

survey area. We obtained oil- and gas-well locations from the RRC to delineate areas of potential oil-fi eld salinization caused by past discharge of produced water into pits, migration of saline water from deeper formations, and leakage of injected saline water. Signifi cant salinization should be accompanied by an increase in electrical conductivity near the salinity source. Magnetic fi eld data can be used to verify well locations and locate either unknown or mislocated wells. Conductivity data can then identify conductivity anomalies consistent with salinization. When combined, these data help distinguish natural and oil-fi eld salinity sources (Paine and others, 1999).

Groundwater Depth and QualitySterling survey water depths range from 4 to 94 m, averaging 26 m. Surface elevation varies

signifi cantly from below 700 m in the Lacy Creek valley to more than 800 m on the plateau. The elevation of the water table generally mimics surface elevation. Calculated airborne geophysical exploration depths exceed reported well depths of 6 to 120 m. Wells produce from an alluvial aquifer or from the Antlers Sand (Ashworth and Coker, 1993; Renfro, 1993) (Figure 3a). Water is dominantly fresh to slightly saline. Reported TDS concentrations range from 238 (fresh) to 26,035 mg/L (very saline) but only 4 of 53 analyses have TDS values higher than 3,000 mg/L.

One key to the potential success of airborne EM is the relationship between TDS concentration and the electrical conductivity of water. Water samples from the survey area show that the measured conductivity increases linearly at a rate of 0.2 mS/m per 1 mg/L increase in TDS concentration (Figure 3a). Groundwater conductivity is low because TDS concentrations are low. Only three samples have conductivities that exceed 500 mS/m, suggesting that rock properties will strongly infl uence measured ground conductivities where TDS values are low. Highly conductive ground (more than 200 mS/m) is expected only where groundwater is moderately to very saline.

Magnetic Field DataMagnetic fi eld strength ranged from 49,464 to 49,660 nanoteslas (nT) over the survey area. After

removal of the regional reference fi eld, the residual magnetic intensity map (Figure 4) shows a few linear anomalies corresponding to major pipelines and hundreds of small anomalies that are tens of meters across

Figure 3: Relationship between total dissolved solids (TDS) concentration and specifi c conductivity of samples from (a) 52 water wells within the Sterling survey area and (b) 135 water wells within the El

Paso survey area. Data from the Texas Water Development Board (TWDB). Water salinity classifi cation from Robinove and others (1958).

Con

duct

ivity

(mS/

m)

Total dissolved solids (mg/L)0 1000 2000 3000 4000

Slightly salineModerately

salineFresh

500

400

300

200

100

0

600

700(a) (b)

Con

duct

ivity

(mS/

m)

Total dissolved solids (mg/L)0 10,000 20,000 30,000

Slig

htly

sal

ine

Verysaline

Fresh4000

2000

0

6000

Mod.saline

conductivity = 0.207 × TDS + 0.001r = 0.999conductivity = 0.207 × TDS + 0.001r = 0.999

conductivity = 0.183 × TDS - 1.78r = 0.973conductivity = 0.183 × TDS - 1.78r = 0.973

726

and a few to several tens of nT in magnitude. Most correspond to oil- and gas-well locations obtained from RRC or to evidence of a well on recent aerial photographs. A few known wells, particularly those that are midway between adjacent fl ight lines, were not detected. The airborne magnetometer distinguishes wells best where they are farther apart than about 100 m. At closer spacings, such as those common in Parochial Bade fi eld, the signatures of several adjacent wells combine to form a single anomaly. Individual, densely clustered wells cannot be distinguished without tighter fl ight spacings or ground measurements.

Airborne EM DataApparent conductance maps constructed from airborne-survey data show relatively low values at

higher elevations where poorly conductive Edwards limestones cap the plateaus (Eifl er and Barnes, 1974). At lower elevations, where limestones are thin or absent and more conductive alluvium and underlying Antlers Sand and older strata are present, conductances are higher. Highest conductances are located along Lacy Creek in the western part of the survey area and within Parochial Bade fi eld. Elevated conductances are found where shallow water wells have high TDS values.

The 108,630 conductivity-depth transforms can be combined to produce images of apparent conductivity at selected depths. In general, conductivities within the upper 300 m in the survey area are relatively low. The highest apparent conductivities are observed on depth slices from the upper few tens of meters. For example, conductivities at 30-m depth are elevated northwest of Sterling City and in Parochial Bade fi eld (Figure 5). The highest measured TDS value, 26,035 mg/L in Parochial Bade fi eld, is reported within this area of elevated conductivity. Lower TDS values characteristic of fresh water are reported at this depth elsewhere in the survey area, where apparent conductivities are low.

Figure 4: Residual magnetic fi eld intensity measured during the Sterling airborne survey. Also shown are oil- and gas-well locations and pipeline locations from the RRC.

Residual magnetic field (nT)-1940

-2149

158

163

U.S. 87 163

158

mi0

0 km

3

4

SterlingCity

N

Airborne survey

Oil or gas well

Lacycyc

C k .

727

Quantifying the Relationship between Conductivity and Water SalinityQualitative comparisons of apparent conductivities at various depths with TDS values from water

wells show relatively good agreement at depths less than 110 m. Beyond this depth, there are no wells for comparison. Areas with relatively high apparent conductivities at given depths tend to encompass wells with relatively high TDS water at that depth; areas with relatively low apparent conductivities at given depths tend to be near wells producing relatively low TDS water from those depths. We attempted to quantify this relationship by extracting an apparent conductivity value at the location of the well and at a depth that was shallower than the well but deeper than the water level. We found 47 wells in the Texas Water Development Board (TWDB) database for which there was a well depth, a water level, and a TDS value. We used the reported locations of these wells to obtain an apparent conductivity from the airborne geophysical data.

Reported TDS values show relatively poor correlation with apparent conductivities at the specifi c well location and interpreted sample depth (Figure 6), apparently contradicting the reasonable geographic agreement among TDS values plotted on the conductivity maps. There are several possible explanations for the lack of a better quantitative relationship between TDS and apparent conductivity. First, most of the TDS values are for fresh to slightly saline water. At low TDS values, the conductivity of water-saturated sediments and rocks is more strongly infl uenced by sediment or rock type than it is at higher TDS values. Second, chemical analyses of water samples date from between 1961 and 2001 and may not refl ect current salinity. Third, water-well locations are less accurate than the location of the conductivity profi les. Fourth, it is diffi cult to identify the actual stratal depth that contributed the water sample from the available data. Fifth, apparent conductivity values were determined from gridded line data rather than from the nearest actual measurement. Depending on the well distance from the original measurement, the apparent conductivity value may not represent the best estimate obtainable from the airborne geophysical data at the well.

Figure 5: Apparent conductivity at a depth of 30 m in the Sterling survey area. Also shown are locations of wells with TDS values from samples interpreted to be from this depth range.

Apparent conductivity 179

8

158

163

U.S. 87 163

158

mi0

0 km

3

4

SterlingCity

N

Airborne survey

Lacycyc

C k .

356

458

404

1220

474

317

303

1829

26035 356

458

404

1220

474

317

303

1829

26035

551551

337337

Total dissolved solids (mg/L)252

728

Extent and Intensity of SalinizationSterling survey geophysical data show that ground conductivity is generally low, consistent with the

distribution and predominance of fresh to slightly saline water in the alluvial and Antlers aquifers. There has been signifi cant oil and gas exploration and production, but the salinization extent is limited. Elevated conductivity consistent with shallow salinization that has impacted the alluvial and Antlers aquifers is present in Parochial Bade fi eld, particularly near abandoned saltwater disposal pits. Elevated conductivities in the fi eld and along the draw and creek near the fi eld suggest that salinization related to oil-fi eld activity has occurred.

El Paso County Groundwater Resource Study

El Paso County, Texas’ westernmost county, is situated in the arid western mountain region of the state. Like in much of the southwestern U.S. desert, there is a shortage of available surface water to meet increasing municipal, industrial, and agricultural water demands of an estimated 2001 county population of 690,000 (U.S. Census Bureau, 2003). Because of the importance of adequate water to the health and prosperity of this heavily populated area, numerous State and Federal studies have examined its groundwater resources (e.g., Knowles and Kennedy, 1956; Meyer and Gordon, 1972; Alvarez and Buckner, 1980; Gates and others, 1980; Ashworth, 1990). We conducted geophysical studies to delineate potentially favorable groundwater resources in the upper few hundred meters of the Hueco Bolson in eastern El Paso County

Figure 6: Relationship between TDS concentration in groundwater samples from the Sterling survey area and apparent conductivities calculated from airborne data for depths below the water level and

above the well depth. Water data from TWDB.

�����������������������������

� ����������� ������ ������ ������ ������

����������������������������

���

���

���

���

���

�

������

����������������������������

�����

���������������

729

(Figure 1c) that might be considered as a future water supply supplement. The survey covers 372 km2 that includes areas interpreted to contain potential groundwater resources of relatively low TDS concentrations at shallow depths (Mullican and others, 1997). Favorable groundwater resource areas within the survey area will have low electrical conductivity (fresh water in sandy or gravelly sediments), whereas areas with poor water-resource potential will have high conductivity (saline water or clayey sediments).

Geologic SettingThe El Paso survey is within the Hueco Bolson, a fault-bounded West Texas desert basin surrounded

by the Franklin Mountains to the west, the Hueco Mountains to the east, and the Diablo Plateau. Rocks, unconsolidated to slightly friable sediments, and the landscape record Precambrian to recent geologic events (Henry and Price, 1985; Muehlberger and Dickerson, 1989). The geologic framework for water resources has been mostly infl uenced by events that began about 24 to 30 mya when the regional stress regime became extensional and normal faulting uplifted the Franklin and Hueco Mountains relative to the Hueco Bolson.

Sediments within the basin refl ect the development of the Rio Grande and its ancestors (Albritton and Smith, 1965; Strain, 1966, 1971; Hawley and others, 1969; Stuart and Willingham, 1980; Mack and Seager, 1990; Gustavson, 1991). The ancestral Rio Grande system drained into Lake Cabeza de Vaca, which encompassed much of the basin. Breaching by about 2.2 mya resulted in a through-fl owing river system. Basin-fi ll deposition associated with the ancestral river system continued into the early Pleistocene to about 0.78 mya (Mack and others, 1998). Since then, periods of downcutting, deposition, and stability have been recorded in the Rio Grande and large arroyo terraces, alluvial-fan and piedmont deposits, and soils.

Basin-fi ll deposits thicken westward toward the Franklin Mountains to 1.5 to 2.5 km (Mattick, 1967; Ramberg and others, 1978; Seager, 1980; Collins and Raney, 1991, 2000). Cliett (1969) reported the best water-bearing sediments to be thick fl uvial gravel and sand deposits that occur along the east side of the Franklin Mountains. North-striking normal faults cut the basin fi ll and are expressed at the surface as subtle sand-covered scarps (Collins and Raney, 1991, 2000). These faults may have partly controlled the basin’s surface drainage and sedimentation, although geometries of coarser basin-fi ll intervals are unknown.

Groundwater Depth and QualityWe examined well and water depths to ensure that the geophysical exploration depth is suffi cient

to reveal information about groundwater. Water depths are generally between 90 and 130 m and correlate strongly with surface elevation. The Hueco Bolson aquifer is under water-table conditions (Gates and others, 1980). Water-quality data show that most groundwater samples from wells in eastern El Paso County are fresh to slightly saline (Figure 3b). The most recent TDS analyses range from 213 to 8,079 mg/L, averaging 1,159 mg/L. Samples from about half of the wells are classifi ed as fresh. Multiple-depth samples from some wells show increasing TDS concentrations with depth. Because most of the sediments fi lling the Hueco Bolson have naturally low conductivities (generally less than 100 mS/m), and because the conductivity of most water samples is higher than 100 mS/m, water quality will strongly infl uence ground conductivity measurements.

Magnetic Field DataMagnetic fi eld data were used to delineate faults that infl uence the subsurface distribution of

sediments. Local magnetic fi eld features are enhanced by subtracting the calculated regional fi eld (IGRF 2001.67) from the measured and corrected values. The residual magnetic-fi eld intensity map displays difference values ranging from 660 nT below IGRF along the west margin, where the Hueco Bolson is the deepest, to values as high as 165 nT below IGRF in the southeast and northeast areas, where basin-fi ll sediments are thin and bedrock is exposed. Several linear anomalies coincide with major pipelines.

730

Fugro produced shallow-depth transforms of the magnetic fi eld data (Figure 7) that emphasize several curvilinear features that correspond well to faults mapped and inferred from aerial photographs and fi eld mapping (Collins and Raney, 1991, 2000). These features are hydrologically signifi cant because they relate to the development and evolution of the Hueco Bolson and may have infl uenced drainage patterns during basin deposition.

EM DataWe used EM data to detect subsurface changes in electrical conductivity that reveal information

about groundwater quality and depth and the type of sediment hosting the water. High conductance might be caused by the presence of saline water, particularly where the water table is shallow. Relatively fresh groundwater would be expected where conductance is low. In this area, where groundwater depths are generally 100 m or more, depth transforms of the EM data are more useful than simple apparent conductances because the inversions produce estimates of apparent conductivity that vary with depth.

We examined apparent conductivity patterns on horizontal slices at 10-m depth intervals from 10 to 200 m below the surface. Depth slices between 10 and 90 m are above the water table across most of the area. At these depths, conductivities are relatively low and are more infl uenced by variations in moisture and clay content than by the dissolved constituents in pore water. Conductivity patterns correspond strongly

Figure 7: Shallow-depth transform of magnetic fi eld data acquired in the El Paso survey area. Also shown are faults mapped in the area by Collins and Raney (1991, 2000).

Shallow-depth transform (nT)9

-10

Mapped faultN

0 4 mi

0 4 km

I–10

HorizonCity

Loop

375

U.S. 62MontanaVista

Ft. Bliss

HorizonCity

U.S. 62MontanaVista

Ft. Bliss

731

to geologic features. Superimposed on the 70-m depth slice (Figure 8a) are known and inferred faults with surface expressions. Areas of similar but low apparent conductivity are separated from other areas of similar but high apparent conductivity by abrupt boundaries that coincide with the faults. In most instances, conductivities are lower on the downthrown side of the fault. Low conductivities are also associated with mapped bedrock outcrops. Higher conductivities are likely to be found where clayey sediments have been deposited; poorly conductive areas are likely to have higher sand, gravel, or caliche contents.

At depths below the water table, favorable groundwater resources should appear on conductivity depth slices as areas of low apparent conductivity. On the 120-m conductivity depth slice (Figure 8b), low-conductivity areas form three north-northwest to south-southeast trending zones (east, central, and west) that are separated by similar-trending zones with higher apparent conductivities. As depth increases, the conductive zones expand until depths of 160 m or more are reached, at which point only the eastern low-conductivity zone covers a signifi cant total area. Below 100 m, the central zone of low conductivity that was prominent at shallower depths narrows considerably and is barely discernible on the 150-m depth slice. Other signifi cant areas of relatively low conductivity are found in the northwest corner, an area northwest of Horizon City, and a large north-south-trending area in the north and east quadrant, where bedrock is exposed.

Figure 8: (a) Apparent conductivity at a depth of 70 m in the El Paso survey area. Also shown are faults mapped in the area by Collins and Raney (1991, 2000). Labels “U” and “D” denote interpreted

upthrown and downthrown sides of a fault. (b) Apparent conductivity at a depth of 120 m. Also shown are TDS values for water samples that have water levels above this level and are screened at this level

and interpreted favorable exploration areas for relatively low-TDS groundwater.

��� ���

�������������������������������

��

������������������������

��

� �

����

� �

��

� ���

��

��

��

��

����

��

��

��

��

����

�����������

��������

�������

������������

���������

�����������

�������

������������

���������

�

� ����

� ����

�������������������������������

��

�����

�����

�����

�����

����

�����������

��������

�������

������������

�����������

�������

������������

������������������

���������������������������������

����

����

��������

��������

����

����

���

����

����

����

��������

����

����

���

���

���

�������

�������

���

��� ���

����

���

���

�������

�������

��� ���

������

����

����

��������

��������

����

����

����

��������

����

����

����

���� ����

����

����

���

������

����

���� ����

�

� ����

� ����

732

Figure 9: Comparison of TDS values from one low-TDS water well and one high-TDS water well with conductivity profi les calculated for transients acquired during the El Paso airborne survey nearest the

well locations.

Relationship between Water Quality and ConductivityWater quality can dominate measured conductivity within the water-saturated zone. To illustrate the

correlation between apparent conductivity measured by an airborne instrument and the TDS concentration in groundwater, we selected two wells (4915404 and 4915509, Figure 9) that were located near fl ight lines, had similar depths, and had substantially different TDS concentrations. Well 4915509 has a reported water depth of 125 m, a slotted casing from 119 to 140 m, and a fresh TDS value of 410 mg/L. Well 4915404, located about 6 km west of well 4915509, has a water depth of 105 m, two slotted casing sections from 107 to 125 m and from 180 to 192 m, and a slightly saline TDS value of 1,935 mg/L. A conductivity profi le located about 30 m from the fresh-water well 4915509 shows consistently low apparent conductivity values throughout the depth range potentially sampled by the adjacent well (Figure 9). A conductivity profi le located about 70 m from the more saline water well 4915404 shows apparent conductivity values that are considerably higher than those near well 4915509 throughout the depth ranges that could have produced the more saline water.

A more quantitative assessment of the relationship between water quality and apparent conductivity can be carried out by plotting the TDS value reported for a well against the apparent conductivity indicated by the conductivity profi le located closest to the well (Figure 10). In most cases, there is more than one conductivity value for each TDS value because apparent conductivity slices at more than one depth fall between depths reported for the water level and the top and bottom of slotted casing. If all wells are used regardless of TDS value, the accuracy of the well location, or the distance between the closest conductivity

200 300 400 5001000

Apparent conductivity (mS/m)

0

50

100

150

200

250

Dep

th (m

) Well 4915404TDS = 1935 mg/L

TDEM profile near well 4915404

TDEM profile near well 4915509

Well 4915509TDS = 410 mg/L

Water level

Screen top

Screen bottom

Well bottom

733

Figure 10: Relationship between TDS concentration in groundwater samples from the El Paso survey and apparent conductivities calculated from airborne data for depths below the water level and within

the screened interval of each well. (a) Data pairs and best-fi t relationship using all available wells; (b) data pairs and best-fi t relationship using only those wells that have been accurately located using

GPS and are within 100 m of an airborne measurement; and (c) data pairs and best-fi t relationship using wells that meet the criteria listed in (b) and have TDS concentrations greater than 1,000 mg/L.

(a)

(b)

(c)

3000200010000Total dissolvTotal dissolvT ed solids (mg/L)

3000200010000Total dissolvTotal dissolvT ed solids (mg/L)

3000200010000Total dissolvTotal dissolvT ed solids (mg/L)

300

200

100

0

Ap

par

ent

con

du

ctiv

ity

(mS/

m)

300

200

100

0

Ap

par

ent

con

du

ctiv

ity

(mS/

m)

300

200

100

0

Ap

par

ent

con

du

ctiv

ity

(mS/

m) conductivity = 0.130 × TDS - 111.3

r = 0.813n = 37

conductivity = 0.0578 × TDS + 18.4r = 0.739n = 59

conducconductivity = 0.0393 tivity = 0.0393 × TDS + 52.5r = 0.430r = 0.430n = 182n = 182

734

profi le and the well, there is only a vague relationship between water quality and apparent conductivity (Figure 10a). For these 182 data pairs, there is a hint of the expected increase in apparent conductivity that should accompany increasing TDS values, but the correlation coeffi cient is low at 0.430. A better fi t is obtained when we restrict the data pairs used in the analysis to the most accurately located wells (located by TWDB using GPS instruments) and only those wells located within 100 m of a fl ight line (Figure 10b). The straight-line fi t through the 59 remaining data pairs has a reasonably good correlation coeffi cient of 0.739.

Most of the increase in apparent conductivity occurs at relatively high TDS values (Figure 10b), suggesting that, for relatively fresh groundwater, the apparent conductivity is within a range that is as strongly infl uenced by sediment or rock properties as it is by ionic concentrations. Taking data only from wells that meet the strict location and distance criteria and further removing TDS values less than 1,000 mg/L, the correlation coeffi cient for a linear fi t to the 37 remaining data pairs improves to 0.813 (Figure 10c).

Favorable Groundwater Exploration AreasIn a simplifi ed sense, we seek to identify laterally and vertically extensive areas at or below current

groundwater levels that have low apparent conductivities, which indicate fresh groundwater saturating coarse and potentially productive sediments. The most favorable groundwater exploration areas can be delineated on the basis of airborne and ground-based geophysical data, the distribution of fresh to moderately saline water in wells, depths to water and slotted well casing, and presence of geological features within the Hueco Bolson. This analysis indicates that there is more confi dence associated with identifying likely low-salinity groundwater than there is in identifying highly productive zones.

At the most relevant depths of 110 to 130 m, there are two extensive areas with low apparent conductivity. The largest is area A (Figure 9), which covers about 23 km2 on the depth slice at 120 m. It is the broad zone of low conductivity trending north-northwest to south-southeast across the survey area and passing west of Montana Vista. It appears to be bounded on the west, and perhaps on the east, by faults that have dropped this area downward relative to adjacent blocks. Few lithologic data are available, but a driller’s log from a well near the margin of this area indicates coarse sediments in this depth range. One possible geological interpretation is that this feature represents a former drainageway that focused deposition of coarse sediment moving southward toward the ancestral Rio Grande.

Area B, located in the northwest part of the study area (Figure 9), may represent the eastern extension of an extensive area of fresh water interpreted to exist on the eastern fl ank of the Franklin Mountains (Gates and others, 1980). Here the low-conductivity zone covers an area of about 8 km2 within the survey area, migrates westward with depth, and extends to a depth of at least 160 m.

Airborne and ground-based geophysical data and water-well data indicate that fresh groundwater exists in the study area. These resources are underlain by saline water and do not appear to be laterally or vertically extensive enough to support local high-volume production without accompanying water-level decline and salinity increase.

Conclusions

We combined airborne and ground-based geophysical studies with an analysis of water-well data to identify groundwater salinization in Sterling County, Texas, and favorable groundwater resources in El Paso County, Texas. In the Sterling study, airborne magnetic fi eld data reveal accurate locations of most oil and gas wells. Many magnetic anomalies more accurately represent well position than do locations obtained from agency records. Wells located nearly equidistant from fl ight lines spaced at 100 m produce anomalies that can be missed by airborne instruments. Magnetic anomalies caused by wells spaced closer than the fl ight-line spacing can appear as a single large anomaly.

735

Sterling airborne EM data show that ground conductivities are relatively low. Low conductivities in shallow, water-bearing strata indicate low TDS concentrations typical of the fresh to slightly saline values reported from most of the water wells. Local areas of elevated conductivity, particularly in and around oil fi elds, suggest that TDS values of groundwater have been increased locally by oil-fi eld activities. Airborne and ground-based geophysical surveys indicate that abandoned saltwater disposal pits have elevated conductivities indicative of salinization.

In the El Paso groundwater resource assessment, ground-based geophysical measurements demonstrated that (1) EM data are infl uenced by water content and water quality, (2) the saturated zone is within the exploration depth range of EM instruments, and (3) groundwater salinity increases with depth. El Paso magnetic fi eld and TDEM data helped delineate basinal faults that may infl uence groundwater fl ow and resource quality. Conductivity-depth slices produced from airborne TDEM data revealed prominent geological features above the water table. Below the water table, apparent conductivities correlate reasonably well to water quality for accurately located wells near fl ight lines. The most favorable groundwater resource areas are two low-conductivity zones identifi ed on conductivity-depth slices within the upper part of the water-saturated section (110 to 130 m). Fresh groundwater is present in the survey area but is unlikely to be extensive enough to support high-volume production without signifi cant water-level decline and salinity increase.

Acknowledgments

The Sterling County salinization project was funded by the Texas Water Development Board (TWDB) and Upper Colorado River Authority (UCRA) under contract number 2000-483-349. The Sterling County Underground Water Conservation District (SCUWCD) provided additional funds. Project direction was provided by Edward Angle at TWDB and Fred Teagarden at UCRA. Brent Christian at TWDB and Scott Holland and Bill Humble at SCUWCD assisted with fi eld studies. The El Paso County groundwater resource project was funded by the General Land Offi ce (GLO) under Interagency Cooperation Contracts No. 01-546R and 02-306R. Project direction was provided by Jeff Frank of the GLO. TWDB staff William Mullican, Robert Mace, and Brent Christian assisted in the project design and provided water well data. Airborne geophysical data were acquired and processed by Andrew Marshall, Leo Denner, Al Proulx, and Richard Williams at Fugro Airborne Surveys and Bruce Waines, Steven Christeas, Nick Craig, and Brendon Fisher at Voyageur Airways. Lana Dieterich edited the manuscript.

References

1. Albritton, C. C., Jr., and Smith, J. F., Jr., 1965, Geology of the Sierra Blanca area, Hudspeth County, Texas: U.S. Geological Survey Professional Paper 479, 131 p.

2. Alvarez, H. J., and Buckner, 1980, Ground-water development in the El Paso region, Texas, with emphasis on the resources of the lower El Paso Valley: Texas Department of Water Resources, Report 246, 346 p.

3. Ashworth, J. B., 1990, Evaluation of ground-water resources in El Paso County, Texas: Texas Water Development Board, Report 324, 25 p.

4. Ashworth, J. B., and Coker, Doug, 1993, Test hole investigation of a contamination site in Sterling County, Texas: Texas Water Development Board, Open File Report prepared for the Railroad Commission of Texas, 114 p.

5. Cliett, Tom, 1969, Groundwater occurrence of the El Paso area and its related geology, in Cordoba, D. A., Wengerd, S. A., and Shomaker, John, eds., Guidebook of the border region, Chihuahua and the United States: New Mexico Geological Society Twentieth Field Conference, p. 209-214.

736

6. Collins, E. W., and Raney, J. A., 1991, Tertiary and Quaternary structure and paleotectonics of the Hueco Basin, Trans-Pecos Texas and Chihuahua, Mexico: The University of Texas at Austin, Bureau of Economic Geology, Geological Circular 91-2, 44 p.

7. Collins, E. W., and Raney, J. A., 2000, Geologic map of west Hueco Bolson, El Paso region, Texas: The University of Texas at Austin, Bureau of Economic Geology, Miscellaneous Map No. 40, scale 1:100,000, 25-p. text.

8. Eifl er, G. K., Jr., and Barnes, V. E., 1974, San Angelo sheet: The University of Texas at Austin, Bureau of Economic Geology, Geologic Atlas of Texas, scale 1:250,000.

9. Frischknecht, F. C., Labson, V. F., Spies, B. R., and Anderson, W. L., 1991, Profi ling using small sources, in Nabighian, M. N., ed., Electromagnetic methods in applied geophysics—applications, part A and part B: Tulsa, Society of Exploration Geophysicists, p. 105-270.

10. Gates, J. S., White, D. E., Stanley, W. D., and Anderson, H. D., 1980, Availability of fresh and slightly saline ground water in basins of westernmost Texas: Texas Department of Water Resources, 108 p.

11. Gustavson, T. C., 1991, Arid basin depositional systems and paleosols: Fort Hancock and Camp Rice Formations (Pliocene-Pleistocene), Hueco Bolson, West Texas and adjacent Mexico: The University of Texas at Austin, Bureau of Economic Geology Report of Investigations No. 198, 49 p.

12. Hawley, J. W., Kottlowski, F. E., Seager, W. R., King, W. E., Strain, W. S., and LeMone, D. V., 1969, The Santa Fe Group in south-central New Mexico border region, in New Mexico Bureau of Mines & Mineral Resources Circular 104, p. 52-76.

13. Hefford, S., 2001a, Airborne magnetic and MEGATEM II electromagnetic multicoil survey, El Paso County: Fugro Airborne Surveys, Ottawa, Canada, Logistics and Processing Report, Job 744, 21 p.

14. Hefford, S., 2001b, Airborne magnetic and MEGATEM II electromagnetic multicoil survey, San Angelo, Sterling County: Logistics and Processing Report, Fugro Airborne Surveys, not consecutively paginated.

15. Henry, C. D., and Price, J. F., 1985, Summary of the tectonic development of Trans-Pecos Texas: The University of Texas at Austin, Bureau of Economic Geology, Miscellaneous Map No. 36, scale 1:500,000, 8-p. text.

16. Kaufman, A. A., and Keller, G. V., 1983, Frequency and transient soundings: Elsevier, Amsterdam, Methods in Geochemistry and Geophysics, No. 16, 685 p.

17. Knowles, D. B., and Kennedy, R. A., 1956, Ground-water resources of the Hueco Bolson, northeast of El Paso, Texas: Texas Board of Water Engineers Bulletin 5615, 265 p.

18. Mack, G. H., Kottlowski, F. E., and Seager, W. R., 1998, The stratigraphy of south-central New Mexico: in Mack, G. H., Austin, G. S., and Barker, J. M., eds., Las Cruces country II: New Mexico Geological Society Guidebook, 49th Field Conference, p. 135-154.

19. Mack, G. H., and Seager, W. R., 1990, Tectonic control on facies distribution of the Camp Rice and Palomas Formations (Pliocene-Pleistocene) in the southern Rio Grande rift: Geological Society of America Bulletin, v. 102, no. 1, p. 45-53.

20. Mattick, R. E., 1967, A seismic and gravity profi le across the Hueco Bolson: Texas Geological Survey Research 1967, Professional Paper 575-D, p. D85-D91.

21. Meyer, W. D., and Gordon, J. D., 1972, Development of ground water in the El Paso District, Texas, 1963-1970: Texas Water Development Board, Report 153, 51 p.

22. Muehlberger, W. R., and Dickerson, P. W., 1989, A tectonic history of Trans-Pecos Texas, in Muehlberger, W. R., and Dickerson, P. W., eds., Structure and stratigraphy of Trans-Pecos Texas: 28th International Geological Congress, Field Trip Guidebook T317, p. 35-54.

737

23. Mullican, W. F., III, Dutton, A. R., Mace, R. E., and Blum, Martina, 1997, Reevaluation of ground-water resources on State Lands in eastern El Paso county: The University of Texas at Austin, Bureau of Economic Geology, topical report prepared for the Texas General Land Offi ce, 17 p.

24. Paine, J. G., 2002, Airborne geophysical assessment of salinization in the Lacy Creek area, Sterling County, Texas: The University of Texas at Austin, Bureau of Economic Geology, report prepared for the Upper Colorado River Authority under contract number 2000-483-349, 55 p.

25. Paine, J. G., and Collins, E. W., 2002, Evaluating potential groundwater resources on State Lands in El Paso County, Texas using airborne geophysics: The University of Texas at Austin, Bureau of Economic Geology, report prepared for the General Land Offi ce under contract no. 02-306R, 87 p.

26. Paine, J. G., Dutton, A. R., and Blum, M., 1999, Using airborne geophysics to identify salinization in West Texas: The University of Texas at Austin, Bureau of Economic Geology Report of Investigations No. 257, 69 p.

27. Parasnis, D. S., 1973, Mining geophysics: Amsterdam, Elsevier, 395 p.28. Parasnis, D. S., 1986, Principles of applied geophysics: Chapman and Hall, 402 p.29. Ramberg, I. B., Cook, F. A., and Smithson, S. B., 1978, Structure of the Rio Grande rift in southern

New Mexico and West Texas based on gravity interpretation: Geological Society of America Bulletin, v. 89, no. 1., p. 107-123.

30. Renfro, W. C., 1993, Water well contamination investigation on the Price Ranch, Sterling County, Texas: Railroad Commission of Texas, 23 p.

31. Robinove, C. J., Langford, R. H., and Brookhart, J. W., 1958, Saline-water resources of North Dakota: U.S. Geological Survey, Water-Supply Paper 1428, 72 p.

32. Seager, W. R., 1980, Quaternary fault system in the Tularosa and Hueco Basins, southern New Mexico and West Texas, in Dickerson, P. W., Hoffer, J. M., and Callender, J. F., eds., Trans-Pecos region, southeastern New Mexico and West Texas: New Mexico Geological Society Guidebook No. 31, p. 131-135.

33. Spies, B. R., and Frischknecht, F. C., 1991, Electromagnetic sounding: in Nabighian, M. N., ed., Electromagnetic methods in applied geophysics—applications, part A and part B: Tulsa, Society of Exploration Geophysicists, p. 285-386.

34. Strain, W. S., 1966, Blancan mammalian fauna and Pleistocene formations, Hudspeth County, Texas: Texas Memorial Museum Bulletin No. 10, 55 p.

35. Strain, W. S., 1971, Late Cenozoic bolson integration in the Chihuahua tectonic belt, in Hoffer, J. M., ed., Geologic framework of the Chihuahua tectonic belt: West Texas Geological Society Publication 71-59, p. 167-173.

36. Stuart, C. J., and Willingham, D. L., 1980, Late Tertiary and Quaternary fl uvial deposits in the Mesilla and Hueco Bolsons, El Paso area, Texas: Sedimentary Geology, v. 38, no. 1/4, p. 1-20.

37. U.S. Census Bureau, 2003, www.census.gov.38. West, G. F., and Macnae, J. C., 1991, Physics of the electromagnetic induction exploration method,

in Nabighian, M. N., ed., Electromagnetic methods in applied geophysics—applications, part A and part B: Tulsa, Society of Exploration Geophysicists, p. 5-45.

738