Geophysical Investigation of the Arbuckle-Simpson Aquifer, Southern Oklahoma

Erin K. Lewallen, Kumar Ramachandran, and  Bryan Tapp

SUMMARY

The Arbuckle-Simpson aquifer in Johnston County, Oklahoma is investigated by near-surface geophysical surveying. This study seeks to understand the subsurface of a small area of this carbonate aquifer by investigating how faults affect groundwater hydrology. Electrical resistivity soundings were conducted, and using a combination of resistivity curve-fitting and inversion analysis, electrical properties and depth to the groundwater table at each sounding have been estimated. The resistivity data indicate a water table deepening to the south across the northern fault of the Mill Creek block. This is independently corroborated by groundwater well data from the area and supports the interpretation that a southward-dipping fault may confine the aquifer beneath the Mill Creek block. The resistivity data enable formulation of a possible hydro-geologic model of the subsurface.

INTRODUCTION

The Arbuckle-Simpson aquifer in southern Oklahoma is a highly fractured karst carbonate aquifer, and the control of faults and fractures on subsurface fluid movement is not well-understood. The purpose of this geophysical study is to investigate the possible fault control in a small area of the aquifer focused around Pilot Springs (Figure 1) to characterize these faults and deter-mine whether they act as barriers or conduits for groundwater movement. The site chosen for detailed investigation is a fault-bounded carbonate system located near Pilot Springs and the town of Mill Creek, Johnston County, in the southwest region of the Hunton Anticline (Figure 1). This area is comprised primarily of Ordovician limestones, dolomites, shales, and sand-stones of the Simpson and Arbuckle groups. The Simpson Group Oil Creek and Joins (Ooj) Formation is characterized by basal sandstone with overlying limestone and shale. The Cool Creek and McKenzie Hill (Ocm) and the West Spring Creek and Kindblade (Owk) formations are part of the Arbuckle Group and consist predominantly of limestones, dolomites, and thin sandstones.

The primary purpose of this study is to determine the extent to which resistivity can image the subsurface in this portion of the Arbuckle-Simpson aquifer. By imaging the subsurface, we hope to determine how faults and fractures control the behavior of groundwater flow around Pilot Springs, the geometry of the northern fault bounding the Mill Creek block, and the ex-tent to which the surficial geologic map of the area matches geophysical observation. These goals have significance in furthering both geologic and structural knowledge of this region and in better understanding the aquifer hydrodynamics.

Figure 1: Location Map showing Resistivity Sounding Points and Groundwater Well Locations. Map was created in ArcView using GIS data layers obtained from the Oklahoma Water Resources Board Web site.

RESISTIVITY DATA ANALYSIS

Five resistivity soundings were performed in different locations within the study area (Figure 1). Soundings 1 and 2 were performed using a Wenner array configuration, while Soundings 3, 4, and 5 used a Schlumberger array configuration.

Analysis of the resistivity sounding data was carried out by means of manual two-layer curve-fitting as well as IPI2win public domain software, created and published by Moscow State University. The manual two-layer curve-fitting was used to estimate the resistivity values of the upper two layers and the depth to the first interface for each sounding. The IPI2win program uses measured resistivity values to construct a one-dimensional inverse model of the subsurface at a sounding point. It does this through a regularized inversion algorithm. The problem posed by inverse modeling is a nonlinear one; because the number of unknown variables exceeds the number of equations, the problem is underdetermined and an infinite number of solutions can be fit to the data. The problem must therefore be regularized by imposing external constraints based on the initial model structure. Model parameters are chosen to give the smoothest model: the model that contains the fewest features necessary to fit the data. Beginning with the input data, the program performs a series of iterations in which it narrows down the possible models to a best-fit model of maximum smoothness.

Figure 2: Inversion Models for Soundings 1 to 5 are shown in plots ‘a’ to ‘e’. The y-axis represents resistivity, while the x-axis represents electrode spacing. The black curve represents the data line, the blue line represents the final model, and the red curve represents the forward response of the model.

P1S1 (Sounding 1) Analysis

Sounding 1 was performed using a Wenner array configuration with a maximum outer electrode spacing of 900 ft., or 274.3 m. The theoretical survey depth is estimated to be 34 m. This sounding is best fit by a two-layer model of the subsurface.

P1S2 (Sounding 2) Analysis

Sounding 2 was performed using a Wenner array configuration with a maximum electrode spacing of 1275 ft., or 388.6 m. The theoretical survey depth is estimated to be 49 m. This sounding is best fit by a three-layer subsurface model.

P1S3 (Sounding 3) Analysis

Sounding 3 was performed using a Schlumberger array configuration with a maximum electrode spacing of 1500 ft., or 457.2 m. The minimum theoretical survey depth is estimated to be 57.2 m. This sounding is best fit by a three-layer subsurface model.

P1S4 (Sounding 4) Analysis

Sounding 4 was performed using a Schlumberger array configuration with a maximum electrode spacing of 1800 ft., or 548.6m. The minimum theoretical survey depth is estimated to be  68.6 m. This sounding is best fit by a four-layer subsurface model.

P1S5 (Sounding 5) Analysis

Sounding 5 was performed using a Wenner array configuration with a maximum electrode spacing of 1200 ft., or 365.8 m. The minimum theoretical survey depth is estimated to be 45.7 m. This sounding is best fit by a four-layer subsurface model.

RESULTS

The five resistivity sounding curves obtained are interpreted as being caused by two-, three-, or four-layer subsurface structure. .

Table 1: Final Resistivity Model Summary

	Resistivity and Depth to base of hn
P1S1		ρ1 =80Ωm ρ2 =240 −270Ωm	h1 =1.5m
P1S2		ρ1 =20 −25Ωm         ρ2 =800 −1900Ωm ρ3 =25 −30Ωm	h1 =0.5 −1m  h2 =7 −14m
P1S3		ρ1 =40 −45Ωm                  ρ2 =2100 −5000Ωm ρ3 =70Ωm	h1 =3 − 4m    h2 =35 −55m
P1S4		ρ1 =70 −160Ωm          ρ2 =1400 −5700Ωm ρ3 =25 −55Ωm ρ4 =10000Ωm	h1 =3 −4m    h2 =35 −55m   h3 =36 −70m
P1S5		ρ1 =225Ωm           ρ2 =1000 −1200Ωm ρ3 =40 −55Ωm ρ4 =10000Ωm	h1 =6 −7m       h2 =11 −12m h3 =20m

Figure 3: Diagram of Subsurface Resistivity Models Relative to Surface Geology. Depth and resistivity values for each sounding are averaged from Table 1.

Figure 4: Model of Fault Geometry Relative to Resistivity Soundings. See Figure 3 for scale and resistivity color key.

For all five soundings, the first resistivity layer is characterized by relatively low resistivity values and thicknesses. This layer in each case can be interpreted as representing the variable properties of the very shallow subsurface, including the dirt and gravel roads beside which these soundings were taken as well as an upper soil layer with relatively high water con-tent, most likely due to rainwater accumulation.

The high resistivity values of the second layer in the case of Soundings 2, 4, and 3 represent a dry rock unit. These values are generally consistent with the range expected of limestones and dolomites, the predominant rock types in this area.

The significantly lower third layer resistivity values for Sound-ings 2, 4, and 3 indicate high water content in these rocks; these layers may represent the same rock unit as the second layer saturated with water or a different water-saturated rock unit. Because resistivity properties depend on many factors and can vary widely within a given rock type, the resistivity method cannot be used to uniquely determine lithology, and thus neither interpretation can be ruled out. However, in either case, the significant drop in resistivity at this interface between the second and third layers probably reflects the approximate depth to the groundwater table at these sounding points.

The contrast in the resistivity models of Soundings 3 and 5 (Figures 3 and 4) is also significant. The difference in prop-erties between these soundings suggests the presence of a dis-continuity not identified on the surficial geologic map. This may reflect the presence of an unmapped fault or a lithostrati-graphic facies change.

Correlation With Well Data

To independently confirm the interpretation of groundwater level based on resistivity, water levels from monitored ground-water wells in the study area were examined; data for these wells are available on the Oklahoma Water Resources Board Website. Figure 1 displays the locations of groundwater wells with available water level information.

The two northernmost wells, 94677 and 29279, are located in the Oil Creek and Joins (Ooj) Formation. Recent water level information for well 29279 is not available, but the depth at which the driller first encountered water in the year 1991 was approximately 9.1 m. Water depth at well 94677 was measured at 7.5 m in June 2005 and 8.7 m in June 2006. Although farther north than Sounding 2, these wells show a relatively shallow groundwater table in this area, consistent with the depth (7-14 m) to the low-resistivity layer at this sounding.

Farther south, Sounding 4 is located close to well 92340. The only water level information found for this well is that in February 2005, no water was encountered at a depth of approximately 13.1 m. This is consistent with the low-resistivity layer of Sounding 4 occurring no shallower than 12 m. Well 92339 is located between Soundings 4 and 3 and approximately a mile to the east. Here, water depth was measured at 25.8 m in June 2005 and 24.9 m in June 2006. Water depth at this well is shallower than the low-resistivity layer range offered for Sounding 3 (35-55 m), but it lies within the range of that of Sounding 4 (12-34 m). Data from these two wells is thus consistent with the resistivity interpretations and confirms the trend found in the resistivity soundings of increasing aquifer depth toward the south.

Finally, the southernmost three wells lie in the Ooj Formation, west of Sounding 1. No recent water level data could be found for these wells; however, the water table measured at these wells in the 1970s was between 3 and 6 m. Although this data is relatively old and must be used with caution, it is consistent with the interpretation at Sounding 1 of a very shallow low-resistivity layer. It also provides an interesting correlation with the relatively shallow groundwater depth found in the northern Ooj Formation in both the well and the resistivity data.

Overall, the water table depths available from well data in the study area correlate well with resistivity data; this independent corroboration allows considerable confidence to be placed in the resistivity groundwater interpretation.

The interpretation derived from the resistivity data has implications for the nature of both Pilot Springs and Pennington Creek. The relatively low pH of the water emanating from the springs and locations along the creek is not consistent with a carbonate origin, and instead implies a silicate source. If the deepening low-resistivity layers of the northern three sound-ngs reflect a southward dip to this northern fault zone, the presence of the Simpson Group Oil Creek and Joins Formation beneath the Mill Creek block would provide such a silicate source (Figures 3 and 4). Fractures or faults extending down to the main thrust could provide conduits along which water that would otherwise be confined to the deeper Simpson Group aquifer could ascend through the overlying Arbuckle Group formations, emerging at the surface to produce Pilot Springs and feed Pennington Creek.

CONCLUSIONS

Electrical resistivity surveys carried out in the Arbuckle-Simpson aquifer near Pilot Springs provide insight into both the hydro-geologic and the structural environment present in this area of the aquifer. The trend of a deepening low-resistivity layer from Sounding 2 in the north to Sounding 3 farther south (Figures 3 and 4) likely reflects a deepening groundwater table, which is consistent with a south-dipping northern fault delineating the overlying Arbuckle Group formations from an underlying confined Simpson Group aquifer. Within this interpretation, the Mill Creek block is likely an isolated system, interacting with the northern silicate-based aquifer only through dissolution faults and fractures. Such features might allow water to ascend directly from the underlying aquifer to the surface. This could account for the presence of Pilot Springs and Pennington Creek and the properties of the water emanating from them, which imply a silicate rather than a carbonate source.