Geohydrologic and water-quality characterization of a fractured-bedrock test hole in an area of Marcellus Shale gas development, Sullivan County, Pennsylvania

The stratigraphy, water-bearing zones, and quality of groundwater were characterized in a 1,400-ft-deep test hole drilled during 2013 in fractured bedrock in Sullivan County, Pa., by collection and analysis of measurements made during drilling, geophysical logs, and depth-specific hydraulic tests and water samples. The multidisciplinary characterization of the test hole was a cooperative effort between the Pennsylvania Department of Natural Resources, Bureau of Geological Survey (BGS), and the U.S. Geological Survey (USGS). The study provided information to aid the bedrock mapping of the Laporte 7.5-minute quad-rangle by BGS to help quantify the depth and character of fresh and saline groundwater in an area of shale-gas exploration (described in this report), which could help gas operators protect groundwater resources.

The Laporte test hole was drilled with air-hammer methods in an upland setting in the headwaters of Loyalsock Creek in the Glaciated High Plateau section of the Appalachian Plateaus physiographic province. Bedrock residuum and till were penetrated from land surface to 8.5 ft, the Huntley Mountain Formation of Mississippian and Devonian age was penetrated from 8.5 to 540 ft, and the Catskill Formation of Devonian age was penetrated from 540 to 1,400 ft. Fractures, determined from optical televiewer, acoustic televiewer, and video logs, were commonly encountered to 200 ft bls (below land surface), then decreased exponentially with depth, except at a highly fractured zone from 637 to 644 ft bls. Most fractures were along bedding planes and had a strike of about 243 degrees and dip about 4 degrees to the northwest, consistent with the test-hole location on the north limb of the Muncy Creek anticline. Few fractures were noted below 650 ft.

The depths of fresh and saline water-bearing fracture zones were identified in the test hole by geophysical-log analysis and were verified by pumping samples from zones isolated with packers and by collecting samples in the open hole with a wire-line point sampler. Six water-bearing zones associated with single or multiple fractures were identified at depths of 130–135, 180, 267–275, 425, 637–644, and 1,003 ft bls. Under ambient conditions, fresh water entered the hole from fractures at 130-135 and 180 ft bls, flowed downward and exited at fractures from 267–275, 425, and 637–644 ft. When pumped at 16.2 gal/min, most of the water from the open test hole was contributed from the fracture at 180 ft bls. Transmissivity, estimated from analysis of the specific-capacity data and flowmeter logs, is about 850 ft²/d for the entire open hole, and about 60 percent of the transmissivity is contributed from the fracture zone at 180 ft bls. The hydraulic heads in the deep water-bearing zones at 425 and 637–644 ft were about 100 ft lower than hydraulic heads in shallow water-bearing zones at 180 ft bls and above, indicating a large downward vertical hydraulic gradient.

Water samples pumped from fracture zones isolated by packers at and above the water-bearing zone at 450 ft bls were fresh with dissolved-solids contents of 105 mg/L or less. The sample isolated at 637–644 ft bls was probably affected by leakage around packers, but the specific-conductance samples collected during drilling that were believed to be representa-tive of the fracture zone at 637–644 ft bls indicated slightly saline water. Below the 637–644 ft zone, a flowmeter log in the open hole did not detect any vertical flow, and the temperature log approached the geothermal gradient, indicating little ambient fluid flow and minimal fracture transmissivity below this depth. A petrophysical-log analysis using estimates of formation water resistivity from Archie’s Equation indicated an apparent transition from fresh to saline water in the sandstones occurs between 450 to 900 ft bls, with saline water indicated below 900 ft.

Small seeps of saline water were delineated at 958, 989, and 1,003 ft bls by a time series of specific-conductance logs, and a discrete-point water sample at 990 ft bls with total dissolved-solids concentration of 19,900 mg/L verified that highly saline water was present below 900 ft bls. Occurrence of saline water at a depth of about 900 ft bls is below altitude of streams within 3 to 5 miles of the test hole but is about 930 ft above the altitude at the mouth of Loyalsock Creek where is enters the West Branch Susquehanna River at Montours-ville, Pa. The depth to saline water in this test hole is close to depths estimated at two other deep test holes drilled by the BGS in upland settings in Bradford and Tioga Counties in north-ern Pennsylvania.

The saline water from 990 ft bls had a chemical composition similar to Appalachian Basin brines that had been diluted with fresh water. Predominant ions in the saline water were sodium, chloride, and calcium. Trace constituents of strontium, bromide, barium, lithium, and molybdenum were all more than 5,000 times greater than in freshwater samples from 167 or 270 ft bls. Methane concentration in the saline water sample from 990 ft was 120 mg/L. The concentration ratios of methane to higher-chain hydrocarbon gases and isotopic ratios of ¹³C/¹²C and ²H/¹H of methane indicate that the gases are likely of thermogenic origin. In the sample from 990 ft bls, the ¹³C/¹²C of methane was less negative (-34.81 per mil) than ¹³C/¹²C of ethane (-37.1 per mil). Isotopic reversals such as this are generally found in gases from rocks older than the Catskill Formation, so its recognition in a natural upland setting at relatively shallow depth could be important when interpreting isotopic results to identify the origin of stray gas in the area.