Geological Survey Water-Supply Paper 1475-D Geology and Occurrence of Ground Water at Jewel Cave National Monument, South Dakota

On the basis of preliminary field investigations, a test hole was drilled in the NW1/4SE1/4NE1/4 sec. 2, T. 4 S., R. 2 E. to determine the lithologic and water-bearing characteristics of the strata beneath the monument. Lack of subsurface information and other considerations indicated that the test hole might as well be close to the service area and be of sufficient diameter to permit finishing as a well in the event water-bearing rooks were penetrated. A test hole 8 inches in diameter and suitable for casing with heavy-duty 6-inch pipe was begun May 6, 1959, and completed on August 25, 1959, at a total depth of 700 feet. The location of the test well With respect to the service area is shown on figure 25.

The test hole penetrated 432.5 feet of Pahasapa limestone before reaching the underlying Englewood limestone. As revealed by the detailed long on page 151 the Pahasapa consists chiefly of finely crystalline to dense limestone and dolomite, and thin layers of chert and clay at irregular intervals. The 20 feet of clay and large calcite crystals in the interval 170 to 190 feet apparently represents a clay-filled extension of Jewel Cave.

Although some scattered intergranular porosity occurs in the limestone and dolomite of the Pahasapa, the formation contains very little water at the monument. The interval 390 to 400 feet supplied barely enough water (about 100 gallons in 12 hours) for the first drilling run each day.

From 435 to 475 feet the drill penetrated finely crystalline pink to maroon Englewood limestone. Virtually no porosity was observed in any of the cuttings from this interval, and the upper 10 feet of the formation contained appreciable red silt and clay. The Englewood limestone is not considered to be an aquifer at the site of the test hole.

Directly beneath the Englewood in the interval 475 to 485 feet, the drill penetrated 10 feet of clean white sandstone containing sand grains that were poorly sorted, and well rounded to subrounded. The upper few feet of the sandstone was coarser grained and softer than the lower; consequently, drilling progress was more rapid in the upper part than in the lower.

A significant amount of water entered the hole from this 10-foot sandstone interval. After standing overnight, the measured static water level was 389.6 feet. After the sandstone had been completely penetrated at 485 feet, the hole was bailed for 44 minutes at an average rate of about 12.5 gpm with a maximum drawdown of about 87 feet. The water level at the end of 44 minutes of bailing was about 476 feet. Fifteen minutes after bailing stopped the water level had recovered to 423 feet.

Before the bailing test was started a sample of water was tested for pH, iron, chloride, and hardness with a water-testing kit designed for field use. Results of the testing are approximate and are as follows:

In the interval 485 to 665 feet the drill penetrated 180 feet of dolomite, siltstone, limestone, and sandstone comprising the Deadwood formation. The uppermost 115 feet of the Deadwood consists of brown and green limestone and dolomite with siltstone at 500 and 575 feet. No porosity was observed in any samples of limestone and dolomite of the Deadwood and they apparently contained little or no water.

The lower part of the limestone in the Deadwood was very compact at 593 feet as indicated by the description of the sample from this depth. The driller reported drilling through a "boulder" from 593 to 595 feet.

The upper 5 feet of the sandstone section of the lower part of the Deadwood was also hard and apparently was not water bearing. Of the 65 feet of sandstone in the lower part of the Deadwood only about 15 to 20 feet appeared clean and permeable enough to yield water. The samples from the sandstone intervals 630 to 635 feet and 640 to 650 feet were moderately clean and moderately well cemented; most of the water yielded by the Deadwood formation probably entered the hole from these two intervals.

After the entire thickness of the Deadwood formation had been penetrated at 665 feet, test drilling was continued for an additional 35 feet into the underlying Precambrian quartz-biotite schist. This schist is compact and, with the exception of the upper few feet which is generally weathered, is not an aquifer.

Upon completion of the test hole, a second bailing test was made. The static water level at the beginning of the test was about 390 feet, about the same as that for the sandstone at 475 to 485 feet. Thus, leakage from one zone to another probably is insignificant. After bailing for 70 minutes at an average rate of 14 gpm the water level was about 500 feet; a drawdown of about 110 feet. Because of the wet condition of the hole and the relatively great depth to water, accurate measurements of the water level during recovery were not possible. Measurement by use of the bail line, however, indicated that recovery was slow. About 20 hours was required for the water level to recover to 390 feet.

The test hole was cased with 700 feet of 6-inch (outer diameter) heavy-duty pipe. Slots were cut in the pipe in a regular pattern and were set opposite the sands at 475 to 485 feet and 605 to 665 feet. In order not to miss any possible thin water-bearing strata the casing was perforated from 475 to 495 feet and from 605 to 670 feet. Perforations were thus set opposite the uppermost 5 feet of Precambrian schist and all except the uppermost 5 feet of the sandstone in the lower part of the Deadwood formation. The casing was not perforated opposite the minor water-bearing zone in the Pahasapa limestone at 390 to 400 feet.

After the test well was completed, a cylinder pump was inserted in the well. During September 22-26, 1959, the well was tested at different pump settings for short intervals at rates that ranged from 10 to 26 gpm. The longest period of pumping that the contractor was able to maintain was 18 hours beginning at 5:35 p.m. on September 25. During this period the pumping rate ranged from 13 to 20 gpm and averaged 16 gpm. Drawdown and recovery data, necessary to predict a sustained yield for the well, were not obtainable because access to the water surface was prevented by the pump rods.