CONTENTS

Abstract
Introduction 

Purpose and Scope
Hydrogeology
Acknowledgments 

Water-Level Measurements

Quality-Assurance Flags 
Temperature Effects 

Sources of Water-Level Fluctuations 

Precipitation
Barometric Pressure and Earth Tides
Seismic Events and Underground Nuclear Tests
Pumpage

Analysis of Water Levels
Summary
References Cited
Appendix 1
Appendix 2

SOURCES OF WATER-LEVEL FLUCTUATIONS

Water-level fluctuations in wells on Pahute Mesa are caused by a number of natural and human factors. Natural factors include infiltration of precipitation, barometric pressure, Earth tides, and seismic events caused by tectonic activity. Human factors include ground-water pumpage, underground nuclear testing, and recovery from drilling effects. Some of these factors, such as infiltration of precipitation have relatively slow response times but may cause long-term changes in water levels. Other factors, such as seismic events, are relatively instantaneous but generally have no lasting effect on water levels.

Precipitation

The climate at the NTS is arid to semi-arid. Precipitation is controlled largely by altitude with greater amounts of precipitation falling in high-altitude regions. Precipitation monitoring stations maintained by the National Oceanic and Atmospheric Administration (NOAA) are located on Pahute Mesa (PM1)⁶, at an altitude of 6,550 ft, and on Rainier Mesa (A12), at an altitude of 7,490 ft (fig. 2) (Douglas Soule', National Oceanic and Atmospheric Administration, written commun., 1999). Mean annual precipitation at the Rainier Mesa station (13.2 in/yr) was about 1.7 times greater than at the Pahute Mesa station (7.7 in/yr) for 1964-94. Annual precipitation and cumulative departure from normal precipitation at these two stations for 1964-98 are shown in figure 3. The plot of cumulative departure from normal precipitation on Pahute Mesa shows excess precipitation from 1982 to 1984 (cumulative departure line has positive slope), variable precipitation from 1985 to 1987, and in general, a precipitation deficit from 1988 to 1996.

D'Agnese and others (1997, p. 54-55) mapped large parts of Pahute and Rainier Mesas that are within the boundaries of the NTS as having moderate ground-water recharge potentials; recharge in these areas was estimated to be 7 to 25 percent of precipitation using a modified empirical precipitation-recharge relation developed by Maxey and Eakin (1949). Other areas near Pahute Mesa mapped with moderate to high recharge potentials include Kawich Range to the north, Belted Range to the northeast, and Timber and Shoshone Mountains to the south. Much of the surrounding area that comprises the Death Valley regional ground-water flow system is assumed to have little or no recharge potential.

Depth to water beneath Pahute Mesa is typically 1,800 to 2,400 ft below land surface. Considering the substantial depth to water, short-term seasonal fluctuations in precipitation on Pahute Mesa are not likely to have an effect on water levels. Long-term fluctuations in precipitation (such as multiple years of below-average precipitation) on Pahute Mesa and on recharge areas to the north may affect water levels. In shallow alluvial aquifers in east-central Nevada, water levels have shown a response to long-term (10 years) drier- or wetter-than-normal periods of precipitation (Dettinger and Schaefer, 1995). In deeper aquifers (greater than 1,000 ft below land surface), water levels also may show evidence of responding to drier- or wetter-than-normal periods of precipitation. On the east side of the NTS, water levels in the regional Paleozoic carbonate aquifer may correlate, after a lag time of about 3 years, to departures from normal precipitation (Daniel Bright, U.S. Geological Survey, written commun., 1999). Southwest of Pahute Mesa, at Yucca Mountain, Lehman and Brown (1996) suggested precipitation as a possible cause of apparent cyclic water-level fluctuations in wells penetrating volcanic rocks at depths of 1,200 to 4,000 ft.

Barometric Pressure and Earth Tides

Changes in barometric pressure and Earth tides affect water levels in wells screened in confined aquifers under Pahute Mesa. Most of the change in water level from barometric fluctuations is caused by changes in air pressure translated down the open well rather than through the unsaturated zone. Typically, an increase in barometric pressure will cause a lower water level, and a decrease will result in a higher water level. This relation is clearly illustrated in continual water levels from well PM-2, in which the barometric pressure is almost a mirror-image of the water level (fig. 4). Nearly all the short-term water-level fluctuations in this well, which are typically several tenths of a foot in magnitude, are caused by changes in barometric pressure.

Seasonal differences in barometric pressure can affect water levels, resulting in higher water levels in the winter and lower levels in the summer. In addition, the magnitude of short-term fluctuations in water levels caused by barometric pressure tend to be greater in the winter than in the summer. Long-term (10-year) trends in water levels, however, are not likely to be caused by barometric pressure. At a meteorological site near Mercury on the south end of the NTS, barometric pressure did not substantially change from 1986 to 1996 (Daniel Bright, U.S. Geological Survey, written commun., 1999).

Earth tides are caused by the forces exerted on the Earth's surface by the Moon and the Sun. Changes in ground-water level resulting from Earth tides are actual diurnal fluctuations of the head in the aquifer. As a result of Earth tides, water levels will peak near moonrise and moonset, and be lowest near the upper and lower culmination of the Moon (Ferris and others, 1962). The effects of Earth tides on water levels in well PM-2 are evident on the water-level plot (fig. 4) that was corrected for the effects of barometric pressure. (Water levels were corrected for barometric-pressure changes using a method outlined by Brassington, 1988, p. 81-84.) Earth tides cause water levels in well PM-2 to fluctuate several hundredths of a foot, which is about an order of magnitude less than fluctuations caused by barometric pressure.

Seismic Events and Underground Nuclear Tests

During the last 50 years, earthquakes on Pahute Mesa have been caused by nuclear tests and tectonic activity. Through September 1992, 828 underground nuclear tests were detonated at the NTS (U.S. Department of Energy, 1994). Of these, 85 were beneath Pahute Mesa (table 4) and 62 were beneath Rainier Mesa. Nuclear tests beneath Pahute Mesa were detonated between 1965 and 1992 in volcanic rocks; most were near (within two cavity radii) or below the water table. Most large-yield tests (200 kilotons and larger) done at the NTS were beneath Pahute Mesa (Laczniak and others, 1996). Tests beneath Rainier Mesa were typically small (less than 20 kilotons) and above the water table. In addition to earthquakes caused by nuclear testing, natural earthquakes have occurred on Pahute Mesa, which lies in the tectonically active Basin and Range Physiographic Province.

From 1950 to 1998, approximately 350 earthquakes of magnitude 4-5, 155 earthquakes of magnitude 5-6, and 19 earthquakes of magnitude 6 or greater occurred within about a 70-mi radius of the NTS (University of Nevada, Reno, Seismological Laboratory, 1998). Most of these recorded earthquakes, especially those greater than 5, occurred as a result of nuclear tests in the underground test areas of the NTS. Of the 19 recorded earthquakes with a magnitude greater than or equal to 6.0, 17 coincide with nuclear detonations on Pahute Mesa (table 4).

Earthquakes caused by tectonic or nuclear-test activity result in energy waves being propagated through the ground. As the series of waves pass through the earth, ground-water levels may be temporarily affected. This phenomenon was documented at Yucca Mountain following a 7.5-magnitude earthquake that occurred in 1992 in Landers, Calif., and a 5.6-magnitude earthquake, occurring a day later, under Little Skull Mountain at the southern end of the NTS--the largest recorded natural earthquake within the NTS boundary (O'Brien, 1993, p. 9). Small to moderate fluctuations (several inches to 3 ft) in ground-water levels lasting for less than an hour to as much as 6 months were recorded. The long-term changes in water levels near Yucca Mountain were believed to be caused by a change in the regional strain field from the nearby Little Skull Mountain earthquake. Because earthquakes typically only cause small short-term fluctuations in water levels, wells that are monitored infrequently (monthly or less often) will rarely show evidence of these fluctuations. Only wells with continual water-level recorders, such as well PM-2 which has 2 years of record, are likely to document an earthquake. On Pahute Mesa, no observed water-level fluctuations were attributed to an earthquake caused by tectonic activity.

The magnitude and duration of changes in water levels and hydraulic properties caused by nuclear tests under Pahute Mesa are affected by the size of, and distance from, a test. Wells in close proximity to a test may have more pronounced and long-lasting effects. Close to a detonation, changes in rock permeability, storativity, fluid pressures, and temperature can cause large changes in water levels that sometimes last for years before equilibrium is once again reached (Laczniak and others, 1996). In the case of well U-19v PS 1D, which is a re-entry hole into the U-19v test-hole cavity caused by the June 6, 1973, Almendro test (table 4), the water level declined in excess of 800 ft shortly after the test as a result of expulsion of ground water from the test cavity. The water level rose about 200 ft above pre-shot levels following the test, rising 1,029 ft from September 1973 to September 1998 (see well hydrograph U-19v PS 1D in appendix 1) as a result of slow infilling of the test cavity. Further from a detonation, water levels are affected less but changes still can be relatively large. For example, the Benham nuclear test registered as a magnitude 6.4 earthquake (U-20c, table 4). Approximately a month following the test, the water level in well UE-20f (about 3 mi from the shot hole) was still elevated about 50 ft (fig. 5). Six years later, the water level remained 17 ft above the pre-shot water level. However, with two additional large detonations (Jorum and Handley) occurring in the vicinity of well UE-20f within 1.5 years of the Benham test, the water-level pattern shown in figure 5 is more complicated than presented. Sustained effects on water levels in well UE-20f from the Jorum nuclear test were noted by Dudley and others (1971, p. 49). Dudley and others also documented short-term (4 hours) oscillations in the confined fluid pressure in well UE-20f after the Handley test. Amplitudes of fluid-pressure oscillations, measured in feet of water, were from 15 to more than 300 ft. After 2 months, the pressure response indicated the water level was still elevated about 4 ft; however, this was reported to be in the range of instrument error.

Because most of the wells on Pahute Mesa do not have water-level records going back to the 1960's and 1970's, determining how many water levels might be affected by past nuclear tests is difficult. Without pre-test baseline water levels, it is extremely difficult to determine if a water level might be elevated several feet to more than 100 ft because of a past nuclear test. Also, without baseline data, distinguishing between a declining water-level trend that is the result of a 20- to 30-year-old nuclear test and a decline caused by some other factor is not easy. As an example of the difficulty in determining the cause of an anomalously high water level, well U-19ab has a water level that is elevated about 250 ft above regional levels. This water level may be from a shallow localized flow system or from a perched aquifer, in which water levels are naturally elevated above the regional flow system. High water levels might be expected in a well such as U-19ab, which only penetrates about 225 ft of saturated material, and which, except for a 10-ft interval of densely welded tuff, consists exclusively of low-permeability non-welded tuffs in the lower 700 ft of the well. These tuffs may impede recharge water moving downward into the regional aquifer, which would result in steep downward hydraulic gradients and a naturally elevated water level. Alternatively, the water level in well U-19ab, which was drilled in 1979, may be elevated because of the nearby (0.25 mi) Pool nuclear test, which was detonated in 1976. Other wells with elevated water levels that are within 1 mi of a large nuclear test (resulting in a magnitude 6.0 or greater earthquake) include U-19aq, U-19az, U-19bh, U-19v PS 1D, and U-20bc. Water levels in many wells on Pahute Mesa, however, are not elevated and within 1 mi of a large nuclear test.

Pumpage

Seven wells (designated with a "WW" in their name) on or near Pahute Mesa have been used for water withdrawal at the NTS. Total annual and monthly withdrawal data from 1983 to 1998 for the Pahute Mesa area are shown in figure 6. Withdrawals from 1963 to 1998 (excluding 1972-82 for which no data is available) for the three largest-producing wells are shown in figure 7 (D.B. Wood, U.S. Geological Survey, written commun., 1999). These three wells, WW-8, UE-19c WW, and U-20 WW, have supplied most of the water for Pahute Mesa activities. The remaining four wells--UE-19e WW, UE-19gS WW, U-20a 2 WW, and UE-20h WW--were only pumped for 1 to 4 years and have not been used since 1967. Total withdrawal from these wells was approximately 125 million gallons (Wood and Reiner, 1996, table 3; D.B. Wood, U.S. Geological Survey, written commun., 1999).

Between 1983 and 1998, approximately 2.29 billion gallons (7,030 acre-feet) of water were withdrawn from volcanic aquifers on or near Pahute Mesa. Water use generally rose from 1983 to 1989; peak annual use was 305 million gallons in 1989 (fig. 6). From 1989 to 1998, water use has generally declined. A large decline in 1993 marked the first full year after a ban on nuclear testing at the NTS. In 1998, water use was only 10 percent of peak use in 1989.

⁶ The precipitation station, PM1, is not at the same location as well PM-1.