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

WATER-LEVEL MEASUREMENTS

Periodic measurements of water levels in wells on Pahute Mesa have been collected by the USGS since the early 1960's and are maintained in the USGS National Water Information System (NWIS) database. Hydrographs showing periodic measurements of water levels from 65 wells on or near Pahute Mesa are presented in appendix 1; locations of these wells are on figure 2. Wells in appendix 1 are presented in the same order as they are listed in table 1--that is, they are ordered alphabetically within each NTS area number. Water levels are shown as depths below land surface. The land-surface altitudes of all wells listed in table 1 were surveyed as part of well construction and are rounded to the nearest foot of the reported value. Some boreholes on Pahute Mesa contain multiple piezometers or completion intervals, each piezometer or interval consisting of a discrete open interval in the borehole. These discrete open intervals are referred to as "wells" in this report. The 65 wells were chosen for inclusion in this report based on two arbitrary criteria. The criteria were: (1) a minimum of three water-level measurements for a well, and (2) the first and last measurement had to span at least 1 month. These criteria excluded from this report about 30 wells on Pahute Mesa with limited water-level data that are currently in the NWIS database.

Periodic water-level measurements were collected manually by the USGS using calibrated electric-cable units (also known as iron-horse and wire-line devices), calibrated electric tapes, or rarely, a calibrated steel tape. A description of each of these instruments is given by Wood and Reiner (1996, p. 6). Most measurements prior to 1996 were made with an electric-cable unit, whereas more recent measurements were typically made using electric tapes. The tapes and cable units are calibrated annually with a 2,000-ft steel reference tape. The steel reference tape is calibrated periodically by the National Institute of Standards and Technology, and over the period of water-level measurements, has shown an error of 0.23 ft or less over the 2,000-ft length. The water levels in this report are rounded to the nearest tenth of a foot of the reported value. Rounding to the nearest tenth of a foot is not meant to imply the absolute accuracy of the depth-to-water measurements, which varies from well to well but is generally within a foot or less of the true depth. Rather, water-level depths were rounded to a tenth of a foot to maintain relative changes in water levels which might be helpful in trend analysis.

In addition to the USGS water-level measurements, supplemental water levels determined from fluid-density geophysical logs are stored in the NWIS database. These data have been provided by private contractors working at NTS, and are reported to the nearest foot below land surface. The accuracy of these water-level measurements is difficult to verify, primarily because (1) the actual land-surface datum used for each measurement is uncertain, (2) a small chance exists that the fluid level measured was not the water level, and (3) the fluid measurements were not required to be extremely accurate for the purposes for which they were used. In general, the water-level measurements from the geophysical logs are considered to have an accuracy of 1 to 2 ft.

Water-level fluctuations on a smaller, more refined scale were examined in well PM-2 using a pressure transducer and electronic data logger. Hourly water-level and barometric data were collected in well PM-2 from August 1996 through September 1998. The water-level data are maintained in the USGS NWIS database. Transducer data are calibrated periodically with manual water-level measurements using a calibrated tape.

Quality-Assurance Flags

Most of the water-level data used in this report were quality assured prior to this study in preparation for inclusion in USGS data reports or as a prerequisite to being entered into NWIS. However, as a part of this study, data were rechecked for errors, especially water levels that appeared on hydrographs as outliers. Water levels that were not necessarily in error but which did not represent static conditions, did not appear to be representative of water levels in the regional aquifer system (as defined by O'Hagan and Laczniak, 1996), or required some other qualification were flagged to provide an explanation as to what the measurement represents (table 3, appendix 2). Water levels affected by a nuclear test (flagged "N") are discussed in a later section of the report ("Seismic Events and Underground Nuclear Tests"). The remaining quality-assurance flags are discussed here.

Most of the non-static water levels in appendix 2 are flagged as affected by pumping of the well prior to or during the measurement ("R" flag) or equilibration of water levels after well construction or development ("C" flag). Water levels affected by pumping typically occurred in water-supply wells. Water levels affected by well construction may show a rising or falling trend that can last for hours to years before water levels stabilize. Wells in which the permeability of the water-producing zones is low or wells open to perched zones with limited water supplies are especially susceptible to the effects of well construction. Brikowski and others (1993) note that in drilling an emplacement hole on Pahute Mesa, "several million liters" of water may be introduced to the formation. This drilling water or water draining down the borehole from a perched zone may cause artificially high water levels in a well that may take a long time to dissipate in a low-permeability formation. For example, artificially high water levels in wells U-20ax and U-20be dropped more than 40 ft in the first 4 months after well completion; water levels in well U-20ax took approximately 10 months to fully equilibrate (appendixes 1 and 2). In contrast to artificially high water levels in the vicinity of a well after drilling, water levels may be artificially low following dewatering of the formation during well development. In well U-19bh, an artificially low water level rose about 40 ft in the first 6 months after the well was developed. From 1992 to 1998, water levels in well U-19bh rose an additional 12 ft (appendixes 1 and 2)--most likely the result of continued equilibration after well construction. In addition to well U-19bh, seven other wells in appendix 2--ER-19-1-2, U-19ax, U-19bh, U-20az, U-20bb (shallow), U-20bb (deep), and UE-20ab--do not have a water-level measurement that represents conditions after the well equilibrated from well construction (that is, all water levels for these wells are flagged with a "C").

Water levels that were elevated above or depressed below regional ground-water levels by more than 75 ft were flagged with an "H" or "L", respectively. Elevated water levels, which are common in wells on some areas of Pahute Mesa (Blankennagel and Weir, 1973; Brikowski and others, 1993; O'Hagan and Laczniak, 1996), may result from upward or downward hydraulic gradients, perched or semi-perched conditions, or other local conditions near the well. Most of the elevated water levels that are noted by O'Hagan and Laczniak (1996) were found in shallow wells (less than 500 ft of water above the bottom of the open interval). Because aquifers with elevated water levels may be isolated from the regional flow system, water levels in these aquifers may not be useful for providing information about long-term trends in ground-water levels or for determining regional directions of ground-water flow. Static water levels in 14 of the 65 wells listed in appendix 1 are greater than about 75 ft above regional water levels drawn by O'Hagan and Laczniak (1996); water levels in 11 of these wells are greater than 150 ft above the regional water levels. Only one of the wells listed in appendix 1, ER-19-1-1, is greater than 75 ft below regional water levels. Determining wells with water levels that differ by moderate amounts (less than 75 ft) from regional water levels is difficult with the present well network.

Boreholes with large borehole deviations (flagged with an "X") have measured depths to water that are greater than the true depth to water. Large borehole deviations, resulting in large water-level corrections, were found in both post-shot holes (U-19v PS 1D and U-20n PS 1DD-H) included in this report. These holes were intentionally drilled at a slant into or near test-hole cavities. Water levels in these two post-shot holes (appendixes 1 and 2) have been corrected for borehole deviations using the following two equations:

where tvd is the slant-corrected true vertical depth of the water, in feet, and
md is the measured depth of the water, in feet.

Equation 1 was derived by triangulating to the water-level depth from a top and a bottom hole location. Equation 2 was derived by triangulating to the water-level depth using gyroscopic survey measurements taken at 50-ft increments. Slant-hole corrections to water levels in well U-19v PS 1D have ranged from 144 to 217 ft (depending on the water-level depth), whereas, corrections to well U-20n PS 1DD-H have ranged from 46 to 47 ft. Small deviations from vertical in a borehole may have little effect on water levels and no practical effect on calculations of head in the aquifer or on determinations of water-level trends. A survey of readily available documentation of borehole deviations from about 25 percent of the non-post-shot wells listed in table 1 indicated measured depths at the water table ranged from near zero to 0.5 ft greater than true vertical depths. Corrections of this magnitude were not applied to water levels and are not considered important when constructing a potentiometric map of the study area or when determining trends.

The completion interval for well ER-19-1-2 (middle) has two access tubes (small and large) that can be used to measure water levels. Because the tubes are supposed to access the same open interval, the water levels are expected to be identical. However, for an unknown reason, possibly an obstruction in the open interval separating one access tube from the other, the water levels differ by about 6 ft. These water levels are flagged "AT" to indicate the discrepancy. Only water levels from the large access tube are entered into the NWIS database.

Some water levels are flagged as anomalous ("A" flag) or dry ("D" flag). Anomalous values indicate water levels that are either unexplained outliers (possibly bad data or an unknown condition affecting the water level) or short-term fluctuations that are not part of a regional long-term trend. For example, in well PM-2, a 15-ft rise in water level was measured in 1993 (appendix 1) that was not detected in other wells in the area; the rise was attributed to either localized recharge in a crater created by the Schooner test about 900 ft southeast of well PM-2 or possibly seepage down the borehole (Russell and Locke, 1997, p. 3; appendix 1). Only one well, U-19ax, has water-level measurements that indicate a dry hole.

Temperature Effects

Water levels in a well can be affected by the temperature of the water in the well. An anomalously high water temperature will result in water that has a relatively low density. For a given pressure head, warm (low density) water causes the water level in a well to rise higher than it would with "normal-temperature" water; likewise, cool (high density) water will cause the water level to be lower.

In some cases, water levels may need to be corrected for spatial and temporal variations in water temperature. Correcting for spatial differences in temperature among wells can be important when comparing heads for hydraulic-gradient calculations. Where ground-water gradients are low relative to the magnitude of temperature corrections, uncorrected hydraulic heads in wells may result in determinations of erroneous directions of ground-water flow. Temporal variations of water temperature in a well must be kept in mind when determining the cause of a water-level trend in that well. A decreasing or increasing water temperature with time, which might occur because of a nearby nuclear test or changes in natural hydrothermal gradients on Pahute Mesa, will cause an apparent decreasing or increasing water-level trend.

Adjusting water levels for temperature effects, or more precisely, density effects, requires information that is not available for most wells on Pahute Mesa. Although temperature is likely to be the most important factor affecting borehole fluid density in wells on Pahute Mesa, other factors such as dissolved and suspended solids, nonaqueous phase liquids (most likely associated with drilling of a well or leakage of oil from pumps), the compressibility of water caused by pressure from the overlying water column, and gravitational effects can influence the density. To adjust for temperature effects, the zone or zones of inflow to the well must be known, and a temperature distribution in the well from the point of lowest inflow to the top of the water column is needed. Determining the zone(s) of inflow is especially critical in wells with several thousand feet of open hole because correcting the density of the entire open hole (which is necessary if the water is entering from the bottom of the hole) compared to correcting the density of water only above the open interval (which is necessary when most or all water is entering near the top) can make a large difference in the final adjustment.

Water levels in appendix 2 were not flagged to indicate temperature effects because of insufficient data. Most wells on Pahute Mesa do not have accurate information on zones of inflow to the well in combination with an adequate vertical temperature profile of the hole. To be adequate for temperature corrections, the profile must extend from the top of the water column to the lowest zone of inflow and must be made after the water in the well equilibrates with formation water temperatures. In addition, water temperature is not measured with each water-level measurement. Therefore, temperature corrections of water levels with time are not possible with most historical data.

A cursory examination of about 30 wells on Pahute Mesa with available water-level and temperature data indicate that water levels in most wells would not be greatly affected by temperature if corrected to 95^oF. Wells with large corrections (greater than 10 ft) are those with long water columns (greater than 1,500 ft of water above the assumed point of inflow) in combination with mean water-column temperatures exceeding 105^oF. Examples of wells with potentially large corrections are PM-2, UE-20f, and U-19v PS 1D. Temperature corrections of water levels in wells with less than 1,000 ft of water above the bottom of the open interval⁵ (see appendix 2) are likely to be relatively small, regardless of the temperature of the water. In contrast to this cursory examination of water-level corrections, adjustments of water levels in boreholes on Pahute Mesa for temperature effects was examined by Pottorff and others (1987). In this more comprehensive study of temperature data, they calculated water-level adjustments to 37 boreholes on Pahute Mesa, including 8 boreholes used in this study, using temperature log data. Differences between measured and temperature-corrected water levels ranged from about 0 to 50 ft. Pottorff and others corrected water levels to standard temperature (60^oF); however, correcting water levels to a higher temperature more typical of temperatures on the mesa, as was done in the cursory examination for this report, would result in much smaller adjustments to most heads. Pottorff and others also corrected the entire water-column length in each borehole, from the top of the water to the bottom of the borehole, which also would potentially over-correct the water levels.

The potential effect on water levels from water-temperature variations with time is illustrated with well PM-1. In this well, where the maximum water-level fluctuation is about 7 ft (appendix 1), a small change in water temperature over several years could have caused the 7-ft change in water level. Well PM-1 has approximately 5,450 ft of water above the top of the open interval and 5,630 ft of water above the bottom of the open interval. According to Blankennagel and Weir (1973, table 8), the temperature in well PM-1 was 149.9^oF about 77 ft below the bottom of the open interval and the water in the well had an estimated temperature gradient of 1.22^oF/100 ft. This gradient would result in temperatures of 80.2^oF at the top of the water column, 146.7^oF at the top of the open interval, and a mean water-column temperature above the open interval of about 113.5^oF. The following equation, described by Winograd (1970), can be used to calculate a water-level change resulting from a temperature change:

n is the measured water column length above the point of inflow (assumed to be the top of the open interval, in this example);

s is the specific weight (or density) of water in the column at the mean water-column temperature and hydrostatic pressure; and

s' is the specific weight (or density) of water at the new temperature and identical hydrostatic pressure.

For the purposes of the following example, the density of distilled water for a given temperature at 1 bar, obtained from Lide (1992, p. 6-10), will be used.

The density of water, s, in well PM-1 at a mean temperature of 113.5^oF is 0.9900 gm/cc (grams per cubic centimeter). Assuming that the mean temperature in the well rose 5^oF to 118.5^oF, the new density of water, s', would be 0.9888 gm/cc. The measured water-column length above the assumed point of inflow, n, is 5,450 ft. Solving the above equation for n', the length of the water column after a 5^oF rise in temperature would be 5,456.6 ft, resulting in an increase in water level of 6.6 ft.

⁵ The length of the water column above the bottom of the open interval is provided in appendix 2 to indicate maximum water-column length that could be affected by temperature (and therefore, the maximum temperature correction that might be applied), assuming the point of inflow to the well was at the bottom of the open interval.