Nevada Water Science Center


Aquifer Tests

Contact Information

Phil Gardner
Groundwater Specialist
Phone: (775) 887-7664
Email:pgardner@usgs.gov

 

Mailing Address
USGS
Nevada Water Science Center
2730 N. Deer Run Rd.
Carson City, NV 89701

 

Nevada Water Science Center
Information

Home Page Surface Water Groundwater Water Quality Research Contact Us

 

 

Wells ER-20-8 #2 main and ER-EC-11 main, Nevada Test Site

Primary Investigator: Keith Halford

Well Data

USGS Site ID
Local Name Altitude Uppermost
Opening
Lowermost
Opening
Primary Aquifer Transmissivity
(ft2/d)
371135116282601 ER-20-8 #2 main 1680 2263 VOLCANIC ROCKS 110000
371151116294101 ER-EC-11 main 3184 4101 VOLCANIC ROCKS 40000

 

Aquifer Test

All Aquifer Test Files (zip)

ER-20-8 #2 main and ER-EC-11 main

Aquifer Test Report and Appendixes (zip) || Groundwater levels (NWISweb)

 

Introduction

The U.S. Geological Survey (USGS) analyzed two multi-well aquifer tests on Pahute Mesa conducted by Navarro Nevada Environmental Services (NNES) in calendar years 2009 and 2010. The wells that were pumped during these aquifer tests were ER-20-8 #2 main (located in the ER-20-8 well cluster) and ER-EC-11 main (located at the ER-EC-11 well site) (Figure 1). Well ER-20-8 #2 main produced water from lavas within the Scrugham Peak and Benham aquifers and was pumped in November and December 2009. Well ER-EC-11 main produced water from welded tuffs within the Tiva Canyon and Topopah Spring aquifers and was pumped in April and May 2010.

During well development and aquifer testing, continuous water-level data from observation wells were collected by NNES at up to seven well sites (Figure 1). Water-level responses to pumping were monitored in hydrostratigraphic intervals that were similar and vertically separated from the pumped intervals. This was possible because most sites had multiple wells at different completion intervals. Water levels in some of the observation wells were monitored beginning in June 2009.

The two aquifer tests were analyzed simultaneously because they affected overlapping volumes of aquifer. By analyzing the tests together, a consistent set of hydraulic properties was estimated. Estimates of transmissivity and storage properties for the volcanic rocks at Pahute Mesa are needed to constrain hydraulic properties in groundwater flow and contaminant transport models at the Nevada National Security Site (formerly Nevada Test Site).

Description of Well Network

The seven well sites monitored by NNES during the two multi-well aquifer tests are located at and in the vicinity of the Nevada National Security Site (Figure 1). Table 1 provides location information for the 18 observation wells at these 7 well sites that were monitored by NNES during multi-well aquifer testing.

In addition to the NNES observation wells, the USGS has collected water-level data at three background observation wells on Pahute Mesa (Figure 1). One of these wells, ER-20-6 #3, was instrumented in 2008, and was originally used to record water-level changes during aquifer testing at well U-20 WW (U.S. Geological Survey, 2009). In November 2009, data collection began in two additional observation wells to monitor background water-level changes at Pahute Mesa; these two wells—PM-3-1 and U-19bk—are assumed to be unaffected by well development and aquifer testing on Pahute Mesa.

Well Construction

Most wells monitored by NNES (Table 1) were completed within the last ten years. The exceptions are ER-20-1, ER-20-5 #1, and ER-20-5 #3, which were completed in the 1990´s. Table 2 and Figure 2 provide detailed information on well construction for all monitored wells. Construction information was obtained primarily from completion reports and written communications from the U.S. Department of Energy, NNES, Navarro-Intera (N-I), Stoller-Navarro Joint Venture (SNJV), and USGS.

 


Figure 1. Location of well sites and geologic structures associated with multi-well aquifer tests at Pahute Mesa, Nevada National Security Site and vicinity, 2009-2010.

 

 

Table 1 . Site information for wells monitored during multi-well aquifer testing at Pahute Mesa, 2009-2010

 

Well name: Names are listed in alphabetical order. Bold part of name is well site, shown on Figure 1.
Type of well: P, Pumping; O, Observation; B, Background observation
U.S. Geological Survey site identification number: Unique 15-digit number identifying well.
Latitude/Longitude: Latitude and longitude coordinates, referenced to North American Datum of 1927.
Land-surface altitude: Altitude, referenced to National Geodetic Vertical Datum of 1929.
Reference: DOE, U.S. Department of Energy; N-I, Navarro-Intera; SNJV, Stoller-Navarro Joint Venture; USGS, U.S. Geological Survey. Construction and hydrostratigraphic information from these references are presented in Table 2.

Well Name

Type of well

U.S. Geological Survey site identification number

Latitude (degrees, minutes, seconds)

Longitude (degrees, minutes, seconds)

Land-surface altitude
(feet)

Reference(s)

ER-20-1

O

371321116292301

37°13´20.8"

116°29´29.4"

6,180.9

DOE (1997a); USGS (2010)

ER-20-5 #1

O

371312116283801

37°13´12.2"

116°28´37.8"

6,241.8

DOE (1997b)

ER-20-5 #3

O

371312116283802

37°13´11.2"

116°28´37.9"

6,241.9

DOE (1997b)

ER-20-6 #3

B

371533116251801

37°15´33.1"

116°25´17.5"

6466.0

DOE (1998)

ER-20-7

O

371247116284502

37°12´47.0"

116°28´44.8"

6,208.9

DOE (2010a)

ER-20-8 main

O

371135116282601

37°11´35.1"

116°28´26.3"

5,848.3

N-I (2010a);
DOE (2010b)

ER-20-8 deep

O

371135116282602

37°11´35.1"

116°28´26.3"

5,848.3

N-I (2010a);
DOE (2010b)

ER-20-8 intermediate

O

371351116282603

37°11´35.1"

116°28´26.3"

5,848.3

N-I (2010a);
DOE (2010b)

ER-20-8 shallow

O

371351116282604

37°11´35.1"

116°28´26.3"

5,848.3

N-I (2010a);
DOE (2010b)

ER-20-8 #2 main

P, O

371135116282701

37°11´34.9"

116°28´26.9"

5,848.8

N-I (2010b);
DOE (2010b)

ER-20-8 #2 piezometer

O

371135116282702

37°11´34.9"

116°28´26.9"

5,848.8

N-I (2010b);
DOE (2010b)

ER-EC-1

O

371223116314701

37°12´22.7"

116°31´47.1"

6,025.6

DOE (2000a); USGS (2010)

ER-EC-6 deep

O

371120116294803

37°11´19.6"

116°29´48.1"

5,604.4

DOE (2000b); SNJV (2009);
USGS (2010)

ER-EC-6 intermediate

O

371120116294804

37°11´19.6"

116°29´48.1"

5,604.4

DOE (2000b); SNJV (2009);
USGS (2010)

ER-EC-6 shallow

O

371120116294805

37°11´19.6"

116°29´48.1"

5,604.4

DOE (2000b); SNJV (2009);
USGS (2010)

ER-EC-11 main

P,O

371151116294101

37°11´51.2"

116°29´41.1"

5,656.3

N-I (2010c)

ER-EC-11 deep

O

371151116294102

37°11´51.2"

116°29´41.1"

5,656.3

N-I (2010c)

ER-EC-11 lower intermediate

O

371151116294103

37°11´51.2"

116°29´41.1"

5,656.3

N-I (2010c)

ER-EC-11 upper intermediate

O

371151116294104

37°11´51.2"

116°29´41.1"

5,656.3

N-I (2010c)

PM-3-1

B

371421116333703

37°14´20.7"

116°33´36.6"

5822.8

USGS (2010)

U-19bk

B

371714116230301

37°17´14.4"

116°23´03.1"

6669.9

USGS (2010)


Table 2 . Well construction and hydrostratigraphic units open to wells monitored by NNES during multi-well aquifer testing at Pahute Mesa, 2009-2010


Well name: Names are listed in alphabetical order.
Drilled depth: Total depth, in feet below land surface, that the well was drilled.
Top and bottom of open casing: Depth, in feet below land surface, of the top and bottom of openings in well casing. The openings may be perforated or screened intervals. N/A, no open casing—open interval is uncased open hole.
Top and bottom of open annulus: Depth, in feet below land surface, of the top and bottom of open annulus. Open annulus includes: (1) the space between the well casing and borehole that is either empty or filled with sand and/or gravel; or (2) uncased open hole deeper than the well casing and shallower or equal to well depth.
Hydrostratigraphic units: Hydrostratigraphic units in contact with open casing or open annulus. Hydrostratigraphic units in bold type are the primary water-producing unit(s) for the well. FCCU, Fluorspar Canyon confining unit; BA, Benham aquifer; UPCU, Upper Paintbrush confining unit; SPA, Scrugham Peak aquifer; MPCU, Middle Paintbrush confining unit; TCA, Tiva Canyon aquifer; LPCU, Lower Paintbrush confining unit; TSA, Topopah Spring aquifer; CHZCM, Calico Hills zeolitic composite unit; CFCM, Crater Flat composite unit.
Water-level altitude: Altitude, in feet, referenced to the National Geodetic Vertical Datum of 1929. Water-level altitudes presented for wells at the ER-20-8, ER-20-8 #2, and ER-EC-11 well sites are the most recent measurements reported in Navarro-Intera reports.

Well Name

Drilled depth

Top and bottom of open casing

Top and bottom of open annulus

Hydrostratigraphic units

Water-level altitude; measurement date

ER-20-1

2,065

N/A

1,940-2,065

TCA

4192.3
(05/18/2009)

ER-20-5 #1

2,823

2,315-2,374

2,249-2,655

TSA, CHZCM

4186.8
(11/03/1995)

ER-20-5 #3

4,294

3,430-3,882

3,348-3,954

CHZCM

4181.9
(02/07/1996)

ER-20-7

2,936

2,360-2,875

2,292-2,924

LPCU, TSA, CHZCM

4188.9
(06/27/2009)

ER-20-8 main

3,442

2,486-2,912
3,127-3,298

2,440-2,940
3,070-3,442

MPCU, TCA,LPCU, TSA, CHZCM

4181.3
(08/22/2009)

ER-20-8 deep

3,442

3,141-3,302

3,070-3,440

LPCU, TSA, CHZCM

4181.4
(09/17/2009)

ER-20-8 intermediate

3,442

2,498-2,909

2,440-2,940

MPCU, TCA,LPCU

4182.0
(09/17/2009)

ER-20-8 shallow

3,442

2,088-2,119

1,614-2,150

BA, UPCU, SPA

4180.9
(09/03/2009)

ER-20-8 #2 main well

2,338

1,680-2,263

1,626-2,338

BA, UPCU, SPA, MPCU

4180.7
(09/08/2009)

ER-20-8 #2 piezometer

2,338

1,663-2,234

1,626-2,338

BA, UPCU, SPA, MPCU

4181.4
(11/16/2009)

ER-EC-1

5,000

2,298-2,821
3,348-3,760
4,448-4,750

2,258-2,863
3,286-3,776
4,399-4,895

BA, UPCU, TCA, LPCU, TSA, CHZCM, CFCM

4170.4
(05/19/2009)

ER-EC-6 deep

5,000

3,437-3,811
4,420-4,904

3,392-3,820
4,369-5,000

TSA, CHZCM, CFCM

4178.2
(05/19/2009)

ER-EC-6 intermediate

5,000

2,194-2,507

2,138-2,510

UPCU, TCA

4178.7
(05/19/2009)

ER-EC-6 shallow

5,000

1,628-1,870

1,606-1,948

BA

4179.1
(05/19/2009)

ER-EC-11 main

4,149

3,184-3,374
3,644-4,101

3,196-3,385
3,590-4,148

TCA, LPCU, TSA, CHZCM

4180
(10/16/2009)

ER-EC-11 deep

4,149

3,641-4,093

3,590-4,148

LPCU, TSA, CHZCM

4180.0
(11/12/2009)

ER-EC-11 lower intermediate

4,149

3,159-3,378

3,196-3,385

TCA

4179.2
(11/12/2009)

ER-EC-11 upper intermediate

4,149

2,678-2,991

1,662-3,024

FCCU, BA

4179.3 (11/12/2009)


Figure 2 . Well construction and hydrostratigraphic units penetrated by wells monitored by NNES during multi-well aquifer testing at Pahute Mesa, 2009-2010. Construction diagrams grouped as shown on three hydrostratigraphic sections in Figure 4.

 

 

Figure 2. (continued).

 

 

Figure 2. (continued).

 

 

Figure 2. (continued).

 

 

Hydrogeology

The wells monitored by NNES during multi-well aquifer testing at Pahute Mesa are completed in Tertiary volcanic rocks. The volcanic rocks of Pahute Mesa are dominated by lavas and tuffs of rhyolitic composition (Laczniak and others, 1996). Geologic structures at Pahute Mesa include normal faults with surface exposure and buried structural zones and caldera margins.

A buried structural feature affecting the hydrostratigraphy of wells monitored during multi-well aquifer testing at Pahute Mesa is the Northern Timber Mountain Moat Structural Zone (NTMMSZ) (Figure 3). The NTMMSZ is a buried west-northwest trending fault zone. The NTMMSZ is a down-on-the-southwest fault zone, with rock displacements of more than 1,000 ft (U.S. Department of Energy, 2010a). The area that is bounded on the north by the NTMMSZ and on the south by the Timber Mountain caldera complex structural margin is referred to as “the Bench area" (U.S. Department of Energy, 2009) (Figure 3 and Figure 4). Observation well sites located on the Bench area are ER-20-8 cluster, ER-EC-1, ER-EC-6, and ER-EC-11; sites ER-20-1, ER-20-5 cluster, and ER-20-7 are located north of the Bench area (Figure 3 and Figure 4).

Observation wells on the Bench area penetrate about 1,400 to 1,800 ft of unsaturated rock, and those north of the Bench area penetrate about 2,000 feet of unsaturated rock (Figure 2 and Figure 4). The water table in wells on the Bench area occurs in the Timber Mountain aquifer (TMA), Benham aquifer (BA), or Fluorspar Canyon confining unit (FCCU). The water table in wells north of the Bench area occurs in the Tiva Canyon aquifer (TCA) or Lower Paintbrush confining unit (LPCU).

The wells on the Bench area were constructed to monitor water levels from four major water-producing hydrostratigraphic units (Figure 2 and Figure 4): the BA, Scrugham Peak aquifer (SPA), TCA, and Topopah Spring aquifer (TSA). The Calico Hills zeolitic composite unit (CHZCM) and Crater Flat composite unit (CFCM) also supply water to observation wells on the Bench area (Table 2). Major water-producing hydrostratigraphic units in wells north of the Bench area are the TCA and TSA, with some production from the CHZCM.

The major water-producing hydrostratigraphic units in the aquifer-test area are the rhyolitic lava flows and the welded ash-flow tuffs. The BA and SPA are lava-flow aquifers. These aquifers comprise rhyolitic lava flows with some intervals of vitrophyre, pumiceous lava, and flow breccias (U.S. Department of Energy, 2010b). The TCA and TSA are welded-tuff aquifers. The ash-flow tuffs in these aquifers are partially to densely welded, with some nonwelded layers and local zeolitization. The CHZCM and CFCM are composite units of rhyolitic lava-flow aquifers and nonwelded tuff confining units with local to common zeolitization (Laczniak and others, 1996, p.11; U.S. Department of Energy, 1997b; 2000b). The CFCM also contains welded-tuff aquifers.

 

Figure 3. Location of hydrostratigraphic sections through wells monitored for multi-well aquifer tests at Pahute Mesa, 2009-2010.

 

 

 Figure 4. Hydrostratigraphic sections A-A´, B-B´, and C-C´, Pahute Mesa.

 

 

 Figure 4. Continued.

 

 

 Figure 4. Continued.

 

 

Data Collection

ER-20-8 #2 Aquifer-Test Data

Monitoring of the pumping well (ER-20-8 #2 main) and observation wells at Pahute Mesa (Table 1; Figure 1) began as early as June 2009. However, aquifer-test analysis of pumping from ER-20-8 #2 main was limited to data collected from the pumping well and 14 observation wells from November 27, 2009 through January 19, 2010 (Table 3). The distance between observation and pumping wells ranged from 50 to 17,000 ft. The observation wells were generally north and west of the pumping well with bearings relative to the pumping well of about 260–360 degrees.

Data from the pumping and observation wells provided the drawdown, recovery, and discharge rates needed for analysis of the aquifer test. Water level and pumping rate were monitored in the pumping well and water levels were measured continuously in the observation wells. The data used in the analysis can be found in appendix A (see attached Microsoft Excel spreadsheet).

Multiple aquifer tests were attempted using well ER-20-8 #2 main as the pumping well (Figure 5; Table 4). The longest constant rate test lasted about 101 hours and was conducted from 12/14/2009 at 10:06 to 12/18/2009 at 15:23. The discharge rate during this aquifer test was approximately 130 gallons per minute (gal/min), with a total withdrawal during this pumping period of about 790,000 gallons. For about the first 24 hours of this test, the discharge values recorded by the data logger were about 20 gal/min larger than those from the in-line flowmeter (U.S. Department of Energy, written communications, 2009). It was determined that the in-line flowmeter values were correct and the data logger readings were adjusted.


Table 3 . Wells monitored during ER-20-8 #2 main aquifer test


Well name: Names listed in order of distance from pumping well ER-20-8 #2 main.
Distance from pumping well: Distance, in feet, from ER-20-8 #2 main.
Bearing relative to pumping well: True bearing, in degrees, from pumping well to observation well.

Well name

Distance from pumping well, in feet

Bearing relative to pumping well

ER-20-8 #2 main

0

N/A

ER-20-8 deep
ER-20-8 intermediate
ER-20-8 shallow

50

67°

ER-EC-11 deep
ER-EC-11 lower intermediate
ER-EC-11 upper intermediate

6,220

285°

ER-EC-6 deep
ER-EC-6 intermediate
ER-EC-6 shallow

6,740

257°

ER-20-7

7,450

349°

ER-20-5-3

9,800

355°

ER-20-5-1

9,900

355°

ER-20-1

11,770

336°

ER-EC-1

16,870

287°

 


Figure 5. Water levels and discharge from ER-20-8 #2 main during well development and aquifer testing, November to December, 2009.

 

 

Table 4. General pumping schedule from well ER-20-8 #2 main during multi-well aquifer testing at Pahute Mesa, November to December, 2009


Start date/time and End date/time: Start and end of pumping from Navarro Nevada Environmental Services daily drilling reports.
Approximate pumping duration: Time, in minutes, that pump was turned on, rounded to 2 significant figures.
Approximate discharge rate: Approximate discharge, to the nearest ten gallons per minute, of the pumping well between the start and end time. Value estimated from data collected from an in-line flowmeter.
Approximate total discharge: Approximate discharge, to the nearest ten thousand gallons, of the pumping well between the start and end time. Value based upon data collected from in-line flowmeter.

Start date/time

End date/time

Aquifer-test description

Approximate pumping duration (minutes)

Approximate discharge rate (gallons per minute)

Approximate total discharge (gallons)

Notes

11/28/2009 14:30

12/3/20009 14:40

Well development

7,200

90-170

840,000

Step drawdown testing completed on 11/29-30/2010.

12/10/2009
10:21

12/10/2009
15:40

Constant rate

320

130

35,000-40,000

Test suspended to install transducer in ER-20-8 shallow. Flowmeter problems make approximate total discharge values questionable.

12/11/2009
18:10

12/13/2009
1:20

Constant rate

1,900

130

250,000

Test suspended due to pump failure.

12/14/2009
10:06

12/18/2009
15:23

Constant rate

6,100

130

790,000

Datalogger recorded incorrect discharge rate for the first 23 hours of the aquifer test.

 

 

ER-EC-11 Aquifer-Test Data

Monitoring of the pumping well (ER-EC-11 main) and observation wells at Pahute Mesa (Table 1; Figure 1) began as early as June 2009. However, aquifer-test analysis of pumping from ER-EC-11 main was limited to data collected from the pumping well and 12 observation wells from April 30 to June 7, 2010 (Table 5). The distance between observation and pumping wells ranged from 0.5 to 11,000 ft. The observation wells were located in all azimuthal quadrants around the pumping well ER-EC-11 main.

Data from the pumping and observation wells provided the drawdown, recovery, and discharge rates needed for analysis of the aquifer test. Water level and pumping rate were monitored in the pumping well and water levels were measured continuously in the observation wells. The data used in the analysis can be found in appendix B (see attached Microsoft Excel spreadsheet).

Multiple aquifer tests from pumping well ER-EC-11 main are summarized in Figure 6 and Table 6. The longest constant-rate test lasted about 194 hours and was conducted from 5/10/2010 at 13:15 to 5/18/2010 at 15:10. The discharge rate during this aquifer test averaged 310 gal/min, with a total of about 3.6 million gallons of water withdrawn during the test.

 

Table 5. Wells monitored during ER-EC-11 main aquifer test


Well name: Names listed in order of distance from pumping well ER-EC-11 main.
Distance from pumping well: Distance, in feet, from ER-EC-11 main.
Bearing relative to pumping well: True bearing, in degrees, from pumping well to observation well.

Well Name

Distance from pumping well

Bearing relative to pumping well

ER-EC-11 main
ER-EC-11 upper intermediate

0

N/A

ER-EC-6 deep
ER-EC-6 intermediate
ER-EC-6 shallow

3,250

190°

ER-20-8 deep
ER-20-8 intermediate
ER-20-8 shallow

6,250

105°

ER-20-7

7,250

39°

ER-20-1

9,130

ER-20-5 #3

9,650

32°

ER-20-5 #1

9,670

32°

ER-EC-1

10,700

287°

Figure 6. Water levels and discharge from ER-EC-11 main during well development and aquifer testing, April and May, 2010.

 

 

Table 6. General pumping schedule from well ER-EC-11 main during multi-well aquifer testing at Pahute Mesa, April and May, 2010


Start date/time and End date/time: Start and end of pumping from Navarro Nevada Environmental Services daily drilling reports.
Approximate pumping duration: Time, in minutes, that pump was turned on, rounded to 2 significant figures.
Approximate discharge rate: Approximate discharge, to the nearest ten gallons per minute, of the pumping well between the start and end time. Value estimated from data collected from an in-line flow meter.
Approximate total discharge: Approximate discharge, to the nearest ten thousand gallons, of the pumping well between the start and end time. Value based upon data collected from in-line flow meter.

Start date/time

End date/time

Aquifer-test description

Pumping duration (minutes)

Approximate discharge rate (gallons per minute)

Approximate total discharge (gallons)

04/30/2010
11:20

05/03/2010
11:20

Well development

4,300

290

1,250,000

05/03/2010
13:15

05/04/2010
16:40

Step drawdown

1,600

190-320

430,000

05/10/2010
13:15

05/18/2010
15:10

Constant rate

12,000

310
(rate began at about 318 gal/min and ended at about 300 gal/min)

3,580,000

05/18/2010
15:10

05/19/2010
15:00

Constant rate

1,400

180

260,000


 

Water-Level Modeling & Drawdown Estimation

Drawdowns from pumping of wells ER-20-8 #2 main and ER-EC-11 main were detected and estimated by modeling water levels in the pumping and observation wells. This approach was necessary because environmental (naturally occurring) water-level fluctuations of more than 0.2 ft exceeded maximum drawdowns from pumping in most observation wells. Drawdowns were differentiated from environmental fluctuations by modeling synthetic water levels that simulated environmental water-level fluctuations and pumping effects. Environmental water-level fluctuations were simulated by summing individual time-series of barometric pressure, tidal potential, and background water levels (Halford, 2006a). Pumping responses were simulated by superposition of Theis solutions.

Environmental water-level fluctuations were simulated with time series of barometric pressure, earth tides, gravity tides, and water levels from background wells ER-20-6 #3 and PM-3-1 (Figure 1). These two wells had complete record during the period of analysis and were unaffected by pumping from ER-20-8 #2 main and ER-EC-11 main. Water levels from background wells were critical because they were affected by tidal potential–rock interaction, imperfect barometric coupling, and seasonal climatic trends. A linear trend additionally was incorporated where instrument drift was apparent. This drift could not be corrected adequately using manual measurements because most depths to water exceeded 2,000 ft below land surface. At these depths, the inherent error of the manual measurements was similar to the magnitude of the drift correction.

Pumping responses from wells ER-20-8 #2 main and ER-EC-11 main were modeled with a superposition Theis model, where multiple pumping periods were simulated by superimposing multiple Theis (1935) solutions. Superposition Theis models served as simple transform functions, where step-wise pumping records were translated into approximate water-level responses. Numerical experiments have confirmed that superposition Theis models closely approximate water-level responses through hydrogeologically complex aquifers. This approach will herein be discussed as the superposition Theis model.

Synthetic water levels were fit to measured water levels by minimizing a sum-of-squares objective function (Halford, 2006a). Amplitude and phase were adjusted in each time series that simulated environmental water-level fluctuations. Transmissivity and storage coefficient were adjusted in the superposition Theis model. Estimated values of transmissivity and storage coefficient generally were not valid estimates of aquifer properties because the assumptions of the underlying Theis solution were violated grossly.

Environmental water-level fluctuations and pumping effects were modeled simultaneously because water-level records prior to development and testing were less than a few weeks. These were short periods relative to a combined well development, aquifer testing, and recovery period of about two months. Reliable estimation of drawdowns without simulating pumping effects requires fitting synthetic water levels during a period prior to development. This fitting period should be more than 3 times greater than the drawdown estimation period (Halford, 2006a).

Drawdown estimates were the summation of the superposition Theis models and differences between synthetic and measured water levels. Drawdown estimates were presented as maximum values because many of the drawdowns at observation wells were small relative to environmental water-level fluctuations. Synthetic water levels matched measured water levels with root-mean-square errors between 0.01 and 0.02 ft in wells where temperature changes and well-losses were not factors. Pumping effects were detected definitively where the maximum drawdown estimate in a well exceeded 0.05 ft. This drawdown threshold was established relative to the fitting errors between synthetic and measured water levels.

ER-20-8 #2 Aquifer Test

Water-level change resulting from the pumping of well ER-20-8 #2 main was approximated using simplified pumping steps. Nineteen pumping steps (Figure 7) were used to generate a superposition Theis model for near-field observation wells (ER-20-8 well cluster and ER-20-8 #2 main). Pumping steps were reduced to nine (Figure 7) for superposition Theis models in far-field observation wells. A simplified schedule of pumping steps was considered acceptable for far-field wells because water levels do not respond to high-frequency changes in pumping.

 

Figure 7. Pumping from ER-20-8 #2 main during well development and aquifer testing, November and December, 2009. Data were binned into 19 and 9 pumping steps for use in the superposition Theis models and numerical mode

 

 

Drawdowns were estimated in the pumping well and each of the 14 observation wells that were monitored during the aquifer test (Table 7). Drawdowns were estimated with water-level modeling in each well. Water levels for well ER-EC-11 deep show an upward drift (Appendix A), apparently caused by a poorly functioning transducer, and were not analyzed for drawdown.

Synthetic water levels in the pumping well simulated frictional well loss and thermal expansion of the water column. Water-level changes that resulted from water-temperature changes during pumping were relatively minor and could be neglected. However, a large component of the water-level decline in the well resulted from well loss rather than from drawdown in the aquifer. Maximum drawdown in the aquifer from the superposition Theis model non-uniquely ranged from 0.4 to 4 ft because well loss and drawdown in the aquifer were correlated. Transmissivity estimates were greater than 10,000 ft2/d.

Estimated drawdowns for all monitored wells (Table 7) were relatively large (0.2 ft or greater) in four wells, small (0.03-0.08 ft) in three wells, and below detection in seven wells. Table 7 also presents the relative degree of certainty that the estimated drawdown is large enough to be distinguished from the background noise in the data; that is, is a pumping response (superposition Theis model) necessary to adequately synthesize the measured water levels. A well with a low relative certainty indicates that drawdown may or may not have occurred in the well; if drawdown occurred, it is poorly constrained and probably has a magnitude that is equal to or less than the estimated maximum drawdown.

 

 

Table 7. Estimated maximum drawdowns in observation wells from pumping in well ER-20-8 #2 main during multi-well aquifer testing at Pahute Mesa, November 2009 to January 2010


Estimated maximum drawdown: Maximum drawdown was estimated by matching measured water levels in the observation well to a synthetic curve of nonpumping and pumping responses. The pumping response was estimated with a superposition Theis model.
Relative certainty that drawdown occurred: A relative scale indicating likelihood that estimated drawdown is large enough to be observed above background noise in data. High, very likely; Low, possible, but drawdown also could be zero; N/A, not applicable.

Well name

Estimated maximum drawdown (feet)

Relative certainty that drawdown occurred

ER-20-8 #2 main

1.

High

ER-20-8 deep

0.2

High

ER-20-8 intermediate

0.2

High

ER-20-8 shallow

0.3

High

ER-EC-11 deep

Not estimated

N/A

ER-EC-11 lower intermediate

0.08

Low

ER-EC-11 upper intermediate

<0.05

N/A

ER-EC-6 deep

<0.05

N/A

ER-EC-6 intermediate

<0.05

N/A

ER-EC-6 shallow

0.08

Low

ER-20-7

<0.05

N/A

ER-20-5-3

<0.05

N/A

ER-20-5-1

<0.05

N/A

ER-20-1

<0.05

N/A

ER-EC-1

<0.05

N/A

 

Figure 8 shows three observation wells illustrating estimated drawdowns that are relatively large (top plot), small (middle plot), and undetected (bottom plot). Using well ER-20-8 intermediate as an example, the drawdown response resulting from pumping of ER-20-8 #2 main is relatively large and clearly defined. Measured water levels from 11/24/2009 to 1/18/2010 were simultaneously fitted to time series of barometric pressure, earth and gravity tides, water levels in background wells PM-3-1 and ER-20-6 #3, and a superposition Theis model. Slight differences between the estimated drawdowns and the superposition Theis model show most of the water-level changes in ER-20-8 intermediate have been explained (Figure 8). Plots of estimated drawdowns for all wells in Table 7 can be viewed in individual Excel files stored under the "Detrends" subdirectory in the ER-20-8 #2 aquifer-test archive.

 

Figure 8. Examples of measured and synthetic water levels, superposition Theis models, and drawdown estimates in three observation wells. Drawdown estimate is the summation of superposition Theis model and differences between measured and synthetic water levels. Drawdowns resulted from pumping in well ER-20-8 #2 main during multi-well aquifer testing, November 2009 to January 2010.

 

ER-EC-11 Aquifer Test

Water-level change resulting from the pumping of well ER-EC-11 main was approximated using simplified pumping steps. Thirteen pumping steps (Figure 9) were used to generate a superposition Theis model for the observation wells. These 13 steps were sufficient to adequately calculate the pumping response in all observation wells. The pumping history of well ER-EC-11 main was further simplified to six pumping steps (Figure 9) for use in the numerical model simulation (see "Groundwater-Flow Models" section).

 

 

Figure 9 . Pumping from ER-EC-11 main during well development and aquifer testing, April and May, 2010. Data were binned into 13 and 6 pumping steps for use in the superposition Theis models and numerical model, respectively.

 

 

Drawdowns were evaluated for the pumping well and each of the 12 observation wells that were monitored during the aquifer test (Table 8). The superposition Theis model that resulted from the fitting process was used to estimate the drawdown. Estimated drawdowns were relatively large (0.2 ft or greater) in eight wells and small (0.08-0.1 ft) in five wells (Table 8). The relative degree of certainty that the estimated drawdown is large enough to be distinguished from background noise in the data also is provided in Table 8; that is, is a pumping response (superposition Theis model) necessary to adequately synthesize the measured water levels. A well with a low relative certainty indicates that drawdown may or may not have occurred in the well; if drawdown did occur, it is poorly constrained and probably has a magnitude that is equal to or less than the estimated drawdown. High relative certainty indicates high probability that drawdown occurred in the well.


Table 8 . Estimated drawdowns in observation wells from pumping in well ER-EC-11 main during multi-well aquifer testing at Pahute Mesa, April and May, 2010


Estimated drawdown: Drawdown was estimated by matching measured water levels in the observation well to a synthetic curve of nonpumping and pumping responses. The pumping response was estimated with a superposition Theis model.
Relative certainty that drawdown occurred: A relative scale indicating likelihood that estimated drawdown is large enough to be observed above background noise in data. High, very likely; Moderate, more probable than not; Low, possible, but drawdown also could be zero.

Well name

Estimated drawdown (feet)

Relative certainty that drawdown occurred

ER-EC-11 main

4.

High

ER-EC-11 upper intermediate

0.9

High

ER-EC-6 deep

0.5

High

ER-EC-6 intermediate

0.4

High

ER-EC-6 shallow

0.5

High

ER-20-8 deep

0.2

Moderate

ER-20-8 intermediate

0.2

Moderate

ER-20-8 shallow

0.08

Low

ER-20-7

0.2

High

ER-20-1

0.09

Low

ER-20-5 #3

0.1

Moderate

ER-20-5 #1

0.1

Moderate

ER-EC-1

0.1

Moderate

 

Figure 10 shows three observation wells that illustrate different magnitudes of estimated drawdowns. Using well ER-EC-6 deep as an example, the drawdown response resulting from pumping of ER-EC-11 main is relatively large and clearly defined. Measured water levels from 4/25/2010 to 6/8/2010 were simultaneously fitted to time series of barometric pressure, earth and gravity tides, water levels in background wells PM-3-1 and ER-20-6 #3, and a superposition Theis model. Slight differences between the estimated drawdowns and the superposition Theis model show that most of the water-level changes in ER-EC-6 deep have been explained (Figure 10). Plots of estimated drawdowns for all wells in Table 8 can be viewed in individual Excel files stored under the “Detrends" directory in the ER-EC-11 aquifer-test archive.

 

Figure 10. Examples of measured and synthetic water levels, superposition Theis models, and drawdown estimates in three observation wells. Drawdown estimate is the summation of superposition Theis model and differences between measured and synthetic water levels. Drawdowns resulted from pumping in well ER-EC-11 main during multi-well aquifer testing, April and May, 2010.

 

 

 

Synthetic water levels in the pumping well (ER-EC-11 main) simulated frictional well loss and thermal expansion of the water column, which dominated measured water-level changes. Maximum drawdown in the aquifer non-uniquely ranged between 1 and 5 ft, whereas total water-level decline in the well exceeded 100 ft (Figure 11A). Furthermore, during the constant-rate test, pumping rates declined from about 318 to 300 gal/min (Figure 9). Concurrently, water temperature increased as the well was pumped. These two factors resulted in a water-level rise of more than 20 ft after the constant-rate test began (measured water level from about 5/11/2010 to 5/18/2010 on Figure 11A). This rise was much greater than a maximum drawdown estimate in the aquifer of 5 ft (superposition Theis models of drawdown estimates on Figure 11B). Because the effects from well loss, varying water temperatures, and slightly changing pumping rates are large and correlated with aquifer drawdown, it is impossible to estimate a unique Theis response of the aquifer drawdown in this well. By using transmissivities of 20,000 ft2/d and 100,000 ft2/d to estimate aquifer drawdown, synthetic curves can be generated that reasonably match measured water levels (note that differences between measured and simulated water levels are similar for both transmissivities shown on Figure 11B). The transmissivity around ER-EC-11 main is uncertain and greater than 20,000 ft2/d.

Figure 11. Estimates of aquifer drawdown in well ER-EC-11 main that were derived from superposition Theis models with varying transmissivity. Estimated drawdowns are compared to (A) measured and synthetic water levels and (B) differences between measured and synthetic water levels using a smaller vertical scale.

 

 

Well ER-EC-11 upper intermediate is open to the Benham aquifer, which is hydraulically separated from the deeper pumped zone by the upper Paintbrush confining unit (Figure 2). Water levels in ER-EC-11 upper intermediate were dominated by temperature effects during the aquifer test in May 2010. The water temperature was unstable because heat was transferred from the heated water column in the pumping well (during pumping) through the well annulus to the water column in ER-EC-11 upper intermediate. The effect of temperature on water level was much larger and opposite in direction to the drawdown in the aquifer (Figure 12). Prior to the constant-rate aquifer test that began on 5/10/2010, water levels were declining in ER-EC-11 upper intermediate as a result of a cooling water column following the previous step-drawdown test on 5/4/2010 (Table 6). Before water levels and temperature could equilibrate, the water column was again heated during the constant-rate test, resulting in a measured water-level rise of about 3 ft (Figure 12). This measured rise is caused by a large upward response as a result of the heated water column (“maximum temperature effect" in Figure 12) that is offset by a smaller downward response as a result of aquifer drawdown (“maximum drawdown" in Figure 12). Because these two responses are negatively correlated, it was impossible to estimate an accurate drawdown in the aquifer. Rather, maximum drawdown of about 1.3 ft (Figure 12) was constrained by the maximum estimated average temperature of the water column in ER-EC-11 upper intermediate using limited water-temperature information. The minimum drawdown likely is greater than 0.5 ft, based on similar drawdowns in the distant ER-EC-6 observation wells (Table 8).

Figure 12. Maximum estimate of aquifer drawdown in well ER-EC-11 upper intermediate based on the maximum temperature effect expected in the water column.

 

Aquifer-Test Analysis

Numerical models were created and simulated for each of the aquifer tests to obtain estimates of hydraulic properties for the volcanic rocks underlying Pahute Mesa. Numerical methods were used to estimate hydraulic properties because the groundwater beneath the study area flows through a complexly layered sequence of volcanic-rock aquifers and confining units that have been faulted into distinct structural blocks. The observation wells were vertically separated and distant from the pumping wells. Therefore, water-level responses in the observation wells could not be analyzed with a simple analytical method, such as the Theis solution (Theis, 1935), because assumptions were violated grossly.

Hydraulic properties of aquifers, confining units, and fault structures were estimated by interpreting pumping responses from multiple aquifer tests using a single, three-dimensional hydrogeologic framework and multiple groundwater-flow models. Multiple groundwater-flow models allowed grid refinement near each pumping well and different pumping schedules to match each aquifer test. The hydrogeologic framework model and groundwater flow models were designed so the hydraulic properties of fault structures could be estimated. Multiple groundwater-flow models also facilitate incorporating additional aquifer tests because each test is interpreted with an independent groundwater-flow model. This allows for each model to be created, tested, and debugged individually and assures that simulated drawdowns and sensitivities are computed and extracted correctly before combining with all previous model comparisons.

Hydrogeologic Framework Model

Hydraulic properties were distributed spatially with a single, three-dimensional hydrogeologic framework that was simplified from refined cross sections (Figure 4). The hydrogeologic framework was simplified by reducing the nineteen mapped hydrostratigraphic units (HSUs) to eight simplified HSUs (Table 9). HSUs were simplified to create a more efficient and simple model and to focus on the major aquifers and confining units that are thought to be hydraulically important in the study area.

The TCVA, THLFA, THCM, TMA, and PVTA units (Table 9) are omitted because they are above the water table. The joint Benham and Scrugham Peak aquifers (BA/SPA) simplified HSU incorporate the original BA and SPA units. A simplified Upper Paintbrush confining unit (sUPCU) combined the original UPCU and MPCU units. A simplified Calico Hills zeolitic composite unit (sCHZCM) aggregated the original CHZCM, IA, and BRA units. A simplified Crater Flat composite unit (sCFCM) also was created and combined the original CFCM, BFCU, and PBRCM units.

 

Table 9. Categorization of hydrostratigraphic units to simplified hydrostratigraphic units.


Modeled displacements of simplified HSUs along fault structures were limited to the Northern Timber Mountain Moat Structural Zone (NTMMSZ) and the Unnamed Buried Structure (UBS) (Figure 13). Displacements of more than 1,000 ft were modeled along the NTMMSZ and UBS (Figure 4). Displacements along all other fault structures were considered minor and were not simulated explicitly.

Fault structures were differentiated as 200-ft wide zones within each simplified HSU so that the effects of faulting on hydraulic conductivity could be tested. These zones straddled the faults and defined hydraulic conductivity on both sides of a fault. The zones Fault West, Fault East, and Fault South were the buffers around NTMMSZ west of UBS, NTMMSZ east of UBS, and the UBS, respectively (Figure 13). A single hydraulic-conductivity value was assigned to each fault zone.

Hydraulic conductivity was distributed throughout each of the eight simplified HSUs with pilot points, which are points in the model domain where hydraulic properties are estimated (RamaRao and others, 1995). Pilot points were assigned to all eight simplified HSUs, where present, at 19 mapped locations (Figure 13) for a total of 115 pilot points. Hydraulic properties were interpolated from pilot points with kriging to node locations as defined for each groundwater-flow model (Doherty, 2008b). The spatial variability of log-hydraulic conductivity was defined with an isotropic, exponential variogram, where the specified range was 15,000 ft and no nugget was specified.

Hydraulic conductivities were not estimated independently at 63 of the 115 pilot points because drawdowns were below detection, 0.05 ft, near these pilot points. These hydraulic conductivities were defined by pilot point values closer to the pumping wells that were sensitive to pumping effects. Drawdowns did not exceed 0.05 ft at distances of more than 3 miles from well sites ER-20-8 #2 and ER-EC-11 or in the sCFCM.

Discontinuity and displacement were modeled by splitting simplified HSUs across the NTMMSZ and UBS faults and combining HSU pieces in groundwater-flow model layers. Discontinuities were created by splitting simplified HSUs into North, Southeast, and Southwest model areas that were defined by the NTMMSZ and UBS (Figure 13). Simplified HSU pieces were displaced vertically to appropriate groundwater-flow model layers, which approximated continuous horizons across faults. For example, the TSA, FCCU, and BA/SPA occur in the North, Southwest, and Southeast model areas, respectively, of a single groundwater-flow model layer. This approach uses relatively coarse vertical discretization that is groundwater-flow model dependent. Hydraulic-property estimates likely are affected by this, consequently, structural inconsistencies were avoided with a common vertical discretization in all groundwater-flow models.

Figure 13. Location of flow-model areas, fault zones, and 18 of the 19 map locations of pilot points that were simulated in the numerical models for the ER-20-8 #2 and ER-EC-11 aquifer tests (the 19th map location, U-19bg, is 3.5 mi east-northeast of ER-20-6).

 

 

Groundwater-Flow Models

Drawdowns from each multiple-well aquifer test were interpreted with a three-dimensional MODFLOW model (Harbaugh and others, 2000), where each model was centered on the pumping well and rows paralleled the NTMMSZ fault. Each model grid extended laterally about 200,000 ft away from the pumping well. All models were about 4,200 ft thick and extended vertically from the water table to sea level (Figure 14). Rows and columns were assigned widths of 3 ft at the pumped well and at the intersection of the NTMMSZ and UBS. Columns and rows expanded successively by a factor of 1.25 away from the pumped well until a mid-point between well and faults or the model edge was reached. Rows and columns in each model differed, but ranged between 125 and 128 (Table 10). All external boundaries were no-flow. Changes in the wetted thickness of the aquifer system were not simulated because the maximum drawdown near the water table was small relative to the total thickness. Simulation periods were subdivided into stress periods that simulated simplified pumping schedules for ER-20-8 #2 (Figure 7) and ER-EC-11 (Figure 9).

 

Table 10. Number of layers, rows, columns and stress periods, simulated period, and volume pumped for each MODFLOW model.

 

 

 

 

 

Stress Periods

Period Simulated

Volume Pumped, in Millions of Gallons

Pumping Well

Layers

Rows

Columns

Begin

End

ER-20-8 #2 main

12

125

127

9

11/28/2009

01/11/2010

1.9

ER-EC-11 main

12

126

128

6

04/30/2010

06/15/2010

5.6

 

All groundwater-flow models were discretized vertically into 12 layers that were defined with the water table and simplified HSU contacts. Layer 1 was 1-foot thick to better approximate drainage from the water table. Contact altitudes were defined by stratigraphy at well sites ER-20-1, ER-20-5, ER-20-7, ER-20-8, ER-EC-1, ER-EC-11, and ER-EC-6 and inferred contacts in cross sections (Figure 4). Some simplified HSUs were subdivided into multiple layers so that a single HSU can abut multiple HSUs along fault structures as occurs with the FCCU abutting the BA/SPA, UPCU, and TCA (Figure 14). HSUs were assigned to groundwater-flow model layers by area (Table 11) primarily to preserve the juxtaposition of units as conceptualized in hydrostratigraphic sections (Figure 4). A common vertical discretization was used to ensure consistency of simulated displacements of HSUs in all groundwater-flow models.

 

Figure 14. Vertical discretization and assignment of simplified hydrostratigraphic units to layers in numerical flow models analyzing ER-20-8 #2 and ER-EC-11 aquifer tests.

 

 

Table 11. Simplified hydrostratigraphic unit (HSU) assignments by model layer and area. Model areas are shown on Figure 13

Parameter Estimation

Hydraulic-conductivity distributions and a global, vertical-to-horizontal anisotropy were estimated by minimizing a weighted composite, sum-of-squares objective function. Fifty-two of the 115 pilot points that define these distributions were adjusted with PEST (Doherty, 2008a). Differences between measured and simulated observations defined the goodness-of-fit or improvement of calibration. These differences, residuals, were weighted and summed in the objective function,

where
x is the vector of parameters being estimated,
nobs is the number of observations that are compared,
(Ôi) is the ith simulated observation,
(oi) is the ith measurement or regularization observation, and
wi is the ith weight.

 

Although the sum-of-squares error serves as the objective function, root-mean-square (RMS) error was reported because RMS error was compared easily to measurements. Root-mean-square error is,

 .

Measurement and regularization observations controlled model calibration. The models used 3,558 drawdowns as measurement observations. Estimated hydraulic conductivities were guided by regularization observations to preferred conditions in areas that were insensitive to measurement observations. This approach is Tikhonov regularization (Doherty, 2008a).

Measured drawdowns are the superposition Theis models plus differences between synthetic and measured water levels. For example, the bold, green lines in Figure 10 are the superposition Theis models. The small, black crosses in Figure 10 are the estimated drawdowns which are the unexplained noise plus the superposition Theis model.

The number of measurement observations was reduced by averaging drawdowns from a well every 6 hours. Averaging reduced the number of measurement observations from more than 17,000 to 3,558 and suppressed high-frequency noise. Totals of 1,993 and 1,565 measured drawdowns were compared during the ER-20-8 #2 and ER-EC-11 aquifer-test analyses, respectively. Reliable drawdowns that were not affected by pumping losses, heating effects, or abridged records totaled 3,388 and were observed in 23 wells. These reliable observations were assigned weights of 1, except at the ER-EC-6 well cluster during the ER-EC-11 aquifer test. Weights of 0.6 were assigned to large measured drawdowns from the ER-EC-6 cluster to reduce sensitivity to these observations.

Compromised observations in wells that were affected by pumping losses and heating effects were assigned small weights so that hydraulic-conductivity estimates were minimally affected. Measured drawdowns in well ER-20-8 #2 main during the ER-20-8 #2 aquifer test were uncertain because of the strong correlation between pumping losses and aquifer response; transmissivity was estimated to exceed 10,000 ft²/d. Assigned weights of 0.05 reflected the large uncertainty associated with these observations. Measured drawdowns in wells ER-EC-11 main and ER-EC-11 upper intermediate during the ER-EC-11 aquifer test were very uncertain because of pumping losses and heating effects. Assigned weights of 0.002 and 0.05, respectively, reflected the large uncertainty associated with these observations.

Tikhonov regularization limited hydraulic-conductivity estimates at pilot points to reasonable values (Doherty, 2003). Sharp differences between nearby values in similar simplified HSUs were penalized to ensure relatively continuous hydraulic-conductivity distributions. Unrealistic hydraulic-conductivity distributions were avoided by limiting the fit between measured and simulated observations (Fienen and others, 2009). This irreducible, weighted measurement error combined measurement and numerical model errors.

Regularization observations were equations that defined preferred relations between hydraulic-conductivity estimates. Regularization observations affected calibration most where the models were insensitive to measurement observations. Imposing preferred states, such as homogeneity, with regularization observations was preferable to rigid zones of assigned homogeneity. This was because hydraulic-conductivity estimates could differ where dictated by measured drawdowns.

Homogeneity within simplified HSUs was the primary preferred relation between pilot points and was applied equally to fault-structure zones and to undifferentiated (not part of a fault structure) zones within a simplified HSU. Hydraulic conductivity of the fault-structure zones could depart from values in the undifferentiated zones if dictated by misfits between simulated and measured drawdowns. Minimum variance of hydraulic-conductivity estimates in all HSUs that were classified as confining units was an additional preferred relation. Similar preferred relations also were defined for HSUs that were classified as aquifers. About 490 regularization observations constrained hydraulic-conductivity estimates with these preferred relations.

Goodness of Fit and Investigated Volumes

RMS was used to gauge the misfit between simulated and measured drawdowns. The RMS for the models was 0.03 ft, which was similar to the drawdown-detection threshold of 0.05 ft. Simulated and measured drawdowns visually agreed in all 23 wells with reliable observations (Appendix C). Differences between simulated and measured drawdowns in well ER-EC-6 shallow during the ER-20-8 #2 and ER-EC-11 aquifer tests are shown in Figure 15. The fit between simulated and measured drawdowns was best and worst in wells ER-EC-1 and ER-20-7, respectively, during the ER-EC-11 aquifer test. The RMS error of 0.05 ft for well ER-20-7 was a small discrepancy relative to maximum simulated and measured drawdowns of 0.14 and 0.25 ft, respectively. The fit in well ER-20-7 during the ER-EC-11 aquifer test could have been improved, but the hydraulic-conductivity estimate for the sCHZCM at ER-20-7 would exceed 20 ft/d.

Figure 15 . Simulated and measure drawdowns in well ER-EC-6 shallow during the ER-20-8 #2 and ER-EC-11 aquifer tests.

 

 

The investigated volumes of these aquifer tests reasonably can be defined by the volumes where simulated drawdowns exceed 0.05 ft at the end of pumping. The investigated-volume threshold of 0.05 ft was supported by the drawdown-detection threshold of 0.05 ft and the overall RMS error of 0.03 ft. The defined volume with a threshold of 0.05 ft also could be interpreted as a minimum investigated volume if based on model misfit in distant observation wells where the maximum measured drawdown does not exceed 0.1 ft. This is because the RMS error in these distant observation wells is 0.01 ft, which could support a drawdown threshold of less than 0.05 ft.

A greater volume of aquifer was investigated by the ER-EC-11 aquifer test than the ER-20-8 #2 aquifer test (Figure 16). The volume investigated was greater, in part, because cumulative pumpage was three times greater during the ER-EC-11 aquifer test (Figure 7 and Figure 9). Aquifer properties around well sites ER-20-8 #2 and ER-EC-11 also affected the investigated volume. Well ER-20-8 #2 main is completed in the BA/SPA, which locally has transmissivities of 100,000 ft2/d and is unconfined. Well ER-EC-11 main was completed in the TCA and TSA, which locally has transmissivities of 2,000 ft2/d and is confined. Drawdowns could propagate far from well site ER-EC-11 because of confined conditions and because the BA/SPA is 300 ft above the top of the pumped interval (Figure 4).

 

Figure 16. Simulated drawdowns from pumping wells ER-20-8 #2 and ER-EC-11 at the end of pumping during each aquifer test. Drawdown threshold is 0.05 ft.

 


Hydraulic-Property Estimates

Estimated transmissivities for four selected groupings of HSUs at various pilot-point locations are presented in Table 12. Estimated and assigned hydraulic conductivities for all pilot points are given in  Table 13. Visual inspection of these tables shows that the BA/SPA, which comprises lava flows, is the dominant aquifer in the model area. The estimated hydraulic conductivity of this unit ranges from 4 to 150 ft/d, and is generally one to two orders of magnitude larger than any of the other simplified HSUs modeled ( Table 13). The BA/SPA accounts for much more than half of the bulk transmissivity where present; at well site ER-20-8, it accounts for 95 percent of the transmissivity (Table 12). The dominance of the transmissivity by the BA/SPA and the anomalously large hydraulic conductivity and transmissivity in the area of ER-20-8 (Appendix D) may be caused by the presence of the SPA (Figure 4), which is limited primarily to the area east of ER-20-8 (Prothro and Drellack, 1997).

 

Table 12. Estimated transmissivities of selected groupings of hydrostratigraphic units at seven mapped locations simulated in the ER-20-8 #2 and ER-EC-11 numerical flow models.

[All values are in feet squared per day; na is Not Applicable; FCCU is the Fluorspar Canyon confining unit; BA/SPA is the Benham and Scrugham Peak aquifers; Paintbrush Tuffs includes the Upper Paintbrush confining unit (UPCU), Middle Paintbrush confining unit (MPCU), Tiva Canyon aquifer (TCA), lower Paintbrush confining unit (LPCU), and Topopah Spring aquifer (TSA); sCHZCM includes the Calico Hills zeolitic composite unit (CHZCM), Belted Range aquifer (BRA), and Inlet aquifer (IA)]

Transmissivity of the BA/SPA near well site ER-20-8, 100,000 ft²/d, is relatively uncertain because the water-level record in well ER-20-8 shallow is poor. Water level measurements were affected by instrument drift, potential offsets, and limited record. A rising trend of from 0.03 to 0.05 ft/d exists in well ER-20-8 shallow that is not present in wells ER-20-8 intermediate and ER-20-8 deep (Figure 17). Recovery was not observed and offsets of more than 0.2 ft potentially were introduced by not measuring water levels during the first week of recovery. Environmental water-level fluctuations could not be estimated independently of pumping effects. This was because water levels were not monitored in well ER-20-8 shallow until more than 1,080,000 gallons had been pumped from well ER-20-8 #2 main. Less than 45 percent of the cumulative pumpage occurred while well ER-20-8 shallow was instrumented.

 Figure 17 . Water-level changes in wells ER-20-8 shallow, ER-20-8 intermediate, and ER-20-8 deep and pumpage from ER-20-8 #2 during ER-20-8 testing.

 

A vertical-to-horizontal anisotropy of 1 was estimated but cannot be considered an independent estimate. This is because the estimate was constrained by an assigned maximum vertical-to-horizontal anisotropy of 1. A non-unique estimate was expected because vertical-to-horizontal anisotropy was correlated highly with horizontal hydraulic conductivity.

The estimates of hydraulic conductivity in the Paintbrush tuff units (TSA, LPCU, TCA, and sUPCU), which comprise alternating layers of welded tuff aquifers and nonwelded tuff confining units, showed little difference between the aquifers and confining units ( Table 13). Estimates ranged from 0.5 to 5 ft/d, but averaged 2 ft/d for all Paintbrush tuffs throughout the model area (Appendix D). The bulk transmissivity of the Paintbrush tuffs (900–3,600 ft2/d) was relatively small compared to the total transmissivity estimated for the model area (Table 12).

Hydraulic-conductivity differences between the aquifers and confining units within the Paintbrush tuff units would have been less had hydraulic-conductivity estimates in the sUPCU and LPCU not been constrained by a maximum of 2 ft/d. Estimates at 7 of the 30 pilot points in the sUPCU and LPCU were constrained. These constraints were estimated from flow log and single-well aquifer test interpretation in wells ER-EC-1 and ER-EC-4 (Garcia and others, 2010).  

The sCHZCM, comprising interlayered tuffs and lava flows, had comparable or slightly higher estimates of hydraulic conductivity and transmissivity than the Paintbrush tuffs (Table 12 and  Table 13). These estimates compare well with estimates of hydraulic conductivity and transmissivity for the CHZCM in an aquifer test at U-20 WW. The bulk transmissivity estimate from the U-20 WW test was 3,700 ft2/d (U.S. Geological Survey, 2009).

The estimates of hydraulic conductivity in the fault zones differed little from estimates distributed throughout each HSU outside of the fault zones ( Table 13). This suggests that the primary hydraulic influence of these fault structures may be to juxtapose aquifers and confining units, rather than to create barriers or conduits for flow in a zone along the fault plane.

 

 Table 13 . Estimated and assigned hydraulic conductivities for the 115 pilot points that were simulated in the ER-20-8 #2 and ER-EC-11 numerical flow models.

References Cited

Doherty, J., 2008a, PEST: Model-Independent Parameter Estimation. Brisbane, Australia: Watermark Numerical Computing.

Doherty, J., 2008b, PEST Groundwater Data Utilities. Brisbane, Australia: Watermark Numerical Computing.

Doherty, J., and J.M. Johnston, 2003, Methodologies for calibration and predictive analysis of a watershed model: Journal of American Water Resources Association, v. 39, no. 2, p. 251-265.

Fienen, M., Muffels, C., and Hunt, R., 2009, On constraining pilot point calibration with regularization in PEST, Ground Water, v. 47, no.6, p. 835–844.

Garcia, C.A., Halford, K.J., and Laczniak, R.J., 2010, Interpretation of flow logs from Nevada Test Site boreholes to estimate hydraulic conductivity using numerical simulations constrained by single-well aquifer tests: U.S. Geological Survey Scientific Investigations Report 2010–5004, 28 p.

Halford, K.J., 2006a, Documentation of a spreadsheet for time-series analysis and drawdown estimation: U.S. Geological Survey Scientific Investigations Report 2006-5024, 38 p.

Harbaugh, A.W., Banta, E.R., Hill, M.C., and McDonald, M.G., 2000, MODFLOW-2000, the U.S. Geological Survey modular ground-water model -- User guide to modularization concepts and the Ground-Water Flow Process: U.S. Geological Survey Open-File Report 00-92, 121 p.

Laczniak, R.J., Cole, J.C., Sawyer, D.A., and Trudeau, D.A., 1996, Summary of hydrogeologic controls on ground-water flow at the Nevada Test Site: U.S. Geological Survey Water-Resources Investigations Report 96-4109, 59 p.

Navarro-Intera, LLC, 2010a, Written communication prepared for U.S. Department of Energy-National Nuclear Security Administration Nevada Site Office, Subject: “Pahute Mesa ER-20-8 Well Data Report, Preliminary Revision No.: 0", April 2010.

Navarro-Intera, LLC, 2010b, Written communication prepared for U.S. Department of Energy- National Nuclear Security Administration Nevada Site Office, Subject: “Pahute Mesa ER-20-8#2 Well Data Report, Preliminary Revision No.: 0", April 2010.

Navarro-Intera, LLC, 2010c, Written communication prepared for U.S. Department of Energy-National Nuclear Security Administration Nevada Site Office, Subject: “Pahute Mesa ER-EC-11 Well Data Report, Preliminary Revision No.: 0" , April 2010.

Prothro, L.B. and Drellack, S.L., Jr., 1997, Nature and extent of lava-flow aquifers beneath Pahute Mesa, Nevada Test Site: U.S. Department of Energy Report DOE/NV/11718-156, 50 p.

RamaRao, B.S., de Marsily, G., and Marietta, M.G., 1995, Pilot point methodology for automated calibration of an ensemble of conditionally simulated transmissivity fields 1. Theory and computational experiments: Water Resources Research, v. 31, no. 3, p. 475–493.

Stoller-Navarro Joint Venture, 2009, Written communication prepared for U.S. Department of Energy-National Nuclear Security Administration Nevada Site Office, Subject: “UGTA Morning Report: Well ER-EC-6", May 12-14, 2009.

Theis, C.V., 1935, The relation between the lowering of the piezometric surface and the rate and duration of discharge of a well using groundwater storage: Am. Geophys. Union Trans., vol. 16, pp. 519-524.

U.S. Department of Energy, 1997a, Completion report for well ER-20-1 [DRAFT]: U.S. Department of Energy Report DOE/NV/11718-XX/UC-700.

U.S. Department of Energy, 1997b, Completion report for well cluster ER-20-5: U.S. Department of Energy Report DOE/NV-466/UC-700.

U.S. Department of Energy, 1998, Completion report for well cluster ER-20-6: U.S. Department of Energy Report DOE/NV-467/UC-700.

U.S. Department of Energy, 2000a, Completion report for Well ER-EC-1: U.S. Department of Energy Report DOE/NV/11718—381.

U.S. Department of Energy, 2000b, Completion report for well cluster ER-EC-6: U.S. Department of Energy Report DOE/NV/11718-360.

U.S. Department of Energy, 2009, Phase II corrective action investigation plan for Corrective Action Units 101 and 102--Central and western Pahute Mesa, Nevada Test Site, Nye County, Nevada: U.S. Department of Energy Report DOE/NV--1312, Rev. 2, 255 p.

U.S. Department of Energy, 2010a, Completion report for well ER-20-7, Corrective Action Units 101 and 102: Central and Western Pahute Mesa: U.S. Department of Energy Report DOE/NV--1386.

U.S. Department of Energy, 2010b, Completion report for wells ER-20-8 and ER-20-8 #2, Corrective Action Units 101 and 102: Central and Western Pahute Mesa, [DRAFT]: U.S. Department of Energy Report DOE/NV--XXXX.

U.S. Geological Survey, 2009, 'Nevada Water Science Center Aquifer Tests' webpage, accessed August 31, 2010, at http://nevada.usgs.gov/water/aquifertests/index.htm

U.S. Geological Survey, 2010, 'USGS/U.S. Department of Energy Cooperative Studies in Nevada' webpage, accessed August 6, 2010, at http://nevada.usgs.gov/doe_nv

 

 

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo USA.gov logo U.S. Department of the Interior | U.S. Geological Survey
URL: http://nevada.usgs.gov/barcass/index.htm
Page Contact Information: Nevada Water Science Center Web Team
Page Last Modified: February 25, 2015 -->