Observation well network design for pumping tests in unconfined aquifers

Xunhong Chen, James Goeke, Jerry F. Ayers, Scott Summerside

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

This paper presents a method for determining locations of observation wells to be used in conducting pumping tests in unconfined aquifers. Sensitivity coefficients, the distribution of relative errors, and the correlation coefficients between four aquifer parameters (horizontal and vertical hydraulic conductivities Kr and Kz, storage coefficient S, and specific yield Sy) are used as the criteria for the design of observation well networks and the interpretation of pumping tests. The contours of the relative errors over a vertical profile are very useful in selecting the "best" location of an observation well. Results from theoretical analyses suggest that a wide range of locations is suitable for the determination of Kr and that good locations for the determination of Kz and S may be poorly suited for the determination of Sy. Consideration must be given to the position and lengths of the pumping well screen in the selection of observation well locations. For a given location, the quality of test data can be improved by using high pumping rates and frequent sampling of drawdowns. We found that a minimum of two and preferably three observation locations are needed along a given transect. Results of the four parameters from a single well analysis may contain higher uncertainties. However, composite analyses of multiple observation wells can reduce the correlation between the four aquifer parameters, particularly between Kr and Sy, thus, improving the quality of parameter estimation. Results from two pumping tests conducted at sites located in Nebraska were examined with regard to the proposed methodology.

Original languageEnglish (US)
Pages (from-to)17-32
Number of pages16
JournalJournal of the American Water Resources Association
Volume39
Issue number1
DOIs
StatePublished - Feb 2003

Fingerprint

network design
unconfined aquifer
Aquifers
pumping
well
aquifer
drawdown
Hydraulic conductivity
vertical profile
test
hydraulic conductivity
Parameter estimation
transect
Sampling
methodology
sampling
Composite materials
parameter

Keywords

  • Aquifer hydraulic parameters
  • Observation well locations
  • Pumping tests
  • Sampling frequency
  • Sensitivity analysis
  • Unconfined aquifers

ASJC Scopus subject areas

  • Ecology
  • Water Science and Technology
  • Earth-Surface Processes

Cite this

Observation well network design for pumping tests in unconfined aquifers. / Chen, Xunhong; Goeke, James; Ayers, Jerry F.; Summerside, Scott.

In: Journal of the American Water Resources Association, Vol. 39, No. 1, 02.2003, p. 17-32.

Research output: Contribution to journalArticle

Chen, Xunhong ; Goeke, James ; Ayers, Jerry F. ; Summerside, Scott. / Observation well network design for pumping tests in unconfined aquifers. In: Journal of the American Water Resources Association. 2003 ; Vol. 39, No. 1. pp. 17-32.
@article{7f3bafa858a34944a99087a523cda6bf,
title = "Observation well network design for pumping tests in unconfined aquifers",
abstract = "This paper presents a method for determining locations of observation wells to be used in conducting pumping tests in unconfined aquifers. Sensitivity coefficients, the distribution of relative errors, and the correlation coefficients between four aquifer parameters (horizontal and vertical hydraulic conductivities Kr and Kz, storage coefficient S, and specific yield Sy) are used as the criteria for the design of observation well networks and the interpretation of pumping tests. The contours of the relative errors over a vertical profile are very useful in selecting the {"}best{"} location of an observation well. Results from theoretical analyses suggest that a wide range of locations is suitable for the determination of Kr and that good locations for the determination of Kz and S may be poorly suited for the determination of Sy. Consideration must be given to the position and lengths of the pumping well screen in the selection of observation well locations. For a given location, the quality of test data can be improved by using high pumping rates and frequent sampling of drawdowns. We found that a minimum of two and preferably three observation locations are needed along a given transect. Results of the four parameters from a single well analysis may contain higher uncertainties. However, composite analyses of multiple observation wells can reduce the correlation between the four aquifer parameters, particularly between Kr and Sy, thus, improving the quality of parameter estimation. Results from two pumping tests conducted at sites located in Nebraska were examined with regard to the proposed methodology.",
keywords = "Aquifer hydraulic parameters, Observation well locations, Pumping tests, Sampling frequency, Sensitivity analysis, Unconfined aquifers",
author = "Xunhong Chen and James Goeke and Ayers, {Jerry F.} and Scott Summerside",
year = "2003",
month = "2",
doi = "10.1111/j.1752-1688.2003.tb01558.x",
language = "English (US)",
volume = "39",
pages = "17--32",
journal = "Journal of the American Water Resources Association",
issn = "1093-474X",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - Observation well network design for pumping tests in unconfined aquifers

AU - Chen, Xunhong

AU - Goeke, James

AU - Ayers, Jerry F.

AU - Summerside, Scott

PY - 2003/2

Y1 - 2003/2

N2 - This paper presents a method for determining locations of observation wells to be used in conducting pumping tests in unconfined aquifers. Sensitivity coefficients, the distribution of relative errors, and the correlation coefficients between four aquifer parameters (horizontal and vertical hydraulic conductivities Kr and Kz, storage coefficient S, and specific yield Sy) are used as the criteria for the design of observation well networks and the interpretation of pumping tests. The contours of the relative errors over a vertical profile are very useful in selecting the "best" location of an observation well. Results from theoretical analyses suggest that a wide range of locations is suitable for the determination of Kr and that good locations for the determination of Kz and S may be poorly suited for the determination of Sy. Consideration must be given to the position and lengths of the pumping well screen in the selection of observation well locations. For a given location, the quality of test data can be improved by using high pumping rates and frequent sampling of drawdowns. We found that a minimum of two and preferably three observation locations are needed along a given transect. Results of the four parameters from a single well analysis may contain higher uncertainties. However, composite analyses of multiple observation wells can reduce the correlation between the four aquifer parameters, particularly between Kr and Sy, thus, improving the quality of parameter estimation. Results from two pumping tests conducted at sites located in Nebraska were examined with regard to the proposed methodology.

AB - This paper presents a method for determining locations of observation wells to be used in conducting pumping tests in unconfined aquifers. Sensitivity coefficients, the distribution of relative errors, and the correlation coefficients between four aquifer parameters (horizontal and vertical hydraulic conductivities Kr and Kz, storage coefficient S, and specific yield Sy) are used as the criteria for the design of observation well networks and the interpretation of pumping tests. The contours of the relative errors over a vertical profile are very useful in selecting the "best" location of an observation well. Results from theoretical analyses suggest that a wide range of locations is suitable for the determination of Kr and that good locations for the determination of Kz and S may be poorly suited for the determination of Sy. Consideration must be given to the position and lengths of the pumping well screen in the selection of observation well locations. For a given location, the quality of test data can be improved by using high pumping rates and frequent sampling of drawdowns. We found that a minimum of two and preferably three observation locations are needed along a given transect. Results of the four parameters from a single well analysis may contain higher uncertainties. However, composite analyses of multiple observation wells can reduce the correlation between the four aquifer parameters, particularly between Kr and Sy, thus, improving the quality of parameter estimation. Results from two pumping tests conducted at sites located in Nebraska were examined with regard to the proposed methodology.

KW - Aquifer hydraulic parameters

KW - Observation well locations

KW - Pumping tests

KW - Sampling frequency

KW - Sensitivity analysis

KW - Unconfined aquifers

UR - http://www.scopus.com/inward/record.url?scp=0037297106&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037297106&partnerID=8YFLogxK

U2 - 10.1111/j.1752-1688.2003.tb01558.x

DO - 10.1111/j.1752-1688.2003.tb01558.x

M3 - Article

AN - SCOPUS:0037297106

VL - 39

SP - 17

EP - 32

JO - Journal of the American Water Resources Association

JF - Journal of the American Water Resources Association

SN - 1093-474X

IS - 1

ER -