A study of the long-term, ground-water-quality monitoring needs at Wurtsmith Air Force Base near Oscoda, Michigan, was undertaken to support the U.S. Air Force Installation Restoration Program. The investigation develops a statistically-based design to increase the efficiency of ground-water-quality monitoring. The design is based on the use of site-specific monitoring objectives and constraints together with a statistical model of the possible time-varying contaminant concentrations.
To implement the design, a short-time interval over which the mean concentration is likely to be constant is specified for each site. Then, prior estimate of the mean concentration during this interval are computed using a trend function describing the historical concentrations. Parameters of the function can be estimated on the basis of techniques that are applicable to censored (some measurements below the analytical detection limit) and uncensored data. Because the concentration measurements are likely to be unequally spaced over time, prior estimates of the covariance structure are based on variogram analysis of the detrended historical concentrations. Computation of sample size and the effect of covariance on the sample size needed to satisfy monitoring objectives is illustrated with data on trichloroethylene concentrations in ground water at the Base.
Publication
Holtschlag, D.J., 1991, Installation Restoration Program, Phase II,
Confirmation/Quantification, Stage 2, Wurtsmith Air Force Base, Michigan: A
sequential sampling design for long-term, ground-water-quality monitoring: U.S.
Geological Survey Administrative Report.
Back to USGS, WRD Michigan Home Page
U.S. Department of the
Interior, U.S. Geological
Survey
Maintainer: Webmaster
<gs-w-milns_webmaster@usgs.gov>
Updated On: Thursday, September 21, 2000
Privacy Statement ||
Disclaimer
URL: http://mi.water.usgs.gov/reports/djhwafb.fm.html