Use of Kriging to Estimate the Wolfcampian and San Andres Potentiometric Surfaces, Palo Duro Basin, Texas Panhandle

Access full-text files




Smith, D. Anderson
Orr, E. D.

Journal Title

Journal ISSN

Volume Title



Kriging is a statistical method historically used to estimate mineral reserves and more recently applied to estimate spatially-distributed or regionalized variables in hydrologic studies, such as ground-water fluid potential or transmissivity. In estimating these regionalized variables, kriging serves two main purposes. Firstly, it can generate a contour map of the variable with an associated measure of error. Secondly, it can optimize the location of additional samples, particularly valuable in acquiring data for ground-water potential in deep formations, which often involves drilling expensive wells.

Kriging has been previously employed to estimate water elevation in studies such as those conducted at the Department of Energy Hanford Reservation in the Pasco Basin, Washington (Doctor, 1979), and in a study of the multilayer aquifer system underlying Venice, Italy (Volpi and Gambolati, 1979). In both cases, actual water level measurements were utilized.

This report discusses the use of kriging methods by The University of Texas, Bureau of Economic Geology (UT/BEG), in the context of nuclear waste isolation feasibility studies in the Palo Duro Basin, funded by the Department of Energy under contract DE-AC97-80ET46615. Kriging has been employed by UT/BEG to estimate the ground-water potentiometric surfaces of the permeable San Andres cycle four dolomite and the Wolfcampian aquifer.

The San Andres Formation study utilized water level measurements and drill-stem-test (DST) pressures. Kriging was applied to minimize the variation of the DST data and reveal the regional potentiometric surface. The Wolfcamp study relied solely on DST pressures, with kriging employed to model the variation in the data and generate a contoured map of the potentiometric surface.

The kriging techniques employed by the Bureau of Economic Geology involve three steps: (1) variogram analysis, (2) kriging, and (3) contouring krige block estimates. Computer programs used for calculating the empirical variogram statistics, krige block estimates, and krige block variances were obtained from Dr. Young C. Kim and are detailed in Knudsen and Kim, 1978. For a comprehensive understanding of regionalized variables theory and kriging, please refer to Knudsen and Kim, 1978. A brief overview of the techniques used at the Bureau follows.


LCSH Subject Headings