Comparison of Instream Methods for Measuring Hydraulic Conductivity in Sandy Streambeds

Authors

Matthew K. Landon, U.S. Geological Survey
David L. Rus, U.S. Geological Survey
F. Edwin Harvey, University of Nebraska - LincolnFollow

Document Type

Article

Date of this Version

12-2001

Comments

Published in GROUND WATER Vol. 39, No. 6, November-December 2001 (pages 870-885).

Abstract

Streambed hydraulic conductivity (K) values were determined at seven stream transects in the Platte River Basin in Nebraska using different instream measurement techniques. Values were compared to determine the most appropriate technique(s) for use in sandy streambeds. Values of K determined from field falling- and constant-head permeameter tests analyzed using the Darcy equation decreased as permeameter diameter increased. Seepage meters coupled with hydraulic gradient measurements failed to yield K values in 40% of the trials. Consequently, Darcy permeameter and seepage meter tests were not preferred approaches. In the upper 0.25 m of the streambed, field falling- and constant-head permeameter tests analyzed with the Hvorslev solution generally had similar K values that were significantly greater than those determined using the Hazen grain-size, Bouwer and Rice slug test for anisotropic and isotropic conditions, and Alyamani and Sen grain-size methods; median differences between these tests and the Hvorslev falling-head 60 cm diameter permeameter were about 8,9, 17, and 35 rn/day, respectively. The Hvorslev falling-head permeameter test is considered the most robust method for measuring K of the upper 0.25 m of the streambed because of the inherent limitations of the empirical grain-size methods and less sediment disturbance for permeameter than slug tests. However, lateral variability in K along transects on the Platte, North Platte, and Wood Rivers was greater than variability in K between valid permeameter, grain-size, or slug tests, indicating that the method used may matter less than making enough measurements to characterize spatial variability adequately. At the Platte River tributary sites, the upper 0.3 m of the streambed typically had greater K than sediment located 0.3 to 2.5 m below the streambed surface, indicating that deposits below the streambed may limit ground water/surface water fluxes. The Hvorslev permeameter tests are not a practical measurement approach for these greater depths. Thus, selection of a method for measuring streambed K needs to consider the vertical location of the sediments that are most likely to limit the rate of ground water/surface water interaction.