HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Figures Only Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Farmani, M. B. Articles by Keers, H. Search for Related Content PubMed Articles by Farmani, M. B. Articles by Keers, H. GeoRef GeoRef Citation Agricola Articles by Farmani, M. B. Articles by Keers, H. Related Collections Ground Penetrating Radar, GPR Tomography Inverse Procedures/Parameter Estimation

ORIGINAL RESEARCH

Estimation of Unsaturated Flow Parameters Using GPR Tomography and Groundwater Table Data

^a Dep. of Geosciences, Univ. of Oslo, 1022 Blindern, 0315 Oslo, Norway
^b Norwegian Institute for Agricultural and Environmental Research, Frederik A. Dahls vei 20, 1432 Ås, Norway
^c Schlumberger Cambridge Research, Madingley Rd., Cambridge CB3 0EL, UK
^* Corresponding author (m.b.farmani{at}geo.uio.no).

Received 8 November 2007.

Flow parameters in the vadose zone of an ice-contact delta near Oslo's Gardermoen Airport were estimated by inverse flow modeling combining time-lapse ground-penetrating radar (GPR) travel-time (TT) tomography and groundwater (GW) table data. Natural water infiltration from the 2006 snowmelt was used as the upper boundary condition in forward flow simulations. The lower boundary condition was the outflux of water from the vadose zone derived by time derivative of the GW table. The inverse flow modeling was done using different combinations of time-lapse GPR TT tomography and time series of the level of the GW table. The purpose was to assess the improvements in parameter estimations and the information value of the different data sets. Flow parameters estimated by conditioning only on time-lapse GPR TT tomography capture the development of the wet front but fail to simulate the GW table fluctuations. Inverse flow modeling conditioned on only the GW table did not simulate the wet front correctly but decreased the objective function better than conditioning on time-lapse GPR data. If the inversion is conditioned on both time-lapse GPR data and the GW table fluctuations, the final estimates of the flow parameters are close to the estimates from the inversion conditioned on only GW table data. This result was obtained because the moving GW table was monitored continuously in time while the GPR data were sampled only three times during the infiltration event; thus, observations of the GW table had higher weights in the objective function than did observations derived from GPR tomograms. Finally, we did forward flow modeling with the estimated parameter sets and compared the flow paths with an independent tracer experiment performed at the field site in 1999. The results showed that anisotropy in the intrinsic permeability was an important parameter that should be considered when simulating the flow paths. However, volumetric soil water content distribution is not strongly related to the anisotropy of intrinsic permeability. Therefore, anisotropy cannot be correctly estimated by inverse flow modeling conditioned on volumetric soil water content only.

Abbreviations: EM, electromagnetic • GPR, ground-penetrating radar • LM, Levenberg–Marquardt • SCE, shuffled complex evolution • 2D, two-dimensional • 3D, three-dimensional