Measuring material properties of a wet and draining snowpack using crosshole radar tomography

TitleMeasuring material properties of a wet and draining snowpack using crosshole radar tomography
Publication TypeConference Proceedings
Year of Conference2002
AuthorsWilliams, M. W., and Knoll M.
Conference Name70th Annual Western Snow Conference
Series TitleProceedings of the 70th Annual Western Snow Conference
Date PublishedMay 2002
PublisherWestern Snow Conference
Conference LocationGranby, Colorado
KeywordsLiquid water, Meltwater, Flow paths, Pulse radar, Dielectric constant
Abstract

Movement of liquid water through snow is one of the least understood aspects of snow hydrology. Most methods of investigating melt water flow through snow involve sparse or invasive sampling. This makes it difficultto study scale and time-dependent processes such as the evolution of preferential flowpaths. Imaging of internalsnow pack properties using radar wavelengths shows promise as a non-invasive technique. Here we report on aproof-of-concept experiment conducted to evaluate the potential of crosshole radar tomography to image and characterize the material properties of a wet and draining snow pack. The experiment was conducted using a pulseradar system with 250 MHz borehole antennas, each placed in 3 PVC pipe, 5 m in length. One pipe was located atthe bottom of the snow pack and the second at the top of the snow pack; both pipes were placed perpendicular tothe fall line of the snow field. Travel velocities ranged by more than a factor of two, from about 125 m u,s'' to 300m u,s'1, showing that material properties of the snow pack were influencing the propagation of radar waves throughthe snow pack. We then inverted the traveltime data to estimate the dielectric constant for each grid cell using acurved ray traveltime tomography algorithm. The estimated dielectric constants ranged from 1 to 4, similar to the 2.3 to 3.1 measured in a nearby snow pit using a Denoth Wetness Sensor. Conversion of the dielectric constants toliquid water content yield values of 1-3, slightly less than the 1.5 to 4.5 estimated by the Denoth WetnessSensor.

URLsites/westernsnowconference.org/PDFs/2002Williams.pdf