¾«Æ·SMÔÚÏßӰƬ

Petrzelka, Jennifer L ¹ ; Erickson, Tyler A ² ; Williams, Mark W ³

¹ University of Colorado and INSTAAR
² Michigan Tech Research Insitute and Department of Civil and Environmental Engineering
³ University of Colorado and INSTAAR

In areas containing seasonal snowpacks, snowmelt contributes significantly to the hydrological cycle. Thus, quantifying the spatial distribution of flow through a snowpack is essential to accurate hydrograph interpretation and representation in snowmelt runoff modeling. Movement of liquid water through snowpacks is generally recognized to occur in distinct flow paths rather than as uniform flow through a homogeneous porous medium. Most methods of investigating meltwater flow through snow involve sparse or invasive sampling. This makes it difficult to study scale and time-dependent processes such as the evolution of preferential flowpaths.

McGurk and Marsh (1995) developed a method for quantifying the spatial distribution of flowpaths using a metal framework with an attached blade that sliced uniform sections of a snowpack. Using light transmission, photographs were taken of the thick-cut sections to identify flow paths. The metal framework design of McGurk and Marsh inspired the development of the snow guillotine. The structure of the snow guillotine allows the ability to take cross-sectional slices at 1-cm intervals of a snowpit where dye tracer has been applied. Photographs are taken of each cross-section over a one meter distance. Application of image processing and geostatistical analysis allows collection of high resolution (1cm3), three-dimensional data on meltwater flow through a snowpack.

Three snow guillotine experiments were conducted over the course of two melt seasons. Datasets obtained from the experiments were analyzed to address the following questions: What portion of the snowpack transmits meltwater? Under what conditions do vertical meltwater pathways occur? Are vertical meltwater flowpaths continuous across layer boundaries? Can dye tracer experiments be used to identify snowpack layer transitions that cannot be easily identified from snowpit sampling? Results indicated that the fraction of the snowpack transferring liquid water was highly variable between experiments, as well as within individual experiments with respect to depth. All three experiments illustrated preferential flowpaths, with the majority of vertical flow occurring in the upper 20-55cm of the snowpack, while no preferential flowpaths are apparent below 100cm. Layer interfaces were found to significantly affect the volume of dye, indicating dominance by lateral flow at these boundaries. These findings were supported by the decrease in probability with depth of finding vertical flow and an increase in the probability of finding lateral flow at layer interfaces. Preferential flowpaths decreased in distinctness with time after the snowpack reached isothermal conditions at 0oC. Implementation of the snow guillotine significantly improves field measurements of meltwater flow while providing three-dimensional, quantitative data of unprecedented spatial resolution.

McGurk, B. J. and P. Marsh, Flow-finger continuity in serial thick-sections in a melting Sierran snowpack, in Biogeochemistry of Seasonally Snow-Covered Catchments, edited by K.A. Tonnessen, M.W. Williams, and M. Tranter, pp. 81-88, International Association of Hydrological Sciences, Wallingford, Oxfordshire, UK, 1995.