remarkably constant in direction. Two direc- 
tions are important: (a) storm winds, which 
bring new snow, are generally from the west and 
(b) prevailing winds, which primarily serve to 
redistribute snow on the ground, are from the 
east. The two aircraft runways in the Prudhoe 
Bay area are aligned with the prevailing winds. 
Also, the lakes in this region have a pronounced 
tendency to become elongated in a direction 
perpendicular to the prevailing winds; an expla- 
nation of this was provided by Carson and 
Hussey (1962). West winds are slightly more 
important for snow drifting in the Prudhoe Bay 
area, while the east winds are more important 
for moving dust. Because of drifting, the snow 
thickness varies from almost nothing up to 2m 
in this region. Before discussing such variations, 
it will be useful for us to discuss the techniques 
of measurement and, in the next section, to 
consider the physical characteristics of the snow 
and some of the processes involved in_ its 
melting. 
Three methods of measurement were used to 
observe the snow itself: 
1. Pit studies: Standard snow stratigraphy 
studies as used in Greenland and parts of Arctic 
Alaska (Benson 1962, 1967, and 1969) were 
used to make detailed studies of the snowpack 
(Figs. 7 and 9). Thermometers and density 
sample tubes (500 cm? volume) were inserted 
horizontally into the exposed pit wall (Fig. 7). 
The samples were taken in such a way that 
density values could be calculated for each layer. 
The layers were plotted with sharp boundaries 
to indicate the stratified nature of the snow. 
2. Vertical cores: A large number of verti- 
cal cores were made by driving an aluminum 
tube, sharpened on one end, through the entire 
snowpack. Snow was removed from around the 
tube, a metal plate was inserted at the base, and 
the contents were placed into a plastic bag for 
measurements of average snow density, dust 
content, and electrical conductance of the 
derived meltwater. Before emptying the con- 
tents into the bag, the base of the sample was 
examined and any loose soil or grass was re- 
moved; if this material could not be removed 
another sample was taken. This sampling tech- 
nique was used on all of the traverses and dust 
collection sites. It has the advantage of sampling 
Fig. 2. 
Fig. 3. 
15 
Photo of snow, 6 September 1977. 
Pes 
Photo of snow, 13 April 1972. 
