WMOD00504

• Title: Papers Presented at the 17th Annual Canadian Meteorological and Oceanography Congress
• Date: 5/1/1983
• Weather Modification - Doc Type: Report

-

---. -, .-.. -~

value of about one station per 800 sq. km. Secondly,
most of the stations are at lower elevations with only
a few above the contour. These characteristics
underscore the difficulty in providing a detailed and
confident assessment of the snow climate.

The data sources vary substantially as to length of
record and seasonal coverage. The AES climate sta-
tions generally have a longer record length. Figure 3
indicates this by showing the number of A.E.S. stations
as a function of measurement record length. A bi-modal
distribution is evident with peaks of station installation
activity being at 20 and 60 years benchmarks. For
climate assessment, the AES network record length
seems quite adequate.

The different requirements and seasonal character
of the snow surveys result in a data set more limited
and variable than that of the A.E.S. stations. This
variability, evident in both record length and number
of stations, is shown for the Alberta and B.C. snow
survey sites (Fig. 4). The generally longer record of B.C.
data, particularly in the early months of the survey, is
apparent.

3.0 RESULTS

3.1 Seasonal Snow Variation

A plot of the calculated average monthly snowfalls
for all 56 mountain climate stations within the project
region shows that snowfall increases from a minimum
in the autumn to a monthly maximum of 50 cm in
January (Fig. 5). This mid-winter maximum may reflect
the higher efficiency of precipitation processes with
colder temperatures and sufficient atmospheric water
supply. The plateau of 30 cm to 40 cm through the late
winter and early spring months gives way to the rapid
decrease in late spring. The significant spring contribu-
tion suggests that the spring temperature-moisture
"seeding window" may occur during periods of high
natural precipitation.

Snowfall frequency (Fig. 6) follows a pattern similar
to that of snowfall amount. The January peak of about
10 snow-days is followed by a decline that accelerates
rapidly during the late spring. The lower April snow-day
frequency, combined with the higher snowfall amount,
suggests that in the late spring fewer but larger
snowfalls occur.

Snow water equivalent data give a more limited view
of actual snowfall, but they do provi<;1e relevant infor-
mation on the seasonal character of snowpack water
storage. The seasonal variation of the combined

Alberta, B.C. and Montana data (Fig. 7) shows tln~t over
half (approximately 250 mm) of the maximum
snowpack accumulates before the measurement pro-
gram begins in January. Additional contributions
throughout the winter-spring period results in a peak
snowpack of 500 mm on May 1. A rapid depletion of
the snowpack in late spring, which results in an in-
crease in water flow, is also evident.

3.2 Snow Variation With Elevation

The complex topography and higher elevations of
the mountain regions substantially affect snowfall
characteristics. The cooler temperatures at higher
elevations enhance snowfall occurrence and snow-
pack persistence. Station data were examined as to
their variability with elevation to determine the rela-
tive importance of this factor. A plot of average
annual snowfall against station elevation for the
mountain AES climate stations (Fig. 8) shows a strong
trend to increased snowfall amounts at higher eleva-
tions. Snowfall at lower elevation sites (e.g. 700 m MSL)
may be less than one-half the annual snowfall at higher
sites (e.g. 1500 m MSL). Included in Figure is is the
calculated linear regression relation which predicts
snow depth from elevation. The high, significant cor-
relation coefficient (0.73) suggests that about 50% of
the annual snowfall variation is explained by station
elevation.

The seasonal variation of snowfall with elevation as
measured by the correlation coefficient is more pro-
nounced (Fig. 9). The twin peaks of high cOHielations
in the early autumn and late spring may reflect the role
of cooler temperatures at higher elevations during those
periods. Snow at higher elevations would settle and re-
main first, and melt last, at higher elevation sites. The
mid-winter low correlation reflects the ubiquitous cold
temperatures within the region. The still-significant cor-
relation during the warmer season may reflect the in-
fluence of complex topography and cooler tE~mpera-
tures at higher elevations.

The elevation-snow relation obtained snow survey
data is less well-defined than that for average annual
snowfall. Greater variability is indicated, particularly
with the inclusion of the Montana data (Fig. 10). The
Montana values seem to be consistently higher than
those for Alberta and B.C., and may reflect differences
in network criteria, data measurement procedures and
quality control. The correlation coefficient of 059 (0.38
with Montana) suggests that only about 35% of the
snow depth variation can be explained by sitl~ eleva-
tion. The complex effects of topography on snow
storage may overshadow the effect of eleva.tion on
snow precipitation patterns.

4