Operational Use of the New Swiss SNOWPACK Model:
A Method for Improved Winter Precipitation Estimates
at High Alpine Sites
Michael Lehning1 and Perry Bartelt1
The new SNOWPACK model developed at the Swiss Federal Institute for Snow and Avalanche
Research is a 1D momentum, mass, and energy balance model. Its current version includes impor-
tant features such as a numerical solution of the instationary heat transfer and creep/ settlement
equations, a complete surface energy balance, phase changes, water transport, and snow microstruc-
ture development (metamorphism). The microstructural parameters are linked to the thermal con-
ductivity and the snow viscosity. The Lagrangian finite element solution allows for a realistic repre-
sentation of the layered snowcover structure.
The model is already in operational use. It calculates the local snowcover characteristics for approx-
imately 40 automatic Swiss high alpine snow and weather stations located at altitudes between 2000
and 3000 m a.s.l. The stations provide the necessary model input air temperature, humidity, wind
speed, surface temperature, reflected short wave radiation, and total snow height. These remote sta-
tions operate autonomously on solar energy. The main purpose of the stations and the associated
model calculations is to provide weather and snowcover information to the avalanche warning ser-
vice. One of the key parameters for avalanche warning as well as hydrology is the amount of new
snow deposited during a certain time interval. Because only the total snow height is measured and no
direct precipitation measurements are possible (no external power supply), the precipitation rate is
calculated from the increase in the snowcover height, taking into account the modeled settlement
rates of the snowpack. This requires a sophisticated data control routine because the snow height
measurements are often erroneous. In addition, a model for the estimation of the new snow density is
necessary. The method has been implemented into the SNOWPACK model and evaluated against
measurements at the Weissfluhjoch experimental site.
The evaluation has been carried out for the winter seasons 96/97 and 97/98. The estimated amounts
of new snow from the model as well as the modeled total water equivalent have been compared
against manual measurements of new snow amounts (daily) and total water equivalent (twice per
month) and against the automatic measurements from the standard precipitation gauge as used by the
Swiss Weather Service. The gauges are especially equipped for high alpine snow precipitation. With-
out any parameter fitting, the agreement between the model estimates and the manual measurements
is very good, while the automatic precipitation measurements typically underestimate the precipita-
tion by 30%. We conclude that our method to combine controlled measured snowpack heights with
our density estimation plus the modeled snowpack settling gives reliable winter snow precipitation
rates with a high temporal (30 minutes) and spatial resolution at high alpine sites. Such information
is desperately needed by hydrologists, meteorologists, and avalanche specialists, but could not be
obtained with conventional methods until now.
1 Swiss Federal Institute for Snow and Avalanche Research, Flelastrasse 11, CH-7260 Davos Dorf, Switzer-
land. Correspond to "lehning@slf.ch".
64