Ten Years of Monitoring Areal Snowpack Using NOAA-AVHRR Radiometry
and Ground Measurements in the Southern Alps
Roberto Ranzi1, Giovanna Grossi1, and Baldassare Bacchi1
Cost-efficiency and fine temporal resolution of images from the satellite-borne NOAA-AVHRR
sensor indicate this source of information as a suitable candidate for monitoring snow cover in alpine
areas. This is important for the estimation of the water storage at the beginning of the snowmelt
season especially in view of irrigation, hydropower production, and water supply. The information
can also be used for the "internal" validation of snowmelt models. As a result of a long-term study,
ten years of snow cover area depletion curves have been estimated using remote sensing in seven
watersheds of size larger than some hundreds of square kilometres in the Southern Alps.
Coupling of satellite imagery with detailed topographic data and some ground measurements of
snowpack depth and density is shown to provide reliable regional estimates of snow water equiva-
lent in the Southern Alps. The study area (15809 km2) includes the major left-side tributaries to the
Po river, from the Sarca to the Ticino river, thus covering part of the Trentino-Alto Adige, Lombar-
dia, and Piemonte Regions in Italy and Switzerland. NOAA-AVHRR images are used for estimating
the extension of snow cover, and cloud cover is filtered out by using the relationship between snow
cover and topography. For the purpose, we used automated topographic data retrieved from a Digital
Elevation Model (DEM) with space gridding of about 230 230 m2 and resampled with a 250-
250-m2 grid spacing.
The validation of the distributed technique is carried out by comparing the basin estimates it pro-
vides with those obtained from the lumped water balance equation in some of the selected water-
sheds, computed for different snowmelt seasons. The satellite-based procedure is shown to result in
a slight underestimation of the water balance-based estimate of the water equivalent.
In spite of many uncertainties and open problems, remotely sensed data seem to provide realistic
estimates of snow water equivalent on alpine watersheds when ground measurements are available,
and topographic effects are properly accounted for. The long-term statistics of the presence of snow
in different altitude and aspect zones that can be derived using the proposed technique might provide
useful information for the management of water resources and also for studying physical, chemical,
and biological processes in mountain areas.
1
Dipartimento di Ingegneria Civile, Universit di Brescia, via Branze 38/40, 25123 Brescia, ITALY
89
Previous Page
