ANALYSIS OF CONTINUOUS SNOW TEMPERATURE PROFILES FROM AUTOMATIC WEATHER STATIONS IN AOSTA VALLEY (NW ITALY) M. Maggioni, G. Filippa, E. Zanini, M. Freppaz Dipartimento di Scienze Agrarie, Forestali e Alimentari and NatRisk-LNSA, University of Torino, Italy ABSTRACT: Snow temperatures are strictly related to the heat balance of the snowpack and are commonly used to calculate the temperature gradient, which is one of the driving factors for snow metamorphisms. In this work we analyze the snow temperature profiles generated from the continuously recorded data from five automatic weather stations located at an elevation ranging between 2000 and 2800 m asl in Aosta Valley (Northwest Italy) in the period 1998-2009. The aims of this work are: a) to evaluate the influence of solar radiation on snow temperature profiles and on the calculation of thermal gradients; b) to compare the temperature profiles from the automatic weather stations with manually recorded data; c) to define an objective method to define the onset of an isothermal snowpack. We present some scientific findings but also some practical outputs, useful e.g for the Avalanche Warning Services: a) 60 cm is the surface layer thickness where relevant daily temperature changes occur; b) 8.00 am is the best time to calculate the thermal gradient; c) there is good agreement between the automatic and manual temperature profiles; d) contour plots may be an easy method to assess when the snowpack reaches isothermal conditions. 1. INTRODUCTION Snow temperatures are strictly related to the heat balance of the snowpack and are commonly used to calculate temperature gradients. Temperature gradients are responsible for the snow metamorphisms which can lead the snowpack towards stability or instability conditions. Therefore, it is common among the Avalanche Warning Services to periodically measure, beside other physical properties, the snow temperature profile in specific locations, in order to know the snowpack structure and its possible evolution. In some cases, the snow temperatures are continuously registered from automatic weather stations or are calculated by models, which are able to describe the snowpack structure only on the basis of few snow and meteorological data, such as for example air temperature, solar radiation, surface snow temperature (e.g. Brun et al., 1989; Morland et al., 1990; Jordan, 1991; Lehning et al, 1999). It is this last variable that is commonly used by models to determine the thermal gradient within the snowpack. Also when using data from manual snow profiles, the temperature gradient is calculated considering the snow surface temperature, the snow/soil interface temperature and the snow depth. But the snow surface temperature greatly varies during day and night time. Diurnal temperature fluctuation within the top portion of the snowpack is the result of the net energy balance at the snow surface, which includes different contributions. Among them the most relevant are the radiation fluxes: short wave radiation flux and net long wave radiation flux (Gray and Male, 1981). McClung and Sharer (1993) suggested that short wave radiations can penetrate within the snowpack to a depth of 10-20 cm. In other works (for ex. Fierz et al., 2008) the same threshold of 20 cm is reported. We found higher values only in Ohara and Kavvas (2006): the thickness of what they call “active layer”, where the snow temperature varies daily, reached about 60 cm. However, the depth of short wave penetration is strictly related to the properties of the surface snowpack layers (Bakermans and Jamieson, 2006). Therefore we think it is not possible to find a general rule to determine the thickness of Correspondig author address: Margherita Maggioni, Dipartimento di Scienze Agrarie, Forestali e Alimentari and NatRisk-LNSA, University of Torino, Via L. Da Vinci 44, 10095 Grugliasco (TO), Italy; tel. +39 011 6708522, fax. +39 011 6708692; email: margherita.maggioni@unito.it Proceedings, 2012 International Snow Science Workshop, Anchorage, Alaska 1017