GNGTS 2017 - 36° Convegno Nazionale

GNGTS 2017 S essione 3.2 651 Fig. 1 - Location of the five study sites in the Alpine chain (upper central panel); survey map of Aletsch glacier (a), Forni glacier (b), La Mare glacier (c) and Pian di Neve and Lobbia glaciers (d). Neve and Lobbia glaciers (Fig. 1d). In November 2013, passive seismic measurements and two mutually orthogonal RES surveys were conducted on the Forni glacier (Fig. 1b). In order to test the reliability of the HVSR method on deep and shallow glaciers, RES surveys and passive seismic measurements on the Aletsch glacier (April 2013 and May 2014; Fig. 1a) and on the La Mare glacier (April 2015 and May 2015; Fig. 1c) were carried out. The passive seismic measurements were conducted using different broadband seismometers, a Guralp, a Lennartz and a Nanometrics Trillium, whose corner periods are 30, 5 and 20 s, respectively. We also tested a Tromino sensor, whose corner period is not declared by the manufacturer. The Guralp and Lennartz sensors were connected to a Reftek data acquisition system, while the Trillium sensor was coupled to a Centaur data system. The two systems include an external GPS for positioning and timing purposes. In order to optimize the coupling, the sensors were protected using insulating covers and placed on a plexiglas or metal base directly in contact (when possible) with the ice, at the bottom of deep holes dug or excavated right into the snow and firn. The hole depth ranged from half a meter to ∼ 4 m depending on the snow coverage. This particular coupling greatly improved data quality, by reducing the noise generated by the wind and by protecting the sensor from the sun, which can lead to thermal drifts. The sensors were always oriented parallel to the ice flow direction. Discussion and conclusions. The HVSR technique has been tested on passive seismic data from several Alpine glaciers in the Adamello and Ortles-Cevedale massifs (Italy) and in the Bernese Oberland Alps (Switzerland), and from the WIS (west Antarctica). Results were validated using active seismic, GPR and geoelectric methods, meeting modern standards and widely used in exploration geophysics. The resonance frequency in the H/V spectra correlates well with the ice thickness at each considered site, in a wide range from a few tens of meters to over 800 m. In almost all the examined cases, the directional H/V spectra show that the resonance frequency is nearly azimuth independent while the maximum peak amplitude lies at low azimuths, denoting a longitudinal polarization of the recorded wavefield. Moreover,