GNGTS 2023 - Atti del 41° Convegno Nazionale

Session 3.2 ___ GNGTS 2023 frequency-domain parameters. The temporal rate of these events can indeed supply additional information on the rock glacier stability and ongoing modifications at the daily and seasonal scales. Results A summary of the obtained passive seismic results after a complete year of almost continuous recording is reported in Fig. 2. Seismic parameters are shown in comparison with air temperature (Fig. 2a) and precipitation (Fig. 2b) recorded during the same period at the meteorological station of Goillet Lake (green square in Fig. 1b). Ambient seismic noise spectral analysis of the stations located on the back and white lobes revealed amplification in different frequency bands. As an example, the H/V spectral ratio is shown in Fig. 2c for station S2. A first frequency peak, located around 8 Hz, was systematically recorded during the whole monitored period and related to the rock glacier-bedrock interface at a depth of approximately 25 m below the surface. This peak showed a value of 6 Hz on the white lobe (S4), probably indicating a deeper bedrock interface in this compartment. In both cases, no variations in the interested frequency band were depicted over time. By contrast, a second peak in the H2/V2 plot is depicted around 20 Hz at the beginning of the monitoring period (Fig. 2c). This peak shows significant fluctuations over time in both amplitude and frequency values. The trend of this peak is negatively correlated with the air temperature variations (Fig. 2a). As a consequence, it was interpreted as the fluctuation in the depth of the rock glacier active layer. Lower values of this second peak are indeed found in the summer months, likely indicating a deeper interface between the unfrozen and the frozen material within the rock glacier. At the end of the summer, with decreasing air temperature, the frequency values start to increase, mirroring the decrease in depth of the unfrozen shallow layer. During winter months, this second frequency peak is almost absent, corroborating the hypothesis of a completely frozen body. The spectral results are supported by the outcomes of ambient seismic noise cross-correlation. The velocity changes depicted between station S2 and S1 are shown in Fig. 2d with the related correlation coefficients (Fig. 2e), computed on the vertical components and in the frequency band 10-15 Hz, with respect to the average cross-correlogram of the first days of monitoring. Minor velocity changes, negatively correlated with the precipitation amount, are found during the warm months. At the end of the summer, the inner velocity of the rock glacier significantly increases (up to +15%) as a result of the freezing processes. Different cluster of microseismic events were recognized during the same period. The related peak frequency is shown in Fig. 2f, again for the station S2. A first cluster of low-frequency events, showing low-amplitude and low-duration was recorded through the monitored period (blue in Fig. 2f). These events were interpreted as related to the basal movements and sliding of the rock glacier. A second cluster of events (purple in Fig. 2f) exhibited peak frequencies variable with temperature and showing a trend similar to ambient seismic noise results (Fig. 2c and Fig. 2d).