The seismic network consists of eight 3-component geophones (4.5 Hz, GS11D, Geospace) installed in an area of about 250 x 150 meters on both sides of the top scarp at Åknes.
Since August 2006, data are recorded in triggered mode and automatically transferred from Åknes to NORSAR. There, the data are submitted to an automatic event detection (STA/LTA > 4 on at least 10 channels within a common time window of 1s). Each event is also automatically classified into one of the eight following classes:
- Slopequake: low-frequency (“Slope_LF”), high-frequency (“Slope_HF”), tremor-like (“Slope_tremor”) , successive slopequakes (“Slope_multi”). HF and successive slopequakes may be related to shearing on the sliding plane; LF slopequakes to fracture opening or formation of new cracks, and tremor-like slopequakes to water flow in cracks during thawing.
- Regional earthquake (“Regional”)
- “Noise” and electric “Spike”
More details on the automatic classifier implementation as well as on the interpretation of the different classes can be found in Langet & Silverberg (2023). Relevant information is summarised in the sketch below.
A follow-up work by Lee et al. (2022) showed that the automatic classifier performances can be further improved.