Systematic identification of seismic signatures from medium-intensity landslides in the Arunachal Pradesh region, Northeast India, and their accurate locations
摘要
Seismic recordings offer high temporal resolution and are sensitive to remote ground vibrations. Recent studies have shown that seismic identification of landslides is useful for early detection and mitigation. However, most research in this topic is focused on high-volume landslides (> 500,000 m3) or rockfalls. In this study, we perform seismic analysis of medium-intensity landslides (volume < 10,000 m3) in the Arunachal Pradesh region. For this purpose, a network of closely spaced broadband stations (~ 3 km average spacing) was established near Itanagar in June 2022. Additionally, data from regional seismic networks operating in Arunachal Pradesh during 2009–2020 were also utilised. First, a landslide catalogue is compiled from various sources. Then, seismic data corresponding to the landslide dates were scanned to identify and validate landslide signals, followed by their seismic location using a source migration approach. Of the 22 analysed landslides during 2009–2020, seismic signals from 18 could be identified at one or two stations of the regional networks. The 2 analysed landslides near Itanagar during 2022–23 were clearly detected at almost all stations of the closely-spaced network, enabling accurate locations within an error of ~ 400 m. The waveform characteristics of these two landslides were found to be very similar. The accuracy of the landslide location is seen to be dependent on the central frequency of the signal recorded at all stations and their propagation velocity. This study acts as a stepping stone towards future automatic detection and classification of more landslides in the region and beyond.