Monitoring giant landslide detachment planes in the era of big data analytics

0481928 - ÚSMH 2018 RIV CH eng C - Conference Paper (international conference)
Blahůt, Jan - Rowberry, Matthew David - Balek, Jan - Klimeš, Jan - Baroň, I. - Meletlidis, S. - Martí, Xavier
Monitoring giant landslide detachment planes in the era of big data analytics.
Advancing Culture of Living with Landslides. Volume 3 Advances in Landslide Technology. Cham: Springer, 2017 - (Mikoš, M.; Arbanas, Ž.; Yin, Y.; Sassa, K.), s. 333-340. ISBN 978-3-319-53486-2.
[World Landslide Forum /4./. Ljubljana (SI), 29.05.2017-02.06.2017]
R&D Projects: GA ČR(CZ) GJ16-12227Y
Institutional support: RVO:67985891
Keywords : monitoring networks * giant landslides * San Andrés Landslide * El Hierro * Canary Islands
OECD category: Geology

A small mesh of sensors which monitor movements across detachment planes of the giant San Andrés Landslide on the northeastern lobe of El Hierro in the Canary Islands was established in 2013. In this paper we present the results obtained over a two year period spanning from October 2013 to October 2015. Our results demonstrate that the detachment planes are affected by sinistral strike slip displacements and subsidence of the depleted mass of the landslide. While these general trends are consistent the movements recorded at particular monitoring points differ in detail as one site is characterised by progressive strike slip and dip slip trends while another is characterised by movement pulses and reversals in the sense of movement. These findings contrast markedly with suggestions that the giant landslide is inactive and demonstrate that its reactivation is a possibility which cannot be dismissed categorically. Big data analytics have been used to identify interdependence between the recorded movements and a range of climatic and geophysical variables such as seismic data, tidal data, and geomagnetic data. We have found that the recorded movements correlate only weakly or moderately with climatic and seismic parameters but strongly to the horizontal and vertical intensity of the magnetic field. These findings are rather unexpected and we emphasise that special care must be taken in pushing the conclusions of a purely numerical analysis. The advantages of adopting a big data mindset led us to make significant improvements to the instrumental infrastructure in early 2016. These incremental improvements to the small mesh of sensors are driven partly by our desire to understand the kinematic behaviour of landslide itself and partly by our desire to explore the potential of big data analytics in geoscientific research.
Permanent Link: http://hdl.handle.net/11104/0277377