EGU23-13287
https://doi.org/10.5194/egusphere-egu23-13287
EGU General Assembly 2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Landslide mechanisms unraveled by RFID monitoring with a Machine Learning approach

Arthur Charléty1, Mathieu Le Breton2, Eric Larose1, and Laurent Baillet1
Arthur Charléty et al.
  • 1Université Grenoble Alpes (UGA), ISTerre, Failles, ST MARTIN D HERES, France (arthur.charlety@univ-grenoble-alpes.fr)
  • 2Geolithe, Crolles, France

 Radio-Frequency Identification (RFID) shows great potential for earth-sciences applications [1], notably in landslide surface monitoring at high spatio-temporal resolution [2] with meteorological robustness [3]. Ten 865MHz RFID tags were deployed on part of a landslide (Harmalière) and continuously monitored for 12 months by a station composed of 4 reader antennas. 2D relative localization was performed using a Phase-of-Arrival approach [4,5], and compared with optical reference measurements.

    The spatio-temporal accuracy of the method allowed for a thorough exploration of the landslides mechanisms during a 6-months period of activity. Laplacian clustering was applied to the RFID data and groups of tags with coherent behavior were identified, allowing a fine description of the kinematic motion of the landslide blocks and various mass transfer mechanisms. Each identified block can be monitored individually. 
    
    Different deformation zones were highlighted on the monitored zone. The surface movement was initiated by the topmost blocks, transferring after several weeks to the bottom of the monitored zone. This opens the way to building a landslide mechanical model in order to interpret the acquired data.

    RFID landslide monitoring allows dense observation of ground surface movements at a centimeter scale and with sub-hourly time precision, and new results bring a finer understanding the the landslides inner mechanisms.

 

References :

[1] M. Le Breton, F. Liébault, L. Baillet, A. Charléty, E. Larose, and S. Tedjini,
“Dense and long-term monitoring of earth surface processes with passive
rfid—a review,” Earth-Science Reviews, p. 104225, 2022.

[2] M. Le Breton, L. Baillet, E. Larose, E. Rey, P. Benech, D. Jongmans, F. Guy-
oton, and M. Jaboyedoff, “Passive radio-frequency identification ranging, a
dense and weather-robust technique for landslide displacement monitoring,”
Engineering geology, vol. 250, pp. 1–10, 2019.

[3] M. Le Breton, L. Baillet, E. Larose, E. Rey, P. Benech, D. Jongmans, and
F. Guyoton, “Outdoor uhf rfid: Phase stabilization for real-world appli-
cations,” IEEE Journal of Radio Frequency Identification, vol. 1, no. 4,
pp. 279–290, 2017

[4] A. Charléty, M. Le Breton, E. Larose, and L. Baillet, “2d phase-based rfid lo-
calization for on-site landslide monitoring,” Remote Sensing, vol. 14, no. 15,
p. 3577, 2022. 

[5] P. V. Nikitin, R. Martinez, S. Ramamurthy, H. Leland, G. Spiess, and
K. Rao, “Phase based spatial identification of uhf rfid tags,” in 2010 IEEE
International Conference on RFID (IEEE RFID 2010), pp. 102–109, IEEE,
2010

How to cite: Charléty, A., Le Breton, M., Larose, E., and Baillet, L.: Landslide mechanisms unraveled by RFID monitoring with a Machine Learning approach, EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-13287, https://doi.org/10.5194/egusphere-egu23-13287, 2023.