ARATTANO M., 2002, The use of the Automatic Inclinometric System (AIS) for landslide early warning_ the case of Cabella Ligure (North -Western Italy)., Physics and chemistry of the earth (2002) 27 (2002): 1545–1550. doi_10.1016/S1474-7065(02)00175-4,
URL: http://www.cnr.it/prodotto/i/239255

Following the flood events occurred in the 1993 autumn, a landslide affecting the village of Montaldo di Cosola was reactivated. Montaldo di Cosola is a small village located in North-Western Italy, on the Appennines Mountains. Historical data report of at least another similar episode occurred about one century ago. The landslide takes place over debris deposits, which are about 20-35 m deep, produced by an ancient landslide event. Underneath these debris deposits there is an altered layer of the calcareous-marly substratum of the Antola formation. The landslide, almost certainly triggered by the incision of the torrent Rio Montaldo, has a probable rotational geometry and takes place through slow and progressive continuous movements. The landslide typology impedes a precise dimensional definition. Only the order of magnitude of the main scarp width, which is of hundreds of meters, and the order of magnitude of the landslide volume, which is of some hundreds of thousands of cubic meters, can be indicated. A complex monitoring network has been installed in the area. This network includes an automatic inclinometric system (AIS) that automatically performs measurements, twice a day, along the entire length of a pipe that is 45 m deep. Measurements are taken every 50 cm and are then broadcasted to a remote station located in Torino. After the exceptional meteorological events occurred in October 2000, the AIS measured signicant movements that have not yet stopped. On the basis of these available data it has been possible to locate the sliding surface at a depth of about 13 m at the level of a sandy-silty layer with some clay lenses inside. The movements measurements performed so far reveal a good correlation with the rainfalls. In particular it has been possible to identify a time lag of about 8-9 days between the occurrence of a rainfall peak and the corresponding peak in the recorded movements that were produced by these rainfalls. © 2002 Elsevier Science Ltd. All rights reserved.

Data from https://intranet.cnr.it/people/