Slope response to effective rainfall of a large, complex rock-slide in flysch material

Abstract

The Camugnano landslide, located in the Province of Bologna, affects an area of 9.1×105 m2. It is considered a complex phenomenon involving both roto-translational kinematics within the Camugnano Formation’s arenaceous flysch and translational slides that affect the degraded products of the flysch and other clay materials in the lower sector of the unstable slope. The activity of the landslide is characterized by very slow displacement rates (4 cm/year), with occasional acceleration events in the past. The most recent significant event occurred in 2014. In response, various surveys and monitoring activities were conducted, including the use of boreholes, geophysical surveys, displacement measurements through inclinometers, Ground Navigation Satellite System (GNSS), and Robotic Total Station (RTS), and groundwater monitoring through pressure transducers installed in piezometers. The aim of this note is to summarize the monitoring results and compare them with the trend of effective rainfall over multiple years. To determine effective precipitation, the Thornthwaite formulation was used, based on mean daily temperature and daily cumulative rainfall data from the Diga del Brasimone gauge station, located approximately 5 km from the landslide site. The inclinometer monitoring helped identify the sliding surfaces and obtain time series data by integrating displacements on these surfaces. GNSS campaigns conducted periodically revealed displacement rates ranging from 30 to 70 mm/year, generally decreasing progressively from 2014 to 2022. RTS monitoring indicated slow but detectable movements within the landslide. Then comparing GNSS displacements with precipitation data, it was found that displacement rates remained relatively constant regardless of variations in effective precipitation. This trend was also observed when analysing inclinometer data. The groundwater monitoring showed seasonal fluctuations, with peak levels occurring during the winter period in most piezometers. However, no long-term groundwater trend was observed. The displacement record highlights a variable sensitivity of the slope to effective rainfall with respect to the last major reactivation event. Temporary, localized accelerations are in fact recorded after 2014, gradually decreasing in magnitude in the next years. These findings suggest that mitigation strategies could by modified according to this style of activity and indicate that assumptions linking rainy or dry periods to variations in movement acceleration or deceleration may be overly simplistic.