Abstract This study reports a comparative analysis of the environmental conditions and micro-seismicity recorded on a rock wall resulting from an intense meteorological event. The findings are focused on a quarry wall located in the Acuto Field Laboratory (Central Italy), where multi-parametric environmental monitoring is operating and an Artificial intelligence Camera Prototype has been placed to detect rock falls reaching a railway target. Six accelerometers were installed to detect micro-seismic events caused by the expected strong thermal transient caused by the Buran storm on February 2018. Within a few hours, a steep decrease in the average air and rock mass temperature down to 8 °C was recorded, and −4 °C and −8 °C were reached for the rock and air temperatures, respectively. A total of 103 micro-seismic events were analysed with respect to both rainfall and thermal forcing: while no correlation with rainfall was reported, the steep thermal transient was responsible for the strain effect that occurred during the heating phase of the rock mass following the Buran storm. An elastic deformation event with a maximum daily amplitude of 165 μ strain was recorded by the strain gages installed on the mm-joints due to the rock heating and cooling caused by the variation in temperature. The collected evidences show the relevance of short thermal transients in modifying stress conditions within rock masses and their relationship to a peculiar micro-seismic response. The main outcomes established the key role played by integrated monitoring systems to better understand the relationship between vibrational behaviour and environmental forcings in terms of understanding the precursors to rock failure.
Seismometer arrays have been widely applied to record collapse by controlled explosion in mines and caves. However, most underground failures are natural events, and because they can occur abruptly, underground failures represent a serious geological hazard. An accelerometric array installed on 4 September 2008 has been used to manage the geological risk of the Peschiera Springs drainage plant of Rome's aqueduct, which is located in the Central Apennines approximately 80 km from Rome, Italy. The plant occupies a karstified carbonatic slope that is extensively involved in gravitational deformations, which are responsible for underground failures such as cracks and collapses. To distinguish among different types of recorded events, an automated procedure was implemented based on the duration, peak of ground acceleration (PGA) and PGA variation in the recordings of the plant's accelerometric stations. The frequencies of earthquakes and micro-earthquakes due to underground failures are, in general, well correlated. Nevertheless, many underground failure sequences can be directly associated with the continuous deformations that affect the slope. The cumulative Arias intensity trend derived for the underground failures combined with the failure and earthquake frequencies enabled the definition of a control index (CI) that identifies alarming or emergency conditions. The CI can be used as a tool for managing the geological risk associated with the deformational processes that affect the drainage plant.
An integrated monitoring system is operative in the Peschiera Springs slope (Central Apennines, Italy) to manage the landslide risk related to the plant of Rome aqueducts. Since 2008, an accelerometric network has been operating in order to integrate the stress-strain monitoring system. Nowadays about 1300 microseismic signals due to instabilities have been recorded; these events can be distinguished in failures and collapses. Whereas the failures are related to the rock mass deformation, the collapses are mainly associated with the aquifer discharge changes (about 16–21 m3/s). A Control Index (CI), based on the frequency of occurrence and the cumulative energy of the recorded local instabilities was tested for providing three levels of alert. In 2014, a nanoseismic array (Seismic Navigation System) was installed inside the drainage plant that is contributing to identify sequences of microseismic pre-failure events, allowing to assess the related landslide hazard.