**3. Materials and Methods**

The study was based on the geological and geomorphological analysis of the Coroglio-Trentaremi sea cliff by combining field work with analysis of detailed scale topographic maps and orthophotos, and stratigraphical data from deep boreholes.

Geomorphological analysis has been carried out by means of a detailed scale topographic map (Technical Map of the Regione Campania at scale 1:5000), Google Earth images and field work.

This analysis allowed us to reconstruct the main geomorphological features of the sea cliff, and to map the main landform indicators of slope instability. We gave particular attention to detachment niches and landslide bodies and to the presence of notches and cavities at the base of the sea cliff. The latter are indicators of undercutting by waves, which is undoubtedly the most important factor in causing coastal retreat [4,58]. The rate of undercutting is controlled by the complex and wide-ranging behavior of geological materials and by the great variability in their geotechnical properties. According to Budetta [24], the increasing depth of the notch into fractured rock mass, because of wave erosion, is responsible for the spreading of the shear stresses towards the top of the cliff and for the destabilization of the overlying slope. We also considered the presence of ancient quarries (mainly roman in age) for tuff extraction which, accordingly to Ruberti et al. [27], may also be considered as instability factors, especially if involved in the shear stress propagation caused by waves undercutting.

Geological analysis was conducted by means of field work, both inland and by the sea using small boats that allowed to reach some areas not accessible by foot (e.g., most of the sea cliff base), and borehole data [59]. The combination of field and borehole data allowed us to derive 6 geological cross-sections representative of the subsurface stratigraphical setting of the Coroglio-Trentaremi sea cliff. Furthermore, field work also allowed us to define the structural setting of some portions of the sea cliff by establishing 7 structural stations. Structural stations have been established directly on the cliff by climber geologists along 10 vertical rope descents [59]. These structural data have been combined with field data collected at the base of the sea cliffs. As a result, we have been able to establish the relationships between bedding, sea cliff trend, and fractures by defining dip and dip direction and to propose failure mechanisms for each structural station.

By comparing geological and geomorphological collected data, we used a heuristic approach to identify and select two main groups of predisposing factors which are listed in Table 1. The first group includes factors that may influence landslide intensity (e.g., sea cliff height, I1; volume of landslide body, I2; volume of detachment niches, I3; volume of blocks or projecting sectors, I4), and factors that may increase the weakness of a sea cliff (e.g., fracture spacing, W1; persistence of beating fractures, W2; volume of caves at the sea cliff base, W3; distance between the shoreline and the base of active sea cliff, W4). The volume of the caves has been derived by expeditive surveys with a laser pointer. The distance

between the present shoreline and the base off the cliff considers the presence/absence of a debris body that may act as protection from the wave action. Each factor includes four group values, which have been weighted according to their attitude to instability.

**Table 1.** Instability (I) and weakness (W) factors adopted to evaluate the geomorphological zonation to landslide along a sea cliff, with relative weight.


To evaluate the susceptibility to landslide of a sea cliff, we have crossed the weights of the Intensity (I) and the Weakness (W) factors (Table 2). This matrix suggests that high magnitude (M) landslides may occur when high weight I and W factors occur. As an example, either a very high sea cliff or a sea cliff with a high volume of blocks or projecting sectors coupled with large caves at the sea cliff base may contribute to high magnitude landslides (M3 class in Table 2). On the other hand, a low height sea cliff carved in poorly fractured rocks and with no caves at the sea cliff base may contribute to low magnitude landslides (M1 class in Table 2).

**Table 2.** Matrix for the evaluation of landslide magnitude (M) by crossing weights of both the Intensity (I) and Weakness (W) factors. Red, orange and yellow colors indicate high (M3), moderate (M2) and low (M1) magnitude classes.

