Table 4 Summary of the main results of the seismic noise analyses carried out on Malta Island for both site effects and landslide characterisation

Galea et al. ²³

Anchor Bay, northwestern part of Malta Island

HVSR

The H/V curves show peaks caused by both the underlying geology (i.e., those at 1–2 Hz linked to the UCL overlaying the BC) and the mechanical vibration modes of partly or wholly detached blocks (i.e., those at 4–5 Hz or at 9–10 Hz).

The polarisation analysis indicates that the 1.5 Hz peak is not associated with any particular polarisation direction (confirming the lithological origin of the random-direction sources of noise), while peaks between 2.0 Hz and 10.0 Hz show strong directivity and linearity at sharply defined frequencies (confirming their origin linked to mechanical vibrational modes of blocks).

The results of polarization analysis do not coincide with the results obtained from GPS monitoring (Mantovani et al.⁵³), but polarization analyses do not represent long-term mass movements but rather high frequency modal vibrations, which reflect the degrees of freedom of individual blocks. If blocks are only partly detached from the mainland, they are limited in their degrees of freedom and the dominant direction of polarization (at frequencies up to 10.0 Hz) is generally normal to the cliff edge, or to the large scale fractures. If blocks are fully detached, the polarization directions indicate the existence of more degrees of freedom.

Finally, the amplitude of H/V peaks can give information on the state of the detachment: large amplitudes are associated to strong horizontal translational motion present in a state of early detachment, while smaller, or no H/V peaks, but clearly polarized and linear particle motion at well-defined frequencies, or frequency bands indicate vertical sliding.

Iannucci et al.²⁴

Selmun Promontory, northeastern part of Malta Island

HVSR

The H/V data acquired on the UCL plateau show a peak in the narrow range of 1.5–2.0 Hz, followed by a sharp dip of the spectral ratio, considered representative of the UCL–BC–GL sedimentary sequence.

Data collected on the stable plateau present only one peak that does not show evidence of polarisation. On the other hand, data acquired within and in proximity of the unstable zones have additional peaks in the range 3.0–60.0 Hz that show a marked linearity and strong polarisation in the same dip direction of the joint system.

The first resonance frequency of a selected block, calculated using the equation proposed by ref. ²², is in agreement with the experimental data. Unfortunately, this result is not reliable because the authors use wrong input values (i.e., they use a Vp = 800 m s^-1, but indeed this value is for shear waves, as in ref. ⁴⁰ that the authors themselves use as reference).

Iannucci et al.²⁵

Selmun Promontory, northeastern part of Malta Island

HVSR, ESAC, continuous seismic noise monitoring, Seismic

Navigating System (SNS) array

H/V results from²⁴.

From the joint inversion of ESAC and H/V data, the Vs values of UCLand BC are obtained and are in agreement with previous studies.

SNS array, with an aperture smaller than the conventional one, allowed to detect and locate 20 natural microseismic events.

Long-term monitoring confirms that the polarised frequency at 3.3–3.5 Hz is related to the vibrational behaviour of the rock block, as it is characterised by a prevalent sub-horizontal oscillation.

Panzera et al.³⁷

Xemxija Bay, northeastern part of Malta Island

HVSR

The H/V curves analysis highlights the existence of three different zones: (1) a stable zone, in which the H/Vs show only a dominant peak at about 1.5 Hz linked to the interface between BC terrains and GL (Globigerina Limestone) rock masses; (2) a zone, close to the cliff area, characterized by the presence of peaks at a frequency >9 Hz linked to the presence of fractures and of blocks almost detached from the cliff and therefore free to oscillate; (3) a zone, on the landslide body, which shows both bimodal peaks at 1–3 Hz (that can be associated with the contact between the rockfall and detritus unit and the BC formation, as well as with the interface between BC and GL formations) and peaks >9 Hz (associated with detached blocks almost free to oscillate).

The polarisation analysis makes evident that the prevailing angle of directional site effects, especially in the fractured and in the landslide zone, appears constant in the NE direction in the range 1–10 Hz, whereas directional effects become more randomly distributed at frequencies greater than 10 Hz.

Panzera et al.³⁸

Whole Malta Island (16 sites)

HVSR, MASW

Surface waves analysis allowed for obtaining the Vs30 features of the main outcropping lithotypes.

Seismic noise analysis highlighted that apart from GL formation, all the other formations show important spectral ratio peaks and the UCL sites are characterised by peaks in the frequency range 1.0–3.0 Hz (related to the presence of a BC layer between the UCL and GL formations).

Vella et al.³⁹

Whole Malta Island (202 sites)

HVSR

Seismic noise analysis shows that UCL sites (which overlay the BC formations) have a H/V peak in the range 1–2 Hz.

Farrugia et al.⁴⁰

Whole Malta Island (7 sites)

HVSR, ESAC

The H/V data acquired at the site where UCL directly overlays GL is flat, i.e., does not show the 1–2 Hz peak visible in all the sites where the BC are present below the UCL.

From the joint inversion of ESAC and H/V data, the Vs values of the different lithologies are obtained.

Pischiutta et al.⁴⁹

Rabat (central-western part of Malta) and Golden Bay (northwestern part of Malta)

HVSR, active and passive seismic arrays

The H/V data show an isotropic, i.e., not polarised, peak in the range 1–2 Hz that is caused by the impedance contrast between the GL unit and the other two overlaying units (i.e., UCL and BC).