Search

The Victoria Lines Fault (island of Malta) is a >15 km-long and N260°-striking segmented normal fault-system, which is probably inactive since the late Pliocene. In the westernmost part, the Fomm Ir-Rih segment displays comparable geologic throw and escarpment height (~150–170 m), moreover its hangingwall hosts thin patches of Middle Pleistocene clastic continental deposits (red beds), which are poorly preserved elsewhere. We acquired two seismic transects, by collecting ambient vibration recordings, processed by using horizontal-to-vertical spectral ratios, complemented by one high-resolution 2-D refraction tomography survey crossing this fault where it is locally covered by red beds and recent colluvial deposits. We found a resonance peak at ~1.0 Hz in the hangingwall block, whereas clear peaks in the range ~5.0–10.0 Hz appear when approaching the subsurface fault, and we relate them to the fractured bedrock within the fault zone. The best-fit tomographic model shows a relatively high-Vp shallow body (Vp 2200–2400 m/s) that we relate to the weathered top of the Miocene Upper Coralline Limestone Fm., bounded on both sides by low-Vp regions ( 230 m/s above the weathered top-bedrock. Our results depict a clear seismic signature of the Victoria Lines Fault, characterized by low seismic velocity and high amplification of ground motion. We hypothesize that, during the Middle Pleistocene, faulting may have affected the basal part of the red beds, so that this part of the investigated complex fault-system may be considered inactive since ~0.6 Myr ago.

Seismic site response analysis is an important procedure used to obtain parameters such as peak ground motion values. These are needed by the engineering community for reliable seismic design, analysis and retrofitting of structures. One of the inputs for such analysis is the shear-wave velocity profile ( V S ) of the site. Surface wave methods are increasingly being used to obtain these profiles. However, these are characterised by uncertainties related to the measurements and the model used for interpretation (e.g. the inadequacy of 1D assumption). Moreover, the inversion procedure is also highly non-unique because numerous profiles are able to fit the experimental dispersion curve/s. Even though the uncertainties are not commonly included in site response analysis, their propagation from the dispersion curves to the inversion procedure can lead to significant differences in the site response analysis results. In a previous study, V S profiles were obtained at 20 sites on the Maltese islands (Central Mediterranean) using the Horizontal-to-Vertical Spectral Ratio (H/V), the Extended Spatial Auto-Correlation (ESAC) technique and the Genetic Algorithm. The geology of all the sites is characterised by a layer of clay, which can be up to 75 m thick, buried under the Upper Coralline Limestone, a fossiliferous coarse grained limestone. The effect of a thick buried low V S layer, which gives rise to a velocity inversion, in ground response is not well understood and/or documented. The above concerns raised the need for carrying out tests with the aim of better clarifying and understanding which parameters have the most influence on site response analysis and to assess the consequences of inversion non-uniqueness on ground motion amplification and response. From the first part of the study, it was observed that the thickness of the clay layer has the greatest effect on the ground response results. By obtaining ground motion parameters from equivalent profiles, the consequences of the uncertainty in the inversion process, which can be serious at times, are revealed.

We performed geophysical investigations in the northwestern sector of the island of Malta to reconstruct velocity-depth models and provide shear-wave velocity profiles. We have chosen two sites, one located in Rabat (Malta) and another in the Golden Bay area. We used both active (seismic and electrical 2D-tomography, Multichanel Analysis of Surface Waves – MASW) and passive (2D arrays and single-station measurements using ambient noise) geophysical methods. Consistently with previous studies performed in this part of Malta, we have found that both sites are characterized by site resonance in the frequency range 1–2 Hz as an effect of the local lithostratigraphic succession that shows an impedance contrast at about 60–90 m depth. This resonance effect can have important implications on both seismic hazard as well as seismic risk evaluation of the region since the amplified frequency range coincides with the resonance frequencies typical of 5–10 storey buildings which are very diffuse in the Maltese Islands, especially after intense recent urbanization. We also highlight the importance of performing seismic velocity measurements even at rock sites. As an example, the Golden Bay site would be classified as class “A” following the EuroCode EC8 when considering only the outcropping lithology represented by limestone rocks. Conversely the subsoil characterization provided by this study has revealed that this site falls in the EC8 class “B”, stressing the importance of direct geophysical measurements since the a-priori assignment to A-class could lead to wrong estimates in evaluating the site response.

The Maltese islands in the Central Mediterranean are composed of a simple 4-layer sedimentary sequence of Oligocene–Miocene limestones and clays, highly disturbed by synsedimentary to recent tectonic activity, so that the surface outcrops present a variety of lithologies, as well as complex geomorphology. The seismic history of the islands includes a number of damaging earthquakes, the largest intensity since 1500 being EMS-98 VII–VIII in 1693. In this study a first seismic site response survey was undertaken over the whole archipelago, by recording ambient noise and applying the H/V (Horizontal-to-Vertical Spectral Ratio) technique. The results show a clear correlation between site response and outcrop geology. In general, where the two oldest layers, the Lower Coralline Limestone and the Globigerina Limestone outcrop, the H/V is flat above 0.5 Hz. A thick layer of clays and marls (Blue Clay) deposited originally on top of the Globigerina Limestone is preserved in the western half of the archipelago, and this strongly influences the site response. Where the clay outcrops at the surface, it produces the highest amplitudes, and peak frequencies ranging between 2 and 5 Hz, whereas all sites on the hard Upper Coralline Limestone, which overlies the clay in many regions, still exhibit H/V peaks at a consistent frequency between 1 and 2 Hz, which we attribute to the presence of the underlying clay. Results were validated by a second measurement survey and numerical modelling. The site effect induced by the clay layer has implications for the assessment of seismic risk, as well as for the validity of using VS30 as a proxy for site class and site amplification.