Comparative assessment of nonlinear static and dynamic methods for analysing building response under sequential earthquake and tsunami

This paper presents a comprehensive comparison of different dynamic and static approaches for assessing building performance under sequential earthquakes and tsunami. A 10-storey reinforced concrete seismically designed Japanese vertical evacuation structure is adopted as a case study for the investigation. The case study building is first assessed under sequential earthquake and tsunami nonlinear response history analyses: the first time this is done in the literature. The resulting engineering demand parameters are then compared with those obtained when the analysis procedure is systematically simplified by substituting different static approaches for the nonlinear response history analyses in both the earthquake and tsunami loading phases. Different unloading approaches are also tested for the cases when an earthquake pushover is adopted. The results show that an earthquake nonlinear response history analysis, followed by a transient free vibration and a tsunami variable depth pushover, provides the best alternative to full dynamic analyses in terms of accuracy and computational efficiency. This structural analysis combination is recommended and has the advantage that it does not require the tsunami inundation time history to be known in advance. The proposed double pushover approach is instead deemed only suitable for the collapse assessment of regular low to mid-rise buildings and for the development of collapse fragility functions. An important observation made is that sustained earthquake damage seems not to affect the tsunami resistance of the case study building when the fully dynamic analysis is carried out for the sequential loading. This observation will be the subject of future work.     

Carbonate sedimentation and hydrodynamic pattern on a modern temperate shelf: The strait of Bonifacio (western Mediterranean)

The sedimentary features of the inner-middle shelf of the strait of Bonifacio (western Mediterranean) were analyzed to evaluate the relationship between the production and transport of biogenic carbonate sediments and the basin morphology and hydrodynamics. A three-dimensional hydrodynamic modeling was performed in order to simulate the influence of waves and currents at seabed level. Superficial sediments were collected at depths ranging from 5 to 80 m and were analyzed for grain size, mineralogical composition and skeletal carbonate composition. Posidonia oceanica seagrass meadows border the coasts in a narrow strip on both sides of the strait down to a depth of 40 m. At greater depths, the seabed is characterized by the presence of plateaus and ridges which are controlled by outcropping bedrock morphology.     

Physical modelling of tsunami using a new pneumatic wave generator

The physical simulation of tsunami in the laboratory has taken a major leap forward with the construction and testing of a new wave generator, capable of recreating scaled tsunami waves. Numerical tools fail to reproduce tsunami nearshore and onshore processes well, and physical experiments in large scale hydraulic facilities worldwide have been limited to the generation of solitary waves as an (controversial) approximation for evolved forms of tsunami. The new concept in wave generation presented herein is born of collaboration between UCL's Earthquake and People Interaction Centre (EPICentre) and HR Wallingford. It allows for the first time the stable simulation of extremely long waves led either by a crest or a trough (depressed wave). This paper presents the working concepts behind the new wave generator and the first stages of testing for verifying its capacities and limitations. It is shown that the new wave generator can not only reproduce solitary waves and N-waves with large wavelengths, but also the 2004 Indian Ocean Tsunami as recorded off the coast of Thailand (“Mercator” trace).     

Physically based dynamic run-out modelling for quantitative debris flow risk assessment: a case study in Tresenda,northern Italy

Quantitative landslide risk assessment requires information about the temporal, spatial and intensity probability of hazardous processes both regarding their initiation as well as their run-out. This is followed by an estimation of the physical consequences inflicted by the hazard, preferentially quantified in monetary values. For that purpose, deterministic hazard modelling has to be coupled with information about the value of the elements at risk and their vulnerability. Dynamic run-out models for debris flows are able to determine physical outputs (extension, depths, velocities, impact pressures) and to determine the zones where the elements at risk can suffer an impact. These results can then be applied for vulnerability and risk calculations. Debris flow risk has been assessed in the area of Tresenda in the Valtellina Valley (Lombardy Region, northern Italy). Three quantitative hazard scenarios for different return periods were prepared using available rainfall and geotechnical data. The numerical model FLO-2D was applied for the simulation of the debris flow propagation. The modelled hazard scenarios were consequently overlaid with the elements at risk, represented as building footprints. The expected physical damage to the buildings was estimated using vulnerability functions based on flow depth and impact pressure. A qualitative correlation between physical vulnerability and human losses was also proposed. To assess the uncertainties inherent in the analysis, six risk curves were obtained based on the maximum, average and minimum values and direct economic losses to the buildings were estimated, in the range of 0.25–7.7 million €, depending on the hazard scenario and vulnerability curve used.     

Compressional behaviour of CaNiSi2O6 clinopyroxene: bulk modulus systematic and cation type in clinopyroxenes

A single-crystal of composition CaNiSi₂O₆ (space group C2/c) was investigated at high pressure up to about 7.8 GPa by X-ray diffraction. The unit-cell parameters were measured at 18 different pressures. The P-V data were fitted by a third-order Birch-Murnaghan equation of state V₀=435.21(1) ų, K ₀=117.6(3) GPa and K^=6.4(1). The linear axial compressibilities _a, _b, _c and a sin are 2.14(1), 3.00(1), 2.43(1) and 1.63(1) × 10–3 GPa^–1. Comparing the compressibility data with other CaM1Si₂O₆ pyroxenes we suggest that the empirical K × V = constant relationships are followed in C2/c pyroxenes only if the same valence electron character is shared.