An earthquake-source-based metric for seismic fragility analysis

The seismic fragility of a system is the probability that the system enters a damage state under seismic ground motions with specified characteristics. Plots of the seismic fragilities with respect to scalar ground motion intensity measures are called fragility curves. Recent studies show that fragility curves may not be satisfactory measures for structural seismic performance, since scalar intensity measures cannot comprehensively characterize site seismicity. The limitations of traditional seismic intensity measures, e.g., peak ground acceleration or pseudo-spectral acceleration, are shown and discussed in detail. A bivariate vector with coordinates moment magnitude m and source-to-site distance r is proposed as an alternative seismic intensity measure. Implicitly, fragility surfaces in the (m, r)-space could be used as graphical representations of seismic fragility. Unlike fragility curves, which are functions of scalar intensity measures, fragility surfaces are characterized by two earthquake-hazard parameters, (m, r). The calculation of fragility surfaces may be computationally expensive for complex systems. Thus, as solutions to this issue, a bi-variate log-normal parametric model and an efficient calculation method, based on stochastic-reduced-order models, for fragility surfaces are proposed.     

Investigation on regional attenuation of Vrancea (Romania) intermediate-depth earthquakes

This paper aims at investigating possible regional attenuation patterns in the case of Vrancea(Romania) intermediate-depth earthquakes.Almost 500 pairs of horizontal components recorded during 13 intermediate-depth Vrancea earthquakes are employed in order to evaluate the regional attenuation patterns.The recordings are grouped according to the azimuth with regard to the Vrancea seismic source and subsequently,Q models are computed for each azimuthal zone assuming similar geometrical spreading.Moreover,the local soil amplification which was disregarded in a previous analysis performed for Vrancea intermediate-depth earthquakes is now clearly evaluated.The results show minor differences between the four regions situated in front of the Carpathian Mountains and considerable differences in attenuation of seismic waves between the forearc and backarc regions(with regard to the Carpathian Mountains).Consequently,an average Q model of the type Q(f) = 115×f~(1.25) is obtained for the four forearc regions,while a separate Q model of the type Q(f) = 70×f~(0.90) is computed for the backarc region.These results highlight the need to evaluate the seismic hazard of Romania by using ground motion models which take into account the different attenuation between the forearc/backarc regions.     

Empirical relationships between macroseimic intensity and instrumental ground motion parameters for the intermediate-depth earthquakes of Vrancea region,Romania

Natural Hazards - Empirical relationships between MSK macroseismic intensity and several engineering ground motion parameters are developed using the macroseismic and instrumental data available...     

Using Artificial Neural Networks to Predict the Presence of Overpressured Zones in the Anadarko Basin,Oklahoma

Many sedimentary basins throughout the world exhibit areas with abnormal pore-fluid pressures (higher or lower than normal or hydrostatic pressure). Predicting pore pressure and other parameters (depth, extension, magnitude, etc.) in such areas are challenging tasks. The compressional acoustic (sonic) log (DT) is often used as a predictor because it responds to changes in porosity or compaction produced by abnormal pore-fluid pressures. Unfortunately, the sonic log is not commonly recorded in most oil and/or gas wells. We propose using an artificial neural network to synthesize sonic logs by identifying the mathematical dependency between DT and the commonly available logs, such as normalized gamma ray (GR) and deep resistivity logs (REID). The artificial neural network process can be divided into three steps: (1) Supervised training of the neural network; (2) confirmation and validation of the model by blind-testing the results in wells that contain both the predictor (GR, REID) and the target values (DT) used in the supervised training; and 3) applying the predictive model to all wells containing the required predictor data and verifying the accuracy of the synthetic DT data by comparing the back-predicted synthetic predictor curves (GRNN, REIDNN) to the recorded predictor curves used in training (GR, REID). Artificial neural networks offer significant advantages over traditional deterministic methods. They do not require a precise mathematical model equation that describes the dependency between the predictor values and the target values and, unlike linear regression techniques, neural network methods do not overpredict mean values and thereby preserve original data variability. One of their most important advantages is that their predictions can be validated and confirmed through back-prediction of the input data. This procedure was applied to predict the presence of overpressured zones in the Anadarko Basin, Oklahoma. The results are promising and encouraging.     

The effect of dispersion on the stability of density-driven flows in saturated homogeneous porous media

This work is a continuation of Musuuza et al. [37] in which a stability criterion for density-driven flow in a saturated homogeneous medium was derived. The criterion predicted the stability of a system as a function of the density and viscosity contrasts, the magnitude of the flow velocity and the concentration gradients for flow aligned orthogonal to gravity. It could not accurately predict stability transition with increasing velocity, a failure we attributed to dispersion effects that were not included. Small-scale dispersion and molecular diffusion, the main stabilising mechanisms in homogeneous media can stabilise flow if the instability wavelengths are below a certain cutoff. The width of the mixing zone is also central in controlling the range of wavelengths that persist into fingers. We propose a method of quantifying the cutoff wavelength and the width of the mixing zone, which are incorporated into the earlier criterion as constituents of the dispersive part. The earlier criterion is reformulated in terms of the Rayleigh number and with the dispersive part added, we attempt to predict the number of fingers formed which is directly related to the physical stability of the system. The inclusion of the Rayleigh number and dispersion into a single stability criterion provides new insight in the way dispersion affects vertical flow systems. Stability numbers computed with the new criterion are in reasonable agreement with numerical simulations for a range of physical variables. The numerical computations are performed with the software package d³f, which uses the cell-centred finite volume and the implicit Euler methods for the spatial and temporal discretisations, respectively. The admission of the density and dispersivities as inputs into the criterion makes it usable in practical problems.     

The stability of density-driven flows in saturated heterogeneous porous media

This work concludes the investigations into the stability of haline flows in saturated porous media. In the first part [33] a stability criterion for density-driven flow in a saturated homogeneous medium was derived excluding dispersion. In the second part [34], the effects of dispersion were included. The latter criterion made reasonable predictions of the stability regimes (indicated by the number of fingers present) as a function of density and dispersivity variations. We found out that destabilising variables caused an increase in the number of fingers and vice versa. The investigation is extended here for the effects of the medium heterogeneity. The cell problem derived via homogenization theory [20] is solved and its solution used to evaluate the elements of the macrodispersion tensor as functions of time for flow aligned parallel to gravity. The longitudinal coefficient exhibits asymptotic behaviour for favourable and moderately unfavourable density contrasts while it grows indefinitely for higher density contrasts. The range of densities stabilised by medium heterogeneities can thus be estimated from the behaviour of the coefficient. The d³f software program is used for the numerical simulations. The code uses the cell-centred finite volume and the implicit Euler techniques for the spatial and temporal discretisations respectively.     

Mixed finite element discretization and Newton iteration for a reactive contaminant transport model with nonequilibrium sorption: convergence analysis and error estimates

We present a numerical scheme for reactive contaminant transport with nonequilibrium sorption in porous media. The mass conservative scheme is based on Euler implicit, mixed finite elements, and Newton method. We consider the case of a Freundlich-type sorption. In this case, the sorption isotherm is not Lipschitz but just Hölder continuous. To deal with this, we perform a regularization step. The convergence of the scheme is analyzed. An explicit order of convergence depending only on the regularization parameter, the time step, and the mesh size is derived. We give also a sufficient condition for the quadratic convergence of the Newton method. Finally, relevant numerical results are presented.     

Assessment of seismic risk scenarios for Bucharest,Romania

In this paper, seismic risk scenarios for Bucharest, the capital city of Romania, are proposed and assessed. Bucharest has one of the highest seismic risk levels in Europe, and this is due to a combination of relatively high seismic hazard and a building stock built mainly before the devastating Vrancea 1977 earthquake. In this study, the seismic risk of Bucharest is assessed using the most recent information regarding the characteristics of the residential building stock. The ground motion amplitudes are evaluated starting from random fields obtained by coupling a ground motion model derived for the Vrancea intermediate-depth seismic source with a spatial correlation model. The seismic risk evaluation method applied in this study is based on the well-known macroseismic method. For several structural typologies, the vulnerability parameters are evaluated based on a damage survey performed on 18,000 buildings in Bucharest after the March 1977 earthquake. Subsequently, the risk metrics are compared with those from other studies in the literature that apply a different risk assessment methodology in order to gain a better view of the uncertainties associated with a seismic risk study at city level. Finally, the impact of several Vrancea intermediate-depth earthquake scenarios is evaluated and the results show that the earthquake which has the closest epicenter to Bucharest appears to be the most damaging.     

Mechanisms to limit higher mode effects in a controlled rocking steel frame. 1: Concept,modelling, and low‐amplitude shake table testing

Controlled rocking steel frames have been proposed as an efficient way to avoid the structural damage and residual deformations that are expected in conventional seismic force resisting systems. Although the base rocking response is intended to limit the force demands, higher mode effects can amplify member design forces, reducing the viability of the system. This paper suggests that seismic forces may be limited more effectively by providing multiple force‐limiting mechanisms. Two techniques are proposed: detailing one or more rocking joints above the base rocking joint and providing a self‐centring energy dissipative (SCED) brace at one or more levels. These concepts are applied to the design of an eight‐storey prototype structure and a shake table model at 30% scale. A simple numerical model that was used as a design tool is in good agreement with frequency characterization and low‐amplitude seismic tests of the shake table model, particularly when multiple force‐limiting mechanisms are active. These results suggest that the proposed mechanisms can enable better capacity design by reducing the variability of peak seismic force demands without causing excessive displacements. Similar results are expected for other systems that rely on a single location of concentrated nonlinearity to limit peak seismic loads. Copyright © 2012 John Wiley & Sons, Ltd.