B. Benito

AN APPROACH TO THE MEASUREMENT OF THE POTENTIAL STRUCTURAL DAMAGE OF EARTHQUAKE GROUND MOTIONS

SUMMARY The quantification and prediction of damage due to di⁄erent seismic actions to structure types of di⁄erent strength is an important problem not yet solved in the Earthquake Engineering field. In addition, owing to the fact that macroseismic information cannot be used directly in dynamic calculations, a new problem appears when these are the only kind of data available. Thus, there is a need to estimate a parameter to relate the energy of the ground motion and the damage occurrence, and eventually achieve a better seismic risk assessment. After the study and review of some representative potential damage parameters, attention has been paid to the Arias intensity (unfiltered and filtered in certain frequency ranges) and the Cumulative Absolute velocity (CAV) as the parameters to evaluate the energy of movement, and to relate them with the observed damage. The data used to infer these correlations have been provided by the ENEA-ENEL (Italy). The information consists of strong motion records from the Campano Lucano (1980), Umbria (1984) and Lazio-Abruzzo (1984) earthquakes, and data of damage to buildings in the vicinity of recording instruments (within a maximum radius of 300 m, where the soil conditions remain constant). In this paper, some relations have been obtained to quantify the damage level for di⁄erent seismic inputs. The results suggest that unfiltered Arias intensity and CAV (for calculation threshold 20 cm/s2) correlate well with the macroseismic information used. Best fits are obtained between the quoted parameters and the observed damage in type A structures.

Triggering of destructive earthquakes in El Salvador

We investigate the existence of a mechanism of static stress triggering driven by the interaction of normal faults in the Middle American subduction zone and strike-slip faults in the El Salvador volcanic arc. The local geology points to a large strike-slip fault zone, the El Salvador fault zone, as the source of several destructive earthquakes in El Salvador along the volcanic arc. We modeled the Coulomb failure stress (CFS) change produced by the June 1982 and January 2001 subduction events on planes parallel to the El Salvador fault zone. The results have broad implications for future risk management in the region, as they suggest a causative relationship between the position of the normal-slip events in the subduction zone and the strike-slip events in the volcanic arc. After the February 2001 event, an important area of the El Salvador fault zone was loaded with a positive change in Coulomb failure stress (.0.15 MPa). This scenario must be considered in the seismic hazard assessment studies that will be carried out in this area.

An Overview of the Damaging and Low Magnitude Mw 4.8 La Paca Earthquake on 29 January 2005: Context, Seismotectonics, and Seismic Risk Implications for Southeast Spain

This article presents an overview of the La Paca earthquake of magnitude mbLg 4.7, which occurred on 29 January 2005, with its epicenter located near the town of Avile´s in the Murcia region in southeast Spain. Despite its low magnitude, the earthquake caused important damage in two towns of the epicentral area, La Paca and Zarcilla de Ramos. These areas recorded intensities of VI–VII (European Macroseismic Scale, 1998) and sustained estimated economic losses amounting to 10 million €. Aftershocks continued for more than 2 weeks, producing considerable alarm in the population and mobilizing emergency services from the whole region. The La Paca seismic series is the third registered in the region in the past 8 years, being preceded by the Mula (1999) and southwest Bullas (2002) seismic series. These main events had also low magnitudes (mbLg 4.8) and caused damage levels similar to the 2005 earthquake. The case is an example of a moderate seismic zone where low-magnitude and frequent earthquakes have important implications on the seismic hazard and risk of the region. Although these are not the largest expected earthquakes, they have yielded important information for improving the knowledge of the seismic activity of the area. With this aim in mind, different topics have been analyzed from a multidisciplinary perspective, including seismicity, local tectonics and surface geology, focal mechanisms, macroseismic effects, and ground motion. Results indicate a local tectonic interpretation, consistent with a strike-slip focal mechanism, the confirmation of a triggering process between the 2002 and 2005 earthquakes, a geotechnical and ground-motion characterization for the damaged sites (supporting local amplification effects and estimated peak ground acceleration values of �0.1g), and an understanding of damage patterns in relation to local building trends. The results may be used as guidelines for future revisions of the Spanish Building Code (Norma de la Construccio´n Sismorresistente Espan˜ola [NCSE-02], 2002). The study results should contribute to risk mitigation in a region where strong-motion records from the maximum expected earthquakes are not available. This approach can be extended to other regions with similar seismic backgrounds and a lack of strong-motion records.

Relationships Between M w and Other Earthquake Size Parameters in the Spanish IGN Seismic Catalog

An important part of the preparation and homogenization of the seismic catalog used in the recent seismic hazard assessment study of Spain is the development of relationships among the different scales used to measure earthquake size. The objective is to convert all earthquake size data in the original catalog to a single magnitude scale, the moment magnitude Mw, in order to have a set of events with an uniform comparable size measurement. These new relationships are based on regression analysis between Mw and the other units used in the catalog in different epochs. The reduced major axis regression scheme is used, because it is the most suitable method for symmetric treatment of the variables involved in the fits. The new relationships obtained for Spain, Mw as a function of mbLg; Mw as a function of mb; and Mw as a function of Imax, are presented and their applicability limits and accuracy are discussed. The results obtained could have other practical uses in regional seismicity analysis.

Analysis of the spatial and temporal distribution of the 2001 earthquakes in El Salvador

This paper presents a study of the spatial and temporal distribution of the large destructive earthquakes that occurred in El Salvador during January and February 2001, together with the static stress transfer after each main shock, associated with their respective rupture processes. The sequence began with the magnitude M W 7.7 earthquake of 13 January, located off the western Pacifi c Coast in the subduction zone between the Cocos and Caribbean plates. One month later, a second destructive earthquake of M W 6.6 occurred in the Caribbean plate farther inland, the epicenter of which was located near San Pedro Nonualco. This shock was linked to the local faults beneath the volcanic arc and also produced signifi cant damage. The two main shocks and their aftershock sequences, together with other minor events that followed successively, produced unusually intense activity in the zone, in a short interval of time. The aims of this study are to document the spatial and temporal evolution of each seismic sequence and also to understand the possible interaction between the different events. We have inferred that some events with M > 5 triggered other shocks with the same or different origin (subduction zone or local crustal faults). The Coulomb stress transfer has been studied, and some models developed, using the rupture parameters derived from the geometric distribution of aftershocks. These results suggest the existence of a dynamic interaction, since the 13 February event occurred in a zone where the Coulomb stress increased following the January 13 event. Subsequently, some further events with magnitude around M W 5 in turn were located in other zones of increased stress associated with the two previous large earthquakes.

Ground-Motion Characterization of Low-to-Moderate Seismicity Zones and Implications for Seismic Design: Lessons from Recent Mw ∼4.8 Damaging Earthquakes in Southeast Spain

We address the ground-motion characterization of three damaging earthquakes that occurred in the low-to-moderate seismicity region of Murcia (southeast Spain) and compare our results with current earthquake-resistant provisions to establish whether those provisions are consistent with the available data or need to be revised. The analyzed series are the 1999 Mula ( m b Lg 4.8, I EMS = VI), the 2002 southwest Bullas ( m b Lg 4.8, I EMS = VI), and the 2005 La Paca ( m b Lg 4.7, I EMS = VII). Recorded peak ground accelerations are low (pga 20 km). We find that local factors control the distribution of ground motions, possibly in combination with propagation effects. Characteristic spectral shapes, representative of ground motions at each station, are observed. Several ground-motion predictive models and a simulation method are used for estimating accelerations and response spectra in the most damaged towns, where no records exist. The different methods consistently predict similar response spectra for the epicentral areas. Recorded and predicted normalized spectral shapes exceed the design spectral shape of the Spanish Building Code ncse-02 for all soil categories for intermediate– high frequencies (above about 3 Hz). Moreover, the ncse-02 absolute response spectra are possibly exceeded by the corresponding predicted spectra for the epicentral areas in the same frequency range. Predominant frequencies of common soils and critical periods of most conventional buildings in the Region of Murcia also lie in this frequency range. These factors could explain, in part, the observed damage. The results raise questions about the definition of the ncse-02 design spectra, and specifically about the design spectral shape used. The approach followed in this work may be used to calibrate other national seismic codes, especially of regions with similar characteristics: moderate seismicity and limited availability of ground-motion data.

Object-based urban structure type pattern recognition from Landsat TM with a Support Vector Machine

ABSTRACT This study evaluates the potential of object-based image analysis in combination with supervised machine learning to identify urban structure type patterns from Landsat Thematic Mapper (TM) images. The main aim is to assess the influence of several critical choices commonly made during the training stage of a learning machine on the classification performance and to give recommendations for classifier-dependent intelligent training. Particular emphasis is given to assess the influence of size and class distribution of the training data, the approach of training data sampling (user-guided or random) and the type of training samples (squares or segments) on the classification performance of a Support Vector Machine (SVM). Different feature selection algorithms are compared and segmentation and classifier parameters are dynamically tuned for the specific image scene, classification task, and training data. The performance of the classifier is measured against a set of reference data sets from manual image interpretation and furthermore compared on the basis of landscape metrics to a very high resolution reference classification derived from light detection and ranging (lidar) measurements. The study highlights the importance of a careful design of the training stage and dynamically tuned classifier parameters, especially when dealing with noisy data and small training data sets. For the given experimental set-up, the study concludes that given optimized feature space and classifier parameters, training an SVM with segment-shaped samples that were sampled in a guided manner and are balanced between the classes provided the best classification results. If square-shaped samples are used, a random sampling provided better results than a guided selection. Equally balanced sample distributions outperformed unbalanced training sets.

Using the damage from 2010 Haiti earthquake for calibrating vulnerability models of typical structures in Port-au-Prince (Haiti)

After the 2010 Haiti earthquake, that hits the city of Port-au-Prince, capital city of Haiti, a multidisciplinary working group of specialists (seismologist, geologists, engineers and architects) from different Spanish Universities and also from Haiti, joined effort under the SISMO-HAITI project (financed by the Universidad Politecnica de Madrid), with an objective: Evaluation of seismic hazard and risk in Haiti and its application to the seismic design, urban planning, emergency and resource management. In this paper, as a first step for a structural damage estimation of future earthquakes in the country, a calibration of damage functions has been carried out by means of a two-stage procedure. After compiling a database with observed damage in the city after the earthquake, the exposure model (building stock) has been classified and through an iteratively two-step calibration process, a specific set of damage functions for the country has been proposed. Additionally, Next Generation Attenuation Models (NGA) and

Seismic hazard assessment of the Province of Murcia (SE Spain): analysis of source contribution to hazard

\hbox {Vs}^{30}