S33B-01
Microtremors study applying the SPAC method in Colima state, Mexico.
One of the main parts of seismic risk studies is to determine the site effect. This can be estimated by means of the microtremors measurements. From the H/V spectral ratio (Nakamura, 1989), the predominant period of the site can be estimated. Although the predominant period by itself can not represent the site effect in a wide range of frequencies and doesn't provide information of the stratigraphy. The SPAC method (Spatial Auto-Correlation Method, Aki 1957), on the other hand, is useful to estimate the stratigraphy of the site. It is based on the simultaneous recording of microtremors in several stations deployed in an instrumental array. Through the spatial autocorrelation coefficient computation, the Rayleigh wave dispersion curve can be cleared. Finally the stratigraphy model (thickness, S and P wave velocity, and density of each layer) is estimated by fitting the theoretical dispersion curve with the observed one. The theoretical dispersion curve is initially computed using a proposed model. That model is modified several times until the theoretical curve fit the observations. This method requires of a minimum of three stations where the microtremors are observed simultaneously in all the stations. We applied the SPAC method to six sites in Colima state, Mexico. Those sites are Santa Barbara, Cerro de Ortega, Tecoman, Manzanillo and two in Colima city. Totally 16 arrays were carried out using equilateral triangles with different apertures with a minimum of 5 m and a maximum of 60 m. For recording microtremors we used short period (5 seconds) velocity type vertical sensors connected to a K2 (Kinemetrics) acquisition system. We could estimate the velocities of the most superficial layers reaching different depths in each site. For Santa Bárbara site the exploration depth was about 30 m, for Tecoman 12 m, for Manzanillo 35 m, for Cerro de Ortega 68 m, and the deepest site exploration was obtained in Colima city with a depth of around 73 m. The S wave velocities fluctuate between 230 m/s and 420 m/s for the most superficial layer. It means that, in general, the most superficial layers are quite competent. The superficial layer with smaller S wave velocity was observed in Tecoman, while that of largest S wave velocity was observed in Cerro de Ortega. Our estimations are consistent with down-hole velocity records obtained in Santa Barbara by previous studies.
S33B-02
Damage in the town of Miahuatlan by Oaxaca earthquake of September 30, 1999
Instituto de Ingeniería Coordinación de Ingeniería Sismológica Universidad Nacional Autónoma de México jccsa@pumas.ingen.unam.mx On September 30, 1999 (11:31 local time) a 7.4-magnitude earthquake occurred along the coast of the southern state of Oaxaca (by its proximity called Puerto Escondido earthquake) resulting of subduction of the Cocos plate under the North American continental plate. Reported fatalities 30 people and much more injured. The intense movement felt by people in Mexico City with great scare, caused considerable damage in churches of Oaxaca, as religious representative monuments. Its behavior was some stronger and better earthquake resistant characteristics. Very much houses made of adobe widely used in a poorest state were damaged. Many heavy parapets fell over the street. From 440 km to Mexico City (with light damage) was registered maximum horizontal acceleration of 28 cm/s2, also near to 137 km from the epicenter with 196 cm/s2 on Oaxaca City, were the damage was concentrated, with more than 260 historical building. Constructions moderns were not damaged. To south of Oaxaca the rural town of Miahuatlan was damaged in your adobe houses and some of them destroyed with wood roofs supporting clay tiles, your principal church suffer the collapse of the tower in the side left and some failures in the interior of this church. The fallen superior part of its left tower was projected toward the left side and the other tower of right side not collapsed. Crack in the walls bordering the base of the towers (over the ceiling) as a form of stresses acting along to small walls with cracks in X form (showing effects in this walls to the action of inverted pendulum), as indication of the movement on the four directions. Also was observed some cracks in the small arcs of the towers and in a very high parapet upon of the entrance in the upper in the front of the church without cracks or collapse. It is showed much of the failure in adobe houses characteristic from damage for earthquake. Others adobe houses reinforce in your base not suffered damage due to this and wide walls.
S33B-03
Development of Tsunami Hazard Maps along the Pacific Coast of Nicaragua
We developed Tsunami hazard maps in four areas with an extension of 30 km each along Pacific coast of Nicaragua, based on the observed worst historical Tsunami inundation record, assuming that the future Tsunami earthquake with similar magnitude would occur again in these areas as the worst case. This is a part of the result obtained by the technical cooperation "The Study for Establishment of Base Maps for GIS in the Republic of Nicaragua" executed during the year from 2004 to 2006 by Japan International Cooperation Agency with INETER as a counterpart agency upon the request by the government of the Republic of Nicaragua. By integration of available topography information that exist in various institutions in Nicaragua and additional study, we developed topography model using 2700m, 900m, 300m, and 100m sized grid system. The program used for the simulation was based on the one developed by Tohoku University and published by UNESCO (1997). Historical records of Tsunami in Nicaragua were documented since mid-19th century. Among them, the 1992 Nicaragua Tsunami earthquake was the largest one. The methodology used here was verified using tide gauge records, inundation records, and inundation areas observed during the 1992 Nicaragua Tsunami. Further, "the worst case scenario" for each one of the four study areas was estimated moving the same dimension of fault along the coast. Finally, Tsunami hazard map was made that visualizes the spatial distribution of simulated maximum wave height, maximum inundation depth, and elevation of non inundated area. We expect that the resulted maps will serve as basic reference for Tsunami disaster prevention planning by related Nicaraguan governmental institutions, local government in charge, and communities etc.
S33B-04
Applications of a GIS on Georisks for Nicaragua and Central America
A GIS on Georisks in Nicaragua was developed in the last years at the Instituto Nicaraguense de Estudios Territoriales (INETER) in cooperation with Federal Institute for Geosciences and Natural Resources (BGR, Germany). This GIS includes extensive topographical coverage and a large part of the data obtained in Nicaragua in many recent projects on natural hazard, vulnerability and risk. It contains numerous data on elements under risk, for instance cadastre data of Nicaraguan cities. Activities include the integration of the GIS with the monitoring and early warning systems of INETER to update certain parts of the data base continuously and in real time. The GIS is used on a routine basis at INETER and the GIS data base provides an efficient starting point for multiple new projects on georisks in Nicaragua which after their termination deliver their products to the GIS to assure the continuous growth of the system. Local universities, governmental institutions, local administrations, NGOīs make use of the GIS data base. Examples of important datasets are the seismicity data of Nicaragua with around 30,000 events, landslide coverage with 17,000 events, seismic vulnerability of 212,000 buildings in Managua city, seismic microzonation data of several towns, and multidisciplinary hazard and vulnerability data for 30 municipalities in Western Nicaragua. An interdisciplinary group of Nicaraguan geoscientists, informatics engineers and GIS specialists at INETER was trained to develop and use the GIS in their daily work. Web mapping services were put onto INETERīs website to provide the general public in Nicaragua with direct access to the data. Based on the experience in Nicaragua a regional GIS on Georisks for Central America is under development in cooperation with other institutions in El Salvador, Guatemala and Honduras.
S33B-05
Development of Earthquake Hazard Maps in Managua, Nicaragua
We developed 1/50,000 scale earthquake hazard maps in Managua by deterministic and probabilistic approach, compiling available data. This is a part of the result obtained by the technical cooperation - The Study for Establishment of Base Maps for GIS in the Republic of Nicaragua - during the year from 2004 to 2006 executed by Japan International Cooperation Agency with INETER as a counterpart agency upon the request by the government of the Republic of Nicaragua. We firstly collected and studied available earthquake catalogues. Among these catalogues, the historical earthquake catalogue by INETER (1505 - 1992) and instrumental earthquake catalogue by INETER (1993 - 2001) are the most comprehensive. Therefore these catalogs are selected as the base catalog and corrected and improved using other catalogues. Finally, these catalogs are unified, and then separated into two new catalogs namely Volcanic Catalogs and Non-volcanic Catalog. Then we considered three types of scenario earthquakes. For earthquake scenario from active fault, we used Aeropuerto Fault and Cofradia Fault. The location and magnitude of each fault are determined using USGS fault map and empirical formula on its length and magnitude. For earthquake scenario by volcanic earthquake, we used earthquake from Masaya volcano (M=6.0) and the one from Apyoque volcano (M=6.0). Magnitudes of these earthquakes are estimated from the past reports of the hazards. As for the probabilistic approach, based on the newly improved the Non-volcanic Catalog, hazard curve analysis is performed at the Center of Managua City. As a result, the 100-years-return period earthquake is obtained as 110 gal with the standard deviation of 28 gal. For the ground motion attenuation, three types of attenuation laws were tested to estimate maximum accelerations and MM Intensities at Managua by major earthquakes. As a result, we found that combined law of Joyner-Boore (1981) and Young et al. (1997) are appropriately applicable to the historical records and accelerations of both non-volcanic and volcanic events. As for the site effect, boring logs from 173 boreholes are collected and studied. We found that there is no obvious variety of the subsurface soil characteristics and their thicknesses at Managua City area. Through the study, the Study area is classified into three types based on the average S wave velocity over upper 30m standardized by NEHRP (FEMA302). Finally, earthquake hazard maps were prepared representing ground motion by peak ground acceleration, using 500 m square grid system.
S33B-06
Seismic Microzonation of the City of Cali (Western Colombia)
The city of Cali is located in the western margin of the Cauca Valley in the flat area between the Western and Central cordilleras of the Colombian Andes, at 70 km east of the Eastern Pacific Subduction Zone. Even though present seismic activity associated with nearest faults is low, historical records demonstrate that earthquakes have caused damage in the city going up to intensity VIII (EMS). Those earthquakes have had origin on diverse sources: the intermediate-depth Benioff zone, near and far continental crustal faults and the Pacific Subduction Zone. Taking into account the location of the city and the seismologic history of the region, neotectonic and seismological studies extending over a region of about 120000 km2 were required to compute seismic hazard. Construction of the geotechnical model of the city included detailed geological mapping, geophysical profiling, single station ambient vibration essays and the deployment of a 12-stations accelerographic network. Geotechnical properties of the soils were determined by mechanical perforations, CPTU (piezocone) and CPT (static penetration) essays, flat plate dilatometer (DMT) tests and down-hole essays which were complemented in the Lab by analysis of consolidation and static and cyclic three-axial essays. As a result, ten geotechnical zones were outlined and characterized. Finally, expected ground motions were calculated at 39 sites in the city using numerical modeling methods.
S33B-07
Dynamic Soil Modeling of Caracas Valley for the Simulation of the 1967 Caracas Earthquake
We developed a ground model in Caracas metropolitan area in Venezuela, to evaluate seismic amplification of the site and to develop earthquake scenarios. This is a part of the results obtained from a study on "Disaster Prevention Basic Plan in the Metropolitan District of Caracas in the Bolivarian Republic of Venezuela", executed by Japan International Cooperation Agency. Geotechnical borings database development in Caracas was underway that contains description of geological texture, ground water level, and SPT-N value. The maximum depth of boreholes database with geotechnical information is about 30m, whereas the deepest depth of the basin may reach more than 300 m. Wells and seismic refraction surveys results were available for estimation of the soil characteristics of deeper deposit and thickness distribution of the alluvial deposit. Micro-tremor measurements in large plain area in Caracas valley were carried out in 500m mesh by FUNVISIS and researchers from Japanese universities with the total number of 329 points. The depth of bedrock was estimated from deep wells and seismic refraction surveys results. The average shear wave velocity of the deposit above the bedrock was estimated from micro-tremor measurement. Shear wave velocity of geological layers in top 30 m or less were estimated from average SPT-N values of each layer using empirical relationship between soil types, N-value and shear wave velocity developed in Japan, For the non-linear properties of soils, such as damping and rigidity versus shear strain, appropriate curves according to soil types developed in Japan were used. The ground model in each mesh was calibrated by comparing the predominant period of H/V (Horizontal/Vertical) spectrum ratio of micro-tremor with the predominant period of theoretical amplification curve computed from the ground model. The selected 287 borehole data were applied to each nearer mesh for the modelling of shallow subsurface soils. To verify the reasonableness of the methodology and the ground model, the possibility of the reproduction of the 1967 Caracas earthquake was firstly tested. Scenario fault was located between two major sub-events, separated by 42 km, with a mechanism of strike slip with Mw= 6.6. An empirical attenuation formula that involves near-field source mechanism and different ground conditions, and which is applicable to earthquakes with large magnitudes, was used to estimate the ground motion on bedrock. The input wave was selected from worldwide strong motion database, generated from an earthquake with the same mechanism and compatible magnitude, recorded on bedrock at a distance compatible to the distance between the fault and the study area. Higher seismic intensities were observed in Palos Grandes and San Bernardino area during the 1967 Caracas earthquake. This was attributed to the thick deposit in those areas. The seismic intensity calculated using an empirical relation between spectrum intensity (SI) and seismic intensity (MMI) was able to reproduce the higher intensities in the two areas with application of the developed one dimensional dynamic ground model. This methodology was applied to simulate other earthquake scenarios as well.
S33B-08
Modelling of the Seismic Ground Motion in Barquisimeto and Cabudare cities
A 2D modeling of the seismic response is obtained using modeling techniques that allow the computation of synthetic seismograms, corresponding to two different seismic source models. The application of the modal summation technique, extended to laterally heterogeneous structural models, to the area of Barquisimeto and Cabudare allows us to create synthetic seismograms that can be used for the study of the local response along of four profiles The calculation has been made by a hybrid method: Modal summation in the regional anelastic mode (1D) where the source is located, and the finite difference in the local sedimentary structure (2D). The realistic modeling of ground motion is a very important base of knowledge for the preparation of ground shaking scenarios, as well as for the assessment of ground motion parameters for seismic microzonation.
S33B-09
USGS Training in Afghanistan: Modern Earthquake Hazards Assessments
Afghanistan is located in a tectonically active region where ongoing deformation has generated rugged mountainous terrain, and where large earthquakes occur frequently. These earthquakes can present a significant hazard, not only from strong ground shaking, but also from liquefaction and extensive land sliding. The magnitude 6.1 earthquake of March 25, 2002 highlighted the vulnerability of Afghanistan to such hazards, and resulted in over 1000 fatalities. The USGS has provided the first of a series of Earth Science training courses to the Afghan Geological Survey (AGS). This course was concerned with modern earthquake hazard assessments, and is an integral part of a larger USGS effort to provide a comprehensive seismic-hazard assessment for Afghanistan. Funding for these courses is provided by the US Agency for International Development Afghanistan Reconstruction Program. The particular focus of this training course, held December 2-6, 2006 in Kabul, was on providing a background in the seismological and geological methods relevant to preparing for future earthquakes. Topics included identifying active faults, modern tectonic theory, geotechnical measurements of near-surface materials, and strong-motion seismology. With this background, participants may now be expected to educate other members of the community and be actively involved in earthquake hazard assessments themselves. The December, 2006, training course was taught by four lecturers, with all lectures and slides being presented in English and translated into Dari. Copies of the lectures were provided to the students in both hardcopy and digital formats. Class participants included many of the section leaders from within the AGS who have backgrounds in geology, geophysics, and engineering. Two additional training sessions are planned for 2007, the first entitled "Modern Concepts in Geology and Mineral Resource Assessments," and the second entitled "Applied Geophysics for Mineral Resource Assessments."
S33B-10
Source and site effects implications on Bucharest (Romania) microzonation
The Bucharest metropolitan area is one of the most vulnerable cities of the World to earthquakes. Although the city is situated at a distance of around 150 km from the epicentral area, the damage caused by many Vrancea intermediate-depth earthquakes was extremely severe. Historical information over one thousand years suggests a rate of 2-3 damaging earthquakes per century. For example, the 4 March 1977 event produced the collapse of 32 buildings with 8-12 levels, while more than 150 old buildings with 6-9 levels were seriously damaged. Since then the occurrence of 3 other earthquakes (1986 / M=7.1; 1990 /M=6.9; 2004 /M= 6.0) demonstrated that the Vrancea seismic activity is continuing, permanently threatening the Bucharest City area. The studies done after 1977 earthquake had shown the importance of the surface geological structure upon ground motion parameters and emphasized the need for new methods of quantifying the site effects. The main purpose of our study is to analyze the influence of the source and local site conditions upon the soil response in Bucharest, in case of large (M greater than 7) Vrancea subcrustal earthquakes. To this purpose, we use on one hand, geological, geotechnical and geophysical information, including in situ measurements of shear wave velocities and data from ambient noise measurements, small-to-moderate earthquakes and large events, on the other hand. The Neogene sedimentary cover, as obtained on the basis of borehole information, undergoes a slight descend from south to north, accompanied by an increase of the deposits thickness in the same direction from about 150 m to 350 m. The predominant periods of oscillation of subsurface layers over Bucharest territory range between 1.0 and 1.9 s, increasing from south to north, in correlation with the constant increasing of the thickness of the Quaternary cohesionless deposits. The dominant resonance in the period range of 1- 2 s is obtained both using ambient noise data and earthquake data. The spectral analysis of the recorded waveforms (response spectra, power spectral density, H/V spectral ratios) shows two predominant periods, one around 0.4 s, other around 1.4 s, independently of site position in the city area. They correlate very well with the local structure layering and parameters, as shown by the theoretical amplification curves. In addition, the seismic source radiation for the largest shocks matches the local structure resonance at 1-2 s, which makes the seismic response in this range to be exceptionally enhanced as compared with the amplification observed for smaller events. Therefore, the disastrous damage reported for the high-tall buildings in Bucharest is a result of both subsurface sedimentary structure and source radiation from Vrancea shocks with magnitude exceeding 7. Our results outline two main features of high significance for earthquake engineering practice: (1) inadequacy for the Bucharest city case of the standard procedure that limits the investigation depth in order to set soil dynamic characteristics to uppermost 30m; the local response during the large Vrancea earthquake is controlled by the entire package of Quaternary cohesionless deposits which are significantly thicker than 30m beneath Bucharest and (2) difficulty to delineate zones with different local amplifications; thus, for Bucharest urban area and strong subcrustal Vrancea earthquakes one can refer rather to regional effects than "local effects".