Probabilistic Seismic Hazard Maps Research Articles

Probabilistic seismic hazard maps and intra-plate earthquake hazard reduction

Probabilistic seismic hazard maps and intra-plate earthquake hazard reduction

2003年7月26日宮城県北部地震M6.4 が提起した地震動予測地図と活断層評価の課題

2003年7月26日宮城県北部地震M6.4 が提起した地震動予測地図と活断層評価の課題

OUTLINE OF STRONG-MOTION EVALUATION IN NATIONAL SEISMIC HAZARD MAPPING PROJECT OF JAPAN

The National Research Institute for Earth Science and Disaster Prevention (NIED) has carried on the special research project 'National Seismic Hazard Mapping Project of Japan' to support the preparation of the National Seismic Hazard Maps made by the Headquarters for Earthquake Research Promotion. In this article, the outline of methodologies to evaluate strong-motion both for the probabilistic seismic hazard maps and for the scenario seismic hazard maps is presented.

Precarious rock evidence for low near-source accelerations for trans-tensional strike-slip earthquakes

This paper describes precarious rock evidence for low ground motions associated with extensional sections of strike-slip faults. Recent evidence from physical and numerical models and data regressions has indicated that ground motion for extensional strike-slip regions may be lower than for strike-slip faults with a large fault-normal tectonic stress component, and for thrust faults in general. Data from compressional strike-slip and thrust earthquakes dominates the database used in most regression curves for ground acceleration, and in the calculation of current probabilistic seismic hazard maps. Therefore, estimates of ground accelerations on these seismic hazard maps may be too high for sites near extensional sections of strike-slip faults. This paper discusses precariously balanced rock data from three areas near extensional sections of strike-slip faulting: (1) the region of the Honey fault, California, with an active Holocene fault, (2) the Red Rock Canyon region of the Garlock fault, near a dilatational step-over, and (3) the region just south of Beaumont, California, near the Hemet dilatational step-over in the San Jacinto fault. These are all active strike-slip faults, with at least a few large earthquakes in the Holocene, and, in the case of the San Jacinto example, historic large earthquakes ( M=7). Thus, the precarious rocks at these sites are evidence of relatively low ground motions associated with extensional strike-slip faulting. The results of this study could be very important in developing more detailed seismic hazard maps in the future.

Probabilistic PGA and Arias Intensity maps of Kyrgyzstan (Central Asia)

New probabilistic seismic hazard and Arias Intensity maps have beendeveloped for the territory of the Kyrgyz Republic and bordering regions.Data were mainly taken from the seismic catalogue of Kyrgyzstan and partlyfrom the world seismic catalogue. On the base of seismicity and activetectonics, seismic zones were outlined over the area. For these,Gutenberg-Richter laws were defined using mainly instrumental data, butregarding also historical events. Attenuation of acceleration inside the targetarea could not be determined experimentally since existing strong motiondata are insufficient. Therefore, empirical laws defined for other territories,principally Europe and China, were applied to the present hazardcomputations. Final maps were calculated with the SEISRISKIII programaccording to EUROCODE8 criteria, i.e. for a period of 50 years with90% probability of non-exceedance. For long-term prediction, 100 yearsmaps with 90% probability of non-exceedance have been developed. Theprocedure used for seismic hazard prediction in terms of PGA (PeakGround Acceleration) was also applied to Arias intensities in order to beable to define regional seismogenic landslide hazard maps.

Central US earthquake catalog for hazard maps of Memphis, Tennessee

An updated version of the catalog that was used for the current national probabilistic seismic-hazard maps would suffice for production of large-scale hazard maps of the Memphis urban area. Deaggregation maps provide guidance as to the area that a catalog for calculating Memphis hazard should cover. For the future, the Nuttli and local network catalogs could be examined for earthquakes not presently included in the catalog. Additional work on aftershock removal might reduce hazard uncertainty. Graphs of decadal and annual earthquake rates suggest completeness at and above magnitude 3 for the last three or four decades. Any additional work on completeness should consider the effects of rapid, local population changes during the Nation's westward expansion.

Methodological considerations of probabilistic seismic hazard mapping

This paper presents a methodological discussion of several issues involved with the development of maps of seismic hazard. The points made are illustrated with worked examples, using Scotland as an illustrative case. The issues treated are divided under three headings: matters relating to the difference between hazard maps and site studies; matters concerned with the technical issues of mapping, and matters relating to the use to which hazard maps will be put. It is concluded that a hazard map cannot be an all-purpose substitute for site-specific studies, owing to the impracticality of ensuring all-round conservatism in a hazard map, and the lower level of detail (more broad-brush approach) in a regional mapping study. Also, since users of a hazard map are not necessarily going to be engineers, consideration should be given to the provision of maps expressed in parameters other than physical measures of ground motion. Intensity is useful here, since it relates to actual earthquake experience and to damage. One can also move to making maps of generic seismic risk even before one has data on the distribution of exposure and vulnerability. Discussion is made of the issue of testing the validity of hazard maps against real experience, with examples. If a map can be shown to accord with real observations, then it can be treated with greater confidence by users.

Probabilistic seismic hazard maps for the Japanese islands

Probabilistic seismic hazard maps are constructed by utilising a new methodology called the parametric-historic method, since it combines the best features of the ‘deductive’ ( Cornell CA. Engineering seismic risk analysis. Bull Seism Soc Am 1968;58;1583–606) and ‘historic’ ( Veneziano D, Cornell CA, O'Hara T. Historic method for seismic hazard analysis. Elect Power Res Inst, Report NP-3438, Palo Alto, 1984) procedures. The technique has been developed specifically for the estimation of seismic hazard at individual sites without the subjective judgment involved in the definition of seismic source zones, when specific active faults have not been mapped or identified, and when the causes of seismicity are not well understood. The combination of historical and instrumental data is permitted. The historical part of the catalogue contains only the strongest events, whereas the complete part can be divided into several subcatalogues, each assumed complete above a specified threshold of magnitude. Uncertainty in the determination of magnitude has also been taken into account. The maximum credible magnitude, m max, is of paramount importance in this approach. The seismic hazard maps are based on a long-term earthquake history (599–1997) compiled of the catalogues of Utsu (Utsu T. Catalog of large earthquakes in the region of Japan from 1885 through 1980. Bull Earthq Res Inst, Univ Tokyo, 1982;57:401–63), Usami (Usami T. Materials for comprehensive list of destructive earthquakes in Japan. Tokyo: Tokyo Press, 1996) and JMA for the Japanese islands. The analysis is based on subregions at a grid size of 0.05° along the Japanese islands, for each of which peak ground accelerations and spectral accelerations for natural frequencies of 1, 3, 5 and 10 Hz, are predicted and mapped to occur at a 10% probability in 50 years.

The Umbria–Marche case: some suggestions for the Italian seismic zonation

Probabilistic seismic hazard maps, in terms of spectral acceleration and uniform hazard response spectra at given sites, considering local soil conditions, represent a much more complete estimate of the seismic hazard than the traditional maps in terms of peak ground acceleration or macroseismic intensity. This is particularly true when the requests of urban planners and engineers have to be met. The present analysis shows how some hazard parameters, such as the effective peak acceleration and the spectrum intensity, can well synthesise the overall information available from traditional probabilistic studies, but also suggests that soil condition is a first-order ingredient for effective seismic hazard mapping at national level. Three Italian towns, damaged by the 1997 Umbria–Marche earthquake sequence, are considered as example to demonstrate that: (1) soil condition dependent uniform hazard spectra well approximate actual spectra recorded during some events of the seismic sequence; (2) for these localities, the design spectrum of the present Italian seismic code does not seem adequate. These considerations have induced the Italian scientific community to propose an updating of the national seismic zonation on the basis of several hazard parameters, that are described in this paper.

New Seismic Zoning Maps for Norway, the North Sea, and the United Kingdom

The methods adopted to evaluate seismic hazard may vary significantly from country to country, even where the countries share a common border and the regional seismic zones overlap. Apart from differences in method, the underlying seismological and geological databases maintained in neighboring countries may differ substantially in scope, reliability, and interpretation, and thus cause further international disparities in the conduct and output of seismic hazard assessment. For reasons both technical and administrative, it is rare to establish a collaborative industrial venture involving independent organizations from several countries aimed at joint probabilistic seismic hazard mapping for a region spanning a common international border. All too often, hazard contour maps (with some regulatory significance) display a discontinuity across frontiers, which is explained by an international breakdown in human communication rather than being attributable to any natural geological fracture. A bizarre effect of this “nationalism” is that the hazard maps at the national borders deviate from the neighbor country maps in terms of results as well as presentation means (and, more understandably, in terms of design rules). In the North Sea this is far from an academic discussion. Oil rigs are clustered along the Viking Graben on both the British and Norwegian sides, and a distance below what is geologically significant may entail a large difference in seismic hazard estimates and certainly in design regulations (supported by seismic hazard zoning maps). It is evident that this situation is confusing and disturbing for interested parties, including the industry, which wants to apply earthquake-safe design in the border areas. Norway developed, in parallel with the first petroleum installations in the North Sea, its first earthquake design regulations through the Norwegian Petroleum Directorate (NPD). Dating back to 1977, the first regulations specified earthquake loading to be of the “accidental” type (in contrast to an “environmental” load). …

Seismicity, seismotectonics and seismic hazard of Italy

A first generation of probabilistic seismic hazard maps of the Italian country are presented. They are based on seismogenic zoning deriving from a kinematic model of the structural tectonic units and on an earthquake catalogue with the foreshock and aftershock events filtered out. The following ground motion parameters have been investigated and mapped using attenuation equations based on strong-motion recordings of Italian earthquakes: peak ground acceleration and velocity; Arias intensity; strong motion duration; and the pseudovelocity and pseudoacceleration spectral values at 14 fixed frequencies both for the vertical and the largest horizontal component. A Poissonian model of earthquake occurrence is assumed as a default and the hazard maps are presented in terms of ground motion values expected to be exceeded at a 10% probability level in 50 years (return period 475 years) according to the requirement of Eurocode 8 for the seismic classification of national territories, as well as in terms of exceedance probabilities of selected ground motion values. Finally, as a tentative study, the use of hybrid methods (implementing both seismogenic zones and structures), renewal processes (including earthquake forecasting) and the influence of site effects (as the basis for the planning of earthquake scenarios) were explored.

Geology in the 1996 USGS Seismic-hazard Maps, Central and Eastern United States

The current (1996) national probabilistic seismic-hazard maps utilize information about geologic structure and tectonics of the central and eastern U.S. to compensate for uncertainty that arises from the short seismicity record. Geology was incorporated into the maps mainly as seven source zones that are delineated in three distinct ways. The North American stable continental region is divided into two large zones, the sparsely seismic Precambrian craton and the more active Phanerozoic rim. Five other source zones are much smallerqthe Wabash Valley source zone is within the craton, whereas the Reelfoot Rift, eastern Tennessee, Charleston, and Charlevoix source zones are in the Phanerozoic rim of the continent. We document these zones and explain and justify their use. The seven zones provide a foundation from which we suggest a criterion for including more geology in future maps.

Dating precariously balanced rocks in seismically active parts of California and Nevada

Precariously balanced boulders that could be knocked down by strong earthquake ground motion are found in some seismically active areas of southern California and Nevada. In this study we used two independent surface-exposure dating techniques—rock-varnish microlamination and cosmogenic 36 Cl dating methodologies—to estimate minimum- and maximum-limiting ages, respectively, of the precarious boulders and by inference the elapsed time since the sites were shaken down. The results of the exposure dating indicate that all of the precarious rocks are >10.5 ka and that some may be significantly older. At Victorville and Jacumba, California, these results show that the precarious rocks have not been knocked down for at least 10.5 k.y., a conclusion in apparent conflict with some commonly used probabilistic seismic hazard maps. At Yucca Mountain, Nevada, the ages of the precarious rocks are >10.5 to >27.0 ka, providing an independent measure of the minimum time elapsed since faulting occurred on the Solitario Canyon fault.

Benefits of scenario ground motion maps

Earthquake hazards can be communicated with both probabilistic seismic hazard maps and “deterministic” maps showing estimated ground motions expected from a subset of the possible earthquakes. This paper describes the latter class of maps as “scenario” ground motion maps. The principle behind the scenario maps is to portray a median estimate of ground motions from events that are reasonable, and regionally the most significant. The intent is that the scenario map answers the common question: “If the fault ruptures, what do you expect?” This paper proposes rules for construction of scenario maps, argues that a parallel map (a scenario probability map) should be constructed giving estimates of the annual recurrence rates of the scenario ground motions, and compares probabilistic and scenario ground motions for areas with high and moderate seismic hazards. Uses for scenario maps include educating the public about the earthquake hazards, some engineering applications and hazard response planning.

Predominant seismic source distance and magnitude maps for Los Angeles, Orange, and Ventura Counties, California

Abstract Hazard deaggregation at each grid point of a probabilistic seismic hazard map can provide maps of the most probable seismic source distance and magnitude. Maps of predominant seismic source distance and magnitude for PGA, and 5% damped spectral acceleration (SA) at 0.3-, 1.0-, and 2.0-sec periods are produced for the first time for Los Angeles, Orange, and Ventura counties in southern California. These maps show that large earthquakes on faults local to any site generally dominate the 10% probability of exceedence in 50- and 100-yr hazards in southern California. Major faults, like the San Andreas, San Jacinto, and Elsinore faults, dominate mainly in their vicinity. At spectral periods beyond 2 sec, earthquakes on these major faults can dominate out to larger distances.

Earthquake hazard in Indian region

The causes of earthquakes and their incidence in the Indian region have been briefly reviewed. The basic elements of probabilistic seismic hazard analysis, which is a basis for making pragmatic decisions for strategies for hazard reduction and mitigation, are discussed. A probabilistic seismic hazard map of the Indian region, which delineates the seismic hazard in various regions in terms of peak ground accelerations expected to be exceeded with a probability of 10% in any 50-year period, is presented. The need and scope for further research are outlined.

Treatment of Parameter Uncertainty and Variability for a Single Seismic Hazard Map

The inputs to probabilistic seismic hazard studies (seismic source zones, earthquake rates, attenuation functions, etc.) are uncertain, being based on subjective judgments and interpretations of limited data. In the face of this uncertainty, we consider (a) how one might “reasonably” determine the ground-motion levels to show on a single probabilistic seismic hazard map, and (b) the extent to which uncertainty in the calculated levels can be meaningfully represented on such a map. If the “best guess” estimates of the earthquake rate, the Gutenberg-Richter b-value and the maximum magnitude for a single source zone are regarded as uncorrelated and the uncertainty in each parameter can be regarded as symmetric about the estimated value, then the probabilistic ground-motion levels calculated using these best estimates represent both most likely values and also approximate mean values.

Probabilistic Seismic Hazard Mapping of the Mississippi Embayment

Abstract Seismic hazard is calculated for a grid of sites on the Mississippi embayment and adjacent areas. The seismic zonations, seismicity parameters, and maximum magnitudes for these calculations were synthesized from the EPRI (Electric Power Research Institute) and LLNL (Lawrence Livermore National Laboratory) interpretations, augmented by information from recent relevant studies. Attenuation functions developed for this study include the effects of source size and crustal structure. The soil amplification factors include the effects of surficial geology, soil-column thickness, and soil nonlinearity. Seismic hazard maps, including the effect of site conditions, are presented for spectral accelerations at 1 and 10 Hz and for peak acceleration, for average return periods of 100, 500, and 1000 years. These maps show significantly lower ground-motion amplitudes than those reported by Algermissen et al. (1991). Because of these large differences, it is suggested that code- development decisions be made after consideration of both studies.

Probabilistic seismic hazard analysis and damage assessment in Andalusia (Spain)

Data are provided for use in earthquake engineering, urban planning and civil defense. The investigation has been applied to Andalusia, the most seismically hazardous area of Spain. First, a new probabilistic seismic hazard map is calculated by using geological information and the data contained in the seismic catalog and the atlas of isoseismal maps. The hazard map has been compared with others obtained earlier by various other researchers. Second, the extent of the damage was derived according to a simple methodology using as a database the geographical and statistical information on buildings and population (obtained from the Spanish census) and considering several assumptions regarding the percentage and vulnerability of each type of construction. Finally, a probabilistic analysis of the values obtained in the assessment of the seismic hazard and damage is included.

A seismic hazard map of India and adjacent areas

We have produced a probabilistic seismic hazard map showing peak ground accelerations in rock for India and neighboring areas having a 10% probability of being exceeded in 50 years. Seismogenic zones were identified on the basis of historical seismicity, seismotectonics and geology of the region. Procedures for reducing the incompleteness of earthquake catalogs were followed before estimating recurrence parameters. An eastern United States acceleration attenuation relationship was employed after it was found that intensity attenuation for the Indian region and the eastern United States was similar. The largest probabilistic accelerations are obtained in the seismotectonic belts of Kirthar, Hindukush, Himalaya, Arakan-Yoma, and the Shillong massif where values of over 70% g have been calculated.