Search

The hypothesis1 of subduction of oceanic plates under active continental margins has become a fundamental idea in plate tectonics. Yet the evidence for the existence of oceanic plates beneath continental borders is still largely circumstantial. In seismology, the presence of a new layer or structural unit is taken as confirmed only after seismic phases corresponding to reflections or refractions from that layer have been identified on seismograms. I report here the presence of a mantle phase, interpreted as a seismic wave refracted from a dipping interface on records of aftershocks from the intermediate earthquake of 24 October 1980 in the Mixtec Highlands of south-central Mexico which may be direct evidence on the structural position of the subducted plate under the Mexican active continental margin.

Se presenta un método para la estimación del riesgo sísmico en regiones cuya historia de sismicidad es incompleta y para las cuales no se cuenta con datos de aceleración, basado en simulaciones pseudoaleatorias del proceso sísmico con un mínimo de suposiciones. La aplicación de este método a la estimación del riesgo sísmico para la planta nuclear de Laguna Verde, México (basada en suposiciones simplistas), indica que una estimación pesimista del riesgo para un período de 50 años es del orden de 6.0%, para la aceleraci6n de OBE de 0.12g y del orden de 0.97% para la aceleraci6n de SSE de 0.24g.

We propose two alternate structural models featuring a dipping oceanic plate in the subduction zone of southwestern Mexico, based on the relocation, by the Master Event method, of 77 aftershocks of the 1980 Huajuapan, Mexico earthquake, recorded by the Mexican Continental Aperture Seismic Network (RESMAC). The aftershock locations roughly define the rupture surface of the main event as a normal fault dipping towards the continent. The rupture involved a thickness of about 25 km in the top of the subducted plate. These results are consistent with the fault plane solution. Both models feature a waveguide in the top of the subducted plate, which may account for the large relative amplitudes of plate waves relative to direct waves at all stations, and the large high frequency content observed at a station on the coast, compared with stations inland.

A permanent seismographic array has been established around the Gulf of California. Solar-powered stations are operating at Caborca, Bahia de los Angeles, Guaymas, Topolobampo and La Paz. Trailer stations using local 60-cycle power are operating at San Felipe, Rio Hardy, Rancho Meling, and Ensenada. The array is unique in the following aspects: (1) It is the only array operating in close proximity to an active region of sea-floor spreading; (2) its sophisticated low-power design and use of solar energy made siting choices simpler so that local noise sources could be minimized; (3) it is the first accurately timed array to operate in northwest Mexico. The array has already provided important information on seismicity in the Gulf of California and has aided in several sonobuoy studies of swarms and aftershock sequences. As more data accumulate, many of the important questions concerning the seismicity, tectonics, and structure of the Gulf of California will be answered.

In 1970 we began a study of kinship structure in an upper-class family in Mexico City that ranged over five generations of men and women, including 118 nuclear families.1 These were the descendants of Carlos Gomez (1825-76), a small-town merchant of the state of Puebla, who had three children by his first, criollo wife, and six by his second, Indian wife. From this nuclear family emerged a number of branches, within which socioeconomic stratification developed. One of the "mestizo" sons, who

El método propuesto corrige la inhomogeneidad en la ditribución azimutal de estaciones, adaptando los residuales a una función coseno del azimut. A los arribos tempranos se les asigna mayor peso que alos arribos tardíos. Los pesos son optimizados experimentalmente mediante el uso de datos de grandes exposiones. En lugar de tabular tiempos de viaje se utilizan aproximadamente polinomiales; por lo tanto, los nuevos programas requieren considerablemente menos tiempo de computación y de almacenamiento de datos. La explosión Longshot del 20 de Octubre de 1965 ha sido relocalizada. El desplazamiento epicentral de Longshot es normal al eje de la trinchera y mayor que el reportado anteriormente. Esto significa que el término de fuente de Longshot no puede ser adecuadamente explicado por diferencias de temperatura entre la placa hundida y el mmanto que la rodea.

Numerical experiments on simulating ruptures of a one-dimensional plate boundary suggest that certain basic features of seismicity (including multiple ruptures and aftershock sequences) can be adequately reproduced if one assumes the existence of an elementary particle (or ‘geon’) of fault dynamics. Some general consequences of corpuscular assumptions in the earth sciences are discussed, with particular reference to the estimation of hazard.

The standard regression model for hypocenter determination contains four parameters: latitude correction, longitude correction, focal depth correction, and origin time correction. Similarity between the function ∂t/∂Δ and ∂t/∂H for values of Δ < 20° and again for Δ > 25° causes linear dependence between the coefficient vectors, provided that the observations are contained within a relatively narrow azimuthal range. The regression is overparameterized, and nothing can be inferred about the solution even when the procedure does not diverge. A solution can always be obtained in such cases by reducing the number of parameters (e.g., by restricting the focal depth); but such a solution remains dependent on the error in focal depth.Because of the geometry of the Earth's subduction zones largely surrounding the continental masses, many earthquake observations are clustered within an azimuthal range of the order of 90°. Thus, the problem of overparameterization is inherent in the location model as set up originally by Geiger. For example, the data set for LONGSHOT is shown to contain 150 observations in the azimuth range 350° to 90°, including all European and North American stations. Hence, the LONGSHOT least-squares location is poorly conditioned. In order to obtain well-conditioned solutions, the hypocentral location model must be reformulated in such a way that the epicenter is determined independently of the focal depth, thus avoiding overparameterization.

The occurrence of cyclical instabilities along plate boundaries at regular intervals suggests that the process of earthquake causation differs in some respects from the model of elastic rebound in its simplest forms. The model of tectonic feedback modifies the concept of this original model in that it provides a physical interaction between the loading rate and the state of strain on the fault. Two examples are developed: (a) Central Chile, and (b) Mexico. The predictions of earthquake hazards for both types of models are compared.

The Popayán earthquake of 1983 was a highly destructive, high mortality event (mb = 5.5). It was probably caused by the rupture of the Pubenza fault, a NNW-trending feature mapped on the basis of cracks in pavements and breaks in underground water pipes. Significant damage to modern construction was confined to an area within 1 km of the fault trace in the urban area. Near-field observations of upthrow and other effects suggest that the earthquake featured a high-amplitude pulse of the order of 1 g in the vicinity of the fault trace. Losses were of the order of $500 million.

The Colima earthquake (magnitude 7.5) occurred just inland from the Middle America Trench, 110 km south of the Volcan de Colima and 160 km southeast of Manzanillo, Colima, Mexico. Damage at several cities and towns was severe, 30 people were killed, and hundreds were injured. Four days after the earthquake, a six-station portable seismograph array was set up in the epicentral area as part of a cooperative program between UCSD, the University of Mexico, and the Mexican Federal Power Commission. From about 330 aftershocks recorded in the following 212 weeks, accurate locations were obtained for 50. One large aftershock had a magnitude of 6.2, the others range in local magnitude from 1.5 to 4.5. The locations outline a region approximately 90 km long and 60 km wide, in nearly the same location as the aftershock zone inferred by Kelleher et al. (1973) for the 1941 earthquake. The focal depth of the aftershocks (ranging from 2 to 30 km) and the fault-plane solutions for the main event indicated a shallow dipping thrust plane (about 30°). The seismic moment estimated from mantle rayleigh waves is 3 × 1027 dyne-cm. The pattern of aftershocks was used to estimate the source dimensions. From the moment and source dimensions the average slip was estimated to be about 1.4 m, corresponding to a stress drop of about 8 bars. The occurrence of this earthquake is discussed in terms of the general seismicity of the Middle America Trench, the convergence rate predicted by plate tectonics, and the use of seismic gap theory for earthquake prediction. The fact that this earthquake may have been in the zone of the 1941 earthquake rather than in the adjacent seismic gap, suggests that caution must be taken in using seismic gap theory to predict earthquakes in the region. It further suggests that in the adjacent seismic gap a large earthquake may be eminent, and thus the gap may be an important area for deploying seismic instruments.

A major simplification of teleseismic epicenter location procedures can be achieved by regression in the space domain. If distance residuals are used instead of travel-time residuals the normal equations become linear. Observations must be weighted by ∂t/∂▵; however, linear weighted regression is extremely fast as compared to unweighted nonlinear regression. The new program features several other innovations; separate estimation of focal depth, self-initialization using the nearest station as trial epicenter, and second-degree approximations to the Herrin tables instead of travel-time tables in core. On the basis of a relocation of event LONGSHOT, the new routine is found to be about ten times faster, absolutely convergent, and more compact than any of the standard programs in current use.

DIFFICULTIES have arisen in attempting to reconcile the tectonic evolution of continental borders with the theory of ocean floor spreading. In the case of the Chilean coast, sea floor spreading has apparently been confirmed by magnetic evidence and disproved by the presence of undisturbed sediments in the trench. Scholl, von Huene, and Ridlon1 have suggested that there are only four alternative solutions to this difficulty, none of which is supported by experimental evidence.

Plotting the seismic travel time from a point as a function of azimuth and distance yields a surface called a “chronoid”. The chronoid for the standard earth is a surface of revolution, termed the “standard chronoid”. Local anomalies are defined as the set of coordinate transformations required to bring the local chronoid into least-square coincidence with the standard chronoid. In general this is accomplished by means of two mutually orthogonal translations, plus one rotation in an azimuthal plane. These operations give rise to three distinct components of the local anomaly; a geophysical interpretation of each of the anomaly components is provided.

Summary Earlier work on preferred patterns of transition of earthquake activity between seismic regions has been inconclusive. The problem is restated in terms of 51 seismic regions. A series of 4164 earthquakes covering the period 1904-1954 is analyzed for randomness of transition between consecutive events. Using the multinomial distribution as a null hypothesis, contingency tables are derived for the sequence of events M ≥ 6.5 and M ≥ 7.0. It is found that the hypothesis of randomness is rejected for both sequences at the 0.975 significance level. This result is attributed to the presence of aftershocks.

The 1970 Peru earthquake was a multiple shock. The first widely-recorded shock had a magnitude of about 5; within 20 sec there were at least four larger earthquakes, at least two of which exceeded magnitude 7. Focal mechanisms appear to be complicated, with at least one major shock yielding compressions at distances of 25° to 30° to the northeast. Most aftershocks occur at depths of 40 to 70 km; their first motions are sometimes reversed with respect to the main shocks. The seismic history of the region includes one comparable earthquake in 1725 but no events of similar magnitude thereafter. The town of Ranrahirca was not destroyed by the 1970 mud flow but by an earlier event in 1962.

Travel times from earthquakes or explosions contain both positive and negative systematic errors. Positive skews in travel-time residuals due to epicenter mislocation, and negative skews due to lateral inhomogeneity in the Earth, are analyzed. Methods for travel-time estimation are critically reviewed. Recent travel-time tables, including the J-B tables, are within the range of root-mean-square travel-time fluctuations; the J-B tables are systematically late but cannot be reliably improved by least-square methods. Effects of lateral inhomogeneity at teleseismic distances can be estimated by chronoidal methods independently of standard tables, but the available explosion data are insufficiently well-distributed in azimuth and distance for this purpose.

The San Salvador earthquake of May 3, 1965 was preceded by a local seismic swarm of three months duration. The main shock was destructive in a densely populated area of not more than 15 km in radius; the same general area was damaged in the earthquakes of 1576, 1659, 1798, 1839, 1854, 1873, 1880, 1917, and 1919. Over 120 casualties were reported. The epicenter has been located on the south rim of the Median Trough, a post-Pliocene structure which accounts for the high seismic and volcanic activity in the region. The observed intensity is attributable to shallow focal depth and to the presence of thick inhomogeneous beds of fluviatile pumice. The tectonic setting and shallow subsurface factors should be recognized in future building codes and zoning regulations.

A catalog of major Chilean earthquakes (magnitude presumed greater than 7.5) is presented. This catalog is abstracted from a punched-card list of Chilean earthquakes with over 15,000 entries. Each event is accompanied by a description of effects, including tsunami observations, and estimates of epicentral location and magnitude. Major earthquakes in Chile occur in a small number of source areas. These are aligned offshore and along the range faults between the Coast Range and the Central Valley. In the Central Chile region, between Valparaiso and Concepcion, major earthquakes occur chiefly inland. From Concepcion to the south the major epicenters occur offshore. Each focal area produces a predictable pattern of seismic and tsunami effects.

Al graficar el tiempo de recorrido sísmico desde un punto como función de azimut y distancia se tiene una superficie llamada "cronoide", La cronoide para una tierra estandar es una superficie de revolución, llamada la "cronoide estándar". Las anomalías locales se definen como el conjunto de transformaciones de coordenadas requeridas para llevar a una coincidencia de mínimos cuadrados la cronoide local con la cronoide estándar. En general, esto se obtiene por medio de dos traslaciones ortogonales mutuas, más una rotación en un plano azimutal. Estas operaciones dan lugar a tres componentes distintas de la anomalía local. Una de ellas, la "anomalía de sesgo", introduce un error en el residuo del tiempo de recorrido como lo defini6 Jeffreys. Se demuestra que el valor esperado del residuo es mayor que cero. La suposici6n de que el residuo medio es cero lleva a una sobresti-maci6n sistemática del tiempo de recorrido estándar.

Seismological results on the destructive Peru earthquake (M = 7.5) include: determination of epicenter, isoseismal map, tsunami observations, and some correlations between intensity and geologic factors. The epicenter was located off the coast of central Peru, about 190 kilometers northwest of Lima. Isoseismals were elongated in the direction of the coast line. The intensity exceeded VIII in the region of Huacho, and reached VII in Lima. There are indications, however, that accelerations and frequencies were both higher than would be expected in the regions for which the Modified Mercalli Scale was developed. Local high intensities were frequently correlated with the occurrence of recent alluvium. Analysis of tsunami travel times shows that the Peru Trench acts as a high-velocity channel which refracts wave energy to the coast in advance of the expected arrival over the direct path.

The Benioff torsion apparatus for creep testing of slender cylindrical specimens is described. Continuous creep and creep recovery curves for granodiorite and gabbro at room temperature and atmospheric pressure are given. For constant-torque tests of up to 1 week's duration, the results are closely represented by the equation \documentclass{aastex} \usepackage{amsbsy} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{bm} \usepackage{mathrsfs} \usepackage{pifont} \usepackage{stmaryrd} \usepackage{textcomp} \usepackage{portland,xspace} \usepackage{amsmath,amsxtra} \usepackage{wasysym} \pagestyle{empty} \DeclareMathSizes{10}{9}{7}{6} \begin{document}$$\epsilon (t) = \frac{\sigma}{\mu} [1 + q\ {\mathrm{ln}}\ (1 + at)],$$\end{document}, which was found valid for low stresses, up to 0.05 per cent of the rigidity modulus. The results are compared with early torsion creep experiments by A. A. Michelson. The strain behavior of rocks at low stresses is discussed.

A positive correlation is established between the time of day and the number of victims in Chilean earthquakes. This correlation is attributed to a 24-hr cycle of dwelling occupancy. The incidence of foreshocks lowers the number of casualties. These observations are of interest to earthquake engineering design and to emergency measures for the protection of human populations in earthquakes.

THIS communication outlines a probability model which seems to provide an adequate basis for making predictions concerning the occurrence of largest earthquake magnitudes over time: (a) The number of earthquakes in a year is a Poisson random variable with mean α; (b) X, the earthquake magnitude, is a random variable distributed with cumulative distribution function

The slope b of the magnitude distribution of earthquakes is proportional to the inverse of the mean magnitude in a region. Systematic changes of b with depth in some subduction zones are interpreted as evidence that the downgoing slabs are not continuous down to depths of 600 km. The Tonga-Kermadec and Japan subduction zones exhibit no significant variation of b with depth and are therefore proposed as prototypes of continuous subduction zones.

An attempted use of seismic gap observations to predict a large earthquake in Oaxaca, Mexico is discussed. The observations were initially published in a scientific journal and were subsequently distorted by nonscientists, who predicted a major earthquake and tsunami to take place at Pinotepa Nacional, Oaxaca on 23 April 1978. Public reactions and property losess sustained by individuals and communities were comparable to those expected from an actual earthquake. A revision of epicenter locations from the NOAA data file revealed that a number of earthquakes did occur in the alleged gap but had been excluded because their reported focal depth was in excess of 60 km. It is shown that the probability that the number of earthquakes in two consecutive time intervals of a stationary Poisson process differs by an amount which would be reported as a ‘seismic gap’ is of the order of 5% or more for Oaxaca. This means that spurious ‘seismic gaps’ would be observed in one out of 20 data runs. The possibility of detecting a true interval of abnormal quiescence in a random earthquake sequence appears to be fairly remote in this case.

The paper contains, 'Research on the Central California Large-Scale Seismic Array.' The objectives of the study are as follows: Maintain and operate the telemetered network of stations in the seismically active Coast Range; Conduct research on the seismic data recorded from this network. This research should include the following: (a) Focal mechanism and aftershock characteristics of small earthquakes. (b) Propagation of S waves across the network. (c) P-wave coherence from teleseisms. (d) Improved travel-time curves. (Author)

Coupling between an evanescent P wave and a Rayleigh wave can occur when the two modes share the same phase velocity at an interface. The result is a new, distinctive monochromatic mode of large amplitude and finite duration that may reverberate in a bounded layer, generating stationary patterns due to constructive and destructive interference. Such modes may be responsible for earthquake damage in the Valley of Mexico and perhaps in other sedimentary basins vulnerable to seismic disasters on soft ground.

As an addendum to J. Urrutia Fucugauchi's (Eos, 63, June 8, 1982, p. 529) excellent analysis of why things go wrong in Latin American geophysics, I submit that funds in whatever form are not the only answer. In Mexico over the past decade there has been a reasonable availability of funds, yet no dramatic increase in the quality or quantity of geophysical research was detected. Graduate scholarships have even gone begging for applicants in the earth sciences! Leadership is the big problem. National plans and forecasts for science and technology continue to ignore this central fact. They want to generate hundreds, nay thousands, of middle-level scientists while providing no incentive for excellence. As others have found out long before us, this approach is doomed from the start.

A new seismic array is currently being installed throughout Mexico. Resmac, which stands for Red Sismica Mexicana de Apertura Continental, or Mexican continental aperture seismic array, will use a nationwide microwave communication network as a vital element in a modern, all-digital, integrated system of seismic data acquisition and processing. The Mexican microwave network spans the country from Tijuana to Cozumel and from coast to coast: hence it is called a ‘continental aperture array.’ Eventual cooperative projects now under the consideration of the Central American nations make a southward extension to Panama a possibility. At present, however, the Resmac project is sponsored exclusively by five agencies of the Mexican federal government, there being no financial or technical contribution from outside sources. The executive agency is the National University of Mexico, which is in close cooperation with the Department of Communications and Transportation of the federal government.

This brief introductory textbook fills a definite need. There have been no accessible references in Spanish on general geophysics. The level of the book requires only a knowledge of high school physics, according to the preface; however, this is not so much a high school text as a reference guide for the general reader. The book contains four chapters: (1) Gravity and the Shape of the Earth, (2) Earthquakes and the Interior of the Earth, (3) The Magnetic field, and (4) Origins and Evolution of the Earth. The latter chapter contains most of the relevant information on plate tectonics and dynamics of the lithosphere. The sequence of chapters is somewhat awkward, since the idea of lithospheric plates is already introduced in Chapter 2. In fact, the discussion of plate tectonics contains several inconsistencies: in Figure 4-4, for example, volcanoes are shown as plumes rising from the Benioff zone and erupting offshore, on the continental slope. The text compounds the confusion: ‘Because of the dip angle of the sinking plate, these volcanoes are found behind the coast line. Continental rocks in the collision zone are pushed up to form large mountain ranges in some places, such as the Andes.’

Statistical prediction is an extension of Fourier spectrum analysis of time series. Historical attempts at prediction of earthquakes have centered on the question of periodicity. A survey of work in this field is given, both preceding and following critical investigations by Jeffreys. Present evidence indicates earthquakes to be uncorrelated events; observed deviations from the Poisson distribution, previously attributed to dependence in time, may be an effect of spatial inhomogeneity. It is fallacious to infer properties of the earthquake process on the basis of the shape of the distribution function of number of events in time. Some restrictions are shown to apply to the use of direct extrapolation methods in earthquake time series. Multiple prediction is a possibility; however, the actual measurement of geophysical variables for prediction purposes has progressed very slowly. Direct measurements of stresses and strains at depth are needed; values based on the theoretical release of strains by earthquakes are insufficient. Plotting the square roots of earthquake energies yields no specific strain information of predictive value. The estimation of earthquake risks is different, in principle, from the prediction problem. Mathematical techniques are available for risk estimation. A simplified model of the earthquake process is suggested, which may satisfy elementary requirements for purposes of engineering, city planning, and earthquake insurance.