A real-time prediction model for macroseismic intensity in China

The macroseismic intensity spatial distribution is an important input for most rapid loss modeling and emergency work. Data from a total of 175 earthquakes (Ms ≥ 5.0) in China from 1966 to 2014 were collected, and the rapid assessment method of macroseismic intensity distribution was studied. First, simple relationships among the epicentral intensity, magnitude, and focal depth were established. A greater amount of database is used in this study than that in a previous work (Fu and Liu in Sci R 4(5): 350-354 (1960), Mei in Chin J Geophys 9(1): 1–18 (1960), and Yan et al. in Sci Chin 11: 1050-1058 (1984)), and the studied earthquakes all occurred in the last 50 years, providing more accurate and uniform parameter information. As the seismic intensity-attenuation relationship is traditionally used to estimate the intensity distribution, the macroseismic intensity-attenuation relationship for mainland China was fitted by the earthquake data collected in this region. The deviation of the intensity assessment by the macroseismic intensity-attenuation relationship was examined for 43 earthquakes (Ms ≥ 6.0). In addition, seismic damage emergency assessment work requires the isoseismal lines to be constantly modified according to the updated information. Therefore, an improved ellipse intensity-attenuation model was proposed in this study, completed by the establishment of a semimajor axis and semiminor axis length matrix. Based on the initial value of the length matrix obtained by the regression of historical data and survey data from the site, the least mean squares (LMS) algorithm is used to revise the length matrix. In the end, the practicability of this method is verified by a case study of the Lijiang 7.0 earthquake.

Automatic methods for generating seismic intensity maps

For many years the modified Mercalli (MM) scale has been used to describe earthquake damage and effects observed at scattered locations. In the next stage of an analysis involving MM data, isoseismal lines based on the observations have been added to maps by hand, i.e. subjectively. However a few objective methods have been proposed (by e.g. De Rubeis et al., Brillinger, Wald et al. and Pettenati et al.). The work presented here develops objective methods further. In particular the ordinal character of the MM scale is specifically taken into account. Numerical smoothing is basic to the approach and methods involving splines, local polynomial regression and wavelets are illustrated. The approach also allows the inclusion of explanatory variables, for example site effects. The procedure is implemented for data from the 17 October 1989 Loma Prieta earthquake.

Automatic methods for generating seismic intensity maps

For many years the modified Mercalli (MM) scale has been used to describe earthquake damage and effects observed at scattered locations. In the next stage of an analysis involving MM data, isoseismal lines based on the observations have been added to maps by hand, i.e. subjectively. However a few objective methods have been proposed (by e.g. De Rubeis et al., Brillinger, Wald et al. and Pettenati et al.). The work presented here develops objective methods further. In particular the ordinal character of the MM scale is specifically taken into account. Numerical smoothing is basic to the approach and methods involving splines, local polynomial regression and wavelets are illustrated. The approach also allows the inclusion of explanatory variables, for example site effects. The procedure is implemented for data from the 17 October 1989 Loma Prieta earthquake.

Parameters estimation of intensity decay relationships

This paper presents a methodology that analyses a set of observed intensities and estimates the parameters of an adopted attenuation law directly using the data points. A procedure was developed to define and evaluate the equivalent radii Di of the isoseismal lines. From these data it is possible to derive the parameters of the attenuation law. Moreover a validation procedure was developed to measure the capability of intensity decay relationships to reproduce the observed intensities. A case study of 55 earthquakes, divided into 9 subsets, of similar-attenuation zones, was analysed, using, as attenuation law, the one proposed by Grandori (1987, l991) to estimate either the parameters for each single earthquake or the parameters of an average intensity decay relationship for the similar-attenuation zones. The calculated intensity decay relationships result in 60- 70% of correctly reproduced points for most intensity data maps analysed. Analysing the similar-attenuation zones and different earthquakes simultaneously, the parameters of attenuation laws obtain results with a lower percentage of correctly reproduced points. The proposed methodology seems to be effective and suitable to reach practical results in parameters estimation of intensity decay relationships.

A case of the influence of radiation pattern on peak accelerations

The strong ground shaking of the 23 November 1980 earthquake in southern Italy seems to have been conditioned by the dimension of the source, its focal mechanism, and by the distance from the shallow portion of the source. There was only a low, and doubtful, directivity effect. These results come from a comparison of the azimuthal distribution of the recorded peak ground horizontal accelerations with that of the total, dimensionless, radiation pattern of S waves in the horizontal plane at each site (radiation from the closest point of the fault, and appropriate azimuth and take-off angles were considered). The recorded maxima were obtained from hodogram plots of each couple of automatically digitized horizontal components in 13 stations with negligible local site effects at a distance of up to 78 km from the epicenter. The analysis indicates the strong influence of the strike-slip component on the azimuthal distribution of motion. The fault mechanism best fitting the recorded maxima is as follows: strike 318°, dip 64°, rake 317°. This picture does not change if acceleration maxima in the frequency bands 0.1 to 5 Hz, 1 to 5 Hz, or 1 to 2 Hz are used. In a segment of the southern Apennines, where the strong-motion energy radiation in the near/intermediate field of a repetitive series of shocks from the seventeenth century up to 1980 seems to be controlled by the gross features of the source, it could be useful to include radiation patterns into algorithms for regional seismic hazard calculations. Conversely, because of the fact that drawing isoseismal lines results in a smoothing of at least the very local site effects, it might be possible to infer information about the gross features of the sources of historical earthquakes from macroseismic data.

A macroseismic study and the implications of structural damage of two recent major earthquakes in the Jordan Rift

abstract A detailed macroseismic study of the July 11, 1927 earthquake was carried out. A quantitative analysis of damage data provided a correlation for estimating (MM) intensities: I = 6.4 + 1.2 log (percentage of damaged houses). Using axis lengths and areas bounded by the ensuing isoseismal lines, the depth (16 to 28 km) of the event was estimated, and its probable epicenter located near Damiya bridge on the Jordan river. A comparison with the equivalent parameters, inferred from instrumental records, shows agreement between both sets of results and thus confirms the validity of the approach used in the macroseismic study. The same procedure of studying macroseismic data was applied to an earlier (January 1, 1837) destructive earthquake. It was found that this latter event orginated in the upper crust, eastward of Safed, with a 6.25 to 6.5 magnitude. The isoseismals of both these major earthquakes are elongated in a north-south direction, along the major structural trend in the area. The southern coastal plain of Israel seems generally less vulnerable to Jordan Rift Valley earthquakes than inland regions of similar epicentral distances, though local pockets of anomalous intensities are observed for both earthquakes.

IV. On the Leicester earthquake of August 4, 1893

On August 4, 1893, at 6.41 p. m., an earthquake of intensity nearly equal to 6 (according to the Rossi-Forel scale) was felt over the whole of Leicestershire and Rutland and in parts of all the adjoining counties. The disturbed area was 58 miles long, 46 miles broad, and contained an area of about 2066 square miles. The direction of the longer axis (about W. 40° N. and E. 40° S.) and the relative position of the isoseismal lines show that the originating fault, if the earthquake were due to fault-slipping, must run in about the direction indicated, passing between Woodhouse Eaves and Markfield, and heading towards the north-east.