The Jabalpur earthquake (23.08o N, 80.06o E, H = 36 km) is the first event in the Indian Peninsular shield region to be well recorded by a newly installed, 10-station, broadband seismographic network. Using these data, we estimate Q of Lg wave in the Indian shield region as Q = 508f 0.48 (1 f 20 Hz). The corrected source spectrum, with M0 = 5.4 x 1024 dyne-cm (reported in the Harvard CMT catalog) and an 2-source model, requires a stress parameter, p, of ~420 bar to explain high-frequency spectral level. The computed seismic energy from the records is 7.4 x 1020 erg, which yields an apparent stress and Brune stress drop of 62 and 270 bars, respectively.
The analysis of the Jabalpur earthquake provides some elements for the estimation of ground motions during future earthquakes in the Indian shield region. Based on the results of the Jabalpur earthquake and on studies of tectonically similar region of eastern North America, we assume that (1) the sources follow an 2 model; (2) S waves dominate at R < 100 km, and Lg waves dominate at R 100 km; (3) Q(f ) = 508f 0.48; (4) the ground motion is a bandlimited, finite-duration, Gaussian white noise; and (5) the effective duration of the ground motion equals fc−1 + 0.05R, where fc is the corner frequency. We apply random vibration theory (RVT) to compute various measures of ground motion, such as Amax and Vmax. At near-source distances, the source finiteness is approximately taken into account. The attenuation curves for 5.5 Mw 7 and for p of 100-400 bar are presented. As expected, the predicted values (with p ~ 420 bar) agree reasonably well with the limited Jabalpur data. An Amax of ~150 gal is predicted in the epicentral region of this earthquake. The predicted curves imply p 100 bar for the Latur earthquake of 1993 (Mw = 6.1; H = 2.5 km) to explain Amax < 1 g and the reported isoseismal intensities in the epicentral region. For Koyna earthquake of 1967 (Mw 6.3), the inferred Amax and Vmax from isoseismal intensities and the recorded strong motions at the Koyna dam site are in agreement with the prediction curves for p ~ 100 bar. The RVT predictions seem reasonable but need validation from more strong-motion data, which is presently lacking.