We study the source time function (STF) and radiated seismic energy (ER) of the Mw 7.6 Bhuj earthquake using the empirical Green's function (EGF) technique. Our estimations of the STF and ER are based on teleseismic P waves and regional seismograms, respectively. We find that the STFs as a function of azimuth have a similar shape and nearly constant duration of 18 sec. This suggests that the rupture propagation was essentially radial. The STFs show a sharp rise in the first 6 sec. The ER estimated from the EGF technique is 2.1 × 1023 erg. We find that ER's computed from integration of corrected velocity-squared spectra of teleseismic P waves and regional seismograms are in excellent agreement with the ER obtained from the EGF technique. Since the seismic moment, M0, is 3.4 × 1027 dyne cm, we obtain ER/M0 = 6.2 × 10-5. The radiation efficiency, νR, during the Bhuj earthquake was low, about 0.23. The sharp rise of the STFs and νR = 0.23 can be explained by Sato and Hirasawa's (1973) quasi-dynamic, circular source model with an effective stress of ∼ 300 bar and the ratio of rupture to shear-wave velocity, VR/b,of ∼ 0.5. The corresponding estimate of slip velocity at the center of the fault is 156 cm/sec. VR/b ∼ 0.5 is in reasonable agreement with the duration of the STF and the reported dimension of the aftershocks, as well as with the results of inversion of teleseismic body waves.
The observations may also be explained by a frictional sliding model, with gradual frictional stress drop and significant dissipation of energy on the fault plane. This model requires an average dynamic stress drop of about 120 bar and VR/b ∼ 0.7 to explain both the rapid rise in the first 6 sec of the STFs and, along with a static stress drop of 200 bar, the observed ER/M0. High static stress drop is a common feature of most crustal earthquakes in stable continental regions. An examination of the available data, however, does not suggest that most of them also have relatively low radiation efficiency.