are steep in the second subevent resembling it to be nearly a
pure strike-slip event, whereas the north dipping nodal
plane of the first subevent is shallow (Fig. 7c). Moreover,
in the second subevent the compression direction is rotated
to the first quadrant, however, the position of T-axis in
both subevents does not change significantly. Since the second
subevent is purely strike slip event it could not produce
strong amplitudes at teleseismic distances (Lay and Wallace,
1995) as evidenced in broadband seismograms too.
Considering the time lag between the two subevents signature
of pP phase (if any) due to the second subevent might
have been concealed by the noise in the later parts of the
seismogram. Further the total mechanism solution looks
very similar to the first subevent and is consistent with
those published earlier. In Table 2 we present the result
obtained in this study and compare our solution with
already published results.
3.2. Slip model estimation
To estimate the slip model, we first compare the slip vectors
of the two nodal planes with the convergence vector of
India with respect to Eurasia and then choose one nodal
plane which is consistent with the convergence vector as
the fault plane. From NUVEL-1 (DeMets et al., 1990)
the convergence vector of India at this location is estimated
51.6 mm/yr along N14.6E. Slip vector of the northerly
dipping nodal plane (219, 41, 4) is consistent with the convergence
vector, so we assumed this nodal plane to be the
fault plane and run the inversion in this fault plane as
described in the following paragraph.
We choose total rupture length of 39 km and width of
21 km assuming a rupture velocity of 2.8 km/s and total
rupture time of 14 s. We then divided the rectangular
fault into a matrix of size 13 · 7 where each sub-fault has
a dimension of 3 km · 3 km and each point source is
located at the center of the sub-fault. We assumed an isosceles
triangle with half duration of 3 s as the source time
function and placed the source fault at the best estimated
hypocentral depth of 46 km during the point source estimation.
Total number isosceles triangle used in inversion is 3.
For the estimation of slip at each sub-fault we assumed the
source region a Poisson solid with rigidity of 36 GPa.
Inversion results revealed a slip model that is consistent
with the point source estimation. The estimated slip model
is presented in Fig. 8a where the upper panel shows the