Abstract:
"Pend S-wave travel times from shallow earthquakes with epicenters
in the northwestern, southwestern and southern directions free, India as
recorded hy Indian ohservatorles have heen studied with the objective of
determining the regional travel-time curves and the •«.!. velocity
structure. The travel time data for the various directions have heen
fitted to anumber of plausible 11.. segments on ahigh speed computer,
using anew statistical method developed by Kaila and Narain (1970).
P.travel time curves thus determined in the northwestern, south-
western and southern directions fr- India show considerable departures
from the Jeffreys-Bullen (1940) travel time tables although they agree
quite well with those of Herrin et al (1968). S-travel time curves in
these three directions also reveal some differences with respect to
the d-B (1940) travel time curves. In the northwestern direction, ^
changes in the slope of the *-travel tta. curves occur at 21°, 33°, 49 ,
54° and 67° corresponding to velocity discontinuities in the mantle at
depths of 590.10, 860±20, 1300.20, 1420±5O and 1880.130 km below the
crust, true velocities underneath them being respectively 10.95, 11.67,
U.JO. 12.48 and 12.97 km/sec. In the southwestern direction,^changes
ln the slope of the r-travel time curves occur at 17°, 24°, 41 , 51
and 64° corresponding to velocity discontinuities at 330.10, 730.20,
,50.20, 1400.40 and 1860.60 km and with true velocities below them
being respectively 9.18, U.49, 11.62, 12.44 and 13.08 Wsec.
It was not possible to determine the first segment of the travel time
curve in the southern direction owing to non-availability of data in
that epicentral distance range. However, subsequent changes in the slope
of the P-travel time curves beginning from the second segment were
observed at A = 25 , 39 and 49.. Assuming the velocity below the
Mohorovicic discontinuity to be the same in this case as that obtained
in the southwestern direction, velocity discontinuities were then computed
along the southern direction. These occur at depths of 380+10, 760+20,
940+20 and 1230+30 km below the crust and the true velocities underneath
them were found to be 9.54, 11.53, 11.72 and 12.11 km/sec. The upper
mantle velocities immediately below the Mohorovicic discontinuity were
found to be 8.44 km/sec for P and 4.58 km/sec for S in the northwestern
and 8.24 km/sec for P and 4.56 km/sec for S in the southwestern directions.
These upper mantle velocities are considerably higher than those used in
the 1940 Jeffreys-Bullen tables (7.90 km/sec for P-waves and 4.29 km/sec
for S-waves for A between 2 to 10 ). Similar high upper mantle
velocities were also obtained In the Hindukush region from the deep
earthquakes data using a new analytical method developed by Kaila (1969a).
Jeffreys-Bullen residuals for P-waves In the southwestern and
southern directions reveal finest an identical picture which is quite
different from that in the northwestern direction. In the northwestern
direction, beyond A = 2000 km, a majority of the residuals are negative
with an average shift of about 3 to 4 seconds with respect to the J-B
travel time curves. This is quite the opposite of what was observed
in the northernly direction from India where these residuals are found
to be positive with an average excess value of about 4 seconds.
The slopes of S versus P-residual curves in the three directions from
India are found to vary from 1,8 to 2.5. This, in other words, means
that the ratio of the P and S velocities in the mantle in these direct
ions is not constant and equal to 1.8 as found by many workers for
crustal layers. The study of J-B residuals reveal lateral variations
in the mantle velocity structure in the Indian subcontinent.
The velocity discontinuities in the mantle in the three directions
from India occur at varying depths. If an average picture of the mantle
velocity structure in the three directions is taken, the first velocity
discontinuity is found to occur between 360 to 410 km depth from the
surface of the earth (except in the northwestern direction where no
such discontinuity is observed). The second discontinuity is found at
depths between 630 to 790 km followed by a third discontinuity lying at
a depth of 390 to 980 km. The fourth and the fifth discontinuities lie
at depths of 1330 to 1430 km and at about 1900 km respectively. As the
depth ranges of these discontinuities are much larger than three times
the standard deviation of their depth values, it is concluded that the
depths of these discontinuities do in reality vary in different
directions.
Furthermore, the P-wave amplitudes from nuclear explosions in the
o o
epicentral distance range of 1 to 98 , which could be satisfactorily
represented by discontinuous curves also support the idea of velocity
discontinuities in the mantle."