Dix - 55 - Seismic Velocities From Surface Measurements

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 15

SEISMIC VELOCITIES FROM

SURFACE MEASUREMENTS
C. Hewitt Dix 1955 Geophysics, 20(1), 68-86.
Seismic Imaging and modelling
presented by
Zacharia SHITAKWA HOIDI_16770632

Contents

Introduction
Estimation of seismic interval velocity
Basic Assumptions
Field techniques
Generalization
Dix equation
Other factors to consider
Summary and conclusions
Introduction
The purpose of this paper, Dix-1955, is the discussion
of field and interpretive techniques that can be
used to accurately determine seismic interval
velocity before the commencement of drilling.
Accurate velocity determination will lead to
accurate structural interpretation and consequently
gain some lithological information.
Dix formulation is closer to practical exploration
scenario and though laborious it is sound.
It involves interpretation by successive
determination of interval velocities from top layer to
bottom.



Estimation of Seismic Interval Velocities

Seismic interval velocity is the average propagation
velocity through a depth or time interval (V
int
= Z
i
/T
i
).

It is estimated from root-mean-square velocities (V
rms
)
through the called Expanding Spread Technique (EST)
Average velocity is the average of all the interval velocities
from the surface to the depth of a particular horizon.

Used with small angle approximations in the moveout
formula Vrms closely predict travel times for any offset
For a single layer, it is apparent that V
nmo
and V
avg
are equal.
Variations from the model may cause severe problems,
example dipping beds, and rapid lateral velocity changes
within a layer can affect seismic velocity calculations.

Basic assumptions

1. Multiple horizontal layer medium of constant velocity
known as root-mean-square velocity (Vrms)

2. Travel times related to nearly vertical incidence/
straight line ray paths t
i
is one-way vertical travel-
time through bed i


3. Source-receiver distances must be kept small
relative to the distances to reflecting interface
Thus for short distances, we can state that V
rms
= V
nmo
,
provided there is no dip on the beds.
4. Solution of the Dix equation assumes that the zero-
offset ray paths to the (n-1)th and nth reflectors
follow a common path.

Field technique
This is a natural extension of
continous profiling technique.
Done using the shot
combination alongside:1
st
shot
regular symmetric split, the
second linked to the first and
the third to the second.
The result is a continuous tied
range of distance x from O
and the separation between
S.P 3 and 4.
The tie link of each new record
is through centre point B.

Generalization
First layer
and
Where T
x
, is the reflection time at x and V is the
layer velocity. z is the depth.
A plot of X
2
-T
x
2
, will give a slope 1/V
2
with T
0
as the
intercept.
Consider equation of a straight line y=mx +c
Second Layer
If we plot X
2
-T
x
2
, for layer 2 we obtain a hyperbola
curving towards small values of T
x
.

1
Generalization




Figure 1 Figure 2
We consider minimum distance and time paths
A true ray path will follow a minimum time path (SGDHF) which
means that SG shorter GD longer
The solution of traveltime (ABCDE) in a multilayer model produces
and equation of the infinite series form T
x
2
=A+BX
2
+CX
4
+...
where A is equal to T
o
2

B obtained from the above series by differentiation of the eqn
w.r.t X
2




Generalization
Second layer_approach 1
We need Using it, we can determine

1
which we need for determining V
2
. The
derivative would give an uncertainity in the
range of sin
1
.
We use the tangent LM to the T
x
2
-x
2
at x
1.
The eqn.
of
Differentiating the tangent equation w.r.t x at x
1

gives the derivative above which we manupilate
to get

2
Generalization
Second layer
From
1
, SG and FH can be computed together
with the time to be removed from T (travel time
through the 1st layer) to obtain the time from G to
D to H. This gives T
xR

Similarly, from
1
, 2AG can be computed, and
this can be subtracted from x to give x
R
.
Plotting (T
x
)
R
2
(x)
R
2
we obtain a straight line
whose eqn.
The gradient of the line gives V
2
and the
intercept gives the depth time (T
0
)
R

The thickness is given as
3
4

Approach 2-Dixs equation
ABC and CDE we can
define tan
1
nd tan
2
as

Tan
1
=

1
2
and
Tan
2
=

2
2

z
1
=

1
2
and z
2
=

2
2

Using the right angle triangle theorem we arrive at
5
Dix equation
Divide through by x
1
+x
2
, and taking the limit x
1
+x
2
0
then multiplying by we obtain the eqn.

The above equation is valid only for x values near
zero,
For n layers, we can generalize equation 6 above as
and
If we get the difference in eqns. 7 we obtain


Note that and therefore

6
7
8
1
V V

n n
A A
n
A n
V V

Multiples reflections
Every reflector produces multiples, we however
easily pick out those multiples that result from the
ground-air interface and weathered layer- high
velocity interface.
Must be identified and excluded from the data for
accurate velocity determination
Dipping Layers
For a single dipping layer the slope of T
x
2
X
2
graph
is given as
For layer 2 parameters determined by removing the
effects of the first layer
Curvature
Effects of curvature are most difficult to correct, in
practice graphyical methods are employed.










Other factors to consider
Summary and Conclusions

Dix equation calculates interval velocities from
average velocities assuming vertical incidence in
horizontal layers and small source-receiver
distances relative to the distances to the reflective
interfaces.
Use x
2
-T
2
plot to determine rms velocities to each
layer to get interval layer velocities and thicknesses.
If the geology of an area is known, seismic velocities
may be more easily interpreted.
RMS velocity >= Average velocity.
Further calculations based on rms velocity for
arbitrary and nonuniform dipping lead to NMO
velocity.
THANK YOU FOR YOUR ATTENTION

You might also like