Seismic and Seismology

Final report
Nicolas Bedoya Jauregui 201413649
Luis Armando Blanco Bohórquez 201223834
Sergio Esteban Silva Sánchez 201414836
Carlos David Valencia Pacheco 201215300

Abstract
Data seismic was acquired from a field trip during the course of Seismic and Seismology
of the University of Los Andes. This data was aquired to understand the differente ways of
acquiring reflection and refraction waves. At the same time, a data processing was mae using
data from CREWES consortium (Consortium for Research in Elastic Wave Exploration
Seismology); 4 different exercises were made in order to understand the way of seismic data
procesing. The first exercise is an introduction to the modules and commands of CREWES,
how to read *.segy files and plot seismic data. The second exercise gives an introduction to data
sorting in its possibilities (Common Receiver, CMP and Common Offset sorting). Exercise 3
gives an understanding and development of velocity analysis, Normal Moveout corrections and
stacking velocities. Finally exercise 4 seismic data is migrated creating velocity models and
using Kirchhoff migration which uses the integral form of the wave equation to carry out the
process.

Introduction
Seismic studies are a fundamental tool in Geosciences investigation, this research method provides
the possibility of knowing the structure, characteristics and arrangement of the different layers
located in the subsurface. All this, based on the behavior of acoustic waves produced by a source,
which can be an explosion, blows of hammers, among others. That could be reflected or refracted
with a change of impedance between the different layers, causing changes in the travel - time path
of the waves, that are received by arrays of geophones or sensors. This process, allow us to know
in a better way about the subsurface structure, what can be used in academic or industrial finals,
like in petroleum exploration.

There are two main ways to analyze seismic data: seismic refraction and seismic reflection.
The first one, is based on the observation of arrival times of the first movements of the terrain in
variable points, generated, how it was mentioned, by a source of energy. In this way, the set of data
obtained in the surveys consist in a series of time vs distance, that are interpreted in terms of depth
at interfaces between layers and in terms of the propagation velocities of the P wave in each layer.
Where the first curve that appear and that have a origin in the point 0,0 is the direct wave, and
the other travel-time curves that appear with different slopes are refracted waves. These velocities

Page 1 of 38
 Seismic & Seismology

are controlled by the elastic parameters describing the material. The principal problems that this
kind of survey could have is that: first, requires larger arrays with a huge source of energy if the are
really depths surveys, and, only is functional if the velocity increases with depth. If it decreases,
the refracted wave will not return to the surface to be measured. It is important to note that there
is a critical angle where the wave start to travel in the layers interface after the refraction, and it
is described by:
v1
sin θc =
v2
On the other hand, in seismic reflection surveys the analysis is based on the energy of the vibra-
tions after the movement of the soil have started. Specifically the target is to measure the terrain
movements induced by the reflection of waves in the interfaces of layers, that are generated in a
specific point. In this way, there is studied information related to the amplitude and shape of the
wave path. In this kind of surveys, it s needed to process in a huge way the different information
obtained through a corrections, like normal-moveout correction, stacking, among others.

In this field report, it will be showed the methodology used in a seismic survey made in the
Sabana of Bogotá. Like the data collected have different mistakes, it will be used a example of
CREWES data to show the different processing that it is necessary to use after a survey. Even so,
it will be enunciated the methodology used during the measurement of data in the field and the
geological setting of the area.

Important definitions
The following definitions were taken from the Encyclopedic dictionary of applied geophysics (Sheriff,
2002).

• Common source gather: A display of seismic traces that share a source.

• Common midpoint (CMP): In multichannel seismic acquisition, the point on the surface
halfway between the source and receiver that is shared by numerous source-receiver pairs.

• Common offset: Pertaining to traces that have the same offset, or distance between source
and receiver.

• Segy: Standard file format developed by the Society of Exploration Geophysicists (SEG) for
storing geophysical data.

• Kirchhoff migration: Backpropagation (or continuation) of the seismic wavefield from the
region where it was measured.

• Seismic trace: The seismic data recorded for one channel.

• Seismic Sorting: Is the way to make the coordinate transformation required into CMP
gathers.

• Header: The location, acquisition and processing parameters, and other pertinent informa-
tion attached to a well log, seismic record and traces.

• Common receiver positions: A display of seismic traces that share a receiver.

Page 2 of 38
 Seismic & Seismology

• Fold: A measure of the redundancy of common midpoint seismic data, equal to the number
of offset receivers that record a given data point or in a given bin and are added during
stacking to produce a single trace.

• Normal moveout correction: The effect of the separation between receiver and source on
the arrival time of a reflection that does not dip, abbreviated NMO.

• Stacking velocity: The distance-time relationship determined from analysis of normal move-
out (NMO) measurements from common depth point gathers of seismic data.

• Mute: To remove the contribution of selected seismic traces in a stack to minimize air waves,
ground roll and other early-arriving noise.

Geologic setting
The Sabana of Bogotá is conformed by a complex structural system composed of failed SSW-NNE
course synclines. In all this region sediments that date from the Middle Cretaceous until the
Quaternary arise. To this period of time belong: the Villeta and the Guadalupe Group and the
formations: Guaduas, Cacho, Bogotá, Regadera, Usme, Tilatá, Sabana and Tunjuelo.

On the other hand, the Sabana is characterized by the presence of lagoon deposits, fluvial and
glacial-alluvial deposits. These have been deposited in the center and the edges of the basin during
the last 3.5 million years. To this group belong the Tilatá (Pliocene), Sabana (Pliocene, Pleistocene
and Holocene) Formations, and Tunjuelo (Pleistocene and Holocene) Formations.

In the greater valleys there are long synclines and in the part of the mountainous hills anticlines
narrow. All these structures present inverse faults that mainly border the anticlines. Within these
structures, the Bogotá Fault is observed, which, despite its importance, is not studied.

Of the most characteristic formations of the region is the Guadalupe Formation which is a
discontinuous regional aquifer with a medium permeability and a high saturation scatter. This
training serves as a water supply, which is of a medium-good quality.Like the previous one, the
Tilatá, Sabana and Tunjuelo Formations contain continuous, semi-confined, unconsolidated regional
aquifers; With great thicknesses of saturation and good quality of the water.

Figure 1: Geology of the Sabana de Bogotá.

Page 3 of 38
 Seismic & Seismology

As we see this region has a great hydrological importance, which is exploited by thousands of
wells that currently generate severe problems. Currently, the potentiometric level of the ground-
water of this basin is descending at rates between 3 and 5 m / year. The drop in the pressure level
is one of the main causes of the disturbing phenomena of the region.

Methodology
The field trip was made to a place located between Cota and Suba, specifically at the coordinates
4.77152, −74.12544. The location did not have topographic changes and there was a straight road,
which allowed a lot the data acquisition. The array chosen consisted of 50 geophones located with
a distance of 10 m each, and two different sources that were located with a distance of 20 m each
taking into account a distance of 1 m with the nearest geophone (Figure 3).

Figure 2: a) Main base. b) Position of the geophones throughout the way. c) Photo of a geophone.

Figure 3: Basic array with 50 stations separated to 10 m and a source separated to 1 m of the first
station and then put to 20 m of separation.

The array made in the field for the acquisition was common shot gather, this is used when the
traces of the gather come from a single shot and many receivers. Based on the data acquisition, we
could find that with one explosion there was approximately 4 - 5 geophones that register the wave.
Each geophone has a box with an antenna (Figure 2), that communicates it with the base (Figure
2). In specific this boxes were on if the user put it face up, with the antenna pointing towards
itself, and it could be off putting it face down parallel to the ground. In such a way that with each
detonation the data of all the geophones was recorded and stored in the main computer of the base.

Page 4 of 38
 Seismic & Seismology

There were two different sources of energy in the field:

a) Hammer and plate
This method is one of the most common sources for geophysicists in shallow depth applications
(geophysicists in shallow depth applications). It is cheaper than other types of sources, it does
not required so much logistic and it’s acquisition of the data is faster than others methods. One
of the most important factors of this method is the weight of the hammer, in this case we use a
10 Kg hammer, the small weight of the hammer will affect the energy content of the pulse and
increase in that way the need for vertical stacking. The other important factor is the plates that
are constructed in different sizes by various enduring materials. Some consideration that need
to be taken for a production of high frequency signal are: the light and the surface wide of the
plate, it must be light enough in order to accelerate after the strike (according to the momentum
conservation principle) and must be wide enough, to reduce its plastic penetration in the ground.
In the fieldwork conditions, the hammer and plate system, in order to act as a seismic source,
normally needs a vertical stacking of at least 5 repeating shots.

Figure 4: photo of the processing for generating the waves. Hitting the plate with a hammer.

Explosives
The explosive materials as seismic sources in shallow reflection seismic are mostly based on the
explosion of a small quantity of seismic dynamite (Atanackov, 2013), in this study it was used
black gunpowder cartridges. Some studies (Ziolkowski & Lerwill, 1979), have demonstrate that
the resolution was clearly better after the use of the small quantities of explosives. Generally, the
explosive charges are placed between 6 and 76 metres below ground, in a hole that is drilled with
dedicated drilling equipment for this purpose. However, the configuration that was used allowed
to locate the charge 45 cm above the surface, it was detonated by a remote control and the record
of information started seconds after the detonation time.

Page 5 of 38
 Seismic & Seismology

Day 1

Figure 5: photo

In this day, it was made a survey of 500 m, at the coordinates:

• Initial point: 4.77052o , -74.125226o .

• Final point: 4.766071o , -74.124283o .

During the measurement there were used the two sources already mentioned: hammer and
explosives. But, to achieve a higher resolution in the data it was changed to only use explosives.
It is important to take into account that by problems in the connection of the geophones with the
base we did not do more seismic lines of survey.

Day 2

Figure 6: Path oh the survey day 2. IP: Initial point, FP: Final point. In could be represented the
3 lines made during the survey.

In this day, the seismic line of acquisition was continued. The start location of the measure was
taken in the point located to 400 m (geophone 40) of the first day start location with the aim of
make an interpolation of the data taken along each measurement line. When this line was finished,
there was two translations more of the geophones to make new measurements: one with a length
of 500 m and the second one with a length of 250 m. In this way, there was made a total line with
a length of 1050 m approximately (Figure 6). The important points in this survey are located in:

• Initial point 1: 4.766950o , -74.124447o .

• Final point 1: 4.762520o , -74.123525o .

• Initial point 2: 4.763399o , -74.123704o .

Page 6 of 38
 Seismic & Seismology

• Final point 2: 4.758979o , -74.122817o .

• Initial point 3: 4.759832o , -74.123010o .
• Final point 3: 4.757583o , -74.122585o .

The lines 1 and 2 were made with explosives, while the line 3 was made with hummer.

Crewes work
The following shows the code for the three exercises of Crewes implemented in a MATLAB code,
the solution and their different results.

Exercise 1b - Reading and plotting seismic data using CREWES software

The following exercises gives the specific commands for using some software modules, developed by
the CREWES consortium (Consortium for Research in Elastic Wave Exploration Seismology).
The first thing is to use the module plotseis, where the input needed is the data in a *.segy
file (this is done by using the module readsegy). The following code gives solution to this first
exercise, part b, using the previous commands and modules and adjusting the graphs at will shown
below.

1 %Universidad de Los Andes

2 %Departamento de Geociencias
3 %Sı́smica & Sismologı́a
4 %Proyecto Final
5 %Sergio Silva - Carlos Valencia - Luis Blanco - Nicolas Bedoya
6 %---------------
7 clc; % Limpia el tablero de comandos.
8 clear all; %Borra todas las variables almacendas.
9

10 addpath(genpath('intro_seismology/crewes'))
11 addpath(genpath('intro_seismology/system'))
12

13 % Exercise 1b
14 %
15 FP=readsegy('data/shotground.segy'); %Lee el archivo SEGY
16 tama=size(FP)% Encuentra el tama~
no de la matriz
17

18 figure1=figure%Genera figura
19 plotseis(FP)% Utiliza el comando plotseis para graficar la matriz
20

21 figure3=figure% genera figura

22 plotseis(FP(150:250,20:40))%Delimita en los intervalos en x y y de la matriz
23

24 figure3=figure%Genera figura
25 plot(FP(:,1));%grafı́ca unicamente una traza sismica.
26

27 figure4=figure

Page 7 of 38
 Seismic & Seismology

28 % dt=0.004;
29 % tps=(0:size(FP,1)-1)*dt;
30 %Se redefinieron los lı́mites x y y de acuerdo a la distancia y tiempo.
31 plotseis(FP,0.004:0.004:512*0.004,100:100:109*100)
32 %Nombre del eje y - tiempo
33 ylabel('time (s)','Interpreter','latex','fontsize',16);
34 %Nombre del eje x - distancia
35 xlabel('Source Geophone Distance (m)', 'Interpreter','latex','fontsize',16);
36 %Imagen de las trazas
37 plotimage(FP,0.004:0.004:512*0.004,100:100:109*100)
38 %Nombre del eje y- tiempo
39 ylabel('time (s)','Interpreter','latex','fontsize',16);
40 %nombre del eje x - distancia
41 xlabel('Source Geophone Distance (m)','Interpreter','latex','fontsize',16);

Figure 7: Plotseis of the shotground.segy.

Page 8 of 38
 Seismic & Seismology

Figure 8: Plotseis from time samples from 150 to 250 and traces from 20 to 40.

Figure 9: Plot of one trace of shotground.segy.

Page 9 of 38
 Seismic & Seismology

Figure 10: Plotseis of shotground.segy with vertical axes in time (s) and horizontal axis in meters.

Figure 11: Plot image of shotground.segy with vertical axes in time (s) and horizontal axis in
meters.

Exercise 2 - Sorting of Seismic Data

The following exercise, explains the importance of sorting, and its different possibilities. Data
sorting takes place based on information that is usually given in the SEGY header. In this header
the positions of the sources and geophones are stored. When using headers in combination with
the data, separate headers need to be defined. Each trace will have one header. Data sorting can
take different possibilities, the ones shown in this exercise are:

• Sorting to Common Receiver Positions (CRP’s): Data is sorted so one gets all the shots
together in an increasing shot position, beloning to one receiver position.

Page 10 of 38
 Seismic & Seismology

• Sorting to Common Midpoint Positions (CMP’s): this sorting is based on the field geometry
information, each individual trace is assigned to the midpoint between the shot and receiver
locations associated with that trace and those traces with the same midpoint location are
grouped together, making up a CMP gather.

• Sorting to Common Offsets: This is a gather in which we collect all those source-receiver
pairs that have a certain offset in common. If we have a fized distance between source and
receiver we will have as many traces in our common-offset gather as there are shots.

New commands used in this section are: sorthdr, plothdr, analysefold and sortdata.
Plots are shown below.

1 %% EXERCISE 2
2

3 H_sht=importdata('data/H_sht.mat');%Se importa la matriz del header

4

5 plothdr(H_sht)%Graficar la matriz cargada.

6

7 %Sorting Common Receiver Positions (1)

8

9 H_CR=sorthdr(H_sht,4); % Se organizo respecto a la posición de common receiver

10 plothdr(H_CR)% Grafica la matriz sorteada anteriormente
11

12 %Sorting CMP (2)

13 H_CMP=sorthdr(H_sht,5);% Se organizo respecto a la posición de common Midpoint
14 plothdr(H_CMP)% Grafica la matriz sorteada anteriormente
15

16 %(3)
17 H_sht_An_CMP=analysefold(H_sht, 5); %Analiza el numero de trazas por CMP.
18

19 %Sorting to Common Offsets (4)

20 H_CO=sorthdr(H_sht,2); % Se organizo respecto a la posición de common receiver
21 plothdr(H_CO)% Grafica la matriz sorteada anteriormente
22

23 %(5)
24 [Head_pos, Head_folds]=analysefold(H_CO, 2); %Analiza numero de trazas por CMO
25

26 %(6)
27 clear('H_CMP','H_CO','H_CR','H_sht_An_COF','H_sht_An_CMP','H_sht');
28 %Sorting data and headers
29 [seis, Hea_Tripli, geo_tripli]=segyread('intro_seismology/data/tripli.segy');
30

31 %Cargamos el arhivo segy Tripli con el header y el data

32 Figure_P_Select =figure
33 [P_CMPgather_Tripli, P_H_CMPgather_Tripli]=selectcmp(seis,Hea_Tripli, 150);
34

35 plothdr(Hea_Tripli)
36 H_tripli_An_CMP_8=analysefold(Hea_Tripli, 5)

Page 11 of 38
 Seismic & Seismology

37

38 %Sorteamos el data de Tripli incial, no el del shot seleccionado

39 [Tr_Data_CMP,Tr_H_SRT]=sortdata(seis, Hea_Tripli, 5);
40

41 Figure_S_Select=figure
42 [S_CMPgather_Tripli,S_H_CMPgather_Tripli]=selectcmp(Tr_Data_CMP,Tr_H_SRT, 1500);
43 %Volvemos a seleccionar un gather y este mismo plotea
44

45 [Tr_Data_SRT_CO,Tr_H_SRT_CO]=sortdata(seis, Hea_Tripli, 2);

46 %Sorteamos el Data a Common Offset
47

48 %Se plotea el header de los datos sorteados al Common Offset

49 plothdr(Tr_H_SRT_CO);

Figure 12: Header plot of data hs ht.mat.

Page 12 of 38
 Seismic & Seismology

Figure 13: Header plot product of sorthdr to a common receiver.

Figure 14: Header plot product of sorthdr to a common midpoint.

Page 13 of 38
 Seismic & Seismology

Figure 15: CMP fold get by analysis fold, this for the position of gathers such as CMP’s gathers.

Figure 16: Analysefold offset using the output of function analysefold.

Page 14 of 38
 Seismic & Seismology

Figure 17: CMP gather for midpoint. Plot apart of the data output of sortdata.

Figure 18: Header plot based on the tripli data header.

Page 15 of 38
 Seismic & Seismology

Figure 19: Plot middle CMP.

Figure 20: Plot header of the plot hdr based on the sortdata into a common offset.

Page 16 of 38
 Seismic & Seismology

Exercise 3 - Velocity analysis, NMO and stack.

In this exercise a velocity model of the subsurface will be built by performing Normal Moveout
(NMO) analysis to CMP’s. Also it will be investigated the effect of muting during NMO, and see
how stacking the CMP-gathers after NMO-correction into single traces suppresses the noise present
in seismic data. As it has been mentioned, the NMO describes the effect that the distance between
a seismic source and a receiver (the offset) has on the arrival time of a reflection in the form of an
increase of time with offset. Where the relationship between arrival time and offset is hyperbolic
and the traveltime equation is:

x2
t2 = t20 +
v2
And the NMO correction is given by:

4h2
t2 = t2n +
v2
To carry it out this correction, the different data obtained from the segy was plotted with the
help of the seisplot function. Where, each reflector showed a curve that decreased with depth,
precisely because of the change with the distance between each receiver and the source. Then,
it was determined the speed in which the different reflectors stoped of decay and showed a more
flat linear behavior. Then, the velocity found was put in the function nmov, with the objective of
applies NMO to a single CMP-gather, with linear interpolation. Where the output of the function
is a CMP-gather with NMO. Also, it was used the function nmovt, which did the same function of
nmov, but the velocity used is a velocity-time log.
It is important to take into account that there are two ways of find the velocity of zero offset
time used to the correction: the stacking velocity. The first one, is to make different manual
speed changes based on the distance between the gathers (Figure 25), in order to flat the reflectors
curve. On the other hand, the second consists into use the semblance function 26, which generates
a semblance panel for wave propagation velocities ranging from vmin to vmax , with a stepsize of
velstep (27).

1 %% EXERCISE 3 - Velocity analysis, NMO and stack

2

3 %load the dataset tripli.segy

4 [seis_3,Hea_Tripli_3,geo_tripli_3]=segyread('intro_seismology/data/tripli.segy');
5 Hea_An_CMP_3=analysefold(Hea_Tripli_3, 5); %Analiza el numero de trazas por CMO
6 [Tr_Data_CMP_3,Tr_H_SRT_3]=sortdata(seis_3, Hea_Tripli_3, 5);
7

8 %Seleccionamos el CMP-distance of 800 m

9

10 [CMPgather_Tripli_3,H_CMPgather_Tripli_3]=selectcmp(Tr_Data_CMP_3,
11 Tr_H_SRT_3, 808);
12 %Plot de los datos seleccionados en el cmp
13 %Se plotean los headers del CMP para verificar un solo shot-gather
14 seisplot(CMPgather_Tripli_3)
15 plothdr(H_CMPgather_Tripli_3)
16

17 %Aplicación de la correción del NMO con velocidad constante al CMP escogido

Page 17 of 38
 Seismic & Seismology

18 NMOedCMP_3 = nmo_v(CMPgather_Tripli_3, H_CMPgather_Tripli_3, geo_tripli_3, 2000)

19 plotseis(NMOedCMP_3)
20 %Ahora se aplica el NMO_vt con los picks de tiempo y velocidad anteriores
21 %para 6 refelctores más el inicio y el final
22

23 NMO_vtedCMP_3 = nmo_vt(CMPgather_Tripli_3, H_CMPgather_Tripli_3, geo_tripli_3);

24

25 % Se observan los cosos negros de la gráfica de semblance

26 semblance(CMPgather_Tripli_3, H_CMPgather_Tripli_3, geo_tripli_3, 1400, 6000, 5)
27

28 % 1: Near the beginnig

29 [CMPgather_Tripli_3_1,H_CMPgather_Tripli_3_1]=selectcmp(Tr_Data_CMP_3,
30 Tr_H_SRT_3,520);
31

32 % 2: Near the end

33 [CMPgather_Tripli_3_2, H_CMPgather_Tripli_3_2]=selectcmp(Tr_Data_CMP_3,
34 Tr_H_SRT_3, 2488);
35

36 % 3: Middle
37 [CMPgather_Tripli_3_3, H_CMPgather_Tripli_3_3]=selectcmp(Tr_Data_CMP_3,
38 Tr_H_SRT_3, 1512);
39

40 % 4: First between the start and the middle

41 [CMPgather_Tripli_3_4, H_CMPgather_Tripli_3_4]=selectcmp(Tr_Data_CMP_3,
42 Tr_H_SRT_3, 1088);
43

44 % 5: Second between the middle and the end

45 [CMPgather_Tripli_3_5, H_CMPgather_Tripli_3_5]=selectcmp(Tr_Data_CMP_3,
46 Tr_H_SRT_3, 1912);
47

48 %Parte Semblance para todos los CMP's

49

50 %1: Near the beginnig

51 semblance(CMPgather_Tripli_3_1, H_CMPgather_Tripli_3_1, geo_tripli_3, 1200,
52 6000, 5);
53

54 %2: Near the end

55 semblance(CMPgather_Tripli_3_2, H_CMPgather_Tripli_3_2, geo_tripli_3, 1200,
56 6000, 5);
57

58 %3: Middle
59 semblance(CMPgather_Tripli_3_3, H_CMPgather_Tripli_3_3, geo_tripli_3, 1200,
60 6000, 5);
61

62 %4: First between the start and the middle

63 semblance(CMPgather_Tripli_3_4, H_CMPgather_Tripli_3_4, geo_tripli_3, 1200,
64 6000, 5);

Page 18 of 38
 Seismic & Seismology

65

66 %5: Second between the middle and the end

67 semblance(CMPgather_Tripli_3_5, H_CMPgather_Tripli_3_5, geo_tripli_3, 1200,
68 6000, 5);
69

70 %NMO vt para cada uno de los CMP graher, según sembance análisis
71

72 %1: Near the beginnig

73 NMO_vtedCMP_3_1 = nmo_vt(CMPgather_Tripli_3_1, H_CMPgather_Tripli_3_1,
74 geo_tripli_3);
75

76 %2: Near the end

77 NMO_vtedCMP_3_2 = nmo_vt(CMPgather_Tripli_3_2, H_CMPgather_Tripli_3_2,
78 geo_tripli_3);
79

80 %3: Middle
81 NMO_vtedCMP_3_3 = nmo_vt(CMPgather_Tripli_3_3, H_CMPgather_Tripli_3_3,
82 geo_tripli_3);
83

84 %4: First between the start and the middle

85 NMO_vtedCMP_3_4 = nmo_vt(CMPgather_Tripli_3_4, H_CMPgather_Tripli_3_4,
86 geo_tripli_3);
87

88 %5: Second between the middle and the end

89 NMO_vtedCMP_3_5 = nmo_vt(CMPgather_Tripli_3_5, H_CMPgather_Tripli_3_5,
90 geo_tripli_3);
91 % Utilizamos generatevmodel para generar el modelo de velocidades con los
92 % outputs del analyzefold inicialmente hecho.
93

94 vmodel_Tripli = generatevmodel(Tripli_pos, geo_tripli_3);

95 %-------------------------------------------
96

97 % Improving signal quality after NMO: Muting over-stretched signals

98

99 %Gráficar con 1 y sin 1 para el smute

100 NMOedCMP_Middle_sin1 = nmo_v(CMPgather_Tripli_3_3, H_CMPgather_Tripli_3_3,
101 geo_tripli_3,
102 1800)
103

104 figure1 = figure

105 plotseis(NMOedCMP_Middle_sin1)
106

107 NMOedCMP_Middle_con1 = nmo_v(CMPgather_Tripli_3_3, H_CMPgather_Tripli_3_3,

108 geo_tripli_3, 1800, 1)
109 figure2 = figure
110 plotseis(NMOedCMP_Middle_con1)
111

Page 19 of 38
 Seismic & Seismology

112 %Stacking the NMO-ed data---------------

113

114 NMOedCMP_3_4stack = nmo_v(CMPgather_Tripli_3, H_CMPgather_Tripli_3,

115 geo_tripli_3, 1800)
116

117 stack = stackplot(NMOedCMP_3_4stack, geo_tripli_3)

118 hold on
119 plotseis(NMOedCMP_3_4stack)
120

121 [Tr_Data_CMP_34zo,Tr_H_SRT_34zo]=sortdata(seis_3, Hea_Tripli_3, 5);

122

123

124 zosection_Tripli_0 = nmo_stack(Tr_Data_CMP_34zo, Tr_H_SRT_34zo,

125 Tripli_pos, Tripli_folds, geo_tripli_3, vmodel_Tripli);
126

127 zosection_Tripli_1 = nmo_stack(Tr_Data_CMP_34zo, Tr_H_SRT_34zo,

128 Tripli_pos, Tripli_folds, geo_tripli_3, vmodel_Tripli, 1);
129

130 zosection_Tripli_2 = nmo_stack(Tr_Data_CMP_34zo, Tr_H_SRT_34zo,

131 Tripli_pos, Tripli_folds, geo_tripli_3, vmodel_Tripli, 2);
132

133 plotimage(zosection_Tripli_0)
134

135 plotimage(zosection_Tripli_1)
136

137 plotimage(zosection_Tripli_2)
138

139 % Preparation for the next exercise

140

141 %save('file_out','zosection_Tripli_0','vmodel_Tripli');

Page 20 of 38
 Seismic & Seismology

Figure 21: Generates a plot of CMP gathers according to its midpoint, in this case 808 m.

Figure 22: Seisplot of sorted data product of the CMP distance of 808 m.

Page 21 of 38
 Seismic & Seismology

Figure 23: Header plot of sorted data product of the CMP distance of 808 m.

Figure 24: Plotseis of the result after NMO correction. It was determined that the best velocity
for the third reflector was 2000 m/s.

Page 22 of 38
 Seismic & Seismology

Figure 25

Figure 26: Semblance graph obtained for the CMP of 808 m.

It was taken another 5 CMP’s to analyze them and built the velocity averaging in between
them:

Page 23 of 38
 Seismic & Seismology

Figure 27: Near the beginning, 520 m, fold 8.

Figure 28: Near the end, 2488 m, fold 8.

Page 24 of 38
 Seismic & Seismology

Figure 29: In the middle, 1512 m, fold 8.

Figure 30: Between the beginning and the middle, 1088 m, fold 44.

Page 25 of 38
 Seismic & Seismology

Figure 31: Between the middle and the end, 1912 m, fold 44.

Respective semblances, the velocities and rates are in the code:

Figure 32: Semblance near the beginning, 520 m, fold 8.

Page 26 of 38
 Seismic & Seismology

Figure 33: Semblance near the end, 2488 m, fold 8.

Figure 34: Semblance in the middle, 1512 m, fold 8.

Figure 35: Semblance between the beginning and the middle, 1088 m, fold 44.

Page 27 of 38
 Seismic & Seismology

Figure 36: Semblance between the middle and the end, 1912 m, fold 44.

Corrected images from the velocities obtained in the semblances:

Figure 37: Corrected image between the middle and the end, 1912 m, fold 44.

Page 28 of 38
 Seismic & Seismology

Figure 38: Corrected image between the middle and the end, 1912 m, fold 44.

Figure 39: Corrected image between the middle and the end, 1912 m, fold 44.

Page 29 of 38
 Seismic & Seismology

Figure 40: Corrected image between the middle and the end, 1912 m, fold 44.

Figure 41: Corrected image between the middle and the end, 1912 m, fold 44.

Page 30 of 38
 Seismic & Seismology

Figure 42: Generate v model is used to create a velocity model using the analysefold outputs and
the velocity-time pickups for the 5 CMP’s.

Figure 43: Improving signal quality after NMO.

Page 31 of 38
 Seismic & Seismology

(a) (b)

Figure 44: NMO-ed CMP gather next to stacked traze

Figure 45: NMO stack for a zero-offset section with a mute factor of 2.

Exercise 4 - Zero-offset Migration using a Velocity Model

In this exercise, seismic data will be migrated for some simple cases and for the dataset. For these
datasets, it is necessary to create velocity models. After the NMO correction there is a reduce of
the data volume, in this way there is necessary to carry out a migration that solve this problem.
Through the migration there is re-located the different events in the respective space or time loca-
tion, taking into account the real geometrical position and not the recorded one. In this way, there
is improved the accuracy of the data shown and found. During this process, there is a collapse of
the diffractions that increased the spatial resolution. What is showed during the matlab process
because there were two bow ties in the pre-migration images that after this process showed the
presence of two synclinals.

The migration used was Kirchhoff migration, which uses the integral form of the wave equation
to carry out the process. Its implementation reduces to stacking the data along curves that trace
the arrival time of energy scattered by image points in the earth. The function used is kirkmig,
which has as output the migrated time section and the respective time and distance.

1 %% EXERCISE 4 - Zero-offset Migration using a Velocity Model

2

Page 32 of 38
 Seismic & Seismology

3 % Migration of point diffractors------------

4

5 [seis_4, Hea_Points_4, geo_Points_4]=

6 segyread('intro_seismology/data/points.segy');
7 %Cargamos el arhivo segy
8

9 Points_rela=readsegy('intro_seismology/data/points.segy');
10

11 Points_size=size(Points_rela)
12

13 plotseis(Points_rela,0.004:0.004:1001*0.004,50:50:141*50)
14 %Se redefinieron los lı́mites x y y de acuerdo a la distancia y tiempo
15 ylabel('time (s)','Interpreter','latex','fontsize',16);%Nombre del eje y tiempo
16 xlabel('Source Geophone Distance (m)', 'Interpreter','latex','fontsize',16);
17 %nombre del eje x - distancia
18 %Se realiza la migración de Kirchhoff en el tiempo.
19

20 %Se toma la velocidad de 1500 a partir dde los 1592.35 hallados con la
21 %pendiente
22 [arymig_points,tmig_points,xmig_points]=kirk_mig(Points_rela,3000,0.004,50);
23

24 seisplot(arymig_points)
25

26 %Migration of dipping reflectors

27

28 Planes_rela=readsegy('intro_seismology/data/planes.segy');
29

30 Planes_size=size(Planes_rela)
31

32 plotseis(Planes_rela,0.004:0.004:512*0.004,10:10:128*10)
33 %Se redefinieron los lı́mites x y y de acuerdo a la distancia y tiempo
34 ylabel('time (s)','Interpreter','latex','fontsize',16);
35 %Nombre del eje y - tiempo
36 xlabel('Source Geophone Distance (m)', 'Interpreter','latex','fontsize',16);
37 %nombre del eje x - distancia
38

39 %Se realiza la migración de Kirchhoff en el tiempo.

40 [arymig_planes,tmig_planes,xmig_planes]=kirk_mig(Planes_rela,3000,0.004,10);
41

42 % plotseis(arymig_planes)
43 seisplot(arymig_planes)
44

45 % Migration of stacked dataset tripli

46

47 Tripli_size=size(zosection_Tripli_1)
48

49 plotseis(zosection_Tripli_1,0.004:0.004:351*0.004,8:8:276*8)

Page 33 of 38
 Seismic & Seismology

50 %Se redefinieron los lı́mites x y y de acuerdo a la distancia y tiempo

51 ylabel('time (s)','Interpreter','latex','fontsize',16);
52 %Nombre del eje y - tiempo
53 xlabel('Source Geophone Distance (m)', 'Interpreter','latex','fontsize',16);
54 %nombre del eje x - distancia
55

56 %Se realiza la migración de Kirchhoff en el tiempo.

57 imagesc(vmodel_Tripli)
58 colorbar
59

60 [arymig_tripli,tmig_tripli,xmig_tripli]=kirk_mig(zosection_Tripli_1,
61 vmodel_Tripli,0.004,8);
62

63 plotimage(arymig_tripli)

Figure 46: points.segy with horizontal axis in meters and vertical axis in seconds.

Page 34 of 38
 Seismic & Seismology

Figure 47: Kirchhoff migration of pointsdata with constant velocity of 3000 m/s, 0.004 seconds of
time sampling and 50 meters of spatial sampling. It is evident that when it is applied this correction
the means you need to focus the energy to where it comes from: in this case the apices.

Figure 48: planes.segy with horizontal axis in meters and vertical axis in seconds.

Page 35 of 38
 Seismic & Seismology

Figure 49: Kirchhoff migration of planesdata with constant velocity of 3000 m/s, 0.004 seconds of
time sampling and 10 meters of spatial sampling.

Figure 50: tripli.segy with horizontal axis in meters and vertical axis in seconds, taking into account
the velocity model.

Page 36 of 38
 Seismic & Seismology

Figure 51: Result of the Kirchhoff migration using the velocity model with the no migrated section.

Now, there was made an updated of the velocity model one more time, using the geometry now
know subsoil geometry for the dataset tripli.segy. To get an improve in the Kirchhoff migration,
the result is:

Figure 52: Result of the Kirchhoff migration using the velocity model with the no migrated section,
improved.

Conclusions
From the previous work the following conclusions can be said:
• An appropriate seismic survey must be done in order to have an appropriate seismic data to
process.
• The processed data was interesting as an academic assignment; nevertheless it is proposed
next time to have an appropriate seismic survey to interpret real seismic data from our own
survey.

Page 37 of 38
 Seismic & Seismology

• It is necessary for the seismic data to have an organized structure such as headings, with
correct positions and data

• It is evident that we need to follow a set of steps that must be done in order, to get a correct
solution. Normal Moveout, then Mute then Stacking and then Kirchhoff migration.

• A velocity of 3000 m/s was used instead of the proposed velocity of 2000 m/s for the Kirchhoff
migration in order to obtain a better graphic. Figure 49

• We found a better precision with the semblance method than with the visual method and the
quality of the graphs was improved.

• Some errors can be derived from selected data in a qualitative and arbitrary way.

References
Atanackov, J. (2013). Field comparison of seismic sources for high resolution shallow seismic
reflection profiling on the ljubljana moor. Acta Geodynamica et Geomaterialia, 19-40. doi:
10.13168/agg.2013.0002
Margrave, G. F. (2015). Practice on seismic on data processing. CREWES .
Sheriff, R. E. (2002). Encyclopedic dictionary of applied geophysics. Society of exploration geo-
physicists.
Ziolkowski, A., & Lerwill, W. E. (1979). A simple approach to high resolution seismic profiling for
coal. Geophysical Prospecting, 27 (2), 360-393.

Page 38 of 38