En Tutorcdp

Tutorial on Seismic Reflection CDP
Data Processing
in the RadExPro Plus software
(Edition of 21.11.2007.)

DECO Geophysical, OOO
Moscow State University Science Park,
Leninskiye gory 1-77,
119992 Moscow, Russia
Tel./Fax: (+7 495) 930 94 14
E-mail: [email protected]
Internet: www.radexpro.ru

Content

INTRODUCTION.....................................................................................................................................................................3
Data input, assigning geometry, binning....................................................................................................................3
Creating a RadExPro Plus project ............................................................................................................................3
Loading raw data into the project..............................................................................................................................5
Geometry assigning and binning..............................................................................................................................11
Import of coordinates of source and receiver points from a text file......................................................................... ............12
Calculation of the distances between the shot and receiver points, coordinates of CDP points, binning.................... ...........16

Sorting the data by CDP and control of the assigned geometry ..............................................................................18
DATA ANALYSIS AND TRACE BY TRACE PROCESSING...................................................................................................................24
Sorting traces by CDP and analysis of the wave field..............................................................................................24
Correction for amplitude attenuation.......................................................................................................................27
Spectrum spreading..................................................................................................................................................28
Bandpass filtering.....................................................................................................................................................30
Trace amplitude equalization...................................................................................................................................31
Assigning muting parameters...................................................................................................................................33
Top muting................................................................................................................................................................36
Execution of the preprocessing flow.........................................................................................................................38
VELOCITY ANALYSIS AND STACKING.......................................................................................................................................39
Preparation of the data for velocity analysis, super-gathering................................................................................39
Velocity analysis.......................................................................................................................................................42
Stacking....................................................................................................................................................................52
Display of the stack...................................................................................................................................................57

Introduction
This tutorial is intended for the users, who begin to process seismic reflection CDP data in the
RadExPro Plus program. All standard stages of basic CDP processing are discussed, from the
introduction of geometry to stacking, that is the so-called minimal processing sequence. It is
assumed that the user is already familiar with the theory of the CDP reflection method and with the
fundamental technology of processing such data.
The processing is conducted on an example of the real data, which can be downloaded from our
Web-site: http://radexpro.ru/upload/file/tutors/CDP/inpdata.zip
The archive contains initial data for the work: a fragment of an on-shore seismic profile, recorded in
SEG-Y format (file line_1.sgy), with the trace headers containing source point and receiver point
numbers, and two ASCII files, rec_geom.txt and sou_geom.txt, containing coordinates of the
receivers and sources, respectively.
Furthermore, you can load the final project, which is a result of executing all steps, described in the
tutorial: http://radexpro.ru/upload/File/tutors/CDP/MyProject.zip
Note that the facilities of the software, of course, are not limited to the minimal processing sequence
described here. We consciously did not consider more complicated tasks such as, for example,
horizontal velocity analysis, migration, calculation and analysis of seismic attributes, etc. You can
find the information about these and other procedures of data processing and analysis in the "User
Manual" to the program.

Data input, assigning geometry, binning
Creating a RadExPro Plus project
All reflection data processing in the RadExPro Plus is performed within the framework of projects.
A project is a complex of initial data, intermediate and final processing results, processing flows, all
organized into a common database used by the software. The projects are stored in separate folders
on disk. When a project is created, the new folder for it is created automatically. A project can be
transferred from one computer to another by simple copying of the folder (in case that all the data
used are stored inside this folder).
.
Let us create a new processing project. Start-up the software. To do this select RadExPro Plus
Total 3.75 command in the Windows Start menu.

3

When the RadExPro Plus starts, the Project Manager dialog opens, containing the list of
registered projects.

Click the button New Project and select a parental folder on the disk, where the subdirectory with
the project will be created. After this, in the appearing window, enter the new project name.

Make sure that the option Create subfolder is selected and click Ok. In the selected folder, the
subdirectory with the project name will appear. Also the project will appear in the list of available
(registered) projects.

4

Select it and click Ok.

The main window of the RadExPro program, containing the project tree, will appear. By now this
tree is empty.

Loading raw data into the project
Using Windows Explorer, enter the project folder:

5

In this folder, create a subdirectory Data and copy the initial data there.

Storing the data inside the folder of the project allows the software to use relative paths to the data
files instead of the absolute ones. This makes the transfer of projects from one computer to another
much easier.
Return to the main window of the RadExPro program. A RadExPro database has 3 structural
levels. The upper level corresponds to a study area, the middle level - to a line, and the lower one to a processing flow. Right-click with the mouse on the yellow circle (that is the root of the project
tree), select option Create new area in the context menu, and enter a name of the area where the
field work was conducted. .

The following figure shows the window, where the name of the area should be entered :

6

Similarly, right-click with the mouse on the yellow rectangle with the name of the area, select
Create line command and create a new line. The name of the line is assigned the same way.

The database allows storage of several areas within one project. Each area may contain several
lines, each line may contain several processing flows that are applied to the data of this line.
The same way you created the area and the line, create a processing flow and name it 010 - data
load. It is recommended to that you start the name of each flow with a number. The process of
seismic data processing has several stages, carried out consecutively. Since the RadExPro Plus
program sorts the structural elements of the database alphabetically, it is reasonable to number the
flows so that they would be displayed in an accurate logical sequence.

Double-click the left mouse button on the flow name to enter the flow editing mode. The window of
the flow editor will open. On the left side of the window is the flow itself (it is empty so far), on the
right side is the library of available processing routines (modules), divided into groups.

7

Create a flow, consisting of the modules SEG-Y Input and Trace Output (both are located in
group Data I/O ). This flow will read the data from a SEG-Y- file and record them into the project
database a "dataset"-object of the database..
Modules are added to the flow one by one. To add a module into the flow just drag it over from the
library on the right to the area of the flow on the left. At this moment, a parameter dialog of the
module will be opened (later, when the module is already in the flow, the same routine parameter
dialog can be called up by a double-click of the mouse on the module name in the flow). While
already in the flow, the routines can be moved up and down relative to each other by simple dragand-drop.
In Data I/O group find the SEG-Y Input module and add it into the flow. When adding the routine,
assign the parameters of data reading in the opened dialog box. Click the Add... button and select
our test data file line_1.sgy. For training purposes, the information about source and receiver
positions in this file is stored not in the standard SEG-Y trace header words, but in the reserved
space of the trace headers. In order to read the traces header information recorded in non-standard
positions within the SEG-Y trace headers, enter a remap of the header fields. For this, switch on the
Remap header value option and enter the following text RECNO,4I,,181/ SOURCE,4I,,185/.

8

Remap of header fields. Some seismic data formats allow trace header remapping, that is storage in
the trace headers of some values, not provided by the standard. Sometimes the values are recorded
in a non-standard format of number representation or in a non-standard position within the header.
As a rule, contemporary software packages provide a possibility to indicate explicitly from what
byte from the beginning of a header and in what format to read the values saved this way. In the
training file line_1.sgy, the header fields containing the number (picket) of source and receiver
position are recorded as integer 4-byte numbers in bytes 181-184 and 185-188 are saved this way.
The remap string shown above will allow for reading them from there and saving into the RECNO
and SOURCE header fields of the RadExPro Plus database.
RadExPro header fields. The RadExPro software uses its own set of header fields for storing
auxiliary information about seismic traces. The values of header fields are associated with the trace
and can be perceived an array of named variables linked to the trace.
When creating a new project, the default set of header fields generated is quite similar to the SEGY trace header. (See the correspondence of the RadExPro header fields and SEG-Y trace header in
the description of module SEG-Y Output in the "User’s manual"). However, further the header
fields can be edited – you can add new fields, remove or re-name the existing ones.
A part of the header fields are standard and it is strongly not recommended to change their meaning
(for example field DT must always store the value of the sample interval). Other fields can be used
at your discretion. In the new (or existing, but not used) header fields different information can be
recorded, for example the arival time of a wave as picked on this trace. You can perform
mathematical operations with the values of the header fields, convert them into picks, etc.

9

After the SEG-Y Input module, add into the flow the module Trace Output, which will save the
data from the flow into the database. Name the object, which will contain these data as “line 1 –
raw” and place it at the second level of the database into the Line 1 (as shown in the following

figure).
Besides, for the control, after module Trace Output, add into the flow the Screen Display module.
The obtained flow must appear as follows:

For executing the flow select the Run menu command. As a result, the Screen Display window
should open, showing the data being entered, while the data will be read from the disk file and
recorded into the database. The window Screen Display, which should appear on the screen, is
shown below.
10

Important!: When the volume of the data read from the file is great (comparable or exceeds the
volume of RAM or just around 1 Gb or more), it is necessary to use Framed mode, which allows to
read the data into the memory not as a whole, but by pieces (frames). To turn into this mode and to
determine the size of the frames use the Framed mode... command available from the flows editor
main menu.
Note about the names. A name of any object of the database (seismic dataset, processing flow, etc.)
must reflect its essence, but not consist just of several letters. The name of a seismic dataset should
consist of 2 parts –the identifier of initial data and the processing stage of the data. Thus, when
entering the field data, the name line 1 - raw was selected here.
Geometry assigning and binning
Assigning geometry to seismic data means that for each trace a number of values are determined,
which, then, are saved in the specified header fields of the dataset in the project database. The list of
the necessary values and the header fields corresponding to them are given below:
1. Source point (also called 'shot point' or SP) number (header field SOURCE)
2. SP coordinate(SOU_X)
3. Receiver point (RP) number (RECNO)
4. RP coordinate (REC_X)
5. Distance between the source and the receiver (OFFSET) and the modulus of this value
(AOFFSET)
6. Unique field record number (FFID)
7. Channel number (CHAN )

11

Note. The list given above corresponds to one-dimensional geometry. Generally speaking, the
header fields SOU_X, SOU_Y and REC_X, REC_Y can be used for describing the coordinates of
SP and PP. However, as the observations in the training file were carried out along one line only, it
is proposed to use only one coordinate X, X axis is directed along the profile.
While the FFID, CHAN, SOURCE and RECNO header fields were filled when reading the data
from the initial SEG- Y file, the coordinates of SP and RP should be imported from the test files,
and the distances between SP and RP should be calculated.
In practice, absolutely any combination of the filled trace headers can be met. For example, the data
can be transmitted into processing with empty headers. In this case they should be formed with the
use of tools, available in the processing software package.
The situation, in which the initial seismic data contain the numbers of field records and channels,
while the connection between the field record numbers and SPs, as well as between the channel
numbers for each shot and the RPs is to be calculated is rather widespread. However, for
simplification of geometry assigning for training purposes, the data already contain the numbers of
pickets shot points and receiver points, and only the coordinates are to be imported..
Import of coordinates of source and receiver points from a text file
For manipulations with trace header field values, including import of the values from ASCII tables,
in the RadExPro software the Geometry Spreadsheet tool is used.

Select the Database/Geometry Spreadsheet... entry in the menu.
Then, in the opened dialog, select the seismic dataset, geometry of which is to be edited.

12

The following figure shows the Geometry Spreadsheet window.

In order to display the required header fields (all header fields declared in the database already
exist, but are not displayed) use the Fields/Add fields... option of the menu In the opened dialog
box, keeping the Ctrl key pressed, select the following header fields: SOURCE, SOU_X, RECNO,
REC_X

13

As a result the header editor window shall look as following:

Select the Tools/Import menu command. The dialog with the import parameters will open. Open
the file sou_geom.txt there and define the rules of filling header fields. For this add to the Matching
fields list the field SOURCE (for that, click the appropriate Add button and select it from the list).
To the Assign Fields list add the SOU_X field . Then specify the columns of the text file from
which the indicated fields should be read. They shall be set in the text lines under Set column
buttons. (By the way, if you set the cursor into the appropriate column and click Set column, the
number of the column will be automatically entered there). Finally, the range of the lines should be
14

indicated, from which the program will obtain the values. This shall be done in Lines field: set
From and To values. The correct settings are shown on the figure below.

When importing the values of the header fields from the text file, the program works as follows. For
each line of the text file, all the matching fields as well as the assign fields are read from the
specified columns. In the specified seismic dataset all the traces with the values of the matching
header fields being precisely equal to the values from the read line are determined. Then, the values
from the read line are entered into the assign fields of these traces.
Important!: Among other things, this means that the matching fields are better to be integer (this
fact shall be taken into account when the files with the geometry are formed).
Click the OK button to assign the header fields. The result is shown below.

15

Import the coordinates of the receiver points from the rec_.geom.txt file in the same way.

Calculation of the distances between the shot and receiver points, coordinates of
CDP points, binning
Using the Fields/Add fields menu entry add to the spreadsheet the following fields: OFFSET
(source- receiver distance), AOFFSET (the absolute value of the offset), CDP_X (coordinate of the
CDP), CDP (CDP point number).

16

For calculation of the specified values use the Trace Header Math tool, dedicated to mathematical
operations with header field values. The Trace Header Math is available from the menu
Tools/Header Math.
In the open dialog box enter the following expressions:

and click the OK button.
The numbers of the CDP gathers will be calculated, on the basis of the CDP coordinate and desired
size of a bin. As the distance between the receivers was ~25 m, and the shot interval was ~50 m, the
bin size should be selected as 12.5 m. For calculation of the CDP numbers, in the same Trace
Header Math window enter the following expression:

17

The resulting table should look like the following:

To save the changes in the database on exit from the Geometry Spreadsheet select Yes, or use the
Edit/Save changes menu option.

Sorting the data by CDP and control of the assigned geometry
To verify the correctness of geometry assigned, make the following.
Create a new flow, name it 020 - geometry check

18

Construct a processing flow, consisting of the Trace Input and Screen Display modules.

The Trace Input should enter the data into the flow sorted by, primarily, CDP and, secondarily,
OFFSET (hereinafter, we will use CDP:OFFSET notation to describe this sorting). For this, set the
routine parameters as shown on the following figure.

19

Due to this selection of the Sort Fields, the traces will be entered into the flow ordered by their
CDP number. Within each ensemble with identical CDP number, the traces will be ordered by the
OFFSET.
The text line specifying a trace selection mask for each of the sorting keys should be entered in the
Selection field. Selection parameters for each of the keys are separated by a colon. In this case, *:*
means that for each of the two sorting keys, all available traces will be entered into the flow sorted
according to the sorting key values in ascending order.
In the Screen Display module assign the parameters as shown on the figure below.

20

If the Ensemble boundaries parameter is switched on, the ensembles of traces will be separated on
the screen by empty spaces. The ensemble in the RadExPro is defined by the first sorting key
assigned in the Trace Input, that is, in this case, by the value of the CDP field.
Click the button Axis... and assign the following parameters of the axes:

Execute the flow using the Run command of the menu. The raw data sorted by CDP:OFFSET
(CDP gathers) will be displayed on the screen.

21

To control geometry assigning display the theoretical travel times of reflected wave calculated from
header field values. To do it use the Tools/Approximate/Hyperbola (reflection) entry of the
Screen Display menu.

Use the default parameters of the travel-time hyperbola:

22

On the screen the travel time curve of a wave reflected from the boundary of a half-spaces is
depicted by a blue line. Current parameters of the medium and the boundary are displayed in the
upper left corner of the Screen Display window.
Using the arrows on the keyboard (right/left) change the velocity in the medium, until the
approximate coincidence of the blue line and the observed first arrivals is achieved.

If it is possible to attain good matching of the observed direct wave by the theoretical travel times
for all CDP gathers, then the distances between the SP and the RP were calculated correctly.
If the matching cannot be attained (for example, the observed travel times appear to be shifted
relative to the theoretical positions), this indicates an error in the geometry of the traces. Then
assign the geometry again or find and correct the error some other way.

23

Data analysis and trace by trace processing
Sorting traces by CDP and analysis of the wave field
To preprocess the seismic data, create a new flow in the RadExPro project, as shown on the figure.

Put the Trace Input and Screen Display routines into the 030 - preproc flow.
Enter the following parameters in the Trace Input dialog:

The 0-10000(10):* as a sorting rule means the following:
 From all CDP gathers, which fall into the range of 0-10000 (but these are all CDP gathers in
24



the dataset), only those with the numbers divisible by 10, will be taken;
Within the CDP ensembles, the traces will be sorted by the values of OFFSET in ascending
order.

This sorting is necessary now to decrease the amount of data (in 10 times) on the stage of testing of
the processing parameters. Because of this sorting it will be possible to control the result of the
processing not on a single ensemble but on a number of CDP gathers evenly selected along the line.
Assign the parameters of the Screen Display to have some 3-5 gathers on the screen. The flow will
look as follows:

Execute the flow, the CDP gathers will appear on the screen.
Examine the gathers on the screen and try to identify the types of the observed waves. Find the
direct wave, reflected waves, surface waves.
Estimate the velocity of the direct wave, the group velocity of the surface waves. For this, use the
the Screen Display “Line”-tool allowing to fit theoretical travel times of the direct wave to the data.
Since velocity is calculated as distance from the source divided by the arrival time, first it is
necessary to indicate the header field, which will be used for calculation of the distance between the
traces when calculating the apparent velocity. Use the point of menu Tools/Approximate/Line
Header word... and select the field OFFSET.

25

For evaluating the apparent velocity use Tools/Appoximate/Line. In order to fit the line to the data
assign the beginning of the approximated section on the screen by left click of the mouse, then the
end of the section - by right click. The current value of the apparent velocity will be shown in the
green line in the top left corner of the screen.

Record the obtained values of velocities into a text file. It can be made automatically. For that,
while the Line-tool is active, select Tools/Approximate/Save parameters menu entry.

Additional window with an editable text field will open. Copy the current velocity there by pressing
Ctrl+Q while the Screen Display window is in focus. You can copy as many values as you like.
26

Comments can be added to the copied values in the window, just type them manually. When you are
finished with the velocities, save the file, selecting the command File/Save in the menu of
additional window and then close the window.

Estimate the spectrum of different parts of the record. For this use the Tools/Spectrum/Average
menu command.

Correction for amplitude attenuation
To compensate amplitude attenuation add the Amplitude Correction module into the flow. Place it
between the modules Trace Input and Screen Display. Changing the parameters of the module and
each time executing the flow, you can test different types of amplitude corrections. For purposes of
the subsequent processing, spherical divergence correction is a good choice. In order to apply this
correction, set the parameters of the module as shown on the figure below.

27

Compare the appearance of the data before and after the spherical divergence correction. For this,
run the flow two times - with the active and "commented" Amplitude Correction module. (to
“comment” the module right-click it with the mouse). As a result, two Screen Display windows
appear on the screen, one of which contains corrected data, the other one – initial data.
Spectrum spreading
After the correction for spherical divergence, add the module Predictive Deconvolution to the
flow. When the purpose of the predictive deconvolution is spectrum spreading, it is reasonable to set
prediction gap equal to one sample, to select the filter length close to the length of the wavelet and
to assign the window for the calculation of the deconvolution operator in such a way that it contains
the target reflections. Basing on these considerations, some initial parameters can be assigned as
follows:

28

Experiment with different parameters, research how the level of white noise influences the result.
The following two figures demonstrate the data before and after the use of predicting
deconvolution.

Before the deconvolution:

29

After the deconvolution:

Bandpass filtering
Bandpass filtering is to be used after the deconvolution, aimed to decrease the level of lowfrequency and high-frequency noise and to shape the spectrum of the trace to achieve a simple
wavelet. Add the module Bandpass Filtering into the flow after the deconvolution routine. In the
parameters of the module select Ormsby’s filter with the following parameters: 5-10-40-80 Hz.
Several CDP gathers after the bandpass filtering with these parameters are shown on the following
figure.

30

Trace amplitude equalization
The amplitudes, recorded by each seismic receiver is influenced, among other things, by the
conditions around the source and the receiver. When the data is not aimed for the dynamic
interpretation (for example, for the purposes of AVO- analysis) it is not necessary to use the
complex procedures of the surface-consistent amplitude corrections, instead you can try to reduce
the differences between traces by the simple traces equalization. To do this add the module
Amplitude Correction into the flow again with the following set of the parameters:

31

Note, that the window we select here for estimation of the average amplitude of each trace contains
the target reflections and does not contain trace sections before the first arrivals.
At the present moment the flow looks as follows:

32

Assigning muting parameters
Since our processing is aimed to obtain a seismic section of reflected waves, in this case, direct
wave is, obviously, shall be considered as noise. The most effective method of suppression of this
noise is the top muting from the beginning of the trace to the time, equal to the direct wave arrival
time, plus some time after it, containing the wavelet of the direct wave.
In order to assign this muting, re-sort the traces in the module Trace Input temporarily in the order
OFFSET:CDP. To do it, change the parameters of the module Trace Input as follows:

Execute the flow; the traces, sorted out in ascending order of the OFFSET header field will be
shown on the screen. Such a gather (called, common-offset gather) allows convenient assigning the
muting time, which will be suitable for all CDP gathers.
To do it, create a new pick object (Tools/Pick/New Pick menu command of the Screen Display),
then pick the time of muting approximately as shown on the figure (direct wave arrival time plus
~100-200 ms).

33

A pick in the RadExPro is the collection of the time values matched by two header fields, as it is
considered that it is always possible to find two header fields that will unambiguously identify a
trace (for example, number of CDP and offset, or shot point number and channel number). In this
case, however, we want the time of the muting to be suitable for all CDP gathers and depend only
on the offset. Therefore we must bind the pick to only one header field - OFFSET.
To do it, select the Tools/Pick/Pick Headers entry of the Screen Display menu

34

In both columns of the open window select OFFSET:OFFSET

Then click the OK button and save the pick with the Tools/Pick/Save As... command of the Screen
Display menu. Indicate the name of the pick as top_mute.

35

Top muting
In the flow 030 - preproc return to the initial sorting (CDP:OFFSET) changing respectively the
Trace Input parameters.
Add the module Trace Editing to the end of the flow (before the Screen Display module) with the
following parameters:

On the second tab specify the horizon that will define muting, as a database pick top_mute, which
was saved at the previous stage. The Horizon tab must look like this:

Now the flow looks like the following:

36

Execute it and make sure that the result looks approximately like this:

37

Execution of the preprocessing flow
At this point you may consider that the parameters of preliminary data processing are selected and
you can execute the flow with the complete dataset. To do it change the parameters of the
Trace Input in such a way that now the data from all CDP points enter the flow:

As the volume of data in the flow is rather large now, we have to execute it frame by frame. In the
window of the flow editor select the Flow mode menu entry and assign the frame size so that the
data of each frame were completely fit to the available RAM memory:

Add the module Trace Output to the end of the flow in order to save the results of the preliminary
data processing. In the parameter dialog of this routine create a new dataset with the name line 1 –
preproc where the data from the flow will be saved to.
38

Comment the Screen Display.
Finally the processing flow appears as follows:

Execute the flow.

Velocity analysis and stacking
Preparation of the data for velocity analysis, super-gathering
As a rule, in order to increase the signal-to-noise ratio and obtain more coherent velocity spectra,
velocity analysis is carried out on ensembles consisting of several adjacent CDP gathers (the so
called super-gathers), rather than on separate single CDP gathers.
Another peculiarity of the data preparation to the velocity analysis is that the data must possess
maximum signal-to-noise ratio, while the high vertical resolution of the record and the recovery of
the dynamics are not important. Therefore, when preparing the data to the velocity analysis such
procedures as automatic gain control in a relatively short window and bandpass filtering in a
relatively narrow band (aimed to keep only the part of the spectrum with the the maximum signalto-noise ratio) are often used.
Create the flow 040 – velocity analysis.

39

Place the Super Gather routine, which will form the super-gathers, in the beginning of the flow.
It is reasonable to select the Super Gather parameters approximately as shown on the following
figure.

That is, the velocity analysis will be carried out with an interval of 50 CDP points, each supergather will contain10 adjacent individual CDP gathers. The super-gathers are formed of the
preprocessed dataset line 1 - preproc prepared at the previous stage.
If you add the module Screen Display into the flow after the Super Gather you will be able to see
how the resulting super-gathers look like.

40

The following figure shows the result of executing such a flow.

It is easy to see that, due to the use of a substantially larger volume of data, the phases of the
reflected waves are outlined much more confidently than on single CDP gathers..
41

Velocity analysis
Comment the Screen Display module or remove it from the flow. For conducting the velocity
analysis, add the Interactive Velocity Analysis routine to the end of the flow.
Let us discuss the assigning order and the reasonable values of the parameters of the module.
First, it is necessary to specify, where to save the resulting velocity pick. The output velocity field
can be stored either in a text file or in a database “velocity-pick” object. We recommend that the
output velocities are always saved as a database object. (Then all possible manipulations with
them, such as export/import, can be made using the with the special tool Database Manager,
available from the Database/Database Manager menu of the RadExPro main window).
Therefore, in the module parameter dialog select the Output Velocity tab and make sure that the
option Database - picks is selected.

Click the Browse button corresponding to this option and in the appeared dialog box specify the
database object name, where the velocity pick is to be stored, as shown on the following figure.

42

It is reasonable to store the velocity picks at the second level of the database, corresponding to the
line.
Now specify the same velocity pick as an input velocity function. It will be useful if you decide to
return to the velocity analysis later and to continue the work with the pick. For this in the Input
Velocity tab assign the same parameters as on the previous one.

Important!: When assigning a new velocity pick, its name must be specified first on the Output
Velocity tab and only then - on the Input Velocity tab. Otherwise the program reports an error
message.
43

Now select the tab Super gather and switch off the Bin offsets option. This option allows for
summing up the traces with the close offsets before calculating the velocity spectrum. This would
result in much faster calculation of the velocity spectrum, however its coherency somewhat
deteriorates.

Select the tab Semblance to assign the parameters of the velocity spectrum calculation: the start and
end velocities, the velocity step and the time step. The velocity spectrum will be calculated as a
normalized for the specified range of velocities.

44

The parameters by default for the considered data set are completely acceptable.
The remaining dialog tabs are responsible for display parameters of different elements of the
velocity analysis window.
It is reasonable to keep the parameters by default on the Semblance Display tab responsible for the
parameters of velocity spectrum display.

45

It is reasonable to display the traces of the current super-gather for which the velocity analysis is
performed in color. Select any palette, in which the traces before and after the normal move-out
(NMO) corrections will be conveniently observed (by default the gray-scale pallet is selected, here
we are going to replace it by the black-white-orange pallet). For this, first select the Color option of
the Display mode in the tab Gather.

46

Then click the Palette button and in the appeared dialog box click on the Load palette button.

A set of the predefined palettes is stored in the folder, where the RadExPro package is installed, in
PALETTES subdirectory. Select the palette blkwtord.pal.

47

The result should be as shown in the following figure.

The dynamic stack (FLP Display) and the panel of the constant velocity stacks (CVS Display) can
be conveniently observed when the traces are displayed in wiggle trace/variable area (WT/VA)
mode. Therefore the parameters for them can be kept as they are set by default.

48

After assigning the parameters, execute the flow.
As a result, the interactive velocity analysis window will appear. It consists of 4 parts (from left to
right): the semblance (velocities spectrum), the current super-gather, the dynamic stack and the
constant velocity stack panels.
The dynamic stack consists of the traces, obtained as a result of stacking of the CDP gathers of the
current super-gather corrected for NMO with with the current velocity pick. When the velocity pick
changes, the stack is dynamically recalculated.
The constant velocity panels show a series of stacked traces for a series of constant velocities. Each
stack fragment here is obtained from the CDP gathers of the current super-gather corrected for
NMO with one or another constant velocity.

49

Velocity pick is made on the semblance panel. When picking, follow the maximums of the energy
of the semblance. An example of a reasonable velocity pick for the above figure is shown below:

50

Additionally, the module Interactive Velocity Analysis allows accomplishing the following
actions, useful for the QC of the velocity picking:
●

The button N on the tool bar switches on the mode, when the current super-gather is
dynamically corrected for NMO with the current velocity pick. At that, the travel time
curves of the reflections ideally must become rectified (see the figure below).

51



Clicking the Dix button of the tool bar you may display the interval velocities calculated for
the current velocity pick according to the Dix’s formula (light-blue blocked curve on the
figure). By the way, switching this option on and changing insignificantly the stacking
velocity (barely moving one of the points of the velocity pick), you can see how extremely
unstable is the conversion of the RMS velocities (that is what we actually pick) into the
interval velocities. You can see that even the slightest changes in the stacking velocity can
lead to sometimes catastrophic changes in the interval velocities. This effect is the greater,
the less is the analyzed interval.

After the velocity pick for this super-gather position is assigned, it is possible to pass to the next
position, clicking the button with the right arrow on the tool bar. If you carry out the velocity
analysis in the framed mode, you can move through the super-gathers to the right or left within the
current frame, otherwise – within the entire dataset.
Carry out the velocity analysis for all super-gathers, then before exiting the module press the Save
button on the tool bar in order to save the created velocity pick.
Stacking
Create the flow 050 – stack.

52

The flow must contain the following routines:
Trace Input, which enters the line 1 - preproc dataset into the flow in the CDP:OFFSET sorting
order.

The NMO/NMI module to correct the traces for the NMO. The parameters of the module should be
53

assigned as follows. Select the NMO mode in the NMO tab, set muting on the signal tension = 30
(i.e. those parts of the traces, which as a result of the NMO correction will be extended to more than
30%, will be reset to zero).

In the Velocity tab select the velocity pick, which was obtained as a result of the velocity analysis
performed before.

After the NMO/NMI module in the flow, place the Ensemble Stack routine. This module stacks
horizontally all the traces within each ensemble in the flow. Since in this case in the Trace Input at
the beginning of the flow the CDP header field was selected as the first sorting key, the CDP gathers
will be considered as ensembles.

54

Assign the parameters of the Ensemble Stack module as shown on the following figure.

Finally, the Trace Output module should be the last one in the flow, it will save the results to the
line 1 – stack dataset, which is also reasonable to be created at the second structural level of the
database.

The flow looks now as shown on the following figure:

55

For executing this flow, switch on the framed mode. Set any reasonable frame width and make sure
that the Honor ensemble boundaries option is on. This will ensure that each frame is completed
until the last trace of the last ensemble in the frame.

Execute the flow.

56

Display of the stack
Create the flow 060 – view stack.

The flow should consist of the modules Trace Input and Screen Display. The Trace Input should
read the obtained stack (line 1 - stack) sorted by CDPs.
The result should look approximately as shown on the following figure.

57

58