Seismic Refraction Data Processing Software: Seismo Version 2.7

INSTRUCTION REPORT GL-90-2
$,'r SOFTWARE:
SEISMIC REFRACTION DATA
SEISMO PROCESSING
VERSION 2.7
Cby
S iE COPY Donald E. Yule, Michael K.
Sharp
Geotechnical Laboratory
(DEPARTMENT OF THE ARMY
Waterways Experiment Station, Corps of Engineers
3909 Halls Ferry Road, Vicksburg, Mississippi 39180-6199
00
Nj
N
#EL1ECTE,
(KqJ9.',
F September 1990
Final Report
Approved for Public Release; Distribution Unlimited
Prepared for DEPARTMENT OF THE ARMY

US Army Corps of Engineers
Washington, DC 20314-1000
and
US Army Engineer Topographic Laboratory

LABORATORY Fort Belvoir, Virginia 22060-5546
This software and manual are distributed "AS IS" and without
warranty as to performance. Because of the many uses to which
this software may be put and the variety of hardware used In
conjunction with it, no warranty of fitness for a particular purpose is
offered. While the developers have invested considerable time to
create a high quality product, the user mast assume the risk of
using this software.
IBM is a registered trademark of International Business Machines

Corporation.
DOS Is copyrighted by International Business Machines
Corporation.
HP LaserJet is a registered trademark of Hewlett Packard
Company.
Quickpak is a registered trademark of Crescent Soltware, inc.
Destroy this report when no longer needed. Do not return it to

the originator.
The findings in this report are not to be construed as an official

Department of the Army position unless so designated by other
authorized documents.
The contents of this report are not to be used for

advertising, publication, or promotional purposes.
Citation of trade names does not constitute an
official endorsement or approval of the use of
such commercial products.
Unclassified
SECURITY CLASSIFICATION OF THIS PAGE
IForm Approved
REPORT DOCUMENTATION PAGE OMB No. 0704.0188
Ia. REPORT SECURITY CLASSIFICATION 1b, RESTRICTIVE MARKINGS
Unclassified
2a. SECURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION IAVAILABILITY OF REPORT
Approved for public release; distribution
2b. DECLASSIFICATION I DOWNGRADING SCHEDULE unlimited.
4. PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANIZATION REPORT NUMBER(S)
Instruction Report GL-90-2

6a. NAME OF PERFORMING ORGANIZATION 6b, OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION
USAEWES (If applicable)
Geotechnical Laboratory
6C. ADDRESS (City, State, and ZIP Code) 7b. ADDRESS (City, State, and ZlP Code)
3909 Halls Ferry Road

Vicksburg, MS 39180-6199
Ba.-NAME OF FUNDING ISPONSORING j8b. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER
ORGANIZATION (If applicable)
See reverse.
8c. ADDRESS (City, Sate, and ZIP Code) 10. SOURCE OF FUNDING NUMBERS
PROGRAM PROJECT TASK WORK UNIT
ELEMENT NO. NO. NO. ACCESSION NO.
See reverse.
17, TITLE (Include Security ClaOs1nc8ation)
Seismic Refraction Data Processing Software: SEISMO Version 2.7

12. PERSONAL AUTHOR(S)
Yule, Donald E.; Sharp, Michael

'13a,.
TYPE OF REPOR T K.E
13b. TIME COVERED TO A EO EO T(er M nh a) 1.P G O N
Final report FROM TOSeptember 1990 139
1'6. SUPPLEMENTARY NOTATIO'N
Available from the National Technical Information Service, 5285 Port Royal Road, Springfield,
VA 22161.
17 - COSATI CODES 18. SUBJECT TERMS (Continue on reverse if necessary and identify by block number)
-- FIELD GROUP SU8.GROUP Seismic refraction
.... Engineering geophysics
K 19. ABSTRACT (Continue on reverse if necessary and idepf), by block number)

A software package entitled T'SEISMO"I~as een developed to perform seismic refraction
data reduction, interpretation, and presentation. The software package will do both forward
and inverse modeling of refraction data, calculation of results, display of results, and final
report figure presentation of results. The basic concepts of seismic refraction surveying and
the procedures for forward and inverse modeling of refraction data are presented. In
addition, several case histories of data processed with the softwar e are presented.
)I
tb-
20. DISTRIBUTION/AVAILABILITY OF ABSTRACT "21 ABSTRACT SECURI Y CLASSIFICATION

E
E "LlHTEDl SCAM AS OPT
Eli ,Mi ^ iCIC!, - ,- Unla...4.
. a
2"2a.NAME OF RESPONSIBLE INDIVIDUAL 22b TELEFHONE (include Area C'ode) 22c. OFFICE SYMBOL
DD Form 1473, JUN 86 Previous editions are obsolete. SECUrITY CLASSIFICATION OF THIS PAGE
Unclassified
Unclassified
8 a. NAME OF FUNDING/SPONSORING ORGANIZATION (Continued).

US Army Corps of Engineers
US Army Engineer Topographic Laboratory
8c. ADDRESS (Continued).

Washington, DC 20314-1000
Fo-t Belvoir, VA 22060-5546
Unclassified
PREFACE
This report describes the user guidelines for implementing the seismic
refraction data analysis and presentation software SEISMO. This user's guide
contains information about data input, calculation procedures, output,
graphics, and presentation. Two appendices are included that explain the
concepts of the two types of modeling available with seismic refraction data,
inverse and forward modeling. Three examples are included in a third appendix
to illustrate the material in this text.
Partial funding for development of the program came from Headquarters,
US Army Corps of Engineers (HQUSACE), and part came from the US Army Engineer
Topographic Laboratory (ETL) under work performed by the Water Detection
Response Team (WDRT). Funding from the HQUSACE was provided to make program
improvements, corrections, modifications, and documentation under the Numer-
ical Modeling Maintenance Program. Funding from ETL was provided to estab-
lish a program that could be used by the WDRT.
SEISMO was initially developed by Mr. D. E. Yule with assistance from
Mr. M. K. Sharp. Corrections, modifications, and improvements were performed
by Mr. Sharp and Mr. Yule. This report was prepared by Mr. Sharp.
The project was accomplished under the general supervision of Dr.
William F. Marcuson III, Chief, Geotechnical Laboratory, US Army Engineer
Waterways Experiment Station. Direct supervision was provided by Mr. J. R.
Curro, Chief, Engineering Geophysics Branch, GL, Dr. Mary Ellen Hynes, Chief,
Earthquake Engineering and Seismology Branch, GL, and Dr. A. G. Franklin,
Chief, Earthquake Engineering and Geosciences Division, GL.
COL Larry B. Fulton, EN, was Commander and Director of WES during the
investigation. Dr. Robert W. Whalin was Technical Director.
D~ts a
Unafwll4uved (3
S.....ust.ifi.itt.
~~~bistribult #.....
vailab3IL.U CWt
[Di t s#Cl.
K
CONTENTS
Page
PREFACE.............................................................. 1I
CONVERSION FACTORS, NON-SI TO SI UNITS OF MEASUREMENT...................4
PART I: INTRODUCTION.................................................S5
Manual Layout.................................................. 5
Typographic Ccnvention..........................................6
PART II: REQUIREMENTS................................................ 7
Overview....................................................... 7
Basic Steps to Using SEISMO......................................8
Hardware and Software Requirements...............................8
Plotters and Printers Supportod.........I.........................9
Software Requirements............................................9
PART III: PROGRAM OVERVIEW............................................11
Starting the Program............................................11
The Main Menu................................................... 12
Control Menus................................................... 13
PART IV: FILE MENU................................................... 15
Build (New)..................................................... 15
Read (Existing)................................................. 19
Write (To Disk)................................................. 20
Print (Data).................................................... 21
PART V: PLOT MENU.................................................... 25
Screen.......................................................... 25
Plotter......................................................... 26
File (To Plotter)............................................... 30
Overlay Plot.................................................... 30
PART VI: EDIT MENU................................................... 31
Data....................... I....................................31
Inverse Model................................................... 31
2
Page
PART VII: invMOD MENU................................................ 33
Input........................................................... 33
Screen.......................................................... 35
Calculate....................................................... 37
Terrain/Shot Corrections.........................................38
PART VIII: forMOD MENU.................................. .............. 43
Input.......................................................... 43
Plot........................................................... 45
Export.......................................................... 46
PART IX: RESULTS MENU................................................ 47

Screen......................................................... 47
Printer......................................................... 48
Final Plate..................................................... 49
PART X: SYSTEM MENU.................................................. 51
Quit...............................51
Shell to DOS.................................................... 51
Help........................................................... 52
Install......................................................... 52
APPENDIX A: FORWARD MODELING..........................................Al
APPENDIX B: INVERSE MODELING..........................................Bl
APPENDIX 0:EXAMPLES TO ILLUSTRATE THE USE OF SEISMO..................Cl
3
CONVERSION FACTORS, NON-SI TO SI (METRIC) UNITS OF MEASUREMENT
Non-SI units of measurement used in this report can be converted to SI

(metric) units as follows:
Multiply By To Obtain
degrees (angle) 0.1745329 radians
inches 2.54 centimeters
feet 0.3048 meters
4
SEISMIC REFRACTION DATA PROCESSING SOFTWARE
SEISMO VERSION 2.7
PART I: INTRODUCTION
1. The software package "SEISMO" will perform seismic refraction data
reduction, interpretation, and presentation. SEISMO will do both forward and
inverse modeling of refraction data, calculation of results, display of
results, and final report figure presentation of results. The basic concepts
of seismic refraction surveying and the procedures for forward and inverse
modeling of refraction data are presented in Appendixes A and B, respectively.
2. This program is not presented as a state-of-the-art procedure for
seismic refraction data reduction and modeling. The program uses an algorithm
that has been in use for several years (time intercept method). Although more
complicated algorithms, such as the general reciprocal method, exist, these
algorithms also require a more sophisticated understanding of seismic theory.

In addition, the more complicated algorithms require more data collection.
The algorithm used in this program is a means of obtaining and reducing
seismic data, which will suffice for most engineering applications.
Manual Layout
3. Part I of this document gives an overview of SEISMO, the basic steps

to running the pLogram, hardware and software requirements, and the
installation procedure. Part II explains how to start the program, and
discusses the main and control menus, including their operation. Parts III -
X explain in detail how to use SEISMO to process data and obtain results.
These chapters explain the main menus along with all their options, giving a
step-by-step description of how each one functions in the program. Appendix A
gives the formulas and conventions used in the forward modeling routine.
Appendix B gives the formulas for seismic refraction depth calculations,
terrain corrections, and shot depth corrections. Appendix C gives examples of
field work performed and analyzed with the SEISMO program.
5
Typographic Convention
4. The following is a list of the typographic convention used

throughout this guide.
"press" Key indicated should be pressed.
<return> Enter or return key should be pressed.
"type" Type the information requested and press

enter or return.
<Esc> Press key labeled "Esc" on the keypad.
S"The control character indicating

that the control key (Ctrl) should be
pressed and held down while pressing the
second indicated key.
6
PART II: REQUIREMENTS
5. This section of the report gives an overview of SEISMO's features

and functions, and describes the hardware and software configuration the
system must have for SEISMO to run properly. The last portion of this section
explains in general terms how to use the program. The individual menus and
features of each menu are described in detail in subsequent parts of this
report.
Overview
6. The SEISMO program was written to provide an aid to analyzing and

presenting surface seismic refraction test data. The basic philosophy for
this program is to provide a user-friendly, yet powerful, tool that would
allow reducing data in a field environment with a portable computer ( no
pencils, paper, or calculator). This is not an expert system, in that it
depends on user knowledge and experience to model the data. With SEISMO one
can:
a. Perform both forward and inverse modeling.
b. Have complete control over the modeling of data:
Keyboard entry of velocities and intercepts.
Full-screen digitization of model from data.
c. Easily enter field data.
d. Perform editing of both input data and input models:
Full-screen editing of data.
Full-screen editing of model in both tabular and
graphic form.
e. Store data.
f. Have multiple choices of output:
Large 5 by 7 plot of data and model
Report-ready plot of data, model, and calculated
results.
Print of input model and calculated results.
Storage of finished plot to disk.
g. Perform arrival-time corrections:
Shot point depth.
Terrain variations.
h. Receive on-line help for SEISMO operations.
i. Have easy access to the Disk Operating System (DOS) environment
from within program.
7
Most of the functions have error-trapping capabilities that will alert the
user to errors or possible errors.
Basic Steps to Using SE(SMO
7. The following is a step-by-step procedure for the input, modeling,

and calculation of seismic refraction data. Each item first describes the
process in words and then in SEISMO command terminology.
a. Enter the field data, geophone distance versus arrival
time.
File Build
b. Display the data on the screen and digitize the model (or
alternately enter the model from the keyboard).
invMod Screen or invMOD Input
c. Calculate the model.
invMOD Calculate
d. Get printed or plotted output of the data and results.
Results Printer or Plot Plotter
e. Store the data and model before exiting program.
File Write
Hardware and Software Requirements
8. The SEISMO program was developed under Microsoft(MS)-DOS using

Microsoft Quickbasic version 4.5 and QuickPak Professional. Version 2.7
of the program supports differing hardware setups using the installation
routine to customize the program for the user's particular setup (see
paragraph 71).
General requirements IBM Personal Computer or "compatible."

256K Random access memory availab]e.
MS-DOS 2.0 or higher.
Plotter that interprets Hewlett Packard (HP) graphic
language (GL) commands.
Printer
Color graphics adapter (CGA), extended graphics
adapter(EGA), or very high resolution graphics
adapter (VGA) giaphics card.
SEISMO will run on a computer that has a math co-proccssor installed as well
8
as one that does not. The math co-processor is not necessary, since the
program was compiled as if a co-processor did not exist. This makes SEISMO
more flexible for varying computers and configurations. It does not, however,
noticeably slow the program execution by not utilizing a math co-processor,
since the program is not computationally intensive.
Plotters and Printers Supported
9. The plotter must be connected to either serial port 1 or 2 (coml or

com2). This requires the computer to have an RS-232-type serial port to
transmit the data to the plotter. All plotting commands coming from the
program are in HP-GL language. This requires the plotter to be either an HP
plotter or one that will interpret such commands. By using the option to plot
to a file (paragraph 33) there is the availability to produce plots on the HP
LaserJet or a similar device. The program will support any printer connected
to a parallel (LPT) port.
Software Requirements
10. The SEISMO program consists of one program disk. The program is
not write-protected and can be copied. 'le program can be run from either a
floppy or hard disk, but the files that end in the extension "pic" must be
present in the directory where the program is running. The files contained on
the SEISMO disk are listed below.
Name of File Purpose of File
SEISMO.exe Invokes the SEISMO

program
Scrnl.pic Control menu for data
input
Scrn2.pic Control menu for model

input
Scrn3.pic Control menu to

input/edit data
9
Name of File Purpose of File
Scrn4.pic Control menu for

plotter input

installation

plotter input
Helpl.pic Display of help screen
Ilelp2.pic Continuation of help

screen
10
PART III: PROGRAM OVERVIEW
Starting the Program
11. To begin the program, the user types "SEISMO" and presses return.
The program will then display a screen that has the program information
header. This screen will remain for 2 seconds before the main menu appears.
The information header contains the name of the program and the version
number. It also contains the author's names as well as phone numbers where
they may be reached in case of difficulties. Before the main menu appears,
the program queries the printer to determine if it is on-line. If the printer
is not ready to teceive output, the program will display a screen as shown in
Figure 1. The length of time the screen remains will depend on the type of
hardware used. The user chooses the appropriate response and continues with
the program. The program then establishes the system configuration and
several defaults. The plotter port is set to be coml, the printer port to be
iptl, and the units to be feet. These choices can be changed in the
installation routine discussed in Part X. The program also queries the
monitor and establishes the graphics capabilities (CGA, EGA, VGA, etc.).
PRINTER NOT READY:
TO TRY AGAIN < RETURN >
TO ACCEPT < Esc >
Figure 1. Printer error menu
11
The Main Menu
12. A display of the main menu is presented in Figure 2. The top

portion of the menu presents the programming information I-eader. It gives the
program name, a brief description of the program, and version level. At
the center of the screen is a line entitled 'Menus,' followed by the words
FILE, PLOT, EDIT, invMOD, forMOD, RESULTS, and SYSTEM. Each of these words
represents a different group of options aqailable in the p'ogram. Under the
word FILE a selection icon consisting of four red smiling faces should appear.
****** SEISMO ******

Seismic Refraction Data Analysis and Presentaion Software
******** v2.7 Apr 90 ********
To Select Menus UseT"rrow keys

Menus: I FILE I PLOT I EDIT I invMOD I forMOD I RESULTS I SYSTEM I
Options: @@
Build (new) <<< select option

Read (existing) enter Highlighted letter
Write (to disk)
Print (data)
Figure 2. The main menu
This selection icon (cursor) indicates the position on the Menus line. The
cursor can be moved zo the next heading by pressing the <right/left> arrow
keys or the <enter> key. The cursor wraps around whe - the end of the Menus
line is reached. In other words, if the cursor is under FILE, to move to
SYSTEM the user presses the left arrow once, and the cursor will "jump" to
SYSTEM. As the cursor is moved through the Menus choices, the options under
each are updated. The choices are easy to locate, they are on the left of the
screen under the word Options. To perform the option, the user presses the
letter corresponding to the highlighted letter in the option. This is stated
to the right of the options list. When an option is selected, the program
will branch to the control menu that corresponds to that particular option. A
12
description of the various options available under each menu is presented in
detail in the corresponding chapters.
Control Menus
13. After selecting an option from one of the menus discussed in

paragraph 11, the program branches to the control menu for that option. The
control menus for each option all follow the same rules. Movement within the
control menus and exit from the menus are consistent from one option to the
other. The control menus will have questions that require input by the
user from the keyboard. To the right of each question will appear a
highlighted box, where the answer to the question is entered. This box will
be referred to as a field. Two types of movement are allowed in the control
menu, movement within a field and movement between fields. An example of a
typical control menu is given in Figure 3. The following keys are active in
the control menus.
Movement between fields:

Key Action
Down arrow Moves the cursor to the next field
(this movement may be down in some
cases or to the right in others).
Up arrow Moves the cursor to the next field

(tL±is movement may be up in some
cases or to the left in others).
PgDn Moves the cursor to the last field

on the screen.
PgUp Moves the cursor to the first field

on the screen.
Enter Moves the cursor down or right to

the next field in the sequence.
Esc Aborts entry and exits the screen
13
Movement within a field:
Key Action
Any alpha/ Enters any character permitted by
numeric key the field, a number or letter.
Left arrow Movement to the left.
Right arrow Movement to the right.
Home Cursor moves to the far left of
the bo:.
End Cursor moves to the far right of
the box.
Ins Switches between insert and

typeover mode. On most monitors the
cursor appears as a block in the
insert mode and an underscore in
the typeover mode.
Del Deletes the character under the

cursor.
Backspace Movement to the left.
Space bar Movement to the right.
Enter Completes a field entry and moves

the cursor to the next field.
After entering a field, a message to assist in answering the question may be

printed at the bottom of the screen. This does not occur for every field, but
it does for most of the fields. Some of the fields already have answers in
them. These answers are default values that the program will use unless they
are changed. These answers can be changed by using the editing keys described
above.
14
PART IV: FILE MENU
14. This is the first menu in the Menus line. It is designed to be the
default menu when the program is first started. As such, the red selection
bar should appear beneath the word FILE. A list of options (begin with red
first letter) available for this menu should also appear. The options
available are:
Build (new)
Read (existing)
Write (to disk)
Print (data)
The screen should appear as that in Figure 2. If it does not, the user should
check the selection bar. If it is not under the word FILE, the <right/left>
arrow keys are used to position the cursor there. The user selects the option
to perform by typing the red letter corresponding to that option.
Build (New)
15. This option is invoked by typing the letter <b>. The control menu
for this option should appear and resemble that of Figure 3. As the name
implies, this option is used to build a new file. This would consist of
entering the time-distance field data for the refraction line. Before the
actual data are entered, however, there are some questions to answer.
[ NTROL MENU FOR DATA 1N1U
IDENTIFICATION:
UNE LENGTH:
NUMBER OF GEOPHONES FORWARD: REVERSE:
<Esc> exit screen

I '
Figure 3. Control menu for data input
15
Identification
16. This information is used to help identify the time-distance data
that will be entered. The identificationi will be written to disk along with
the data. This is to assist in recognizing the data at some future date. All
keys are active in this field.
Line length
17. The line length is defined to be the distance from shot point to
shot point. If only a forward shot is done, then the line length is the
distance from the shot to the last geophone. An example of the correct line
length, given a spread with 12 geophones spaced 10 ft* apart and a source at
both the forward and reverse end, spaced 10 ft from the nearest geophone, is
presented in Figure 4.
LINE LENGTH
0 OWNEUUEEUEEE. 0
GEOPHONES
SHOT POINTS
line length - 10 ft + (12-i)*i0 + 10 ft - 130 ft

Figure 4. Line length calculation
Number of geophones
18. This is the number of geophones in the spread. The program can
handle up to 48 geophones. There are two inputs to complete this question.
The user enters the number of forward geophones (48 maximum), and the number
of reverse geophones (48 maximum). Normally the number of forward geophones
will equal the number of reverse geophones for a particular spread since they
are usually coincidental. However, this need not be the case. To exit the
screen and begin data input the user presses the escape key as instructed at
the bottom of the screen.
* A table of factors for converting non-SI units of measurement to SI

(metric) units is found on page 4.
16
Data input screen
19. This is the screen where the actual field data are entered (time-
distance). The screen should appear as that in Figure 5. The screen contains
several fields, with headings of DISTANCE and TIME. The cursor will be in the
first field under the hegding DISTANCE. The user enters the necessary
distance and time data, and presses <enter> or the <up/down> arrow keys after
each entry. The cursor will then move to the next field. When a value is
entered in the distance field, the program uses that value to predict the next
distance value. The predicted distance will automatically be entered by the
program in the upcoming distance field. The user either accepts the predicted
distance value by pressing enter or the down arrow, or changes the value. The
program will then use the new value to predict the next distance value. To
change a value, the editing keys (Ins, Del, etc.) that are active inside a
field are used. The status message in the lower left corner of the screen is
updated to indicate which data point is being entered (or should be entered).
It also reports if the forward or reverse data set is being entered. In the
same location is a box indicating which screen is active and how many screens
are necessary to complete data input. In the lower right corner of the screen
is information about exiting the routine. To exit the input screen, the user
presses <Esc>. This will store -ne data in memory and return to the main
menu. There are two other options to help in manipulating the screen if
multiple screens are needed for data input. If more than 24 points are being
1
entered, t en multiple scieens are used. The user pages through the screens
by using the "'PgUp or ^PgDn keys.
Line length errors
20. At the conclusion of data input and upon exiting the routine, the
program will perform a check of the line length. The line length is
calculated using the number of geopaones and the distance between those
geophones. If the calculated distance in this check does not correspond to
that entered in the build file routine, then the program will print an error
message and correct the line length. The program then uses the corrected line
length for all subsequent operations.
17
DATA INPUT SCREEN
DISTANCE TIME DISTANCE TIME DISTANCE TIME DISTANCE TIME
_____________________< <Esc> exit screen

<OtrI PgUp> previous screen
<OtrI PgDn> next scree n
Figure 5. Data input screen
18
Read (Existing)
21. This routine is used to read an existing data file from disk (hard
or floppy). Upon entering this routine, a question appears that asks the name
of the file to access. There will also be a comment about the extension. All
data files written from SEISMO have a ".ref" extension added to them. This
will assist in recognizing data files produced from within SEISMO. The user
types in the name of the file (without extension) and pre:3ses <enter>. The
program will search for the file and read the data. If just the file name is
entered, the program searches in the current directory only. To retrieve data
from another directory or another drive, the user enters the complete path
with the file name. If an incorrect name is entered the program will print a
screen as shown in Figure 6.
FILE .REF DOES NOT EXIST

< RETURN > TO TRY AGAIN
< Esc > TO ABORT
< D > TO DIR ALL .REF FILES
Figure 6. File listing routine
The program prints that there are no ".ref" files at the location specified
and presents three choices. The user presses <return>, at which point the
program will ask for the file name again. Choice two is to press <Esc> and
19
abort the whole process, which then returns control to the main menu. The
third choice is to press < D > to get a listing of 1l ".ref" files in the
directory specified. If this option is selected, the SEISMO directory screen

(Figure 7) will appear. Enter the directory Lo st.arch for any ".ref" files.
By pressing <return> a listing of the current directory will appear. To
search another directory or drive, the user types in the full path name to
that directory including the ".ref" extension and then follows the directions
on the screen to input the file.
5 file(s) found
FILE1 .REF
FILE2.REF
FILE3.RER
FILE4.REF
FILES.REF
Filename: filel .ref
Use cursor keys to select file then press <ENTER>
Figure 7. List of matching files display
Write (To Disk)
22. This routine writes data to the specified drive\directory\file

name. Here again, the user does not enter an extension; SEISMO will add a
".ref" extension. If no drive\directory is specified, the file is written to
the current directory. The file name is limited to eight characters, not
counting the extension. All MS-DOS rules must be followed. The file contains
the following information.
20
title,#forward geophones,#reverse geophones,line length,#layers
distance(l),time(l)
distance(2),time(2)
distance(last),time(last)
forward velocity(l),forward intercept(l)
forward velocity(last),forward intercept(last)

reverse velocity(l),reverse intercept(l)
reverse velocity(last),reverse intercept(last)
The velocity and intercept information pertains to the inverse model which
will be discussed in Part VI. If the file already exists, a message will
appear to that effect, and the user follows the information on the screen to
continue.
Print (Data)
23. This routine will send a print of the selected data currently in
memory to the installed printer. There are three types of data printouts; the
uncorrected time-distance data, the terrain-corrected data, and the forward
model calculated data. The printouts will resemble those in Figures 8, 9, and
10, respectively. If there are no data in memory, the program will print a
message to that effect and return to the main menu.
21
SEISMIC DATA SUMMARY #
DATA FILE NAME : WSB4R7.REF

DATA IDENTIFICATION : wsb4r7
THERE ARE : 24 FORWARD POINTS 24 REVERSE POINTS
LINE LENGTH 50 ft
*** FORWARD POINTS *** *** REVERSE POINTS ***
Distance Time Distance Time

ft msec ft msec
2.0 4.0 2.0 4.0

4.0 7.0 4.0 7.0
6.0 8.0 6.0 8.5
8.0 8.5 8.0 9.0
10.0 9.0 10.0 10.0
12.0 11.0 12.0 11.0
14.0 11.5 14.0 12.0
16.0 15.0 16.0 12.5
18.0 14.0 18.0 14.0
20.0 14.3 20.0 15.5
22.0 15.5 22.0 16.0
24.0 16.5 24.0 16.5
26.0 17.0 26.0 17.0
28.0 18.0 28.0 17.5
30.0 18.5 30.0 19.0
32.0 20.5 32.0 20.0
34.0 21.0 34.0 21.0
36.0 22.0 36.0 22.0
38.0 24.0 38.0 23.5
40.0 25.0 40.0 25.0
42.0 27.5 42.0 27.0
44.0 28.0 44.0 28.0
46.0 29.0 46.0 30.0
48.0 30.0 48.0 31.0
END OF DATA
MODEL SAVED WITH DATA
*** FORWARD MODEL *** *** REVERSE MODEL ***
Velocity Intercept Velocity Intercept

ft/sec msec ft/sec msec
542 0.0 512 0.0

1836 3.7 1780 4.0
Figure 8. Printout of uncorrected time-distance data
22
##### SEISMIC DATA SUMMARY ###W
DNTA FILE NAME : WSB4R7.REF

THERE ARE 24 FORWARD POINTS 24 REVERSE POINTS
LINE LENGTH : 50 ft
e** FORWARD POINTS *** *** REVERSE POINTS ***
Dist UTine CTime Cor CVel Dist UTime CTime Cor CVel
ft msec msec ft ft/sec ft msec msec ft ft/sec
2.0 4.0 3.9 0.5 527.0 2.0 4.0 4.0 0.0 527.0
4.0 7.0 7.9 0.5 527.0 4.0 7.0 7.0 0.0 527.0
6.0 8.0 8.9 0.5 527.0 6.0 8.5 8.5 0.0 527.0
8.0 8.5 9.4 0.5 527.0 8.0 9.0 9.9 0.5 527.0
10.0 9.0 9.9 0.5 527.0 10.0 10.0 10.9 0.5 527.0
12.0 11.0 12.8 1.0 527.0 12.0 11.0 11.9 0.5 527.0
14.0 11.5 13.3 1.0 527.0 14.0 12.0 12.9 0.5 527.0
16.0 15.0 16.8 1.0 527.0 16.0 12.5 13.4 0.5 527.0
18.0 14.0 15.8 1.0 527.0 18.0 14.0 14.9 0.5 527.0
2C.0 14.3 16.1 1.0 527.0 20.0 15.5 16.4 0.5 527.0
22.0 15.5 17.3 1.0 527.0 22.0 16.0 16.9 0.5 527.0
24.0 16.5 18.3 1.0 527.0 24.0 16.5 17.4 0.5 527.0
26.0 17.0 17.9 0.5 527.0 26.0 17.0 18.8 1.0 527.0
28.0 18.0 18.9 0.5 527.0 28.0 17.5 19.3 1.0 527.0
30.0 18.5 19.4 0.5 527.0 30.0 19.0 20.8 1.0 527.0
32.0 20.5 21.4 0.5 527.0 32.0 20.0 21.8 1.0 527.0
34.0 21.0 21.9 0.5 527.0 34.0 21.0 22.8 1.0 527.0
36.0 22.0 22.9 0.5 527.0 36.0 22.0 23.8 1.0 527.0
38.0 24.0 24.9 0.5 527.0 38.0 23.5 25.3 1.0 527.0
40.0 25.0 25.9 0.5 527.0 40.0 25.0 25.9 0.5 527.0
42.0 27.5 28.4 0.5 527.0 42.0 27.0 27.9 0.5 527.0
44.0 28.0 28.0 0.0 527.0 44.0 28.0 28.9 0.5 527.0
46.0 29.0 29.0 0.0 527.0 46.0 30.0 30.9 0.5 527.0
48.0 30.0 30.0 0.0 527.0 48.0 31.0 31.9 0.5 527.0
END CIF DATA
-** FORWARD MODEL **A *** REVERSE MODEL ***
Velocitj Intercept Velocity Intercept

5.2 0.0 512 0.0

1836 3.7 1780 4.0
Figure 9. Printout of terrain and shot point corrected data
23
SEISMIC DATA SUMMARY #####
DATA FILE NAME : WSB4R7.REF

THERE ARE 24 FORWARD POINTS 24 REVERSE POINTS
LINE LENGTH : 240 ft

ft msec ft msec
10.0 11.1 10.0 11.1

20.0 18.4 20.0 18.4
30.0 22.4 30.0 22.4
40.0 26.4 40.0 26.4
50.0 29.0 50.0 29.0
60.0 30.3 60.0 30.3
70.0 31.7 70.0 31.7
80.0 33.0 80.0 33.0
90.0 34.3 90.0 34.3
100.0 35.7 100.0 35.7
110.0 37.0 110.0 37.0
120.0 38.3 120.0 38.3
130.0 39.7 130.0 39.7
140.0 41.0 140.0 41.0
150.0 42.3 150.0 42.3
160.0 43.7 160.0 43.7
170.0 45.0 170.0 45.0
180.0 46.0 180.0 46.0
190.0 47.0 190.0 47.0
200.0 48.0 200.0 48.0
210.0 49.0 210.0 49.0
220.0 50.0 220.0 50.0
230.0 51.0 230.0 51.0
240.0 52.0 240.0 52.0
END OF DATA

900 0.0 900 0.0

2500 10.4 2500 10.4
7500 22.4 7500 22.4
10000 28.0 10000 28.0
Figure 10. Printout of forward model calculated data
24
PART V: PLOT MENU
24. The second menu in the menus line contains the plotting options.
When the cursor is beneath the word PLOT, the screen should appear as that in
Figure 11. There are four options available from the PLOT menu. The options
are Screen, Plotter, File, and Overlay plot. Each option is discussed in
detail below. The user selects the option by typing the highlighted letter
corresponding to the option.
****** SEISMO ******

Seismic Refraction Data Analysis and Presentalon Software
******** v2.7 Apr 90 ********
.......................................................................................................-
To Select Menus Use7--rrow keys

Options:
Screen < select option

Plotter enter Highlighted letter
File
Overlay plot
Figure 11. Main menu (plot)
Screen
25. This option is used to plot the time-distance data to the screen.
It is also used to display the model, if one exists, or enter the model, if
one does not exist. The screen should appear as that presented in Figure 12.
If a CGA screen is present, the presentation will be in black and white. If
an EGA or VGA screen is present, the presentation will be in color. The
primary function of this routine is to be able to see data before making a
hard copy. However, it is possible to input a model with this routine as
explained in detail in paragraph 45.
25
CURSOR ACTIVE <AX> CLEAR MODEL <Esc> MAIN MENU
W+
DISTANCE
DIST TIME I XTIC YTIC I-LAYER VELOCITY I LAYER VELOCITY
Figure 12. Screen model -digitization
Plo tter
26. This routine is used to make a hard-copy plot-of the data-. This
would include the time-distance values as well as the model, if one exists.
This -routine does not present any calculated results (depths to interfaces- and
true velocities). The routine produces a 5- -by 7-in. plot of the -input data.
This routine -was added -to give the user a working copy of -the data and model
before finalizing the results. Upon -entering-this option-, the program
branches to the plotter -control menu which is -described in paragraph 28. An
example plot is presented in Figure- 13. At the top of the plat is a banner
that contains information relative to the plot. The first line contains the
title entered from the plotter control menu and described in the following
sections. The second line contains the name -of the file- where the information
is contained-. This is very useful as a means of keeping track of where data
are located.
Plott:er control menu
27. This menu is -used to set parameters -required by the plotter. The
plotter control menu should resemble that in Figure 14. There are eight
questions on the plotter control menu that require input from the user. These
26
CD
pi 3
z
m4
w I
0 Cl 0
NJ 41
U)>-4
J-)
0 0
ii
IL"-
r%H
1 0
H
xI
zw
HNI
27H
questions involve sce.iing and labeling for the plot. Each item is described
in detail in the sections fol.,2.wing.
T~LE: CONTOL MENU FOR PLOTTER INPUTj1

TITLE: .. .
X.AXIS LABEL:
Y-AXIS LABEL:
X-TIC INTERVAL
Y-TIC INTERVAL
DO YOU WISH TO PLOT MODEL:
SEND OUTPUT TO: FILE NAME: = .pit
< Esc > exit screen
< Ctrl PgDn > begin plot
Figure 14. Control menu for plotter input
Title
28. This is the title that will actually be displayed on the plot.
There are no restrictions as to what may be entered in the title field. All
keys are active for this particular field. However, there are limitations on
how long the title can be. Typing outside the field, denoted by the
highlighted box beside the word TITLE, is not allowed. There are a total of
50 character spaces in the title field. If a title on the plot is not
desired, the user leaves the field blank.

X-axis and y-axis labels
29. These are the labels for the axis on the plot. Both these fields
should have default values in them. The x-axis is normally labeled "Distance,
ft" and the y-axis is labeled "Time, msec." These labels can of course be
changed, and if metric units are being used, will have to be changed. Here
again, the field will accept any character, and input is limited only by the
size of the field.
X-tic and y-tic intervals
30. These variables control the number of tics on each respective axis.
Default values for these fields are given in feet. Default values are not
28
intended to be used for all cases. For example, if there is a line length of
1,200 ft and selected default x-tic intervals of 10 ft are used, then 120 tics
will appear across the x-axis. On the plot this will look like a solid dark
line after the tic labels are added. In this case, the user should select a
larger tic interval for the axis.
Model option
31. There is a choice in this menu to either plot the model or not plot
the model. This is convenient in that it allows the user the option of
sketching in his own model. This may be necessary in cases where the model
cannot be reasonably discerned from the screen digitizing routine. The
answer to this question is a "yes" or "no" and is entered by typing <y> or
<n>. The default selection is "yes."
Plot destination
32. The next two questions in the plotter control menu concern the
destination of the plot. The plot can be sent directly to the plotter, or it
can be sent to a file for later use. This option is convenient if a plotter
is not readily accessible. Remembering that SEISMO was developed as a useful
tool for field data collection and interpretation, there will be times when a
plotter is not available. Plots can still be made, however, and saved to a
file to be plotted at some later date. The answer to the question "SEND
OUTPUT TO" is <p> for plotter or <f> for file. The field has a default answer
of "p." If <f> is selected, then the user inputs a file name or uses the
default file name. Normally there will be a default name in the "FILE NAME"
field. This name will coincide with the name entered for data storage from
the "Build" routine. The name can be changed, but all plot files have a
".plt" extension. This extension will be added by the program and need not be
typed in. If the user does not want to plot to a file, then this option can
be left either blank or at the default value.
Begin the plot
33. To complete the plotter control menu input and continue with the
plot, the user presses < ^
PgDn>. The plotter will be activated and a plot
should be made. This requires the plotter to be connected to the appropriate
serial port and to be on-line. If the plotter is not ready, an error will be
generated and an error message displayed. To exit the screen and abort the
plotting process, the user presses <Esc>.
29
File (To Plotter)
34. This option will read a plot file from the specified
drive\directory\file name and send the data to a plotter. The same routine
used to read a data file and discussed in paragraph 22 is utilized here. The
functions and operations are exactly as described in that section, the only
difference being that the user is working with ".plt" files here instead of
".ref" files. The user should be sure that the plotter is connected and on-
line before performing this option. This routine handles both the 5-by 7-in.
working plots discussed in paragraph 27 as well as the final plate plots
discussed in paragraph 66.
Overlay Plot
35. This routine will allow the plotting of more than one set of data
points on the same plot. This refers to hard copy plots only, and not to
plots on the screen. If more than one set of data is to be plotted to the
same plot, this is the routine to follow. The user reads in the original data
set and makes a plot as described in paragraph 27 or paragraph 66. Leaving
the plot on the plotter, the user reads in a second data file. The user moves
to the PLOT menu and selects this option (overlay plot). A message to the
effect that the overlay plot is working will be displayed on the screen. The
data will then be plotted on the plot already in the plotter. This can be
done as many times as the user wishes for as many data sets as desired on one
plot. However, all data sets are plotted on the scale (distance versus time)
dictated by the original plot. This routine applies to data only, not data
and models, only one model per plot.
30
PART VI: EDIT MENU
36. The edit menu is used to edit both data and inverse models. Upon
selecting this menu (EDIT) the screen should appear as does Figure 15. These
are not input options, but are used to edit existing data or models.
****** SEISMO ******

******** v2.7 Apr 90 ********
------------------------------------------------------------------------------------
To Select Menus Use-arrow keys
Menus: I FILE I PLOT I EDIT I lnvMOD I forMOD I RESULTS I SYSTEM I
Options:
Data < < < select option

Inverse model enter Highlighted letter
Figure 15. Main menu (edit)
Data
37. These are the time and distance data tha, -ere entered from the
FILE menu (build or read) and described in the sectior3 above. An example of
the menu to edit data is shown in Figure 5. This is the same menu used for
the data input routine, and follows the guides expressed in paragraph 20.
Inverse Model
38. The second edit option involves the inverse model. For a complete
description of the inverse model and the means of input, see Part VI. The
menu used to edit the model is the same as that described in paragraph 41 for
inputting the model from the keyboard. The menu should appear as that of
Figure 16. The control menu for model editing (and input) is described in
detail in Part VI which can be consulted for the function of each field and
31
the appropriate response. If a model has been entered to edit, then that
model should be displayed in the control menu. The number of layers,
velocities, and time intercepts will be in their appropriate fields. To edit
the model, the keys as described in paragraph 12 can be used to move to the
field and perform the correction. If adding or deleting a layer, the user
should be sure that the number in the NUMBER OF lAYERS field reflects the
change.
CONTROL MENU FOR MODEL INPUT

NUMBER OF LAYERS:
MODEL--- > (B)OTH (F)ORWARD (R)EVERSE
-FORWARD R.REVERSE
APPARENT VEL TIME INTER APPARENT VEL TIME INTER
LAYER 1: E R EKK =
LAYER12: ZEE=
LAYER 3:
LAYER 4:
LAYER 5:
<Esc> exit screen
Figure 16. Control menu for aodel input
32
PART VII: invMOD MENU
39. This menu concerns the various options related to the inverse model
as well as the terrain and shot point corrections. The menu should appear as
that in Figure 17. The program will do two types of modeling, inverse and
forward. Inverse modeling is the entry of apparent velocities and intercepts
with subsequent calculation of the model (depths to interfaces and true
velocities). Forward modeling is the entry of depths and true velocities with
subsequent calculation of apparent velocities and intercepts. The normal
process of data reduction would be the input of the inverse model. This
chapter deals with the input and -iinipulation of the inverse model.
****** SEISMO ******

SeismI(, Refraction Data Analysis and Presentalon Software
******** v2.7 Apr 90 ********
-----------------------------------------------------------------------------------
To Select Mern :,
'JseT-rrow keys
Menus: I FILE IPLOT I EDIT I invMOD I forMOD IRESULTS I SYSTEM
Options:
Snput (from keyboard) < < < select option
creen (digitize)
Calculate enter Highlighted letter
Terrain/shot corrections
Figure 17. Main menu (invMOD)
Input
40. This option concerns inputting the inverse model from the keyboard.
7pon selction of tLi option, the co.ro. menu for . acd
in Figure 16 will appear. This menu has two questions followed by enough
fields to input up to five layers forward and reverse.
33
Number of layers
41. This is the number of layers that will be used to model the data.
The program will handle up to five layers. For most cases there will be an
equal number of forward and reverse layers. If, however, there is a case
where this is not true, the modeling will have to be done in two steps. This
would involve modeling the layers for the forward traverse and calculating
them, then modeling the layers for the reverse traverse and calculating them.
This is accomplished by using the "MODEL" option from the model input screen.
Such a case could occur, for instance, if there is a layer that pinches out
(consult Appendix A for a discussion of this and other such occurrences).
Model
42. SEISMO can model three separate cases: both a forward and reverse
traverse, forward traverse only, and reverse traverse only. There are times
when only a forward shot of a seismic line is performed. If this is the case,
that can be modeled in SEISMO. It is not recommended that this be a common
field practice, and the user should always try to conduct both a forward and
reverse shot for the seismic line. By doing a forward shot only, the program
can only interpret the interfaces as being horizontal with no indication oF
possible dipping layers. When entering a forward traverse model only, the
program will make the reverse traverse model equal to the forward model input.
It is this feature that will allow the input model to not have equal forward
and reverse layers, by splitting the model into two parts. The interpretation
will then have to be combined, understanding, of course, that it is based on
assumed horizontal layers. The default answer for this field is "B,"
indicating that both a forward and reverse traverse will be modeled.
Velocities and intercepts
43. The velocities that are entered are the apparent velocities,
meaning that they are those determined from the field data. The velocities
must increase with depth, to prevent any errors. This means that layer 1
velocity is less than layer 2 velocity, layer 2 velocity is less than layer 3
velocity, and so on. The velocities are entered in feet/second or
meters/second. The time intercepts are the projections of the velocities
(straight line segments) back to the time axis. For the first layer this
value will always be a zero. The time intercepts are always entered in
milliseconds.
34
Screen
44. This routine is used to input the model from the screen by
digitizing breakpoints. This screen is shown in Figure 9. The cursor for
this screen is a "+" sign and should be located in the center of the screen.
To determine where the breakpoints are, the user draws straight line segments
through the data points. The breakpoints to be digitized are located at the
points where the line segments intersect (change in slope). This concept is
presented in Figure 18. The points are digitized by moving the cursor to the
point, and pressing the appropriate key. The keys active for this screen and
the function they perform are described below. The movement of the cursor is
controlled in steps determined by the number keys. To change the step size,
the user presses a number from one to nine.
KEY FUNCTION
Right arrow Moves the cursor to the right.
Left arrow Moves the cursor to the left.
Up arrow Moves the cursor up.
Down arrow Moves the cursor down.
Enter Returns the cursor to the middle of

the screen.
Fl-F5 These function keys are for entering the

forward breakpoints, this allows entry
of five layers.
F6-FlO These function keys are for entering the

reverse breakpoints, this allows entry
of five layers.
Esc The escape key will exit the routine at

any time.
^x This will clear the model but not the

data.
35
VELOC1TY LAYER TWO
FOR WARD BREAK POINT

LAYER ONE
VELOCITY LAYER ONE
DISTANCE
Figure 18. Breakpoint concept
As mentioned above, the escape key is used to exit the routine. It is

possible to exit the routine at any time, but the user should be aware that
the model on the screen at the point of exit will be saved. After completion
of model digitization, the user presses <Esc> to return to the main menu.
A
Warning! If the user presses x to clear the model digitized on screen, this
will also clear the model read from disk. For example, suppose a file was
read from disk, then the data and model were displayed on screen with the
"screen" routine above. The forward layer is modified slightly, but not
desired to be saved. By pressing AX the entire model is erased. To recall
the model, the user reads the file from disk again.
Information panel
45. At the bottom of the screen are displays that provide useful
information to aid the user in this screen. The location of the cursor is
displayed at all times under the headings "dist" and "time." This is the x,y
position of the cursor relative to the scale used to make the plot. The tic
marks on the plot are not labeled, but the tic mark spacings are given under
the headings xtic, ytic. More times than not, these will be some unusual
number such as xtic = 31.3. The user should not be concerned about odd
36
scales, as the location of the cursor can always be found by checking the
cursor location displays. The displays that show the digitized model input
layer velocities are located on the right side of the information panel.
There should be four headings; layer, velocity, layer, and velocity. Then a
breakpoint is digitized, two velocities are being changed: the one behind the
breakpoint and the one in front of the breakpoint. The display updates those
velocities and the corresponding layer, which is especially useful when
editing the model.
Onscreen editin5 of the model
46. After entering d model, whether by keyboard or screen digitization,
SEISMO is extremely useful in editing that model. The user has complete
control of where the line segments, reprasenting velocities, should be
located. To edit a model, the user moves the cursor to where the new
breakpoint (between iinR segments) should be and presses the appropriate
function key. Again, fnrward breakpoints are controlled by functien keys 1-5,
and reverse breakpoints by function keys 6-10. As the model ic edited, two
things will be apparent. First, the display at the lwer right of the screen
will print the new velocities for the appropriate layer. Secondly, the old
line segments will "disappear" and the new line segments will be drawn in.
The program does not restrict where a breakpoint may be entered.
Calculate
47. This option is used, iri the. csse of the inverse model, to calculate
the true velocities and depths to interfaces from the input model. If there
is no model in memory, the program will print out a message to that effect and
the calculation will not take place. If there is a model present, the program
will print out a message that it is calculating. When the calculation is
complete, control is returned to the main menu. The calculation is performed
using the time-intercept method described in Appendix A. The calculation
routine has an error checking function that alerts the user of possible
problems. Theoretically, the forward and reverse total travel time
(reciprocal time) for each layer should be the same. The program computes
this and alerts the user to any discrepancies. For each layer, the travel
time forward is compared with the travel time reverse. If the d~fference
between the two is greater than 10 percent, the program will print a message
37
to that effect and pause the program. The user presses <enter> to continue
the calcu]hcion. It is then up to the user to either accept the error or
rework the model. It is suggested that the model be reworked, such that the
10-percent travel time error is not surpassed.
Terrain/Shot Corrections
48. This option allows the correction of data for varying shot point
depths and varying geophone changes in elevation. In order to utilize this
routine, the number of forward geophones must equal the number of reverse
geophones, and they must be coincidental. A refraction line should be run in
an area as nearly horizontal (constant elevation) as possible. When this is
not possible, this routine will correfct the measured arrival times to a common
datum plane. The datum plane is arbitrary and is chosen by the user. The
subsequent corrected arrival times will be the times as if the geophones were
located along a horizontal plane extending from the datum point. The program
will also correct the arrival times for varying shot depths, effectively
correcting the arrival times to a surface source. Prior to the start of the
corrections menu, a display will appear that resembles Figure 19. In order to
use the routine, the velocity of the first two layers and the first geophone
receiving refracted arrivals for the forward and reverse spread must be known.
The display (Figure 18) will print the first and second layer velocity
presently in memory. These velocities come from the calculations performed on
the inverse model. If they do not exist, the program will print zeros as the
velocities. The user may accept the velocities printed, input new velocities,
or exiL the routine. The geophone receiving the first refracted arrival can
be obtained by plotting the raw data and observing the breaks in slope. The
following paragraphs explain the function of the options in the correction
menu. The correction menu will appear as shown in Figure 20.
Scale
49. At the left of the screen is the elevation scale. This scale
initially ranges from 5 to -5 in increments of one half. For most normal.
Operation-, 1i C!1--- will be sufficient to cover the elevation changes.
.- To
change the scale, the user presses AS, and answers the question that appears.
The entered scale factor is multiplied by the original scale to obtain the new
scale. The new scale will then be displayed in place of the old scale.
38
THIS ROUTINE REQUIRES :
TRUE VELOCITIES OF
FIRST TWO LAYERS
GEOPHONE RECIEVING FIRST REFRACTED ARRIVAL
FIRST LAYER TRUE VELOCITY SECOND LAYER TRUE VELOCITY

0 0
< return > TO ACCEPT VELOCITIES
< Esc > TO ABORT
< I > TO INPUT NEW VELOCITIES
>>>> SELECT
ENTER FIRST LAYER TRUE VELOCIT

ENTER SECOND LAYER TRUE VELOCITY
GEOPHONE RECEIVING FIRST FORWARD REFRACTED ARRIVAL
GEOPHONE RECEIVING FIRST REVERSE REFRACTED ARRIVAL
Figure 19. Pre-correction menu
ELEVATION < Esc > EXIT W/O SAVE

ft < ^A > SAVE & APPLY CORR
< AQ > SAVE W/O APPLYING
4 o
3 SHOT PT
2 CORRECTION'
< ENTER > TO ACCEPT
-
INITIAL SCALE -5 TO 5 VELOCITY FOR CORRECTION
CURRENT SCALE = INT * SF VELOCITY =
.4 SF=
< S > TO CHANGE SCALE FACTOR <A V > TO CHANGE VELOCITY
TERRAIN AND SHOT DEPTH CORRECTION MENU
Figure 20. Correction menu
39
At the bottom left of the screen is printed information to assist in
determining the status of the scale. The initial scale is printed since all
subsequent scales are referenced to this scale. Then the current scale
calculation procedure is displayed, to remind the user of how the present
scale was obtained. The current scale factor, which is changed as the scale
changes, is also displayed. Example: The scale is as presented initially (5
to -5), the user presses the ^s keys, and enters the new scale factor. The
initial scale is multiplied by the scale factor and the new scale is
displayed. If a scale factor of "2" was entered, then the new scale will
range from 10 to -10 in increments of one. The scale can be changed to any
factor desired by utilizing the change scale function. There is no limit as
to how small or large the scale will become; however, there is a limit to the
degree corrections can be reasonably applied to refraction data. There is a
small horizontal arrow pointing to the zero elevation location. This arrow is
used to select the elevation that will be entered for the shot point or
geophone. Recall that this number is the elevation above or below the datum
plane. The arrow is controlled by the up/down arrows on the keypad and will
"wrap" at the top or bottom.
Datum plane
50. Located at approximately the center of the screen is the datum
plane. There should be two "x's" which represent the shot points for the
refraction line. Between the x's are small squares that represent the
geophones. There should be as many squares as there are geophones for the
spread being corrected. The maximum number of geophones that the program
can handle from this menu is 48. Upon entering the screen, there will be a
vertical arrow pointing to the first shot point. This arrow marks the shot
point or geophone that is being corrected. It is controlled by the
<right/left> arrow keys on the keypad and will perform the "wrap" function.
Elevation corrections
51. Located in the center of the screen above the datum plane is
information that will be useful to the user as the corrections are applied.
The first line informs the user where along the datum plane he is located. If
the arrow points to a shot point, the message line reflectb that; if it points
to a geophone, the message reflects that. The second line gives the
correction as entered by the user. To enter the corrections, the user places
the vertical arrow over the shot point or geophone to be corrected. The user
40
places the horizontal arrow at the elevation to be entered and presses
<enter>. If proceeding from left to right across the screen, the program will
update the vertical arrow. After pressing <enter>, the vertical arrow is
moved to the next location on the datum plane. After all the corrections are
entered (or during entry), the entries can be reviewed. As the vertical arrow
4
is moved across the datum plane, the informat on entered for each point is
displayed. These can of course be edited by re-entering a new elevation.
Velocities used in corrections
52. In the lower right-hand corner of the screen is located information
about the velocities used for the corrections. The velocity can be changed by
pressing ^v. Each geophone has its own velocity used to make the correction,
and that velocity is displayed for each geophone as the vertical arrow points
to it. The velocity is in feet or meters per second. This feature can be
used to correct for lateral changes in the top layer velocities.
Exit module
53. At the top right of the screen are located the exiting routines.
The first option is to press <Esc>, which is an exit without saving. This
option is used to abort the process and return to the main menu. When this
option is performed, no information is retained from the correction menu.
The second option is to press ^a to save the corrections and apply them to the
original data. Be aware that once the corrections are applied, the original
data are lost and the corrected data become current in the programs memory.
It is recommended that two sets of data be kept, one file containing the
original data and a second file containi.. the corrected data. The third
option, pressing q, will simply save the corrections in the programs memory
but will not apply them to the original data. When exiting the program the
corrections are of course lost. All three options will return control to the
main menu after performing their indicated function.
Recommended procedure
54. If there are data to be corrected for elevation changes in shot
points or terrain, this is the recommended procedure. The user enters the
time-distance data and a model (from keyboard or by screen digitization). The
user then calculates the results of the model and saves the original
uncorrected data. He selects the terrain/shot point corrections option and
enters the correction data as described above. He uses the Aa option to exit
the routine and apply the corrections. The user replots the data (corrected
41
now) and edits the model to fit the corrected data. This information is
saved under a different file name to keep it separate from the original
(uncorrected) data.
42
PART VIII: forMOD MENU
55. See paragraph 40 for a definition of the forward modeling process.

When this menu is selected, the screen should appear as that in Figure 21.
This menu will present the options available to the user to perform forward
modeling. Forward modeling is useful in that it gives a picture of what can
be expected from given field conditions. With this option, the user can have
a tool to assist in planning and laying out the test procedure. This is also
useful in determining if a layer can be detected by the seismic refraction
technique. The general procedure is to enter the forward model, calculate,
and then either plot the results or export them to the invMOD routines. The
program calculates results consisting of the apparent velocities and intercept
times for each layer. The program gives output for both the forward and
reverse directions. If there is no dip angle for any of the interfaces, then
the velocities and intercepts for both the forward and reverse directions will
be the same.
****** SEISMO
******** v2.7 Apr 90 ********
To Select Menus Use--rrow keys

Menus: I FILE I PLOT I EDIT I invMOD I forMOD I RESULTS !SYSTEM I
Options:
Input <<< select option
Plot
to iMod)
Export (vel & int enter Highlighted letter
Figure 21. Main menu (forMOD)
Tnni"t
56. When this option is selected, the control menu for forward model
input will be displayed and it should appear as shown in Figure 22. This menu
43
is used to input the data necessary for the program to compute results of the
forward model. The menu input fields are: the number of layers, the line
length, distance interval to calculate arrival times, and the layer
information. The individual sections of this menu will be discussed in detail
in the following sections.
MEUFOR FORWARD MODELING DATA INP~lI
NUMBER OF LAYERS: 11 I LINE LENGTH:

DISTANCE INTERVAL TO CALCULATE ARRIVAL TIME:
LAYER VELOCITY LAYER THICKNESS DIP ANGLE
r --------------------------------------------------------
1ST I ]
2ND
3RD -- I I:
4TH
5TH F
<Esc> exit screen
Figure 22. Menu for forward modeling data input
Number of layers and line length

57. This is the number of layers to be modeled and the length of the
geophone spread. The program will handle up to five layers and any spread
length. The line length is as described in paragraph 16.
Distance interval
58. This value is used by the routine to calculate corresponding
arrival times over the length of the line at the selected interval. These
calculations are based on the model input. If this value is left blank, then
no arrival time calculation will be performed.
Layer information
59. The box in the center of the screen (Figure 22) contains the fields
for inputting th1 lay- inf.rmation. There are three types of information for
each layer; the layer velocity, the layer thickness and the dip angle. The
layer velocity is the true velocity in feet or meters per second. The layer
thickness is self-explanatory and entered in consistent units. The user
44
should not enter a thickness for the last layer, since that thickness is
unknown. The dip angle is entered in degrees, and is taken as the angle of
the layer interface from the forward end to the reverse end. A positive angle
represents a downward slope, whie a negative angle represents an upward
slope. The dip angle for the first layer is that of the ground surface and is
set by the program to be 0 deg. Most of the errors in using this routine come
from entering too large a dip angle for the line length entered. See Appendix
B for an explanation of forward modeling calculations. Upon exiting the
routine, the calculation of the forward and reverse model will be performed
and displayed on screen. The display will resemble Figure 25, with the input
data being true velocities, depths to interfaces, and dip angles, and the
calculated data being interface velocities and arrival times.
Plot
60. The plotter control menu should appear as shown in Figure 23. The
options are the same as described in paragraph 28.
L CONTROL MENU FOR PLOTTER INPUTJ
rITLE:
X-AXIS LABEL:
Y-AXIS LBEL:
X-TIC INTERVAL
Y-TIC INTERVAL
DO YOU WISH TO PLOT MODEL:
SEND OUTPUT TO: FILE NAME: .pit
< Esc > exit screen
< Ctri PgDn > begin plot
Figr 9 Plotter input menu
45
Export
61. This option is used to transport the calculated apparent velocities
and intercepts into the inverse modeling routine. As such, the forward model
input data must first be calculated; otherwise, there will be nothing to
export. The program will print a warning if exportation is attempted before
calculating. Exporting a forward model is a useful technique to give the user
a starting inverse model that can be edited to best fit the data.
46
PART IX: RESULTS MENU
62. The results menu is the means of obtaining calculated output data
from the program. The screen should appear as that in Figure 24. There are
three options available from this menu; screen, printer, and final plate.
Each one will be discussed in detail in the following sections. The results
obtainable from the program are calculated results and can come from either
the inverse or forward modeling routines.
****** SEISMO *****

******** v2.7 Apr 90 ********
To Select Menus Use: arrow keys

Options:
Screen < < < select option
Printer enter Highlighted leter
Final Plate
Figure 24. Results menu
Screen
63. This option presents a table of the calculated results from either
the forward or inverse modeling routines to the screen. Figure 25 presents an
example of such a screen. There are two blocks of information, one that
contains the input data and one that contains the calculated results. If
results from the inverse modeling routine are being displayed, then the input
data will be apparent velocities and intercepts. The calculated results will
be true velocities and depths to interfaces. If results from the forward
modeling routine are being displayed, then the input data will be true
velocities and depths to interfaces. The calculated results will be apparent
47
velocities and intercepts. The user presses <enter> to continue the program.
TITLE
* INPUT DATA *** *** COMPUTED SEISMIC PROFILE *
FORWARD REVERSE LAYER TRUE DEPTH

LAYER VEL TIME VEL TIME # VEL FOR REV
# ft/s msec ft/s msec ft/s ft ft
1 500 0.0 500 0.0 1 500
2 2500 2.7 2500 2.5 2 2500 1.5 1.5
3 5000 5.6 5000 5.5 3 5000 3.0 3.0
I I
L --------------------------------- a ----------------- --------
PRESS < RETURN > TO CONTINUE ?
Figure 25. Example output to screen
Printer
64. This option sends the results to the printer. The printout will
appear exactly as that described in the paragraph above and represented in
Figure 26. ...............................................
TITLE
* INPUT DATA ***
FORWARD REVERSE
LAYER VEL TIME VEL TIME
# ft/s msec ft/s msec
1 500 0.0 500 0.0
2 250 2.7 2500 2.5
3 5000 5.6 5000 5.5
*** COMPUTED SEISMIC PROFILE ***
LAYER TRUE DEPTH

# VEL FOR REV
ft/s ft ft
1 5O
2 2500 1.5 1.5
3 5000 3.0 3.0
Figure 26. Example output to printer
48
Final Plate
65. This consists of a plot of both the data and model as well as a
listing of the calculated results. There is also a graphical representation
of the resulting profile. This plot was designed to be used as a final figure
in a report. When this option is selected, the program branches to the final
plate plotter control menu. This menu is shown in Figure 27, and operates as
described in paragraph 28. The only difference in this screen and that
described in the above-mentioned paragraph is the absence of the plot model
option. The model is always plotted and therefore it does not appear as an
option in the plotter control menu for the final plate. An example of a final
plate is shown in Figure 28.
CONTROL MENU FOR PLOTTER INPUT
TITLE:
X-AXIS LABEL:
Y-AXIS LABEL:
X-TIC INTERVAL
Y-TIC INTERVAL
SEND OUTPUT TO: FILE NAME: .plt
< Esc > exit screen

< Ctrl PgDn > begin plot
Figure 27. Control menu for final plate plot
49
I.-n
0 'U 0 1).0
IL9E Wi x c
w 0 Ix
aSrl I' N UN
r
L 04J M t H S c
0 .0 41i
a 0
r 0 ul 0
L3u m DS 43
L IL U. 1 To I
H
2 \)C M'~ 0IL 0 3 41
2L0r (U1vt I a c
J > ToE.u
X0 ri
w co
4J
-4
-4
114
C-1O
500
PART X: SYSTEM MENU
66. This menu contains options relating to the program environment and
interfacing with the DOS operating system. The screen should appear as shown
in Figure 29. There are four options available from this menu; quit, shell,
help, and install. Each option will be described in detail below.
****** SEISMO ******

v2.7 Apr 90 ********
To Select Menus Use: arrow keys
Menus: j FILE I PLOT I EDIT I invlOD I forMOD I RESULTS I SYSTEM I

Options: @
Quit < < < select option

Shell to DOS enter Highlighted letter
Help
Install
Figure 29. Main menu (System)
Ouit
67. This option is used to conclude the SEISMO session and exit the
program. This function will return control to the DOS operating environment.
If no data file was written during the session, a warning will sound along
with a question about quitting. This i to alert the user that data will be
lost if the program is exited before saving the input data or model.
68. This option allows temporary exit from the program to perform DOS
commands and then re-enter the program. The data are conserved while outside
51
the program. To return to the program, the user type exit and presses
<enter>.
69. This option will give on-line information, which is also contained
in this manual. It is greatly reduced, but provides important information
about movement within the main menu, selection of options, control menus, and
inverse and forward modeling. The help information is presented in two
screens. Instructions to exit the menu or proceed to the next menu are
presented at the bottom of the screen. Screen one is shown in Figure 30, and
screen two is shown in Figure 31.
Install
70. The install menu is used to change certain default parameters in

the program. The menu is shown in Figure 32, and consists of the following
fields; port for the plotter output, port for the printer output, and the
units.
HELP MENU
" Move thru the main menu by using the <right/left> arrow keys
* Select options by typing the first letter of the option (red)
" Control menus

-move between fields by pressing <enter>, <up/down> arrow keys
- move to last field by pressing <PgDn>
- move to first field by pressing <PgUp>
- move within fields by using <right/left> arrow keys
all editing keys are active within a field
- exit control menu by pressing <Esc>
<Esc> exit screen

<Ctri PgUp> previous screen
<Ctrl PgDn> next screen
Figure 30. Help menu
52
It HELP MENU CONTINUED I
* Inverse modelling routine active keys

^
F1 - enter layer 1 for F6 - enter layer 1 rev L - cursor to left edge
^
F2 - enter layer 2 for F7 - enter layer 2 rev R - cursor to right edge
F3 - enter layer 3 for F8 - enter layer 3 rev " U - cursor to top edge
F4 - enter layer 4 for F9 - enter layer 4 rev " D - cursor to bottom edge
^
F5 - enter layer 5 for F1O- enter layer 5 rev M - cursor to middle
" <PgUp> - move up 15 pixels " <right arrow> - move right 15 pixels
^ <PgDn> - move down 15 pixels " <left arrow> - move left 15 pixels
Arrow keys move indicated direction 1 pixel " X - clears model
...........................
............................................ .....................................................................
* Forward modelling instructions

.velocities for each layer are 'true' velocitieb
- thicknesses are layer only, not cumulative from ground surface
- do not enter a thickness for the last layer <Esc> exit screen
-the dip angle for layer one should be zero <Ctrl PgUP> previous screen
<Ctri PgDn> next screen
Figure 31. Help menu continued
PORT FOR PLOTTER
PORT FOR PRINTER
UNITS
<Esc> exit screen
Figure 32. Install menu
53
APPENDIX A
FORWARD MODELING
Reference:
1. Butler, Gangi, Wahl, Yule, Barnes, 1982. Analytical and Data

Processing Techniques for Interpretation of Geophysical Survey
Data with Special Application to Cavity Detection, Miscellaneous
Paper GL-82-16, U.S. Army Engineer Waterways Experiment Station,
CE, Vicksburg, Ms.
EXCERPT FROM REFERENCE 1
Computer Modeling and Interpretation of

Seismic Refraction Data
119. Two problems can be identified related to the seismic

refraction method: (a) the direct problem, and (b) the inverse problem.
The direct problem involves the determination of arrival times as a
function of distance from a hypothetical source for various seismic
events and a specified model; the model consists of layer velocities,
layer thicknesses, and interface dips. For a given set of arrival
time versus distance data from a refraction survey, the inverse problem
involves the determination of the parameters of a subsurface model
from the data itself. Two FORTRAN computer programs have been developed
which solve the direct and inverse problems of the seismic refraction
method, REFRUIR and REFRINV, respectively. REFRDIR is useful as a
modeling tool for planning field surveys, e.g., geophone spacing and
A3
survey line length required to investigate a possible subsurface model.
REFRINV is very useful for refraction interpretation, where the data
indicate the possibility of multiple, dipping layers.
120. The theoretical basis or the two programs (REFIU.R and
REFRINV) is a set of equations relating to refracted waves (see, for
example, Officer, 1957). These equations are given in the following
sections describing the programs. They consist of: an equation for
the horizontal phase velocity at the surface, equations corresponding
to Snell's law at each interface, and an equation to compute the
intercept times for the refracted waves.
121. Because the layer interfaces are assumed to be planar,
th
the travel time for the wave refracted along the n interface (or
th
top of the n layer) is given by
tn (x) = T o,n + x/v nn (14)
Thus, given the horizontal phase velocity, vn , and the intercept time,
thn
To,n for the n refraction, the travel time can be computed at all
source-to-receiver distances, x
REFRDIR
122. The horizontal phase velocity and the intercept time are
computed for each refraction event using the program RIRIR. In this
case, the P-wave velocities, the interface dips, and the layer "thick-
nesses" must be given. This program gives the intercept times and the
horizontal phase velocities for both the direct and the reverse profile.
The reverse profile is that array layout in which the shot is on the
opposite end of the array relative to the direct profile.
123. In computing the intercept times and the horizontal phase
velocities, program REFRDIR uses the angles, 0±in , that the rays make
with the normals to the interface. These angles, as well as other
important parameters (for example, the layer "thicknesses"), are
illustrated in Figure 60.
124. REFRDIR is a modeling program. it assumes: (a) thaL
the velocity variation in the earth increases monotonically with depth,
A4
SOURCE RECEIVER
\~1 14o;-
1 ,
+H 132"
81+4 8 Z,
n=2 +823....
+ ,
. 2
n=4" o.4
Figure 60. Assumed geometry for seismic refraction paths in

common-strike, multiple dipping layers
(b) that it can be approximated by layers of constant velocity but with

varying thicknesses, (c) that the interfaces between the layers are
planar, and (d) that the interfaces have the same strike (see Figure 60).
If REFRDIR is given an input model containing a velocity inversion
(i.e., a decrease in velocity with depth), it will print an error message
to that effect and stop.
125. The input data for the program are: (a) the number of
laycers in the model, (b) the layer thicknesses relative to the direct-
and .... .....ofil sh ..... ( PUT a MMTNUS ; see rgure 1) , and
(c) the dip angles of the interfaces. The air/ground interface is the
first one and it is assumed to be horizontal. Any velocity and distance
A5
units may be used, as long as they are consisten't, but the (dip) angles
must always be given in degrees. REFRDIR computes and prLnts out:
(a) the intercept times, and (b) the horizontal phase velocities of
the refractions from each interface for tile direct and reversed profiles.
The following equations are used by REFRDIR:
sin 6nl,n = an-i/a n (15)
= -- (16)
n-l,n n-l,n n-l,n
sin -i-l,n (a ~-/a!)

i sin (6±,n +
-i i) (17)
+
++ a/sin(0 + + ) (18)
Vn s (8
,n -
+ + n-2 +
T:- (211-- t/a )cos 0 + (I +/ (Cos 0.
on n-l n-i n-l,n
+ cos 0. ) (19)
The a's are the layer velocities, the H's are the layer thicknesses,
the v's are the horizontal phase velocities, and the T's are the
intercept times. The superscripts + or - refer to the direct and
reversed profiles, respectively. See Figure 60 for the indexing and
angle definitions.
126. For a given layer, computation is started by using Equa-
tion 15 to solve for sin e . This, in turn, is used in Equation
17 to generate sin 6- through sin 0- The phase velocities
n-2,n l,n "
and intercept times are then compt-ed using Equations 18 and 19. To
clarify the use of these equations, the calculations for the general
case of three interfaces have been carried out in Appendix i. A listing
of REFRDIR is also given in Appendix I.
127. Input REFRDIR is extensively docwuuented ind prompting
messages appear before each input. All input is in free-field format.
The input sequence is given below:
A6
Input No. 1: NLAYRS
NLAYRS--numbers of layers in the model
Input No. 2: SPRDLN
SPRDLN--the spread length, i.e., the length of the
refraction survey line
Input No. 3: (ALPIIA(N), DELTA(N), 1IPLUS(N), N=l, NLAYRS)
ALPILA(N)--velocity in layer N
DELTA(N)--dip angle (in degrees) of the N th inter-
face, which is the top surface of layer N
DELTA (1) = dip of ground surface of model =
0
HPLUS(N)--thickness of N th layer for the direct
profile (see Figure 60)
Input No. 3 is repeated for each layer.
Input No. 4:
Input "1" if only forward profile to be plotted, and
input "2" to plot both forward and reverse profiles.
128. Output. REFRDIR output consists of tabulated and plotted
versions of the results of the computation. The tabulated output gives
the horizontal phase velocities (apparent velocities) and intercept
times for the forward and reverse profiles. For the plotted output,
the apparent velocities and intercept times are used to define line
segments, and forward and reverse time-distance plots are produced.
PLOT2 (Appendix B) is used to produce the plots.
129. Example. Figures 61 and 62 illustrate the use of REFRDIR
to solve a direct seismic refraction problem. The hypothetical model is
shown in Figure 61, and the spread length was selected as 150 m. Com-
puted time intercepts, see Figure 62, are indicated on the plot in
Figure 61. This example is important because it illustrates that
intuitive predictions of interface dip from apparent velocities or of
apparent velocities from dip may be incorrect. For the example in
Figures 61 and 62, intuition and experience might lead one to expect that
the apparent velocity in the forward direction for interface 3 would be
greater than V 3 ; however, the results from REFRDIR (see Figure 62) show
that this is not the case.
A7
320.
0.30609012.IO
Liz
3000./
Fiue. TheIae moe adfrt rivl im-iSU C 1~

.prdue30. 60. 0. 20.lEO
A8~
INPUT NUMBER OF LAYERS IN MODEL
=3
INPUT LENGTH OF SHOT SPREAD

= 150
FOR EACH LAYER INPUT A LINE ENTRY WITH THE FOLLOWING DATA
SEPARATED BY COMMAS: LAYER VELOCITY,DIP OF INTERFACE,
LAYER THICKNESS
= 500,0,15
= 1500,5,15
= 3000,-5,
VELOCITIES ARE IN METERS/SEC (OR FT/SEC), TIMES ARE IN

MILLISECONDS AND LAYER THICKNESSES ARE IN METERS (FT).
THERE ARE 3 LAYERS IN THIS MODEL
LAYER NO. VELOCITY DIP (DEGREES) H+ H-

1 500. 0. 15.0 2.0
2 1500. 5.0 15.0 40.9
3 3000. -5.0
***** CALCULATED RESULTS *****
THE HORIZONTAL PHASE VELOCITIES AND THE INTERCEPT TIMES FOR

THE DIRECT (+) AND REVERSED (-) PROFILES.
INTERFACE NO. PHASE VEL+ PHASE VEL- T+ T-

2 2000. 1207. 56.6 7.5
3 3896. 2498. 76.4 55.1
Figure 62. Tabular output from REFRDIR for case shown in Figure 61
A9
APPENDIX B
INVERSE MODELING
Reference:
1. Department of the Army, 1979. Geophysical Exploration,

Engineer Manual 1110-1-1802, Office, Chief of Engineers,
Washington, D. C.
EM 1110-1-1802
31 May 79
APPENDIX B
SEISMIC REFRACTION SURVEYING
B-I. Principle. The refraction method consists of measuring the travel

times of compressional and sometimes shear waves generated by an impul-
sive energy source to points at various distances along the surface of
the ground. The energy source is usually a small explosive charge or a
sledgehammer blow. The energy is detected, amplified, and recorded so
that its time of arrival at each point can be determined. The instant
of the impact or explosion, the "zero time," is also recorded along with
the ground vibrations arriving at the detectors (geophones). The raw
data, therefore, consist of travel times and distances, the travel time
being the interval between the zero time and the instant that the de-
tector begins to respond to the disturbance. This time-distance infor-
mation is then processed to obtain an interpretation in the form of
velocities of wave propagation and struture of the subsurface strata.
The process is illustrated schematically in figure B-1. All measure-
ments are made at the surface of the ground, and the subsurface struc-
ture is inferred from interpretation methods based on the laws of wave
propagation.
80- .
60•
U S
40 .
E
- 20
FTr2ve0I-times first impulse
of
0 100 200 300 400 500

2,500 ft/sec Distance., ft
- 8)Recording equipment
_ , Explosive charge in sh lo.oe eet r
,Detecto0 fts
02
9, 100 ft/sec <; 0.. : . ... -...

Sei smi c ray path for
Tirst arriving signal"s
Redpath 2 0
Figure B-1. Schematic of seismic refraction survey (after
B-1
EM 1110-1-1802
31 May 79
a. Wave Propagation in Elastic Media.
(1) A disturbance in an elastic medium propagates through two

types of waves: compression waves, in which the deformations of the
medium consist of alternating compression and rarefaction; and shear
waves, in which the deformations of the medium are pure shear. At the
surface of an elastic medium (ground surface), a third type of wave, a
surface (Rayleigh) wave, is also generated. The amplitude of the dis-
placements in a Rayleigh (R) wave is relatively large at the surface but
dies out rapidly with depth. The compression wave is commonly called
the "primary" wave, or the "P-wave," while the shear wave is called the
"secondary," or "S-wave." In layered systems, other types of waves are
generated along the boundaries between layers and resemble surface waves.
However, they are not important to refraction surveys and are not con-
sidered in this manual.
(2) Fluids, or viscous media, differ from elastic media in that

they cannot propagate S-waves, and only P-waves are observed directly.
(3) The P-wave is propagated at the highest velocity and is fol-

lowed by S- and R-waves, in that order. As a result, the wave train
received by a detector at a large distance from the source resembles
that shown in figure B-2.
R- WAVE
z P-WV S-WAVE
I-I
U 0
TIME
Figure B-2. Hypothetical wave train recorded by a geophone at

the surface of a homogeneous elastic medium at some distance
from the source (prepared by WES)
(4) In refraction surveying, only the P-wave is usually used.

Thus, the travel time is determined by the arrival of the P-wave at the
detector, or geophone. This is always the first arrival.
b. Refraction.
(1) General. The propagation of seismic energy through elastic

media and across boundaries between layers is described by essentially
B-2
EM 1110-1-1802
31 May 79
the same laws that govern the propagation of light rays through trans-
parent media. According to Huygens' Principle, waves in a homogeneous
medium radiate from a point source as expanding spheres, and every point
on a wave front is the source of a new wave that moves away from it as
an expanding sphere. Thus, a signal can travel between two points over
an infinite number of different paths, each requiring a different amount
of travel time. A ray path represents the shortest travel time and is
thus the path of the first arriving signal. It is everywhere perpen-
dicular to the wave front. The refraction or angular deviation that
a ray undergoes when passing from one material to another depends upon
the ratio of the transmission \elocities of the two materials. The law
that describes the refraction of light rays in terms of transmission
velocities, known as Snell's law, also applies to the refraction of
seismic waves in linear elastic media. This law, together with the
phenomenon of "critical incidence," is the physical foundation of seis-
mic refraction surveys.
(2) Refraction and reflection. When a P-wave meets a boundary

between two media that have different elastic properties, four distinct
new waves are generated at the point of contact. These new waves are:
(a) a P-wave reflected back into the first medium, (b) a P-wave re-
fracted in the second medium, and (c and d) similarly reflected and
refracted S-waves. The newly generated S-waves, traveling at lower
velocities than the P-waves, do not affect observations of first ar-
rivals and can be ignored. The total energy in the four new waves is
equal to that in the incident wave. However, since the energy of vi-
bration is proportional to the square of the amplitude, the sum of tht
amplitudes of the reflected and refracted waves may be larger than the
amplitude of the incident wave.
(3) Snell's law. Snell's law and critical incidence are illus-
trated in figure B-3, which shows a medium with a velocity v, , under-
lain by a medium with a higher velocity v2 . According to Snell's law,
the relation between the angle of incidence i of the incident P-wave
and the angle of refraction r of the refracted P-wave is given by
sin i Vl1
s- (B-1)
sin r v2
where i and r are measured with respect to a perpendicular to th-

interface. More generally, this relation gives the directions of the
ray paths for all four of the waves generated at the interface. If one
substitutes for v 2 the velocity of either the P-wavu ur thI Z-wave in
either of the two media, then r is the corresponding angle of reflec-
tion or refraction.
B-3
EM 1110-1-1802
31 May 79
SOURCE
RAY PATHS
RAY PA THS
LAYER 1
VELOCITY = VELOCIY
va
vANGLE = =CRITICAL
OF
1
INCIDENCE
CP
LAYER 2
VELOCITY = v
2
Figure B-3. Refraction of ray paths across a

boundary between two elastic media (prepared
by WES)
(4) Critical refraction.
(a) The critical angle of incidence a is the angle of incidence

for which r = 90 deg, so that the refracted ray follows the boundary
between the media. For angles of incidence smaller than the critical
angle, most of the compressional energy is transmitted into the higher
velocity medium. Where the angle of incidence is greater than the
critical angle, the energy is almost totally reflected back into the
lower velocity medium. At critical incidence, sin r is unity; so
the critical angle of incidence is given by
sin a = v (B-2)
2
(b) The particular case of the critical angle of incidence is

fundamental to the derivation of the formulas for refraction explora-
tion. The critically refracted ray travels along the boundary between
the two redia at the higher of the two velocities, and it continually
generates in the lower velocity layer seismic waves, called head waves,
that depart from the boundary at the angle of critical incidence (i.e.
are rr c.ted bacl, nto the lower velocityylayer). h angle of crit-
ical incidence may therefore also be called the angle of critical
refraction.
B-4
EM 1110-1-1802
31 May 79
(c) Figure B-4 is a wave-front diagram, showing the positions of

the front of the disturbance at times t, t 2 , t3 , and so on, as it
propagates through a two-layer medium. The diagram shows that detectors
at distances greater than some critical distance xc from the source
will receive the refracted signal (head wave) as the first arrival,
while nearer geophones will first receive waves transmitted directly
through the surface layer.
SHOT
2 HEAD
3 WAVE
7-7
t8
Figure B-4. Positions of wave front in a two-layer medium at

times tl, t2 ,...(prepared by WES)
(d) The critical distance, often called "crossover distance" in

the geophysics literature, can be identified from a plot of travel time
of the first arrival versus distance (fig. B-5). For the simple, ideal-
ized case shown, the curve representing the direct wave is a straight
line with a slope equal to the reciprocal of the velocity of the surface
layer. For those ray paths refracted through the second layer, fig-
ure B-5 shows that the distance traveled in the surface layer is the
same for all geophones. Therefore, the travel-time difference from one
geophone to the next is the time required for the wave to travel in the
lower layer along a horizontal path whose length is the same as the
distance between the two geophones. The curve representing the refracted
wave is thus a straight line with a slope equal to the reciprocal of
the velocity of the wave in the second layer. The critical distance
is indicated by the break in slope.
(5) Determination of layer thickness. At a point located exactly

at the critical distance from the source, the direct and refracted
waves arrive at the same time. One can write expressions for the travel
times to this point by each of the two paths and, by equating the times,
solve for the depth D1 to the top of the second layer, in terms of
Xc , with the following result:
B-5
EM 1110-1-1802.
31 May 79
TIME, T i 2,.,.
CRITICAL DISTANCE, x
c
01,x
DISTANCE
SHOT
F
V2
(IN az= V
Figure B-5. Simple two-layer case

with plane, parallel boundaries, and
corresponding time-distance curve
(after Redpath 2 0 )
D1 = x2 c vv 22 +- Vl
1 (B-3)
12 2 + vl
Equation B-3 is valid for the simple system shown in figure B-4 or B-5,
i.e., a semi-infinite medium overlain by a single layer of lower veloc-
ity and uniform thickness, and for the depth to the first interface in
a multiple-layer system. The actual field condition may be more com-
plicated in several ways, such as irregular interfaces, dipping inter-
faces, and more complicated velocity distributions. The treatment of
such cases will be discussed in paragraph B-4. In every case, it relies
on the concepts of the path of the first arriving signal and of the
critical angle of ray incidence.
B-2. Apparatus.
a. Seismic Units. A seismic unit consists basically of ampli-

fier(s) and associated electronic circuitry, geophones, seismic cable,
B-6
EM Iii0-I-i$0!
31 May 79
seismic energy source, blaster, if the energy source is an explosive,

and recording device. The amnplifier's function is to boost the signal
detected by the geophones to a level suitable for input to the record-
ing device. Sophisticated seismic units may include, in addition to
amplifiers, filters (high-cut and low-cut), 60-Hz reject circuitry, a
test circuit for geophone continuity, timebreak circuitry, and coarse
and fine amplifier gain controls. Most seismic units are battery
operated, either on 6 or 12 volts. Commercially available seismic unitz
are many and varied; the choice of the best unit depends on the appli-
cation for which it is intended and involves factors such as frequency
of use, length of seismic lines, type of seismic source to be used, and
the complexity of the anticipated project.
(1) Single or multichannel seismic units are readily available.

Seismic units with more than 24 channels are not common but can be
obtained on special order.
(a) The normal single-channel seismic units are very portable and
easy to operate. They are used primarily with nonexplosive seismic
sources and for seismic lines no longer than 100 to 150 ft. In some
single-channel seismographs, "signal enhancement" is used to improve
record quality. This is done by a memory circuit, which algebraically
sums and stores the records of seismic signals produced by repeated
operaticns of the seismic source. The repeated part of the record,
which represents the useful signal, builds up with repeated summation,
while random noise tends to cancel, so that the signal-to-noise ratio
is improved. This feature improves performance in noisy environments
and increases by about a factor of two the length of seismic lines that
can be run with nonexplosive sources.
(b) Multichannel seismic units can either be very portable and

simple to operate or otherwise, depending primarily on the amount of
signal conditioning equipment included. Although multichannel units
work well with either explosive or nonexplosive sources, the recording
device that would ordinarily mate with the unit may impose limitations
on the type of seismic source that can be used. Signal enhancement is
also available for multichannel units. In addition, a test circuit for
geophone continuity and the timebreak circuitry should be included on
all multichannel units.
(2) In selecting a seismic unit, amplifier gains should be con-

sidered. A 90-decibel (db)* full-scale amplification is fficze,.
* A gain of +f ini r P t -,
output signal over the input signal. A gain of 90 db, therefore, mean::
an amplification ratio of 104 5, or about 32,000, in signal amplitudes.
B-7
EM 1110-1-1802
31 May 79
A signal that is too weak to be observed at 90 db generally cannot be

distinguished from the ambient noise. Intermediate selection of gains
from 0 to 90 db should be in steps no coarser than 6 db. The timebreak
circuit should provide to the recording device a time break (zero-time)
signal, which is obtained from the rupture (opening) of the bridgewire
of the electric initiating device. Filters are particularly useful
when seismic tests are conducted near high-voltage lines, machinery, or
other sources of extraneous noises or motions. High-cut filters block
out frequencies higher, and low-cut filters frequencies lower, than a
chosen value. For example, if the low-cut filter is set at 25 Hz and
the high-cut filter at 80 Hz, all signal components with frequencies
below 25 or above 80 Hz would virtually be eliminated from the recorded
signal. Three selections of filters on the low-cut side (such as 10,
25, and 45 Hz) and three on the high-cut side (50, 80, and 120 Hz) are
usually sufficient.
(3) A geophone test circuit measures the coil resistance of each

geophone. This provides for detection of such conditions as a loose
connection between a geophone and the seismic cable, a foreign material
(e.g., a sand grain) preventing proper contact between geophone and
cable takeout, or a short circuit in the cable or geophone.
b. Recorders. The various display or recording devices may in-

clude timing lights for display of travel time in binary form, a direct
digital readout of travel times, a seismic signal and time marker dis-
play on a cathode ray tube (CRT), or a recording of the time marker,
seismic signal, and a set of timing lines on photographic film or
oscillograph paper.
(1) Single-channel seismic units usually use timing lights, a

direct digital readout, or a CRT display. Some recording devices with
the CRT display have a camera attachment or a plug-in strip chart re-
corder so that a permanent record of the seismic signal can be obtained.
Where timing lights or direct digital readouts are used, the only
permanent record is in the field notes made by the operator.
(2) The recorders used with multichannel seismic units are oscil-
lographs or magnetic tape recorders (analog or digital). An oscillo-
graph normally includes: (a) a galvanometer for each recording channel;
(b) an oscillator to produce timing pulses, displayed on the record is
timing lines; and (c) in other than the simpler seismic units that make
a record on a stationary photographic film, a paper drive with sensi-
tized recording paper. The paper drive should be adjustable, so that
aL l:ast two and prefer ably thrOC r-cording speeds (about 15, 30, and
50 in./sec) are available. The speed adjustment is necessary to pro-
vide a time scale that is suitable for the types of site materials
B-8
EM 1110-1-1802
31 May 79
and the geophone spacings used. Timing lines should be at intervals on

the seismic record of not more than 10 msec.
(3) Galvanometers should have a broadband frequency response with

a range of at least 0-300 Hz. On some seismic recorder systems, the
galvanometers require periodic checks for alignment. This is done by
activating the galvanometers with an externally supplied voltage pulse,
obtaining a recording, and checking the time indicated by each galva-
nometer for the pulse. Any time differences noted in comparing the re-
sponses of all the galvanometers in the recorder indicate galvanometer
misalignment, which can be corrected with an alignment tool.
c. Geophones.
(1) General. In refraction seismic surveying, geophones are

generally vertically oriented, velocity-type transducers. The sensing
elements usually consist of a magnet, rigidly attached to the geophone
housing, and a surrounding spring-supported coil. Relative motion of
the magnet and the coil produces an electrical signal. Firm contact
with the ground is provided by a spike in the base of the geophone hous-
ing. A ground conductor may be included in the geophone and is desirable
for use in environments (especially arid regions) in which static elec-
tricity may cause noise problems. Further, consideration should be
given to sensitivity, durability, frequency response characteristics,
damping, and impedance matching with the seismic unit. There are trade-
offs among these factors, and the best choice for a particular applica-
tion usually involves some compromise. For example, the more sensitive
a geophone is, the heavier and less durable it is.
(2) Sensitivity. A sensitivity of about 1.0 volt/in./sec is ad-

equate for most refraction work and can be achieved in a geophone that
is relatively rugged and light in weight.
(3) Frequency response. Natural frequencies of the geophones

should range from 4.5 to 14 Hz (8-Hz average) and have a reasonably
uniform frequency response up to about 300 Hz. At frequencies below
the natural frequency, there is a sharp drop in the sensitivity; there-
fore, geophones should be selected on the basis of the lowest frequency
that may have to be recorded. Good sensitivity to the higher frequen-
cies is desirable because it contributes to the sharpness of the re-
sponse to the first arriving seismic signal and thus to precision in
picking the arrival times.
for the coil to stop moving after the excitation ceases and so helps
B-9
31 May 79
to inhibit response to noise, but it also diminishes sensitivity. Gen-

erally, geophones for refraction work should be undamped.
(5) Impedance. Electrical impedance matching between geophone and

seismic unit is required because a serious mismatch (more than an order
of magnitude difference in impedance) will result in excessive signal
loss at the interface. In most cases, geophones with coil resistances
in the 100- to 1000-ohm range will provide satisfactory matching of
impedance.
(6) Waterproofing. The geophone housing should be waterproofed;

in addition, geophone connectors should not be vulnerable to damage by
water and should be made of a rust-resistant material.
d. Seismic Cables. A seismic cable has a set of conductors that

carry the electrical signals from the geophones to the seismic unit.
The conductors are housed in a jacket that can be made of such mate-
rials as neoprene, rubber, plastic, or urethane. The number of con-
ductors needed depends on the number of data channels (i.e. the number
of geophones) to be recorded. This also dictates the number of takeouts
(connection points for geophones) required on the cable. Takeouts
should be waterproof and rust- and corrosion-resistant. The cable
length, spacing and number of takeouts, and type of connectors on the
ends of the cable can be specified by the purchaser. The cable should
be at least 200 ft long between the connector and the first takeout on
each end to ensure that sufficient distance can be placed between the
seismic source and the seismic unit. Cables may be subjected to numer-
ous detrimental effects, such as extreme heat and cold, water, snow,
ice, abrasion (by rock, cactus, sagebrush, and other abrasive materials),
and careless handling. Specifications for a seismic cable should in-
clude such characteristics as diameter, weight, flexibility, tensile
strength, and resistance to temperature extremes, abrasion, scraping,
cutting, tearing, cracking, and moisture. Seismic cables, as well as
geophones, should have a ground conductor to minimize noise or cross-
feed problems that may be encountered in arid regions.
e. Seismic Energy Sources.
(1) Nonexplosive sources. The energy for a seismic signal may be

provided by sources such as a sledgehammer, a drop weight, air or gas
guns, or a hydraulic force between an anchor and an inertial mass.
Most commonly used are sledgehammers and drop weights, because of their
simplicity and portability. They are usually used with a striking
plate made of metal or wood. The time of impact (zero-time) is obtained
from a microswitch attached either to the source or the striking plate,
or from a geophone placed close to the point of impact. These sources
deliver relatively little energy, as compared with explosives, and
B-10
EM 1110-1-1802
31 May 79
seismic lines are generally limited to 100 to 200 ft in length (with
corresponding depth of investigation about 30 to 70 ft). However, with
the use of signal enhancement seismographs, this distance and depth
range can probably be increased two-fold. Other types of nonexplosive
seismic sources (air and gas guns, etc.) are more complex and require
auxiliary equipment, such as air and gas tanks, compressors, and
generators.
(2) Explosive sources. The performance of explosives is charac-

terized primarily by strength (energy content), density, and detonation
velocity. Strength refers to the percentage by weight of nitroglycerin
(NG) in a straight dynamite (one which has only NG as the explosive)
and usually ranges between 20 and 90 percent, the balance being pri-
marily an inert absorbent, such as diatomaceous earth. Other explosives
are given strength ratings referenced to the straight dynamite standard.
The density of dynamites is commonly expressed as the number of 1-1/4-
by 8-in. cartridges contained in a 50-lb case, which ranges from 85 to
205, while the density of other explosives is expressed in grams per
cubic centimetre and ranges from 0.4 to 2.0. The detonation velocity
is the velocity at which the shock wave of the explosion travels through
a column of explosives, usually from 4,000 to 26,000 ft/sec for various
kinds of explosives. These three properties contribute most to the
force and power output of an explosive. To obtain P-wave data for engi-
neering applications, the best performance will be given by explosives
with the highest values of strength, density, and velocity. Generally,
explosives can be categorized as dynamites, primers, and blasting
agents, all of which are used extensively in refraction seismic work.
In this application, electric devices are universally used to initiate
the detonation.
(a) Electric initiating devices. These devices are the means by

which the main explosives are initiated and the zero time is obtained.
The primary types are electric blasting caps and exploding bridgewire
detonators. Since accurate determination of the zero-time is extremely
important in refraction seismic work, the in.tiating device should be
instantaneous, i.e., there should be no time lag between the rupture of
the bridgewire and the detonation of the device. Exploding bridgewire
detonators (which require a special firing power supply) meet this cri-
terion, as do instantaneous "seismograph-type" electric blasting caps.
Either of these two types of initiating devices may be used satisfacto-
rily in refraction seismic surveys, but delay caps should be avoided.
(b) Dynamites. Most dynamites are the NG-type, although some

dynamites are non-NG. AlL dynamites are cap sensitive, i.e., they can
be detonated by electric initiating devices. Their performance for
refraction seismic test use ranges from poor to good. The user should
B-11
EM 1110-1-1802
31 May 79
refer to manufacturers' specifications to determine which dynamites have

characteristics that are desirable in seismic use (strength, 50 to 90
percent; density, 110 cartridges or higher; and velocity, at least
14,ooo ft/sec). Although dynamites are easily obtained and can be an
excellent P-wave source, they have severe safety drawbacks, particu-
larly the NG-type. For this reason, the use of dynamite for refraction
seismic testing is not recommended. Unsafe characteristics of dynamites
include sensitivity to shock and friction, very high flammability, and
generation of toxic fumes. Other undesirable characteristics in most
dynamites include low resistance to water (wetting desensitizes them so
that they will not detonate) and susceptibility to freezing (upon thaw-
ing they become very unstable and dangerous to handle).
(c) Primers. Primers contain a relatively safe non-NG explosive

and can be used to initiate noncap-sensitive explosives (blasting
agents). They may also serve as the main explosives. All primers are
cap sensitive, and their performance characteristics are excellent
(strength, 60 percent and higher; density, 1.4 grams/cm 3 and higher;
velocity, 20,000 ft/sec and higher). A common primer used in refraction
seismic work is detonation cord. Explosive manufacturers' specifi-
cations should be consulted for the characteristics of various primers.
Generally, primers are not as readily available as dynamites and are
relatively expensive. Since their excellent performance and relative
safety outweigh these factors, primers are highly recommended for re-
fraction seismic work.
(d) Blasting agents. These products consist of a mixture of non-

explosive ingredients and, therefore, are not classified as explosives.
They cannot be detonated by electric initiating devices but can be made
to detonate by using high-strength primers. The performace of blast-
ing agents as seismic sources ranges from poor to good; however, explo-
sive manufacturers make a seismograph type of grade of blasting agent
(commonly known as nitro-carbo-nitrate) that is satisfactory for seis-
mic use. Manufacturers' specifications should be referred to for char-
teristics of blasting agents. The seismic-type blasting agents may not
be as readily available as dynamites. However, they are considerably
cheaper and much safer. For these reasons, they are recommended for
seismic use, in conjunction with primers, when a main explosive consist-
ing of more than 1 lb is required. A related product, available com-
mercially under various trade names, is used frequently in refraction
seismic testing. The product consists of two components, a flammable
liquid (nitromethane) and a powder (ammonium nitrate), in separate con-
tainers. They are not classified as explosives until the components are
mixed, shortly before being used; therefure, the two separte components
may be transported and stored without restrictions. Safety tests con-
ducted on mixed specimens yielded no detonations from impact or friction
B-12
EM 1110-1-1802
31 May 79
sensitivity, or rifle bullet tests, but the mixed product is cap sensi-
tive, so that it can be detonated by electric initiating devices. A
typical agent of this type has a strength of 70 percent, a density of
1.1 grams/cm 3 , and a detonation velocity of 18,000 ft/sec. This product
is highly recommended as an explosive source for refraction seismic
work.
f. Blasters. A blaster is a device that provides an electrical

pulse to initiate the detonation through an electric blasting cap. The
best and most reliable blaster is the capacitive discharge type. Such
a blaster uses dry cell batteries to charge a bank of capacitors, which
then discharge the electrical energy to the detonator. The capacitive
discharge blaster should deliver as much current (at least 0.5 amp) as
is practically possible to the detonator. For seismograph-type electric
caps, current should be from 0.5 to 4.0 amp. In addition to firing the
detonator, the blaster also gives a timebreak outpiit to the seismic unit.
When the current from the blaster ruptures thu bridgewire of the detona-
tor and thus produces an open circuit, an electrical pulse is trans-
mitted to the seismic unit and recorded as the zero time or time of
detonation.
B-3. Field Procedure.
a. Layout of Seismic Surveys.
(1) Planning. Preliminary planning and design of a seismic survey

should be done with due consideration of the objectives of the survey
and the characteristics of the site. Important factors include topog-
raphy, geology, the presence of noise generating activity, on-site util-
ities, hazards (such as power lines), structures, the depth of seismic
investigation needed, and operational constraints (such as restrictions
on the use of explosives). It is good practice to obtain as much of the
relevant information as possible prior to mobilization of the field
party and to make a systematic visual inspection of the site upon ar-
rival. The best orientation of seismic lines is determined by the known
geology of the site, topography (the ground surface should be as nearly
level as possible), utilities (power lines, pipes, etc.), and the
planned locations of the structures. The length of the seismic lines
should be initially about four times the required depth of investigation
but may be adjusted on the basis of early results. It is normal prac-
tice to run all lines in both a forward and a reverse direction to per-
mit the detection of dip in the subsurface strata and to minimize the
number of assumptions required in the interpretation of the data.
(2) Field layout. Prior to shooting, seismic lines should be

marked with stakes at the shot points and geophone locations. Surveys
B-13
EM 1110-1-1802
31 May 79
should be made to obtain the elevation of each shot point and geophone
location and the azimuth of the line, and the location of the line
should be tied to some well-defined feature on a map of the site. Ac-
curacy of elevations and horizontal locations should be to within 0.5 ft
or less. Elevations should be referenced to the project datum in every
case except preliminary reconnaissance surveys, since correlation of
seismically determined layer depths with boring information requires a
common datum. In the case of reconnaissance surveys, it may be suffi-
cient to refer elevations to an arbitrary datum (or a temporary bench
mark) since only elevation differences, rather than absolute elevations,
are required for interpretation. The seismic instrumentation, which is
usually in a truck or van, should be at least 200 ft from each shot
location, if explosives are used. It is also advisable for the equip-
ment to be about 100 ft from the nearest geophone to minimize noise from
movements of the operator or his assistants.
(3) Geophone spacing. Geophone spacing is determined primarily

by the desired degree of definition of the subsurface layers and by the
length of the seismic line. If the layers are relatively thin and the
seismic lines are short, small geophone intervals are necessary; but if
the layers are thick and seismic lines are long, greater distances be-
tween geophones are practical. On the basis of a 12-channel seismic
system and seismic lines less than 200 ft long, geophone spacings from
10 to 15 ft would be sufficient. Seismic lines greater than 200 ft can
be run with 25-ft geophone spacings. Only for seismic lines over about
1000 ft should greater geophone spacings be used, and then segments of
the line should have 25-ft spacings. Very long seismic lines can be
run in segments to avoid excessive distances between geophones. For
example, if a 600-ft line is to be run with a 12-channel seismic system
and 25-ft geophone spacings, the entire 600-ft line would first be
staked with two shot points at each end. This spacing of the geophones,
at 25-ft intervals from one end, would put the last geophone at 300 ft.
A shot would than be fired at each end of the 600-ft line to obtain one
segment in both forward and reverse directions. The geophones would
then be moved to cover the remaining 300 ft, and the other two shots
would be fired to obtain the second segment of the total line. Where
the seismic line is long, particularly if geophone spacings are 25 ft
or more, auxiliary short lines with small geophone spacings should be
run along the primary line, at each shot point location and in the mid-
dle of t), line, to provide more information on the near-surface mate-
rials. 1' simplicity, the primary seismic lines are generally run with
equal geophone intervals.
b. Conduct of Tests.
(1) Mechanical energy sources. Because of their relatively
B-l4
EM 1110-1-1802
31 May 79
low energy output, nonexplosive sources are used primarily in shallow

exploration and for S-wave velocity srveys. Usually a sledgehammer is
used with a wood or metal striking plate. If a zero-time device is not
attached to the seismic source, the striking plate should be as close
as possible to the geophone on the end of the line. After a record is
obtained, the striking plate is moved to the opposite end of the line,
and the procedure is repeated to obtain the reverse record.
(2) Explosive energy sources.
(a) Shot holes. If explosives are used, shot holes are required
to contain the energy of the explosive in the ground and for the safety
of personnel. The holes are usually drilled by hand augers. They
should be from 2.5 to 6 in. in diameter and usually from 3 to 10 ft
deep, depending on the size of the explosive charge. A 3-ft-deep hole
will usually contain the energy (prevent blowout of material) from a
1-lb charge, and a 10-ft-deep hole can contain the energy of a 10-lb
charge, depending on geological conditions. After the armed charge is
placed in the shot hole, soil material, preferably sand, is used to
backfill the hole. Some materials may require tamping to ensure that
as much material as possible covers the charge. Backfilling helps to
contain the energy in the soil and is a safety factor in that it aids
in preventing ejection of debris from the shot hole.
(b) Test procedure. Any person using explosive materials should

be cognizant of the hazards that may be caused by faulty or careless
handling of these materials. Regulations of Federal, state, and local
governments govern the storage, transportation, and use of explosive
substances, and the qualifications of personnel who use them. The
operator should follow all applicable regulations and should be fully
qualified. Following regular, systematic procedures during seismic
surveys helps to assure the safety of persons and property and the ac-
quisition of good data. The normal procedure for a seismic test is as
follows:
Step 1. The shot holes are prepared.
Step 2. Firing lines are run from the blaster to each shot point,
and each firing line is shorted at the blaster end.
Step 3. The external shorting plug (a plug that is inserted in the

blaster to complete the firing circuit) is removed from the blaster and
is held by the person who is to arm the charge.
Step 4. One of the firing lines is connected to the blaster.
B-15
EM i10-1-1802
31 May 79
Step 5. The weather conditions are evaluated for the possibility

of thunderstorms in the next hour; if no danger exists, the area is
cleared of personnel and vehicles and radio transmission is stopped. If
the weather is adverse, the sequence is delayed until conditions improve.
Step 6. The unarmed explosive charge and detonator are carried

separately from the approved storage location to the shot point by the
person who is to arm the charge and an explosive handler.
Step 7. A blast.r- -1'v9nometer is used to check the resistance

of the detona+ the firing line is shorted.
aced in the shot hole, and the

ho3
Step 9. The ".,d firing line are again checked with the
blasting galvanometer, and the detonator lead wires are tied to the
firing line.
Step 10. The firing line should be secured near the mouth of the
borehole to eliminate the possibility of its being carried by the ex-
plosion into an overhead hazard, such as a high-voltage power line.
Step 11. Amplifier gains are adjusted to minimize background noise.

Gains are usually adjusted so that those for the geophones nearest the
shot are relatively low. High gains on these amplifiers may result in
saturation of the system and cross-feed between channels.
Step 12. The area of the shot is cleared, and he person who has
armed the charge then gives the external shorting plug to the operator,
who places it in the blaster.
Step 13. The operator checks to see that the recorder is on, fires
the charge, and secures the record.
Step 14. The blaster verifies that detonation has occurred; when
it is judged safe, he enters the area alone while being watched by an-
other crew member, checks for any explosive that might not have been
detonated, and if none is present, gives an "all-clear" signal.
c. Sources of Problems. In some instances, problems are encoun-

tered in refraction work that lead to degradation of data quality.
These may include:
(1) Background noise. If wind noise is troublesome, it can be

minimized by burying the geophones. Machinery noise can sometimes be
B-16
EM 1110-1-1802
31 May 79
eliminated by electrically filtering the output of the geophones.
(2) Lack of seismic energy coupling. Deeper shot holes and better
backfilling procedure may help to minimize this problem. The shot hole
could be too large in proportion to the diameter of the explosive and
thus produce voids at the point of the shot.
(3) Wet weather. Geophones and cable should be waterproof. Be-

cause connectors on either end of the seismic cable are not usually
waterproof, they must be protected by water-resistant c,)vers, such as
plastic bags or polyethelene film. Water in one of these connectors
may cause cross-feed and 60-Hz interference. Impact of rain on the
geophcnes may cause high noise levels on the seismic records and can
mask tiie true signals. Field operations should cease when any lightning
is present in the area.
(4) Imr.oper placement of geophones. This i! a common problem and

is usually cat3ed by carelessness. The result is excessive background
noise on the record and poor detection of the seismic signal. The
location of thc geophone should be cleaned of any vegetation, and the
-pike on the geophone should be pushed firmly into the ground to make
the contact as tighu as possible. Many times, the top 2 or 3 in. of soil
is very loose and should be scraped off so that the geophone can be im-
planted into firm soil.
(5) Frozen ground. A frozen surface layer will give erroneously

high v. ues of velocity (approximately 6000 ft/ser) for the near-surface
material.64 The problen can sometimes be overcome by breaking up the
froze, ground with a pick, provided it is not excessively thick.
(6) Metallic tapes. ± metallic measuring tapes are used for the
layout of seismic lines, they should be taken up before shooting. If
left in place, they may provide a high-velocity short-circuit path for
the seismic signal, giving spurious high-velocity values for the soil.
B-4. Interpretation.
a. General. If a particular set of subsurface conditions is pos-

tulated, including thicknesses, geometry, and velocities, a theoretical
time.-versus-distance plot can always be constructed. However, the
inverse of this process is not generally possible. For a given time-
distance plot, the interpreter cannot infer a uniquely possible set
of subsurface conditions. Because of this difficulty, conventional
interpretation methods rely on the use of hypothetical, idealized
models, such as the simple two-layer case with a horizontal interface
and vI < v2 , the multiple-layer case with horizontal boundaries and
B-17
EM 1110-1-1802
31 May 79
velocities increasing with depth, or more 2omplicated alternative models

that may include fault offsets, dipping strata, and the like. The model
chosen is the one whose hypothetical time-distance plot most nearly
matches the field data.
b. Two-Layer Case. Figure B-5 illustrates the simple two-layer

case, with plane, parallel boundaries, with velocities uniform within
each layer, and with the deeper layer having a greater velocity than the
surface layer. The upper boundary represents the ground surface. As
shown, the boundaries are horizontal, but the seismic response is
exactly the same for inclined, parallel boundaries. For such a case,
it is necessary only to observe that depths or thicknesses determined in
refraction surveys are always measured normal to the layer boundary and
are not necessarily vertical. The velocities vI and v 2 of the two
layers are determined from the slopes of the two straight-line segments
on the uime-distance plot, the velocity being equal to the reciprocal of
the slope. The layer velocity may be computed from the time and dis-
tance intervals, At and Ax , between any two points on the line seg-
ment corresponding to the layer. Thus,
v = AX
At
(B-h)
c. Critical Distance. The critical distance xc can be picked

from the intersection of the two straight-line segments. The depth D
to the second layer is given by equation B-3, which is repeated here
for convenience:
x c Vv 2 - v I
D1 2= v + v1
d. Intercept Time. An alternative method of interpretation is

provided by the intercept time Ti., which is the intercept of the
second line segment on the time axis, as shown in figure B-5. The
depth D1 can be expressed in terms of the intercept time as
D i 2 (B-5a)
1 2 2
2 2 v1
An equivalent form of this equation is
Tiv ]
1 2 cos (B-b
B-18
EM 1110-1-1802
31 May 79
where a is the critical angle of incidence, defined by Snell's law,

as in equation B-2. To obtain the result in this form it is helpful to
note that since the critical angle of incidence a is given by
sin a = V1 /V2 , one can write
2
cos a= - (B-6)
and v1
v2
tan a = (B-7)
In practice, the intercept time is usually more convenient to use than

the critical distance, although the two methods are completely equiva-
lent. The critical distance method offers two advantages: (1) travel-
time errors that affect all geophone records equally (such as a blasting
cap delay) dc not affect che critical distance or the apparent veloc-
ities, and (2) the critical distance is useful in estimating the length
of the seismic line require. to detect a subsurface layer at a partic-
ular depth.
e. Critical Distance Versus Depth. Equation B-3 can be used to

construct a graph showing the length of a seismic line (relative to the
depth of the first layer) required to detect refractions from the under-
lying layer, as a function of velocity ratios. As shown in figure B-6,
this can be useful in planning a seismic survey if velocity ratios can
be estimated. The graph is also helpful in illustrating the effect of
velocity contrasts. For example, assume that there is about 15 ft of
overburden with a velocity of 2500 ft/sec, and that it is underlain by
a shale with a velocity of about 5500 ft/sec. How long should the
seismic line be to ensure adequate coverage for mapping the overburden
thickness? If the velocity ratio v2 /vl is 2.2, then xc/Dl is ap-
proximately 3.25 and xc is about 50 ft. The line should be at least
60 ft long to assure that xc is well defined by the time-distance
plot.
f. Multiple-Layer Case.
(1) Figure B-7 illustrates a simple multiple-layer case. It is

assumed that the layers have plane, parallel boundaries (including the
ground surface), velocities are uniform within each layer, and layer
velocities increase with depth. The time-distance plot has a line
I
B-19
EM 1110-1-1802
31 May 79
10
9
7--
0 6
50 Figure B-6. Plot of ratio of
0~
t-0 critical distance to depth of
> 4 -first layer as a function of
0 0 velocity contrast (after
2 0)
-)3 -Redpath
2
Critical distance xc
Depth of first layer' D
*2)*
Ti }4nec 1v
e
t nceti
Dis3
(V2 ) 2
3 (V4V3 v
B-V2
EM 1110-1-1802
31 May 79
segment corresponding to each layer and having a slope equal to the re-
ciprocal of the layer velocity. Each interface has a critical distance
and a corresponding intercept time. In figure B-7 and in the equations
below, the subscript identifying the critical distance and intercept
time is that of the layer lying immediately below the interface.
(2) The thickness D1 of the first layer is found by using the

two-layer case and either the intercept time Ti2 of the second line
segment or the critical distance Xc2 determined by the first two line
segments. This thickness is used in computing that of the next lower
layer, D2 , as follows:
2
2v 2
The equivalent of equation B-8, in terms of critical distance, is
= v2 2- j /# v
The computation can be extended to deeper layers by use of either of

the general equations n~l
nvn
n- 2 v v
T.D vv2 - D vn v (B-8)
Dn v2 2 \ v 2 2
2 v v j n+l -v
and
Xcn+l V+ 1 v v
vn Kn n v - vv
2 2 l (B-II)
incmutt
and itcn be t ene o tee nt l -
inthi c tehc s of layr. Since tee eauattoon
hisfor th thcknsse ofshallower layers, the computa-

in cotai
tion begins with the first layer and progresses downward. Equations
B-21
EM 1110-1-1802
31 May 79
B-10 and B-11 can be used to obtain equations for cases with particular
values of n , or they may be used in the general form in computer codes
for processing seismic data.
g. Two-Layer Case with Dipping Interface.
(1) The geometry of this case and the corresponding time-distance

plot are presented in figure B-8. In developing the model for this
Total time, Tt
or
reciprocal time
--------------------- --
Slopee=
Down-dip I V2d
T1 d
____
'N - Sl pe
"i____ = l/
__Up. v __d__p
Distance
Dd......
Y= dip angle"
Figure B-8. Example of dipping in-
terface and concepts of "reverse
shooting" and "apparent velocity"
20
(after Redpath )
case, assume that the interface between layers 1 and 2 is a plane dip-
ping at an angle y with the ground surface, that the velocities are
uniform within layers, and that v2 is greater than v1
(2) It is not possible with a shot in only one direction along a

seismic line to determine whether the interface is parallel Lo the
ground surface or not; moreover, if the interface does dip, the slope
of the time-distance curve does not give the true velocity for layer 2.
B-22
EM 1110-1-1802
31 May 79
For these reasons, refraction seismic lines should always be run in
both forward and reverse directions, i.e. with a shot point at each end
of the line.
(3) The diagram in figure B-8 shows a shot point at each end of
the seismic line, as used in reverse shooting, and also the ray path of
the shortest travel time between the ends of the line. This path lies
in a plane normal to the plane of the interface, and the cross section
of figure B-8 is in the same plane. The distances D u and Dd are
the perpendicular distances from the two shot points to the plane of
the interface. To avoid confusion, and to maintain consistency, the
subscript u is used where the seismic line extends up-dip from the
shot point, and d when the seismic line extends down-dip from the
shot point. Thus, the distance Dd refers to the shot point located
at the up-dip end of the line, and Du to the shot point located at
the down-dip end of the line.
(4) The angle y in figure B-8 is the apparent dip, or the

component of dip in the plane of the figure. It is very close to the
true dip angle 6 of the interface if the seismic line is laid out
within about 10 deg of the direction of the true dip.
(5) The time-distance plot is made by plotting in both directions

with each shot point in turn as the origin. As noted in the plot (fig.
B-8), the velocity of the surface layer is correctly represented by the
reciprocal slope of the first line segment, but the apparent velocity
of the second layer is different for the two directions. The critical
distance also depends on whether the shot point is at the up-dip or the
down-dip end of the line. The total travel time Tt from one end of
the line to the other, however, must be the same for both directions,
since this time merely represents travel in opposite directions along
the same path. This time is sometimes called the reciprocal time. Any
disparity in the reciprocal time is an indication of a mistake in data,
plotting, computation, or layout of the line.
(6) The effect of the dip on the apparent velocity can be seen
by observing that the path of the first arrival is determined by the
critical angle of incidence, as given by Snell's law, and for a given
shot point, the length of the travel path in the upper (low-velocity)
layer is different for each geophone location. With the shot point at
the up-dip end of the line (shooting down-dip), the travel path in the
upper layer becomes increasingly longer for more distant geophones, and
the resulting apparent velocity of layer 2 is lower than the true veloc-
ity. With the shot point at the down-dip end of the line (shooting
up-dip), the reverse is true and the apparent velocity is greater than
the true velocity.
B-23
EM 1110-1-1802
31 May 79
(7) It is relatively simple to write an expression for the travel

time Tu or Td to any geophone from either shot point and to find
the derivative dTu/dx or dTd/dX , which is equal to the slope of the
time-distance plot. If the respective slopes are denoted m u and md ,
the following relE.sions apply:
m 1 1 sin (a - y) (B-12a)
V2u vI
=
a __L=
V2d vI1 sin (a + y) (B-12b)
where a is the critical angle of incidence according to Snell's law,

and the other quantities are as shown in figure B-8. Equations B-12a
and B-12b can be rewritten to solve for the angles a and y , as
follows:
d + sin
2= 1 vlmu (B-13a)
2a =sin - vmi v
2ysin 1 vlmd - sin-1 vlm u (B-13b)
The perpendicular distances from the shot points to the interface are
given in terms of intercept times by
VlTu
iu(B-14a)
Du = 2 cos a
D=VlTid (B-14b)
d 2 cos a
(8) The velocity v2 can be computed from v and the critical

angle a , according to Snell's law (equation B-2):
vI1
sin a =
Alternatively, it can be computed from the relation
v2 = (v2v~ d )cos y (B-15)
B-2h
EM 1110-1-1802
31 May 79
Where the dip is small, cos y is nearly equal to unity, and an approx-
imation for v2 that is good enough for most cases can be computed by
taking cos y equal to 1.
(9) Since the angle y is an apparent dip angle, which depends

on the direction of the seismic line relative to the direction of dip,
two seismic lines in different directions are required if the true dip
is needed. Each line must be shot in both forward and reverse
directions.
(10) In figure B-9, the lines 0-1 and 0-2 are two seismic lines,
and the angle between them is denoted by 0 . Lines O-d and O-s are
d
I
! !2
I
!
I
!
I
/ /
Figure B-9. Determination of dip and

strike by two seismic refraction lines
(prepared by WES)
in the dip and strike directions, respectively. The angle B is the

unknown angle between the dip direction and line 0-1. The angle of dip
is denoted 6 . The direction of dip relative to 0-1 and the dip angle
are given by
B-25
EM 1110-1-1802
31 May 79
tan- si
ny (B-16a)
(sin y (B-16b)
where y and y are the apparent dip angles along lines 0-1 and
0-2, respectively. If 0 is close to 90 deg (i.e., line 0-1 is close
to the strike), equation B-16b may not, as it stands, give an accurate
value of 6 . In that case, substitution of y for y and - 0
for 0 in equation B-16b will allow the dip angle to be computed by
reference to line 0-2.
(11) In practice, this method of determining dip is infrequently

used. If the dip angle is small (not exceeding 10 deg), it is suffi-
ciently accurate for practical purposes to use the depths D and Dd
as the vertical depths from the shot points to the interface. Also,
dipping interfaces can conveniently be dealt with by the delay-time
method, which may also be used for the interpretation of irregular or
curved surfaces.
h. Delay Times. It has been pointed out that true refractor

velocities cannot be determined by firing a shot at only one end of a
seismic line, but that they can if arrival times are recorded from both
ends. Using both forward and reversed lines, however, offers another
significant advantage in that the true velocities and the thicknesses of
layers can be computed on a much more detailed basis by means of delay
times. Under favorable circumstances, depths can be determined beneath
each geophone to allow mapping of irregular and dipping boundaries.
(1) Velocity determination. The velocity v 2 of an irregular or

dipping refractor can be obtained, as shown in figure B-10, by plotting
the difference between the arrival times tDl and tD2 , for the for-
ward and reversed shots, for each geophone against distance. It can be
easily shown that the arrival time difference is given by
S- x+ constant (B-17)
Dl D2 v2
It follows that differences in arrival times plotted against distance,

x , will form a line, the reciprocal slope of which is half the velocity
of the refractor. In figure B-10, arrival-time differences for a two-
layer case have been plotted above and below an arbitrary reference
line. The arrival times being differenced must represent arrivals from
the same refractor. The method is frequentyluly useful forUeer -
L
whether the arrivals have been refracted from the same layer. A devia-
tion of the points from a straight line may indicate either that there
B-26
EM 1110-1-1802
31 May 79
300
250
200
ci DIFFERENCE INY
wi ARRIVAL TIME ARIAR
7- ISO * REFERENCE
DIFFERENCE IN
A R R IV A L TIM Ev2
( / ,
1002 (2 2 1/003 0 1
50
I I
0200 400 600 800
DISTANCE, FT
6400 FT/SEC
Figure B-10. True velocity of refractor from differences of

arrival times (prepared by WIES)
B-27
EM ili0-i-1802
31 May 79
exists a lateral variation of the refractor velocity, that the times

being differenced represent arrivals refracted from different layers,
or that the times represent both refracted and direct arrivals (note
in figure B-10 the deviation of the points at 50 and 100 ft from the
dashed line).
(2) Two-layer case.
(a) The meaning of the term "delay time" is illustrated by fig-

ure B-11, in which the delay time is defined at the shot point and at
the detector. The delay time is the difference between the actual
travel time, over the refracted path through both layers, and the hypo-
thetical time that would have been required, at the refractor velocity
v , along the normal projection of this path on the interface. It is
assumed that the interface is not highly irregular or steeply dipping,
so that the distance along the interface between shot point and detector
is approximately equal to the straight-line distance from shot to detec-
tor. The total delay time At is then given by
At Att
= tD - -(-8 (B-18)
D v2
where t is the arrival time at the detector. The total delay time
D a delay time AtD
consists of a delay
at the detector, that AtS at the shot point and
so time
At = At + At (B-19)
S D (-9
For the pulse traveling up to the detector in figure B-11, the delay
time has been defined as
Shot Detector
EF E'F
v
Delay time at shot point =A ts-
=A GH GH'
Delay time at detector D Vl v2
Figure B-l. Definition of delay time (after Redpath 2 0 )
B-28
EM 1110-1-1802
31 May 79
GH GH ' B-0
AtD -
viI vi'(B-20)
2
This definition leads to the following relations between delay time

and depth:
At v
DD= D 1 (B-21a)
cos in - v )
or
AtDv v2
D D (B-21b)
Both forms are equivalent and may be used interchangdably. The same
relations apply to the depth at the shot point, with the substitution of
At and D . The depth DD actually represents the depth of a point
S is closer
that s
to the shot point (point G in fig. B-11), where the ray
path is refracted at the interface between layers. The horizontal dis-
tance from the detector to this point, if the dip of the interface is
small, is the distance GH , expressed as
GH' = DD tan a (B-22)
where a is the critical refraction angle (i.e., a = arc sin v 1 /V2).
(b) In order to use equations B-21a and B-21b to find depths of a

refractor, it is necessary to be able to separate the shot point and
detector delay times from At . If one of the depths is known from
boring data, the corresponding delay time can be computed from equa-
tions B-21a and B-21b and the rest of the delay times for the line can
be obtained by applying equations B-18 and B-19. If no depths are
known, the determination of delay times will require the use of a re-
versed seismic line, as shown in figure B-12, which shows the arrival
times at the geophones from shots at both ends of a s(ismic line. The
total travel time from one end of the line to the other (the "reciprocal"
time) has been designated Tt and should be the same for both shots.
The arrival times at one of the (arbitrarily selected) geophones from
the two shots SP, and SP2 have been designated tDl and tD2 , re-
spectively. If it is assumed that the distance along the refractor
surface is nearly equal to the straight-line distance at the ground
surface. and that the distances CD and ED (fig. B-12) are nearly
equal, it can be shown that:
B-29
EM 1110-1-1802
31 May 79
100
s0 -.
'N I
20 -
Li 0 to 0-0 0 o 60 70 80 90 10
40
ILI
DITACE FT,
CI E
303
EM 1110-1-1802
31 May 79
D 2 (Dl + D2 t )
where AtD is the delay time for the chosen geophone. This delay time
is, in effect, an average for the forward and reversed lines. There-
fore, AtD and the depth computed from it refer to the actual geophone
location. It is important to note that Tt must represent the total
time, or reciprocal time, for signals refracted at the same layer that
is represented by the arrival times tl and t . If the reciprocal
D .D2
time obtained represents some deeper layer, it is necessary to use one
or more additional shot points on the line.
(c) As shown in figure B-12, a plot of delay time versus distance

is similar to a profile of the refractor surface. The delay times shown
with an open circle (fig. B-12) were obtained by using equation B-23.
Those shown with a closed circle represent geophone locations at which
a refracted first arrival was obtained in only one direction. Delay
times were computed, therefore, from equations B-18 and B-19, after
equation B-19 and a delay time for a geophone in the middle of the line
were used to find the shot-point delay times. The delay times marked by
closed circles are "migrated" toward their respective shot points, ac-
cording to equation B-22.
(3) Multiple-layer case.
(a) The delay-time method can be extended to cases in which there

are three or more layers, although three layers are usually the prac-
tical limit because of loss of accuracy and resolution as the number of
layers increases. An important limitation is that the method can be
applied only where there is "overlap" between arrivals refracted from
the same layer. In other words, the actual length of the seismic line
should be such that the forward and reverse arrival times at a geophone
where delay time is to be computed represent signals refracted through
the same layer. If a lack of overlap for deep refractors is noted in
the field by reducing and plotting the data immediately after the shots,
it is often possible to remedy the situation by firing additional shots
off one or both ends of the line (i.e., beyond the ends of the geophone
array) and in line with the array. It is also common in refraction in-
vestigations with long lines and multiple layers to have no overlap of
refractions from the second layer. Resolution of this situation re-
quires that some intermediate shots be fired along the length of the
seismic line. The use of intermediate shot points is a good general
practice. It will result in better control of first-layer depths and
velocities and fewer assumptions are required to interpret the raw data.
(b) If the delay times in the first layer are subtracted from the
B-31
EM 1110-1-1802
31 May 79
corresponding arrival times and the differences plotted, the result is,
in effect, a new time-distance curve that is equivalent to placing the
shot and detectors at the top of the second layer. This plot of re-
duced arrival times can be interpreted in terms of delay times in the
same manner as previously described. Some useful examples of the use
of delay times are given by Redpath 20 and Black et al.ll
i. Continuous Increase of Velocity with Depth. The case of veloc-

ity continuously increasing with depth is rare but may occur in such
materials as hydraulic fill or dredge spoil. This type of situation
can exist because of a finely stratified geology, a progressive decrease
of weathering with depth, or a gradually increasing density with depth
in soft soil resulting from consolidation under its own weight. A con-
tinuous increase of velocity with depth will produce a time-distance
curve similar to the one shown in figure B-13. The ray paths are curved,
and in the special case of a linear increase of velocity with depth,
they are circular arcs.
Slope = 1/'v,
Slope- I/ v
Distance
=
Vx velocity (from time-distance curve)
corresponding to any distance x < s
from shot point.
x =distance from shot point.

Figure B-13. Schematic of ray paths and time-distance curve for
20 )
continuous increase of velocity with depth (after Redpath
- B-32
EM 1110-1-1802
31 May 79
(1) This type of time-distance curve can be transformed into a

curve of velocity versus depth by means of the "Herglotz-Weichert-
Bateman" equation:20,36
~0
Ds = f cosh- -i dx (B-24)
where
D = depth at which velocity is v

s velocity (from time-distance curve) corresponding to
distance s from shot point
v X = velocity x < stime-distance
(from curve) corresponding to any
from shot point
distance
x = distance from shot point
The integral is evaluated graphically by choosing a particular distance

s and the corresponding velocity vs , obtained from the tangent to the
time-distance curve at s , and plotting values of cosh-l(vs/vx )
against x , where x is any distance less than s . The area under
the resulting curve (obtained numerically or with a planimeter) is
multiplied by 1/n to determine the depth zs at which the velocity
is vs . The integration is repeated for decreasing values of s , and
the graph can then be constructed of the variation of velocity with
depth. The method is tedious, although the integration can also be done
numerically, with the aid of a digital computer, and it is justified
only when there is a smooth, continuous increase of velocity with depth.
Often, gradual increases of velocity can be interpreted by assuming that
the curved time-distance plot is made up of a few straight-line segments.
Interpretation can be aided in this case, also, by a digital computer,
number of straight-line seg-
using equations B-10 and B-11. Any desired number
ments could be used. However, the largestthat could be meaning-
ful would be one for each geophone, or each recorded arrival time.
(2) A special case of occasional interest is that of velocity

increasing linearly with depth. The only situation that will be con-
sidered here is the one in which such a layer is at the ground surface.
The velocity for this case is given by
v v + kz (B-25)
where
v = velocity at dc't z
vo = velocity at ground surface
k = rate of velocity increase
B-33
EM 1110-1-1802
31 May 79
It has been determined that the ray path in such a medium is a circular
arc with a center at a distance v /k above the surface. According to
Dobrin, 3" the lowest point on the path is at a depth
V_ okx\
Dmax k + - (B-26)
and the travel time t for this path is given by
-1 k
sinh (B-27)
0
Moreover, the velocity at depth D is equal to the reciprocal of

the slope of the time-distance curve at distance x
(3) The parameters vo and k can be obtained from the time-

distance plot and equation B-27. Any two distances, x and x , and
the corresponding times, t andt , can be chosen from the part of
the time-distance plot corresponding to the surface layer. Two simul-
taneous equations can then be written
2 -1 kx1
tl k sinh 2- (B-28a)
kx
2
t2 =sinh-v -1 2 (B-28b)
0
and solved by trial to obtain vo and k . A convenient way to do this

is to write equation B-27 in the form
kx
v =
0 2 sinh kt
2
and to plot a curve of v o versus k for one set of time and distance
values. This is repeated for a second time and distance, and the in-
tersection of the two curves will give the required values of vo and
k . While v. and k can be determined from only two pairs of time
and distance values, the exercise shouild be repeated with several sets
of values in order to obtin te best average and to determine how wel
the data fit the assumed model.
B-34
EM 1110-1-1802
31 May 79
(4) If a surface layer with a linear increase of velocity is

underlain by a layer with a velocity v2 , higher than any in the sur-
face layer, the time-distance plot will have a straight-line segment
with a slope equal to 1/v 2 The time intercept T. for this line
segment can be obtained by projecting it back to the time axis in the
same manner as previously described. The relation between the thickness
D1 of the surface layer and T. is given by:
2 1v2 v20 2
T. = coshvv c +ckDh -
V + /v
kD 2
+ j0+o (B-30)
The velocity v2 and the parameters v o and k are obtained as

described above. The thickness D1 can only be determined by trial,
since equation B-30 cannot be solved explicitly for D
j. Correction for Shot Depth and Elevation Differences.
(1) In all of the foregoing discussions, it was assumed that the

shot was placed at the ground surface and that the ground surface, or
at least the positions of the geophones, could reasonably be represented
as a plane. The ground surface was not necessarily required to be hori-
zontal but was taken as a reference plane; all positions and dips of
subsurface boundaries were thus determined with respect to that refer-
ence plane. In practice, the shot is frequently placed at a depth of
several feet in a borehole, and the ground surface may have enough
irregularity to produce significant errors if calculations are made as
though it were a plane. Also, it may be desirable to refer the data to
a horizontal reference.
(2) The elevation and shot depth corrections may be treated

together, as shown in figure B-14. The shot and detector are reduced
to a common datum by adding or subtracting a delay time corresponding
to the elevation difference. The elevation correction, which is to be
subtracted from the arrival time value, is
(e- h + E-2d)v 2 - v2
Elevation correction VlV2
(B-31)
12
B- 35
EM 1110-1-1802
31 May 79
where
e = elevation of ground surface at shot point above sea level

h = depth of shot below ground surface
E = elevation of geophone above sea level
d = elevation of datum plane above sea level
All depths and thicknesses are then obtained with respect to the datum
plane. Frequently it is sufficient to correct only for the depth of
the shot. In this case, a very good approximation is obtained by the
following rule: Add one half of the shot depth to the apparent thick-
ness of the surface layer as computed from formulas based on the surface
shots.
DETECTOR
utACE
UN
AFTER~c TIMECTIO
SBADD TIME C ADOTIME O
V2
• SEA LEVEL
Figure B-lb. Elevation correction for the two-layer case; d is

the elevation of the datum plane, e the elevation of the ground
at the shot point, E the elevation of the detector, and h the
depth of the shot below the ground surface (after Dobrin3 6 )
k. Examples of Time-Distance Plots. In the following paragraphs,

examples are given of some normal and problem conditions and the time-
distance plots that would be encountered.
(1) Buried shot. If the shot is buried, the arrival time to at

the ground surface over the shot point is equal to h/v1 , where h is
B-36
EM 1110-1-1802
31 May 79
the shot depth. The theoretical time-distance plot for the surface
layer would be a hyperbola, with a time-intercept t , asymptotic to
the line with slope 1/v1 , as shown in figure B-15. ° In most actual
field cases, the geophone spacing would not be close enough to define
the hyperbola well, but the time intercept may be apparent.
0/
: /
to, //
//
O-x
DISTANCE
Figure B-15. Effect of burial of shot

on initial portion of time-distance
plot (prepared by WES)
(2) Soil of low velocity. Occasionally soils are encountered with

P-wave velocities of a few hundred feet per second. Such a velocity is
less than the velocity in air, which is 1127 ft/sec under standard
conditions. If the shot is at the surface, the first segment of the
time-distance plo-. will indicate a velocity of about 1100 ft/sec, be-
cause the first-arriving signal will be one that is propagated through
the air (fig. B-16). While apparent velocities of deeper refractors
will not be affected, the depths to these deeper layers will be incor-
rectly computed because the surface layer velocity is erroneously taken
to be 1100 ft/sec. The true velocity of the surface layer can be
obtained by using a buried shot and computing the velocity from the
arrival times and slope distances from the shot to nearby geophones.
The depth of the surface layer can be computed from the time intercept
and the true velocities. Note that the use of the critical distance
B-37
EM 1110-1-1802
31 May 79
T
VI,.
DISTANCE
SP vair = 1100 FT/SEC
• ::. ,,, *::. ,', = 600 FT/SEC•
. .. . .. . V2 = 5000 FT-SEC'?:.":';
Figure B-16. Time versus distance plot for sur-

face layer with velocity less than air (prepared
by WES)
method would give an erroneous depth in this case. Wherever an apparent

surface layer velocity of about 1100 ft/sec is encountered, the pos-
sibility that air waves are being observed should be considered. This
condition can sometimes be detected by the shape of the recorded wave
forms. If the geophones are vertically oriented, an air-wave arrival
may produce a first break in the direction opposite to that given by
a signal transmitted through the soil.
(3) Groundwater table. The compression wave velocity of water

is about 4800 ft/sec, and the velocity of loose, saturated soil is
usually in the vicinity of 5000 ft/sec. Therefore, a horizontal
interface with a velocity of about 5000 ft/sec below it is frequently
indicative of a groundwater table. Thiz can sometimes be determined
by observation of the slower shear waves. Since the propagation vc-
locity of shear waves is nearly independent of water conten t, there is
no change in shear wave velocity at the water table, and no refraction
of the shear wave.
B-38
EM 1110-1-1802
31 May 79
( ) High or low points on rock surface. Abnormally high or low
points in the buried rock surface may plot below the average velocity
line, where there is a rise in the rock profile, or above the velocity
line where a large depression exists, as shown in figure B-17. Also,
points will plot above or below the line where an uneven ground surface
lies over a smooth rock surface, unless an elevation correction is
applied.
Tt 0Tt
GROUND SURFACE.
. . .. ... ...-.. ° .,-.°%°.. . o,.
i............
• I , 1 ROC L .- *. . , '. .. *
Figure B-17. High or low points on rock

surface (prepared by WES)
(5) Change in dip. A change in dip of the refractor surface along

the line of the seismic profile is illustrated in figure B-18. The
change in dip is indicoiLed by a change in slope of the time-distance
plot. Such a curve could also be produced by a change in slope of the
ground surface, if elevation corrections are not applied.
(6) Buried rock ledge. Beyond the end of a buried rock ledge, the
upward turn of the time-distance curve, as shown in figure B-19, indi-
cates that low-velocity materials exist beyond the end of the ledge.
(7) Subsurface stream channel. A subsurface deeply incised

stream channel or cliff will give unusual velocity curves, as illus-
trated in figure B-20. The sloping edge of the channel is indicated by
a reduced apparent velocity on the shot in the downslope direction and
an increased apparent velocity in the opposite direction. This kind of
B-39
EM 1110-1-1802
31 May 79
GROUND-SURFACE
Figure B-18. Change in dip of refractor

surface (prepared by WES)
GROUND SURFACE
............................
~~~.
xxx x x .x~x
..
. x xxx
Figre A"9 Buie rokle
--(prepared-by
- - -S)
EM 1110-1-1802
31 May 79
v2 (true) I (true
............
.. ... GROUND SURFACE
Figure B-20. Subsurface stream channel

(prepared by WES)
condition can usually be resolved by the use of delay times, except in
the vicinity of very steep or irregular surfaces.
(8) Subsurface cliff. For this case, the time-distance plot in

figure B-21 is somewhat similar to that in figure B-20. However, the
slope of the break of the rock surface is very steep, and there is a
definite up-or-down offset in the time-distance curve.
(9) Offset in time-distance curve. An offset in the time-distance

curve can be caused by any one of several conditions. Among the possi-
bilities are: (a) erroneously plotting arrivals due to vibration
cycles later than the first arrival, usually because of weak first
arrivals or excessive background noise; (b) a vertical offset in the
refractor surface, as shown in figure B-21; and (c) the presence of a
fractured fault zone, vertical mud seam, open cavern, sinkhole, or
grike across the seismic line. In case (b), the sense of the offset is
reversed when the direction of the shot is reversed (fig. B-21), while
in case (c) the offset is positive for both directions (fig. B-22).
(10) Subsurface fault. A forward and reverse spread across

steeply dipping or faulted massive bedrock overlain by alluvium would
B-41
EM 1110-1-1802
31 May 79d
2
- Y2
V2
Tt V2
GROUND SURFA CE Figure B2.Subsurface cliff

(prepared by WES)
.......
'. ...
ROCK
n ie
Figure~~~~ B-2v22e
B-_2
EM 1110-1-1802
31 May 79
yield time-distance curves similar to those shown in figure B-23. The

apparent velocities v I , v 2 , and v 3 are true velocities.
V3 > V2 > "I
,... "
zz Tt
-GROUND SURFACE
:: ....
:::::~~~...'....':'.................. '.'.....::.....
::~ : $::
<b.. !i:. ' . . .': :::.y: .':::
....... ++. ++
......
----------- ... . ± . . . .+ + +
+ + + + + + +
----- 2.----------4 + + V3 + +4.
+@ +@
Figure B-23. .ubsurface fault

(prepared by WES)
(11) Buried ridge. A vertical or steeply dipping buried ridge of

high-velocity rock aligned at an angle to the direction of the survey
line may give a time-distance plot similar to that shown in figuxe B-24.
Unless an outcrop is found that will provide an indication of thickness
and direction of strike, several survey lines may be required to define
the feature.
1. Typical Velocities of Geologic Materials. Table B-l, reproduced

from superseded EM 1110-2-1804, lists typical longitudinal, oi compression,
wave velocities of various soil and rock materials. They are presented
here to give some idea of what velocities mean in terms of geologic mrdia.
B-5. Limitations, Accuracy, and Resolution.
a. General. A fundamental limitation in the seismic refraction

method lies in the fact that a given set of arrival time and distance
data cannot be associated with a unique set of subsurface conditions.
In principle, for a given system of velocity distribution in the
B-43
EM 1110-1-1802
31 May 79
]i ~ <~,v
Tt T
GROUND SURFACE
L PROFILE
BURIED RIDGE OF STEEPLY DIPPING

ROCK ALIGNED AT AN ANGLE TO THE
DIRECTIONS OF SURVEY
-- SURVEY LINE
tios
f elciy it deth Hoevr th problem'..:met.with'..in.seismic'.
PLAN VIEW
Figure B-24. Buried ridge (prepared by WES)
subsurface, it is always possible to synthesize a one-and-only time

versus distance plot, no matter how much the subsurface conditions are
complicated by faults, peculiar boundary shapes, or unusual distribu-
tions of velocity with depth. However, the problem met with in seismic
exploration is the inverse--that of proceeding from time-distance plots
to evaluation of subsurface conditions--and uniqueness of the solution
.is not assure . J canAot c e reduced by additional shootig
with some part of the layout changed; a simple example of this is the
use of reversed lines to distinguish between horizontal and dipping
EM 1110-1-1802
31 May 79
Table B-1. Velocity of Propagation of Seismic

Waves in Subsurface Materialsa
FEET PER FEET PERJ

MATERIALS SECOND MATERIALS SECOND
Top SoiLs: GRR.NIT~I:

LIGHT AND DRY 600 TO 900 SIERAA NEVADA RANGE.
MOIST. LOAMY OR SILTY 1.000 TO 1.300 CALIFORNIA (IN ROAD CUTSI
CLAYEY 1.300 TO 2.000 DEOICOMPOSEDD 1.540
RED CLAY IN COLORADO (A) 1.630 BADLY FRACTURED AND
SEMI.CONSDL bATED PARTLY DECOMPOSED 2.200
SANDY CLAY C) 1.250 TO 2.150 SOFTENED AND PARTLY CE.
WET LOAM 2.500 COMPOSED OUT SLIGHTLY
SEAMED 10,500
CLAY. DENSE AND WET *SOLID0 AND MONOLITHIC
DEPENDING ON DEPtM 13.000 To 5.900 70 EET DEEP 1&500
RARILEr. oR GfRAREtL) 1.70TO2.0
1.9
TO2.60 IEW HAMPSHIRE (CCOu.
CMNCSAD(01 PANISON OF
'1VElOCITI1ES
CEMNTO
SND2.800 to 3.200 1IT DILNG LOGS)
SAND CLAY C)3.200 TO 3.800 BADLY RROGGN AND
CEMENTED SAND CLAY W 3.800 TO 4.200 WEATHtRCOT FREQUENTLY
ONLY CHIPS AND FRAGMENTS
WATEH SATURATED SAND R)4.600 RECVYCME. SE.AET OF
SAND 4.600 TO 8.400 CURE LONGER. RUT PEA.
TRERING HAS PEN"ETRTtD
CLAY. CLAYEY SANDSTONE CR) 590 AOUT I/4 INHCHON ,ACHw
GLACIAL TILL Cc1IDEOF OTHE J1OINT PLANES
UPPER SUSOEHANNA () 5.600 to 7,400 ON WCH A FLM Of HE.
GLAIA AIRANE A)SISUAL CLAY H4AS FORMED 3.000 TO 8.000
DEPOSIT. DHY.CALIFoRNIA 12.500 to 5.000 JOINT PLANES SHOW RUT
GLACIAL MORAINE DEPOSIT. LITTLE S!CN OF WEATH.
SAI.RATED.CALIFORNIA 5.000 10 7.000 INC. EVER THOUGH THEY
CEMIENTED LAVA ARC OPEN 10,000 TO13,000
ACGLOMERATE. CAIONA 5,00T6000 ENTIRELY UNWEATHERED
LOOSE RVCK.TALUS 1 .250 TO 2,100 AN. NSE1AMIO &6000T0,0,000
WEATHERED AND f*ACTURCD-ROGH I.bOO TO 10.000 GHANODIORIYE(8 5.0
S.ALET I1ASALT-CANAL ZONE.REATHERED
OLENIANGY RIVER. OH.IO ,9.000 TO 11.000 AND FRACTURED 9.000 TO 14 000
UPER SUSOUCT.ANNA(C 10.200 TO12.800 LIMESTONE. rOLOMITE. MCTA.
CANALIC C) 700 TUPHIC ROCRS. MASSIVE
PANAMA CNLZON 7 To 8.000 RCSR1 16.400 TO20200
LIANCOS. COLORADO () 2.600 TO 2.960 Di:AAG.OIN RED Of BROAD
POMVEVyS4ALESCN4NANDOAH IPIVFR. SOUTH CAROLINA 19.700
RIVER . WCATNEVEC 14.000 TO~6.500 GRIENStONE. TIGHT SEAMIED.
ROMNENSHALE5HNANoA
RIVERA. COOD
CALIFORNIACA) I 6.100
12'000 GREENSTONE.SLIGHtY SEAMED
JIOHNMARSHALL DAM SITE 2.900 TO 4.250 -_CALIFORNIA ______J 13.300
PHYLITC.YOPK. PA. (0 1000 To 11.000 tIiQC
SANOSTONET(e 17.200 to 7.900
OCAORIAN.UpFp(p I A)Peported by7C. A,. W.iI..Ax. V. S. Bureau
S')SOVEHANNA (C) 14.000 of Public Roads
CANAL ZONE. PACIFIC (ND 7.000 TO 9.000
1
COLORADO. DIENSE. HARD. FoRpot fIIprAGchyia
AND CONTINUOUS PITH orRorofep tlGohyia
FCE TEAMS (A) 7.250 Loperlnental Survey in Australia
C1LORAOO. CONTAINING
RCAMT.ENED SCAMS AND CC)7Pe ted by A. E. Wood, Corps of
0,
SOFT AREAS. (A) 4.725 rptinoee:
SMAOKY11ILL RIVER, KANSAS 6.000 TO 7.500
SANOOTONE CONGLOIICRATECR) 8.000 (0yL). b . C~Uo
CHALK' £njinfors
FORT RANDALL OAMITTE
ARovc RATER TABLE 6.300 TO 7.000
rORT RANDALE OAMSITE
a Prepared by WES from information given in superseded EM 1110-2-1802

dated September 1948.
B-4 5
EM 1110-1-1802
31 May 79
interfaces. In some other situations, surface shoc "g alone, no matter
how it is arranged, cannot resolve all ambiguitieE . some direct in-
formation, such as borin6 data, is required. Because of this inherent
limitation in the seismic method, a seismic survey alone can never be
considered definitive. Properly integrated with other complementary
explorations methods, however, seismic ':-Lction suLveying is a highly
effective, accurate, and cost-effective meti.od of ottaining subsurface
information. In the following paragraphs, some specific limitations and
sources of error are discussed.
b. Velocity Reversals.
(1) Throughout the preceding discussions, it has been explicitly

assumed that the velocities of the soil or rock layers increase with
depth. If in the field a layer is encountered with material of higher
velocity above it, the low-velocity layer cannot be detected ' surface
refraction surveys, and the computed depths of deeper layers will be too
great. Although the most common condition is that of velocity increas-
ing with depth, there are many kinds of situations in which velocity
reversals can occur. Examples can be found in the cases of sands over-
lying organic bog soils, glacial till overlying lacustrine sediments, or
basalt flows in alluvial sediments.
(2) Figure B-25 illustrates a hypothetical four-layer case in

which layer 2 has a lower velocity than the surface layer. The ray path
is shown for a first arrival that is refracted through layer 4. The
path is governed by Snell's law, as always, and the ray is refracted
toward the normal in layer 2. There is no possible ray path represent-
ing a first arrival with its deepest part in layer 2; so there is no
indication of layer 2 in the time-distance plot. The time-distance
plot, in fact, is indistinguishable from that of a three-layer system
with velocity increasing downward.
(3) The effect of a low-velocity layer is to make the computed

depths greater tnan actual depths. This is due to the fact that the
time required for the signal to traverse layers 1 and 2 is greater than
it would be if v2 were equal to v I , while the depths are computed
as though the whole distance from the surface to layer 3 were traversed
at vl , the higher velocity. If the presence and velocity of a low-
velocity layer are known, the layer can be compensated for in the com-
putations, but its existence must be determined by means of a more
direct method, such as an uphole velocity survey.
c. Blind Zones.
(1) A blind zone, or hidden layer, is a layer that cannot be
B-46
EM 1110-1-1802
31 May 79
E'3 4)4
Vl
Distance
(Vi)
(V2)
(V3)
(V4)
VI > V2 < v 3 < V 4
Figure B-25. Velocity reversal and cor-

responding time-distance curve (after
20 )
Redpath
detected by the refraction seismograph because of insufficient velocity

contrast or thickness. This inability cannot be remedied by any change
in -he layout of the geophones, but only by applying information from
borings. In most cases, the blind zone will lie between the surface
and a high-velocity layer.
(2) An example of this type of problem is given by Soske. 1 Fig-
ure B-26 depicts an example of a three-layer case with the layers having
good velocity contrasts. In the wave-front diagram, note that the head
wave from the high-velocity layer 3 overtakes the head wave from layer 2
at some distance from the shot. If layer 2 is thin, that distance may
be less than the critical distance of layer 2. The example in figure
B-26 shows the case where the overtaking distance is just equal to the
critical distance of the intermediate, 8500-ft/sec layer. Thus, it
shows the minimum thickness of the intermediate layer that would have
to exist before its presence could be detected by first arrivals, re-
gardless of the geophone layout. However, there is no indication of
the 8500-ft/sec layer on the time-distance curve; it would be necessary
that 4.te 'lay i 1-2 b at 'Cat 70 ft thick in order that
refractions from it be recorded as first arrivals.
• : B-117
EM 1110-1-1802
31 May 79
125 1/26, 500 ft/sec
100 1/,0 V/- 1/1 6,000 ft/sec

E 75
a 50
2-
0
0 100 200 309 400 500 700 800
-Distance - ft t
Time mse 7
4 5Q - ' 4,500 ft/sec
200 16, 000 f t/sec
Figure B-26. Wave-front diagram and maximum undetectable

thickness of blind zone (courtesy of Society of Exploration
Geophysicists71 )
Geophone
ctsi- " '00=35.70
Figure B-27. Schematic of shallow, ir-

regular refractor surface producing
different first-layer travel times at
same geophone from shots on either side
of it (from Redpath20
B-48
EM 1110-1-1802
31 May 79
(3) The problem can be overcome by firing a shot in a deep hole

so that arrivals from the intermediate layer are recorded at the sur-
face. Of course, it would be necessary that the existence of the in-
termediate layer be known beforehand because of some other source of
information, such as a drill hole, before this wculd normally be
attempted. If a time-distance curve shows a very large velocity con-
trast between the first and second layers, the existence of a blind
zone might be suspected. The error that results from not knowing the
existence of a blind zone is that the computed depth 'to the refracting
layer is too shallow. However, even in the worst case, it is doubtful
whether the error would approach 50 percent. A nomograph that can be
used to determine the maximum possible thickness of blind zones is given
20
by Redpath.
(4) The hidden layer problem can be a serious drawback to shallow

surveys with the refraction seismograph. The possibility of such a
layer's existence emphasizes the desirability of using an exploratory
drill hole in conjunction with seismic surveys whenever possible.
d. Accuracy and Resolution.
(1) Generally, as the depth of the investigation becomes shallower,

the resolution limits of the method are approached, and the inherent
uncertainties become more important. A large proportion of refraction
surveys for engineering purposes are concerned with depths of 50 to
75 ft and total times of 50 to 100 msec. It is apparent then that a
millisecond is a large unit of time, and arrival times must be picked
with an accuracy of a millisecond or less. This means that the charge
weight, its surrounding medium and coupling, the amplifier gains, and
the placement of the geophones are all important factors in obtaining
the sharp breaks required for accurate timing of first arrivals. An
error of 1 msec corresponds to only 1 ft of depth with a first-layer
velocity of 2,000 ft/sec, but the error would be 5 ft for a first-layer
material with a velocity of 10,000 ft/sec.
(2) Generally, in working with short lines of 500 ft or less, ob-

taining sharp breaks will not be a problem, unless there is difficulty
in burying sufficient explosive deep enough. Because of the attenuation
of the explosive impulse with increasing distance, and particularly be-
cause of the relatively greater attenuation of the high-frequency com-
ponents, the arrivals tend to become less distinct and more rounded with
increasing distance from the shot. The objective is to pick arrival
times to within 1 msec or less, using what appears to be the onset of
the pulse. The presence of background noisc, such
. s that generated by
vehicle traffic, or weak arrivals caused by poor charge coupling, can
B-h9
EM 1110-1-1802
31 May 79
make accurate determinations difficult.
(3) The effect of a 1-ft error in the surveyed elevation cf a shot

point or geophone is the same as a l-msec error in timing for a 2000-
ft/sec material. Any effort spent in obtaining arrival times to a
higher degree of precision would be wasted if surveying errors of this
magnitude existed. The needed degree of precision varies with the
nature of the project for which the survey is made, so no general rule
can be stated. However, the precision needed for a survey should be
considered in pla'ning the fieldwork, and all measurements should be
performed with equivalent degrees of refinement.
(4) It is necessary to maintain close control of overburden veloc-

ities and depths, particularly to distinguish travel time spent in the
overburden from time spent in underlying, higher velocity layers. This
is particularly so when there are more than two layers. Unconsolidated
overburden deposits frequently show lateral changes in velocity over
relatively short distances, and it is almost always desirable to fire
at least one intermediate shot between the ends of the line.
(5) Irregularities in a shallow bedrock surface have a more pro-

nounced effect on accuroy than irregularities in a deeper horizon. A
highly irregular refractor surface will tend to make delay-time methods
somewhat inaccurate because the times being compared represent different
depths, depending on the direction of the shot. This situation is de-
picted in figure B-27. The effect of using delay times to compute
depths will be to "smooth" a highly irregular surface. The situation
is aggravated if the velocity contrasts are not large, i.e., if the
horizontal offsets between the geophone and the point at which the ray
emerges from the refractor (=Z tan a) are large.
(6) Further complications arise if the rock surface is not only

eroded and irregular but also weathered. In this case, the rock surface
may not be a well-defined boundary but rather a zone of transition. In
the case of layered rock, some of the layers-may be better media for the
transmission of seismic energy, and the refractions may not necessarily
be from the top of the rock. Also, as distance from the shot increases,
the higher frequency (i.e. shorter wavelength) energy is progressively
absorbed; because of the signal's increasing wavelength, the energy may
refract from progressively thicker beds. It is necessary to bear in
mind that the wavelength becomes a significant factor when the refractor
is of a limited thickness. Long-wavelength seismic signals do not "see"
thin beds.
(7) Some materials may be anisotropic, i.e., there may be differ-

ences between horizontal and vertical velocities, or even between
B-50
EM 1110-1-1802
31 May 79
horizontal velocities in different directions. These differences may be
as large as 40 percent in some materials. Rocks with finely bedded
structure, such as sandstones and shales, generally exhibit some degree
of anisotropy.
B-51
APPENDIX C
EXAMPLES TO ILLUSTRATE THE USE OF SEISMO
Examples to Illustrate the Use of Seismo
1. This appendix contains three examples of cases where

seismo was used for data reduction and presentation. Each case
was chosen to present and explain a certain aspect of the
programs data reduction capabilities. Example one (Figures 1-4)
gives a case where a forward traverse only was done (shotpoint on
one end of the spread only), this case is presented to explain
how seismo handles a single sided traverse. Example two (Figures
5-12) gives a case where a forward and reverse traverse were
performed (shotpoint on both ends of the line). This case also
presents the use of a shorter line on one or both ends
(overburden line) to serve as a more precise measurement of the
near surface layering. This case is repeated with the units
changed from feet to meters. Example three (Figures 13-17)
presents a case in which the terrain correction routine was
utilized to correct the data for elevation changes.
2. Examle . The case presented as example 1 is shown in

Figures 1 through 4. Figure 1 is a printout of the data reduced
from information collected in the field. For this particular
case, there were twelve geophones in the spread. A shotpoint was
placed on one end of the line only, referred to as the forward
traverse. Notice that the printout indicates zero reverse points
beside the heading REVERSE POINTS. The data is located in a file
named KWAJl.REF, and has the identifier kwajl. The total line
length from shotpoint to last geophone is 42 ft, the geophones
are spaced 2 ft apart, and the shotpoint to first geophone
distance is 2 ft. The file also contains a model that was saved
with the data, and is shown at the bottom of the figure. This
model information is input only, meaning that the results, layer
velocities and depths to interfaces, would have to be calculated
through the program using the invMOD and Calculate options.
3. A plot of the data from example 1 is shown in Figure 2.

This is the working plot, 5x7 inches, used to present the data on
a large scaie to aid in determination of the layering. The plot
also contains a working model represented by the solid lines
running through the data. Notice that there is a reverse model,
even though there is no reverse data (no shotpoint at the far
end). This illustrates the method seismo uses to calculate the
results. If only a forward model exist (or alternately a reverse
model only) the program will make the reverse model equal to the
forward model because both are necessary to calculate the
C3
##### SEISMIC DATA SUMMARY #####
DATA FILE NAJE : a:KWAJ1.REF

DATA IDENTIFICATION : kwajl
LINE LENGTH : 42 ft

ft msec ft msec
2.0 1.2 0.0 0.0

4.0 2.9 0.0 0.0
6.0 4.1 0.0 0.0
8.0 5.5 0.0 0.0
10.0 6.5 0.0 0.0
12.0 7.9 0.0 0.0
17.0 11.0 0.0 0.0
22.0 12.5 0.0 0.0
27.0 13.6 0.0 0.0
32.0 14.2 0.0 0.0
37.0 15.2 0.0 0.0
42.0 16.2 0.0 0.0
END OF DATA

1543 0.0 1543 0.0

4721 7.5 4721 7.5
Figure i. Printout of time-distance data for example 1.
C4
01I
LI
H IL i
U)
H
ILI
IL:3 4
WI 0
IL rI
IC
HIu I
y In
I~I
SH ~L
XI
C5 I
kwajl
*** INPUT DATA *** *** COMPUTED SEISMIC PROFILE ***

----------------------------- ----------------------------- I
I FORWARD REVERSE LAYER TRUE DEPTH
ILAYER VEL TIME VEL TIME # VEL FOR REV
ft/s msec ft/s msec ft/s ft ft
1 1543 0.0 1543 00 1 1540

2 4721 7.5 4721 7.5 2 4720 6.0 6.0
Figure 3. Printout of input data and calculated results for example i.
C6
m 1'c
M 11 ci
LUvl 0
0 0 0 Mn
4 1
a a me4
L. E w C a14
XH L a 1 0xa
2 0rl\ IW L a S
02 0 u4 Ll
4 cu a .4.4
0)
/ -0
/ .4-
N 4J
rl
/ 1L
* / /
/
C7-
o
*E
results. This in affect means that the layers can only be
interpreted as being horizontal with no indication of dipping.
The plot also contains the identifier, filename where data is
located, program version number and authors.
4. Figures 3 and 4 contain the calculated results of the

input data and model, with Figure 4 being the report ready
presentation. The input data are velocities and intercept times
for each layer, the calculated results are the true velocity for
each layer and the depth to each interface. Notice that the
input data for the reverse traverse has been made to equal that
of the forward traverse, and that the calculated depth to
interface for the forward and reverse end of the spread are the
same. This is indicative of a single sided spread. Figure 4
contains the plot and the results, with the computed results
presented as an idealized profile.
5. Example 2. The case presented as example 2 is shown in

Figures 5-12. Figure 5 is a printout of the data located in a
file named KWAJ12M.REF, and with the identifier kwajl2. The
seismic refraction spread contained 12 geophones and extended,
from shotpoint to shotpoint, 110 ft. In addition, there was a
short line, 42 ft, run at the reverse end of the spread which
contained 12 geophones and is referred to as the overburden line.
This line had a shotpoint at one end only, geophones spaced 2 ft
apart, and a distance from shotpoint to first geophone of 2 ft.
This line was performed to better delineate the near surface
layering. Doing an overburden spread is common practice,
especially when the distance between geophones on the main spread
are large. Too large a geophone spacing can result in missing
the first layer if it is of a thickness less than the distance
between the shotpoint and first geophone. In this case the
overburden line data was added to the end of the file, it could
also reside in a separate file and plotted on the plot using the
overlay routine. This file also contains a model printed at the
bottom of the page.
6. The 5x7 inch working plot of the data is presented in

Figure 6. The overburden line can be seen plotted along with the
main spread data at the far end of the spread. Notice that the
overburden line in addition to the main spread gives nine points
in the first layer on the reverse end as compared to two points
for the Lirst layer on the forward end. The file location and
identifier for the spread is located at the top of the plot.
7. The results of the data for example two are presented
08
# SEISMIC DATA SUMMARY
DATA FILE NAME : A:KWAJ12M.REF

DATA IDENTIFICATION : kwajl2
LINE LENGTH : 110 ft

ft msec ft msec
-- -----------------------------------------------
5.0 4.3 5.0 2.9
10.0 7.9 10.0 6.2
20.0 10.3 20.0 12.2
30.0 13.2 30.0 14.2
40.0 14.9 40.0 16.1
50.0 17.0 50.0 17.8
60.0 19.0 60.0 19.7
70.0 20.9 70.0 20.9
80.0 22.3 80.0 22.8
90.0 24.2 90.0 24.5
100.0 25.9 100.0 25.2
105.0 26.9 105.0 26.2
2.0 1.2
4.0 2.9
6.0 4.1
8.0 5.5
10.0 6.5
12.0 7.9
17.0 11.0
22.0 12.5
27.0 13.6
32.0 14.2
37.0 15.2
42.0 16.2
END OF DATA

1347 0.0 1586 0.0

5095 7.1 5202 8.2
6217 9,7 6725 10.9
Figure 5. Printout of time-distance data for example 2.
C9
z
0
cc <
w I
> (n C
U) UJ
H J
W :3 L 3
LC
I-.
4
0-
4
U)
-1
liiio
kwajl2
*** INPUT DATA *** I*** COMPUTED SEISMIC PROFILE ***

--------------------------------------------- I - ------------
II ...........
ILA~iR 'L TIME VEL TIME I VEL FOR REV
# t/s msec ft/s msec ft/s ft ft
1 1347 0.0 1586 0.0 1
2 5095 7.1 5202 8.2 2 5150 5.5 6.5
3 6217 9.7 6725 10.9 3 6460 16.0 17.0
Figure 7. Printout of input and calculated data for example 2.
Cll
K..ID
ri I
4 a I 0 c
> 0 tro I IL0

1% 3 0 0 0 c 's
__ A
L.Lr2L
1
0I..
ID10u ___
00
CL
IL
r
/c
1~I/
C12
##### SEISMIC DATA SUMMARY
DATA FILE NAME : kwajmetr.ref

DATA IDENTIFICATION : kwajl2
LINE LENGTH : 34 m

m msec m msec
1.5 4.3 1.5 2.9

3.0 7.9 3.0 6.2
6.0 10.3 6.0 12.2
9.1 13.2 9.1 14.2
12.1 14.9 12.1 16.1
15.2 17.0 15.2 17.8
18.3 19.0 18.3 19.7
21.3 20.9 21.3 20.9
24.4 22.3 24.4 22.8
27.4 24.2 27.4 24.5
30.5 25.9 30.5 25.2
32.0 26.9 32.0 26.2
0.6 1.2
1.2 2.9
1.8 4.1
2.4 5.5
3.0 6.5
3.7 7.9
5.2 11.0
6.7 12.5
8.2 13.6
9.8 14.2
11.3 15.2
12.8 16.2
END OF DATA

m /sec msec m /sec msec
411 0.0 483 0.0

1553 7.1 1586 8.2
1895 9.7 2050 10.9
Figure 9. Printout of time-distance data for example 2 in meters.
C13
I E
tH / Lc
Hi a
E
~i I
UP,
/4-, 0I
zX
IL L
H +;
I~wi
LL~
U)sw
001
kwajl2
V *** INPUT DATA *** I*** COMPUTED SEISMIC PROFILE *** I

------------------------------ ----------------------------- I
t FORWARD REVERSE LAYER TRUE DEPTH
ILAYER VEL TIME VEL TIME VEL FOR REV
#I m s insec m /s msec rn/s m m I
1 411 0.0 483 0.0 1 450
2 1553 7.1 1586 8.2 2 1570 1.5 2.0
3 1895 9.7 2050 10.9 3 1970 5.0 5.5
ti
VI
I I
Figure Ii. Prinout of results for example 2 in meters.
C15
> m t I E
C.> u 3 c
(r0%. 0 9
z 10 I r
0 L
Flr~ ri.
0)
43
CD
C16)
in Figures 7 and 8. This particular case revealed a three layer
system for the site, with the last layer extending to an unknown
depth, limited by the spread length. Also notice that the layers
are dipping downward (positive angle) from distance equal 0 ft to
distance equal 110 ft. This is shown in the idealized profile.
8. This case has been repeated in Figures 9-12 with the

units changed to meters as opposed to feet. The case is the same
as described above, except that the printouts have changed from
units of ft and ft/sec to m and m/sec, and all distances are in
meters.
9. E The case presented as example 3 is shown in

Figures 13-17. Figure 13 is a printout of the data as reduced
from the field. The data are located in a file in directory QB4
named WSB4R8.REF with the identifier wsb4r8. The spread had 24
geophones spaced 2 ft apart ,nd a shotpoint at each end spaced 2
ft from the first geophone fcrward and reverse respectively. The
total length of the spread was 50 ft. There was a model saved
with the data, and it is printed at the bottom of Figure 13.
This case is presented to show how the terrain/shotpoint
correction routine functions.
10. The 5x7 inch working F)ot of the data is presented in

Figure 14. It can be clearly seen that the data are affected by
terrain. At a distance of 10 ft from the forward shotpoint, the
arrival times suddenly drop off from the strait line segment
placed through the data. This same effect can be seen in the
case of the reverse data. The arrival times are all less than
would be expected, which indicates that they must have either
passed through a fast zone or had a shorter travel path. In this
case the effects of terrain are producing shorter travel paths (
a dip in the topography) and hence the reduction in travel times.
Therefore, the data were terrain corrected in the program to
remove the effects of topography on the data set. There was no
correction made for shotpoint depth since the source was placed
at the surface, however, that could also be included in the
correction.
11. The results of the terrain corrected data are

presented in Figures 15-17. The printout of the terrain
corrected data is given in Figure 15. The printout contains the
usual file location, line information, data, and model. The data
are distances from source to each geophone (Dist), original
uncorrected time (UTime), corrected time (CTime), elevations
entered (Cor), and velocities used in correction routine (CVel).
C17
SEISMIC DATA SUMMARY #####
DATA FILE NAME : \QB4\WSB4R8.REF

LINE LENGTH : 50 ft

ft msec ft msec
2.0 3.0 2.0 4.0

4.0 4.5 4.0 5.0
6.0 6.0 6.0 6.8
8.0 7.0 8.0 7.5
10.0 8.0 10.0 8.5
12.0 7.5 12.0 9.0
14.0 8.0 14.0 10.0
16.0 10.0 16.0 10.5
18.0 11.0 18.0 11.5
20.0 12.5 20.0 12.5
22.0 14.0 22.0 13.0
24.0 16.5 24.0 14.5
26.0 16.0 26.0 16.5
28.0 17.5 28.0 18.0
30.0 18.5 30.0 19.5
32.0 21.5 32.0 20.5
34.0 22.0 34.0 22.0
36.0 23.0 36.0 23.0
38.0 24.0 38.0 24.0
40.0 26.0 40.0 25.0
42.0 27.5 42.0 26.0
44.0 28.0 44.0 28.0
46.0 30.0 46.0 29.0
48.0 32.0 48.0 31.0
END OF DATA
*** FORWARD MODEL * *** REVERSE MODEL ***

688 0.0 542 0.0

1694 2.4 1776 2.6
Figure 13. Printout of time-distance data for example 3.
018
101
w II
H
ii
I 1
itr
C"
'L U
~H/
0H
MB Irc I
/
Z (nilc
h4
0 .4
(-UU)31
019
##### SEISMIC DATA SUMMARY #####
DATA FILE NAME : \QB4\WSB4R8.REF

LINE LENGTH : 50 ft
Dist UTime CTime Cor CVel Dist UTime CTime Cor CVel
ft msec msec ft ft/sec ft msec msec ft ft/sec
2.0 3.0 3.0 0.0 615.0 2.0 4.0 4.0 0.0 615.0
4.0 4.5 4.5 0.0 615.0 4.0 5.0 5.0 0.0 615.0
6.0 6.0 6.0 0.0 615.0 6.0 6.8 6.8 0.0 615.0
8.0 7.0 7.0 0.0 615.0 8.0 7.5 7.5 0.0 615.0
10.0 8.0 8.0 0.0 615.0 10.0 8.5 8.5 0.0 615.0
12.0 7.5 9.8 1.5 615.0 12.0 9.0 10.0 0.0 615.0
14.0 8.0 10.3 1.5 615.0 14.0 10.0 11.0 0.0 615.0
16.0 10.0 12.3 1.5 615.0 16.0 10.5 11.5 0.0 615.0
18.0 11.0 13.3 1.5 615.0 18.0 11.5 12.5 0.0 615.0
20.0 12.5 14.8 1.5 615.0 20.0 12.5 14.8 1.5 615.0
22.0 14.0 16.3 1.5 615.0 22.0 13.0 15.3 1.5 615.0
24.0 16.5 16.5 0.0 615.0 24.0 14.5 16.8 1.5 615.0
26.0 16.0 18.3 1.5 615.0 26.0 16.5 16.5 0.0 615.0
28.0 17.5 19.8 1.5 615.0 28.0 18.0 18.3 1.5 615.0
30.0 18.5 20.8 1.5 615.0 30.0 19.5 19.8 1.5 615.0
32.0 21.5 21.5 0.0 615.0 32.0 20.5 20.8 1.5 615.0
34.0 22.0 22.0 0.0 615.0 34.0 22.0 22.3 1.5 615.0
36.0 23.0 23.0 0.0 615.0 36.0 23.0 23.3 1.5 615.0
38.0 24.0 24.0 0.0 615.0 38.0 24.0 24.3 1.5 615.0
40.0 26.0 26.0 0.0 615.0 40.0 25.0 25.0 0.0 615.0
42-.0 27.5 27.5 0.0 615.0 42.0 26.0 26.0 0.0 615.0
44.0 28.0 28.0 0.0 615.0 44.0 28.0 28.0 0.0 615.0
46.0 30.0 30.0 0.0 615.0 46.0 29.0 29.0 0.0 615.0
48.0 32.0 32.0 0.0 615.0 48.0 31.0 30.0 0.0 615.0
END OF DATA

688 0.0 542 0.0

1694 2.4 1776 2.6
Figure i5. Printout of corrected time-distance data example 3.
C20
I *1
> U)
I I m*
0 '0
S 4.'
wt
I~ L
L
I *u
LL 7
0~ 0
zI 4/
___ J 10
<021
wsb4r8
*** INPUT DATA *** *** COMPUTED SEISMIC PROFILE ***

-----------------------------------------------
LAYER VEL TIME VEL TIME VEL F., REV
# ft/s msec ft/s msec ft/s ft ft
1 688 0.0 542 0.0 1 620

2 1694 2.4 1776 2.6 2 1730 1.0 1.0
l I I
Figure 17. Printout of results from example 3.
022
Notice that the geophones from distance 12 ft to distance 30 ft
(forward) have all been corrected to a common datum plane, which
was selected to be the elevation of the shotpoint. In other
words, there was a dip in the line at this location and the
geophones were elevated to the new location selected as the datum
plane by adding 1.5 ft to them. The program takes this 1.5 ft
and the velocity of 615 ft/sec to calculate the correction in
msec to be added or subtracted from the data. Figure 16 is a
plot of the data with the corrections applied. Comparing this
plot with that in Figure 14, a marked difference in the arrival
times can be seen. Figure 17 is a printout of the calculated
results for this model.
023

Seismic Refraction Data Processing Software: Seismo Version 2.7

Uploaded by

Copyright:

Available Formats

Seismic Refraction Data Processing Software: Seismo Version 2.7

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Seismic Refraction Data Processing Software: Seismo Version 2.7

Uploaded by

Copyright:

Available Formats

INSTRUCTION REPORT GL-90-2

Approved for Public Release; Distribution Unlimited

Prepared for DEPARTMENT OF THE ARMY

US Army Engineer Topographic Laboratory

IBM is a registered trademark of International Business Machines

Destroy this report when no longer needed. Do not return it to

The findings in this report are not to be construed as an official

The contents of this report are not to be used for

4. PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANIZATION REPORT NUMBER(S)

Instruction Report GL-90-2

3909 Halls Ferry Road

17, TITLE (Include Security ClaOs1nc8ation)

Seismic Refraction Data Processing Software: SEISMO Version 2.7

Yule, Donald E.; Sharp, Michael

K 19. ABSTRACT (Continue on reverse if necessary and idepf), by block number)

20. DISTRIBUTION/AVAILABILITY OF ABSTRACT "21 ABSTRACT SECURI Y CLASSIFICATION

8 a. NAME OF FUNDING/SPONSORING ORGANIZATION (Continued).

8c. ADDRESS (Continued).

CONVERSION FACTORS, NON-SI TO SI UNITS OF MEASUREMENT...................4

PART IX: RESULTS MENU................................................ 47

Non-SI units of measurement used in this report can be converted to SI

SEISMO VERSION 2.7

1. The software package "SEISMO" will perform seismic refraction data

reduction, interpretation, and presentation. SEISMO will do both forward and

inverse modeling of refraction data, calculation of results, display of

modeling of refraction data are presented in Appendixes A and B, respectively.

2. This program is not presented as a state-of-the-art procedure for

complicated algorithms, such as the general reciprocal method, exist, these

algorithms also require a more sophisticated understanding of seismic theory.

The algorithm used in this program is a means of obtaining and reducing

seismic data, which will suffice for most engineering applications.

3. Part I of this document gives an overview of SEISMO, the basic steps

installation procedure. Part II explains how to start the program, and

step-by-step description of how each one functions in the program. Appendix A

Appendix B gives the formulas for seismic refraction depth calculations,

terrain corrections, and shot depth corrections. Appendix C gives examples of

field work performed and analyzed with the SEISMO program.

4. The following is a list of the typographic convention used

"press" Key indicated should be pressed.

<return> Enter or return key should be pressed.

"type" Type the information requested and press

<Esc> Press key labeled "Esc" on the keypad.

S"The control character indicating

5. This section of the report gives an overview of SEISMO's features

6. The SEISMO program was written to provide an aid to analyzing and

Basic Steps to Using SE(SMO

7. The following is a step-by-step procedure for the input, modeling,

Hardware and Software Requirements

8. The SEISMO program was developed under Microsoft(MS)-DOS using

General requirements IBM Personal Computer or "compatible."

Plotters and Printers Supported

9. The plotter must be connected to either serial port 1 or 2 (coml or

SEISMO.exe Invokes the SEISMO

Scrn2.pic Control menu for model

Scrn3.pic Control menu to

Scrn4.pic Control menu for

Scrn5.pic Control menu for

Scrn6.pic Control menu for

Helpl.pic Display of help screen

SEISMO

* FORWARD POINTS * * REVERSE POINTS *

* FORWARD MODEL * * REVERSE MODEL *

- FORWARD MODEL A * REVERSE MODEL *

* FORWARD POINTS * * REVERSE POINTS *

* FORWARD MODEL * * REVERSE MODEL *

SEISMO