A tutorial for processing a 2D land line using

Seismic Unix

Karl Schleicher

ABSTRACT

This paper described how to process a public 2D land line data set though a verybasic processing sequence using Seismic Unix. The data from the north slopesof Alaska has good signal, although it may be suitable for testing ground rollattenuation programs. The detailed steps download the data from the internetand process it are described. You should be able to follow these steps and recreatemy results. I hope this will accelerate the testing and validation of programsdeveloped in research groups.The tutorial starts with background information like data downloading, regionoverview, data aquisition, and previous processing. The scripts and custom pro-grams to translate to seismic unix, load headers, qc, apply preprocessing (in-cluding velocity filtering and decon), cdp gather, velocity analysis, and stack arelisted and described.Header loading is not described in other Seismic Unix documentation. One ap-proach to header loading is described here. The velocity analysis script is betterthan I found in previous papers.My effort to process these data is ongoing. I have not computed and appliedresidual statics. I have not migrated the data. My current results are not as goodas obtained when the data was originally processed by GSI in 1981. I would liketo provide my most current results to anyone who wants to use publicly availablesoftware to process seismic data.This paper is the beginning of an effort to build a library of open-access seismicdatasets and processing scripts using open-source geophysical software.

INTRODUCTION

Seismic Unix (SU) is one of the best known open software packages for seismic pro-cessing. The distribution includes some example processing scripts and there are twobooks that teach how to use the software, Geophysical Image Processing with SeismicUnix (?) and Seismic Processing with Seismic Un*x (Forel et al., 2005) books coverbroad subjects including basic Unix commands, setting up a user environment, basicseismic processing, and more advanced scripting for seismic processing. You can fol-low the text and recreate some seismic processing. These books to not provide muchinformation about creating trace headers or preprocessing.

2

This paper assumes you are familiar with Seismic Unix. I concentrate on providingexample scripts and programs to process a 2D land line using the processing system.I describe the data acquisition and previous data processing. I provide the customprograms I used to load the trace headers. The scripts for preprocessing, velocityanalysis, and stacking are provided and explained. These scripts provide the detailedparameters used by the Seismic Unix programs. The parameters for deconvolution,velocity filter, mute, moveout velocity, and scaling can adapted for use by otherseismic processing systems. Currently I have not applied residual statics or migration.The results in the deep section are similar to the results obtained in 1981. The shallowsection is not as good as previously obtained.

This paper provides detailed steps that describe how to download the data fromthe internet and process it using seismic unix. You should be able to follow thesesteps and recreate my results. The scripts, data, and programs can be modified toallow you to validate your own ideas about seismic processing. This should acceleratethe testing and validation of new seismic research.

BACKGROUND INFORMATION ABOUT THE DATASET

President Warren Harding created the Naval Petroleum Reserve Number 4 in 1923.It was renamed the National Petroleum Reserve, Alaska (NPRA) in 1976. Figure 1is a map of the region and Figure 2 maps the lines collected and processed between1974 and 1981. I selected and processed line 31-81 because it was a short line fromthe most recent acquisition season.

The files I found for line 31-81 are:

L23535.SGY unprocessed seismic data in segy format

L23536.SGY

L23537.SGY

3181O.PDF the Observer’s log

3181.SGY segy of the 1981 final stack

3181S.PDF the surveyor’s log

3181.SPT the textfile of shotpoint, latitude, longitude.

S609.JPG A 1981 final stack in jpeg format.

S609.TIF A higher resolution stack in tif format

I downloaded these files into my directory, /home/karl/data/alaska/31-81/clientfiles.I was able to view the large JPG and TIF files in OpenOffice.org draw (other programs

3

had problems). I zoomed the display of S609.TIF to look at the side label thatdescribes the acquisition and processing parameters.

The data was acquired 96 trace, 12 fold with a dynamite source. The shotpointinterval is 440 ft and the receiver interval is 110 ft. The average shot depth is 75 ft.The elevation plot on the section header shows the elevation is about 200 ft and thereis little variation.

The processing applied by GSI in 1981 was:

1. Edit and true amplitude recovery

2. spherical divergence and exponentiation (alpha - 4.5 db/s, initial time - 0 s,final time - 4.5 s)

3. Velocity filtering dip range -12 to 4 ms/trace

4. Designature deconvolution

5. Time-variant scaling (unity)

6. Apply datum statics

7. Velocity estimation

8. Normal moveout correction

9. Automatic Residual Statics application

10. First Break Suppression100 ms at 55 ft380 ms at 2970 ft700 ms at 5225 ft

11. Common depth point stack

12. Time Variant filtering 16-50 Hz at 700 ms14-45 Hz at 1400 ms12-40 Hz at 2000 ms8-40 Hz at 400 ms

13. Time variant scaling

DATA LOADING AND INITIAL DATA QC

I found little documentation about loading data and creating trace headers in preciouspublications. This section provides a detailed description.

4

I created a script file to load the first segy file, /home/karl/jobs/alaska/31-81/load/view1.job.This file is listed in Appendix 1.

The way I developed this script is by pasting a command or two into a terminalwindow and picking information from the print of displays to write the next few com-mands. The segyread command translates data to SU format. The surange commandprints: 3030 traces: tracl 1 3030 (1 - 3030)tracr 1 3030 (1 - 3030)fldr 101 130 (101 - 130)tracf 1 101 (1 101)trid 1ns 3000dt 2000f1 0.000000 0.000000 (0.000000 - 0.000000)

The print indicates there is little information in the input headers. There are 31shot records (fldr 101 - 130) and each record has 101 traces (tracf 1 - 101). Thesuximage produces Figure 1 and the zoom plot Figure 2.

The first 10 records in the file are test records. These tests records are the unuisualtraces in the left third of Figure 1. The Observer Log indicates the first shotpointis field file identifier (ffid) 110. There are also 5 auxiliary traces in each shot record.These auxiliary traces can be seen o nfigures 4 and 5.

The suxmovie command allows all the shots record on the line to be displayed.The parameter n2 is set to the number of traces in each shot record, which was basedon the tracf range printed in surange (1 - 101). The movie loops continuously andcan be stopped by pressing s. Once stopped, you can move forward or backward usingthe f and b. An example screen is shown in Figure 5. This plot allowed me to seethat traces 1-96 were the data channels and 97-101 were the auxiliary traces.

The final command in the view1.job file is the sugethw. This prints the fldr andtracf header keys. These are the keys I used to map the data to the shotpoint andreceiver locations.

I wrote a custom java program to load the trace headers. The code is not elegant,but it is straightforward to write and it works. It is listed in Appendix 2. This programrequire files to define the shotpoint elevation and the FFID/EP relationship. The sur-veyor log, downloaded as: /home/karl/data/alaska/31-81/clientfiles/3181O.PDF de-fines the shotpoint elevations. The observers log, downloaded as: /home/karl/data/alaska/31-81/clientfiles/3181O.PDF describes the relationship between the su header keys ffidand ep (these are called REC. and SHOTPOINT by in the observers log). Thisinformation was typed into two files listed in appendices 3 and 4.

The program to load the headers was run by the script: /home/karl/jobs/alaska/31-81/load/hdrload1.job This file is listed in appendix 5.

The other two segy files were loaded by scripts view2.job, view3.job, hdrload2.job,

5

0

1

2

3

4

5

1000 2000 3000

Figure 1: First 300 traces on the first tape.

6

0

0.5

1.0

1.5

2.0

2.5

3.0

1000 1050 1100 1150 1200 1250

Figure 2: Zoom of figure 1.First look at the data. Notice the ground roll and goodsignal. There are 101 traces in each shotpoint, 5 auxiliary traces and 96 data traces.

7

and hdrload3.job in the same directory, /home/karl/jobs/alaska/31-81/load. Thesescripts differ from the previous scripts only by the names of the input and output file.

The stack file from the previous process was translated to SU format and displayedusing the viewqcstack.job script listed in appendix 6. This is a straight forwardexecution of segyread and suximage. Display of the checkstack is in Figure 6.

SHOT RECORD VELOCITY FILTER

I applied Velocity filtering (sudipfilt) to remove the groundroll. The receivers leadingthe shotpoint were processed separately from the trailing receivers. This allowed anasymmetrical dip filter to be applied (-15,5 ms/trace). These parameters are looselybased on the 1981 processing that used (-12,4). Sudipfilter was intended for poststack processing, so a loop is required in the script to divide the data into individualshotrecords and seperate the positive and negative offsets. The script also includessuxmovie of the velocity filtered data. It is listed in appendix 7. Figure 7 is anexample shot record after velocity filtering,

SHOT RECORD EDIT, MUTE, AND CDP SORT

There is a bad shotpoint I removed. I also applied a mute and sorted the data to CDPorder in the same script. The script is listed in Appendix 8. Figure 8 is a display oftwo cdp gathers.

VELOCITY INTERPRETATION

I used a long script that combined several su program for velocity interpretation. Thisscript is listed in Appendix 9. I combined the scripts from Forel et. al. 2005 (iva.shsection 7.6.7.3 and velanQC.sh section 8.2.2.2). The script is more practical, becauseyou can start from no velocity function, or review and update and existing velocityfield. The script hos more capabilities, but it is tidier and a little shorter than Forel’sscript.

Figure 9 is an example velocity analysis plot. There are four components of theplot, semblance plot, constant velocity stack (CVS) plot, cmp gather without NMOand cmp gather after NMO. The semblance plots and CVS plot have the velocityfunction overplotted.

The velocities I picked are listed on Appendix 10.

8

STACK

The script that stacks the data is in Appendix 11. This script also applied deconusing the command supef. The script also applied AGC. I wanted to start the decondesign gate start time to increase with offset so the shallow data near the mute wasnot included. The only way I knew how to do this us SU was to apply linear moveoutto move the design gate start time to a constant time. This was accomplished byremoving the sign from the offset header and computing a time static from the offset.Supef was bracketed with sustatic to apply and remove the offset dependent static.I compared the stack created in SU to the original processing results created in 1981(figures 10 and 11).

COMPARISON OF THE SU AND THE 1981 RESULTS

Figures 10 and 11 are the results created using SU and the results obtained in 1981.Some of the differences in the processing sequences are:

1. The 1981 processing used GSI’s designature process. This source signatureremoval applies the same operator to all traces in a shotrecord. The filteris designed differently from conventional deconvolution and the two types ofdeconvolution will not produce data with the same wavelet.

2. There are many differences in the application of AGC. For example the GSIused to apply AGC before velocity filtering inverse AGC after velocity filter.

3. GSI used diversity stack to control the high amplitude noise bursts.

4. The original processing includes residual statics.

Considering all these differences, I think the results are surprisingly similar, especiallybelow 400 ms. I think residual statics account for most of the differences. The shallowresult on the SU processing is worse than the 1981 processing.

In general, I found the SU software hard to use. Sudipfilter was not intended forshotrecord velocity filtering, so a collection of program were required. Many of myinput errors were not trapped by the code. I was disappointed that the processingapplied 30 years ago produced better results than I obtained using SU. SU is a usefulprototyping environment, but it falls short of the commercial packages I have used.

PLANS

Some of the ideas I have to continue this work include:

1. Adding more processes, especially residual statics and migration.

9

2. Process using other software (Open source of proprietary.

3. Working on different datasets (marine, 3D, ...).

4. Make scripts and intermediate results available on the internet.

5. Improve open source software.

CONCLUSIONS

I have provided a set of scripts the process a land line using SU. These scripts includethe processing stages:

1. data load

2. trace header creation

3. velocity filtering

4. cdp gather

5. velocity analysis

6. stack

Although this sequence omits important processes like residual statics and migra-tion, it illustrates processing stages not presented by Forel or Stockwell.

My processing results can be recreated by others since the processing scripts areincluded in appendixes of this paper and the data can be downloaded from the inter-net.

The task of loading trace headers is no addressed in other publications. Thecustom java program included in this paper is one way to complete this processingstage. The velocity analysis script (iva.sh in Appendix 9) improves the scripts inForel’s book. I was able to pick the velocities on this line using the script.

Some of the ideas I have to continue this work include:

1. Adding more processes, especially residual statics and migration

2. Process using other software (Open source of proprietary.

3. Working on more data (Marine, 3D, ...)

4. Make scripts and intermediate results available on the internet

5. Improve open source

10

REFERENCES

Forel, D., T. Benz, and W. D. Pennington, 2005, Seismic data processing with seismicun*x: Society of Exploration Geophysicists.

APPENDIX A

SCONSTRUCT FILE

1 from r s f . supro j import ∗2 import r s f . p ro j3 # need to add the v e l o c i t y ana l y s i s s c r i p t4 # need to add migrat ion5 # need to t r y s ud i v s t a c k6

7 ####################8 # Fetch the da t a s e t and conver t to mu l t i p l e r s f f i l e s9 ####################

10

11 segyread = ’ ’ ’12 segyread b f i l e=${TARGETS[ 1 ] } h f i l e=${TARGETS[ 2 ] }13 tape=$SOURCE endian=%d14 ’ ’ ’ % (1− l i t t l e e n d i a n ( ) )15

16 rawsegy=[ ’ L23535 ’ , ’ L23536 ’ , ’ L23537 ’ ]17 for f i l e in rawsegy :18 r s f . p ro j . Fetch ( f i l e+’ .SGY ’ ,19 s e r v e r=’ http :// certmapper . c r . usgs . gov ’ ,20 top=’ n e r s l /NPRA/SEISMIC/1981/31 81 ’ ,21 d i r=’DMUX’ )22 Flow ( [ f i l e , f i l e+’ . bin ’ , f i l e+’ . asc ’ ] ,23 f i l e+’ .SGY ’ , segyread , s td in =0)24

25

26 r s f . p ro j . Fetch ( ’ 31 81 PR .SGY ’ ,27 s e r v e r=’ http :// certmapper . c r . usgs . gov ’ ,28 top=’ n e r s l /NPRA/SEISMIC/1981/31 81 ’ ,29 d i r=’PROCESSED ’ )30

31

32 # conca t ina t e the input f i l e s33 Flow ( ’ l i n e ’ , rawsegy ,34 ’ cat L23535 . su L23536 . su L23537 . su ’ ,35 s td in =0)

11

36

37 # d i s p l a y the f i r s t 3000 t r a c e s38

39 Result ( ’ f i r s t ’ , ’ l i n e ’ ,40 ’ suwind count=3000 | suximage perc=90 ’ )41

42 Result ( ’ z oomf i r s t ’ , ’ l i n e ’ ,43 ’ ’ ’44 suwind key=t r a c l min=1000 max=1250 tmin=0 tmax=3 |45 suximage perc=9046 ’ ’ ’ )47

48 #Resu l t ( ’ f i r s tmo v i e ’ , ’ l i n e ’ ,49 # ’ ’ ’50 # suxmovie n2=101 loop=1 fframe=1 dframe=151 # t i t l e =”record %g” perc=2052 # ’ ’ ’ )53

54 Result ( ’ f i r s t r e c 2 4 ’ , ’ l i n e ’ ,55 ’ ’ ’56 suwind key=t r a c l min=2324 max=2424 tmin=0 tmax=6 |57 suximage perc=9058 ’ ’ ’ )59 Flow ( ’ h d r f i l e . txt ’ , ’ l i n e ’ ,60 ’ sugethw output=geom key=t r a c l , f l d r , t r a c f ’ )61

62 #Flow ( ’ In terpText . c l a s s ’ , ’ In terpText . java ’ ,63 # ’ ’ ’64 # javac ${SOURCE}65 # ’ ’ ’ , s t d i n=0)66

67 #Flow ( ’ h d r f i l e 1 . t x t ’ , [ ’ h d r f i l e . t x t ’ , ’ In terpTex t . c l a s s ’ ] ,68 # ’ ’ ’69 # java InterpTex t70 # ’ ’ ’ , s t d i n=0)71

72 Flow ( ’ h d r f i l e 1 . txt ’ , [ ’ h d r f i l e . txt ’ , ’ InterpText . py ’ ] ,73 ’ ’ ’74 . / InterpText . py75 ’ ’ ’ , s t d in =0)76

77 Flow ( ’ b i n a r y h d r f i l e 1 . dat ’ , ’ h d r f i l e 1 . txt ’ ,78 ’ ’ ’79 a2b n1=1380 ’ ’ ’ )

12

81

82 keys=’ ep , sx , sy , gx , gy , cdp , t r a c f , o f f s e t , ’ \83 ’ s e l ev , ge lev , s s ta t , gstat , t s t a t ’84

85 Flow ( ’ a l l s h o t s ’ , [ ’ l i n e ’ , ’ b i n a r y h d r f i l e 1 . dat ’ ] ,86 ’ ’ ’87 sushw88 i n f i l e=b i n a r y h d r f i l e 1 . dat89 key=%s |90 sushw91 key=f192 a=0 |93 suwind94 key=cdp95 min=−99999896 max=999999 |97 suwind98 key=ep99 min=−999998

100 max=999999 |101 suwind102 key=ep103 r e j e c t =149104 ’ ’ ’ % keys )105

106 #Resu l t ( ’ t e s t ’ , ’ a l l s h o t s ’ ,107 # ’ ’ ’108 # suxmovie n2=96 loop=1 fframe=1 dframe=1109 # t i t l e =”a l l s h o t s record %g” perc=20110 # ’ ’ ’ )111

112

113 # d i s p l a y t r a c e s 1010−1250 f i r s t 3100 ms ( or 1550 samples )114 #Resu l t ( ’ second ’ , ’ l i n e ’ ,115 # ’ ’ ’116 # window f2=1010 n2=240 n1=1550 |117 # agc rec t1=250 rec t2=100 |118 # grey t i t l e =”F i r s t 3000 t r a c e s ” p c l i p=70119 # ’ ’ ’ )120

121 #Fetch ( ’31 81 PR .SGY ’ ,122 # serve r=’ h t t p :// certmapper . cr . usgs . gov ’ ,123 # top=’ n e r s l /NPRA/SEISMIC/1981/31 81 ’ ,124 # di r=’PROCESSED ’)125

13

126 # conver t s t a c k and view i t :127 f i l e=’ checkstack ’128 Flow ( [ f i l e , f i l e+’ . bin ’ , f i l e+’ . asc ’ ] ,129 ’ 31 81 PR .SGY ’ , segyread + ’ ’ ’130 | sushw key=ntr a=534 | sushw key=d1 a=0131 ’ ’ ’ ,132 s td in =0)133

134 Result ( ’ checkstack ’ , ’ checkstack ’ ,135 ’ suximage t i t l e =”checkstack ” perc=90 ’ )136 Result ( ’ zoomcheckstack ’ , ’ checkstack ’ ,137 ’ ’ ’138 suwind tmin=0 tmax=3 |139 suximage t i t l e =”zoom checkstack ” perc=90140 ’ ’ ’ )141

142 Flow ( ’ a l l v e l f i l t s h o t ’ , ’ a l l s h o t s a l a s k a v e l f i l t . sh ’ ,143 ’ . / ${SOURCES[ 1 ] } 100 155 ’ )144

145 #Resu l t ( ’ m o v i e v e l f i l t ’ , ’ a l l v e l f i l t s h o t ’ ,146 # ’ ’ ’147 # suxmovie n2=96 loop=1 fframe=1 dframe=1148 # t i t l e =”record %g” perc=20149 # ’ ’ ’ )150

151 Result ( ’ v e l f i l t r e c 2 4 ’ , ’ a l l v e l f i l t s h o t ’ ,152 ’ ’ ’153 suwind key=t r a c l min=2324 max=2424 tmin=0 tmax=6 |154 suximage perc=90155 ’ ’ ’ )156

157 # sergey did t h i s , but i t d id not work . I j u s t used a s t r i n g158 #tnmo=[0000 , .339 , .607 , .707 , .872 ,1 .125 ,159 # 1.358 ,1 . 652 ,2 . 091 ,2 . 349 ,2 . 859 ,6 . 000 ]160 #vnmo=[7752 ,8074 ,7964 ,7781 ,8124 ,08754 ,161 # 08931 ,09721 ,10425 ,10742 ,12518 ,16552]162 #163 #tnmo = ’ , ’ . j o i n (map( s t r , tnmo ))164 #vnmo = ’ , ’ . j o i n (map( s t r , vnmo))165

166 tnmo=’ 0 0 0 0 , . 3 3 9 , . 6 0 7 , . 7 0 7 , . 8 7 2 , 1 . 1 2 5 , 1 . 3 5 8 , 1 . 6 5 2 , 2 . 0 9 1 , 2 . 3 4 9 , 2 . 8 5 9 , 6 . 0 0 0 ’167 vnmo=’ 7752 ,8074 ,7964 ,7781 ,8124 ,8754 ,8931 ,9721 ,10425 ,10742 ,12518 ,16552 ’168 Flow ( ’ v e l f i l t c d p s m u t e ’ , ’ a l l v e l f i l t s h o t ’ ,169 ’ ’ ’170 s u so r t cdp |

14

171 sunmo172 tnmo=%s173 vnmo=%s |174 sumute \175 xmute=−5225 ,−2970 ,−55 ,55 ,2970 ,5225 \176 tmute = . 7 0 0 , . 2 8 0 , . 1 0 0 , . 1 0 0 , . 2 8 0 , . 7 0 0 \177 mode=0 \178 key=o f f s e t |179 sunmo180 tnmo=%s181 vnmo=%s182 i n v e r t=1183 ’ ’ ’ % (tnmo , vnmo , tnmo , vnmo) )184

185 #Resu l t ( ’ movie cdps ’ , ’ v e l f i l t c d p smu t e ’ ,186 # ’ ’ ’187 # suxmovie n2=96 loop=1 fframe=1 dframe=1188 # t i t l e =”record %g” perc=20189 # ’ ’ ’ )190

191 Result ( ’ cdp250−251 ’ , ’ v e l f i l t c d p s m u t e ’ ,192 ’ ’ ’193 suwind key=cdp min=250 max=251 tmin=0 tmax=3 |194 suwigb perc=90195 ’ ’ ’ )196

197 Flow ( ’ v e l f i l t cdp snmo ’ , [ ’ v e l f i l t c d p s m u t e ’ , ’ vpick . txt ’ ] ,198 ’ ’ ’199 suchw200 key1=t s t a t201 key2=o f f s e t202 a=−200203 b=.0861204 | s u s t a t i c205 hdrs=1206 s i gn=1207 | supe f208 maxlag=.2209 pno i se =.01210 mincor =.25211 maxcor=5212 | s u s t a t i c213 hdrs=1214 s i gn=−1215 | sunmo

15

216 par=vpick . txt217 | suga in218 agc=1219 wagc=.1220 | sumute221 xmute=55 ,2970 ,5225222 tmute =.100 , . 280 , . 700223 mode=0224 key=o f f s e t225 ’ ’ ’ )226

227 #Resu l t ( ’ mov i e v e l f i l t c dp snmo ’ , ’ v e l f i l t c d p snmo ’ ,228 # ’ ’ ’229 # suxmovie n2=96 loop=1 fframe=1 dframe=1230 # t i t l e =”v e l f i l t c d p snmo record %g” perc=20231 # ’ ’ ’ )232

233 Flow ( ’ brute s tack ’ , ’ v e l f i l t cdp snmo ’ ,234 ’ ’ ’235 sus tack236 key=cdp237 normpow=1238 ’ ’ ’ )239

240 Result ( ’ b rute s tack ’ , ’ b rute s tack ’ ,241 ’ ’ ’242 suximage t i t l e =”brute s tack ” perc=90243 ’ ’ ’ )244

245 Result ( ’ zoombrutestack ’ , ’ b rute s tack ’ ,246 ’ ’ ’247 suwind tmin=0 tmax=3.0 |248 suximage t i t l e =”brute s tack ” perc=90249 ’ ’ ’ )250

251 # do inve r s e nmo, 1 func t i on nmo and compare r e s i d u a l s t a t i c s .compare

252 # ve l a n a l y s i s a f t e r r e s i d u a l s t a t i c s .253

254 Flow ( ’ s t a t i c s i n ’ , ’ v e l f i l t cdp snmo ’ ,255 ’ ’ ’256 suwind tmin=1.0 tmax=3.0 |257 suresamp r f =2258 ’ ’ ’ )259

16

260 Flow ( ’ s o u r c e s t a t i c . dat r e c s t a t i c . dat ’ ,261 ’ s t a t i c s i n ’ ,262 ’ ’ ’263 s u r e s s t a t \264 s s o l=s o u r c e s t a t i c . dat265 r s o l=r e c s t a t i c . dat266 n i t e r=5267 nt rac e s =5280268 ntp ick =100 samples269 nshot=167270 nr=668271 nc=636272 s f o l d =96273 r f o l d =96274 c f o l d =48275 verbose=1276 ’ ’ ’ , s tdout =0)277

278 Result ( ’ s o u r c e s t a t i c ’ , ’ s o u r c e s t a t i c . dat ’ ,279 ’ xwigb n1=167 t i t l e =”source s t a t i c ” ’ )280 Result ( ’ r e c s t a t i c ’ , ’ r e c s t a t i c . dat ’ ,281 ’ xwigb n1=668 t i t l e =”r e c e i v e r s t a t i c ” ’ )282

283 Flow ( ’ r s t a ck ’ ,284 ’ v e l f i l t cdp snmo s o u r c e s t a t i c . dat r e c s t a t i c . dat ’ ,285 ’ ’ ’286 sushw key=t s t a t a=0287 | sushw key=s s t a t a=0288 | sushw key=gs ta t a=0289 | s u s t a t i c290 hdrs=3291 s o u f i l e=${SOURCES[ 1 ] }292 r e c f i l e=${SOURCES[ 2 ] }293 ns=167294 nr=668295 no=96296 | sus tack297 key=cdp298 normpow=1299 ’ ’ ’ )300

301 Result ( ’ r s t a ck ’ , ’ r s t a ck ’ ,302 ’ suximage t i t l e =”r s ta ck ” perc=90 ’ )303

304 Result ( ’ zoomrstack ’ , ’ r s t a ck ’ ,

17

305 ’ ’ ’306 suwind tmin=0 tmax=3.0 |307 suximage t i t l e =”r s t a ck ” perc=90308 ’ ’ ’ )309

310 Flow ( ’ s t k v e l 1 . intp2 . bin ’ , ’ 010 Ve loc i ty In te rp IVA . sh ’ ,311 ’ ’ ’312 . /010 Ve loc i ty In te rp IVA . sh313 ’ ’ ’ )314

315 Result ( ’ v f i l e ’ , ’ s t k v e l 1 . intp2 . bin ’ ,316 ’ ’ ’317 ximage n1=3000 d1=.002 f2 =101 d2=1 cmap=rdb11318 ’ ’ ’ )319

320 Flow ( ’ mig ’ , [ ’ r s t a ck ’ , ’ s t k v e l 1 . intp2 . bin ’ ] ,321 ’ ’ ’322 s u f r a c power=.5 s i gn=1 |323 suktmig2d v f i l e=s t k v e l 1 . intp2 . bin dx=55 h o f f s e t=0324 ’ ’ ’ )325

326 Result ( ’ mig ’ , ’ mig ’ ,327 ’ ’ ’328 suximage t i t l e =”mig” perc=90329 ’ ’ ’ )330 Result ( ’ zoommig ’ , ’ mig ’ ,331 ’ ’ ’332 suwind tmin=0 tmax=3.0 |333 suximage t i t l e =”mig” perc=90334 ’ ’ ’ )335

336 # i t l o o k s l i k e one o f f s e t i s used f o r the whole o f f s e t ga ther337 # documentation says use su f rac b e f o r e migrat ion338 Flow ( ’ premig ’ , [ ’ v e l f i l t cdp snmo ’ , ’ vpick . txt ’ , ’ s t k v e l 1 . intp2 . bin ’ ] ,339 ’ . / mig . sh ’ )340

341 Flow ( ’ premigstack ’ , ’ premig ’ ,342 ’ ’ ’343 s u so r t |344 sus tack345 key=cdp346 normpow=1347 ’ ’ ’ )348

349 Result ( ’ premigstack ’ , ’ premigstack ’ ,

18

350 ’ ’ ’351 s u f i l t e r f =4 ,8 ,200 ,240 |352 suximage t i t l e =”premigstack ” perc=90353 ’ ’ ’ )354 Result ( ’ zoompremigstack ’ , ’ premigstack ’ ,355 ’ ’ ’356 s u f i l t e r f =4 ,8 ,200 ,240 |357 suwind tmin=0 tmax=3.0 |358 suximage t i t l e =”premigstack ” perc=90359 ’ ’ ’ )360

361 Plot ( ’ premigmovie ’ , ’ premig ’ ,362 ’ ’ ’363 suxmovie n2=536 loop=1 f f rame=1 dframe=1364 t i t l e =”record %g” perc=20365 ’ ’ ’ )366

367 End ( )