EdgeTech's SEG-Y to SIOSEIS's SEG-Y
One of the most important differences in the X-Star's SEG-Y is that the data are "raw matched filtered" or "evelope data" and are complex numbers rather than real numbers. SIOSEIS process GAINS parameter TYPE 7 converts complex numbers to real numbers. Using process prout to dump SEG-Y trace header words 1, 3, 4, 5, 6, 7 shows several other differences: SIOSEIS ver 99.6 (5 September 1999) (C) Regents of U.C. procs diskin prout end diskin ipath tape_6.raw.segy end *** WARNING *** RP is zero and RP trace number is non-zero. DISKIN thinks this file is sorted by RP. Use DISKIN parameters NO and TR or SORT to read properly. Use process HEADER to set the rp trace number to 0. end prout indices l1 l3 l4 l5 l6 l7 fno 0 lno 99999 ftr 0 ltr 999 end end end **** 0 ERRORS IN THIS JOB **** 12308. 0.0000 0.0000 0.0000 0.0000 1.0000 12308. 0.0000 1.0000 0.0000 0.0000 1.0000 12309. 0.0000 0.0000 0.0000 0.0000 1.0000 12309. 0.0000 1.0000 0.0000 0.0000 1.0000 12310. 0.0000 0.0000 0.0000 0.0000 1.0000 12310. 0.0000 1.0000 0.0000 0.0000 1.0000 12311. 0.0000 0.0000 0.0000 0.0000 1.0000 12311. 0.0000 1.0000 0.0000 0.0000 1.0000 12312. 0.0000 1.0000 0.0000 0.0000 1.0000 12312. 0.0000 0.0000 0.0000 0.0000 1.0000 12313. 0.0000 1.0000 0.0000 0.0000 1.0000 12313. 0.0000 0.0000 0.0000 0.0000 1.0000 12314. 0.0000 0.0000 0.0000 0.0000 1.0000 12314. 0.0000 1.0000 0.0000 0.0000 1.0000 12315. 0.0000 1.0000 0.0000 0.0000 1.0000 12315. 0.0000 0.0000 0.0000 0.0000 1.0000 l1 l3 l4 l5 l6 l7 rec no shot no shot tr energy pt CDP no CDP tr The EdgeTech documentation for "X-STAR FULL SPECTRUM SUB-BOTTOM PROFILER", page 4-10 defines their use of the SEG-Y trace header as: Integer No Description 0-1 Trace (record) number (integer - 32 bit) 14 Trace identification . . 93 NMEA hour 94 NMEA minutes . . 100-101 Trace scale factor (floating point) Note that there is no millisecond. There are several models of EdgeTechs, this one apparently has two hydrophone outputs, one a single phone and the other beamformed from four hydrophones. Use the single phone for very shallow horizons. Notice that SEG-Y word four seems to be a trace number within the shot, though the trace order seeme to be reversed occasionaly. SIOSEIS wants the shot number to be in SEG-Y word 4. SIOSEIS also expects the traces to be numbered 1, 2 rather than 0, 1. Another problem is that SIOSEIS uses word 7, the CDP trace number to indicate that the data are sorted by CDP or RP, so EdgeTech's 1 in word 7 will be a problem. Another "gotcha" is contained in the SEG-Y readable header that says "Envelope Sample interval = A/D/ Sample Interval*2". The SEG-Y data is envelope data and not A/D data, so the SEG-Y sample interval is wrong for the data in the file! lsd tape_6.raw.segy 1 11 SHOT TR RP TR ID RANGE DELAY NSAMPS SI YR DAY HR MIN SEC 0 0 0 1 1 0 0 3976 20 1999 238 22 18 5 0 1 0 1 1 0 0 3976 20 1999 238 22 18 7 0 0 0 1 1 0 0 3976 20 1999 238 22 18 7 0 1 0 1 1 0 0 3976 20 1999 238 22 18 7 0 0 0 1 1 0 0 3976 20 1999 238 22 18 7 0 1 0 1 1 0 0 3976 20 1999 238 22 18 7 0 0 0 1 1 0 0 3976 20 1999 238 22 18 7 0 1 0 1 1 0 0 3976 20 1999 238 22 18 7 0 1 0 1 1 0 0 3976 20 1999 238 22 18 7 0 0 0 1 1 0 0 3976 20 1999 238 22 18 7 0 1 0 1 1 0 0 3976 20 1999 238 22 18 7 EdgeTech has used SEG-Y unused bytes 201-204 (EdgeTech 100-101) to store the "Trace scale factor". SIOSEIS uses the same bytes (SIOSEIS word 51) as an "end-of-gather" flag. SIOSEIS process HEADER parameters: header fno 0 lno 999999 ftr 0 ltr 99999 l3 = l1 l4 = l4 + 1 l7 = 0 i59 = i59 * 2 r49 = r49 * 2 r55 = r51 r51 = 0 end end will changed the EdgeTech SEG-Y to SIOSEIS SEG-Y. l3 = l1 copies the EdgeTech record number to the shot number. l4 = l4 + 1 increments the shot trace number so it is 1 or 2 rather than 0 or 1. l6 = 0 sets the CDP trace number to 0. i59 = i59 * 2 corrects the SEG-Y sample interval r49 = r49 * 2 corrects sioseis's floating point sample interval diskin calculated. r55 = r51 copies the EdgeTech trace scale factor to word 55. r51 = 0 sets SIOSEIS's end-of-gather flag to zero. SIOSEIS process GAINS converts complex traces to real traces: gains type 7 end end The EdgeTech trace scalar MUST be applied since it can change on every trace. The new SIOSEIS process WEIGHT parameter HDR will multiply all trace amplitudes by the trace scale factor: weight fno 0 lno 999999 hdr 55 end end The script to read and convert EdgeTech files is: sioseis << eof procs diskin header gains weight diskoa prout end diskin ipath tape_6.raw.segy end end header fno 0 lno 99999 ftr 0 ltr 99999 l3 = l1 l4 = l4 + 1 l7 = 0 r55 = r51 r51 = 0 end end gains type 7 end end weight fno 0 lno 999999 hdr 55 end end diskoa opath data end end plot nibs 75 vscale 150 nsecs .1 tlines .01 .05 .1 srpath sunfil def .02 end end prout fno 0 lno 99999 ftr 0 ltr 999 end end end eofBACK Go to the list of seismic processes. Go to SIOSEIS introduction.