variable receiver interval in vintage data

Oil and natural gas exploration -- geology and geophysics
Post Reply
Silver Member
Silver Member
Posts: 19
Joined: Thu Jun 09, 2016 8:17 am
Location: cyberspace

variable receiver interval in vintage data

Post by Tesseract »

I was asked to review some 1970's data, to decide if it would be worth reprocessing. One thing that stuck out was that the receiver interval was not constant. It was 100 m on the near traces, 50 m on the far traces and 75 m in the middle of the streamer. Now the coarse sampling on the near traces seemed rather odd to me. You are only getting 3-fold on the shallow section. Then it occurred to me that they didn't have demultiple technologies back in olden times, and biasing the offset distribution to the far offsets was an attempt to discriminate against multiples. Could there be any other reason for doing this?

VIP Member
VIP Member
Posts: 652
Joined: Sat Mar 24, 2012 11:35 pm

Re: variable receiver interval in vintage data

Post by GuyM »

I'd guess in the 1970's they were limited on channel count - maybe 48 or even 24 channels?

One reason for a tighter spacing at the end of the streamer might be to try and avoid spatial aliasing on the steeper dip tails of the reflection hyperbolae.
Digital signal processing was being applied in the 1970s, and potentially this was the issue.

While the last big dataset I processed from the 1970's had 24 channels x 100m regular spacing (it was '73 and '74 vintage) people were trying all sorts of things.

We got pretty good results using the same shot- and receiver- domain interpolation routines that we routinely use for 2DSRME in modern processing; we lifted the data from 24 fold (50m pop increment) up to 96 fold in the CDP domain, so that you could apply the full modern set of demultiple and imaging routines.

It polished up *very* nicely, although from memory we still had to use an inner trace mute in a few places to nail the residual multiple.

Post Reply
  • Similar Topics
    Last post