Welcome to Petroleum Geology Forums

This is a free online community that aims to bring petroleum professionals and geologists together and share valuable knowledge. Registration is easy so become a member now for instant free access.
  • Petroleum Geologists can stay up to date with industry related topics and exchange ideas and concepts.
  • Upstream Oil and Gas Consultants get a chance to share their expertise and gain exposure to land future projects.
  • Geology students and graduates can join the discussion and get into contact with potential future employees.

  >> Register Now

Post new topic Reply to topic  [ 2 posts ] 
Author Message

Joined: Mon Dec 02, 2013 2:57 pm
Posts: 3
Hi guys,

complex structural sitiuation and sparse data.

Only one well in a relativ big survey, no sonic and density data.
A correct horizon identification over the great distance with several facies variations isn't possible.

Therefore i want to calibrate the "soft" velocity data (from the stacking process) with the "hard" data (from the well).

My question:
how is the best way to calibrate checkshot data with interval velocities?

My current processing stage:

1. Vrms velocity function gridded into a Vrms velocity cube
2. Vrms velocity cube converted into Vint velocity cube

Thank you very much in advance for your answer.

Mon Dec 02, 2013 3:44 pm

Joined: Wed Sep 01, 2010 10:06 pm
Posts: 107
I know how to do it in Petrel, so here's the workflow. In other software there might be other routes...

First sample the stacking velocity datapoints into a regular 3D grid. Use upscaling process to do this.

Then use the petrophysical modeling tools to interpolate the velocity values. Krigging with a 1D trend along Z might be the way to go.

You can then also sample the checkshot into this grid and co-krig these checkshot velocity to the interpolated stacking velocities. Use a high correlation factor of 0.95.

The checkshot velocity will now be interpolated away from the well usign the variations observed in the stacking velocity. You can then pass this velocity property directly into the velocity modeling in Petrel.

There are other routes, but this is probably the most robust and most appropriate.

Note that you will be gridding average velocities so you need to ake sure you have a 1D vertical trend applied through data analysis. Petrel velocity modeling can only take in average velocities and therefore it's best to work with these velocities and to not use interval velocities (although you can convert them).

Let us know how you get on!

Wed Jan 01, 2014 3:34 pm
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 2 posts ] 

Search for:
Jump to:  

Content on EPGeology.com is intended for personal use only and to supplement, not replace, professional judgment. EPGeology.com disclaims any and all liability for your use of its content. As most of our content is supplied by our users we can not check copyright, and stress that copyright remains at the original owner. If you suspect copyright infringement please use the contact form to report it.
Contact || © EPGeology.com. || Powered by phpBB Asteroid Mining