Thursday, 1 October 2015

'Can Envelope help us understand the subsurface' by David M Roberts (Guest writer)

In response to an earlier blog post we published about whether 'Envelope can help us understand the subsurface', this week we have the pleasure of introducing a guest writer. David M Roberts of 3-DMR kindly agreed to put his opinions and findings in a blog post for us to publish. We are currently investigating David's findings so look out for our reply soon. 

A comment on 'Can Envelope help us understand the subsurface' (Friday 31 July 2015 blog) by David M Roberts 3-DMR 25th Sep 2015

The rather dramatic change in the reflection geometry between the migrated data and the 'Envelope' attribute in the small part of the New Zealand 3D featured in the above referenced blog rather alarmed me (Fig 1). 

Figure 1: Migrated data (left) and GeoTeric 'Envelope' attribute volume (right)

The introduction of steeply dipping 'progrades' by 'Envelope' (Fig 1) looked rather suspicious as there is no indication of these steep deeps on the migrated data (Fig 1). The bright reflections in the centre of the section on the migrated data also exhibit rather strange cyclic amplitude variations at several levels, which could perhaps be caused by sub-seismic faulting .. or what?

So I had a good look at a larger cube of the migrated data using GeoTeric, GeoProbe and the DecisionSpace 3D viewer. The new data link between GeoTeric and DecisionSpace makes it very easy to work with all three applications, which are now closely integrated. I created a standard FD RGB blend volume and also an HDFD RGB blend using GeoTeric, and managed to replicate the dipping reflectors (shown coloured cyan) in the HDFD RGB blend volume (Fig 2).
Figure 2: GeoTeric HDFD RGB blend
The first thing I noticed was that the same steeply dipping reflectors were evident wherever the cyan coloured interval appeared, at several other places on the section (Fig 2). The steep dips sometimes occurred in the opposite direction. Mmmhhhh.

Figure 3: Migrated data using a black and white colour table

Examining the migrated dataset using a simple black and white colour table revealed that steeply dipping high frequency noise is evident in the data (Fig 3), and is particularly noticeable above and within the 'steep dip' area featured in the blog.

There is also sub-seismic faulting evident in the crestal part of the fault block, but these faults are actually quite well defined, and I don't think are directly related to the steep dips.

I have therefore concluded that the steep dips indicated by the 'Envelope' volume (Fig 1) are probably not real, and their origins almost certainly lie in the steeply dipping seismic noise which I assume has been dramatically enhanced by the 'Envelope' algorithm.

I have suggested that the GeoTeric team should have another look at this to try and understand exactly what has happened here, so that interpreters are aware of such potential 'interpretation pitfalls' in future.

I would like to thank GeoTeric staff for providing this opportunity to use their exciting software to examine these data. 

David M Roberts is a Consultant in Visualisation and Exploration and Development Geoscience at 3-DMR.

No comments:

Post a Comment