Anisotropic PSTM and seismic well ties

[jefry123]

Hi,

As I realized, in Pre-Stack Depth migration because of the interval velocity obtained, its more accurate algorithm and also depth section as the output, there is a more correlation between seismic section and well data and misties are less that PSTM. In anisotropic PSDM, including the Thomson parameters results in even better images according to well ties.

In anisotropic PSTM a parameter is included in calculations which accounts for far-offset hockey stick and consequently a higher fold stack is obtained. So, I think it could only affects the resolution of the image and not the depth of reflectors.

I was wondering if anisotropic PSTM could also yield an image with a better correlation to well markers' depths than PSTM?

[GuyM]

It depends....

There's a number of reasons you could have a seismic-well mistie:

- the time/depth conversion in the well is inaccurate
You need both sonic logs and checkshots over the depth range to be tied; checkshots give the coarse-but-accurate conversion, and the sonic log once calibrated for drift (due to ultra-sound frequencies being deployed and the sensor length) gives the deail

- the well is in the wrong place(!)
Most common issues are vertical (KB or DF height and so on) or incorrect conversions from Lat/Long to East/North (that is to say the wrong geodetic datum is used or assumed in the conversion)

- the seismic dips are not correctly imaged
This is down to the imaging algorithm and model; that is to say how well it handles refraction at layer boundaries, refraction within a layer and whether sufficient detail is included within the model (and correct) for the image to be in the correct lateral position

So - dealing with the latter case and your specific question, in theory preSDM will take into account all refraction (and hence be a true "vertical ray zero offset section) however that depends on the accuracy of the model, and the sophistication of the algorithm in terms of travel-time calculations and image formation.

PreSTM doesn't usually take into account refraction, but some of the anisotropic forms of preSTM do allow for curved-rays that address someof the refraction issues, as well as (as you say) correcting for the hockey stick.

Anisotropic preSDM should theortically be matching in depth to the well, since it is the difference between the vertical and horizontal velocities within each rock layer that require us to calibrate seismically-derived velocities (or depths) to wells as a separate phase. The variation between seismically-derived depths and well depths without this anisotropic correction can be as much as 20%.