Incorporating Different Data Types in Velocity Modelling

Pranav Audhkhasi
Indian School of Mines, Dhanbad

Description:

Velocity modelling is an essential and challenging aspect of seismic interpretation. Better and more accurate depth imaging entirely relies on the accuracy of the velocity model. Based on their accuracy, data types can be categorised into different categories. Most accurate data or hard data are the well logs available. Semi-hard data include the VSP/Checkshot data and soft data is the processed (migrated) seismic data available. Incorporating the data by deriving the velocity information from them and calibrating them logically can yield better velocity models.

Application:

Identified picks from the well logs and their corresponding interpreted horizons from the seismic data are used to compute pseudo interval velocities. The one-to-one correspondence between the depth picks from well logs and time markers from seismic sections can be established by implementing seismic to well tie. Interval velocity plots from VSP/Checkshot data are directly used to derive the interval velocity plots which will be used. The RMS velocities derived from time migrated seismic sections are Dix inverted to give interval velocities which are used in the model. Keeping the accuracy of the data, the velocities obtained from the seismic will be calibrated with respect to those from VSP/Checkshot which will be further calibrated with pseudo interval velocities from the well logs. Interpreted time horizons from the seismic can also be used for additional structural control. Different options for model interpolation are available like Delaunay triangulation, Inverse distance and Normalized inverse distance, of which the latter has been used.

Results and Conclusions:

By incorporating different data types and calibrating them logically, more accurate velocity models can be derived and better depth imaging is possible with no or least lateral and vertical mispositioning, which are the two most common time to depth conversion errors.

Significance:

Since PSDM is a relatively expensive and time consuming procedure, although more accurate than PSTM, majority of the oil industry relies on the latter. This simple technique of velocity modelling coupled with any depth conversion scheme, the simplest being the vertical stretch technique, can serve as an alternate to PSDM. Hence it can be stated as poor man’s PSDM.

AAPG Search and Discovery Article #90182©2013 AAPG/SEG Student Expo, Houston, Texas, September 16-17, 2013