next up previous print clean
Next: Interval Velocity Estimation Up: How to improve a Previous: How to improve a

Problems involved

The introduction of well data in the velocity model demands some extra care because seismic data provide a relatively high horizontal resolution while well data provide a very narrow vertical resolution but a poor horizontal resolution. Well velocities are direct rock properties, estimated directly from the earth, while seismic velocities are indirect rock properties estimated from traveltimes. If the combination of seismic velocities and well velocities were possible, migration results would be favorably improved.

The problem is how to manage both velocities to get better seismic images. Well velocities may be interpolated to obtain a horizontal velocity model. This velocity model can be correlated with the velocity model from seismic data, resulting in a new velocity model from both kinds of information. The interpolation problem, may be solved using geostatistics. As a conclusive example, Lee and Xu (2000) show how geostatistics helped to improve the imaging of a geopressure zone in the Gulf of Mexico. Other interpolation techniques Fomel (1997) can be implemented in order to obtain a satisfactory $\it well$ $\it 
velocity$ $\it model$.Steering filters prove effective in interpolating velocities obtaining a horizontal velocity model from well data Clapp et al. (1997, 1998); Clapp and Biondi (1998), but it will be necessary to regularize well velocities in order to allow a satisfying the correlation between the velocity model from seismic and from well to produce good results. In the next section, I am going to discuss a method to obtain interval velocities from RMS velocities.


next up previous print clean
Next: Interval Velocity Estimation Up: How to improve a Previous: How to improve a
Stanford Exploration Project
4/27/2000