System for utilizing seismic data to estimate subsurface lithology

1. A method of geophysical exploration of a subsurface region of interest, comprising:

utilizing an unsupervised learning network to organize seismic data representing a subsurface region of interest;

correlating a portion of said organized seismic data with lithological data from a well bore located in said subsurface region of interest; and

applying said correlation to said seismic data to estimate lithology in said subsurface region of interest.

2. The method of claim 1 wherein said unsupervised learning network is a self organizing feature map.

3. The method of claim 1 wherein said unsupervised learning network is a Kohonen network.

4. A method of geophysical exploration of a subsurface region of interest, comprising:

applying a plurality of seismic data attributes for measurement location from a seismic data set from a subsurface region of interest to a Kohonen network to organize said seismic data set into a plurality of seismic Kohonen classes;

selecting a subset of said organized seismic data set representative of the earth's subsurface in the vicinity of a well bore penetrating said subsurface region of interest;

correlating Kohonen classes of said subset of said organized seismic data set with classes of lithological data from said well bore to generate a correlation between Kohonen classes and lithological classes; and

applying said correlation to said seismic data set to estimate lithology of said measurement locations.

5. The method of claim 4 wherein said seismic data attributes comprise semblance, amplitude-versus-offset and attenuation.

6. The method of claim 4 wherein said lithological data comprise volume shale and acoustic impedance.

7. A method of geophysical exploration of a subsurface region of interest, comprising:

applying a plurality of lithology values for measurement location from a well bore penetrating a subsurface region of interest to a Kohonen neural network to organize said lithology values into a plurality of lithology Kohonen classes;

utilizing said lithology Kohonen classes to establish ranges of a lithology value;

applying a plurality of seismic data attributes for measurement location from a seismic data set from said subsurface region of interest to a Kohonen network to organize said seismic data set into a plurality of seismic Kohonen classes;

selecting a subset of said organized seismic data set representative of the earth's surface in the vicinity of said well bore penetrating said subsurface region of interest;

correlating Kohonen classes of said subset of said organized seismic data set with classes of lithological data from said well bore to generate a correlation between Kohonen classes and lithological classes, wherein said ranges of a lithology value are utilized in establishing boundaries of said lithology classes; and

applying said correlation to said seismic data set to estimate lithology of said measurement locations from said subsurface region of interest.

8. The method of claim 7 wherein said lithology values are volume shale and acoustic impedance.

9. The method of claim 7 wherein said seismic data attributes comprise semblance, amplitude-versus-offset and attenuation.

10. A device which is readable by a digital computer having instructions defining the following process and instructions to the computer to perform said process:

utilizing an unsupervised learning network to organize seismic data representing a subsurface region of interest;

correlating a portion of said organized seismic data with lithological data from a well bore located in said subsurface region of interest; and

applying said correlation to said seismic data to estimate lithology in said subsurface region of interest.

11. A device which is readable by a digital computer having instructions defining the following process and instructions to the computer to perform said process:

applying a plurality of seismic data attributes for measurement location from a seismic data set from a subsurface region of interest to a Kohonen network to organize said seismic data set into a plurality of seismic Kohonen classes;

selecting a subset of said organized seismic data set representative of the earth's subsurface in the vicinity of a well bore penetrating said subsurface region of interest;

correlating Kohonen classes of said subset of said organized seismic data set with classes of lithological data from said well bore to generate a correlation between Kohonen classes and lithological classes; and

applying said correlation to said seismic data set to estimate lithology of said measurement locations.

12. A device which is readable by a digital computer having instructions defining the following process and instructions to the computer to perform said process:

applying a plurality of lithology values for measurement location from a well bore penetrating a subsurface region of interest to a Kohonen neural network to organize said lithology values into a plurality of lithology Kohonen classes;

utilizing said lithology Kohonen classes to establish ranges of a lithology value;

applying a plurality of seismic data attributes for measurement location from a seismic data set from said subsurface region of interest to a Kohonen network to organize said seismic data set into a plurality of seismic Kohonen classes;

selecting a subset of said organized seismic data set representative of the earth's surface in the vicinity of said well bore penetrating said subsurface region of interest;

correlating Kohonen classes of said subset of said organized seismic data set with classes of lithological data from said well bore to generate a correlation between Kohonen classes and lithological classes, wherein said ranges of a lithology value are utilized in establishing boundaries of said lithology classes; and

applying said correlation to said seismic data set to estimate lithology of said measurement locations from said subsurface region of interest.