System and method for lateral statistical estimation of rock and fluid properties in a subsurface formation

1. A computer-implemented method of estimating reservoir attributes as a function of position in a subsurface volume of interest, the method being implemented in a computer system that includes one or more physical computer processors, non-transient electronic storage, and a graphical user interface, comprising:

a. obtaining, from the non-transient electronic storage, a geophysical dataset and a geological dataset representative of a subsurface volume of interest;

b. obtaining, from the non-transient electronic storage, a parameter model, the parameter model having been trained using training data, wherein the geological dataset includes well data of one or more wells in the subsurface volume of interest and the training data includes the well data;

c. picking, with the one or more physical computer processors, at least one surface in the geophysical dataset;

d. assigning stratigraphic meaning to the at least one surface based on the geological dataset;

e. identifying at least one region of interest on the at least one surface;

f. generating, with the one or more physical computer processors, statistical seismic attributes for the at least one region;

g. identifying, with the one or more physical computer processors, statistically significant critical attributes among the statistical seismic attributes by applying the parameter model to generate response variable maps for the at least one region;

h. generating, with the one or more physical computer processors, uncertainty maps of the at least one region for each of the critical attributes and uncertainty maps for the response variables; and

i. displaying, via the graphical user interface, the response variable maps and the uncertainty maps.

2. The method of claim 1 further comprising using the response variable maps and the uncertainty maps to manage hydrocarbon production from the subsurface volume of interest.

3. The computer-implemented method of claim 1 , wherein the parameter model comprises random forest machine learning.

4. The method of claim 1 further comprising:

a. correlating, with the one or more physical computer processors, each of the critical attributes and response variables to generate a correlation plot;

b. displaying, via the graphical user interface, the correlation plot;

c. generating, with the one or more physical computer processors, an impact ranking of the critical attributes to the response variables based on the correlation plot to generate a ranking seriatim; and

d. displaying, via the graphical user interface, the ranking seriatim.

5. A computer system, comprising:

one or more physical computer processors;

memory; and

one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions that when executed by the one or more processors cause the system to:

a. obtain, from non-transient electronic storage, a geophysical dataset and a geological data set representative of a subsurface volume of interest;

b. obtain, from the non-transient electronic storage, a parameter model, the parameter model having been trained using training data, wherein the geological dataset includes well data of one or more wells in the subsurface volume of interest and the training data includes the well data;

c. pick, with the one or more physical computer processors, at least one surface in the geophysical dataset;

d. assign stratigraphic meaning to the at least one surface based on the geological dataset;

e. identify at least one region of interest on the at least one surface;

f. generate, with the one or more physical computer processors, statistical seismic attributes for the at least one region;

g. identify, with the one or more physical computer processors, statistically significant critical attributes among the statistical seismic attributes by applying the parameter model to generate response variable maps for the at least one region;

h. generate, with the one or more physical computer processors, uncertainty maps of the at least one region for each of the critical attributes and uncertainty maps for the response variables; and

i. display, via the graphical user interface, the response variable maps and the uncertainty maps.

6. A non-transitory computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by an electronic device with one or more processors and memory, cause the device to:

a. obtain, from non-transient electronic storage, a geophysical dataset and a geological dataset representative of a subsurface volume of interest;

b. obtain, from the non-transient electronic storage, a parameter model, the parameter model having been trained using training data, wherein the geological dataset includes well data of one or more wells in the subsurface volume of interest and the training data includes the well data;

c. pick, with the one or more physical computer processors, at least one surface in the geophysical dataset;

d. assign stratigraphic meaning to the at least one surface based on the geological dataset;

e. identify at least one region of interest on the at least one surface;

f. generate, with the one or more physical computer processors, statistical seismic attributes for the at least one region;

g. identify, with the one or more physical computer processors, statistically significant critical attributes among the statistical seismic attributes by applying the parameter model to generate response variable maps for the at least one region;

h. generate, with the one or more physical computer processors, uncertainty maps of the at least one region for each of the critical attributes and uncertainty maps for the response variables; and

i. display, via the graphical user interface, the response variable maps and the uncertainty maps.