Generating an earth model from spatial correlations of equivalent earth models

1. A method of modeling a subterranean formation, the method comprising:

generating a plurality of earth models of the subterranean formation based on at least one measurement of the subterranean formation at a borehole;

projecting at least one second earth model upon a well path of the borehole;

selecting a model of the plurality of earth models based on a spatial correlation between at least one parameter of the selected model and a corresponding at least one parameter of the projected at least one second earth model, wherein the projecting the at least one second earth model upon the well path of the borehole comprises performing a discrete cross-correlation between the at least one parameter of the selected model and the corresponding at least one parameter of the projected at least one second earth model over a portion of the well path of the borehole; and

geosteering based on the selected model based on the spatial correlation between the at least one parameter of the selected model and the corresponding at least one parameter of the projected at least one second earth model.

2. The method of claim 1 , wherein:

a misfit of each model of the plurality of earth models is less than a particular threshold value, the misfit based on a difference between the at least one measurement of the subterranean formation and a predicted measurement based on the corresponding model; and

the selection based on the spatial correlation improves a likelihood that the selected model is more geologically accurate than at least one other of the plurality of earth models, relative to selecting based on the misfit alone.

3. The method of claim 2 , wherein the selected threshold is based on a noise level corresponding to the at least one measurement.

4. The method of claim 1 , wherein the at least one measurement comprises a resistivity measurement.

5. The method of claim 4 , wherein generating the plurality of earth models comprises performing a plurality of resistivity inversions based on the at least one measurement, the plurality of resistivity inversions comprising at least a stochastic resistivity inversion or a deterministic resistivity inversion.

6. The method of claim 1 , wherein projecting the at least one second earth model comprises spatially extrapolating or interpolating the corresponding at least one parameter from at least one second well path to the well path of the borehole.

7. The method of claim 6 , wherein the spatial extrapolation or interpolation is based on at least a spatial function or a spatially statistical function.

8. The method of claim 1 , wherein the at least one second earth model corresponds to at least one well adjacent to the borehole, or comprises a geocellular earth model.

9. The method of claim 1 , wherein, among the plurality of earth models, the selected model has a strongest degree of spatial correlation with the projected at least one second earth model.

10. The method of claim 1 , wherein the selected model has a degree of spatial correlation that is above a particular threshold.

11. The method of claim 1 , wherein selecting the model comprises performing at least a cross-correlation, a convolution, or a coherence between the at least one parameter of the selected model and the corresponding at least one parameter of the projected at least one second earth model.

12. The method of claim 1 , wherein the portion of the well path is based on a sensitivity of a source of the at least one measurement.

13. A logging system comprising:

an electromagnetic logging tool that collects at least one measurement of a subterranean formation at a borehole; and

a processing system that:

generates a plurality of earth models of the subterranean formation based on the at least one measurement of the subterranean formation;

projects at least one second earth model upon a well path of the borehole;

selects a model of the plurality of earth models based on a spatial correlation between at least one parameter of the selected model and a corresponding at least one parameter of the projected at least one second earth model, wherein the processing system projects the at least one second earth model upon the well path of the borehole by performing a discrete cross-correlation between the at least one parameter of the selected model and the corresponding at least one parameter of the projected at least one second earth model over a portion of the well path of the borehole; and

geosteers based on the selected model based on the spatial correlation between the at least one parameter of the selected model and the corresponding at least one parameter of the projected at least one second earth model.

14. The logging system of claim 13 , wherein:

a misfit of each model of the plurality of earth models is less than a particular threshold value, the misfit based on a difference between the at least one measurement of the subterranean formation and a predicted measurement based on the corresponding model; and

the selection based on the spatial correlation improves a likelihood that the selected model is more geologically accurate than at least one other of the plurality of earth models, relative to selecting based on the misfit alone.

15. The logging system of claim 13 , wherein:

the processing system projects the at least one second earth model by spatially extrapolating or interpolating the corresponding at least one parameter from at least one second well path to the well path of the borehole; and

the spatial extrapolation or interpolation is based on at least a spatial function or a spatially statistical function.

16. The logging system of claim 13 , wherein the at least one second earth model corresponds to at least one well adjacent to the borehole, or comprises a geocellular earth model.

17. The logging system of claim 13 , wherein, among the plurality of earth models, the selected model has a strongest degree of spatial correlation with the projected at least one second earth model.

18. The logging system of claim 13 , wherein the processing system selects the model by performing at least a cross-correlation, a convolution, or a coherence between the at least one parameter of the selected model and the corresponding at least one parameter of the projected at least one second earth model.