Geologic fault seal characterization

1 . A method comprising:

stochastically generating at least fifty realizations for a subsurface geologic environment by sampling distributions for one or more parameters that characterize the subsurface geologic environment, wherein the at least fifty realizations represent different results for an actual, physical characteristic of the subsurface geologic environment, wherein the one or more parameters comprise pressure, temperature, flowrate, composition, or any combination thereof associated with the subsurface geologic environment, and wherein each of the at least fifty realizations is generated by:

modeling petrophysical properties of the subsurface geologic environment based on the one or more parameters;

performing a juxtaposition analysis of the subsurface geologic environment based on the one or more parameters;

performing a fault clay content prediction for the subsurface geologic environment based on the one or more parameters;

estimating a fault capillary threshold pressure for the subsurface geologic environment based on the one or more parameters; and

computing a sealing capacity for the subsurface geologic environment based on the one or more parameters, wherein each of the at least fifty realizations is generated based on a parameter value for each of the one or more parameters for modeling the petrophysical properties, the performing the juxtaposition analysis, the performing the fault clay content prediction, the estimating the fault capillary threshold pressure, and the computing the sealing capacity, wherein the parameter value is different for at least two of:

the modeling of the petrophysical properties;

the juxtaposition analysis;

the fault clay content prediction;

the estimation of the fault capillary threshold pressure; or

the computation of the sealing capacity;

ranking the one or more parameters with respect to influence on the different results;

generating one or more result predictions using a trained machine learning model for variations in values of at least a portion of the one or more parameters having a ranking above a threshold, wherein the trained machine learning model is trained using at least a portion of the at least fifty realizations and their corresponding different results, wherein the at least a portion of the at least fifty realizations corresponds to the portion of the one or more parameters, and wherein the one or more result predictions comprise one or more indications of one or more physical objects being present in the subsurface geologic environment; and

controlling a drill bit forming a borehole within the subsurface geologic environment based on the one or more result predictions.

2 . The method of claim 1 , wherein ranking the one or more parameters comprises implementing a predictive model that determines an influence of each of the one or more parameters on the different results.

3 . The method of claim 2 , comprising implementing gradient boosting to generate the predictive model.

4 . The method of claim 3 , wherein the gradient boosting generates the predictive model from an ensemble of predictive models.

5 . The method of claim 1 , wherein the trained machine learning model comprises an input layer, one or more hidden layers, and an output layer.

6 . The method of claim 5 , wherein the trained machine learning model is a multilayer perceptron model.

7 . The method of claim 1 , wherein the actual, physical characteristic of the subsurface geologic environment at least in part controls accumulation of hydrocarbons in a trap of the subsurface geologic environment.

8 . The method of claim 7 , wherein the trap is formed in part by a fault seal, and wherein the accumulation of hydrocarbons in the trap depends at least in part on characteristics of the fault seal.

9 . The method of claim 1 , wherein the actual, physical characteristic of the subsurface geologic environment is a free water level in the subsurface geologic environment.

10 . The method of claim 9 , wherein the free water level is indicative of fault sealing ability of a fault in the subsurface geologic environment.

11 . The method of claim 1 , wherein the sampling comprises implementing a Latin hypercube sampling technique.

12 . The method of claim 1 , wherein the one or more parameters comprise parameters in a continuous parameter space.

13 . The method of claim 1 , wherein the one or more parameters comprise at least one parameter in a discrete parameter space.

14 . The method of claim 13 , wherein the at least one parameter in the discrete parameter space comprises an analysis technique parameter that is selectable from one or more a different analysis techniques.

15 . The method of claim 1 , comprising rendering a visualization of at least a portion of the subsurface geologic environment to a display, wherein the visualization comprises indicia of uncertainty in at least one of the one or more parameters.

16 . The method of claim 1 , wherein the one or more result predictions comprise one or more predictions for at least the actual, physical characteristic of the subsurface geologic environment.

17 . The method of claim 1 , comprising increasing the number of the at least fifty realizations based on one or more of an analysis of the ranking and an analysis of the one or more result predictions.

18 . The method of claim 1 , wherein the one or more physical objects of the one or more result predictions comprise a free water level, an elevation of a juxtaposition window, or both.

19 . A system comprising:

a processor;

a memory accessible by the processor; and

processor-executable instructions stored in the memory and executable to instruct the system to:

stochastically generate at least fifty realizations for a subsurface geologic environment by sampling distributions for one or more parameters that characterize the subsurface geologic environment, wherein the at least fifty realizations represent different results for an actual, physical characteristic of the subsurface geologic environment, wherein the one or more parameters comprise pressure, temperature, flowrate, composition, or any combination thereof associated with the subsurface geologic environment, and wherein each of the at least fifty realizations is generated by:

modeling petrophysical properties of the subsurface geologic environment based on the one or more parameters;

performing a juxtaposition analysis of the subsurface geologic environment based on the one or more parameters;

performing a fault clay content prediction for the subsurface geologic environment based on the one or more parameters;

estimating a fault capillary threshold pressure for the subsurface geologic environment based on the one or more parameters; and

computing a sealing capacity for the subsurface geologic environment based on the one or more parameters, wherein each of the at least fifty realizations is generated based on a parameter value for each of the one or more parameters for modeling the petrophysical properties, performing the juxtaposition analysis, performing the fault clay content prediction, estimating the fault capillary threshold pressure, and computing the sealing capacity, wherein the parameter value is different for at least two of:

the modeling the petrophysical properties;

the juxtaposition analysis;

the fault clay content prediction;

the estimation of the fault capillary threshold pressure; or

the computation of the sealing capacity;

rank the one or more parameters with respect to influence on the different results;

generate one or more result predictions using a trained machine learning model for variations in values of at least a portion of the one or more parameters having a ranking above a threshold, wherein the trained machine learning model is trained using at least a portion of the at least fifty realizations and their corresponding different results, wherein the at least a portion of the at least fifty realizations correspond to the portion of the one or more parameters, and wherein the one or more result predictions comprise one or more indications of one or more physical objects being present in the subsurface geologic environment; and

control a drill bit forming a borehole within the subsurface geologic environment based on the one or more result predictions.

20 . One or more non-transitory computer-readable storage media comprising computer-executable instructions executable to instruct a computing system to:

stochastically generate at least fifty realizations for a subsurface geologic environment by sampling distributions for one or more parameters that characterize the subsurface geologic environment, wherein the at least fifty realizations represent different results for an actual, physical characteristic of the subsurface geologic environment, wherein the one or more parameters comprise pressure, temperature, flowrate, composition, or any combination thereof associated with the subsurface geologic environment, and wherein each of the at least fifty realizations is generated by:

modeling petrophysical properties of the subsurface geologic environment based on the one or more parameters;

performing a juxtaposition analysis of the subsurface geologic environment based on the one or more parameters;

performing a fault clay content prediction for the subsurface geologic environment based on the one or more parameters;

estimating a fault capillary threshold pressure for the subsurface geologic environment based on the one or more parameters; and

computing a sealing capacity for the subsurface geologic environment based on the one or more parameters, wherein each of the at least fifty realizations is generated based on a parameter value for each of the one or more parameters for modeling the petrophysical properties, performing the juxtaposition analysis, performing the fault clay content prediction, estimating the fault capillary threshold pressure, and computing the sealing capacity, wherein the parameter value is different for at least two of:

the modeling the petrophysical properties;

the juxtaposition analysis;

the fault clay content prediction;

the estimation of the fault capillary threshold pressure; or

the computation of the sealing capacity;

rank the one or more parameters with respect to influence on the different results;

generate one or more result predictions using a trained machine learning model for variations in values of at least a portion of the one or more parameters having a ranking above a threshold, wherein the trained machine learning model is trained using at least a portion of the at least fifty realizations and their corresponding different results, wherein the at least a portion of the at least fifty realizations correspond to the portion of the one or more parameters, and wherein the one or more result predictions comprise indications of one or more physical objects being present in the subsurface geologic environment; and

control a drill bit forming a borehole within the subsurface geologic environment based on the one or more result predictions.