Processing subsurface data with uncertainty for modelling and field planning

A non-transitory, computer-readable medium storing instructions that, when executed by at least one processor of a computing system, cause the computing system to perform operations, the operations comprising:

• performing an oilfield operation using a set of sensors to collect input data from a subterranean volume;
• wherein the input data specifies a weight on bit parameter associated with the operation, a torque on bit parameter associated with the operation, a pressure parameter associated with the operation, a flow rate parameter associated with the operation, a rotary speed parameter associated with the operation, a well log, a porosity parameter associated with the operation, a fluid composition parameter associated with the operation, and a permeability parameter associated with the operation,
• wherein the set of sensors comprises a geophone and a gauge,
• wherein the subterranean volume comprises a shale layer, a carbonate layer, and a sand layer, and
• wherein the operation is a drilling operation, a wireline operation, or a seismic survey operation;
• receiving the input data representing the subterranean volume;
• generating a multi-domain model of the subterranean volume using the input data;
• statistically sampling one or more realizations of the multi-domain model based at least in part on an uncertainty associated therewith to generate sampled realizations, wherein statistically sampling comprises using machine learning, k-means clustering, probability bands, or a combination thereof to select the one or more realizations from among other, non-selected realizations;
• simulating the sampled realizations using a field development planning engine to generate simulated realizations for determining viability and profitability of a location and a well trajectory in the subterranean volume;
• generating a field development plan based at least in part on the simulated realizations;
• and
• displaying the field development plan and the multi-domain model to a user, wherein the field development plan is used to manage a hydrocarbon resource in the subterranean volume and to inspect an outlier realization.

A computing system, comprising:

• one or more processors;
• and
• a memory system including one or more non-transitory computer-readable media storing instructions that, when executed by at least one of the one or more processors, cause the computing system to perform operations, the operations comprising:
• performing an oilfield operation using a set of sensors to collect input data from a subterranean volume;
• wherein the input data specifies a weight on bit parameter associated with the operation, a torque on bit parameter associated with the operation, a pressure parameter associated with the operation, a flow rate parameter associated with the operation, a rotary speed parameter associated with the operation, a well log, a porosity parameter associated with the operation, a fluid composition parameter associated with the operation, and a permeability parameter associated with the operation,
• wherein the set of sensors comprises a geophone and a gauge,
• wherein the subterranean volume comprises a shale layer, a carbonate layer, and a sand layer, and
• wherein the operation is a drilling operation, a wireline operation, or a seismic survey operation;
• receiving the input data representing the subterranean volume;
• generating a multi-domain model of the subterranean volume using the input data, wherein generating the multi-domain model comprises:
• generating an ensemble of a plurality of realizations of a first model based at least in part on the input data, an uncertainty of the input data, and an uncertainty of the first model;
• generating a plurality of second realizations of a second model based at least in part on the ensemble of the plurality of realizations and an uncertainty of the second model;
• and
• including the plurality of second realizations in the ensemble in connection with the realizations of the first model, wherein the first model comprises a model of at least one physical characteristic of the subterranean volume, and wherein the second model comprises a commercial model, an economic model, or a combination thereof;
• statistically sampling one or more realizations of the multi-domain model based at least in part on an uncertainty associated therewith to generate sampled realizations, wherein statistically sampling comprises using machine learning, k-means clustering, probability bands, or a combination thereof to select the one or more realizations from among other, non-selected realizations;
• simulating the sampled one or more of the realizations using a field development planning engine to generate simulated realizations for determining viability and profitability of a location and a well trajectory in the subterranean volume;
• and
• generating a field development plan based at least in part on the simulated realizations;
• and
• displaying the field development plan and the multi-domain model to a user, wherein the field development plan is used to manage a hydrocarbon resource in the subterranean volume and to inspect an outlier realization.