HPC / Subsurface Simulation

Seismic Simulation

Seismic simulation processes massive seismic datasets to model wave propagation through the subsurface. It supports subsurface imaging, resource identification and risk assessment in oil and gas, geothermal energy, carbon storage and broader subsurface engineering.

Wave Propagation Subsurface Imaging HPC Resource Identification Uncertainty

What It Is

Seismic simulation uses physical and numerical models to understand how seismic waves travel through rock layers, faults, salt bodies and reservoirs. By comparing simulated wavefields with measured seismic data, geoscientists can infer subsurface structures that cannot be directly observed.

It is closely related to seismic imaging, seismic inversion and Full Waveform Inversion. The objective is not to create a perfect picture of the subsurface, but to reduce uncertainty and support better exploration and development decisions.

Seismic simulation illustration showing wave propagation and subsurface layers
Seismic simulation models wave propagation through subsurface layers to support imaging, inversion and resource identification.
≈
Definition Seismic simulation is the numerical modelling of seismic wave propagation through the subsurface to support imaging, inversion and geological interpretation.

Key Pain Points

Seismic workflows are among the most data- and compute-intensive processes in the energy sector. The difficulty comes from the combination of large data volumes, complex physics and non-unique geological interpretation.

Pain Point Data scale Seismic surveys can generate extremely large datasets that require specialized storage, transfer and processing pipelines.
Pain Point Noise and uncertainty Measured signals contain noise, acquisition artifacts and ambiguities that can obscure true subsurface structures.
Pain Point Compute intensity High-resolution wave simulations and inversion methods require large-scale parallel computing and memory bandwidth.
Pain Point Model assumptions Results depend on geological assumptions, velocity models and numerical parameters that may not fully represent reality.

Simulation Process

Seismic simulation transforms raw measurement data into interpretable subsurface insight. The workflow usually combines field acquisition, signal processing, numerical modelling and expert interpretation.

1
AcquisitionSeismic waves are generated and recorded using specialized sources and sensors.
2
PreprocessingNoise filtering, signal correction, normalization and quality control prepare the data.
3
ModelingWave propagation is simulated through numerical subsurface models.
4
InversionMeasured and simulated data are compared to update subsurface properties.
5
InterpretationReservoirs, faults, traps and geological structures are interpreted for decision-making.

Simulation Methods

Different methods balance physical accuracy, compute cost and interpretability. High-resolution methods can reveal more detail, but they are more expensive and sensitive to data quality.

MethodFull Waveform InversionUses full seismic wavefields to iteratively improve subsurface velocity and property models.
MethodReverse Time MigrationBack-propagates seismic wavefields to produce detailed images of subsurface reflectors.
MethodRay-based modellingUses simplified wave paths for faster analysis and large-scale exploration studies.
MethodElastic wave simulationModels compressional and shear waves for more physically realistic subsurface representation.

Role of High Performance Computing

Seismic simulation is one of the classic HPC workloads in energy. Large 3D domains, fine grid resolution, many source-receiver combinations and iterative inversion loops create enormous compute and storage demand.

HPC Component Role in Seismic Simulation
Compute clusters Run massively parallel wave simulations across large subsurface domains.
GPU acceleration Speeds up modelling, imaging and inversion workloads.
High-performance storage Handles repeated reads and writes of very large seismic volumes.
Workflow orchestration Coordinates preprocessing, simulation, inversion and interpretation pipelines.

Applications

Seismic simulation is most established in oil and gas exploration, but the same computational principles are increasingly relevant for geothermal projects, carbon storage and subsurface risk analysis.

ApplicationOil & Gas ExplorationIdentifies reservoirs, traps, faults and potential drilling targets.
ApplicationGeothermal EnergyMaps heat reservoirs, fracture networks and subsurface flow structures.
ApplicationCarbon StorageSupports monitoring of CO₂ injection, plume movement and containment risk.
ApplicationSubsurface Risk AnalysisDetects faults, unstable structures and geohazards that influence project safety.

Key Performance Metrics

Seismic simulation should be evaluated by both computational performance and geological usefulness.

ComputeSimulation runtimeTime required to complete modelling, imaging or inversion workflows.
QualityImage resolutionAbility to distinguish subsurface structures at the target depth.
ModelData misfitDifference between simulated and measured seismic data.
DecisionInterpretation confidenceHow strongly the simulation supports resource identification or risk reduction.

Limitations & Practical Considerations

Seismic simulation does not produce an exact representation of the subsurface. Multiple geological models can explain similar seismic observations, especially when data coverage is sparse or noisy.

Results must be validated with well logs, rock physics, geological interpretation and uncertainty analysis. HPC can increase resolution and throughput, but it cannot remove uncertainty by itself.

Wiki note: Avoid implying that seismic simulation directly “finds” resources. A better framing is that it improves subsurface imaging, reduces uncertainty and supports resource identification decisions.