Show button Hide button Support

Structural Uncertainty Modeling

In classic reservoir modeling the structural framework is built as a best-case estimate of horizons and faults based on the available input data. Since the input data is commonly assumed to be accurate, the resulting structure model is used in the decision-making process when planning an economically optimal depletion strategy. However, since both horizons and faults have significant uncertainties associated with their locations and features, a deterministic structural model ignores both the risks and the possibilities associated with our limited knowledge of these important reservoir characteristics.

RMS™ Structure Uncertainty Modeling tools are designed to incorporate and mitigate these aspects. The solution aims at helping the decision maker acknowledge and quantify the uncertainties associated with the modeling of a structural framework. Sensitivity studies of key parameters, such as horizon depths and fault locations, are facilitated through a transparent, powerful, and fully reproducible way of modeling. Fault and horizon uncertainty tools tightly linked to structure modeling and 3D gridding have been added, making it fast and easy to build several geological scenarios for investigating the effect of structure uncertainty.

Structure Uncertainty Benefits

  • Explore fault sensitivity easily using fault uncertainty modeling tools 
  • Handle data efficiently by taking cross domains data as input for studying horizon sensitivity 
  • Adjust structure model quickly and accurately with new data using a simple and short workflow consisting of powerful horizon uncertainty modeling tools
  • Analyze volume uncertainty comprehensively with the provided realistic structural scenarios.   

Structure Uncertainty Features


Example of an uncertainty envelope representing positional uncertainty of a fault

Fault uncertainty modeling allows changes to the position, strike, dip and throw of faults in an integrated structure model. Since the fault uncertainty tool is tightly integrated with structure modeling and 3D gridding, the user can rapidly build these models in full, to investigate the scenarios corresponding to uncertainty in the input data. Perturbing fault parameters like throw dip, strike and location in a Fault Uncertainty Modeling job is also accessible from the RMS Uncertainty Management module, allowing fast investigations of multiple fault scenarios, where dependency between faults is also properly accounted for.


Example of updates between time domain input model (dark grey lines -horizons, red - faults) and depth domain output model (filled zones) after running horizon uncertainty modeling

Horizon uncertainty modeling makes consistent horizons in depth by including realistic uncertainties on all the input data. Trends for thickness and velocity intervals are used as input for calculating horizons; the horizons can also be conditioned to, or made consistent with, well data. Prior uncertainties can be included on all the input and the posterior uncertainties are calculated for the output horizons. Horizon Uncertainty Modeling in RMS is highly integrated with COHIBA, a computer program developed by the Norwegian Computing Center that handles surfaces efficiently, and enables input horizons to be adjusted to additional data while at the same time calculating uncertainties on the adjusted horizons. The strength of the Horizon Uncertainty Modeling functionality is that it is possible to use data from different data sources in the conditioning, including velocities, isochores and well data, while acknowledging that there are uncertainties on all the input data.

Horizon uncertainty modeling is based on an existing horizon model. Horizon models in both the time and depth domain can be used. Furthermore, horizon uncertainty modeling handles faulted horizons which are treated semi-independently by COHIBA and RMS. After COHIBA has adjusted the horizon patches, RMS imports the patches and builds them back to a consistent structure model with faults and horizons. The main output from horizon uncertainty modeling is thus a structure model with both an updated fault model and a horizon model. The faults in the fault model shift with the adjusted horizons. 



Moving faults or horizons can obviously change the volumes in place. With the fault and horizon uncertainty tools, sensitivity studies in RMS are easy and straightforward to set up and run. By using the workflow concept in RMS, a range of different scenarios can be built and gridded by simply varying the input parameters and running the workflow.

The histogram shows how the bulk oil volume is affected by moving a fault with defined uncertainty in the horizon model with 11 realizations. RMS also provides Tornado charts showing the impact of the different uncertain parameters, which can be used as a decision support tool.

Bulk volume histogram colored with realizations
Tornado plot to investigate which are the most important uncertainties