Skip to content Skip to main navigation Skip to footer

This dataset was designed to study the impact of channel position, geometry and variations properties such as porosity, permeability on reservoir performance. This dataset represents a simplified depiction of fluvial depositional setting while preserving key geological features and variability.

Main characteristics

  • 3D «cubes» of properties (porosity and permeability).
  • Model dimensions: 16 × 12 × 10 cells.
    Grid resolution: 50 × 50 × 1 m.
  • A total of 5,000 realizations were generated. Half of the models contain one channel, and the other half have two channels. Each realization is unique.

A commercial Petrel subsurface software (SLB™) was used to create the set of models. The process of creating models for the dataset structured as follows. Stochastic Object Modelling was chosen as the geological modelling process, as it is the most preferable for modelling fluvial systems. Two workflows were implemented to create the dataset.

The workflow simulates the propagation of channel/non-channel facies in 3D volume. In addition to the spatial distribution of channels, their parameters were also varied. The workflow results in porosity and permeability grids (*. GRDECL format).

Uncertainty parameters

To create diversity in the models, the following parameters were varied:

  • Width, curvature amplitude, and orientation (dip angle) of the channels.
  • Channel positioning within the model was altered, allowing the study of how geometry influences the reservoir flow and fluid properties.
  • Spatial distributions of porosity and permeability within and between channels were utilized.
ScenarioSingleDouble
Number of channelsonetwo
Width[300; 500][300; 500]
Thickness[10; 20][10; 20]
Wavelength[1000; 2000][500; 1000]
Amplitude[0; 300][500; 900]
Orientation90120

The petrophysical relationship between porosity and permeability was defined by a deterministic function with some slope variation and bias to provide some uncertainty.

The petrophysical relationship between porosity and permeability follows a “standard” exponential relationship. It should be noted that the uncertainty range for the entire dataset is quite broad, which can be attributed to the inherent uncertainty in the petrophysical relationship of the parameters. However, each individual dataset realization has a much smaller variability.

Porosity/Permeability petrophysical relationship for the channel facie for some random model samples. Left: y-axis in millidarcy, right: the same relationship, but the y-axis in log(millidarcy)
Examples of FLUVIAL synthetic dataset: a single channel scenario (A) vs the double channel scenario (B) are considered. Realization (C) in which only one channel is visually observed, but in reality, two channels merged into one due to the stochastic component of the object modelling algorithm

Usage rights

  • The dataset is provided exclusively for educational and research purposes.
  • Access to the dataset does not give the right to redistribute or share the data in any way.
  • Every research paper or other materials built on this dataset should include a reference to DOI 10.3997/2214-4609.202244105.

Get access

Please contact us at ShirokovAV@hw.tpu.ru to request access.