3D grids are essential tools used to organize and visualize subsurface data in a structured format. The process typically involves the interpolation of scattered data points (e.g., well tops or seismic picks) across a predefined grid.

Grids are built using interpolation techniques that estimate unknown values between data points. The grid can vary in resolution, based on the interpolation steps defined by the user, allowing flexibility in adapting to different geological complexities.

To begin the 3D gridding process in g-Space navigate to the Data Manager, find the 3D Grids section. This is where grid data will be organized and stored.

To create a grid, right-click the 3D Grids folder and select Create new grid. In Select grid type, choose S-Grid or Tetrahedral and click OK.

To create a tetrahedral grid, first run a structural model with at least two horizon result maps. In Create Tetrahedral Grid, enter the Grid name, select the Structural model, and choose a Layering type. For Proportional layering, set the Layers per zone; for Parallel to Top or Parallel to Bottom, set the Layer thickness. Set the horizontal interpolation steps, then click OK to create the grid.

For an S-Grid, enter the Grid name, choose an optional Polygon and whether to Extrapolate in the polygon, set the horizontal interpolation steps and Margin, and select the Input type. The Faults group defines whether faults are taken into account when the grid surfaces are built.

• Polygon: If available, you can select a polygon to limit the grid to a particular area. Choose “Not required” if the grid is intended to cover the full extent of the data.
• Extrapolate in the polygon: Selected by default, and used only when a Polygon is selected. It rebuilds the grid surfaces by re-interpolating the input values that fall inside the polygon. Clear it to read the values directly from the input horizons or maps at each grid node instead, so that the surfaces follow the source data. Either way the grid stays empty outside the polygon and wherever the input data has no values.
• Interpolation Step X [ft/m] and Interpolation Step Y [ft/m] These define the resolution of the interpolated data. Smaller steps result in higher grid resolution, capturing finer geological details but increasing computational demand. Larger steps provide a more generalized view but are faster to compute.

Notes

Geological Context: In regions where fine stratigraphic details or thin reservoirs are critical (e.g., offshore channels, thin carbonate layers), smaller steps (10-20 ft/1-5 m) may be appropriate. In contrast, for larger, homogenous geological structures (e.g., broad anticlines or synclines), larger steps (50-100 ft/20-50 m) may suffice.

• Margin [ft/m]: This parameter sets an additional boundary around the grid. A larger margin might be useful in areas where you expect the geology to extend beyond known data points.

In Faults, Mode defines how faults are used when the grid surfaces are built:

• Ignore faults: the default. The grid surfaces are built without using faults.
• Use selected faults: shows the Select faults list, where you select the check box of every fault to apply. If no fault is selected, the grid is built without using faults.
• Use fault model: shows the Fault model list, where you select one existing fault model.

Select faults and Fault model offer only the objects that match the domain of the selected Input type: time faults and time fault models for Time horizons and Time maps, depth faults and depth fault models for Depth horizons and Depth maps. Changing the Input type rebuilds both lists and clears any selection that is not available in the new domain.

For both fault modes, Clear radius [ft/m] sets how far from a fault trace the grid values are recalculated, so that the surface is no longer interpolated across the fault. Faults classified as Normal also leave a gap along their trace; fault types are assigned in the Faults branch of the Data Manager.

In Create grid, click Run to create the S-Grid. Click Save to workflow to store the setup as a workflow task, or Close to dismiss the window. Run and Save to workflow leave the window open.

To review or change the created S-Grid's physical extent, right-click the grid and select Settings. The Grid parameters window controls the grid's position, rotation, spacing, and size. It also provides the same Faults group as the creation window, so fault handling can be changed after the grid exists. Here the fault and fault model lists follow the grid's own domain, and applying a change rebuilds the grid surfaces with the current fault settings.

• Start X [ft/m] and Start Y [ft/m]: These values define the starting coordinates of the grid in the X and Y directions.
• Step X [ft/m] and Step Y [ft/m]: These values determine the spacing between the grid nodes along the X and Y axes. Smaller grid steps result in a finer grid, while larger grid steps produce a coarser, more general representation.

Notes

Geological Context: For detailed geological structures, use smaller grid steps (e.g., 10-20 ft/1-5 m). Larger grid steps (e.g., 50-100 ft/20-50 m) can be used for large-scale structural features.

• Angle [degree]: This defines the rotation of the grid relative to the coordinate system.

Notes

Grid rotation is essential for aligning the grid with key features to improve accuracy and efficiency:

1. Tectonic Structure Alignment: Rotate the grid to match elongated tectonic features, minimizing the number of cells inside the reservoir contour while preserving horizontal resolution.
2. Well Patterns: Aligning the grid with injection and production wells ensures more direct fluid flow paths, reducing filtration time and increasing efficiency.
3. Fault Orientation: Rotating the grid along fault trends prevents grid distortion near faults, maintaining smooth geological transitions.

• Length X [ft/m] and Length Y [ft/m]: These define the total dimensions of the grid, controlling how far the grid will extend in the X and Y directions. Adjust these based on the size of your study area.

In Grid parameters, click Run to apply the changes, Save to workflow to preserve the setup, or Close to dismiss the window. Run and Save to workflow leave the window open.

Right-click a grid in the 3D Grids section of the Data Manager to work with it. Available actions depend on the grid type. The options below describe an S-Grid. Tetrahedral grids omit the S-Grid-only Edit zone, Update, Export to ECLIPSE, QC, and Upscale seismic volume actions. Their menu instead provides Export to VTK, Export to GRDECL, Export to GOCAD TSolid, Convert to Geobodies, and Mesh optimization.

Settings: Opens Grid parameters, where you can change Start X, Start Y, Step X, Step Y, Angle, Length X, Length Y, and the fault settings.

Edit zone: Opens Manage grid. Add or remove zones with + and -, select the Top and Bottom, choose a Layering type, and set Layers count or Layers thickness as applicable. Click Run to apply the changes, Save to workflow to preserve the setup, or Close to dismiss the window. Run and Save to workflow leave the window open.

Notes

The number of layers in a zone, is influenced by several factors:

1. Geological Heterogeneity: The vertical variability in properties, like the average thickness of reservoir layers, dictates layer thickness. For instance, if the average layer thickness is 1 meter, grid cells larger than 50 cm shouldn't be used to capture details effectively. This is assessed using histograms of layer thickness.
2. Well Data Sampling Interval: Typically, well data is sampled at intervals of 0.2 to 0.4 meters, which affects layer thickness.
3. Total Thickness: The entire thickness of the interval to be subdivided impacts the number of vertical layers required.
4. Computer Limitations: The computational power available may restrict the grid resolution that can be practically implemented.

Upscale Log Data: Creates cell values from a selected LAS curve. Choose the wells, curve and curve type, averaging method, cutoffs, minimum cell coverage, null handling, and sampling direction. Click Run to store the result under Upscaled cells, or Save to workflow to preserve the setup for reuse.

Propagate Upscaled Properties: Creates a grid property from an item under Upscaled cells. Select the source upscaled curve and the modeling and propagation settings. Use Edit per-zone settings... when different zones need different parameters, then click Run. Save to workflow preserves the setup for reuse.

Update: Refreshes or recalculates the grid, applying any recent changes in data or grid parameters.

Export to ECLIPSE: Allows the user to export the grid to the ECLIPSE format, which is widely used for reservoir simulation and modeling. This function is critical for taking the grid data from g-Space into reservoir simulation software.

Rename: Renames the selected grid in the Data Manager for organizational purposes.

Remove: Deletes the selected grid from the project.

Show grid report: Shows the report of the selected grid in the Properties table of the Table view. For an S-Grid, the report displays general parameters of the grid, such as dimensions, resolution, and volume, and provides detailed information for each zone within the grid. For a tetrahedral grid, it holds only a single Entire grid row with the cell count and volume, followed by one statistics block with the rows Min, Max, Average, Median, Rms, and Most common and one column per grid property; there are no Step X and Step Y values, no Top, Bottom, or Layers columns, and no per-zone tables. The data from the grid report can be copied directly or saved as a .CSV file

Upscale seismic volume: For an S-Grid, samples a selected seismic volume into the grid. Enter a Name, select Seismic data and Value extraction, set the modeling and propagation parameters, and click OK.

Advanced Options and Applications

Polygon Clipping: If you are working in a specific region of interest, using a polygon to limit the grid’s extent can reduce computation time and ensure focus on relevant data.

Time vs. Depth Grids: Grids can be created in both time and depth domains. A grid is built from time or depth horizons and maps and keeps its native domain — a grid built from time inputs stays in time, with no depth conversion applied. Depth grids incorporate well data and velocity models for more accurate geological interpretations. Well logs (LAS) can be upscaled onto a time-domain grid provided the wells carry a checkshot (time-depth relationship), which positions the log samples in time.

3D grid can be visualized in 3D depth view as well as in section view. Visualization settings can be configured in the Visual Settings, in the example below, lithological fill type is selected for the Grid zones.