Best Practices for Gridding Drillhole Data
Gridding algorithms have evolved significantly since their inception and have become powerful tools in the geomodelling toolbelt. However, determining the optimal interpolation parameters for your data can be challenging. In the first blog of this series, we examined best practices for gridding concentration data. In this post, we will delve into best practices for gridding vertical well data.
Challenges of Gridding Drillhole Data
The skilled geoscientist typically has an understanding of what the subsurface landscape looks like long before generating any maps or models. The goal for data interpolation is to use collected drillhole data and gridding algorithms to accurately reflect and build on that understanding. When working with well data there are four key data characteristics to consider:
- Three-dimensional data
Data collected at different depths in a well has X, Y, Z, and C coordinates. This increases the complexity of data interpolation and decreases the number of available algorithms. - Uneven data distribution
Well locations can be sparse in some areas and prolific in other areas. This variable XY spacing is often combined with dense, uniformly spaced data in the Z direction. - Multiple data collection methods
Different instruments, such as Laser Induced Fluorescence (LIF) and Cone Penetration Testing (CPT), generate different types of data with different distributions along the length of the well. - Impact of natural phenomena
Natural subsurface phenomena such as soil horizons can impact the preferred flow of monitored materials
Understanding these challenges helps identify which gridding parameters to focus on when creating your 3D grid file or matrix.
Surfer model illustrating the horizontal and vertical data distribution for an example drillhole dataset
Drillhole Gridding Best Practices
1. Select the right data format during import
Before gridding 3D well data, it’s important to consider the type of data that you have down-the-hole. Different data collection methods produce two common data types that differ based on the way the data is spaced along the well path: point data and interval data.
Point data, which is typically chemical concentration data, Laser Induced Fluorescence (LIF) data, and Cone Penetration Testing (CPT) data, can be very dense and has little spacing between data points. Interval data, which is typically lithology data, is spaced farther apart. Each interval, defined by a From depth to a To depth, represents a layer of soil or a section of constant value.
The first step in any gridding process is importing the data. During this step, make sure you select the data type that aligns with your data. If only the from or to values from interval data are imported, this can result in an upward or downward bias in the interpolation results.
Example of interval data layout and import in Surfer
2. Select the gridding algorithm
The next thing to consider when gridding 3D drillhole data is the gridding method. 3D gridding generally does not have as many algorithms available so careful use of the available properties is key. When gridding vertical well data, Inverse Distance will produce the best results 99% of the time.
The Inverse Distance to a Power gridding algorithm is the most universal. This method is fast but has the tendency to generate concentric spheres around high and low values unless you increase the Smooth value. One particularly important feature of Inverse Distance for well data is the ability to specify anisotropy, where weights can be applied to the grid nodes in specific directions.
3. Apply Search and Anisotropy Parameters
When gridding drillhole data, it is essential to account for the preferred orientations of natural phenomena by setting the Search Neighborhood and Anisotropy parameters.
When setting the search parameters, adjust the search ellipse to extend farther in the X and Y directions than in the Z. This is a good way to account for dense data in the Z direction and sparse data in the XY directions common with drillhole point data. Adjusting the search in this fashion can also minimize the blending of data in the areas where intervals meet.
Setting the anisotropy parameters enhances the impact of the search settings by defining the preferred or anticipated orientation. Anisotropy applies preferential weighting in a specific direction to data points within the search ellipse during the gridding process.
For most drillhole datasets, the Anisotropy should set to Anisotropic with an influence ellipse set to accommodate the data distribution. For well data the X Length and Y Length should be set to approximately 10 to 100 times the Z Length values. For dense down-hole data, the Z Length values should be set to a relatively small value encompassing 10-100 points along the well. For interval data, the Z Length value will depend on whether each interval should impact the others. A larger Z Length value will ensure more values are considered during interpolation.
4. Adjust the grid resolution
Grid resolution is similar to image resolution in that the number of nodes or pixels in each direction will determine how accurately smaller features are expressed in the results. Also similar to images, the higher the resolution the larger the file size. The is particularly important to consider with 3D grids because a 50x50x50 grid file contains 125,000 grid nodes (pixels) and doubling the number of nodes in one direction doubles the grid nodes total to 250,000.
To determine the best resolution for your 3D grid file, consider the accuracy required for meaningful interpretation of the results. If every measurement must be represented, the spacing in the Z direction may be quite small while the spacing in the XY direction may remain large in comparison.
Complete 3D model illustrating the interpolation results for key drillhole interval data
Gridding vertical drillhole data comes with some unique challenges, but a good understanding of your data and use of the right tools and settings can ensure accurate results. Use these four best practices the next time you’re modeling drillhole data:
- Import the data appropriately so that all values are considered during interpolation.
- Select the right gridding method for your data.
- Apply search and anisotropy settings to account for your data density and the impacts of natural phenomena.
- Adjust the resolution to align with your data density and accuracy requirements.
Not a Surfer user? Download the free trial to see what it can do for you.
