↑ Top
Table of Contents

Full tensor gravity gradient cookbook (C18)

This cookbook has two examples of simple and direct steps for reducing a raw full tensor gradient (FTG) dataset that is located in a time series of observations.

Many aspects are novel to Intrepid V6.

Sample data location

Location of sample data for Cookbooks

Where install_path is the path of your INTREPID installation, the project directory for the Cookbooks sample data is install_path/sample_data/cookbooks.

For example, if INTREPID is installed in
C:/Intrepid/Intrepid 6.1.0f3e954b6ca6_x64,
then you can find the sample data at
C:/Intrepid/Intrepid 6.1.0f3e954b6ca6_x64/sample_data/cookbooks

For information about installing or reinstalling the sample data, see the relevant section in “About the sample data for the INTREPID Cookbooks” in Using INTREPID Cookbooks (C12).

For a description of INTREPID datasets, see Introduction to the INTREPID database (G20). For more detail, see INTREPID database, file and data structures (R05).

The .task files and data for this cookbook are in subfolders of this location.

Aurizonia example

This example workflow is based upon LookHeedMartin GGI tensor survey data.

The .task files and datasets for this cookbook are in the subfolder {install_path}sample_data/cookbooks/tensors/FullTensorGravityGradients_FTG\Aurizonia.

The data is sourced from Bell Geoscience, and the Brazilian oil exploration company is also thanked for making this available.

The setting is shallow, swampy, sand dunes, with near surface gas.

The Data as delivered from Bell, is Aurizonia_Block_5_Air-FTG.gdb, and also the RAW, ASCII version.

The dataset has already had a lot of prior processing by Bell – Free Air gravity gradient componenets (Txx_fa etc), Terrain Correction gravity gradient components (TC_Txx_100 etc), and the complete Bouguer components (Txx_BG_200 etc), where the terrain correction at a density of 2.0 g/m**3, is removed. In this case, critical examination of the contractor supplied products fairly rapidly indicates areas where Full tensor processing shows issues to work on. eg density should be expected in a normal range, say 2 to 2.4 questions then raised eg was the DTM at high enough resolution? Was the Terrain Correction model done assuming density of 1.0 etc

Earlier stage processing is not developed for this dataset. (It is demonstrated for the magnetic tensor gradients workflows).

We recommend that you do most of the processing on the tensor gradient field, not on its components. Therefore you should switch to a tensor field as soon as possible. The six degrees of freedom are in play in material science. In a geology setting, solid mechanics response to stress loadings is rarely isotropic. The Full Tensor Gravity Gradient remote sensing measure is one way to sense what rocks are where, and to locate the fractures.

This cookbook demonstrates most of the gravity gradient processes available for these standard enhancements and processes for FTG.

Follow these steps (Aurizonia)

Here is a set of steps that encapsulate the simple and direct steps to take to reduce a Raw FTG dataset, which is located in a time series of observations.

  1. Form a Full tensor signal from its mixed gradient observed parts.
  2. Use a Full tensor Noise reduction 1D filter on a moving window of 11 points, or more (Check the profile plots)
  3. Gather whole of survey directional biases for the gradients
  4. Look at by flight statistics, and the possibility of the biases – correct applied.
  5. Look at classical heading corrections by line and direction, but this time the full gravity tensor, under damp iteratively some adjustments.
  6. Create a Crossover network of ties/lines and create statistical measures of mis-closures of the signal
  7. Loop level adjust the misclosures, and create an adjustment for each line of the survey, and reduce the errors, removing the mis-closure and the line to line mis-fits.
  8. Demonstrate how to do a very rigorous time/spatial corrections that allow a drift in the observing instrument system, and other small error adjustments.
  9. Full tensor gridding, with a finer cell size than one quarter the line spacing – up to 10 cells!
  10. Demonstrate residual error streaking, using a decorrugate tool, as flight lines at bearing of 135, need a rotated grid as the corrugations are usually in the primary line direction.
  11. Micro level from the decorrugated grid back into the Line Dataset, creating a micoLevelled Field
  12. Perform terrain correction options
  13. Take the best gravity tensor grid and estimate the gravity grid, using FFT methods,
  14. Take the tensor grid, and then look for sources using Euler Deconvolution. This is followed by a clustering step to aggregate the discrete solutions, while looking for "Faults/Contacts" structures.
  15. Other interpretation options – search for zones dominated by 2D sources, do a cluster analysis.
  16. Estimate the Vertical component of Gravity using one or more of the tensor components – see the difference, depending upon how many components you use.

Hagar example

The .task files for this example are in the subfolder tensors/Process_Hagar_Tasks.

Follow these steps (Hagar)

The example workflow has the following steps:

  1. Import the raw data into INTREPID
  2. Split into lines,
  3. Form a full tensor signal from its mixed gradient observed parts.
  4. Use a Full tensor Noise reduction 1D filter on a moving window of 11 points
  5. Gather whole of survey directional biases for the gradients
  6. Create a crossover network of ties/lines and create statistical measures of mis-closures of the signal
  7. Loop level adjust the mis-closures, and create an adjustment for each line of the survey, and reduce the errors, removing the mis-closure and the line to line mis-fits.
  8. Demonstrate how to do a very rigorous time/spatial corrections that allow a drift in the observing instrument system, and other small error adjustments.
  9. Full tensor gridding, with a finer cell size than one quarter the line spacing – up to 10 cells!
  10. Demonstrate residual error streaking, using a decorrugate tool