↑ Top
Table of Contents

Radiometrics processing (C07b)

This cookbook demonstrates the stages in processing radiometric data.


In this Cookbook task files will be used to demonstrate a Radiometrics processing workflow, each task file’s parameters will be detailed here.

At the end of this Cookbook you will have:

  1. Viewed the grid datasets provided, showing the original data and the results of the radiometrics processing.
  2. Used the Multi-Channel Processing tool to compute NASVD spectral smoothing on the raw 256 spectra data
  3. Used the Multi-Channel Processing tool to perform remaining full spectra corrections and extract standard K, U, Th and Total Count windows.
  4. Used the Standard3 tool to compute Compton stripping, height corrections and radio-element conversions on K, U, Th and Total Count fields.
  5. Gridded the processed radiometric fields.

Preliminary notes

You must have at least an evaluation licence to run the Radiometrics worked examples. Contact Intrepid Geophysics or one of our agents if you require an evaluation license.

We assume that you understand:

  • How to use 3D Explore to visualise grids.
  • If necessary for your choice of optional steps, how to use the INTREPID Gridding tool. For instructions, see Gridding (T22a).

Solution datasets and optional steps

For each stage of this worked example we have provided solution datasets and task files that will run you through the operations.

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
then you can find the sample data at

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).

Location of dataset files

In this worked example we assume that the example datasets reside in directory {install_path}/sample_data/cookbooks/radiometrics.

If the sample data is not present or is modified

If the sample data is not present as described above you will need to reinstall your copy of INTREPID, if you no longer have the installer you first used, you can download it from our website. Go to https://www.intrepid-geophysics.com/downloads.


The data you process in this cookbook was acquired over the 1:100,000 Buffalo map sheet, as part of the Victorian Initiative for Minerals and Petroleum (VIMP). The data custodians are Geoscience Victoria, Department of Primary Industries, Melbourne, Australia. The data is freely available through the Geoscience Australia GADDS (Geophysical Archive Data Delivery System).

This data was acquired by a Bell helicopter equipped with an Exploranium GR820 256 channel spectrometer. These instruments utilise real-time automatic gain stabilisation based on tracking of the Thorium peak to control the tendency of the spectrometer to drift. The gamma ray detector had a total crystal volume of 16.8 litres. The sampling interval for the radiometric data was 1.0 second (approx 45 metres sample spacing). The survey line spacing was 200 metres and the flight lines were flown in an E-W direction. The mean survey flying height was 80 metres.


View the grid datasets provided

Under radiometrics/R_GRIDS, view the three supplied grids kroi, uroi, troi, of the window data supplied.

These grids show the standard radiometric windows, extracted from the International Atomic Energy Agency (IAEA) defined regions of interest (K,U,Th) in the raw 256 channel spectrum data. They represent the data at a completely unprocessed stage.

To view the grids, open 3D Explore and select File > Open Ternary Grid Dataset


  • Potassium kroi as the red channel
  • Thorium throi as the green channel
  • Uranium uroi as the blue channel

At this early stage of processing, the count rate representing each K,U,Th window is significantly affected by counts from the adjacent windows. The data is also contaminated by background radiation contributions from the aircraft, cosmic rays, and the presence of Radon gas escaping from the ground. You can see the effect of Radon most clearly in the U grid.

The count rate is also affected by the ground clearance of the detector, as well as the volume of the detector.

In order to reduce the data to a more meaningful state, such that we can draw valid conclusions as well as make sensible comparisons with other surveys, we need to carry out a series of corrections to the raw data. Central to applying these corrections are a series of calibrations which are carried out separately before, during, and after the survey. The calibration data must be prepared, and the results of these calibrations are stored in a special file which INTREPID uses during the processing of the 256 channel data. Accurate radiometric data processing is impossible without properly collected calibration data.

In INTREPID, radiometric data processing is done using the following tools:

  • Multichannel Processing for 256 channel data corrections
  • Standard3 Corrections for corrections to extracted window data
  • Maximum Noise Fraction for an alternative spectral smoothing method to NASVD

Task Files and Processing Steps

These task file can all be run using Run Task File from the right-click menu on each file. The files with the exception of C0, C6 and C9, can be easily run using the Open with Task File option instead. This will open the tool into it’s GUI, where you will be able to inspect and modify the parameters used more intuitively.

To inspect settings used, run the task file with Open With... and navigate to Process > Select Corrections. In the window that appears select the button that corresponds to the setting you wish to inspect/change.

Data Preparation


This task file will modify LiveTime data, create a Cosmic field from a channel in the spectrum and it will conditionally delete rows containing no Radiometrics data. It achieves this using the Spreadsheet Editor.

NASVD Smoothing


This task file will perform NASVD smoothing on the input spectrum as well as saving out the Principal Component Spectra.

The NASVD process is performed along Flight and an integration period of 400.

The Principal Components can be view using 3D Explore.

Note that both NASVD and MNF smoothing methods support line or flight based processing, and also spectra stacking, a method which boosts the Signal/Noise ratio.

Livetime Correction


Perform a livetime/deadtime correction using the livetime field in the database, other instruments may not record a livetime measurement, however a DeadTime constant (each pulse) can be used instead.

Energy Calibration


The task file defines ranges of channels to cover radio-peaks, these are typically referred to as windows, to calibrate.

Cosmic and Aircraft Background Removal


Estimates a cosmic background spectrum from an observed cosmic field in the database, the field must have the alias of Cosmic.

With 256-channel processing, the background corrections can be done as full spectrum corrections. Single window coefficients are replaced by spectra, which have a unique value for each channel in the spectrum. These spectra are stored in the calibration file. Consequently the background corrections can be performed more accurately than is possible using windowed data.

Radon Removal


Removes Radon signal from the signal by estimating radon content in the atmosphere. This cookbook uses radon constants c1,c2,c3,c4 as 1.95, 0.71, 0.0268 and -0.0179 respectively and assumes a terrain clearance of 80m

INTREPID uses the spectral ratio’ing method described by Minty (1992) to remove radon from the 256 channel spectra data. It examines the ratio of counts in the 0.609 MeV Radon window to the Bi214 (U) window to estimate the amount of Radon contamination. The process requires a 256 channel pure Radon spectrum in the calibration file.

The constants C1 and C2 and the integration period control the extent of radon removal from the data. C3 and C4 are normally set to zero.

Increasing C1 estimates that the contribution from Radon to the Radon window is larger, so removes a higher proportion of counts. Decreasing C2 estimates that there is a smaller Compton scattering contribution from Uranium in the Radon window.

The Compton stripping ratios alpha, beta and gamma and terrain clearance parameters are used to remove K, U, Th contributions from the raw Radon count rate.

Note: This process does NOT perform Compton stripping, even though the stripping ratios are required. This process also requires the survey terrain clearance.

View Radon Removal effects


View the effect of Radon Removal on the Potassium, Thorium and Uranium Fields. The impact of Radon removal more heavily impacts the profile of Uranium, as expected.


Create corrected spectrum comparison


Subtracts the raw spectrum from the Energy Calibrated Spectrum,

View Spectral Changes


Visualises the changes in spectrum due to Background and Radon removal.


View Spectral Changes from all stages



3 Channel Standard Processing


Converts the fields with Counts Per Second to ground concentrations.

In this case the conversions are:

  • CPS Per Total Count = 1.0
  • CPS Per Potassium Percent = 54.48
  • CPS Per Uranium PPM = 6.21
  • CPS per Thorium PPM = 3.14

Create Potassium, Thorium and Uranium Grids


This gridding task will iterate over the Potassium, Thorium and Uranium fields and produced homogenized grids to be views separately, or as part of a RGB visualisation.

View Corrected Ternary Grid



Optional Steps after completion

If you have completed the main exercise and have some spare time, try adjusting the following parameters and examine any differences in the results.

The following table suggests some adjustments you can make.



Implication and Effects



Estimates that the contribution from Radon to the Radon window is larger, so removes a higher proportion of counts.



Estimates that there is a smaller Compton scattering contribution from Uranium in the Radon window.

Integration time
(bunch size)


Uses shorter line segments to calculate the Radon contribution. If you make it too small, you degrade the data.

See Minty (1997) for a description of the derivation of C1, C2, C3 and C4

You can also save the Radon peak as a dataset field by selecting it in the Channels to be saved dialog box. Gridding and inspecting this field may give you a clearer picture of where Radon is affecting the survey data.

Determine channel positions of IAEA peaks

The Energy Calibration process requires high and low channel bounds for the main photo peaks. These must be determined accurately from the data, otherwise the process may fail to find the main peaks.

In 3D Explore choose to open the dataset and right click in the workspace. Choose to Create a Spectral Plot for "data"

Performing NASVD or MNF smoothing

You may wish to perform NASVD or MNF smoothing before the multi-channel corrections. On a large dataset the spectral smoothing could take several hours. Both NASVD and MNF processes are straightforward, so we have omitted detailed steps.

If you process by flight instead of by line, it ensures that INTREPID uses a higher number of samples to gather statistics and aid the signal de-noising process. This is most beneficial for Uranium estimates.

You may be able to further enhance results by combining local spectra to ‘boost’ the signal/noise statistics. There is an option to do this in INTREPID.

What is in the radiometrics sample data cookbook?

The radiometrics cookbook contains the following datasets

The radiometrics line dataset cookrad_new..DIR

This radiometrics dataset is a subset of a larger dataset that was originally flown over the 1:100,000 Buffalo map sheet in the Eastern Highlands of Victoria, Australia.

The test dataset is a helicopter survey with specifications:

  • 200 metre line spacing
  • 55 metre terrain clearance
  • GR820 spectrometer with 16.8 litres of crystal detector
  • 1 second sample rate
  • 40 metres/second aircraft speed
  • Real time Differential GPS navigation

The cookrad_new..DIR dataset contains the following fields









Cosmic count (total counts > 3 MeV) recorded by the spectrometer in real time



East-West location






GPS height


Date Field





Flight number



Line number



Line type



Terrain clearance corrected for temperature and pressure (also called Effective Height)


HourMinuteSecond time format


Julian Date


Latitude as AGD66


Latitude as WGS84



Live time recorded by the spectrometer in real time


Original LiveTime before correction


Longitude as AGD66


Longitude as WGS84



North-South location



Raw data recorded by the spectrometer in real time and summed for K window



Final processed Potassium data



PseudoFlight Field which can be generated to group data that were recorded on same flight/conditions.



Terrain clearance


256 band

Raw 256 channel spectra recorded by the spectrometer in real time






Raw data recorded by the spectrometer in real time and summed for Th window



Final processed Thorium data






Raw Total Count (0.4–3.0MeV) recorded by the spectrometer in real time and summed



Final processed Total Count data



Raw data recorded by the spectrometer in real time and summed for U window



Final processed Uranium data

Spectra Calibration file

A spectra calibration file, cosmic_radon_jwf.asc_data contains the aircraft background, cosmic background and Radon calibration spectra for the data. This file exists in the directory {install_path}/config/calibration_spectra.


See “References” in Multi-channel gamma ray spectrometric processing (C07)