Vendor column: Qualitative models and their application to mining – “How you gonna get any gold if you’re not diggin’?”

Skip to Navigation


  • Published: Jun 13, 2012
  • Author: Daniel Shiley
  • Channels: Infrared Spectroscopy / Chemometrics & Informatics
thumbnail image: Vendor column: Qualitative models and their application to mining – “How you gonna get any gold if you’re not diggin’?”

Daniel A. Shiley – Senior Application Chemist, SummitCAL Solutions Team, ASD Inc.

This is the second of a four-part series on chemometric applications for materials analysis.

Those of you who have watched the “Gold Rush Alaska” television show on the Discovery Channel might recognize the quote. This is an old attitude, much different than today’s commercial mining companies, which are all very concerned with obtaining information before they begin to dig. The data derived from near-infrared (NIR) analysis often plays an important role in development of this much needed data.

Alteration defined—Rock does change!

In certain geologic conditions, rock may become altered or changed from one mineral to another. During the alteration process, metals such as gold, silver or copper may also be deposited or concentrated in the alteration zone. The presence or absence of certain minerals may be indicative of these alterations. The presence of the alterations themselves does not guarantee that mineralization (an area where metal has been deposited) will have occurred. Additional testing is needed to confirm that a particular alteration has been mineralized because NIR cannot measure the metal content of rock. However, in most cases if there has been no alteration, there will likely be no mineralization. So the tracking of the alteration can provide an understanding of where there may be mineralization, so even though we cannot measure the target metal, NIR can provide very important information that can be used to create an overall mine plan.

Development process explained—Application of technology improves success rates

The mine site development process can be broken down into three time periods: early exploration, late exploration and mine operations. During each of these stages NIR plays a role.

During early exploration, NIR data from an airborne or spaceborne sensor may be obtained to begin to identify a broad area of exploration. Data is commercially available from many sources for this purpose. Ground truth data collected in the field with a portable spectrometer is used to develop the key that relates spectra from the overflight sensor to the spectra collected on the ground. Ground truth data is related via GPS coordinates to the overflight data. Once the ground truth data have been obtained, a wide-area assessment map can be created using the overflight NIR data. If interpretation of the wide-area assessment data indicates a likely area for further study, then a field campaign can be developed in which a team of geologists use a portable NIR instrument, such as the TerraSpec® 4 mineral analyzer, to measure the surface rocks in the identified target area at discrete intervals. The NIR data that has been collected in the field is then used with a program such as The Spectral Geologist™ (TSG®), developed by CSIRO and marketed by AusSpec International, to interpret the spectral signatures of the rocks that are present. If this area continues to look promising, a drilling plan is then developed based on a combination of the geologist’s field observations and on the interpreted spectral data.

Exploratory drilling produces solid core samples, which can be tested using NIR on the solid rock, or the sample can be crushed and composited to represent an interval associated with normal mining operations. Typically measurement of mineralogy is expensive and time consuming, so NIR can be used to produce mineralogy measurement at a frequency that would otherwise not be practical or affordable for the exploration project. In this manner, NIR provides an economical and rapid means of characterizing a large number of samples. High-throughput analysis of samples is a task well suited for NIR, and we find that NIR is used for this purpose in many diverse industries. Once spectra have been collected on the drill core samples, these can be interpreted using TSG alone, or in combination with other mining-specific programs such as Leapfrog Mining software. Leapfrog is used for geologic modeling to visually display the combined information from a variety of sources to aid in interpretation of complex data sets that will be used to determine if it is feasible to create a mining operation at the deposit.

During the late exploration phase, a decision is made to undertake additional drilling in order to develop a block model, or to refine the model developed in the early exploration phase. This drilling program is on a much larger scale than in the early exploration stage and is typically done to provide an enhanced level of detail to facilitate refinement of the block models. Again, at this point, NIR is instrumental in providing additional information on the mineralogy of the samples. During this development stage more samples may also be submitted for reference mineral determination. As with the early exploration phase, the NIR models that are used are still qualitative as they continue to provide an indication as to which minerals are present or absent, but quantitative NIR models may begin to be created. At this stage, the block models ideally should include more quantitative mineral data.

In the production phase, once a mine has been developed, additional samples from blast holes need to be characterized to aid in short-term planning. Although qualitative or semi-quantitative values can be used, typically more of a quantitative result is preferred. Creation of quantitative models will be discussed in future articles.

Qualitative NIR model information—What can be accomplished?

Figure 1

Figure 1. Screen shot of TSG comparison of a sample that contains kaolinite Screen shot of TSG is shown above. The sample spectrum is shown in rainbow colors based on the intensity of the spectral response. Also shown on this plot is an overlay of the mineral that has been identified.

Figure 2

Figure 2. Screen shot of the Summary screen in TSG.

In the summary plot above of a single drill core, the changing mineralogy can easily be seen. This information is useful and important for development of the block models that will be used with this deposit.

The qualitative NIR information used for mining exploration and mining production typically is not the result of a multivariate model, but instead is usually from either peak identification programs utilizing a spectral library search or spectral unmixing programs. These programs can be used to identify the predominant mineral or minerals that are contributing spectral features. Not only is there an indication of minerals that may be present in the sample, but also some indication whether these minerals are increasing or decreasing in quantity. However, the values that are produced by these programs are called spectral abundance, which should not be confused with concentration. Spectral abundance simply means that of 100 percent of the spectral signature, some percentage of each mineral that has been identified is contributing to this total observed signal. Remember that not every mineral has a diagnostic spectral signature in the NIR region, so some minerals cannot be identified using NIR. As an example, a rock sample that contains 30% muscovite, 6% kaolinite, 10% pyrite and 40% quartz would be identified by the qualitative model as containing only muscovite and kaolinite and even the relative percentages of these may not be in the correct order. The values produced by these programs are not as important as the identification of the pathfinder minerals. As previously mentioned, the minerals are indicators of the alteration or proximity to the alteration.

Qualitative mineralogy results based on NIR are important for early and late exploration and can even be used to aid in short term planning of an operating mine. In future articles, we will discuss quantitative models and their application predicting mineralogy using NIR.


  1. Goetz, A.F.H., Curtiss, B., Shiley, D.A., 2009. Rapid gangue mineral concentration measurement over conveyors by NIR reflectance spectroscopy. Minerals Engineering 22 (2009), 490-499.


Article by Daniel Shiley

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.

Social Links

Share This Links

Bookmark and Share


Suppliers Selection
Societies Selection

Banner Ad

Click here to see
all job opportunities

Copyright Information

Interested in separation science? Visit our sister site

Copyright © 2017 John Wiley & Sons, Inc. All Rights Reserved