GLENCORE’S GEORGE FISHER MINE, QUEENSLAND AUSTRALIA

A geological model is only as reliable as the data it’s built on. Huge resources are poured into gathering as much useful data as fast as possible for that very reason. However, delays in moving that data from point-of-collection to interpretation results in decisions made with an inaccurate and out-of-date understanding of the project. To avoid this, geologists and mining companies aim to create automated fast workflows from data collection through to 3D interpretation and modelling.

Glencore’s George Fisher Mine (GFM) addressed this issue by deploying a workflow using Minalyze’s Minalyzer CS drill core scanner and Seequent’s Leapfrog Geo modelling software. Minalyzer is used by GFM for geochemical analysis of drill core, which is then rapidly imported into Leapfrog Geo for 3D interpretation and modelling.

Seequent and Minalyze worked with GFM to further improve this workflow, using existing functionality in both tools. The result: Leapfrog users can now access the raw XRF results in the Minalyze Minalogger web service at any time, through a URL link in the Drillhole Database.

In the words of Nick Spanswick, Superintendent Mineral Resources at Glencore’s George Fisher Mine, “The challenges we were facing were around maintaining data quality, while not reducing the volume collected. We also needed to ensure continuity between geologists, and to try and undertake quality assurance sooner rather than later.”

Glencore mine site

The Project

The George Fisher Mine (GFM) is located near Mount Isa in North West Queensland, and has been operated by Glencore since 2013. Producing approximately 3 million tonnes of ore per annum makes GFM one of the largest zinc, lead and silver mines globally.

The weakly metamorphosed and stratiform geology of GFM is deformed by a regional northsouth trending fold belt. The two main deposits comprise interbedded dolomite, shales, siltstones and mudstones, with orebodies heavily faulted. Deformation is pervasive, extending down to the scale of a 5×5 metre development drive. The large number of units (80 different domains), the visual similarity of units, and the structural deformation make logging difficult. Tuffaceous marker beds (TMBs) provide useful markers for interpreting core. But correlating these geological features between drilling is challenging without geochemical analysis.

“The challenges we were facing were around maintaining data quality, while not reducing the volume collected. We also needed to ensure continuity between geologists, and to try and undertake quality assurance sooner rather than later.”

Nick Spanswick, Superintendent Mineral Resources,
Glencore’s George Fisher mine

The Situation

Accurate drillhole logging is essential for generating resource estimation models, which determines the thickness, grade and resource reserves of the orebodies. These models ultimately feed into the economic analysis used for mine planning and scheduling, meaning logging errors for a single hole can have profound impacts on the overall mine.

Historically, geologists at GFM were printing sections with previous drilling, and domaining the geology using any surrounding assays and TMBs (which had been identified through hydrochloric acid). Once the geology was logged, potentially economic domains were sent off for analysis at the lab. Once assays were returned 2 to 4 weeks later, the domaining was reviewed.

With so many highvalue decisions riding on the quality of these domains and models, it was essential that the interpretation of drill core and domaining of the resulting data was as accurate as possible.

For Glencore’s geologists, creating a workflow that improved logging consistency and data movement efficiency was key. They were spending significant amounts of time validating models against the core logs and assay results. Additionally, lessons learnt through logging review were not making it back to the core shed fast enough, with the same errors occurring in new core.

Nick Spanswick comments, “Poor data collection leads to large amounts of re-work at the modelling stage. The best time to get it right is at the beginning. However, with complex orebodies there are always going to be issues and the need for verification of data and interpretations. The easier it is to check core logs and assay results, the quicker the modelling process is going to be.”

“Poor data collection leads to large amounts of re-work at the modelling stage”

Figure 1: Glencore_P49 model
Figure 2: Leapfrog assays

The Response

Glencore introduced the Minalyzer CS drill core scanner for on-site geochemical analysis of drill core. They combined this with Seequent’s Leapfrog Geo modelling software (which was already in use at GFM). With the addition of the Minalyzer, their workflow is as follows:

  1. The core is scanned in the coreshed by the Minalyzer CS, producing very fine-scale geochemical analysis in 10 cm intervals.
  2. The core is then logged, with the assay data from the Minalyzer CS assisting domaining of units, primarily through identification of the high potassium TMBs. This highly detailed analytical data allows for improved domaining through better identification of domains when logging.
  3. Once the core is logged, the geochemistry data is composited within the geological intervals.
  4. Newly scanned and composited XRF and logging data is imported or reloaded into the Drillhole Database of Leapfrog Geo, as either a csv file or via the ODBC link. Underground pit mapping and structural measurements can also be appended to this data.
  5. The metamorphosed zinc-lead-silver stratiform deposit is modelled in Leapfrog Geo using the stratigraphic sequence tool, and the Fault System is used to model the N-S trending sub-vertical faults that cross-cut the deposit.

    • The implicit nature of these surfaces and resulting models mean that Leapfrog Geo will re-run and update when new data is added to the inputs. This means new logged data from the Minalyzer CS is reflected in the model with minimal effort.
  6. Drillhole data is updated from the Minalyzer CS and validated daily, with results available in 3D within Leapfrog Geo.
  7. These are then used to produce an updated geological model in just a few hours, with the latest wireframes exported to Deswik for blast engineers and mine planners.

Collaboration between Seequent and Minalyze identified an easy step to improve this workflow, using mostly existing functionality in both products:

  1. In Leapfrog Geo, drillhole tables support columns of URLs, which can link to internet-hosted data or file paths for local data. Minalyze modified the software in Minalyze CS to produce an additional URL tag for each logged interval.
  2. This URL points to the logged data within the Minalogger web-hosted coretray viewer.
  3. This URL is imported into Leapfrog Geo, along with the geochemical logging. Leapfrog Geo users can select drilling in 3D, and choose to follow the URL to view what the original core looked like during the Minalyze scan. The Minalogger tool presents a high-resolution scan of the core as well as overlays for all geochemical mapping.

Easy access to this information during the domaining process helps validate logging, and identify errors before they propagate too far along the process.

GFM’s Nick Spanswick comments, “Wire framing complex orebodies with multiple domains will always have problematic areas and errors in logging. Checking and editing lithologies and domains quickly, by viewing geological models alongside high-resolution core photos and assays, improves this.“

Minalyzer in Core Shed
Core tray

The outcome

The GFM geology team has seen several benefits from integrating their core interpretation and model domaining workflows.

Introducing the Minalyzer CS has meant logging time has decreased by up to 50%, by providing a rapid and objective high-resolution geochemical reference to guide the logging. The incorporation of this highly detailed analytical data in Leapfrog Geo, results in improved domaining through better selection of the economic portions of the orebody. Better granularity of orebody geochemistry and shape allows flexibility of economic stope shapes, based on varying metal conditions. Additional time savings are gained by linking the core photography, as the geologist does not have to spend time searching for the relevant images and chemistry in the database or network drive. Users can quickly and easily validate oddities in a model by checking the geochemistry information, resulting in increased quality assurance and robustness of the geological model.

Minalogger

Pairing the Minalyze workflow with Leapfrog Geo allows for up to date models within hours, as opposed to standard processes with results not available for weeks or months. The immediacy of data interpretation by geologists and the subsequent use by blast engineers and mine planners results in better decisions being made faster, increasing efficiency of resource utilisation.

GFM’s Nick Spanswick concludes, “This dual and dynamic workflow of Minalyze’s geolytical core scanner and Seequent’s Leapfrog Geo software means an increased understanding in poorly tested areas of the resource by rapid access to results leading to better decision making.”

Minalyze CTO Mikael Arthursson, says “We’re excited to see the Minalyze analytical data is improving the workflows of our clients and can be seamlessly transferred into Leapfrog Geo. The integration of these two workflows is a natural progression as both software solutions complement each other’s drive to help geologists uncover valuable insights from their geological data.”

Seequent’s Partner Integration Manager, Sam Bain, says, “Delivering rapid analysis and visualisation for drilling projects is crucial for mining and exploration projects of today. As companies collect increasing amounts of data, understanding and reacting to this information as fast as possible will result in better decisions and huge gains in productivity. Collaboration between the hardware and software suppliers is crucial to making sure data moves as fast as possible, as can be seen in this case. We look forward to continuing to work alongside Glencore and Minalyze in the future.”

“This dual and dynamic workflow of Minalyze’s geolytical core scanner and Seequent’s Leapfrog Geo software means an increased understanding in poorly tested areas of the resource by rapid access to results leading to better decision making.”

Nick Spanswick, Superintendent Mineral Resources, Glencore’s George Fisher mine
Figure 3: Glencore_P49 assay domains