string(10) "[Minerals]"

Advanced Analysis of the QUEST-South Stream Sediment Geochemical Data, British Columbia

Lead Researcher(s):  D. C. Arne

Project ID:  2018-016

Key Research Organization(s):  Telemark Geosciences Pty Ltd

Project Location:  Southern BC

Strategic Focus Area:  Minerals

Summary



This research project, managed by Telemark Geosciences, has generated a series of predictive maps based on probability estimates for various mineral deposit types in BC’s South Central Region. The research is based on combining existing stream sediment geochemical analyses and various classes of mineral deposit types from the BC MINFILE database.

The regional stream sediment geochemistry dataset was previously compiled for the QUEST-South Catchment Basin Analysis and Stream Sediment Exploration project, which was completed in 2011. The multi-element geochemistry was integrated with MINFILE deposits, prospects, occurrences and anomalies. The application of advanced multivariate statistics and machine learning methods resulted in maps where increased mineral resource potential was identified with measures of likelihood (probability).

The Need

Publicly funded regional stream sediment sampling projects can be used by the mineral exploration sector to provide improved regional geochemical information. Analysis of this information can help companies, governments, communities and Indigenous groups make decisions about mineral exploration activities.

However, the mix of sands, clays and gravels collected from a single point in a stream (a “stream sediment sample”) may have travelled a significant distance, including from tributaries, over a long period of time. As a result, sediments may have mixed geochemical signals from many rock types within the catchment basin that can mask geochemical or physical clues related to a mineral deposit.

Analyzing the chemistry of the sediments in a laboratory can indicate the presence of mineral deposits, and a more detailed statistical analysis of the data can provide more robust information about the location of potential deposits.

Using advanced multivariate statistical analysis techniques and machine learning, this project relates stream sediment geochemistry with geology and geography to determine the most likely deposit type that a stream sediment sample in the study area originated from.

Project Goals

This project fits under Geoscience BC’s Strategic Objective of ‘Identifying New Natural Resource Opportunities’ and the goal to:

  • Undertake research that adds value to existing or ongoing data sets through ground-truthing studies, data interpretation and mining camp compilations.

Specifically, this project:

  • Advanced analysis of the QUEST-South data;
  • Produced predictive gridded and thematic maps;
  • Generated exploration targets;
  • Utilized machine learning technology for Regional Geochemical Sample data;
  • Refined the QUEST-South geochemical dataset; and
  • Raised awareness of machine learning applications in BC mineral exploration.

Project Benefits

This project used data collected over an area in South Central BC where interpretation of stream sediment data will benefit mineral exploration efforts.

The new geoscience information generated by this project may spark new mineral exploration activity in the region. If undiscovered mineral deposits do occur in this area, the information generated by this project will help the mineral exploration sector, communities, Indigenous groups and governments to make informed mineral exploration decisions.

Survey Area

The project analyzed previously collected samples from across South Central BC, including the Cariboo and Thompson-Nicola Regional Districts, an area that includes the communities of Williams Lake, 100 Mile House, Cache Creek, Kamloops, Merritt, and operating mines such as Highland Valley Copper, Copper Mountain and New Afton.

What Was Found?

The researchers used information from geological maps, stream catchment data, and mineral occurrence data from the British Columbia Geological Survey’s MINFILE mineral deposit database to ‘train’ machine learning algorithms to make interpretations about the geochemical data.

The machine learning algorithms categorized the geochemical data to account for the influence of the rocks the stream encountered in the catchment, and the possible effects this had on the stream sediment sample.

The predictions that came from this complex process were used to produce new mineral deposit probability maps that show areas where specific mineral deposits are most likely to occur.

Deliverables