EXPLORATION TECHNOLOGY
Minex with AI
New technology such as applying AI techniques are having a big effect on exploration strategies of both juniors and majors . Plus while drilling remains the backbone of mineral exploration , new approaches including better use of national scale work from datasets to existing core analysis can also save time and money reports Paul Moore
Artificial Intelligence ( AI ) in mineral exploration
is actually something that has been around for some time but is now really gaining traction thanks to the application of the latest technology and computing power , plus there are a host of startups that are now focussing on it , meaning juniors are now looking at AI as one of their first main strategies to firm up projects as opposed to just an add on .
Windfall Geotek based in Brossard , Quebec is a mining services company and a leader in the use of AI and advanced knowledge-extraction techniques since 2005 in the mining sector . Its system is called CARDS ( Computer Aided Resource Detection System ) and by implementing this technology , it says near-immediate savings of 3-5 % on exploration and development are attainable . Pattern recognition algorithms identify predictive data sets . It is based on the MCubiX-KE ( Knowledge Extraction ) data mining engine . Analysis of historic data can include : n Geophysical surveys : MAG , EM , IP , Gravity ,
Radiometry , etc n Chemical Assay Results : Drill Holes & Surface
Samples ( Grab , Chip & Channel ) n Geochemical Surveys : Lake Bottom Sediment ,
Stream Sediment , Soil & Till n Geology : Rock Type , Faults , Lineaments ,
Alterations , etc . n Topography : Digital Elevation Models ( DEM , DTM
& SRTM ) n Satellite imagery : Lineaments / Stress , Alterations , Radiance , Geobotany , etc A breakdown into two databases allows for areawide comparative analysis based on both known and unknown characteristics . “ Unlike previous rulebased analytics , CARDS is unbiased by any particular geologic model ; learns and adapts and can make predictions on any mineral deposit type represented in the data set .”
The result the company says is an unprecedented and rapid ability to identify underground targets with
high accuracy including all forms of mineral deposits .
All data is entered into CARDS as geo-referenced data points . Each point in the database is linked to its own set of characteristics ( variables ) extracted from a variety of sources . For example : n Proximity to mineral occurrences / mineralised drill holes n Geophysical surveys : MAG , EM , IP , gravity , radiometry n Geochemical surveys : rock , soil , lake bottom , drill hole assays n Satellite imagery n Geological maps : rock type , alteration n Digital elevation models n Proximity to lithological contacts / specific intrusive suites n Proximity to interpreted lineaments / mapped faults and shear zones CARDS identifies the positive points ( drill holes
5410500 5410000 5409500 5409000 5408500 5408000
5407500
471000
471000
E
471500
471500
E E
472000
042A14
E
472000
E
3.6 km
E!(!(!(!(!( !(!(!(!(!(!(
472500
E
Open
472500
E
473000
CRAWFORDE
!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(
E
CRAW18-02
!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(
E
CRAW18-03
473000
E
E
GoldSpot Discoveries combines proprietary technology with traditional domain expertise , offering a front-to-back service solution to its partners
and MDIs ) according to established thresholds for each of the mineralisations sought . “ By using a moving window , neighbouring patterns around each point are captured and expressed by new calculated variables for each primary exploration layer . In the analysis of each point in the database , the characteristics of all points within a specified distance ( neighbourhood ) are weighted into the evaluation of that point . Therefore the combination of their limited characteristics and their proximity to points with other significant characteristics similar to that of known positive points is identified .”
Windfall Geotek recently announced that it has been commissioned by Capella Minerals Ltd to deploy CARDS . The contract will identify VMS ( Cu- Zn ) exploration targets over Capella ’ s Kjøli project
Windfall Geotek ' s CARDS technology has also been successfully used to identify targets for Canada Nickel Company ' s world class Crawford nickel-PGM project in Ontario
473500
E
!(
CRAW18-01
E CRAW18-04
473500
E
474000
E
CARDS Main Nickel AI Target at 80 % Similarity
500 0 500 Meters
NAD 1983 UTM Zone 17N 474000
E
474500
E
474500
E
E
E
E
475000
475000
!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!(!( E
Legend
Crawford _ March 2019 _ DDH _ Projected _ Assays LUCAS Ni (%)
E E
0.391 - 0.669 0.293 - 0.388 0.193 - 0.290 0.05 - 0.190
E
475500
Crawford _ March 2019 _ DDH _ Collars
DDH Historical Collars
!( Ni Positive Training Points ( 0.2 % Ni )
!( DDH all Historical Projected Assays
Nickel CARDS Ageo Prediction Model
Similarity (%) 100
80
475500
5410500 5410000 5409500 5409000 5408500 5408000 5407500
10 International Mining | FEBRUARY 2021