Qubec: Mineral Potential Maps

Five years ago Gologie Qubec began to create a new tool for geologists, using existing databases. It would require no expensive fieldwork, but instead would apply the instincts of an experienced geologist and the power of computer systems, to predict the locations of mineralization. Daniel Lamothe, project manager in charge of mineral potential evaluation, began work on the Systme de production des cartes de potentiel minral (SPCPM), which today employs one full-time geologist and several associate advisors.

“The principle is that you can analyze the potential of an area if you consider many parameters in the context of a metallogenic model,” says Lamothe, who has 20 years of field mapping experience and admits to liking computers.

The stumbling block in such a project is the slow, expensive assembling of databases, but this is exactly what Qubec’s Sigom mining information system has been doing for the last 10 years. It now has $5 billion worth of information spanning a century available via Internet, and more being added each year. “For the last 20 years Gologie Qubec geologists were enrolled to use Geofiche [computer entries] to describe outcrops on field projects,” says Lamothe. “That’s why we have such large field databases.” He adds: “Many companies use Geofiche now.”

Two metallogenic models were eventually built during SPCPM development: volcanogenic massive sulphide (VMS) and iron oxide-copper-rare earth elements (Olympic Dam-type). It was easiest to start using the VMS model because of the wealth of literature from around the world, and the Abitibi was a natural location, with so many well-known deposits. Lamothe and metallogenist Claude Dion created the first maps in 2000 at 1:250,000 scale.

They used an expert system in a process called ‘overlay superposition’. In this knowledge-driven approach, the evaluator arbitrarily chooses the most important attributes (‘parameters’) of a location (such as rock type, or lineaments in Landsat images or magnetic vertical gradient maps, etc.). Each parameter, therefore, forms an overlay, and every pixel shows the amount of it in a definite portion of the study area. The mapmaker then assigns each layer an arbitrary percentage of emphasis (weight). The overlays are combined to make a mineral potential map.

The value for each overlay can be adjusted until the resulting map comes as close as possible to showing the most likely locations to look for a certain type of deposit. Once the best values are determined for the best parameters, that information can be used to make maps of other areas, and the results should be comparable.

A year later Lamothe began to build an Olympic Dam model for the Lower North Shore region, again using overlay superposition. The four most important parameters he chose were:

nproximity to a heat source;

nproximity to fractures and breaks;

nevidence of potassium, sodium or calcium alteration

nlocation in a certain rock type or proximity to a contact.

In 2004, the system migrated from an in-house expert system platform to the ArcGIS mapping system, so the ArcSDM mineral potential evaluation module created jointly by the GSC and USGS could be used to produce future maps. Lamothe says his group will build a customer-friendly VMS model in about two years. Assuming they have ArcGIS 9, users will be able to create customized potential maps by plugging their own data into a functional flowchart model.

Lamothe and his team are currently working on a project using more scientific methods. Instead of arbitrarily assigning a weight to a parameter, they are looking at parameters near actual deposits and statistically calculating their spatial association. “The project is based on fuzzy logic integration of grids, where the value of each pixel is objectively assessed by a ‘weight-of-evidence’ approach,” says Lamothe. “That kind of treatment is more rigorous and leaves less to the personal biases of the evaluator.”

The team is using the new procedure to make a series of VMS deposit potential maps at 50-m resolution (1:50,000 scale) covering the whole Abitibi region in Qubec. “In many cases we have been surprised by the results,” says Lamothe. “For example, hydrothermal alteration rates a much higher weight than we had originally assumed.”

Gologie Qubec aims to have all 119 Abitibi maps covered by the project completed by March 2005.

Feox-CuREE potential in the Lac Manitou area, Grenville province, Qubec North Shore

Favorability assessment for Olympic Dam/ Kiruna-type of deposits in Qubec’s Eastern Grenville Province. In the Lac Manitou area, the study brought to light two main zones. The northern zone surrounds a group of occurrences forming the Kwyjibo deposit, a Mesoproterozoic Cu-REE-Mo-F-U-Au iron-oxide mineralization 120 km from Sept-les.