Hudbay found 60% more gold at Snow Lake without discovering a new mine

The new Snow Lake mine plan in Canada raises expected gold production from 1.8 to 2.8 million ounces and extends mine life to 2043 — combining fuller mills, higher metallurgical recovery and conversion of resources into reserves.

Canada's Hudbay Minerals released on September 28, 2026 a revised mine plan for the Snow Lake complex in Manitoba. Expected gold production over the life of mine rose to 2.8 million ounces, 60% above the 1.8 million estimated in the 2021 technical report. For 2026 to 2030, projected average output is 185,000 ounces a year, at an average cash cost of US$821 per ounce. Proven and probable reserves reached 27 million tonnes containing about 2 million ounces of gold, and reserve life was extended to 2043. The plan calls for taking the New Britannia mill to 2,300 tonnes per day from 2027 (original design 1,500 t/d, permitted for 2,500 t/d) and the Stall mill to 3,000 t/d by 2030, using idle capacity at a plant sized for 3,800 t/d. The Lalor mine remains the main source until 2037, with the 1901 deposit reaching full production at the end of 2027. The company also pointed to additional potential at the Britannia project, with inferred resources of 8.26 million tonnes at 2.87 g/t, and at a former tailings area, Anderson, estimated at 20 to 30 million tonnes grading about 0.8 to 1.0 g/t gold.

The Snow Lake case is a lesson in where value lies in an operation that already exists. None of the levers in the new plan depends on a spectacular discovery. They are in places any process or planning engineer knows — and that many companies leave for later. The first is idle capacity. The Stall mill was sized for 3,800 tonnes per day and runs at close to 2,300. New Britannia was designed for 1,500 and is permitted for 2,500. Filling a plant that is already paid for costs far less than building another one, and every extra tonne dilutes fixed costs. The second is metallurgical recovery. At Stall, the fraction of gold recovered went from 58% in 2022 to more than 70% today. Twelve percentage points seem small until they are applied over years of production: this is gold that was already being mined, crushed and ground, and that previously went to tailings. The next step, a hot tailings leach project planned for 2028, follows the same logic. The third is discipline in converting resources into reserves. A resource is mineralization with geological evidence; a reserve is the part that pays for itself under the mine plan and the prices adopted. Hudbay cites a historic conversion rate of around 90% at Snow Lake, which lends credibility to the 21 million tonnes of inferred resources still outside the plan. There is also a lesson about assumptions. Reserves use a long-term gold price of US$3,600 per ounce, below 2026 prices. That protects the plan against a drop in the metal, and shows that the increase does not depend on an extraordinary price. Anyone reading a technical report should always look for this number before being impressed by the projected production.

The gold bull cycle has created pressure for growth in the sector, and the most common path to it has been expensive mergers and acquisitions. Snow Lake shows an alternative: growing by optimizing existing assets, with less risk and capital. For managers, the case reinforces the value of periodically revisiting the mine plan with updated operating data, instead of treating it as a static document between technical reports. For process engineers, the recovery at Stall is a reminder that small percentage gains at the mill have a disproportionate effect on reserves and mine life. And for those working at older Brazilian mines, where there are mills with spare capacity and tailings dams or stockpiles with residual grade, the question Snow Lake leaves is simple: how much value is already inside the operation's perimeter waiting to be planned?