Step Dancing at Snap Lake

When it opens on Oct. 1, 2007, Snap Lake will be De Beers’ first producing mine in Canada, and in fact its first mine outside Africa.

The mine, 220 km northeast of Yellowknife, NWT, just south of the treeline, is not large. At full production, which is expected to start in 2008, the ramp-access underground mine and plant will average 3,150 tonnes (t) of ore per day, yielding 3,800 carats (ct) of diamonds, about a thermal mug full (1.4 million ct/yr). But that’s a pretty rich beverage: the diamonds are most recently conservatively valued at US$144 (Cdn$172) per carat, making the daily product worth almost $650,000.

Snap Lake will employ 500 people (300 De Beers employees plus 200 contractors for hospitality and site services), mostly on two-week rotations, with 250-300 on site at a time. CMJ visited the site in mid-August.

Shortly after De Beers acquired the project in 2000, it was anticipating the mine starting up in 2004 with capital costs of $490 million. That was optimistic. The crucial water permit was not granted until May 2004, and De Beers’ board withheld the go-ahead decision until the orebody was better defined. Its complex shape drove up the mine development costs. When the board finally gave its approval in May 2005, a late 2007 startup was set, and a capital cost of $636 million. The timing remains on target, but the cost has since ballooned to $975 million.

The mine has indicated resources of 1.4 million tonnes grading 1.2 ct/t (recoverable), within 300 m of surface. In addition there is an inferred resource of 25 million tonnes grading 1.2 ct/t. That will support a 20-year life, which is just the beginning, according to De Beers Canada‘s vice-president for the NWT, Chantal Lavoie. “There is such potential to increase reserves. We know it’s open at depth and open on the north side. Some think there will be a 30-40-year mine life.”

The mine plan and processing flowsheet have not changed substantially since the November 2004 optimization plan (summarized in an article in CMJ December 2004). What has improved is the knowledge of the orebody.

The 520-million-year-old kimberlite dike outcrops and dips shallowly beneath Snap Lake. Its sublacustrine location means that the ore is not frozen, and that the mine is prone to being wet. The mine plan includes crown pillars to support the back within 100 m of surface. About 7,600 m3/day of water are currently pumped from the mine to surface. This water will be used in the processing plant or treated and released to Snap Lake.

Two phases of test mining (in 2001 and 2004-05), 20,000 m of diamond drilling from surface and underground, and borehole radar surveying are helping the geologists to determine that the dike is not a consistent slab, but really a series of steps, each sloping horizontal to 10 NE, broken up by cross-cutting diabase dikes. The kimberlite averages 2.7 m thick, but ranges from 1-10 m, and appears to have been all emplaced at one time. The horizontal boreholes, spaced 70 m apart, were also used to test for water, faults and bad ground. The radar system had been used by De Beers in South Africa, but the innovation was to move it into drillholes.

According to mine manager Gil Lawson, the steps “actually make mining easier. A 15 continuous angle would be the most problematic angle for mining. Here we can lay out a room and pillar mining plan, followed by paste backfill.” The plan is to lay out five mining blocks of 250,000 tonnes each before mining can begin in October 2007.

Warm weather woes

Construction began at Snap Lake in July 2005. The bulk of supplies for the 2006 construction work were readied to be brought in along the winter road from Tibbitt to Contwoyto (which supplies many operations in the area). But the unusually warm weather last winter resulted in the road lasting two weeks less than its usual eight weeks, and not becoming as thick. Only 1,600 of the scheduled 2,200 loads of supplies could be trucked to Snap Lake; 500 tonnes of the excess material has had to be airlifted to site at twice the cost.

The transportation shortfall had some negative effects on construction. Since the temporary accommodation modules could not be delivered, modules dedicated to other purposes were converted into accommodations, and some accommodations were stick-built–a slower, more costly procedure. This delayed the arrival of more workers and thus the ramping up of operations.

The 52-tonne scrubber (similar to a SAG mill) for the plant could not be brought in because the road was not thick enough. If it can’t come in on the 2007 winter road, it will have to be in half for transport, and reassembled onsite. The concentrator building now has a removable side panel so the scrubber can be lifted in when it arrives.

The transportation and labour problems coupled with increased prices for fuel and construction materials have caused the major overrun on capital costs. The project is still on time for October 2007 startup, but some things have had to be delayed till next year. The silver lining was that the warm weather allowed the concrete work to begin earlier than planned.

Mining and processing

The area that will be mined over the first 15 years encompasses 1 km2. A single portal provides access via a ramp sloping 15, plus there will be three ventilation raises. The ore will be conveyed from underground via a second decline to a coarse ore bin on surface. This will feed the secondary crusher, the product of which will be conveyed into the plant.

In the plant, kimberlite will be crushed, screened, conveyed, pumped, cycloned and sent through dense media separation. The dense mineral concentrate will then enter the “black box recovery area”, a De Beers proprietary diamond recovery system that includes X-ray sorting. This was designed and is being built in South Africa by DRA; it will be reassembled on site by Cosira.

AMEC has done the overall design management for the project. Most of the structural steel was supplied by Permasteel. KeTe Whii Ledcor is the main builder. The construction job will consume 3,600 tonnes of structural steel, more than 10,000 m3 of concrete and 26 km of piping. It is critical that the utility and processing buildings and the utilidor are up, clad and heated by the end of October, so the indoor work can take place over the winter.

The Snap Lake project is already registered as ISO 140001 for environmental management (unusual for a construction site) and in substantial compliance with OHSAS 18001 for health and safety issues. De Beers has concluded impact benefit agreements with three of the four aboriginal groups impacted by the project–the Yellowknives Dene First Nation, the Tlicho Government and the North Slave Mtis Alliance–and is negotiating with the Lutsel K’e Dene First Nation. In 2005 Snap Lake averaged 41% NWT resident employment (exceeding its 40% target), and 17% aboriginal employment.

Will Snap Lake be a success for De Beers? If the ore contains 1.2 ct/t, and the diamonds are valued at Cdn$172/ct, then the ore is worth $206/tonne. The estimated operating costs are $156/tonne, so that leaves an operating profit of $50 per tonne, or $57.5 million per year. The operating profits over a 20-year mine life would be $1,035 million, slightly above the capital cost, not taking into account the acquisition costs. It would appear the mine has to extend its life beyond 20 years to move into the black.

But there’s another consideration.

De Beers has an advanced exploration project at Gahcho Ku, only 80 km away, which will share administration with Snap Lake from De Beers’ Yellowknife office. Perhaps more importantly, the Gahcho Ku project will benefit from the lessons that Snap Lake is teaching the company about permitting, systems and how to work in Canada’s north.

Gahcho Ku
tackles its own problems

If everything goes according to plan, De Beers will open its third Canadian mine late in 2012–Gahcho Ku, 80 km southeast of Snap Lake in the Northwest Territories. Formerly called “Kennady Lake”, the project lies on the AK claims held by a joint venture of De Beers Canada (51% and operator), Mountain Province Diamonds (44.1%) and Camphor Ventures (4.9%). De Beers can increase its interest to 60%.

The three diamond-bearing kimberlite pipes in the mine plan are nearby each other–5034, Hearne and Tuzo. These reflect root to diatreme transition zones of pipes, and resemble those in the Kimberley area of South Africa.

The current indicated resource in the three pipes is 14.4 million tonnes grading 1.64 ct/t (about 23.6 million ct of diamonds), with inferred resources of 17 million t grading 1.35 ct/t (about 22.9 million ct).

A June 2005 study recommends a series of three pits producing ore at a rate of 5,700 t/d (2.1 t/yr). Containment dikes would hold back the waters of the temporarily lowered Kennady Lake during mining. Onsite processing would produce 3 million ct/yr of diamonds over a 15-year mine life. Some of the processed kimberlite and waste rock would be permanently stored in the pits. The capital cost would be $825 million (unescalated), and the operating cost would be $65/t. Engineering studies are attempting to reduce the capital and operating costs.

In November 2005 De Beers applied to the Mackenzie Valley Land and Water boards for permits, and the application was referred the next month to the Mackenzie Valley Environmental Impact ReviewBoard. In March and April 2006, the MVEIRB conducted technical and community issues scoping workshops and hearings as part of the assessment of the project.

The application was ordered to Environmental Impact Review by the MVEIRB in June, on the basis that the project was likely to be a cause for significant public concern. The company responded by asking for a judicial review of the order. “Our understanding of the Mackenzie Valley Management Resource Act is that an Environmental Assessment and an Environmental Impact Review are sequential steps,” said Cathie Bolstad, De Beers Canada’s manager public & corporate affairs for NWT projects. “In asking for this review, we are seeking clarity from the court on the permitting process. Given the Gahcho Ku project is similar to Snap Lake and falls under the same legislation, a thorough Environmental Assessment would be consistent with the permitting process for Snap Lake. It’s important to have certainty that permits issued for the Gahcho Ku project are issued on robust legal basis.”

In July, Mountain Province announced the results of a valuation on Gahcho Ku diamonds (1.47-mm cutoff) by WWW International Diamond Consultants. The 2,600 ct from 5034 were valued at US$101/ct; the 2,508 ct from Hearne were valued at US$54/ct; and the 529 ct from Tuzo (too small a sample to be representative) came in at US$43/ct. The samples included a 9.9-ct diamond valued above US$7,800/ct and a 5-ct stone valued at more than US$4,600/ct.

De Beers is currently involved in a $38.5-million advanced evaluation and permitting program (July 2005 to December 2006), with project management and EPCM provided by AMEC. Thirty-one holes have been drilled to delineate the 5034 North Lobe and Tuzo pipes and to carry out geotechnical drilling for pit design.

Samples are being analyzed at the Saskatechewan Research Council Geoanalytical Lab, and results for 5034 North Lobe will be available before year end.

---