RE: RE: RE: RE: RE: RE: RE: Leaky BoatYes nugget effect tends to be problematic and normally under estimates mill grade by significant margins. This is a well documented characterisic of Abitibi Greenstone Belt deposits. Gold Bullion has been hitting 1 g/t gold using nq drilling when mill grade is known to be 1.72 g/t they have ordered RC rigs to sample a larger cross section to get a better handle on grade.
Below is part of a technical report that documents this problem. Part of NGM ni 43-101. There is a table which shows the variation that occurs between drill hole diameter (it would not paste so it is not included.
From his review of the Dome drilling Rogers further noted that 40 to 60% of holes completed
through multi-vein gold structures and 50 to 80% of holes completed through single vein structures
areas failed to return any gold values in excess of 1.7 grams per tonne in areas that were ultimately
mined. The majority of this drilling was short, close-space holes, drilled from levels and sublevels
to evaluate ore blocks that were planned to be stoped.
At the time of Rogers (1982) analysis, the Dome Mine had produced 42.7 million tons of ore and
11.1 million ounces of gold (Atkinson 1985), which is equivalent to 38.3 million tonnes of in-situ
resource grading 10.3 grams gold per tonne (based on 10% mining dilution, 90% mine recovery,
and 95% mill recovery).
Also, if one reviews the annual production and reserve information for the Dome Mine over its long
history, rarely did the proven reserve base ever exceed 3 years of future production until the 1990’s
when the bulk mining open pit was developed.
Clearly, based on the above analysis, in the evaluation of the geological potential of a
predevelopment-stage gold property such as the Property that is the focus of this Report, emphasis
must be placed upon the tonnage potential of the vein structures, and a review of the overall hit and
miss ratio of the gold intersections obtained during the exploration stage drill program.
Northern Gold Mining Inc., Garrison Gold Property, October 2009 Page 25
None of the great gold mines of the Timmins camp would ever have achieved production if a
requirement of financing of the initial development had been to demonstrate 7 to 8 years of proven
plus probable mineral reserves as defined in NI 43-101.
In the past 4 years, as a result of corporate needs, there have been NI 43-101 reports prepared (a) for
the Dome, Hoyle Pond, and Pamour mines in the Timmins camp, Ontario (Rocque et al 2006,
Couture 2003), (b) for the Campbell and Red Lake Mines in the Red Lake camp, Ontario (Crick et
al, 2006), and (c) the Musselwhite Mine in the Pickle Crow area, Ontario (Mah 2006). The
aforementioned reports provide significant information relating to grade estimation issues in typical
Archean vein-type gold deposits in Ontario. The information is very relevant to the issues relating
to sampling this type of gold deposits by drilling.
All of these operations have sampling issues relating to reconciliation of grades indicated by all
manner of sampling (drill core, chip samples, muck samples from ore cars and trucks, and belt
samples from various points in the mill). Most of these operations treat each vein-type or stoping
area as individual projects and use sophisticated geostatistical software to determine assay indicated
grades, capping grades, etc in order to reconcile resource and reserve estimates with mill
production.
Over the past several decades, geostatisticians (Pitard, 1993a,b, 1998, 2002, Ingamells and Pitard,
1986) have published extensively on gold sampling theory and all recognize that the primary and
most problematic issue is “Nugget Effect”.
Typically, in Archean vein-type gold deposits, gold is very rarely uniformly distributed throughout
the vein structure but rather, occurs as clusters of small particles or single masses of spectacular
“nuggety” gold. This random, unpredictable distribution of gold influences all sampling of gold
mineralization, whether it be (1) exploration drill core, (2) close-spaced underground stope planning
drill core, (3) channel sampling of drift faces, development raises and sublevels, (4) sampling of
mined ore by underground car or truck sampling or belt sampling after primary crushing in the mill.
This problematic sampling issue is what is known as the “Nugget Effect”.
These same statisticians also point out that once the sample is acquired and sent to an assay
laboratory where it is crushed and then subdivided into a smaller sub-sample for fine grinding from
which is ultimately take a smaller sample that is submitted to the assay laboratory (commonly 30
grams of material, know as one-assay-tonne). The whole process, which has been industry
standard for decades, has a high risk that the final one-assay-tonne sample will not be representative
of the material originally sampled.
When one considers (a) the small number of grams of gold that are required to produce economic
grades in a tonne of ore, then considers (b) the small volume of that small amount of gold,
compared to the volume of one tonne of ore and (c) further considers the volume of core in a single
diamond drill hole passing through that tonne of ore, the issues implicit in sampling with drilling for
grade becomes apparent. Table 3 provides a simple summary of the above facts.
Northern Gold Mining Inc., Garrison Gold Property, October 2009 Page 26