Barkerville in the doghouse with BCSC

The British Columbia Securities Commission (BCSC) has halted trading in Barkerville Gold Mines (BGM-V) because a much-anticipated technical report on its Cow Mountain gold deposit is not up to standard.

The company filed the National Instrument 43-101 technical report on Aug. 13, just in time to meet the 45-day filing requirement to back up the press release it issued on June 29, which stated that its Cow Mountain deposit in central B.C. hosts 10.6 million indicated oz. gold at 0.154 oz. per short ton.

The BCSC, which waited almost two weeks after the initial press release before forcing Barkerville to caution investors on the reliability of the data, came down far more swiftly this time. The commission halted trading on Barkerville’s stock before the market opened on Aug. 14, stating that “the report was not prepared in the required form,” and that trading would remain halted until the company files a compliant technical report.

The technical report establishes a somewhat different resource estimate than was outlined in the press release, splitting what was an entirely indicated resource into both indicated and inferred. Cow Mountain is now estimated to contain 57 million indicated tons grading 0.116 oz. gold per ton for 6.6 million oz. gold, and 30.3 million inferred tons at 0.189 oz. gold per ton for 5.7 million oz. gold. The report, unlike the previous release, also shows a capped resource of 57 million indicated tons at 0.075 oz. gold per ton for 4.3 million oz. gold, and 30.1 million inferred tons at 0.107 oz. gold per ton for 3.2 million oz. gold.
The BCSC released no further details as to how exactly the report, prepared by geologist Peter T. George, was non-compliant, but others have raised their own concerns about its contents.

Garth Kirkham, owner of Burnaby, B.C.-based Kirkham Geosystems Ltd. and chairman of the Best Practices Committee at the Canadian Institute of Mining, Metallurgy and Petroleum (CIM), said in a phone interview that the report misses on a number of criteria, and overall is thin on details.

“First and foremost, it does not comply with CIM best practices, which, when you’re doing a 43-101, is a requirement,” Kirkham said, noting that the methods, the validation and verification of data and a whole checklist of things don’t comply.

For Kirkham though, the biggest concern is the sample data on which the resource was based. He sifted through the lengthy sample database in the report that apparently backed up the resource, and found that while the database shows about 125,000 foot-long samples, only 22,355 samples that were plugged into the Amine software by George had positive values  and appear to be used to calculate the resource.

The hundred thousand or so samples that the author appears to have not incorporated into the software show a grade of -1 in the database, or if they were brought in then they appear to be ignored. Ordinarily a -1 means a lack of information on the sample, which often happens when a field geologist does not send the core to be assayed because it doesn’t appear to be mineralized. But that does not mean the -1s don’t need to be factored in.

“He appears to be ignoring the negative ones, so he’s only giving any weight or credence to the assay data that is zero or above,” Kirkham said. “It also looks like he’s smearing the positive data into where there’s negative data.”

“It’s not just an error, it could be considered selective reporting,” Kirkham continued. “You can’t just say all those negative ones, all the stuff that we didn’t sample, actually is mineralized. If that’s the case, then you better substantiate that. And he doesn’t.”
Besides discounting the likely presence of extensive barren rock in the deposit, by using only 22,355 samples, George has established a multi-million ounce indicated gold resource using only 22,355 feet, or 6,814 metres, of core sample data.

And those 22,355 samples are potentially even less representative, with Kirkham noting that George has split up assay intervals from their original lengths into the one-foot intervals, basically reverse compositing them. The result, for example, is that a high-grade, six-foot interval grading 23.95 oz. gold is split into six one-foot intervals in a row grading the same, a three-foot interval carrying 25.03 oz. gold is split into three one-foot intervals grading the same and a two-foot 56.64 oz. gold interval is split into two one-foot intervals grading the same. The splits boost the overall sample size number, thereby possibly skirting around requirements of a minimum of four composites used to interpolate into a block, and they could generally skew the results.

“He’s actually taking intervals of four, five and six feet and making them all a ‘one.’ This has the effect of artificially creating data,” Kirkham said. The report also states that a minimum of four composites should be used for interpolating into a block. By splitting up the samples, in reality only one or two are used.

Kirkham also raised concerns over how George came to his overall grades of 0.116 oz. per ton indicated and 0.189 oz. per ton inferred, noting that the average of the reported grades don’t add up.

“If you take all of that composite data, and put it into Excel and not even include the negative ones, and just do an average, I get 0.085 oz. per ton. I cannot see how you can get a higher value in the estimate than your data, especially considering the immense amount of waste that you are diluting it with,” Kirkham said. He thinks George may have given higher-grade samples a higher weighting, or allowed them to influence the global model in an unconstrained manner, but as with elsewhere in the report, information about methodology is lacking.

As to any measures the CIM might take, Kirkham said the institute is there to set the standards, not to enforce them, but that there are avenues for investigation and discipline if needed.

Meanwhile, geologist Brent Cook of Exploration Insights newsletter — who was one of the few in the industry to publicly raise concerns over the initial resource estimate press release — has also found red flags in the actual report.

“What I’m seeing, what I think he’s doing, is essentially an unconstrained estimate,” Cook said. “Without constraining the high grade, you’re kind of pushing the grade out into areas where either there is no data, or the data’s been ignored.”

Cook cautioned that he has yet to analyze the report in detail, choosing to wait for his weekly newsletter to provide more commentary on the matter, and generally taking a reserved tone.

“I’m not criticizing what he’s done yet, these are just red flags to me that need to be explained,” Cook said.

For Cook, those red flags include the fact that bench number 3,950 in the report contains almost 50% of the overall resource, yet by applying the grade cap, about 70% of those ounces vanish. Just as startling, if the cap is applied to the total resource, 37% of the gold goes away.

“That is odd and suggests grade is being smeared all over the place, and I’m not exactly sure how that happens,” Cook said.
He also pointed to the sensitivity of that bench, which drops down to 895,000 oz. gold indicated and 837,000 oz. gold inferred once the resource is capped.

Overall, Cook said that, like Kirkham, he found the report was missing important details that made it tough to go back and reconstruct what’s been done.

Finally, there is the issue of all that geological potential, upwards of 90 million oz. gold along a 6.4-km trend, according to the author.

“Geological potential” is not a CIM definition, but it is technically allowed under NI 43-101. However, Kirkham no
tes that the guidance given by the securities regulators is that it be “reasonable and realistic, and defensible.” With world gold production at 88 million oz. last year and Canada at 3.5 million oz., Kirkham’s comment on the potential was that “the author of this report is stating that there may be a complete year of global gold output, or up to twenty-five years of total Canadian gold output, within an area roughly four times the size of Stanley Park. In my opinion, this is not reasonable, nor is it realistic, nor is it defensible.”
Trading in Barkerville’s shares will remain halted by the BCSC until more answers are forthcoming. In the weeks leading up to the filing date, the company’s stock price climbed from 70¢ to close at $1.22 on Aug. 13, where it will stay until the company complies with BCSC requirements.

Frank Callaghan, president and CEO of Barkerville, has not responded to multiple requests for comment.