Tuesday, April 24, 2018

Pretium - an exercise for you

This is just a quick response to a comment by Tom. I originally created this Leapfrog viewer file for a previous post on Brucejack, focusing on a small area of the mine that I have put into Leapfrog.

Original post here - link

Download the Leapfrog viewer file from here (link)

The key slides are:

Slide 2 - The PR grades (i.e. the smeared grades) - nice juicy think gold intercepts

Gold fecking everywhere!

Slide 5 - Split grades (i.e. high-grade intervals removed and background grades back calculated) - now more of a needle in a haystack gold hits.

Oh no, its all gone!
Your job is to join the pink bits. Have fun!







54 comments:

  1. Geezus. Too amazing how a picture tells a zillion words. Take another bow AG. Da power of grade smearing is an amazing thang.

    ReplyDelete
  2. Exactly the point I have tried to make over the past five years. To my fellow mining engineers, try to put stope boundaries around that dogs breakfast and mine it at 2900 tpd.

    ReplyDelete
    Replies
    1. So AG you averaged the high grades into the background?? Ok but why for, when it makes a diagram that is not actually useful? The high grade intercepts were very real, left there by nature, as the key economic component of the deposit. The correct thing to do is to leave it all as is, and use MIK kriging, as was done. The mine is having problems but it seems to be an outstanding deposit, easily accessible (without grounwater inflow or rock quality problems, etc), much of it suitable for low cost mining methods. 3800 tpd will lower the cut off grade and relax some economic constraints. It appears to be a lightning rod for attracting criticism, but it's probably going to work out well for the shareholders in the end...

      Argh, someday, maybe soon, I'll start my own blog...

      Paths/midpath

      Delete
    2. Hello Paths,

      the first figure is the PR intercepts where the high-grade intercepts have been combined with the surrounding rock to produce a think high-grade zone.

      The lower figure is where I've back calculated the residual grade of the rock surrounding the narrow high-grade intercepts.

      The idea is to see how much gold is the in the high-grade zones and how little gold is in the surrounding country rock.

      Delete
  3. I came to the same conclusion by looking at the composite statistics in the technical reports. Lots of low grade (below cut-off) surrounding occasional high grade pods. If you check the last discussion on Pretium I think I predicted a new and smaller reserve estimate in the near future as they come to terms with their reconciliation problem.

    ReplyDelete
    Replies
    1. yes, but for people who don't have experience in resource calculations, an image is a lot easier to understand

      Delete
    2. Agreed. It requires a reasonable understanding of statistics. For anyone who has the time and desire I would recommend a book titled "Practical Geostatistics" by Isobel Clark.

      Delete
  4. The problem is Bill, many small retail investors have been taken to the cleaners here by these geologists using their MIK models that no one can understand. And AMC, you guys should also be ashamed of yourselves.

    ReplyDelete
    Replies
    1. So take a good look at a photo of the new CEO Bill and ask yourself if this guy has spent much time underground and whether you will trust him with what he states with respect to tonnes/grade/unit costs/sustaining capital going forward.

      Delete
    2. Well Doug...few if any retail investors understand resource estimation regardless of how it's done. It's unfortunate because that chapter of a technical report is probably the most important by far. To me, MIK is like voodoo magic. Every deposit I've heard about or been involved with that has used it has over estimated grade. But this deposit doesn't lend itself well to any statistical method so where does that leave you as the resource geologist?

      From what I've seen the management team is taking the right actions for the most part. They seem to be focussed on getting a better handle on reconciliation through tighter drilling and more flexibility in their mine plan. I'm not sure they should be looking at expanding mine output until they have a good handle on the resource and reserve. No sense spending more capital on a mine that can't fill the mill. I've been there and done that.

      Delete
    3. I don't think it's appropriate to comment on management capability based on their photos or bios. The market will quickly figure out whether they can do what they say they intend to do. They clearly have some serious challenges ahead of them but it's important to point out that they have met their throughput and recovery targets after only a few months and they are taking steps to meet the grade targets. I have over 35 years of U/G experience and I would do a few things differently, but it's hard to opine on their actions without actually being there and knowing what they know.

      Delete
    4. "this deposit doesn't lend itself well to any statistical method"?? I would have to disagree with that. Irregular gold in an erratic stockwork. Geostatistics is ideal for that situation, and is the only technical approach they have that is reasonable. What was done in the report seemed reasonable, and was well explained.

      The variogram ranges applied to each grade range sounded reasonable, at least they were believable for the situation. One wrinkle is the data plots used to set the variograms were not shown in the report, making it impossible for a reader to verify them. However, trusting the report is not a difficult step as the variogram ranges made sense, and it came from a third party company.

      The actual results are calling that into question, but not in a huge way. Seems to have had more of an operating problem. The company trusted the geostatistics so completely they selected a mining method that was not as flexible, and they didn't set up grade control drilling until they were surprised by problems. In addition to the close spaced drilling, they likely will gradually introduce several changes to the mining method, and perhaps adopt the use of more than just one mining method for differing locations.

      It was more of a situation where they went too far too fast with a high productivity mining method, without paying enough attention to the hard constraints placed by the geology. They are learning the hard way, but as long as they do learn and adapt to better methods and procedures, even a better philosophy then positive change are ahead, I think more so than the market seems to realize.

      Your pal, hehe, Paths/midpath

      Delete
    5. In order for a population to be normally distributed the sample needs to be normally distributed. The basic stats for the assays and composites did not demonstrate that they were normally or even log normally distributed. That's the only evidence I can offer to justify my comment. I agree with the rest of your comments, particularly with respect to what I call definition drilling. Proper stope design based only on surface drilling is quite risky. Close in drilling is necessary to accurately determine stope boundaries. I agree they are doing the right things to get closer to reconciling actual production to their reserve model, and also to improving flexibility for future mining. What we don't know from public information is how much of their problem is due to the inability of statistics to give them an accurate model.

      Delete
    6. Paths, I see you are coming around to the idea that they did not have appropriate mining method for this deposit. Of course mining experience should help resolve the proper method, which is what I have been saying all along. But, I think you must know that there could be a huge cost difference between say long hole bulk stopes and re-suing shrinkage stopes at the other extreme? Or that the production tonnage would be vastly different as well?

      Delete
    7. Bill, they have already done a significant underground infill drill program but even that has proved insufficient especially as it is radial so the spacing will get wider for stopes further away from the drill station. They will now do infill drilling from closer as well as sample blast hole cuttings. That will help to some extent but long hole stoping will permit them to refine mining shapes only so much.

      Delete
    8. Definition drilling does two things. First, it gives a much more accurate spatial description of the potential stoping area. At Kidd we found that ore geometry could be off by as much as 10 meters based on radial drilling from a distance. Footwall drive location could be laid out, but not stopes. Stopes could only be properly laid out once the definition drilling was done from no more than 30 meters distance. Second, the definition drilling will provide a new model to compare against the ore reserve model. This will help determine the accuracy of the reserve model and may cause them to go back and re-do it with new assumptions.

      Delete
    9. Yes I should have better appreciated some of your earlier comments, however, I was referring to more modest changes. The longitudinal option that was suggested, with a serious attention to definition sounds appealing. Perhaps, the occasional use of sublevels to cut the stope height in half, but only if a very irregular location would justify it. At the edges of the ore body where thickness is less and it becomes more irregular, perhaps the use of mining from alimaks could be considered. Not the standard ones (as shown on youtube), but a large mechanism that could carry a serious drilling machine on a wider platform. I don't know, never been there, just thinking about being open to reasonable workable modifications that do not alter the big picture.

      Modifications to a mining method, and occasional inclusion of another mining method does not have to represent a large and costly departure from what is already in place. This mine does not have the bottleneck of hoisting everything through a shaft, so changes are less likely to be disruptive. Also once the mine is at 3800 tpd, there could be more flexibility as economic targets can be met without as much focus on what the grade was.

      They might look for ways to focus all employees on "grams produced, over tonnes of production". People already know that, but whatever they can do to encourage it would help. Maybe a bonus payment system based on how close they got to the stope design outlines, or how much dilution was reduced, how many grams were produced, together with safety, etc instead of other types of incentive payment systems based on advance or tonnes that might be considered.

      Paths/midpath

      Delete
    10. Longitudinal mining is very similar to transverse. It is used for narrower zones to minimize the cost of stope access as one drawpoint can serve three stopes. I hadn't considered alimak mining but yes, it should be considered, particularly for shoots with lots of vertical continuity. One of my best friends owns a company (Manroc) that pioneered that method at Hemlo. They mined stopes up to 150 meters high. He taught me how to drive track drift back in the '70's.

      Delete
    11. Thank you, that's interesting. He has a video on youtube: https://www.youtube.com/watch?v=oMD1mmrWbWc

      Delete
    12. One question could be this - we know that the mine is profitable at 9 g/t, but will the market be happy with that?

      Delete
    13. The mine may viable at 9 grams but the company may not be profitable. Using the 4th qtr as a guide...operating earnings were $27 million. From that you have to deduct $6 million in corporate costs and a further $15 million in interest costs. That leaves them with a tiny cash balance of about $6 million per quarter. They also have a rather large debt to pay back (about $600 million) with roughly half of it current. It brings to mind the plight of Thompson Creek. Not enough cash flow to pay interest and debt. Sadly, it's a common theme in mining. In this scenario (above) they can hope for higher gold price, which appears to be a reasonable expectation given all the debt flooding the world these days.

      Delete
  5. Still they cranked out over 30,000 ounces in March and will probably be OK as a higher tonnage operation. Not a real "gold mine" LOL but still profitable at $1300 gold. IMO CNL will be the real "gold mine"

    ReplyDelete
    Replies
    1. but if everything was working to plan, they would be producing over 40Koz/month

      Delete
    2. Now you're sounding like the banker that financed them!

      Delete
  6. One question for Bill and Doug is:
    What scale could this deposit be mined? Is the bulk approach the 'only' way as to define the individual shoots to any degree of certainty would be expensive?

    ReplyDelete
    Replies
    1. There isn't enough public information to be definitive on this, but Taylor's rule would give you a decent conservative estimate. Assuming their stated 18.9Mt of reserves would translate into about 3,900 mtpd. The big question is whether their reserve number is accurate. It's based on their M&I resource, which we know is not reconciling very well, which is why I suspect their reserves are over stated. In my view they should hold off on any expansion until they have a solid reserve number.

      Delete
    2. Forgot to reply to your second question on appropriate mining method. Bulk mining isn't the only way. Cut and fill is an alternative. Both methods are appropriate and each has its advantages and disadvantages. In general, bulk mining causes more external dilution and is less expensive, while C&F is the opposite.

      It appears that the mine is developing an ore control model, from which they can more accurately decide the appropriate mining method.

      Delete
    3. I probably didn't phrase the question properly - in this situation, where the Ultra high grade zones appear to be narrow, inconsistent, small lateral and vertical extents with a huge nugget effect, could you focus mining on these short, high grade inconsistent type veins, or are they so hard to model that you are pushed towards to less selective mining technique.

      i.e. pushing the operation to a bulk mining technique as you have to get all of the UHG zones to bring the grade up to a sufficiently high level to break even as the costs to UG development for selective mining on such small structures (they may only have a 10-20m vertical and horizontal extent) that are so inconsistent?

      Delete
    4. This comment has been removed by the author.

      Delete
    5. I'm reasonably certain they would have looked at all mining methods when they did their reserves and mine plan, and moved towards their current plan after carefully considering the economics of alternatives like cut and fill or shrink stoping. There are two important economic considerations....NPV and cash flow. (There are other considerations as well but these are the main economic ones.) They would likely have picked the method that gave the best NPV, given that it provided positive cash flow over the LOM.

      Another factor is the experience of the mine design engineer. If he/she has limited operating experience it will drive him/her to rely on technical training more than actual operating results. An experienced engineer with good operating experience might choose a different mine method because he/she would "know" the downsides of picking the more technically correct method. (I'm an operator so I would go for the solution that's worked for me in my past.)

      That said, they now have a heck of a lot more information and should use it to re-assess their initial decisions. There are always lots of assumptions made prior to start-up on things like grade distribution, dilution, recovery, etc. If these are re-examined in light of new information it could lead to a different mining method. It all comes back to the reasons why their reserves are not reconciling. Is it dilution or is it resource model inaccuracy or is it both?

      So what do we outsiders know about the mine so far? Well, we know they have a reconciliation problem. We know they've accelerated their development, which is necessary to improving stope readiness and also to provide definition drill access. We know they are now drilling on 3.75 meter centers vs the old 15 meter centers. Based on this information.....

      My uneducated guess is that they will eventually come back with a smaller reserve base and a combination of mining methods. They might continue with blasthole stoping but with tighter sub intervals and fully developed over and under cuts. (Allows better fragmentation and less dilution) If the 3.75 meter definition drilling provides an ore control model that is significantly different than the ore reserve model then they may need to re-do their reserves and mine plan.

      Delete
    6. Let me try this from another angle. Bulk mining isn't a default method. It works well when the deposit is well behaved, particularly the boundaries. At Hemlo we were mining on structure and the hanging wall was very regular both horizontally and vertically. So we could be pretty sure that our blasthole dilution was minimal. If the hanging or footwall was irregular it would create dilution problems. These could be countered to some degree by smaller stopes or less vertical stope height, but moving in those directions tends to minimize the benefits of mine productivity.

      So in the circumstances you outlined I think they should be considering more selective mining methods, like cut and fill, where you can maintain geological control more easily. Mechanized C&F can be almost as productive as longhole, but it has its own drawbacks. (Manpower are more exposed to ground conditions, among other things.) It's definitely more expensive, but would create less dilution and would probably reconcile a little bit better than longhole. I wouldn't ignore other methods like shrinkage but would only consider it in very narrow stoping areas where grade could offset the dramatic productivity drop. Very hard to find miners that can do that nowadays. Back in the day we were all trained how to do it.

      Delete
    7. Agree on fully developed over and under cuts, to allow rows of parallel blast holes instead of fan shaped rings. If the rock conditions make that difficult, then it could be from two drill drifts, one along the footwall, other along the hanging wall? I thought the drilling will now be on 7.5 m centres, I had not seen anything on 3.75m before? Paths/midpath

      Delete
    8. I thought I read 3.75 meters somewhere but upon re-reading their 1st qtr report it says 5 to 7.5 meters in the section on infill drilling. Yes, you can drill it on HW/FW drifts but it's seldom a problem to fully sill out because the stope dimensions are based on stress analysis and they would have to cable the overcut to prevent deterioration after fully blasting the stope. At one mine we used to survey all holes top and bottom and then design a redrill pattern if we thought fragmentation would be an issue.

      Delete
  7. Angrygeo, Appreciate the follow-up post! The Leapfrog does show just how jumbled this deposit appears to be. It's just not possible to get a proper handle on it from solely drill holes, even tightly spaced. There are good indicators, however, from the drilling that controls do exist.

    For example, note that high grades tend to appear at the edge of the low grade mineralized intercepts a majority of the time. That's quite significant especially since the low grades do have some continuity beyond very locally. So right there you already have one possibility to investigate (mine along the edge of low grade domains). You also have high grades sometimes appear within the lower grade zones but some of this could be due to an imprecise resolution, i.e. there are actually two adjacent or overlapping low grade zones (which could be why they are wider). In other cases these may be structures ("veins") that can be followed for short distances such as the Cleopatra vein. Also I would note the low grade zones do not appear to be linear but rather undulating/curved so that needs to be taken into account as well.

    There are also some isolated areas with only high grade intercepts that may be somehow connected. These would need tighter drill spacing to established a hit rate frequency that could then be reconciled with trial mining to see how much gold these areas actually contain. Until then they should be treated as inferred resources, if at all. The completely isolated high grade intercepts should only be considered as prospects unless nearby drilling repeatedly encounters high grades as well.

    I'm not saying this simplistic observation is a solution as there would be much more needed to determine a reliable reserves number for VOK. Especially if you are smart enough to realize MIK or other geostatistical methods based on drilling cannot alone provide a good handle on this deposit.

    ReplyDelete
    Replies
    1. Hello Tom,

      I'm working with summary drill-data where I have back-calculated the residual grade of the country by removing the influence of the narrow high-grade zones. However, I don't know how accurate this is, it may be that there is a low-grade halo around each vein, which could be used as an exploration tool, or within these low grade zones there is a second, 3-5 g/t vein that wasn't reported as a separate individual intercept. For this reason, I'm not comfortable stating that there is a low grade halo and left it open to 'interpretation'.

      Delete
    2. Hello Angrygeo, Thanks for the clarification, but I still assume they reported a grade for the intervals you back-calculated, and the actual location of the high grade intercepts is correct. If so, and I get the same thing using my model, there seems to be some predictability of the high grade IN RELATION TO the lower grade. I'm not sure the low grade is a "halo" as there are several stages of mineralization at VOK. What I'm interested in is spatial-grade distribution. And there does appear to be some continuity of the high grades, not as "veins" but rather as the rare permissive "zones" in which veinlet swarms tend to carry visible gold (vs. the abundant stockworks where gold is finely disseminated).

      Delete
    3. If there is continuity of high grade along the margins, and they mine it selectively, then the lower grade portions of those zones could likely fall below the economic cut off grade. This could result in a resource reduction.

      Delete
    4. Anonymous, yes...that was my conclusion also, based on the composite stats. That's why I've said that I anticipate a new resource and reserve statement in the near future.

      Delete
  8. http://globenewswire.ca/news-release/2018/05/02/1494725/0/en/Continental-Gold-Drills-High-Grade-Gold-Over-1-200-Vertical-Metres-at-Veta-Sur-and-Confirms-the-Laurel-System-as-a-Major-Discovery-at-Buriticá-Colombia.html

    ReplyDelete
  9. Bill,
    I agree. It's something I have seen numerous times.

    When you take erratic high grade assays and spread them around, you can create a lot of moderate grade material. If that moderate grade is above your economic cut off, you can create a lot of resource ounces that aren't there.

    They do appear to have a great deposit, and it probably has a long future in front of it, but if they can't achieve the reserve grade month by month, then there is a problem that needs to be addressed.

    ReplyDelete
  10. Cut and fill mining will not work here Bill, Tom and Angry simply because this sort of mining needs to be under geological control on a daily basis. Think of the mine geologists with their can of spray paint marking up the hangingwall and footwall daily on multiple headings for the jumbo miners to follow. This simply is not the case here. These guys have never been able to sort out any predominant trend to the mineralization. So the fall back is bulk mining which the ex-CEO acknowledged in multiple presentations. "Take it all", I seem to recall him stating. But the problem here is, if you look at some of the studies issued, AMC shows selective longitudinal blasthole stoping on the lower levels, not bulk tranverse stoping. But to use longitudinal it implies that they understand the mineralizing trend and not only that, it is vertical according to previous study isometrics. And oh by the way must do it at 2,900 tpd ++. This is a fail on both the daily tonnage rate and grade. You cannot have your cake and eat it too.

    ReplyDelete
    Replies
    1. Hello Doug,

      So these issues are going to be amplified if they expand the operations to 3800 tpd?

      Delete
    2. Grade goes even lower since they do not have any sort of short term response underground geology wise or stope boundary wise wrt to a mill starving for feed. I thought this one should settle down to 7 g/t but I now think 5 g/t with downside potential.

      Delete
    3. Doug, I'm not sure if you're saying they can't use cut and fill because they don't have enough geologists or because there's no structure to mark up. Can you clarify? I agree with your comments on grade dropping with rising tonnage. It's the effect of a mine superintendent keeping the mill full regardless of grade. I note that you are in Saskatoon. I grew up there. Small world!

      Delete
    4. There is no structure to mark up Bill. Just the odd random nugget with no particular orientation so the underground geos would only have perhaps some tight spaced drilling to influence their face markups but I have seen no evidence of grade trends even over 1 metre collars. So they are really lost.

      Delete
    5. time will tell, basically this next quarter should be key as their reconciliation and grade control drilling will be in full swing.

      However, I'll be more interested in the update resource report due in fall/winter to see if there will be a massive downgrade,or they'll play ostrich and pretend that everything is just spiffy and the low grade material that they are mining now just means that the rest of the ore will be higher grade , you know, because of math

      Delete
  11. And by the way Bill, Tom and Angry I highly respect your thoughts and conviction on this topic. Paths, looks like you swallowed some Quartermain coolaide but give you 10/10 for the counter arguments.

    ReplyDelete
    Replies
    1. And folks need to recall that these guys are in the primary stoping sequence at this time. You would know what that statement means Bill. In other words, the initial stopes need to be filled up with coarse mill tailings with cement or rock fill with cement in order to mine the ore left in pillars. Perhaps these guys are geniuses but almost every other underground miner I have ever looked at or operated personally has fallen well behind on backfilling year on.

      Delete
  12. You would probably get some kicks out of what I published on the ceo.ca site Bill and others. https://ceo.ca/@ocotilloredux/zinc-mining-ramblings-the-summary There is about 400 pages of info there if you page down, that I am currently updating but keeping to myself since there is no glory in speaking out I guess.

    ReplyDelete
  13. Dear Paths,

    Do you have any comments on the Q1 financials?

    I mean, operating earnings didn't even manage to cover interest paymnets, and at the current rate of production, they'll only fall ~$280M short of paying the debts due at the end of the year.

    I look forward to that surging share price when Pretium have to issue a crap load of shares to settle their debts....

    ReplyDelete
    Replies
    1. The grade in January was so low as to skew the entire quarter. One cannot tell much from it as they are now getting around to using better procedures. It's a high productivity mining method, that will be started to function reasonably well going forward. The improving results later this year should make it easier to obtain financing, or they may have to defer the stream buyout for another year. No problem, all is resolving and looking good.

      Regards, Paths

      Delete
  14. AG..see my post above for my rough financial analysis. I'm no numbers guy but I thought Operating Earnings were $27M, Interest was $15M and Corporate Costs were $6M, leaving $6M in cash for other needs, including debt repayment. My understanding of Current Debt is that it's debt payable in the next 12 months, or in this case by the end of the first quarter of next year they must pay $300M. That would give them three more months than the end of the year to get all those new shares issued. I think SEC requires an updated 43-101 report if they intend to issue any further securities, including equity, bonds or a loan roll over.

    ReplyDelete
  15. Slide 29 in the latest presentation is all you really need to know. It illustrates the stope boundaries based on the current reserve model and then illustrates the stope boundaries based upon the geological control drilling. Most would state, wow, what a great job these guys are doing at grade control. Guys like me state, holy crap, you have dropped off 30% of the stope tonnage planned (the entire slot area for instance and the last two rings.) And you guys are still pulling up 40% short on grade. Good luck with the expansion. This is one dead parrot.

    ReplyDelete