The Place of Geostatistical Simulation through the Life Cycle of a Mineral Deposit

Round 1

Reviewer 1 Report

You are advocating for modern techniques and in doing so provide an overview of the use of simulation in the various stages of the mining cycle.

I would have liked to see more on the final stage, reclamation, an illustrative example such as in the previous sections would have been good.

I would have also liked to read about your views on the use of machine learning techniques which a appear to be an emerging new set of methods, where you see their place and how they might interact with simulation.

In Figure 6, I think the caption either needs a better explanation or the content of the figure needs to be revised. How do you conclude from what is shown that 20-30% of the data spacing is reasonable to resolve contacts?

Figure 2 (3) needs a better explanation and should ideally be linked to the map. Also, while it is clear that green corresponds to a moderate correlation, some key for the colour scale would be good

Line 255: should it not be a high probability of low quality ?

Figure 1 bottom right. This does not look like a tornado chart to me, as is claimed in the text, please provide a better explanation in the caption.

Provide references for IK and UC

There are some minor grammatical errors that the author should be able to identify easily through another proofreading

Author Response

Thanks for the review. I agree about expanding the final stage although I have no convenient figure. The text has been expanded.

I do not want to rant on ML techniques, but I have added some sentences early on. ML is absolutely important, but unlikely to replace what we do - supplement in important ways, but not replace.

Yes - the caption to Figure 6 was sketchy. I expanded. Yes - the color of the correlation matrix on Figure 2 has been explained. Yes - it should be "high probability of low quality" Yes - a tornado chart would be symmetric. The explanation to the plot on Figure 1 was changed.

No references to IK or UC are provided because no mention of them was made.

Reviewer 2 Report

1) In Section 4, it would be nice to include bibliographical references from varied sources for each stage of mining. There have been plenty of developments worldwide (not only Canada) and it is a pity that no reference is provided. Recent references (>2017) are missing.

2) Do geometallurgical and geomechanical modelling also fit in Section 4.2?

3) More emphasis could be done on the contribution of geostatistical simulation to strategic mine planning under uncertainty (maybe as part of the definition of ore reserves in Section 4.2?).

Minor comments

line 1: I wonder whether the paper should be classified as "Article" (research paper) or "Review"

line 50: is uncertainty a "state of nature", or rather a consequence of our ignorance of nature?

line 71: supports -> support

line 106: change a comma to a dot

line 130: "reasonably" may be changed to "approximately"

line 145: "stable" may be changed to "robust"

line 166: "NI 43-101" may be changed to "CRIRSCO"

line 182: gemetallurgical -> geometallurgical

lines 183-184: is the kriging error variance an "unreasonable" assessment of uncertainty?

line 274: 16000 -> 16,000

lines 298-304: I fear that classification based on production volumes is not a standard nowadays; most classifications still rely on geometric criteria (distances to drill holes, etc.)

line 444: authors declare -> author declares

lines 464-478: references 16-23 are not cited in the text.

Author Response

The references have been adjusted. Primarily references to the authors own work have been removed!

Reference to geometallurgical and geomechanical properties have been mentioned. Good point.

Yes - a paragraph on strategic mine planning has been inserted. Another good point.

The paper could be an article or review.

Regarding Line 50 - it is both. Line 71 - corrected. Line 106 - corrected. Line 130 - fair enough - changed. Line 145 - changed. Line 166 - I used 43-101 reports for my analysis, but changed the text. Line 182 - corrected. Line 444 - corrected.

Regarding the kriging variance, it does not account for the proportional effect for most skewed grade distributions - so, not reasonable, but I did not say that.

Regarding classification, I agree that geometric criteria are used, but they aim to have the meaning of some relevant uncertainty - I have clarified.

References have been fixed.