Glerum et al. have coupled rift-modelling software developed in ASPECT to a surface-processes model called FastScape. They use this computational infrastructure to perform a suite of 2D numerical models of extensional-basin formation and sedimentation under different rifting scenarios. They apply a mineral-systems approach to identify characteristics that might be favourable to the formation of clastic-dominated lead-zinc deposits. Their results indicate that architectures involving narrow, asymmetric rifts are most likely to produce a thick pile of sediments that (a) contain suitable lithologies necessary to act as both source and host rocks; (b) have the correct thermal histories; and (c) can be connected to one-another by suitable fluid pathways along faults. They conclude that identification of ancient, narrow margins in the geological record will likely help to reduce the search-space for new deposit discoveries.

I enjoyed reading this thought-provoking manuscript and am impressed by the amount of work that it contains. It certainly raises new possibilities concerning which mechanisms may be responsible for the observed localisation of these deposits within very specific regions. I am in support of publication as is, but have a few optional suggestions and talking points that may potentially add to the story. Feel free to take or leave as you see fit and thanks for the opportunity to read it!

A few general thoughts:

My interpretation of the numbers in Lines 377-378 and focus of Figure 5 is that it is mainly the generation of appropriate “source rocks” and the pathways that connect them to host rocks that are the limiting factor across the different types of rift (i.e. the availability of suitable host rocks is not generally a problem). Is this the case? I don’t think that point is made particularly clearly anywhere (my apologies if this is incorrect).You could potentially also include a host-rock-only count in Figure 6?

Related to this point, do we actually want widespread host rocks? Local ones that are in the vicinity of the brine conduits are probably all we need to concentrate metals into a viable deposit. Having them widespread has the potential to result in metals precipitating in lower concentrations over larger areas, depending on the particular hydrothermal circulation pathways.

I realise that these models do not include melting and it is therefore prudent to not overly speculate on this aspect, but I wonder if they might also provide insights on the potential for generating mafic volcanism? These rocks often act as a source of Zn and particularly Cu, such as the Eastern Creek Volcanics at Mount Isa. Given that the locations and strength of mantle upwelling is modelled, we can infer where in the model domain the solidus is most likely to be crossed. It strikes me that the narrow asymmetric margin more often has a focused mantle upwelling in its vicinity than other margins in the various models? They may therefore be more susceptible to generating volcanic rocks, which would be another potential strength of this particular geodynamic setting.

A further potential benefit of the narrow-margin setting would be proximity of source and host rocks to the adjacent continental platform. This architecture would seem to be ideal for the “brine factory” ideas of Manning & Emsbo (2018) and others – i.e. potential for a nearby evaporative platform to generate brines and a hydrological head to drive them down into the basin.

Please forgive the personal bias, but one of the interpretations we made in Hoggard et al. (2020) was that large clastic-dominated base metal deposits are all located in basins that occur on cratonic margins. The models in this study are all run using regular, Phanerozoic-style (ie. thin) lithosphere as the starting template. It may be that you don’t agree with the craton story, but I’d be interested to hear your take on which aspects of the narrow-margin fertility might translate well to a potential extensional setting on the edge of thick lithosphere? In addition, which aspects of your model results might change if they were run using thicker, potentially depleted lithosphere? For example, I would guess that the basal heat flow would be lower and it would be trickier to get the source rocks into your desired temperature window? At the same time, the solubility of Zn and Pb in brines is still high at lower temperatures, as long as they are oxidised, so I don’t think it would matter much.

Some specific comments:

The Plain Language Summary makes mention of the 3 Myr maximum extent for mineralisation, but this fact is not listed in the abstract. It seems potentially important enough to mention somewhere in there too?

Mccuaig should be McCuaig throughout.

Just checking - is the sand and silt porosity set to zero, as suggested in Table 1? I only ask because I think that renders the e-folding depths redundant.

Lines 294-295: Qualify that the “geotherm and adiabat temperatures match”.

Line 326: Typo “inhered” → “inherited”?

Figs 3, 7 and 8: Add a simple label above panels stating the model type (e.g. “Narrow Asymmetric Rift”) to ease quick comparison.

Figure 4: I found the panels for ore formation mechanisms 2 and 3 a bit small – the first one is better. I also wonder if you could potentially reverse the domain for model NA-9 to make comparison and contrast to NA-4 easier (i.e. have all the rifts migrating in the same direction)?

Figure and text order: Potentially move Figures 7 and 8 ahead of 5 and 6, so a reader sees all the model results prior to the scenario comparison figures. Lines 410-415, Figures 5 and 6 would then naturally fall under a separate subsection after 3.3 called something like “Optimal characteristics for mineralisation” or similar.

There currently is no Section 3.4.

Line 455: typo “at in”.

Overall that paper provides valuable insights into the dynamics of clastic-dominated base metal deposits. The authors cover a lot of ground to bring together geodynamics and a mineral systems analysis, providing an approach itself that will be valuable reading to multiple disciplines. 

The modeling results provide some testable conjectures, such as “[the] conditions for deposit formation can briefly occur in both narrow and wide rifts for at most 3 My”, and that narrow, asymmetric rifts are the most fertile settings for these deposits. The paper contains a thorough discussion of the results that includes comparisons to real geological cases. It lists some assumptions of the model but lacks a discussion of how those assumptions may limit the specific model results and applicability of the conclusions drawn. Great value from the models could be gained by the reader if the temperature field was included in the result images.

Some minor comments to address below.

“In sedimentary basins, key components of the mineral systems model are the metal source, flow conduits for metal-bearing fluids, and a trap for concentrating metals at the deposit site.”

It would be helpful to see some references here, and explicitly state that if these are the ‘key’ processes, what are the secondary processes deemed to be less important to focus on and why.

“Eq (9) 𝑑𝑧 = 20 km is the assumed in-plane extent of the basins to create a source rock volume”

Assumed based on what? This is somewhat important when you are comparing model results to estimates of deposit endowment to validate your results since I don’t see why this value couldn’t be arbitrarily set much higher or lower. Unless it is calibrated using realistic volume estimates? In which case it can’t be used to justify the resulting endowments from the models.

One limitation not discussed is the modeling of near surface deformation, which is a brittle process, using equations that treat the Earth as a viscous fluid. I understand that this is common practice but strictly speaking it should not apply to the upper crust, and is even debatable in some situations at greater depths. Since a portion of the results depend on the model's ability to simulate near-surface faulting, I suggest adding a discussion around this to the limitations section.

Is there a missing section 3.4 or just a typo?

Would be great to see thermal evolution here in the results, alongside composition and strain. Can you say why you thought it was not important to include that? It would help provide some insight into where melting might be focused even if not explicitly predicted.

The sedimentation rates and volumes are discussed and compared with observations but not the resulting sedimentary structure of the basins. This seems like a key prediction of the models also. Do you have any insight into how, in a broad sense, does the predicted sequence of lithologies and their distribution compare to observations?