Properties of Impact-Related Pseudotachylite and Associated Shocked Zircon and Monazite in the Upper Levels of a Large Impact Basin: a Case Study From the Vredefort Impact Structure
Seann McKibbin
Round 1
Reviewer 1 Report
Paragraph starting line 100, about X-ray fluorescence spectrometry: There is no detail on the quantification of the results. Is the technique here semi-quantitative, as for the EDX on the SEM, or only qualitative (used only for intensity maps)?
Line 122: could be good to specify here that the deformed crystals consist of zircon and monazite. It would also be interesting to know what proportion of deformed zircon (and monazite) has been identified (compared to non-deformed grains).
Lines 278 and line 300, about the granular texture of grain 01-14 and grain 01-16. The granular texture is not obvious, although it is clear for the grain 01-19. It appears to be very localized on the edges, in contact with porosity or other interfaces. In this case, can we really speak of a granular texture? It is often in close relation with black pixels (therefore undetermined pixels), one also wonders the reliability of the identification of this granular texture. If it is just a single pixel, it does not define a grain.
Line 304: could you specify here if the planar lamellae could correspond to a twin? (i.e., domains 1 and 2 could be related by a twin law ?). It is specified for the zircon lamellae so it could be stated also for the monazite grain since it is known that twinning is a shock response of monazite. In the discussion it is written that domains 1 and 2 are twinned area so please give more details about the relative orientation of domains 1 and 2.
The discussion seems to me reasonable.
Author Response
Reviewer 1
Paragraph starting line 100, about X-ray fluorescence spectrometry: There is no detail on the quantification of the results. Is the technique here semi-quantitative, as for the EDX on the SEM, or only qualitative (used only for intensity maps)?
This is a semi-quantitative technique, as I have now indicated in the method section. We did not attempt to quantify the mapping results, hence the scales in Figures 4-5 are only relative and the images are used merely for comparison between different phases within the sample.
Line 122: could be good to specify here that the deformed crystals consist of zircon and monazite.
This is indicated accordingly
It would also be interesting to know what proportion of deformed zircon (and monazite) has been identified (compared to non-deformed grains).
I have provided this information under the Results section 4.4 (first paragraph).
Lines 278 and line 300, about the granular texture of grain 01-14 and grain 01-16. The granular texture is not obvious, although it is clear for the grain 01-19. It appears to be very localized on the edges, in contact with porosity or other interfaces. In this case, can we really speak of a granular texture? It is often in close relation with black pixels (therefore undetermined pixels), one also wonders the reliability of the identification of this granular texture. If it is just a single pixel, it does not define a grain.
I have removed the term “granular texture” from the description of these grains.
Line 304: could you specify here if the planar lamellae could correspond to a twin? (i.e., domains 1 and 2 could be related by a twin law ?). It is specified for the zircon lamellae so it could be stated also for the monazite grain since it is known that twinning is a shock response of monazite. In the discussion it is written that domains 1 and 2 are twinned area so please give more details about the relative orientation of domains 1 and 2.
I have mentioned in Results that the domains 1 and 2 are likely microtwins. The twin relationships are given (180'/[101])
The discussion seems to me reasonable.
Thank you!
Reviewer 2 Report
Dear Dr. Kovaleva and Dixon,
I reviewed your Minerals manuscript 973052 “Properties of impact-related pseudotachylite and associated shocked accessory minerals in the upper levels of a large impact basin: a case study from the Vredefort impact structure”.
The manuscript is well written and provides data that should be appealing to the readers of Minerals. While I am concerned that the sample and topic of this study have already been discussed in several other publications (Kovaleva et al. 2018, 2020) I think the study should be suitable for publication after some issues are addressed. Please try to resolve the following questions:
- The authors do not provide any quantitative or “semi-quantitative” data that supports their interpretation that the “pseudotachylite” has a different or similar composition to the “granite” host clast. Their interpretation is based on strictly qualitative comparisons of XRF-mapping that is presented as relative intensities (of X-rays?) without units. To remediate this issue, the authors could produce true quantitative data for the composition of the “granite” host, the “granophyre” melt pockets and the “pseudotachylite”; this could be accomplished with LA-ICPMS. Because this seems crucial for the topic of the manuscript, the application of such a quantitative method should not be “beyond the scope of this study”.
- The “granite” clast is described as microcrystalline quartz with “interstitial biotite” and its texture is “foliated”. These are not characteristics of granite. I recommend revising the petrography of the “granite” clast and entertain the possibility that this lithology could be a metamorphosed sedimentary rock instead. I think this could strengthen the argument for a near-surface location of this lithology during the impact.
- The authors should entertain the possibility that the “pseudotachylite vein” is an impact melt vein that became chilled after intruding a cool clast, which was then incorporated into a larger volume of melt and injected into the crater floor. Their compositional interpretation of the “pseudotachylite vein” contradicts the definition of pseudotachylites. Moreover, I am missing other aspects of pseudotachylite petrology such as branching-off injections that could support the interpretation as a pseudotachylite.
- Figures 10A and 11A (desktop-SEM BSE images of zircons) are of inferior quality, unsuitable for publication. The authors should consider taking replacement images with an improved focus or omitting these images.
I attached a PDF-version of your manuscript with additional comments and suggestions for modification that I hope you will find useful for the revision.
Comments for author File: 
Comments.pdf
Author Response
Dear Editors and dear Reviewer,
Thank you for your comments. You will find my point-by-point response below.
Best regards
Elizaveta
I reviewed your Minerals manuscript 973052 “Properties of impact-related pseudotachylite and associated shocked accessory minerals in the upper levels of a large impact basin: a case study from the Vredefort impact structure”.
The manuscript is well written and provides data that should be appealing to the readers of Minerals. While I am concerned that the sample and topic of this study have already been discussed in several other publications (Kovaleva et al. 2018, 2020) I think the study should be suitable for publication after some issues are addressed.
This sample was indeed described in one paper: Kovaleva et al. 2018 in South African Journal of Geology. Note that this description was preliminary and did not include EBDS maps of accessory minerals, as well as detailed µXRF maps. We also included new petrographic descriptions and images of new thin sections.
Another paper was published in Geology, 2019. That study was dedicated to one “FRIGN” zircon grain. Here we present new EBSD data on zircon and monazite, unpublished before.
I would like to emphasize that this sample is unique in its nature and we will continue to study it with new methods.
Please try to resolve the following questions:
- The authors do not provide any quantitative or “semi-quantitative” data that supports their interpretation that the “pseudotachylite” has a different or similar composition to the “granite” host clast. Their interpretation is based on strictly qualitative comparisons of XRF-mapping that is presented as relative intensities (of X-rays?) without units. To remediate this issue, the authors could produce true quantitative data for the composition of the “granite” host, the “granophyre” melt pockets and the “pseudotachylite”; this could be accomplished with LA-ICPMS. Because this seems crucial for the topic of the manuscript, the application of such a quantitative method should not be “beyond the scope of this study”.
I am not entirely clear what are the reviewer’s doubts about the nature of the vein are based on. Maybe I did not understand the comment. The field relationships, as well as textural, petrographic, mineralogical, and geochemical evidence point at the connection of the vein with the host meta-granitic clast. So, we have no doubts that the vein belongs to the clast and not an offshoot of granophyre, for example.
Also note that we have already established the genetic connection of the vein to the host granite by a semi-quantitative hand-held XRF technique, published in Kovaleva et al., 2018 (SAJG). I have added this information into lines 89-91. It is also mentioned in the first line of the sub-section 5.2. The further possibilities of identifying the composition and their validity/necessity were discussed in the comment to Kovaleva et al. 2018 paper by W.U. Reimold (Reimold et al. 2019) and in our reply to the comment (Kovaleva et al., 2019, SAJG). Our conclusion is that quantitative composition is not necessary and will not change the story or any of the interpretations; and additional analyses could be difficult, time-consuming and not that informative.
- The “granite” clast is described as microcrystalline quartz with “interstitial biotite” and its texture is “foliated”. These are not characteristics of granite. I recommend revising the petrography of the “granite” clast and entertain the possibility that this lithology could be a metamorphosed sedimentary rock instead. I think this could strengthen the argument for a near-surface location of this lithology during the impact.
As stated in the manuscript, the texture of clast is massive (line 163; Figure 2A-B), it is only foliated next to the vein (line 172; Fig. 2D). Obviously, the rock is better to be described as meta-granite (not just granite), as it was shocked and then thermally re-worked. It could legitimately be a metasedimentary rock with granitic composition, as we have identified comparable to granite amounts of Fe, Ti, Al, K (see also Kovaleva et al. 2018). The clast is small and it’s hard to tell what is originally was. I have added your suggestion to lines 165-169.
Regarding the near-surface location of the rock at the moment of impact, there is not a problem for a granite to be derived from near the surface. It is documented that granitic basement was locally exposed, just as it is now in places unrelated to the Vredefort central uplift. In South Africa, North, North-west and West from Vredefort dome, there are multiple granite batholiths and exposures (e.g., Johannesburg Dome, Nelspruit batholith), which have composition and age identical to OGG. 2 billion years ago the picture was the same – west from the structure there are mined iron deposits, which formed 2.2-2.0 Ga and it has been shown that their sedimentology process was controlled by the granitic basement, outcropping at the paleo surface or very close (first hundreds m in depth)
See, for example:
Smith, A.J.B., Beukes, N.J., 2016, Palaeoproterozoic banded iron formation – hosted high-grade hematite iron ore deposits of the Transvaal Supergroup, South Africa. In: The Great Mineral Fields of Africa, M.G.C. Wilson and R.P. Viljoen (eds.), Episodes, v. 30, pp. 269-284.
The same picture is observed NW from Vredefort dome in gold fields of the Wits Supergroup:
Robb, L.J. and Meyer, F.M., 1990. The Nature of the Witwatersrand Hinterland: Conjectures on the Source Area Problem. Economic Geology, 85, 511-536.
- The authors should entertain the possibility that the “pseudotachylite vein” is an impact melt vein that became chilled after intruding a cool clast, which was then incorporated into a larger volume of melt and injected into the crater floor. Their compositional interpretation of the “pseudotachylite vein” contradicts the definition of pseudotachylites. Moreover, I am missing other aspects of pseudotachylite petrology such as branching-off injections that could support the interpretation as a pseudotachylite.
Clearly, it is impact-related melt vein, but pseudotachylite, not granophyre. It was established in work by Kovaleva et al., 2018. The clast is only few cm in diameter, so no branching-off veinlets are observed. The other textural evidence is clearly pointing at the vein’s in situ origin (Fig. 2D, E-F).
Yes, our interpretation is not entirely in line with the classic definition of pseudotachylite, but this is the point of our study. We suggest that at higher levels of impact structure anomalies can be observed. Also, mixture of material in PT veins was documented by various authors before, the most recent – Kovaleva et al. (2020) in Lithos, which we cite in lines 407-408 (“Material mixing in shock-induced pseudotachylites, Vredefort impact structure, South Africa”).
- Figures 10A and 11A (desktop-SEM BSE images of zircons) are of inferior quality, unsuitable for publication. The authors should consider taking replacement images with an improved focus or omitting these images.
Note the size of the grains, only 10-15 µm in diameter, hence the quality of the BSE images. Unfortunately, I can’t take replacement images of these particular grain as the samples are in external laboratory for TEM study; I have only 5 days for revision. I think the current images are useful and show information about the general appearance of the grains (e.g., how grain 01-19 is resorbed), and the arrangement of fractures in grain 01-14.
I attached a PDF-version of your manuscript with additional comments and suggestions for modification that I hope you will find useful for the revision.
The comments from .pdf file are introduced accordingly (see document attached).
Author Response File: 
Author Response.docx
Reviewer 3 Report
L13 - 'However, granophyre (impact melt) dykes in such structures preserve...'
L15 - I would suggest writing 'One such lithic clast from the Vredefort impact structure...' or 'We have sampled one such lithic clast containing...' in order to make it clear that the clast is from this locality.
L23-24 - consistent use of farther/further?
L25 - 'These relationships point to elevated shock pressure...'
L31 and thereafter - 2020 My or 2,020 My
L108 - Sodium or Na? The latter would be consistent with normal scientific usage (except at the beginning of a sentence) and with Mn on line 106
L156 - finely crystalline
L172 - mosaiced or mosaicised?
L178 - most of which (Fe...
L207 - ...inhomogeneous but comparable to that in granite.
L358 - Here it is stated that pseudotachylite is more mafic than the host. Bulk melting of mafic phases (i.e. disequilibrium melting of the whole rock) is indeed one suggestion to explain this. It should be noted that during melting of granitic rocks, an excess of silica could possibly survive as refractory quartz grains. The mention of 'partially molten quartz aggregates' at L155 would be consistent with their leftover refractory character after partial melting.
I would encourage the authors to develop such an idea in one paragraph or so, to support the inferences that they have made by geochemical techniques that are qualitative (or semi-quantititive only). Incongruent melting of a mafic phase such as enstatite to produce a silica-rich melt and refractory forsterite is well documented (e.g. Bowen and Schairer 1936). There may be complementary situations discussed in the literature, in which a peritectic close to an amphibole-like composition generates a silica-poor melt coexisting with refractory silica. Such a process, if relevant, would be of value to the study of impact melts, and a brief survey of published phase diagrams may turn up useful ideas for the community.
Author Response
Reviewer 2
L13 - 'However, granophyre (impact melt) dykes in such structures preserve...'
Corrected
L15 - I would suggest writing 'One such lithic clast from the Vredefort impact structure...' or 'We have sampled one such lithic clast containing...' in order to make it clear that the clast is from this locality.
Added
L23-24 - consistent use of farther/further?
Corrected to “farther”
L25 - 'These relationships point to elevated shock pressure...'
Corrected
L31 and thereafter - 2020 My or 2,020 My
Corrected
L108 - Sodium or Na? The latter would be consistent with normal scientific usage (except at the beginning of a sentence) and with Mn on line 106
Corrected to Na
L156 - finely crystalline
Corrected
L172 - mosaiced or mosaicised?
Corrected
L178 - most of which (Fe...
Corrected
L207 - ...inhomogeneous but comparable to that in granite.
Corrected
L358 - Here it is stated that pseudotachylite is more mafic than the host. Bulk melting of mafic phases (i.e. disequilibrium melting of the whole rock) is indeed one suggestion to explain this. It should be noted that during melting of granitic rocks, an excess of silica could possibly survive as refractory quartz grains. The mention of 'partially molten quartz aggregates' at L155 would be consistent with their leftover refractory character after partial melting.
This is added accordingly.
I would encourage the authors to develop such an idea in one paragraph or so, to support the inferences that they have made by geochemical techniques that are qualitative (or semi-quantititive only). Incongruent melting of a mafic phase such as enstatite to produce a silica-rich melt and refractory forsterite is well documented (e.g. Bowen and Schairer 1936). There may be complementary situations discussed in the literature, in which a peritectic close to an amphibole-like composition generates a silica-poor melt coexisting with refractory silica. Such a process, if relevant, would be of value to the study of impact melts, and a brief survey of published phase diagrams may turn up useful ideas for the community.
This is a well-documented and supported idea that was nicely summarized in the work of Spray (2010), who discussed both impact-generated and tectonic pseudotachylites, and which we cited. We also discussed it in relation to impact-generated melts in detail in some recent papers such as Kovaleva et al., 2018 (SAJG), Kovaleva et al. 2020 (Lithos). So, it’s really not new and I don’t know if we need to tell the same information again (other than the short summary we give in lines 373-382, revised version). I have also recently reviewed a paper by Montheil et al. that experimentally and quantitatively shows the same thing: hydrous and mafic phases melt and refractory phases remain as clasts in friction melts. This paper is accepted and will be soon published in Frontiers Earth Science. So, I do not think that the alternative idea would be necessary to discuss – I am quite convinced about the disequilibrium flash melting!
Reviewer 4 Report
Dear Authors,
Please see the attached PDF with comments, notes, and suggestions on your study.
Kind regards,
Comments for author File: Comments.pdf
Author Response
Reviewer 3
Small corrections throughout the text – accepted
Line 60-61: After "detail" mention which techniques were used on which minerals and why this contribution is important for the planetary science community. Also, try to make the transition from the one sentence to the other more smooth.
The information is added
Line 64: Rephrase here..sample ??
Re-phrased
Line 86: They are not pink in this photo
Corrected
Line 112: What it means chemo-mechanically ?
Means polishing with the colloidal silica agent, it is a standard procedure needed to create a really smooth surface for the EBSD maps
Line 122-123: How did you locate the zircon and monazite grains ? Did you use feature map ping ?
We have located them with the BSE imaging. Specified in text.
Line 156: replace with "fine-grained" or "crystalline"!
Corrected
Line 184: The scale is not well visible, modify it.
The scale is given in panel B, all the other panels are to the same scale; this is specified
Lines 198-199: The K counts in the vein are higher than in the granite,
It’s is actually the same in vein and in granite, but depleted in the foliated zone; I corrected accordingly
Also you dont show any map of Na!
Yes we did not show all the mapped elements simply to spare space (we describe the distribution of ca. 15 elements in the text, it would be difficult to show all of them in a figure); and the maps look relatively similar
Line 200: In Figure 5B-C are documented only the compositional variations of Fe and K not the rest of the elements that you report.
They look similar to each other and there are ca. 15 maps, therefore, all of them are not shown; I have re-phrased the sentence accordingly.
Line 202: What is the pink colour in Fig. 5A ? Or is just red ?
I think, it’s a mixture between red (Mg) and blue (Al), which gives such purplish effect
Line 207: I see patchy distribution of K within the vein and more homogeneous in the granite.
Actually, what we see as homogeneous and depleted is a foliated zone around the vein. The granite is also patchy (see portions at the far left and far right from the vein). The phases are corrected in the figure.
Line 219: The Mg distribution seems to be fairly homogeneous.
I can’t quire agree, it’s enriched in the Cpx needles as indicated in text (see position and configuration of the needles in panel A)
Line 224: From Figure 6 is evident that the vein is enriched in Ca, and Al.
Figure 6 does not show the PT vein, it’s the matrix of granite with veinlets of granophyre. The discussed vein is shown in Figure 7. Sorry for confusion.
Line 236: Make the Size of the letters for the chemical elements the same for all your Figures.
Corrected accordingly
Line 237: Replace with PT vein here and in Figures
Corrected
Lines 247-248: Would be beneficial for the reader to see schematically as a transect across the vein the mm-scale gradients observed in the intensity of zircon shock microstructures relative to the vein.
Such schematic diagram is shown in Kovaleva et al., 2019 (Geology), their Figure 2B; I gave a reference to this paper
Line 278: 1 and 3 are parallel sets of PF's are truly different ?
Yes, they are not parallel but at an acute angle to each other.
Line 307: Domain 2 is purple in the maps, and domain 3 green, please correct.
The pole figures are color-coded as in C (cumulative misorientation map), there, domain 2 is green and domain 3 is blue. We used this color-coding for clarity; in the IPF map, domains 1 and 2 are a little bit similar to each other in color (purple and pink), so not everyone will see a difference in the pole figure
Line 332: Again here a scheme would vastly help the reader to conceptualize the sequence of events reported.
This would be quite complicated scheme, as we describe large-scale processes side-by-side with the micro-scale deformation effects vs pressure, temperature and time, where some of the stages are almost coeval. I agree that it would be very helpful but I honestly do not know myself how to make such scheme!
Lines 382-383: Would be interesting to draw any connections with this study : https://www.sciencedirect.com/science/article/pii/S0012821X18305946?via%3Dihub
I have added this discussion to the section 5.3, thanks for the reference!
Lines 415-417: Recrystallization of twins has been documented also in titanite grains also from a foliated pseudotachylite at the South Range of the Sudbury structure, in our study (https://link.springer.com/article/10.1007/s00410-018-1511-0), in case that you want to cite it.
I, by the way, reviewed this paper, when it was submitted to Geology.
The twin-recrystallization is attributed in the titanite study to the post-shock heating effects but the origin and the effect of twin recrystallization on U-Pb systematics remains speculative and needs ultrahigh spatial resolution geochronology (APT). Would be interesting to theorize a bit about the origin of the neoblasts formed in expense of the twins in your study.
Thank you, I added the reference and the discussion to the section 5.3.
Line 248: Would be interesting to know if the proximity to the vein induced any age resetting in potentially future studies.
Most certainly, we have dated the granular neoblastic zircon near the vein with SII SIMS, and it gave the age of Vredefort impact (with the large error, but still!). That study is submitted to EPSL at the moment. I have added a sentence in this regard.
Author Response File: Author Response.docx
