The Evolution of Permian Mafic–Ultramafic Magma of the Yunhai Intrusion in the Northern Tianshan, Northwest China, and Its Implications for Cu-Ni Mineralization

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Please see the attachment. 

Comments for author File: Comments.pdf

Comments on the Quality of English Language

No additional comments.

Author Response

Comments 1: The only technical suggestion is for the final figure 13a, where primitive magmas in the area of partial melting in the mantle, as well as ultramafic-mafic intrusions in the crust should be marked in green.

Response 1: Thanks for your valuable advice, we have modified the color of the primitive magmas and ultramafic-mafic intrusions in Fig. 13a according to your requirements

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript "The Evolution of Permian Mafic-Ultramafic Magma in the Yunhai Region ..." by Yuxuan Pei and others is well written and interesting. The stated objectives of the authors were to "determine the timing of diagenesis and Cu-Ni mineralization, identify the magmatic source, understand the magma evolution, and further elucidate the Cu-Ni mineralization process" of the Yunhai deposit. Most of these objectives were met, although the diagenesis was not discussed and should be deleted from the list of objectives.  There are, however, a few important changes that the authors should consider to make it better; particularly those related to Figure 13.  A short list of line-by-line comments and questions follow:

Line 98 – Replace "diagenesis and" with "the"

Line- 203 Define "net-textured"

Figure 7d -  Replace "Flood Fasalts" with "Flood Basalts"

Figure 7d – Provide the source of the diagram and each of the five rock fields.

Line 379 – What is the source of the "High MgO basalt field?

Lines 463-470 – This is a run-on sentence that should be divided into two or three sentences.

Lines 511- 515 – This is a reasonable conclusion, but it is not represented in the cross –sectional model (Figure 13). A labeled "Descending slab" needs to be added to Figure 13.

Lines 562-566 – Again, this seems to be a reasonable conclusion, but it is not represented in Figure 13. Instead, a slab or layer labeled "Partial melting" seems to be the source with nothing directly derived from the "Metasomatized subcontinental lithospheric mantle" The authors need to relabel some of the slabs or layers particularly the "Partial melting" layer to clarify the intent.

Lines 624-627 – Readers will wonder why important layered mafic intrusions including the Skaergaard, the Great Dyke, and the Stillwater were not considered.

Figure 13 – This is an important figure but needs considerable work. Where are the ophiolites that are discussed in the text? What is the reason for extension? There is no plume, and there should be flood basalts in such an extensional environment.

Figure 13b – The arrows labeled "Early sulfide segregation at depth" do not point to anything. The process is not clear and is in need of explanation. It should be possible to illustrate the process. Is the ore layer a cumulate or not?

Figure 13a – The layer labeled "Partial melting" seems to be independent from the Matasomatized layer. If it is part of the Metasomatized layer it should have the same blue color and simply circled.

Figure 13 – Is any of the melting due to decompression melting?

Figure 13 – The font is too small to read.

Author Response

Comments 1: Line 98 – Replace "diagenesis and" with "the"

Response 1: It has been modified as required

Comments 2: Line- 203 Define "net-textured"

Response 2: net-textured is formed when many thin veins are interwoven into an irregular net

Comments 3: Figure 7d -  Replace "Flood Fasalts" with "Flood Basalts"

Response 3: It has been modified as required

Comments 4: Figure 7d – Provide the source of the diagram and each of the five rock fields.

Response 4: The sources of the illustrations have been supplemented as requested

Comments 5: Line 379 – What is the source of the "High MgO basalt field?

Response 5: the "High MgO basalt field are from Chai et al., 1992[57].

Comments 6: Lines 463-470 – This is a run-on sentence that should be divided into two or three sentences.

Response 6: The sentence has been divided into three sentences as requested

Comments 7: Lines 511- 515 – This is a reasonable conclusion, but it is not represented in the cross –sectional model (Figure 13). A labeled "Descending slab" needs to be added to Figure 13.

Response 7: Zircon U-Pb age data indicate that the Yunhai intrusions was formed in the early Permian period, which was in the post-collision extension stage. There was no subduction, which occurred before the mineralization.

Comments 8: Lines 562-566 – Again, this seems to be a reasonable conclusion, but it is not represented in Figure 13. Instead, a slab or layer labeled "Partial melting" seems to be the source with nothing directly derived from the "Metasomatized subcontinental lithospheric mantle" The authors need to relabel some of the slabs or layers particularly the "Partial melting" layer to clarify the intent.

Response 8: According to lines 519-527. Evidently, the geochemical characteristics of the Yunhai intrusion, particularly its trace element composition with notable LILE enrichment and HFSE depletion, as well as high LILE/HFSE rations(fig7a and b), indicate a metasomatized subcontinental lithospheric mantle origin. Isotopic data, including high positiveεHf(t) (9.27-15.9) and εNd(t) (2.75-6.56), low (⁸⁷Sr/⁸⁶Sr)_i ratios (0.7037-0.7053) (fig8), and young _T2DM(Hf) ages(435-643Ma), support the hypothesis of a newly metasomatized subcontinental lithospheric mantle. The Yunhai deposit is the product of post-collision and extension environment. The subduction before mineralization caused the metasomatism of subduction fluid in the mantle source region below the study area. The metasomatism transformed mantle source region was partially melted under the influence of orogeny during the Permian post-collision and extension stage, forming the initial magma of Yunhai intrusion. Thank you for your valuable advice

Comments 9: Lines 624-627 – Readers will wonder why important layered mafic intrusions including the Skaergaard, the Great Dyke, and the Stillwater were not considered.

Response 9: The typical mafic intrusions in the world which are affected by crustal contamination

are mentioned in this sentence.

Comments 10: Figure 13 – This is an important figure but needs considerable work. Where are the ophiolites that are discussed in the text? What is the reason for extension? There is no plume, and there should be flood basalts in such an extensional environment. The absence of flood basalt is also indicated in the figuer7d.

Response 10: Reply 7 and Reply 8 for detailed responses. Yunhai intrusion is the product of post-collision extension stage, and subduction occurred before diagenesis and mineralization.

Comments 11: Figure 13b – The arrows labeled "Early sulfide segregation at depth" do not point to anything. The process is not clear and is in need of explanation. It should be possible to illustrate the process. Is the ore layer a cumulate or not?

Response 11: The controversial arrows have been removed from the figure. Fig b is mainly intended to express the early sulfide segregation at depth, and fig c is mainly intended to express the fractional crystallization during magma emplacement. Thank you for your valuable advice.

Comments 12: Figure 13a – The layer labeled "Partial melting" seems to be independent from the Matasomatized layer. If it is part of the Metasomatized layer it should have the same blue color and simply circled.

Response 12: Reply 7 and Reply 8 for detailed responses.

Comments 13: Figure 13 – Is any of the melting due to decompression melting?

Response 13: The tectonic background of this area has been controversial for a long time. Most scholars believe that the tectonic background of this period is post-collision extension environment, but a few scholars believe that the tectonic background of this period is related to the Tarim mantle plume activities, but this part is not discussed in this paper, so this paper only refers to the unified post-collision extension environment of predecessors for subsequent discussion.

Comments 14: Figure 13 – The font is too small to read.

Response 14: The font has been jumped one size

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors!

Please, find my comments in the attached file.

Best wishes.

Comments for author File: Comments.docx

Author Response

Title

Comments 1: of the article is broader than the topic under consideration and claims to summarize the magmatism of the area. In fact, ,it should correspond  to the real content and should be slightle modified : Evolution of Permian-Triassic mafic-ultramafic magma of the Yunhai  intrusion of Northern Tianshan, northern China and its implication for Cu-Ni mineralization.

Response 1: Thank you for your constructive comments. The title of the paper has changed to: The Evolution of Permian Mafic-Ultramafic Magma of the Yunhai intrusion of Northern Tianshan, Northwest China and Its Implications for Cu-Ni Mineralization.

 

Abstract

Comments 2: It should be stated not only economic importance of the Yunhai deposit, but also its role in the understanding the genesis of this metallogeny zone.

Response 2: Thanks for your constructive suggestions, we have made corresponding modifications. This article concludes that the Yunhai deposit in the western part of the JLOB is similar to the typical deposits in the eastern part in terms of petrogeochemistry and age characteristics, which may indicate that the western part has the same metallogenic potential as the eastern part.

 

Comments 3: Phrase “ no intrusions related to Cu-Ni mineralization” must be corrected . “no mineralization associated with intrusions”.

Response 3: This Phrase is not in our abstract

 

Comments 4: When you give data on age of the intrusions, it is necessary to indicate an error of 2σ (±).

Response 4: We have corrected the corresponding age data as required            

 

Introduction

Comments 5: Fig. 1 labeled ‘Intrusions”. What are their compositions? Judging in color, they are granites. Then where are basic-ultrabasic intrusions? They should be labeled too (even out of scale it they are small), not only Cu-Ni deposits.

Response 5: We have added the basic-ultrabasic intrusions to the figure as requested. Thank you for your valuable advice

 

Geological setting

Comments 6: This section must be reorganized: firstly, give geology of this area and then – tectonic. Because geology is primary data while tectonic setting is an interpretation.

Response 6: The logic described in this chapter is to first describe large-scale geological features such as CAOB and then describe small-scale geological features such as JLOB. Thank you for your valuable advice

 

Comments 7: Fig.2. According to geological instructions, gabbro and diorites are shown in green on geological maps. Here they are colored blue. This does not correspond to the generally accepted designations. Moreover, on the scheme at the end of this manuscript (Fig.13) intrusions are shown by green color.

Response 7: The color standard for our geological map is green for gabbro and blue for diorite. The intrusions shown in green in Figure 13 are just a representation Thank you for your valuable advice.

 

Comments 8: Fig.3. The captions to the figure says “outcrops”, but there are no outcrops on photos, there cores ; b) – the gradation contact is not visible because the fracture is drilled there; d,f) maybe, clinopyroxene just is replaced by hornblend?

Response 8: We have corrected the disputed figure captions, and there are indeed changes in mineral composition and content within the range of the white dotted line in Figure b. Corresponding to the problem in Figure d you mentioned, we have re-observed under the mirror and found that the main structure is still inclusion structure., while Figure f is an obvious poikilitic texture. Thank you for your valuable advice

 

Comments 9: The contact between olivine varieties and pyroxenite is not visible on the indicate line but the other contacts are nor shown also they are visible. Should be corrected.

Response 9: Through microscopic observation, we found that there were indeed relatively complete olivine crystal shapes in Figure f at an early stage, but their contact boundaries were not obvious due to hydrothermal alteration

 

Comments 10: Line 165. Fugure 2a, not b.

Response 10: Corrections have been made as requested

Comments 11: Fig.4. a – no scale, b,c – unclear scale.

Response 11: Fig4a is a typical surface oxidized-ore photograph in the exploratory trench work, and Fig. b and c are typical ore structures in the drill-core

Comments 12: Fig. 3d – where is actinilite? It is described for this photo.

Response 12: Through microscopic observation, we found that there were indeed relatively complete olivine crystal shapes in Figure f at an early stage

Comments 13: Minerals abbreviations must be corrected according to international nomenclature. For example, chalcopyrite must be indicated Ccp, not Cpy.

Response 13: Corrections have been made as requested

Comments 14: Line 207, 214 – The Ni-13 orebody (add) represents….

Response 14: Thank you for your valuable comments. We have corrected the orebody label in the paper

Comments 15: Line 218-219 – it is necessary to correct English, not clear.

Response 15: The sentence in this paragraph has been corrected

 

Methods

Comments 16: The section should be reorganized: firstly, give the methods used for rock characteristics (XRF+ICP). In the description of these methods it is not clear which of them needs dissolution – please, check the text. Then you should give the characteristic of Sr-Nd method, and after that put zircon description.

Response 16: The chapter has been reorganized as requested

 

 

Results

Comments 17: What was said about methods also should be done for results: please, start with characterization of rocks and then give data on zircon. Especially since the zircon characteristics are broken into 2 parts (1 and 3).

Response 17: It will be required to describe the characteristics of the zircon before giving the relevant data of the zircon.

Comments 18: Fig.5 shows that some analyses (No 11,18) includes two zones. How does it affect on the result? 

Response 18: Their oscillatory zoning patterns strongly suggest a magmatic origin

Comments 19: Fig. 6. Reference to abbreviation to Fig.3, but this figure does not include Opx, only Hy. Should be corrected. What do grey areas mean?

Response 19: The reference shorthand has been corrected as requested, The shaded areas indicate that the contents of the Yunhai intrusions are mainly controlled by the relative abundances of olivine, orthopyroxene, clinopyroxene, hornblende and plagioclase.

Comments 20: Fig.7. Normalized to ??? Sun, Hoffmann?

Response 20: Standardized data have been annotated in the text

 

Discussion

Comments 21: Two questions arise reading the result interpretations. First one: it is said that all rocks have common magma source judging from the trends (Fig.10; lines 442, 449). Why is so high range of εNd (line 472; Table S5) ? There is a big difference in diorites (2.8-3.1) and peridotites (5-6). This needs to be explained. Maybe, diorites form another massif?

Response 21: The large variation range of εNd may be related to the crustal contamination of Yunhai intrusion. Thank you for your valuable advice.

Comments 22: The second question. The high Cu/Pd ratio is explained by the sulfide segregation (line 602)but it is possible that it was high originally?

Response 22: Previous studies have shown that the D_sul/sil between Cu and Pd is very different, and the Cu/Pd ratio of Yunhai intrusion is much higher than that of the primitive mantle. Therefore, this article tends to suggest that the high Cu/Pd ratio of Yunhai deposit is caused by the sulfide segregation

Comments 23: Line 446-447 – “Intrusions associated with mineralization” must be corrected “mineralization associated with intrusions”.

Response 23: It has been modified as required

Comments 24: Line 517. “geological studies evidence…” What data? No direct data.

Response 24: The geological studies are cited in Mckenzie et al., 1995[40]. The reference shows that melt produced by modest extension in continental regions which are not affected by mantle plums comes from the enriched lithpsphere, and that the solidus of this material is similar to the wet, not the dry, solidus of mantle peridoite. The alkali-rich magmas generated in this way ofren transport garnet and spinel peridotite nodules to the surface.

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript is well written and properly organized; its length is adequate for the topic. The analytical data appear to be of good quality. The contents are certainly of interest for a wide readership, at least in Eastern and Central Asia.

Nevertheless, I cannot recommend publication of the manuscript in its present form. In my opinion, there is a major flaw in the interpretation of the pyroxenites. Although the authors clearly state that they deal with an intrusion that is "well-differentiated" from bottom to top with pyroxenites, gabbro and diorite, they model the formation of these rocks simply by partial melting of a mantle peridotite source, thus treating all rock-types as primary mantle partial melts.

From figure 12 they conclude that the analyzed rocks (mainly pyroxenites) originate from small degrees of partial mantle melting. This is extremely unlikely in view of the major and minor element composition of the pyroxenites. Their high MgO and Ni concentrations indicate that very high degrees of mantle melting would be required in order to generate melts with 30% MgO and more than 2000 ppm Ni. In fact, if these rocks were partial melts, most of them would have to be called (ultramafic) komatiites, a rock-type which is extremely rare in the Phanerozoic. At the high degrees of melting necessary to generate komatiite, the mantle residue would consist of olivine with minor orthopyroxene only, but without garnet (and all sulfides would have entered the melt). Their partial melting modeling in Fig. 12 has therefore no basis and two out of their three conclusions on page 23 are not correct.

In my opinion, the authors should interpret the pyroxenites as cumulates from a (primitive?) magma. It is unfortunate that either there are no (near-)primary magmas (gabbros in Fig. 2?) or the authors have not analyzed such rocks.

Comments for author File: Comments.pdf

Author Response

abstract

Comments 1: Do you really believe that the mass of mineralization can be calculated to this precision? Why not rather write "about 50,000 tons"?

Response 1: We have made corresponding modifications to the unreasonable parts

Comments 2: This character is called "tilde" (as in ñ).  It is often, but incorrectly used to mean approximately (≈).  In your case, however, you mean from A "to" B. You should use an n-dash (–) instead, for example:0.77 – 6.55.The tilde is definitely incorrect.

Response 2: It has been modified as required

Comments 3:Line31-32 This is strange. What are the remaining 97 % of the mantle?

Response 3: This means peridotite with 2% spinel and peridotite with 1% garnet.

Introduction

Comments 4: why switch from tons to megatons, e.g. 125000 t Ni vs. 0.17 megatons Cu?

Response 4: The unit conversion was carried out because of the high reserves of the Huangshandong deposit.

Comments 5: line-96 have led to the discovery of...?

Response 5: It refers to that with the development of geological work, the discovery of new Cu-Ni sulfide deposits in the west section of JLOB, such as Baixintan Lubei.

Comments 6: figure1- Check the scales in b and c!For example, from 42°N to 46°N we have 4°. From South to North Pole we have 180° and the distance is about 20,000 km. Thus, 4 degrees correspond to about 20,000*4/180 = 444 km. Your 500 km scale-bar is much too short, if the latitudes are correct.In c, the distance between 42°20' and 42°00' is one third of a degree, i.e. 111/3 = 37 km. The lenth of your scale-bar is, maybe, 30 km instead of 50.

Response 6: Thank you for your valuable advice. We have modified the figure according to the requirements

Comments 7: figure2- Carboniferous – missing "s" 6 instances altogether! Either"Exploration lines and their numbers"or"Exploration line and its number)

Response 7: The figure has been modified as required

Comments 8: figure3- I don't see photos of outcrops. a and b are photos of drill-cores.

Response 8 We have made some changes to the chart title in question

Comments 9: Line-180± amphibole?

Response 9 The crystal assemblages in question refer to the olivine-pyroxene-plagioclase or olivine-pyroxene-hornblende combinations found in olivine pyroxene.

Comments 10: line188-191 : You have not pointed to Opx in these figures.

Response 10 We re-examined the microscope to distinguish the amount of orthopyroxene from clinopyroxene

Comments 11: figure4c delete "The"

Response 11 The content in the figure has been modified as required

Analytical methods

Comments 12: Line 246-GJ-1 first mentioning of this sample This is also a zircon standard, I guess.

Response 12：GJ-1 is a standard sample for isotope ratio monitoring of zircon

Comments 13 line-248 ICP-MS-DATACAL provide a reference or explain what this is Also, provide the reference for Isoplot.

Response 13：Appropriate references have been provided as requested.

Comments 14: line-267-272 Surely to did not use 0.5 g of sample powder plus 0.5 g of Rh. Do you mean that the amount of Rh added was calculated to provide 1 ppm Rh in the solution used for analysis? Do I understand this correctly – precision and accuracy are about a factor of 2, that is a concentration of 1 ppm Pt could also be 2 ppm or close to 0?

Response 14：We have corrected this controversial description of 0.5g Rh(1ppm). Thank you for your valuable advice

Comments 15: line274-281 You might want to mention which eluent was used to separate Nd from the REE fraction.

Response 15：The corresponding contents have been added as required

Comments 16: figure6 Tell the readers what the source of the compositional variation of the minerals (Ol, Opx, Cpx, Hbl, Pl) is.

Response 16：The source of these data has been indicated in the figure chapter.

Comments 17: figure6 What do the arrows in e, f and g mean?

Response 17：The arrows in the fig e f are representations of correlations between the whole-rock major elements

Comments 18: figure6h What is the source of these lines? To me, (Mg+Fe)/Ti in olivine looks a bit low at the low end. (Mg+Fe)/Ti of 100 would indicate Ti of about 0.3 % in olivine (30 % Mg).

Response 18：The lines in the figure6h refer to a reference that has been added to the figure title[59]

Comments 19: line338 remain consistent In the abstract you have defined LREE (not LREEs) to mean light rare earth elements.

Response 19：The inaccuracies in the sentence have been corrected

Comments 20: Mg# of 0.8 are outside the range of basalts. You should therefore discuss whether the pyroxenites represent cumulates. This is also strongly suggested by the high to very high Ni concentrations. Whether they contain intercumulus melt, I don't know. The diorites are melts (they probably have low Ni).Could the pyroxenites represent cumulates from the diorites or a primitive magma from which the diorites evolved?

Response 20: As for the situation you mentioned, we have supplemented and discussed itMg# of all samples of Yunhai intrusions vary from 0.50-0.93, and the variation range is large. It is generally believed that it is caused by the crystal fractionation of the magma. If Mg#=0.68-0.73 represents the Mg# range of primary magma that is in equilibrium with mantle peridotite[32], it can be seen that Mg#=0.78-0.83 of olivine pyroxene and hornblende pyroxene are mainly formed by the accumulation of mineral facies of early magma crystallization, and Mg# of diorite is 0.50-0.55. It shows that diorite is mainly formed by evolutionary magma.

Comment21：The steep HFS "negative anomalies" could simply indicate that Nb and Ta are incompatible for pyroxenes and only moderately compatible for Hbl and not indicate an OIB character (again in contrast to the diorites).

Response 21：As for what you say about the negative anomaly of this HFSE, it may indeed exist, but in terms of isotopes, most of the samples in Figure 8 fall into the region where MORB and OIB coincide and are also significantly distinguished from OIB. Therefore, the geochemical characteristics of Yunhai deposit are quite different from OIB in terms of both trace elements and isotopes. Thank you for your valuable advice.

Comments 22: line347- Are you sure that a difference of 0.02 is analytically significant??

Response 22：The difference of 0.02 is indeed not very significant but it does not affect the gradual decrease in Eu/Eu^* values from olivine pyroxene to hornblende pyroxene to diorite

Comments 23: figure7a- The name "E-MORB" points to the line for OIB! "the Permian"...delete "the" in Figs. a and b

Response 23：The problems in the figure have been modified as required

Comments 24: 1.figure8- 1. The information that the Sr and Nd isotopic ratios have been calculated for the Permian belongs into the figure caption.

2. The Nd isotopic compositions of EM I and EM II appear to me unbelievably low, Sr isotopic composition of EM II unbelievably high. Where did you get these numbers from?
3. Explain the lines with numbers and arrows at the end.

What kind of upper and lower crust is this – the local crust, average crust? As ref. 25 is incorrect, where do the data come from. Rudnick and Gao have, as far as I know, not reported Sr and Nd isotopes.

4. "the Permian"... – delete "the" in both figures

Response 24：1. The relevant information is described in the paragraphs, see lines 382-384.

The crustal data here refer to ordinary continental crustal data.

2. Since EMI and EM2 data have no impact on the full discussion, we have decided to delete this controversial point after late consideration.
3. The citations for crustal data here refer to common crustal data, and we have corrected the citations for crustal isotope data.
4. The content in the figure has been modified as required.

discussion

Comments 25: line-434 I don't know what this means. Fig. 10 simply shows variation diagrams of Sm, La, Nb, Ta vs. Zr.

Response 25：In the process of hydrothermal alteration, some large ion lithophile elements such as Cs and K are easy to move, while REE and some HFSE such as Nb and Ta are weak. Among them, Zr and Th have the weakest activity. Therefore, we can judge the activity of other elements by studying their relations with Zr and Th. Sm, La, Nb, Ta have obvious positive correlation with Zr. It shows that REE, Nb and Ta are basically inactive during hydrothermal alteration.

Comments 26: figure10- What do the arrows in these figures mean? Note that their direction is opposite from those in Fig. 6 (starting with diorites and ending with pyroxenites).

Response 26：These arrows indicate the correlation between rare earth elements and Nb, Ta and Zr

Comments 27: line468- Are you sure? On the bottom of p. 468 the authors write: "Basaltic magmas derived from such a source peridotite must have Mg-values 68-75" where "such a source" means peridotite with Mg^# 88 – 89.

Response 27：After our later verification, there is no problem with Mg^# . Mg^# data of primary magma are quoted in Frey et al., 1978[32]

Comments 28: line540-543 meaning unclear to me I suppose you mean peridotite with either 2% spinel or 1% garnet. Otherwise you would have to explain what the other 99 or 98% are.

Response 28：we have revised to “peridotite with about 2% spinel and 1% garnet”.

Comments 29: figure12- In the figure caption you should clearly state that the numbers related to the rock-type (harzburgite, lherzolite, peridotite) are the amounts of garnet or spinel left after the degrees of partial melting (in %) that are written along the various curves.However, in my opinion this kind of modeling is entirely useless because you do not deal with primary or near-primary magmas.

Response 29：This figure can be used to judge the degree of partial melting and the composition of the source region, and can be used for the subsequent discussion on the composition and properties of the source region of the magma. This figure shows that the primary mantle of Yunhai intrusion contains peridotite with about 2% spinel and 1% garnet and has undergone partial melting of 2-10% degree.We have also made corrections to the caption. Thank you for your valuable advice.

Comments 30: line614- fractional crystallization or magma crystallization and separation of minerals?

Response 30：The problem has been modified to magma crystallization and separation of minerals

Comments 31: line670- concentrations or elemental ratios?

Response 31：This refers to the concentration of the PGE element

Comments 32: figure13a- write: lithospheric mantle and asthenospheric mantle figure13d- "vein", not "vine"

Response 32：The corresponding problems in the figure have been modified

Colclusion

Comments 33: line698- conclusions 2 and 3 not warranted

Response 33：Conclusion 2 is discussed in section 6.3, and Conclusion 3 is discussed in section 6.4

References.

Comments 34:line741- either "Miner Deposita" (abbreviated) or "Mineralium Deposita" (full name)

Response 34：Changes have been made to the abbreviated name of the journal

Comments 35: There are journals with abbreviations you use that are not familiar to me, such as GR, Nw geol, Xj Geol, Jaap, Gssp, Ags. Either use common abbreviations or spell the names out.

For example, the correct citation for Sun & McDonough (1989) might be:

S.-s. Sun & W. F. McDonough (1989) Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Magmatism in the Ocean Basins, A. D. Saunders & M. J. Norry [Eds.], Special Publication 42, Geological Society, 313–345

Response 35：We checked the abbreviations of the journals cited online.

Comments36:line-778 :This reference does not exist! I am aware of:

1. L. Rudnick & S. Gao (2005) Composition of the continental crust. In: The Crust, R. L. Rudnick [Ed.], Treatise on Geochemistry 3, Elsevier–Pergamon, Oxford, 1–64

However, this paper does not deal with isotopes.

Response 36：After our later verification that the reference did not contain crustal isotope data, I changed the reference cited by corresponding crustal isotope data

table

Comments 37: tableS3- The ordering of oxides and elements is strange. Normally, major elements are ordered in the sequenceSiO2, TiO2, Al2O3, Fe2O3, FeO, MgO, CaO, Na2O, K2O, P2O5

(that is, with the exception of P, tetravalent, trivalent, divalent and monovalent metals).

Trace elements are ordered according to their atomic numbers, that is V first, U last. Feel free to use a separate block for the REE.

Response 37：The major elements have been rearranged as required, and the trace elements have been rearranged by atomic number

Comments 38: tableS3-There are no major element data for Y-10, Y-18, Y30 and Y-31 even though you have plotted data for these samples in Fig. 6.

Response 38：The data in our suitable figure 6 are the data in the supplementary tableS3 and the previous research data cited

Comments 39 The references cited in the chart are missing

Response 39：We have added the missing references in the supplementary tables

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

 The revised version of the manuscript "The Evolution of Permian Mafic-Ultramafic Magma in the Yunhai Region ..." by Yuxuan Pei and others is about the same as the previous draft with only minimal changes made.  Most of my suggestions and questions have been adequately addressed although a few important changes were not made.  In particular, the cross-section (Figure 13) is still in need of revision. The cross section should represent Figures 2 with some of the components of Figure 1.  The Permian ophiolites are not represented, and the cross-section does not look realistic. So, I repeat; what is the reason for extension? There is no plume or back-arc tectonism or reason for partial melting of metasomatized subcontinental lithosphere.  The best way to assess the petrology of the lopoliths that are sketched would be to analyze the chill-zones. Most large layered basic intrusions such as the few that are listed on lines 624-637 and the Skaergaard and Bushveldt are tholeiitic norites or gabbros that have undergone fractional crystallization, not intrusions of "mafic-ultramafic" magma as labeled in Figure 3.  Their tectonic setting is also unlike Figure 13.  Why is that?  

Author Response

Comments 1: In particular, the cross-section (Figure 13) is still in need of revision. The cross section should represent Figures 2 with some of the components of Figure 1.  The Permian ophiolites are not represented, and the cross-section does not look realistic. So, I repeat; what is the reason for extension? There is no plume or back-arc tectonism or reason for partial melting of metasomatized subcontinental lithosphere.  The best way to assess the petrology of the lopoliths that are sketched would be to analyze the chill-zones. Most large layered basic intrusions such as the few that are listed on lines 624-637 and the Skaergaard and Bushveldt are tholeiitic norites or gabbros that have undergone fractional crystallization, not intrusions of "mafic-ultramafic" magma as labeled in Figure 3.  Their tectonic setting is also unlike Figure 13.  Why is that? 

Response 1: Thank you for your valuable advice and so many opinions have given me a lot of inspiration and I have benefited a lot. We systematically analyzed the relevant data of Tarim basalt rock samples, and systematically compared the rock series, rock geochemical characteristics, isotope characteristics and rock genesis. It is found that Yunhai rock samples and Tarim basalt samples are quite different. For detailed discussion, see Section 6.5. Therefore, we believe that the tectonic setting of the Yunhai deposit is not related to the Tarim mantle plume, but is the product of post-collision extension environment, and the corresponding dynamic model is established. For detailed discussion, see Section 6.5.

 

Reviewer 4 Report

Comments and Suggestions for Authors

The authors have made minor changes in their revised version. My major criticism, that most of their samples represent cumulates and not melts, that renders their modeling in Fig. 12 meaningless, has not been considered. Instead, they have added six lines on page 11 where they agree with my concern. Why they have not corrected their mantle melting model on page 21, I do not know. The accompanying letter in which they have commented on the reviewer's criticism was not made available to me. I have made a comment to Fig. 12 in order to show how they might save their model. Plotting data for cumulates in partial melting diagrams without any justification is clearly inappropriate, to say the least.

Also, the arrows in figures 6 and 10 are still not explained and contradict each other. If the authors do not know what the arrows mean, they should delete them, so that only lines remain or also eliminate the lines.

Comments for author File: Comments.pdf

Author Response

Abstract

Comments 1: Although there is a region in the upper mantle where spinel and garnet will co-exist, you might better write:..."peridotite with 2% spinel and/or 1% garnet"Spinel has almost no influence on the trace elements that you discuss. Therefore ist does not matter whether spinel is present or not.

Response 1: We have made corresponding modifications according to your suggestions. Thank you for your valuable comments

 

5.2 Whole-Rock Major and Trace element

Comments 2: This new (red) section is not well written. Consider the following:

Mg# of all samples from the Yunhai intrusions show a large range from 0.50 to 0.93. If Mg# between 0.68 and 0.73 represent primary basaltic magmas in equilibrium with mantle peridotite [32], the olivine pyroxenites and hornblende pyroxenites (Mg# between 0.79 – 0.84, Table S3) must represent cumulate rocks from primary or near-primary magmas. The diorites (Mg# 0.52 – 0.58, Table S3) may then represent the fractionated magmas.

Response 2: We have revised the red section according to your suggestions, thank you for your valuable suggestions

Comments 3: figure-6 I still have no idea what the meaning of the arrows in e, f and g is. If you do not know this either, delete the lines with arrows.

Response 3: We have deleted the disputed arrows from figure E, figure F, figure G

Comments 4: figure-10 Again, I have not idea what the arrows mean. Either delete the arrow-heads or explain them!

The correlations are quite good. You could try to define a primary magma composition on each of these lines, probably closer to the diorites than to the cumulate rocks. From this primary composition two lines each will originate, one line towards the diorites (indicating evolution of the fractionating magma) and another line towards the cumulates).

Response 4: The arrows in the figure were originally intended to indicate the correlation between high field strength elements, light rare earth elements and Zr, but in view of the good correlation between high field strength elements and light rare earth elements and Zr in the figure, we deleted the arrows in the figure

 

6.3. Nature of Mantle Source and Primitive Magma

Comments 5: The new (red) sentence is meaningless in this context. Mantle xenoliths are only found in volcanic or subvolcanic rocks that quickly move through the crust. Otherwise, due to the density contrast, the mantle rocks would sink in the magma column.

In crustal magma chambers, mantle xenoliths would probably sink to the bottom (like your cumulate rocks).

Response 5: We have deleted the red sentence from the paragraph as you requested

 

Comments 6: As I argued in my previous review, you cannot use these figures for your cumulate rocks. You can, maybe, use them for the diorites.

In this case you would have to argue, that, during magma evolution from primary basaltic magmas to your diorites, the ratios of elements you discuss here will not be much changed.

For example, La, Sm and Yb are incompatible elements for pyroxenes and amphibole (and highly incompatible for Ol), with La being more incompatible than Sm and Yb. During magma evolution, La/Sm and La/Yb are thus expected to increase slightly in the fractionated magma (and be lower in the cumulates, which is what your data show).

Hf is more incompatible for these minerals than Lu (compare the relative position of these elements in Fig. 7; the arrangement of elements has been derived from the generation of basalts in the mantle). Hence, Lu/Hf expected to slightly decrease during fractional crystallization in the remaining melts – your diorites (and be higher in the cumulates), again what is indicated by your data.

 

I also urge you to provide details of your modeling. To calculate the curves, you need input data, such as mineral/melt partition coefficients, the melting process (total equilibrium melting, fractional melting etc.), mineral abundances in the mantle source and fractions of minerals that enter the melt) and element concentrations in the sources. This information could be supplied in a table in the electronic appendix.

Figure12-a how much garnet in the gt-lherzolite?

 

Response 6: Thank you very much for your meticulous review and valuable suggestions. Regarding your request for details of our modeling, we regrettably find ourselves in a challenging position. The modeling presented in our paper was not conducted by our team, but rather referenced from Sun et al. (2019), which discusses the Baixintan mafic-ultramafic complex in the Jueluotage structural belt of the North Tianshan. After studying the theoretical principles of mantle magma partial melting processes (see Supplementary Table 6), we incorporated their relevant figure as Figure 12 in our paper. Despite our thorough and careful examination of the simulation details in their paper, we were unable to obtain comprehensive data, as the cited work does not provide detailed model parameters. This limitation prevents us from fully addressing your queries. However, given the striking similarities between the Yunhai and Baixintan deposits in terms of fundamental geology, petrological assemblages, mineralization types, mineral compositions, metallogenic processes, and tectonic settings, we potentially think this model reasonably reflects the characteristics of the mantle source region in the Yunhai mining district.

 

The feasibility of using REE ratios to model the mineral composition of the mantle source region is described in detail in the reference McKenzie et al., 1991[68]. The reference suggests that Inverse theory is used to calculate the melt distribution required to produce the rare earth element concentrations in a wide variety of terrestrial and extra-terrestrial magmas. The concentrations of the major and trace elements in the source regions are assumed to be the same as those for the bulk Earth, and the peridotite mineralogy calculated from the mineral compositions by least squares. Rare earth element oartition coefficients are then used for inversion, assuming the melt generation is by fractional melting. The mean composition of the magmas is taken to be an estimate of the average composition of the melt. Among them, the partition coefficient of mineral/melt is quoted from Harrison, 1981 For specific analysis and demonstration, and corresponding mathematical models and formulas, see reference [68], which also demonstrates the feasibility of this method from different tectonic environments The specific data used in this reference have been added to Table S6

 

Conclusion

The concentrations of the major and minor elements in source regions are assumed to be the same as those for bulk Earth, and the peridotite mineralogy calculated from the mineral compositions by least squares. Rare earth element partition coefficients are then used for inversion, assuming the melt generation is by fractional melting. The mean composition of the magmas is taken to be an estimate of the average composition of the melt.

 

Comments 7: This statement is too strong in view of the lack of primary magmas in your samples suite.Maybe repclace by "may have been derived"

Response 7: We have revised the sentence according to your suggestion, thank you for your valuable comments

 

Reference

Comments 8: Commonly used journal abbreviations can be found, for example, at:

https://woodward.library.ubc.ca/woodward/research-help/journal-abbreviations/

Please, check for consistency. Either use abbreviated or full names. Names such as GCA, Jaap or Gr are neither abbreviated nor full names.

Response 8: Thank you for providing the query URL, we have changed the name of the references to abbreviations according to the requirements of the journal

Table S3

Comments 9: Table S3:

at the end, replace "loss of ignition"by "loss on ignition"

Response 9: We have corrected the problem as requested

 

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

The changes made to the revised manuscript by Yuxuan Pei and others seem to adequately answer most of my remaining questions.