the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Petrogenesis and tectonic setting of late Paleoproterozoic diorites in the Trans-North China Orogen
Abstract. The Xiong’er volcanic rocks and mafic dike swarms mark a significant magmatic event after the amalgamation of the North China Craton (NCC) in the Paleoproterozoic, yet their tectonic origins remain controversial. Several Paleoproterozoic diorite intrusions have received widespread attention recently. Their genesis and geological significance are crucial for understanding the evolution of the NCC. In this study, we report zircon U-Pb ages and geochemical data of the Jiguanshan diorite. The diorites in the Trans-North China Orogen, including the Jiguanshan diorite, have comparable element and isotopic geochemical characteristics. The weighted mean average of initial 87Sr/86Sr and εNd(t) values is 0.7052 ±0.0003 and -6.5 ±0.2, respectively. The initial Pb isotope compositions of these diorite samples do not show significant enrichment of radiogenic lead. In terms of Sr-Nd-Pb isotope compositions and Nb/Ta, Ba/Th, and Sr/Th ratios, these diorites differ from the Xiong’er volcanic rocks and mafic dike swarms. Our results suggest that these diorites originated from the basaltic lower crust, rather than from the enriched subcontinental lithospheric mantle. Whole-rock and zircon trace element geological tectonic diagrams indicate that the diorites formed in a rift environment. These diorites mark a crustal-origin rock shift from orogenic-related magmatism to intraplate magmatism during the post-collisional extensional stage.
- Preprint
(3025 KB) - Metadata XML
-
Supplement
(113 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2115', Anonymous Referee #1, 17 Jul 2025
Dear editor,
I have carefully read the article “Petrogenesis and tectonic setting of late Paleoproterozoic diorites in the Trans-North China Orogen”.
The authors systematically studied the ca. 1.78–1.75 Ga Jiguanshan diorite and other coeval diorites from the southern margin of the NCC. These diorites are inferred to have similar geochemical characteristics and may have formed during a single magmatic episode. This, together with the whole rock geochemical and Sr-Nd-Pb isotopic compositions, as well as zircon Hf isotopic compositions, enabled a better understanding of the tectonic environment and evolution of the NCC during the late Paleoproterozoic.
The science of this manuscript looks fine to me, but I also noticed some spellings and grammar mistakes, along with incomplete sentences. Furthermore, I would also like to suggest the authors slightly re-organize structures. Therefore, I believe this manuscript can be published after a careful revision.
Some of my major concern,
1. Please rearrange the abstract. Key data such as ages of the investigated diorite are not presented. The first sentence is also misleading, the Xiong’er volcanic rocks are not the focus of this study. The inferred conclusions are not supported by data. Please present more data patterns in the abstract.
2. Rearrange the Introduction
The scientific questions mentioned in the introduction are not clearly presented, and not closely linked with your discussion.
Verifying the necessity of your research is very important.
It would be better if you can slightly revise the introduction, or perhaps you can revise the discussion.
3. The research background should be more detailed. Give us more information on the assembly of the north China craton, related metamorphic and igneous records, as well as related structures. Tell us more about the Xionger group, the Taihua group, their relations.
4. I believe that models for the distribution of the tectonic units of the north China craton are very much varied. There are some other models that are not mentioned by the article, but should be mentioned.
5. line 38 what is the ancient basement rocks? Be specific
6. line 40-41, not logically correct.
7. I don’t see strong evidence of rifting setting. Please expand your discussion on the tectonic settings.
If the diorites and the xionger groups were coeval, but the xionger volcanics rocks were formed in an arc setting. How can you explain the within plate setting of these diorites.
Figure 1, revise.
Figure 2, any good field photo?
Citation: https://doi.org/10.5194/egusphere-2025-2115-RC1 -
AC1: 'Reply on RC1', Jun He, 08 Sep 2025
Dear Editors and Reviewers,
We would like to express our sincere gratitude for taking the time to review and evaluate our manuscript. The suggestions from the reviewers were extremely insightful and constructive. We have carefully considered each comment and have made the corresponding revisions to enhance the quality of the manuscript. Please find below our detailed explanations.
We sincerely hope that our research meets your expectations for publication. All authors have reviewed the revised manuscript and agree with its resubmission.
Yours sincerely,
Zhiyi Wang and Jun He on behalf of the other authors
Response to reviewer
Reviewer 1:
- Please rearrange the abstract. Key data such as ages of the investigated diorite are not presented. The first sentence is also misleading, the Xiong’er volcanic rocks are not the focus of this study. The inferred conclusions are not supported by data. Please present more data patterns in the abstract.
We have revised the abstract according to your comments to better highlight the key work and contributions of this study.
- Rearrange the Introduction
The scientific questions mentioned in the introduction are not clearly presented, and not closely linked with your discussion. Verifying the necessity of your research is very important. It would be better if you can slightly revise the introduction, or perhaps you can revise the discussion.
We have modified it accordingly. The introduction has been revised to better highlight the key issues addressed in this paper.
- The research background should be more detailed. Give us more information on the assembly of the north China craton, related metamorphic and igneous records, as well as related structures. Tell us more about the Xiong’er group, the Taihua group, their relations.
We have substantially enhanced the research background section with expanded discussions on: (1) the assembly processes of the North China Craton; (2) the models for the distribution of the tectonic units of the North China Craton; and (3) detailed information of the Xiong’er and Taihua Groups. These revisions will provide clearer context for our work.
- I believe that models for the distribution of the tectonic units of the north China craton are very much varied. There are some other models that are not mentioned by the article, but should be mentioned.
We have modified it accordingly.
- line 38 what is the ancient basement rocks? Be specific
We have modified it accordingly.
- line 40-41, not logically correct.
We have modified it accordingly.
- I don’t see strong evidence of rifting setting. Please expand your discussion on the tectonic settings.
If the diorites and the Xiong’er groups were coeval, but the Xiong’er volcanics rocks were formed in an arc setting. How can you explain the within plate setting of these diorites.
In the revised manuscript, we have incorporated additional geological background information. At the beginning of the “Tectonic Setting” section, we explicitly evaluated whether the diorite samples exhibit typical geochemical features of intermediate rocks from convergent plate margins. Our results do not support a subduction-related origin. Building on these findings, we further discuss the petrogenesis of the diorites in relation to the Xiong’er volcanic rocks, and place them within the regional tectonic evolutionary framework.
- Figure 1, revise.
We have updated Figures 1 to ensure the figure reflect the most current understanding of the regional geology.
- Figure 2, any good field photo?
Thank you for the suggestion. We've enhanced the field photographs.
Citation: https://doi.org/10.5194/egusphere-2025-2115-AC1
-
AC1: 'Reply on RC1', Jun He, 08 Sep 2025
-
RC2: 'Comment on egusphere-2025-2115', Anonymous Referee #2, 29 Jul 2025
This paper identify and presents new Zircon U-Pb isotopic age and geochemical analyses of the ca. 1.78 Ga Jiguanshan diorite in the North China Craton. The rocks likely formed from basaltic lower crust in a rift setting. The paper is in general well written, the topic is suitable for this journal, and the discussion also seems reasonable, except some interpretations may also need more evidence. Therefore, it needs moderate revision considering the following comments.
Detailed comments:
- Lines 30-33, I agree with that the studied diorite formed in a rift setting. However, it is hard to say they mark a crustal-origin rock shift from orogenic-related magmatism to intraplate magmatism. How to define or constrain a tectonic transition?
On the other hand, I suggest the authors distinguish between rift (intraplate) and post-collisional settings, as the latter is continuation of a collisional setting, but the former is not.
- Introduction part, lines42-47, since the previous research has been very clear that “...the craton experienced multiple rift phases, with the Xiong’er rift being the first rift formed after the assembly, resulting in the formation of the c. 1780 Ma Xiong’er volcanic rocks and contemporaneous mafic dyke swarms”, why do the origin and tectonic setting of the Xiong’er volcanic rocks remain controversial?What is the crux or cause of these dispute? Please add related content.
- Line 83, please change “fault”to “Fault”.
- Fig.1, the legend does not correspond to the pattern in the figure. Please recheck.
- Lines 185-191, please add MSWD values for the calculated ages.
- Besides, the zircon spots with ages of ca. 1.6 Ga and 1.93 Ga may not belong to the same population as the zircons with ages of ca. 1.77-1.74 Ga. So they should be discussed separately instead of being taken into account.
- Line 200, add “wt. %”after 5.57, i.e., from 5.57 wt. % to 6.03 wt. %.” Please check the whole text for the similar problems.
- Line 240,the SiO2contents are different from those in line 199, why?
- Discussions on the sources of the diorite, such as lines 315-320, I recommend to use the isotopic data first, which is more effective to eliminate relevant source regions.
- Line 368, please delete the superfluous “.”.
- Lines 423-426, in fact, most of the samples fell into the rift/ anorogenic environment fields, consistent with the previous discussion. But why do the author say ”These may indicate that the post-collisional extension during this period may ultimately lead to rift evolution continuously and progressively.” How and when lead to?
- About the tectonic setting, similar to comment 1, it is still necessary to distinguish whether the ca. 1.78 Ga diorite was formed in a rift or a post-collision environment? Meanwhile,please comprehensively define the tectonic background in combination with other coeval rock combinations (e.g., granitoids) and research cases of the same period.
Citation: https://doi.org/10.5194/egusphere-2025-2115-RC2 -
AC2: 'Reply on RC2', Jun He, 08 Sep 2025
Dear Editors and Reviewers,
We would like to express our sincere gratitude for taking the time to review and evaluate our manuscript. The suggestions from the reviewers were extremely insightful and constructive. We have carefully considered each comment and have made the corresponding revisions to enhance the quality of the manuscript. Please find below our detailed explanations.
We sincerely hope that our research meets your expectations for publication. All authors have reviewed the revised manuscript and agree with its resubmission.
Yours sincerely,
Zhiyi Wang and Jun He on behalf of the other authors
Reviewer 2:
- Lines 30-33, I agree with that the studied diorite formed in a rift setting. However, it is hard to say they mark a crustal-origin rock shift from orogenic-related magmatism to intraplate magmatism. How to define or constrain a tectonic transition?
On the other hand, I suggest the authors distinguish between rift (intraplate) and post-collisional settings, as the latter is continuation of a collisional setting, but the former is not.
We have replaced the phrase “mark a shift” with the more cautious formulation “could be indicative of a potential transition”. On one hand, we emphasized the transitional geochemical characteristics of the diorites. On the other hand, we supplemented the regional tectonic background and evolutionary history in the text, highlighting the significance of the diorites. These revisions have been made to clarity of our argument.
- Introduction part, lines 42-47, since the previous research has been very clear that “...the craton experienced multiple rift phases, with the Xiong’er rift being the first rift formed after the assembly, resulting in the formation of the c. 1780 Ma Xiong’er volcanic rocks and contemporaneous mafic dyke swarms”, why do the origin and tectonic setting of the Xiong’er volcanic rocks remain controversial? What is the crux or cause of these dispute? Please add related content.
We have modified it accordingly. We have expanded the geological background and cause of these disputes in the revised manuscript.
- Line 83, please change “fault”to “Fault”.
We have modified it accordingly.
- Fig.1, the legend does not correspond to the pattern in the figure. Please recheck.
Lines 185-191, please add MSWD values for the calculated ages.
We have modified it accordingly.
- Besides, the zircon spots with ages of ca. 1.6 Ga and 1.93 Ga may not belong to the same population as the zircons with ages of ca. 1.77-1.74 Ga. So they should be discussed separately instead of being taken into account.
We used zircon trace element analysis to constrain the tectonic setting. Outliers identified by Isoplot were excluded during data processing. All analytical spots with La contents >1 ppm were removed to ensure data reliability. Some zircons show evidence of Pb loss in Fig. 4. Nevertheless, the average 207Pb/206Pb ages still exhibit a relatively concentrated distribution in Fig.4. Few outliers cannot undermine the overall coherence. Furthermore, all diorite samples in this study were processed using consistent data processing, and the large number of valid data points significantly reduces the potential impact of individual anomalies on the general interpretation.
- Line 200, add “wt. %”after 5.57, i.e., from 5.57 wt. % to 6.03 wt. %.” Please check the whole text for the similar problems.
We have modified it accordingly.
- Line 240, the SiO2 contents are different from those in line 199, why?
In the Results section, the contents described refer specifically to the samples from the Jiguanshan diorite. In the Discussion, we compare them with the compositional ranges of regional diorites of similar age.
- Discussions on the sources of the diorite, such as lines 315-320, I recommend to use the isotopic data first, which is more effective to eliminate relevant source regions.
We provide a comprehensive analysis of the source of the diorites by integrating multiple lines of evidence, including elemental geochemistry and isotopic compositions. By comparing isotopic characteristics with those of potential source reservoirs, such as the basement rocks of the Taihua Group, the contemporaneous Xiong’er volcanic rocks, regional mafic dike swarms, and crustal rocks’ characteristics, we conclude that these diorites were derived from partial melting of middle to lower crustal rocks.
- Line 368, please delete the superfluous “.”.
We have modified it accordingly.
- Lines 423-426, in fact, most of the samples fell into the rift/ anorogenic environment fields, consistent with the previous discussion. But why do the author say “These may indicate that the post-collisional extension during this period may ultimately lead to rift evolution continuously and progressively.” How and when lead to?
We have enhanced the discussion by incorporating additional regional geological context and a more detailed analysis of the tectonic evolution. Following the ~1.85 Ga collisional event, the North China Craton entered a prolonged post-collisional extensional stage. Initial crustal thickening and remelting generated extensive crust-derived granites. Subsequent slab breakoff and gravitational collapse induced mid- to upper-crustal extension and felsic magmatism. By ~1.78 Ga, continued lithospheric thinning led to asthenospheric upwelling and melting of subduction-modified lithospheric mantle. The magmatism then evolved into A-type granites and alkaline rocks, marking a transition to an anorogenic intracontinental extensional setting. The 1.78 Ga crust-derived diorites exhibit transitional features, retaining orogenic signatures while shifting toward an intraplate affinity, reflecting prolonged extension after craton amalgamation.
- About the tectonic setting, similar to comment 1, it is still necessary to distinguish whether the ca. 1.78 Ga diorite was formed in a rift or a post-collision environment? Meanwhile, please comprehensively define the tectonic background in combination with other coeval rock combinations (e.g., granitoids) and research cases of the same period.
In the revised manuscript, we have enhanced the regional geological background, summarized tectonic models and debates related to other coeval rock. Furthermore, we have added a concluding section discussing the tectonic evolution and significance of the diorites. These revisions improve the clarity of our arguments.
Citation: https://doi.org/10.5194/egusphere-2025-2115-AC2
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
414 | 42 | 7 | 463 | 11 | 32 | 41 |
- HTML: 414
- PDF: 42
- XML: 7
- Total: 463
- Supplement: 11
- BibTeX: 32
- EndNote: 41
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1