the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Oct 2025
Comparative study of Low-grade metamorphic Precambrian supracrustal rocks and HP–UHP Rocks in the South Altyn Tagh: Insights into subduction-exhumation
Abstract. Low-grade metamorphic (LGM) rocks are widespread in high- to ultrahigh-pressure (HP–UHP) subduction zone yet frequently neglected in orogenic evolution. Establishing their spatiotemporal relationship with HP–UHP rocks and comparing protolith affinities are key to deciphering subduction zone architecture and exhumation dynamics. Here we investigate LGM Precambrian supracrustal rocks in the South Altyn Tagh (SAT) through field investigations, chronology and geochemical analysis, and comparison with HP–UHP rocks. Granites emplaced at 933–898 Ma, exhibiting crustal melting and syncollisional granite affinities, serving as robust markers for Rodinia convergence, consistent with protolith of regional HP–UHP granitic gneiss. Mafic dyke emplaced at ~806 Ma, exhibiting within-plate basalt (WPB) affinities, serving as markers for regime transition from collision to extension, consistent with protolith of regional eclogite and garnet pyroxenite. (Meta-)sedimentary rocks deposited during 939–932 Ma, exhibiting Taxidaban Group (Central Altyn block, CAB) affinities. Results reveal these LGM rocks lack significant Cambrian metamorphic (HP–UHP) overprinting but share protolith ages and characteristics with HP–UHP units, indicating shared formation origins yet distinct pre-subduction tectonic affiliations. This comparison implies that these supracrustal rocks may represent the non-subducted overlying plate of the SAT Early Paleozoic subduction zone. Synthesizing our data with existing metamorphic records, we propose that the current spatiotemporal distribution of LGM and HP–UHP rocks in the SAT resulted from: (1) Early Paleozoic whole-slab continental subduction, followed by (2) differential exhumation and late-stage modification.
- Preprint
(7272 KB) - Metadata XML
-
Supplement
(73 KB) - BibTeX
- EndNote
Status: open (until 18 Nov 2025)
-
CC1: 'Comment on egusphere-2025-4830', Axe Tom, 28 Oct 2025
reply
-
AC1: 'Reply on CC1', Tuo Ma, 29 Oct 2025
reply
(1) The coexistence of high-grade metamorphic rocks with low-grade or unmetamorphosed rocks is a common phenomenon in the South Altyn.
(2) In the field, LGM rocks and the high-pressure to ultrahigh-pressure rocks of the South Altyn exhibit fault contacts or unconformable contacts.
(3) No characteristic high-grade metamorphic minerals such as garnet, kyanite, or sillimanite have been observed in the petrography of LGM rocks.
(4) Zircons from LGM rocks show no evidence of zircon growth at 500 Ma.
(5) The trace element patterns of zircons from LGM rocks do not indicate elemental partitioning associated with the growth of characteristic metamorphic minerals such as garnet.
Therefore, we propose that these LGM rocks did not undergo deep subduction.
Citation: https://doi.org/10.5194/egusphere-2025-4830-AC1
-
AC1: 'Reply on CC1', Tuo Ma, 29 Oct 2025
reply
-
CC2: 'Comment on egusphere-2025-4830', Emma watson, 29 Oct 2025
reply
Model constraints and dynamic implications for the SAT subduction zone architecture:
The study aims to clarify the relationship between LGM and HP-UHP rocks to provide key constraints on resolving uncertainties in the SAT subduction zone architecture (such as whether the LGM rocks represent the overlying plate material, and the compositional consistency between the upper and lower plates). Based on your new findings and the systematic review leading to the proposed model of "consistent subduction but differential exhumation", how does this model specifically explain the present spatiotemporal distribution pattern of metamorphic rocks with varying grades in the SAT region?Citation: https://doi.org/10.5194/egusphere-2025-4830-CC2 -
AC2: 'Reply on CC2', Tuo Ma, 29 Oct 2025
reply
Previous studies and the findings of this paper indicate that the HP-UHP granitic gneisses and the LGM granites share the same protolith age and geochemical characteristics, and their formation is related to the assembly of the Rodinia supercontinent. The HP-UHP eclogites and the LGM mafic rocks share the same within-plate basalt affinity and have similar protolith formation ages, likely associated with the breakup of the Rodinia supercontinent. Therefore, we propose that the protoliths of these two rock types are identical. In the CC1 discussion, we elaborated on the entirely different metamorphic processes experienced by the LGM rocks and the HP-UHP rocks: the HP-UHP rocks underwent deep subduction, while the LGM rocks did not; therefore, we propose that the non-subducted rocks represent the upper plate material of the subduction zone.
The model of "consistent subduction" but "partitioned exhumation and modification" applies to the high-grade metamorphic rocks preserved in the South Altyn that exhibit different metamorphic grades, as shown in Fig. 13. Although these rocks currently display various characteristics in petrography, such as amphibolite facies, granulite facies, and eclogite facies, detailed mineralogical studies reveal that their metamorphic grades can all be traced back to the eclogite facies, with the peak metamorphic age concentrated around 500 Ma. Therefore, we conclude that these rocks experienced "consistent subduction" but "partitioned exhumation and modification".
Citation: https://doi.org/10.5194/egusphere-2025-4830-AC2 -
AC4: 'Reply on CC2', Tuo Ma, 29 Oct 2025
reply
In the east section of South Altyn (e.g., Danshuiquan, Yinggelisayi), HP–UHP rocks form dome-like structures and experienced eclogite-facies metamorphism at ~500 Ma, followed by near-isothermal decompression and HP granulite-facies retrogression at ~480 Ma (Gai et al., 2022). Partial melting enhanced buoyancy, supporting diapiric exhumation (Gai et al., 2024). In contrast, the western regions (e.g., Jianggalesayi, Younusisayi) show mixed HP–UHP and LGM rocks in banded distributions, with eclogite-facies metamorphism at ~500 Ma but younger granulite-facies retrogression (460–450 Ma) and cooling-decompression paths, indicating subduction-channel exhumation. These spatial and metamorphic differences confirm divergent exhumation processes: diapirism in the east and subduction-channel exhumation in the west.
Gai, Y. S., Liu, L., Zhang, G. W., et al.: Differential exhumation of ultrahigh-pressure metamorphic terranes: A case study from South Altyn Tagh, western China, Gondwana Research, 104, 236–251, 2022.
Gai, Y. S., Ma, T., Liu, L., et al.: Partial melting of HP–UHP felsic gneiss in the South Altyn Tagh reveals the rapid exhumation of a deeply subducted slab, Lithos, 488–489, 107835, https://doi.org/10.1016/j.lithos.2024.107835, 2024.
Citation: https://doi.org/10.5194/egusphere-2025-4830-AC4
-
AC2: 'Reply on CC2', Tuo Ma, 29 Oct 2025
reply
-
CC3: 'Comment on egusphere-2025-4830', Sun scott, 29 Oct 2025
reply
The study indicates that the LGM and HP-UHP rocks share consistent formation ages and protolith characteristics, yet it is inferred that they might have belonged to distinct tectonic units prior to Early Paleozoic deep subduction. On what specific chain of evidence for protolith comparison is this key inference based (e.g., detailed detrital zircon provenance comparison, discrimination of protolith formation environments, and field geological evidence such as their spatial distribution and contact relationships)? How does this evidence support the core model that they experienced "consistent subduction" but "partitioned exhumation and modification", rather than alternative tectonic scenarios (e.g., they originally belonged to the same stratigraphic sequence which subsequently underwent differential metamorphic responses during subduction)?
Citation: https://doi.org/10.5194/egusphere-2025-4830-CC3 -
AC3: 'Reply on CC3', Tuo Ma, 29 Oct 2025
reply
Previous studies and the findings of this paper indicate that the HP-UHP granitic gneisses and the LGM granites share the same protolith age and geochemical characteristics, and their formation is related to the assembly of the Rodinia supercontinent. The HP-UHP eclogites and the LGM mafic rocks share the same within-plate basalt affinity and have similar protolith formation ages, likely associated with the breakup of the Rodinia supercontinent. Therefore, we propose that the protoliths of these two rock types are identical. In the CC1 discussion, we elaborated on the entirely different metamorphic processes experienced by the LGM rocks and the HP-UHP rocks: the HP-UHP rocks underwent deep subduction, while the LGM rocks did not; therefore, we propose that the non-subducted rocks represent the upper plate material of the subduction zone.
The model of "consistent subduction" but "partitioned exhumation and modification" applies to the high-grade metamorphic rocks preserved in the South Altyn that exhibit different metamorphic grades, as shown in Fig. 13. Although these rocks currently display various characteristics in petrography, such as amphibolite facies, granulite facies, and eclogite facies, detailed mineralogical studies reveal that their metamorphic grades can all be traced back to the eclogite facies, with the peak metamorphic age concentrated around 500 Ma. Therefore, we conclude that these rocks experienced "consistent subduction" but "partitioned exhumation and modification".
Citation: https://doi.org/10.5194/egusphere-2025-4830-AC3
-
AC3: 'Reply on CC3', Tuo Ma, 29 Oct 2025
reply
-
RC1: 'Comment on egusphere-2025-4830', Anonymous Referee #1, 11 Nov 2025
reply
Ma et al. present a study of low-grade metamorphic rocks from the Southern Altyn Tagh. They incorporate zircon geochronology and whole rock geochemical data and compare these against data from the HP-UHP rocks within the same terrane. They interpret the low-grade rocks to have similar protoliths to the high-grade rocks but to have not witnessed the high-grade metamorphism at all instead preferring an upper plate origin for the low-grade rocks with the high-grade rocks emplaced either as diapirs (eastern section) or along channel (western section). Overall the data is of a reasonable quality and their interpretations seem mostly valid, below are some small technical questions and attached are some annotations to the PDF with stylistic improvements.
Zircon dating
It is not stated in the main text nor the analytical methods how concordance in the zircon isotopic data is being determined.
You should also state somewhere what your error reporting in figures and text is. I presume from the supplementary tables it is 1 s.e. but it should be stated in the text.
Also, were there no discordant analyses? You should put all data even those you consider discordant into your data tables.
How robust really are your maximum depositional ages, the only sample with a high number of analyses (16A96) has two concordant late Neoproterozoic zircons; the remaining samples have very limited datasets (n<30) and small populations can easily have been missed. So, what is the significance of the late Neoproterozoic ages and why are they ignored in the text? I realise that you might point to the early Neoproterozoic igneous rocks to suggest these ages are spurious, but following your logic of the tectonic model it would also seem that the Altyn Complex should not be a coherent sedimentary sequence, post early Neoproterozoic sedimentary cycles (that should exist given your supposed >100 myr of rifting) could be obscured by the metamorphic overprinting?
Geochemical data and interpretation
How valid is it to make interpretations based on the use of mobile major oxides and trace elements in rocks that have undergone metamorphism? The trace element data clearly show that the samples are only superficially similar.
Also, is it not unusual that crust-derived S-type granites do not contain significant populations of older xenocrystic zircons?
Tectonic model
Is there robust evidence for an oceanisation stage between Central and Southern Altyn blocks as depicted in Fig. 14? If you suggest there are common protoliths to the HP-UHP rocks and the low-grade metamorphic rocks it would seem simpler to suggest a highly stretched continental crust rather than common origin, c.400 myr of ocean separation, and then re-assembly at more or less the same position
-
RC2: 'Comment on egusphere-2025-4830', Anonymous Referee #2, 13 Nov 2025
reply
This paper presents new geochemical and geochronological data from the South Altyn Tagh, making it may of interesting for the regional researchers. However, after reading the paper, I found the reasoning is not conclusive, and the significance of paper is not clear, and hence lack of novel insights into the region.
My main concern is that the paper's scope is overly broad, making it difficult to identify a specific research question. While a comparative geological history between HP–UHP lenses and their host rocks could be of broad interest to the community, such a focus is not adequately substantiated in the current study, which highlights a comparison between supracrustal rocks and HP-UHP rocks in the South Altyn Tagh (SAT). It has so far not sufficiently justified why this comparison is scientifically meaningful. By definition, supracrustal rocks are not expected to share a co-genetic evolution with deeply subducted HP-UHP rocks, right? In this context, a more focused research question is essential; otherwise, the current argumentation limits the novelty of this work.
Another concern involves the unclear geological context of both the supracrustal and HP–UHP rocks discussed in the study. First, the spatial relationship between these two units remains poorly defined, leading to potential confusion. For example, in lines 149–152, the supracrustal rocks (labeled LGM) and the HP–UHP rocks are described as distributed in northern and southern domains, separated by fault contact. In contrast, lines 582–585 describe the HP–UHP rocks as occurring in dome-like structures, while lines 598–599 suggest they may form a banded or zonal distribution pattern. Such inconsistencies undermine the clarity and interpretability of the geological framework. Second, the classification of supracrustal rocks appears ambiguous. Specifically, placing granites within this category—and the description of them as lower-grade rocks (e.g., line 150)—is conceptually problematic. Furthermore, although the supracrustal rocks are generally interpreted as lacking significant Cambrian metamorphic overprint, elsewhere in the text (lines 508–518) similar rocks are reported to have experienced deep continental subduction under HP–UHP conditions. These contradictory interpretations create substantial uncertainties. To strengthen the manuscript, it is critical to establish a clear and internally consistent description of the nature of the rocks throughout the text.
Yet, the proposed shared origins between the investigated supracrustal rocks and HP–UHP rocks, based solely on whole-rock geochemical comparisons (specifically major and trace elements), appears somewhat arbitrary. To enhance the robustness of this interpretation, I strongly recommend that the authors incorporate corroborative evidence from isotopic geochemistry, thereby strengthening the credibility of their conclusions.
In addition, the data of this study are insufficient to constrain the role of Rodinia supercontinent evolution (e.g., its amalgamation or breakup) in the studied tectonic processes. Given that this aspect extends beyond the evidential scope of the current dataset, I recommend that the authors omit speculative interpretations related to Rodinian geodynamics from their tectonic model.
Some specific comments:
1) L11, please clarify the term “low-grade metamorphic rocks”. In most literatures, this classification is mainly defined by temperature, but it should be noted that low-grade metamorphic rocks are not necessarily petrogenetic opposites to HP-UHP rocks (for instance blueschist).
2) L44, again, the authors should provide a clear definition for “low-grade metamorphic rocks” as used in this paper. It is not sound to categorize “gneisses” as low-grade rocks.
3) L58-59, please clarify “deep to ultra-deep continental subduction zone”;
4)L129-130, The statement regarding which geological unit preserves the record of the Rodinia supercontinent assembly and the Proto-Tethys Ocean evolution is ambiguous. Please specify whether this refers to one or both of them.
5) L152, It seems that the Fig. 1c could not tell the spatial dimensions of the “two rock units”, and a better revised map is recommended.
6) L154, Grouping the felsic and mafic intrusions as low-grade rocks is conceptually problematic, as intrusive are not, by definition, metamorphic. If you want, you could say ... rocks from the low-grade sequence; Additionally, Fig. 2c appears a repetition of 2b, one of them should be removed.
7) Fig. 3, The scale bars in all photomicrographs are illegible. Please replace them with clear, high-contrast versions.
8) Fig. 4 it is not necessary to repeat the sample names all the time;
9) L375-376, please clarify “sandy rocks deposited by continental crust”, which is vague.
Citation: https://doi.org/10.5194/egusphere-2025-4830-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 149 | 57 | 25 | 231 | 20 | 7 | 8 |
- HTML: 149
- PDF: 57
- XML: 25
- Total: 231
- Supplement: 20
- BibTeX: 7
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The text points out that the widely distributed LGM rocks in the SAT lack reliable Cambrian HP metamorphic records. To distinguish between the two competing hypotheses of "deeply subducted but overprinted" versus "never deeply subducted and tectonically interlard", what specific geological, petrological, or geochemical criteria (e.g., field contact relationships, mineral inclusions, trace element geochemistry, geochronology and trace element analysis of zircon/monazite) did this study employ to effectively identify potential early deep subduction signals in the LGM rocks that may have been obscured by later overprinting? How does this evidence rule out the possibility that the LGM rocks never underwent deep subduction?