Does Large-Scale Crustal Flow Shape the Eastern Margin of the Tibetan Plateau? Insights From Episodic Magmatism of Gongga-Zheduo Granitic Massif.

The mechanisms driving crustal deformation and uplift of orogenic plateaus are fundamental to continental tectonics. Large-scale crustal flow has been hypothesized to occur in eastern Tibet, but it remains controversial due to a lack of geologic evidence. Geochemical and isotopic data from Cenozoic igneous rocks in the eastern Tibet-Gongga-Zheduo intrusive massif, provide a way to test this model. Modeling results suggest that Cenozoic magmas originated at depths of ∼30-40 km, the depth that crustal flow has been postulated to occur at. Detailed isotopic analyses indicate that the igneous rocks are derived from partial melting of the local Songpan-Ganzi crust, arguing against a long-distance crustal flow. Episodic magmatism during the Cenozoic showing a repeated shifting of magmatic sources can be correlated with crustal uplift. The continued indentation of the Indian Block and upwelling of the asthenosphere contribute to the crustal deformation, magmatism, and uplift.

Introduction

The Tibetan Plateau was created by the India‐Asia continental collision during the Cenozoic and is a natural laboratory to test models of continental tectonics. The mechanisms of crustal deformation, uplift, and outward expansion of the plateau are among the most controversial and unresolved aspects of the collision. The Asian block has experienced relatively diffuse deformation, located far away from the Indo‐Asian suture zone (England & Houseman, 1986; England & Molnar, 1997). Some researchers have ascribed crustal thickening and outward expansion of the plateau as a result of escape tectonics along major strike‐slip faults (Molnar & Tapponnier, 1975; Tapponnier et al., 2001). Others have proposed that the indentation of India is driving large‐scale lower crustal flow and redistributing mass, to cause crustal thickening and outward expansion of the plateau (Clark & Royden, 2000; Royden et al., 1997).

The large‐scale (>1,000 km) crustal flow model can explain why there is high topography in the absence of significant crustal shortening and a low topographic gradient along the eastern margin of the Tibetan Plateau (EMTP; Figure 1; Clark & Royden, 2000; Royden et al., 1997; Schoenbohm et al., 2006). The central Tibetan Plateau, with high elevations and thickened crust, provided a lateral pressure gradient required for the crustal flow (Clark & Royden, 2000; Royden et al., 1997). Furthermore, low‐velocity and high‐conductivity zones with high Poisson's ratios have been reported in the mid‐lower crust of the EMTP and Northern Qiangtang terrane (NQT), interpreted to be mobile mid‐lower crust (Figure 1; Bai et al., 2010; Bao et al., 2015; Kong et al., 2016; Q. Y. Liu et al., 2014; C.‐Y. Wang et al., 2010). Therefore, a crustal flow “channel” was proposed to extend from the NQT to the EMTP, at a depth of ∼25–40 km (Figure 1; Bai et al., 2010; Bao et al., 2015; Clark, Bush, et al., 2005; Clark & Royden, 2000).

Figure 1

Maps of the Tibetan Plateau and Gongga‐Zheduo granitic massif and conceptual models for crustal thickening and tectonic uplift in eastern Tibet. (a) Distribution of geophysical anomalies and Cenozoic magmatic rocks in the Tibetan Plateau. Vs perturbation at 30 km depth are from Y. Yang et al. (2012). Pn low‐velocity anomalies at the uppermost mantle are from Zhou and Lei (2016). Light‐yellow arrows mark the inferred location of hypothesized flow channels (Bai et al., 2010). The distribution of magmatic rocks is from Chung et al. (2005) and Hou et al. (2006). Numbers represent suture zones and faults: 1‐Eastern Kunlun‐Animaqing suture zone; 2‐Jinshajiang suture zone; 3‐Longmuco‐Shuanghu suture zone; 4‐Bangong‐Nujiang suture zone; 5‐Indus‐Yarlung Zangbo suture zone; 6‐Ganzi‐Litang suture zone; 7‐Kunlun fault; 8‐Longmenshan thrust fault; 9‐Xianshuihe‐Xiaojiang fault; 10‐Ailaoshan‐Honghe fault; 11‐Sagaing fault. (b) Large‐scale crustal flow model with crustal thickening caused by channelized exotic crustal materials (after Clark, Bush, et al., 2005; Clark & Royden, 2000). (c) Soft crust model with crustal thickening caused by local crustal diffusive deformation. (d) A simplified geological map of the Gongga‐Zheduo granitic massif. Yellow circles denote sample locations. The ages in red are data from this study. The ages in black are from previous studies (Lai & Zhao, 2018; H. Li & Zhang, 2013; H. Li, Zhang, et al., 2015; Searle et al., 2016).

Crustal flow is a solid‐state process (e.g., mylonitic shear zone) that initiates under appropriate stress conditions and when temperatures exceed 400°C–500°C (e.g., Kruse et al., 1991; MacCready et al., 1997; McKenzie et al., 2000). For a large‐scale crustal flow (>1,000 km), higher crustal temperatures are required (i.e., >700°C; Kruse et al., 1991). The Tibetan plateau has high modern geothermal gradients (>25°C/km), which are thought to be related to limited erosion and high radiogenic heat production (Whittington et al., 2009). The high heat flow provides a basis for the large‐scale crustal flow model and also explains magmatism which involves partial melting of the lower middle crust (H. Li & Zhang, 2013; Searle et al., 2016).

Here, we present a test for the large‐scale crustal flow hypothesis in the EMPT using the isotopic signature of magmatic rocks. If large‐scale crustal flow from the NQT occurs, distinct isotopic signatures of that terrane are expected in crustal‐derived or crustally contaminated melts. The Gongga‐Zheduo granitic massif, the largest intrusion in the EMTP, has a complex magmatic history from the Mesozoic (∼220–170 Ma) to Cenozoic (∼40–5 Ma) (Figure 1; Lai & Zhao, 2018; H. Li & Zhang, 2013; H. Li, Zhang, Zhang, Dong, & Zhu, 2015; Roger et al., 1995; Searle et al., 2016; Y.‐Z. Zhang et al., 2017). It provides an opportunity to test the large‐scale crustal flow model. Previous studies have tried to use such a method to trace crustal flow in the southern Tibet (e.g., King et al., 2007), but the conclusion was debated because the lower crust of Himalaya has similar isotopic compositions to the Lhasa terrane (e.g., L. Zeng et al., 2011). The Sr‐Nd‐Hf‐O isotopic compositions of the local crust in the EMTP compared to the NQT show clear differences and crustal‐derived magmas from the EMTP and NQT can be distinguished (Figure S1 in Supporting Information S1; de Sigoyer et al., 2014; S. Li et al., 2021; Long et al., 2015; Peng et al., 2015; Song et al., 2021; Q. Wang et al., 2016; Y.‐C. Zeng et al., 2020; Zhao et al., 2018).

Geology of the Gongga‐Zheduo Granitic Massif

The Gongga‐Zheduo granitic massif (Figure 1) intruded into Triassic turbidites in the Songpan‐Ganzi terrane and outcrop along the west side of NNW‐SSE trending Xianshuihe Fault. The eastern margin of the massif is an Oligocene mylonite‐migmatite zone, intruded by Pliocene dikes (H. Li & Zhang, 2013; Y.‐Z. Zhang et al., 2017).

The main part of Gongga‐Zheduo massif is subdivided into two units, the Gongga granites in the south and the Zheduo granites in the north (Figure 1). The Gongga granites are mainly composed of Triassic quartz diorite to monzogranite and minor Miocene syenogranite (H. Li, Zhang, et al., 2015). The Zheduo granites consist of mainly Miocene syenogranite with some Jurassic syenogranite (Lai & Zhao, 2018; H. Li, Zhang, et al., 2015; Roger et al., 1995), and newly discovered fine‐grained monzogranite and leucogranitic dike (Figures S2–S4 in Supporting Information S1). The Cenozoic magmatism here reflected a complex deformation history related to the Xianshuihe Fault: compression during the Eocene, transition from compression to strike‐slip during the Miocene, and large‐scale shearing during the Pliocene (e.g., Y. Chen et al., 2020; Y.‐Z. Zhang et al., 2017). See Supporting Information for detailed geologic descriptions.

Methods

Detailed analytical and modeling methods are presented in the Supporting Information.

Analytical Methods

Whole‐rock major and trace elements were analyzed at the Wuhan Sample Solution Analytical Technology Co., Ltd., Wuhan, China. Zircon U‐Pb dating and in situ Sr‐Nd‐Hf‐O isotopic analyses were performed at the Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, China.

Thermodynamic Modeling

Thermodynamic modeling was performed using a Gibbs free energy minimization approach using the software Perple_X (version 6.9.1) to constrain melting temperatures and pressures. The average composition of Neoproterozoic mafic rocks from the western margin of the Yangtze Craton and Paleozoic to Mesozoic metasedimentary rocks from the Songpan‐Ganzi terrane were selected as representative of the local EMTP source compositions. A system of Na2O–CaO–K2O–FeO–MgO–Al2O3–SiO2–H2O–TiO2–O2 (NCKFMASHTO) and the data set of hp633ver were used for the thermodynamic modeling based on the mineral assemblages and bulk rock compositions, assuming Fe3+/(Fe3+ + Fe2+) value as  0.2 for basaltic rock and 0.3 for metasedimentary rock, respectively (Forshaw & Pattison, 2021; Pourteau et al., 2020). The water contents are estimated based on the average value of loss‐on‐ignition. The equilibrium mineral assemblages and melt compositions are exported at discrete P‐T points for every 10°C and 0.1 GPa using a Perple_X‐based program Rcrust (Mayne et al., 2016). The uncertainty of P‐T estimates are ±1 kbar and ±50°C at the 2‐sigma level (Palin et al., 2016).

Results and Discussion

Geochronology of the Gongga‐Zheduo Granitic Massif

Our zircon U‐Pb dating results indicate that the Gongga‐Zheduo granitoid rocks were emplaced from 214 to 4 Ma (Figures S5–S6 in Supporting Information S1, Table S1 in Supporting Information S1), similar to previously published data. There are five main magmatic episodes in the EMTP, which are 215–200 Ma (Gongga monzogranite and leucogranite), 173–172 Ma (Zheduo porphyritic to coarse‐grained syenogranite), ∼50–30 Ma (Zheduo fine‐grained monzogranite, mylonitized granite and related leucogranite), ∼20–10 Ma (Gongga‐Zheduo coarse to fine‐grained syenogranite), and ∼5–3 Ma (Zheduo leucogranite; H. Li & Zhang, 2013; H. Li, Zhang, et al., 2015; Searle et al., 2016; Y.‐Z. Zhang et al., 2017).

Petrogenesis of the Gongga‐Zheduo Granitic Massif

Whole‐rock geochemistry data and isotopic compositions are listed in Table S2–S5 in Supporting Information S1. All samples are high in SiO2 (67.6–76.0 wt.%) and most are peraluminous (A/CNK = 0.97–1.11; molar Al2O3/(CaO + Na2O + K2O)) (Figure S7 in Supporting Information S1) and can be divided into two subgroups according to their K2O/Na2O ratio (Figure 2). Subgroup 1, including Gongga Triassic leucogranite, Zheduo Jurassic syenogranite, and Gongga‐Zheduo Miocene syenogranite, has high K2O content (4.40–6.98 wt.%) and K2O/Na2O ratio (1.22–2.59), and low MgO content (0.11–0.51 wt.%) and Mg# value (Figure 2 and Figure S7 in Supporting Information S1). Subgroup 2, including Gongga Triassic monzogranite, Zheduo Eocene‐Oligocene granite, and Zheduo Pliocene leucogranite, has relatively lower K2O content (1.30–4.53) and K2O/Na2O ratio (0.24–1.20), and higher MgO content and Mg# value (0.13–1.32 wt.%; Figure 2 and Figure S7 in Supporting Information S1). Subgroup 1 displays strongly negative Ba, Eu, and Sr anomalies, whereas Subgroup 2 shows negligible or slightly Eu anomalies with enrichment in Ba and Sr (Figure S8 in Supporting Information S1). Subgroup 1 has a high (La/Yb)N ratio (17.0–264; N denotes the chondrite values from Sun & McDonough, 1989) with a low Sr/Y ratio (4.72–86), in contrast to the Subgroup 2 samples ((La/Yb)N = 8.09–101; Sr/Y = 26–187; Figure 2). The in situ mineral isotope analyses indicate that Subgroup 1 and Subgroup 2 have distinct isotopic features (Figure 2 and Figure S9 in Supporting Information S1). Subgroup 1 has enriched Sr‐Nd‐Hf isotopic compositions (87Sr/86Sr(t) = 0.7083–0.7155; ε Nd(t) = −12.0 to −3.7; ε Hf(t) = −11.2 to +3.2) with elevated zircon δ 18O values (8.2–11.1‰). Subgroup 2 has relative depleted Sr‐Nd‐Hf isotopic compositions (87Sr/86Sr(t) = 0.7041–0.7061; ε Nd(t) = −8.2 to +0.3; ε Hf(t) = −1.0 to +7.8) with mantle‐like zircon δ 18O values (4.2–7.3‰), except for the Zheduo Oligocene leucogranite showing slightly enriched isotopic compositions (87Sr/86Sr(t) = 0.7078–0.7088; ε Nd(t) = −8.2 to −4.2; ε Hf(t) = +0.4 to +3.1; δ 18O = 8.0–9.0‰; Figure 2).

Figure 2

Geochemical and isotopic characteristics of the Gongga‐Zheduo granitic massif. (a) K2O/Na2O vs. SiO2 (wt.%) diagram. (b) MgO (wt.%) vs. SiO2 (wt.%) diagram. Fields of metabasaltic and eclogite melt, and metabasaltic and eclogite melt hybridized with peridotite are after Q. Wang et al. (2006). (c) Sr/Y vs. (La/Yb)N diagram. Subscript N denotes chondrite‐normalization. (d) δEu vs. Ba/Nb diagram. δEu = EuN/[(SmN * GdN)^0.5]. (e) εNd(t) vs. 87Sr/86Sr(t) diagram. The isotopic compositions of potential source rocks were calculated at 10 Ma. (f) δ 18O (‰) vs. εHf(t) diagram. The δ 18O (‰) values for the mantle are from Bindeman (2008). Modeling parameters are listed in Table S6 in Supporting Information S1.

Three main petrogenetic models have been proposed for the formation of granitic rocks, including (a) melting of crustal materials (i.e., Patiño Douce, 1999); (b) mixing of crustal‐derived melts and mantle‐derived melts (i.e., J.‐H. Yang et al., 2007); and (c) differentiation of mantle‐derived melts (i.e., Castillo, 2012). Our samples have no mafic enclaves and show low MgO content (most <1 wt.%) and Mg# values (<50), suggesting that magma mixing may not have been significant (Figure 2). Their evolved radiogenic isotope values indicate that they did not form by closed‐system fractionation of depleted mantle‐derived melts. Triassic and Cenozoic shoshonitic rocks are present in the EMTP and have been interpreted to be lithospheric mantle‐derived (Q. Chen et al., 2017; Hou et al., 2006; B. Xu et al., 2021). However, these shoshonitic melts cannot produce medium to high‐K granitic magma by closed‐system fractional crystallization or significant upper crustal assimilation. Subgroup 1 samples have high and uniform δ 18O values, indicating they are sourced from metasedimentary rocks (Figure 2). Subgroup 2 samples show mostly mantle‐like δ 18O values (Figure 2), but are felsic, without mafic to intermediate components. Therefore, we interpret the Gongga‐Zheduo granitic rocks to have been derived from melting of crust.

The contrasting isotopic and geochemical compositions between Subgroups 1 and 2 indicate that they have different crustal sources. Sr, Nd, and Hf isotopic compositions of Subgroup 1 samples are comparable to Songpan‐Ganzi metasedimentary rocks and S‐type granites, and lower crustal rocks of NQT (Figure 2). However, their elevated δ 18O values suggest that they are not sourced from the lower crust of NQT (Figure 2; Long et al., 2015; Song et al., 2021; Q. Wang et al., 2016). The isotopic compositions better match the metasedimentary rocks of the Songpan‐Ganzi terrane, which are local to the EMTP (Figure 2; de Sigoyer et al., 2014; Roger et al., 1995). Sediment‐derived melts at low‐pressure exhibit strongly negative Eu anomaly, as well as high La/Yb relative to Sr/Y ratios, reflecting both plagioclase and garnet are residuals (Moyen, 2009; Patiño Douce, 1999; Q. Wang et al., 2016).

The Sr‐Nd isotopic compositions of Subgroup 2 samples are more depleted than all crustal materials of NQT (Figure 2), arguing against a derivation from the NQT. Their Sr‐Nd‐Hf‐O isotopic compositions resemble Neoproterozoic mafic rocks from the western margin of Yangtze Craton, which is considered as the basement in the EMTP (de Sigoyer et al., 2014; Zhao et al., 2018). Their high Sr/Y relative to La/Yb ratios without strongly negative Eu anomaly indicates the residual minerals are mainly amphibole and garnet. Hence, we interpret both Subgroup 1 and Subgroup 2 to have involved local crustal sources in the EMTP and that those sources have remained the same throughout Mesozoic to Cenozoic time, the age range of our samples.

Thermodynamic and trace element modeling provide additional constraints on the petrogenesis of the samples. Here, we focus on the Cenozoic granites, which were emplaced during the tie frame of postulated crustal flow (Figure 3). The residual mineral assemblages, melting degrees, and major element compositions of melts were obtained using the thermodynamic calculations of Connolly (2009). Trace element modeling was conducted using a simple batch melting model (Shaw, 1970). Cenozoic Subgroup 1 granites could be generated by 26%–49% melting of metasedimentary rocks at 0.8–1.0 GPa, 740°C–840°C (including uncertainty; Figure 3). Cenozoic Subgroup 2 granites could be formed by 10%–23% melting of Neoproterozoic mafic rocks at 1.0–1.3 GPa, 770°C–890°C (including uncertainty; Figure 3). Good fits were obtained for both major and trace elements, constraining their source characteristics and melting conditions (Figures 2 and 3).

Figure 3

Thermodynamic and trace element modeling. (a, c) Simplified P‐T phase diagram for the average composition of metasedimentary rocks from the Songpan‐Ganzi Basin and Neoproterozoic mafic rocks from the western margin of the Yangtze Craton, calculated with water contents of 3.7 wt.% and 2.2 wt.%, respectively, corresponding to dehydration melting (Table S7 in Supporting Information S1). The red solid line and red dashed line mark calculated solidus and water saturation of the system, respectively. Brown dashed lines show the calculated degree of melting (wt.% of melt); purple dashed lines and orange dashed lines represent garnet and plagioclase proportion (wt.%) in the residue, respectively. Solution models: G‐Green et al. (2016); FL‐Fuhrman and Lindsley (1988), W/WPH‐White et al. (2014), HP‐Holland and Powell (2011). (b, d) Trace element patterns of granitic melts calculated at specific P‐T conditions and the average composition of potential source rocks. The Kd used for modeling are presented in Table S9 in Supporting Information S1. The blue, pale orange, and red shaded areas represent the overall compositional range of the Subgroup 1 Miocene granites, Subgroup 2 Eocene‐Oligocene granites, and Subgroup 2 Pliocene granites, respectively.

Large‐Scale Crustal Flow or Episodic Crustal Thickening/Uplift

Previous thermochronological studies conducted in the EMTP demonstrated a major phase of rapid uplift during the Late Miocene to Pliocene (∼12–4 Ma), which was proposed to be related to large‐scale crustal flow (Clark, House, et al., 2005; Schoenbohm et al., 2006; E. Wang et al., 2012; H. Zhang et al., 2016; Y.‐Z. Zhang et al., 2017). In consideration of the time required for crustal thickening (∼20 m.y.), the large‐scale crustal flow was interpreted to have started at ∼40–30 Ma (Clark, House, et al., 2005; Clark & Royden, 2000). This time is coeval to the onset of magmatism in both the NQT and EMTP (Long et al., 2015; Y.‐C. Zeng et al., 2020). Previous paleo‐elevation studies proposed that the NQT reached ∼5,000 m elevation by the Eocene (F. Hu et al., 2020; Q. Xu et al., 2013), indicating a potential lateral pressure gradient existed. Our modeling results show that the Cenozoic magmas in the EMTP originated at ∼30–40 km depth (Figure 3), similar to the depth of proposed large‐scale crustal flow (Bai et al., 2010; Clark, Bush, et al., 2005; Clark & Royden, 2000).

Collectively, all the requirements for large‐scale crustal flow during the Cenozoic seem to be met. However, Cenozoic granites in the EMTP have different isotopic compositions from the crustal rocks of NQT (Figure 2). If crustal flow exists, the isotopic composition of flowed materials is unlikely to be modified by local partial melts because: (a) deep‐crustal derived magma is scarce in the eastern Songpan‐Ganzi terrane (e.g., H. Li & Zhang, 2013); (b) melting degree of local crust should be lower than 5% (the lower limit of melt volume within the crustal flow; Bai et al., 2010; Hacker et al., 2014), which is lower than limit for melt extraction (Brown, 2013). Our data show no evidence that crustal materials derived from the NQT are found at ∼30–40 km depth in the EMTP prior to the Pliocene, our youngest sample. Geophysical observations suggest that crustal flow may be occurring at present (Bai et al., 2010; Bao et al., 2015; C.‐Y. Wang et al., 2010), but such young crustal flow cannot result in high elevations in the EMTP (Clark & Royden, 2000). Therefore, our results do not support long‐distance crustal flow, but do not rule out regional scale (<200 km) crustal flow. The disordered and weaker crustal anisotropy in central Tibet, compared to the plateau margins, is also inconsistent with the large‐scale crustal flow (Bao et al., 2020).

We interpret temporal changes in the isotopic data from Cenozoic granites in the EMTP to reflect changes to the magmatic sources during three episodes of magmatism (Figure 4).

Figure 4

Temporal changes in the characteristics of Cenozoic magmatism in the EMTP. (a) Probability density plot of Gongga‐Zheduo granitic massif, Mianning‐Dechang alkaline rocks, and Batang‐Dali alkaline rocks. Data sources are listed in Table S10 in Supporting Information S1. (b) Exhumation history of the EMTP. Data of Longmenshan and Gongga‐Jiulong areas are from E. Wang et al. (2012) and H. Zhang et al. (2016), respectively. (c–f) Initial 87Sr/86Sr, εNd(t), εHf(t), δ 18O (‰) vs. age (Ma) of Gongga‐Zheduo granitic rocks. Three magmatic episodes, shown by orange, green and purple dashed boxes, are documented. Symbols for rock units are the same as Figure 2.

The first magmatic episode (∼50−30 Ma) is characterized by Subgroup 2 granites with depleted isotopic compositions and mantle‐like δ 18O values (Figure 4, orange box). This episode is generally coeval with Eocene‐Oligocene alkaline magmatism and carbonatites in the EMTP, that is, Mianning‐Dechang (Hou et al., 2006; B. Xu et al., 2021) and Batang‐Dali magmatism (Chung et al., 1998; B. Xu et al., 2021). These lithospheric mantle‐derived magmas could have provided heat for locally melting of metabasaltic rocks in the middle‐lower crust, which may also weaken the lithosphere. During this period, early crustal uplift has been documented in some areas in the EMTP, which is related to crustal thickening by compression (E. Wang et al., 2012; H. Zhang et al., 2016). Compressive thickening is supported by the Late Eocene to Oligocene fold and thrust in eastern Tibet (Cao et al., 2020; H. Li & Zhang, 2013) and high paleo‐elevations (Hoke et al., 2014; S. Li, Currie, Rowley, & Ingalls, 2015).

The second magmatic episode (∼20–10 Ma) is characterized by a shift to Subgroup 1 granites with more enriched isotope ratios and elevated δ 18O values (Figure 4; green box). Thermochronology data suggests this was a period of general stability with relatively slow uplift rates (E. Wang et al., 2012; H. Zhang et al., 2016; Y.‐Z. Zhang et al., 2017). The isotopic data suggests that magmas from this episode involved the most supracrustal metasedimentary material with minimal mantle‐involvement. In addition, the magmatism was active prior to the onset of the Xianshuihe strike‐slip fault zone (∼13–9 Ma; S. Wang et al., 2009; Y.‐Z. Zhang et al., 2017). Hence, mantle‐derived magma or shear heating are not likely the reason for this episode of magmatism. In turn, magmatism may have softened the crust and facilitated to the strike‐slip movement of Xianshuihe Fault (J. Yang et al., 2020). Low exhumation rates during this time also argue against a decompression melting (E. Wang et al., 2012; H. Zhang et al., 2016). Therefore, the most likely heat source for episode 2 magmatism is radiogenic heating after crustal thickening (Bea, 2012).

The third magmatic episode (∼10–4 Ma) is characterized by Subgroup 2 granites (Figure 4, purple box), but their higher SiO2 content, Rb/Sr, and Th/La ratios with lower La content reflect higher degrees of fractionation than granites of episode 1 (Figure S7 in Supporting Information S1). The EMTP was experiencing uplift and exhumation during this time (Clark, House, et al., 2005; E. Wang et al., 2012; H. Zhang et al., 2016). There are several lines of evidence to suggest that this is related to upwelling of asthenospheric mantle including (a) ∼12 Ma alkaline rocks in the Mianning‐Dechang area (B. Xu et al., 2021), (b) abnormally high lithospheric heat flow, mantle signatures of 3He/4He from hot springs (S. Hu et al., 2000; M. Zhang et al., 2021), (c) upper mantle low‐velocity anomalies (Z. Huang et al., 2019; W. Wang et al., 2021), and (d) dynamic support for high elevations (Bao et al., 2020). Geophysical observations suggest that northwestward downwelling of the Indian Block (Z. Huang et al., 2019) and regional lithospheric delamination (W. Wang et al., 2021) could account for the upwelling of the asthenosphere. We suggest that mantle upwelling caused (re)melting of metabasaltic rocks in the EMTP during episode 3.

Conclusions

The Gongga‐Zheduo granitic massif contains Mesozoic (∼214–172 Ma) to Cenozoic (∼50–4 Ma) granitoid rocks and helps to constrain crustal compositions and sources in the EMTP. The granitoid rocks can be divided into two subgroups according to their geochemical and isotopic characteristics. Subgroup 1 was derived from partial melting of metasedimentary rocks of Songpan‐Ganzi terrane, and Subgroup 2 was derived from partial melting of metabasaltic rocks of western margin of the Yangtze Craton. Evidence for crustal materials derived from the NQT were not observed and we suggest that the mid‐lower crust in the EMTP consists entirely of locally derived crustal rocks. Changes in the magmatic sources during the Cenozoic correlate well with changes in uplift and exhumation. Cenozoic magmatism was primarily controlled by mantle‐crust interactions, which in turn may have modified the lithospheric (especially crustal) rheology in eastern Tibet. The continued indentation of India and changes in crustal rheology of Asia shaped the present eastern boundary of the Tibetan Plateau.

Supporting Information S1

Click here for additional data file.