Source-To-Sink Links Between East Asia and Taiwan from Detrital

Geochemistry, Geophysics, Geosystems

RESEARCH ARTICLE Source-To-Sink Links Between East Asia and Taiwan From 10.1029/2018GC007576 Detrital Zircon Geochronology of the Oligocene Key Points: Huagang Formation in the East China Sea • The Min River or similar drainage system is the likely source for Shelf Basin Eocene–Oligocene strata in western Taiwan Wei Wang1,2,3 , Tandis Bidgoli3,4 , Xianghua Yang1,2, and Jiaren Ye1,2 • Oligocene sandstones from the East China Sea Shelf Basin contain 1Key Laboratory of Tectonics and Petroleum Resources, Ministry of Education, China University of Geosciences, Wuhan, abundant Precambrian zircons, likely 2 3 sourced from the Yangtze and North China, Faculty of Earth Resources, China University of Geosciences, Wuhan, China, Kansas Geological Survey, University of 4 China blocks Kansas, Lawrence, KS, USA, Department of Geological Sciences, University of Missouri, Columbia, MO, USA • Miocene strata from western Taiwan may be reworked from Oligocene units in East China Shelf Basin during Abstract Researchers have long suspected a source-to-sink link between the Yangtze Block (and/or North Miocene inversion China Block) and Eocene-Miocene passive margin strata in Taiwan, and various models have been proposed to explain similarities in provenance and the presence of Archean grains. The East China Sea Shelf Basin, Supporting Information: whose evolution has been neglected in most models, resides between potential mainland China source • Supporting Information S1 • Table S1 regions and the sediment sink in Taiwan, and occupies the corridor through which sediment would have been transported. Here we present 378 new concordant detrital zircon U–Pb ages from the Oligocene Correspondence to: Huagang formation in the Xihu Sag, eastern East China Sea Shelf Basin. We also present a compilation of T. Bidgoli, published detrital zircon data from major rivers across mainland China to evaluate potential sediment source [email protected] regions. Detrital zircons in the Huagang formation range in age from 33 ± 0.4 to 2842 ± 43 Ma, with major clusters at 100–300 and 1,600–2,200 Ma and small early-middle Paleozoic, Neoproterozoic, and Citation: Paleoproterozoic to Archean peaks. Visual and statistical comparison of age distributions suggest close Wang, W., Bidgoli, T., Yang, X., & Ye, J. – (2018). Source-to-sink links between relationships between the Min River and Eocene Oligocene samples from western Taiwan and between the East Asia and Taiwan from detrital zir- Oligocene Huagang formation and Miocene strata in western Taiwan. Combined with the tectonic and con geochronology of the Oligocene depositional evolution of the region, these observations suggest that Eocene–Oligocene sediments in Taiwan Huagang formation in the East China Sea Shelf Basin. Geochemistry, were mainly sourced from the Min River or similar drainage in the Cathaysia Block. However, the similarities Geophysics, Geosystems, 19, 3673–3688. between the Huagang formation and Miocene strata are best explained by reworking of the Oligocene https://doi.org/10.1029/2018GC007576 sediment in the Xihu Sag and transportation to western Taiwain during Miocene tectonic inversion of the East China Sea Shelf Basin. Received 27 MAR 2018 Accepted 18 AUG 2018 Accepted article online 13 SEP 2018 Published online 4 OCT 2018 1. Introduction The Cenozoic topographic and drainage evolution of East Asia is complex and intimately tied with the India-Asia collision, the Pacific plate subduction, and with climate change (Clark et al., 2004; Hu et al., 2015; Northrup et al., 1995). The major rivers across southeastern China like the Yangtze, Red, Pearl, and Min rivers (Figure 1) are dynamic recorders of these tectono-climatic changes, providing important constraints on the timing of uplift of Tibet (e.g., Clift et al., 2006, 2008), rifting of the South China Sea (e.g., Shao et al., 2015; Wang et al., 2017), and intensification of the East Asian Monsoon (e.g., Clift et al., 2004). However, obtaining long and continuous records of continental erosion and deposition from these terrestrial systems has remained a challenge (Zhang et al., 2017). To circumvent this issue, researchers have taken advantage of outcrops on the island of Taiwan (Figure 1), where a relatively complete succession of Cenozoic passive margin strata is preserved. These strata are in many ways the ideal proxy for landscape change as they preserve an uninterupted record of Eocene- ©2018. The Authors. to-Miocene continental erosion (Deng et al., 2017; Lan et al., 2016; Zhang et al., 2017). The source-to-sink This is an open access article under the history of these units has been evaluated using thousands of zircon U–Pb ages from outcrop and modern terms of the Creative Commons Attribution-NonCommercial-NoDerivs sediment samples, from both Taiwan island and mainland China to the west, in an attempt to build a more License, which permits use and distri- robust understanding of the relationship between the two regions (see summary in Deng et al., 2017). bution in any medium, provided the Detrital zircon ages from strata exposed in the Western Foothills and Hsuehshan Range, western Taiwan, original work is properly cited, the use is – non-commercial and no modifications document variable sediment sources and a pronounced change in provenance between the Eocene or adaptations are made. Oligocene, dominated by Mesozoic and Paleozoic zircons, and the Miocene, with more complicated age

WANG ET AL. 3673 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Figure 1. Topographic map of East Asia showing major tectonic elements and river systems. ECSSB, East China Sea Shelf Basin; TXB, Taixi Basin; TXNB, Taixinan Basin; and XH, Xihu sag. The background topographic map is modified from general bathymetric chart of the oceans (Gebco) world ocean bathymetry. The Qinling-Dabie Belt divides the North China block (NCB) and the South China block (SCB). The Jiangshan-Shaoxing fault (JSF) separates the Yangtze block (YB) and Cathaysia block (CB). The Zhenghe-Dapu fault (ZDF) separates the Cathaysia block into west (WCB) and east (ECB) parts. Major faults are from Wang et al. (2013). Yellow circles show the location of the samples cited in this study. Red stars correspond to boreholes sampled in this study. Inset figure shows simplified tectonic divisions of China, modified from Geng et al. (2011).

spectra that include Mesozoic, Paleozoic, Proterozoic, and Archean grain populations (Deng et al., 2017; Lan et al., 2016; Zhang et al., 2017). Although the age data provide insight into the provenance of these Eocene-to-Miocene units in Taiwan, multiple models for sediment transport from the Cathaysia, Yangtze, and North China blocks (Figure 1) have been proposed for Taiwan (Deng et al., 2017, and references therein) and are debated. While most workers agree that a river similar to the modern Min River, with a drainage largely conﬁned to the Cathaysia Block, is the most probable source for Eocene–Oligocene sediment to a proto-Taiwan (Lan et al., 2016; Deng et al., 2017; Xu, 2017), the notable increase in the abundance of Proterozoic and Archean zircons in Miocene sandstones from the Western Foothills suggest sediment was derived, in part, from the Yangtze and/or North China blocks, where ancient basement is exposed (Lan et al., 2016; Zhang et al., 2017). Lan et al. (2016) speculated that a larger paleo-Min River may have been responsible for delivery of Miocene sediment, sourced from near the boundary of the Yangtze and Cathaysia blocks to Taiwan, until its headwaters were captured by the Yangtze River sometime in the Neogene. Miocene strata in the Hengchun Peninsula in southern Taiwan had a different source area from similar age units in the Western Foothills and Hsuehshan Range. This observation cannot be reconciled with the paleo-Min River model and it suggests, instead, that a drainage similar to the modern Yangtze River may have been established prior to the middle Miocene (Zhang et al., 2017). Deng et al. (2017) conﬁrmed a Yangtze and North China Block source for

WANG ET AL. 3674 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Miocene strata in the Western Foothills and proposed that drainage reorganization in east China combined with longshore currents were responsible for transport of Yangtze Block sediment to the Taiwan area. The model, however, was immediately challenged by Xu (2017), who inferred that the longshore current would have been too weak to carry heavy minerals large distances, suggesting instead that the sediment was likely derived from a paleo-Min River. Although the signature of the Yangtze Block and possibly North China Block is evident in Miocene strata in western Taiwan (Deng et al., 2017; Lan et al., 2016; Zhang et al., 2017), how sediment was transported to a proto-Taiwan remains uncertain. Sediment from mainland China would have certainly been transported through the broad East China shelf, but the evolution of the East China Sea Shelf Basin (ECSSB), which occupies the bulk of this region, has not been adequately accounted for in provenance studies and source-to-sink models (Figure 1). Zhang et al. (2017) cited limited data from Paleogene-Eocene samples in the western ECSSB and from one Oligocene sample in the eastern part of the basin (Fu et al., 2015; Yang et al., 2006), but the low number of grains analyzed, particularly for Oligocene strata in the Eastern Depression Belt (EDB), hindered interpretations. More recent provenance investigations into Oligocene units also provide clues (Qin et al., 2017; Wang et al., 2017), but additional analysis is needed. Given that the Oligocene-to- Miocene coastline of southeast China was east of its present location (Deng et al., 2017), any discussion of source-to-sink links between the Yangtze and other river drainages and western Taiwan must consider the dynamics of this shelf basin and its role in sediment delivery to Taiwan. In this paper, we examine the provenance of the Oligocene Huagang formation in the Xihu Sag, the largest subbasin of the ECSSB (Figure 1). The data are aquired from eight boreholes (six core samples and two cutting samples) distributed across four structures in the Central Uplift Zone of the Xihu Sag and offer a rare opportunity to examine strata, now buried to depths greater than 3,500 m, in the corridor between potential mainland China sources to the west and Taiwan to the east (Figure 2). Using new and published detrital zircon U–Pb ages, we show that the age distributions of Oligocene samples in the ECSSB are most similar to the Yangtze River and rivers in the North China Block, suggesting a strong link between the two regions, but are also camparable to Miocene strata in Taiwan. When combined with the tectono-stratigraphic evolution of the ECSSB, the data suggest that the provenance of Miocene strata in Taiwan are best explained by reworking of the Oligocene or younger strata in the Xihu Sag and transportation to western Taiwain during tectonic inversion of the ECSSB. The results also provide further constraints on the evolution of the Yangtze River and its establishment as a major sediment transportation system prior to 24 Ma.

2. Geological Setting The East China Sea continental shelf is made up of five major tectonic units that record the evolution of the region (Suo et al., 2015). From west to east, they include the Min-Zhe Uplift, ECSSB, Taiwan-Sinzi belt (also called the Diaoyudao Uplift-Fold Belt), Okinawa Trough, and Ryukyu Arc and Trench (Figure 1). The Mesozoic Min-Zhe Uplift is located along the southeastern coast of China. The Taiwan-Sinzi belt, which resides between the ECSSB and the Okinawa Trough, is a paleotopographic high made up of Cretaceous and older metamorphic rocks and younger volcanic intrusions (Sibuet & Hsu, 2004; Wang et al., 2011). The Okinawa Trough is a middle Miocene-to-recent back-arc basin to the Ryukyu Arc (Letouzey & Kimura, 1986). The ECSSB is the largest offshore basin in southeastern China and covers the bulk of the East China shelf (Figure 1). It is located near the zone of convergence between the Eurasian and Philippine Sea plates (Cukur et al., 2011; Zhou et al., 2001; Figure 1). Structurally, the ECSSB can be divided into three major, first-order tectonic units that are roughly parallel the east China shoreline. From west to east, they are the Western Depression Belt, the Central Uplift Belt, and the EDB (Figure 2). The tectonic evolution of the ECSSB can be divided into several stages: (1) extension and rifting, (2) compression and inversion, and (3) thermal subsidence stages (Suo et al., 2015; Zhang et al., 2016). However, the timing of these stages differs between the Western Depression Belt and the EDB, with deformation in the Western Depression Belt preceding Eocene-to-Miocene deformation in the EDB (Suo et al., 2015; Zhang et al., 2016). These first-order tectonic units have been further divided into a series of depressions or subbasins, bounded by structural highs. The EDB, also called Zhedong Depression, the focus of this study, is made up, from north to south, of the Fujiang, Xihu, and Diaobei depressions, among which the Xihu Sag is the largest and best explored (Ye et al., 2007; Figure 2). This subbasin is filled with a thick succession of

WANG ET AL. 3675 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Figure 2. Structural elements of (a) ECSSB, modified from Suo et al. (2015) and (b) Taiwan, modified from Chen et al. (2000), Huang et al. (2012), Lan et al. (2016), and Deng et al. (2017). The Lishan fault separates deep marine trench-slope sequences of the central range and Hengchun peninsula from passive margin sediment exposed in the Hsuehsahn range and western foothills of Taiwan. (c) Cross-sectional profile A-A0 of the ECSSB (line location shown in a) from seismic data, modified from Suo et al. (2015). The box with blue dotted lines indicates the central anticline zone of Xihu sag where we sampled.

Paleocene to Quaternary units, including the Eocene Pinghu formation; Oligocene Huagang formation; Miocene Longjing, Yuquan, and Liulang formations; Pliocene Santan formation; and Pleistocene Donghai

Group (Suo et al., 2015; Ye et al., 2007; Figure 3). The Oligocene Huagang formation, between the T30 and T20 unconformities, ranges in thickness from 1,000 to 1,800 m. Because it is a primary target for in hydrocarbon exploration, it relatively rich with data compared to other formations in the Xihu Sag

(Ye et al., 2007). The unconformity T30, at the base of the Huagang formation, is the boundary between synrift and postrift successions (Suo et al., 2015; Yang et al., 2014; Figure 3). The Huagang formation represents two tectonic cycles or third-order sequences that consist of ﬂuvial, deltaic, and lacustrine facies (Ye et al., 2007), of which the lower part is composed of siltstone and mudstone and the upper part consists of sandstone interbedded with mudstone and coal seams (Ye et al., 2007). More recently, thick and deeply buried lowstand-system-tract sandstones have been recognized in the Huagang formation and are targeted for hydrocarbon exploration, providing a unique opportunity for research (Chen, 2015). Following deposition of the Huagang formation, sediment delivery expanded and crossed the western boundary of the basin onto the ﬂanks of the subbasin, like on the Haijiao Uplift (Yang et al., 2014). Taiwan Island is located off the southeast margin of the Asian continent, at the junctures between the Eurasian and Philippine Sea plates and between the Manilla and Ryuku trenches (Figure 1). The island may

WANG ET AL. 3676 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Figure 3. Generalized Cenozoic stratigraphy, unconformities, facies, and tectonic events (movements) of the EDB in the ECSSB, modiﬁed from Suo et al. (2015).

be divided into ﬁve major north-south oriented tectonic units. From west to east they include the Coastal Plain, Western Foothills, Hsuehshan Range, Central Range, and Coastal Range (Ho, 1986; Huang et al., 2006, 2012; Figure 2). The Coastal Plain, Western Foothills, and Hsuehshan Range are made up of early-to-middle Cenozoic passive-margin successions that developed in response to extension and seaﬂoor spreading in the South China Sea (Ho, 1986; Huang et al., 2012). In the middle Miocene, subduction of South China Sea oceanic lithosphere beneath the Philippine Sea Plate initiated, forming the Luzon Arc and an accretionary complex in what are now the Coastal and Central ranges, respectively. Subduction progressed to arc– continent collision at ~6 Ma (Ho, 1986; Huang et al., 2006), during which passive margin sequences were deformed and uplifted, forming a fold-thrust belt in the Western Foothills and crustal-scale pop-up structure in Hsuehshan Range (Clark et al., 1993). Underthrust Eurasian crust makes up the eastern part of the Central Range (Ho, 1986; Huang et al., 2006, 2012).

3. Samples and Methods We collected eight samples of the Oligocene Huagang formation from eight boreholes in the north-central anticlinal zone of the Xihu Sag. The details of the borehole samples are provided in Table 1. Because boreholes with core are rare, only 125 g of rock were obtained for each sample. Samples were separated using standard mechanical, heavy liquids, and electromagnetic mineral separation techniques. Zircon grains were picked under binocular microscopes, mounted in epoxy, and polished for U–Pb spot analyses (32-μm spot size). Cathodoluminescence images were used to evaluate the internal structure of grains. Laser ablation- inductively coupled plasma mass spectrometry (LA-ICPMS) was carried out at the State Key Laboratory of Geological Processes and Mineral Resources. LA-ICPMS analytical procedures are outlined in Wang et al. (2017), and detailed procedures for the laser and mass spectrometric systems and data reduction are provided by Liu et al. (2008, 2010). Zircon grains older than 1,000 Ma were assessed by 207Pb/206Pb ages due to larger amounts of radiogenic Pb present, while grains younger than 1,000 Ma were determined using the more reliable concordant 206Pb/238U age (Compston et al., 1992). All ages reported use 1σ absolute propagated uncertainties.

WANG ET AL. 3677 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Table 1 Borehole Number, Stratigraphic Position, and Depth of Detrital Zircon Samples Used in This Study Borehole number Depth (m) Formation Sample type Dated grains Concordant ages Reference

L1 3329–3339 Oligocene Huagang formation Cutting 53 50 This study L2 3711–3755 Oligocene Huagang formation Cutting 50 49 This study H2 3842 Oligocene Huagang formation Core 45 42 This study H3 3964 Oligocene Huagang formation Core 48 46 This study G2 3768 Oligocene Huagang formation Core 52 49 This study G1 3848 Oligocene Huagang formation Core 51 47 This study Y3 3137 Oligocene Huagang formation Core 52 49 This study Y1 2826 Oligocene Huagang formation Core 52 46 This study Unknown Unknown Oligocene Huagang formation Unknown Unknown 71 Yang et al. (2006)

Detrital zircon age distributions are presented as histograms, discretizing data according to bin size (50 Ma), and kernel density estimation (KDE) plots, grouping data by bandwidth as a smooth curve (Vermeesch, 2012). Although both approaches allow for visual/qualitative comparisons of samples and potential sediment sources, direct comparison and quantitative discrimination of samples was accomplished through cumulative probability plots (CPP), which reveal differences in the proportions of grains among samples (Gehrels, 2011; Malusà et al., 2013, 2016), and multidimensional scaling (MDS) analysis, which uses goodness-of-ﬁt and closeness criteria to evaluate similarities and differences among samples (Vermeesch, 2013).

4. Results 4.1. Detrital Zircon U–Pb Ages From the Oligocene Huagang Formation Of the 403 detrital zircon grains analyzed, a total of 378 grains yielded ages with a concordance greater than 90%. Kolmogorov–Smirnov (K-S) tests (Smirnov, 1939; Vermeesch, 2013; Young, 1977) were used to

evaluate the similarity between the uncut distributions and those obtained with a 90% cutoff level. VK-S was calculated following the method of Malusà et al. (2013, 2016) and is greater than 0 (0.075), indicating that differences between uncut and cut distributions are statistically not signiﬁcant. The age distribution of zircon grains from Huagang formation sandstones are shown as histograms in Figure 4. The ages range from 33 ± 0.4 to 2842 ± 43 Ma and are characterized by two major clusters at 100–300 and 1,600– 2,200 Ma. Samples also contain small early-middle Paleozoic (400–540 Ma), Neoproterozoic (540– 1,000 Ma), and Paleoproterozoic to Archean (>2,300 Ma) peaks. Although the number of grains analyzed for each sample is limited, with n < 50 for most samples, the age distributions agree well with the published detrital zircon data (n = 71) obtained from the same unit by Yang et al. (2006) (site is roughly marked in Figure 2). The full suite of isotopic measurements and U–Pb ages are provided in Table S1 in the supporting information.

4.2. Compilation of Published Detrital Zircon U–Pb Ages From Modern River Sands Although the timing of tectonomagmatic events and geologic age distributions across East Asia are well known (e.g., Wang et al., 2013, and references therein), comparing bedrock ages of potential source areas with detrital zircon age distributions of Eocene to Miocene units in the ECCB and Taiwan is not straightfor- ward. The Cathaysia, Yangtze, and North China blocks have complex tectonic histories that include Precambrian, middle Paleozoic (Kwangsian; 465–400 Ma), Triassic (Indosinian; 270–195 Ma), and/or Jurassic-Cretaceous (Yanshanian; 180–90 Ma) orogenic events, resulting in a patchwork of bedrock of variable age (Wang et al., 2013). The relative contributions of these potential zircon sources are difﬁcult to predict from their exposure areas. Recent studies have shown that detrital zircon data from modern river sands may be a good proxy for expected age distributions of zircon-bearing bedrock source areas (e.g., Deng et al., 2017; Liu et al., 2017; Zhang et al., 2017). These data better account for complications within a given catchment (e.g., zircon fertility, sensitivity to weathering and erosion, drainage networks effects, and mixing and dilution; summary in Deng et al., 2017 and Malusà et al., 2013, 2016) and can be directly compared to other detrital zircon data sets through statistical tests and analyses (e.g., Vermeesch, 2013). In the following

WANG ET AL. 3678 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Figure 4. Histograms (50-Ma bin size) of detrital zircon U-P ages of samples from this study and Yang et al. (2006).

sections, we present a compilation of published detrital zircon data from the major rivers in southeast China (Figure 1), which provides a robust framework for evaluating potential source regions. 4.2.1. River in the North China Block The North China Block is host to the Yellow River, which flows through several terranes (e.g., western Qinling Orogen, the Qilian Orogen, and the North China Craton) before terminating in the Bohai Sea (Yang et al., 2009). The North China Craton contains some of the oldest cratonic rocks in the world (Gao et al., 2004; Liu et al., 1992) and river sands from the lower reach of the Yellow River reflect this with two major Precambrian age clusters at 2.1–2.5 and 1.6–2.0 Ga (Yang et al., 2009). Because the modern Yellow River flows through several terranes and it is unknown when the river came into its present morphology, we incorporate data from two tributaries of the Yellow River (Jing and Luo rivers; Diwu et al., 2012) and from the Luan and Yongding rivers (Yang et al., 2009), which are restricted to the North China Craton, to characterize this ancient block given its geographic proximity to the continental margin (Figure 1). As shown in Figure 5a, the data reveal pronounced Paleoproterozoic-Neoarchean and Paleoproterozoic clusters, diagnostic of cratonal sediment sources in the North China Block. Additionally, there are two Paleozoic age peaks, made up principally of Ordovician and Permian grains. 4.2.2. Yangtze River The Yangtze River is the largest river in Asia and its evolution has attracted considerable attention (He et al., 2013; Zhang et al., 2017; Zheng et al., 2013). The river’s inception is controversial, with estimates ranging from middle Eocene to Pleistocene (see summaries in Zheng et al., 2013 and Zhang et al., 2017). He et al. (2013) dated more than 2200 detrital zircon grains from modern river sands, covering much of the river’s drainage. Similarities between modern river sands and lower-to-middle Miocene sandstones (~24–10 Ma) in Nanjing area, located in the lower reaches of the river, indicate that the modern Yangtze River was likely in place prior to ∼23 Ma (Zheng et al., 2013). Here we combine the data from the lower reaches of the river. As shown in Figure 5b, the age spectrum is characterized by pronounced early Mesozoic, Neoproterozoic, and Mesoproterozoic peaks, and subordinate early Paleozoic and Paleoproterozoic peaks (He et al., 2013). The Neoproterozoic grains differentiate sediment sources in the Yangtze Block from the North China and Cathaysia blocks (Zhang et al., 2017).

WANG ET AL. 3679 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Figure 5. Kernel density estimation plots (Vermeesch, 2012) for the zircon U–Pb ages of modern river sands from the East China mainland, Eocene to Miocene passive margin strata in western Taiwan Island, and Oligocene strata in the Xihu sag. Data sources: North China block rivers (Yang et al., 2009; Diwu et al. (2012), Yangtze River is from He et al. (2013), Taiwan Island is from Lan et al. (2016) and Zhang et al. (2017), Ou River is from Xu et al. (2007), the Jiulong River is from Xu et al. (2014) and Zhang et al. (2017), the Min River is from Xu et al. (2014, 2016), and the Pearl River is from Zhao et al. (2015) and Xu (2017).

4.2.3. Min River The Min River is the largest river along the southeast coast of mainland China (Figure 1). The headwaters of the river cross a Precambrian terrane (Wuyishan terrane), which includes a Paleoproterozoic complex (Badu Complex, 1,830–1,890 Ma), the oldest Cathaysia block basement, and middle Paleozoic granitoids (Kwangsian granitoids; Wang et al., 2013; Xu, 2017; Xu et al., 2016). The watershed also crosses a major belt

WANG ET AL. 3680 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

of Jurassic-Cretaceous (Yanshanian) igneous rocks, where Triassic (Indosinian) granites are also developed (Xu, 2017; Xu et al., 2016). Figure 5f shows that the resulting age spectrum, based on data from Xu et al. (2014, 2016) and Zhang et al. (2017), is complicated, with Paleoproterozoic, middle Paleozoic, Triassic, and Jurassic-Cretaceous clusters (Figure 5f). 4.2.4. Ou River and Jiulong River The Ou and Jiulong rivers occur north and south, respectively, of the Min River. Their catchments are restricted to the area east of the Zhenghe-Dapu Fault, in the eastern Cathaysia Block, where Jurassic- Cretaceous granites and volcanic rocks are exposed (Yanshanian magmatism; Wang et al., 2013; Figure 1). The age spectrum of the Ou River, based on data from Xu et al. (2007), is characterized by a large early Cretaceous peak and subordinate Triassic and Paleoproterozoic peaks, with the Paleoproterozoic peak implying influence from metamorphic basement exposed in the headwaters of the river (Wuyi or Nanling terranes; Wang et al., 2013; Xu, 2017). The Jiulong River is similar, but with a more pronounced Triassic peak, which likely reflects Indosinian granitoids located near the headwaters of the river (Xu et al., 2016). 4.2.5. Pearl River The Pearl River, also known by its Chinese name Zhujiang, crosses much of southern China and is made up of watersheds of the Xi (West), Bei (North), and Dong (East) river system. Due to the large size and variability of the drainage area, we follow Shao et al. (2016) and divide the river’s drainage into western and northeastern tributaries, using data from Xu et al. (2007) and Zhao et al. (2015) (Figures 5g and 5h). The northeastern Pearl River is represented by detrital zircons of the Bei and Dong rivers, which mainly flow across Jurassic-Cretaceous igneous rocks and sporadic Triassic and middle Paleozoic granites. The age spectrum for this part of the river is characterized by an Early Cretaceous (97 Ma) peak and two subordinate age clusters of ~140–170 and ~400–500 Ma, corresponding to the Indosinian and Kwangsian orogenies, respectively (Figure 5h). The western part of Pearl River is represented by detrital zircon from the Xi River. In the northwest, the river crosses the Yangtze Block, but in the south, the river crosses middle Paleozoic (Kwangsian) and Triassic (Indosinian) granitoids in the southern Cathaysia Block (Yunkai terrane; Zhao et al., 2015; Shao et al., 2016). The age distribution for this part of the river is more complicated than the northeastern Pearl River, with strong late Permian to Middle Triassic and middle Paleozoic age groups and a near absence of Cretaceous (late Yanshanian) grains (Figure 5g). A notable Neoproterozoic peak is also present and may reflect sediment delivery from Yangtze Block in the northwest.

5. Discussion 5.1. Source-To-Sink Links Between River Systems, the ECSSB, and Taiwan Visual inspection of the KDE plots (Figure 5) reveals that the detrital zircon data from the modern rivers and sink areas can be divided into two groups. The North China Block rivers, Yangtze River, Oligocene strata in the Xihu Sag, and Miocene strata in Taiwan all contain pronounced Precambrian age peaks, while the other age distributions are characterized by prominent Phanerozoic age peaks and minor age populations older than 500 Ma. From the CPP (Figure 6), it is clear that the detrital zircon age distributions from the Eocene– Oligocene strata in Taiwan are most similar to the age distributions of the Min River, with similar proportions of grains suggesting a strong source-to-sink link. Kolmogorov–Smirnov (K-S) tests (Smirnov, 1939; Vermeesch, 2013; Young, 1977) were also used to evaluate the similarity between cumulative grain-age dis-

tributions of Eocene–Oligocene and Miocene strata in Taiwan and potential sources in Figure 6. VK-S was calculated following the method of Malusà et al. (2013, 2016) and is less than 0, indicating that differences

between the compared age distributions are statistically signiﬁcant. However, the VK-S of Eocene– Oligocene strata in Taiwan versus the Min River is closest to 0 (0.0107), suggesting a higher degree of simi-

larity between the two, which is also reﬂected by their proximity in the MDS analysis (Figure 7). The CPP, VK-S (0.0858, closest to 0 compared with others) analysis, and MDS plots also suggest a close relationship between Oligocene samples from the Xihu Sag and Miocene strata from Taiwan. The MDS analysis also reveals several other important relationships. For example, the western Pearl and Yangtze rivers show a close relationship that may relate to the large numbers of Neoproterozoic grains in samples, a diagnostic feature of the Yangtze Block that both rivers cross (Zhang et al., 2017; Zhao et al., 2015; Figure 7). The Min River and Northeastern Pearl River are also close to one another and both ﬂow across the

WANG ET AL. 3681 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Figure 6. Cumulative probability plots of zircon U–Pb ages of modern river sands from major rivers in Southeast China and of strata from Taiwan Island and the Xihu sag. Similarity between distributions is also evaluated by the K–S method, comparing the maximal distance between cumulative frequency curves (bottom) with the critical value for a 0.05 signiﬁ- cance level (Malusà et al., 2013, 2016).

Cathaysia Block. A similar relationship is also observed for the Ou and Juilong rivers, both of which have catchments restricted to the southeast coast (Figure 7). Although the data and analyses suggest that Eocene–Oligocene strata in Taiwan were likely derived from the Min River drainage or a similar source area in the Cathaysia Block, the close relationship between Oligocene strata in the Xihu Sag and Miocene strata in Taiwan implies a possible linkage between the two. The relationship could be explained by a similar source like the Yangtze River or mixture of Yangtze River and rivers in the North China block; however, an alternative that has not been considered for the provenance of Miocene strata in Taiwan is reworking of Oligocene or younger strata in the ECSSB. Considering this possibility, the evolution of the ECSSB is crucial to decipher potential sediment transport models. 5.2. Min River Source for Eocene–Oligocene Strata in Taiwan Although interpretations differ on controlling factors for the provenance change between Eocene–Oligocene and Miocene sediments to the proto-Taiwan region, sediment derivation from the adjacent Cathaysia block and a Min River-like drainage for Eocene–Oligocene strata is widely accepted (Deng et al., 2017; Lan et al., 2016; Xu, 2017; Zhang et al., 2017). Compared to rivers like the Ou and Jiulong, which are areally restricted, the upper reaches of Min River extend into the interior of the Cathaysia Block and into the Wuyishan

WANG ET AL. 3682 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

terrane, where ancient metamorphic basement is exposed (Xu, 2017). This is reﬂected in the age spectrum of the Min River, which is mixed and contains the signature of both the young coast and old interior (Xu, 2017). When compared with the detrital zircon age distribution of the Eocene– Oligocene strata from Taiwan, the pronounced middle Paleozoic (Kwangsian) age peak, the similarity in Phanerozoic and Proterozoic age peaks (Figures 5e and 5f), the overlap in the CPP and the proximity in the MDS analysis seem to conﬁrm the Min River or a similar drainage as the most likely source for sediment.

5.3. Cause for Similarity Between the Oligocene ECSSB and Miocene Taiwan Strata The high degree of similarity between the Min River and the Eocene– Oligocene strata in western Taiwan reflects a robust relationship between the two. Some researchers hypothesize that changes in the age spectra of Eocene–Oligocene and Miocene strata in western Taiwan may simply be the consequence of an evolving Min River (Lan et al., 2016; Xu, 2017). However, the radical change in components, including the introduction of three major Precambrian clusters with peaks at 774, 1,771, and Figure 7. Nonmetric multidimensional scaling map of U–Pb ages from sam- 2,424 Ma, diagnostic of the Yangtze and North China blocks, are difficult ples of modern river sediment, Miocene sequences in the western foothills to account for by simple drainage expansion (Figure 5d). The low contribu- and the Hengchun peninsula, and old Yangtze River sediments. The stress tion of late Mesozoic (Yanshanian) grains, which are most likely derived value is 0.035, indicating high goodness of fit (Vermeesch, 2013). from rocks of similar age that are widely distributed along the southwest coast, are also difficult to reconcile with such a model. Instead, the similarities in the age spectra of Miocene strata in western Taiwan, Oligocene strata in the Xihu Sag, the Yangtze River, and rivers in the North China Block, and their proximity in the MDS plots suggests a closer relationship among these samples. 5.3.1. Pre-24-Ma Initiation of the Yangtze River? Detrital zircon U–Pb data are limited in the ECSSB and previous studies have mainly focused on Paleocene strata in the West Depression Belt (Fu et al., 2015; Yang et al., 2006). These studies show Mesozoic-dominated age spectra, consistent with derivation from the Min-Zhe Uplifts to the west and local structural highs within the basin. Beginning in the Eocene, however, deposition migrated to the EDB (Suo et al., 2015; Zhang et al., 2016), but little data are available for this region. The limited published age data from ECSSB lack significant Precambrian grain populations, diagnostic of ancient cratonal sources for sediment (Yang et al., 2006; Figure 8a), and their stark contrast with Miocene units in the Western Foothills and the Hsueshan Range, which contain large numbers of >600-Ma grains, have been used to infer that the Yangtze River was likely established before the middle Miocene (Zhang et al., 2017). The new data from this study suggest a high degree of similarity between the age spectra of Oligocene strata in the EDB and modern sediment from Yangtze River and North China block rivers (Figures 5a–5c). A possible interpretation from these data is that sediment to the Xihu Sag was derived from a mixture of these potential source regions and may lend support to the existence of the Yangtze River or similar river in the Oligocene or earlier. Although the timing of establishment is debated (e.g., see summary of Deng et al., 2017; Zhang et al., 2017), a pre-24-Ma Yangtze River has been suggested by several studies (e.g., Clift et al., 2008; Zhang et al., 2017; Zheng et al., 2013). Such an interpretation may be consistent with recognized expansion of Oligocene deposits over the Min-Zhe Uplift belt to the west (Yang et al., 2014) and may imply an increase in sediment to the basin, supplied from drainage systems in East Asia. Alternatively, it is also possible that Oligocene sediment in the Xihu Sag and EDB were sourced from nearby structural highs, highs to the north, and/or from the Taiwan-Sinzi belt to the east (Qin et al., 2017; Wang et al., 2011; Yang et al., 2010). Several recent provenance studies have concluded that the Precambrian zircons in Oligocene units in the Xihu Sag were sourced from basement to the north of the Xihu Sag (Wan & Zhang, 2016; Qin et al., 2017). Although this possibility cannot be evaluated directly because the age of potential basement sources is unknown, Proterozoic rocks that are an eastward extension of the South China Block may underlie the region (Wang et al., 2011; Yang et al., 2010). Early international boreholes located close

WANG ET AL. 3683 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Figure 8. Schematic diagrams illustrating the model of sediment delivery. Paleogeographic reconstructions of the East Asian continental margin are based on Deng et al. (2017) and ages used in pie charts are from Zhang et al. (2017). The yellow lines in the pie charts separate the Precambrian age populations from the Paleozoic- Mesozoic ones.

to the Hupingjiao Uplift, north of Xihu Sag, encountered a series of gneiss and schist (JDZ-V-2 by Japan, KV-1 by South Korea) interpreted to be Precambrian in age (Wang et al., 2011; Yang et al., 2010). Additional age data are needed, however, to conﬁrm or reject this hypothesis. 5.3.2. Reworking of Oligocene Strata in ECSSB? The new data from this study show that, in addition to likeness with sediment from the Yangtze River and North China Block rivers, detrital zircon U–Pb ages of Oligocene samples from the Xihu Sag are remarkably similar to those of Miocene strata from western Taiwan Considering the durability of zircon and its potential for recycling, a possible explanation, not considered by previous provenance studies, is that Miocene strata in Taiwan are derived, at least in part, from reworked Oligocene or younger strata in the ECSSB. Given the proximity of the ECSSB to Taiwan, such an interpretation seems possible, but is dependent on the tectonic and sedimentary evolution of the ECSSB and a viable dispersal path for sediment. Although much of the Cenozoic history of the ECSSB is characterized by rifting and thermal subsidence, inversion of earlier rift-related structures is well documented across the East China shelf and may have made con- ditions favorable for recycling and transportation of sediment to a proto-Taiwan (Suo et al., 2015; Zhang et al., 2016; Zhou et al., 2001). Inversion largely initiated in the Oligocene and was primarily concentrated on the eastern part basin, along the Taiwan-Sinzi belt (Diaoyudao Uplift-Fold Belt; Wang et al., 2011). In the Miocene, compression and inversion migrated to the Xihu Sag, in the northern part of the basin, forming a series of NE trending anticlinal structures in areas of former subsidence and sedimentation (Guo et al., 2015; Suo et al., 2017; Zhang et al., 2016; Zhou et al., 2001). Reversal of earlier normal faults and fold development were most intense in the central part of the Xihu Sag, as evidenced by 300 to 1,500 m of erosion across structures in the Central Uplift Zone (Zhang et al., 2016; Zhou et al., 2001), and diminished to the south. The southward tilt of the basin (Zhang et al., 2017), paired with structural highs to the east in the Tanwan- Sinzi belt, which may have blocked major eastward transport of sediment, created a corridor through which sediment may have been transported. However, the cause of the inversion remains controversial and several models are proposed. Some workers cite changes in the rate and direction of subduction between the India and Eurasian plates and/or Paciﬁc and Eurasian plates as the primary cause (Zhang et al., 2016). Guo et al. (2015) suggested that the subduction of Philippine Sea Plate beneath Asia caused the western margin of Okinawa Trough to be pushed westward and

WANG ET AL. 3684 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

facilitated the inversion of structures in the ECCSB. It has also been suggested that the inversion is related to the collision between the Philippine Sea and South China Sea plates (Suo et al., 2017). Although the cause of the inversion is obscure, the primary dynamic source is inferred to be northwestward motion of the Philippine Sea Plate, which dominated during the Miocene (Gutscher, 2001; Suo et al., 2017). Changes in sea level and depositional environments recorded in Eocene to Miocene sedimentary rocks in the ECSSB and in the nearby Taixi and Taixinan basins, located between the southeastern coast and present-day Taiwan, also provide clues to the spatial and temporal evolution of the region and sediment dispersal, and may explain differences in the provenance of Eocene–Oligocene and Miocene units in Taiwan. Sea level in the ECSSB was at its highest during the Eocene, as reflected by a north directed marine transgression. The northern part of the basin was characterized by a semi closed bay environment (Wang et al., 2011), while strata in the southern part of the basin developed neritic-facies (Zang et al., 2016). The facies and sedimentary configuration indicate that the basin, at this time, was strongly influenced by marine processes. However, by the Oligocene, major lowering of sea level led to the development of fluvial, in the north, and lacustrine sequences, in the south (Wang & Zhu, 1992; Zang et al., 2016, and references therein) and a dramatic eastward shift in the position of the shoreline. This change in shoreline position likely enabled transport and delivery of extrabasinal sediment from continental drainage systems to west to the ECSSB. In contrast, the Eocene to Oligocene Taixi Basin, located between mainland China and Taiwan, developed terrestrial sequences that suggest the paleo-shoreline was east of its current position (Zang et al., 2016). Such a configuration may have allowed rivers like the paleo-Min River to provide abundant sediment from the Cathaysia Block to proto-Taiwan to the east. From middle Oligocene to late Miocene, the basin transitions from terrestrial to marine depositions, a result of increasing influence of subsidence and associated marine transgressions in the South China Sea, that may have ultimately limited the influence of nearby continental drainage systems (Zang et al., 2016). The Taixinan Basin to south sees a similar pattern, but marine deposition may have started earlier there, perhaps in the middle Eocene (Li et al., 2016; Zhang et al., 2015). 5.4. New Model for Sediment Transport The detrital zircon age data, when placed into the context of the tectonostratigraphic development of East Asia, the South and East China seas, and Taiwan, permit inferences about the evolution of drainage systems and the source-to-sink history of the region. Based on our analysis, we propose the following model for sediment delivery to Taiwan: (1) During the Eocene, the rate of convergence between the Pacific and Eurasian plates reduced to 30–40 mm/year (Northrup et al., 1995) and a period of intense rifting initiated in the South China Sea and ECSSB. The depositional center of the ECSSB was located in the EDB, with the Xihu Sag, in the northern part of the EDB, characterized by a semiclosed bay (Wang et al., 2011), while neritic facies associated with a south-to-north transgression in sea level developed in the south (Suo et al., 2015; Zang et al.,2016). West tilting of East Asia is inferred during this period and therefore, major eastward flowing drainage systems present today may not have been in place (Deng et al., 2017; Suo et al., 2015). The EDB at this time was characterized by the Taiwan-Sinzi belt to the east (west of the East China Sea shelf outer margin high at this time) and the Min-Zhe, Haijiao, and Yushan uplifts and intervening basins to the west (Wang et al., 2011; Figure 8a). Sediments, mainly sourced from these highs, were transported by small rivers restricted to the coast and deposited into separate deep basins (sags; Fu et al., 2015). At the same time, lacustrine/delta deposits developed south of the ECSSB in the Taixi basin suggesting that the South China Sea remained closed (Zang et al., 2016). Drainage systems may have connected with rivers in the adjacent Cathaysia Block (Min River or similar river) and transported sediment to the proto-Taiwan region (Lan et al., 2016; Xu, 2017; Figure 8a). (2) In the Oligocene, activity along rift-related faults in the EDB slowed or ceased and the depression entered a thermal subsidence phase (Suo et al., 2015; Yang et al., 2014). Depositional environments around the Xihu Sag were predominantly fluvial deltaic and lacustrine (Yu et al., 2017; Zang et al., 2016), and the area of deposition expanded, covering structural highs to the west (Figures 2 and 8b), and likely linked with the drainage systems to the west, which allowed for access of sediments from the interior of the East Asian continent (Figure 8b). Although the timing of reversal of East Asia’s topography to the modern eastward tilt is not clear, major drainage captures that signal the reversal occurred prior to 24 Ma and maybe as early as 30 Ma (Clift et al., 2006). The Yangtze River may have come into its present

WANG ET AL. 3685 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

morphology at a similar time (Zhang et al., 2017; Zheng et al., 2013). The Huagang formation in the Xihu Sag has signatures of both the Yangtze River (Neoproterozoic) and rivers in the North China Block (Mesoproterozoic and Archean grains), which implies that drainage systems that collected material from these regions of ancient craton were established by the Oligocene. In the Late Oligocene, the northwestward compression, largely from subduction of the Philippine Sea Plate, triggered structural inversion (Guo et al., 2015; Suo et al., 2017). (3) In the Miocene, fluvial systems dominated the ECSSB due to retreat of the East China Sea and inversion- related uplift of the basin (Zang et al., 2016). Uplift in the northern ECSSB initiated in the early Miocene and resulted in up to 2,000 m of erosion of the sedimentary cover and basement rocks (Cukur et al., 2011). Continued inversion into the late Miocene resulted in the development of an extensive fold-thrust belt and an additional 900 m of erosion (Cukur et al., 2011). At the same time, large-scale reor- ganizations of drainage systems like the Yangtze River occurred (Zhang et al., 2017; Zheng et al., 2013) and delivered additional sediment to the East China Sea area. Neogene sediments along the Yangtze River are mainly fluvial and suggest that a large river system may have connected such deposits with an emergent East China Sea shelf. Materials transported by the Yangtze River combined with intense uplift and reworking of Oligocene and Miocene strata in the northern ECSSB would have provided a considerable amount of sediment. Sediment delivery to the proto-Taiwan area would have been facilitated by southward tilt of the East China Sea shelf (Suo et al., 2017; Zhang et al., 2016) In contrast, seafloor spreading in the South China Sea and subsequent transgression significantly influenced the depositional environment of the northern South China Sea basin (Clift et al., 2002). From the late Oligocene to Miocene, the Taixi and Taixinan basins are dominated by marine deposition (Zang et al., 2016). Thin-bedded limestones also developed near the base of the Taixinan Basin, suggesting that clastic transport from mainland China was relatively rare at this time (Zang et al., 2016; Figure 8c).

6. Conclusions In this study, we report 378 new detrital zircon U–Pb ages from core and cuttings samples of Oligocene sandstones in the Xihu Sag in the EDB of the ECSSB. Using a compilation of published detrital zircon data from modern sediment from rivers across the East Asia, which provide a clear framework for potential source regions, we show that the age distributions of sediment from the North China Block and Yangtze rivers, Oligocene sandstones from Xihu Sag, and Miocene strata from western Taiwan are similar, implying a source-to-sink relationship. In contrast, the detrital zircon age distributions from the modern Min River corre- late well with those of Eocene–Oligocene strata in the western Taiwan, conﬁrming earlier interpretations that the Min River or similar drainage was the likely source for sediment. The results, when combined with the tectonic and sedimentary evolution of the ECSSB, suggest that the Yangtze River was a major source for Oligocene sediment to the East China shelf and was thus established Acknowledgments as a major sediment transportation system prior to 24 Ma. Reworking of these Yangtze River-derived units The data used are listed in the references, tables, and supporting in the Xihu Sag and delivery to proto-Taiwain was facilitated by tectonic inversion of the ECCSB, a conclusion information. The full suite of data is also that may resolve the long-standing debate on the origin of ancient zircons in Miocene strata in available through the IEDA western Taiwan. (Interdisciplinary Earth Data Alliance) EarthChem Library (www.earthchem. org) at http://doi.org/10.1594/IEDA/ 111206. This study was supported by References the National Science and Technology Chen, C. H., Ho, H. C., Shea, K. S., Lo, W., Lin, W. H., Chang, H. C., et al. (2000). Geologic Map of Taiwan (1/500,000): Central Geological Survey. Major Projects of China (grants Taipei, Taiwan: Ministry of Economic Affairs. 2016ZX05024-002-003, 2016ZX05027- Chen, L. L. (2015). Sedimentary analysis of humid fan in Oligocene Huagang formation of the northern Xihu Sag (in Chinese with English 001-005, and 2017ZX05032-001-004) abstract). Complex Hydrocarbon Reservoirs, 8(4), 1–6. https://doi.org/10.16181/j.cnki.fzyqc.2016.04.001 and the China Scholarship Council. We Clark, M. B., Fisher, D. M., Lu, C., & Chen, C. (1993). Kinematic analyses of the Hsüehshan range, Taiwan: A large-scale pop-up structure. appreciate the positive cooperation Tectonics, 12(1), 205–217. https://doi.org/10.1029/92TC01711 with Shanghai Offshore Oil & Gas Clark, M. K., Schoenbohm, L. M., Royden, L. H., Whipple, K. X., Burchﬁel, B. C., Zhang, X., et al. (2004). Surface uplift, tectonics, and erosion of Company and SINOPEC and technical eastern Tibet from large-scale drainage patterns. Tectonics, 23, TC1006. https://doi.org/10.1029/2002TC001402 support from State Key Laboratory of Clift, P. D., Blusztajn, J., & Nguyen, D. A. (2006). Large-scale drainage capture and surface uplift in eastern Tibet-SW China before 24 Ma Geological Processes and Mineral inferred from sediments of the Hanoi Basin, Vietnam. Geophysical Research Letters, 33, L19403. https://doi.org/10.1029/2006GL027772 Resources. We also thank Lynn Watney Clift, P. D., Layne, G. D., & Blusztajn, J. (2004). Marine sedimentary evidence for monsoon strengthening, Tibetan uplift and drainage evolution for his support in developing collabora- in East Asia. In P. Clift (Ed.), Continent-ocean interactions within East Asian marginal seas, Geophysical Monograph Series (Vol. 149, tions between the China University of pp. 255–282). Washington, DC: American Geophysical Union. Geosciences and the Kansas Geological Clift, P. D., Lee, J. I., Clark, M. K., & Blusztajn, J. (2002). Erosional response of South China to arc rifting and monsoonal strengthening; a record Survey. from the South China Sea. Marine Geology, 184(3–4), 207–226. https://doi.org/10.1016/S0025-3227(01)00301-2

WANG ET AL. 3686 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Clift, P. D., Long, H. V., Hinton, R., Ellam, R., Hannigan, M. R., Tan, M. T., et al. (2008). Evolving east Asian river systems reconstructed by trace element and Pb and Nd isotope variations in modern and ancient Red River-song Hong sediments. Geochemistry, Geophysics, Geosystems, 9, Q04039. https://doi.org/10.1029/2007GC001867 Compston, W., Williams, I. S., Kirschvink, J. L., Zichao, Z., & Guogan, M. A. (1992). Zircon U-Pb ages for the early Cambrian time-scale. Journal of the Geological Society of London, 149(2), 171–184. https://doi.org/10.1144/gsjgs.149.2.0171 Cukur, D., Horozal, S., Lee, G. H., Kim, D. C., Han, H. C., & Kang, M. H. (2011). Structural evolution of the northern East China Sea Shelf Basin interpreted from cross-section restoration. Marine Geophysical Research, 32(3), 363–381. https://doi.org/10.1007/s11001- 011-9114-4 Deng, K., Yang, S., Li, C., Su, N., Bi, L., Chang, Y. P., & Chang, S. C. (2017). Detrital zircon geochronology of river sands from Taiwan: Implications for sedimentary provenance of Taiwan and its source link with the East China mainland. Earth-Science Reviews, 164,31–47. https://doi.org/ 10.1016/j.earscirev.2016.10.015 Diwu, C., Sun, Y., Zhang, H., Wang, Q., Guo, A., & Fan, L. (2012). Episodic tectonothermal events of the western North China craton and north Qinling Orogenic Belt in Central China: Constraints from detrital zircon U–Pb ages. Journal of Asian Earth Sciences, 47, 107–122. https://doi. org/10.1016/j.jseaes.2011.07.012 Fu, X., Zhu, W., Chen, C., Kai, Z., & Xu, C. (2015). Detrital zircon provenance of upper Mingyuefeng formation in west slope of Lishui-Jiaojiang sag, the East China Sea (in Chinese with English abstract). Earth Science, (12), 1987–2001. Gao, S., Rudnick, R. L., Yuan, H. L., Liu, X. M., Liu, Y. S., Xu, W., et al. (2004). Recycling lower continental crust in the North China craton. Nature, 432(7019), 892–897. https://doi.org/10.1038/nature03162 Gehrels, G. (2011). Detrital Zircon U-Pb Geochronology: Current Methods and New Opportunities. In C. J. Busby & A. Azor (Eds.), Tectonics of sedimentary basins: Recent advances (pp. 45–62). Oxford, UK: Wiley-Blackwell. Geng, S., You, G., & Wang, Z. (2011). Simplified tectonic division map of China (1:20,000,000 scale). Institute of Geology, Chinese Academy of Geological Sciences. Guo, Z., Liu, C. Y., & Tian, J. F. (2015). Structural characteristics and main controlling factors of inversion structures in Xihu depression in Donghai Basin (in Chinese with English abstract). Earth Science Frontiers, 22(3), 059–067. Gutscher, M. A. (2001). An Andean model of interplate coupling and strain partitioning applied to the flat subduction zone of SW Japan (Nankai trough). Tectonophysics, 333(1–2), 95–109. https://doi.org/10.1016/S0040-1951(00)00269-9 He, M., Zheng, H., & Clift, P. D. (2013). Zircon U–Pb geochronology and Hf isotope data from the Yangtze River sands: Implications for major magmatic events and crustal evolution in central China. Chemical Geology, 360, 186–203. Ho, C. S. (1986). A synthesis of the geologic evolution of Taiwan. Tectonophysics, 125(1-3), 1–16. https://doi.org/10.1016/0040-1951(86)90004-1 Hu, X., Garzanti, E., Moore, T., & Raffi, I. (2015). Direct stratigraphic dating of India-Asia collision onset at the Selandian (middle Paleocene, 59 +1/1 Ma). Geology, 43(10), 859–862. https://doi.org/10.1130/G36872.1 Huang, C. Y., Yen, Y., Zhao, Q., & Lin, C. T. (2012). Cenozoic stratigraphy of Taiwan: Window into rifting, stratigraphy and paleoceanography of South China Sea. Chinese Science Bulletin, 57(24), 3130–3149. https://doi.org/10.1007/s11434-012-5349-y Huang, C. Y., Yuan, P. B., & Tsao, S. J. (2006). Temporal and spatial records of active arc-continent collision in Taiwan: A synthesis. Geological Society of America Bulletin, 118(3-4), 274–288. https://doi.org/10.1130/B25527.1 Lan, Q., Yan, Y., Huang, C. Y., Santosh, M., Shan, Y. H., Chen, W., et al. (2016). Topographic architecture and drainage reorganization in Southeast China: Zircon U-Pb chronology and Hf isotope evidence from Taiwan. Gondwana Research, 36, 376–389. https://doi.org/ 10.1016/j.gr.2015.07.008 Letouzey, J., & Kimura, M. (1986). The Okinawa trough: Genesis of a back-arc basin developing along a continental margin. Tectonophysics, 125(1–3), 209–230. https://doi.org/10.1016/0040-1951(86)90015-6 Liu, C., Clift, P. D., Carter, A., Böning, P., Hu, Z., Sun, Z., & Pahnke, K. (2017). Controls on modern erosion and the development of the Pearl River Basin since the Eocene. Marine Geology, 394,52–68. https://doi.org/10.1016/j.margeo.2017.07.011 Liu, D. Y., Nutman, A. P., Compston, W., Wu, J. S., & Shen, Q. H. (1992). Remnants of >3800 Ma crust in the Chinese part of the Sino-Korean craton. Geology, 20(4), 339–342. https://doi.org/10.1130/0091-7613(1992)020<0339:ROMCIT>2.3.CO;2 Liu, Y., Hu, Z., Gao, S., Günther, D., Xu, J., & Gao, C. (2008). In situ, analysis of major and trace elements of anhydrous minerals by LA-ICP-MS without applying an internal standard. Chemical Geology, 257(1–2), 34–43. https://doi.org/10.1016/j.chemgeo.2008.08.004 Liu, Y.-S., Hu, Z.-C., Zong, K.-Q., Gao, C.-G., Gao, S., Xu, J., & Chen, H.-H. (2010). Reappraisement and refinement of zircon U-Pb isotope and trace element analyses by LA-ICP-MS. Chinese Science Bulletin, 55(15), 1535–1546. https://doi.org/10.1007/s11434-010-3052-4 Malusà, M. G., Carter, A., Limoncelli, M., Villa, I. M., & Garzanti, E. (2013). Bias in detrital zircon geochronology and thermochronometry. Chemical Geology, 359(6), 90–107. https://doi.org/10.1016/j.chemgeo.2013.09.016 Malusà, M. G., Resentini, A., & Garzanti, E. (2016). Hydraulic sorting and mineral fertility bias in detrital geochronology. Gondwana Research, 31,1–19. https://doi.org/10.1016/j.gr.2015.09.002 Northrup, C. J., Royden, L. H., & Burchfiel, B. C. (1995). Motion of the Pacific plate relative to Eurasia and its potential relation to Cenozoic extension along the eastern margin of Eurasia. Geology, 23(8), 719–722. https://doi.org/10.1130/0091-7613(1995)023%3C0719:MOTPPR% 3E2.3. CO;2 Qin, L., Liu, J., Li, S., Zhao, H., & Chang, Y. (2017). Characteristics of zircon in the Huagang formation of the central inversion zone of Xihu sag and its provenance indication (in Chinese with English abstract). Petroleum Geology & Experiment, 39(4), 498–504. Sibuet, J. C., & Hsu, S. K. (2004). How was Taiwan created? Tectonophysics, 379(1-4), 159–181. https://doi.org/10.1016/j.tecto.2003.10.022 Smirnov, N. V. (1939). On the estimation of the discrepancy between empirical curves of distribution for two independent samples. Bull. Math. Univ. Moscou, 2(2), 3–14. Suo, Y., Li, S., Cao, X., Li, X., Liu, X., & Cao, H. (2017). Mesozoic-Cenozoic inversion tectonics of East China and its implications for the subduction process of the oceanic plate (in Chinese with English abstract). Earth Science Frontiers, 24(4), 249–267. Suo, Y. H., Li, S. Z., Zhao, S. J., Somerville, I. D., Yu, S., Dai, L. M., et al. (2015). Continental margin basins in East Asia: Tectonic implications of the Meso-Cenozoic East China Sea pull-apart basins. Geological Journal, 50(2), 139–156. https://doi.org/10.1002/gj.2535 Vermeesch, P. (2012). On the visualisation of detrital age distributions. Chemical Geology, 312-313, 190–194. https://doi.org/10.1016/j. chemgeo.2012.04.021 Vermeesch, P. (2013). Multi-sample comparison of detrital age distributions. Chemical Geology, 341, 140–146. https://doi.org/10.1016/j. chemgeo.2013.01.010 Wan, Y., & Zhang, N. (2016). Analysis on provenance in Huagang formation of Xihu depression in the East China Sea basin (in Chinese with English abstract). Journal of Yangtze University (Natural Science Edition), 13(35), 24–27. Wang, G. C., & Zhu, W. L. (1992). Cenozoic sedimentary environment in East China Sea basin (in Chinese with English abstract). Acta Sedimentologica Sinica, 10, 100–108.

WANG ET AL. 3687 Geochemistry, Geophysics, Geosystems 10.1029/2018GC007576

Wang, P., Zhao, Z. G., Zhang, G. C., & Zhang, J. W. (2011). Analysis on structural evolution in Diaoyu Islands Folded-Uplift Belt, East China Sea basin and its impact on the hydrocarbon exploration in Xihu sag. Geological Science & Technology Information, 30(4), 65–72. Wang, W., Ye, J., Bidgoli, T., Yang, X., Shi, H., & Shu, Y. (2017). Using detrital zircon geochronology to constrain Paleogene provenance and its relationship to rifting in the Zhu 1 depression, Pearl River Mouth Basin, South China Sea. Geochemistry, Geophysics, Geosystems, 18, 3976–3999. https://doi.org/10.1002/2017GC007110 Wang, Y. J., Fan, W. M., Zhang, G. W., & Zhang, Y. H. (2013). Phanerozoic tectonics of the South China block: Key observations and contro- versies. Gondwana Research, 23(4), 1273–1305. https://doi.org/10.1016/j.gr.2012.02.019 Xu, X., O’Reilly, S. Y., Griffin, W. L., Wang, X., Pearson, N. J., & He, Z. (2007). The crust of Cathaysia: Age, assembly and reworking of two terranes. Precambrian Research, 158(1–2), 51–78. https://doi.org/10.1016/j.precamres.2007.04.010 Xu, Y. (2017). Comment on “Detrital zircon geochronology of river sands from Taiwan: Implications for sedimentary provenance of Taiwan and its Source link with the east China mainland” by Deng K, Yang SY, Li C, Su N, Bi L, Chang YP, Chang SC. [Earth-Science Reviews 164, 31–47]. Earth-Science Reviews, 168, 232–234. Xu, Y., Sun, Q., Yi, L., Yin, X., Wang, A., Li, Y., & Chen, J. (2014). Detrital zircons U-Pb age and Hf isotope from the western side of the Taiwan Strait: Implications for sediment provenance and crustal evolution of the northeast Cathaysia block. Terrestrial, Atmospheric and Oceanic Sciences, 25(4), 505. https://doi.org/10.3319/TAO.2014.02.18.01(TT) Xu, Y., Wang, C. Y., & Zhao, T. (2016). Using detrital zircons from river sands to constrain major tectono-thermal events of the Cathaysia block, SE China. Journal of Asian Earth Sciences, 124,1–13. https://doi.org/10.1016/j.jseaes.2016.04.012 Yang, C. H., Wang, Y. M., Zhang, L., Wan, L. F., Li, K., Zhou, X. H., et al. (2014). Delimitation of Cenozoic rift-depression transitional unconformity in Xihu depression, East China Sea Shelf Basin (in Chinese with English abstract). Marine Origin Petroleum Geology, 19(3), 18–26. Yang, J., Gao, S., Chen, C., Tang, Y., Yuan, H., Gong, H., et al. (2009). Episodic crustal growth of North China as revealed by U–Pb age and Hf isotopes of detrital zircons from modern rivers. Geochimica et Cosmochimica Acta, 73(9), 2660–2673. https://doi.org/10.1016/j. gca.2009.02.007 Yang, W. D., Cui, Z. K., & Zhang, Y. B. (2010). Geology and Minerals of the East China Sea (in Chinese). Beijing: Ocean Press. Yang, X. H., Li, A. C., Qin, Y. S., Wu, S. G., Wu, Z. X., & Zhang, J. P. (2006). U-Pb dating of zircons from Cenozoic sandstone: Constrain on the geodynamic setting of East China Sea Shelf Basin, marine geology & quaternary. Geology, 26(3), 75–86. Ye, J., Qing, H., Bend, S. L., & Gu, H. (2007). Petroleum systems in the offshore Xihu Basin on the continental shelf of the East China Sea. AAPG Bulletin, 91(8), 1167–1188. https://doi.org/10.1306/02220705158 Young, L. (1977). Proof without prejudice: Use of the Kolmogorov-Smirnov test for the analysis of histograms from flow systems and other sources. Journal of Histochemistry & Cytochemistry, 25(7), 935–941. Yu, X. H., Li, S. L., Cao, B., Hou, G. W., Wang, Y. F., & Huangfu, Z. Y. (2017). Oligocene sequence framework and depositional response in the Xihu depression, East China Sea shelf basin (in Chinese with English abstract). Acta Sedimentologica Sinica, 35, 300–314. Zang, Y. B., Li, S. Z., Guo, L. L., Suo, Y. H., Somerville, I. D., Zhao, S. J., et al. (2016). Similarity and differentiation between the East China Sea Shelf Basin and Cenozoic basins in the northeast South China Sea. Geological Journal, 51(S1), 304–317. https://doi.org/10.1002/gj.2826 Zhang, G., Li, S., Suo, Y., & Zhang, J. (2016). Cenozoic positive inversion tectonics and its migration in the East China Sea Shelf Basin. Geological Journal, 51(S1), 176–187. https://doi.org/10.1002/gj.2809 Zhang, H., Shao, L., Zhang, G., Chen, S., & Wu, G. (2015). Distribution and petroleum geologic significance of Eocene marine strata in the South China Sea (in Chinese with English abstract). Earth Science, 40(4), 660–670. Zhang, X., Huang, C., Wang, Y., Clift, P. D., Yan, Y., Fu, X., & Chen, D. (2017). Evolving Yangtze River reconstructed by detrital zircon U-Pb dating and petrographic analysis of Miocene marginal sea sedimentary rocks of the western foothills and Hengchun peninsula, Taiwan. Tectonics, 36, 634–651. https://doi.org/10.1002/2016TC004357 Zhao, M., Shao, L., & Qiao, P. (2015). Characteristics of detrital zircon U-Pb geochronology of the Pearl River Sands and its implication on provenances (in Chinese with English abstract). Journal of Tongji University, 43, 915–923. https://doi.org/10.11908/j.issn.0253- 374x.2015.06.018 Zheng, H., Clift, P. D., Wang, P., Tada, R., Jia, J., He, M., & Jourdan, F. (2013). Pre-Miocene birth of the Yangtze River. Proceedings of the National Academy of Sciences, 110(19), 7556–7561. https://doi.org/10.1073/pnas.1216241110 Zhou, Z., Jiang, J., Liao, Z., Yang, F., & Shang, K. (2001). Basin inversion in Xihu depression, East China Sea. Gondwana Research, 4(4), 844–845.

WANG ET AL. 3688