River is an important part of the surface morphology, and its sediments record rich information about the whole river basin (Kang et al., Reference Kang, Li, Wang and Shao2009; Yang et al., Reference Yang, Zhang and Wang2012; Zheng et al., Reference Zheng, Clift, Wang, Tada, Jia, He and Jourdan2013). The provenance tracing of river sediments is helpful to further understand the tectonic environment, paleoclimate, weathering process and drainage pattern during the sedimentary period, which is of great significance to the understanding of river source-sink system (Nesbitt and Young, Reference Nesbitt and Young1982; Algeo and Maynard, Reference Algeo and Maynard2004; Tribovillard et al., Reference Tribovillard, Algeo, Lyons and Riboulleau2006; Hofer et al., Reference Hofer, Wagreich and Neuhuber2013; Verma and Armstrong-Altrin, Reference Verma and Armstrong-Altrin2016; Zhang et al., Reference Zhang, He, Wang, Clift and Rits2019a, Reference Zhang, Wan, Clift, Huang, Yu, Zhang, Mei, Liu, Han, Nan, Zhao, Li, Chen, Zheng, Yang, Li and Zhang2019b; Quek et al., Reference Quek, Lee, Ghani, Lai, Roselee, Lee, Iizuka, Lin, Yeh, Amran and Rahmat2021; Zhang et al., Reference Zhang, Lu, Zhou, Cui, Zhang, Lv and Chen2022a; Lin et al., Reference Lin, Liu, Liu, Chen and Li2022; Sun et al., Reference Sun, Chen, Wu, Liu, Jin and Wei2022; Li et al., Reference Li, Qian, Xu, Hou, Zhang, Chen, Chen, Qu and Ren2023a).
Provenance analysis, which allows for the diagnosis of the source areas of sediments and the establishment of relationships between source and sink areas through specific geological processes, has been widely used in many fields of the earth sciences (Weltjea and Eynatten, Reference Weltje and Eynatten2004; Walsh et al., Reference Walsh, Wiberg, Aalto, Nittrouer and Kuehl2016; Sun et al., Reference Sun, Xie, Kang, Chi, Du and Wang2021; Wu et al., Reference Wu, Xie, Kang, Chi, Sun and Wei2022). Because of their specific sources, the element geochemical characteristics of sediments can provide important information for deciphering their provenances, tectonic processes and environmental evolution (Yang et al., Reference Yang, Yim and Huang2008; Singh, Reference Singh2009; Clift et al., Reference Clift, Wan and Blusztajn2014); Sr-Nd isotopic composition, especially Nd isotope, is largely unaffected by geological processes and therefore has greater potential for determining sediment sources and transport patterns (Asahara et al., Reference Asahara, Takeuchi, Nagashima, Harada, Yamamoto, Oguri and Tadai2012; Lim et al., Reference Lim, Jung, Xu, Jeong and Li2015; Carter et al., Reference Carter, Griffith, Clift, Scher and Dellapenna2020); heavy minerals record in detail the characteristics of all geological events in the process of sediment transport from source to sink, and especially stable heavy minerals can be well preserved after physical and chemical weathering, and thus can be used as one of the important indicators for provenance identification (Pettijohn et al., Reference Pettijohn, Potter and Siever1987; Mange and Wright, Reference Mange and Wright2007; Yang et al., Reference Yang, Wang, Guo, Li and Cai2009); strong resistance to weathering and abrasion, as well as the ability to retain the characteristics of the source region, make the detrital zircon U-Pb dating widely used in sediment provenance tracing (Bao et al., Reference Bao, Chen and Li2013; Lin et al., Reference Lin, Liu, Wu, Wang, Zhao, Chen, Li and Liu2020; Jones et al., Reference Jones, Orovan, Meffre, Thompson, Belousova, Cracknell, Everard, Bottrill, Bodorkos and Cooke2022). However, it has been proved that a variety of factors, including chemical weathering, sedimentary cycle and source rock types, have a significant impact on the above indexes, which has caused great confusion and challenges for the establishment of large-scale drainage source-sink systems and provenance tracing study (Cao et al., Reference Cao, Liu, Shi, Wei and Zhong2019; Liyouck et al., Reference Liyouck, Ngueutchoua, Armstrong-Altrin, Sonfack, Ngagoum, Bessa, Bela, Tsanga and Wouatong2023). In recent years, although the studies about the factors (chemical weathering, sedimentary recycling and source-area parent rock types) influencing composition of fluvial sediments have been performed in the process of the establishment of source-sink systems in many large drainage systems (Chi et al., Reference Chi, Liu, Hu, Yang and He2021; Huyan and Yao, Reference Huyan and Yao2022; Huyan et al., Reference Huyan, Zhang, Wang, Lu and Liu2023; Li et al., Reference Li, Qian, Xu, Hou, Zhang, Chen, Chen, Qu and Ren2023a; Liyouck et al., Reference Liyouck, Ngueutchoua, Armstrong-Altrin, Sonfack, Ngagoum, Bessa, Bela, Tsanga and Wouatong2023; Saha et al., Reference Saha, Roy, Khan, Ornee, Goswami, Idris, Biswas and Tamim2023), but there is still a lack of relevant studies in Northeast China.
The Songhua River system is one of the most important rivers in Northeast China (Qiu et al., Reference Qiu, Wang, Makhinov, Yan, Lian, Zhu, Zhang and Zhang2014; Wang et al., Reference Wang, Wang, Ran and Su2015a). As important tributaries of the Songhua River, the sediments of the Lalin River (LR) and the Jilin Songhua River (JSR) have a profound influence on the trunk Songhua River basin (Liu et al., Reference Liu, Liu, Zhang, Zhao, Li, Hu, Ding, Lang and Li2013). The source tracing of river sediments is of unique value for the study and division of the basin-mountain coupling system (Matenco and Andriessen, Reference Matenco and Andriessen2013; Arató et al., Reference Arató, Obbágy, Dunkl, Józsa, Lünsdorf, Szepesi, Benkó, Molnár and Eynatten2021). The source areas of the two rivers are the Zhangguangcai Range and the Changbai Mountain, respectively, in which the strata of different ages and various types of parent rocks are interleaved. Accordingly, they are natural geological laboratory for studying the influence of various factors on the source-sink system during the process of sediment transport and deposition. Although the predecessors have studied the elements geochemistry of soil, surface water and groundwater as well as the control factors of the chemical weathering of the sediments in the LR and the JSR basins (Li, Reference Li2008; Liu et al., Reference Liu, Liu, Zhang, Zhao, Li, Hu, Ding, Lang and Li2013; Wang et al., Reference Wang, Hu, Teng, Zhan and Zhai2021), the various indicators have not yet been integrated to trace the source from the perspective of Quaternary geology. Instead, the limited studies focused on the discussion of soil, chemical weathering degree and element differentiation characteristics by element geochemical indexes (Li, Reference Li2008; Liu et al., Reference Liu, Liu, Zhang, Zhao, Li, Hu, Ding, Lang and Li2013; Wang et al., Reference Wang, Hu, Teng, Zhan and Zhai2021). Although Li (Reference Li2010) employed detrital zircon U-Pb age and Hf isotope to study the sediment provenance of the JSR, only one index was selected, and the conclusion inevitably overlooked some information about the parent rock in the source area. At present, no conclusions have been drawn about the extent to which various factors influence the composition of the two river sediments in the process of transport and deposition.
To address the key scientific issues above, the river sediments of the LR and the JSR were selected for the analysis of element geochemistry (major, trace and rare earth elements), Sr-Nd isotopes, heavy minerals and zircon U-Pb dating data to compare the composition differences of the two river sediments with similar provenance areas but different parent-rock types. On this basis, the effects of chemical weathering, sedimentary cycle and parent rock composition on the river sediments are discussed. In addition, in order to reveal the tectono-magmatic events in the eastern Songnen Plain in geological past, we discuss the response of the detrital zircon U-Pb ages of the LR and the JSR sediments on crustal growth and cratonization of the North China Craton (NCC), the closure of the Paleo-Asian Ocean, subduction and rollback of the Paleo-Pacific plate, respectively. This work is helpful to trace the provenance from source to sink in the basin-mountain coupled system and provides an important reference for the study of the Songhua River drainage evolution history.
1. Environmental setting
1. a. Physical geographical setting
The LR, a first-level tributary of the right bank of the trunk Songhua River (Gao et al., Reference Gao, Gao, Cui, Sun, Han, Liu, Huang, Dong and Pang1993), originates from the Shilazi Mountain at the west foot of the Beiyin Mountain in the Zhangguangcai Range, with a total length of 355 km and a drainage area of 2.18 × 104 km2, runs from southeast to northwest and flows into the trunk Songhua River in Fuyu City (Fig. 1) (Wang et al., Reference Wang, Wang and Zhai2019).
Figure 1. Digital elevation model map of the research area, showing the location of the sampling points and the basin range of the Lalin River and the Jilin Songhua River.
The JSR originates from the Tianchi (41° 44’ N, 124° 26’ E), the main peak of the Changbai Mountain, with a total length of 958 km and a drainage area of 7.34 × 104 km2, and flows into, from southeast to northwest, the trunk Songhua River in Songyuan city together with the Nenjiang River (Fig. 1) (Lv et al., Reference Lv, Wang, Li and Ye2017). The two rivers are located in the middle temperate zone and have a continental monsoon climate with an average annual temperature of 2.9–4.2°C and annual precipitation of 550–750 mm, mainly from June to September. In winter, the basin is under the control of the Siberian continental air mass, and the northwest monsoon prevails. In summer, it is affected by the southeast monsoon from the ocean often with heavy rain in that time (Zhou and Yu, Reference Zhou and Yu1984; Gao and Li, Reference Gao and Li2011).
1. b. Geological setting
Northeastern China (1,520,000 km2) is located in the eastern part of the Central Asian Orogenic Belt (CAOB) (Sengör et al.,Reference Sengör, Natal’in and Burtman1993; Wu et al., Reference Wu, Ye and Zhang1994; Jahn et al., Reference Jahn, Wu and Chen2000), at the junction of the NCC and the Siberian Craton (Fig. 2a) (Liu et al., Reference Liu, Li, Ma, Feng, Guan, Li, Chen, Liang and Wen2021a; Liu et al., Reference Liu, Ma, Feng, Li, Li, Guan, Chen, Zhou and Fang2022), with widely exposed granite as its prominent feature (JBGMR, 1988; IMBGMR, 1992; HBGMR, 1993). It is composed of several micro-continental blocks and suture zones between them (Fig. 2b, Wu et al., Reference Wu, Sun, Li and Wang2001; Li, Reference Li2010; Wang et al., Reference Wang, Tong, Huang, Zhang, Guo, Li, Wang, Eglington, Li, Zhang, Donskaya, Petrov, Zhang, Song, Zhang and Wang2022), and was significantly affected by the late Paleozoic closure of the Paleo-Asian Ocean and the Mesozoic Paleo-Pacific subduction and retreat (Wang and Mo, Reference Wang and Mo1996; Wu et al., Reference Wu, Sun, Li, Jahn and Wilde2002; Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011; Liu et al., Reference Liu, Li, Ma, Feng, Guan, Li, Chen, Liang and Wen2021a), as different from other regions in the CAOB.
Figure 2. (a) Tectonic map of Northeast Asia (Wang et al., Reference Wang, Tong, Huang, Zhang, Guo, Li, Wang, Eglington, Li, Zhang, Donskaya, Petrov, Zhang, Song, Zhang and Wang2022). Abbreviations: CAOB = Central Asian Orogenic Belt; MOS = Mongol-Okhotsk Suture; C. = Central; (b) Tectonic division sketch map of NE China (Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011). F1 = Xinlin-Xiguitu Fault; F2 = Hegenshan-Heihe Fault; F3 = Solonker-Xar Moron-Changchun Fault; F4 = Yitong-Yilan Fault; F5 = Jiayin-Mudanjiang Fault; F6 = Dunhua-Mishan Fault; F7 = Chifeng-Kaiyuan Fault; (c) Geologic map of the source area of the Lalin River and the Jilin Songhua River, showing the basin range and the types of the source rocks.
The Songliao Basin was developed in the extensional tectonic environment, with a very high heat flow, and overlay the Songnen-Zhangguangcai Range Massif (SZRM) during Mesozoic (Ma Reference Ma1987; Yang et al., 1997). Most of the basement of the Songliao Basin is composed of Paleozoic and Mesozoic substrata (Wu et al., Reference Wu, Sun, Li and Wang2001, Reference Wu, Sun, Li, Jahn and Wilde2002), and has the characteristics of thin crust (29∼36 km thick) and thin lithosphere (60∼80 km thick) (Ma Reference Ma1987; Yang et al., Reference Yang, Mu, Jin and Liu1996). The Changbai Mountain and Zhangguangcai Range are located on the eastern tectonic unit of the SZRM and span the Dunhua-Mishan, Solonker-Xar Moron-Changchun and Chifeng-Kaiyuan suture zones. The massive Triassic igneous rocks in the eastern SZRM and the western Jiamusi Massif were formed by bidirectional subduction of the Mudanjiang Ocean between the SZRM and Jiamusi Massif (Wang et al., Reference Wang, Xu, Meng, Cao and Gao2012; Wei, Reference Wei2012; Shao et al., Reference Shao, Li and Tang2013; Yang et al., Reference Yang, Ge, Zhao, Yu and Zhang2014; Guo et al., Reference Guo, Xu, Yu, Wang, Tang and Li2015; Lv et al., Reference Lv, Xiao, Feng, Li, Deng, Ren and Zheng2015; Wang et al., Reference Wang, Xu, Xu, Gao and Ge2015c; Yang et al., Reference Yang, Ge, Zhao, Dong, Xu, Wang, Ji and Yu2015; Zhao et al., Reference Zhao, Ge, Yang, Dong, Bi and He2018; Zhang et al., Reference Zhang, Chen, Zhao, Tang, Li, Feng and Kong2022b). The Zhangguangcai Range was formed by the subduction and subsequent overthrust of Silurian oceanic crust (Wang et al., Reference Wang, Xu, Meng, Cao and Gao2012; Shao et al., Reference Shao, Li and Tang2013; Zhang et al., Reference Zhang, Chen, Zhao, Tang, Li, Feng and Kong2022b), and it has successively experienced an early arc magmatic stage (220-179 Ma) and a late metamorphic-deformation stage (193-165 Ma) (Shao et al., Reference Shao, Li and Tang2013). The formation of this orogenic belt marks the transition from the Mesozoic Paleo-Asiatic tectonic domain to the Pacific tectonic domain (Shao et al., Reference Shao, Li and Tang2013), and the existence of the granite in this area is also closely related to the collision event (Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011). The source rocks in the Zhangguangcai Range are mainly Triassic-Cretaceous intermediate-acid rocks (40.65%) and Devonian-Permian sandstone, conglomerate and granite (6.69%) (Fig. 2c and Table 1), with zircon U-Pb ages being in the range of Triassic (45.05%) and Jurassic (36.88%) period (Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011; Xu et al., Reference Xu, Pei, Wang, Meng, Ji, Yang and Wang2013; Wang et al., Reference Wang, Xu, Xu, Gao and Ge2015c).
Table 1. Areal percentages (%) of geological units in the Lalin River (LR) and Jilin Songhua River (JSR) catchments. The data here are calculated by ArcMap based on the digitized geological map of Northeast China and the digital elevation model.
The Changbai Mountain Range tectonically belongs to the terrigenous accretion belt on the northeast margin of the NCC (Qiao et al., Reference Qiao, Huang, Zhou and Gao2016), and its source-rock types are mainly Triassic-Cretaceous intermediate-acid rocks (23.29%), Neoarchean granite-gneiss (8.50%), Cretaceous sand conglomerate, monzonite, conglomerate and volcanic rock (8.10%), Pleistocene basalt (7.74%) and Permian intermediate-felsic rocks and granite (6.46%) (Fig. 2c and Table 1). The zircons U-Pb ages of the Changbai Mountain Range are marked by a conspicuous Precambrian peak (about 51.40% of the total ages, especially 21.86% for 2.5 Ga), and meanwhile also show a considerable number of Paleozoic (29.77%) and Mesozoic (18.84%) ages (Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011; Wang et al., Reference Wang, Xu, Pei, Wang and Guo2016b; Zhang et al., Reference Zhang, Zhang, Zeng, Liu and Cui2017). The Tianchi in the Changbai Mountain is a young bimodal volcano and is located in the uplifted area between the Songliao Basin and the Japan Sea back-arc basin (Liu et al., Reference Liu, Chen, Guo, Gou, He, You, Kim, Sung and Kim2015), the uplift of which occurred during the middle-upper Pliocene to Pleistocene (Wang et al., Reference Wang, Li, Wei and Shan2003). It experienced three stages early shield formation (2-1 Ma), middle cone formation (1-0.02 Ma) and late eruption (Liu et al., Reference Liu, Fan, Zheng, Zhang and Li1998). A series of faulting and uplifting activities caused several eruptions of basaltic magma from the mantle magma chamber to the surface. After the volcanic eruption, a large area of basalt platform was formed around the Tianchi volcanic cone (Horn and Schmincke, Reference Horn and Schmincke2000; Wang et al., Reference Wang, Li, Wei and Shan2003; Siebert et al., Reference Siebert, Simkin and Kimberly2010; Wei et al., Reference Wei, Liu and Gill2013; Liu et al., Reference Liu, Chen, Guo, Gou, He, You, Kim, Sung and Kim2015; Qiao et al., Reference Qiao, Huang, Zhou and Gao2016).
2. Materials and methods
2. a. Sampling
The fine-grained components of river sand can represent the average composition in a large area after being strongly transported and mixed by wind and water power (Xie et al., Reference Xie, Kang, Chi, Du, Wang and Sun2019a, Reference Xie, Kang, Chi, Wu, Wei, Wang and Sun2020). In order to constrain the sediment composition and provenance of the LR and the JSR, a total of 15 samples were obtained from the river point bars (Fig. 1), 7 from the JSR and the rest from the LR. All the samples obtained in this study are river sands near riverbed, lithologically different from the black soils through which the river flows, thus reflecting the composition of the source-area parent rocks. In addition, the samples are far away from urban population activity areas.
2. b. Analytical methods
In this paper, the obtained sediments were taken for elemental geochemistry, heavy mineral, Sr-Nd isotope and detrital zircon U-Pb dating analysis. After air-dried at room temperature, the bulk samples were dry-screened using standard sieve to obtain <63 μm fraction for element and Sr-Nd isotope analysis, <63 μm, 63–125 μm, 125–250 μm fractions for heavy mineral analysis and 63–250 μm fraction for U-Pb dating of detrital zircon.
2. b.1. Element geochemistry
Major elements were analyzed by a standard X-ray fluorescence spectrometer (AL104, PW2404) on fused glass beads. The detection limit is ∼0.01 wt% and analytical precision (relative standard deviation) is <1%. Loss on ignition was obtained by weighing before and after 1 h of heating at 950 °C. The tests for trace and rare earth elements were performed by an inductively coupled plasma mass spectrometer (ICP-MS, Finnigan MAT ElementI) with relative standard deviations of less than ± 5% and ± 1%, respectively. The external calibration was carried out by using Chinese National Standard soil reference samples (GSS-28). The sample pretreatment and testing of elemental geochemistry were carried out at the State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences (Wuhan).
2. b.2. Heavy mineral analysis
Three grain-sized fractions (<63 μm, 63–125 μm, 125–250 μm) were used to characterize the heavy mineral composition of the LR and the JSR. The sub-samples were first weighed after drying in the oven of <60 °C, elutriated repeatedly with clear water in a washing pan for the removal of impurities and a dispersion of mineral grains and then separated with bromoform (density = 2.88) to collect heavy minerals. The collected heavy minerals were further separated, using magnetic and electromagnetic methods, into diamagnetic, paramagnetic and ferrimagnetic groups, and then weighed separately again. The identification for the heavy mineral groups was performed using stereomicroscopy and a polarized microscope, with identification numbers of each mineral species totalling up to 1000 grains. The weight percentage of heavy minerals was calculated. Details for heavy mineral analysis have been described by Xie et al. (Reference Xie, Kang, Chi, Wu, Wei, Wang and Sun2020). The pretreatment and heavy mineral analysis of the samples were carried out at Chengxin Geological Service Co., Ltd., Langfang City, Hebei Province, China.
2. b.3. Sr-Nd isotope
The sub-samples of <63 μm fraction were soaked in 0.5 mol/L acetic acid solution at room temperature for 4h (Biscaye et al., Reference Biscaye, Grousset, Revel, Van der Gaas, Zielins and Kukla1997; Wang et al., Reference Wang, Yang, Chen, Zhang and Rao2007; Chen et al., Reference Chen, Li, Yang, Rao, Lu, Balsam, Sun and Ji2007) in order to eliminate the influence of secondary carbonates on Sr isotope composition and ensure that only detrital minerals in sediments were analyzed (Wang et al., Reference Wang, Yang, Chen, Zhang and Rao2007; Újvári et al., Reference Újvári, Varga, Ramos, Kovács, Németh and Stevens2012), and then dried and ground to 200 mesh. The Sr (87Sr/86Sr) and Nd (143Nd/144Nd) isotopic ratios were determined by thermal ionization mass spectrometry. The Sr and Nd isotopic ratios of the sub-samples were corrected for mass fractionation using 86Sr/88Sr = 0.1194 and 146Nd/144Nd = 0.7219, respectively. The accuracy of the instrument was tested with the international standard samples NBS987 and JMC, the measured values of which were 87Sr/86Sr = 0.710250 ± 7 (2σ) and 143Nd/144Nd = 0.512109 ± 3 (2σ), respectively. Chemical analysis blanks are <1 ng for Sr and <50 pg for Nd. The Nd isotopic composition is expressed by the εNd (0) value, which is calculated as follows: εNd(0) = {[(143Nd/144Nd) sample/(143Nd/144Nd) CHUR] – 1} × 10000.
Among them, (143Nd/144Nd) CHUR uses the modern value of chondrite uniform reservoir (CHUR) values of 143Nd/144Nd = 0.512638. The pretreatment and testing were carried out at the Key Laboratory of Mineral Resources in Western Gansu Province, Lanzhou University, China.
2. b.4. U-Pb dating of detrital zircon
Two 63-250 μm sub-samples were selected for zircon U-Pb dating. Zircon particles were extracted by conventional heavy liquid and magnetic separation methods, and then about 1000 grains were randomly selected by hand-picking under a binocular microscope. About 300 zircon particles were randomly selected to be fixed in epoxy discs and then polished to about half their thickness to expose its internal micro-structure. Transmitted electron, backscattered electron and cathode luminescence photographs were taken to observe their internal micro-textures. In the process of selecting potential target spot sites for laser denudation, inclusions and fracture sites are avoided as much as possible. Due to the influence of crystallization temperature, magmatic zircons usually have oscillating rings, the different positions of which also affect the zircon ages (Hanchar and Miller, Reference Hanchar and Miller1993; Corfu et al., Reference Corfu, Hanchar, Hoskin and Kinny2003). Therefore, we try to avoid laser denudation points or ion beams passing through oscillating bands with different characteristics and choose bands with relatively uniform characteristics to avoid meaningless mixing ages (Wu and Zheng, Reference Wu and Zheng2004).
Zircon U-Pb dating was performed by laser denudation-inductively coupled plasma mass spectrometry (LA-ICP-MS) at the Key Laboratory of Mineral Resources in Western Gansu Province, Lanzhou University, China. The instrument was calibrated with the reference material NIST SRM 610 of synthetic silicate glass. The 91500 standard zircon and Plesovice (PLE) standard zircon were used as external and internal standards for age calibration monitoring (Wiedenbeck et al., Reference Wiedenbeck, Allé, Corfu, Griffin, Meier, Oberli, Quadt, Roddick and Spiegel1995; Pearce et al., Reference Pearce, Perkins, Westgate, Gorton, Jackson, Neal and Chenery1997; Slàma et al, Reference Sláma, Košler, Condon, Crowley, Gerdes, Hanchar, Horstwood, Morris, Nasdala, Norberg, Schaltegger, Schoene, Tubrett and Whitehouse2008). The spot diameter is 30 μm, the denudation time is 40 s and the denudation depth is about 20 μm. The isotope ratio and element content data of the zircon grains were analyzed by Glitter 3.0 software, and the error of each age data was controlled within 1σ (Ludwig, Reference Ludwig2003). The data of standard lead were calibrated by Andersen (Macro) (Andersen, Reference Andersen2002), and the Probability Density Plot and age histogram were drawn by Isoplot 3.0 programme (Ludwig, Reference Ludwig2003). The 206Pb/238U and 207Pb/235U ages were used for ages younger and older than 1 Ga, respectively, and only ages with discordance within ±10% were accepted, resulting in a total of 257 zircons with concordant ages.
3. Results
3. a. Element geochemistry
3. a.1. Major elements
The major element distribution in the studied river sediments shows strong uniformity (Fig. 3a), but the contents of some major elements are slightly different. Specifically, the LR and JSR sediments contain similar contents of some major elements, such as SiO2 (70.1–74.6% and 64.8–68.7%), Al2O3 (12.2–14.1% and 13.1–13.9%), Na2O (3.0–3.7% and 2.5–3.1%), K2O (2.8–3.0% and 2.9–3.1%), TiO2 (0.6–0.8% and 0.7–1.0%), P2O5 (0.1-0.1% and 0.1–0.2%), FeO (0.8–1.4% and 0.7–1.5%). However, some major elements are different, such as Fe2O3 (2.2–3.2% and 4.1–5.2%), MgO (0.6-0.9% and 1.0–1.3%), CaO (1.5–1.9% and 1.7–3.5%), MnO (0.1–0.2% and 0.05–0.10%). Compared with Upper Continental Crust (UCC), SiO2 contents in the sediments of the two rivers are similar, and TiO2 and MnO are significantly enriched. Al2O3, Na2O and K2O are slight losses, and MgO, CaO and P2O5 are significant losses. In the sediments of the two rivers, the depletion of Fe2O3, MgO and CaO of the JSR sediments is smaller than that of the LR sediments, while the MnO content of the LR is significantly higher than that of the JSR. A detailed major element composition of the two river sediments is listed in Supplementary Table S1.
Figure 3. The normalized patterns for elements and chondrite-normalized rare earth element patterns for the Lalin River and Jilin Songhua River sediments. UCC and Post Archean Australian shale (PAAS) patterns are given as a reference. UCC, PAAS and chondrite values are after Taylor and McLennan (Reference Taylor and McLennan1985).
3. a.2. Trace elements
The characteristics of trace elements in the sediments of both the rivers show obvious similarities (Fig. 3b). Relative to UCC, the high field strength element Y (23.9–41.4 and 27.1–31.2 ppm for the LR and JSR, respectively, similarly hereinafter), Zr (273–1774 and 198–514 ppm), Th (7.5–19.7 and 10.8–18.2 ppm) are enriched in the LR and JSR sediments. Large ion lithophilic elements are enriched in Cs (3.30-4.96 and 3.88-6.89 ppm), Ba (544–784 and 562–635 ppm) and Pb (18.7–27.4 and 19.3–25.8 ppm), and depleted in Rb (84–102 and 99–111 ppm) and Sr (226–253 and 226–274 ppm) in the sediments of the two rivers. Compared with UCC (1.50 ppm), the Mo (0.15–0.49 and 0.32–1.00 ppm) loss of both rivers is obvious. Compared with the JSR, the LR has significant loss of Nb (13.7–19.4 and 17.2–34.2 ppm), a slight loss of V (38.8-63.1 and 66.0-89.5 ppm) and Zn (33.8–58.5 and 71.3–80.8 ppm) and an enrichment of U (2.32–6.28 and 1.84–2.54 ppm). A detailed trace element composition of the river sediments is listed in Supplementary Table S1.
3. a.3. Rare earth elements (REE)
The contents of total rare earth elements (ΣREE) in the sediments are 125–223 ppm for the LR and 163–227 ppm for the JSR, higher than those in UCC (146 ppm) and PAAS (Post Archean Australian shale, 185 ppm). The patterns of rare earth elements in the two rivers are similar to those in UCC and PAAS, and show a right-leaning pattern (that is, ‘the left side is steep and the right side is slow’ trend, with (La/Yb)N values of 5.5–7.7 for the LR and 6.2–9.2 for the JSR), showing enrichment of light rare earth elements (110-195 ppm for the LR and 143-206 ppm for the JSR), depletion of heavy rare earth elements (16-28 ppm for the LR and 19–22 ppm for the JSR) and significant negative Eu anomaly (0.50-0.71 for the LR and 0.50–0.64 for the JSR), as indicated in the chondrite-normalized REE patterns (Fig. 3c). The Ce anomalies are not clear with the values of 0.97–1.0 for the LR and 0.88–1.1 for the JSR. A detailed rare earth element composition of the two rivers is listed in Supplementary Table S1.
3. b. Heavy mineral
In total, 15 types of heavy minerals were detected in the two river sediments, including zircon, apatite, rutile, anatase, leucoxene, titanite, hornblende, tourmaline, garnet, epidote, pyroxene, ilmenite and ferromagnetic minerals (haematite-limonite, magnetite, magnetic haematite), etc. A detailed heavy mineral composition of the <63 μm, 63–125 μm and 125–250 μm is listed in Supplementary Table S2. The heavy mineral composition is very different in the two rivers and in different grain-sized fractions (Fig. 4). In the <63 μm fraction, the LR consists mainly of hornblende (44.02%) and epidote (13.50%), followed by zircon (11.01%) and magnetite (8.03%), and other minerals account for 23.44%; the JSR is dominant in haematite-limonite (29.19%), magnetic haematite (18.44%), epidote (16.87%) and hornblende (13.12%). In the 63-125 μm fraction, hornblende in the LR increases significantly up to 65.42%, while the JSR is dominated by hornblende (37.63%) and epidote (25.87%). In the 125-250 μm fraction, hornblende occupies an absolute dominance (73.95%) in the LR, and the JSR has a higher content of hornblende (52.67%) and epidote (28.23%).
Figure 4. Heavy mineral abundances of the Lalin River and Jilin Songhua River sediments at <63 μm, 63–125 μm and 125–250 μm fractions, expressed as relative weight percentage (wt%).
3. c. Sr-Nd isotopic composition
The 87Sr/86Sr ratios in the LR and JSR sediments are similar, with a value of 0.7099–0.7106 and 0.7105–0.7112, respectively. The εNd (0) values can distinguish well among the two river sediments, −6.2812–8.5831 for the LR and −8.1149 to −10.2411 for the JSR (Fig. 5). A detailed Sr-Nd isotopic composition of two rivers is listed in Supplementary Table S1.
Figure 5. Sr-Nd isotopic composition in the Lalin River and Jilin Songhua River sediments.
3. d. U-Pb dating of detrital zircon
Th/U ratios of most of the zircons are >0.1 (Fig. 6), indicating their magmatic origin, and a total of 257 zircon grains with Concordia ages were produced (Fig. 7). The zircon age patterns of the two rivers are clearly different (Fig. 8a-d). Specifically, the zircon ages of the LR sediments are mainly Mesozoic and Paleozoic (66–252 Ma, 252–541 Ma), whereas, in addition to the Mesozoic and Paleozoic zircons, a considerable number (36.92%) of Paleoproterozoic and Neoarchean (1.6–2.5 Ga, 2.5–2.8 Ga) zircons occur in the JSR sediments, suggesting their notably different age of the parent rocks in the sediment source areas. A detailed U-Pb dating of detrital zircon of the river sediments is listed in Supplementary Tables S3 and S4, respectively.
Figure 6. Diagram of zircon ages vs. Th/U ratios from the Lalin River and the Jilin Songhua River. The high Th/U ratios (>0.1) of most of the analyzed zircons suggest a magmatic source, while the low Th/U ratios (<0.1) suggest a metamorphic source.
Figure 7. U-Pb Concordia diagrams (a-b) for the detrital zircon grains of the Lalin River and Jilin Songhua River sediments.
Figure 8. PDP plots for the detrital zircon ages of the two rivers. (a) the Lalin River in this study; (b) the Lalin River in this study (only containing less than 500 Ma); (c) the Jilin Songhua River in this study; (d) the Jilin Songhua River in this study (only containing less than 500 Ma); (e) igneous zircon ages in the Great Xing’an Range compiled from Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011; Liu et al., Reference Liu, Lv, Chen, Yang, Wang and Sun2014; Zhang et al., Reference Zhang, Wu, Jiang, Guo, Yang, Liu, Cui, Li, Li, Si and Zhang2015; Wang et al., Reference Wang, Du, Xu, Chen, Sun, Lu, Hou and Cheng2018; Song et al., Reference Song, Du, Na, Pang, Yang, Liu, Sun, Tao and Ge2019; Wang and Yang, Reference Wang and Yang2019; Chen et al., Reference Chen, Li, Yu, Wang, Zhen and Gong2021; Jia et al., Reference Jia, Ren, Xu, Ma, Zhang and Shi2022; (f) igneous zircon ages in the the Lesser Xing’an Range compiled from Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011; Li et al., Reference Li, Zhang, Sun and Lv2019; Wang et al., Reference Wang, Pei, Xu, Cao, Wang and Zhang2016a; (g) igneous zircon ages in the Zhangguangcai Range compiled from Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011; Xu et al., Reference Xu, Pei, Wang, Meng, Ji, Yang and Wang2013; Wang et al., Reference Wang, Xu, Xu, Gao and Ge2015c; (h) igneous zircon ages in the Changbai Mountains compiled from Wu et al., Reference Wu, Sun, Ge, Zhang, Grant, Wilde and Jahn2011; Wang et al., Reference Wang, Xu, Pei, Wang and Guo2016b; Zhang et al., Reference Zhang, Zhang, Zeng, Liu and Cui2017; (i) igneous zircon ages in the Songnen Plain basement compiled from Wang et al., Reference Wang, Zhang, Zhang, Miao, Li, Xie, Meng and Liu2006; Li et al., Reference Li, Chen, Siebel, Wu, Zhu, Shan and Sun2012; Song et al., Reference Song, Zhou, Guo and Li2012; Jia et al., Reference Jia, Jiang, Wang and Gao2016; Huang et al., Reference Huang, Zhu, Wang and Cheng2019; Zang et al., Reference Zang, Li, Ning and Wang2023.
4. Discussion
4. a. Effects of chemical weathering on the river sediments
The element geochemical composition of sediments has been widely used in exploring the parent-rock types and provenance tracing of source areas (Cox et al., Reference Cox, Lowe and Cullers1995; Jahn et al., Reference Jahn, Gallet and Han2001; Ding et al., Reference Ding, Sun, Yang and Liu2001), but the composition can be affected by chemical weathering and even suffers alteration (Shao et al., Reference Shao, Yang and Li2012; Clift et al., Reference Clift, Wan and Blusztajn2014; Li et al., Reference Li, Zhang, Sun and Lv2019). Therefore, it is necessary to evaluate the effects of chemical weathering on the geochemical indices of sediments before source tracing (McLennan et al., Reference Mclennan, Taylor and Eriksson1983; Feng and Kerrich, Reference Feng and Kerrich1990; Roddaz et al., Reference Roddaz, Christophoul, Zambrano, Soula and Baby2012).
Chemical Index of Alteration (CIA) is one of the important indexes to measure the chemical weathering degree of sediments (Nesbitt and Young, Reference Nesbitt and Young1982; Shao et al., Reference Shao, Yang and Li2012; Huyan and Yao, Reference Huyan and Yao2022). The formula is: CIA = [Al2O3/(Al2O3 + CaO* + Na2O + K2O)] × 100. In the expression, the principal component refers to the mole fraction, where CaO* is the molecular weight of CaO in the silicate part (Mclennan, Reference McLennan1993a). Larger CIA values correspond to higher degrees of chemical weathering (Fedo et al., Reference Fedo, Nesbitt and Young1995). In this study, the CIA values of the LR sediments are similar to those of the JSR sediments, with the range of 51.4–57.6, indicating incipient chemical weathering degree, which is comparable to the chemical weathering level of Quaternary aeolian sediments and river sediments in this area (CIA = 51.2–60.0, Xie et al., Reference Xie, Yuan, Zhan, Kang and Chi2018a; Xie et al., Reference Xie, Yuan, Zhan, Kang, Chi and Ma2018b; Zhao et al., Reference Zhao, Xie, Chi, Kang, Wu, Sun and Wei2023).
Because K-metasomatism may interfere with the determination of CIA values, the use of CIW (Chemical Index of Weathering, CIW = [Al2O3/(Al2O3 + CaO* + Na2O)] × 100) can effectively avoid the migration of K element during diagenesis or metamorphism (Harnois,Reference Harnois1988). The CIW values of the LR and JSR sediments are 59.7–62.7 and 61.5–66.1, respectively, showing a low degree of chemical weathering. Plagioclase Index of Alteration (PIA) is also used for a correction for K-metasomatism (Fedo et al., Reference Fedo, Nesbitt and Young1995), and calculated as follows: PIA = [(Al2O3 - K2O)/(Al2O3 + CaO* + Na2O - K2O)] × 100. The PIA values (52.0–60.2) of the two rivers show low chemical weathering levels. In the A-CN-K compositional space constructed for this study (Fig. 9), the samples are plotted close to UCC and parallel to the A-CN line, indicating no effect of K-metasomatism and low levels of chemical weathering (Nesbitt and Young, Reference Nesbitt and Young1984; Fedo et al., Reference Fedo, Nesbitt and Young1995).
Figure 9. A-CN-K (Al2O3-CaO*+Na2O-K2O) ternary diagram (in molecular proportion, after Nesbitt and Young, Reference Nesbitt and Young1984), characterizing weathering trend of the Lalin River and Jilin Songhua River sediments. UCC and Post Archean Australian shale values are from Taylor and McLennan (Reference Taylor and McLennan1985).
Weathering Index of Parker (WIP = [(2Na2O/0.35) + (MgO/0.9) + (2K2O/0.25) + (CaO*/0.7)] × 100, Parker, Reference Parker1970) is more sensitive to the accumulation of quartz and zircon caused by sedimentary cycle and sorting. The WIP values of the two rivers are similar but have little difference, ranging from 56.0 to 65.5, which indicates low levels of the chemical weathering. In addition, Rb/Sr ratio was also widely used to determine the degree of chemical weathering (Chen et al., Reference Chen, An and Head1999; Jin et al., Reference Jin, Cao, Wu and Wang2006; Xu et al., Reference Xu, Liu and Wu2010; Chang et al., Reference Chang, An, Wu, Jin, Liu and Song2013; An et al., Reference An, Lai, Liu, Fan and Wei2018). The Rb/Sr ratios of the LR and JSR are 0.38 and 0.42, respectively, much lower than the PAAS values (Taylor and Bradley, Reference Taylor and Bradley1985), indicating incipient degrees of chemical weathering (Wang et al., Reference Wang, Zhu, Dong, Chen, Su, Liu and Wu2020).
In the process of chemical weathering, some immobile element references (e.g., Al and Ti) are usually selected to calculate the mass migration coefficient of each element (Nesbitt, Reference Nesbitt1979). In this study, the high contents of Al element in the sediments indicate no clear leaching of Al element (Supplementary Tables S1). Therefore, Al element is selected as the immobile element reference in this study to evaluate the relative loss or enrichment degree of elements in chemical weathering process by using the material balance coefficient (Xchange). The formula for material balance coefficient is:
Xchange = (Xws/Yws)/(Xpr/Ypr) − 1
In the formula, Xws and Xpr are the concentration of elements in the sample and UCC, respectively, and Yws and Ypr are the concentration of immobile elements in the sample and UCC (i.e., Al), respectively. The results show a clear leaching of Mg and Ca, but no clear leaching of Na and K (Fig. 10). Cautions are that in the source rocks of the two rivers, the proportion of acid igneous rocks is the largest (Table 1), and the Mg content is relatively low. Therefore, the relative loss of Mg is due to the difference of source-rock types rather than chemical weathering.
Figure 10. Bivariate plots showing relative depletion or enrichment of major elements relative to element Al reference.
In conclusion, both two river sediments are marked by incipient chemical weathering, thus indicating a weak effect of chemical weathering on the composition of the river sediments. It draws a conclusion that the geochemical composition of the two river sediments can be employed for the source-area tracing.
4. b. The influence of sedimentary recycling on the river sediments
The sorting and recycling of minerals during the transport and sedimentation processes of sediments will affect the material composition of river sediments, and even mask the provenance information from the parent rocks of source area (Condie, Reference Condie1991; Cullers and Podkovyrov, Reference Cullers and Podkovyrov2000; Schneider et al., Reference Schneider, Hornung, Hinderer and Garzanti2016). Therefore, it is necessary to evaluate the effect of sedimentary recycling on river sediments.
SiO2/Al2O3 ratio is often used to evaluate sediment maturity, with a positive correlation with sediment maturity (El-Bialy, Reference El-Bialy2013; Armstrong-Altrin et al., Reference Armstrong-Altrin, Nagarajan, Balaram and Natalhy-Pineda2015). In magmatic rocks, the variation range of this index is small, from about 3 in basic rocks (such as basalt) to about 5 in acidic rocks (such as granite and rhyolite) (Lemaitre, Reference Lemaitre1976; Roser et al., Reference Roser, Cooper, Nathan and Tulloch1996). Therefore, a ratio greater than 5 or 6 indicates the maturity and sediment cycles, while a ratio greater than 7 indicates the strong maturity of sediments (Xie et al., Reference Xie, Lu, Kang and Chi2019b). The mean values of SiO2/Al2O3 in the LR and the JSR are 5.79 and 5.03, respectively, indicating low maturity.
The Index of Compositional Variability (ICV) is usually used to evaluate the maturity characteristics of sediments and distinguish the sedimentary recycling process (Cox et al., Reference Cox, Lowe and Cullers1995; Cullers and Podkovyrov, Reference Cullers and Podkovyrov2000; Armstrong-Altrin et al., Reference Armstrong-Altrin, Nagarajan, Balaram and Natalhy-Pineda2015; Wang et al., Reference Wang, Zhang, Dai and Lan2015b; Perri et al., Reference Perri, Dominici, Pera, Chiocci and Martorelli2016; Huyan et al., Reference Huyan, Yao and Xie2021). The formula is: ICV = (CaO + K2O + Na2O + Fe2O3 + MgO + TiO2 + MnO)/Al2O3. In the formula, CaO represents the CaO in all components, and Fe2O3 represents the total iron content. ICV>1 indicates a low maturity of sediments, the first deposition under tectonic activity, while ICV<1 indicates a high maturity of sediments, the first cycle under high chemical weathering or strongly affected by recycling (Weaver, Reference Weaver1989; Cox et al., Reference Cox, Lowe and Cullers1995; Tobia et al., Reference Tobia, Al-Jaleel and Ahmad2019). The ICV values of the LR and the JSR are similar, with the mean values of 1.0 and 1.2, respectively, indicating their low maturities and/or low degree of recycling characteristics.
WIP, combined with CIA indicator, is more accurate to identify the degree of recycling (Fig. 11a) (Garzanti et al, Reference Garzanti, Padoan, Andò, Resentini, Vezzoli and Lustrino2013; Garzanti and Resentini, Reference Garzanti and Resentini2016). The CIA/WIP ratios of the LR and JSR sediments are less than 2, which proves their initial sedimentary cycling (Garzanti et al., Reference Garzanti, Padoan, Andò, Resentini, Vezzoli and Lustrino2013). In the CIA-WIP binary diagram (Fig. 11a), the sediments in this study are plotted along the trend line of chemical weathering of UCC, indicating a poor degree of sedimentary recycling. In mafic-felsic-weathering (Ohta and Arai, Reference Ohta and Arai2007) ternary diagram (Fig. 11b), the data points of the LR and JSR sediments are plotted along the igneous rock trend line, close to UCC and far away from the W vertex, indicating a low degree of weathering and a low sedimentary recycling (Ohta and Arai, Reference Ohta and Arai2007; Xie et al, Reference Xie, Lu, Kang and Chi2019b; Li et al., Reference Li, Qian, Xu, Hou, Zhang, Chen, Chen, Qu and Ren2023a).
Figure 11. Discriminant diagrams for sedimentary recycling in the Lalin River and Jilin Songhua River sediments. (a) CIA vs. WIP plot distinguishing first-cycle and recycling sediments (after Garzanti et al.,Reference Garzanti, Padoan, Andò, Resentini, Vezzoli and Lustrino2013). Because quartz dilution strongly affects WIP but not CIA, the CIA/WIP plot readily reveals quartz enrichment in sediments due to recycling sedimentation. The studied samples plot well below UCC weathering trend line, reflecting some degrees of sedimentary recycling; (b) The MFW (Mafic-Felsic-Weathering) plots, served as depicting tendency and intensity of weathering (after Ohta and Arai, Reference Ohta and Arai2007) as well as distinguishing first-cycle from recycling processes (after Ohta, Reference Ohta2008). The studied sediments plot along a trend close to igneous rock compositional trend and to the F-W line, but far away from W apex, suggesting low degree of chemical weathering and dominant felsic and minor intermediate rocks in their primitive source areas. In addition, the weathering trend of these sediments does not extend towards the W vertex but instead, arrays along igneous rock compositional trend towards the M vertex, indicating that they cannot be represented by the weathering of the igneous rocks, i.e. they are of nature of recycling sedimentation; (c) Th/Sc v. Zr/Sc bivariate plot (after McLennan et al., Reference McLennan, Hemming, Mcdaniel and Hanson1993b) for the sediments of the Lalin River and the Jilin Songhua River, identifying whether the sediments are derived from recycled sedimentation. UCC and Post Archean Australian shale values are from Taylor and McLennan (Reference Taylor and McLennan1985). M = exp.(−0.395*ln(SiO2) + 0.206*ln(TiO2)–0.316*ln(Al2O3) + 0.16*ln(Fe2O3) + 0.246*ln(MgO) + 0.368*ln(CaO*) + 0.073*ln(Na2O)–0.342*ln(K2O) + 2.266). F = exp. (0.191*ln(SiO2)–0.397*ln(TiO2) + 0.02*ln(Al2O3)–0.375*ln(Fe2O3)–0.243*ln (MgO) + 0.079*ln(CaO*) + 0.392*ln(Na2O) + 0.333*ln(K2O)–0.892). W = exp. (0.203*ln(SiO2) + 0.191*ln(TiO2) + 0.296*ln(Al2O3) + 0.215*ln (Fe2O3) – 0.002*ln(MgO)–0.448*ln(CaO*)–0.464*ln(Na2O) + 0.008*ln(K2O)–1.374).
Th/Sc-Zr/Sc binary diagram can well evaluate the potential effect of sedimentary processes (e.g., sorting and recycling) on sediments (McLennan et al., Reference Mclennan, Taylor, Mcculloch and Maynard1990; Xie et al., Reference Xie, Lu, Kang and Chi2019b). Zr/Sc ratio commonly shows a clear increase but a far less increase of the Th/Sc ratio during multiple recycled processes (McLennan et al., Reference Mclennan, Taylor, Mcculloch and Maynard1990; McLennan et al., Reference McLennan, Hemming, Mcdaniel and Hanson1993b). In the Th/Sc-Zr/Sc binary diagram (Fig. 11c), the LR sediments array along zircon addition direction with a deviation from the magmatic composition trend line, whereas the JSR sediments are plotted close to the magmatic composition trend line slightly with zircon addition trend, indicating a minimal effect of sedimentary processes.
4. c. Control of the river sediments by source rocks
Source rocks are one of the most important factors affecting the composition of river sediments, which in turn indicates that the composition of river sediments also has a good response to sediment provenance properties and their tectonic settings (Gabo et al., Reference Gabo, Dimalanta, Asio, Queaño, Yumul and Imai2009; Jian et al., Reference Jian, Guan, Zhang and Feng2013). In this section, we use mineralogy, elemental geochemistry and detrital zircon U-Pb dating to distinguish the types of parent rocks in the source area and explore the extent of their influence on the river sediments.
4. c.1. Change characteristics of heavy minerals
The LR and JSR sediments show the notable differences in some of the characteristic heavy minerals (Fig. 4), e.g., haematite-limonite, magnetic haematite, epidote, apatite, hornblende, magnetite, pyroxene and zircon. The different parent rocks in the source areas and later weathering and alteration are responsible for these differences. Large areas of mafic rocks (e.g., basalts) are outcropped in the source area and course of the JSR (Fig. 2c), and they are depleted in zircon and are highly susceptible to weathering and alteration, thus resulting in more weathering-alteration minerals (e.g., haematite, limonite, magnetic haematite) occurring in the JSR sediments (Fig. 4). In addition, the formation of epidote is a result of the dynamic metamorphism of basic igneous rocks and/or the potassium-metasomatism of hornblende (Marmo, Reference Marmo1979). The large area of basalt outcrops in the source area of the JSR and the multiple tectonic movements in geological history (Liu et al., Reference Liu, Chen, Guo, Gou, He, You, Kim, Sung and Kim2015; Qiao et al., Reference Qiao, Huang, Zhou and Gao2016) have provided the high epidote content in the JSR.
Significantly, the marked differences in the heavy mineral composition can be observed in the different grain-sized fractions, most notably in hornblende and epidote (Fig. 4). More specifically, hornblende is clearly enriched in the coarse-sized fractions (especially 125–250 μm) compared to the <63 μm in the two rivers, indicating its grain size dependency being controlled by hydrodynamic sorting (Mange and Maurer, Reference Mange and Maurer1992). However, epidote presents the contrasting enrichment patterns in the two rivers, enrichment in <63 μm for the LR and in 125–250 μm for the JSR, which cannot be explained by mineral sorting. In addition to mineral sorting during sedimentary transportation, initial size of the heavy minerals in source areas also affects grain size dependency of heavy mineral composition. In this sense, more caution should be exercised when using heavy minerals for source tracing, as characterized by other observations (Zhang et al., Reference Zhang, Geng, Pan, Hu, Chen, Wang, Chen and Zhao2020).
4. c.2. Change characteristics of Sr-Nd isotopes
Sr isotopes are subject to potential factors, e.g., chemical weathering, diagenesis and particle size effects during weathering, transport and deposition (Walter et al, Reference Walter, Hegner, Diekmann, Kuhn and Rutgers van der loeff2000; Rao et al., Reference Rao, Yang, Chen and Li2006; Rao et al., Reference Rao, Chen, Tan, Jiang and Su2011), thus providing less provenance information (Chen et al., Reference Chen, Li, Yang, Rao, Lu, Balsam, Sun and Ji2007). However, Nd isotopes are very little affected by non-source factors (Jones et al., Reference Jones, Halliday, Rea and Owen1994; Grousset and Biscaye, Reference Grousset and Biscaye2005; Rao et al., Reference Rao, Yang, Chen and Li2006; Chen et al., Reference Chen, Li, Yang, Rao, Lu, Balsam, Sun and Ji2007; Asahara et al., Reference Asahara, Takeuchi, Nagashima, Harada, Yamamoto, Oguri and Tadai2012) and thus have been widely used as a powerful source tracer (Goldstein et al., Reference Goldstein, O’Nions and Hamilton1984; Goldstein and Jacobsen, Reference Goldstein and Jacobsen1988; Aubert et al., Reference Aubert, Probst, Stille, Stille and Viville2002; Jiang et al., Reference Jiang, Xiong, Frank, Yin and Li2019).
The two river sediments share the similar Sr isotopic ratios, as also is the case for their similar degree of chemical weathering and sedimentary recycling, whereas Nd isotopic ratios distinguish well between the two river sediments, with significantly positive Nd isotopic ratios for the LR (Fig. 5). This observation once indicates the source-area dependency of Nd isotopic ratio. Notably, both the two river sediments show the significant variations of the εNd (0) values, with an inter-sediment variation in the same river exceeding 2ε, which is well in excess of the threshold (1ε units) of experimental error (Xie et al., Reference Xie, Kang, Chi, Wu, Wei, Wang and Sun2020). These findings indicate that Nd isotopic variations between sediments in the same river are essentially triggered by the sustained denudation and thus addition of different parent rocks in process of sediment transportation. This is an indication of the heterogeneity of the river sediments.
4. c.3. Change Characteristics of U-Pb ages in detrital zircons
The two river sediments are of the drastically different detrital zircon U-Pb age characteristics (Fig. 8), with a clear peak age at ∼2.5 Ga for the JSR. The 100-400 Ma-aged zircons generally occur in the mountains surrounding the Songnen Plain, e.g., the Great Xing’an Range, the Lesser Xing’an Range, the Zhangguangcai Range and the Changbai Mountains, despite slightly different peak age. Accordingly, the two rivers have the similar age pattern at 100–400 Ma intervals, despite a few sub-peak ages in the JSR. The Precambrian detrital zircons only occur in the basement of the Songnen Plain and the northeastern NCC (Fig. 8), but almost lacking in the young orogenic belt (Great Xing’an Range, Lesser Xing’an Range and Zhangguangcai Range). The notable peak age of ∼2.5 Ga in the JSR but not in the LR indicates the detrital contributions of the northeastern NCC to the JSR, as evidenced by the widely outcropped Neoarchean parent rocks (e.g., amphibolite, granulite, gneiss and granite, Fig. 2c) in the headwater. Accordingly, the lithological differences in the headwater area have led to significantly different detrital zircon U-Pb age patterns in the sediments of the two rivers.
Notably, the U-Pb age composition of the detrital zircons in the two rivers does not well match the exposed area ratios of different types of the source rocks in the basin (Fig. 8a-d and Table 1). This discordance can be attributed to the differences such as power and range of fluvial erosion, as well as zircon fertility in source areas.
4. c.4. Evaluation of immobile element ratios in identification of the parent-rock nature
Some immobile trace elements and their ratios are usually used to depict the parent-rock properties (felsic/mafic source) of sediments (Cullers et al., Reference Cullers, Basu and Suttner1988; Condie and Wronkiewicz, Reference Condie and Wronkiewicz1990). For example, high-field strength elements La, Th, Zr and U are mainly enriched in acidic (felsic) parent rocks (Cullers, Reference Cullers1995), and transition elements Sc, Co, Cr and Ni are mainly enriched in basic (mafic) rocks (Feng and Kerrich, Reference Feng and Kerrich1990). In the discrimination diagrams characterizing parent-rock nature with immobile elements in this study (Fig. 12), the two river sediments cluster together in the felsic source region, indicating their felsic parent rocks in origin. This is true for the LR sediments because their source area is lithologically dominated by late Triassic-Cretaceous granite and intermediate-acid igneous rocks (Fig. 2c). However, this is not the case for the JSR sediments. In the upstream area southeast of Jilin city, basalts are sporadically exposed in the intermediate-acid igneous rocks, especially in the headwater near Fusong county, where basalts are widely exposed, whereas the downstream area northwest of Jilin city is covered by the sporadic intermediate-acid igneous rocks and widely exposed Quaternary sediments. The basalt information occurring in the upper reaches of the JSR is not reflected in the commonly used discrimination diagrams of parent-rock natures (Fig. 12), and we speculate that these basalt records are likely to have been masked by the records from the widely outcropped intermediate-acid igneous rocks. In this sense, the employment of immobile element ratios to the parent-rock nature identification needs to be treated with caution.
Figure 12. Discrimination diagrams illustrating source-rock nature with immobile trace elements for the Lalin River and Jilin Songhua River sediments. (a) Co/Th vs. La/Sc plot after Gu (Reference Gu1994); (b) La/Th–Hf diagram after Floyd and Leveridge (Reference Floyd and Leveridge1987); (c) Discriminant diagram of TiO2%-Zr; (d) Provenance discrimination function diagram (after Roser and Korsch, Reference Roser and Korsch1986). The studied samples plot across felsic igneous and quartzose sedimentary provenance fields, suggesting characteristics of sediment recycling and felsic components in their ultimate source area. (e) Discriminant diagram of Th/Co-La/Sc. DF1 = 30.638 × TiO2/Al2O3 – 12.541 × Fe2O3/Al2O3 + 7.329 × MgO/Al2O3 + 12.031 × Na2O/Al2O3 + 35.42 × K2O/Al2O3 – 6.382. DF2 = 56.500 × TiO2/Al2O3 – 10.879 × Fe2O3/Al2O3 + 30.875 × MgO/Al2O3 – 5.404 × Na2O/Al2O3 + 11.112 × K2O/Al2O3 – 3.89.
4. d. Implications for regional tectonic and magmatic activities in the Southeastern Songnen Plain
The river sediments obtain the attention of a wide range of scholars due to their records of information about source-area tectonic and magmatic activities using detrital zircon ages (Jie et al., Reference Jie, Gao, Yuan, Gong, Zhang and Xie2007; Han et al., Reference Han, Guo, Chen, Huang, Zong, Liu, Hu and Gao2017; Liang et al., Reference Liang, Gao, Hawkesworth, Wu, Storey, Zhou, Li, Hu, Liu and Liu2018; Li et al., Reference Li, Zhang, Li, Liu and Qin2023b). The headwater regions of the LR and the JSR are tectonically located at the eastern part of the CAOB and the northeastern end of the NCC, respectively. Accordingly, the study area involves the geological evolution history of tectonic and magmatic events, e.g., the formation of the Precambrian NCC, the closure of the Paleo-Early Mesozoic Asian Ocean and the subduction and retreat of the Paleo-Pacific plate in the Mesozoic.
It is generally accepted that peak ages of detrital zircons correspond to a certain scale of tectonic-magmatic events that occurred in the source area during this period (Kong et al., Reference Kong, Guo, Wan, Liu, Wang and Chen2022), as different from biased geochemical methods (Rudnick and Gao, Reference Rudnick and Gao2003). In this study, the main age peaks of 119–306 Ma in the LR (Fig. 8a, b) as well as 120–368 Ma and 2480–2667 Ma in the JSR (Fig. 8c-f) are the comprehensive responses to the multi-stage tectono-magmatic events in the different periods in the study area.
The early Precambrian period was the rapid growth stage of the Earth’s continents, and 70% of the continental crust was formed during the Archean period (Condie, Reference Condie and Condie1994, Reference Condie1998). The NCC, one of the oldest cratons in the world, underwent complex multi-stage tectonic evolution (Kusky et al., Reference Kusky, Li and Santosh2007; Wan et al., Reference Wan, Liu, Xu, Dong, Wang, Zhou, Yang, Liu and Wu2008, Reference Wan, Liu, Dong, Xie, Kröner, Ma, Liu, Xie, Ren and Zhai2015, Reference Wan, Dong, Ren, Bai, Xie, Liu, Xie and Liu2017), with age back as far as 4.0-3.8 Ga (Wan et al., Reference Wan, Song, Liu, Li, Yang, Zhang, Yang, Geng and Shen2001, Reference Wan, Xie, Dong and Liu2020, 2021a, 2021b). Paleoarchaean-early Neoarchaean (3.6-2.6 Ga) continental growth was dominated by plate floor support or mantle overturning activities, which caused large-scale magmatic activity and metamorphism, and Trondhjemite, Tonalite and Granodiorite (TTG) rocks began to be found at many sites in NCC (Wan et al., Reference Wan, Dong, Xie, Wang, Song, Xu, Wang, Zhou, Ma and Liu2012, Reference Wan, Dong, Ren, Bai, Xie, Liu, Xie and Liu2017). However, pre-Mesoarchean (2.8 Ga) mainly existed in the eastern, southern and central landmasses divided by Wan et al. (Reference Wan, Liu, Dong, Xie, Kröner, Ma, Liu, Xie, Ren and Zhai2015). The three main tectono-magmatic events occurred in the NCC since the Neoarchean, i.e., extensive crustal accretion at 2.9-2.7 Ga, crustal growth and cratonisation marked by numerous tectonic and metamorphic-magmatic events at 2.5 Ga, and the final formation of the craton at 1.8 Ga (Shen et al., Reference Shen, Wu and Geng1999; Zhai and Bian, Reference Zhai and Bian2000; Zhao et al., Reference Zhao, Wilde, Cawood and Sun2001, Reference Zhao, Sun, Wilde and Li2005; Kusky and Li, Reference Kusky and Li2003; Kröner et al., Reference Kröner, Wilde, O’Brien, Li, Passchier, Walte and Liu2005; Kusky et al., Reference Kusky, Li and Santosh2007; Geng et al., Reference Geng, Shen and Ren2010; Zhai, Reference Zhai2010; Geng et al., Reference Geng, Du and Ren2012; Diwu et al., Reference Diwu, Sun and Wang2012; Hu et al., Reference Hu, Zhai, Peng, Liu, Diwu, Wang and Zhang2013; Wang et al., Reference Wang, Xie, Chi, Kang, Wu, Sun and Liu2023). Among them, the most important magmatic-thermal events occurred at 2.5 Ga (Kröner et al., Reference Kröner, Wilde, O’Brien, Li, Passchier, Walte and Liu2005; Shen et al., Reference Shen, Geng, Song and Wan2005; Zhao et al., Reference Zhao, Sun, Wilde and Li2005; Diwu et al., Reference Diwu, Sun and Wang2012), as is the specificity of NCC different from other ancient continents (Windley, Reference Windley1995; Liu et al., Reference Liu, Guo, Lu and Diwu2009). The upper reaches of the JSR, e.g., the areas near Huadian, Jingyu and Fusong county, are lithologically marked by the widely outcropped Neoarchean amphibolite, granulite, gneiss and granite (Fig. 2c), which, combined with the prominent peak age at 2.5–2.6 Ga of the detrital zircons in the JSR (Fig. 8), confirms the strong tectono-thermal events in that period in the northeastern NCC. This result also corroborates that 2.5–2.6 Ga event is one of the strongest tectono-magmatic events in the NCC (Kröner et al., Reference Kröner, Wilde, O’Brien, Li, Passchier, Walte and Liu2005; Zhao et al., Reference Zhao, Sun, Wilde and Li2005; Liu et al., Reference Liu, Guo, Lu and Diwu2009; Diwu et al., Reference Diwu, Sun and Wang2012; Wang et al., Reference Wang, Xie, Chi, Kang, Wu, Sun and Liu2023), which induced an important crustal growth and thus its cratonisation. Notably, the 2.9-2.7 Ga and 1.8 Ga events are not documented in the northeastern NCC.
The formation of the Solonker-Xar Moron-Changchun suture zone represents the formation of the southern edge of the CAOB and the final closure of the Paleo-Asian Ocean, closely related to the collision between the NCC and the Siberian Craton (Xu et al., Reference Xu, Pei, Wang, Meng, Ji, Yang and Wang2013; Wang et al., Reference Wang, Guo, Zhang, Yang, Zhang, Tong and Ye2015d). The timing of the final closure of the Paleo-Asian Ocean has aroused wide controversy (Xiao et al., Reference Xiao, Windley, Hao and Zhai2003, Reference Xiao, Windley, Huang, Han, Yuan, Chen, Sun, Sun and Li2009; Windley et al., Reference Windley, Alexeiev, Xiao, Kröner and Badarch2007; Ren et al., Reference Ren, Zhang, Sukhbaatar, Hou, Wu, Yang, Li and Chen2023; Wang et al., Reference Wang, Li, Xiao, Zheng, Wang, Jiang and Brouwer2024), with variable ages at Carboniferous (Chen et al., Reference Chen, Wang, Li, Evans and Chen2022), Lower Permian (Li, Reference Li2006), Upper Permian-Lower Triassic (Sun et al., Reference Sun, Suzuki, Wu and Lu2005; Li et al., Reference Li, Gao, Sun, Li and Wang2007; Cao et al., Reference Cao, Xu, Pei, Wang, Wang and Wang2013), Lower Triassic (Wu et al., Reference Wu, Zhao, Sun, Wilde and Yang2007a) and Middle Triassic (Wang et al., Reference Wang, Li, Xiao, Zheng, Wang, Jiang and Brouwer2024). We conclude, therefore, that the minor detrital zircon peak age at Carboniferous-Triassic (300-230 Ma, Fig. 8) in the JSR responds to the closure event of the Paleo-Asian Ocean. It is universally recognized that the Paleo-Pacific plate began to subduct westward beneath the Eurasian continent at the Jurassic period (Wu et al., Reference Wu, Yang, Lo, Wilde, Sun and Jahn2007b; Tang et al., Reference Tang, Xu, Wang and Ge2018a; Li et al., Reference Li, Xu, Zhou, Wang, Ge and Sorokin2020; Li et al., Reference Li, Xu, Zhang and Tang2024) and rollback eastward at the Early Cretaceous period (Tang et al., Reference Tang, Sun, Mao, Yang and Deng2018b; Ma and Xu, Reference Ma and Xu2021). This subduction and rollback triggered the large-scale magmatic activity in the late Mesozoic in northeast China, which made the Mesozoic igneous rocks widely developed in the upper reaches of the LR and the JSR (Fig. 2c). Therefore, the prominent peak ages of detrital zircons in the two rivers, ranging from 200 to 150 Ma (Fig. 8) are a good response to the subduction event of the ancient Pacific Plate. Compared to very poor age group at ∼120 Ma in the LR (Fig. 8), slightly pronounced peak age at ∼120 Ma in the JSR documents the rollback event of the Paleo-Pacific plate. The notable difference between detrital zircon ages and resultant records of regional tectono-magmatic events for the two rivers is determined by the different parent rocks in the source areas.
5. Conclusions
The sediments from the LR and the JSR are compared in terms of geochemistry (element and Sr-Nd isotopic composition), heavy mineral and detrital zircon U-Pb dating. The major conclusions are drawn as follows:
(1) The two rivers have similar geochemical compositions (such as elements and Sr isotopes) as well as weak chemical weathering and cycling characteristics, indicating that the geochemical composition of the river sediments is mainly controlled by climate rather than provenance.
(2) Significant differences exist in the detrital zircon U-Pb ages, Nd isotopic ratios and to a certain extent heavy mineral composition for the two rivers, indicating source control. The heterogeneity of the river sediments is indicated by the significantly variable Nd isotope ratios. The information from basic parent rocks could not be identified by element geochemical methods due to the dilution of widely exposed acidic rocks.
(3) The clear U-Pb multi-peak ages of the detrital zircons from the JSR record the crust growth and cratonization of the NCC during the 2.5 Ga period. The minor peak age of the Carboniferous-Triassic (300-230 Ma) detrital zircons in the JSR responds to the Paleo-Asian Ocean closure event. The prominent age of the detrital zircons in the two rivers is 200-150 Ma, which is a good response to the subduction event of the Paleo-Pacific plate. Compared with the very poor age group of ∼120 Ma in the LR, the slight peak age of ∼120 Ma in the JSR records the rollback event of the Paleo-Pacific plate. The significant differences in the detrital zircon ages and regional tectono-magmatic records between the two rivers are determined by the different parent rocks in the source areas.
Supplementary material
The supplementary material for this article can be found https://doi.org/10.1017/S0016756824000517
Acknowledgements
The geochemical analysis was supported by the State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences (Wuhan), heavy minerals were detected by Chengxin Geology Company, Langfang, Hebei, and the U-Pb dating analysis of detrital zircons was measured by Ms. Xiaoli Yan, Lanzhou University. Thanks to Fengzhan Guo and Xiaoyu Han for their participation in the discussion of the paper writing.
Financial support
This study was financially supported by the National Natural Science Foundation of China (Grant: 42171006 and 41871013).
Competing interests
The authors declared that they have no conflicts of interest in this work. We declare that we do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted.
Open access
