j ourna l homepage: www.e lsev ie r .com/ locate /chemgeo
The intensity of chemical weathering: Geochemical constraints frommarine detrital sediments of Triassic age in South China
Ming-Yu Zhao, Yong-Fei Zheng ⁎CAS Key Laboratory of Crust-Mantle Materials and Environments, School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China
A geochemical study of major-trace elements in detrital sediment and carbon–oxygen isotopes in carbonate wascarried out for a marine stratigraphic profile of Early Triassic that is composed of argillaceous limestone andcalcareous mudstone in the Lower Yangtze basin, South China. The results place constraints on the geochemicalbehaviors of various elements in the detrital sediment that was deposited in the residual Paleotethyan seawater.This leads to establishment of new geochemical proxies for chemical weathering of continental crust. In terms ofthe correlations between element concentrations and their variations in the profile, the elements are categorizedinto four groups with respect to the difference in their geochemical behaviors. The first group is composed of Al,Th, Sc, Be, In, Ga, K, Rb and Cs that are tightly correlated due to their immobility during chemical weathering. Thesecond group is composed of Ca and Na that show opposite variation trends with Th and Sc, on account of theirmobile behavior in theweathering profile. The third group is composed of highfield strength elements such as Ti,Nb, Ta, Zr and Hf that are closely correlatedwith each other because they were primarily taken up by heavymin-erals from sedimentary provenance. The fourth group is composed of redox sensitive elements such as Co, Cu, Fe,Mn and Ni that are correlated with S and thus mainly hosted by sulfides. Th, Sc, Ca and Na were not amenable tochanges in sedimentary provenance, and thus are selected to establish the new proxies for chemical weathering.These are composed of logarithmic parameters such as log(Th/Ca), log(Sc/Ca), log[Th/(Na/5 + Ca)] and log[Sc/(Na/5 + Ca)]. They exhibit synchronous increases at the Permian–Triassic boundary, the middle Griesbachianand the early Smithian, indicating the enhancements of chemical weathering. High proxy values approachingthe values for the extremely weathered product of granodiorite occurred in the middle to late Griesbachianand early Smithian, demonstrating the occurrences of extreme chemicalweathering and verywarmpaleoclimatein those periods. These paleoclimatic changes are concordantwith results from geochemical studies elsewhere inthe world. Therefore, the intensity of chemical weathering can be indicated by the new geochemical proxies forthe different properties of elements in marine detrital sediments.
Marine sedimentary rocks record the overall information on theevolution of the Earth's surface in the geological history. Chemicalprecipitates (e.g., carbonate, sulfate, phosphate, organic material andpyrite) record the geochemistry of seawater, including temperature,composition, productivity and redox status. On the other hand, detritalsediments record the composition of provenance rocks from thecontinental crust due to chemical weathering. In fact, the chemicalweathering is a process that converts atmospheric CO2 and silicaterocks to alkalinity and divalent cations, which are then buried on theseafloor as chemical precipitates (e.g., Berner et al., 1983; Kump et al.,2000). As such, the chemical weathering has played an important rolein regulating the stability of Earth's climate (e.g., Kump and Arthur,1997; Molnar, 2004). While much attention has been paid to the
chemical precipitates from the seawater, rare studies have been devot-ed to the geochemistry of terrigenous detritus linked to the chemicalweathering.
The chemical weathering is a major mechanism that partitionselements between crustal rocks and natural water (e.g., Taylor andMcLennan, 1985). It is primarily controlled by the climate in the sourceregion, but its effects on the composition ofweathered products are alsodictated by the property of crustal rocks. Elements with differentmobil-ity are differentiated during chemical weathering, and the degree ofchemical differentiation is determined by the intensity of chemicalweathering. If some elements in weathered terrigenous detritus sufferminor effect from further differentiation during fluvial transport, theratios between them in ultimate marine sediments can be used togain information about the intensity of chemical weathering. After re-moving chemical precipitates such as carbonate, element ratios of detri-tal materials such as Al/Ti, Al/K, Al/Na, Ca/Ti, K/Ti, Na/Ti, Sc/Ti, Rb/Sr,La/Lu and La/Sm in mixed carbonate and detrital sediments have beenused to decipher the intensity of chemical weathering (Li et al., 2003;
Wei et al., 2003, 2006). In addition, element X/Al ratios such as K/Al,Mg/Al, Rb/Al and Ti/Al for untreated mixed carbonate and detrital sedi-ments also have been used to trace the intensity of chemicalweathering(Zabel et al., 2001; Clift et al., 2008; Sun et al., 2008; Tian et al., 2011).However, no such kinds of geochemical studies were performed on an-cient marine sediments before the Quaternary.
A number of studies have utilized the Chemical Index of Alteration(CIA) of ancient detrital sediments to trace the change in chemicalweathering (Nesbitt and Young, 1982; Scheffler et al., 2003; Rieu et al.,2007; Yan et al., 2010). Because CIA is calculated by the relation ofmolar [Al2O3/(Al2O3 + CaO⁎ + Na2O + K2O)] × 100 where CaO* onlyrepresents the CaO in silicate minerals (Nesbitt and Young, 1982),individual variables may have impact on the calculated CIA values. Forinstance, K2O concentrations may be complicated by secondary enrich-ment (Nesbitt and Young, 1989; Fedo et al., 1996), and Al2O3 concentra-tions may be affected by chemical precipitates (Murray and Leinen,1996; Kryc et al., 2003). As a consequence, the CIA index is unlikelyapplicable to all sedimentary environments.
Because of the difference in the mobility of various elements duringthe chemical weathering of crustal rocks (Zhao and Zheng, 2014), theintensity of chemical weathering in the source regionmay be quantifiedby measuring the difference. The abundances and ratios of elements inmarine detrital sediments, if properly organized, can be used for thispurpose. Due to the complexity of geological processes, however, asingle index for chemical weatheringmay be subjected to a lot of inter-ferences, and thus not applicable to all sedimentary environments.Therefore, a comprehensive understanding of the general behaviorsand interferential factors of various elements in marine detritalsediments is necessary in order to establish geochemical proxies forthe intensity of chemical weathering.
In this paper, we present a combined study of geochemicalapproaches to understand the behaviors of major and trace elementsin marine detrital sediments from the Lower Yangtze basin of EarlyTriassic in South China. The target region was a residual Paleotethyansea between two converging continental blocks at that time, with theseawater of euxinic property for marine deposition. On the otherhand, a geochemical study of paleosols indicates enhanced chemicalweathering in the Early Triassic (Sheldon, 2006). In order to quantifythe intensity of chemical weathering, we select the appropriate ele-ments that are sensitive to chemical weathering and only have sufferedminor influence from the other factors. In doing so, the carbon andoxygen isotope compositions of impure carbonates in the sequencewere also analyzed for stratigraphic comparison. As such, the recordsof chemical weathering are used to investigate their implications forthe paleoclimatic change in the Early Triassic. Consequently, the resultsprovide insights into the relationship between tectonism, chemicalweathering and paleoclimate in this period.
2. Geological setting and samples
The South China Block is composed of three landmasses of theCathaysian terrane, the Jiangnan orogen and the Yangtze Craton(Zheng et al., 2013). It was located in the eastern edge of thePaleotethyan Ocean during the Late Paleozoic (e.g., Luo et al., 2010).There were two large basins of the Lower Yangtze and Qian-Gui-Xiang, and a connected platformwith carbonates and detrital sedimentsat that time (Fig. 1A). Due to the progressive convergence between theSouth China Block and the North China Block during the Late Paleozoic,continental collision took place in the Triassic to result in the Qinling-Tongbai-Hong’an-Dabie-Sulu orogenic belt (Wu and Zheng, 2013). Asa consequence, sedimentary environments in the Middle-LowerYangtze basins would be evolved from a broad Paleotethyan ocean toa narrow Paleotethyan sea during the Paleozoic (Zheng et al., 2013).Stratigraphic sequences may vary from deposition of deep water to de-position of shallow water in this period, with a residual Paleotethyansea in the Early Triassic prior to the final collision between the two
continental blocks. Marine deposition in these basins exhibits the fluc-tuation between transgression and retrogression phases due to variablerates of the continental convergence in the Late Permian to EarlyTriassic. Consequently, euxinic seawater is expected for the marine de-position prior to the continental collision in the Middle-Lower Yangtzebasins.
This study dealswith sedimentary rocks of Early Triassic from the sub-urb of Chaohu city in Anhui province (Fig. 1B). It is about 180 km awayfrom the Meishan section in the Zhejiang province (Fig. 1A), which isknown as the Global Stratotype Section and Point of the Permian–Triassicboundary (PTB). The Chaohu area was located in the central part of theLower Yangtze basin in the Late Paleozoic (Feng et al., 1997). The WestPingdingshan section was selected for this study because it has receivedextensive studies of chemostratigraphy, magnetostratigraphy and bio-stratigraphy of conodonts, ammonoids and bivalves (Tong et al., 2002,2007; Zhao et al., 2007; Sun et al., 2009). The results frombiostratigraphicstudies indicate that this section was entirely deposited in a marineenvironment.
The West Pingdingshan section contains four sedimentary forma-tions, which are named as Dalong, Yinkeng, Helongshan and Nanlinghuin ascending order (Fig. 2). The Dalong Formation is characterized by si-liceous shale, with a depositional age of the latest Permian (Fig. 2). Thekey index fossil for the lowermost Triassic,Hindeodus parvus, has not yetbeen found in the West Pingdingshan section. However, the PTB in thissection can still be defined based on the “boundary sequence set” (Penget al., 2001; Sun et al., 2009). The contact between theDalong Formationand the overlying Yinkeng Formation consists of “boundary clay beds”,“boundary limestone” andmudstone in ascending order. The “boundaryclay beds” in the uppermost part of the Dalong Formation can be corre-lated with beds 25 and 26 of the Meishan section, whereas the 25-cm-thick “boundary limestone” in the lowermost part of the YinkengFormation accordswith bed 27 of theMeishan section. The Yinkeng For-mation is dominated by alternations of thin- tomedium-bedded argilla-ceous limestone and mudstone in the lower part and mudstoneintercalated with thin-bedded calcareous mudstone in the upper part(Fig. 2). The overlaying Helongshan Formation is covered by a landslipin the lower part, and consists of alternations of thin-bedded argilla-ceous limestone and mudstone in the upper part (Fig. 2). In the top ofthe section, the Nanlinghu Formation is represented by thick-beddedlimestone, intercalated by thin- to medium-bedded limestone (Fig. 2).The Griesbachian–Dienerian boundary and the Dienerian–Smithianboundary are within the Yinkeng Formation and are defined by the am-monoid Prionolobus and the conodont Neospathodus waageni, respec-tively (Tong et al., 2002; Zhao et al., 2007; Sun et al., 2009), whereasthe Smithian–Spathian boundary situated in the Helongshan Formationis defined by the conodontNeospathodus triangularis (Tong et al., 2002).
A total of 92 samples were collected along the West Pingdingshansection in an interval of about 1 m. It started from a GPS position of31°38'03''N and 117°49'41''E from the Dalong Formation upwards. Inorder to fulfill the analyses of carbonate C-O isotopes and detrital ele-ment concentrations simultaneously, argillaceous limestone and calcar-eous mudstone were collected preferentially. For the intervalscontaining no calcareous component, mudstonewas collected. Samples11CH133 and 11CH134 came from levels below the “boundarysequence set”, followed by samples 11CH135 and 11CH136 collectedfrom the “boundary clay beds” and the “boundary limestone”,respectively.
3. Methods
All the limestone and argillaceous limestone samples were exam-ined under amicroscope to ensure that they are free of visible recrystal-lization or metamorphism. The limestone, argillaceous limestone,calcareous mudstone and mudstone samples were crushed into smallchips, and the fresh chips containing no veins were picked out andground into powder for geochemical analyses.
Fig. 1. (A) Paleoenvironmental reconstruction of South China during the Induan stage (modified after Feng et al., 1997). (B) Geological map of the Pingdingshan area at Chaohu in theLower Yangtze basin of South China, showing the location of the West Pingdingshan section.
113M.-Y. Zhao, Y.-F. Zheng / Chemical Geology 391 (2015) 111–122
3.1. Carbon and oxygen isotope compositions of carbonate
Other than the mudstone devoid of carbonate components, all thesamples were analyzed for carbon and oxygen isotopes. To determinethe C and O isotope compositions of carbonate, samples of 80–800 μgwere reacted with 103% orthophosphoric acid at 72 °C for about 2 husing a GasBench II device attached to a Thermo Fisher Scientific MAT253 mass spectrometer. The detail analytical procedures can be foundin Zha et al. (2010). The carbon and oxygen isotope compositions areexpressed by the conventional δ13C and δ18O notations in permil (‰)relative to Vienna PeeDee Belemnite (VPDB). The working standardmaterial in these analyses was NBS-19, with δ13C =1.95‰ andδ18O=−2.20‰. Reproducibility of the δ13C and δ18O analyses is betterthan ±0.1‰ and ±0.2‰, respectively.
3.2. Major and trace elements in detritus
To determine the major and trace element contents of carbonate-free residues, 2–50 g samples were decarbonated by 2 N hydrochloricacid for 12 h. In the present study, terms “detritus” and “detrital sedi-ment” represent the insoluble residues after leaching by HCl solution.The carbonate-free residues were first neutralized by a stepwise centri-fugation process and then dried at 75–85 °C and eventually weighed.Such a chemical process can remove most of the chemical precipitatesin the sedimentary rocks, such as carbonates, Fe-Mn oxyhydroxidesand absorbedmaterials except organic materials, sulfur and biogenic si-licious materials (Wei et al., 2003, 2006). This treatment can be wellused to gain information about the composition of terrigenous detritus.About 0.25 g carbonate-free residues after chemical treatment were
Fig. 2.Profiles for high-resolution records ofmarine carbonate δ13C values (relative to VPDB) andTh, Sc, Na andCa concentrations indetrital sediments fromtheWest Pingdingshan sectionin the Lower Yangtze basin, South China. PTB denotes the Permian–Triassic boundary. Zonation of ammonoids comes from Zhao et al. (2007).
completely digested by the HNO3-HClO4-HF-HCl mixture. The resultantsolutions were evaporated to dryness and re-dissolved in HCl, spikedand diluted for ICP-AES and ICP-MS analyses, respectively. The ICP-MSinstrument used in these analyseswas Perkin Elmer Elan 9000. Two cer-tified materials were used in the present analyses: (1) basalt BHVO-2,which was provided by the United States Geological Survey (USGS),and its trace element composition was published by Raczek et al.(2001); (2) limestone GBW07120, which was provided by the Instituteof Geophysical and Geochemical Exploration (IGGE) at the ChineseAcademy of Geological Sciences, and its trace element compositionwas published by Wang et al. (2001). The analytical results for thetwo certified materials are listed in supplementary Table S1. Throughmonitoring by the standards, the external precision (1σ) of the presentanalyses was guaranteed to be better than ±10%.
3.3. Organic carbon and sulfur in detritus
In order to analyze the contents of organic carbon and sulfur in detri-tal sediments, samples were pretreated following the same proceduresas the element analyses. The organic carbon content of detritus wasmeasured on an Elementar Vario EL cube with a combustion tube anda reduction tube at 950 °C and 550 °C, respectively. Reproducibilityand accuracy for all the analyses of organic carbon content were en-sured to be better than ±0.3%. The sulfur content of detritus was mea-sured by Elementar Vario EL III with a combustion tube and a reductiontube at 1150 °C and 850 °C, respectively. Reproducibility and accuracyfor all the analyses of sulfur content are better than ±0.5%.
4. Results
4.1. Carbon isotope chemostratigraphy
The carbonate δ13C values of the study section vary from −6.5 to5.4‰ (Fig. 2 and Table S2). The δ13C curve exhibits a general trend
that is similar to the previous result from the same profile (Tong et al.,2007). The δ13C value is as low as−5.9‰ at the “boundary limestone”,corresponding to the upper part of the negative C-isotope shift spanningbeds 25–27 in theMeishan section (Cao et al., 2002; Xie et al., 2007) andin parallel to the end-Permian excursion that was extensively reportedfrom global successions (e.g., Grasby and Beauchamp, 2008; Korte andKozur, 2010). There is a positive δ13C excursion from −5.9‰ to−4.4‰ above the “boundary limestone”, followed by a decrease to−6.5‰ at 3.5 m above the PTB (Fig. 2 and Table S2).
After the two negative shifts near the PTB, δ13C values increase to−1.8‰ at the middle of Griesbachian sequence. Although a negativeδ13C shift is found in the middle of Griesbachian in the Qian-Gui-Xiangbasin (Payne et al., 2004; Meyer et al., 2011) and the Arabian platform(Clarkson et al., 2012), only a small negative shift from −1.8 to−2.6‰ occurs in theWest Pingdingshan section, followed by a carbon-ate gap (Fig. 2 and Table S2). The δ13C curve reveals a gradually increas-ing trend throughout the Dienerian (Fig. 2 and Table S2), consistentwith the previous results from the same site (Tong et al., 2007). A re-bound peak of δ13C appears near the Dienerian–Smithian boundary,followed by a gradually negative shift from 0.9‰ to −4.7‰ in theearly Smithian (Fig. 2 and Table S2). At the Smithian–Spathian bound-ary, the δ13C values recover from −2.7‰ to 5.4‰, followed by a highδ13C plateau in the early Spathian (Fig. 2 and Table S2). The majornegative and positive δ13C excursions in the early and late Smithianare both in accordance with the contemporaneous sequences in SouthChina (Payne et al., 2004; Tong et al., 2007; Meyer et al., 2011).
4.2. Major and trace elements in detritus
Thorium and Sc concentrations in the detrital sediments are in theranges of 3.8–12.6 ppm and 7.0–15.5 ppm, respectively (Fig. 2 andTable S3). The variations of Th and Sc are tightly coupled (with a corre-lation coefficient r = 0.77). They decrease from the Late Permian to the“boundary clay beds” interval, and then remain low till the first increase
115M.-Y. Zhao, Y.-F. Zheng / Chemical Geology 391 (2015) 111–122
in the middle of Griesbachian (Fig. 2 and Table S3). After the decreasesin the Griesbachian–Dienerian boundary, a clearly reversed variationpattern occurs between these two elements and δ13C (Fig. 2). The vari-ations of Ca and Na are also tightly coupled except in the early Spathian(Fig. 2 and Table S3). They also have high concentrations in the LatePermian, followed by a decrease in the “boundary clay beds” and thenan increase in the “boundary limestone” (Fig. 2 and Table S3). Thevariation patterns of Ca and Na in the Early Triassic generally followthe δ13C but reverse to Th and Sc. In the early Spathian, the variationsof Ca and Na are decoupled, i.e., the Ca contents gradually increase butthe Na contents gradually decrease.
The contents of Ti, Nb, Ta, Zr and Hf exhibit parallel variation trendsin the profile (Fig. 3 and Table S3). They remain relatively stable till themiddle Smithian, followed by a dramatical increase as much as 100% inthe late Smithian (Table S3 and Fig. 3).
In order to find new geochemical proxies for the intensity ofchemical weathering, we have managed to select six logarithmicparameters log(Th/Ca), log(Sc/Ca), log(Al/Ca), log[Th/(Na/5 + Ca)],log[Sc/(Na/5 + Ca)] and log[Al/(Na/5 + Ca)] from numerical combina-tions. Their values for the Early Triassic sediments are presented in Fig. 4and Table S3, which are generally higher than the corresponding valuesfor upper continental crust (Rudnick and Gao, 2003) and post-Archeanshales (Table S4 and Fig. 4). Their variation patterns are similar to eachother in the entire profile, with increases in the “boundary clay beds”,the middle Griesbachian and the early Smithian (Fig. 4 and Table S3),matching the negative δ13C shifts at these times. The log(Th/Ca),log(Sc/Ca) and log(Al/Ca) values for the late Griesbachian and theSmithian approach those for extremely weathered products of granodi-orite (Fig. 4, and Tables S3 and S4). Further results on theother elementsin detritus will be discussed in the next section.
4.3. Organic carbon and sulfur contents of detritus
The organic carbon contents of detritus vary between 0.2% and 2.9%withmost values lower than 1%,whereas its sulfur contents are relative-ly higher with a range from 0.0 to 6.6% (Fig. 6B). Regarding the C-S
Fig. 3. Profiles of Ti, Nb, Ta, Zr and Hf concentrations in detrital sediments from theWest Pingddenotes the Permian–Triassic boundary.
relationship, the two samples of the latest Permian are plotted intothe region of normal seawater. In contrast, many of the Early Triassicsamples are in the region of euxinic seawater, generally above themean correlation line for normal marine sediments.
5. The classification of elements in marine detrital sediments
Chemical weathering is composed of complex interactions betweenthe lithosphere, the atmosphere, and the hydrosphere that occur in thebiosphere and that are powered by solar energy. It can be described asthe processes of dissolution, hydration, hydrolysis, oxidation, reduction,and carbonatization. All of these processes are based on the rules ofphysicochemistry, and they lead not only to the destruction of parentminerals and to thepassing of elements from theminerals into solutionsand suspensions, but also to the formation of mineral and chemicalcomponents. While some of these components are soluble in surfacewater, the other components are insoluble and thus only physicallytransported by terrestrial water from provenances to seawater. This isattributable to a series of differences in the solubility of various ele-ments in surface water and thus to the mobility of elements duringchemicalweathering. Based on the variations of element concentrationsand the correlations between elements in themarine detrital sedimentsof Early Triassic from the Lower Yangtze basin, this study classifiesthe elements in the marine detrital sediments into four groups,i.e., immobile elements, mobile elements, easily sorted elements andredox sensitive elements.
5.1. Immobile elements
Immobile elements are conservative during chemical weatheringgiven that they are concentrated in refractory minerals or stronglyabsorbed by clays. The concentrations of these elements in the weath-ered products and corresponding sediments increase after intensivechemical weathering as a result of dissolution and transport of mobileelements. Scandium is one of the most immobile elements duringchemical weathering (Middelburg et al., 1988; Condie et al., 1995;
ingshan section. Notice that substantial increases occur in the latest Smithian interval. PTB
Fig. 4. Profiles of geochemical parameters log(Th/Ca), log(Sc/Ca), log(Al/Ca), log[Th/(Na/5+Ca)], log[Sc/(Na/5+Ca)], log[Al/(Na/5+Ca)], Th/Sc, CIA and CIW for detrital sediments fromtheWest Pingdingshan section. The dotted lines denote the values for post-Archean shales, which are calculated from the data of Taylor andMcLennan (1985). PTB denotes the Permian–Triassic boundary.
Nesbitt andMarkovics, 1997), with a very low solubility in surface water(Taylor andMcLennan, 1985). Therefore, the correlations between Sc andother elements can be used to identify the elements with the immobileproperty. Elements such as Al, Th, Be, In, K, Rb, Cs and Ga exhibit consid-erable correlations with Sc (Table S5), and their variation trends in thestudy profile are similar to each other (Fig. 2 and Table S3), indicatingtheir relatively immobile properties during chemical weathering, consis-tent with those documented from weathering profiles.
Aluminum, Be and Ga are conservative during chemical weatheringof felsic to mafic rock (Chesworth et al., 1981; Middelburg et al., 1988;Nesbitt and Wilson, 1992; Nesbitt and Markovics, 1997). However, un-like Th and Sc, no organized variation trend is found for Al in the study
Fig. 5. Cross plots and liner fitness of Nb-Ti, Ta-Ti, Zr-Ti and Hf-Ti in all
profile, which may be caused by the disproportional influence fromchemical precipitates given that the enrichment of non-terrigenous Alhas been documented in marine sediments (Murray and Leinen, 1996;Kryc et al., 2003).
Thorium is also very insoluble in aqueous solutions and hasbeen regarded as the most immobile element during chemicalweathering (Middelburg et al., 1988; Braun et al., 1993; Ma et al.,2007; Zhao and Zheng, 2014). It was hypothesized that Th could bechemically leached from the surface layer and enriched in the deeppart of weathering profile (Nesbitt and Markovics, 1997; Kurtz et al.,2000). Because Th has very low solubility in natural water, however, itis not able to be dissolved and transported by surface water (Kaufman,
detrital sediments from the West Pingdingshan section at Chaohu.
117M.-Y. Zhao, Y.-F. Zheng / Chemical Geology 391 (2015) 111–122
1969; Langmuir and Herman, 1980). In this regard, the so-called Thmobility could be realized by the transport of tinny Th-rich minerals bysurface water within the weathering profile. As such, this kind of Thmobilization has no bearing on its mobility during chemical weathering.
During chemical weathering, Rb and Cs are retained in weatheredproducts due to their large ionic radii, whereasK can be removedby sur-face water under certain conditions (Nesbitt et al., 1980). However, theconsiderable correlation between K and Sc in the study sedimentsindicates that either K has not been significantly removed fromthe weathered products or the sediments have received extensiveK-replacement (Nesbitt and Young, 1989; Fedo et al., 1996). Althoughno study of In mobility during chemical weathering has beenperformed, the correlation between In and Sc suggests an immobilebehavior of In during chemical weathering.
5.2. Mobile elements
Calcium and Na aremainly hosted by plagioclase in magmatic rocks.They are highly mobile during chemical weathering of either basalt orgranite (Nesbitt et al., 1980; Chesworth et al., 1981; Middelburg et al.,1988; Babechuk et al., 2014). Carbonates in the study samples were re-moved by digestion, thus Ca in the detrital sediments is mainly hostedby silicates. Except for the decoupling in the early Spathian (Fig. 2)that may result from the transformation of plagioclase types in sourcerocks, a regular covariation between Ca and Na matches their coupledgeochemical behavior during chemical weathering. There are oppositevariation trends between Ca-Na and Th-Sc in the Early Triassic sedi-ments (Fig. 2 and Table S3). This may indicate the mass conservationduring chemical weathering in the source region, i.e., the increases ofimmobile elements would be balanced by the decreases of mobileelements. The opposite variation trends suggest that, like Th and Sc,Ca and Na also have relatively simple behaviors during chemicalweathering and had sufferedminor influences from the other processesprior to the formation of detrital sediments. Therefore, a combinedstudy of these two groups of elements may provide important informa-tion on the intensity of chemical weathering.
Some divalent elements such as Mg, Sr and Ba are mobile duringchemical weathering (Nesbitt et al., 1980; Chesworth et al., 1981;Middelburg et al., 1988). However, their mobility is not revealed bytheir correlations and variation trends in the presently studied detritalsediments. This may be ascribed to their secondary enrichment in chem-ical precipitates (Nesbitt et al., 1980; Nesbitt and Young, 1989; Fedo et al.,1996).
5.3. Easily sorted elements
Tetravalent Zr andHf have similar ionic radii (0.84 Å and 0.83 Å) andthus similar geochemical behaviors, primarily hosted by zircon. Like-wise, Ti4+, Nb5+ and Ta5+ also have similar ionic radii (0.605, 0.64and 0.68 Å) and similar geochemical behaviors, mainly residing in rutileand other Ti-bearing minerals. These high field strength elements(HFSE) are immobile during chemical weathering (Middelburg et al.,1988; Nesbitt and Markovics, 1997) as a consequence of the refractori-ness of zircon and rutile. As zircon and rutile are both heavyminerals insedimentary provenance, it is imaginable that they may suffer from hy-draulic sorting (the grain size effect) during transport by surface water.A study of river suspended sediments does indicate that Ti, Nb, Ta, Zrand Hf are all “well-sorted” elements (Bouchez et al., 2011).
The Ti, Nb, Ta, Zr and Hf contents of detrital sediments are highlycorrelated (Table S5 and Fig. 5), in accordance with their similargeochemical behaviors during chemical weathering. Hence, these ele-ments can be used to detect the sorting effect of terrigenous detritus.The concentrations of these elements are relatively stable in the entireprofile (Fig. 3 and Table S3), but substantial increases occur in the latestSmithian samples.
5.4. Redox sensitive elements
The oxidation state and solubility of redox sensitive elements aresensitive to redox status in sedimentary environments. These elementslargely reside in silicates if the sedimentary environment is oxic.However, they can largely be fixed by sulfides in sulfidic sediments orby organic materials in suboxic or anoxic but non-sulfidic sediments(Algeo and Maynard, 2004; Tribovillard et al., 2006). Thus, the correla-tions between organic carbon and sulfur contents and element concen-trations can be utilized to examine if the elements are hosted by organicmaterials or sulfides.
As presented in Table S5, elements such as Co, Cu, Fe, Mn and Nishow considerable correlations with sulfur rather than organic carbon,indicating that these elements were either precipitated as independentsulfide minerals or trapped in Fe-sulfides. A very tight correlation isfound between Fe and S in the detritus of Early Triassic (Table S5 andFig. 6A). Itsmolar Fe/S ratios are close to 2, suggesting that Femainly ex-ists in the form of pyrite. The fitting line in Fig. 6A has an intercept of0.88%, much less than the mean concentration of silicate Fe in the sedi-ments of continental margin (Raiswell and Canfield, 1998; Poulton andRaiswell, 2002). This indicates that Fe rarely exists in silicates as a resultof severe chemical weathering. Euxinic sediments usually have lowerC/S than normal marine sediments and thus the redox status of marinebasin can be detected by the C-S relationship (Berner and Raiswell,1983). Most of the samples lie near or above the mean correlation linefor normal marine sediments, which has a slope of about 2.8 (Fig. 6B).This suggests that euxinic environments occurred in the Lower Yangtzebasin during the Early Triassic.
6. Geochemical proxies for the intensity of chemical weathering
During chemical weathering of the continental crust, water-solubleelements are chemically dissolved in terrestrial water and transportedinto seawater (Zhao and Zheng, 2014), whereas water-insolubleelements are physically transported by the terrestrial water, oceaniccurrent orwind into the seawater. The difference in the solubility of var-ious elements in surface water is correlated with the difference in themobility of these elements during the chemical weathering. Thus,chemical and physical sediments in marginal basins not only recordthe intensity of chemical weathering in continental margins, but alsotrace the source region of terrigenous materials.
Nevertheless, the chemical compositions of detrital sediments maybe influenced by chemical precipitates such as biogenic silicious mate-rials, organic materials and pyrite. Due to the extinction of radiolariansand siliceous sponges during the Permian–Triassic transition, the EarlyTriassic is known as an interval with a “chert gap” (Knoll et al., 2007;Chen and Benton, 2012). Therefore, the contents of biogenic siliciousmaterials in the sediments of Early Triassic should beminor. The organiccarbon and sulfur contents in the residues are also low (Table S3 andFig. 6), and many of the elements, including Al, Th, Sc, Ca, Na, Ti, Zrand Nb, have very low concentrations in biogenic silicious materials, or-ganic materials and pyrite. Except for some redox sensitive elements,the correlations between element concentrations and the contents oforganic carbon and sulfur also indicate that most of elements are nothosted by organic materials and pyrite (Table S5). Thus, the concentra-tions of these elements are nearly identical to the corresponding valuesof terrigenous detritus. Moreover, it was found that most of theelements that are not concentrated in chemical precipitates displaythe nearly same variation pattern in bulk sediments and residual detri-tus after treatment by hydrochloric acid (Wei et al., 2003, 2006).
From the analyses on the relationships and variation trends ofelements in Section 5, it is found that, other than the redox sensitive ele-ments that are trapped in sulfide minerals, the geochemical compositionof detrital sediments is mainly controlled by the property of source rocksand the intensities of chemical weathering and hydraulic sorting. There-fore, the geochemical composition of detrital sediments can be used to
Fig. 6. (A) Cross plot of total Fe and S contents in detrital sediments from the West Pingdingshan section. The data of the latest Permian sequence are not included in the linear match.(B) Cross plot of sulfur and organic carbon in detrital sediments from the West Pingdingshan section. The yellow shading denotes the region of normal marine sediments (after Bernerand Raiswell, 1983). The line denotes the mean correlation line for normal marine sediments, with a slope of 2.8.
Fig. 7. Plot of La-Th-Sc in detrital sediments from the West Pingdingshan section. Theregion representing post-Archean shales is taken from Taylor and McLennan (1985),and the regions representing detrital sediments derived from active and passive continen-tal margins, including continental and oceanic arcs, are taken from Cullers (1994).
decipher the intensity of chemical weathering if the influences from theother factors are evaluated properly.
Elements with variable mobilities exhibit different degrees of lossduring chemical weathering. Since the level of this chemical differenti-ation is primarily controlled by the weathering intensity, the ratiosbetween elements with different mobilities can be used to quantifythe change of chemical weathering. So far, a lot of element ratios havebeen used to trace the intensity of chemical weathering, including CIA,Chemical Index of Weathering (CIW), Al/K, Mg/Al, Rb/Al, Rb/Sr, Ca/Ti,K/Ti, Na/Ti, Sc/Ti and Al/Ti (Nesbitt and Young, 1982; Harnois, 1988;Zabel et al., 2001; Li et al., 2003; Wei et al., 2003, 2006; Clift et al.,2008; Sun et al., 2008; Tian et al., 2011). CIW is defined as CIW =molar [Al2O3/(Al2O3 + CaO* + Na2O)] × 100 (Harnois, 1988). Owingto the complexity of geological systems, the chemical composition ofdetrital sediments may be dictated by various processes and diverse in-terferences. Thus, a single index can hardly be applied to awide range ofgeological conditions and may draw an incorrect conclusion if the be-haviors of selected elements are not fully understood. The CIA may suf-fer influences from K-replacement and the enrichment of non-terrigenous Al. Likewise, the CIW can also be influenced by the enrich-ment of non-terrigenous Al, and its sensitivity decreases dramaticallyat high degrees of chemical weathering due to deceleration of Ca andNa losses (Nesbitt and Young, 1984). In addition, K, Mg, Rb and Ti areamenable to replacement (Nesbitt and Young, 1989; Fedo et al., 1996)or the addition of eolian dust (Yarincik et al., 2000; Martinez et al.,2007). As a consequence, the ratios containing these elements mayalso be influenced in certain cases. Therefore, for any given sediments,the basic geochemical behaviors and possible interferences of selectedelements should be fully understand in order to develop reasonableproxies. As argued in Section 5, immobile elements Th and Sc as wellasmobile elements Ca and Na can be best used to establish geochemicalproxies for chemical weathering since they have received minor inter-ferences. However, before talking about the proxy potential of theseelements, it needs to evaluate the possible influences from the sourcevariation and hydraulic sorting.
The very immobile elements suffer minor loss during chemicalweathering, and thus the ratios between them are sensitive to the com-position of sedimentary provenance rather than the intensity of chemi-cal weathering. Indeed, McLennan et al. (1993) found that Th/Sc ratiosvarymore than 100 fold in detrital sedimentswith different provenanceand hence maintained that it is a sensitive and simple index for theoverall source of terrigenous detritus. As presented in Fig. 4 andTable S3, Th/Sc ratios are nearly constant with no major variations inour studied profile, indicating minor changes in source rocks.
Light REE have high correlation coefficients with Th/Sc (Table S5).Although studies of weathering profiles indicate LREE immobilityduring chemical weathering (Nesbitt, 1979; Middelburg et al., 1988),no correlation is found between LREE and Sc in the present study. Thismay result from the large variations of LREE in source rocks. The sensi-tivity of LREE to source rocks indicates that the Th-Sc-La system is alsoinstrumental in identification of the change in the provenance. Howev-er, in the La-Th-Sc ternary diagram (Fig. 7), nearly all the samples fallwithin the range of active and passive continental margins, and thetwo groupswith different Th concentrations overlap. Thus, the variationof provenance is not the major cause for the regular variations of Th, Sc,Ca and Na.
Thorium is mainly hosted by fine minerals like clays in chemicallyweathered products, but Ca, Na, Ti and Zr are primarily concentratedin the coarse part of terrigenous detritus (Bouchez et al., 2011), so neg-ative correlations should be found between Th and Ti-Zr whereas posi-tive correlations should be found between Ca-Na and Ti-Zr if thehydraulic sorting is a dominant factor on their concentrations. However,
Fig. 8. Cross plots of Th-Zr and Ca-Zr in detrital sediments from theWest Pingdingshan section. The triangles denote thedatawith highZr values thatmay result from eolian contribution inthe latest Smithian, and the diamonds denote the data from the other intervals. The data with high Zr values in the latest Smithian are not included in the linear matches.
119M.-Y. Zhao, Y.-F. Zheng / Chemical Geology 391 (2015) 111–122
no such correlations are observed (Table S5 and Fig. 8). Rejection of theanalyses with very high Zr concentrations (N200 ppm) yields a weakpositive correlation between Th and Zr along with a weak negative cor-relation between Ca and Zr (Fig. 8), concordant with their behaviorsduring chemical weathering rather than hydraulic sorting. Thus, theeffect of hydraulic sorting is minor.
According to the above arguments, the changes of Th, Sc, Ca and Naare mainly controlled by chemical weathering. In order to erase the in-terference factors such as the dilution effect of chemical precipitates andto increase sensitivity during high degrees of chemical weathering, thelogarithms of element ratios such as log(Th/Ca), log(Sc/Ca), log[Th/(Na/5+ Ca)] and log[Sc/(Na/5+ Ca)] are utilized to quantify the inten-sity of chemical weathering (Fig. 4 and Table S3). The data of log(Al/Ca)and log[Al/(Na/5 + Ca)] are also shown for comparison (Fig. 4 andTable S3). Number 5 is used to calibrate the concentration of Na becausethe concentration of Na is roughly 5 times as high as Ca in the analyticalsamples. The six logarithmic parameters have generally higher valuesfor the Early Triassic sediments than those for upper continental crustand post-Archean shales (Table S4 and Fig. 4), indicating severe chemi-cal weathering in this period. This is in accordance with the warmpaleoclimate indicated by oxygen isotope results of conodont apatitein the Early Triassic (Sun et al., 2012). The six logarithmic parametersincrease across the PTB, indicating the enhancement of chemicalweathering that is coincident with the largest mass extinction in theEarth's history. This enhancement of chemicalweathering is also consis-tent with the geochemical result from paleosols (Sheldon, 2006).
The variation trends for the six logarithmic parameters are quite uni-form, with increases at the PTB, the middle Griesbachian and the earlySmithian, respectively, corresponding to the intervals with negativeδ13C shifts (Figs. 2 and 4). Although the negative δ13C shift during thelate Griesbachian is not prominent in the study section, it was foundin the other sections from South China (Payne et al., 2004; Meyeret al., 2011). These patterns manifest that more severe chemicalweathering occurred during the intervalswith negative δ13C excursions.This correlation can be attributed to significant input of the terrigenousmaterial from the continental crust to marginal seawater (Zhao andZheng, 2014).
7. Implications for paleoclimatic change
As revealed by laboratory and field studies, the intensity of chemicalweathering ismainly governed by temperature and precipitation on thecontinent (e.g. Bluth and Kump, 1994; Kump et al., 2000; Gislason et al.,2009). Since temperature and precipitation are also closely related, it isdifficult to separate their impacts on the chemical weathering of conti-nental crust (White and Blum, 1995; Kump et al., 2000; Gislason et al.,
2009). In practice, the simultaneity of paleoclimatic warming and en-hanced chemical weathering has also been documented from extensivestudies (e.g., Burton and Vance, 2000; Ravizza et al., 2001; Cohen et al.,2004; Oxburgh et al., 2007; Finlay et al., 2010; Theiling et al., 2012). Thepresent study can also link the intensity of chemical weathering topaleoclimatic changes.
The six logarithmic parameters log(Th/Ca), log(Sc/Ca), log(Al/Ca),log[Th/(Na/5 + Ca)], log[Sc/(Na/5 + Ca)] and log[Al/(Na/5 + Ca)] ex-hibit increased values across the PTB (Fig. 4). This demonstrates the en-hanced chemical weathering, a viewpoint also evidenced by increasedsedimentation rates and changed mineralogical compositions in thesame region (Algeo and Twitchett, 2010). Furthermore, more severechemical weathering in the Early Triassic was also revealed by chemicaland mineralogical changes of paleosols (Retallack, 1999; Retallack andKrull, 1999; Sheldon, 2006; Sheldon and Tabor, 2009). The intensity ofchemical weathering increases across the PTB (Fig. 4), indicating apaleoclimatic warming event. This is supported by the oxygen isotoperesults of conodont apatite that suggest a temperature increase ashigh as 8 °C (Joachimski et al., 2012).
The six logarithmic parameters exhibit dramatically increasedvalues in themiddle Griesbachian and the Smithian (Fig. 4). They are as-sociated with synchronous increases of Th and Sc but decreases of Naand Ca (Fig. 2). The log(Th/Ca), log(Sc/Ca) and log(Al/Ca) values forthese intervals approach the values for extremely weathered productsof granodiorite (Fig. 4, and Tables S3 and S4). Consistently, the oxygenisotope results of conodont apatite at the middle to late Griesbachianand the Smithian reveal a prohibitively high temperature as high as38 °C (Sun et al., 2012). High values for the six logarithmic parametersare also evident in the “boundary clay beds” (Fig. 4), which may be as-cribed to a short term over-hot event that was not revealed by the oxy-gen isotope results. The relatively low values for the six logarithmicparameters occur during the early Griesbachian, the Dienerian and theearly Spathian, in coincidence with the relatively low temperatures atthese time intervals (Sun et al., 2012).
Enhanced chemical weathering in the Early Triassic is consistentwith the negative δ13C shifts in the carbonate (Fig. 2). The negativeδ13C shifts could be caused by release of CH4 or CO2 due to massive vol-canism (e.g. Sobolev et al., 2011) or thermal metamorphism of organic-rich sediments (e.g., Svensen et al., 2009). Although volcanic eruptionnear the Lower Yangtze basin at that time is indicated by beds of volca-nic ash at the PTB in shallow water sections (e.g., Shen et al., 2011), nomassive volcanism has been found in this region. Instead, there wasthe progressive convergence between the South China Block and theNorth China Block in the late Paleozoic (Wu and Zheng, 2013), resultingin the continental collision in the Middle Triassic along the Dabie-Suluorogenic belt. This left a residual Paleotethyan sea in the Early Triassic.
Because of the subduction of the South China Block beneath the NorthChina Block during the Late Permian to the Early Triassic (Zheng et al.,2013), the Lower Yangtze basin would receive more and more terrige-nous detritus from chemical weathering of the elevated continental mar-gin. This is evidenced by elevated sedimentary rates (~30–90 m/m.y−1)in the study region during the Early Triassic (Algeo and Twitchett,2010). It is the continental convergence that has led to the warmpaleoclimate and thus enhanced chemical weathering of the continentalmargin above the subduction zone. This tectonism is temporally coupledwith warming of the paleoclimate, resulting in the increased input of ter-rigenous materials into the residual Paleotethyan seawater. As such, theenhanced weathering may be the basic mechanism for transport ofmore 13C-delepted organic carbon and dissolved inorganic carbon fromthe continental crust to the marginal sea, leading to the negative δ13Cshifts in the euxinic seawater. With regard to the origin of marine sedi-ments, it is thus important to distinguish thedeposition environments be-tween vast oceanicwater andmarginal seawater (Zheng, 2012).With theprogressive convergence between the two continental blocks during theLate Permian to Early Triassic, the Paleotethyan ocean became thePaleotethyan sea. This led to the increased influx of terrigenous materialsfrom chemical weathering of the uplifted continentalmargin of the NorthChina Block. As a consequence, the both transgression and retrogressionphases occur in the marine deposition of this period in the Middle toLower Yangtze regions.
The mass extinction at the PTB represents the largest biotic crisis inthe Phanerozoic record (Retallack, 1995; Erwin et al., 2002). Massivesoil erosion on the continents is indicated by a shift from fine-grainedmeandering to conglomeratic braided fluvial facies (Newell et al.,1999; Ward et al., 2000), by redeposition of soil clasts (Retallack,2005), and by increased sedimentation rates in terrestrial successions(Retallack, 1999). The signature of this erosional event has been recog-nized in marine successions in the form of soil-derived biomarkers(Sephton et al., 2005; Xie et al., 2007). The sedimentological study ofAlgeo and Twitchett (2010) indicates that sediment delivery wasfocused on shallow-marine shelves adjacent to landmasses, consistentwith derivation through subaerial weathering. The present study dem-onstrates a genetic link between tectonism and paleoclimate prior tothe final closure of the Paleotethyan sea between the two colliding con-tinental blocks in the Triassic. It is the continental convergence that ledto themassive transfer ofmineral and organicmatter from the terrestri-al to themarine realm in this transitional stage. The large increase in ter-rigenous sediment flux is prominent, which would have a profoundeffect onmarginalmarine ecosystems through higher nutrient availabil-ity, increased turbidity, and smothering of benthic organisms. This cou-pling between tectonism and paleoclimate may be a basic cause for themass extinction at the Permian–Triassic transition.
8. Conclusions
Since elements with similar geochemical behaviors tend to be corre-lated with each other in their abundances, the correlations betweenelements as well as the variations of element concentrations in asedimentary profile can be used to understand the general behaviorsand possible interference factors for each element in marine detritalsediments. Elements Al, Th, Sc, Be, In, Ga, Rb and Cs are immobile duringchemical weathering, whereas Ca and Na are mobile. These two groupsof elements can be selected to establish geochemical proxies for the in-tensity of chemical weathering if they received minor influences fromother factors. Titanium,Nb, Ta, Zr andHfwere hosted by heavymineralsin sedimentary provenance. Cobalt, Cu, Fe, Mn and Ni were mainlyhosted by sulfide minerals.
Thorium, Sc, Ca and Na are selected to establish new geochemicalproxies for the intensity of chemical weathering on account of their or-ganized variation trends and relatively simple geochemical behaviors.The concentrations of Th, Sc, Ca and Na were not affected by changesin the provenance of terrigenous detritus or the effect of hydraulic
sorting. Indices based on element ratios can suppress interferencessuch as dilution by chemical precipitates and thus increase the sensitiv-ity. Therefore, logarithmic parameters log(Th/Ca), log(Sc/Ca), log[Th/(Na/5+ Ca)] and log[Sc/(Na/5+ Ca)] are used to quantify the intensityof chemical weathering. These indices increase simultaneously at thePermian–Triassic transition, the middle Griesbachian and the earlySmithian, demonstrating that the intensity of chemical weathering in-creased at these time intervals. The proxy values for the middle to lateGriesbachian and the early Smithian get close to the values for the ex-tremely weathered products of granodiorite, pointing to enhancedchemical weathering and warm paleoclimate at these times. Further-more, the covariation between chemical weathering and paleoclimatecan be linked to the tectonism in this period.
Acknowledgments
This study was supported by the fund from the Natural ScienceFoundation of China (41221062). We thank Yan’an Shen, Jun Yan andYan-Yan Zhao for their assistance with field sampling. Thanks also goto Xiangping Zha for his assistance with stable isotope analyses. Weare grateful to two anonymous reviewers for their constructive com-ments that greatly helped in the improvement of the presentation. Dr.Michael E. Böttcher is thanked for his editorial handling.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.chemgeo.2014.11.004.
Algeo, T.J., Twitchett, R.J., 2010. Anomalous Early Triassic sediment fluxes due to elevatedweathering rates and their biological consequences. Geology 38, 1023–1026.
Babechuk, M.G.,Widdowson, M., Kamber, B.S., 2014. Quantifying chemical weathering in-tensity and trace element release from two contrasting basalt profiles, Deccan Traps,India. Chem. Geol. 363, 56–75.
Berner, R.A., Raiswell, R., 1983. Burial of organic carbon and pyrite sulfur in sedimentsover Phanerozoic time: a new theory. Geochim. Cosmochim. Acta 47, 855–862.
Berner, R.A., Lasaga, A.C., Garrels, R.M., 1983. The carbonate-silicate geochemical cycle andits effect on atmospheric carbon dioxide and climate. Am. J. Sci. 283, 641–683.
Bluth, G.J.S., Kump, L.R., 1994. Lithologic and climatologic controls of river chemistry.Geochim. Cosmochim. Acta 58, 2341–2359.
Bouchez, J., Gaillardet, J., France-Lanord, C., Maurice, L., Dutra-Maia, P., 2011. Grain sizecontrol of river suspended sediment geochemistry: clues from Amazon River depthprofiles. Geochem. Geophys. Geosyst. 12 (3), Q03008.
Braun, J.J., Pagel, M., Herbillon, A., Rosin, C., 1993. Mobilization and redistribution of REEsand Thorium in a syenitic lateritic profile: a mass-balance study. Geochim.Cosmochim. Acta 57, 4419–4434.
Burton, K.W., Vance, D., 2000. Glacial-interglacial variations in the neodymium isotopecomposition of seawater in the Bay of Bengal recorded by planktonic foraminifera.Earth Planet. Sci. Lett. 176, 425–441.
Cao, C.Q., Wang, W., Jin, Y.G., 2002. Carbon isotope excursions across the Permian-Triassicboundary in the Meishan section, Zhejiang Province, China. Chin. Sci. Bull. 47,1125–1129.
Chen, Z.Q., Benton, M.J., 2012. The timing and pattern of biotic recovery following theend-Permian mass extinction. Nat. Geosci. 5, 375–383.
Chesworth, W., Dejou, J., Larroque, P., 1981. The weathering of basalt and relative mobil-ities of the major elements at Belbex, France. Geochim. Cosmochim. Acta 45,1235–1243.
Clarkson, M.O., Richoz, S., Wood, R.A., Maurer, F., Krystyn, L., McGurty, D.J., Astratti, D.,2012. A new high-resolution δ13C record for the Early Triassic: insights from theArabian Platform. Gondwana Res. 24, 233–242.
Clift, P.D., Hodges, K.V., Heslop, D., Hannigan, R., Van Long, H., Calves, G., 2008. Correlationof Himalayan exhumation rates and Asian monsoon intensity. Nat. Geosci. 1,875–880.
Cohen, A.S., Coe, A.L., Harding, S.M., Schwark, L., 2004. Osmium isotope evidence for theregulation of atmospheric CO2 by continental weathering. Geology 32, 157–160.
Condie, K.C., Dengate, J., Cullers, R.L., 1995. Behavior of rare-earth elements in apaleoweathering profile on granodiorite in the Front Range, Colorado, USA. Geochim.Cosmochim. Acta 59, 279–294.
Cullers, R.L., 1994. The controls on the major and trace element variation of shales, silt-stones, and sandstones of Pennsylvanian-Permian age from uplifted continentalblocks in Colorado to platform sediment in Kansas, USA. Geochim. Cosmochim.Acta 58, 4955–4972.
Fedo, C.M., Eriksson, K.A., Krogstad, E.J., 1996. Geochemistry of shales from the Archean(~3.0 Ga) Buhwa Greenstone Belt, Zimbabwe: implications for provenance andsource-area weathering. Geochim. Cosmochim. Acta 60, 1751–1763.
Feng, Z.Z., Bao, Z.D., Li, S.W., 1997. Lithofacies Paleogeography of Middle and LowerTriassic of South China (in Chinese with English abstract). Petroleum IndustryPress, Beijing (198 pp.).
Finlay, A.J., Selby, D., Gröcke, D.R., 2010. Tracking the Hirnantian glaciation using Osisotopes. Earth Planet. Sci. Lett. 293, 339–348.
Gislason, S.R., Oelkers, E.H., Eiriksdottir, J., Torsannder, P., Oskarsson, N., 2009. Direct evi-dence of the feedback between climate and weathering. Earth Planet. Sci. Lett. 277,213–222.
Grasby, S.E., Beauchamp, B., 2008. Intrabasin variability of the carbon-isotope recordacross the Permian-Triassic transition, Sverdrup Basin, Arctic Canada. Chem. Geol.253, 141–150.
Harnois, L., 1988. The CIW index: a new chemical index of weathering. Sediment. Geol.55, 319–322.
Joachimski, M.M., Lai, X., Shen, S., Jiang, H., Luo, G., Chen, B., Chen, J., Sun, Y., 2012. Climatewarming in the latest Permian and the Permian-Triassic mass extinction. Geology 40,195–198.
Kaufman, A., 1969. The 232Th concentration of surface ocean water. Geochim. Cosmochim.Acta 33, 717–724.
Korte, C., Kozur, H.W., 2010. Carbon-isotope stratigraphy across the Permian-Triassicboundary: a review. J. Asian Earth Sci. 39, 215–235.
Kryc, K.A., Murray, R.W., Murray, D.W., 2003. Al-to-oxide and Ti-to-organic linkages inbiogenic sediment: relationships to paleo-export production and bulk Al/Ti. EarthPlanet. Sci. Lett. 211, 125–141.
Kump, L.R., Arthur, M.A., 1997. Global chemical erosion during the Cenozoic: weatherabil-ity balances the budgets. In: Ruddiman, W.F. (Ed.), Tectonic Uplift and ClimateChange. Plenum Press, New York, pp. 399–426.
Langmuir, D., Herman, J.S., 1980. Themobility of thorium in natural waters at low temper-atures. Geochim. Cosmochim. Acta 44, 1753–1766.
Li, X.H., Wei, G., Shao, L., Liu, Y., Liang, X., Jian, Z., Sun, M.,Wang, P., 2003. Geochemical andNd isotopic variations in sediments of the South China Sea: a response to Cenozoictectonism in SE Asia. Earth Planet. Sci. Lett. 211, 207–220.
Luo, G.M., Kump, L.R., Wang, Y.B., Tong, J.N., Arthur, M.A., Yang, H., Huang, J.H., Yin,H.F., Xie, S.C., 2010. Isotopic evidence for an anomalously low oceanic sulfateconcentration following end-Permian mass extinction. Earth Planet. Sci. Lett.300, 101–111.
Ma, J.L., Wei, G.J., Xu, Y.G., Long,W.G., Sun, W.D., 2007. Mobilization and re-distribution ofmajor and trace elements during extreme weathering of basalt in Hainan Island,South China. Geochim. Cosmochim. Acta 71, 3223–3237.
Martinez, N.C., Murray, R.W., Thunell, R.C., Peterson, L.C., Muller-Karger, F., Astor, Y.,Varela, R., 2007. Modern climate forcing of terrigenous deposition in the tropics(Cariaco Basin, Venezuela). Earth Planet. Sci. Lett. 264, 438–451.
Meyer, K.M., Yu, M., Jost, A.B., Kelley, B.M., Payne, J.L., 2011. δ13C evidence that high prima-ry productivity delayed recovery from end-Permian mass extinction. Earth Planet.Sci. Lett. 302, 378–384.
Middelburg, J.J., Van der Weijden, C.H., Woittiez, J.R.W., 1988. Chemical processes affect-ing the mobility of major, minor and trace elements during weathering of graniticrocks. Chem. Geol. 68, 253–273.
Molnar, P., 2004. Late Cenozoic increase in accumulation rates of terrestrial sediment:how might climate change have affected erosion rates? Annu. Rev. Earth Planet. Sci.32, 67–89.
Murray, R.W., Leinen, M., 1996. Scavenged excess aluminum and its relationship to bulktitanium in biogenic sediment from the central equatorial Pacific Ocean. Geochim.Cosmochim. Acta 60, 3869–3878.
Nesbitt, H.W., 1979. Mobility and fractionation of rare earth elements during weatheringof a granodiorite. Nature 279, 206–210.
Nesbitt, H.W.,Markovics, G., 1997.Weathering of granodioritic crust, long-term storage ofelements in weathering profiles, and petrogenesis of siliciclastic sediments. Geochim.Cosmochim. Acta 61, 1653–1670.
Nesbitt, H.W., Wilson, R.E., 1992. Recent chemical weathering of basalts. Am. J. Sci. 292,740–777.
Nesbitt, H.W., Young, G.M., 1982. Early Proterozoic climate and plate motions inferredfrom major element chemistry of lutites. Nature 299, 715–717.
Nesbitt, H.W., Young, G.M., 1984. Prediction of some weathering trends of plutonic andvolcanic rocks based on thermodynamic and kinetic considerations. Geochim.Cosmochim. Acta 48, 1523–1534.
Nesbitt, H.W., Young, G.M., 1989. Formation and diagenesis of weathering profiles. J. Geol.97, 129–147.
Nesbitt, H.W., Markovics, G., Price, R.C., 1980. Chemical processes affecting alkalis andalkaline earths during continental weathering. Geochim. Cosmochim. Acta 44,1659–1666.
Newell, A.J., Tverdokhlebov, V.P., Benton, M.J., 1999. Interplay of tectonics and climate ona transverse fluvial system, Upper Permian, southern Uralian foreland basin, Russia.Sediment. Geol. 127, 11–29.
Oxburgh, R., Pierson-Wickmann, A.C., Reisberg, L., Hemming, S., 2007. Climate-correlatedvariations in seawater 187Os/188Os over the past 200,000 yr: evidence from theCariaco Basin, Venezuela. Earth Planet. Sci. Lett. 263, 246–258.
Payne, J.L., Lehrmann, D.J., Wei, J.Y., Orchard, M.J., Schrag, D.P., Knoll, A.H., 2004. Largeperturbations of the carbon cycle during recovery from the end-Permian extinction.Science 305, 506–509.
Peng, Y., Tong, J., Shi, G.R., Hansen, H.J., 2001. The Permian-Triassic boundary stratigraphicset: characteristics and correlation. Newsl. Stratigr. 39, 55–71.
Poulton, S.W., Raiswell, R., 2002. The low-temperature geochemical cycle of iron: fromcontinental fluxes to marine sediment deposition. Am. J. Sci. 302, 774–805.
Raczek, I., Stoll, B., Hofmann, A.W., Jochum, K.P., 2001. High-precision trace element datafor the USGS reference materials BCR-1, BCR-2, BHVO-1, BHVO-2, AGV-1, AGV-2,DTS-1, DTS-2, GSP-1 and GSP-2 by ID-TIMS and MIC-SSMS. Geostand. Newslett. 25,77–86.
Raiswell, R., Canfield, D.E., 1998. Sources of iron for pyrite formation inmarine sediments.Am. J. Sci. 298, 219–245.
Ravizza, G., Norris, R.N., Blusztajn, J., Aubry, M.P., 2001. An osmium isotope excursion as-sociated with the late Paleocene thermal maximum: evidence of intensified chemicalweathering. Paleoceanography 16, 155–163.
Retallack, G.J., 1995. Permian-Triassic extinction on land. Science 267, 77–80.Retallack, G.J., 1999. Postapocalyptic greenhouse paleoclimate revealed by earliest Triassic
paleosols in the Sydney Basin, Australia. Geol. Soc. Am. Bull. 111, 52–70.Retallack, G.J., 2005. Earliest Triassic claystone breccias and soil-erosion crisis. J. Sediment.
Res. 75, 679–695.Retallack, G.J., Krull, E.S., 1999. Landscape ecological shift at the Permian-Triassic bound-
ary in Antarctica. Aust. J. Earth Sci. 46, 785–812.Rieu, R., Allen, P.A., Plötze, M., Pettke, T., 2007. Climatic cycles during a Neoproterozoic
“snowball” glacial epoch. Geology 35, 299–302.Rudnick, R.L., Gao, S., 2003. Composition of the continental crust. Treatise Geochem. 3,
1–64.Scheffler, K., Hoernes, S., Schwark, L., 2003. Global changes during Carboniferous-Permian
glaciation of Gondwana: linking polar and equatorial climate evolution by geochem-ical proxies. Geology 31, 605–608.
Sephton, M.A., Looy, C.V., Brinkhuis, H., Wignall, P.B., de Leeuw, J.W., Visscher, H., 2005.Catastrophic soil erosion during the end-Permian biotic crisis. Geology 33, 941–944.
Sheldon, N.D., 2006. Abrupt chemical weathering increase across the Permian-Triassicboundary. Palaeogeogr. Palaeoclimatol. Palaeoecol. 231, 315–321.
Sheldon, N.D., Tabor, N.J., 2009. Quantitative paleoenvironmental and paleoclimatic re-construction using paleosols. Earth Sci. Rev. 95, 1–52.
Sun, Y., Wu, F., Clemens, S.C., Oppo, D.W., 2008. Processes controlling the geochemicalcomposition of the South China Sea sediments during the last climatic cycle. Chem.Geol. 257, 240–246.
Sun, Z., Hounslow, M.W., Pei, J., Zhao, L., Tong, J., Ogg, J.G., 2009. Magnetostratigraphy ofthe Lower Triassic beds from Chaohu (China) and its implications for the Induan-Olenekian stage boundary. Earth Planet. Sci. Lett. 279, 350–361.
Sun, Y., Joachimski, M.M., Wignall, P.B., Yan, C., Chen, Y., Jiang, H., Wang, L., Lai, X., 2012.Lethally hot temperatures during the Early Triassic greenhouse. Science 338,366–370.
Svensen, H., Planke, S., Polozov, A.G., Schmidbauer, N., Corfu, F., Podladchikov, Y.Y.,Jamtveit, B., 2009. Siberian gas venting and the end-Permian environmental crisis.Earth Planet. Sci. Lett. 277, 490–500.
Taylor, S.R., McLennan, S.M., 1985. The continental crust: its composition and evolution.Blackwell, Oxford (312 pp.).
Theiling, B.P., Elrick, M., Asmerom, Y., 2012. Increased continental weathering flux duringorbital-scale sea-level highstands: evidence from Nd and O isotope trends in MiddlePennsylvanian cyclic carbonates. Palaeogeogr. Palaeoclimatol. Palaeoecol. 342, 17–26.
Tian, J., Xie, X., Ma, W., Jin, H., Wang, P., 2011. X-ray fluorescence core scanning records ofchemical weathering and monsoon evolution over the past 5 Myr in the southernSouth China Sea. Paleoceanography 26, PA4202.
Tong, J., Qiu, H., Zhao, L., Zuo, J., 2002. Lower Triassic inorganic carbon isotope excursion inChaohu, Anhui Province, China. J. China Univ. Geosci. 13, 98–106 (in Chinese withEnglish abstract).
Tong, J., Zuo, J., Chen, Z.Q., 2007. Early Triassic carbon isotope excursions from SouthChina: proxies for devastation and restoration of marine ecosystems following theend-Permian mass extinction. Geol. J. 42, 371–389.
Tribovillard, N., Algeo, T.J., Lyons, T., Riboulleau, A., 2006. Trace metals as paleoredox andpaleoproductivity proxies: an update. Chem. Geol. 232, 12–32.
Wang, C.S., Gu, T.X., Chi, Q.H., Yan,W.D., Yan, M.C., 2001. New series of rock and sedimentgeochemical reference materials. Geostand. Newslett. 25, 145–152.
Ward, P.D., Montgomery, D.R., Smith, R., 2000. Altered river morphology in South Africarelated to the Permian-Triassic extinction. Science 289, 1740–1743.
Wei, G.J., Liu, Y., Li, X.H., Shao, L., Liang, X., 2003. Climatic impact on Al, K, Sc and Ti in ma-rine sediments: evidence from ODP Site 1144, South China Sea. Geochem. J. 37,593–602.
Wei, G.J., Li, X.H., Liu, Y., Shao, L., Liang, X.R., 2006. Geochemical record of chemicalweathering and monsoon climate change since the early Miocene in the SouthChina Sea. Paleoceanography 21, PA4214.
White, A.F., Blum, A.E., 1995. Effects of climate on chemical weathering in watersheds.Geochim. Cosmochim. Acta 59, 1729–1747.
Wu, Y.-B., Zheng, Y.-F., 2013. Tectonic evolution of a composite collision orogen: an over-view on the Qinling-Tongbai-Hong'an-Dabie-Sulu orogenic belt in central China.Gondwana Res. 23, 1402–1428.
Xie, S.C., Pancost, R.D., Huang, J.H., Wignall, P.B., Yu, J.X., Tang, X., Chen, L., Huang, X.Y., Lai,X.L., 2007. Changes in the global carbon cycle occurred as two episodes during thePermian-Triassic crisis. Geology 35, 1083–1086.
Yan, D., Chen, D., Wang, Q., Wang, J., 2010. Large-scale climatic fluctuations in the latestOrdovician on the Yangtze block, south China. Geology 38, 599–602.
Yarincik, K.M., Murray, R.W., Peterson, L.C., 2000. Climatically sensitive eolian andhemipelagic deposition in the Cariaco Basin, Venezuela, over the past 578,000years: Results from Al/Ti and K/Al. Paleoceanography 15, 210–228.
Zabel, M., Schneider, R.R., Wagner, T., Adegbie, A.T., de Vries, U., Kolonic, S., 2001. LateQuaternary climate changes in central Africa as inferred from terrigenous input tothe Niger Fan. Quat. Res. 56, 207–217.
Zha, X.-P., Zhao, Y.-Y., Zheng, Y.-F., 2010. An online method combining a Gasbench II withcontinuous flow isotope ratiomass spectrometry to determine the content and isoto-pic compositions of minor amounts of carbonate in silicate rocks. Rapid Commun.Mass Spectrom. 24, 2217–2226.
Zhao, M.-Y., Zheng, Y.-F., 2014. Marine carbonate records of terrigenous input intoPaleotethyan seawater: geochemical constraints from Carboniferous limestones.Geochim. Cosmochim. Acta 141, 508–531.
Zhao, L., Orchard, M.J., Tong, J., Sun, Z., Zuo, J., Zhang, S., Yun, A., 2007. Lower Triassic co-nodont sequence in Chaohu, Anhui Province, China and its global correlation.Palaeogeogr. Palaeoclimatol. Palaeoecol. 252, 24–38.
Zheng, Y.-F., 2012. Metamorphic chemical geodynamics in continental subduction zones.Chem. Geol. 328, 5–48.
Zheng, Y.-F., Xiao, W.-J., Zhao, G.-C., 2013. Introduction to tectonics of China. GondwanaRes. 23, 1189–1206.
