Geochemical interpretation of the Precambrian basement and overlying Cambrian sandstone on Bornholm, Denmark: Implications for the weathering history

    Geochemical interpretation of the Precambrian basement and overlying Cambrian sandstone on Bornholm, Denmark: Implications for the weathering history Lingli Zhou, Henrik Friis, Tian Yang, Arne Thorshøj Nielsen PII: DOI: Reference:

S0024-4937(17)30230-X doi:10.1016/j.lithos.2017.06.019 LITHOS 4351

To appear in:

LITHOS

Received date: Accepted date:

23 July 2015 22 June 2017

Please cite this article as: Zhou, Lingli, Friis, Henrik, Yang, Tian, Nielsen, Arne Thorshøj, Geochemical interpretation of the Precambrian basement and overlying Cambrian sandstone on Bornholm, Denmark: Implications for the weathering history, LITHOS (2017), doi:10.1016/j.lithos.2017.06.019

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Geochemical interpretation of the Precambrian basement and overlying Cambrian sandstone on Bornholm, Denmark: implications for the weathering history

Department of Geoscience, Aarhus University, Høegh-Guldbergs Gade 2, DK-8000 Aarhus C, Denmark

IP

a

T

Lingli Zhou a,b,*, Henrik Friis a, Tian Yang a,c, Arne Thorshøj Nielsend

b

SC R

Geology Department, Trinity College Dublin, Museum Building, Trinity College Dublin, Dublin 2, Ireland

c

School of Geoscience, China University of Petroleum, Qingdao Shandong Province 266580, China Department of Geosciences and Natural Ressource Management,University of Copenhagen, Øster

Voldgade 10, DK-1350 København K, Denmark

MA

* Corresponding author: Ling-li Zhou

NU

d

D

E-mail: [email protected]

AC

CE P

Phone: (+383) 0834824408

TE

Address: School of Natural Sciences, Department of Geology, Trinity College Dublin, Dublin 2, Ireland

1

ACCEPTED MANUSCRIPT Abstract: A geochemical study of the Precambrian basement granites from the Borggård borehole on Bornholm, Denmark, suggests that the granites were moderately weathered (Chemical Index of

T

Alteration-CIA=66–71) during subaerial exposure in a humid climate. Microcline is well preserved,

IP

whereas plagioclase was thoroughly altered to clay minerals (Plagioclase Index of

SC R

Alteration-PIA=93–99) which is likely due to its original Ca-rich composition. The primary Fe-Ti accessory minerals were oxidized to hematite and anatase. Evidence from REE distribution patterns and immobile element ratios, e.g. Zr/Hf and Nb/Ta, between the weathered basement granite from the

NU

Borggård borehole and regional granitoids on Bornholm, constrains the Svaneke Granite as the original basement lithology. A tau (τ) mass transport model (assuming immobile Ti) was applied to

MA

quantify the mass transfer during weathering of the basement granite. The results show a depletion of major elements in the following order: Na>Ca>Mg>Si; Al and Ti are immobile and stay constant; K

D

shows sample dependent enrichment or depletion; Fe is slightly enriched. The Cambrian sandstone

TE

overlying the basement in the Borggård borehole, assigned to the Gadeby Member of the Nexø Formation, is feldspathic litharenite-litharenite in composition. Provenance indicators including

CE P

(Gd/Yb)N, Zr/Hf and Nb/Ta ratios and petrological features indicate that source material was derived from both weathered and fresh basement granite of intermediate composition. The Gadeby Member equivalents in Germany, the basal lower Cambrian Adlergrund Konglomerat Member (AKM) in the

AC

offshore G-14 well north of Rügen, and the approximately coeval Lubmin Sandstein Formation (LSF) from the Loissin-1 borehole, mainland Germany, must have been sourced from a basement with compositions comparable to the intermediate group of the regional granitoids on Bornholm. The source materials for the AKM (CIA=71–72, PIA=94–96), the Gadeby Member in the Borggård borehole (CIA=52–69, PIA=56–99) and an outcrop in Nexø, eastern Bornholm (CIA=52–66, PIA=61–96), have endured similar degrees of weak to moderate weathering but lost most of the plagioclase. The LSF has a comparable weathering history (CIA=63–73), but the plagioclase is better preserved (PIA=65–78). K metasomatism occurred in the basement granite and sandstone in both the Borggård and the G14-1 boreholes, mainly through the conversion of aluminous clay minerals (e.g. kaolinite) to illite, with transformation of plagioclase to K-feldspar occurring locally. The significantly variation of weathering rates of plagioclase and K-feldspar in the basement granite and 2

ACCEPTED MANUSCRIPT the provenance of sandstone from the Borggård borehole are likely due to the different internal crystal structures, a Ca- rich plagioclase original composition of the plagioclase, and the occurrence

T

of weathering in a very humid climate.

AC

CE P

TE

D

MA

NU

SC R

IP

Key words: Weathering; provenance; mass transfer; paleoclimate; Cambrian; Precambrian.

3

ACCEPTED MANUSCRIPT 1. Introduction Weathering by physical and chemical alteration of rocks and minerals at or near the Earth’s surface forms detrital sediment and is also an important process for the development of soil profiles

T

(Birkeland, 1984). Weathering processes in the past were closely constrained by hydrosphere and

IP

atmosphere conditions as recorded by mineral transformations (Nesbitt and Young, 1982).

SC R

Constituents are released as dissolved ions stoichiometrically during mineral transformations, therefore making it possible to determine the complex mineralogical changes by comparing bulk compositions in weathering profiles to those of fresh unaltered rocks. This information can be used to

NU

assess the paleoclimate during Earth’s early evolution (Gay and Grandstaff, 1980; Schau and Henderson, 1983; Holland, 1984; Feakes and Retallack, 1988). In addition to the redox significance,

MA

erosion of weathering profiles and hydraulic sorting in fluvial systems can produce sands and muds, the compositions of which reflect the geochemistry and mineralogy of the source rocks (Nesbitt et al.,

D

1996). The proportion of felsic or mafic contributions to siliciclastic sediments can be discriminated

TE

using provenance sensitive indicators, such as ratios of immobile elements, e.g. La or Th to Co, Sc, or Cr, size of Eu anomaly, combined with the distribution patterns of Rare Earth Elements (REE).

CE P

Geochemical evidence thus has been broadly demonstrated to be useful for interpreting the provenance of siliciclastic sediments, and to unravel the mineralogical transformations that occurred during the weathering of the source rocks (Bhatia, 1983, 1984; Dickinson et al., 1983; Roser and

AC

Korsch, 1988; Condie, 1993; McLennan et al., 1993; Cullers, 1994, 2000; Nesbitt et al., 1996). The Island of Bornholm in the Baltic Sea hosts the only basement exposures in Denmark. It links the exposed crustal provinces in southern Sweden and the Precambrian rocks buried beneath the sedimentary cover in the southern Baltic Sea, Lithuania, northeast Poland and northernmost Germany (Figs. 1A, B, Bogdanova et al., 2006, 2008). The granitoids on Bornholm were formed during a short time interval between 1.47 and 1.44 Ga (Zariņš and Johansson 2009; Waight et al. 2012). Prior to deposition of the overlying Cambrian sandstone the basement rocks were extensively subaerially weathered (Gravesen et al., 2011). However, the paleo-weathering processes, especially for ancient weathering conditions during the Precambrian, are still poorly understood. The basement is overlain by a reddish fluviatile-aeolian sandstone, the Gadeby Member of the Nexø Formation, of assumed early Cambrian age (Nielsen and Schovsbo, 2007). A 12 m-thick red zone is observed to grade from 4

ACCEPTED MANUSCRIPT the top of the weathered basement granite beneath, in Borggård borehole on Bornholm, with petrological features supporting a genetic relationship between the underlying basement and the overlying sandstone. This profile provides a good opportunity for research on the weathering

T

processes and Precambrian climate conditions. Additional samples of the Gadeby Member from the

IP

abandoned Nye Frederiks sandstone quarry in Nexø on eastern Bornholm have also been included in

SC R

the study. The provenance of the Nexø Formation is still unclear. A local provenance was proposed based on a study of heavy minerals (Gry, 1936), but the presence of tourmaline indicates at least some supply from an outside source. An at least partial origin from southern Sweden was proposed by

NU

Jensen (1977) based on a study of the opaque mineral content.

In the present study, the geochemical composition of the lower part of the Nexø Formation and

MA

the underlying weathered basement granite in the Borggård borehole are examined and compared to unweathered granitoids on Bornholm (based on data published by Johansson et al., 2016). The main

D

aim is to clarify the types of weathering process that occurred during the Precambrian including an

2. Geological setting

TE

estimate of the contemporary paleoclimatic conditions.

CE P

Located within the Fennoscandian Border Zone (FBZ) at the southwestern margin of Fennoscandian Shield, Bornholm in Denmark is an uplifted block which that is tectonically important within the plate framework of the aforementioned East European Craton. Two major fault zones, the

AC

Sorgenfrei-Tornquist zone (STZ) and the Teisseyre-Tornquist zone (TTZ), separate the Baltic Shield from the Danish-Polish Basin. Bornholm is situated just east of the Rønne Graben, which offsets the STZ and TTZ (Fig. 1A). The STZ as a structural element is of Paleozoic origin that was reactivated during Triassic-Jurassic extension and Cretaceous-Paleogene compression (Liboriussen et al., 1987; Mogensen, 1994; Hansen et al., 2000; Mogensen and Korstgård, 2003). The old Precambrian TTZ, extending 2000 km from the Baltic Sea to the Black Sea, is generally regarded as an ancient plate boundary between the Precambrian platform in the NE and a belt of Caledonian deformation in the SW (Mazur et al., 2015). The Bornholm horst block is composed of Precambrian crystalline basement overlain by Paleozoic-Mesozoic cover rocks (Milthers, 1930; Callisen, 1934). The Precambrian basement was exposed in northern and central area, whereas the Paleozoic-Mesozoic sedimentary rocks are 5

ACCEPTED MANUSCRIPT distributed across the southern area (Fig. 1C). The Precambrian basement rocks on Bornholm comprise high-grade gneisses and a complex of granitic rocks (Waight et al., 2012) that are intruded by a series of younger dolerite dykes (Holm et al., 2010). At least five separate granitoid intrusions

T

have been identified from the mineralogical compositions, metamorphic imprint and degree of

IP

deformation: the Rønne, Vang and Svaneke granitoids, Almindingen and Hammer granites (Fig. 1C,

SC R

Callisen, 1934; Gravesen, 1996). A migmatitic rock occurs at the border between Svaneke granite and the gneisses, and was named the Paradisbakkerne migmatite (Fig. 1C, Callisen, 1934). Precise zircon U-Pb dating reveals a relatively restricted and contemporaneous period at around 1455±10 Ma

NU

for the granitic magmatism, deformation and metamorphism on Bornholm (Waight et al., 2012). The basement lithologies on Bornholm exhibit only minor variations in mineralogical composition among

MA

them (Table 1; see also Callisen, 1934; Johansson et al., 2016). The main constituents are K-feldspar, plagioclase, quartz, hornblende and biotite, along with minor Ti-magnetite, titanite, apatite, epidote,

D

fluorite and allanite (Micheelsen, 1961). An apparent gap in SiO2 content at around 70 wt. % divides

TE

the Bornholm granitoids into an intermediate group (Rønne, Vang and Svaneke granitoids) and a felsic group (Almindingen and Hammer granites, Johansson et al., 2016). The strong similarity in

CE P

geochemical signatures indicates the Bornholm granitoids belong to a single igneous suite composed of alkali-calcic biotite-hornblende quartz monzonites ranging into more evolved biotite granites (Johansson et al., 2016).

AC

Deposition of sediments was not initiated until the early Cambrian (Nielsen and Schovsbo, 2011). The basement is overlain by fluvial–aeolian reddish sandstone, assigned to the Gadeby Member of the Nexø Formation. The Gadeby Member is in turn overlain by quartzose shallow marine sandstones assigned to the Langeskanse Member of the Nexø Formation and the Hardeberga Formation (Nielsen and Schovsbno, 2007). The Lower Cambrian sandstones were rather deeply buried during the late Silurian-Devonian in a foreland basin associated with the Caledonian collision (Buchardt et al., 1997; Friis et al., 2010). A burial depth of 4–6 km (130-190°C at 30 °C/km) was estimated for the local tuffaceous sandstone on Bornholm from non-annealed zircon ages and annealed apatite cooling ages (Hansen, 1995), but it is likely that the geothermal gradient was significantly higher, reducing burial to some 3–4 km (e.g. Buchardt et al. 1997). Later, in the late Carboniferous–early Permian, Bornholm was subjected to uplift, and at least two stages of uplift 6

ACCEPTED MANUSCRIPT /erosion occurred: one to 3 km depth (100°C) prior to 261 Ma, followed by another uplift/erosion to the surface (arbitrarily set to 0°C) during the last c. 261 Myr. (Hansen, 1995). Subsequent erosion stripped off most of the Paleozoic sedimentary cover from the present-day island of Bornholm, and

IP

island (Vejbæk et al., 1994; Jensen and Nielsen, 1995, Fig. 1C).

T

down-faulted Cambrian to lower Silurian deposits are preserved only on the southern part of the

SC R

The sandstone samples for this study are from the Gadeby Member within the lower part of Nexø Formation in the Borggård borehole located in southeastern Bornholm (Fig. 1A). The Nexø Formation on Bornholm comprises the Gadeby and Langeskanse members (Nielsen and Schovsbo,

NU

2007), and it rests uncomformably on the weathered crystalline Precambrian basement. The Nexø Formation is widely distributed across southern Bornholm, but nearly all exposures are small

MA

abandoned quarries (Hansen, 1936). The formation is 92 m in the Borggård borehole and probably 100–110 m at the Nexø area (Fig. 2, Nielsen and Schovsbo, 2007), but individual exposures rarely

D

span more than a few meters (Hansen, 1936). Trace or body fossils are absent for precise

TE

biostratigraphical dating, but a gradational contact with the overlying Lower Cambrian Hardeberga Formation suggests that deposition of the Nexø Formation occurred during the earliest Cambrian

CE P

(Bruun-Petersen, 1977; Poulsen, 1978; Surlyk, 1980; Nielsen and Schovsbo, 2007, 2011). The Langeskanse Member in the upper part of the Nexø Formation is quartzose, better sorted, only partly red-striped and has a significantly lower clay-matrix content (Fig. 2, Bruun-Petersen, 1972). In

AC

contrast, the lower part of the Nexø Formation on Bornholm, the Gadeby Member consists of reddish, poorly sorted sandstone characterized by abundant feldspar and a high content of clay matrix (Fig. 2, rock types I and II of Hansen, 1936; Bruun-Petersen, 1972). Unweathered sandstone is pinkish, reddish-brown and violet and represent what is traditionally regarded as typical Nexø Formation. A 12 m-thick transitional succession with a gradual upwards decrease in red coloration was observed in the Borggård borehole (c.274–262 m) at the boundary to the overlying more quartzose Langeskanse Member (Fig. 2). A continental origin for the Nexø Formation has been proposed, based on evidences such as red coloration (hematite), and lack of trace and body fossils. The major part of the Nexø Formation was deposited in a fluvial environment (Bruun-Petersen, 1977; Surlyk, 1980), which at times was influenced by aeolian reworking and local deposition of aeolian sand sheets (Clemmensen and Dam, 1993). The incursion of quartsitic layers with glauconite near the top of the Gadeby 7

ACCEPTED MANUSCRIPT Member suggests a gradual change of deposition from a fluvial to a marine environment (Langeskanse Member). The overlying Hardeberga Formation comprises of dominantly well-sorted and strongly cemented orthoquartzites deposited in a shallow marine environment with incursions of

T

fine-grained shelfal mudstones (Fig. 2, Bruun-Petersen, 1972; Hamberg, 1991; Nielsen and

Sampling and analytical techniques

SC R

3.

IP

Schovsbo, 2011).

For this study, a total of 19 samples were collected from the lower part of the Borggård core (294.65–316.0 m), including 12 sandstone samples from the Gadeby Member of the Nexø Formation

NU

(294.65–310.5 m) and 7 basement granite samples (310.6–316.0 m). The fully cored Borggård borehole, which was drilled in 2006, is 316 m deep and penetrated from the Middle Cambrian to the

MA

basement (Fig. 2). The main purpose of this drilling was to determine the composition and thickness of the Lower Cambrian sediments on Bornholm. The sample locations within the Borggård borehole

D

are shown in Figure 2, and the geochemical compositions are listed in Appendix 1.

TE

The 19 samples were crushed and milled into powder <50 µm, and then analyzed by X-ray diffraction (XRD) for mineralogical composition at the Geoscience Department of Aarhus

CE P

University. Powder from the same batches was sent to the Acme Lab in Canada for major-element analysis by X-ray fluorescence (XRF), and for trace-element analysis by Inductively Coupled Plasma Mass Spectrometry (ICP-MS). To determine the loss-on-ignition (LOI), a predetermined amount of

AC

sample was roasted and then fused in a platinum-gold crucible with a commercial lithium tetraborate flux. For the major-element analysis, the molten materials were further cast in platinum molds, and the fused discs were analyzed by XRF to determine the contents of SiO2, Al2O3, Fe2O3, CaO, MgO, Na2O, K2O, MnO, TiO2, P2O5, Cr2O3, Ba, Cu, Ni, Pb, Sr, Zn, Zr and V2O5. To measure the trace element contents, prepared samples were mixed with LiBO2/Li2B4O7 flux in crucibles that were fused in a furnace. The cooled beads were dissolved in ACS grade nitric acid and analyzed by ICP-MS. Back-scattered electron (BSE) images of samples were acquired to distinguish the main mineral phases in polished thin sections by using a JEOL JSM-840 SEM equipped with an EDAX energy dispersive X-ray spectrometer (EDS) at Aarhus University. The standard guide for the performance of energy dispersive X-ray spectroscopy can be referred to in ASTM E 1508, "Quantitative Analysis by Energy-Dispersive Spectroscopy" (DOI: 10.1520/F1375-92R12). Total Carbon (TOT/C) and 8

ACCEPTED MANUSCRIPT Total Sulfur (TOT/S) contents of another set of 19 samples were determined by using a Leco CS230 induction furnace at the Geoscience Department of Aarhus University. For the analysis, induction flux was added to the prepared samples and then ignited in an induction furnace. A carrier gas swept

T

up released carbon to be measured by adsorption in an infrared spectrometric cell. The content of

IP

organic C was obtained by analyzing the insoluble residue after the carbonates were dissolved in

SC R

diluted HCl. The content of S in pyrite was calculated from the concentrations of Fe2O3 and TOT/S. Additional geochemical data from 1) the Mesoproterozoic granitoids on Bornholm (Johansson et al., 2016), 2) the continuation of Mesoproterozoic basement in the offshore G14-1 borehole located

NU

northeast of Rügen, Germany (Fig. 1A, Obst et al., 2004), 3) an outcrop of Gadeby Member in Nye Frederiks Quarry on eastern Bornholm (Fig. 1A), 4) the Adlergrund Konglomerat Member (AKM) in

MA

the G14-1 borehole, and 5) the Lubmin Sandstein Formation (LSF) from Loissin-1 borehole, Fig. 1A) in mainland Germany (Feldrappe et al., 2005), were compiled in this study for comprehensive

D

discussion. The AKM and LSF are German equivalents of the Nexø Formation on Bornholm,

TE

Denmark.

CE P

4. Results

4.1 Mineralogical compositions

The greyish-red basement granite from the Borggård borehole has pegmatite veins, 0.5 to 2 cm

AC

wide (Figs 3A, B). The granite is composed of K-feldspar, quartz, minor biotite and chlorite, with traces of hematite, anatase, pseudobrookite (Figs 3C, D, E and Figs 4A, B, Table 1), apatite, zircon, calcite and fluorite. The grain size varies from medium to coarse (Fig. 3A). Microcline is the dominant feldspar, and has narrow exsolution lamellae on the grain surfaces (Figs 3D, E). Plagioclase grains were mostly altered to clay minerals (Figs 3C, D, E). Biotite is the only mafic mineral distinguished (Fig. 3C). Ti-magnetite, titanite and ilmenite were mostly altered to Fe- and Ti- oxides, and only hematite and anatase were observed in the granite from the Borggård borehole (Fig. 3C and Figs 4A, B). The sandstone of the overlying Gadeby Member in the Borggård borehole is mainly composed of quartz, microcline and illite, minor biotite, and traces of zircon, apatite, Fe-and Ti- oxides (Figs 3E, F, G, H; Figs 4C, D, E). The textural and mineralogical compositions vary for sandstones of the Gadeby Member at different depths in the Borggård borehole. The lower part of Gadeby Member is 9

ACCEPTED MANUSCRIPT poorly sorted and contains angular to sub-angular clasts in the coarse grain fractions. Moving stratigraphically up-sequence the unit becomes progressively more mature and better lithified, being characterized by less detrital feldspar and clay matrix and with more detrital quartz and quartz cement

T

(Figs 3C, F). Plagioclase grains were mostly altered, with their morphology being preserved by illite,

IP

quartz or hematite fillings (Figs 3H, I). Clay minerals, as cements, were locally dissolved, and their

SC R

remnants were observed to be attached to the altered feldspar grains (Fig. 3H). Some quartz grains show pressure solution boundaries against detrital feldspar or biotite grains. Three successive phases of quartz cementation in the Nexø Formation were distinguished from their distinctive Scanning

NU

Electron Microscopy-Cathode-Luminescence (SEM-CL) patterns (Friis et al., 2010). In the sandstone samples examined in this study, quartz cement was mostly observed as fillings in the

MA

porosity left from dissolution of plagioclase and Fe-Ti minerals (Figs 3F, G). Ilmenite and martite grains that are common in the Nexø Formation (Jensen, 1977) were absent in the sandstones

D

examined in the present study. Instead, Fe- and Ti- oxides, e.g., hematite and anatase, were observed

TE

(Figs 3F, G; Figs 4C, D, E). Three types of occurrences of hematite are distinguished: dispersive needles in the clay matrix (HD, Fig. 3E); hematite rims that enclose the detrital feldspar grains (HR,

CE P

Fig. 3E), and fillings of the altered porous feldspar (HF, Fig. 3F). The sandstone samples from the Gadeby Member are of litharenitic composition when plotted in a Quartz-Feldspar-Lithics (QFL) ternary diagram, with minor feldspathic characteristics (Fig. 5). Two main types of lithic fragments

AC

were observed in the Gadeby Member that are easily confused with quartz and feldspar clastic grains after diagenesis. The first type is granite lithic fragments that are mainly composed of a high percentage of quartz and a low percentage of feldspar. The dissolution of the feldspar leaves granular quartz surrounded by pores or illite. The second type is lithic fragments composed of graphic granite, in which the quartz and feldspar are intimately intergrown (Fig. 3H). The optical orientations of quartz elements are identical between the lithic fragments and clastic. After the feldspar has been dissolved, the granite lithic fragments are easily confused with the clastic quartz grains. In addition, some graphic-granite lithic fragments are difficult to distinguish because of their high content of feldspar and low content of vermicular quartz. The lithic fragments of graphic granite are thus commonly overlooked, especially after dissolution of the feldspar. We recorded these grains as lithic fragments during point-counting work even though they represent grains of arkosic composition. The 10

ACCEPTED MANUSCRIPT graphic intergrowth of quartz and feldspar is not observed in the basement granite from the Borggård borehole, but it occurs locally in the Bornholm granitoids (Callisen, 1934) and also sporadically in clasts from sandstones in the lower part of the Gadeby Member exposed elsewhere on Bornholm.

T

4.2 Major element geochemistry

IP

The geochemical compositions of granitoids and sandstone samples included in this study are

SC R

listed in Appendix 1. The sandstones of the Gadeby Member from the Borggård borehole generally have lower SiO2 and Na2O concentrations than samples from the Nye Frederiks quarry, but higher concentrations of Al2O3, Fe2O3 and MgO (Appendix 1). Sandstone samples of the AKM from the

NU

G14-1 borehole have similar concentration of major elements to the Gadeby Member, being characterized with low contents of Na2O and high K2O (Appendix 1). In contrast, the sandstones of

MA

the LSF from the Loissin-1 borehole show high abundance of Na2O and depletion in K2O (Appendix 1).

D

The weathered basement granite from the Borggård borehole contains similar SiO2

TE

concentrations as the basement granite from the G14-1 borehole, but lower concentrations than the orthogneisses, migmatite and other granites on Bornholm (cf. Appendix 1). In contrast, the weathered

CE P

basement granite from the Borggård borehole has slightly higher concentrations of Al2O3, TiO2, FeOt and K2O than the basement granite from G14-1 borehole (Appendix 1). Regarding mobile constituents, such as Ca, Na and Mg, the weathered basement granite from the Borggård borehole

AC

shows significant depletion in comparison to the basement granite from the G14-1 borehole and the orthogneisses, migmatite and other granites on Bornholm (Appendix 1). For the regional Bornholm granitoids and the basement granite from G14-1 borehole, MgO is positively correlated to Al2O3 and CaO, but negatively correlated to K2O, indicating that the Mg is mainly hosted in hornblende, instead of biotite (Figs 6A, B, C). In contrast, MgO is negatively correlated to CaO and positively correlated to K2O in the basement granite and sandstone from the Borggård borehole, indicating that biotite is the primary host of Mg (Figs 6B, C). CaO is positively correlated with Al2O3, TOT/C, and P2O5 in the Bornholm granitoids and the basement granite from the G14-1 borehole (Figs 6D, E, F), suggesting the occurrence of Ca in silicates, carbonate and apatite. In contrast, the CaO concentration in the basement granite from the Borggård borehole is extremely low. All of the sandstone samples have low Ca concentrations. The high correlation of 11

ACCEPTED MANUSCRIPT CaO with TOT/C (Fig. 6E) and low correlation with Al2O3 (Fig. 6D) and P2O5 (Fig. 6F) suggest the presence of a small amount of Ca more likely in the carbonates, instead of silicates. Iron and Ti show a perfect linear correlation for the Bornholm granitoids, the basement granite samples from both the

T

Borggård and the G14-1 boreholes (Fig. 6G), suggesting that there is a genetic correlation between

IP

them. TiO2 and Al2O3 are positively correlated in the Bornholm granitoids and in the basement

SC R

granite from the G14-1 borehole (Fig. 6H), indicating that they are also the host of Ti and Fe in mafic minerals. In contrast, plots of TiO2 vs. Al2O3 for basement granite from the Borggård borehole are off the trend-line (Fig. 6H), indicating a primary contribution of Fe and Ti from Ti-magnetite, titanite

NU

and/or ilmenite, instead of mafic minerals. The sandstone plots scatter in the TiO2 vs. Al2O3 (Fig. 6H) and Fe2O3 vs. Al2O3 (Fig. 6I) diagrams, but mostly fall on the TiO2 vs. Fe2O3 trend-line (Fig. 6G),

MA

demonstrating a control of Fe and Ti from the Fe-Ti oxides. Several sandstone plots for the Gadeby Member in the Borggård borehole deviate from the TiO2 vs. Fe2O3 trend-line, which may be due to

D

disturbance by hydrothermal veins. The basement granite from the G14-1 borehole follows the

TE

fractionation trends of Na2O vs. SiO2 (Fig. 6J) and K2O vs. SiO2 (Fig. 6K) seen in the intermediate group of Bornholm granitoids (cf. Johansson et al., 2016). However, the plots of basement granite

CE P

from the Borggård borehole deviate from the trend-lines showing distinctively low Na2O and high K2O contents (Figs 6J, K). The sandstone plots are highly scattered in the diagrams showing Na2O vs. SiO2 (Fig. 6J) and K2O vs. SiO2 (Fig. 6K). Several sandstone samples from Borggård borehole have

AC

K2O contents comparable to the basement granite beneath, whereas the others show K2O concentrations similar to the Gadeby Member from the Nye Frederiks quarry (Fig. 6K). Sandstones of the LSF have lower K2O concentration than the other sandstone samples (Fig. 6K). The basement granite and sandstone from the Borggård borehole are characterized by significantly lower Na2O, but higher K2O contents than seen in the Bornholm granitoids, 4.3 Weathering indices The Chemical Index of Alteration (CIA), defined as CIA = 100×Al2O3/(Al2O3+CaO*+Na2O+K2O) (in molecular proportions), is used to evaluate the weathering history of rocks (Nesbitt and Young, 1982). CaO* in the formula represents the amount of CaO incorporated in the silicate fraction of the rock by excluding Ca incorporated in apatite and carbonate. Silicate-bound CaO is calculated using the equation (Fedo et al., 1995): 12

ACCEPTED MANUSCRIPT molCaO*(silicates) = molCaO-molCO2(calcite)-0.5molCO2(dolomite)-10/3 mol P2O5(apatite), assuming that all P2O5 is associated with apatite and all inorganic carbon is associated with carbonates. In practice, these assumptions lead frequently to overcorrection, particularly evident for

T

samples in which silicate-bound CaO becomes negative (Bahlburg and Dobrzinski, 2011). In this

IP

study, several sandstone and basement granite samples from the Borggård borehole have negative

SC R

CaO* values. If one considers the high correlation between CaO and TOT/C (Fig. 6E), and the low correlation between CaO vs. Al2O3 (Fig. 6D) and between CaO vs. P2O5 (Fig. 6F), then an extremely low amount of Ca was likely to be incorporated in silicate or apatite, but most of the Ca is hosted in

NU

carbonate. Therefore, the CaO* with negative values were taken as zero for the calculations of weathering indices, which may overestimate the real weathering intensity of the samples in this

MA

study.

The CIA values calculated accordingly vary in the range of 52–69 for the sandstone and 66–71

D

for the basement granite in the Borggård borehole (Appendix 1), demonstrating a weak to moderate

TE

weathering history. The CIA values for the unweathered basement on Bornholm vary between 47–53 (Appendix 1, calculation based on geochemical data in Johansson et al., 2016), suggesting a low

CE P

degree of alteration. The CIA values are 52–66 for sandstones of the Gadeby Member that are exposed at the Nye Frederiks Quarry at Bornholm, 71–72 for sandstones from the AKM in the G14-1 borehole, and 63–73 for sandstone from the LSF in Loissin-1 borehole. The basement granite in the

AC

G14-1 borehole is relatively fresh (CIA=48–56). The CIA index is used to reflect the progressive alteration of plagioclase and potassium feldspars to clay minerals during the weathering of rocks (Nesbitt and Young, 1982). To evaluate the degree of weathering of plagioclase in rocks, the Plagioclase Index of Alteration (PIA) is calculated using the equation: PIA = 100 × (Al2O3-K2O)/(Al2O3+CaO*+Na2O-K2O) (in molar proportions, Fedo et al., 1995), and the results are listed in Appendix 1. The PIA values of the basement granite in the Borggård borehole are much higher than those for the the basement granite in the G14-1 borehole (93–96 vs. 47–59). The PIA values of the sandstones of the Gadeby Member in the Borggård borehole and the exposures at the Nye Frederiks quarry are comparable (56–99 vs. 61–96). The equivalents of Gadeby Mbr, the AKM in the G14-1 borehole and the LSF in the Loissin-1 borehole, have PIA values of 94–96 and 65–78, respectively. Unaltered Bornholm granitoids contain 13

ACCEPTED MANUSCRIPT well-preserved plagioclase (PIA=46–52 for the orthogneisses, 47–49 for the migmatite, and 46–52 for the complex of granitic rocks, Johansson et al., 2016). To summarize, the basement granite from both the Borggård and the G14-1 boreholes were

T

moderately weathered. The K-feldspars in these rocks are relatively fresh. Plagioclase in the

IP

basement granite from the Borggård borehole was intensely weathered, whereas it is well preserved

SC R

in the basement granite from the G14-1 borehole. The source materials of sandstone of the AKM in Germany, the Gadeby Member in the Borggård borehole and the exposure at the Nye Frederiks quarry have all undergone a similar degree of moderate weathering. Plagioclase grains in the source

NU

materials were mostly weathered. In comparison, the LSF from the Loissin-1 borehole contains unaltered plagioclase in source materials. Several samples of the Gadeby Member from the Borggård

MA

borehole and the Nye Frederiks quarry show slight weathering of the source materials, which is comparable to that in the unaltered Bornholm regional granitoids. This may indicate an involvement

D

of fresh batholith rocks in the source material of the Gadeby Member on Bornholm. The high

TE

correlation between PIA and CIA values for orthogeneisses, migmatite and the complex of granitic rocks on Bornholm (Fig. 6L) are in accordance with the minimal weathering in these rocks. In

CE P

contrast, the data for the sandstones and basement granites from the Borggård and G14-1 boreholes deviate from the trend-line of CIA vs. PIA (Fig. 6L), which may result from K-metasomatism that modified the K content in the rocks.

AC

4.4 Rare earth element geochemistry In order to understand the different behaviors of Rare Earth elements (REE) in a post-Archean sea, the REE data of modern seawater, marine sediments, and post-Archean sedimentary rocks are generally normalized, either to so-called post-Archean average Australian shale (PAAS, Taylor and McLennan, 1985) or to North American Shale Composite (NASC, Goldstein and Jacobsen, 1988). The REE concentrations of igneous rocks are normalized to those of chondrite meteorites, representing the composition of the original starting materials of the Earth (McDonough and Sun,

1995). In this study, chondrite-normalized REE patterns are applied for both the igneous and sedimentary rocks in order to have a consistent standard for comparison of element behaviors. The Cerium and Europium anomalies are defined quantitatively as Ce/Ce*=CeN/[(LaN+PrN)/2] and 14

ACCEPTED MANUSCRIPT Eu/Eu*=EuN/[(SmN+GdN)/2] where Ce* and Eu* are the hypothetical concentrations that strictly trivalent Ce and Eu would have (Taylor and McLennan, 1985).

T

Calculations of the Cerium and Europium anomalies of the investigated samples are listed in

IP

Appendix 1. Fractionation of LREE vs HREE is represented by the (La/Yb)N ratios; these values are

SC R

also listed in Appendix 1.

The basement granite and Gadeby Member in the Borggård borehole share similar REE distribution patterns and have comparable Eu anomalies (Appendix 1, Figs 7A, B). In comparison to

NU

the Gadeby Member, the basement granite beneath has high total REE abundance and smaller fractionation of LREE vs HREE. The size orders of the negative Ce anomaly for the upper part of the

MA

weathered basement granite (0.56–0.96) and the lower part of the Gadeby Member (0.46–0.83) in the Borggård borehole are similar (Appendix 1). Small, and mostly negative Ce anomalies in the

D

weathering profiles of the granitoids are widely reported (Condie et al., 1995; Nesbitt, 1979; Duddy,

TE

1980; Mongelli, 1993), and are the same in the upper part of the basement granite from the Borggård borehole. Whereas the Ce anomaly of the lower part of the basement granite (1.07–1.25, Appendix 1)

CE P

is positive and comparable to those for the upper part of the Gadeby Member (1.01–1.26, Appendix 1) in the Borggård borehole. The small variations in the Ce anomaly could perhaps be indicative of the analytical limits of reproducibility of the measurements, instead of any real redox significance. In

AC

the basement samples from the G14-1 borehole the Ce anomaly and (La/Yb)N vary with depth, changing from 1.08 to 2.33, and from 0.80 to 10.27, respectively (Obst et al., 2004). The Ce and Eu anomalies and the La/Yb ratios of unaltered basement rocks from Bornholm, calculated on the basis of geochemical data from Johansson et al. (2016), are listed in Appendix 1 and shown in Figure 7.

The chondrite-normalized REE abundance for all the investigated sandstone samples consistently show enriched LREE and flat HREE patterns (Fig. 7B). The highly fractionated REE pattern and negative Eu anomalies of the sandstones indicate a provenance from felsic-intermediate basement rocks. 4.5 Trace element geochemistry In consideration of the primary terrigenous sources of the sandstones and the crustal origin of the basement, the Upper Continental Crust (UCC) composition is applied in this study for normalization 15

ACCEPTED MANUSCRIPT of the trace element composition. On the spider diagram, the basement granites from the Borggård and the G14-1 boreholes show enrichment of Th, U, Cs, Rb, Ba, and depletion of Sr, Co and Ni, but in variable degrees (Fig. 8A). Differences in the ratio of plagioclase vs K-feldspar and the weathering

T

intensity may account for the variation. The significant loss of Ca, Sr and Na is associated with the

IP

intense weathering of plagioclase in continental rocks. In contrast, Rb, Cs and Ba are fixed in

SC R

K-feldspar in large quantities in the continental weathering profiles by exchange and adsorption onto secondary clays (Nesbitt and Markovics, 1980) and therefore show relative enrichment. Compared to basement granite from the G14-1 borehole, the basement granites from the Borggård borehole are

NU

enriched in Cs and Rb, but depleted in Ba (Fig. 8A). Several basement samples from the Borggård borehole show extremely high HREE concentrations (Fig. 8A). The granites from the the Borggård

MA

and G14-1 boreholes also show variable Nb/Ta and Zr/Hf ratios (Fig. 8A). The Bornholm granitoids exhibit relatively consistent distribution patterns of trace elements, which are similar to the majority

D

of the basement granites from the Borggård and G14-1 boreholes. The Nb/Ta ratio varies among

TE

orthogneiss, migmatite and the complex of granitic rocks on Bornholm (Fig. 8B). The Zr/Hf ratio of unaltered basement rocks changes in a range that overlaps with those of granites from the Borggård

study (Figs 8A, B).

CE P

and G14-1 boreholes (Fig. 8A). Strontium is consistently depleted in all the granitic rocks in this

Samples of the Gadeby Member from the Borggård borehole and Nye Frederiks quarry

AC

generally have identical distributional patterns of trace elements (Figs 8A, B, C). The content of most trace elements is low relative to the granite, which may be due to the dilution effects of quartz and feldspar. The Gadeby Member in the Borggård borehole contains higher concentrations of K, Cs and Rb than the outcrop at Nye Frederiks quarry, which may have been affected by K-metasomatism. The AKM and LSF follow the general trends of trace element distribution patterns of the Gadeby Member, although the LSF is less abundant in K, Rb and Cs (Fig. 8C). 5. Discussion 5.1 Classification of the basement granite in the Borggård borehole Although the REE can be mobilized and fractionated within a weathering profile, especially during the early stages of weathering (Banfield and Eggleton, 1989; Duddy, 1980; Price et al., 1991; Nesbitt, 1979), there are no net losses or gains of specific REE (Duddy, 1980; Nesbitt, 1979). A uniform shale-like REE distribution is shown by the time REE enter the suspended load of major 16

ACCEPTED MANUSCRIPT rivers (Condie, 1991; Goldstein and Jacobsen, 1988; Martin and Meybeck, 1979). The uniform properties of REE, coupled with their low solubility, facilitate the identification of parent rocks in a weathered profile and source materials for sediments when comparing distribution patterns (Nance

T

and Taylor, 1977; Taylor and Hallberg, 1977; Wildeman and Condie, 1973). The Bornholm granitoids

IP

and the basement granites from both the Borggård and G14-1 boreholes show relatively consistent

SC R

REE distribution patterns (Fig. 7A). Based on the chemical variations of most elements with SiO2, the Bornholm granitoids are divided into a felsic group and an intermediate group that are closely related in origin (Johanssen et al., 2016). The geochemical variations among the Bornholm granitoids likely

NU

represent varying degree of progressive crystallization and removal of ferromagnesian minerals, Ti-bearing phases such as ilmenite and titanite, and P-bearing phases such as apatite, as well as

MA

moderate fractionation of plagioclase (Johanssen et al., 2016). As a result of feldspar fractionation, the size order of the Eu anomaly increases along with the SiO2 content for Bornholm granitoids (Fig.

D

9A, Johanssen et al., 2016). Removal of amphibole and titanite from the magma chamber

TE

consequently leads to a trend of decreasing (Gd/Yb)N with increasing SiO2 content (Fig. 9B, Johanssen et al., 2016). It is therefore possible to classify the basement granites from the Borggård

CE P

borehole by comparing their geochemical signatures with those of Bornholm granitoids. Given the weathering of the basement granite in the Borggård borehole, which modified most of the major element concentrations, the SiO2 content is unsuitable as a classification index. However, the ratio of

AC

(Gd/Yb)N and the size of Eu anomaly usually are transferred from bedrock to weathering profile and remain relatively constant (Nance and Taylor, 1977; Taylor and Hallberg, 1977; Wildeman and Condie, 1973). This also seems to be the case on Bornholm, where basement samples from the Borggård borehole generally plots close to the intermediate group of Bornholm granitoids (Figs 9A, B). Elemental pairs, such as Nb-Ta and Zr-Hf, which have the same charge and nearly identical atomic size, show strongly coherent geochemical behavior in earth processes (Shannon, 1976). Thus no or minimal fractionation is expected between the paired elements (Bougault et al., 1979; Hofmann et al., 1986; Jochum et al., 1986; Hofmann, 1988; Sun and McDonough, 1989). The similarity and immobility of these element pairs in a weathering profile thus provide valuable information for constraining parent rocks and sediment provenance (Nesbitt and Markovics, 1997). Despite large 17

ACCEPTED MANUSCRIPT similarities in geochemical signatures of Bornholm granitoids, variations of Nb/Ta and Zr/Hf ratios are still observed among them (Fig. 8B). Zircon fractionation, along with fractional crystallization of granitic magmas, leads to a change in Zr/Hf ratio (Linnen and Keppler, 2002). The intermediate

T

group generally has higher Nb/Ta and Zr/Hf ratios than the felsic group of Bornholm granitoids. On

IP

an Nb/Ta vs. Zr/Hf diagram, data for the basement granites from the Borggård borehole plot close to

SC R

fields for the orthogneisses and the Svaneke Granite on Bornholm (Fig. 9C). The basement granite from the Borggård borehole has petrological features (e.g., coarse grain-size, red colour, and lack of lineation), that best match those of the Svaneke Granite (Table 1, Platou, 1970; Gravesen et al., 2014).

NU

We therefore conclude that the basement granite from the Borggård borehole likely belongs to the Svaneke Granite on Bornholm.

MA

The absence of hornblende in the basement from the Borggård borehole may be an original feature or it may have been lost through alteration and weathering. Biotite is the only mafic mineral

D

observed, and it is partly altered to chlorite. The abundance of microcline modifies the color from

TE

original grey to secondarily red. Despite these difference due to weathering, the homogeneous medium to coarse grain-size of the basement in the Borggård borehole is similar to that of the

CE P

Svaneke Granite.

5.2 Sandstone Provenance

The REE distributions in fine-grained sedimentary rocks are widely used to characterize

AC

sediment sources (Taylor and McLennan, 1985 and references therein). Felsic-intermediate igneous rocks usually contain higher (La/Yb)N ratios and negative Eu anomalies, whilst mafic igneous rocks contain lower (La/Yb)N ratios with little or no Eu anomalies (Cullers, 1994, 2000). Sedimentary rocks inherit the REE signatures of the parental rock with minimal changes during chemical and physical weathering (Condie, 1991, 1993; Koppi et al., 1996; Taylor and McLennan, 1985; Wronkiewicz and Kent, 1989). All the sandstone samples in this study show high (La/Yb)N ratios and negative Eu anomalies that are similar to those of basement granite in the Borggård borehole, indicating a provenance from basement rocks with similar compositions (Fig. 7B, Appendix 1). The Gadeby Member exposed in the Nye Frederiks quarry has (La/Yb)N ratios and a Eu content similar to the LSF , indicating roughly similar provenance areas for these sandstones or, at least, sources of similar composition. Two samples of the AKM have measured (La/Yb)N and Eu anomaly values, which are 18

ACCEPTED MANUSCRIPT outliers on the (La/Yb)N vs. Eu anomaly diagram (Fig. 10A), are omitted from the following discussion. The Gadeby Member exposed in the Nye Frederiks quarry and the LSF plot together with the intermediate Bornholm granitoids as defined by Johansson et al. (2016). The Gadeby Member

T

from the Borggård borehole differ from the other sandstone samples, showing variable (La/Yb)N

IP

ratios but relatively constant Eu levels (Fig. 10A). Fractionation of LREE from HREE may occur

SC R

during the formation of clay minerals, a process in which the LREE are preferentially adsorbed onto the clays, while the HREE are stabilized as complexes in solution (Ronov et al., 1967; Roaldset, 1973; Nesbitt, 1979; Duddy, 1980). The comparable PIA values for the Gadeby Member from the

NU

Borggård borehole, in comparison with the other sandstones analyzed here, excludes the influence of plagioclase weathering as an explanation of the observed large variations in (La/Yb)N. The Gadeby

MA

Member from the Borggård borehole shows higher concentrations of K, Cs and Rb than the other sandstones analyzed. This difference may record a higher proportion of microcline that formed from

D

K-metasomatic replacement of plagioclase. The replacement also resulted in formation of secondary

TE

clay minerals, which accounts for lower LREE concentrations that lead to the fractionation of LREE from HREE. The majority of samples from the Gadeby Member taken from the Borggård borehole

CE P

have and Eu anomalies comparable to those of the weathered basement granites beneath the unaltered intermediate group on Bornholm, while their (La/Yb)N ratios are dispersive which may be due to K-metamorphism. This suggests that the source materials of the Gadeby Member from the Borggård

AC

borehole were likely to be derived from both fresh and weathered intermediate group of the granitoids on Bornholm. For terrigenous clastic sediments, elements with low mobility (i.e. Zr, Hf, Ti, Nb, Ta, Cr and Sn) are preferentially hosted within stable heavy mineral phases. For example, zircon is the main host of Zr and Hf, rutile/ilmenite/anatase host Ti, Nb and Ta, and chrome spinel contains Cr (Preston et al., 1998; Kamber et al., 2005). Weathering and diagenetic processes are less likely to affect the properties of these immobile elements, although a modification of concentrations can be expected from sorting of the heavy minerals during transport (McLennan et al., 1993). However, no or little fractionation occurs between element pairs with identical charge and ionic radii, i.e., Nb-Ta and Zr-Hf (Goldschmidt, 1937), which therefore allows for provenance identification (Cullers, 1988; Wronkiewicz and Condie, 1990; Cullers and Stone, 1991; McLennan, 2001; Svenden and Hartley, 19

ACCEPTED MANUSCRIPT 2002). The Zr/Hf and Nb/Ta ratios of the sandstones analyzed in this study mostly fall within the range of values for the intermediate Bornholm granitoids (Figs 10B, C). The Gadeby Member from Nye Frederiks Quarry (37.65–39.16) and the LSF (36.00–39.16) have similar Zr/Hf ratios to the

T

Gadeby Member from the Borggård borehole (35.04–38.08), and the values are very close to those

IP

for the weathered basement granite in the Borggård borehole (37.54–39.22) as well as the Svaneke

SC R

Granite (31.56–41.78). The Nb/Ta ratios for the Gadeby Member from the Borggård borehole (5.86– 14.54) are scattered. Some of the samples have Nb/Ta ratio (12.35–14.54) overlapping with those of the weathered basement granite beneath (13.76–15.73). The Gadeby Member from Nye Frederiks

NU

Quarry and the remaining samples of the Gadeby Member from the Borggård borehole have Nb/Ta ratios (9.62–12.45) deviating from the values of the underlying weathered basement (13.76–15.73).

MA

However, these values are still in the range of the Svaneke Granite (7.80–16.62), and are partly comparable to those of the intermediate Bornholm granitoids (7.80–18.58, Fig. 10C). These lines of

D

evidence suggest that the source materials of the Gadeby Member in the Borggård borehole are

TE

derived from both weathered and fresh intermediate basement granite on Bornholm. The Gadeby Member from the Nye Frederiks quarry has identical provenance. The geochemical composition of

CE P

the AKM and LSF indicates a provenance from intermediate granitoids like seen on Bornholm. 5.3 Weathering intensity

During chemical weathering, the proportion of alumina to alkalis changes with the degradation of

AC

feldspars and concomitant formation of clay minerals (Nesbitt and Young, 1982, 1984). To evaluate the weathering intensity, major element chemistry is most applicable for determining the weathering level of the source terrain. Several indices, e.g., the Chemical Index of Alteration (CIA, Nesbitt and Young, 1982) and the Plagioclase Index of Alteration (PIA, Fedo et al., 1995), were developed to quantify the degree of weathering. The CIA is a chemical proxy to measure the extent of conversion of minerals to clay minerals, with values of ~50 for unweathered feldspars, ~10 for unweathered clinopyroxene, and ~30 for unweathered hornblende. In unaltered rocks the CIA values range from 30 to 45 for fresh basalt or gabbro, and from 45 to 55 for rhyolite, granite and granodiorite (Nesbitt and Young, 1982). The range is considerably greater for sedimentary rocks, which are composed of complex mineralogical mixtures commonly including phyllosilicates, carbonates or phosphates. Aluminum-rich phyllosilicates have high CIA values, ranging from 75 for muscovite and ~80 for 20

ACCEPTED MANUSCRIPT illite and smectite to 100 for chlorite and kaolinite (Nesbitt and Young, 1982). The PIA values for fresh rocks are ~50, reaching up to 100 for clay minerals such as kaolinite, illite, and gibbsite, consistent with values derived from the CIA equation (Fedo et al., 1995).

T

The basement granite from the Borggård and G14-1 boreholes are moderately weathered as

IP

indicated by their low to medium CIA values (66.52–70.14). The very high PIA values calculated for

SC R

the granites from the Borggård borehole (93.02–98.21), relative to those for basement granites from the G14-1 borehole (47.59–58.83), suggest an intense weathering of plagioclase. The sandstones treated in this study can be divided into three groups based on the PIA and CIA values. Group I

NU

includes a few of the Gadeby Member samples from the Borggård borehole (CIA=52.25–55.19, PIA=56.44–77.66), and the exposure at Nye Frederiks quarry (CIA=52.71, PIA=61.05), which show

MA

relatively low CIA and PIA values comparable to those of fresh Bornholm granitoids (CIA=47.35– 52.62, PIA=46.08–54.28). Group II consists of the majority of the LSF with medium CIA and PIA

D

values (CIA=63.92–72.22, PIA=65.74–77.61), suggesting a moderate weathering of the source

TE

materials. Group III contains the majority of the Gadeby Member samples from the Borggård borehole (CIA=63.74–68.69, PIA=83.50–98.87), and the AKM from G14-1 (CIA=71.45–71.87,

CE P

PIA=94.69–95.13), which are characterized by moderate CIA but extremely high PIA values, similar to those for the basement granites from the Borggård borehole (CIA=66.52–70.14, PIA=93.02– 98.21). The Gadeby Member from the Borggård borehole has CIA and PIA values that overlap with

AC

both the weathered basement granites beneath and the fresh regional Bornholm granitoids, confirming a contribution from both weathered and relatively fresh sources. Weathering intensity is further evaluated using A-CN-K ternary diagrams, which shows empirically and kinetically predictable weathering vectors for various minerals and rock types, thus allowing for graphical interpretation of proportional chemical changes (Nesbitt and Young, 1984; Nesbitt and Wilson, 1992). The overall weathering vector for feldspar decomposition in various parent rocks is supposed to be parallel to the A-CN axis, but the precise vector direction is a function of the relative proportion of plagioclase and K-feldspar, their congruent or incongruent dissolution, and the rate of conservation of aluminous weathering products (Babechuk et al., 2015). On a A-CN-K ternary diagram, the basement granite from the Borggård borehole follows the overall weathering vector for feldspar decomposition, falling on the A-K axis, but deviates from the ideal weathering 21

ACCEPTED MANUSCRIPT trend defined by the fresh Bornholm granitoids (Fig. 11). Should the degree of weathering increase, the vector is predicted to continue until it reaches the A apex when Ca and Na have been completely depleted. However, the data for the weathered basement granites from Borggård borehole spread

T

toward the K apex recording K enrichment due to metasomatism (Fig. 11). The majority of samples

IP

from the Gadeby Member in the Borggård borehole and the LSF follow the same weathering trend as

SC R

the basement granites beneath. The K enrichment through metasomatism can occur by the conversion of aluminous clay minerals (i.e., kaolinite as matrix) to illite, or by the conversion of plagioclase to K-feldspar (Fedo et al., 1995). The basement granites and the majority of the Gadeby Member from

NU

the Borggård borehole have extremely high PIA values, which means few plagioclase were preserved after the weathering for its conversion to K-feldspar. K metasomatism is therefore likely to occur by

MA

the conversion of kaolinite to illite. Several samples of the Gadeby Member from the Borggård borehole and the Nye Frederiks quarry have relatively low PIA and CIA values, but show intense K

D

metasomatism. This may be due to the conversion of plagioclase to K-feldspar, a process that requires

TE

more potassium than the conversion of kaolinite to illite. No K metasomatism is indicated for the AKM (Fig. 11). The source of potassium is possibly internal to granitoids–sandstone units, wherein

CE P

K-feldspar is dissolved during weathering, releasing K for clay illitization in closed-system alteration (van de Kamp, 2016). However, the K-feldspar grains are fresh and intact in both the basement granite and sandstone in the Borggård borehole, only recording features of trivial alteration. Hence,

AC

an internal contribution of potassium for K metasomatism is thought to be minor. External sources in open-system diagenesis, i.e., from circulating brines, are possible to provide the potassium necessary in amounts greater than available from closed-system alteration (van de Kamp, 2016). 5.4 Quantification of mass transfer To quantify mass transfer during weathering of the basement granites from the Borggård borehole, the tau (τ) mass-transport model (Brimhall and Dietrich, 1987; Anderson et al., 2002) was applied to the mass balance calculations, assuming immobile Ti during weathering. In this model, the concentrations (C) of elements (j) in the parent rock (p), relative to that of an immobile index element (i), are used as a normalization to establish the mass changes in the progressively altered rock using the formula (w): τ(i, j)= [(Cj,w)/(Cj,p)]/ [(Ci,w)/(Ci,p)]−1. 22

ACCEPTED MANUSCRIPT When τj=-1, element j is completely removed during weathering; When τj=0, the element is immobile during weathering with respect to the parent rock. The average geochemical compositions of eight Svaneke Granite samples were taken to represent the composition of unaltered parent rock.

T

Titanium is regarded as the immobile index element for τ calculation. The results calculated for

IP

selected elements are presented in Table 2. During weathering of the basement granite in the

SC R

Borggård borehole, Na shows the greatest loss (τNa from −0.97 to −0.98), followed by Ca (τCa from −0.69 to −0.95). Mg (τMg from −0.60 to −0.78) and P (τP from −0.57 to 0.66) are moderately depleted,

whereas Si is slightly depleted (τSi from −0.21 to −0.30). The positive τFe (0.12–0.20) indicates a

NU

slight gain of Fe during weathering. One basement granite sample shows extraordinary high values of τFe=0.52 and τTi=0.33, which may include hydrothermal veins.

MA

The complete transformation of plagioclase to kaolinite during weathering results in a significant release of Na and Ca (Exq. 1, Fig. 12). Mobile Na is easily dissolved in water and removed

D

(τNa=−0.97 to −0.98). However, rather than kaolinite, which is normally the weathering product of

TE

plagioclase, only illite was detected by XRD (Fig. 4). Illite in the weathered basement granite may be largely diagenetic in origin, transforming from kaolinite in an environment saturated with K and Fe

CE P

(Exq. 2, Fig. 12). The pressures and temperatures required for such a transformation are related to its deep burial and heating during late Silurian-Devonian, associated with the Caledonian foreland basin. The transformation of kaolinite to illite not only retained and enriched Fe (τFe=0.12–0.20), but also

AC

captured and gained minor K (τK=−0.12–0.14). Through the transformation from plagioclase to illite, a stoichiometric relationship would be expected between the gain of K and loss of Na (12 Na+ loss: 1K+ gain). For the basement granite in the Borggård borehole, τK is therefore predicted to be at around 0.08 based on a τNa of −0.98. However, several granite samples show higher τK values than 0.08, suggesting other sources of K to the enrichment. An alternative explanation is an involvement of the replacement of plagioclase by K-feldspar (1 Na+ loss: 1K+ gain) (Exq. 3, Fig. 12). For instance, sample Bo 313.15 with τNa=−0.98 and τK=0.16 would require conversion of approximately 16.4% plagioclase to illite, and replacement of 73.6% plagioclase by K-feldspar to reach the mass balance between Na and K. Estimates of mineralogical transformation during the weathering can be made for the basement granites based on the quantitative mass transfer. Oxidation of Mg-Fe minerals formed the clay 23

ACCEPTED MANUSCRIPT minerals, hematite and goethite, and leached Mg2+ and Ca2+ in the basement granites (Exq. 4, Fig. 12). If primary hornblende existed, a positive relationship between τCa and τMg would be expected for the stoichiometric dissolution of hornblende (2τCa: 5τMg). However, the negative linear relationship

T

between τCa and τMg suggests an alternative source for the depletion of Mg and/or Ca. Oxidation of

IP

titanite, an accessary mineral which is common in the Bornholm granitoids but absent in basement

SC R

granite from the Borggård borehole, can also release and deplete Ca2+ (Exq. 5, Fig. 12). Weathering of Ca-rich plagioclase can be another source of Ca depletion (Exq. 1, Fig. 12). Biotite was partly altered to chlorite, a process that captures Ca2+ and releases Fe2+ and K+ (Exq. 6, Fig. 12). Under the oxic

NU

surface environment, the Fe2+ released from weathering reactions combined with O2 and H2O to form goethite (Exq. 7, Fig. 12). Dehydration further transformed goethite to hematite, a mineral observed

MA

in the basement granite from the Borggård borehole (Exq. 8, Fig. 12). Primary ilmenite is absent, but instead its oxidation products, hematite and anatase, are observed (Exq. 9, Fig. 12). τSi values ranging

D

from −0.21 to −0.30 suggest that silica was lost during weathering of silicates, but partly retained in

TE

the secondary clay minerals.

5.5 Weathering differentiation of feldspar

CE P

The basement granites from the Borggård borehole have extremely high PIA values but moderate CIA values, and the data deviate from the CIA-PIA trend (Fig. 6L), indicating that alteration of plagioclase, rather than K-feldspar, was dominant during weathering. This is confirmed

AC

by the petrological features of intact K-feldspar grains and significantly altered plagioclase (Fig. 3). Differential rates of plagioclase and K-feldspar weathering are commonly observed in bedrock and soil environments (Goldich, 1938; White et al., 2001). A fundamental reason for the variations in the weathering rates of feldspars is the different internal permeability developed during crystallization of plagioclase vs. K-feldspar in magmas. Of the silicate minerals, plagioclase has the highest internal porosity in unweathered granite (0.5–1%, Montgomery and Brace, 1975), providing pathways for weathering fluids. During the early stages of magma cooling, plagioclase crystallizes and forms chains of physically linked phenocrysts in granites and other crystalline rocks (Bryon et al., 1995). In contrast, the late nucleation and growth of K-feldspar are restricted to interstices of this crystalline framework and are thus more physically isolated. This interconnection of plagioclase phenocrysts, coupled with relatively high internal porosity, produces a network of conduits by which meteoric 24

ACCEPTED MANUSCRIPT water first penetrates into the pristine granite matrix and initiates plagioclase weathering. Additionally, plagioclase composition affects the development of internal permeability. Experimental dissolution rates of plagioclase suggest that Ca-rich plagioclase is more rapidly

T

dissolved than Na-rich plagioclase (Blum and Stillings, 1995). Experimental dissolution rates of

IP

plagioclase suggest that granite containing oligioclase may weather a factor of 5 times faster than

SC R

granite containing albite (Blum and Stillings, 1995). Variation in climate cannot be an important factor in the weathering of basement granites in the Borggård and G14-1 boreholes located only ~50 km apart. To account for the large variation of PIA values between the basement in the Borggård

NU

(PIA=93–99) and the G14-1 (PIA=47–59) boreholes, compositional differences of plagioclase are therefore more likely. The primary plagioclase in the basement granites from the Borggård borehole

MA

is likely to be richer in Ca than that from the G14-1 borehole.

Climate difference is another variable that affects feldspar weathering rates via water availability

D

and temperature (White and Blum 1995; White et al., 1999). Water causes silicate hydrolysis and

TE

transports soluble weathering products away from reaction sites. Temperature primarily controls weathering rates and mineral solubility (Lasaga et al., 1994). Compared to high temperatures,

CE P

increasing water availability appears to enhance the climate-erosion interactions that control the Na depletion of feldspar (Holland and Turekian, 2010; Rasmussen et al., 2011). During the initiation of chemical weathering, the permeability of fresh granite is extremely low, and weathering is limited by

AC

severe constraints on the fluid flux and the mass of plagioclase that can dissolve before becoming thermodynamically saturated (Holland and Turekian, 2010). Under such conditions, weathering is mostly limited by the availability of water. Over longer time-scales, slow rates of transport-limited weathering occur, resulting in mass loss from the granite. An increase in porosity from the kaolinization of plagioclase also increases the flux of water, which accelerates the saturation-limited weathering, producing a greater porosity and even higher fluid fluxes. Plagioclase weathering is thus accelerated by this feedback. The significant depletion of Na in the basement granite from the Borggård borehole possibly reflects that weathering occurred in a water-saturated environment in a very humid climate. Due to its lower solubility (and not slightly slower reaction kinetics), the rate of K-feldspar weathering is very limited (Holland and Turekian, 2010). The dissolution of K-feldspar was further suppressed by concurrent plagioclase dissolution which produced solutes, principally Si, 25

ACCEPTED MANUSCRIPT and increased the saturation state. K-feldspar, therefore, was resistant to weathering relative to plagioclase in the basement granite from the Borggård borehole. To summarize, the internal permeability developed during crystallization, the Ca-rich

T

composition, and a very humid environment probably account for the intense weathering of

IP

plagioclase compared to K-feldspar in the basement granite from the Borggård borehole. However,

SC R

the variation of sampling depth in the weathering profile and the local fractures and consequent differences in water movement in the rock within the G14-1 and Borggård boreholes may also affect the weathering intensity of the feldspar, which requires more future sampling work to clarify.

NU

6. Summary and conclusions

- The low to medium CIA values of the Precambrian basement in the Borggård borehole on

MA

Bornholm indicate it is moderately weathered. Microcline is well preserved, whereas, the extremely high PIA values suggest an intense weathering of the plagioclase. Fe-Ti accessory minerals were

D

oxidized to hematite and anatase.

TE

- Trace element signatures (REE distribution patterns, Nb/Ta and Zr/Hf ratios) and petrological characteristics suggest that the basement granites in the Borggård borehole is a weathered equivalent

CE P

of Svaneke Granite .

- During weathering of the basement granite in the Borggård borehole, major elements were depleted in the following order: Na>Ca>Mg>Si; Al and Ti were immobile and stay constant; K shows

AC

sample-dependent enrichment or depletion whereas Fe is slightly enriched. -The Precambrian basement in the G14-1 borehole offshore Rügen in Germany shows similar REE distribution patterns as the granites on Bornholm. Compared to granite in the G14-1 borehole, the contemporaneous basement in the Borggård borehole contains intensely weathered plagioclase, which may be due to an original Ca-rich composition. - The Cambrian Gadeby Member of the Nexø Formation in the Borggård borehole is feldspathic litharenite-litharenite in composition, and is genetically related to the basement granite beneath. The Gadeby Member in the Borggård borehole and from the Nye Frederiks quarry on Bornholm were sourced from both weathered and fresh granite of intermediate composition. -The Adlergrund Konglomerat Member (AKM) from the G14-1 borehole and the Lubmin Sandstein Formation (LSF) from the Loissin-1 borehole in Germany, derive from basement rocks 26

ACCEPTED MANUSCRIPT with compositions similar to the intermediate group of the granitoids on Bornholm . - The source materials for the AKM, the Gadeby Member in the Borggård borehole and at the Nye Frederiks quarry on Bornholm have undergone similar degrees of moderate weathering.

T

Plagioclase in the source materials was almost totally weathered. In contrast, the LSF from the

IP

Loissin-1 borehole in Germany contains relatively fresh plagioclase derived from a less weathered

SC R

source.

- K-metasomatism occurred in the basement granite and sandstone from both the Borggård and G14-1 boreholes, mainly by transforming kaolinite to illite, associated with local occurrence of

NU

transformation of plagioclase to K-feldspar. This may have taken place during deep burial in the Caledonian foreland basin during late Silurian-Devonian.

MA

- The significantly differential rates of plagioclase and K-feldspar weathering in the basement granite and the sandstone from the Borggård borehole are likely caused by different internal crystal

Acknowledgments:

TE

D

structures, a Ca-rich original composition of the plagioclase, and a very humid climate.

CE P

We thank Åke Johansson, Tod Waight, and Kristina Zariņš for their generosity in sharing unpublished geochemical data with us in connection with the first submission of this manuscript. Their kindness is of significant importance to completion of this manuscript and to finishing the first author’s Ph.D. We

AC

also acknowledge Nelson Eby (Editor-in-Chief) and all the reviewers for their interests in our work and constructive comments. Appreciation is also given to John Caulfield and an anonymous reviewer for their time and great patience in correcting the grammar of this manuscript. This work also benefited from discussions with Balz S. Kamber. Dong Energy is acknowledged for financial support of the geochemical analyses.

27

ACCEPTED MANUSCRIPT References: Anderson, S.P., Dietrich, W.E., Brimhall, G.H., 2002. Weathering profiles, mass-balance analysis, and rates of solute loss: Linkages between weathering and erosion in a small, steep catchment.

T

Geological Society of America Bulletin 114, 1143-1158.

IP

Babechuk, M.G., Widdowson, M., Murphy, M., Kamber, B.S., 2015. A combined Y/Ho, high field

SC R

strength element (HFSE) and Nd isotope perspective on basalt weathering, Deccan Traps, India. Chemical Geology 396, 25-41.

Bahlburg, H., Dobrzinski, N., 2011. A review of the Chemical Index of Alteration (CIA) and its

NU

application to the study of Neoproterozoic glacial deposits and climate transitions. Geological Society, London, Memoirs 36, 81-92.

MA

Banfield, J.F., Eggleton, R.A., 1989. Apatite replacement and rare earth mobilization, fractionation, and fixation during weathering. Clays and Clay Minerals 37, 113-127. Bhatia, M.R., 1983. Plate tectonics and geochemical composition of sandstones. The Journal of

D

Geology 91, 611-627.

TE

Bhatia, M.R., 1984. Composition and classification of Paleozoic flysch mudrocks of eastern

249-268.

CE P

Australia: implications in provenance and tectonic setting interpretation. Sedimentary Geology 41,

Birkeland, P. W., 1984. Soils and geomorphology. Oxford University Press.

AC

Blum, A.E., Stillings, L.L., 1995. Chemical weathering of feldspars. Chemical Weathering Rates of Silicate Minerals, Min Soc Am Rev Min 31, 291-351. Bogdanova, S., Gorbatschev, R., Grad, M., Janik, T., Guterch, A., Kozlovskaya, E., Motuza, G., Skridlaite, G., Starostenko, V., Taran, L., 2006. EUROBRIDGE: new insight into the geodynamic evolution of the East European Craton. Geological Society, London, Memoirs, 32, pp.599-625. Bogdanova, S.V., Bingen, B., Gorbatschev, R., Kheraskova, T.N., Kozlov, V.I., Puchkov, V.N., Volozh, Y.A. 2008. The East European Craton (Baltica) before and during the assembly of Rodinia. Precambrian Research 160, 23-45. Bougault, H., Joron, J. L., Treuil, M., 1979. Alteration, fractional crystallization, partial melting, mantle properties from trace elements in basalts recovered in the North Atlantic. Deep Drilling Results in the Atlantic Ocean: Ocean Crust, 352-368.

28

ACCEPTED MANUSCRIPT Brimhall, G.H., Dietrich, W.E., 1987. Constitutive mass balance relations between chemical composition, volume, density, porosity, and strain in metasomatic hydrochemical systems: results on weathering and pedogenesis. Geochimica et Cosmochimica Acta 51, 567-587.

T

Bruun-Petersen, J., 1972. Forvitret granit under Neksø Sandstenen. Dansk Geologisk Forening,

IP

Årsskrift for, 61-76 (in Danish).

SC R

Bruun-Petersen, J. 1977. Nedre Kambrium, in: Hansen, M., Poulsen, V. (Eds). Geologi på Bornholm, 2. edition. VARV, pp. 33-39 (in Danish).

Bryon, D.N., Atherton, M.P., Hunter, R.H., 1995. The interpretation of granitic textures from serial

NU

thin sectioning, image analysis and three-dimensional reconstruction. Mineralogical Magazine 59, 203-211.

MA

Buchardt, B., Nielsen, A.T., Schovsbo, N.H., Bojesen-Kofoed, J.A., 1997. Alun skiferen i Skandinavien. Geologisk Tidsskrift, 1-3 (in Danish).

D

Callisen, K., 1934. Das Grundgebirge von Bornholm. Danmarks Geologiske Undersøgelse II Række

TE

50, 266 (in German).

Clemmensen, L.B., Dam, G., 1993. Aeolian sand-sheet deposits in the Lower Cambrian Neksø

CE P

Sandstone Formation, Bornholm, Denmark: sedimentary architecture and genesis. Sedimentary Geology 83, 71-85.

Condie, K.C., 1991. Another look at rare earth elements in shales. Geochimica et Cosmochimica

AC

Acta, 55, 2527-2531.

Condie, K.C., 1993. Chemical composition and evolution of the upper continental crust: contrasting results from surface samples and shales. Chemical Geology 104, 1-37. Condie, K.C., Dengate, J. and Cullers, R.L., 1995. Behavior of rare earth elements in a paleoweathering profile on granodiorite in the Front Range, Colorado, USA. Geochimica et Cosmochimica Acta 59, 279-294. Cullers, R.L., 1988. Mineralogical and chemical changes of soil and stream sediment formed by intense weathering of the Danburg granite, Georgia, USA. Lithos 21, 301-314. Cullers, R.L., 1994. The controls on the major and trace element variation of shales, siltstones, and sandstones of Pennsylvanian-Permian age from uplifted continental blocks in Colorado to platform sediment in Kansas, USA. Geochimica et Cosmochimica Acta 58, 4955-4972. 29

ACCEPTED MANUSCRIPT Cullers, R.L., 2000. The geochemistry of shales, siltstones and sandstones of Pennsylvanian–Permian age, Colorado, USA: implications for provenance and metamorphic studies. Lithos 51, 181-203. Cullers, R.L., Stone, J., 1991. Chemical and mineralogical comparison of the Pennsylvanian

T

Fountain Formation, Colorado, USA (an uplifted continental block) to sedimentary rocks from other

IP

tectonic environments. Lithos 27, 115-131.

SC R

Dickinson, W.R., Beard, L.S., Brakenridge, G.R., Erjavec, J.L., Ferguson, R.C., Inman, K.F., Knepp, R.A., Lindberg, F.A., Ryberg, P.T., 1983. Provenance of North American Phanerozoic sandstones in relation to tectonic setting. Geological Society of America Bulletin 94, 222-235.

NU

Duddy, L.R., 1980. Redistribution and fractionation of rare-earth and other elements in a weathering profile. Chemical Geology 30, 363-381.

MA

Feakes, C.R., Retallack, G.J., 1988. Recognition and chemical characterization of fossil soils developed on alluvium; a Late Ordovician example. Geological Society of America Special Papers

D

216, 35-48.

TE

Fedo, C.M., Nesbitt, H.W., Young, G.M., 1995. Unraveling the effects of potassium metasomatism in sedimentary rocks and paleosols, with implications for paleoweathering conditions and provenance.

CE P

Geology 23, 921-924.

Feldrappe, H., Hahne, K., Rhede, D., 2005. Regionale Stellung und Alter der präpermischen. Zeitschrift der Deutschen Gesellschaft für Geowissenschaften 156, 299-322 (in German).

AC

Friis, H., Sylvestersen, R.L., Nebel, L.N., Poulsen, M.L.K., Svendsen, J.B., 2010. Hydrothermally influenced cementation of sandstone—An example from deeply buried Cambrian sandstones from Bornholm, Denmark. Sedimentary Geology 227, 11-19. Gay, A.L., Grandstaff, D.E., 1980. Chemistry and mineralogy of Precambrian paleosols at Elliot Lake, Ontario, Canada. Precambrian Research 12, 349-373. Goldich, S.S., 1938. A study in rock-weathering. Journal of Geology 46, 17-58. Goldschmidt, V.M., 1937. The principles of distribution of chemical elements in minerals and rocks. The seventh Hugo Müller Lecture, delivered before the Chemical Society on March 17th, 1937. Journal of the Chemical Society (Resumed) 1937, 655-673. Goldstein, S.J., Jacobsen, S.B., 1988. Rare earth elements in river waters. Earth and Planetary Science Letters 89, 35-47. 30

ACCEPTED MANUSCRIPT Gravesen, P., 1996. Bornholm: en beskrivelse af områder af national geologisk interesse. Miljø- og Energiministeriet, Skov- og Naturstyrelsen (in Danish). Gravesen, P., Binderup, M., Nilsson, B., Pedersen, S.A.S., 2011. Geological characterisation of

T

potential disposal areas for radioactive waste from Risø, Denmark. Geological Survey of Denmark

IP

and Greenland Bulletin 23, 21-24.

SC R

Gravesen, P., Nilsson, B., Rasmussen, P., Pedersen, S.A.S., 2014. Borehole logs from the Precambrian basement on Bornholm, eastern Denmark: geology and groundwater flow. Geological Survey of Denmark and Greenland Bulletin 31, 15-18.

NU

Gry, H., 1936. Om Nexø sandstenen og Aakerformationen. Meddelelser fra Dansk Geologisk Forening 9, 27-42 (in Danish).

MA

Hamberg, L., 1991. Tidal and seasonal cycles in a Lower Cambrian shallow marine sandstone (Hardeberga Fm.), Scania, Southern Sweden. Clastic Tidal Sedimentology - Memoir 16, 255-273.

D

Hansen, D.L., Nielsen, S.B., Lykke-Andersen, H., 2000. The post-Triassic evolution of the

TE

Sorgenfrei–Tornquist Zone—results from thermo-mechanical modelling. Tectonophysics 328, 245-267.

CE P

Hansen, K., 1936. Die Gesteine des Unterkambriums auf Bornholm. Danmarks Geologiske Undersøgelse, II Række, 62, pp. 194 (in German). Hansen, K., 1995. Fennoscandian Borderzone: thermal and tectonic history of a tuffaceous sandstone

AC

and granite from fission track analysis, Bornholm, Denmark. Tectonophysics 244, 153-160. Hofmann, A.W., Jochum, K.P., Seufert, M., White, W.M., 1986. Nb and Pb in oceanic basalts: new constraints on mantle evolution. Earth and Planetary Science Letters 79, 33-45. Hofmann, H.D., 1988. Ciliary neuronotrophic factor stimulates choline acetyltransferase activity in cultured chicken retina neurons. Journal of neurochemistry 51, 109-113. Holland, H.D., 1984. The chemical evolution of the atmosphere and oceans. Princeton University Press. Holland, H.D., Turekian, K.K. (Eds.), 2010. Geochemistry of Earth Surface Systems: A Derivative of the Treatise on Geochemistry. Academic Press, pp. 227-230.

31

ACCEPTED MANUSCRIPT Holm, P.M., Pedersen, L.E., Højsteen, B., 2010. Geochemistry and petrology of mafic Proterozoic and Permian dykes on Bornholm, Denmark: Four episodes of magmatism on the margin of the Baltic Shield. Bulletin of the Geological Society of Denmark 58, 35-65.

IP

Bornholm. Bulletin of the Geological Society of Denmark 26, 69-76.

T

Jensen, Aa., 1977. Character and provenance of the opaque minerals in the Nexø Formation,

SC R

Jensen, S.K., Nielsen, S.B., 1995. Estimating amount and timing of Late Paleozoic uplift and erosion in the Rønne Graben, Bornholm, Denmark. Bulletin. Geological Society of Denmark 42, 23-33. Jochum, K. P., Seufert, H. M., Spettel, B., Palme, H., 1986. The solar-system abundances of Nb, Ta,

NU

and Y, and the relative abundances of refractory lithophile elements in differentiated planetary bodies. Geochimica et Cosmochimica Acta, 50, 1173-1183.

MA

Johansson, Å., Waight, T., Andersen, T., Simonsen, S.L., 2016. Geochemistry and petrogenesis of Mesoproterozoic A-type granitoids from the Danish island of Bornholm, southern Fennoscandia.

D

Lithos 244, 94-108.

TE

Kamber, B.S., Greig, A., Collerson, K.D., 2005. A new estimate for the composition of weathered young upper continental crust from alluvial sediments, Queensland, Australia. Geochimica et

CE P

Cosmochimica Acta 69, 1041-1058.

Koppi, A.J., Edis, R., Field, D.J., Geering, H.R., Klessa, D.A., Cockayne, D.J., 1996. Rare earth element trends and cerium-uranium-manganese associations in weathered rock from Koongarra,

AC

Northern Territory, Australia. Geochimica et Cosmochimica Acta 60, 1695-1707. Lasaga, A.C., Soler, J.M., Ganor, J., Burch, T.E., Nagy, K.L., 1994. Chemical weathering rate laws and global geochemical cycles. Geochimica et Cosmochimica Acta 58, 2361-2386. Liboriussen, J., Ashton, P., Tygesen, T., 1987. The tectonic evolution of the Fennoscandian Border Zone in Denmark. Tectonophysics 137, 21-29. Lindström, M., Staude, H., 1971. Beitrag zur Stratigraphie der unterkambrischen Sandsteine des südlichsten Skandinaviens. Geologica et Palaeontologica 5, 1-7. Linnen, R.L., Keppler, H., 2002. Melt composition control of Zr/Hf fractionation in magmatic processes. Geochimica et Cosmochimica Acta 66, 3293-3301. Martin, J.M., Meybeck, M., 1979. Elemental mass-balance of material carried by major world rivers. Marine chemistry 7, 173-206. 32

ACCEPTED MANUSCRIPT Mazur, S., Mikolajczak, M., Krzywiec, P., Malinowski, M., Buffenmyer, V., Lewandowski, M., 2015. Is the Teisseyre‐Tornquist Zone an ancient plate boundary of Baltica?. Tectonics 34, 2465-2477. McDonough, W.F., Sun, S.S., 1995. The composition of the Earth. Chemical geology, 120, 223-253.

T

McLennan, S.M., 2001. Relationships between the trace element composition of sedimentary rocks

IP

and upper continental crust. Geochemistry, Geophysics, Geosystems 2, Paper number

SC R

2000GC000109.

McLennan, S.M., Hemming, S., McDaniel, D.K., Hanson, G.N., 1993. Geochemical approaches to sedimentation, provenance, and tectonics. Geological Society of America Special Papers 284, 21-40.

NU

Micheelsen, H.I., 1961. Leucogranites in the Pre-Cambrian of Bornholm, Denmark. Meddelelser fra Dansk Geologisk Forening 14, 297-307.

MA

Milthers, V., 1930. Bornholms Geologi. Danmarks Geologiske Undersøgelse, V. Række 1, 140 pp. (in Danish).

D

Mogensen, T.E., 1994. Paleozoic structural development along the Tornquist Zone, Kattegat area,

TE

Denmark. Tectonophysics 240, 191-214.

Mogensen, T.E., Korstgård, J.A., 2003. Triassic and Jurassic transtension along part of the

CE P

Sorgenfrei-Tornquist Zone, in the Danish Kattegat. In: Ineson, J.R., Surlyk, F. (Eds): The Jurassic of Denmark and Greenland. Geological Survey of Denmark and Greenland Bulletin 1, 439-458. Mongelli, G., 1993. REE and other trace elements in a granitic weathering profile from “Serre”,

AC

southern Italy. Chemical Geology 103, 17-25. Montgomery, C.W. and Brace, W.F., 1975. Micropores in plagioclase. Contributions to Mineralogy and Petrological 52, 17-28. Nance, W.B., Taylor, S.R., 1977. Rare earth element patterns and crustal evolution—II. Archean sedimentary rocks from Kalgoorlie, Australia. Geochimica et Cosmochimica Acta 41, 225-231. Nesbitt, H.W., 1979. Mobility and fractionation of rare earth elements during weathering of a granodiorite. Nature 279, 206-210. Nesbitt, H.W., Markovics, G., 1980. Chemical processes affecting alkalis and alkaline earths during continental weathering. Geochimica et Cosmochimica Acta 44, 1659-1666.

33

ACCEPTED MANUSCRIPT Nesbitt, H.W., Markovics, G., 1997. Weathering of granodioritic crust, long-term storage of elements in weathering profiles, and petrogenesis of siliciclastic sediments. Geochimica et Cosmochimica Acta, 61, 1653-1670.

T

Nesbitt, H.W., Wilson, R.E., 1992. Recent chemical weathering of basalts. American Journal of

IP

Science 292, 740-777.

element chemistry of lutites. Nature 299, 715-717.

SC R

Nesbitt, H.W., Young, G.M., 1982. Early Proterozoic climates and plate motions inferred from major

Nesbitt, H.W., Young, G.M., 1984. Prediction of some weathering trends of plutonic and volcanic

NU

rocks based on thermodynamic and kinetic considerations. Geochimica et Cosmochimica Acta 48, 1523-1534.

MA

Nesbitt, H.W., Young, G.M., McLennan, S.M., Keays, R.R., 1996. Effects of chemical weathering and sorting on the petrogenesis of siliciclastic sediments, with implications for provenance studies.

D

The Journal of Geology, 525-542.

TE

Nielsen, A.T., Schovsbo, N.H., 2007. Cambrian to basal Ordovician lithostratigraphy in southern Scandinavia. Bulletin of the Geological Society of Denmark 53, 47-92.

CE P

Nielsen, A.T., Schovsbo, N.H., 2011. The Lower Cambrian of Scandinavia: Depositional environment, sequence stratigraphy and palaeogeography. Earth-Science Reviews 107, 207-310. Obst, K., Hammer, J., Katzung, G., Korich, D., 2004. The Mesoproterozoic basement in the southern

AC

Baltic Sea: insights from the G 14-1 off-shore borehole. International Journal of Earth Science (Geologische Rundschau) 93, 1-12. Platou, S.W., 1970. The Svaneke granite complex and the gneisses on East Bornholm. Bulletin of the Geological Society of Denmark 20, 93-133. Poulsen, V., 1978. The Precambrian-Cambrian boundary in parts of Scandinavia and Greenland. Geological Magazine 115, 131-136. Preston, J., Hartley, A., Hole, M., Buck, S., Bond, J., Mange, M., Still, J., 1998. Integrated whole-rock trace element geochemistry and heavy mineral chemistry studies; aids to the correlation of continental red-bed reservoirs in the Beryl Field, UK North Sea. Petroleum Geoscience 4, 7-16.

34

ACCEPTED MANUSCRIPT Price, R.C., Gray, C.M., Wilson, R.E., Frey, F.A., Taylor, S.R., 1991. The effects of weathering on rare-earth element, Y and Ba abundances in Tertiary basalts from southeastern Australia. Chemical Geology 93, 245-265.

IP

granitic terrain. Earth and Planetary Science Letters 301, 521-530.

T

Rasmussen, C., Brantley, S., 2011. Strong climate and tectonic control on plagioclase weathering in

SC R

Roaldset, E., 1973. Rare earth elements in Quaternary clays of the Numedal area, southern Norway. Lithos 6, 349-372.

cycle. Geochemistry International USSR 4, 1.

NU

Ronov, A.B., Balashov, Y.A., Migdisov, A.A., 1967. Geochemistry of rare earths in sedimentary

Roser, B.P., Korsch, R.J., 1988. Provenance signatures of sandstone-mudstone suites determined

MA

using discriminant function analysis of major-element data. Chemical geology 67, 119-139. Schau, M., Henderson, J.B., 1983. Archean chemical weathering at three localities on the Canadian

D

Shield. Precambrian Research 20, 189-224.

TE

Shannon, R.T., 1976. Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallographica Section A: Crystal Physics, Diffraction,

CE P

Theoretical and General Crystallography 32, 751-767. Sun, S.S., McDonough, W., 1989. Chemical and isotopic systematics of oceanic basalts: implications

313-345.

AC

for mantle composition and processes. Geological Society, London, Special Publications 42,

Surlyk, F. 1980. Denmark. In Geology of the European countries, Denmark, Finland, Iceland, Norway, Sweden, 1-50. Dunod. (Published in cooperation with the Comité National Francais de Géologie (C.N.F.G.) on the occasion of the 26th International Geological Congress). Svendsen, J.B., Hartley, N.R., 2002. Synthetic heavy mineral stratigraphy: applications and limitations. Marine and Petroleum Geology 19, 389-405. Taylor, S.R., Hallberg, J.A., 1977. Rare-earth elements in the Marda calc-alkaline suite: an Archaean geochemical analogue of Andean-type volcanism. Geochimica et Cosmochimica Acta 41, 1125-1129. Taylor. S.R. and McLennan. S.M., 1985. The Continental Crust: Its Composition and Evolution. Blackwcll. Oxford, pp. 312. 35

ACCEPTED MANUSCRIPT van de Kamp, P.C., 2016. Potassium Distribution and Metasomatism In Pelites and Schists: How and When, Relation To Postdepositional Events. Journal of Sedimentary Research 86, 683-711. Vejbæk, O.V., Stouge, S., Damtoft Poulsen, K., 1994. Palaeozoic tectonic and sedimentary evolution

T

and hydrocarbon prospectivity in the Bornholm area. Danmarks Geologiske Undersøgelse A34, 23.

IP

Waight, T.E., Frei, D., Storey, M., 2012. Geochronological constraints on granitic magmatism,

SC R

deformation, cooling and uplift on Bornholm, Denmark. Geological Society of Denmark Bulletin 60, 23-46.

White, A.F., Blum, A.E., 1995. Effects of climate on chemical weathering in watersheds. Geochimica

NU

et Cosmochimica Acta 59, 1729-1747.

White, A.F., Blum, A.E., Bullen, T.D., Vivit, D.V., Schulz, M., Fitzpatrick, J., 1999. The effect of

MA

temperature on experimental and natural chemical weathering rates of granitoid rocks. Geochimica et Cosmochimica Acta 63, 3277-3291.

D

White, A.F., Bullen, T.D., Schulz, M.S., Blum, A.E., Huntington, T.G., Peters, N.E., 2001.

TE

Differential rates of feldspar weathering in granitic regoliths. Geochimica et Cosmochimica Acta 65, 847-869.

CE P

Wildeman, T.E., Condie, K.C., 1973. Rare earths in Archean graywackes from Wyoming and from the Fig Tree Group, South Africa. Geochimica et Cosmochimica Acta 37, 439-453. Wronkiewicz, D.J., Condie, K.C., 1990. Geochemistry and mineralogy of sediments from the

AC

Ventersdorp and Transvaal Supergroups, South Africa: cratonic evolution during the early Proterozoic. Geochimica et Cosmochimica Acta 54, 343-354. Wronkiewicz, D.J., Kent, C.C., 1989. Geochemistry and provenance of sediments from the Pongola Supergroup, South Africa: evidence for a 3.0-Ga-old continental craton. Geochimica et Cosmochimica Acta, 53, 1537-1549. Zariņš, K., Johansson, Å., 2009. U–Pb geochronology of gneisses and granitoids from the Danish island of Bornholm: new evidence for 1.47–1.45 Ga magmatism at the southwestern margin of the East European Craton. International Journal of Earth Sciences 98, 1561-1580.

36

ACCEPTED MANUSCRIPT Figure captions: Fig. 1 Regional geology of Bornholm. (A) Location of Bornholm in the transition between the Sorgenfrei-Tornquist Zone and the Teisseyre-Tornquist Zone (Vejbæk et al., 1994). Abbreviations:

T

RG-Rønne Graben; B-Borggård borehole; G-G14-1 borehole; L-Loissin1 borehole; N-Nye Frederiks

(Callisen, 1934; Waight et al., 2012).

SC R

IP

quarry; (B) Location of Bornholm on East European Craton; (C) Local geology of Bornholm

Fig. 2 Lithology for the Borggård borehole, Bornholm, which is the stratotype for the Nexø Formation (including the Gadeby and Langeskanse members) and the Hadeborg Member of the

NU

Hardeberga Formation (Nielsen and Schovsbo, 2007).

Fig. 3 (A, B) Hand specimens of sandstone and basement granite from the Borggård borehole; (C)

MA

Back Scatter Electron (BSE) image of the basement granite from the Borggård borehole, in which anatase and hematite occur as fillings in the dissolved minerals. Sample Bo 311.5 m; (D) Dissolution

D

of feldspar in the fragment of graphic granite. BSE image, Sample Bo-316.0 m; (E) Mineralogical

TE

composition of sandstone of Gadeby Member, Sample Bo 309.35 m. Hematite occurs as fillings in the porosity of quartz and around the mineral boundary. Quartz overgrowths were observed; (F)

CE P

Hematite fillings (pseudomorphs) that have kept the shape of the primary mineral in the Gadeby Member. Illite occurs as cement and was dissolved, with remnants attached to the detrital grain surface. Sample Bo 304 m; (G) Composition of the basal sandstone of the Gadeby Member close to

AC

the basement granite, Sample Bo 309.35 m. The grains are coarse and angular. Hematite and anatase intergrow with each other; (H) Composition of sandstone higher in the Gadeby Member, sample Bo 304 m. An intensely weathered plagioclase is observed, with quartz and illite as remnants. Abbreviations: G-basement granite; S-sandstone; P-Pegmatite; Q-Quartz; Qz-Quartz overgrowth; Mc-Microcline; Bi-Biotite; I-Illite; B-Brookite; H-Hematite; HD-Hematite dispersion; HR-Hematite rims; An-Anatase; HF-Hematite replaced feldspar; Pd-Pressure dissolution; C-clays. Fig. 4 XRD patterns for the basement granite (A, B) and the Gadeby Member (C, D, E) from the Borggård borehole, Bornholm. Fig. 5 Quartz-Feldspar-Lithics (QFL) diagram of the Gadeby Member from the Borggård borehole. Fig. 6 Covariations between major elements and weathering indices for the rocks discussed in this study. 37

ACCEPTED MANUSCRIPT Fig. 7 Chondrite-normalized REE patterns of the (A) basement granite from the Borggård and G14-1 boreholes; (B) Gadeby Member in Borggård borehole and at Nye Frederiks quarry on Bornholm, the Adlergrund Konglomerat Member from the G14-1 borehole and the Lubmin Sandstein Formation

IP

Chondrite-normalized data from Sun and McDonough (1989).

T

from the Loissin-1 borehole; (C) regional granitoids on Bornholm (Johanssen et al., 2016).

SC R

Fig. 8 Upper Continental Crust (UCC) normalized trace-element compositions of (A) basement granite from the Borggård and G14-1 boreholes; (B) regional granitoids on Bornholm (Johanssen et al., 2016); (C) Gadeby Member in the Borggård borehole and at Nye Frederiks quarry on Bornholm,

NU

the Adlergrund Konglomerat Member from the G14-1 borehole and the Lubmin Sandstein Formation from the Loissin-1 borehole. Upper Continental Crust data are from McLennan, 2001.

MA

Fig. 9 Discrimination diagrams of the primary rock of weathered basement granite from the Borggård borehole. (A) (Gd/Yb)N vs. SiO2; (B) Eu anomaly vs. SiO2; (C) Nb/Ta vs. Zr/Hf. Dash arrows in (A)

D

and (B) are the decreasing trends of (Gd/Yb)N and Eu with increasing SiO2 contents. Different shades

TE

of grey in (C) constrain the ranges of Nb/Ta and Zr/Hf ratios for the basement granite from the Borggård borehole (green dots). Data are normalized to the chondrite composition. The intermediate

CE P

and felsic groups are according to Johansson et al. (2016). Fig. 10 Interpretation diagrams of the provenance of sandstone samples. The samples include the Gadeby Member in the Borggård borehole and at the Nye Frederiks quarry on Bornholm, the

AC

Adlergrund Konglomerat Member from the G14-1 borehole and the Lubmin Sandstein Formation from the Loissin-1 borehole. (A) Binary diagram of (La/Yb)N vs. Eu anomaly; (B) Binary diagram of Nb/Ta vs. SiO2; (C) Binary diagram of Zr/Hf vs. SiO2. Areas of dark grey and light grey defines the values of intermediate and felsic groups of Bornholm granitoids, respectively (Johanssen et al., 2016). Fig. 11 A-CN-K ternary diagram of the rocks included in this study. A=Al2O3; CN= (CaO*+Na2O); K=K2O (in molar proportions). CIA=Chemical Index of Alteration. Mineral compositions: Pl-Plagioclase; Ksp-K-feldspar; Ka-Kaolinite; Gi-Gibbsite; Sm-Smectite; Il-Illite, after Nesbitt and Young (1984). Fig. 12 Flow-chart showing reconstructed geochemistry of weathering regimes of the basement granites from the Borggård borehole on Bornholm. 38

ACCEPTED MANUSCRIPT Table captions: Table 1 The petrological characteristics of the granitoids on Bornholm (Micheelsen, 1961) and the basement granites from the Borggård borehole (in weight percent).

IP

T

Table 2 Tau (τ) values for the weathered basement granites from the Borggård borehole.

AC

CE P

TE

D

MA

NU

SC R

Appendix 1. Major and trace element compositions and calculated indices of the rocks for this study

39

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 1

40

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 2 41

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 3 42

AC

Figure 4

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

43

AC

Figure 5

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

44

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 6

45

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 7

46

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 8

47

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 9 48

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 10 49

AC

Figure 11

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

50

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 12

51

ACCEPTED MANUSCRIPT Table 1 Mineral compositions of Bornholm gneisses and granites (after Micheelsen, 1961) and the basement granite from the Borggård borehole (in weight percent) Hornblen de 5 2 8

Biotit e 7 6 7

Min or 5 3 2

Hypersthe ne 0 0 0

Diopsi de 0 0 0

29

30

21

10

5

5

+

+

33 41 36 55

22 18 26 2

27 33 25 30

6 4 7 7

5 3 3 6

0 0 0 0

0 0 0 0

SC R 5 1 2 0

T

Quart z 25 30 23

IP

Plagiocla se 24 22 25

MA

Dark gneiss Light gneiss Paradisb migmatite, dark Rønne granitic diorite Vang granitic diorite Hammer granite Svaneke granite Basement granite from the Borggård borehole

K-feldsp ar 35 38 35

NU

Rock type

AC

CE P

TE

D

Minor: Ti-magnetite, titanite, apatite, epidote, fluorite, allanite, hematite, anatase, pseudobrookite; +: accessory amounts.

52

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Table 2 Tau (τ) values for the weathered basement granite from the Borggård borehole Sample Rock type τ (Ti, Si) τ (Ti, Al) τ (Ti, Fe) τ (Ti, Ca) τ (Ti, Mg) Bo-310.6 m −0.35 −0.07 0.10 −0.83 −0.79 Bo-311.5 m −0.34 −0.10 0.02 −0.90 −0.75 basement Bo-313.15 m −0.32 −0.11 0.00 −0.89 −0.74 granite from Bo-313.7 m −0.22 −0.02 0.10 −0.70 −0.68 Borggård Bo-314.2 m −0.23 −0.04 0.09 −0.92 −0.67 borehole Bo-315.1 m −0.42 −0.23 0.15 −0.96 −0.70 Bo-316.0 m −0.33 −0.15 −0.05 −0.96 −0.66

53

τ (Ti, Na) −0.98 −0.98 −0.98 −0.98 −0.97 −0.98 −0.98

τ (Ti, K) −0.06 −0.04 0.02 0.14 0.11 −0.12 0.04

ACCEPTED MANUSCRIPT Research Highlights:

(3) The weathering regime of Precambrian granites is illustrated.

IP

(2) The fresh equivalent of a weathered basement granite is defined.

T

(1) The genetic relationship between granitoids and sediments is clarified.

SC R

(4) Mass transfer during the weathering of granite is determined quantitatively.

AC

CE P

TE

D

MA

NU

(5) The Precambrian climate is interpreted from the weathering processes.

54