Geological and Atmospheric Sciences Publications Geological and Atmospheric Sciences

1-3-2019 Pervasively anoxic surface conditions at the onset of the Great Oxidation Event: new multi-proxy constraints from the Cooper Lake paleosol Michael G. Babechuk University of Tübingen

Nadine Weimar University of Tübingen

Ilka C. Kleinhanns University of Tübingen

Suemeyya Eroglu Iowa State University

Elizabeth D. Swanner IFoowlalo Swta tthie Usn iaverndsit ay,dd eswitaionnnealr@i wasorktatse .aedut: https://lib.dr.iastate.edu/ge_at_pubs Part of the Atmospheric Sciences Commons, Geochemistry Commons, and the Oceanography See next page for additional authors Commons The ompc lete bibliographic information for this item can be found at https://lib.dr.iastate.edu/ ge_at_pubs/264. For information on how to cite this item, please visit http://lib.dr.iastate.edu/ howtocite.html.

This Article is brought to you for free and open access by the Geological and Atmospheric Sciences at Iowa State University Digital Repository. It has been accepted for inclusion in Geological and Atmospheric Sciences Publications by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Pervasively anoxic surface conditions at the onset of the Great Oxidation Event: new multi-proxy constraints from the Cooper Lake paleosol

Abstract Oceanic element inventories derived from marine sedimentary rocks place important constraints on oxidative continental weathering in deep time, but there remains a scarcity in complementary observations directly from continental sedimentary reservoirs. This study focuses on better defining continental weathering conditions near the Archean-Proterozoic boundary through the multi-proxy (major and ultra-trace element, Fe and Cr stable isotopes, μ-XRF elemental mapping, and detrital zircon U-Pb geochronology) investigation of the ca. 2.45 billion year old (giga annum, Ga) Cooper Lake paleosol (saprolith), developed on a sediment- hosted mafic dike within the Huronian Supergroup (Ontario, Canada).

Throughout the variably altered Cooper Lake saprolith, ratios of immobile elements (Nb, Ta, Zr, Hf, Th, Al, Ti) are constant, indicating a uniform pre-alteration dike composition, lack of extreme pH weathering conditions, and no major influence from ligand-rich fluids during weathering or burial metasomatism/ metamorphism. The loss of Mg, Fe, Na, Sr, and Li, a signature of albite and ferromagnesian silicate weathering, increases towards the top of the preserved profile (unconformity) and dike margins. Coupled bulk rock behaviour of Fe-Mg-Mn and co-localization of Fe- Mn in clay minerals (predominantly chlorite) indicates these elements were solubilized primarily in their divalent state without Fe/Mn-oxide formation. A lack of a Ce anomaly and immobility of Mo, V, and Cr further support pervasively anoxic weathering conditions. Subtle U enrichment is the only geochemical evidence, if primary, that could be consistent with oxidative element mobilization. The leaching of ferromagnesian silicates was accompanied by variable mobility and depletion of transition metals with a relative depletion order of Fe≈Mg≈Zn>Ni>Co>Cu (Cu being significantly influenced by secondary sulfide formation). Mild enrichment of heavy Fe isotopes (δ56/54Fe from 0.169 to 0.492 ‰) correlating with Fe depletion in the saprolith indicates loss of isotopically light aqueous Fe(II). Minor REE+Y fractionation with increasing alteration intensity, including a decreasing Eu anomaly and Y/Ho ratio, is attributed to albite breakdown and preferential scavenging of HREE>Y by clay minerals, respectively. Younger metasomatism resulted in the addition of several elements (K, Rb, Cs, Be, Tl, Ba, Sn, In, W), partly or wholly obscuring their earlier paleo-weathering trends.

The behavior of Cr at Cooper Lake can help test previous hypotheses of an enhanced, low pH-driven continental weathering flux of Cr(III) to marine reservoirs between ca. 2.48-2.32 Ga and the utility of the stable Cr isotope proxy of Mn-oxide induced Cr(III) oxidation. Synchrotron μ- XRF maps and invariant Cr/ Nb ratios reveal complete immobility of Cr despite its distribution amongst both clay-rich groundmass and Fe-Ti oxides. Assuming a pH-dependent, continental source of Cr(III) to marine basins, the Cr immobility at Cooper Lake indicates either that signatures of acidic surface waters were localized to uppermost and typically unpreserved regolith horizons or were geographically restricted to acid-generating point sources. However, in given detrital pyrite preservation in fluvial sequences overlying the paleosol, we propose that the oxidative sulphide corrosion required to drive surface pH(δ53/52Cr: -0.321 ± 0.038 ‰, 2sd, n=34) that cannot be linked to Cr(III) oxidation and is instead interpreted to have a magmatic origin.

The ombc ined chemical signatures and continued preservation of detrital pyrite/uraninite indicate low atmospheric O2 during weathering at ca. 2.45 Ga preserved in the rift-related sedimentary rocks of the Lower Huronian. The quea ous flux from the reduced weathering of mafic ockr s was characterized by a greater abundance of transition metals (Fe, Mn, Zn, Co, Ni) with isotopically light Fe(II), as well as higher Eu/Eu* and Y/Ho. In most models of Precambrian ocean element inventories, hydrothermal fluids are viewed as the

This article is available at Iowa State University Digital Repository: https://lib.dr.iastate.edu/ge_at_pubs/264 main supplier of several metals (e.g., Fe, Zn), although the results herein suggest that a riverine metal supply may have been substantial and that using Eu-excess as a strict proxy for hydrothermal flux may be misleading in near-shore marine sedimentary environments.

Keywords paleosol, Cooper Lake, Huronian Supergroup, stable Cr isotopes, stable Fe isotopes, redox-sensitive trace elements, anoxic weathering, atmospheric oxygenation, U-Pb detrital zircon geochronology

Disciplines Atmospheric Sciences | Geochemistry | Oceanography

Comments This is a manuscript of an article published as Babechuk, Michael G., Nadine E. Weimar, Ilka C. Kleinhanns, Suemeyya Eroglu, Elizabeth D. Swanner, Gavin G. Kenny, Balz S. Kamber, and Ronny Schoenberg. "Pervasively anoxic surface conditions at the onset of the Great Oxidation Event: new multi-proxy constraints from the Cooper Lake paleosol." Precambrian Research (2019). doi: 10.1016/j.precamres.2018.12.029. Posted with permission.

Creative Commons License

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Authors Michael G. Babechuk, Nadine Weimar, Ilka C. Kleinhanns, Suemeyya Eroglu, Elizabeth D. Swanner, Gavin G. Kenny, Balz S. Kamber, and Ronny Schoenberg

This article is available at Iowa State University Digital Repository: https://lib.dr.iastate.edu/ge_at_pubs/264 Accepted Manuscript

Pervasively anoxic surface conditions at the onset of the Great Oxidation Event: new multi-proxy constraints from the Cooper Lake paleosol

Michael G. Babechuk, Nadine E. Weimar, Ilka C. Kleinhanns, Suemeyya Eroglu, Elizabeth D. Swanner, Gavin G. Kenny, Balz S. Kamber, Ronny Schoenberg

PII: S0301-9268(18)30321-8 DOI: https://doi.org/10.1016/j.precamres.2018.12.029 Reference: PRECAM 5251

To appear in: Precambrian Research

Received Date: 9 June 2018 Revised Date: 22 December 2018 Accepted Date: 27 December 2018

Please cite this article as: M.G. Babechuk, N.E. Weimar, I.C. Kleinhanns, S. Eroglu, E.D. Swanner, G.G. Kenny, B.S. Kamber, R. Schoenberg, Pervasively anoxic surface conditions at the onset of the Great Oxidation Event: new multi-proxy constraints from the Cooper Lake paleosol, Precambrian Research (2018), doi: https://doi.org/10.1016/ j.precamres.2018.12.029

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. Pervasively anoxic surface conditions at the onset of the Great Oxidation Event: new multi-proxy constraints from the Cooper Lake paleosol

Michael G. Babechuka-c*, Nadine E. Weimara, Ilka C. Kleinhannsa, Suemeyya Eroglua,d, Elizabeth D. Swannerd, Gavin G. Kennyb,e, Balz S. Kamberb, Ronny Schoenberga aIsotope Geochemistry Group, Department of Geosciences, University of Tübingen, Tübingen, Germany bDepartment of Geology, Trinity College Dublin, Dublin, Ireland cDepartment of Earth Sciences, Memorial University of Newfoundland, St. John’s, Canada dDepartment of Geological & Atmospheric Sciences, Iowa State University, Ames, USA eDepartment of Geosciences, Swedish Museum of Natural History, Stockholm, Sweden

*Corresponding author, present address: Department of Earth Sciences, Memorial University of Newfoundland, St. John’s, Canada. Contact information – email: [email protected]; phone: 1-709-864-6095 Abstract

Oceanic element inventories derived from marine sedimentary rocks place important constraints on oxidative continental weathering in deep time, but there remains a scarcity in complementary observations directly from continental sedimentary reservoirs. This study focuses on better defining continental weathering conditions near the Archean-Proterozoic boundary through the multi-proxy (major and ultra-trace element, Fe and Cr stable isotopes, µ-XRF elemental mapping, and detrital zircon U-Pb geochronology) investigation of the ca. 2.45 billion year old (giga annum, Ga) Cooper Lake paleosol (saprolith), developed on a sediment-hosted mafic dike within the Huronian Supergroup (Ontario, Canada).

Throughout the variably altered Cooper Lake saprolith, ratios of immobile elements (Nb, Ta, Zr, Hf, Th, Al, Ti) are constant, indicating a uniform pre-alteration dike composition, lack of extreme pH weathering conditions, and no major influence from ligand-rich fluids during weathering or burial metasomatism/metamorphism. The loss of Mg, Fe, Na, Sr, and Li, a signature of albite and ferromagnesian silicate weathering, increases towards the top of the preserved profile (unconformity) and dike margins. Coupled bulk rock behaviour of Fe-Mg-Mn and co-localization of Fe- Mn in clay minerals (predominantly chlorite) indicates these elements were solubilized primarily in their divalent state without Fe/Mn-oxide formation. A lack of a Ce anomaly and immobility of Mo, V, and Cr further support pervasively anoxic weathering conditions. Subtle U enrichment is the only geochemical evidence, if primary, that could be consistent with oxidative element mobilization. The leaching of ferromagnesian silicates was accompanied by variable mobility and depletion of transition metals with a relative depletion order of Fe≈Mg≈Zn>Ni>Co>Cu (Cu being significantly influenced by secondary sulfide formation). Mild enrichment of heavy Fe isotopes (δ56/54Fe from 0.169 to 0.492 ‰) correlating with Fe depletion in the saprolith indicates loss of isotopically light aqueous Fe(II). Minor REE+Y fractionation with increasing alteration intensity, including a decreasing Eu anomaly and Y/Ho ratio, is attributed to albite breakdown and preferential scavenging of HREE>Y by clay minerals, respectively. Younger metasomatism resulted in the addition of several elements (K, Rb, Cs, Be, Tl, Ba, Sn, In, W), partly or wholly obscuring their earlier paleo-weathering trends.

The behavior of Cr at Cooper Lake can help test previous hypotheses of an enhanced, low pH-driven continental weathering flux of Cr(III) to marine reservoirs between ca. 2.48-2.32 Ga and the utility of the stable Cr isotope proxy of Mn-oxide induced Cr(III) oxidation. Synchrotron µ- XRF maps and invariant Cr/Nb ratios reveal complete immobility of Cr despite its distribution amongst both clay-rich groundmass and Fe-Ti oxides. Assuming a pH-dependent, continental source of Cr(III) to marine basins, the Cr immobility at Cooper Lake indicates either that signatures of acidic surface waters were localized to uppermost and typically unpreserved regolith horizons or were geographically restricted to acid-generating point sources. However, in given detrital pyrite preservation in fluvial sequences overlying the paleosol, we propose that the oxidative sulphide corrosion required to drive surface pH<4 lagged behind in the region relative to the paleo-environment captured in other early Proterozoic sequences. The entire saprolith exhibits a consistently light stable Cr isotope composition (δ53/52Cr: -0.321 ± 0.038 ‰, 2sd, n=34) that cannot be linked to Cr(III) oxidation and is instead interpreted to have a magmatic origin. The combined chemical signatures and continued preservation of detrital pyrite/uraninite indicate low atmospheric O2 during weathering at ca. 2.45 Ga preserved in the rift-related sedimentary rocks of the Lower Huronian. The aqueous flux from the reduced weathering of mafic rocks was characterized by a greater abundance of transition metals (Fe, Mn, Zn, Co, Ni) with isotopically light Fe(II), as well as higher Eu/Eu* and Y/Ho. In most models of Precambrian ocean element inventories, hydrothermal fluids are viewed as the main supplier of several metals (e.g., Fe, Zn), although the results herein suggest that a riverine metal supply may have been substantial and that using Eu-excess as a strict proxy for hydrothermal flux may be misleading in near-shore marine sedimentary environments.

Keywords: paleosol, Cooper Lake, Huronian Supergroup, stable Cr isotopes, stable Fe isotopes, redox-sensitive trace elements, anoxic weathering, atmospheric oxygenation, U-Pb detrital zircon geochronology 1. Introduction

Paleosols are important geological archives that provide the only means to constrain the signatures of ancient continental weathering processes at their source, offering a complementary perspective to paleo-environmental conditions reconstructed from marine sedimentary rocks. Precambrian paleosols are most widely recognized as an archive of Earth’s atmospheric oxygenation history, a legacy established from the pioneering work of H.D. Holland. This work found bona fide or ‘definite’ paleosols Rye and Holland (1998) to preserve a temporal evolution from Fe(II) loss or redistribution (≥2.45 Ga; Macfarlane et al., 1994a; Utsunomiya et al., 2003; Yang et al., 2002) to partial Fe(II) oxidation (2.45 to 2.25 Ga; Yang and Holland, 2003; see Beukes et al., 2002 for an alternative view) to near-complete oxidation of Fe(II) (<2.25 Ga; e.g., Babechuk and Kamber, 2013; Holland et al., 1989; Zbinden et al., 1988), and helped to lay the foundation for a unidirectional and single-step vision of the ‘Great Oxidation Event’ (GOE) (see summaries in Holland, 2011; Holland, 1984; Rye and Holland, 1998). Despite some documented variability associated with paleosol protolith composition and/or local hydrological conditions, whereby minor oxidation of Fe(II) is found in some low-Fe protoliths ≥2.45 Ga (E.G-Farrow and Mossman, 1988; Sutton and Maynard, 1993), this temporal evolution of paleosol Fe oxidation is consistent with sedimentary mineralogical data. The disappearance of detrital uraninite and pyrite in fluvial-alluvial sedimentary rocks and corresponding emergence of Fe-oxide-bearing red beds (Chandler, 1980; Johnson et al., 2014; Roscoe, 1973; Zhou et al., 2017) overlaps with the 2.45-2.25 Ga transition in paleosol Fe speciation. Further, the very strong reduction in the magnitude of mass-independent sulphur isotope fractionation preserved in sedimentary sulfate and sulfide (Bekker et al., 2004; Papineau et al., 2007; Pavlov and Kasting, 2002), widely accepted as fingerprinting the accumulation of free atmospheric oxygen through the development of stable stratospheric ozone, also occurred in this time interval.

Over the last two decades, refining reconstructions of the timing and duration of the GOE, as well as understanding the underlying cause(s) and impacts of sustained, free atmospheric O2 on Earth’s other systems have remained active areas of research. Most of these focused research efforts have been on marine chemical sedimentary rocks (see Lyons et al., 2014; Robbins et al., 2016; and the references therein), which, with the aid of new geochemical proxies, have revealed a significantly more nuanced model of Precambrian continental redox evolution. Marine authigenic enrichments and stable isotopic fractionations of elements that are only significantly solubilized as an oxyanion under oxic and near- neutral pH conditions (e.g., Cr, Mo, Se, U, Re) offer several refinements to the classic view of the GOE: (1) marine oxygen oases and possible temporally and/or spatially restricted continental oxidative weathering pre-dated the GOE, with well-documented evidence in the Neoarchean (Anbar et al., 2007; Eroglu et al., 2015; Kurzweil et al., 2016; Kurzweil et al., 2015; Wille et al., 2007) and some evidence dating as far back as ca. 2.9 Ga (Bau and Alexander, 2009; Crowe et al., 2013; Eickmann et al., 2018; Planavsky et al., 2014a) or earlier (Frei et al., 2016; Rosing and Frei,

2004; Satkoski et al., 2015); (2) overlapping authigenic Cr-Mo-U enrichments at ca. 2.5 Ga point to incipient stages of free atmospheric O2 accumulation in continental environments leading up to the traditionally recognized GOE (Kendall et al., 2015; Kendall et al., 2010; Konhauser et al., 2011; Partin et al., 2013a; Partin et al., 2013b; Reinhard et al., 2009); and (3) highly dynamic post-GOE Proterozoic redox conditions with possible spikes (e.g., Lomagundi-Jatuli Event) and declines (Bekker and Holland, 2012; Canfield et al., 2013; Frei et al., 2009; Partin et al., 2013a; Partin et al., 2013b) giving way to low levels after ca. 2.1 Ga that were arguably sustained throughout most of the Mesoproterozoic (Cole et al., 2016; Planavsky et al., 2014b). Importantly, apart from the minor Fe oxidation reported in some Archean and early Proterozoic paleosols developed on low-Fe protoliths (e.g., granite, sandstone; Grandstaff et al., 1986; Sutton and Maynard, 1993), atmospheric O2 fluctuations and associated oxidative weathering fluxes remain poorly delineated in the geochemistry of continental deposits.

Paleosols are imperfect mineralogical and geochemical archives, requiring meticulous care in separating pedogenic from post-depositional signatures (e.g., Rainbird et al., 1990) and providing only a limited temporal and spatial resolution of paleo-environmental conditions. Even cases meeting all criteria outlined by Rye and Holland (1998) as a “definite paleosol” (see Section 2.2), rather than another type of alteration horizon, can show a complex trace element and isotopic geochemistry that is difficult to disentangle (e.g., Babechuk et al., 2017). Moreover, there remains a long-standing debate as to whether the aforementioned Fe depletion preserved in paleosols ≥2.45 Ga is primary and indicative of anoxic atmospheric conditions or instead reflects the vulnerability of pedogenic Fe(III)-oxides to later reduction by organic acids or fluids interacting with the weathered substrate during burial diagenesis/metamorphism (Ohmoto, 1997, 1996). Geochemical paleo-redox proxies largely established from the marine sedimentary record offer a new and much-needed test for oxidative weathering reactions in Precambrian paleosols that is independent from Fe speciation (e.g., Crowe et al., 2013) that needs to extend to more key paleosols. Beyond the redox- sensitive elements, several other geochemical proxies of chemical weathering processes have been developed from Phanerozoic profiles over the last two decades, but remain poorly tested on Precambrian analogues. The application of such proxies are needed to better understand ancient continental weathering fluxes with the ability to track the evolving bulk composition of the exposed crust (e.g., Ni, REE+Y; Kamber, 2010; Konhauser et al., 2009; Konhauser et al., 2015) or which played a role in influencing marine microbial evolution (e.g., P, Mo, Ni, Co; see Robbins et al., 2016 for a review). The importance of elemental supply from weathering of mafic-ultramafic rocks, which were more abundant volumetrically in the Archean-Paleoproterozoic (Condie, 1993), is increasingly recognized to have influenced the ancient hydrosphere- lithosphere-atmosphere systems (Beinlich et al., 2018; Kamber, 2010; Konhauser et al., 2009; Large et al., 2018).

Given that the Precambrian continental weathering environment still remains so poorly understood, it is apparent that further detailed studies applying trace metal and stable isotopic paleo-redox proxies are needed to enable more robust constraints on both the intervals of continental oxidative weathering and more accurate reconstructions of biogeochemical links between marine and subaerial continental reservoirs (Murakami et al., 2016). This study revisited the ca. 2.45 Ga Cooper Lake paleosol to provide new perspectives on the early Proterozoic surface environment. The formation of this paleosol is contemporaneous with the deposition of marine sedimentary rocks inferred to record the formation, release, and aqueous transport of oxyanion-forming elements (U, Mo, S) from the weathering of reduced minerals on the continents, as well as the temporary development of acid weathering conditions promoting transport of Cr(III) (Konhauser et al., 2011) prior to the proposed ca. 2.3 Ga MIF-S time stamp of the GOE (Bekker et al., 2004). The metal-rich mafic dike protolith is also ideal for examining the behaviour of both redox-sensitive and bio-essential trace elements and the role of mafic rock weathering in deep time (Konhauser et al., 2009). This paleosol resides stratigraphically near the base of the Huronian Supergroup – a ca. 2.45-2.25 Ga supracrustal succession that records the GOE transition (Roscoe, 1973) – amidst the siliciclastic units containing reduced detrital minerals (pyrite, uraninite) and showing MIF-S signatures. The aims were to investigate weathering conditions in a single continental environment at ca. 2.45 Ga in terms of: (1) using new paleo-redox proxy data (incl. elemental Mo, V, W, U) to test for preserved evidence of oxidative weathering; (2) using Cr elemental and isotopic data to test for any acid- driven leaching of Cr(III) or Cr(III) oxidation; (3) testing unconventional proxies of subaerial weathering reactions (e.g., release of transition metals, Y/Ho fractionation); and, (4) exploring connections between this and other paleosol signatures and those of the contemporaneous marine rock record.

The earliest studies at Cooper Lake found evidence for reducing weathering conditions (e.g., loss of Fe(II), lack of Ce anomaly; Sutton and Maynard, 1993; Utsunomiya et al., 2003) and proposed analogies to older, Archean paleosols. However, more recent data have suggested minor oxidation is recorded through the subtle mobilization of some redox-sensitive elements (U, Mo, Cr) (Murakami et al., 2016). This calls for further constraints from high-precision data to more thoroughly assess paleo-redox conditions. This study addresses this deficit through the sampling of a new suite of ~50 samples and the application of a wide array of analytical techniques (major element, trace element, ferrous iron, stable Fe and Cr isotopes, total C and S, µ-XRF). Collectively, these data provide several new insights into the conditions of the early Proterozoic surface environment and the continental fluxes to the ancient hydrosphere.

2. Geological background

2.1 Huronian Supergroup

The Huronian Supergroup (HS) is a Paleoproterozoic, largely sedimentary succession deposited on the Archean Superior Province as a rift-to-drift sequence (Young et al., 2001; Young, 2015 and the references therein). From base to top, the HS is divided into the Elliot Lake, Hough Lake, Quirke Lake, and Cobalt Groups. Within this depositional framework, the first three Groups represent early rift sequence deposits (informally considered the Lower Huronian) whereas the Cobalt Group (informally considered the Upper Huronian) marks passive margin sedimentation. The Elliot Lake Group is unique from the others in being the only sequence with volcanic effusive rocks (Thessalon Fm.) considered erosional remnants of continental flood basalts (Ketchum et al., 2013) associated with larger scale plume activity (Ciborowski et al., 2015; Heaman, 1997). Locally, Thessalon Fm. mafic units are intercalated with early rift related conglomerates and sandstones of the Livingstone Creek Fm. Above the uppermost sedimentary units of the Elliot Lake Group (Matinenda and McKim Fms.), the subsequent Groups are characterized by three upward fining siliciclastic sequences each beginning with a diamictite unit (Ramsey Lake, Bruce, and Gowganda Fms.). Due to the paucity of felsic volcanic rocks, absolute age constraints on the timing of sedimentation are limited. The maximum depositional age is tightly defined with a U-Pb thermal ionization mass spectrometry (TIMS) age of 2452.5 ± 6.2 Ma from the Copper Cliff rhyolite (Ketchum et al., 2013; see also (Krogh et al., 1984)), a felsic extrusive in the Thessalon Fm. volcanics. The upper age bracket for sedimentation is 2219 ± 4 Ma (Corfu and Andrews, 1986) provided by the Nippising diabase cross-cutting the HS. Rasmussen et al. (2013) dated zircon extracted from a micro-tuffaceous unit in the Gordon Lake Fm. of the Cobalt Group at ca. 2308 ± 8 Ma. If interpreted as a depositional age, it could further constrain timing of sedimentation (and aid significantly with correlating the Huronian units with similar Paleoproterozoic strata from South Africa (Bekker, 2014; Rasmussen et al., 2013), but a magmatic origin of the zircon is not universally accepted (Young, 2015).

Regardless of the gaps in absolute chronology, the most attractive feature of the HS in the context of this study is the well-established relative age transition from anoxic to oxic surface conditions recorded between the Lower and Upper Huronian. This transition has long been appreciated as a key expression of the GOE (Logan, 1857; Roscoe, 1973). The evidence for this includes paleo-redox indicators (Fe, Ce) in paleosols (Nedachi et al., 2005; Prasad and Roscoe, 1996), the loss of redox-sensitive placer minerals, the first appearance of red beds (Chandler, 1980), and the marked enrichment of Mn in the shallow marine environment (Sekine et al., 2011). Applying the transition to a loss of mass- independent sulfur fractionation as a time stamp of the GOE, the study by Papineau et al. (2007) placed the event between the first and second glacial units, although more recent data suggests it either occurred later or cannot be resolved at all within the HS (Cui et al., 2018) or that the GOE may not have been globally synchronous (Philippot et al., 2018).

2.2 Cooper Lake and other sub-Matinenda paleosols

The Cooper Lake paleosol is located in the Algoma District of northern Ontario, Canada, near the border between the Otter and Haughton Townships north of the town of Thessalon. Discovered by Gerald Bennett of the Ontario Geological Survey (Bennett, 1990), the paleosol is known exclusively from a small area of gravel road outcrops adjacent to Boundary Lake but takes its name from the nearby Cooper Lake (Figure 1). The outcrops reveal green-grey sandstone of the Livingstone Creek Fm. unconformably overlain by white-grey basal sandstone and matrix- supported polymict (quartz pebble-dominated and pyritic) conglomerate of the Matinenda Fm. In one outcrop, a mafic dike interpreted as a feeder for the Thessalon volcanic rocks (Bennett, 1991) cross-cuts Livingstone Creek Fm sandstone. There, the Livingstone Creek Fm. and Thessalon Fm. units both show evidence of chemical and mineralogical alteration consistent with paleo-weathering (Bennett, 1990; Sutton and Maynard, 1993; Utsunomiya et al., 2003) and are truncated by the Matinenda Fm. The contacts of the weathered dike with the Livingstone Creek Fm. show evidence of minor shearing and fault displacement. Following the terminology of Rainbird et al. (1990), the paleosols can be considered lithified equivalents of saprolite, referred to as “saprolith”, although they are extensively leached and mica-rich near the unconformity, suggesting a high degree of alteration proximal to the paleo-surface. The Cooper Lake paleosol is part of a larger suite of paleo-surfaces preserved along unconformities in the Elliot Lake Group (Bennett et al., 1991). Most of the prominent and well-documented paleosols are sub-Matinenda and are hosted on Archean basement ranging from felsic plutonic rocks to greenstone (Gall, 1992; Gay and Grandstaff, 1980; Kimberley et al., 1984; Mossman and Farrow, 1992; Prasad and Roscoe, 1996). The Cooper Lake paleosol formation post-dates the onset of sedimentation in the Lower Huronian, suggesting it either belongs to a separate paleo-weathering event from the paleosols developed on Archean rocks (Bennett et al., 1991) or that all sub-Matinenda weathering profiles developed contemporaneously, after sedimentation in localized areas (Young et al., 2001). The age of paleo-weathering at Cooper Lake is approximated as ca. 2.45 Ga (Rye and Holland, 1998) based on the U-Pb age of related felsic volcanic rocks (Ketchum et al., 2013; Krogh et al., 1984) and the rapid sedimentation inferred for the lower rift-related sequence (Long, 2004). Post-burial, the saprolith was subjected to greenschist facies metamorphism associated with the Penokean orogeny (Schulz and Cannon, 2007). Collectively, the documented sub- Matinenda paleosols, including Cooper Lake, meet the physical, mineralogical, and chemical criteria commonly used to identify paleosols from other types of alteration horizon (e.g., Grandstaff et al., 1986; Retallack, 1992; Rye and Holland, 1998). These criteria include in situ profile development on a homogeneous protolith, a combined depth-dependent mineralogical, textural, and chemical evolution of the protolith consistent with known processes of chemical weathering, and sedimentological evidence that the weathered substrate was soft during deposition of the overlying sediments. However, ambiguity remains for some greenstone-hosted paleosols located in the former Denison Mine east of the Cooper Lake paleosol near the town of Elliot Lake. Specifically, irregular immobile element geochemistry makes it difficult to confirm a homogeneous protolith composition (Gay and Grandstaff, 1980; Prasad and Roscoe, 1996; Rye and Holland, 1998).

3. Methods

3.1 Collection and preparation of samples

The outcrop exposure of the mafic-dike hosted paleosol is located at 46o31’46.95’’N, 83o32’46.18’’ W. Sampling depths were measured along the outcrop surface perpendicular to the Matinenda Fm. unconformity. Two parallel cuts into the outcrop spaced by ~10 cm were made with a rock saw to create a channel with a depth of ~10 cm from which samples were extracted with hammer and chisel. The full sampled transect extended to ~550 cm along the outcrop and comprised two slightly offset sections (Figure 1); the first section (samples CLA-002 to CLA-013) is closer to the southern edge of the dike, whereas the second, adjacent section is closer to the northern contact of the dike (CLB-015 to CLD-056), but gradually transitions closer to the centre with increasing depth. The samples most proximal to the northern dike margin, labelled as dike- north (D-N; CLB-015 to CLC-033), were separated from the remaining samples referred to as dike-south (D-S; CLA-002 to CLA-013 and CLC-034 to CLF-061) based on immobile element concentration profiles and their higher degree of alteration than samples at the same depth closer to the dike centre (see Section 4.2). Rounded regions of lighter green coloured paleosol (LCP) were identified at depths of 409 cm (CLC-045L) and 416 cm (CLC-048L) in the D-S section and were sub-sampled for comparison to the embaying dark coloured “bulk paleosol”. The extracted outcrop samples were immediately adjacent to those described and reported in Utsunomiya et al. (2003), but extended to greater depths by excavating overburden at the foot of the outcrop. The excavation revealed samples (CLE-057 to CLF-062) that were unusual in their visibly more altered composition (e.g., greater enrichment of mica) relative to the overlying paleosol, but their full contact relationship with the dike margin remained obscured by the overburden. Even with additional excavation, the paleosol does not extend to a visibly ‘least-altered’ composition. A chlorite-rich dike sample (CL-FD1) seemingly further from the contact with the Matinenda Fm. was extracted from a separate outcrop south of the main paleosol (Figure 1) and is treated as the least-altered reference point (henceforth referred to as offset dike, OD) unless appropriate previously published data from drillcore in the area was more suitable (Sutton and Maynard, 1992; Utsunomiya et al., 2003). A sample of Livingstone Creek Fm. sandstone paleosol (CL-LCSS), at a depth of ~3 m and immediately adjacent to the southern dike-sandstone contact, as well as samples of overlying sandstone (CL-MTSS) and conglomerate (CL-MTCG) were also taken.

Hand samples were split for thin sections and fist-sized samples were reduced to chips using a tungsten carbide jaw crusher. The chips were washed and picked under alcohol to reject any with remaining evidence of modern surface oxidation (e.g., rusting of pyrite grains) or transfer of material from the crusher plates, and pulverized in an agate ball mill for use in all bulk sample geochemical analyses. An aliquot of rock chips from the CL-LCSS and CL-MTCG samples were processed for U-Pb detrital zircon geochronology. The methods, results, and implications of these U-Pb data are peripheral to the main aims of the study and thus reported separately in the Supplementary Material.

3.2 Analytical methods

3.2.1 Synchrotron µXRF mapping

In situ elemental information was obtained by µXRF element mapping on beamline 10-2 at the Stanford Synchrotron Radiation Lighthouse (SSRL). Maps were collected on 30 µm thick polished thin sections from samples CLA-004, CLC-037, and CLD-056. The incident beam was focused on the sample position using a Si(111) monochromator and the XRF signals generated using a 50 µm × 50 µm spot size with a dwell time of 10 ms were collected using a Vortex silicon drift detector. Fluorescence maps for S, K, Ti, and Cr were first generated at 6800 eV and then merged with maps made at 11000 eV for Fe to avoid interference of the Fe Kα lines with the Mn Kβ line. Maps were deadtime corrected and merged in the Microanalysis Toolkit (Webb, 2011). A Gaussian filter was applied to the fluorescence pixels of each element using a full width half maximum of 0.85 pixel steps.

3.2.2 Bulk X-Ray diffraction

Bulk powder mineralogical information was collected without sample pre-treatment using a Bruker D5000 X-Ray diffraction (XRD) system in the

Department of Geology, Trinity College Dublin (TCD). The system used a Cu Kα source and data were collected from scans of 0 to 70° 2θ with 0.02° steps at 40 kV and 40 mA. The Bruker AXS Diffrac software was used in combination with the International Centre for Diffraction PDF database for phase matching. Semi-quantitative relative modal abundance data (Supplementary Table 3) were determined using the reference intensity ratios method.

3.2.3 Bulk rock LOI and XRF

Loss on ignition (LOI) was determined on dried (24 h at 105 °C) powder samples after 2 h at 1100 °C. Major element concentrations were determined using a wavelength dispersive Bruker AXS S4 Pioneer X-Ray fluorescence (XRF) spectrometer (Rh-tube at 4kW) in the Isotope Geochemistry Group, University of Tübingen. For analysis, fused glass beads were prepared from 1.5 g of a separate aliquot of dried sample powder mixed with 7.5 g MERCK Spectromelt A12 (mixture of 66 % Li-tetraborate and 34 % Li-metaborate) and melted at 1200 °C using an Oxiflux system from CBR analytical service. Calibration was performed based on 32 standardized samples (Potts and Webb, 1992) as compiled in

Govindaraju (1989). Analytical uncertainty and detection limits vary and depend on element and sample composition, although major element abundance uncertainties, expressed as 1 relative standard deviation (1rsd), are generally better than 1%.

3.2.4 Total C and S measurements

The total C (TC), N, and S of paleosol samples were measured from a powder aliquot of ~40 mg on a VARIO EL analyzer (Elementar Analysensysteme GmbH, Hanau, Germany) in the Soil Science and Geomorphology Group at the University of Tübingen. Sample powders were weighed into tin foil and analyzed using oxidative heat combustion at 1150 °C with WO3 as a catalyst. The limit of quantification for TC, N, and S were 0.1, 0.03, and 0.05 wt. %, respectively. All N levels in paleosol samples were below quantification. A high organic sediment reference material (OAS IVA 33802150) was measured in triplicate every 30 samples and used as an accuracy and precision test (e.g., results repeated if the 1rsd was greater than 5% in these triplicates). The long-term (n>700) 1rsd on TC, N, and S for this reference material in the laboratory are 1.3%, 2.5%, and 2.4%, respectively, which is adopted as an estimate of the method precision.

3.2.5 Ferrous iron measurements

The ferrous iron content of selected samples was determined using a rapid digestion and colourimetric titration similar to the Pratt Method (e.g., Maxwell, 1968). Powder aliquots ranging from 50 to 150 mg (depending on total Fe concentration) were digested with a mixture of 5 mL each of conc. HF, conc. H2SO4, and water, heated at 170 °C for 12 min. Upon digestion, the sample was transferred immediately into an

Erlenmeyer flask containing a mixture of sub-boiling water (50 mL), saturated H3BO3 (30 mL), and concentrated H3PO4 (6 mL) and topped up to - ~250 mL with sub-boiling water. The mixed solution was titrated volumetrically with 0.1 M KMnO4 , which was calibrated daily against 0.1 M oxalic acid. A measurement of blank and reference materials bracketed every 8-12 sample unknowns. Full procedural blanks were equivalent to ≤0.02 wt.% FeO. Eight measurements of United States Geological Survey (USGS) reference material BIR-1 returned a mean FeO wt.% of 8.25 ± 0.42 (1sd). The concentration agrees well with other reported FeO values in the literature that range from 8.2 to 8.5 wt. % (e.g., Saikkonen, 1993; Schuessler et al., 2008; and the references therein) and the ~5% (1rsd) is adopted as an estimate of the intermediate method precision. All samples were run in duplicate or triplicate if the duplicate FeO abundances deviated by more than 0.5 wt. %, such that all values reported in Table 1 represent a mean of 2-3 measurements.

3.2.6 Element analysis by ICP-QMS

All preparation stages for inductively coupled plasma quadrupole mass spectrometry (ICP-QMS) elemental analysis were undertaken in a clean laboratory at TCD. All acids used in the stages of cleaning and sample preparation were purified from reagent grade through triple-stage, sub- boiling distillation in an elbow-style still (Analab® CleanAcids purification system) and all diluted acids were prepared with ultra-pure water

(resistivity ≥18.2 MΩ) produced from a Millipore® Milli-Q system.

An aliquot of 100 (±5) mg of sample powder was digested over 3-4 days using a 4:1 concentrated HF-HNO3 mixture in sealed Savillex perfluoroalkoxy (PFA) beakers at ~120 oC and prepared for measurement using an analytical protocol similar to one applied previously at Laurentian University (Babechuk et al., 2010; Babechuk et al., 2015; Kamber, 2009; Marx and Kamber, 2010) but modified for analytical applications requiring low sample volumes (Hahn et al., 2015). All data were measured using a ThermoFisher Scientific iCap-Q at TCD with sample solution supplied through a 1 mL Teflon PFA loop at a rate of ~140 µL min-1 from an ESI SC-2 DX autosampler using a microFast syringe uptake system. The iCap-Q was operated in high-sensitivity (STDS) mode using Ni cones with a 2.8 mm skimmer cone insert. Processing of the instrument intensity data was performed offline and included internal standard and external monitor correction for drift and matrix effects, blank subtraction, and oxide/hydroxide/dimer interference correction (Eggins et al., 1997; Ulrich et al., 2010). Corrected intensities were calibrated against the mean of 6 analyses from 4 different digestions of the USGS dolerite reference material W-2a and applying the TCD laboratory’s preferred concentrations (reported in Supplementary Table 5). The mean instrument response from all six W-2a measurements was taken for all analytes apart those known to potentially exhibit a nugget effect from contamination in the supplied powder (e.g., Pb, Sn, Cd; Weis et al., 2006). In the latter cases, analyte intensities were statistically screened using a Q-test to remove a single W-2a digestion if it was anomalously higher than the others (i.e., an outlier). In addition to the trace element experiments, all stock solutions were diluted and measured separately to determine major element abundances (with the exception of Si) and cross-check the XRF concentrations (Section 3.2.4; Supplementary Table 4). The only element showing a significant bias between the methods is Mn with a systematic offset of ~15% higher concentrations in XRF vs. ICP-QMS data. The intermediate precision of the method is monitored using long-term measurement of USGS reference materials (e.g., BIR-1, BHVO-2, BCR-2, AGV-2). The data for AGV-2 and BHVO-2 analyzed from 2013-2016 along with comparisons to GeoReM data (Jochum et al., 2016) are reported in Supplementary Table 5. The 1rsd for most analytes is better than 3%, with precision decreasing for those at very low abundances or known to be heterogeneous in the reference material (e.g., Pb in BHVO-2). Additional trace element method details are reported in the Supplementary Material.

3.2.7 Stable Fe & Cr isotope analysis

All stable isotope preparation and measurement was performed at the University of Tübingen using the protocols described in Schoenberg and von Blanckenburg (2005) and Swanner et al. (2015) for Fe and Schoenberg et al. (2016) for Cr. One batch of samples (B1) was prepared for stable Cr isotope analysis only by adding a 50Cr-54Cr double-spike to sample powders prior to digestion. A second batch (B2) was prepared for combined

Fe-Cr stable isotope analysis by adding the 50Cr-54Cr spike after digestion and extraction of an aliquot mixed separately with a 57Fe-58Fe double- spike. The measurements of both stable Cr and Fe isotopic ratios were undertaken on a ThermoFisher Scientific Neptune Plus multi-collector ICP-MS (MC-ICP-MS). Additional analytical details are available in the Supplementary Material. The full procedural blanks for Fe and Cr were negligible relative to the amounts recovered for unknowns, negating the need for sample blank correction.

The Fe isotope ratios are reported relative to the isotopically certified reference material IRMM-014 in the δ-notation, with per mil δ56/54Fe 56 54 56 54 values determined from: [[( Fe/ Fe)sample/( Fe/ Fe)IRMM-014] – 1] x 1000. Throughout the duration of the study, the Han-Fe standard returned a mean δ56/54Fe of 0.296 ± 0.033 ‰ (2sd, n=6), agreeing well with previously published values from the laboratory (Kurzweil et al., 2016; Moeller et al., 2014; Swanner et al., 2017; Swanner et al., 2015; Wu et al., 2017), and the Tüb-Fe standard returned a mean δ56/54Fe value of -0.372 ± 0.032 ‰ (2sd, n=3), agreeing well with the longer term laboratory mean of -0.376 ± 0.049 ‰ (2sd, n=31).

The Cr isotope ratios are reported relative to the isotopically certified reference material NIST SRM979 in the δ-notation, with per mil δ53/52Cr 53 52 53 52 53/52 determined from: [[( Cr/ Cr)sample/( Cr/ Cr)SRM979] – 1] x 1000. In B1 and B2, the Merck Cr(III) standard returned mean δ Cr values of -0.429 ± 0.007 ‰ (2sd, n=6) and -0.434 ± 0.013 ‰ (2sd, n=5), respectively, agreeing well with previous measurements from the laboratory (Albut et al., 2018; Babechuk et al., 2018; Babechuk et al., 2017; Schoenberg et al., 2016; Wille et al., 2013). The B1 samples yielded slightly higher (Ti+V+Fe)/Cr ratios than the B2 samples, but the overall Cr yields of the B2 samples were lower than B1 due to a sub-sampling approach (Supplementary Materials). Five samples were processed in both B1 and B2 for comparison and yielded identical δ53/52Cr values within the internal 2 standard error uncertainties (2se): CLA-003 (B1: -0.318 ± 0.012; B2: -0.310 ± 0.031), CLA-005 (B1: -0.328 ± 0.015; B2: -0.334 ± 0.031), CLA-008B (B1: -0.323 ± 0.019; B2: -0.312 ± 0.040), CLB-023B (B1: -0.318 ± 0.016; B2: -0.350 ± 0.024), CLC-033 (B1: -0.309 ± 0.011; B2: -0.304 ± 0.018). For the latter samples, only the B1 experiment data are reported since the interference corrections were robust and the measurement statistics show better internal repeatability.

4. Results

4.1 Mineralogy and texture of the paleosol

The main constituent minerals identified by bulk XRD were quartz, chlorite, muscovite, albite, and K-feldspar, similar to observations from two previous studies on the paleosol (Sutton and Maynard, 1993; Utsunomiya et al., 2003). The changes in modal mineralogy with depth are illustrated in Figure 2. Plagioclase (albite) was found between 300-500 cm and in a small, restricted area near a depth of 110 cm (represented by sample CLA-009) where it occurs predominantly as relict laths partially replaced by clays. Potassium feldspar is present in the profile below 500 cm and above 300 cm, but is predominantly restricted to the samples closest to the dike margin. At least some of the K-feldspar must have formed early in the paragenesis since it is partially replaced by white mica, as also described by Sutton and Maynard (1993). The presence of muscovite is highest below 500 cm and closest to the dike margins and the unconformity. In general, the muscovite abundance increases upwards above ~400 cm, and shows a distinct stepped offset in parallel with the change in outcrop sampling transect. The abundance of chlorite is broadly anti-correlated with that of muscovite, being highest between 300-500 cm and decreasing upwards. The abundance variation of quartz broadly parallels that of chlorite. Quartz is present predominantly within “rings” that are ~1 cm in diameter that become more irregular in shape higher in the profile. Trace amounts of carbonate (calcite) and sulphide (pyrite, chalcopyrite) were detected in most samples. In the case of calcite, modal values near to or exceeding 1% were found only in samples between 300-500 cm depth, the OD sample, and in the sub-sampled LCP paleosol (CLC-045). Pyrite is found as disseminated blebs, micro veinlets, thicker (1-2 cm) cross-cutting veins associated with chalcopyrite and quartz, and in partial replacement textures of Fe-Ti oxides. Compared to the profile, the OD sample had a similar mineralogy, but with higher abundance of K-feldspar and chlorite.

4.2 Immobile elements: protolith composition, magma petrogenesis, and mass balance

The high field strength elements (HFSE) tend to remain structurally bound in weathering resistant minerals or are sparingly soluble to insoluble during chemical weathering unless extremes in pH are reached (i.e., pH<4 and pH>10). Thus, the HFSE can be important for establishing extreme (e.g., laterititic/bauxitic) weathering conditions, testing for the introduction of allochthonous material, assessing heterogeneities in the protolith material, and, when verified as immobile, calculating mass fluxes of mobile elements. The chemostratigraphic profiles of selected immobile element concentrations (Al and Nb) and ratios (Al2O3/TiO2, Zr/Nb, Nb/Th, Hf/Ta) are plotted in Figure 3 (a) and summarized in Table 2. The abundance of several HFSE (Al, Ti, Zr, Hf, Nb, Ta, Th) is consistent throughout most of the D-S samples between ~300-500 cm, but increases above 300 cm and below 500 cm (CLE-057 to CLF-062) and in the two LCP samples (CLC-045L and -048L). The HFSE abundances above 300 cm increase with sample proximity to the dike margin (D-N) and show a return to lower levels in the channel offset closer to the centre/southern side of the dike. A near-identical abundance profile for all of these HFSE translates into highly consistent HFSE ratios throughout the entire saprolith, which are indistinguishable between the D-N, D-S, and OD samples (Table 2). A percent change in ratio formula (Eq. 1), where Rs is the

HFSE ratio in the paleosol and Rp is the same ratio in the inferred protolith (OD sample), provides a quantitative evaluation of HFSE ratio consistency (Nesbitt, 1979).

% Change in ratio = 100 x [(Rs – Rp)/Rp] (Eq. 1)

All internal variation of the 4 selected HFSE ratios relative to the OD sample is within ± 12% and the mean % change in Hf/Ta, Zr/Nb, and Nb/Th for all samples (n=53) is very close to unity at -0.25, 0.34, and -0.25, respectively. Ratios of the geochemical twin pairs, Nb-Ta and Zr-Hf, also reveal a high level of consistency at 15.38 ± 0.50 (2sd) and 38.22 ± 0.23 (2sd), respectively, and overlap the OD sample with Nb/Ta and Zr/Hf of 15.50 and 38.33, respectively. Note that the direct overlap in HFSE ratios between the saprolith and OD samples confirms that they have a common magmatic parentage. In the adjacent Livingstone Creek Fm. sandstone, the Zr/Nb (27.51) ratio is higher and the Nb/Ta (8.88), Nb/Th (0.82), and Hf/Ta (6.31) ratios are lower than the mafic dike saprolith. Previous studies of the Cooper Lake paleosol (Sutton & Maynard, 1993; Utsunomiya et al., 2003) measured some of these HFSE with lower precision techniques (e.g., pressed powder XRF). The same overall conclusions of minimal HFSE mobility were reached in earlier studies based on Al-Ti-Zr relationships (Rye and Holland, 1998), although a comparison of these data with those of the present study reveals some offset outside of the reported dispersion. A mean Zr/Nb ratio from Sutton and Maynard (1993) and Utsunomiya et al. (2003) was 18.2 ± 4.4 (excluding one anomalously low and likely misreported Zr value of 30 µg g-1) and 32.4 ± 4.1, respectively. The Utsunomiya et al. (2003) paleosol samples have a mean Nb/Th of 1.2 ± 0.2.

Constant inter-element ratios between Al, Ti, Zr, Hf, Nb, Ta, and Th indicates that these HFSE are all suitable for use in mass balance calculations to assess the gains and losses of other elements. For this study, trace elements are normalized to Nb (mass ratios) and major elements are normalized to Al (molar ratios) unless specified otherwise. When relative depletion or enrichment are considered, the % change in ratio (Eq. 1) approach (Nesbitt, 1979) is applied, where Rp and Rs are X/Nb or Z/Al, and X and Z are a trace or major element, respectively. This strategy is identical to other volume/density-independent mass balance models, such as the mass transport function commonly used in weathering and alteration studies (Anderson, 2002; Brimhall and Dietrich, 1987), where values are related through a factor of 100. The OD sample is taken as the reference protolith unless specified otherwise.

4.3 Major elements: chemostratigraphy, weathering intensity, and metasomatism Paleosol major element data are essential to understanding the magnitude of paleo-weathering and post-depositional alteration. Saprolith element/Al ratios vs. depth are plotted in Figure 3 (b,c). Added to these plots is a reference line for the mean composition of drillcore samples, as calculated from samples ‘Core1271’ and ‘Core1338’ reported in Utsunomiya et al., 2003 and the singular sample in Sutton & Maynard, 1993, as well as a shaded area for the mean and 2sd dispersion of the Unit 6 Thessalon (U6T) volcanic rocks (see Section 5.1.1 for more details) reported in Ketchum et al. (2013). The major element stratigraphic trends broadly parallel variations observed for both the bulk mineralogy and immobile element concentrations and are therefore described with a similar sub-division into sample depths of <300 cm, 300-500 cm, and >500 cm.

The saprolith Ca/Al is low overall relative to the drillcore and U6T volcanic rocks, but reaches the highest values at 300-500 cm where minor carbonate is present. The 300-500 cm depth range also has the highest Na/Al and Si/Al and lowest K/Al, with ratios closely matching the drillcore and U6T volcanic rocks. The latter chemical signatures are matched, respectively, by the highest albite and lowest muscovite/K-feldspar. The K/Al increases below 500 cm and above 300 cm correspond to higher abundances of muscovite, with the K/Al trend above 300 cm showing both the offset in the sampled channel (through a decrease back to lower values) and increasing alteration intensities (higher K/Al) as samples approach the dike contact. In terms of the latter elements, the OD sample differs from the drillcore and U6T volcanic rocks with a lower Ca and Na (lower Ca/Al and Na/Al) and enrichment in K (high K/Al), showing similarity in composition to the <300 cm and >500 cm areas of the saprolith and suggesting it was also subjected to weathering and post-depositional alteration. The Fe/Al, Mg/Al, and Mn/Al exhibit correlated trends, with the coupling of Fe and Mg being strongest (Mg/Al vs. Fe/Al for all samples: r2 = 0.952). The Fe and Mg are at the highest abundance overall and closest to the OD and drillcore samples between 300-500 cm, although both elements show a slight gradual decrease upwards in the 300-500 cm region that is decoupled from Na, reflecting a differential mafic mineral vs. albite weathering rate. A more pronounced decrease in Fe-Mg occurs below 500 cm and above 300 cm, inverse to the trends for K/Al (including the sample transect offset). Notably, the stratigraphic trends of Fe-Mg-Mn above 500 cm depth but before the offset in channel cut (i.e., D-N samples) match very closely to the upward loss of Fe documented in Utsunomiya et al. (2003). The Mn/Al plot in Figure 3 shows both the XRF and ICP-QMS data to illustrate that the same trends are preserved despite the consistent bias towards higher values in the XRF data (Section 3.2.7). The P/Al is near-constant throughout most of the saprolith at values close to that observed in drillcore. However, an increase in P/Al is observed below 500 cm and in the uppermost two samples (CLA-002 and CLA-003). The sub-sampled LCP areas (CLC-045L and CLC0048L) show higher Ca/Al and K/Al and lower Na/Al, Fe/Al, and Mg/Al relative to the surrounding bulk paleosol.

Major element data can be cast into several well-established chemical indices that track progressive alteration by quantifying the loss of mobile elements relative to an immobile element(s) and illustrated with ternary plots to reveal specific mineral weathering or post-depositional alteration vectors (Nesbitt, 1992). Most indices are designed to consider only silicate-hosted Ca (e.g., Fedo et al., 1995). An excellent correlation of the TC and Ca abundances (r2=0.997; not shown) across all saprolith samples indicates that most Ca is hosted in carbonate. However, the CaO is very low overall at less than 1 wt. %, excluding the area of subtle carbonate enrichment between 300-500 cm, where it reaches up to 2.8 wt. %, and the two carbonate-rich LCP samples with 8.70 (CLC-045L) and 5.99 (CLC-048L) wt. %. Thus, it was decided to correct the CaO values (CaO*) only for the small contribution of phosphate prior to use in weathering indices (Fedo et al., 1995). All indices are calculated using molar proportions of major element oxides and assume Al immobility (verified in Section 4.3).

The chemical index of alteration (CIA) tracks predominantly the weathering of plagioclase (loss of Ca and Na) and lesser amounts of pyroxene (loss of Ca) during mafic rock weathering.

CIA = 100 * [Al2O3/(Al2O3+CaO*+Na2O+K2O)] (Eq. 2)

The CIA ranges from 56-90 in all samples of the saprolith, apart from the two LCP samples with lower values of 42 (CLC-045L) and 51 (CLC-048L), reflecting their higher proportion of carbonate. The CIA values in Precambrian paleosols most often underestimate the true extent of Ca and Na depletion due to the post-depositional, metasomatic addition of K. The Cooper Lake saprolith is no exception, where the K and Al concentrations are well correlated (r2 = 0.839) and indicate highest K in the areas with greatest Ca and Na depletion. A post-depositional origin of K is best expressed in the Al2O3-CaO*+Na2O-K2O (A-CN-K) plot (Figure 4), where the deviation of samples from the predicted trend for basalt weathering (parallel to and near the A-CN join) towards the K apex (Fedo et al., 1995; Nesbitt, 1992) is illustrated. The localized Ca-enrichment in the LCP samples is also evident in the A-CN-K plot through their deviation from the main sample trend.

Recalculating the CIA without K (CIA-K; numerically equivalent to the Chemical Index of Weathering, CIW, of (Harnois, 1988)) or using the Plagioclase Index of Alteration (PIA; (Fedo et al., 1995; Harnois, 1988)) more accurately expresses the magnitude of Ca and Na loss.

CIA-K = 100 * [Al2O3/(Al2O3+CaO*+Na2O)] (Eq. 3)

PIA = 100 * [(Al2O3-K2O/(Al2O3+CaO*+Na2O-K2O)] (Eq. 4)

The CIA-K and PIA show higher values in the saprolith (reaching 100; Figure 3) and better demonstrate the contrast with the unweathered U6T volcanic rocks (n=30) of Ketchum et al. (2013) that have mean CIA-K/CIW and PIA values of 45 ± 4 and 43 ± 4, respectively. The Ca and Na depletion also evident in the OD sample is indicated by a CIA-K/CIW and PIA of 99 and 98, respectively.

The loss of mobile elements from mafic minerals can be quantified separately or in combination with Ca and Na using indices that incorporate Mg or Mg plus Fe, such as the Magnesium Index (MgI) or the Mafic Index of Alteration (MIA). Here, the MIA iteration that assumes mobility of Fe in its reduced form of Fe(II), the MIA(R), is used (see confirmation of ferrous iron assumption in Section 4.7) and with the exclusion of K (MIA(R)-K; (Babechuk et al., 2014)). MgI = 100 * [Al2O3/(Al2O3+MgO)] (Eq. 5)

MIA(R)-K = 100 * [Al2O3/(Al2O3+Fe2O3(T)+MgO+CaO*+Na2O)] (Eq. 6)

The MgI, tracking only the loss of Mg relative to Al, ranges from 52 to 79, exceeding the value of the OD sample (53.4) and the U6T volcanics with a mean of 51 ± 4. Similarly, the paleosol exhibits a range in MIA(R)-K values of 33-67 that are generally higher than the OD sample (38) and the U6T volcanics with a mean of 26 ± 2. From a Al2O3-(CaO*+Na2O+K2O)-(Fe2O3(T)+MgO) (A-CNK-FM) plot (after Nesbitt and Wilson, 1992), it is evident that several samples cluster together with minimal Fe-Mg depletion, while those from the more altered areas of the dike plot on a tie- line between the composition of chlorite and muscovite (Figure 4) and are thus fully consistent with the observed mineralogical data (Figure 2; Section 4.2).

Stratigraphically, all aforementioned weathering indices show values in the paleosol higher than the U6T rocks at all depths, but close to the drillcore samples between a depth of 300-500 cm. The variation in weathering intensity with depth matches the aforementioned Fe-Mg-Ca-Na-K trends, whereby higher index values develop above 300 cm and below 500 cm. Accordingly, the overall alteration progression is best tracked with decreasing Mg-Fe (and Fe/Al and Mg/Al), increasing K (and K/Al) (Figure 4), and decreasing Na (Na/Al) specifically in the 300-500 cm zone that reflects changes in albite abundance.

4.4 Extended group of elements with a probable pedogenic vs. post-depositional signature

The ratios of Na/Al and Mg/Al (or Fe/Al) can be plotted against Nb-normalized trace element (TE) ratios (TE/Nb) to screen for elements exhibiting a coupled behaviour of depletion or enrichment. Considering all saprolith samples together, Sr/Nb is correlated with Na/Al and Li/Nb is correlated with Mg/Al, indicating that these elements are associated with albite and chlorite, respectively. Other trace elements lack a correlation with Mg-Fe or Na and are described separately below when relevant to the study.

Several trace element (TE) abundances and TE/Nb ratios show strong positive correlations with K concentrations and K/Al, especially in the samples where the K/Al > 0.2. These are Be, Rb, Cs, Tl, Sn, W, and Ba. In the case of Ba, this is consistent with the late addition of barite that was documented by Sutton and Maynard (1993) to occur in thin veinlets. Several of these elements (e.g., Rb, Cs, K) are redox-insensitive and soluble during modern basaltic weathering (e.g., Babechuk et al., 2014; Nesbitt and Wilson, 1992) and were thus presumably originally leached out of the dike prior to metasomatism. At K/Al < 0.2 there is either more scatter, an undeveloped correlation, or an inverse correlation (e.g., W), which suggests their element geochemistry is locally decoupled from K in some areas of the paleosol.

4.5 Fe redox and isotopic geochemistry Utsunomiya et al. (2003) reported a consistent upward decrease in total Fe (ΣFe) and Fe(II) with minimal change to Fe(III) in the Cooper Lake saprolith, indicating a lack of Fe(III)-oxide development or preservation. The same trend of the total Fe (expressed as ΣFe/Al) being controlled by Fe(II)/Al is found in the new data (Table 1), although a slightly higher Fe(III)/Al is measured in all samples here compared to previous studies (Sutton and Maynard, 1993; Utsunomiya et al., 2003). The molar Fe(II)/Al ratios exhibit a wide range from 0.129 to 0.851, while the molar Fe(III)/Al ratios exhibit a limited range from 0.064 to 0.126 with a mean of 0.102 ± 0.034 (2sd; n=31). Stratigraphically, changes in the Fe(II)/Al (Figure 5) parallel the formerly described trend of ΣFe/Al (Section 4.4; Figure 3).

Iron speciation trends can be evaluated effectively with Fe(II)/Al vs. Fe(III)/Al scatter plots (Figure 6; Ohmoto, 1996). The constant Fe(III)/Al with variation in Fe(II)/Al observed in the Cooper Lake saprolith is the expected trend of a reduced paleosol, i.e., samples extend horizontally towards lower Fe(II)/Al from the protolith without plotting significantly above or below the protolith Fe(III)/Al (Figure 6a). Paleosols with higher Fe(III)/Al, but lower ΣFe/Al than the protolith would be consistent with Fe(III) reduction through a secondary, post-oxidation pathway and were considered a “mixed-type” by Ohmoto (1996). A fully oxidized paleosol with negligible post-depositional Fe(III) reduction would follow a line from the protolith towards higher Fe(III)/Al and lower Fe(II)/Al along a negative slope equivalent to the ΣFe/Al of the protolith, as illustrated in Figure 6b using data from the ca. 1.85 Ga Flin Flon paleosol (Babechuk and Kamber, 2013). Also shown for comparison is the ca. 2.96 Ga Nsuze (or Denny Dalton) paleosol (Crowe et al., 2013) in Figure 6c, which exhibits both enrichment and depletion of Fe(II), a loss of Fe(III) in the most Fe-depleted samples (trend of Fe(III)/Al towards the origin from the protolith), and a single sample with a fully oxidized Fe(III)/Al ratio.

The δ56/54Fe values of the saprolith range from 0.169 to 0.492 ‰, whereas the OD sample (protolith) has a value 0.245 ‰. The δ56/54Fe increases 56/54 upwards in the profile with the same offset between the D-S and D-N samples observed for Fe/Al (Figure 5). Iron isotopic data (also δ FeIRMM-

014) available in a conference abstract (Sreenivas et al., 2008) from the same outcrop returned a range of 0.39 to 0.68 ‰ and a value of 0.30 ‰ for a drillcore sample (99BDL11). The δ56/54Fe (and Fe/Al; Murakami et al., 2016) of the latter drillcore closely match that of the OD sample, suggesting both are comparable and appropriate estimates of the protolith Fe isotopic composition. The protolith is notably heavier than the values of unaltered igneous rocks, including basalts, which have δ56/54Fe values closer to ~0.0-0.1 ‰ (e.g., Beard and Johnson, 2004). Although there are some samples in the dike with lower δ56/54Fe than the protolith, the saprolith δ56/54Fe variation is predominantly in the direction of increasing 56Fe enrichment with progressive loss of Fe+Mg from the system. This is evident through well-developed correlations of δ56/54Fe with 56/54 the MIA(R)-K (Figure 7a). The higher δ Fe ratios (up to 0.68 ‰) in some the unpublished data (Sreenivas et al., 2008) are consistent with their lower Fe/Al compared to those from this study (Figure 7). Importantly, however, both data sets fall on a similar trend of coupled Fe depletion and heavy Fe isotope enrichment.

4.6 Rare earth element plus yttrium (REE+Y) geochemistry The REE+Y data are presented normalized to CI-chondrites (Anders and Grevesse, 1989) in Figure 8a along with the range of patterns from the U6T volcanics (shaded grey). The total concentration of the REE+Y (ΣREE+Y, in µg/g) ranges from 64.3 to 362 in all paleosol samples, which is both above and below the OD sample with ΣREE+Y of 145. The REE+Y patterns across the saprolith are broadly parallel to each other, but greater

LREE over HREE variability is evident through PrCN/YbCN and DyCN/YbCN ratios that range from 0.874 to 5.80 and 0.770 to 1.25, respectively.

Stratigraphically, the ΣREE+Y, PrCN/YbCN, and DyCN/YbCN follow similar patterns, being the lowest and most consistent between 300-500 cm, and showing increases above and below this interval to indicate preferential accumulation of LREE>HREE in more altered areas of the dike. An offset increase in PrCN/YbCN and DyCN/YbCN occurs at the transition between the two channel sections (Section 3.1), followed by a gradual decrease in both ratios towards the dike-sandstone contact. Three REE+Y ratios, EuCN/EuCN*, Y/Ho, and CeCN/CeCN*, were selected for examination in closer detail due to their significance in chemical weathering and alteration tracing.

0.5 The Eu anomaly (EuCN/EuCN*, where EuCN* = (SmCN x GdCN) ) in the Cooper Lake paleosol exhibits a range from 1.02 to 0.76, with the lowest (most negative) anomalies found below 500 cm and above 300 cm. The OD sample is near the lower end of the paleosol sample range with a

EuCN/EuCN* of 0.82. Within the saprolith samples, the Eu anomaly is correlated with some alteration proxies, of which a co-variance of decreasing

EuCN/EuCN* and Mg/Al (Figure 8d) is most prominent.

The paleosol Y/Ho ratios lie within a relatively narrow range of 24.42-26.93 (mean of 25.98 ± 0.58, n=53), with the OD sample (Y/Ho: 24.42) matching the low end of the paleosol values. Despite the narrow range, the measured variability significantly exceeds the estimated precision on the Y/Ho ratio of ~1% (1rsd; Supplementary Table 5). Stratigraphically, changes in Y/Ho shows a similar trend to ΣREE+Y, LREE/HREE fractionation, and EuCN/EuCN* with the lowest Y/Ho ratios present above and below 300 and 500 cm, respectively. The subtle decrease in Y/Ho is broadly correlated with the increasing ΣREE+Y (Figure 8c), as well as some immobile element concentrations (e.g., Al).

0.5 Calculating the Ce anomaly (CeCN/CeCN*) using a geometric projection between La and Pr (i.e., CeCN* = (LaCN * PrCN) ) reveals a limited range from unity at 1.05-0.94 with a saprolith mean of 1.01 ± 0.04 and a value of 1.04 in the OD sample. Using the PrCN/PrCN* vs. CeCN/CeCN* plot of Bau and Dulski (1996) with arithmetic calculations for Pr* and Ce* graphically illustrates the lack of any significant Ce anomaly and shows that the lowest CeCN/CeCN* value of 0.94 (CLE-057) likely reflects a slight overabundance of La (Figure 8b, e) since it is not accompanied by an increase in

PrCN/PrCN*.

4.7 Redox-sensitive trace element geochemistry

Of the redox-sensitive elements, focus here is on Mo, V, U, and W, and plots of Nb-normalized stratigraphic trends are in Figure 5. Chromium is considered separately in Section 4.9. 4.7.1 Molybdenum

The Mo concentration of the bulk paleosol ranges from 0.139 to 0.593 µg/g and the two LCP samples have higher Mo concentrations at 0.523 (CLC-045L) and 0.753 (CLC-048L). The mean Mo concentration of the bulk paleosol, 0.318 ± 0.231 (2sd), scatters near the OD sample concentration of 0.320 µg/g. The overall Mo concentration trend is comparable to that of Mo/Nb where samples have a range and mean of 0.018 to 0.097 and 0.005 ± 0.035 (2sd), respectively, that bracket the OD sample Mo/Nb ratio of 0.046. Stratigraphically, the variation in Mo/Nb scatters consistently throughout the paleosol apart from more deviation to higher Mo/Nb in the D-S samples above a depth of 150 cm (Figure 5).

4.7.2 Vanadium

The V concentration of the bulk paleosol ranges from 302 to 790 µg/g and the V/Nb ratio scatters around a mean of 71.8 ± 16.4 (2sd) that is equivalent to a range of +29 % to -19% from the V/Nb ratio of the OD sample (71.2). Stratigraphically, the variation in V/Nb scatters consistently throughout the paleosol (Figure 5). Some of the V scatter was linked to instances when the iCap-Q measurement signal on 51V+ crossed from pulse counting to analogue detection.

4.7.3 Tungsten

The W concentration and W/Nb ratios of the paleosol samples range from 0.660 to 2.17 µg/g and 0.100 to 0.257, respectively, with a mean W/Nb of 0.166 ± 0.085 (2sd) that scatters near the OD sample W/Nb ratio (0.198). A subset of samples exceeding 1.00 µg/g W show a positive correlation with increasing P. Stratigraphically, these are the samples showing higher P/Al in the uppermost paleosol and below a depth of 500 cm. Remaining samples with higher W were previously noted to correlate with K in more altered areas of the saprolith (Section 4.5).

4.7.4 Uranium

The U concentration in the bulk paleosol ranges from 0.829 to 2.56 µg/g with a mean of 1.30 ± 0.850 µg/g, whereas the two LCP samples show higher values locally of 2.00 µg/g (CLC-045L) and 4.08 µg/g (CLC-048L). Uranium enrichment dominates over U depletion, whereby U shows slight enrichment closer to the dike margin above 300 cm, but returns to values overlapping with the protolith above 100 cm (Figure 5). Between a depth of 300-500 cm, U enrichment is minor (not exceeding 20%) apart from in the two LCP samples recording significantly higher U enrichment (69-118 %). Below 500 cm, the U enrichment is higher than anywhere else in the profile (outside of the LCP samples) with a mean enrichment of 64 ± 9% (1 rsd, n=5). The variations in U relative to redox-insensitive immobile elements Th and Nb are also expressed in Figure 11 with a %Th/Nb vs. %U/Nb and U vs. Th/U plot. Outside of the most enriched samples, the Th/U shows a subtle, gradual stratigraphic increase from the lowest to highest samples (not shown). The mean Th/U ratio of all paleosol samples, 3.51 ± 1.17 (2sd), overlaps with the Th/U of the OD sample (3.94). The LCP samples have amongst the lowest Th/U ratios with 2.38 (CLC-045L) and 1.84 (CLC-048L), with the other U-enriched areas reaching a minimum of 2.21. The mean Th/U is close to the average Archean and Proterozoic cratonic shale ratios of 3.54 and 4.21 reported in Condie (1993) and most samples overlap within the mean Th/U ratio of the U6T volcanics with 4.55 ± 1.46 (2sd). The two sandstone samples, CL-MTSS (Matinenda Fm.) and CL-LCSS (Livingstone Creek Fm.) have similar Th/U ratios of 2.46 and 2.44, respectively.

4.8 Extended 3d-subshell transition metal geochemistry

The geochemistry of Co-Ni-Cu-Zn is presented in stratigraphic element/Nb plots and a box-whisker plot of %TE/Nb ratio changes (including %Fe/Al and Mg/Al for comparison) in Figure 9. The concentration ranges (in µg/g) of Co, Ni, Cu, Zn in the bulk paleosol are 25.51 to 126.2, 37.95 to 153.7, 5.18 to 241.8, and 4.68 to 50.40, respectively. The high concentration variation is matched by TE/Nb variations, indicating true losses and gains. The reference point OD sample experienced loss of other mobile elements (Ca, Sr) (Section 4.4), indicating the potential for it to underestimate the true % metal depletion in the paleosol if it also experienced minor metal loss. Thus, TE/Nb ratios of the protolith sample 99BDL11 from Murakami et al. (2016) is also plotted in Figure 9 for reference. The latter study did not include Nb data, so the mean Zr/Nb ratio from this study (24.39; Section 4.2) was used in combination with their Zr data to calculate Nb and construct Nb-normalized data for Zn, Ni, and Cu (Co data were not reported in Murakami et al., 2016).

The highest magnitude of metal depletion is documented for Zn. The Zn/Nb of both the OD and 99BDL11 samples is higher than nearly all of the paleosol (Figure 9) and the maximum depletion of Zn relative to the OD sample is 90%. The Zn depletion increases if data are recast relative to 99BDL11. The Zn/Nb correlates very strongly with both Fe/Al (r2 = 0.911) and Mg/Al (r2 = 0.913) and Zn shows the same stratigraphic trends of depletion as Fe and Mg, implying leaching of Zn from ferromagnesian silicates. The %Ni/Nb (relative to OD) appears to indicate a greater tendency towards Ni enrichment, but recasting data to sample 99BDL11 shows depletion and indicates that the OD sample underestimates the true protolith Ni abundance (Figure 9). This observation coupled with a moderate correlation of Ni/Nb with Zn/Nb (r2=0.562) indicates that Ni had a similar association with ferromagnesian silicates to Zn. There is minimal Co depletion and several samples showing enrichment relative to the OD sample, but data are unavailable for 99BDL11 to provide clarity in whether the enrichment trends are primarily a function of Co depletion in the OD sample. A weak correlation of Co/Nb with Zn/Nb (r2=0.280) that improves slightly (r2=0.384) with the removal of the three highest Co/Nb values (>20) suggest some leaching from ferromagnesian silicate minerals, but a positive correlation of Co and S in the more Co- enriched samples is evidence for some distribution within sulphides. The behaviour of Cu contrasts highly with the other metals in its evidence for extensive enrichment, which is predominantly restricted to depths below 400 cm (Figure 9). The very wide range of Cu/Nb ratios, equivalent to a -15% to 2960 % change (relative to OD), exhibits no correlation with Zn, Co, Ni, or Fe (and Mg). Recasting %Cu/Nb to 99BDL11 reduces the magnitude of Cu enrichment and shifts more samples to slightly depleted values, but does not account for the most Cu-enriched samples, which retain a clear excess relative to both protolith compositions. Collectively, the aforementioned trends suggest a relative depletion of Fe≈Mg≈Zn>Ni>Co>Cu, where Cu shows the greatest evidence of localized enrichment. Although Cu should exhibit redox-sensitive mobility during chemical weathering (Neaman et al., 2005b) and holds promise as a paleo-redox proxy from marine environments (Chi Fru et al., 2016), the elemental enrichments and presence of secondary chalcopyrite at Cooper Lake indicate that no new constraints on pedogenic Cu behaviour can be made here and Cu data are not considered further.

4.9 Cr geochemistry

The Cr abundance in the bulk paleosol ranges from 1.71 to 3.89 µg/g (mean of 2.46 ± 1.25, 2sd) with stratigraphic variations paralleling those of the HFSE (Section 4.3). The OD sample has a Cr abundance of 2.72 µg/g and the sub-sampled LCP has slightly higher abundances of 2.59 (CLC- 045L) and 4.17 (CLC-048L) µg/g relative to bulk samples at the same depth. Constant Cr/Nb are observed across all paleosol samples, 0.373 ± 0.066 (2sd), match the Cr/Nb of the OD samples (0.388), and show no variation with depth. The surrounding sandstone samples have higher Cr abundances of 31.35 (CL-MTSS) and 30.35 (CL-LCSS) µg/g. Previous studies reported significantly higher Cr in the saprolith of 51 to 83 µg/g (Utsunomiya et al., 2003) and 13 to 24 µg/g (Sutton and Maynard, 1993), although the most recent study reported only slightly higher values ranging from 5.19 to 11.2 µg/g (Murakami et al., 2016). The Cr abundance in CLA-005 was reproducible between two different ICP-QMS instruments with completely different calibration strategies (Supplementary Material), and the ICP-QMS data (Table 1) show good agreement with those derived from the double-spike MC-ICP-MS method (Table 3). The inter-study bias is thus attributed to analytical factors (e.g., Cr-steel milling and overlap of Fe peaks on Cr in XRF analyses) with the true Cr abundance overestimated in earlier studies.

The stable Cr isotope ratios of the saprolith and OD sample are near-constant with a δ53/52Cr ranging from -0.291 to -0.365 ‰ and a mean of - 0.321 ± 0.038 ‰ (2s; n=34). No obvious trend of δ53/52Cr values with depth is evident (Figure 5) and the δ53/52Cr values do not correlate significantly with any major or trace element concentration or ratio. Sub-samples of the LCP (CLC-045L and CLC-048L) did not reveal a consistent trend in their Cr isotope systematics relative to the adjacent bulk paleosol, with CLC-045D and CLC-045L bearing δ53/52Cr of -0.351 ‰ and -0.309 ‰, respectively, and CLC-048D and CLC-048L bearing δ53/52Cr of -0.293 ‰ and -0.310 ‰, respectively. All stable Cr isotope compositions of the saprolith are significantly depleted in the heavier isotopes of Cr relative to the so-called average igneous inventory with a δ53/52Cr of -0.124 ± 0.101 ‰ (Schoenberg et al., 2008).

4.10 Mn-Fe-Cr distribution in paleosol

Three samples (CLA-004: 47 cm; CLC-037: 324 cm; CLD-056; 494 cm) were selected for µ-XRF mapping to represent different areas of the saprolith. The deeper samples bracket the 300-500 cm zone that is highest in Mg-Fe-Na and lowest in K (high chlorite abundance), whereas the shallowest sample is closer to the dike centre in the area with low Mg-Fe-Na and high K (high mica abundance) (Section 4.2 and 4.4). The primary objective of the mapping (Figure 10) was to elucidate the element-mineral associations of Cr-Mn-Fe and their relationship with S, Ti, and K (as proxies for sulphide, Ti-oxides, and K-bearing clay/feldspar, respectively). In CLD-056, a fine-grained groundmass of chlorite that hosts both Fe and Mn surrounds irregular patches of plagioclase and quartz. The correlations observed between Mn and Fe and the two of these elements with Cr are consistent with chlorite being their primary host. However, an independent correlation of Fe with S and a minor correlation of Cr with Ti is consistent with separate controls by the sulphide and Ti-oxides in the sample, respectively. Higher in the profile but still within the least-altered zone, CLC-037 shows similar Fe-Mn-Cr relationships. Chromium is distributed throughout the chlorite-rich matrix, and also within smaller Fe-Ti bearing phases (Fe-Ti oxides) and smaller sulphide-bearing areas are apparent through Fe-S correlations. In the highly altered sample close to the unconformity, CLA-004, it is apparent that K dominates the fine-grained matrix and a separate generation of Fe/Mn-bearing chlorite bounded by quartz is spread throughout the matrix. Chromium is distributed both within the K-rich matrix and in smaller grains also rich in Ti (Ti-oxide) throughout. Clusters of sulphide are again evident through Fe-S relationships. An area of apparent higher Cr abundance around the main cluster of sulphide (top left) is an artefact from sample preparation. Overall, the maps illustrate that Cr has an association with the matrix clay minerals and with Ti-oxide minerals and thus was likely distributed throughout different phases, including mafic silicate minerals such as pyroxene, prior to alteration.

5. Discussion

5.1 New insights into the origin of the Cooper Lake dike, sub-Matinenda paleosol formation, and sedimentation in the Lower Huronian

5.1.1 Implications of constant HFSE ratios in the dike

A full suite of HFSE (Nb-Ta-Zr-Hf-Th) measured at high-precision is a powerful tool for identifying even subtle pre-weathering lithological variations or allochthonous contamination such as aeolian dust or reworked sediment in weathering profiles (e.g., Babechuk et al., 2015) and recognizing unusual (e.g., F-rich) post-depositional fluid alteration signatures in paleosols (e.g., Land et al., 2018). The very limited deviation in all HFSE ratios of the Cooper Lake dike (Table 2; Figure 3a) has several implications: (1) the interaction of the mafic melt with the host sandstone during intrusion led to negligible local assimilation and did not significantly affect these elements; (2) the pH during chemical weathering did not reach the extremes necessary to induce significant HFSE fractionation; (3) the dike had a similar initial bulk composition across all depths prior to any alteration; and, (4) the HFSE remained unaffected throughout post-weathering metasomatism and metamorphism.

Point (1) is important since the presence of K-feldspar at the dike margins suggests interaction of the mafic dike with the host sandstone (Sutton and Maynard, 1993). The immobile element ratios (Zr/Nb, Nb/Ta, Nb/Th, and Hf/Ta) in the Livingstone Creek Fm. sandstone differ from those of the dike (Table 1) and given the sensitivity of these ratios (e.g., Nb/Th) to assimilation, it is unlikely that the interaction implied by K-feldspar left a significant chemical imprint on the dike. Point (2) is relevant to comparing Cooper Lake with other sub-Matinenda paleosols, most notably the greenstone-hosted Denison Mine paleosol, where significant Al-Ti, but not Al-Zr, fractionation is documented (Mossman and Farrow, 1992). The fractionation remains ambiguously related to either transported allochthonous material in the paleosol (Rye and Holland, 1998) or an enhancement in acid weathering intensity after the Archean-Proterozoic boundary (Konhauser et al., 2011). The lack of Al/Ti fractionation at Cooper Lake indicates that acid-generated, extreme weathering conditions at the time did not influence this regolith or that such conditions did not penetrate deeply into weathering profiles (with Cooper Lake not recording a more leached upper horizon). The origin of HFSE fractionations in other sub-Matinenda paleosols should be revisited with a full suite of HFSE to decipher the mineralogical/parent rock or pH-induced controls. In addition to weathering intensity, no allochthonous material addition at Cooper Lake can be inferred and is consistent with rapid precipitation of minerals in available pore spaces suggested previously by Sutton and Maynard (1993). Point (3) validates the use of the HFSE in mass balanced-based interpretations of mobile element cycling during one or more of the three main alteration events documented in the area (Sutton and Maynard, 1993). Point (3) and (4) permit the use of the HFSE in testing the proposed co-genetic link between the Cooper Lake dike and the volcanic effusive rocks of the Thessalon Fm. (Bennett et al., 1991). Ketchum et al. (2013) sub-divided these volcanic rocks into units based on chemical and mineralogical characteristics, and the 5 main high-Mg basaltic or basaltic-andesitic units exhibit distinct groupings based on their HFSE ratios (Zr/Ti, Nb/Th, Zr/Nb). Of these units, the one that is stratigraphically youngest and volumetrically most significant (Unit 6; excluding samples KYT512, KYT513 as per the screening of Ketchum et al., 2013) bears HFSE ratios closest to the Cooper Lake saprolith (Table 2) with mean ± 2sd Zr/Ti, Nb/Th, and Zr/Nb ratios of 0.017 ± 0.006 (n=30), 1.51 ± 0.48 (n=24), and 20.68 ± 6.68 (n=30), respectively. Ratio-based and ternary discrimination diagrams that incorporate the latter elements (plus Yb), such as the Th-Hf-Ta, TiO2/Yb-Nb/Yb, and Th/Yb-Nb/Yb plots (Wood et al., 1980; Pearce et al., 2008), accurately record the mantle source compositions due to the high magmatic incompatibility of the HFSE. These plots (not shown) demonstrate a good overlap of the Cooper Lake saprolith samples with the Unit 6 Thessalon (U6T) volcanic rocks, supporting a co-genetic link between both and also the previous petrogenetic interpretations as within-plate basalts/basaltic andesites.

5.1.2 Overview of paleosol paragenesis and post-depositional alteration

The multiple generations of feldspar in the saprolith reveal distinct alteration events. The earliest of these events appears to be an albitization, also known from elsewhere in the Huronian (Fedo et al., 1997; Sutton and Maynard, 1992), which modified existing plagioclase in both the dike and adjacent sandstone (Sutton and Maynard, 1993). This event perhaps contributed to the depletion of Ca throughout the dike and the drillcore samples in the Cooper Lake area (Utsunomiya et al., 2003). Subsequent erosion led to exposure of the dike and Livingstone Creek Fm. sandstone to subaerial weathering, resulting in breakdown of albite to Al-rich clay minerals (e.g., kaolinite) and removal of remaining Ca and Na, as evident from very high CIA-K and PIA values. These conditions would have also supported extensive weathering of ferromagnesian minerals (e.g., pyroxene) to clay minerals such as the smectites (Babechuk et al., 2014; Eggleton et al., 1987; Nesbitt and Wilson, 1992). Smectites have a high capacity to retain Mg (Harder, 1972) and probably controlled the minimal losses of Mg (and Fe) below 300 cm in the saprolith and closer to the centre of the dike with MgI and MIA(R)-K close to drillcore and U6T samples. However, the increased MgI and MIA(R)-K higher in the saprolith and towards the dike margins implies greater leaching of Mg and Fe(II) and a weathering residue richer in kaolinite (Murakami et al., 2004). The Cooper Lake samples measured by Utsunomiya et al. (2003) showed a gradual and consistent upward depletion of Fe-Mg that matches the trend observed for the D-N samples (Figure 4), but did not capture the same intra-dike chemical variability observed here and by Sutton and Maynard (1993) related to the proximity of samples to the dike margin. Nevertheless, the overall trends are similar to the granite-hosted sub-Matinenda Lauzon Bay paleosol (Sutton and Maynard, 1992), pointing to high levels of element leaching (Ca, Na, Fe, Mg, Mn, Si) from both albite and ferromagnesian minerals that is comparable to other deep, Precambrian weathering profiles (Maynard, 1992). Such conditions were probably supported by elevated atmospheric CO2 and associated lower pH in weathering fluids in the early Proterozoic (Reinhard et al., 2009; Sheldon,

2006), as well as other climatic factors such as high rainfall, although a more quantitative assessment of CO2/pH was not undertaken for this study. The new trace element data, combined with those reported previously (Murakami et al., 2016; Sutton and Maynard, 1993; Utsunomiya et al., 2003), reveal several trace elements also showing a chemical depletion associated with subaerial weathering (Sr, Li, Zn, Co, Ni, Eu) based on correlations with Na and Mg. Several of these elements are mobile during modern basalt chemical weathering (Babechuk et al., 2014; Nesbitt and Wilson, 1992). Other elements show evidence for only subtle mobilization (U, P) or immobility (V, Cr) and the significance of these results is the focus of further discussion in Sections 5.2-5.4.

An enrichment of K is known to have influenced the entire Huronian Supergroup after sedimentary deposition (Fedo et al., 1997 and references therein) resulting in the formation of K-feldspar and micas. The extent of K addition to the Cooper Lake dike evaluated from the trajectory of samples in Figure 4 is consistent with that observed in the siliciclastic samples of the Livingstone Creek and Matinenda Fms. (Young et al., 2001 and references therein), as well as younger sedimentary units of the Huronian Supergroup (Fedo et al., 1997), providing supporting for the regional extent and contemporaneity of the alteration event. Potassium addition was most severe in the Al-rich sections (most weathered and probably most porous) sections of the weathered dike Cooper Lake dike, and is consistent with metasomatic alteration models (Novoselov and de Souza Filho, 2015) and empirical observations from most other Precambrian paleosols (Nesbitt & Young, 1989; Maynard, 1992). Rubidium was previously demonstrated as an element co-enriched with K at Cooper Lake (Sutton and Maynard, 1993) and this study identifies a number of other elements (Cs, Be, Tl, Ba, Sn, In, W) geochemically associated directly (i.e., co-hosted as trace substituent in mica) or indirectly with K, indicating their weathering signatures have been partially to wholly overprinted. A more detailed analysis of the metasomatism is not the focus of this contribution, but attention is drawn to W since there is very little known about this element in ancient paleosols. The potential of applying W geochemistry in Precambrian paleosols as a continental redox tracer was explored in Murakami et al. (2016), but data remain sparse and potentially compromised by sample preparation using tungsten carbide. This study, which applies a high precision analysis technique (see Babechuk et al., 2010), indicates that low-T brines involved in metasomatism contained soluble aqueous W species that were at least partly scavenged and retained in the buried paleosol. Prior to developing W as a deep time redox proxy, more paleosols will need to be carefully screened for post-depositional W effects. Nevertheless, the enrichment of W in the paleosol broadly reflects its fluid mobility, which is well- documented, for example, in subduction zone fluid-transfer processes (e.g., Bali et al., 2012; König et al., 2008) as well as several low-T aqueous fluid reactions such as serpentinization and chemical weathering and transport to the hydrosphere (e.g., Arnórsson and Óskarsson, 2007; Babechuk et al., 2015; Baldwin et al., 2012; Johannesson and Tang, 2009; Peters et al., 2017).

Novoselov and de Souza Filho (2015) predicted that metasomatism could impart post-depositional changes to the Fe(II)/Fe(III) ratio of mafic paleosols and identified the previously published upward decrease of Fe(II)/Fe(III) preserved in the Cooper Lake saprolith (as measured in earlier studies) as an example of metasomatic change to Fe speciation. Unlike the model predictions of these authors, however, the new data reveal that the Fe(III) content is virtually unchanged (normalized to Al) throughout the dike such that the Fe(III)/Al ratio is unmodified during all of the stages of alteration. Thus, although post-depositional Fe(III) reduction may be evident in other paleosols (Crowe et al., 2013; Ohmoto, 1996), it is not evident through the Fe speciation at Cooper Lake. Moreover, an inverse trend of upward increasing Fe(III)/Fe(II) is preserved in the adjacent, contemporaneously developed sandstone profile (Sutton and Maynard, 1993). Both these observations remain consistent with earlier interpretations of subaerial weathering under a very low atmospheric oxygen content where free O had a greater influence on the extent of Fe(II) oxidation in low-Fe rocks over high-Fe rocks (Sutton and Maynard, 1993; Utsunomiya et al., 2003). The very close coupling of bulk rock Fe and Mg and the consistency of Fe/Mg ratio evolution of chlorite grains in the dike are also consistent with experimental anoxic weathering trends (Murakami et al., 2004). Accordingly, it is interpreted here, as in previous studies, that the chlorite-muscovite assemblage (Figure 2) and inverse trend in Mg/Al vs. K/Al space (Figure 4) developed from the K-enrichment of a weathering residue initially depleted in Fe and Mg rather than Fe-Mg loss occurring during metasomatism (Nesbitt, 1992). Finally, the observation of secondary sulphide textures, such as partial replacement of Fe-Ti oxides (Section 4.1), indicates that some reduced S was introduced late in the paragenesis and likely influenced some of the trace metal signatures (notably Cu and possibly Co to a lesser extent). Nevertheless, the Fe(II)/Fe(III) data argue that Fe behaved isochemically and retained its pedogenic signature throughout this sulphidation. The cycling of S in Archean-early Proterozoic paleosols, including Cooper Lake, has recently been considered in terms of S isotopes (Maynard et al., 2013), but given the complexities in the preserved sulphide textures, a more detailed and focused investigation would be required to establish primary vs. secondary trends and associated trace metal and S isotopic budgets.

Ensuing discussion focuses on soluble element behaviour during weathering and fluxes to the hydrosphere. However, it is noted that the geochemical composition (e.g., residually enriched elements) of the most weathered samples would best approximate the composition of material amenable to physical redistribution (e.g., in sediment transported through aeolian or fluvial transport). In view of the spatial complexities of the dike alteration and evidence for displacement along the faulted contacts, the paleosol is less amenable to mass flux calculations that rely on horizontality of the chemical weathering progression. As such, this study focuses on bulk geochemical trends with variations in weathering intensity. Throughout the discussion, microbial processes are not considered to have played a major role in regolith development at Cooper Lake. It is difficult to ascertain biotic effects with the uppermost profile at least partly eroded away. However, a minor to negligible role can be suspected based on all total C being preserved in carbonate and all elements demonstrated to be mobilized preferentially as organic ligand complexes (P; (Neaman et al., 2005a)) retained through the high degrees of leaching.

5.2 Assessment of trace element paleo-redox proxies

5.2.1 Convergence of proxy evidence for pervasively anoxic weathering conditions

The Cooper Lake Fe speciation data (Figures 5 and 6) fit the criteria of a reduced paleosol without explicit evidence for an earlier oxidative event overprinted by organic acids, reducing groundwater, or hydrothermal/metasomatic fluids (Ohmoto, 1996; Rye & Holland, 1998; Sutton & Maynard 1993; Utsunomiya et al., 2003). The leaching of Fe(II) (with Mg and Mn) is therefore linked to the subaerial weathering of ferromagnesian silicates, implying minimal formation of Fe(III)-oxides (and by extension Mn(III)/(IV)-oxides). Importantly, a similar, leached behaviour of Fe and Mn is also observed in other high-Fe, sub-Matinenda paleosols in the area (Mossman and Farrow, 1992; Prasad and Roscoe, 1996), although minor Fe(II) oxidation is observed in two low-Fe, sub-Matinenda paleosols (Gay and Grandstaff, 1980; Sutton and Maynard,

1993). From a global standpoint, a lower surface water pH from high Paleoproterozoic atmospheric CO2 (Sheldon, 2006) would have helped to release and stabilize Fe(II). However, several lines of evidence from the paleosol trace element inventory, as summarized below, suggest it was the reducing nature of pore waters that exerted the strongest control on Fe(II) depletion with inhibited pedogenic Fe/Mn oxide formation.

The coupled leaching of several 3d-subshell transition metals (Mn, Zn, and to a lesser extent Co, Ni) alongside Fe and Mg provides indirect support for their initial localization in ferromagnesian minerals and their coupled release during subaerial weathering. The most parsimonious explanation for a coupled loss of all metals reaching high degree of depletion is an absence of Fe(III)- and Mn(III)/Mn(IV)-oxide formation, since these oxides play a major role in metal retention through surface adsorption and co-precipitation (Cornell and Schwertmann, 2003). However, it is recognized that metal retention would have also depended on the interplay of pH, nature of clay surfaces and incorporation into phases such as Mg/Fe(II)-smectite minerals (Nesbitt and Wilson, 1992), aqueous metal speciation, and the (potential) presence of organic matter. Depletions of Zn and Ni were documented in previous studies of the Cooper Lake diabase and similarly interpreted to record their leaching during pedogenesis (Murakami et al., 2016; Sutton and Maynard, 1993).

The lack of a Ce anomaly (outside of 0.95-1.05) in the Cooper Lake saprolith (Figure 8b/e) is consistent with reducing conditions (see also Utsunomiya et al., 2003). Anomalies of Ce develop as Ce(III) is oxidized to Ce(IV) on oxide (primarily Mn(III)/(IV)-oxide) mineral surfaces to form cerianite and the pH conditions subsequently become favourable for trivalent REE mobilization in preference to Ce(IV) (e.g., Koppi et al., 1996; Marsh, 1991). Although bulk rock analyses can sometimes obscure Ce anomalies developed at a fine scale (Prudêncio et al., 1993), Fe/Mn-rich basaltic substrates are generally efficient in developing Ce anomalies under oxidative conditions, as illustrated by the magnitude of positive and negative Ce anomalies in saprolitic and lateritic profiles from the Deccan Traps (Figure 8b). Moreover, independent and more robust micro- mineralogical support also comes from previous studies, where secondary, LREE-rich accessory phases (e.g., rhabdophane) that do not exhibit decoupling of Ce from other LREE were documented in sub-Matinenda paleosols (Murakami et al., 2001). These phases stand in contrast to the secondary cerianite observed in some post-GOE mafic paleosols (Pan and Stauffer, 2000).

The complete immobility of Cr indicated by the invariant Cr/Nb ratios throughout the entire saprolith and OD sample (Figure 5) suggests an absence of oxidative Cr mobilization. The synchrotron µ-XRF maps (Figure 10) illustrate that Cr is distributed between different phases in variably altered parts of the paleosol, confirming that Cr did not remain crystal lattice locked in a specific, weathering-resistant phase that would isolate the metal from interaction with pore waters (D’Arcy et al., 2016). The maps imply that retention of Cr(III) originally present in ferromagnesian silicates was likely controlled by localized adsorption and incorporation into Mg/Fe(II)-smectites (e.g., (Nesbitt and Wilson, 1992)). Thus, both chemical and mineralogical observations suggest that Cr was not oxidized to more soluble aqueous Cr(VI) species, a process which requires the presence of Mn-oxide surfaces in modern weathering environments (e.g., Oze et al., 2007). Previous results from Cooper Lake indicated slight mobility of Cr (Murakami et al., 2016), but even a minor amount of Cr mobility is not supported by the new data.

The immobility of V indicated by the very limited deviation in V/Nb ratios is comparable to Cr in suggesting it was not removed via oxidation from V(III) to vanadyl or vanadate species and exported to pore waters. The oxidation capacity of V is enhanced after its release and adsorption to mineral surfaces in soils, such that it may poorly reflect the state of pore water oxidation state if it remained sheltered in weathering resistant minerals (typically substituting along with Ti in minerals such as magnetite; Nesbitt and Wilson, 1992). Overall, V mobility can also be influenced by other factors such as variable retention by clays and Fe(III)-oxides (e.g., Wisawapipat and Kretzschmar, 2017), but the immobility relative to Nb supports sustained reducing conditions during weathering.

In summary, a consistent view emerges from the geochemical behaviour of several directly redox-sensitive elements (Fe, Mn, Ce, Cr, V) known to exhibit redox-dependent changes in mobility during weathering (Middelburg et al., 1988) and those known from modern soils to respond indirectly to Fe/Mn-oxides (Co, Ni, Zn). All signatures point to pervasively anoxic pore water conditions (Sutton and Maynard, 1993; Utsunomiya et al., 2003). The overall reducing conditions (i.e., low enough to prevent Fe, Mn, Cr, Ce, Mo, and V oxidation) inferred from the Cooper Lake paleosol are consistent with the presence of reduced detrital minerals in overlying sedimentary units (Long et al., 2011; Pienaar, 1963; Roscoe, 1973; Zhou et al., 2017). The signatures also stand in contrast to those preserved in paleosols higher in the HS, which show evidence for greater Fe and Ce oxidation (Panahi et al., 2000; Rainbird et al., 1990), coinciding with the appearance of continental red beds. As such, new observations remain consistent with a younger (<2.45 Ga) timing, at least in this region, of accumulating atmospheric oxygen beginning to play a major role in redox-sensitive element cycling. 5.2.2 Testing Mo-U as the most sensitive oxidative weathering proxies

The newest research on the timing of the GOE suggests onset at ca. 2.45 Ga, which was contemporaneous with Cooper Lake regolith development. At this time, the marine geochemical record requires at least an episodic supply of oxidized aqueous species of several redox- sensitive elements (e.g., Re, Mo, U, S) from oxidative continental weathering (e.g., Anbar et al., 2007; Bau and Alexander, 2009; Duan et al., 2010; Eroglu et al., 2015; Kendall et al., 2015; Kurzweil et al., 2015; Partin et al., 2013a; Partin et al., 2013b; Philippot et al., 2018; Reinhard et al., 2013; Reinhard et al., 2009). While the primary source of these elements was likely evolved felsic rocks (Mo, U) or sulfides (S, Mo, Re) exposed under a mildly oxygenated atmosphere rather than unmineralized mafic rocks, these marine observations nonetheless predict these elements as most probable to show evidence for mobility in paleosols (Wille et al., 2013).

The near-constant Mo/Nb ratios in the Cooper Lake saprolith (Figure 5) implies insignificant Mo mobility during subaerial weathering and any younger alteration event. These results contrast with Murakami et al. (2016) who found minor leaching of Mo from the dike. Studies on modern basaltic weathering profiles reveal organic matter and Fe-oxides to play major roles in Mo retention (or even enrichment) and that external sources of Mo from groundwater and atmospheric input can be significant (King et al., 2016; King et al., 2014), although HFSE and Fe speciation data at Cooper Lake indicate these factors were negligible. Textural evidence indicates some sulfide minerals (pyrite, chalcopyrite) are secondary and post-date subaerial weathering, but the limited Mo/Nb variation combined with poor correlations between Mo and S in bulk geochemical data (not shown) also eliminates an obvious association of Mo with sulphide minerals. Thus, a caveat is that Mo immobility may be related to its position as a trace substituent in relatively weathering resistant, non-sulfide minerals (e.g., co-hosted with V and Ti). However, some supporting evidence for inhibited Mo mobilization comes from an absence of Mo depletion in glacial tillites higher in the Huronian Supergroup (Gaschnig et al., 2014). Moreover, a limited capacity to release Mo from sulphides via oxidative weathering at the time is also inferred from the presence of detrital pyrite in the overlying Matinenda and Missassagi Fm. fluvial units (Long et al., 2011; Zhou et al., 2017).

There are currently very few Precambrian paleosol studies with high-quality Mo data, as evident through the data compilation by Murakami et al. (2016). Notably, however, Yang et al. (2002) found Mo/Ti variations in the ca. 2.76 Ga basalt-hosted Mt. Roe #2 paleosol that could capture continental Mo mobility under low atmospheric O2 conditions not found in the Lower Huronian. The direct continental fingerprint of oxidative Mo release at ca. 2.45 Ga remains inconclusive and calls for studies of other ca. 2.45 Ga continental sedimentary reservoirs.

Uranium is the only redox-sensitive element showing evidence for mobility and redistribution (Figures 5 and 11). Uranium can be mobilized in its reduced state (U(IV)), but only at low pH and with the presence of certain ligands (e.g., F-), whereas in oxygenated environments U(VI) can be 2+ mobilized more extensively as the uranyl (UO2 ) ion over a wider range of pH (slightly acidic to basic), often as a complex with phosphate or carbonate (Langmuir, 1978). Very low pH is not supported by other geochemical data (e.g., Sections 5.1.1, 5.4), lending support to an oxidative U mobilization process. The subtle variation in U/Nb ratios (from higher values in the deepest samples to lower values in samples closer to the unconformity) may suggest a greater release of U closer to the surface and reductive deposition or adsorption on mineral surfaces in deeper groundwater. However, given the substantial evidence for anoxic conditions prevailing in the dike and lack of areas exhibiting extensive U depletion, it is suggested that some of the depth-dependent U variation is related to mobilization of U from the adjacent Livingstone Creek Fm. sandstone. Oxygen dissolved in pore water would have survived for longer in the sandstone due to a lower reductant-driven O2 demand (Sutton & Maynard, 1993) and could have led to more U release and transport. Subsequent reductive scavenging of U(VI) would have occurred in the dike due to higher availability of Fe(II) and/or adsorption to the more abundant reactive clay surfaces. The high U enrichment (high U/Nb and low Th/U) in the carbonate-rich LCP samples also suggests small scale U(VI) mobility. The coupled and selective enrichment of U with P in the lowermost profile (samples below 500 cm) suggests the residence of U in a phosphate and perhaps coupled U-P mobility. Tungsten is also selectively enriched in this area, indicating it too may have a similar affiliation with phosphate (and thus may locally record a signature not related to burial metasomatism; Section 5.1.2).

Caution is attached to the inference of primary U mobility in the paleosol. Although the adjacent sandstone paleosol appears to show a greater level of Fe oxidation compared to the dike, consistent with its lower total Fe (Sutton & Maynard, 1993), the loss of U from the adjacent sandstone is unsupported by the singular measurement of low Th/U (2.44). This Th/U measurement could reflect a combination of U-rich detritus from highly evolved Archean granites in the area (Robinson and Spooner, 1982) or incomplete digestion of a Th-rich phase (e.g., monazite) during preparation. It is noted that the lowest paleosol region recording the samples with the most extreme U enrichments (not including the LCP samples) are restricted to those extracted from beneath overburden (Figure 12a); the very U-rich (detrital uraninite) rocks of the Matinenda Fm. in the area coupled with modern oxidative processes indicates that recent U scavenging must be considered as a possibility (Albut et al., 2018) despite the effort to secure the freshest possible samples during preparation.

Assuming U signatures at Cooper Lake are pedogenic, data are compatible with some oxidative U cycling in continental environments at ca. 2.45 Ga and thus could be consistent with minor continental U(VI) delivery to oceans via rivers. However, the presence of detrital uraninite (e.g., Pienaar, 1963; Robinson and Spooner, 1982) in the Matinenda Fm. immediately overlying the Cooper Lake paleosol as well as the lack of U depletion in glacial diamictites throughout the Huronian (Gaschnig et al., 2014) would both indicate that oxidative U mobility could not have -3 been widespread and that atmospheric O2 levels were not likely to have exceeded 1.5 x10 PAL (Johnson et al., 2014).

5.2.3 Continental-marine redox connection and atmospheric O2 levels at ca. 2.45 Ga

Recent semi-quantitative atmospheric oxygen models based on the Fe data of sub-Matinenda paleosols (Kanzaki and Murakami, 2016; Murakami et al., 2011), including Cooper Lake, are consistent with the maintenance of limited MIF-S preserved in the lower Huronian in -5 suggesting PO2 at <10 present atmospheric levels (PAL) (Papineau et al., 2007; Pavlov and Kasting, 2002). However, the thresholds of detrital uraninite and pyrite preservation, release of Mo from molybdenite oxidation, and the generation and preservation of Cr(III)-Cr(VI) redox induced -4 Cr isotopic fractionation suggest pre-GOE PO2 fluctuations reaching as high as 3 x 10 PAL (Crowe et al., 2013; Greber Nicolas et al., 2015; Johnson et al., 2014). A combination of empirical, experimental, and theoretical thermodynamic models predict that the release of Mo (and Re) should be more sensitive than U in recording continental oxidative weathering, since sulphides can corrode under conditions that still preserve U(IV)-bearing minerals (e.g., (Kendall et al., 2015; Sverjensky and Lee, 2010). However, marine U enrichments at certain intervals throughout the Archean have been documented (Bau and Alexander, 2009; Frei et al., 2016) and oxidative U release is thermodynamically permissible below a -5 PO2 of 10 PAL (Frei et al., 2016). Reducing conditions are aided in mafic substrates by their higher CO2>O2 demand (quantified as the R ratio; Holland, 1984; Holland and Zbinden, 1988; Pinto and Holland, 1988), indicating that trace element evidence of a mildly oxic atmosphere and upper surface may have remain masked in mafic paleosols until higher atmospheric oxygen levels after ca. 2.3-2.4 Ga were reached. Thus, as advocated in previous studies (e.g., Lalonde and Konhauser, 2015), models connecting oxygenation signatures between continental and marine reservoirs call for localized areas of continental oxidation near the Archean-Proterozoic boundary to supply the oceans with Mo, Cr, and U. An implication is that marine deposits are liable to record a wider (catchment to global) scale weathering signature than what may be preserved in singular paleosols. Alternatively, the onset of the GOE at ca. 2.45 Ga was not globally synchronous (Philippot et al., 2018) and lagged behind in the Huronian Supergroup of North America relative to South Africa and Australia (where IF deposits record the greatest Cr and U enrichments; Konhauser et al. 2011; Partin et al., 2013b). This is plausibly explainable through the abundance of subaerial and marine reductants exposed in the early continental rift basin during the sedimentation captured in the Lower Huronian. These conditions could have remained until the establishment of a more biologically active (oxygen-producing) continental margin, as recorded in the Upper Huronian, while a more extensive amount of surface oxygenation began at ca. 2.45 Ga on other continents.

5.3 Pedogenic REE+Y fractionation and the continental REE+Y flux from mafic landmasses

Despite the geochemical complexities of the dike-sandstone contact, changes to the REE chemistry during the alteration of the dike (Figure 8) are consistent with the well-documented (but complex) fractionation of REE during modern chemical weathering processes (Babechuk et al., 2014; Duddy, 1980; Middelburg et al., 1988; Nesbitt, 1979; Nesbitt and Markovics, 1997). The LREE/HREE fractionation shows LREE-enrichment in the most altered areas of the dike and is consistent with observations of kaolinite-dominated weathering products becoming LREE-enriched (e.g., Nesbitt et al., 1990). The fractionation of REE in other Precambrian weathering profiles has been attributed to pedogenesis (Frei and Polat, 2013; Panahi et al., 2000), but the REE are also demonstrably susceptible to later metamorphic disturbance, as revealed by Sm-Nd isochron ages younger than the known timing of subaerial exposure (MacFarlane et al., 1994b). Correlation of REE enrichment and fractionation with increasing Al and albite alteration at Cooper Lake (Section 4.6) is taken here as evidence for a pedogenic origin of these signatures. The mobility of Y across the spectrum of incipient to extreme basalt weathering has been established for some time (Hill et al., 2000), but more recent studies have demonstrated Y/Ho fractionation may be powerful in tracing REE+Y mobility and scavenging processes in weathering profiles (Babechuk et al., 2015; Thompson et al., 2013) comparable to those in the marine environment (Bau, 1999; Nozaki et al., 1997). Existing experimental data indicates that preferential Ho>Y scavenging (or Y>Ho mobility) in soils should be favoured in the presence of Fe(III)- (oxyhydr)oxides, at intermediate pH, and under organic-poor conditions (Thompson et al., 2013). An evolution towards lower Y/Ho with increased alteration intensity has also been documented in basaltic saprolite and laterite (Babechuk et al., 2015; Thompson et al., 2013). The preservation of Y/Ho fractionation, albeit minor (24.42-26.93), at Cooper Lake implies pedogenic Y>Ho mobility at ca. 2.45 Ga. The role of organic matter in this Y/Ho fractionation cannot be assessed but is presumed to have been negligible. The Fe-depletion (Section 4.5) also suggests a negligible contribution from Fe(III)-oxide surfaces. The transition towards lower Y/Ho occurs progressively in the more altered and REE and Al-enriched areas of the dike (Figure 8c), and indicates that Ho>Y scavenging on secondary clay minerals (predominantly kaolinite) was the primary cause of fractionation. The greater magnitude of Y/Ho fractionation in oxidative Phanerozoic environments (e.g., reaching <20; Babechuk et al., 2015) suggests that paleosol Y/Ho fractionation has potential to track the appearance of pedogenic Fe(III)-oxides across the GOE. Nevertheless, the continental Y/Ho fractionation documented at Cooper Lake is also significant in that it indicates subaerial weathering may have contributed to elevated Y/Ho ratios in river waters, before further Y/Ho fractionation in estuaries and during particle scavenging by Fe- minerals in open seawater (Bau, 1999; Lawrence and Kamber, 2006; Planavsky et al., 2010), prior to the GOE.

Modern mafic weathering profiles have shown a reduction in the Eu anomaly correlated with proxies for plagioclase or clinopyroxene weathering (Babechuk et al., 2014; Banerjee et al., 2016), pointing to the preferential release of Eu into pore waters upon destruction of Eu(II)- bearing phases (e.g., Jin et al., 2017, for shale weathering). In paleosols, attributing Eu anomaly variations to pedogenesis can be more complicated, since hydrothermal (Alderton et al., 1980; Genna et al., 2014) and diagenetic (MacRae et al., 1992) fluids can also alter Eu anomalies. For example, complex Eu anomaly distributions are present in the ca 2.76 Mt. Roe basaltic paleosol known to have a younger disturbance of REE patterns (MacFarlane et al., 1994b). All factors are potentially relevant at Cooper Lake given the multiple episodes of feldspar alteration (Section 5.1.2). However, the correlation of Eu/Eu* with major element chemical weathering indices interpreted to record a pedogenic signature (Figure 8d) is adopted as evidence that Eu anomaly development occurred predominantly with albite and pyroxene alteration to clays during subaerial weathering. Upon release from minerals, reduced pore water conditions would have limited the oxidation of aqueous Eu(II) to Eu(III) and aided with its preferential removal to the other REE. The trend towards progressively more negative Eu anomalies at Cooper Lake (Figure 8d) contrasts with the younger, granitic Ville Marie paleosol higher in the Huronian Supergroup stratigraphy (Panahi et al., 2000). In the latter paleosol, variations in the Eu anomaly are more irregular and presumed by Panahi et al. (2000) to have been influenced by Eu(II) oxidation and greater scavenging subsequent to leaching from feldspar, which is reasonable given the sensitivity of Eu(II) oxidation to Eh at low temperatures (Sverjensky, 1984). Assuming a pedogenic origin of the negative Eu anomaly development at Cooper Lake, it is apparent that ancient siliciclastic sedimentary rocks derived from a highly weathered source (i.e., with significant feldspar decomposition to kaolinite) may possess a more negative Eu than their source rocks, especially those weathered and eroded prior to the GOE. Further, both preferential Eu-bearing mineral weathering and a more reducing atmosphere/weathering fluids ≥ 2.45 Ga may have contributed to an enhanced Eu flux from regolith into the hydrosphere. Teasing out this in situ weathering signature from the siliciclastic record, which also captures transport effects and a strong source composition control, is difficult (Gao and Wedepohl, 1995), but there is some supporting evidence from marine environments for continental weathering controlling marine REE budgets. Positive Eu anomalies of modern hydrothermal fluids and vent-proximal precipitates (Bau and Dulski, 1999; Peter et al., 2003) and a temporal correlation of iron formation Eu anomaly peaks with plume activity through time (Viehmann et al., 2015) are widely adopted as evidence for a hydrothermal-dominated origin of the positive Eu anomalies captured in ancient marine chemical precipitates. However, this applies most prominently to exhalative iron formation and an enhanced supply of Eu to marine basins via continental weathering has been proposed as a contributor to positive Eu anomalies since the earliest studies of trace elements in continental margin iron formation (Fryer, 1977). Shale-normalized REE pattern slopes (LREE/HREE depletion) and positive La-Y-Gd anomalies in marine chemical precipitates are strong arguments for continental REE+Y sources predominating over hydrothermal (assumed to exclude Eu). Importantly, the magnitude of Eu anomaly does not match the proportion of hydrothermal REE apparently required by Nd isotope modelling (Derry and Jacobsen, 1990) and this mass imbalance was noted by others to possibly be reconciled through the delivery of continental REE with a more positive Eu anomaly (Kamber, 2010). The Cooper Lake data strengthen the case for considering continental REE+Y fluxes from the subaerial weathering of mafic landmasses (and other Eu-enriched source rocks) as a contributor to the positive Eu/Eu* in the ancient oceans, especially in continental margin deposits.

5.4 Reduced weathering profiles as a continental source of soluble metals

5.4.1 Evidence for open system Fe mobility and a continental flux of isotopically light aqueous Fe

The Fe isotopic fractionation preserved in modern soils is related to incomplete reductive dissolution of Fe(III)-oxides or ligand-promoted dissolution of primary Fe-bearing minerals (e.g., Brantley, 2001; Yesavage et al., 2012), with rapid Fe(II) oxidation and removal of colloidal Fe(III) imparting only a minimal isotopic effect (Li et al., 2017; Poitrasson et al., 2008; Yesavage et al., 2016). The Fe speciation chemistry of the Cooper Lake saprolith eliminates Fe redox as a major control on Fe isotopic fractionation, leaving the selective removal of isotopically light, dissolved aqueous Fe(II) during incongruent ferromagnesian mineral weathering as the most parsimonious explanation for increasing δ56/54Fe (Figure 7). The lack of saprolith samples with δ56/54Fe isotopically lighter than the protolith (apart from sample CLA-011) can be considered evidence for minimal re-deposition of mobilized Fe(II) at depth, indicating an open system of Fe loss beyond the profile (Yamaguchi et al., 2007). To quantify the isotopic effect associated with Fe(II) removal, paleosol Fe/Al ratios are recast as a fraction of remaining Fe (fFe) (or gained when fFe>1) relative to the inferred protolith (OD) (Eq. 7, the fraction equivalent of the percentage loss of Eq. 1): fFe = 1 – [(Fe/Alprotolith – Fe/Alpaleosol)/Fe/Alpaleosol] (Eq. 7)

56/54 The δ Fe scatters near the protolith Fe isotopic composition (within 0.35-0.15 ‰) in samples where the Fe/Al is within ± 25% (fFe 0.75-1.25) of the OD sample, but increases systematically with the progressive decrease of fFe below 0.80 (>20% loss). The sub-sampled LCP (CLC-045L and CLC-048L) deviates from the trend with lower δ56/54Fe for the magnitude of Fe loss, suggesting heterogeneity at the finer smaller-scale. If the LCP and CLA-011 (with fFe of 1.27) samples are removed, a model fractionation factor (α) using a least-squared regression anchored to the isotopic composition of the protolith is 0.9998 (1000lnα of -0.20). This relationship is illustrated with Eq. 8 and graphically in Figure 7b.

56/54 56/54 3 (α-1) 3 δ Fepaleosol = (δ Feprotolith + 10 )f - 10 (Eq. 8)

The preferential release of isotopically light Fe(II) makes water-logged and anoxic soils, where incongruent and ligand-promoted dissolution processes dominate, the closest modern analogues (e.g., Brantley, 2001; Wiederhold et al., 2006). However, seafloor hydrothermal basalt alteration is also an appropriate analogue, where kinetic leaching of isotopically light Fe(II) is also the primary source of fractionation (Rouxel et al., 2003; Sharma et al., 2001) and apparent fraction factors between leached Fe(II) and the Fe source were reported to range from 0.5-1.3 ‰ 57/54 (δ FeIRMM-14) during alteration of oceanic crust in the Mariana Trench (Rouxel et al., 2003).

There are no other Fe isotopic data available at present from pre-GOE paleosols for comparison to Cooper Lake, but the direction of fractionation is similar to what is observed in some parts of the ca. 2.2 Ga Hekpoort paleosol (Yamaguchi et al., 2007). The latter, post-GOE paleosol has a significantly more complicated architecture of Fe(II)/Fe(III) that includes both Fe-enriched and Fe-depleted zones (Beukes et al., 2002; Rye and Holland, 2000; Yamaguchi et al., 2007; Yang and Holland, 2003) and with isotopic compositions (Yamaguchi et al., 2007) that are difficult to explain without at least some partial oxidation of aqueous Fe(II) to Fe(III) (minimum of 20% assuming no fractionation during release of aqueous Fe(II) from a protolith basalt with δ56/54Fe of 0 ‰). Nevertheless, some fractionation could be attributed to the release of isotopically light Fe(II) in deeper, anoxic portions of the profile (Yamaguchi et al., 2007). Based on the Fe isotopic compositions of siliciclastic sedimentary rocks, Yamaguchi et al. (2005) suggested the influence of this process on the Fe isotopic composition of the hydrosphere was negligible. However, the Cooper Lake data are the first constraints offered from an anoxic paleosol and indicate the capacity for isotopically light aqueous Fe(II) to have been leached from weathering profiles prior to the GOE. Thus, Fe stable isotopic fractionation during incongruent Fe(II) release from ferromagnesian silicates may have been a common feature of basaltic weathering prior to the GOE. Based on the Cooper Lake data, a soluble Fe(II) flux to the hydrosphere may have been isotopically lighter than igneous rocks and also that physical erosion of chemically weathered rock could supply isotopically heavy Fe(II) particles to the hydrosphere.

Aqueous Fe(II) leached from saprolite pore waters is still considered to be an important contribution to the hydrosphere (e.g., Hao et al., 2017). Similar to the discussion on Eu/Eu* (Section 5.3), the potential implications of a soluble source of Fe(II) to rivers from the weathering of Fe(II)- rich mafic (and ultramafic) is considered qualitatively in the context of the better constrained marine signatures. The sources of Fe contributing to the ferruginous oceans and deposition of iron formation in the Precambrian have been debated for decades (e.g., see the summaries in Huston and Logan, 2004 and Lascelles, 2007). Early models favouring continental Fe(II) sources supplied by intense weathering under an anoxic atmosphere (e.g., Lepp and Goldich, 1964) have been widely superseded but models favouring oceanic/hydrothermal sources of Fe (see Bekker et al., 2010; Isley, 1995), based on arguments related to iron formation depositional rates and geochemical signatures that include mantle-like O and Sr isotopic compositions and prevalent positive Eu anomalies. While hydrothermal Fe sources were certainly required and probably dominated, the Cooper Lake paleosol data imply an additional flux of aqueous continental Fe(II) could have been important prior to the GOE and also that a positive Eu anomaly is not necessarily a pure indication for a hydrothermal source (Section 5.3), especially in continental margin deposits. The proportional influence of continental Fe(II) would have presumably been related to the volume of emerged vs. submerged crust and the percentage that was mafic-ultramafic in composition, in addition to a number of paleo-environmental factors that are beyond the scope of this contribution to model. Nevertheless, the ca. 2.5 Ga Dales Gorge continental margin iron formation provides a good example of ambiguity the original sources of elements, continental or seafloor (Pecoits et al., 2009), and where a greater contribution of continentally derived Fe(II) was favoured based on coupled Nd-Fe isotope variations (Li et al. (2015). The Nd-Fe isotope data implicated both a hydrothermal Fe(II) source (isotopically heavier Fe and more radiogenic Nd) and a continental source (isotopically lighter Fe and less radiogenic Nd), with an accompanying Eu anomaly in the latter. The continental Fe(II) source was interpreted to be dissimilatory iron reduction (DIR) of marine sedimentary Fe(III) minerals on the continental shelf (see also Konhauser et al., 2017), but the Cooper Lake paleosol supports the case for another potential continental source of isotopically light Fe(II) via riverine/groundwater discharge. The anoxic leaching and riverine export of isotopically light Fe(II) from subaerially exposed ferromagnesian minerals and variable mixing with aqueous REE sourced from more isotopically evolved sources prior to reaching the oceans could also account for the continental end-member signatures. Thus, subaerial weathering may be an overlooked source of continental Fe(II) contributing to the Fe budget (along with other elements) of continental margin iron formations and other shallow marine chemical sedimentary reservoirs (Eroglu et al., 2018; Eroglu et al., 2017). Free aqueous Fe(II) reaching open marine environments requires that Fe availability exceeded sulphide precipitation in continental environments, which is reasonable assuming that atmospheric S deposition and bacterial sulphate reduction in soils (Maynard et al., 2013) were outpaced by the Fe(II) released from ferromagnesian silicates.

5.4.2 Relative pedogenic mobility and flux of Ni-Co-Zn-Cr to the Early Paleoproterozoic oceans

The chemistry of paleosols can provide a direct examination of relative metal mobility during chemical weathering and, if mobile, help inform the potential continental weathering contributions to marine transition metals budgets. Overall, the Cooper Lake data extend the metal leaching documented in Archean paleosols such as the ca. 2.76 Ga Mt Roe paleosols (Yang et al., 2002) past the Archean-Proterozoic boundary (Murakami et al., 2016). Assuming similar paleo-weathering conditions throughout the Archean and early Proterozoic, the metal-leaching at Cooper Lake also provides support for the mantle-controlled production and subaerial exposure of high-Mg mafic-ultramafic landmasses exerting a strong control on the temporal trends of oceanic metal delivery (Haugaard et al., 2016; Konhauser et al., 2009; Konhauser et al., 2015; Pecoits et al., 2009). A more detailed consideration of the transition metals in paleosols can be found in Murakami et al. (2016), where a data compilation offers a more temporal perspective, although these discussions are not repeated here. Focus here is given to Co-Ni-Zn, as well as Cr due to its proposal to track low pH surface conditions in the early Proterozoic (Konhauser et al., 2011).

Several studies have exploited temporal variations of one or more of the abundances or ratios of Ni-Co-Zn-Cr in fine-grained siliciclastic rocks as a proxy for the evolving composition of the bulk continental crust exposed to weathering and erosion, under the assumption that the metals are poorly soluble and represent the exposed source lithologies (Condie, 1993; Large et al., 2018; Tang et al., 2016). For example, the study of Tang et al. (2016) pointed towards Ni/Co and Cr/Zn ratios as a record of changing proportions of Mg-rich (mafic-ultramafic) rocks contributing to terrigenous shales throughout the Archean-Paleoproterozoic.

In Cooper Lake, Co-Ni show a reasonably well-coupled behaviour with Ni/Co ratios in the paleosol having a mean of 1.21±0.41 (1sd), excluding the two uppermost samples with 3.13-3.79, that overlaps with the OD sample (1.03). Nevertheless, the most Fe-depleted samples trend towards slightly lower Ni/Co and the Nb-normalized trends of both (Section 4.1) provide evidence for their slight depletion of both metals (with Ni>Co) during chemical weathering. In terms of Ni, these data support mobility and continental supply of Ni to the oceans under anoxic conditions that is documented in iron formation (Baldwin et al., 2012; Konhauser et al., 2009). The slight depletion of Co, an important marine nutrient and potential redox tracer (Swanner et al., 2014), also supports model-derived interpretations that anoxic weathering and low S availability in the Archean aided with Co mobility during continental weathering and its aqueous bioavailability (Moore Eli et al., 2018).

The relative mobility of Cr and Zn at Cooper Lake is significantly decoupled, with near-complete immobility of Cr (Section 4.9 and Figure 5) contrasting with the extensive depletion of Zn (reaching ≥90%; Section 4.8 and Figure 9). Accordingly, the Cr/Zn ratios show variability that is correlated to proxies of Fe loss and indicate that anoxic subaerial weathering may promote significant Cr/Zn fractionation, at least in mafic rocks. This implies that caution may be required in assuming faithful source representation of Cr/Zn ratios in terrigenous sediments, where anoxic leaching of Zn would push trends towards higher Cr/Zn than the source. Moreover, the documented mobility of Zn indicates that a break from uniformitarian assumptions of a hydrothermal Zn supply outweighing continental Zn may need to be reconsidered throughout the Archean (Liu et al., 2016; Robbins et al., 2013; Scott et al., 2012).

Paleo-weathering recorded at Cooper Lake occurred within the ca. 2.48-2.32 Ga time frame interpreted by Konhauser et al. (2011) to be influenced by an enhanced, acid-driven flux of continental Cr(III) (aqueous Cr(III) or colloidal particles with adsorbed Cr(III)). This interpretation is based primarily on substantial authigenic Cr enrichments in iron formation that lack the isotopic fingerprint expected from a flux of Cr(VI) (Frei et al., 2009), necessitating an alternative mechanism of Cr mobilization. Konhauser et al. (2011) cite supporting evidence of Cr(III)-enriched continental runoff from variable Cr/Ti ratios in the sub-Matinenda Denison Mine paleosol (Gay and Grandstaff, 1980), a process requiring low pH (<4) conditions or perhaps ligand-aided Cr(III) mobility (Babechuk et al., 2018; Saad et al., 2017). The lack of Cr mobility in the Cooper Lake paleosol, developed contemporaneously (or near-contemporaneously) to the Denison Mine locality, cannot confirm acidic weathering conditions capable of solubilizing Cr(III) in this area. Given the HFSE scatter in the Denison Mine paleosol reported in Gay and Grandstaff (1980), it also remains possible that the Cr/Ti ratios reflect protolith heterogeneity or analytical scatter and thus should be re-examined. Further, it is also re-emphasized that detrital pyrite preservation in fluvial sequences of the Lower Huronian overlying sub-Matinenda paleosols also impart restrictions on the extent of near-surface oxidative sulphide weathering. Nevertheless, the marine Cr enrichments stand and it is entirely possible the requisite low pH conditions for Cr(III) mobility were generated locally from the weathering of more sulphide- and Cr-rich greenstone or in areas where the oxidative weathering preferentially initiated corrosion of exposed volcanogenic massive sulphide (VMS) deposits. The local surface exposure of a VMS deposit is indeed supported by the chemistry of detrital pyrite in the Mississagi Fm. (Long et al., 2011), but, as noted earlier, its presence as detrital grain suggests oxidative chemical corrosion was probably limited. If only deeper weathering conditions are recorded in the Cooper Lake saprolith, observations could indicate that acidity was very surface restricted, with pH buffered in groundwaters to the point of inhibiting Cr(III) solubility in deeper saprolite. Overall, the new results further support the geographically and/or spatially restricted nature of benthic oxygen production envisaged by Lalonde and Konhauser (2015) and noted in Section 5.2.3, as well as provide motivation to examine other early Proterozoic weathering profiles. However, it is also becoming apparent that the role of organic and inorganic ligands in aiding Cr(III) solubility beyond thermodynamic pH boundaries may have been underappreciated in earlier models of Precambrian Cr cycling (Babechuk et al., 2018; Saad et al., 2017), loosening requirements of low pH as the sole driver of continental Cr(III) mobility.

5.5 The origin and implications of isotopically light Cr in the Cooper Lake paleosol

Stable Cr isotopes have gained traction as a Precambrian paleo-redox proxy that provides a more direct tracer of Cr(III)-Cr(VI) redox reactions (Crowe et al., 2013; Frei et al., 2009) than elemental data alone (e.g., (Yang and Holland, 2003)). However, proxy application assumes redox reactions are implied by isotopic excursions beyond the igneous inventory with a δ53/52Cr of -0.124±0.101 ‰ (2sd; Schoenberg et al., 2008). The isotopically light Cr (fractionated to lower δ53/52Cr values than -0.225‰, the 2sd limit of the mean average igneous inventory) in the Cooper Lake saprolith would be interpreted under this model as evidence for isotopically heavy Cr(VI) removal. However, Cr immobility (constant Cr/Nb; Figure 6) and consistency of the Cr isotopic composition throughout the variably altered dike (-0.321 ± 0.038 ‰; Figure 5) are strong evidence that the Cr isotopic signature pre-dates any low-T alteration event.

If the dike stable Cr isotope composition is decoupled from weathering processes, what could have generated the isotopically light signature? Mantle chromite is isotopically heavier than bulk peridotite (Farkaš et al., 2013; Shen et al., 2015) and, given the low Cr abundance of the dike, fractional crystallization of chromite (along with olivine) may have driven the Cr isotope composition of the melt to lighter values. There is also now growing evidence for magmatic mechanisms capable of generating localized, large-magnitude stable Cr isotopic fractionation in the mantle. Xia et al. (2017) examined a comprehensive suite of mantle rocks and found an extreme range of δ53/52Cr in peridotites (-0.51 to 0.75 ‰) with very isotopically light δ53/52Cr in websterite veins (as low as -1.36 ‰). These variations were attributed to partial melting and kinetic effects associated with preferential diffusion of light Cr isotopes from olivine into melt, possibly related to Cr(II)/Cr(III) speciation (e.g., (Jollands et al., 2018), that metasomatically refertilized localized areas of the mantle. Other studies have also demonstrated conditions favourable for Cr mobility in the upper mantle (Klein-BenDavid et al., 2011; Watenphul et al., 2014) that are potentially important in terms of Cr isotope fractionation. Thus, the isotopically light Cr composition of the dike could have been inherited from low-degree partial melting of a refertilized mantle domain or, alternatively, through the kinetic diffusion of light Cr into the melt during its ascent through the crust. These scenarios of fraction crystallization or melting-driven isotopic fractionation in the area could be further tested through the Cr isotopic study of other Thessalon Fm. rocks. Nevertheless, the take home message from Cooper Lake is that the Cr isotope signature is an original magmatic feature of the dike, providing new evidence for non-redox effects inducing significant isotopic fractionation.

The interpretation of Cooper Lake Cr isotope compositions being decoupled from surficial redox leaves these data compatible with the pervasively anoxic conditions implied by Fe speciation and Mo-V-Ce elemental data (Section 5.2.1). However, this interpretation contrasts with the study of a ca. 2.96 Ga mafic paleosol by Crowe et al. (2013), where light Cr isotope excursions were attributed to loss of Cr(VI) and the Fe depletion (see Figure 6c) was attributed to a late paragenetic overprint. The Cooper Lake data imply that isotopically light Cr compositions in paleosols may not be an unambiguous fingerprint of oxidative Cr(VI) formation. It is therefore unclear if Cr isotopes in the ca. 2.96 Ga paleosol capture a transient episode of atmospheric oxygenation (Crowe et al., 2013) exceeding the levels at ca. 2.45 Ga or rather some other Cr isotope fractionating process. Further empirical and experimental studies will continue to refine the paleo-redox tracing capacity of stable Cr isotopes in pre-GOE environments. This is particularly important in view of the limited Cr isotopic variations in sedimentary reservoirs prior to the Mesoproterozoic-Neoproterozoic (Cole et al., 2016; Gilleaudeau et al., 2016; Planavsky et al., 2014b) and the recent studies that have proposed or demonstrated non-redox Cr isotopic effects (Albut et al., 2018; Babechuk et al., 2018; Babechuk et al., 2017; Konhauser et al., 2011; Saad et al., 2017; Wille et al., 2018). Younger post-GOE mafic paleosols such as the ca. 1.9 Ga exposure at Schreiber Beach, Ontario (Frei and Polat, 2013) show Fe retention and coupled mobility of Cr and V along with fractionation in δ53/52Cr and variations in the Ce anomaly, providing a more convincing case for oxidative weathering conditions and advocating a multi-proxy approach to interpreting paleo-redox signatures in paleosols. However, primary signatures may not always be preserved through post-depositional alteration and these effects also need to be evaluated case by case (Babechuk et al., 2017).

6. Conclusions A multi-proxy investigation of the ca. 2.45 Ga Cooper Lake paleosol at the base of the ca. 2.45-2.25 Ga Huronian Supergroup was undertaken on a suite of ~50 new samples, building upon previously published observations (Murakami et al., 2016; Sutton and Maynard, 1993; Utsunomiya et al., 2003) and allowing new atmospheric oxygenation hypotheses established from the contemporaneous marine record to be tested in a subaerial environment. The main observations and conclusions extracted from the new data are:

The complete immobility of several HFSE (Al, Ti, Nb, Ta, Zr, Hf, Th) indicates a lack of extreme pH conditions being reached during weathering and an absence of allochthonous input, contrasting with previous reports of Al-Ti fractionation in other sub-Matinenda paleosols that should be revisited in future studies applying high-precision HFSE measurements.

The coupled losses of Fe, Mg, Mn, Zn, Ni, Co and the immobility or lack of anomaly development for Cr, Ce, Mo, and V all point to pervasively anoxic weathering conditions, and provide new evidence for an early Proterozoic continental flux of the former elements to the hydrosphere.

Uranium shows evidence for slight mobilization, which would support the plausibility of a continental source of U to explain the minor U enrichments in contemporaneous early Proterozoic iron formations (Partin et al., 2013b). However, these data still require that PO2 did not exceed 1.5 x 10-3 PAL to reconcile oxidative U mobility with the preservation of detrital uraninite in overlying siliciclastic units (Johnson et al., 2014). Over-printing by younger U addition in some areas of the paleosol cannot be fully discounted.

The immobility of Cr does not directly support evidence from the marine iron formation record for acidic (pH˂4) subaerial weathering conditions enabling Cr(III) release from land at ca. 2.45 Ga (Konhauser et al., 2011). However, assuming a continental origin of the marine authigenic Cr in early Proterozoic iron formation, the paleosol observations invoke either highly temporal, spatial, or even lithological restrictions on acid-driven Cr(III) mobility.

Internally constant, but isotopically light stable Cr isotope compositions throughout variably weathered and Fe-leached parts of the saprolith provide new evidence that isotopically light stable Cr isotope signatures do not strictly originate from pedogenic Cr(III)-Cr(VI) redox reactions. At Cooper Lake, the isotopic signatures pre-dated chemical weathering and appear to have a high-T origin. Tentatively, the light isotopic signature is attributed to either fractional crystallization of a Cr-rich phase(s), partial melting of an isotopically heterogeneous mantle domain, or kinetic assimilation of Cr into the mafic melt during its residence and ascent in the crust. Similar non-redox signatures will need to be considered carefully in other Precambrian paleosols, in addition to potentially unusual pH or ligand-rich weathering conditions (Babechuk et al., 2018; Babechuk et al., 2017; Saad et al., 2017; Wille et al., 2018) prior to relying on the Cr isotope proxy for paleo-atmospheric information.

Assuming a pedogenic origin of REE+Y fractionation in the saprolith, a preferential depletion of Eu from the breakdown of pyroxene and plagioclase occurred along with subtle Y/Ho fractionation induced by preferential scavenging of Ho>Y. By extension, the aqueous continental flux from the weathering of mafic rocks (and probably other plagioclase-bearing lithologies) may have been characterized by a higher Y/Ho and positive Eu anomaly relative to the protolith.

Fractionation of stable Fe isotopes is documented and interpreted to have occurred during Fe(II) leaching from the dike, presumably associated with the incongruent weathering of ferromagnesian silicates (predominantly pyroxene). These are the first stable Fe isotope data published from a reduced-type paleosol. The residual enrichment of heavy iron isotopes in the paleosol provide new evidence that a continental flux with isotopically light aqueous Fe(II) may have been generated by the weathering of mafic landmasses. The relative contribution of such sources to oceanic Fe budgets should be explored further in future studies, as advocated in early studies of Precambrian iron formation (Fryer, 1977), and the mechanisms should be evaluated further through the study of Archean mafic paleosols.

A summary of the main geochemical signatures produced during the emplacement of the dike and during its subsequent subaerial weathering within its probable paleo-environmental context is presented in Figure 12. Even considering the potential isolation of the paleosol from a mildly oxic atmosphere (required from Cr-Mo-U data of contemporaneous marine deposits), the new paleosol results provide important insights into -3 the continental weathering behaviour of several redox-sensitive elements under pervasively anoxic conditions (atmospheric PO2<1.5x10 to accommodate the preservation of detrital uraninite and pyrite; Johnson et al., 2014). Disparities between marine and continental geochemical archives of oxidative weathering (Lalonde and Konhauser, 2015) remain and could be reconciled by considering low-Fe protoliths (e.g., granitoids) and sulphides as preferentially contributing to marine U and Mo inventories and high-Fe mafic-ultramafic protoliths preferentially contributing to marine transition metal inventories. The early rift basin setting of the Lower Huronian may also have been less conducive for oxygen production (at least production in marine environments) relative to other early Proterozoic sequences preserving IF. Nevertheless, a continental metal flux, supplied primarily by mafic substrates, may have persisted throughout the Archean and early Proterozoic until the point that free atmospheric O2 was high enough to promote complete or near-complete Fe(II) oxidation to Fe(III) (≤ ca. 2.25 Ga) even while accommodating an earlier flux of U and Mo, perhaps very locally, from more easily oxidized, low-Fe(II) substrates and sulphides. Future studies comparing multiple paleo-redox proxies in paleosols developed contemporaneously on contrasting substrates (e.g., (Driese et al., 2011)) will allow geological controls on oxidation potential and element fluxes to be further assessed. As data from other Archean-Proterozoic continental reservoirs is accumulated, increasingly tighter constrains on the spatial, temporal, lithological, and environmental conditions supporting continental oxidative element mobilization and changes of aqueous element fluxes through time will be established. Deciphering the chemical signatures of pedogenesis amidst a complex paragenesis of alteration in paleosols remains a challenge moving forward. However, this study provides a template for the combination of trace element, isotopic, and major element data that can add confidence that certain pedogenic signatures were retained through metasomatic and metamorphic overprinting.

Acknowledgements

Financial support to M.G.B. at the University of Tübingen and analyses for the study were provided by Deutsche Forschungsgemeinschaft (DFG) project SCHO 1071/7-1 to R.S. under the Schwerpunktprogramm 1833 “Building a Habitable Earth”. Use of the Stanford Synchrotron Radiation Lightsource (SSRL), SLAC National Accelerator Laboratory, is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515. M.G.B. was supported by an Ussher Fellowship at Trinity College Dublin for part of this research. T. Chevrier and C. Gordon are thanked for assistance in the field during sampling. R. Goodhue is thanked for his contribution to the XRD data. M. Easton and C. Mealin are thanked for discussion about Huronian geology and geochronology. D. Schöckle and E. O’Beirne are thanked for assistance with XRF results and C. McKenna is thanked for assistance with the trace element analyses. S. Webb at SSRL and M. Halama and C. Roach are thanked for assistance with µ-XRF analysis. M. Wille, S. Flaiz, and P. Kühn are thanked for assistance with the total C-N- S analyses. We thank the constructive comments of two anonymous reviewers, which resulted in a more comprehensive consideration of the data and interpretations.

Figure Captions

Figure 1: Geological sketch map (modified from (Sutton and Maynard, 1993; Utsunomiya et al., 2003) and location of the study area within Ontario, Canada (top left) and more detailed geological sketch map of the outcrop study area (modified from (Bennett, 1990)) showing the relationships between the sedimentary units (Livingstone Creek and Matinenda Fms.) and the mafic dike-hosted paleosol (bottom left). The two slightly offset sampling channels, the first (samples CLA-002 to CLA-013) closest to the unconformity and the second (samples CLB-015 to CLD- 056) at greater depth, are shown in the latter map. A sketch of the full profile showing the subdivided samples into D-N (yellow squares; samples closest to northern margin of the dike, CLB-015-CLC-033) and D-S (blue circles; samples closer to the centre and southern side of the dike, all remaining samples), as discussed in the text, shows the depth range covered and number of samples (right). The wedge of Livingstone Creek Fm. in the depth profile sketch is used to illustrate the proximity of samples to the dike margin. Note that “unconformity” is used here specifically for the contact with the Matinenda Fm. above the Cooper Lake dike. A sample from the outcrop further south of the profile, used as a protolith composition (CLFD-1/OD; pink circle), is illustrated on both the sketch map and profile.

Figure 2: Modal abundances (%) of the major mineral phases identified with XRD of bulk sample powders (quartz: qtz; muscovite: ms; chlorite: chl; microcline: mc; albite: ab) plotted vs. depth. Abundances were divided into arbitrary bins relevant to the range of individual phases. The horizontal stippled lines at ~120 cm and ~ 560 cm represent the depth of the sample channel offset and the break from the profile sensu stricto to the OD sample, respectively.

Figure 3: Stratigraphic trends of (a) immobile element geochemistry (concentration and ratio), (b) major elements (Ca, Na, K, Fe, Mg, Mn) normalized to Al, and (c) major elements (Si, P) and weathering indices (CIA, CIA-K, PIA, MIA(R), MIA(R)-K, MgI). In all figures the yellow squares represent the D-N samples closest to the dike-sandstone contact, blue circles represent the D-S samples closer to the dike centre, and the pink circle represents the OD sample, as in Figure 1. In (a) the stippled vertical line and shaded bar represent the mean ± 1sd calculated from all samples. In (b) and (c) the dotted vertical line is the mean composition of three drillcore samples from the area reported in previous studies (Sutton and Maynard, 1993; Utsunomiya et al., 2003) and the shaded area is the 1sd range of the mean calculated from the U6T volcanic samples reported in (Ketchum et al., 2013).

Figure 4: A-CN-K and A-CNK-FM ternary plots (left) showing trends of Na depletion and Fe+Mg depletion linked to the breakdown of albite and ferromagnesian silicates, respectively. Weathering and metasomatism vectors are shown with blue and pink arrows, respectively, in both ternary plots. The position of mineral compositions for muscovite (mus), illite (ill), plagioclase (pl) and chlorite (chl) are listed for reference and a tie-line is drawn between chlorite and muscovite in the A-CNK-FM plot to show that these minerals dominate the weathering progression. In accordance, scatter plots of Fe/Al and K/Al vs. Mg/Al (right) show that the weathering intensity is tracked with increasing K and decreasing Fe and Mg. Symbols for all paleosol samples are as per Figure 4. Shaded green and orange areas are show the composition of drillcore and Thessalon volcanic rocks (Unit 6), respectively, for comparison. The red crosses and green ticks represent the paleosol compositions reported in previous studies from the same outcrop (Sutton and Maynard, 1993; Utsunomiya et al., 2003). The pink hexagons in the A-CN-K plot are Livingstone Creek Fm. sandstone samples as plotted in Young et al. (2001).

Figure 5: Stratigraphic plots of redox sensitive element geochemistry broken into Cr-Fe (top) and Mo, V, W, and U (bottom). Error bars in the δ53/52Cr and δ56/54Fe plots are 2SE of the individual measurements. The vertical shaded area in the δ53/52Cr plot is the extent of the 2sd range of the average igneous inventory reported in Schoenberg et al. (2008). Symbols are the same as Figure 1, with the exception of the Fe/Al plot, where the colours change to reflect ferrous iron (green), ferric iron (brown), and total iron (orange).

Figure 6: Scatter plots of Fe(II)/Al vs. Fe(III)/Al (after (Ohmoto, 1996)) for the (a) Cooper Lake paleosol, (b) ca. 1.85 Ga Flin Flon paleosol, and (c) ca. 2.96 Ga Nsuze paleosol. The data for the latter two paleosols are from Babechuk and Kamber (2013) and Crowe et al. (2013), respectively. The expected trends on the diagram generated by fully oxidized (O) paleosols, fully reduced (R) paleosols, hydrothermally influenced (H) paleosols, and those recording a mixed signature (M) of partial reduction to an initially oxidized substrate (M) are indicated. The Cooper Lake paleosol falls on a horizontal trend consistent with a reduced paleosol showing no evidence for hydrothermal Fe(III) reduction. The Flin Flon paleosol samples fall on a line with a negative slope that passes through the protolith, consistent with near-complete Fe(II) oxidation without subsequent Fe(III) reduction. The Nsuze paleosol has one fully oxidized sample, several that plot on a near-horizontal array to higher Fe(II)/Al (suggesting Fe(II) redistribution), but also several that plot close to the origin below a horizontal array from the protolith. The latter are consistent with hydrothermal Fe(III) reduction.

56/54 Figure 7: Stable Fe isotope trends in the paleosol. A positive correlation of the δ Fe with the MIA(R)-K tracking Fe-Mg loss (a) links the isotopic variation to ferromagnesian silicate weathering and the correlated enrichment in heavy Fe isotopes with Fe loss suggests these processes are coupled (b). Error bars in (a) are 2se on the δ56/54Fe measurement. The solid line in (b) corresponds to an isotopic fractionation of 1000lnα of - 0.20 as determined from a regression through the δ56/54Fe data excluding the LCP samples and one sample with Fe gain (collectively identified with a cross through the symbols). The shaded area around the regression line represents an envelope of ± 0.10 on the 1000lnα value. The position intersecting stippled lines represent the Fe depletion and Fe isotopic data from unpublished data from the Cooper Lake paleosol (Sreenivas et al., 2008).

Figure 8: REE geochemistry of the Cooper Lake paleosol. (a) CI-chondrite normalized REE plot shows the overlap of most samples with the U6T volcanic samples and broadly parallel patterns throughout the full paleosol sequence. (b/e) Ce-Pr anomaly plots showing the absent Ce anomaly in the paleosol and comparison to a modern oxidized basaltic saprolite. (c) ΣREE+Y vs. Y/Ho showing lower Y/Ho developing with increasing REE concentration. (d) Mg/Al vs. Eu anomaly showing the progressive decrease of Eu anomaly with loss of Mg.

Figure 9: Transition metal geochemistry of the Cooper Lake paleosol. Niobium-normalized Zn, Ni, Co, Cu stratigraphic trends (top) and box- whisker plots showing the aforementioned ratios (recast to a % change relative to the OD sample) along with Fe/Al and Mg/Al. The bars illustrate the mean and range (1st and 3rd quartiles), whereas whiskers extend to the highest and lowest sample values. Three separate bins with their own box-whiskers are divided by depth into samples <300 cm (n=26; green bar), between 300-500 cm (n=18; blue bar), and >500 cm (n=5; yellow bar). The horizontal red lines show the position of the mean composition of three drillcore samples from the area and the vertical red bars show the range in paleosol data reported in Murakami et al. (2016).

Figure 10: Synchrotron µ-XRF maps obtained on thin sections from three different depths in the paleosol. Tricolor maps of Fe-S-Cr, Ti-Cr, and K-

Mn-Cr show the association of Fe-Mn-Cr (chlorite), Fe-S (sulphide), and K-Cr (mica) throughout the weathering progression (see Section 4.12 for more details). PPL = plane polarized light; XPL = cross polarized light.

Figure 11: (a) Plot of %U/Nb vs. %Th/Nb (relative to the OD sample) illustrates the decoupling of U enrichment from Th relative to Nb. The shaded areas indicate ± 10% ratio change from the OD sample. The samples encircled by the dotted represent those extracted from beneath overburden at the foot of the outcrop. (b) Plot of U vs. Th/U plot for all Cooper Lake paleosol samples compared to the ca. 2.96 Ga Nsuze paleosol. The solid grey line and shaded area represent the mean and 2sd of the U6T volcanic rocks (Ketchum et al., 2013) that are the closest geochemical match to the dike. The average Th/U ratio of Archean and Proterozoic cratonic shale of Condie (1993) are also plotted for reference.

Figure 12: Summary plot of the main geochemical constraints, geological setting, and interpreted paleo-environment during weathering and sedimentation in the Lower Huronian. (A) Tectonic setting of the Lower Huronian near ca. 2.45 Ga (adapted from Young, 2015) showing early continental rifting, sedimentation, and plume-related magmatic activity. The stable Cr isotope composition of the dike hosting the paleosol is interpreted to have a magmatic origin, including the two possible mechanisms demonstrated in panels B and C. (B) Crystal fractionation of chromite (and possibly also olivine) preferentially removing heavier stable isotopes of Cr to generate the light isotope composition of the residual magma. (C) Development of isotopically light Cr isotope composition in the magma through the incorporation of partially melted mantle domains with lower δ53/52Cr than the average igneous inventory (Xia et al., 2017). (D) Reconstruction of the surface environment at ca. 2.45 Ga showing weathered and eroded remnants of continental flood basalts and their feeder dikes during sedimentation. (E) Summary of key geochemical findings related to the residual weathered mafic rock, as preserved in the studied Cooper Lake paleosol. (F) Summary of key geochemical findings related to the elements removed from the weathered mafic rock and presumably contributing to the soluble flux to the hydrosphere.

References

Albut, G., Babechuk, M.G., Kleinhanns, I.C., Benger, M., Beukes, N.J., Steinhilber, B., Smith, A.J.B., Kruger, S.J., Schoenberg, R., 2018. Modern rather than Mesoarchaean oxidative weathering responsible for the heavy stable Cr isotopic signatures of the 2.95 Ga old Ijzermijn iron formation (South Africa). Geochimica et Cosmochimica Acta 228, 157-189.

Alderton, D.H.M., Pearce, J.A., Potts, P.J., 1980. Rare earth element mobility during granite alteration: Evidence from southwest England. Earth and Planetary Science Letters 49, 149-165.

Anbar, A.D., Duan, Y., Lyons, T.W., Arnold, G.L., Kendall, B., Creaser, R.A., Kaufman, A.J., Gordon, G.W., Scott, C., Garvin, J., Buick, R., 2007. A Whiff of Oxygen Before the Great Oxidation Event? Science 317, 1903.

Anders, E., Grevesse, N., 1989. Abundances of the elements: Meteoritic and solar. Geochimica et Cosmochimica Acta 53, 197-214.

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, deep catchment. Geological Society of America Bulletin 114, 1143-1158.

Arnórsson, S., Óskarsson, N., 2007. Molybdenum and tungsten in volcanic rocks and in surface and <100°C ground waters in Iceland. Geochimica et Cosmochimica Acta 71, 284-304.

Babechuk, M.G., Kamber, B.S., 2013. The Flin Flon paleosol revisited. Canadian Journal of Earth Sciences 50, 1223-1243.

Babechuk, M.G., Kamber, B.S., Greig, A., Canil, D., Kodolányi, J., 2010. The behaviour of tungsten during mantle melting revisited with implications for planetary differentiation time scales. Geochimica et Cosmochimica Acta 74, 1448-1470.

Babechuk, M.G., Kleinhanns, I.C., Reitter, E., Schoenberg, R., 2018. Kinetic stable Cr isotopic fractionation between aqueous Cr(III)-Cl-H2O complexes at 25 °C: Implications for Cr(III) mobility and isotopic variations in modern and ancient natural systems. Geochimica et Cosmochimica Acta 222, 383-405.

Babechuk, M.G., Kleinhanns, I.C., Schoenberg, R., 2017. Chromium geochemistry of the ca. 1.85 Ga Flin Flon paleosol. Geobiology 15, 30-50.

Babechuk, M.G., Widdowson, M., Kamber, B.S., 2014. Quantifying chemical weathering intensity and trace element release from two contrasting basalt profiles, Deccan Traps, India. Chemical Geology 363, 56-75. Babechuk, M.G., Widdowson, M., Murphy, M., Kamber, B.S., 2015. A combined Y/Ho, high field strength element (HFSE) and Nd isotope perspective on basalt weathering, Deccan Traps, India. Chemical Geology 396, 25-41.

Baldwin, G.J., Turner, E.C., Kamber, B.S., 2012. A new depositional model for glaciogenic Neoproterozoic iron formation: insights from the chemostratigraphy and basin configuration of the Rapitan iron formation. Canadian Journal of Earth Sciences 49, 455-476.

Bali, E., Keppler, H., Audetat, A., 2012. The mobility of W and Mo in subduction zone fluids and the Mo–W–Th–U systematics of island arc magmas. Earth and Planetary Science Letters 351-352, 195-207.

Banerjee, A., Chakrabarti, R., Mandal, S., 2016. Geochemical anatomy of a spheroidally weathered diabase. Chemical Geology 440, 124-138.

Bau, M., 1999. Scavenging of dissolved yttrium and rare earths by precipitating iron oxyhydroxide: experimental evidence for Ce oxidation, Y-Ho fractionation, and lanthanide tetrad effect. Geochimica et Cosmochimica Acta 63, 67-77.

Bau, M., Alexander, B.W., 2009. Distribution of high field strength elements (Y, Zr, REE, Hf, Ta, Th, U) in adjacent magnetite and chert bands and in reference standards FeR-3 and FeR-4 from the Temagami iron-formation, Canada, and the redox level of the Neoarchean ocean. Precambrian Research 174, 337-346.

Bau, M., Dulski, P., 1996. Distribution of yttrium and rare-earth elements in the Penge and Kuruman iron-formations, Transvaal Supergroup, South Africa. Precambrian Research 79, 37-55.

Bau, M., Dulski, P., 1999. Comparing yttrium and rare earths in hydrothermal fluids from the Mid-Atlantic Ridge: implications for Y and REE behaviour during near-vent mixing and for the Y/Ho ratio of Proterozoic seawater. Chemical Geology 155, 77-90.

Beard, B.L., Johnson, C.M., 2004. Fe Isotope Variations in the Modern and Ancient Earth and Other Planetary Bodies. Reviews in Mineralogy and Geochemistry 55, 319-357.

Beinlich, A., Austrheim, H., Mavromatis, V., Grguric, B., Putnis, C.V., Putnis, A., 2018. Peridotite weathering is the missing ingredient of Earth’s continental crust composition. Nature Communications 9, 634.

Bekker, A., 2014. Huronian Glaciation, in: Amils, R., Gargaud, M., Cernicharo Quintanilla, J., Cleaves, H.J., Irvine, W.M., Pinti, D., Viso, M. (Eds.), Encyclopedia of Astrobiology. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 1-8. Bekker, A., Holland, H.D., 2012. Oxygen overshoot and recovery during the early Paleoproterozoic. Earth and Planetary Science Letters 317-318, 295-304.

Bekker, A., Holland, H.D., Wang, P.L., Rumble Iii, D., Stein, H.J., Hannah, J.L., Coetzee, L.L., Beukes, N.J., 2004. Dating the rise of atmospheric oxygen. Nature 427, 117.

Bekker, A., Slack, J.F., Planavsky, N., Krapez, B., Hofmann, A., Konhauser, K.O., Rouxel, O.J., 2010. Iron Formation: The Sedimentary Product of a Complex Interplay among Mantle, Tectonic, Oceanic, and Biospheric Processes*. Economic Geology 105, 467-508.

Bennett, G., 1990. Stratigraphy of the Elliot Lake Group between Sault Ste. Marie and the Quirke Lake Syncline, in: Fenwick, K.G., Giblin, P.E., Pitts, A.E. (Eds.), Report of Activities 1989 Resident Geologists. Ministry of Northern Development and Mines.

Bennett, G., Dressler, B.O., Robertson, J.A., 1991. The Huronian Supergroup and Associated Intrusive Rocks, in: Thurston, P.C., Williams, H.R., Sutcliffe, R.H., Stott, G.M. (Ed.), Geology of Ontario, pp. 549-591.

Beukes, N.J., Dorland, H., Gutzmer, J., Nedachi, M., Ohmoto, H., 2002. Tropical laterites, life on land, and the history of atmospheric oxygen in the Paleoproterozoic. Geology 30, 491-494.

Brantley, S.L., Liermann, L., Bullen, T.D., 2001. Fractionation of Fe isotopes by soil microbes and organic acids. Geology 29, 535-538.

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.

Canfield, D.E., Ngombi-Pemba, L., Hammarlund, E.U., Bengtson, S., Chaussidon, M., Gauthier-Lafaye, F., Meunier, A., Riboulleau, A., Rollion-Bard, C., Rouxel, O., Asael, D., Pierson-Wickmann, A.-C., El Albani, A., 2013. Oxygen dynamics in the aftermath of the Great Oxidation of Earth’s atmosphere. Proceedings of the National Academy of Sciences 110, 16736-16741.

Chandler, F.W., 1980. Proterozoic Redbed Sequences of Canada. Geological Survey of Canada Bulletin 311, 53.

Chi Fru, E., Rodríguez, N.P., Partin, C.A., Lalonde, S.V., Andersson, P., Weiss, D.J., El Albani, A., Rodushkin, I., Konhauser, K.O., 2016. Cu isotopes in marine black shales record the Great Oxidation Event. Proceedings of the National Academy of Sciences 113, 4941.

Ciborowski, T.J.R., Kerr, A.C., Ernst, R.E., McDonald, I., Minifie, M.J., Harlan, S.S., Millar, I.L., 2015. The Early Proterozoic Matachewan Large Igneous Province: Geochemistry, Petrogenesis, and Implications for Earth Evolution. Journal of Petrology 56, 1459-1494. Cole, D.B., Reinhard, C.T., Wang, X., Gueguen, B., Halverson, G.P., Gibson, T., Hodgkiss, M.S.W., McKenzie, N.R., Lyons, T.W., Planavsky, N.J., 2016. A shale-hosted Cr isotope record of low atmospheric oxygen during the Proterozoic. Geology 44, 555-558.

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.

Corfu, F., Andrews, A.J., 1986. A U–Pb age for mineralized Nipissing diabase, Gowganda, Ontario. Canadian Journal of Earth Sciences 23, 107-109.

Cornell, R.M., Schwertmann, U., 2003. The Iron Oxides: Structure, Properties, Reactions, Occurences and Uses, 2nd ed. WILEY-VCH Verlag GmbH & Co., Weinheim.

Crowe, S.A., Dossing, L.N., Beukes, N.J., Bau, M., Kruger, S.J., Frei, R., Canfield, D.E., 2013. Atmospheric oxygenation three billion years ago. Nature 501, 535-538.

Cui, H., Kitajima, K., Spicuzza, M.J., Fournelle, J.H., Ishida, A., Brown, P.E., Valley, J.W., 2018. Searching for the Great Oxidation Event in North America: A Reappraisal of the Huronian Supergroup by SIMS Sulfur Four-Isotope Analysis. Astrobiology 18, 519-538.

D’Arcy, J., Babechuk, M.G., Døssing, L.N., Gaucher, C., Frei, R., 2016. Processes controlling the chromium isotopic composition of river water: Constraints from basaltic river catchments. Geochimica et Cosmochimica Acta 186, 296-315.

Derry, L.A., Jacobsen, S.B., 1990. The chemical evolution of Precambrian seawater: Evidence from REEs in banded iron formations. Geochimica et Cosmochimica Acta 54, 2965-2977.

Driese, S.G., Jirsa, M.A., Ren, M., Brantley, S.L., Sheldon, N.D., Parker, D., Schmitz, M., 2011. Neoarchean paleoweathering of tonalite and metabasalt: Implications for reconstructions of 2.69Ga early terrestrial ecosystems and paleoatmospheric chemistry. Precambrian Research 189, 1-17.

Duan, Y., Anbar, A.D., Arnold, G.L., Lyons, T.W., Gordon, G.W., Kendall, B., 2010. Molybdenum isotope evidence for mild environmental oxygenation before the Great Oxidation Event. Geochimica et Cosmochimica Acta 74, 6655-6668.

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

E.G-Farrow, C., Mossman, D.J., 1988. Geology of precambrian paleosols at the base of the huronian supergroup, elliot lake, Ontario, Canada. Precambrian Research 42, 107-139. Eggins, S.M., Woodhead, J.D., Kinsley, L.P.J., Mortimer, G.E., Sylvester, P., McCulloch, M.T., Hergt, J.M., Handler, M.R., 1997. A simple method for the precise determination of ≥ 40 trace elements in geological samples by ICPMS using enriched isotope internal standardisation. Chemical Geology 134, 311-326.

Eggleton, R.A., Foudoulis, C., Varkevisser, D., 1987. Weathering of basalt: changes in rock chemistry and mineralogy. Clays and Clay Minerals 35, 161-169.

Eickmann, B., Hofmann, A., Wille, M., Bui, T.H., Wing, B.A., Schoenberg, R., 2018. Isotopic evidence for oxygenated Mesoarchaean shallow oceans. Nature Geoscience 11, 133-138.

Eroglu, S., Schoenberg, R., Pascarelli, S., Beukes, N.J., Kleinhanns, I.C., Swanner, E.D., 2018. Open ocean vs. continentally-derived iron cycles along the Neoarchean Campbellrand-Malmani Carbonate platform, South Africa. American Journal of Science 318, 367-408.

Eroglu, S., Schoenberg, R., Wille, M., Beukes, N., Taubald, H., 2015. Geochemical stratigraphy, sedimentology, and Mo isotope systematics of the ca. 2.58–2.50Ga-old Transvaal Supergroup carbonate platform, South Africa. Precambrian Research 266, 27-46.

Eroglu, S., van Zuilen, M.A., Taubald, H., Drost, K., Wille, M., Swanner, E.D., Beukes, N.J., Schoenberg, R., 2017. Depth-dependent δ13C trends in platform and slope settings of the Campbellrand-Malmani carbonate platform and possible implications for Early Earth oxygenation. Precambrian Research 302, 122-139.

Farkaš, J., Chrastný, V., Novák, M., Čadkova, E., Pašava, J., Chakrabarti, R., Jacobsen, S.B., Ackerman, L., Bullen, T.D., 2013. Chromium isotope variations (δ53/52Cr) in mantle-derived sources and their weathering products: Implications for environmental studies and the evolution of δ53/52Cr in the Earth’s mantle over geologic time. Geochimica et Cosmochimica Acta 123, 74-92.

Fedo, C.M., Wayne Nesbitt, H., Young, G.M., 1995. Unraveling the effects of potassium metasomatism in sedimentary rocks and paleosols, with implications for paleoweathering conditions and provenance. Geology 23, 921.

Fedo, C.M., Young, G.M., Nesbitt, H.W., Hanchar, J.M., 1997. Potassic and sodic metasomatism in the Southern Province of the Canadian Shield: Evidence from the Paleoproterozoic Serpent Formation, Huronian Supergroup, Canada. Precambrian Research 84, 17-36.

Frei, R., Crowe, S.A., Bau, M., Polat, A., Fowle, D.A., Døssing, L.N., 2016. Oxidative elemental cycling under the low O2 Eoarchean atmosphere. Scientific Reports 6, 21058. Frei, R., Gaucher, C., Poulton, S.W., Canfield, D.E., 2009. Fluctuations in Precambrian atmospheric oxygenation recorded by chromium isotopes. Nature 461, 250-253.

Frei, R., Polat, A., 2013. Chromium isotope fractionation during oxidative weathering—Implications from the study of a Paleoproterozoic (ca. 1.9Ga) paleosol, Schreiber Beach, Ontario, Canada. Precambrian Research 224, 434-453.

Fryer, B.J., 1977. Rare earth evidence in iron-formations for changing Precambrian oxidation states. Geochimica et Cosmochimica Acta 41, 361- 367.

Gall, Q., 1992. Precambrian paleosols in Canada. Canadian Journal of Earth Sciences 29, 2530-2536.

Gao, S., Wedepohl, K.H., 1995. The negative Eu anomaly in Archean sedimentary rocks: Implications for decomposition, age and importance of their granitic sources. Earth and Planetary Science Letters 133, 81-94.

Gaschnig, R.M., Rudnick, R.L., McDonough, W.F., Kaufman, A.J., Hu, Z., Gao, S., 2014. Onset of oxidative weathering of continents recorded in the geochemistry of ancient glacial diamictites. Earth and Planetary Science Letters 408, 87-99.

Gay, A.L., Grandstaff, D.E., 1980. Chemistry and mineralogy of Precambrian paleosols at Elliot Lake, Ontario, Canada. Precambrian Research 12, 349-373.

Genna, D., Gaboury, D., Roy, G., 2014. Evolution of a volcanogenic hydrothermal system recorded by the behavior of LREE and Eu: Case study of the Key Tuffite at Bracemac–McLeod deposits, Matagami, Canada. Ore Geology Reviews 63, 160-177.

Gilleaudeau, G.J., Frei, R., Kaufman, A.J., Kah, L.C., Azmy, K., Bartley, J.K., Chernyavskiy, P., Knoll, A.H., 2016. Oxygenation of the mid-Proterozoic atmosphere: clues from chromium isotopes in carbonates. Geochemical Perspectives Letters 2, 178-187.

Govindaraju, K., 1989. 1989 Compilation of Working Values and Sample Description for 272 Geostandards. Geostandards Newsletter 13, 1-113.

Grandstaff, D.E., Edelman, M.J., Foster, R.W., Zbinden, E., Kimberley, M.M., 1986. Chemistry and mineralogy of Precambrian paleosols at the base of the Dominion and Pongola Groups (Transvaal, South Africa). Precambrian Research 32, 97-131.

Greber Nicolas, D., Mäder, U., Nägler Thomas, F., 2015. Experimental dissolution of molybdenum‐sulphides at low oxygen concentrations: A first‐order approximation of late Archean atmospheric conditions. Earth and Space Science 2, 173-180. Hahn, K.E., Turner, E.C., Babechuk, M.G., Kamber, B.S., 2015. Deep-water seep-related carbonate mounds in a Mesoproterozoic alkaline lake, Borden Basin (Nunavut, Canada). Precambrian Research 271, 173-197.

Hao, J., Sverjensky, D.A., Hazen, R.M., 2017. A model for late Archean chemical weathering and world average river water. Earth and Planetary Science Letters 457, 191-203.

Harder, H., 1972. The role of magnesium in the formation of smectite minerals. Chemical Geology 10, 31-39.

Harnois, L., 1988. The CIW index: A new chemical index of weathering. Sedimentary Geology 55, 319-322.

Haugaard, R., Ootes, L., Creaser, R.A., Konhauser, K.O., 2016. The nature of Mesoarchaean seawater and continental weathering in 2.85Ga banded iron formation, Slave craton, NW Canada. Geochimica et Cosmochimica Acta 194, 34-56. Heaman, L.M., 1997. Global mafic magmatism at 2.45 Ga: Remnants of an ancient large igneous province? Geology 25, 299-302.

Hill, I.G., Worden, R.H., Meighan, I.G., 2000. Yttrium: The immobility-mobility transition during basaltic weathering. Geology 28, 923-926.

Holland, H.D., 1984. The Chemical Evolution of the Atmosphere and Oceans. Princeton University Press, New Jersey.

Holland, H.D., 2011. Discovering the History of Atmospheric Oxygen, Frontiers in Geochemistry. John Wiley & Sons, Ltd, pp. 43-60.

Holland, H.D., Feakes, C.R., Zbinden, E.A., 1989. The Flin Flon Paleosol and the composition of the atmosphere 1.8 BYBP. American Journal of Science 289, 362-389.

Holland, H.D., Zbinden, E.A., 1988. Paleosols and the Evolution of the Atmosphere: Part I, in: Lerman, A., Meybeck, M. (Eds.), Physical and Chemical Weathering in Geochemical Cycles. Springer Netherlands, Dordrecht, pp. 61-82.

Huston, D.L., Logan, G.A., 2004. Barite, BIFs and bugs: evidence for the evolution of the Earth’s early hydrosphere. Earth and Planetary Science Letters 220, 41-55.

Isley, A.E., 1995. Hydrothermal Plumes and the Delivery of Iron to Banded Iron Formation. The Journal of Geology 103, 169-185.

Jin, L., Ma, L., Dere, A., White, T., Mathur, R., Brantley, S.L., 2017. REE mobility and fractionation during shale weathering along a climate gradient. Chemical Geology 466, 352-379. Jochum, K.P., Weis, U., Schwager, B., Stoll, B., Wilson Stephen, A., Haug Gerald, H., Andreae Meinrat, O., Enzweiler, J., 2016. Reference Values Following ISO Guidelines for Frequently Requested Rock Reference Materials. Geostandards and Geoanalytical Research 40, 333-350.

Johannesson, K.H., Tang, J., 2009. Conservative behavior of arsenic and other oxyanion-forming trace elements in an oxic groundwater flow system. Journal of Hydrology 378, 13-28.

Johnson, J.E., Gerpheide, A., Lamb, M.P., Fischer, W.W., 2014. O2 constraints from Paleoproterozoic detrital pyrite and uraninite. GSA Bulletin 126, 813-830.

Jollands, M.C., O'Neill, H.S.C., Van Orman, J., Berry, A.J., Hermann, J., Newville, M., Lanzirotti, A., 2018. Substitution and diffusion of Cr2+ and Cr3+ in synthetic forsterite and natural olivine at 1200–1500 °C and 1 bar. Geochimica et Cosmochimica Acta 220, 407-428.

Kamber, B.S., 2009. Geochemical fingerprinting: 40 years of analytical development and real world applications. Applied Geochemistry 24, 1074- 1086.

Kamber, B.S., 2010. Archean mafic–ultramafic volcanic landmasses and their effect on ocean–atmosphere chemistry. Chemical Geology 274, 19- 28.

Kanzaki, Y., Murakami, T., 2016. Estimates of atmospheric O-2 in the Paleoproterozoic from paleosols. Geochimica Et Cosmochimica Acta 174, 263-290.

Kendall, B., Creaser, R.A., Reinhard, C.T., Lyons, T.W., Anbar, A.D., 2015. Transient episodes of mild environmental oxygenation and oxidative continental weathering during the late Archean. Science Advances 1, e1500777.

Kendall, B., Reinhard, C.T., Lyons, T.W., Kaufman, A.J., Poulton, S.W., Anbar, A.D., 2010. Pervasive oxygenation along late Archaean ocean margins. Nature Geoscience 3, 647-652.

Ketchum, K.Y., Heaman, L.M., Bennett, G., Hughes, D.J., 2013. Age, petrogenesis and tectonic setting of the Thessalon volcanic rocks, Huronian Supergroup, Canada. Precambrian Research 233, 144-172.

Kimberley, M.M., Grandstaff, D.E., Tanaka, R.T., 1984. Topographic control on Precambrian weathering in the Elliot Lake uranium district, Canada. Journal of the Geological Society 141, 229. King, E.K., Thompson, A., Chadwick, O.A., Pett-Ridge, J.C., 2016. Molybdenum sources and isotopic composition during early stages of pedogenesis along a basaltic climate transect. Chemical Geology 445, 54-67.

King, E.K., Thompson, A., Hodges, C., Pett-Ridge, J.C., 2014. Towards Understanding Temporal and Spatial Patterns of Molybdenum in the Critical Zone. Procedia Earth and Planetary Science 10, 56-62.

Klein-BenDavid, O., Pettke, T., Kessel, R., 2011. Chromium mobility in hydrous fluids at upper mantle conditions. Lithos 125, 122-130.

Konhauser, K.O., Lalonde, S.V., Planavsky, N.J., Pecoits, E., Lyons, T.W., Mojzsis, S.J., Rouxel, O.J., Barley, M.E., Rosìere, C., Fralick, P.W., Kump, L.R., Bekker, A., 2011. Aerobic bacterial pyrite oxidation and acid rock drainage during the Great Oxidation Event. Nature 478, 369-373.

Konhauser, K.O., Pecoits, E., Lalonde, S.V., Papineau, D., Nisbet, E.G., Barley, M.E., Arndt, N.T., Zahnle, K., Kamber, B.S., 2009. Oceanic nickel depletion and a methanogen famine before the Great Oxidation Event. Nature 458, 750-753.

Konhauser, K.O., Planavsky, N.J., Hardisty, D.S., Robbins, L.J., Warchola, T.J., Haugaard, R., Lalonde, S.V., Partin, C.A., Oonk, P.B.H., Tsikos, H., Lyons, T.W., Bekker, A., Johnson, C.M., 2017. Iron formations: A global record of Neoarchaean to Palaeoproterozoic environmental history. Earth-Science Reviews 172, 140-177.

Konhauser, K.O., Robbins, L.J., Pecoits, E., Peacock, C., Kappler, A., Lalonde, S.V., 2015. The Archean Nickel Famine Revisited. Astrobiology 15, 804-815.

König, S., Münker, C., Schuth, S., Garbe-Schönberg, D., 2008. Mobility of tungsten in subduction zones. Earth and Planetary Science Letters 274, 82-92.

Koppi, A.J., Edis, R., Field, D.J., Geering, H.R., Klessa, D.A., Cockayne, D.J.H., 1996. Rare earth element trends and cerium-uranium-manganese associations in weathered rock from Koongarra, Northern Territory, Australia. Geochimica et Cosmochimica Acta 60, 1695-1707.

Krogh, T.W., Davis, D.W., Corfu, F., 1984. Precise U-Pb zircon and badelleyite ages for the Sudbury area, in: Pye, E.G., Naldrett, A.J., Giblin, P.E. (Ed.), The Geology and Ore Deposits of the Sudbury Structure. Ontario Geological Survey, pp. 431-446.

Kurzweil, F., Wille, M., Gantert, N., Beukes, N.J., Schoenberg, R., 2016. Manganese oxide shuttling in pre-GOE oceans – evidence from molybdenum and iron isotopes. Earth and Planetary Science Letters 452, 69-78. Kurzweil, F., Wille, M., Schoenberg, R., Taubald, H., Van Kranendonk, M.J., 2015. Continuously increasing δ98Mo values in Neoarchean black shales and iron formations from the Hamersley Basin. Geochimica et Cosmochimica Acta 164, 523-542.

Lalonde, S.V., Konhauser, K.O., 2015. Benthic perspective on Earth’s oldest evidence for oxygenic photosynthesis. Proceedings of the National Academy of Sciences 112, 995.

Land, J.S., Tsikos, H., Cousins, D., Luvizotto, G., Zack, T., 2018. Origin of red beds and paleosols in the Palaeoproterozoic Transvaal and Olifansthoek Supergroups of South Africa: provenance versus metasomatic controls. Geological Journal 53, 191-202.

Langmuir, D., 1978. Uranium solution-mineral equilibria at low temperatures with applications to sedimentary ore deposits. Geochimica et Cosmochimica Acta 42, 547-569.

Large, R.R., Mukherjee, I., Zhukova, I., Corkrey, R., Stepanov, A., Danyushevsky, L.V., 2018. Role of upper-most crustal composition in the evolution of the Precambrian ocean–atmosphere system. Earth and Planetary Science Letters 487, 44-53.

Lascelles, D.F., 2007. Black smokers and density currents: A uniformitarian model for the genesis of banded iron-formations. Ore Geology Reviews 32, 381-411.

Lawrence, M.G., Kamber, B.S., 2006. The behaviour of the rare earth elements during estuarine mixing—revisited. Marine Chemistry 100, 147- 161.

Lepp, H., Goldich, S.S., 1964. Origin of Precambrian iron formations. Economic Geology 59, 1025-1060.

Li, M., He, Y.-S., Kang, J.-T., Yang, X.-Y., He, Z.-W., Yu, H.-M., Huang, F., 2017. Why was iron lost without significant isotope fractionation during the lateritic process in tropical environments? Geoderma 290, 1-9.

Li, W., Beard, B.L., Johnson, C.M., 2015. Biologically recycled continental iron is a major component in banded iron formations. Proceedings of the National Academy of Sciences 112, 8193-8198.

Liu, X.M., Kah, L.C., Knoll, A.H., Cui, H., Kaufman, A.J., Shahar, A., Hazen, R.M., 2016. Tracing Earth’s O2 evolution using Zn/Fe ratios in marine carbonates. Geochemical Perspectives Letters 2, 24-34.

Logan, W.E., 1857. On the division of the Azoic rocks of Canada into Huronian and Laurentian. Proceedings of the American Association for the Advancement of Science 1857, 44-47. Long, D.G.F., 2004. The tectonostatigraphic evolution of the Huronian basement and the subsequent basin fill: geological constraints on impact models of the Sudbury event. Precambrian Research 129, 203-223.

Long, D.G.F., Ulrich, T., Kamber, B.S., 2011. Laterally extensive modified placer gold deposits in the Paleoproterozoic Mississagi Formation, Clement and Pardo Townships, Ontario. Canadian Journal of Earth Sciences 48, 779-792.

Lyons, T.W., Reinhard, C.T., Planavsky, N.J., 2014. The rise of oxygen in Earth/'s early ocean and atmosphere. Nature 506, 307-315.

Macfarlane, A.W., Danielson, A., Holland, H.D., 1994a. Geology and major and trace element chemistry of late Archean weathering profiles in the Fortescue Group, Western Australia: implications for atmospheric PO2. Precambrian Research 65, 297-317.

MacFarlane, A.W., Danielson, A., Holland, H.D., Jacobsen, S.B., 1994b. REE chemistry and Sm-Nd systematics of late Archean weathering profiles in the Fortescue Group, Western Australia. Geochimica et Cosmochimica Acta 58, 1777-1794.

MacRae, N.D., Nesbitt, H.W., Kronberg, B.I., 1992. Development of a positive Eu anomaly during diagenesis. Earth and Planetary Science Letters 109, 585-591.

Marsh, J.S., 1991. REE fractionation and Ce anomalies in weathered Karoo dolerite. Chemical Geology 90, 189-194.

Marx, S.K., Kamber, B.S., 2010. Trace-element systematics of sediments in the Murray–Darling Basin, Australia: Sediment provenance and palaeoclimate implications of fine scale chemical heterogeneity. Applied Geochemistry 25, 1221-1237.

Maxwell, J.A., 1968. Rock and Mineral Analysis. Inter-science, New York.

Maynard, J.B., 1992. Chemistry of Modern Soils as a Guide to Interpreting Precambrian Paleosols. The Journal of Geology 100, 279-289.

Maynard, J.B., Sutton, S.J., Rumble, D., Bekker, A., 2013. Mass-independently fractionated sulfur in Archean paleosols: A large reservoir of negative Delta S-33 anomaly on the early Earth. Chemical Geology 362, 74-81.

Middelburg, J.J., van der Weijden, C.H., Woittiez, J.R.W., 1988. Chemical processes affecting the mobility of major, minor and trace elements during weathering of granitic rocks. Chemical Geology 68, 253-273.

Moeller, K., Schoenberg, R., Grenne, T., Thorseth, I.H., Drost, K., Pedersen, R.B., 2014. Comparison of iron isotope variations in modern and Ordovician siliceous Fe oxyhydroxide deposits. Geochimica et Cosmochimica Acta 126, 422-440. Moore Eli, K., Hao, J., Prabhu, A., Zhong, H., Jelen Ben, I., Meyer, M., Hazen Robert, M., Falkowski Paul, G., 2018. Geological and Chemical Factors that Impacted the Biological Utilization of Cobalt in the Archean Eon. Journal of Geophysical Research: Biogeosciences 123, 743-759.

Mossman, D.J., Farrow, C.E.G., 1992. Paleosol and Ore-Forming Processes in the Elliot Lake District of Canada, in: Schidlowski, M., Golubic, S., Kimberley, M.M., McKirdy, D.M., Trudinger, P.A. (Eds.), Early Organic Evolution: Implications for Mineral and Energy Resources. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 67-75.

Murakami, T., Ito, J.I., Utsunomiya, S., Kasama, T., Kozai, N., Ohnuki, T., 2004. Anoxic dissolution processes of biotite: implications for Fe behavior during Archean weathering. Earth and Planetary Science Letters 224, 117-129.

Murakami, T., Matsuura, K., Kanzaki, Y., 2016. Behaviors of trace elements in Neoarchean and Paleoproterozoic paleosols: Implications for atmospheric oxygen evolution and continental oxidative weathering. Geochimica et Cosmochimica Acta 192, 203 -219.

Murakami, T., Sreenivas, B., Das Sharma, S., Sugimori, H., 2011. Quantification of atmospheric oxygen levels during the Paleoproterozoic using paleosol compositions and iron oxidation kinetics. Geochimica Et Cosmochimica Acta 75, 3982-4004.

Murakami, T., Utsunomiya, S., Imazu, Y., Prasad, N., 2001. Direct evidence of late Archean to early Proterozoic anoxic atmosphere from a product of 2.5 Ga old weathering. Earth and Planetary Science Letters 184, 523-528.

Neaman, A., Chorover, J., Brantley, S.L., 2005a. Element mobility patterns record organic ligands in soils on early Earth. Geology 33, 117-120.

Neaman, A., Chorover, J., Brantley, S.L., 2005b. Implications of the evolution of organic acid moieties for basalt weathering over geological time. American Journal of Science 305, 147-185.

Nedachi, Y., Nedachi, M., Bennett, G., Ohmoto, H., 2005. Geochemistry and mineralogy of the 2.45 Ga Pronto paleosols, Ontario, Canada. Chemical Geology 214, 21-44.

Nesbitt, H.W., 1979. Mobility and fractionation of rare earth elements during weathering of a granodiorite. Nature 279, 206.

Nesbitt, H.W., 1992. Diagenesis and metasomatism of weathering profiles, with emphasis on Precambrian paleosols, in: Martini, I.P., Chesworth, W. (Ed.), Weathering, Soils & Paleosols. Elsevier Science Publishers B.V., Amsterdam, pp. 127-152.

Nesbitt, H.W., MacRae, N.D., Kronberg, B.I., 1990. Amazon deep-sea fan muds: light REE enriched products of extreme chemical weathering. Earth and Planetary Science Letters 100, 118-123. 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.

Nesbitt, H.W., Wilson, R.E., 1992. Recent chemical weathering of basalts. American Journal of Science 292, 740-777.

Novoselov, A.A., de Souza Filho, C.R., 2015. Potassium metasomatism of Precambrian paleosols. Precambrian Research 262, 67-83.

Nozaki, Y., Zhang, J., Amakawa, H., 1997. The fractionation between Y and Ho in the marine environment. Earth and Planetary Science Letters 148, 329-340.

Ohmoto, H., 1996. Evidence in pre-2.2 Ga paleosols for the early evolution of atmospheric oxygen and terrestrial biota. Geology 24, 1135-1138.

Ohmoto, H., 1997. Evidence in pre-2.2 Ga paleosols for the early evolution of atmospheric oxygen and terrestrial biota: Reply. Geology 25, 858- 859.

Oze, C., Bird, D.K., Fendorf, S., 2007. Genesis of hexavalent chromium from natural sources in soil and groundwater. Proceedings of the National Academy of Sciences 104, 6544.

Pan, Y.M., Stauffer, M.R., 2000. Cerium anomaly and Th/U fractionation in the 1.85 Ga Flin Flon Paleosol: Clues from REE- and U-rich accessory minerals and implications for paleoatmospheric reconstruction. American Mineralogist 85, 898-911.

Panahi, A., Young, G.M., Rainbird, R.H., 2000. Behavior of major and trace elements (including REE) during Paleoproterozoic pedogenesis and diagenetic alteration of an Archean granite near Ville Marie, Québec, Canada. Geochimica et Cosmochimica Acta 64, 2199-2220.

Papineau, D., Mojzsis, S.J., Schmitt, A.K., 2007. Multiple sulfur isotopes from Paleoproterozoic Huronian interglacial sediments and the rise of atmospheric oxygen. Earth and Planetary Science Letters 255, 188-212.

Partin, C.A., Bekker, A., Planavsky, N.J., Scott, C.T., Gill, B.C., Li, C., Podkovyrov, V., Maslov, A., Konhauser, K.O., Lalonde, S.V., Love, G.D., Poulton, S.W., Lyons, T.W., 2013a. Large-scale fluctuations in Precambrian atmospheric and oceanic oxygen levels from the record of U in shales. Earth and Planetary Science Letters 369-370, 284-293.

Partin, C.A., Lalonde, S.V., Planavsky, N.J., Bekker, A., Rouxel, O.J., Lyons, T.W., Konhauser, K.O., 2013b. Uranium in iron formations and the rise of atmospheric oxygen. Chemical Geology 362, 82-90. Pavlov, A.A., Kasting, J.F., 2002. Mass-independent fractionation of sulfur isotopes in Archean sediments: strong evidence for an anoxic Archean atmosphere. Astrobiology 2, 27-41.

Pecoits, E., Gingras, M.K., Barley, M.E., Kappler, A., Posth, N.R., Konhauser, K.O., 2009. Petrography and geochemistry of the Dales Gorge banded iron formation: Paragenetic sequence, source and implications for palaeo-ocean chemistry. Precambrian Research 172, 163-187.

Peter, J.M., Goodfellow, W.D., Doherty, W., 2003. Hydrothermal Sedimentary Rocks of the Heath Steele Belt,Bathurst Mining Camp, New Brunswick: Part 2. Bulk and Rare Earth Element Geochemistry and Implications for Origin*, in: Goodfellow, W.D., McCutcheon, S.R., Peter, J.M. (Eds.), Massive Sulfide Deposits of the Bathurst Mining Camp, New Brunswick, and Northern Maine. Society of Economic Geologists.

Peters, D., Bretscher, A., John, T., Scambelluri, M., Pettke, T., 2017. Fluid-mobile elements in serpentinites: Constraints on serpentinisation environments and element cycling in subduction zones. Chemical Geology 466, 654-666.

Philippot, P., Ávila, J.N., Killingsworth, B.A., Tessalina, S., Baton, F., Caquineau, T., Muller, E., Pecoits, E., Cartigny, P., Lalonde, S.V., Ireland, T.R., Thomazo, C., van Kranendonk, M.J., Busigny, V., 2018. Globally asynchronous sulphur isotope signals require re-definition of the Great Oxidation Event. Nature Communications 9, 2245.

Pienaar, P.J., 1963. Stratigraphy, Petrology, and Genesis of the Elliot Group, Blind River, Ontario: Including the Uraniferous Conglomerate. Department of Mines and Technical Surveys, Canada.

Pinto, J.P., Holland, H.D., 1988. Paleosols and the evolution of the atmosphere; Part II, in: Reinhardt, J., Sigleo, W.R. (Eds.), Paleosols and Weathering Through Geologic Time: Principles and Applications. Geological Society of America.

Planavsky, N., Bekker, A., Rouxel, O.J., Kamber, B., Hofmann, A., Knudsen, A., Lyons, T.W., 2010. Rare Earth Element and yttrium compositions of Archean and Paleoproterozoic Fe formations revisited: New perspectives on the significance and mechanisms of deposition. Geochimica et Cosmochimica Acta 74, 6387-6405.

Planavsky, N.J., Asael, D., Hofmann, A., Reinhard, C.T., Lalonde, S.V., Knudsen, A., Wang, X., Ossa Ossa, F., Pecoits, E., Smith, A.J.B., Beukes, N.J., Bekker, A., Johnson, T.M., Konhauser, K.O., Lyons, T.W., Rouxel, O.J., 2014a. Evidence for oxygenic photosynthesis half a billion years before the Great Oxidation Event. Nature Geoscience 7, 283-286.

Planavsky, N.J., Reinhard, C.T., Wang, X.L., Thomson, D., McGoldrick, P., Rainbird, R.H., Johnson, T., Fischer, W.W., Lyons, T.W., 2014b. Low Mid- Proterozoic atmospheric oxygen levels and the delayed rise of animals. Science 346, 635-638. Poitrasson, F., Viers, J., Martin, F., Braun, J.-J., 2008. Limited iron isotope variations in recent lateritic soils from Nsimi, Cameroon: Implications for the global Fe geochemical cycle. Chemical Geology 253, 54-63.

Potts, P.J., Webb, P.C., 1992. X-ray fluorescence spectrometry. Journal of Geochemical Exploration 44, 251-296.

Prasad, N., Roscoe, S.M., 1996. Evidence of anoxic to oxic atmospheric change during 2.45-2.22 Ga from lower and upper sub-Huronian paleosols, Canada. Catena 27, 105-121.

Prudêncio, M.I., Braga, M.A.S., Gouveia, M.A., 1993. REE mobilization, fractionation and precipitation during weathering of basalts. Chemical Geology 107, 251-254.

Rainbird, R.H., Nesbitt, H.W., Donaldson, J.A., 1990. Formation and diagenesis of a sub-Huronian saprolith: Comparison with a modern weathering profile. The Journal of Geology 98, 801-822.

Rasmussen, B., Bekker, A., Fletcher, I.R., 2013. Correlation of Paleoproterozoic glaciations based on U–Pb zircon ages for tuff beds in the Transvaal and Huronian Supergroups. Earth and Planetary Science Letters 382, 173-180.

Reinhard, C.T., Lalonde, S.V., Lyons, T.W., 2013. Oxidative sulfide dissolution on the early Earth. Chemical Geology 362, 44-55.

Reinhard, C.T., Raiswell, R., Scott, C., Anbar, A.D., Lyons, T.W., 2009. A Late Archean Sulfidic Sea Stimulated by Early Oxidative Weathering of the Continents. Science 326, 713.

Retallack, G.J., 1992. How to Find a Precambrian Paleosol, in: Schidlowski, M., Golubic, S., Kimberley, M.M., McKirdy, D.M., Trudinger, P.A. (Eds.), Early Organic Evolution: Implications for Mineral and Energy Resources. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 16-30.

Robbins, L.J., Lalonde, S.V., Planavsky, N.J., Partin, C.A., Reinhard, C.T., Kendall, B., Scott, C., Hardisty, D.S., Gill, B.C., Alessi, D.S., Dupont, C.L., Saito, M.A., Crowe, S.A., Poulton, S.W., Bekker, A., Lyons, T.W., Konhauser, K.O., 2016. Trace elements at the intersection of marine biological and geochemical evolution. Earth-Science Reviews 163, 323-348.

Robbins, L.J., Lalonde, S.V., Saito, M.A., Planavsky, N.J., Mloszewska, A.M., Pecoits, E., Scott, C., Dupont, C.L., Kappler, A., Konhauser, K.O., 2013. Authigenic iron oxide proxies for marine zinc over geological time and implications for eukaryotic metallome evolution. Geobiology 11, 295-306.

Robinson, A., Spooner, E.T.C., 1982. Source of the detrital components of uraniferous conglomerates, Quirke ore zone, Elliot Lake, Ontario, Canada. Nature 299, 622-624. Roscoe, S.M., 1973. The Huronian Supergroup, a Paleoaphebian succession showing evidence of atmospheric evolution. Geological Association of Canada Special Paper 12.

Rosing, M.T., Frei, R., 2004. U-rich Archaean sea-floor sediments from Greenland – indications of >3700 Ma oxygenic photosynthesis. Earth and Planetary Science Letters 217, 237-244.

Rouxel, O., Dobbek, N., Ludden, J., Fouquet, Y., 2003. Iron isotope fractionation during oceanic crust alteration. Chemical Geology 202, 155-182.

Rye, R., Holland, H.D., 1998. Paleosols and the evolution of atmospheric oxygen: A critical review. American Journal of Science 298, 621-672.

Rye, R., Holland, H.D., 2000. Geology and geochemistry of paleosols developed on the Hekpoort Basalt, Pretoria Group, South Africa.

Saad, E.M., Wang, X., Planavsky, N.J., Reinhard, C.T., Tang, Y., 2017. Redox-independent chromium isotope fractionation induced by ligand- promoted dissolution. Nature Communications 8, 1590.

Saikkonen, R.J., Rautiainen, I.A., 1993. Determination of ferrous iron in rock and mineral samples by three volumetric methods. Bulletin of the Geological Society of Finland 65, 59-64.

Satkoski, A.M., Beukes, N.J., Li, W., Beard, B.L., Johnson, C.M., 2015. A redox-stratified ocean 3.2 billion years ago. Earth and Planetary Science Letters 430, 43-53.

Schoenberg, R., Merdian, A., Holmden, C., Kleinhanns, I.C., Haßler, K., Wille, M., Reitter, E., 2016. The stable Cr isotopic compositions of chondrites and silicate planetary reservoirs. Geochimica et Cosmochimica Acta 183, 14-30.

Schoenberg, R., von Blanckenburg, F., 2005. An assessment of the accuracy of stable Fe isotope ratio measurements on samples with organic and inorganic matrices by high-resolution multicollector ICP-MS. International Journal of Mass Spectrometry 242, 257-272.

Schoenberg, R., Zink, S., Staubwasser, M., von Blanckenburg, F., 2008. The stable Cr isotope inventory of solid Earth reservoirs determined by double spike MC-ICP-MS. Chemical Geology 249, 294-306.

Schuessler, J.A., Botcharnikov, R.E., Behrens, H., Misiti, V., Freda, C., 2008. Amorphous Materials: Properties, Structure, and Durability: Oxidation state of iron in hydrous phono-tephritic melts. American Mineralogist 93, 1493-1504.

Schulz, K.J., Cannon, W.F., 2007. The Penokean orogeny in the Lake Superior region. Precambrian Research 157, 4-25. Scott, C., Planavsky, N.J., Dupont, C.L., Kendall, B., Gill, B.C., Robbins, L.J., Husband, K.F., Arnold, G.L., Wing, B.A., Poulton, S.W., Bekker, A., Anbar, A.D., Konhauser, K.O., Lyons, T.W., 2012. Bioavailability of zinc in marine systems through time. Nature Geoscience 6, 125.

Sekine, Y., Tajika, E., Tada, R., Hirai, T., Goto, K.T., Kuwatani, T., Goto, K., Yamamoto, S., Tachibana, S., Isozaki, Y., Kirschvink, J.L., 2011. Manganese enrichment in the Gowganda Formation of the Huronian Supergroup: A highly oxidizing shallow-marine environment after the last Huronian glaciation. Earth and Planetary Science Letters 307, 201-210.

Sharma, M., Polizzotto, M., Anbar, A.D., 2001. Iron isotopes in hot springs along the Juan de Fuca Ridge. Earth and Planetary Science Letters 194, 39-51.

Sheldon, N.D., 2006. Precambrian paleosols and atmospheric CO2 levels. Precambrian Research 147, 148-155.

Shen, J., Liu, J., Qin, L., Wang, S.J., Li, S., Xia, J., Ke, S., Yang, J., 2015. Chromium isotope signature during continental crust subduction recorded in metamorphic rocks. Geochemistry, Geophysics, Geosystems 16, 3840-3854.

Sreenivas, B., Hirata, T., Murakami, T., 2008. Fe isotope compositions of 2.45 Ga Cooper Lake paleosol. Geochimica Et Cosmochimica Acta 72, 890-890.

Sutton, S.J., Maynard, J.B., 1992. Multiple alteration events in the history of a sub-Huronian regolith at Lauzon Bay, Ontario. Canadian Journal of Earth Sciences 29, 432-445.

Sutton, S.J., Maynard, J.B., 1993. Sediment-hosted and basalt-hosted regoliths in the Huronian Supergroup: role of parent lithology in middle Precambrian weathering profiles Canadian Journal of Earth Sciences 30, 60-76.

Sverjensky, D., A., Lee, N., 2010. The Great Oxidation Event and Mineral Diversification. Elements 6, 31-36.

Sverjensky, D.A., 1984. Europium redox equilibria in aqueous solution. Earth and Planetary Science Letters 67, 70-78.

Swanner, E.D., Bayer, T., Wu, W., Hao, L., Obst, M., Sundman, A., Byrne, J.M., Michel, F.M., Kleinhanns, I.C., Kappler, A., Schoenberg, R., 2017. Iron Isotope Fractionation during Fe(II) Oxidation Mediated by the Oxygen-Producing Marine Cyanobacterium Synechococcus PCC 7002. Environmental Science & Technology 51, 4897-4906.

Swanner, E.D., Planavsky, N.J., Lalonde, S.V., Robbins, L.J., Bekker, A., Rouxel, O.J., Saito, M.A., Kappler, A., Mojzsis, S.J., Konhauser, K.O., 2014. Cobalt and marine redox evolution. Earth and Planetary Science Letters 390, 253-263. Swanner, E.D., Wu, W., Schoenberg, R., Byrne, J., Michel, F.M., Pan, Y., Kappler, A., 2015. Fractionation of Fe isotopes during Fe(II) oxidation by a marine photoferrotroph is controlled by the formation of organic Fe-complexes and colloidal Fe fractions. Geochimica et Cosmochimica Acta 165, 44-61.

Tang, M., Chen, K., Rudnick, R.L., 2016. Archean upper crust transition from mafic to felsic marks the onset of plate tectonics. Science 351, 372.

Thompson, A., Amistadi, M.K., Chadwick, O.A., Chorover, J., 2013. Fractionation of yttrium and holmium during basaltic soil weathering. Geochimica et Cosmochimica Acta 119, 18-30.

Ulrich, T., Kamber, B.S., Woodhead, J.D., Spencer, L.A., 2010. Long-Term Observations of Isotope Ratio Accuracy and Reproducibility Using Quadrupole ICP-MS. Geostandards and Geoanalytical Research 34, 161-174.

Utsunomiya, S., Murakami, T., Nakada, M., Kasama, T., 2003. Iron oxidation state of a 2.45-Byr-old paleosol developed on mafic volcanics. Geochimica et Cosmochimica Acta 67, 213-221.

Viehmann, S., Bau, M., Hoffmann, J.E., Münker, C., 2015. Geochemistry of the Krivoy Rog Banded Iron Formation, Ukraine, and the impact of peak episodes of increased global magmatic activity on the trace element composition of Precambrian seawater. Precambrian Research 270, 165-180.

Watenphul, A., Schmidt, C., Jahn, S., 2014. Cr(III) solubility in aqueous fluids at high pressures and temperatures. Geochimica et Cosmochimica Acta 126, 212-227.

Weis, D., Kieffer, B., Maerschalk, C., Barling, J., de Jong, J., Williams, G.A., Hanano, D., Pretorius, W., Mattielli, N., Scoates, J.S., Goolaerts, A., Friedman, R.M., Mahoney, J.B., 2006. High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS. Geochemistry, Geophysics, Geosystems 7, n/a-n/a.

Wiederhold, J.G., Kraemer, S.M., Teutsch, N., Borer, P.M., Halliday, A.N., Kretzschmar, R., 2006. Iron Isotope Fractionation during Proton- Promoted, Ligand-Controlled, and Reductive Dissolution of Goethite. Environmental Science & Technology 40, 3787-3793.

Wille, M., Babechuk, M.G., Kleinhanns, I.C., Stegmaier, J., Suhr, N., Widdowson, M., Kamber, B.S., Schoenberg, R., 2018. Silicon and chromium stable isotopic systematics during basalt weathering and lateritisation: A comparison of variably weathered basalt profiles in the Deccan Traps, India. Geoderma 314, 190-204. Wille, M., Kramers, J.D., Nägler, T.F., Beukes, N.J., Schröder, S., Meisel, T., Lacassie, J.P., Voegelin, A.R., 2007. Evidence for a gradual rise of oxygen between 2.6 and 2.5Ga from Mo isotopes and Re-PGE signatures in shales. Geochimica et Cosmochimica Acta 71, 2417-2435.

Wille, M., Nebel, O., Van Kranendonk, M.J., Schoenberg, R., Kleinhanns, I.C., Ellwood, M.J., 2013. Mo–Cr isotope evidence for a reducing Archean atmosphere in 3.46–2.76Ga black shales from the Pilbara, Western Australia. Chemical Geology 340, 68-76.

Wisawapipat, W., Kretzschmar, R., 2017. Solid Phase Speciation and Solubility of Vanadium in Highly Weathered Soils. Environmental Science & Technology 51, 8254-8262.

Wu, W., Swanner, E.D., Kleinhanns, I.C., Schoenberg, R., Pan, Y., Kappler, A., 2017. Fe isotope fractionation during Fe(II) oxidation by the marine photoferrotroph Rhodovulum iodosum in the presence of Si – Implications for Precambrian iron formation deposition. Geochimica et Cosmochimica Acta 211, 307-321.

Xia, J., Qin, L., Shen, J., Carlson, R.W., Ionov, D.A., Mock, T.D., 2017. Chromium isotope heterogeneity in the mantle. Earth and Planetary Science Letters 464, 103-115.

Yamaguchi, K.E., Johnson, C.M., Beard, B.L., Beukes, N.J., Gutzmer, J., Ohmoto, H., 2007. Isotopic evidence for iron mobilization during Paleoproterozoic lateritization of the Hekpoort paleosol profile from Gaborone, Botswana. Earth and Planetary Science Letters 256, 577-587.

Yamaguchi, K.E., Johnson, C.M., Beard, B.L., Ohmoto, H., 2005. Biogeochemical cycling of iron in the Archean–Paleoproterozoic Earth: Constraints from iron isotope variations in sedimentary rocks from the Kaapvaal and Pilbara Cratons. Chemical Geology 218, 135-169.

Yang, W.B., Holland, H.D., 2003. The Hekpoort paleosol profile in Strata 1 at Gaborone, Botswana: Soil formation during the great oxidation event. American Journal of Science 303, 187-220.

Yang, W.B., Holland, H.D., Rye, R., 2002. Evidence for low or no oxygen in the late Archean atmosphere from the similar to 2.76 Ga Mt. Roe #2 paleosol, Western Australia: Part 3. Geochimica et Cosmochimica Acta 66, 3707-3718.

Yesavage, T., Fantle, M.S., Vervoort, J., Mathur, R., Jin, L., Liermann, L.J., Brantley, S.L., 2012. Fe cycling in the Shale Hills Critical Zone Observatory, Pennsylvania: An analysis of biogeochemical weathering and Fe isotope fractionation. Geochimica et Cosmochimica Acta 99, 18-38.

Yesavage, T., Stinchcomb, G.E., Fantle, M.S., Sak, P.B., Kasznel, A., Brantley, S.L., 2016. Investigation of a diabase-derived regolith profile from Pennsylvania: Mineralogy, chemistry and Fe isotope fractionation. Geoderma 273, 83-97. Young, G.M., 2015. Did prolonged two-stage fragmentation of the supercontinent Kenorland lead to arrested orogenesis on the southern margin of the Superior province? Geoscience Frontiers 6, 419-435.

Young, G.M., Long, D.G.F., Fedo, C.M., Nesbitt, H.W., 2001. Paleoproterozoic Huronian basin: product of a Wilson cycle punctuated by glaciations and a meteorite impact. Sedimentary Geology 141-142, 233-254.

Zbinden, E.A., Holland, H.D., Feakes, C.R., Dobos, S.K., 1988. The sturgeon falls paleosol and the composition of the atmosphere 1.1 Ga BP. Precambrian Research 42, 141-163.

Zhou, L., McKenna, C.A., Long, D.G.F., Kamber, B.S., 2017. LA-ICP-MS elemental mapping of pyrite: An application to the Palaeoproterozoic atmosphere. Precambrian Research 297, 33-55.

Table 1. Bulk rock major and trace element abundances and selected molar ratios for the Cooper Lake dike-hosted paleosol and adjacent sedimentary rocks

CL CL C- C- CL CL CL CL CL 04 04 Sa CL CL CL CL CL CL CL CL CL CL CL CL CL CL CL CL CL CL CL CL C- C- CL C- CL CL CL C- CL CL CL CL C- CL CL CL 5 CL CL 8 CL CL CL CL CL CL CL CL CL CL CL CL CL CL mpl CL- CL- A- A- A- A- A- A- A- A- A- A- A- A- A- B- B- B- B- B- B- B- 02 02 C- 03 C- C- C- 03 C- C- C- C- 04 C- C- C- LI C- C- LI C- C- C- C- D- D- D- D- E- F- F- F- F- - e MT LC 00 00 00 00 00 00 00 00 00 01 01 01 01 01 01 01 02 02 02 02 4/0 6/0 02 0/0 03 03 03 5/0 03 03 03 04 1/4 04 04 04 GH 04 04 GH 04 05 05 05 05 05 05 05 05 05 06 06 06 CL ID SS SS 2 3 4 5 6 7B 8A 8B 9 0 1 2 3 5 6 8A 0 1 3A 3B 25 28 9 31 2 3 4 36 7 8 9 0A 2 3 4 5 T 7 8 T 9 0 1 2 3 4 5 6 7 8 0 1 2 FD

De 22. 31. 89. 10 11 12 12 14 15 12 12 13 15 16 18 20 24 25 26 27 28 29 30 31 32 33 34 35 37 39 40 40 40 42 41 41 42 43 44 45 42 46 47 49 51 52 53 54 55 65 pth 5 5 47 60 73 5 97 6.5 1.5 0 9.5 2 1 0 4 6 6.5 9 9.5 4 5.5 8 4.5 6 8.5 9 9 7 4 5.5 2 2 9 5.5 1.5 9 9 1 5.5 5.5 4 7 2 3 4 4.5 7.5 3.5 0 0.5 9 9 9 0 (cm

)

Sa nds San Pal Pal Pal Pal Pa Pa Pa Pa Pa Pa Pa Pa Pal Pal Pal Pal Pal Pal Pal Pal Pa Pa Pa Pa Pal Pal Pa Pa Pa Pa Pa Pa Pa Pa Pa Pa Pal Pa Pa Pal Pa Pa Pa Pa Pa Pa Pa Pa Pal Pal Pal Pal Pal Roc ton dst eo eo eo eo leo leo leo leo leo leo leo leo eo eo eo eo eo eo eo eo leo leo leo leo eo eo leo leo leo leo leo leo leo leo leo leo eo leo leo eo leo leo leo leo leo leo leo leo eo eo eo eo eo dik k e one sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol sol e

Livi ngs ton Des Mat e crip ine Cre D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- D- LC D- D- LC D- D- D- D- D- D- D- D- D- D- D- D- D- tion nda ek S S S S S S S S S S S S S N N N N N N N N N N N N N S S S S S S S S S S P S S P S S S S S S S S S S S S S OD

unit

Si wt. 73. 76. 45. 47. 50. 45. 45. 47. 49. 54. 53. 45. 40. 42. 38. 51. 50. 50. 53. 50. 50. 51. 43. 45. 43. 41. 49. 49. 46. 50. 55. 56. 56. 57. 57. 55. 52. 49. 43. 52. 50. 47. 59. 56. 56. 55. 55. 55. 54. 52. 39. 44. 52. 53. 54. 44.

O2 % 20 92 66 04 31 25 10 33 81 29 10 26 09 66 93 53 35 75 22 09 08 31 09 78 70 60 67 57 21 38 24 66 57 05 18 12 17 38 41 22 58 40 34 29 55 64 67 70 52 04 78 25 45 25 82 83

Ti 0.1 0.1 2.2 2.0 2.0 1.7 1.5 1.5 1.4 1.4 1.4 1.4 1.4 1.3 1.7 2.0 2.0 2.0 1.7 1.9 1.8 1.9 1.5 1.4 1.3 1.5 1.8 2.1 1.2 1.2 1.2 1.2 1.2 1.2 1.1 1.1 1.1 1.2 1.6 1.2 1.2 2.5 1.1 1.1 1.2 1.2 1.2 1.1 1.2 1.2 1.7 1.8 1.8 1.9 1.8 1.6

O2 " 8 7 2 1 1 1 9 7 7 6 2 3 5 0 2 8 2 4 6 2 6 1 3 3 6 7 2 6 7 3 2 1 2 0 7 9 9 1 7 0 8 9 5 7 1 5 1 9 3 6 7 7 9 0 6 1

Al2 15. 11. 22. 21. 22. 19. 17. 16. 16. 15. 15. 16. 17. 16. 18. 22. 22. 22. 19. 20. 20. 20. 17. 15. 16. 18. 20. 20. 15. 14. 13. 13. 13. 13. 12. 13. 14. 15. 15. 14. 16. 18. 12. 13. 13. 14. 14. 13. 14. 14. 18. 20. 20. 20. 19. 16.

O3 " 53 41 16 28 72 24 62 36 19 71 59 38 00 41 48 92 18 51 27 75 24 82 24 84 06 01 00 45 95 58 62 42 17 05 84 94 52 33 98 91 48 20 74 78 73 24 10 95 06 70 78 50 75 21 06 93

Fe

2O 1.2 2.9 13. 14. 9.2 18. 19. 19. 18. 15. 16. 21. 25. 23. 24. 8.0 9.5 9.2 11. 12. 12. 11. 21. 19. 21. 22. 13. 12. 20. 17. 13. 13. 13. 13. 13. 15. 17. 16. 10. 17. 16. 8.0 13. 15. 14. 15. 15. 15. 15. 17. 22. 16. 9.8 9.8 9.8 20.

3(T) " 0 3 96 29 9 57 71 15 04 13 46 90 46 84 73 4 2 2 46 33 99 10 41 46 92 56 47 94 75 66 92 33 28 13 55 73 21 99 32 05 36 1 24 34 87 40 06 31 70 14 63 45 4 6 8 05

Fe 1.6 9.9 5.2 13. 14. 10. 12. 20. 20. 4.1 5.7 5.4 7.3 16. 18. 9.1 13. 10. 13. 12. 7.6 13. 5.1 11. 11. 11. 13. 18. 11. 6.1 6.5 14. O " 5 4 4 57 33 9 22 38 57 8 4 5 3 85 71 5 47 75 28 9 3 28 9 65 26 73 06 13 89 2 9 29

" Mn 0.0 0.0 0.1 0.0 0.0 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.1 0.0 0.0 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.1 O 6 7 0 9 7 1 1 1 0 0 1 1 2 1 2 8 7 9 9 8 9 8 2 3 2 1 9 9 2 2 2 2 2 2 2 1 1 2 7 1 0 3 0 0 1 0 0 1 0 1 2 9 8 8 8 0

Mg 0.7 1.1 3.6 3.6 2.5 4.3 4.9 4.6 4.4 3.7 3.9 5.3 6.2 5.8 6.0 2.3 2.7 2.7 3.1 3.4 3.5 3.0 5.7 5.2 5.8 6.0 3.8 3.6 5.5 4.6 3.6 3.4 3.4 3.3 3.5 4.2 4.7 4.7 3.0 4.8 4.6 2.4 3.7 4.4 4.2 4.4 4.3 4.4 4.5 4.9 6.4 5.0 3.1 3.1 3.0 5.8 O " 3 2 3 8 4 4 0 9 5 1 7 2 4 5 5 9 1 0 8 5 6 6 4 2 3 8 2 9 9 7 3 3 4 7 7 9 2 2 3 7 9 9 4 0 2 1 1 1 6 5 8 6 2 5 9 3

Ca 0.0 0.4 0.3 0.3 0.3 0.4 0.7 0.3 0.3 0.2 0.2 0.2 0.2 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.4 1.4 0.7 0.3 0.3 0.3 0.4 1.6 2.3 2.4 2.7 2.7 2.3 0.8 1.0 1.9 8.7 0.5 0.3 5.9 1.3 0.4 0.8 0.4 0.8 0.7 0.8 0.7 0.3 0.3 0.3 0.3 0.3 0.2 O " bdl 6 4 9 4 1 3 2 6 3 9 5 4 4 0 4 4 5 0 2 0 1 8 5 8 0 0 0 9 9 9 5 8 1 4 3 5 3 0 3 9 9 3 4 4 0 6 9 3 6 1 5 5 6 6 7

Na 0.1 0.4 0.0 0.0 0.0 0.0 0.1 0.5 1.1 2.0 2.0 1.1 0.4 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.7 0.4 0.3 0.0 0.0 1.5 2.3 3.3 3.3 3.2 3.1 2.9 2.5 1.9 1.2 0.1 1.8 1.5 0.0 2.4 2.1 2.6 2.4 2.5 2.4 2.2 1.6 0.0 0.0 0.0 0.0 0.1

2O " 2 9 9 7 9 8 2 5 2 1 5 4 1 2 9 9 9 7 8 8 8 0 1 0 0 7 7 4 5 3 8 6 5 2 0 8 8 2 0 5 9 3 1 0 5 4 7 7 5 6 8 8 8 4

K2 7.0 5.4 5.9 5.6 7.3 4.1 3.5 2.9 2.7 2.3 1.8 1.5 1.3 1.9 1.8 7.6 7.1 7.3 5.3 5.7 5.4 6.1 3.0 2.9 2.6 2.3 5.1 5.4 1.4 0.8 0.3 0.3 0.2 0.3 0.4 0.6 1.1 2.3 5.7 1.5 2.7 6.7 0.9 1.4 0.9 1.2 1.0 0.8 1.0 1.6 2.3 4.5 6.2 6.0 5.6 3.5 O " 3 4 9 0 4 1 9 1 8 6 6 8 8 0 6 6 7 0 9 8 5 6 2 5 9 6 4 5 9 0 8 6 9 4 0 6 0 7 6 0 1 7 4 3 4 1 5 9 4 0 3 8 5 5 2 3

P2 0.0 0.0 0.3 0.3 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.2 0.2 0.2

O5 " 2 6 4 1 7 2 0 9 9 8 7 8 8 8 2 7 7 7 3 5 4 4 9 7 8 2 4 4 8 6 6 5 5 5 5 7 6 5 9 6 6 5 5 6 6 6 6 6 5 6 2 8 9 8 9 1

LO 1.8 1.5 4.1 4.2 3.8 4.7 4.7 4.5 3.9 3.5 3.6 4.6 5.2 4.9 5.6 3.8 4.0 3.8 4.0 4.0 4.1 3.9 5.1 5.0 5.1 5.2 4.4 4.2 4.6 4.7 4.5 4.4 4.7 4.4 4.4 3.8 4.3 5.1 9.4 4.1 4.2 7.3 3.8 3.6 3.6 3.5 4.0 4.1 4.1 4.3 5.8 4.9 3.9 3.8 3.8 4.5 I " 2 3 9 1 8 4 6 7 6 3 3 5 5 7 2 0 6 0 3 8 9 4 3 7 8 2 9 9 5 4 3 7 0 3 0 2 2 1 8 4 3 1 7 0 1 7 3 0 1 7 0 3 5 4 4 4

To tal ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. 0.1 ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. 0.2 0.1 ≤0. ≤0. ≤0. ≤0. 0.3 0.4 0.5 0.5 0.5 0.4 0.1 0.2 0.3 1.8 ≤0. ≤0. 1.1 0.2 ≤0. 0.1 ≤0. 0.1 0.1 0.1 0.1 ≤0. ≤0. ≤0. ≤0. ≤0. ≤0. C " 10 10 10 10 10 10 10 2 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 9 5 10 10 10 10 3 9 1 7 5 8 5 0 9 6 10 10 6 5 10 5 10 6 4 6 3 10 10 10 10 10 10

To tal ≤0. ≤0. ≤0. ≤0. 0.3 0.3 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.4 0.1 0.1 0.1 0.1 0.2 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.2 0.1 0.2 0.1 0.1 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.1 0.1 0.0 0.2 0.0 S " 05 05 05 05 3 6 9 5 0 0 2 3 0 3 5 5 7 4 9 7 7 9 1 1 6 0 6 1 6 4 3 5 5 6 6 1 4 6 2 2 0 6 2 3 4 0 1 1 0 5 7 4 9 8 6 8

61 60 55 61 61 56 50 41 43 57 68 61 69 56 56 58 53 59 57 58 61 53 57 65 59 59 50 40 30 28 27 28 29 37 41 47 42 48 48 45 34 45 37 40 36 36 39 46 70 68 60 58 55 78 ng/ 127 164 94 58 35 81 83 53 71 53 51 18 15 98 42 44 88 10 46 31 75 42 51 24 61 96 56 02 02 59 36 12 90 64 43 83 24 32 37 57 80 63 73 34 57 70 67 97 31 33 47 50 03 99 35 58 Li g 40 80 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Be " 10 12 12 12 12 10 10 11 10 10 402 252 43 97 02 90 54 38 33 31 27 25 27 27 41 54 12 20 99 53 98 75 41 30 29 45 89 85 23 17 14 14 13 14 13 16 18 29 50 24 35 69 14 16 15 16 15 16 17 23 61 87 09 98 20 41 0 5 0 96 0 10 27 24 29 85 70 60 31 78 76 0 0 0 02 0 86 0 11 22 14 13 17 61 02 12 32 34 77 33 64 07 71 19 76 21 49 39 72 98 74 79 60 57 75 13 98 41 0 0 0 47

15 15 11 17 24 24 23 23 21 18 14 13 11 17 15 19 18 18 18 16 12 µg/ 36 33 42 26 66 40 84 13 62 87 30 31 14 45 43 43 33 34 32 36 13 53 28 20 31 33 56 39 90 65 83 38 30 99 70 96 74 75 49 49 89 83 24 45 83 61 99 66 19 28 39 37 64 10 Na g 594 566 8 4 9 9 7 71 97 0 0 16 60 56 6 9 0 8 9 5 7 2 94 72 46 95 1 2 0 0 0 0 0 0 0 0 0 84 5 0 0 5 0 0 0 0 0 0 0 0 0 7 2 2 4 12

76 68 69 59 53 55 48 44 45 46 46 43 68 75 70 73 62 65 65 67 49 41 37 52 70 92 38 41 37 34 35 35 34 36 35 42 62 37 44 89 32 32 31 33 33 34 34 38 65 67 69 71 71 55 ng/ 236 220 71 26 26 51 81 52 76 18 63 07 63 69 11 93 23 21 78 95 50 39 86 50 10 34 25 43 82 01 24 49 37 97 60 11 79 28 96 89 61 44 54 32 85 01 69 12 89 95 61 55 20 75 76 63 Sc g 5 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

13 12 12 10 99 97 90 89 86 87 89 80 10 12 12 12 10 11 11 12 94 87 83 96 11 13 76 74 72 71 71 72 69 71 70 72 10 72 75 15 68 71 72 76 72 71 73 77 11 11 11 11 11 10 90 43 46 45 00 74 37 10 75 99 92 19 71 96 68 71 88 88 34 08 02 57 34 14 22 22 33 77 93 49 82 18 45 73 73 94 14 40 81 83 74 19 87 42 63 79 24 23 05 82 95 82 53 16 897 810 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Ti " 000 700 00 00 00 00 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 00 0 0 0 0 00 00 0 0 0 0 0 0 0 0 0 0 00 0 0 00 0 0 0 0 0 0 0 0 00 00 00 00 00 00

66 60 64 59 52 49 45 41 41 44 45 41 62 65 65 66 55 59 64 60 48 40 40 52 61 78 47 37 32 32 34 32 30 32 37 43 50 36 44 64 30 30 30 31 32 34 33 37 58 61 62 62 60 50 140 127 10 45 97 56 65 20 48 10 29 10 42 92 03 73 01 22 70 50 08 41 10 70 09 61 68 96 01 93 74 81 70 92 98 11 93 15 77 27 44 35 65 20 46 26 67 04 84 23 59 37 23 07 06 02 V " 50 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

313 303 33 32 32 26 25 23 21 21 22 22 24 21 27 35 35 35 30 31 30 31 23 21 19 22 26 33 18 18 17 18 17 18 18 18 19 18 25 19 19 41 17 18 18 20 19 19 19 21 34 38 29 31 31 27 Cr " 50 50 52 09 08 24 10 55 62 62 16 92 48 36 78 77 00 10 47 62 84 06 85 50 89 98 98 09 84 14 56 12 08 83 16 38 04 64 86 47 61 72 44 82 69 81 74 02 88 55 83 93 91 62 25 24

µg/ 36 34 18 52 64 60 53 45 48 61 68 66 64 18 21 22 25 28 30 25 66 71 67 58 31 30 62 65 64 60 63 63 60 54 60 67 97 58 54 60 50 47 51 48 49 49 53 54 62 41 22 23 22 52 Mn g 14 72 4 9 9 0 0 9 4 5 1 9 2 0 1 6 4 7 3 0 6 0 1 7 1 5 4 4 1 1 2 9 9 1 8 8 9 8 4 2 2 6 4 1 1 3 0 6 0 7 9 1 3 2 6 1

25 28 79 96 91 67 56 47 50 75 83 86 11 41 84 37 45 44 52 57 70 64 80 74 66 79 78 57 60 65 72 12 10 85 12 85 99 69 56 10 52 61 58 51 56 58 52 63 58 67 53 38 47 50 ng/ 197 129 51 54 12 93 56 49 19 28 77 46 47 78 16 45 48 26 23 87 20 22 54 85 09 21 61 75 50 09 55 86 18 61 44 34 62 34 14 26 64 70 72 38 19 62 59 32 79 06 81 56 73 18 19 99 Co g 0 60 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 00 0 00 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0

96 89 63 13 15 11 10 81 88 12 14 12 13 56 65 61 76 70 73 58 11 10 11 12 75 72 10 82 64 60 56 54 50 58 69 79 60 72 74 51 47 59 58 60 57 62 66 79 11 86 39 37 38 52 603 143 70 38 80 60 37 81 73 14 10 16 59 88 98 14 87 72 18 47 29 41 92 49 55 18 03 48 61 31 37 45 68 05 30 68 33 00 60 72 69 48 83 12 02 90 96 18 35 60 29 42 41 95 99 63 Ni " 8 40 0 0 0 00 00 00 00 0 0 00 00 00 00 0 0 0 0 0 0 0 00 00 00 00 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0

Cu " 553 996 45 15 26 27 66 10 16 77 89 51 24 16 14 10 33 13 83 10 61 50 73 23 33 18 66 49 46 30 14 14 21 25 27 60 38 33 16 30 18 10 42 71 90 44 42 77 68 14 15 24 19 10 66 72 5 50 09 92 05 86 11 55 16 90 03 82 05 36 28 48 80 76 14 42 46 42 50 73 61 04 46 00 32 83 99 38 89 07 96 41 62 03 14 95 87 89 49 99 14 64 20 84 09 84 85 18 18 13 07 60 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 0

16 16 18 24 25 23 19 21 30 37 33 34 12 12 14 11 50 26 30 32 16 13 29 24 20 18 18 18 19 22 24 24 24 23 19 22 21 22 21 22 23 24 32 21 12 35 157 508 08 48 70 34 79 31 57 22 52 26 33 53 94 46 61 62 09 49 73 40 40 94 54 47 35 61 40 95 26 32 07 82 28 87 59 01 95 96 67 36 11 59 74 48 75 47 02 23 12 87 95 96 84 90 Zn " 1 2 0 0 41 0 0 0 0 0 0 0 0 0 0 77 67 57 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 65 0 0 92 0 0 0 0 0 0 0 0 0 0 47 59 0 0

42 39 36 29 27 24 23 20 21 25 30 27 35 35 34 35 31 33 32 33 25 21 23 31 35 36 22 20 17 16 16 16 17 19 20 21 21 20 21 27 16 18 18 18 18 18 18 20 34 37 35 34 32 28 202 129 85 47 90 43 34 67 15 78 43 83 27 17 79 55 96 59 42 45 25 01 49 94 31 05 52 41 72 34 28 28 51 74 03 75 50 58 97 54 59 24 33 56 25 67 37 66 98 70 45 19 27 42 08 56 Ga " 40 80 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

11 10 13 20 24 38 709 11 14 85 59 57 42 33 39 39 53 39 44 78 36 66 26 34 30 37 67 37 31 42 32 60 76 43 31 72 78 67 03 76 53 71 80 69 62 50 33 56 62 55 41 46 44 39 48 21 60 55 30 45 22 As " 479 2 14 54 25 36 84 20 71 27 84 55 46 16 82 69 51 88 76 17 44 07 06 70 76 19 38 79 29 32 0 0 0 0 92 75 95 29 0 85 51 0 91 98 00 26 04 91 30 90 03 07 89 51 96 80

27 25 33 18 11 76 62 64 48 25 21 21 88 36 33 34 27 26 26 28 67 37 21 81 23 25 15 10 12 17 50 15 29 72 23 11 20 12 24 33 32 30 15 277 204 73 73 78 78 78 27 50 66 34 15 05 30 61 26 84 57 73 83 17 78 00 51 08 24 80 57 56 82 62 59 75 71 86 12 09 90 77 06 30 99 98 89 76 86 75 81 99 62 03 33 45 07 14 08 Rb " 600 900 00 00 00 00 00 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 0 0 0 0 00 00 0 24 26 83 71 0 01 0 0 0 00 0 0 00 06 62 64 86 56 83 0 0 00 00 00 00 00 00

11 10 11 10 10 16 17 22 21 13 11 10 22 13 20 37 52 52 57 57 49 29 25 27 52 23 21 24 32 21 29 25 27 27 28 22 10 11 12 12 14 400 435 31 86 03 31 96 88 27 91 15 52 93 91 83 39 98 97 78 75 75 79 99 45 31 96 90 87 12 89 63 56 91 48 76 58 65 46 85 36 46 97 39 74 56 40 58 58 30 51 95 53 40 58 67 50 Sr " 60 20 0 0 0 0 0 0 0 0 0 0 49 97 30 0 37 68 66 64 24 63 0 0 0 81 90 93 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 16 0 0 0 0 0

53 42 39 33 30 28 29 28 26 24 27 25 39 43 38 40 26 30 38 33 32 27 33 44 44 39 22 22 24 22 23 23 23 23 39 19 25 22 18 38 22 22 21 21 21 20 22 22 31 45 52 44 33 34 636 895 57 09 38 33 23 76 68 48 52 66 85 96 54 71 41 45 55 24 75 33 55 40 85 22 29 66 05 97 47 57 21 38 70 03 64 41 88 55 15 38 97 71 37 61 33 61 31 68 84 76 55 33 93 07 Y " 9 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

23 21 21 17 16 16 15 15 14 14 14 13 17 21 20 21 18 20 19 20 16 15 14 16 19 22 12 12 12 12 12 12 11 12 12 12 17 12 12 26 11 12 12 13 12 12 12 13 18 19 19 20 19 17 983 113 77 47 31 81 84 69 17 09 53 63 86 51 81 93 91 53 43 44 51 48 08 07 40 47 24 37 95 93 61 15 15 34 75 42 54 39 04 55 69 94 97 18 81 22 89 46 96 19 58 84 95 11 54 08 Zr " 30 800 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

439 413 92 84 89 71 64 66 63 65 62 59 61 54 70 88 87 87 75 82 79 84 68 64 67 65 82 86 57 51 52 51 50 50 48 49 49 52 68 57 51 48 50 53 54 51 51 52 55 75 76 79 84 76 70 Nb " 4 6 29 69 74 10 15 09 99 63 91 36 25 89 09 58 01 42 53 16 80 57 38 21 14 35 23 11 78 06 44 00 52 57 57 35 80 20 84 00 74 10 60 48 16 55 66 74 56 54 20 40 89 19 24 28 87 0

26 22 49 58 48 43 36 37 38 28 59 48 31 27 36 37 30 37 28 38 40 32 57 56 39 38 29 26 30 27 24 17 15 20 21 24 52 27 25 75 28 26 23 22 19 17 24 27 15 31 31 19 13 32 Mo " 59 390 8 3 6 4 5 7 2 9 0 7 3 3 0 9 9 2 2 6 8 0 9 4 5 7 9 6 2 7 7 0 6 2 3 4 0 8 3 1 4 3 0 5 9 5 5 2 2 4 0 8 7 9 9 0

Cd " 28 34 79 71 68 64 54 54 47 46 51 51 53 47 56 80 62 68 65 62 75 72 56 50 54 54 59 83 48 42 44 47 44 44 38 38 43 45 59 39 40 86 41 42 39 46 42 40 40 42 69 66 64 67 61 56

11 11 14 14 13 13 11 12 13 14 15 14 15 11 15 13 16 16 19 12 In " 35 27 8 2 4 95 62 45 38 39 34 28 30 29 66 3 6 6 3 9 2 8 2 36 32 67 4 8 29 27 24 25 25 29 30 29 32 50 2 37 58 6 27 25 24 26 25 26 25 32 90 5 5 0 4 1

23 22 29 17 11 80 72 79 65 37 34 29 13 22 21 21 18 22 23 28 78 37 19 13 37 41 20 14 18 15 22 41 33 41 46 10 28 63 14 43 23 19 22 22 19 17 24 41 19 34 41 40 37 16 Sn " 755 690 58 06 98 56 75 9 6 7 7 3 5 9 21 46 54 79 56 92 80 12 9 8 9 24 75 21 3 0 2 6 3 1 0 5 8 48 36 9 16 61 7 9 8 3 6 8 4 1 99 17 88 41 05 46

25 26 42 47 37 32 34 34 34 34 32 33 41 30 35 31 37 33 35 37 37 31 35 34 36 39 33 28 30 33 31 33 36 36 43 38 70 36 38 11 37 35 37 30 30 36 35 40 37 49 49 42 52 37 Sb " 139 219 1 3 8 1 0 6 6 3 0 7 8 9 1 6 0 0 8 8 6 4 6 4 5 2 6 9 0 2 4 6 7 3 3 7 2 9 4 4 1 41 0 2 1 8 7 5 8 1 1 6 0 8 2 5

435 438 46 43 56 30 19 13 11 12 95 64 61 53 14 56 53 54 42 44 41 47 10 62 37 13 39 38 36 26 28 28 31 35 33 36 39 83 19 59 12 30 31 24 25 26 27 27 35 49 20 40 54 52 50 41 Cs " 9 8 91 96 22 40 47 66 93 21 6 2 7 4 74 49 22 25 91 97 49 12 24 3 2 72 56 13 3 2 8 9 5 8 0 2 0 3 98 7 46 45 8 7 9 0 2 8 1 0 34 30 72 87 67 64

58 54 71 41 27 19 18 18 15 85 75 74 29 77 70 71 53 56 54 61 16 10 68 21 53 60 55 41 46 45 58 76 83 96 96 17 39 99 20 52 50 48 45 48 43 46 52 65 27 48 65 63 59 45 878 723 66 23 62 51 40 05 06 99 75 98 66 05 94 14 80 24 52 93 42 58 89 69 15 10 97 11 55 34 22 64 24 73 14 11 59 14 27 18 19 27 60 81 03 18 32 34 27 41 42 95 02 83 85 84 Ba " 500 800 00 00 00 00 00 00 00 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 0 00 00 00 0 0 0 0 0 0 0 0 0 00 00 0 00 00 0 0 0 0 0 0 0 0 00 00 00 00 00 00

31 19 21 14 10 11 13 37 29 18 21 14 15 17 18 21 27 25 13 26 37 38 24 10 15 22 10 42 54 49 24 16 148 276 58 25 06 50 82 90 26 66 65 52 92 27 85 77 96 32 19 01 55 34 43 57 36 61 61 58 94 71 85 70 68 60 51 60 66 90 17 80 89 11 64 67 73 78 73 69 79 90 92 66 53 15 22 36 La " 20 00 0 0 0 0 0 22 0 66 73 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 09 0 08 29 13 98 99 15 65 95 0 45 87 0 76 70 10 43 55 81 41 51 0 0 0 0 0 0

71 43 48 33 25 20 25 14 14 30 85 66 42 48 33 35 38 41 48 62 55 29 58 83 86 56 20 24 19 15 15 13 11 13 14 18 31 17 19 47 13 13 15 16 15 14 16 18 22 92 12 10 54 38 336 565 12 98 13 32 58 71 59 88 56 75 76 19 37 71 77 14 56 72 67 63 76 00 04 46 67 99 26 08 28 80 15 52 15 05 22 67 17 21 82 14 67 89 51 21 67 68 54 62 02 49 24 84 27 07 Ce " 60 50 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 00 0 0

10 10 11 12 15 14 428 682 93 57 61 43 33 26 33 19 18 39 89 84 54 62 43 45 49 53 62 79 71 37 74 79 35 72 26 30 24 20 19 17 14 16 18 24 40 21 25 60 17 18 19 20 19 18 21 24 29 10 70 00 69 49 Pr " 0 8 01 32 50 29 17 77 11 33 85 26 0 28 32 42 49 41 87 28 35 15 53 44 15 0 0 99 35 94 82 26 84 39 44 96 39 07 04 97 47 48 49 22 68 87 88 64 22 08 93 0 0 0 86 50 38 23 25 17 13 11 13 15 43 33 22 25 18 18 20 21 25 31 28 15 29 43 45 29 10 12 10 10 16 10 25 12 48 62 56 28 20 170 252 50 76 49 87 89 33 88 82 80 97 73 84 45 50 07 88 24 75 44 85 54 16 50 11 29 69 87 71 36 86 82 74 62 71 78 08 87 91 52 56 73 75 81 85 82 78 87 99 32 00 98 30 46 54 Nd " 30 70 0 0 0 0 0 0 0 32 43 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 25 20 33 05 09 53 0 0 54 0 0 49 39 52 74 24 47 41 47 0 0 0 0 0 0

10 13 11 S 321 451 90 58 60 44 35 29 34 22 21 37 89 70 53 61 45 48 47 51 59 71 63 37 65 93 98 67 27 30 26 22 21 20 17 19 24 24 41 23 24 61 19 19 20 21 20 20 22 25 32 24 13 76 63 51 m " 8 9 26 61 30 15 39 80 33 80 89 21 20 49 96 16 88 24 49 87 94 01 25 18 71 90 25 73 37 49 36 51 91 48 83 58 76 54 26 19 98 74 62 91 94 52 68 08 47 18 67 0 0 0 69 64

111 24 16 15 12 10 89 10 73 70 11 25 20 17 16 12 13 13 14 17 19 19 12 20 26 26 18 91 93 79 71 67 67 62 68 10 87 14 83 87 19 67 74 72 75 69 68 78 88 10 26 32 29 16 14 Eu " 676 6 52 18 85 69 44 8 40 5 7 50 46 80 12 87 97 92 15 39 33 20 10 68 23 86 49 55 6 7 6 1 9 1 5 0 39 4 41 6 6 88 3 5 0 2 6 9 3 1 27 43 88 32 99 73

11 211 319 94 67 67 53 45 38 41 33 31 38 70 58 61 70 58 61 50 55 66 67 60 42 64 88 88 67 32 33 32 28 28 27 26 26 46 28 45 29 27 64 27 27 26 26 26 25 28 31 43 90 27 98 63 59 Gd " 1 1 31 31 07 30 57 18 44 03 14 59 32 63 56 54 73 92 58 65 25 32 53 54 80 99 58 40 52 83 29 22 23 39 53 92 01 28 05 73 41 35 22 51 42 90 35 91 76 02 65 67 0 45 86 17

14 11 11 91 82 68 73 63 58 61 88 78 10 12 10 11 80 86 10 97 93 70 97 13 13 10 55 56 55 50 51 50 50 49 94 48 69 53 44 10 50 50 47 48 47 46 51 55 81 13 15 13 99 10 Tb " 220 358 80 29 00 4 4 5 0 9 6 4 2 4 46 16 44 05 1 1 68 8 0 8 6 37 00 64 2 3 7 5 0 6 6 4 6 0 9 1 9 14 5 4 5 1 2 6 8 8 1 31 94 64 0 17

116 186 93 72 70 59 54 46 49 45 41 40 50 46 67 78 67 71 49 53 67 59 58 45 59 81 79 67 36 37 37 35 35 35 36 34 65 31 44 35 29 64 35 35 32 32 33 32 35 37 53 82 97 83 62 65 Dy " 6 2 87 78 23 59 57 91 40 10 22 04 03 26 66 44 34 59 14 14 62 37 45 62 68 60 29 78 69 24 97 10 44 36 26 96 41 89 17 50 63 92 81 07 86 96 00 73 38 70 18 83 16 62 21 11

20 16 15 13 12 11 11 11 10 94 10 10 15 17 14 15 10 11 14 12 12 10 12 17 16 15 84 87 91 84 86 87 88 86 14 75 10 84 70 14 86 84 79 81 81 78 85 88 12 18 20 18 13 13 Ho " 231 342 75 32 31 05 15 07 71 01 21 4 85 10 21 35 95 86 48 56 74 90 71 32 88 07 97 29 9 4 3 5 4 3 1 7 77 7 02 7 6 84 0 3 8 0 2 6 1 6 23 06 91 15 77 95

61 48 44 37 35 34 35 34 32 29 32 30 45 51 44 46 30 33 42 38 37 31 37 47 49 45 26 27 28 27 27 27 27 27 42 23 30 26 22 46 26 26 25 25 25 24 26 27 37 52 59 52 40 39 Er " 684 892 41 78 25 82 45 17 52 52 15 73 45 57 26 08 34 47 04 54 72 67 36 35 09 50 24 89 50 31 56 13 32 71 74 28 04 89 81 01 21 33 49 00 53 85 79 73 58 74 25 35 69 93 32 68

T 94 76 67 57 54 54 55 55 51 48 51 47 69 79 69 71 44 51 65 60 57 49 55 69 74 73 42 43 45 43 43 43 43 43 61 38 50 40 36 76 42 41 41 41 42 40 42 44 59 79 87 78 60 59 m " 112 130 9 2 5 9 4 1 8 2 3 3 4 8 2 4 1 6 9 0 6 8 9 6 8 7 4 1 5 6 2 4 8 4 7 2 4 8 6 7 2 3 6 2 6 8 6 6 7 3 1 4 7 9 9 4

62 51 43 36 35 35 36 36 34 32 33 31 44 52 45 47 28 32 42 40 37 32 35 43 47 48 28 29 30 29 28 28 28 28 37 25 34 27 24 52 28 27 27 28 28 27 28 29 39 50 55 51 39 38 Yb " 773 827 93 49 63 99 01 89 42 36 17 21 62 26 58 18 54 00 42 93 67 16 24 81 55 52 21 85 59 51 30 00 90 64 84 74 66 88 24 17 48 07 24 41 88 48 80 35 70 72 89 95 42 13 25 82

Lu " 123 127 98 79 65 55 52 54 55 56 52 49 52 48 67 79 69 71 42 49 64 61 56 50 53 63 70 75 43 45 46 45 44 43 44 43 54 39 53 41 37 80 43 42 43 44 44 43 44 46 63 77 82 78 59 57 3 9 7 6 6 5 7 1 9 9 4 8 7 7 6 7 4 2 3 7 3 8 4 8 5 3 8 9 9 1 5 7 5 6 3 9 3 7 5 4 9 1 0 6 9 5 2 2 0 4 9 1 9 7

259 294 61 55 55 46 43 43 39 39 38 38 38 35 46 57 54 55 48 53 51 53 42 39 37 43 50 58 34 33 33 31 31 32 30 32 32 32 44 32 33 69 31 31 33 34 33 32 34 34 48 51 52 52 50 44 Hf " 3 1 53 41 18 63 71 08 93 69 52 33 92 41 50 52 97 90 34 45 14 74 24 45 87 34 16 44 11 75 19 71 96 62 62 45 58 23 37 70 07 91 43 91 68 79 75 69 14 74 58 65 35 80 73 57

61 55 56 48 44 44 41 40 39 39 40 36 47 59 56 57 49 54 52 55 44 40 39 43 50 57 35 34 33 33 33 33 31 32 32 33 46 33 34 71 32 32 34 35 34 33 34 35 51 49 52 53 50 45 Ta " 438 466 3 2 4 7 2 1 1 8 6 5 0 2 4 0 7 1 3 1 1 1 4 5 5 8 8 8 4 3 8 2 2 3 9 7 9 2 3 8 4 6 1 5 0 1 1 6 2 2 1 0 1 2 9 4

13 12 15 99 66 65 69 71 70 76 79 71 87 21 18 20 14 15 15 18 75 78 92 98 14 14 69 73 98 10 99 96 89 93 81 68 80 84 63 11 95 95 97 99 93 91 89 88 10 19 20 21 17 13 W " 544 502 65 53 45 7 4 7 5 9 7 1 1 0 7 49 48 42 12 91 61 40 4 3 1 4 81 07 8 3 9 06 3 3 1 3 8 6 0 7 2 98 2 2 9 2 3 4 0 3 98 22 29 66 45 89

134 101 15 14 19 10 87 51 45 48 48 41 24 28 69 19 19 18 14 15 14 16 42 32 32 56 14 14 61 47 27 30 22 15 47 53 66 53 90 54 83 13 48 51 58 31 36 56 53 67 64 13 17 16 17 81 Tl " 0 1 67 47 21 57 0 4 4 8 4 1 7 8 4 92 12 68 73 24 10 51 3 4 5 1 98 82 3 0 8 6 8 9 5 8 6 4 4 0 2 57 8 3 4 2 3 9 5 4 8 66 74 05 36 7

737 715 24 27 27 22 25 22 25 34 32 20 26 17 24 21 21 19 17 18 21 22 22 20 19 24 30 28 18 22 40 37 45 48 29 25 22 17 22 19 16 31 23 19 22 19 22 19 19 18 19 30 37 81 25 14 Pb " 8 6 42 17 45 82 26 52 97 76 30 87 92 99 93 62 09 04 15 42 06 77 77 24 53 98 00 55 39 80 25 56 67 48 44 62 88 75 11 47 15 37 86 91 22 23 01 51 57 27 89 09 87 10 85 65

465 506 67 58 58 49 46 46 42 42 41 41 42 38 49 57 57 59 51 56 54 57 46 42 40 47 54 62 36 36 35 34 34 34 33 34 34 34 47 35 35 75 33 34 35 37 36 35 36 36 50 56 55 56 53 47 Th " 6 0 52 42 93 11 94 36 96 63 21 13 05 25 84 98 10 57 38 33 06 85 33 01 98 18 72 96 49 20 16 13 35 62 10 76 85 29 49 06 83 28 57 22 85 22 17 26 17 92 26 81 79 43 68 77

189 207 15 14 15 12 10 11 10 97 98 10 13 10 17 15 14 15 13 14 18 20 13 11 11 13 16 18 10 92 85 84 82 83 85 94 10 10 20 10 11 40 94 10 10 10 10 98 10 10 19 21 21 25 22 12 U " 1 2 43 73 09 81 92 15 06 4 0 70 74 36 26 64 89 08 24 78 16 43 34 64 62 86 02 81 04 8 6 4 9 5 5 6 20 14 00 66 05 81 7 31 15 87 16 4 00 51 94 95 69 55 00 11

mol ΣF ar e/ rati 0.0 0.1 0.4 0.4 0.2 0.6 0.7 0.7 0.7 0.6 0.6 0.8 0.9 0.9 0.8 0.2 0.2 0.2 0.3 0.3 0.4 0.3 0.7 0.7 0.8 0.8 0.4 0.4 0.8 0.7 0.6 0.6 0.6 0.6 0.6 0.7 0.7 0.7 0.4 0.7 0.6 0.2 0.6 0.7 0.6 0.6 0.6 0.7 0.7 0.7 0.7 0.5 0.3 0.3 0.3 0.7 Al o 5 6 0 3 6 2 1 5 1 1 7 5 6 3 5 2 7 6 8 8 1 4 9 8 7 0 3 0 3 7 5 3 4 4 7 2 6 1 1 3 3 8 6 1 9 9 8 0 1 4 7 1 0 1 3 6

Fe (II) 0.1 0.3 0.1 0.5 0.6 0.0 0.4 0.5 0.8 0.7 0.1 0.1 0.1 0.2 0.6 0.7 0.3 0.6 0.5 0.6 0.6 0.3 0.6 0.2 0.5 0.5 0.0 0.5 0.6 0.6 0.4 0.2 0.2 0.6 /Al " 0 3 6 0 2 0 9 6 5 9 3 8 7 5 9 4 2 6 6 5 0 4 3 0 8 7 0 9 3 8 1 1 3 0

Fe " (III 0.0 0.1 0.1 0.1 0.1 0.0 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.0 0.1 0.1 0.0 0.1 0.0 0.1 0.1 0.0 0.1 0.1 0.0 0.1 0.0 0.0 0.1

)/A 6 0 0 2 3 0 2 2 1 6 9 9 9 9 0 6 9 2 9 1 1 7 0 8 1 2 0 2 1 8 0 9 8 6 l

Mg 0.0 0.1 0.2 0.2 0.1 0.2 0.3 0.3 0.3 0.3 0.3 0.4 0.4 0.4 0.4 0.1 0.1 0.1 0.2 0.2 0.2 0.1 0.4 0.4 0.4 0.4 0.2 0.2 0.4 0.4 0.3 0.3 0.3 0.3 0.3 0.3 0.4 0.3 0.2 0.4 0.3 0.1 0.3 0.4 0.3 0.3 0.3 0.4 0.4 0.4 0.4 0.3 0.1 0.2 0.2 0.4 /Al " 6 2 1 2 4 9 5 6 5 0 2 1 6 5 1 3 5 5 1 1 2 9 2 2 6 3 4 3 4 0 4 2 3 3 5 9 1 9 4 1 6 7 7 0 9 9 9 0 1 3 4 1 9 0 1 4

K/ 0.5 0.5 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.2 0.2 0.2 0.1 0.3 0.3 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.4 0.1 0.1 0.4 0.0 0.1 0.0 0.1 0.0 0.0 0.0 0.1 0.1 0.2 0.3 0.3 0.3 0.2 Al " 3 6 2 1 8 5 4 1 0 8 4 1 0 4 2 9 8 8 3 3 1 5 0 2 0 5 0 1 1 6 3 3 3 3 4 6 9 8 2 2 9 3 9 2 8 0 9 7 9 3 5 6 5 5 4 4

CI A " 67 63 77 78 74 81 80 77 76 72 74 82 89 86 90 73 74 74 76 76 77 75 81 71 78 85 78 77 78 66 58 57 56 56 58 70 71 66 42 74 74 51 64 72 68 72 68 70 70 73 88 80 75 75 76 81

CI A- K/ CI 10 10 10 10 10 10 10 10 10 10 W " 99 94 99 0 99 99 97 90 89 82 82 90 96 96 0 0 99 99 0 99 99 99 96 82 91 97 0 99 84 69 59 58 57 57 60 73 75 75 51 81 85 64 68 78 72 77 72 73 74 80 0 0 0 0 0 99

PI 10 10 10 A " 98 88 99 99 99 99 97 88 87 79 79 89 96 95 0 99 99 99 99 99 99 99 95 79 89 97 99 99 83 68 58 57 56 57 59 72 74 71 38 79 82 51 66 76 70 75 71 72 73 78 0 0 99 99 99 98

Mg I " 89 80 71 70 78 64 59 58 59 63 61 55 52 53 55 79 76 77 71 70 69 73 54 55 52 54 67 69 53 55 60 61 60 61 59 56 55 56 68 55 58 74 57 55 56 56 56 56 55 54 53 62 72 72 71 53

MI

A[R

] " 60 50 47 46 53 41 37 36 37 39 37 35 33 33 36 54 52 52 48 47 46 49 35 33 33 35 46 46 33 32 33 33 32 33 32 34 33 33 31 34 37 38 34 34 34 35 34 34 34 34 36 42 50 49 48 35

MI

A[R

]-K " 85 68 55 53 65 46 41 39 39 41 39 36 34 35 37 67 63 64 55 55 54 58 37 35 35 37 52 54 34 33 33 33 33 33 33 35 34 35 35 36 40 45 35 36 35 36 35 35 35 35 38 47 59 58 57 38

Table 2: Immobile element ratios of paleosol with comparison to previous studies and nearby volcanic rocks

Paleosol - D-N Paleosol - D-S Paleosol - All Lita Litb OD Drillcorec Thessalon Volcanic Rocksd

n=13 n=40 n=53 n=3 n=24 or 30*

Al2O3/TiO2 mean 10.98 11.18 11.13 12.04 11.48 10.55 12.11 15.18

± 1sd 0.54 0.99 0.90 0.8 1.36 2.07 4.0

Nb/Ta mean 15.47 15.35 15.38 15.50 12.03

± 1sd 0.58 0.47 0.50 1.80

Zr/Hf mean 38.17 38.24 38.22 38.33 43.57

± 1sd 0.15 0.25 0.23 10.3

Zr/Nb mean 24.18 24.46 24.39 18.2 32.1 24.31 25.1 20.68

± 1sd 1.1 0.9 0.95 4.4 4.1 9.4 3.34

Nb/Th mean 1.48 1.46 1.47 1.1 1.47 1.4 1.51

± 1sd 0.07 0.05 0.06 0.2 0.24

Hf/Ta mean 9.79 9.81 9.8 9.83 6.54

± 1sd 0.15 0.18 0.17 1.56 a: Sutton and Maynard (1993) b: Utsunomiya et al. (2003) c: mean from Sutton and Maynard (1993) and Utsunomiya et al. (2003) d: Ketchum et al. (2013) 'Unit 6' volcanic rocks, excluding 2 samples (KYT512, KYT513)

*mean of either 24 or 30 samples depending on available HFSE data

Table 3. Stable Cr and Fe isotope data of Cooper Lake dike-hosted paleosol

Depth Sample ID (cm) Description δ53/52Cr 2se Cr [µg/g] δ56/54Fe 2se

CLA-003 31.5 D-S -0.318 0.012 2.99 0.421 0.022

CLA-004 47 D-S -0.315 0.015 3.54

CLA-005 60 D-S -0.328 0.015 2.49 0.332 0.024

CLA-006 73 D-S -0.311 0.013 2.24

CLA-007B 89.5 D-S -0.308 0.017 2.32 0.318 0.022

CLA-008A 97 D-S -0.322 0.014 2.02

CLA-008B 106.5 D-S -0.323 0.019 2.07 0.304 0.020

CLA-009 111.5 D-S -0.362 0.044 2.04 0.330 0.021

CLA-010 120 D-S -0.325 0.014 2.10

CLA-011 129.5 D-S -0.291 0.028 2.09 0.312 0.022 CLA-013 151 D-S -0.310 0.014 2.55

CLB-015 120 D-N -0.329 0.020 3.31 0.492 0.020

CLB-016 124 D-N -0.353 0.022 3.09 0.487 0.020

CLB-018A 136 D-N -0.324 0.013 3.48 0.476 0.023

CLB-023B 204 D-N -0.318 0.016 2.80 0.407 0.022

CLC-024/025 245.5 D-N -0.303 0.025 2.14

CLC-029 264.5 D-N -0.324 0.014 1.94 CLC-033 299 D-N -0.309 0.011 3.23 0.395 0.019

CLC-037 324 D-S -0.356 0.017 1.75 0.245 0.025

CLC-040A 352 D-S -0.310 0.015 1.77

CLC-044 401.5 D-S -0.359 0.030 1.65 0.228 0.021

CLC-045 409 D-S -0.351 0.032 1.78 0.206 0.021

CLC-045 LIGHT 409 LCP -0.309 0.027 2.95 0.228 0.019

CLC-048 415.5 D-S -0.293 0.015 1.96

CLC-048 LIGHT 415.5 LCP -0.310 0.024 3.78 0.326 0.021

CLC-051 442 D-S -0.316 0.015 1.96

CLC-052 453 D-S -0.301 0.038 1.87 0.190 0.021 CLD-053 424 D-S -0.293 0.038 1.65 0.187 0.019

CLD-054 464.5 D-S -0.314 0.015 1.88

CLD-056 493.5 D-S -0.303 0.030 1.90 0.169 0.026

CLE-057 510 D-S -0.323 0.014 3.23

CLF-058 520.5 D-S -0.344 0.030 2.82 0.248 0.021

CLF-062 559 D-S -0.321 0.010 2.85

CL-CLFD-1 OD -0.325 0.014 2.26 0.245 0.022

Outcrop area map (modifed from Sutton & Maynard, 1993)

Archean granite

Matinenda Fm. N +100 Outcrop area

0

Cooper Lake Canada

Gravel Road -100 Map Border area Sault Ste. Post-Matinenda Rocks Marie Thessalon U.S.A. -200 0 1 km Lake Huron -300

Shearing Unconformity -400

Channel sampling N -500

Gravel Road -600 Outcrop depth (cm) relative to unconformity contact Matinenda Fm.

Mafic Dike (Thessalon Fm.)

Livingstone Creek Fm. 100 m Archean Granite

Sketch map of outcrop (modified from Bennett, 1990) Qtz Ms Chl Mc Ab 0

100

200

300

Depth (cm) 400

500

600 n.d. n.d. n.d. n.d. n.d. 1-2% 2-5% <10% <10% <10% <10% 5-10% 10-20% 20-30% 30-40% 10-30% 30-50% 50-70% 10-20% 20-30% 30-40% 10-20% 10-20% 20-30% 30-40% 0 D-N D-S (a) immobile elements 100

200

300

Depth (cm) 400

500

600 LCP OD 14 16 18 20 22 5 6 7 8 9 1011 7 9 11 13 22 24 26 1.3 1.5 1.7 9.5 10.0 10.5 Al2O3 wt. % Nb [ppm] Al2O3/TiO2 Zr/Nb Nb/Th Hf/Ta 0

XRF (b) major elements (ME)

100 ICP-MS

200

300

Depth (cm) 400

500

600

1 2 3 4 5 1 2 3 4 5 1 2 3 4 2 4 6 8 1 2 3 4 5 0 2 4 6 8 1012 Ca/Al (x 101) Na/Al (x 101) K/Al (x 101) Fe/Al (x 101) Mg/Al (x 101) Mn/Al (x 103)

0 (c) ME + weathering indices CIA

100 CIA-K

200

300

MIA Depth (cm) 400 MIA-K 500

600

2 3 4 1.5 2.5 3.5 4.5 40 60 80 40 60 80 30 40 50 60 50 60 70 2 Si/Al P/Al (x 10 ) CIA/CIA-K PIA MIA(R)/MIA(R)-K MgI Al2O3 1.0 D-S D-N 0.8 LCP ill OD mus U (’03) 0.6 S&M (’93) Cooper Lake drillcore

pl 0.4 Thessalon molar Fe/Al volcanics (Unit 6)

0.2 Al2O3

0.0 CIA 0.5 ill CaO* mus K2O +Na2O 0.4

0.3

0.2 molar K/Al

chl 0.1 MIA(R)

CaO* 0.0 Fe2O3(T) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 +Na2O +MgO molar Mg/Al +K2O ΣFe Fe(II) Fe(III) 0 D-S D-N 100 LCP OD

200

300 Depth (cm) 400

500

600

0.2 0.3 0.4 0.5 0.6 5 0 5 0 5 0 5 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 .4 .4 .3 .3 .2 .2 .1 0 0 0 0 0 0 0 Cr/Nb ------molar Fe/Al δ56/54Fe ‰ δ53/52Cr ‰ 0

100

200

300 Depth (cm) 400

500

600

.00 .02 .04 .06 .08 .10 .12 50 60 70 80 90 100 .05 .10 .15 .20 .25 .30 .10 .15 .20 .25 .30 .35 .40 Mo/Nb V/Nb W/Nb U/Nb a. Cooper Lake paleosol (ca. 2.45 Ga) b. Flin Flon paleosol (ca. 1.85 Ga) c. Nsuze paleosol (ca. 2.96 Ga) 1.0 D-S U (2003) oxidized horizons sericitic paleosol (upper) D-N S&M (1993) unoxidized horizon chloritic paleosol (lower) OD protolith greenstone protolith greenstone 0.8

ΣFe/Al = 0.756 0.6 Fe(III)/Fe(II)O = 1 Fe(III)/Fe(II)O = 1 Fe(III)/Fe(II)O = 1 ΣFe/Al = 0.670

M M M ΣFe/Al = 0.545 Fe(III)/Al 0.4 R R R H H ΣFe/Al = 0.484 H paleosol trends paleosol trends paleosol trends 0.2

0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Fe(II)/Al Fe(II)/Al Fe(II)/Al

0.6 D-S D-N a 0.5 LCP OD

0.4 Fe ‰

0.3 56/54 δ

0.2

0.1 30 35 40 45 50 55 60 65 70

MIA(R)-K 0.8 1000lnα = -0.20 ± 0.10 b

0.6

Fe ‰ 0.4 56/54 δ 0.2

Fe loss Fe gain 0.0 0.1 0.5 1 2 fraction of remaining Fe (f) 1000 1.8 D-S D-N a b OD 1.6

100 1.4

1.2 Ce/Ce*

10 1.0

CI-chondrite normalized 0.8

Deccan Traps saprolite 1 0.6 La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu 0.7 0.8 0.9 1.0 1.1 1.2 1.3 Pr/Pr* 28 1.1 1.10 c d +Ce -La e

27 1.0 1.05

26 0.9 1.00 Y/Ho Eu/Eu* Ce/Ce*

25 0.8 0.95

dike centre dike margin offset dike +La -Ce 24 0.7 0.90 100 200 300 400 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.90 0.95 1.00 1.05 1.10 ΣREE+Y [µg/g] molar Mg/Al Pr/Pr* 0 D-N D-S LCP 100 OD

200

300 Depth (cm) 400

500

600

0 2 4 6 8 0 5 10 15 20 25 30 0 5 10 15 20 25 30 0 10 20 30 Zn/Nb Ni/Nb Co/Nb Cu/Nb

300 <300 cm depth 300-500 cm depth >500 cm depth Drillcore (M et al. 2016) 200 Paleosol range (M et al. 2016)

100

0 % change in ratio relative to OD -100 Fe Mg Zn Ni Co Cu

CLA-004: 47 cm CLC-037: 324 cm CLD-056: 493.5 cm

PPL XPL

Fe

Cr

-

S -

S Cr Fe

Ti

Cr

- Ti Cr

Mn

Cr

-

Mn -

K Cr K 20 D-N D-S a LCP OD 10

0 %Th/Nb

-10

-20 -50 0 50 100 150 %U/Nb Archean cratonic shale b 8 Proterozoic cratonic shale Average Unit 6 Thessalon ca. 2.9 Ga Nsuze Paleosol 6 Th/U 4

2

0 1000 2000 3000 4000 5000 U [ppb] E F Immobile/insoluble elements Soluble elements lost from retained in weathered mac rocks: mac rocks to the hydrosphere: Al, Ti, Nb, Ta, Zr, Hf, Th, P, Ca, Na, Mg, Fe(II), Zn, Ni, V, Cr, Mo Co, Li, Sr, *K, *Rb, *Cs HFSE ratios, δ53/52Cr = magma Y/Ho, Eu/Eu* > protolith source source δ56/54Fe < protolith source δ56/54Fe > magma source

remnant subaerial D lava ows

relict local benthic feeder dikes oxygen production?

pervasively anoxic atmosphere magma: δ53/52Cr < -0.124 ‰ D B A B C crystals (ol+chr): δ53/52Cr > -0.124 ‰ Archean crust Archean crust (lower plate) (upper plate)

C partial melts Plume

mantle domains with δ53/52Cr < -0.124 ‰ Highlights

Elemental Ce-Cr-V-Mo data indicate anoxic conditions, U shows minor mobility

Fe geochemistry fit criteria of a reduced paleosol with loss of isotopically light Fe(II)

Loss of transition metals, Y>Ho to hydrosphere

Isotopically light Cr originating from magmatic processes and decoupled from surface conditions