PALEOMAGNETISM OF THE VOROGOVKA GROUP: SUBSTANTIATING THE VENDIAN GEOMAGNETIC PHENOMENON (Yenisei Ridge)

The Neoproterozoic era is rightfully considered one of the most eventful periods of geological history. Notably, available data indicates radical changes in the behavior of the geodynamo, which by the end of the Neoproterozoic time led to a disruption of the dominating axial dipole structure of the Earth’s magnetic field. Consequently, measured paleomagnetic poles for this period do not correspond to the actualistic model of the Geocentric Axial Dipole (GAD), which distorts the information on the paleogeography and tectonics of continental units the terminal Precambrian [Shatsillo et al., 2020; Metelkin et al., 2022]. Scientific debates are focused on: the causes of these global transformations in the geospheres, the actual configuration of the geomagnetic field, the duration of this period of its anomalous state, and on approaches to tectonic interpretation of the complex set of paleomagnetic determinations [Domeier et al., 2023].

For instance, the paleomagnetic database gathered for the Siberian craton features two discrete groups of coeval late Precambrian–early Cambrian poles with varying degrees of reliability. The first group was mainly described in publications by Soviet paleomagnetologists; its early Cambrian pole was calculated in [Khramov et al., 1982] and was confirmed multiple times in subsequent works [Pavlov et al., 2018]. In the present paper, this group of paleomagnetic directions (poles) is referred to as KHR. These paleomagnetic poles are situated to the southeast of Australia, and in the GAD model they reflect relatively high latitudes of the Southern hemisphere, especially for northern regions of Siberia. The second group of poles located close to the eastern coast of Madagascar was first substantiated in [Kirschvink and Rozanov, 1984] (hereinafter – KRS), and corresponds to an equatorial position of the craton [Kazansky, 2002; Metelkin et al., 2005, 2012]. Several recent publications on Siberian paleomagnetism confirmed the existence of this puzzling combination of paleomagnetic directions, as well as proved at the level of individual samples that both the KHR and KRS components were recorded synchronously (on the scale of the accumulation of a sedimentary sequence with related detrital magnetization) [Shatsillo et al., 2005; Pavlov et al., 2018; Vinogradov et al., 2023]. According to our own hypothesis, this phenomenon was caused by an abrupt decrease in the strength of the main component of the geomagnetic field, which we equate with the bipolar KRS component. Modern estimates of paleointensity confirm a fall of the virtual dipole moment value at least by an order of magnitude compared to the present-day one [Bono et al., 2019; Shcherbakova et al., 2020; Metelkin et al., 2022]. We assume that this decrease could have been even more significant, to the point that in individual brief time periods the dominating field was that of the global magnetic anomalies (GMA), while the fixation of the monopolar KHR component reflects the Antarctic anomaly, which was closest according to reconstructions [Metelkin et al., 2022]. Because of the detrital mechanism of magnetization, the frequently alternating ultra-brief periods when the dipole component was reduced to values, comparable to, and lower than the anomalous field were recorded as both the KHR and KRS components synchronously. This hypothesis is supported by ultra-frequent reversals identified in many sedimentary sections aside from Siberia’s [Popov et al., 2002; Shatsillo et al., 2015; Bazhenov et al., 2016; Levashova et al., 2021].

According to our data, the time range of the supposed anomalous state of the geomagnetic field was 580–530 Ma [Vinogradov et al., 2023]. The most accessible deposits of this age on the Siberian craton occur on its southwestern periphery. In this study, we attempt to verify the above-mentioned ideas on the example of the paleomagnetic record in the Vorogovka Group of the Yenisei Ridge. The Vendian (Ediacaran) age of this sedimentary sequence has long been controversial [Khomentovsky et al., 1972; Sovetov and Blagovidov, 1996; Sovetov, 2001; Sovetov and Le Heron, 2016]. Here we show the entire volume of geological and isotope-geochemical data obtained in recent years, as well as results of paleomagnetic comparisons, which confirm the Vendian age of the group in addition to providing new constraints for deciphering the Vendian geomagnetic phenomenon.

The terrigenous-carbonate rocks defined as the Vorogovka Group occur locally and comprise the Vorogovka depression in the northwest of the Yenisei Ridge (Fig. 1). The group unconformably overlies variously altered volcanogenic-sedimentary and igneous formations of mainly island-arc origin, including ophiolite sheets, collectively regarded as the Isakovka terrane – a fragment of the elongated Yenisei accretionary belt [Vernikovsky et al., 1999; 2003; Metelkin et al., 2007]. According to available geochronological estimates, the evolution of the island-arc system took place 700–630 Ma, and it’s accretion to the margin of the Siberian paleocontinent took place 620–600 Ma [Vernikovsky et al., 1999, 2003; Kuzmichev et al., 2008; Vernikovskaya et al., 2020, 2023].

There is no consensus on the age, formation causes and paleogeography of the Vorogovka depression, however most researchers agree that the Vorogovka Group accumulated in conditions of a marginal continental basin and that the composition reflects its evolution from an alluvial plain to sea shelf with carbonate sedimentation replacing a terrigenous one [Sovetov and Blagovidov, 1996; Vernikovksy et al., 2009; Kochnev et al., 2019]. The sedimentary rocks have many landslide-related and rhythmically laminated features, typical for slopes with high rates of sedimentation. This, along with the relatively limited occurrence of the group can be evidence of accumulation in conditions of a narrow trough (aulacogen) [Sovetov and Le Heron, 2016].

Bottom to top, the Vorogovka Group includes the Severnaya Rechka, Mutnina, and Sukhaya Rechka formations (Fig. 2). Their most representative sections are located in the lower reaches of the Vorogovka R. and were described in many publications [Semikhatov, 1962; Khomentovsky et al., 1972; Sovetov and Blagovidov, 1996; Kochnev et al., 2019]. Our study precisely concerns these sections. For paleomagnetic analysis we took over 300 oriented samples from outcrops on the Vorogovka riverbanks in 23 sites (Fig. 1, 2).

The Sukhaya Rechka Formation begins with gritstone and conglomerate with coarse crossbedded and trough-like layering. Some outcrops in the lower part of the formation have breccia layers that are interpreted as glacial diamictites [Sovetov and Le Heron, 2016]. However, Kochnev et al., [2019] showed an interpretation with a normal sedimentary succession without glacier involvement. Lithological-sedimentological reconstructions show that this coarse-grained sequence was deposited in streams and channels. The middle part of the section is composed of interlayering sandstone and siltstone, transitioning to a characteristic unit of dark siltstone and mudstone and then to limestone. The thickness of the formation in the most complete sections reaches 1300 m.

The main targets for paleomagnetic research are 8 km upstream from a nameless island (Fig. 1). The outcrop consists of alternating gritstone, medium-coarse-grained sandstone, silty thin-slabby siltstone, thin-slabby sandstone. On the right bank of the Vorogovka R. (sites 14ek08 and 14ek09) medium-grained grey-green sandstone units were sampled. In site 14ek08 of the Severnaya Rechka Fm. we took sample К-50-14 for detrital zircon geochronology and for determination of the Nd model age of the provenance rocks (Fig. 2).

Similar greenish-grey and reddish-grey sandstones were sampled ~400 m downstream on the left bank (sites 14ek10, 14ek11, and 14ek12). Here the sequence forms an isoclinal anticline fold with the northern limb overturned.

From the limestone unit in the upper part of the formation, ~1.5 km upstream of the mouth of the Mutnina R. on the left bank we sampled dark-grey laminated microphytolitic limestones with a noticeable amount of medium-grained sand (sites 14ek13 and 14ek14).

The Mutnina Formation gradually replaces the Severnaya Rechka formation and has a typical flysch structure [Sovetov and Blagovidov, 1996; Sovetov and Le Heron, 2016]. Its main part consists of rhythmically laminated greenish-grey polymictic sandstone, gradually replaced by limestone upsection. The thickness of the formation is approximately 900 m (Fig. 2).

The lowermost part of the formation was sampled on the left bank of the Vorogovka R. ~600 m upstream of the mouth of the Mutnina R. (Fig. 1), where there is an outcrop of greenish-grey polymictic massive, medium-fine-grained sandstone (site 14ek17). A similar sandstone interbed was sampled 300 m downstream (site 14ek15), and about 50 m downstream of the Mutnina R. mouth (site 14ek16). Further two km downstream of the Mutnina R. mouth and upsection we sampled a unit of alternating grey massive limestone with a large amount of sand material (site 14ek19) and thin-slabby, wavy-laminated and crossbedded grey calcareous sandstone (site 14ek20). About 1 km downstream we sampled a unit of massive greenish-grey and dark-grey fine-grained sandstone (site 14ek21) and a further one km downstream – light-grey limestone and calcareous sandstone. Outcrop 14ek15 of the Mutnina Fm. sandstones was chosen for detrital zircon age geochronology (smp. К-76-14) and Nd model age of provenance rocks determination (Fig. 2).

The upper part of the formation is composed of mainly grey, greenish-grey sandstone and was sampled 1–1.5 km upstream of the Sukhaya R. on both limbs of the socalled Mutnina anticline on the right bank of the Vorogovka R. (sites 14ek25, 14ek26, 14ek27, and 14ek28). The sedimentation environment indicators here include distinct channel casts and turbidite flow traces.

The Sukhaya Rechka Formation is composed mainly of limestone, which are often argillaceous with high fraction of siliceous clastic material, interbeds and lenses of sandstone (which increase in thickness upsection) and intra-formation conglomerate. The thickness of the formation reaches 2200 m (Fig. 2).

The targets of paleomagnetic research are located in two main territories. The first is in outcrops on the right bank of the Vorogovka R., about 6 km upstream of the mouth of the Sukhaya R. (Fig. 1). Here we sampled light-grey laminated limestones (site 14ek05). The second area is 1.2 km upstream of the mouth of the Vorogovka R., on the left bank. Here we sampled an extensive outcrop of alternating sandstone and limestone of the upper part of the formation with three sandstone units sampled in total. The first one is composed of fine-grained, light-grey argillaceous, slabby and unevenly laminated sandstone (site 14ek22). The second unit is fine-grained, grey, greenish-grey, wavy-laminated sandstone 50 cm thick (site 14ek23) and separated from the first one by a layer of limestone with graded bedding and a large amount of terrigenous material. The third sandstone unit is medium-grained, grey and thin-slabby, with uneven hummocky bedding (site 14ek24), 5 m upsection. From the lower part of the Sukhaya Rechka Fm. on the right bank of the Vorogovka R. (Fig. 1), we sampled a fine-grained, grey sandstone layer with carbonate matrix ~1.5 m thick (K-32-14) for determination of detrital zircon age and Nd model age of provenance rocks (Fig. 2).

For a high-precision reference point of the age of paleomagnetic determinations, it is important to substantiate the age of the Vorogovka Group and prove its tectonic unity with the Siberian craton, or the possibility that it formed in a different tectonic unit. For this goal, the gathering of paleomagnetic samples from bottom layers of all three formations was supplemented with samples of sandstones to determine the age ranges of sedimentation based on U–Th–Pb dating of zircon by LA-ICP-MS method. We also took samples of carbonate rocks from the upper part of the Severnaya Rechka Fm. and the entire section of the Sukhaya Rechka Fm. to estimate the Nd model age of provenance rocks for the clastic material.

Methodology. Zircon extraction was done in the Center for Collective Use for Multielement and Isotope Studies (CCU MII) SB RAS (Novosibirsk) following standard procedure based on combining magnetic and heavy liquid separation. Zircon monofractions for isotopic analysis was separated manually under a binocular microscope. The inner structure of the grains was studied on a LEO 1430VP SEM with a Detector Centaurus attachment in cathodoluminescence mode in the CCU MII SB RAS. The LA-ICP-MS investigations were done in GEOKHI RAS (Moscow) on a Thermo Finnigan Element XR mass spectrometer with a U – 213 laser ablation system with a crater diameter of 30–40 µm. The zircon standards GJ and 91500 were used for calibration and control. The procedure is described in detail in [Kostitsyn and Anosova, 2013]. The resulting data was reduced in Glitter software [van Achterbergh et al., 1999]. For zircons older than 1000 Ma, the crystallization age of the primary rock was accepted as the one calculated from the ²⁰⁷Pb/²⁰⁶Pb ratio, and for younger zircons – the ²⁰⁶Pb/²³⁸U ratio. Hereinafter when discussing the obtained results we use the U–Pb ages with >95% concordance.

The isotope-geochemical Sm–Nd investigation was done in GEOKHI RAS. The Sm and Nd contents were determined by the isotope dilution method from [Revyako et al., 2012]. Mass spectrometer measurements of Sm and Nd isotope composition were done simultaneously for various ions of the same element on a Triton multicollector instrument in static mode using Rhenium double filaments. The JNd-1 standard was used for control and reproducibility of isotope analyses of Nd (¹⁴³Nd/¹⁴⁴Nd = 0.512114 ± 22, 2σ, N = 20). Isotope ratios were normalized to ¹⁴⁶Nd/¹⁴⁴Nd = 0.7219. The inaccuracy of Nd isotope composition measurements did not exceed 0.005 % in individual analyses. Total in-lab blank during work with the samples (May 2015) was 0.2 ng for Nd, and 0.03 ng for Sm. When calculating ε_Nd and the T(DM) model age modern values were used for CHUR (chondrite uniform reservoir) – ¹⁴³Nd/¹⁴⁴Nd = 0.512638, ¹⁴⁷Sm/¹⁴⁴Nd = 0.1967 from [Jacobsen and Wasserburg, 1984] and DM (depleted mantle) – ¹⁴³Nd/¹⁴⁴Nd = 0.513151, ¹⁴⁷Sm/¹⁴⁴Nd = 0.2136 from [Goldstein and Jacobsen, 1988].

Studies of Sr isotope composition in carbonate rocks were done in the CCU MII SB RAS following standard procedure: strontium and rubidium were separated by ion-exchange chromatography in fused silica columns on Dowex AG W50x8 cation exchange resin with a 200–400 mesh and a 2N HCl eluent. Rubidium and strontium content in the carbonate fraction was determined by isotope dilution using ⁸⁵Rb and ⁸⁴Sr indicators. Measurements of Rb and Sr contents were done in a MI-1201AT multicollector mass spectrometer in the CCU MII SB RAS. The strontium isotope composition was determined on a Triton Plus multicollector instrument in the Geoanalitik CCU (Ekaterinburg). The determination precision for isotope ratios was controlled by parallel measurements for the SRM-987 isotope standard (Sr 0.710240 ± 7 (2σ av., n = 8)) in every sample series.

Results. In the base layers of quartz-feldspar sandstone of the Severnaya Rechka Fm. (smp. К-50-14), the main population of detrital zircon grains has ages in the range 600–680 Ma with a peak at 650 Ma (Fig. 3a). A less pronounced population has an age of 740 Ma. The second most numerous zircons group is a population with Paleoproterozoic ages in two clusters 1.73–1.76 Ga and 1.83–1.98 Ga, as well as individual grains aged 2.03, 2.32, 2.42, and 2.54 Ga. The youngest group (5 grains) has a weighted average age of 586 ± 11 Ma (Fig. 3b), which means that the lower sedimentation age limit for the Vorogovka Group is early Vendian.

In the quartz-feldspar sandstone with argillaceous matrix from the lower part of the Mutnina Fm. (smp. К-76-14), the zircons differ significantly. The first of the two populations has a Neoproterozoic age with two peaks at 820 Ma and 900–940 Ma. The second population consists of grains of Archean–Paleoproterozoic boundary 2.45–2.68 Ga. There is a small amount of grains of Paleoproterozoic age: 1.85–1.95 Ga. The lower sedimentation age is limited by a small population of zircon with a weighted average age 791 Ma (Fig. 3 b).

In the lower part of the Sukhaya Rechka Fm., which is composed of mainly carbonate rocks, from a quartz-feldspar sandstone interlayer (smp. К-32-14), the separated zircon grains are comparable in age to the Severnaya Rechka Fm. sandstones. The main population has an age in the range 604–764 Ma with a peak at 630 Ma. Apparently, it represents the nearly uninterrupted record of the prolonged tectonomagmatic activity in the provenance area (Fig. 3 a).

Only two zircons with ages typical of the Mutnina Fm. (848 Ma and 862 Ma) were identified. The second most populous zircon group has Paleoproterozoic ages: 1.74–2.08 Ga with a peak at 1.85 Ga, and individual zircons (13 grains) with ages ranging from 2.28 Ga to 3.15 Ga. The youngest group (5 grains) has a weighted average age of 569 ± 26 Ma (Fig. 3 b), which coincides with estimates of the lower age limit of sedimentation for the Severnaya Rechka Fm.

Our study of the age of detrital zircons in the Sukhaya Rechka Fm. was accompanied by detailed C and Sr isotope characterization of the entirety of its carbonate interval. We also studied the upper limestone layer of the Severnaya Rechka Fm. Analysis of the obtained geochemical data led to a selection of limestone samples with minimal content of terrigenous admixture, insignificant secondary alterations and preserved initial isotope composition, which made them suitable for isotopechemostratigraphic investigation. The measurements showed that the ⁸⁷Sr/⁸⁶Sr ratio is in the range of 0.70813–0.70828, and δ¹³C values are in the range from –0.7 to +1.8‰. This indicates a Vendian age for these rocks, which can be correlated to the level of the Khatyspyt Formation of the Olenek uplift [Vishnevskaya et al., 2017].

Similar results were reported in [Kochnev et al., 2019] with noticeably more detailed isotope curves for a somewhat wider range of the section, including the upper part of the Mutnina Fm. Moreover, these curves were supplemented by data on the isotopic composition of lead. The calculated Pb–Pb isochron age of carbonate rocks of the Severnaya Rechka Fm. is 580 ± 40 Ma (MSWD = 1.4), and of the Sukhaya Rechka Fm. – 565 ± 90 Ma (MSWD = 1.1) [Kochnev et al., 2019].

The concurrent age estimate for the Vorogovka Group from three independent isotope methods does not leave room for doubt about it being not older than 580 Ma. The upper age limit can be estimated from an unconformable overlap by the Lebyazh’ya Formation, which is constrained to the uppermost Vendian – lower Cambrian by biostratigraphic data [Kochnev and Karlova, 2010; Kochnev et al., 2019]. In the General stratigraphic chart of Russia, its age as accepted as not younger than 535 Ma [Stratigraphic…, 2006].

Our analysis of lithological and stratigraphic data lets us assume that the accumulation of the Vorogovka Group took place in a shallow coastal marine basin of the Siberian paleocontinent, which in turn was the main supplier of clastic material. Specifically, this is indicated by multiple zircon populations of Archean and Paleoproterozoic ages from 3.15 to 1.74 Ga, typical for the Siberian craton. This is also supported by data on the old model age of the provenance rocks: Severnaya Rechka Fm. – T(DM) = 2.45 Ga, ɛ_Nd = –8.2, Mutnina Fm. – T(DM) = 2.2 Ga, ɛ_Nd = –5.1, Sukhaya Rechka Fm. – T(DM) = 2.1 Ga, ɛ_Nd = –3.4.

As for the provenance rocks for the Neoproterozoic detrital zircon populations, the closest regions of the southwestern border of the Siberian platform can be considered. Specifically, these zircon-bearing plutonic and volcanic formations correspond to these known intervals: for 560–590 Ma – the adakite-gabbro-anorthosite complex of the South Yenisei Ridge [Vernikovskaya et al., 2017]; for 600–650 Ma – multiple dikes of the dolerite-diorite-leucogranite association [Vernikovskaya et al., 2019; 2020] and the granitoids of the Tatarka complex [Vernikovskaya et al., 2007; 2013], as well as the Mara and Chapa volcanic complexes [Letnikova et al., 2021; Izokh et al., 2024]; for 700–760 Ma – the granitoids of the many plutons of the Ayakhta and Glushikha complexes [Vernikovskaya et al., 2007; 2023], including the Chernaya Rechka pluton [Likhanov and Reverdatto, 2019] and the rhyolites of the Kovriga complex [Nozhkin et al., 2008]; for 800–960 Ma – the granitoids of the Teya complex [Vernikovsky et al., 2016], the Gusyanka, Eruda, Kalama, and Middle Tyrada plutons [Nozhkin et al., 2023], plagiogneisses of the Garevka metamorphic complex [Likhanov et al., 2022]. Many other Neoproterozoic magmatic formations correspond to the main formation stages and alteration of continental crust in the region.

Thus, our geochronological and isotope investigations, combined with geological data, let us to estimate the sedimentation time of the Vorogovka Group in the range 580–535 Ma, after termination of tectonic events caused by the formation of the accretionary structure of the Sayan-Yenisei margin of the Siberian paleocontinent. Therefore, the primary paleomagnetic directions possibly preserved in the Vorogovka Group directly correspond to the Vendian – early Cambrian interval of anomalous geomagnetic field and can be compared to paleomagnetic data for the entire territory of the Siberian craton.

Methodology. Laboratory paleomagnetic experiments and processing of the obtained results were done in the laboratories of Geodynamic and Paleomagnetism of the Central and Eastern Arctic of NSU and Geodynamics and Paleomagnetism of IPGG SB RAS following standard procedures [Tauxe, 2010]. Pilot samples underwent incremental temperature (T-) demagnetization, however, alternating field (AF-) demagnetization proved more effective, and was consequently used on most of the studied samples. The number of demagnetization steps varied from 14 to 20 depending on ongoing results. Measurements of directions and value of the natural remanent magnetization (NRM) vector were done in a lab space shielded from the external geomagnetic field: For AF-demagnetization – on a 755 SRM cryogenic magnetometer (2G Enterprises, USA) with a built-in demagnetizer, and for T-demagnetization – a special shielded MMT-D80A oven-type demagnetizer (Magnetic Measurements Ltd., UK). Data processing was done in the PMTools online application [Efremov and Veselovskiy, 2023]. The composition of the magnetic minerals in the rocks was determined from studying the k/temperature relationship k(T) in open air on a MFK1–FA Kappabridge multifunction instrument with a CS-3 apparatus (AGICO, Czech Republic). These experimental results were supplemented by observation on a Merlin auto-emission SEM (Carl Zeiss, Germany) equipped with an AZtec X-Max energy dispersion spectrometer in the Institute of geology and petroleum technologies of KFU. The SEM investigations mainly involved sounding of the magnetic grains separated from limestones. The spectrometer resolution was 127 eV, measurement precision was 0.01–1% and depended on the state of the object studied. In all cases, we analyzed the surface of identified magnetic grains, and the sounding depth was less than 1 µm. Imaging the surface features was done at acceleration voltage 5 keV to improve the depth of field. Element analysis was performed at acceleration voltage 20 keV and a flange focal distance of 9 mm, which allowed avoiding minimal errors.

Scalar magnetic characteristics and magnetic mineralogy. The scalar magnetic characteristics differ primarily for the studied sandstones and limestones, which is obviously due to different concentration of magnetic minerals in these rocks. In general, the Vorogovka Group is characterized by k values in the range 10^–6–10^–3 SI units. At the same time, the limestones usually have the lowest k, with values varying from 10^–6 to 2 × 10^–4. In sandstones, the k values are from 2 × 10^–4 to 10^–3, with an average of 2–3 × 10^–4 SI units. The NRM value is from 0.1 to 30 mA/m. The Koenigsberger ratio (Qn = NRM/k × H, where H – mean strength of the present-day geomagnetic field at ~40 A/m) changes on the NRM/k bilogarithmic plot from less than 0.01 to slightly above 1 (Fig. 4 a).

On the k(T) curves, the k value falls abruptly at ~580°С, then gradually up to ~ 680°С until full demagnetization, which corresponds to Curie temperatures of magnetite and hematite respectively (Fig. 4,b). At the same time, the thermomagnetic curve is irreversible and shows an abrupt rise of k during cooling, with a new peak at temperatures ~400°С. This could indicate the formation of new magnetic phases of titanomagnetite and magnetite at high temperatures during the experiment. Such new formation through chemical changes (oxidation) of the primary magnetic phase is more typical of red rocks [Dunlop and Ozdemir, 1997]. Nonetheless, in our samples this mechanism is confirmed by the appearance of a bend on thermomagnetic cooling curve L2 close to the Verwey transition about –150°С after heating to 700°С (Fig. 4 b). The irreversible character of the k(T) curves was one of the main reasons for choosing AF-demagnetization as the main method for identifying the NRM components.

Detailed characteristics of the elementary composition of studied magnetic grains and SEM images a given in the Supplementary materials. SEM investigations in all studied samples showed multiple magnetite grains of various shapes and compositions. Most of them are weakly altered particles up to 50 µm in size with flat faces, sometimes retaining a crystalline habit (Fig. 5, grains 1–4, 11), which we relate to terrigenous input. Often there are grains of almost ideal spherical shapes, 15–30 µm in diameter and smooth hummocky-crystalline and acicular-crystalline topography (Fig. 5, grains 5, 6, 8). There are also individual Fe grains with a small (<1%) admixture of Mn, Cr, and Ni as folded flakes, elongated or isometric grains with lateral or transverse grooves and uneven edges, or as folded shavings (Fig. 5, grains 7, 10). Additionally, there are numerous cases of a Ni-Fe-Cr alloy as isometric or elongated lamellas, including large ones (>500 µm) with melted edges (Fig. 5, grains 9, 12). We consider those particles, along with the spherules, to be results of interstellar dust precipitation [Stankowski et al., 2006; Korchagin et al., 2010; Kuzina et al., 2016; Sungatullin et al., 2017]. Their cosmic origin is also supported by a lack of Ti admixture, which would be typical of volcanic particles of similar shapes and composition [Szöőr et al., 2001]. The significant amount of interstellar material in all samples lets us conclude that its occurrence was a background process during the accumulation of the Vorogovka Group.

Severnaya Rechka Formation. The figurative points for the sandstones of the formation on the NRM/k plot form a clearly defined subvertical trend across the Qn lines (Fig. 4,a). According to experimental data, such a distribution type is more typical of rocks with dominating chemical magnetization [Nagata, 1961]. Possible significant alteration of magnetic minerals after sediment diagenesis, which also degrades the paleomagnetic signal, is also indicated by low Qn values, meaning the induced magnetization dominates over the remanence. The same negative prognosis concerning the preservation of the paleomagnetic signal can be made for the limestones. Only the limestone in site 14ek13 is different as its figurative points plot in the lower left part of the diagram. Although the NRM and k values are the lowest for these samples, the points of NRM/k values form a trend along the line Qn = 1 (Fig. 4 a). Such a distribution is typical for rocks with detrital magnetization and reflects the dependence of measured parameters on the concentration of magnetic material [Nagata, 1961].

In all studied samples, either when heated to 280 °С or subjected to alternating magnetic field of 30 mT, a normal polarity component of high inclination is identified, hereinafter referred to as VSC (viscous component) (Fig. 6 a; Suppl. Mat.). The fold test for mean VSC vectors is negative, therefore they obviously do not reflect the ancient field, and are most probably of thermoviscous origin. In sandstones at the base of the formation (sites 14ek08 and 14ek09), it is the only regular component, while in the high-coercivity/high-T part of the demagnetization spectrum the remanence vector changes its direction chaotically. In overlapping sandstone and limestone layers (sites 14ek10, 14ek11, 14ek12, and 14ek14), after unblocking of VSC in more than half of the samples a regular metachronous component is determined (hereinafter MTC), which usually corresponds to the characteristic magnetization. It has a northwestern declination and exclusively positive inclination (Table). The unblocking temperature interval for MTC is from ~280 °C to 500–570 °C and the highest value of unblocking alternating field is 80 mT. This component was analyzed together with analogous MTC vectors in the Mutnina Fm. (further in the text).

Limestones from outcrop 14ek13 demonstrate a different behavior of NRM during demagnetization. In addition to VSC, a bipolar component is identified in the temperature interval 400–560 °C, which corresponds to known KRS directions. Out of 12 studied samples component analysis yielded five vectors with southwestern declination and mainly negative inclination as well as three vectors with northeastern declination and positive inclination (Table). This component was analyzed along with KRS directions in the Sukhaya Rechka Fm., where it is registered in all sites (further in the text).

Mutnina Formation. On the NRM/k plot, the studied samples can be conventionally divided into two groups. The first and largest group mainly consists of sandstone samples with a clear trend across the Qn lines and is characterized by a wide dispersion of NRM from 0.3 to 30 mA/m but a narrow k range from ~3 to 6 × 10^–4 SI units (Fig. 4 a). Such behavior indirectly indicates a chemical nature of magnetization. The figurative points of the second group, the sandstones from the limbs of the Mutnina anticline (sites 14ek25, 14ek26 and 14ek28), show a mostly linear distribution along the line Qn = 0.1. Such a distribution reflects a dependence of the scalar magnetization values mainly on the concentration of magnetic minerals, which is indirect evidence of detrital magnetization [Nagata, 1961].

As preliminary conclusions from the NRM/k plot, most of the formation rocks, except sandstones 14ek25, 14ek26 and 14ek28, and excluding the low-coercivity (soft) VSC, contain only one characteristic, regular MTC component (Table). The median destructive field (MDF) in these samples was ~15 mT, unblocking alternating field values were up to 120 mT in sandstone and up to 60 mT in limestone (Fig. 6 a, b).

In outcrops 14ek25, 14ek26 and 14ek28, the VSC component was not identified, and the MDF is somewhat lower – from 7 mT. At the same time, in the low-coercivity interval (up to 15 mT) of AF-demagnetization, first the MTC component is destroyed, then a component corresponding to KHR is registered. Both these components have a similar declination, however KHR has a noticeably shallower inclination (Fig. 7,a). Coercivity spectra of KHR and MTC or at least some other component often intersect. For instance, the KHR component in site 14ek26 sandstones can only be identified using methods of individual vectors and great circles combined analysis [McFadden and McElhinny, 1988] (Fig. 7,a). It is possible that the “unidentified” component whose coercive spectrum significantly intersects KHR, is actually KRS. Two facts indicate that the components are not fully separated. First, KHR is not a characteristic component. Second, during demagnetization, usually at alternating field above 30 mT, most samples display a deviation of the remanence vector along a great circle arc either into the NE quadrant of the stereoplot at positive inclination, or towards the SW quadrant at negative inclination, which is precisely towards KRS (Fig. 7 a).

Sukhaya Rechka Formation. The distribution of figurative points on the NRM/k plot is distinguished by a mainly concentric relationship. Namely, the sandstones are, as a rule, magnetically softer than sandstones. The k values vary from 0.35 to 12 × 10^-4 SI units. NRM ranges from 0.1 to 10 mA/m. the Qn values are on average 0.9–0.12 (Fig. 4 a). All these observations indirectly indicate detrital magnetization.

During AF-demagnetization, MDF was ~10 mT, which is somewhat lower than usual values for the other formations. Nearly all samples in the low-coercivity interval (up to ~15 mT) have a VSC component (Suppl. Mat.). One third of the studied samples under alternating field between 20 to 100 mT clearly show a regular bipolar KRS component on the vector endpoint diagrams (Fig. 8,a). In the remaining samples KRS can be identified on the stereoplots by the deviation of the remanence vector projection along a great circle arc (Fig. 8,a). Individual vectors and great circles combined analysis yields a mean KRS vector in all 5 sampling sites of the Sukhaya Rechka Fm. (Fig. 8 b; Table).

The results of this study let us confirm that in addition to the present-day viscous component the NRM of the Vorogovka Group contains three regular components.

The MTC vectors have been registered in most of the studied sections of the Severnaya Rechka and Mutnina formations and have a regular distribution in geographic coordinates with exclusively positive inclinations. The simulated parametric fold test [Watson and Enkin, 1993] for all mean MTC directions for the studied outcrops shows that maximum precision is achieved in geographic coordinates at –6.6% untilting (Fig. 6,c). The correlative fold test by [Enkin, 2003] is also negative (DS Slope: –6.25 % ± 20.23 %). The paleomagnetic pole calculated by averaging all virtual geomagnetic poles (VGP) assuming a reversed polarity of the geomagnetic pole during fixation of magnetization (Plat = –46.6°, Plong = 153.2°, A₉₅ = 8.3°) is close to the Cambrian interval of the apparent polar wander path (APWP) for Siberia from [Metelkin et al., 2012] (Fig. 9). Thus we have all the evidence needed to confirm that the MTC component is metachronous and related to the Cambrian regional remagnetization on the post-orogenic stage of evolution of the Yenisei accretionary belt. Probably, right after the filling of the Vorogovka depression, which overlaps the accretionary basement of the Isakovka segment of the Yenisei belt, the region underwent subsequent deformations that led to a new fold system disrupting the initial structure of the depression. This was followed by regional heating or chemical alteration, during which a significant part of the terrigenous-carbonate section of the Vorogovka Group was partially remagnetized.

The KHR component has been identified in three sandstone outcrops of the Mutnina anticline. The precision parameter for mean KHR directions in stratigraphic coordinates is significantly higher than in geographic ones Ks/Kg = 120.8/34.3 (Table; Fig. 7 b). This indicates that this component was recorded before deformation and is therefore close to the primary detrital magnetization. Nonetheless, formal criteria of the fold test [МcElhinny, 1964], including the simulated parametric version of the test [Watson and Enkin, 1993] (maximum precision at 122 % untilting with the 95 % confidence ellipse from 60.8 % to 145.7 %) and the correlative test from [Enkin, 2003] (DS Slope: 137.9 % ± 754.1%) do not provide a definite result. The cause could be in the critically low amount of determinations (n = 3) in the tested batch, as well as the above-mentioned assumption of incomplete division of KHR. Such cases have already been reported for the upper Vendian – lower Cambrian part of the Chekurov section (northeastern Siberian platform) [Pavlov et al., 2004], the lower Cambrian sections of the Udzha-Anabar region [Pasenko et al., 2020], the upper Vendian sections of the Angara R. region [Vinogradov et al., 2023] and the southern Baikal region [Shatsillo et al., 2005, 2006].

The KRS component has been identified in one outcrop of the Severnaya Rechka Fm. and in all five studied outcrops of the Sukhaya Rechka Fm. Individual KRS vectors defined directly by component analysis have a bipolar distribution even within a single sampling site. The reversal test for individual vectors is positive, the angle between mean directions of each batch after correction for the same polarity is 5.1° with a critical 16.85° (Fig. 8,c), which corresponds to a ‘C’ classification in [McFadden and McElchinny, 1990]. The pre-folding age of KRS is supported by the ratio Ks/Kg = 107.2/13.7 when comparing site-mean vectors, which is significantly higher than 4.85 for n = 6 at 99 % confidence level by [McElchinny, 1964]. The simulated parametric fold test [Watson and Enkin, 1993] shows maximum precision at 125.5 ± 18.4% untilting (Fig. 8,c). The correlative test [Enkin, 2003] is also positive (DC Slope: 126.12 ± 45.52%). Maximum precision at 126% untilting could be due to the complexity of performing precise bedding attitude measurements, especially in conditions of poor exposure and wavy or crosscutting bedding. Therefore, we can assume that the KRS component corresponds to the geomagnetic field during the accumulation of the Vorogovka Group. The calculated paleomagnetic pole (Plat = –31.6°, Plong = 47.4°, A₉₅ = 4.7°) is close to the southern coast of Madagascar (Fig. 9) and close to the corresponding late Vendian – early Cambrian poles published in [Kirschvink and Rozanov, 1984; Kravchinsky et al., 2001; Kazansky, 2002; Shatsillo et al., 2005, 2006; 2015, 2018; Pavlov et al., 2018; Metelkin et al., 2022; Vinogradov et al., 2023].

For illustrative comparison of the obtained paleomagnetic directions with those available for the territory of Siberia we gathered all currently known determinations dated as Vendian – early Cambrian (600–520 Ma) (Suppl. Mat., Table S2). To avoid duplicates during analysis and exclude those that do not correspond to minimal requirements (obtained without detailed incremental demagnetization and/or component analysis) demanded currently by reliability criteria for paleomagnetic data [van der Voo, 1990; Pechersky and Didenko, 1995] we do not use results obtained before 1990. The only exceptions are two composite determinations that defined the issue under discussion. Thus, the “classic” KHR is represented by an aggregated determination for lower Cambrian Siberian rocks from [Khramov et al., 1982] (pole no. 17 in Table S2 of Suppl. Mat.), and the “classic” KRS is represented by a determination from reference Vendian – lower Cambrian sections along the Lena R. from [Kirschvink and Rozanov, 1984] (pole no. 20 in Table S2 of Suppl.). As a result, the analyzed database of Vendian – early Cambrian paleomagnetic determinations includes 53 poles. Corresponding sampling sites and positions of paleomagnetic poles are shown on Fig. 10. Out of all the data, 19 determinations correspond to KHR vectors and 29 to KRS vectors. To explain their coeval coexistence, two hypotheses have been proposed until now. Adherents to the first one proposed that KRS reflects the Vendian GAD field, and KHR is the result of Cambrian remagnetization (in some cases pre-folding) [Kravchinsky et al., 2001; Kazansky, 2002; Metelkin et al., 2012]. Supporters of the second one propose the contrary – that KHR corresponds to the Vendian GAD field, and KRS is also primary but related to an anomalous state of the field. Specifically, short-lived periods of KRS-field are assumed to have existed when the magnetic dipole had an oblique, mid-latitude or equatorial orientation [Pavlov et al., 2004, 2018; Shatsillo et al., 2020].

We believe both points of view to be partially correct. Specifically, following [Kazansky, 2002; Metelkin, 2012], we have to accept that some KHR vectors have metachronous components. For example, in the studied Vorogovka Group rocks a monopolar, post-folding MTC component is found, with a paleomagnetic pole that is not different from the classic KHR one (Fig. 9). There is also a reliably validated metachronous magnetization also corresponding to the KHR vector in variously aged pre-Vendian rocks, including in gabbro-dolerite dikes and sills of the Ust-Angara complex of the South Yenisei Ridge, the Karagas Group and the Nersa gabbro-dolerite complex of the Biryusa Sayan region (determinations 56–58 on Fig. 10) [Metelkin et al., 2005, 2010]. However, according to the regional locations seen from analyzing the accumulated database of Vendian – Cambrian determinations, the metachronous magnetization coinciding with KHR is not widespread across the entire Siberian paleocontinent. Its traces are distinctly identified only in the southwestern margin (ref. Fig. 10). This indicates that the causes of remagnetization were very local to this region and not related to the territory of the whole paleocontinent. Considering the Cambrian age of this event, we should assume that the re-recording of the paleomagnetic signal could be linked to the geodynamically active accretion of Vendian – Cambrian island arcs, which have been reconstructed in many publications [Zonenshain et al., 1990; Didenko et al., 1994; Kazansky, 2002; Parfenov et al., 2003; Dobretsov and Buslov 2007; Vernikovsky et al., 2009, 2016; Gordienko, 2006; Metelkin, 2013; Gordienko et al., 2023].

On the other hand, we agree with the conclusions in [Pavlov et al., 2004, 2018; Shatsillo et al., 2020] that the KHR component in many studied Siberian sections is primary, and that the explanation of the Vendian geomagnetic phenomenon is related to the unusual behavior of the geomagnetic field. First of all, analysis of the accumulated database shows that a reliably validated primary KHR dominates in the youngest Cambrian (!) rocks age approximately 530 Ma and younger. Probably, this magnetization truly corresponds to a “ordinary” GAD field that existed after the end of the epoch of anomalous Vendian – early Cambrian field. This same “ordinary” field was the background for the discussed remagnetization in the southwestern Siberian paleocontinent. However, in Vendian rocks KHR is rather the reflection of the anomalous, non-GAD field.

Factual estimates published in recent years showing the paleointensity during the Vendian – early Cambrian time 580–530 Ma indicate that the value of the virtual dipole moment fell at least an order of magnitude compared to the current one [Bono et al., 2019; Shcherbakova et al., 2020; Metelkin et al., 2022]. At the same time, according to [Popov et al., 2002; Shatsillo et al., 2015; Bazhenov et al., 2016; Levashova et al., 2021] the frequency of magnetic reversals abruptly increased. Obviously, these two facts are connected. Specifically, the reversal mechanism implies a decrease in the value of the magnetization vector [Glatzmaier and Roberts, 1995]. In this case, the monopolar character of KHR registered by many researchers is a peculiar anomaly, unlike the bipolar KRS, which, as we believe, really reflects the Vendian (!) GAD field. The KRS determinations in the available database mainly correspond to the interval ~580–540 Ma. Within the margin of error, the coordinates of the KRS pole we obtained for the Vorogovka Group are not different from the KRS pole for the Taseeva Group [Vinogradov et al., 2023], which has been dated at ~570 Ma. Age estimates of the studied sedimentary rocks confirm the view that the corresponding paleomagnetic signal was recorded during this time. According to our hypothesis that has already been confirmed for sedimentary and igneous rocks of the Olenek uplift [Metelkin et al., 2022] and the Taseeva Group in the southern Yenisei Ridge [Vinogradov et al., 2023], the KHR component in the Vendian rocks should be related to the Antarctic anomaly. Our model implies the Global Magnetic Anomalies (GMA) have been relatively stationary for a long time, at least the last 600 Myr. We assume that during periods of strongest decrease of the virtual dipole moment the value of the field generated by the GMAs was higher than that of the GAD field. The fast alternation of ultra-brief episodes of dominating anomalous (GMA) and ordinary (GAD) field led to the preservation of an unusual paleomagnetic record, when the KRS component corresponds to the GAD pole and the KHR component – to magnetization due to a GMA.

According to this model, the distribution of KRS vectors would be primarily caused by secular variations of the dipole field. Accepting the GMAs as relatively stationary (with the GAD field conventionally null), the KHR paleomagnetic vectors should ideally have the highest precision parameter. However, this is not so: the coordinates of the KHR paleopoles supposedly related to the GMA are just as diverging as those of KRS (Fig. 10). We think there is a simple explanation. First, the observed KHR is in fact the sum total of the dominating GMA field and some fraction of GAD field, whose input varies. Second, the accepted stationary state of the GMAs is conventional, therefore the magnetization center can still undergo a slight drift. Nonetheless, we presume that the GMA field variations are minimal. In this case, the KHR paleopoles can be transferred to the GAD coordinates system by overlapping the current center of the Antarctic GMA with the GAD field, which is to say, the geographic South Pole. The Euler rotation calculated for this procedure is 100° clockwise around a point with coordinates 74°N, 276°E [Metelkin et al., 2022]. After this correction, the discussed Siberian KHR poles are now located close to their KRS analogs and generally fill the “gap” between the Vendian (~580–540 Ma) KRS pole for the Vorogovka Group and the early Cambrian (~530 Ma) GAD-compliant KHR poles.

• 1) Analysis of the detrital zircon ages in the Vorogovka Group sandstones and their C and Sr isotopic characteristics combined with geological data let us estimate the age of sedimentation in the range 580–535 Ma, after the accretionary structure of the Sayan-Yenisei margin of the Siberian paleocontinent had finished forming.

• 2) In addition to the viscous magnetization, the Vorogovka Group carbonate-terrigenous rocks include three components of different origins and magnetization source. One of them corresponds to the KRS direction, and the other two – to the KHR group of Vendian – early Cambrian poles.

• 3) The anomalous paleomagnetic record typical for Vendian – lower Cambrian rocks of Siberia and measured in the Vorogovka Group sections confirms the geological and geochronological age estimates. The calculated paleomagnetic pole corresponding to the GAD model is close to known poles aged approximately 580–560 Ma, which allows narrowing the sedimentation time to this range.

• 4) Our analysis of the accumulated paleomagnetic determinations database for Vendian – lower Cambrian rocks of Siberia confirms that the KRS poles correspond to the Vendian GAD field that had ultra-frequent reversals. Monopolar KHR directions measured in the same sections are divided into three groups. The first one occurs only in lower Cambrian sections and corresponds to the GAD field at ~530 Ma. The second group was recorded in pre-Vendian and Vendian sections and corresponds to a Cambrian (530 Ma and younger) remagnetization, the traces of which are found mainly in the southwestern margin of the Siberian craton, specifically the MTC component in the Vorogovka Group. The third group is the record of the dominating anomalous source, presumably the long-lived and conventionally stationary Antarctic GMA.

These results supplement the existing paleomagnetic database for the Vendian–early Cambrian interval for Siberia and agree with our previously proposed hypothesis [Metelkin et al., 2022] that the Vendian geomagnetic phenomenon reflects an abrupt weakening of the main dipole component of the geomagnetic field. This resulted in individual episodes when the strength of the axial dipole field fell below the strength of the GMAs. The frequent alternation between brief episodes of mainly anomalous and normal dipole field caused the observed unusual paleomagnetic record. Our analysis of paleomagnetic directions confirms the idea that the epoch of anomalous geomagnetic field took place approximately between 580 Ma and 530 Ma.

This study was funded by grant no. 24-17-00057 from the Russian Science Foundation (processing results of rock-magnetic and paleomagnetic experiments). The research topic is in accordance with programs of fundamental research of the RAS: FWZZ-2022-0001 (IPGG SB RAS, paleotectonic reconstructions based on paleomagnetic data), FWZN-2022-0036 (IGM SB RAS, provenance study from detrital zircon), and the state assignment of GEOKHI RAS (interpretation of Sr isotope data in carbonate rocks); as well as of the Ministry of science and higher education of Russia, program FSUS-2025-0008 (NSU, analysis of the evolution and modeling of the structure of the geomagnetic field at the Precambrian–Paleozoic boundary).

We are grateful to the staff of the Interdisciplinary Center for Analytical Microscopy of Kazan Federal University for performing the SEM investigations.