Early Ediacaran Magmatism in the Yenisei Ridge and Evolution of the Southwestern Margin of the Siberian Craton

We present the first comprehensive data for the early Ediacaran stage of evolution of the western active continental margin of the Siberian Craton (Yenisei Ridge). U-Pb (SHRIMP-II SIMS) data for zircons from dikes of picrodolerites, quartz diorites, and leucocratic granites show that they were emplaced at 625 ± 5, 623 ± 8, and 626 ± 5 Ma, respectively, which indicates a narrow time window of Ediacaran magmatic events. The mafic tholeiitic rocks have OIB and E-MORB mantle components in their magmatic sources. Mineralogical and geochemical observations showed that the mantle-sourced mafic melts assimilated some crustal material, forming Th-enriched dikes of intermediate composition and K- and Rb-enriched felsic rocks. The possible geodynamic conditions for the formation of these early Ediacaran dikes are shown and a geodynamic model is presented for the development of the Yenisei Ridge orogen from the late Cryogenian to the late Ediacaran.


Introduction
The Ediacaran is one of the key evolution stages for fold belts surrounding the Siberian Craton from the north, west, and south. It is in this period that strong tectonic transformations caused by accretionary-collisional events manifested on the Taimyr Peninsula, in the Eastern Sayan and the Transbaikalia Region [1][2][3][4][5]. As became clear in recent years, the Ediacaran stage was one of the main periods in the tectonic evolution of the Yenisei Ridge-A key structure in the western margin of the Siberian Craton [6], whose accretionary-collisional structure still needs to be further investigated. In addition, the Neoproterozoic-Paleozoic accretionary processes on the western margin of Siberia represent the early stages of formation of the Central Asian Orogenic Belt that are still not fully clear. The area adjacent to the western border of the Yenisei Ridge serves as the basement for the Fore-Yenisei sedimentary basin [7], which is essential for understanding the relationships between the platform cover and the folded margin of the Siberian paleocontinent in this time interval.
Previous studies established the features of the tectonic evolution of the Yenisei Ridge orogen in the pre-Ediacaran time-From the transformation of the passive continental margin of the Siberian craton in an active one in the Tonian at~800 Ma, when the Central Angara terrane collided with the Siberian Craton and caused the formation of syncollisional (760-750 Ma) and postcollisional (750-720 Ma) granites [4,6,8,9]. Then from the early Cryogenian to the early Ediacaran (711-628 Ma), a subduction zone evolved along the western margin of Siberia. This led to the synchronous formation of island arcs, which subsequently accreted to the continent, and an active continental margin complex that formed on the Craton in the back-arc position [10][11][12][13]. Unlike the Tonian and Cryogenian-early Ediacaran magmatic events that are well recorded in collisional granites and suprasubductional rocks, the middle Ediacaran magmatism on the Yenisei Ridge was less pronounced and so was less studied. The only formation that was identified indicating the evolution of the subduction system is a late Ediacaran adakite-gabbro-anorthosite association (576-546 Ma) in the South Yenisei Ridge [14].
Therefore, in this study we present new data on the early Ediacaran magmatic stage in the evolution of the western margin of Siberia within the Yenisei Ridge on the example of dike complexes. We aim to determine the geodynamic formation setting of these dike complexes, which will help to complement our tectonic reconstruction of the whole Yenisei Ridge orogen. The origin of Neoproterozoic dikes and associating igneous rocks in the fold belts of the margins of the Siberian Craton is a debatable topic. These magmatic complexes were considered in connection to various geodynamic settings, such as continental rifting during the breakup of the Rodinia supercontinent [15,16], or an active continental margin of Siberia [10,14,[17][18][19][20][21] on the early formation stages of the Central Asian fold belt. To solve these discussions, we will give particular attention to the nature, composition, and age of the Neoproterozoic mafic, intermediate, and felsic dikes.

Geological Setting
Our studies were performed in the southern part of the Yenisei Ridge in the junction zone of the Tatarka-Ishimba and Yenisei sutures along 150-160 km of outcrops along the banks of the Yenisei River (Figures 1 and 2).
The mafic rocks from dikes (picrodolerites and dolerites) were studied in the east bank of Yenisei River, 12-20 km upstream of the Angara River mouth. They form a group of 20 bodies with thicknesses up to 3 m (Figures 2 and 3c). Host rocks in this locality are deformed deposits of the Gorevka Formation composed of limestones, calc-argillaceous shales, and calcareous sandstones (NP 1-2 ) [22]. The 718-Ma (U-Pb age) Strelka pluton granites intrude the sedimentary succession, and thus represent the minimal age for this unit [23]. The carbonate rocks are folded with steeply dipping (80 • ) axial planes towards the west-northwest (285 • -308 • ) or the west-southwest (247-260 • ). These dikes are spatially associated with the axial planes of the folds. Individual dikes and host limestones are in some places cut by low amplitude (<1 m) reverse faults and shallow thrusts with up to 5 m displacement, with fault planes dipping to the west-northeast.
The intermediate and felsic dikes (quartz diorites and leucocratic granites) in the region were studied on the left bank of the Yenisei River, 2 km upstream of the Abeyka River mouth and 2.5 km upstream of the Yarlychikha River mouth, respectively (Figures 2 and 3a,b). The dikes are non-deformed and intrude Neoproterozoic (Cryogenian-Ediacaran) rocks of the Predivinsk terrane: Island arc metagabbroids of the Yarlychikha pluton and metavolcanics associated with plagiogranites of the Yagunov pluton. The volcanogenic sequence of calc-alkaline series is composed of calc-alkaline basalts, basaltic andesites, dacites, rhyodacites, rhyolites, as well as acidic and mafic tuffs (Figure 2b). Results of U-Pb dating of zircons for metarhyolites and plagiogranites placed their formation at 637 ± 5 and 628 ± 3 Ma, respectively [10,11]. The main deformations in these host rocks have a northeastern vergence, showing accretion and obduction processes of attachment of the island arc blocks to the Siberian Craton margin [24]. That is to say, the host rocks, which have a foliation (S 1 ) dipping to 226-235 • at angles of 72-77 • and 234-235 • at angles 35-50 • , are cut by faults deformed into folds with axial planes dipping to the east-southeast (95-103 • ) at angles 10-80 • . Later strike-slip-related deformations manifest as superimposed west-dipping smaller folds and south-dipping hinges (173-185 • , angles 37-65 • ) that are oriented at an angle close to 45 • to the S 1 foliation strike. In addition, in the northeastern part of the Yarlychikha pluton there is a north-northeastern dipping banding in the metagabbroids with kinematic indicators showing sinistral (left-lateral) strike-slip displacements.

Analytical Methods
The results of studies of the chemical composition of the minerals are presented in the Supplementary Materials (Tables S1-S5). These studies were done using a MIRA 3 LMU scanning electron microscope (Tescan Orsay, Brno, Czech Republic) in the Center of Collective Use of Multielement and Isotope Studies of V.S. Sobolev Institute of Geology and Mineralogy SB RAS (Novosibirsk, Russia) (analysts N.S. Karmanov and M.V. Khlestov). The instrument is fitted with an INCA Energy 450 XMax 80 microanalysis system (Oxford Instruments Nanoanalysis Ltd., High Wycombe, UK). The accelerating voltage was 20 kV, and the cathode-ray current was 1.5 nA. The time compilation was 15-20 s, and the resulting spectra were processed automatically. Relative standard deviation during analysis of the main components (concentration >10-15% when recalculated for corresponding oxides) did not exceed 1.5%, and for components with concentrations 1-10 wt.%-no more than 2.0%. Studies of secondary components with concentrations below 1 wt.% were qualitative. FeO and Fe 2 O 3 were obtained by stoichiometric calculation. The whole-rock Mg-number was calculated as Mg# = Mg 2+ /(Mg 2+ + Fe 2+ ).
Results of whole rock X-ray fluorescence and ICP-MS analyses for the dikes are given in Table 1.
All ICP-MS analyses were performed on an Agilent 7500ce and an Element2 double-focusing sector field mass analyzer (ref. Table 1) in A.P. Vinogradov Institute of Geochemistry SB RAS (Irkutsk, Russia) in accordance with the technique described in [28]. Chemical preparation of the samples varied depending on their composition. Mafic rocks and some intermediate rocks were digested in open systems with a mix of acids (HF/HNO 3 /HClO 4 ). Mafic rocks with low elements contents were digested in autoclaves (using a mix of HF/HNO 3 under high pressure with added HClO 4 ). Felsic rocks were fused with anhydrous lithium metaborate. Weighted samples were 100 mg in mass, all HNO 3 solutions had 2-3% acidity, final dilution factor of the samples solutions was 4000-5000 for open system and autoclave dissolution and 10,000 for fusion. Autoclave digestion was performed on a NPVF ANKON-AT analytical system. Fusion was performed using a VCHG-4 high-frequency induction furnace ("OOO Kristall" factory, Taganrog, Russia). HNO 3 , HCl, and HF acids were either of ultrahigh purity according to GOST or of the Suprapure brand, their purification was done on a subPUR/duo PUR Milestone microwave laboratory system, and then a Savillex DST-1000 sub-boiling distillation system. Water was purified using a Millipore-ELIX-3 (Millipore SA, France) system. Element concentrations were determined using graduation by certified CLMS-1-4 solutions (SPEX, USA) with element concentrations 0.1, 1.0, and 5.0 ng/mL and signal drift control using 103 Rh as inner standard. Detection limits were estimated by standard method: C contr + 3σ, where C contr is the mean value of the control experiment, and σ is the relative mean square deviation of measurements (when recalculating for hard sample the dilution coefficient was taken into account). The relative error standard for repeatability (sr, %) was estimated from independent repeated analyses of samples (new dissolution, different analysts, measurements in different days). For state standard reference samples of different compositions, sr was in the range of 5-15% for elements with concentrations above the detection limit.
Correctness of analytical results was checked against approved standard rocks samples: JB-2 (basalt), JGb-1 (gabbro), JG-3 (granodiorite) from Japan; STM-1 (syenite) from the USA; SG-3 (alkaline agpaitic granite), ST-2A (trap), SGD-2A (essexite-gabbro) from IG SB RAS, Russia. U-Th-Pb isotope analyses of zircons from the studied dike rocks (picrodolerite, quartz diorite and leucocratic granite) isolated by standard methodology using heavy liquids and magnetic separation were carried out on the SHRIMP-II SIMS of the Centre of Isotopic Research of VSEGEI (Saint Petersburg, Russia). The zircons were cast in Buehler EpoKwick resin along with standard Temora [31] and 91500 zircons [32]. Then the zircons were half-sectioned and finally polished. Transmitted and reflected light images as well as BSE and CL microphotographs were taken for guidance during analyses.
In-situ U-Pb analyses were performed, applying a secondary electron multiplier in peak-jumping mode following the procedure described in [33]. The analytical data were acquired for nine mass-stations ( 196 Zr 2 O to 254 UO 2 ) with four to five mass-spectra per individual analysis. The Temora zircons were used as the primary Pb/U standard [31], while the 91500 zircon [32] was employed as U concentration and secondary standard, allowing Pb, U, and Th concentrations calculation. Each 5th analysis was done on the Temora zircon, making no less than 10 analyses per analytical session, and ran from a single mount or >20 if two mounts were analyzed during a single session. Uncertainties of the Temora measurements for the entire analytical session are given in the Table 2 footer. The 91500 zircon has been analyzed twice (at least) in the beginning and the end of the analytical session, revealing instrumental drift (if any occurred). The collected results were processed with the SQUID v1.13 [34] and ISOPLOT/Ex 3.22 [35] software. Common lead was corrected using measured 204 Pb/ 206 Pb and model values of Stacey and Kramers [36].

Petrography and Mineralogy
Picrodolerites have a medium-grained structure and an ophitic texture and porphyry phenocrysts of plagioclase, clinopyroxene, and olivine 0.5-1 mm in size. Plagioclase has a labradorite composition (X An from 0.51 to 0.69) and contains low contents of K 2 O, MgO, and FeO as well as BaO in individual samples (Supplementary Materials Table S1, Figures 4a and 5a). In rims, the labradorite is replaced by andesine and albite. Clinopyroxene corresponds to augite with a composition approaching diopside (Table S2, Figure 4a). This mineral has Mg# up to 0.78 and insignificant contents of TiO 2 , Na 2 O, and Cr 2 O 3 . Phlogopite-annite (grains up to 0.03 mm in size) is pleochromatic in light-brown and brownish-red colors, located in interstices between grains of rock-forming minerals and replaced by chlorite (Table S3, Figure 4b Dolerites are fine-and medium-grained with ophitic texture mainly consisting of plagioclase and pyroxene and a small amount of dark mica. The plagioclase (grains up to 0.3 mm) corresponds to labradorite (X An = 0.65-0.50) with trace contents of FeO, MgO, and K 2 O (Table S1, Figures 4a and 5a). Labradorite has reaction rims of albite. Clinopyroxene (grains up to 0.5 mm) is represented by augite close in composition to the augite in picrodolerites (Table S2, Figures 4d and 5b).

Geochemistry
The geochemical particularities of the mafic, intermediate, and felsic rocks reflected both their similarities and differences ( Table 1) (Table 1). In addition, some of the samples from the first group with the highest MgO content (~12 wt.% and higher) correspond to picritic rocks according to the classification of Le Bas [45]. The mafic rocks have high Na2O/K2O ratios (2.0-5.1), and of the FeO*/MgO-SiO2 diagram [46], they plot in the field of the tholeiitic series ( Figure  7). These rocks have significant similarities in distributions of REE and on spider diagrams ( Figure  8) Figure 8a). The rocks of the first group in most samples are distinguished by their higher (La/Yb)N values and closer similarity of trace element distributions to that of OIB. These dike rocks are enriched in large-ion elements-Ba (up to 3451 ppm) and Rb (up to 46.6 ppm) with concentrations that are close to such rocks in OIB, as well as Sr (up to 408 ppm) with concentrations close to transitional rocks between OIB and E-MORB (Table 1, Figure 8b). The mafic rocks have high concentrations of high field strength elements: REE, Th, U, Nb, Ta, Hf, and Zr-Close to those of rocks between OIB and E-MORB.

Geochemistry
The geochemical particularities of the mafic, intermediate, and felsic rocks reflected both their similarities and differences ( Table 1) (Table 1). In addition, some of the samples from the first group with the highest MgO content (~12 wt.% and higher) correspond to picritic rocks according to the classification of Le Bas [45]. The mafic rocks have high Na 2 O/K 2 O ratios (2.0-5.1), and of the FeO*/MgO-SiO 2 diagram [46], they plot in the field of the tholeiitic series (Figure 7). These rocks have significant similarities in distributions of REE and on spider diagrams (Figure 8) Figure 8a). The rocks of the first group in most samples are distinguished by their higher (La/Yb) N values and closer similarity of trace element distributions to that of OIB. These dike rocks are enriched in large-ion elements-Ba (up to 3451 ppm) and Rb (up to 46.6 ppm) with concentrations that are close to such rocks in OIB, as well as Sr (up to 408 ppm) with concentrations close to transitional rocks between OIB and E-MORB (Table 1, Figure 8b). The mafic rocks have high concentrations of high field strength elements: REE, Th, U, Nb, Ta, Hf, and Zr-Close to those of rocks between OIB and E-MORB.
Minerals 2020, 10, x FOR PEER REVIEW 15 of 24 The felsic rocks show the most differentiated trace elements distributions. They have high positive anomalies for large-ion elements (K, Rb, and in some samples-Ba) with concentrations higher than equivalents for OIB, as well as Sr with concentrations close to E-MORB. They also have low negative anomalies for most high field strength elements (especially Ti). Concentrations of Th, U, Ta    The felsic rocks show the most differentiated trace elements distributions. They have high positive anomalies for large-ion elements (K, Rb, and in some samples-Ba) with concentrations higher than equivalents for OIB, as well as Sr with concentrations close to E-MORB. They also have low negative anomalies for most high field strength elements (especially Ti). Concentrations of Th, U, Ta, and Hf are close to rocks between OIB and E-MORB. The felsic rocks either do not have or have moderate Eu anomalies and are distinguished by low values of Σ REE (up to 21.56 ppm) and moderate (La/Yb)N ratios (2.3-4.46). These rocks show a similar inclination of the REE distributions ((La/Yb)N = 2. 3-4.46) with rocks between OIB and E-MORB (Figure 8a).
The intermediate rocks have high concentrations of Ba, HREE, Hf, and Zr (higher than in OIB) and, together with mafic rocks, are characterized by (Tb/Y)N ratios between 1.4 and 1.73, which is close to OIB values. Concentrations of Th in these rocks are similar to those of island arc basalts. The intermediate rocks have flat REE distributions, no Eu anomalies, and LREE concentrations close to rocks between OIB and E-MORB.

Zircon Geochronology
The results of zircon geochronological studies are given in Figure 9 and Table 2. The ages discussed in the text and error ellipses in diagrams with the concordia are 2σ. Ratios and corresponding ages in Table 2 are given with 1σ errors.
The felsic rocks show the most differentiated trace elements distributions. They have high positive anomalies for large-ion elements (K, Rb, and in some samples-Ba) with concentrations higher than equivalents for OIB, as well as Sr with concentrations close to E-MORB. They also have low negative anomalies for most high field strength elements (especially Ti). Concentrations of Th, U, Ta, and Hf are close to rocks between OIB and E-MORB. The felsic rocks either do not have or have moderate Eu anomalies and are distinguished by low values of Σ REE (up to 21.56 ppm) and moderate (La/Yb) N ratios (2.3-4.46). These rocks show a similar inclination of the REE distributions ((La/Yb) N = 2. 3-4.46) with rocks between OIB and E-MORB (Figure 8a).
The intermediate rocks have high concentrations of Ba, HREE, Hf, and Zr (higher than in OIB) and, together with mafic rocks, are characterized by (Tb/Y) N ratios between 1.4 and 1.73, which is close to OIB values. Concentrations of Th in these rocks are similar to those of island arc basalts. The intermediate rocks have flat REE distributions, no Eu anomalies, and LREE concentrations close to rocks between OIB and E-MORB.

Zircon Geochronology
The results of zircon geochronological studies are given in Figure 9 and Table 2. The ages discussed in the text and error ellipses in diagrams with the concordia are 2σ. Ratios and corresponding ages in Table 2 are given with 1σ errors.
Zircons from picrodolerite sample YR16-4 are mostly dipyramidal-short prismatic, euhedral and subeuhedral, clear or slightly colored grains (Figure 9b). Individual grains can have a rounded form. There are few mineral or melt inclusions. Sometimes a concentric growth zoning can be observed in optical photos in transmitted light. CL images show that their structure consists in a combination of rounded (possibly due to reabsorption) dark central cores and bordered by lighter rims with concentric or sector growth zoning. U/Pb SIMS analysis of 25 grains yielded a 206 Pb/ 238 U variation in the range 613-646 Ma ( Table 2). Excluding two of the lowest values (probably due to Pb loss) and the highest three (possible contamination, considering the rounded shape of grains), the concordant age is 625 ± 5 Ma (N = 20, MSWD = 0.01, probability = 0.92) (Figure 9a). This date is most probably the crystallization age of the studied picrodolerite.
Zircons from quartz diorite sample A09-11 are mostly subeuhedral and euhedral dipyramidal-prismatic grains, sometimes with ditetragonal dipyramid faces (Figure 9d). They have a noticeable amount of melt and mineral inclusions. Their inner structure, as observed in CL images, has a dark central core, in which concentric growth zoning often cannot be detected at the selected imaging parameters. Contours of the central parts are typically rounded, apparently due to partial reabsorption. The cores are bordered by light CL rims with concentric oscillatory and sector growth zoning. The structure and morphology of the zircons indicate their magmatic origin. Obtained 206 Pb/ 238 U analysis results are scattered, possibly due to Pb losses. Excluding the three lowest values, the concordant age is 623 ± 8 Ma (N = 7, MSWD = 2, probability = 0.16) (Table 2, Figure 9c), which corresponds to the time of crystallization for the quartz diorite.
Zircons in leucocratic granite sample M08-3 are mainly subeuhedral and euhedral, dipyramidal-prismatic grains, weakly colored, clear or translucent. The tips of pyramids show growth structures (Figure 9f). CL images show varying cores, usually rounded with different zonation types, and surrounded by dark rims. Growth zoning is unclear on CL images but is discernible on some optical microphotos. Three zircons yield significantly older (825 ± 12 Ma, 752 ± 9 Ma and 738 ± 12 Ma) ages in their cores (1.2, 2.2, 3.1) and have low U, Th, and 206 Pb* contents that are anomalous for this sample-These results were not counted in calculations. Excluding two results with the highest U contents and decreased 206 Pb/ 238 U ratios indicating Pb loss from metamict domains, the concordant age for 9 results is 626 ± 5 Ma (N = 9, MSWD = 0.045, probability = 0.83) (Table 2, Figure 9e). This age determines the time of magmatic crystallization of the leucocratic granite.

Discussion
Results of geologic, mineralogical-petrographic, geochemical, and U-Pb geochronological investigations allowed us to identify an association of dikes of Ediacaran age (626-623 Ma). This complements available data on the Ediacaran magmatic history of the Yenisei Ridge and the entire western margin of the Siberian Craton. The rocks of this association, localized in the junction zone of Figure 9. Concordia diagrams and CL images for zircons from picrodolerite sample YR16-4 (a,b), quartz diorite sample A09-11 (c,d) and leucocratic granite sample M08-3 (e,f). Red circles in the zircon CL images indicate the U/Pb isotopic analysis spots. The blue ellipse is the weighted-mean error ellipse.

Discussion
Results of geologic, mineralogical-petrographic, geochemical, and U-Pb geochronological investigations allowed us to identify an association of dikes of Ediacaran age (626-623 Ma). This complements available data on the Ediacaran magmatic history of the Yenisei Ridge and the entire western margin of the Siberian Craton. The rocks of this association, localized in the junction zone of the Tatarka-Ishimba and Yenisei sutures in the South Yenisei Ridge, are diverse in petrography-From picrodolerites-dolerites to quartz diorites and leucocratic granites, which in itself indicates that at this time, magmatism was affected by both mantle and crustal sources. Detailed mineralogical studies showed that these rocks contain Mg-enriched rock-forming minerals and Th-, P-, Cr-, Ti-, and Zr-containing auxiliary minerals. These particularities, as well as the concentrations of large-ion and high field strength elements (Ba, Rb, Sr, Th, U, REE, Nb, Ta, Hf, and Zr) in the studied mafic tholeiitic rocks confirm the input of OIB and E-MORB mantle components in their magmatic source, as shown in (Figure 8). During their evolution, the mantle-sourced mafic melts assimilated some crustal material and formed Th-enriched dikes of intermediate composition and K-and Rb-enriched felsic rocks. Like the mafic rocks, the tholeiitic intermediate rocks retain close concentrations of large-ion and high field strength elements, such as Sr, Ce, Pr, Nd, Sm, and P. Whereas the felsic varieties have a slight similarity of concentrations with mafic rocks that can be observed for individual samples in Sr and Hf content, but also with the intermediate rocks in their Nb and Ta concentrations. The crustal material was probably of metasedimentary nature and pre-Cryogenian in age, which is indicated by U-Pb isotope studies of zircon cores from felsic rocks. On tectonic discrimination diagrams for Th/Yb vs. Nb/Yb from [47] (Figure 10a) and Th/Yb vs. Ta/Yb from [48] (Figure 10b), the mafic rocks appear in transition fields between E-MORB and OIB and between oceanic tholeiitic (TH) and alkalic (ALK) fields respectively. On the first diagram some of the samples plot in the field of volcanic arcs rocks, close to MORB-OIB. This agrees with the possibility of suprasubduction formation conditions for these rocks. Taking into account the composition data for pyroxenes from the dikes studied in the present paper, they could have formed in conditions of an active continental margin. The studied dikes are mostly non-deformed but intrude significantly deformed and weakly metamorphosed sedimentary rocks (metasandstones and marmorized limestones) and igneous rocks (gabbroids). The deformations in the latter are explained by previous tectonic events-the collision of the Central Angara terrane with the Siberian Craton and subsequent accretion of island arc blocks to it [10]. Weak deformations observed in the dikes themselves and in host metasedimentary rocks are related to subsequent tectonothermal events that manifested during the entire Ediacaran in conditions of the evolving transform margin. At the final stage, these tectonothermal processes took place synchronously with the emplacement of adakite dikes and associating gabbro-anorthosites 576-546 Ma that were formed in conditions of a transform active continental margin [14].  Thus, the formation of the dikes in the junction of the Tatarka-Ishimba and Yenisei sutures of the South Yenisei Ridge likely took place during the entire Ediacaran in a transform margin setting due to oblique subduction in the late Cryogenian [18,21]. Studies of the particularities of magmatism in this region thus help to hypothesize a unified geodynamic model ( Figure 11) for the southwestern margin of Siberia during the Ediacaran. The emplacement of supra-subductional island arc complexes in the Cryogenian formed by oblique subduction was followed by magmatism caused by strike slips between plates and the transition of the continental margin into a transform one, which was mainly manifested by widespread dike complexes of diverse composition ( Figure 11). Figure 10. Tectonic discrimination diagrams for the mafic studied rocks. (a) Th/Yb vs. Nb/Yb from [47]; (b) Th/Yb vs. Ta/Yb from [48]. Abbreviations are the types of mafic volcanic rocks: MORBmiddle ocean ridge basalts; N-MORB-normal of middle ocean ridge basalts; E-MORB-enriched of middle ocean ridge basalts; OIB-ocean islands basalts; WPB-within plates basalts; ALK-alkalic; TH-oceanic tholeiitic; VAB-volcanic arcs basalts; IAT-island arc tholeiitic; CAB-calk-alkaline basalts; SHO-shoshonitic; TR-transitions between ALK and TH.

Conclusions
This paper presents new data for the early Ediacaran stage of magmatism in the evolution of the western margin of the Siberian Craton on the example of a dike complex on the Yenisei Ridge. The discussions on the formation conditions of these rocks are ongoing in the literature. We hope that these data and our conclusions can lead to progress in understanding the tectonomagmatic evolution of the western margin of the Siberian Craton in the Ediacaran: Within fragments of the structures of the southwestern margin of the Siberia craton, dikes of various igneous rocks have been identified. The mafic dike rocks formed from enriched mantle sources of OIB and E-MORB types. During the evolution of the mantle-sourced mafic melts there was assimilation of crustal material, which led to the formation of Th-enriched dikes of intermediate composition and K-and Rb-enriched felsic rocks.
U-Pb (SHRIMP-II SIMS) data for zircons from picrodolerites, quartz diorites, and leucocratic granites show that these dikes were emplaced between 626 and 623 Ma, which indicates magmatic events in a narrow time interval in the early Ediacaran.
Our new data for igneous rocks of the early Ediacaran of the Yenisei Ridge allow us to propose a geodynamic model for the Cryogenian-Ediacaran development stage of the whole southwestern active continental margin of the Siberian Craton ( Figure 11). The model shows that in this early Ediacaran stage, subduction was interrupted, slab rollback took place, and the slab began to break off. This created conditions for the introduction of enriched mantle melts that contributed to the formation of the studied mafic rocks. The active continental margin that was formed as a result of oblique subduction subsequently underwent strike-slip displacements between plates and transitioned to a transform stage, which created conditions for the formation of an adakite-gabbro-anorthosite association.