Revisiting the Permian Stratigraphy of the Kuznetsk Coal Basin (Siberia, Russia) Using Radioisotopic Data: Sedimentology, Biotic Events, and Palaeoclimate

Abstract

The radioisotopic dating of five stratigraphic levels within the Permian succession of the Kuznetsk Coal Basin refined the ages of the corresponding stratigraphic units and, for the first time, enabled their direct correlation with the International Chronostratigraphic Chart, 2024. The analysis revealed significant discrepancies between the updated ages and the previously accepted regional scheme (1982–1996). A comparison of regional stratigraphic units’ durations with estimated coal and siliciclastic sediment accumulation rates indicated that the early Permian contains the most prolonged stratigraphic hiatuses. The updated stratigraphic framework enabled re-evaluating the temporal sequence of regional sedimentological, volcano–tectonic and biotic events, allowing for more accurate comparison with the global record. Palaeoclimate reconstructions indicated that during the early Permian, the Kuznetsk Basin was characterised by a relatively warm, humid, and aseasonal climate, consistent with its mid-latitude position during the Late Palaeozoic Ice Age. In contrast, the middle-to-late Permian shows a transition to a temperate, moderately humid climate with pronounced seasonality, differing from the warmhouse conditions of low-latitude palaeoequatorial regions. The latest Lopingian reveals a distinct trend toward increasing dryness, consistent with global palaeoclimate signals associated with the end-Permian crisis.

1. Introduction

The Permian Period was a time of profound geological, climatic, and biotic transformations recorded in sedimentary basins worldwide [1,2,3]. Understanding the timing and nature of these processes requires accurate and regionally constrained stratigraphic frameworks [4,5,6]. In the Kuznetsk Coal Basin (Western Siberia), regarded as a reference region for the stratigraphy of the Siberian palaeocontinent, the regional stratigraphic scheme currently in use—developed over 40 years ago [7]—remains essentially unchanged [8], despite significant advances in geochronology [9]. In particular, the lack of reliable radioisotopic age constraints has limited the direct correlations with the International Chronostratigraphic Chart and has restricted the integration of regional sedimentological data and biotic events into the global record [10].

U-Pb radioisotopic dating has been increasingly applied to coal-bearing successions worldwide, offering critical constraints for stratigraphic correlation and basin-scale sedimentary reconstructions. In Euramerica, high-precision age determinations have refined chronostratigraphic models for basins in North America [11], Western Europe [12,13], and Eastern Europe, particularly the Donets Basin [14]. In the Gondwanan realm, such approaches have proven valuable in Australia [15,16], South America [17,18], and China [19,20], where integration with palaeontological and sedimentological records has enabled more detailed insights into the evolution of coal-bearing basins.

The present study combines previously published [21,22,23] and new U–Pb zircon ages from five stratigraphic levels within the Permian succession of the basin, enabling refinement of the regional stratigraphic framework. The results provide new insights into stratigraphic discontinuities, event correlation, and palaeoclimatic conditions in the mid-latitudes of the Siberian palaeocontinent during the Permian.

The problem of correlating the regional Permian scheme of the Kuznetsk Basin with the Standard Global Chronostratigraphic Scale (SGCS) has been debated for decades [24,25,26,27,28]. Despite the high degree of investigation of the Permian coal-bearing succession in the Kuznetsk Basin [29,30], its direct correlation with the marine-based SGCS (International Chronostratigraphic Chart, 2024) [31], as well as with the General Stratigraphic Scale of Russia (GSSR) [32], remains tentative.

Direct biostratigraphic correlation with the SGCS is impeded by the absence of marine fossil groups—such as conodonts, fusulinids, and ammonoids—that serve as key global chronostratigraphic markers. At the same time, correlation with the Russian GSSR, whose middle and late Permian subdivisions are also primarily based on continental deposits [33], is complicated by the palaeogeographic and biogeographic isolation of the Kuznetsk Basin during the Late Palaeozoic. This isolation resulted in a high degree of endemism in its floral and faunal assemblages [34,35,36].

This study aims to refine the geochronological boundaries of the Permian deposits of the Kuznetsk Coal Basin and to revise the regional stratigraphic framework based on new radioisotopic ages. As part of the study, U–Pb zircon dating was performed on two newly sampled stratigraphic levels using CA-ID-TIMS and LA-ICP-MS, with additional and integrating previously published dates from three other levels. The ages obtained were compared with the International Chronostratigraphic Chart (ICC), enabling direct correlation. A comparative analysis was made between the updated stratigraphic framework and the officially adopted 1982–1996 regional stratigraphic scheme. We focused on evaluating the duration of stratigraphic hiatuses and assessing the completeness of the sedimentary record within regional units. Finally, we used the refined chronology to interpret the sequence of regional sedimentological, biotic, and palaeoclimatic events in the broader context of global environmental changes during the Permian.

2. Geological Setting

The Kuznetsk Coal Basin is one of the largest in Siberia. It is located in the northern part of the Altai–Sayan Folded Area, between the Caledonian and Hercynian orogenic belts (Figure 1A,B). According to current geological models, the Kuznetsk Basin represents a Late Palaeozoic foreland depression formed during the accretion and collision of the Siberian palaeocontinent with the terranes of the Palaeo-Asian Ocean [37,38]. The Caledonian structures of the Kuznetsky Alatau and Gornaya Shoria mark its eastern and southern boundaries. At the same time, its western and northern margins are adjacent to the Hercynian mobile belts of the Salair and Kolyvan–Tomsk zones [39].

Basin formation began in the Middle Devonian, during the early stages of Hercynian orogeny, and was accompanied by the ingression of a marine basin. Marine sedimentation continued until the Visean. Starting from the Serpukhovian, several phases of folding changed the depositional environment from marine to continental, leading to the accumulation of clastic and coal-bearing deposits, which continued through to the end of the Permian. Subsequent tectonic uplift led to block deformation and partial erosion of the previously deposited sequences. The present configuration of the basin was established in the Early Cretaceous, following the tectonic stability of the region [40].

General Characteristics of the Coal-Bearing Succession: The upper part of the Carboniferous and the entire Permian system in the Kuznetsk Basin consists of continental clastic and coal-bearing deposits with a total thickness of 7000–8000 m. The most significant thicknesses are observed in the central and western areas of the basin (Figure 1C), where the most complete stratigraphic sections are located.

Lithologically, the succession includes sandstones, siltstones, and mudstones, containing thick coal seams and thin carbonaceous interbeds. Subordinate occurrences of conglomerate lenses are typically associated with the base of sedimentary cycles. The average coal content in the succession ranges from 1 to 6%, although it may reach up to 14.5% in certain formations [40].

Figure 1. Overview map (A) with outlines of the tectonic scheme (B) showing the position of the study area within the Altai–Sayan fold belt and (C) the locations of studied sections; tectonic scheme is simplified from [41,42,43].

Figure 1. Overview map (A) with outlines of the tectonic scheme (B) showing the position of the study area within the Altai–Sayan fold belt and (C) the locations of studied sections; tectonic scheme is simplified from [41,42,43].

The coal-bearing succession is divided into the Balakhonka and Kolchugino Groups, which contain subordinate subgroups, formations, and regional stages. The boundary between these groups is lithologically defined by a thick (up to 1000 m) barren interval, referred to as the Kuznetsk Subgroup [20]. Biostratigraphically, this boundary coincides with the replacement of two significant floral assemblages: the Balakhonka flora (dominated by cordaitoids) is replaced by the Kolchugino flora (characterised by ferns, pteridosperms, and cordaitoids) [10,29,30]. Our recently published radioisotopic data indicate that the transition from Balakhonka to Kolchugino floras occurred in the late Kungurian [23].

The Balakhonka Group comprises siliciclastic and coal-bearing strata ranging in thickness from 1300 to 3200 m and corresponds to the Pennsylvanian (Lower Balakhonka Subgroup) and Cisuralian (Upper Balakhonka Subgroup).

The Lower Balakhonka Subgroup is 900–980 m thick in the Kemerovo reference area (Figure 1A). It consists predominantly of sandstones (39%), siltstones (38%), mudstones (19%), and coals (2%), with minor lenses of gritstone and conglomerate, along with carbonate and sulphide concretions. The coal content is low, with numerous thin coal seams. Fine-grained sandstones dominate, frequently interbedded with siltstones to form rhythmic, millimetre-scale lamination.

The Upper Balakhonka Subgroup, 700 to 2000 m thick, comprises three formations (Fms): Promezhutochnaya (Transitional), Ishanova, and Kemerovo. In the type section (Kemerovo region), the subgroup is 700–1000 m thick and consists of sandstones (50%), siltstones (40%), mudstones (6%), and coals (4%), with minor occurrences of gritstones and concretions. The subgroup is characterised by pronounced lateral and vertical thickness and lithological composition variations, including substantial coal content and type heterogeneity across the basin [40].

The Kolchugino Group, up to 6000 m thick, spans the upper part of the Cisuralian and the entire Guadalupian and Lopingian series of the Permian system. It includes seven Fms: Starokuznetsk and Mitina (Cisuralian), Kazankovo–Markina, Uskat, Leninsk (Guadalupian), and Gramoteino and Tailugan (Lopingian).

The lower boundary of the group is lithologically defined in the roof of the upper productive coal seam of the Upper Balakhonka subgroup (coal seam Aralicheva I) and generally corresponds to a subregional depositional hiatus. The upper boundary is marked by replacing coal-bearing strata with volcanogenic and siliciclastic rocks of the Abinsk Group, mainly of Triassic age. In complete sections, a gradual transition from the coal-bearing Tailugan Fm to the tuffaceous–clastic Maltseva Fm is observed, making the boundary between the Kolchugino and Abinsk Groups transitional and conventional.

Pyroclastic material and associated clay interbeds—including tuffs, tuffites, montmorillonite clays, and kaolinitic tonsteins—are important components of the Permian coal-bearing strata of the Kuznetsk Basin [44,45]. These clays, primarily composed of kaolinite and montmorillonite, are widely interpreted [46,47] as alteration products of volcanic ash. Among them, tonsteins—hard, usually light-coloured, kaolinite-rich layers commonly found within coal seams—are particularly interesting due to their mineral composition and suitability for high-precision geochronology.

The origin of these ash-derived layers is linked to volcanic activity in surrounding fold belts [48]. In siliciclastic settings, volcanic ash was transformed into plastic montmorillonite clays, whereas in swampy peat-forming environments, it formed kaolinitic tonsteins [47,48]. These are distinguishable from host coals by their light colour, lack of bedding, and occasional presence of idiomorphic zircon grains.

Several studies have examined these pyroclastic layers’ mineralogical and geochemical characteristics, including rare earth and radioactive element concentrations [49,50].

Palaeobiogeographic context: Angaraland: The close palaeogeographic position of the Kuznetsk Basin to the Siberian Platform during the Late Palaeozoic resulted in remarkable floral and faunal similarities between these regions. This vast area was characterised by unique, largely endemic plant and animal assemblages that differed significantly from Euramerica (Laurussia) and Gondwana assemblages. The high degree of floral and faunal endemism formed the basis for defining a separate palaeobiogeographical province—the Angaran Realm (or Angaraland), encompassing biotically related regions across the Siberian palaeocontinent [34].

The Kuznetsk Basin is traditionally considered the reference region of Angaraland. The Carboniferous and Permian floral and faunal assemblages preserved in the basin provide a framework for reconstructing the evolutionary history of continental ecosystems throughout Angaraland and are widely used in interregional biostratigraphic correlations [10,25,28,29,30].

Despite Angaraland’s pronounced floral and faunal endemism, available data on the distribution of specific taxonomic groups among terrestrial plants, conchostracans, non-marine ostracods and bivalves, and freshwater fishes suggest episodes of intermittent biotic exchange between Angaraland, Euramerica, and Gondwana. These exchanges likely occurred during discrete intervals, enabled by the temporary opening of migration corridors [22,27,29,50].

Changes in the Permian stratigraphy: The International Stratigraphic Scale of the Permian was revised in 2004, introducing a threefold division—the Cisuralian, Guadalupian, and Lopingian series. Over the past two decades, considerable progress has been made in defining global chronostratigraphic boundaries. Scientific efforts have led to the formal ratification of GSSPs for most Permian stages, the establishment of precise radioisotopic age constraints, and the refinement of boundaries with the Carboniferous and Triassic systems [1,2,3].

Corresponding updates have been incorporated into the General Stratigraphic Scale of Russia (GSSR), albeit with particular regional distinctions. While the Cisuralian series corresponds to its international counterpart in both name and extent, the middle and upper parts of the Permian are represented by the Biarmian and Tatarian series, respectively, units that diverge from the International Chart in both terminology and temporal range [32,33].

The revision of the International Permian timescale coincided with a decline in geological mapping activity within the Kuznetsk Basin and, more broadly, in the coal-bearing regions of the Altai–Sayan fold belt. As a result, these updates have not been adequately reflected in regional stratigraphy.

The regional stratigraphic scheme of the coal-bearing strata of the Kuznetsk Basin, formally adopted in 1982 [7] and only slightly revised thereafter [8], remains the main working framework for both the Kuznetsk Basin and adjacent regions [28,29,30]. The scheme is highly detailed, comprising groups, subgroups, formations, and regional stages defined based on floral and faunal evidence. Extensive data from cores of thousands of exploration wells, numerous mine sections, and natural outcrops support its resolution. However, much of this empirical knowledge is becoming outdated and increasingly needs revision using modern techniques. The last comprehensive summary of regional stratigraphic research was published over three decades ago in the volume ‘Kuzbass—A Key Region in the Stratigraphy of Angaraland…’ [30].

Despite updates to the International Timescale, current State geological mapping of the Kuznetsk Basin and adjacent coal-bearing regions [51] continues to be based on the 1982–1996 regional stratigraphic scheme [7,8]. In some cases, i.e., the Geological Map of the Kuznetsk Basin [52], the stratigraphic subdivision of the Permian has been aligned with the International Timescale and/or the GSSR. However, this alignment is formal, mainly achieved by ‘mechanically’ transferring the threefold division of the Permian into the framework of the 1982–1996 regional scheme [53].

Thus, the modern International Chronostratigraphic Scale and the outdated regional scheme coexist parallelly within regional practice. This dual framework requires refinement using independent methods, primarily high-precision radioisotopic dating.

The problem of stratigraphic continuity: Most researchers consider the Upper Palaeozoic coal-bearing strata of the Kuznetsk Basin to be a continuous sedimentary record, interrupted only by a single local unconformity at the boundary between the Balakhonka and Kolchugino groups [29,54]. The duration of this unconformity is usually estimated to be much shorter than that of the stratigraphic units it separates [8].

Since the early 1990s, it has been argued that palaeontologists have underestimated potential hiatuses in the coal-bearing succession. For example, Sivtchikov [55] noted that efforts to construct an artificially continuous sequence of faunal and floral assemblages often resulted in incorrect taxonomic definitions and unreliable correlations. More recently, using radioisotopic dating in the adjacent Minusinsk Basin, we estimated the duration of one such hiatus to be approximately 7 Ma [56].

3. Materials

The tonstein samples used for zircon extraction and new radioisotopic dating derive from the Ishanova Fm, coal seam XXX (sample G23-18), South Mezhdurechensk coal field, and the Kemerovo Fm, coal seam VI (sample G23-15), Mezhdurechensk coal field. In addition, sample G23-1 was collected from coal seam XI (Kemerovo Fm). LA-ICP-MS U–Pb analysis showed that this sample contains only detrital zircons (Figure 1 and Figure 2).

This study also considers previously published ²⁰⁶Pb/²³⁸U age data: (a) Maltseva Fm, volcanic ash beds in the lower part of the formation (252.78 ± 0.06 Ma, 252.65 ± 0.08 Ma and 252.33 ± 0.08 Ma) [21]; (b) Tailugan Fm, tonsteins in the lowermost coal seam 78 (257.0 ± 1.3 Ma, 256.6 ± 0.4 Ma) [22]; and (c) Starokuznetsk Fm, volcanic ash bed in the middle part (276.9 ± 0.4 Ma) [23].

The material used for biostratigraphic analysis and reconstruction of the sequence of biotic events was studied using original field collections and museum specimens examined between 2022 and 2025. Information on the main collections and fossil preservation is given below.

The ostracods come from the monographic collections of M.O. Mandelstam, stored in the Aprelevka branch of the All-Union Research Geological Petroleum Institute, coll. 223, 233, 238, 245, 265, 283, 251. They are represented mainly by single valves, moulds, and imprints, and the material is usually poorly preserved.

Insects were studied from original collections and material housed in the Paleontological Institute of the Russian Academy of Sciences (PIN RAS), Moscow. Examined specimens include coll. 742, 840, 1079 (Lower Balakhonka); coll. 504, 679, 966, 1197, 1424 (Upper Balakhonka); coll. 1292 (Ishanova Fm); and coll. 503, 598, 746, 1283, 1435, 5922 (Starokuznetsk Fm). Insect remains are represented mainly by isolated elytra, more rarely by body fragments of apterygotes and disarticulated parts (legs, sclerites) of pterygote orders.

Non-marine bivalves were studied using material from several institutional collections: the Central Scientific Research Geological Prospecting Museum (CSRGP Museum), St. Petersburg (D.M. Fedotov coll. 4672); the Geological Museum of the Territorial Geological Information Fund (GMTGIF), Novokuznetsk (P.A. Tokareva coll. 1342); and the Central Siberian Geological Museum, Novosibirsk (O.A. Betekhtina coll. 295).

Fish remains were examined from the collections of the CSRGP Museum (coll. 2409) and the PIN RAS (coll. 612, 1288, 5784, 5797). The material includes abundant isolated scales, teeth, bones, and complete skeletons.

Plant remains were studied from the collections of the CSRGP Museum (coll. 573, 4898, 5374, 6307, 8269, 9247, 9259) and the GMTGIF (coll. 2553, 2579, 2771). The assemblages are represented by mineralised wood fragments and impressions of vegetative and reproductive organs.

4. Methods and Approaches

The methodology used for U–Pb radioisotopic zircon dating is detailed in Supplement S1. The present study employed Chemical Abrasion, Isotope Dilution, and Thermal Ionisation Mass Spectrometry (CA-ID-TIMS) to obtain high-precision ages from individual zircon grains. Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) was used to rapidly screen zircon age populations based on large sample sets.

The methodology for estimating sedimentation rates (in m/Ma) for coal and siliciclastic deposits, which forms the basis for calculating the expected duration of geological unit accumulation, is described in detail in our previous works [22,57]. This study assumes that the coal accumulation rate is 0.025 mm/year, close to the lowest values reported [22,58]. At this rate, a 100-metre-thick coal seam would require approximately 4 million years to accumulate. Assuming a compaction ratio of 10:1 for peat-to-coal conversion, the inferred rate of peat accumulation is 0.25 mm/year. This estimate is comparable to observed peat accumulation rates in Holocene mires of modern Europe and Siberia [59,60]. According to calculations, over the past 9200–8780 years, the average annual rate of vertical peat accumulation in Western Siberia ranged from 0.21 to 0.57 mm/year [61].

Field observations suggest that most siliciclastic sediments in the coal-bearing succession of the Kuznetsk Basin were deposited rapidly, during avalanche-like events comparable to modern mudflows—essentially instantaneously in geological terms [39]. Nevertheless, in this study, we do not disregard the duration of siliciclastic accumulation and assume a rate of 0.2 mm/year—the minimum value recorded for modern siliciclastic deposition in the Gulf of Maine on the Atlantic coast of North America [62]. At this rate, and assuming uninterrupted deposition, approximately 200 m of sediment would accumulate over one million years.

Total content of depleting microcomponents in coal (ΣDm): This term refers to the cumulative content of coal microcomponents—primarily fusinite—that do not become plastic when heated, do not melt or swell, and do not release liquid or gaseous products [63]. Fusinite, one of the major inertinite macerals, is formed from woody material either by partial microbial oxidation of organic matter under aerobic conditions or as a result of wildfire and/or associated thermal alteration of peat. The wildfire-related origin is currently the prevailing interpretation among researchers, especially palaeoecologists (for a detailed review, see [64]). Peat-forming plant communities that contribute woody material increase ΣDm due to higher fusinite content, regardless of origin. Units of ΣDm are expressed in wt%, with values normalised to the maximum observed content on an ash-free basis. At this rate, and assuming uninterrupted deposition, approximately 200 metres of sediment would accumulate in one million years.

Biotic events were identified based on key fossil localities within formations and subunits, and their levels and ranges were then correlated with the regional scheme.

5. Results

5.1. Radioisotopic Dating

Five radioisotope-dated levels are currently recognised within the Permian succession of the Kuznetsk Basin. Three of these—located within the Maltseva, Tailugan, and Starokuznetsk Fms—have been published previously [21,22,23]. The present study provides the first U–Pb age constraints for two additional levels within the Ishanova and Kemerovo formations (Figure 2).

Single CA-ID-TIMS analyses have very high precision, with measurement uncertainties of ca. 0.1–0.2% corresponding to ± 0.2 to ± 0.6 Ma for most measurements (Supplement S2). For all five dated samples considered in this study, we selected a cluster of concordant U–Pb zircon dates to calculate the weighted mean age in cases where individual measurements yielded divergent results. Such deviations may reflect Pb loss or inherited components. All weighted mean ages are reported with the associated z-error to ensure comparability with results obtained in other laboratories and using different dating methods [65].

Ishanova Fm: High-precision CA-ID-TIMS dating of zircon from a kaolinitic tonstein layer within coal seam XXX (sample G23-18), located in the lower part of the Ishanova Fm, yielded an age of 292.00 ± 0.43 Ma. This mean age is based on the youngest coherent group of 5 zircons while 12 further zircons revealed older ages up to 320 Ma that probably indicate different degrees of inheritance (Figure 2; Supplement S2). Complementary LA-ICP-MS U–Pb analyses of the same sample produced consistent results (Supplement S3). Two main zircon age populations were identified: one at ~323 Ma and a second at ~292 Ma. The younger cluster yielded a weighted mean ²⁰⁶Pb/²³⁸U age of 292.5 ± 1.8 Ma (n = 22).

The CA-ID-TIMS age of 292.00 ± 0.43 Ma suggests that coal seam XXX, and thus the lower boundary of the Ishanova Formation—stratigraphically positioned at the base of coal seam XXXI—correlates with the middle Sakmarian.

Kemerovo Fm: High-precision CA-ID-TIMS dating of zircon from a kaolinitic tonstein layer within coal seam VI (sample G23-15), situated in the middle part of the Kemerovo Fm, yielded an age of 285.6 ± 0.4 Ma. This mean age was calculated from four zircons with identical ages. Five further zircons have slightly younger ages between 285 and 280 Ma, which we interpret as slight Pb loss. Two additional zircons with high analytical errors yielded ages at ca. 288–289 Ma, which probably indicates some inheritance (Supplement S2). This age indicates that the corresponding stratigraphic level falls within the upper Artinskian.

LA-ICP-MS U–Pb dating of a sedimentary interbed from coal seam XI (sample G23-1), initially interpreted as a tonstein, revealed a wide range of zircon ages (Supplement S3). Phanerozoic zircons dominate the sample, while Precambrian grains occur only sporadically (~1571, ~748, and ~595 Ma). Among the Palaeozoic zircons, Cambrian–Ordovician ages are most frequent (74%, 29 grains), with statistically significant peaks at ~524, ~491, and ~456 Ma. Devonian zircons (25%, 10 grains) form a distinct age cluster at ~406 Ma. One younger grain yielded an age of ~334 Ma.

All zircons are prismatic crystals or fragments measuring 150–200 μm, suggesting a relatively proximal source area. Cathodoluminescence imaging shows well-developed oscillatory zoning, typical of magmatic zircons. Th/U ratios further support this interpretation between 0.1 and 1.0. The broad and polymodal age distribution indicates that the sample represents a detrital sedimentary layer rather than a primary tonstein.

Figure 2. Generalised stratigraphic section of the Upper Balakhonka Subgroup (Cisuralian) of the Kuznetsk Coal Basin (A). Star symbols indicate new radioisotopic datings. (B) Single-zircon-grain U-Pb CA-ID-TIMS analyses shown as ²⁰⁶Pb/²³⁸U weighted mean dates (left column) and Concordia diagrams (right column) for both samples, G23-15 and G23-18, and two secondary standards, 91500 and Temora 2, analysed in parallel with the unknowns. In the weighted mean plots, each vertical bar represents a single zircon analysis with its 2σ analytical (internal) uncertainty; grey bars were excluded from the weighted mean calculation. Vertical green boxes denote the weighted mean age with its z-component uncertainty. The uncertainty of the weighted mean CA-ID-TIMS dates is reported as ±x/y/z, where x = 2σ internal, y = 2σ external (including tracer calibration), and z = 2σ external (including ²³⁸U decay constant uncertainty) [65]. Blue bars for the two zircon standards indicate published ages and their uncertainties (a—[66]; b—[65]; c—[67]; d—[68]). In the Concordia diagrams, green ellipses represent data points used in the ²⁰⁶Pb/²³⁸U weighted mean calculation; black ellipses were excluded.

Figure 2. Generalised stratigraphic section of the Upper Balakhonka Subgroup (Cisuralian) of the Kuznetsk Coal Basin (A). Star symbols indicate new radioisotopic datings. (B) Single-zircon-grain U-Pb CA-ID-TIMS analyses shown as ²⁰⁶Pb/²³⁸U weighted mean dates (left column) and Concordia diagrams (right column) for both samples, G23-15 and G23-18, and two secondary standards, 91500 and Temora 2, analysed in parallel with the unknowns. In the weighted mean plots, each vertical bar represents a single zircon analysis with its 2σ analytical (internal) uncertainty; grey bars were excluded from the weighted mean calculation. Vertical green boxes denote the weighted mean age with its z-component uncertainty. The uncertainty of the weighted mean CA-ID-TIMS dates is reported as ±x/y/z, where x = 2σ internal, y = 2σ external (including tracer calibration), and z = 2σ external (including ²³⁸U decay constant uncertainty) [65]. Blue bars for the two zircon standards indicate published ages and their uncertainties (a—[66]; b—[65]; c—[67]; d—[68]). In the Concordia diagrams, green ellipses represent data points used in the ²⁰⁶Pb/²³⁸U weighted mean calculation; black ellipses were excluded.

The newly obtained and previously published radioisotope dates from tonsteins have been used to update the regional Permian scheme. These data accurately represent the stratigraphic succession and its defining characteristics (Figure 3). For clarity, the updated scheme is correlated with the 2024 ICC. Figure 3 also displays interval-based information on coal content (%), number of coal seams and their cumulative thickness, total content of depleting microcomponents (ΣDm), and the total thickness of each stratigraphic unit.

5.2. Coal Accumulation

Coal accumulation patterns show marked contrasts between the early and middle–late Permian intervals. Although both successions are of comparable duration (c. 20 Ma), they differ significantly in coal seam development. The early Permian sequence (total thickness 870 m) contains approximately 60 coal seams, with a cumulative coal thickness of up to 95 m. In contrast, the middle–late Permian sequence (c. 3000 m thick) comprises around 200 coal seams, with a cumulative thickness of up to 320 m.

The available data indicate two distinct phases of coal accumulation, reflecting contrasting sedimentary regimes. These phases are separated by a long (≥4 Ma) depositional hiatus, partly represented by the Kuznetsk Group coal-barren interval (Figure 3).

The total content of depleting microcomponents in coal (ΣDm) also exhibits distinct values and trends between the early and middle–late Permian intervals. The early Permian interval is characterised by generally elevated ΣDm values with a downward trend. In contrast, the middle–late Permian interval shows predominantly lower ΣDm values that gradually increase upward, culminating in a pronounced rise near the Guadalupian–Lopingian boundary (Figure 3).

The Accumulation Duration of Different Rock Types panel of Figure 3 illustrates the estimated durations of coal and siliciclastic deposition. The height of each shaded bar, representing a specific rock type, reflects the accumulation period, calculated using the sedimentation rates assumed in this study: 0.025 mm/year (25 m/Ma) for coal and 0.2 mm/year (200 m/Ma) for siliciclastic rocks.

Calculations indicate that in the upper part of the succession (Gramoteino and Tailugan Fms), the observed thicknesses of coal and siliciclastic intervals are broadly consistent with near-continuous sedimentation. For example, based on the assumed sedimentation rates, the 4.2 Ma duration of the Tailugan Fm would allow for the accumulation of approximately 100 m of coal (4.2 Ma × 25 m/Ma) and 840 m of siliciclastics (4.2 Ma × 200 m/Ma). These values closely match the observed coal thickness (80–100 m) and the siliciclastic component (650–670 m).

For the Gramoteino Fm, the estimated thicknesses are ~60 m of coal (2.5 Ma × 25 m/Ma) and ~500 m of siliciclastics (2.5 Ma × 200 m/Ma), slightly exceeding the observed values of 40–45 m and 400 m, respectively. This discrepancy likely reflects current uncertainty regarding the exact position of the lower boundary of the Gramoteino Fm.

Discontinuous sedimentation and/or the most prolonged stratigraphic gaps—reducing the completeness of the geological record—are evident in the early Permian of the Kuznetsk Basin. A representative example is the Promezhutochnaya (Transitional) Formation. Its basement approximately coincides with the global boundary age of 298.9 ± 0.15 Ma [31], while its upper boundary is constrained at around 292.0 Ma, according to a new CA-ID-TIMS date from the lowermost part of the overlying Ishanova Fm. Assuming uninterrupted accumulation throughout ~6.9 Ma, the expected thicknesses would be ~165 m of coal (6.9 Ma × 25 m/Ma) and ~1380 m of siliciclastics (6.9 Ma × 200 m/Ma). In contrast, the observed values—~20 m of coal and ~440 m of siliciclastics—are less than one-third of the predicted totals, indicating probable significant stratigraphic gaps likely due to erosion, non-deposition, or both.

For the upper part of the Kemerovo Formation—from coal seam VI to the top of coal seam I—the estimated depositional time is about 7.5 Ma (Figure 3). Based on the assumed sedimentation rates, this interval could theoretically accommodate up to 180 m of coal (7.5 Ma × 25 m/Ma) and 1500 m of siliciclastics (7.5 Ma × 200 m/Ma). The observed thicknesses are approximately 30 m of coal and 200 m of siliciclastics, indicating that the preserved strata in this interval are 6–7 times thinner than the theoretical estimates.

5.3. Biotic Events

Biotic events aligned with the Kuznetsk Basin’s updated Permian Regional Stratigraphic Scheme reveal a markedly uneven temporal distribution (Figure 4). The highest concentration of events is associated with the coal-free Kuznetsk Group (Late Kungurian; Starokuznetsk and Mitina Fms) and, as expected, with the Permian–Triassic boundary interval.

Vegetation and invertebrate gigantism—notably among Cordaitanthales, Sphenophytes, and non-marine bivalves—is recorded in three early Permian Fms—the Promezhutochnaya (Transitional), Ishanova, and Kemerovo, as well as in the late Permian Gramoteino Fm.

Faunal migrations—including ostracods, bivalves, and possibly fishes, whose diversity markedly increased during this time—are recorded within the coal-free interval of the Kuznetsk Group, corresponding to the Late Kungurian. This interval also marks the first appearance of xerophytic callipterids.

The dominance of drought-tolerant, small-leaved ‘sulcial’ cordaitoids with thin furrows along the veins on the upper side of the leaf and numerous false veins is characteristic of the Lopingian strata.

The Permian–Triassic boundary is marked by a profound biotic turnover across all fossil groups—plants, ostracods, insects, non-marine bivalves, and fishes. Of particular significance is the floral transition; the Permian Kolchugino Flora gives way to the Triassic Korvunchan Flora, which encompasses nearly all major lineages of early Mesozoic higher plants.

6. Discussion

The U-Pb CA-ID-TIMS radioisotope ages, both from this study and from earlier publications [21,22,23], provide, for the first time, a direct correlation between the Permian Regional Stratigraphic Scheme of the Kuznetsk Basin and the International Chronostratigraphic Chart (2024) [31]. A revision of existing sedimentological and biotic data, integrated into the updated stratigraphic framework, enables reconstruction of the temporal sequence of regional depositional settings and their comparison with the global event record. This updated framework is illustrated in Figure 5. For comparison, the previous 1982–1996 Regional Stratigraphic Scheme [7,8] is also shown. Several key differences between the two schemes are discussed below.

The lower boundaries of most regional units have been shifted to older stratigraphic levels. New U–Pb radioisotopic ages for five Fms—Ishanova, Kemerovo, Starokuznetsk, Tailugan, and Maltseva—directly support this adjustment. For adjacent units lacking direct age constraints, boundary positions were inferred from their relative stratigraphic context. As a result, correlations between regional formations and stages of the ICC have been significantly modified. Importantly, the downward shift in the unit boundaries is consistent with previous biostratigraphic interpretations proposed by Siberian palaeontologists [29,70]. In the updated framework, the lower boundaries of units constrained by radioisotopic data no longer coincide with ICC stage boundaries, thereby probably reducing their convention and enhancing precision (Figure 5).

The early Permian (Cisuralian) part of the revised framework now includes five regional units, reflecting the reassignment of the Kuznetsk Group—comprising the Starokuznetsk and Mitina Fms—to the upper Kungurian (corresponding to the Ufimian regional stage of the GSSR). Radioisotope data and biostratigraphic evidence support this interpretation of the Kuznetsk Group’s stratigraphic position [23,29,70].

In the Lopingian (upper) interval of the regional stratigraphic scheme, the most significant revisions involve the Tailugan and Maltseva Fms. Radioisotopic dating indicates a Wuchiapingian age for the base and most of the Tailugan Fm [22], and a Changhsingian age for the basal strata of the Maltseva Fm [21].

Several issues remain unresolved and require further investigation. First, many Fms—including the Promezhutochnaya (Transitional), Mitina, Kazankovo–Markina, Uskat, Leninsk, and Gramoteino—currently lack radioisotope age constraints, making their stratigraphic position and extent provisional. Second, the duration of individual Fms varies considerably, from as little as ~1 Ma (Mitina Fm) to as much as ~9 Ma (Kemerovo Fm).

Stratigraphic hiatus and its relationship to tectonics, volcanism, and climate: The new radioisotope age constraints and revised regional scheme highlight the probability of stratigraphic discontinuities and an incomplete sedimentary record, especially in the lower part of the Permian coal-bearing succession (Figure 3 and Figure 4).

Direct evidence of an incomplete stratigraphic record—extending even into the coal seams—is provided by our LA-ICP-MS and CA-ID-TIMS U–Pb zircon dating of a siliciclastic interbed within coal seam XI (Kemerovo Fm). The presence of both Precambrian (~1571, ~748, and ~595 Ma) and Early Palaeozoic (~524, ~491, and ~456 Ma) zircon grains in this thin layer indicates that the sediment was significantly reworked from older, exposed source rocks. The occurrence of this detrital layer within the coal seam suggests a temporary interruption in peat accumulation, likely caused by non-depositional or erosional processes.

The occurrence of reworked siliciclastic layers within coal seams highlights episodes of erosion that temporarily disrupted peat accumulation and introduced hidden gaps into the geological record. These interruptions, often overlooked, may record the interplay of local environmental shifts and broader regional controls on sedimentation and preservation (Figure 6).

Tectonic activity is key in influencing erosion, as it can directly affect landscape stability and indirectly control peat accumulation and clastic supply. Its impact is particularly evident in the early Permian part of the coal-bearing succession, where signs of erosion and gaps in sedimentation are most pronounced, as tentatively indicated by the occurrence of thick sandstone bodies and abrupt facies changes.

The early Permian part of the coal-bearing succession is generally composed of poorly sorted siliciclastic rocks composed mainly of quartz (>40–45%), clasts of effusive volcanic rocks, and their tuffs (15–25%), as well as feldspars, siliceous rocks, and plagioclase feldspars [71]. Against this background, certain stratigraphic intervals within the Promezhutochnaya (Transitional) and Ishanova Fms contain thick (up to 120 m) volcaniclastic sandstone and siltstone packages. These rocks are characterised by a dominance (>50%) of angular clasts of effusive rocks and their tuffs, weak cementation, and a characteristic greenish colour [72]. The volcanic and tuffaceous material is interpreted as the product of erosion from lower-to-middle Devonian volcanic complexes exposed in the surrounding highlands [40]. The predominance of effusive and tuffaceous rock fragments, combined with poor sorting, green colour, and weak lithification, suggests rapid sediment transport and high accumulation rates, likely associated with fluvial avulsions or short-lived depositional pulses [73].

Volcanism in the uppermost part of the coal-bearing succession of the Kuznetsk Basin, near the Permian–Triassic boundary, has been well documented in numerous studies [74,75], which link it to the activity of the Siberian Large Igneous Province (LIP) [76,77]. Regionally, it is expressed by an increased content of volcanogenic material in the uppermost part of the Tailugan Fm, as well as by the presence of at least two basalt units—probably lava flows, each several metres thick—within the Maltseva Fm [73].

Figure 6. Synopsis of regional volcano–tectonic, biotic, and palaeoclimate events compared to the global Permian record. Global events adapted from [6,78,79].

Figure 6. Synopsis of regional volcano–tectonic, biotic, and palaeoclimate events compared to the global Permian record. Global events adapted from [6,78,79].

Sills associated with the development and emplacement of the Siberian LIP and trap magmatism are widespread along the margins of the Kuznetsk Basin, particularly in areas adjacent to surrounding fold belts (e.g., the Tom–Usa region). Intrusions of hypabyssal mafic sills (diabase and dolerite; see Figure 2) are common within the Promezhutochnaya (Transitional), Ishanova, and Kemerovo Fms. Their thickness is highly variable, locally reaching up to 120 m, and they often split into several tabular bodies 20 to 40 m thick. Radioisotopic K–Ar dating suggests emplacement ages between 209 and 270 Ma [40].

To better understand the causes of stratigraphic discontinuities, it is necessary to consider the palaeoclimatic conditions that influenced the sedimentation dynamics, erosion, and peat-forming environments within the basin.

Permian Palaeoclimate: Regional Trends and Phases: The Permian climate history of the Kuznetsk Basin reveals a progression from humid and relatively stable conditions in the Cisuralian to increasingly dry and unstable regimes in the Lopingian. Shifts in climate are reflected in changes in coal accumulation patterns, vegetation and faunal composition, and the intensity of wildfire activity. These trends are summarised in

Table 1. Palaeoclimatic indicators by interval of time and their interpretation.

Stratigraphic Interval	Sedimentary and Biotic Features	Interpretation of Setting	Climate Interpretation	Landscape and Sedimentary Environments
Late Permian, Wuchiapingian– Changhsingian coal-bearing succession (Gramoteino and Tailugan Fms)	Sedimentary: – Coal accumulation: 30–40 seams, 80–100 m total thickness; – Elevated content (25– 30 wt%) of depleting microcomponents, including fusinite, in coal. Biotic: – DOM: ‘sulcial’ cordaitoids with thin furrows along the veins on the upper side of the leaf and numerous false veins, as well as Mesozoic-type pteridosperms; – Monogeneric assemblages of non-marine bivalves; – Peak diversity of fish and insects in the later half of the phase; – Notable decline in xylophagous Coleoptera; – Extinction of cordaitoids, other Permian plants, ostracods, non-marine bivalves, and fishes at the end of the phase.	– Local but intense peat accumulation in stable ecological niches – High wildfire frequency – Frequent seasonal drought – Ecological instability and stress, leading to biocoenotic simplification – Temporary improvement of conditions in freshwater ecosystems – Catastrophic change in terrestrial and aquatic environments	Humid Intercalated wet and dry seasons Dryness Seasonality Global hot house at the end of phase	Dry uplands: – reduced vegetation cover; – simplified plant communities; – dominance of ‘sulcial’ cordaitoids and Mesozoic-type pteridosperms; – pronounced dryness; – frequent wildfires. Transitional zone: – open woodlands and shrublands; – flammable biomass actively transported into peatlands; – widespread wildfires; – peak insect diversity prior to the end-Permian crisis, with diminishing distinctions between regional entomofaunas. Peatlands: – localised but stable peat accumulation; – frequent wildfires, with potential gaps in peat accumulation and erosion; – proximity of forest vegetation to peatlands. Freshwater systems (lakes and channels): – monogeneric non-marine bivalve assemblages indicate ecological degradation of aquatic systems; – peak fish diversity in the latter part of the phase suggests temporary environmental stabilisation prior to extinction events; – the phase concludes with the extinction of Permian plants, invertebrates, and fishes. Fluvial systems: accumulation of clastic sediments; widespread erosion.
Middle Permian, Wordian–Capitanian coal-bearing succession (Leninsk Fm)	Sedimentary: – Coal accumulation represented by ~20 seams, with a total thickness of 35–50 m; – Low content (<10 wt%) of depleting microcomponents (including fusinite) in most of the interval; – Sharp increase in fusinite content to 25–30 wt% near the top of the section. Biotic: – LAD: Rufloria; – presence of distinct growth rings in fossil wood; – Im: fish from Euramerica; – FAD: new fish genera.	– Fragmented and unstable peatlands – Relatively high humidity for most of the phase – Intensifying dryness and more frequent wildfires towards the end of the stage.	Humid Humid Episodic dry phases Declining humidity and increasing dryness Seasonality Intermittent wet phases, floods, and ephemeral water bodies	Dry uplands: – sparse vegetation dominated by xerophytic forms (cordaitoids, callipterids); – trees with well-defined growth rings. Transitional zone: – open shrublands and woodlands; – partial preservation of woody vegetation. Peatlands: – locally developed; – unstable peat accumulation; – probable periodic drying; – wildfires rare in the early phase, becoming more frequent toward the end. Freshwater systems (lakes and channels): – likely increased seasonal water-level fluctuations; – episodic hydrological connections between neighbouring biogeographical provinces. Fluvial systems: accumulation of clastic sediments; widespread erosion.
Middle Permian, Roadian coal-bearing succession (Kazankova-Markina and Uskat Fms)	Sedimentary: – Coal accumulation: 95–145 seams; total thickness 35–120 m; – Low content (<10 wt%) of depleting microcomponents, including fusinite, in coal. Biotic: – FOD of the subgenus Rufloria; DOM of Tungussocarpus; – Small-leaved and ‘sulcial’ cordaitoids; callipterids (Callipteris); – Presence of distinct growth rings in fossil wood; – FOD of Permochoristidae (Mecoptera); – LOD of Palaeoptera (except Odonata), Archescytinidae; – Decline in Homoptera abundance; dominance of Prosbolidae among Homoptera; – Development of coarse elytral sculpture and thickening in Coleoptera (DOM: Schizocoleidae; Rhombocoleidae); – DOM: ostracods and non-marine bivalves that had previously migrated into the basin; – D: freshwater fishes.	Vigorous plant growth; peatlands with anaerobic conditions favourable for peat accumulation; Rare and/or low-intensity wildfires; Absence of forest vegetation in proximity to peatlands; Peat-forming and wetland ecosystems; Dry episodes; Mosaic of peatlands interspersed with open woodlands or/and shrublands; Seasonal variation in temperature and humidity; Seasonal conditions with stable biotic communities; Stable freshwater environments; High resilience of aquatic ecosystems.	Humid Humid Seasonal, short dry episodes Seasonality Stability	Dry uplands: – small-leaved, ‘sulcial’ cordaitoids and callipterids adapted to drought; – colonisation by drought-tolerant insect groups; – strongly seasonal climate; – infrequent wildfires. Transitional zone: – lowland areas with open woodlands or/and shrublands; – patchy vegetation cover; – low woody productivity; – abundance of Mecoptera (Permochoristidae), dominance of Prosbolidae and schizophoroid beetles. Peatlands: – active peat accumulation; – peat-forming vegetation; – stable humidity, absence of wildfires; – resilient ecosystem, sensitive to climatic seasonality. Freshwater systems (lakes and channels): – stable hydrological regime with seasonal fluctuations. Fluvial systems (perennial and intermittent): accumulation of clastic sediments; widespread erosion.
Early Permian, Late Kungurian coal-free succession (Starokuznetsk and Mitina Fms)	Sedimentary: – Coal accumulation virtually absent or removed by erosion.	Peat failed to accumulate or was eroded by seasonal floods.	Water balance disturbance; unstable moisture	Dry uplands: – xerophytic callipterids with narrow, folded leaves; narrowing of sphenopsid stems; – moisture deficit and unstable hydrological conditions. Open lowland landscapes: – sparse woodland, shrub vegetation, mosaic plant communities; –dominance of Homoptera (indicative of seasonal biocenoses); insects active during short wet intervals. Swampy areas and ephemeral channels: – minimal peat accumulation; – frequent erosion and reworking of sediments; – predominantly clastic deposition; – seasonal flooding and ephemeral water bodies; Freshwater systems (lakes and channels): – immigration of freshwater invertebrates from other palaeobiogeographic provinces; – temporarily stable aquatic environments. Fluvial systems: accumulation of clastic sediments; widespread erosion.
	Biotic: – FAD: xerophytic callipterids; – LOD: Blattodea; – Dom: Hemiptera: Homoptera; Mecoptera, and Grylloblattodea; – Im: ostracods from Kazakhstan; – Im: bivalves from Euramerica; – Im: Coleoptera from Euramerica; – FAD: new fish genera.	Plants adapted to drought and intense solar radiation. Loss of humid, shaded forest biotopes; disappearance of forest ecosystems Insects associated with seasonally active plant communities. Emergence of temporary hydrological routes between biogeographical provinces. Transformation of ecosystems and changes in freshwater environments.	Progressive dryness; rising temperatures; decreasing precipitation Intensification of dryness Seasonal climate with a short wet period Intermittent wetting; flood events; ephemeral water bodies Temporary and permanent well-aerated water bodies
Early Permian, Asselian, Sakmarian, and early Kungurian coal-bearing succession (Promezhutochnaya (Transitional), Ishanova, and Kemerovo Fms)	Sedimentary: – Coal accumulation: 20–40 seams with a total thickness of 55–65 m;	Active vegetation growth; extensive peat-forming wetlands under persistently anaerobic conditions	Humid and stable regime	Wet lowland forested plains: – presence of giant trees (lacking growth rings); – high biomass productivity; – development of forest litter supporting Blattodea dominance. Transitional zone: – forests adjoin peatlands; – accumulation of woody debris as habitat for cockroaches and a potential source of combustible material during wildfires. Peatlands: – persistently wet, anoxic conditions; – occurrence of charred plant remains and fusinite, indicating recurrent wildfire activity. Freshwater systems (lakes and channels): – habitats for non-marine bivalves and fish; – stable water levels, not prone to desiccation; – proximity to wetlands, with potential input of carbonaceous material to peatlands Fluvial systems: accumulation of clastic sediments; widespread erosion.
	– High content of depleting microcomponents (ΣDm <30–70 wt%), including fusinite.	Periodic droughts triggering wildfire events Presence of forested vegetation in proximity to peatlands	Short dry episodes
	Biotic: – Plant gigantism; – Absence of growth rings in fossil wood; – Gigantism: non-marine bivalves; – Dom: Blattodea.	Absence of climatic stressors (e.g., frost, drought) Continuous, non-seasonal plant growth throughout the year Well-developed freshwater ecosystems with stable water supply	Warm, humid, and stable Little or no evidence of seasonality

Note. D:—diversity; DOM:—dominance; FAD: (FOD:)—first appearance (occurrence) datum; LAD: (LOD:)—last appearance (occurrence) datum; Im:—immigration event from another realm; ΣDm—total content of depleting microcomponents in coal; wt% (ash-free basis, normalised).

, which integrates sedimentological and biotic indicators (Figure 3 and Figure 4) to characterise the prevailing settings during different time intervals.

Early Permian, Asselian, Sakmarian, and early Kungurian coal-bearing succession (Promezhutochnaya (Transitional), Ishanova, and Kemerovo Fms): The climate during this phase was relatively warm and humid, with minimal annual temperature variability. This interpretation is supported by the absence of distinct growth rings in fossil wood, the dominance of cockroaches (Blattodea) in insect assemblages, and the widespread distribution of peatlands [80,81,82]. The gigantism observed in plants and non-marine bivalves suggests long-term ecological stability in terrestrial and freshwater ecosystems and probably elevated atmospheric oxygen levels [83].

High fusinite content in coal indicates frequent wildfire activity, likely associated with short dry episodes and the proximity of forest vegetation to peat mires. Together with elevated oxygen levels, these factors would have favoured ignition and the incorporation of charred material into accumulating peat. This pattern is consistent with observations from the Xingtai Coalfield (Middle Permian) in the North China Basin, where wildfires occurred despite generally humid conditions [84].

Early Permian, Late Kungurian coal-free succession (Starokuznetsk and Mitina Fms): The climate was temperate, semi-humid, with pronounced seasonality, and an unstable hydrological regime, including periodic flooding. Short humid episodes probably interrupted long periods of relatively warm and dry conditions. Peat accumulation was either suppressed or disrupted by erosion. The appearance of xerophytic callipterids—migrants from lower latitudes—in the plant assemblage suggests increasing drought stress [85]. Other indicators of dry climate include narrowing of sphenopsid stems, reduced leaf size in cordaitaleans, thicker leaf textures, and denser venation patterns [29].

Evidence of episodic connections with neighbouring biogeographical provinces is provided by the immigration of freshwater ostracods (from Kazakhstan) and non-marine bivalves (from Euramerica), suggesting temporary hydrological routes during wetter phases. Insect assemblages also reflect climate variability. Thermophilic Blattodea are completely absent (reappearing only in Maltseva Fm), while the fauna is dominated by Hemiptera (Homoptera), Mecoptera, and Grylloblattodea [82], taxa probably associated with riparian vegetation and understorey habitats. Among the Homoptera, the hydrophilic Scytinopteridae dominate. These insects are thought to have been adapted to moist, marginal environments and may have been capable of temporary submersion [86,87]. The phase marks the first significant expansion of Coleoptera, dominated by xylophagous Permocupedidae with distinct cupedoid elytra [88]. Flood-induced erosion produced spatially heterogeneous and dynamic landscapes.

Middle Permian, Roadian coal-bearing succession (Kazankovo-Markina and Uskat Fms): The climate was temperate, moderately humid, and seasonal, with short dry episodes. The presence of growth rings in fossil wood, the coexistence of xerophytic and hydrophytic plant forms, and the persistence of long-lived peat-forming wetlands support this.

Insects transitioned from typical early Permian assemblages to the Late Permian ones. In particular, the large open-habitat Palaeoptera disappears almost completely. Among the Mecoptera, the Permochoristidae became dominant [82]. Within Homoptera, the Prosbolidae dominate—a group inhabiting a variety of plant hosts, including cordaitaleans [86]. Among Coleoptera, there is an increase in forms with schizophoroid-type elytra (families Schizocoleidae, Rhombocoleidae) and a decrease in cupedoid types (families Permocupedidae, Taldycupedidae) [88]. The dominance of schizophoroid beetles—characterised by thick, coarsely sculptured elytra—along with the absence of moisture-dependent Archescytinidae and the reduced presence of Scytinopteridae (Hemiptera: Homoptera), supports the interpretation of episodic dry conditions [82,88].

Peat-forming wetlands indicate stable, though seasonally variable, humidity. Small-leaved Rufloria and mosses such as Polyssaievia and other moisture-dependent taxa dominated the mire vegetation. The low fusinite content in coal (<10%) suggests infrequent wildfire activity and the absence of extreme droughts. Freshwater ecosystems appear stable and diverse, supporting rich assemblages of ostracods, non-marine bivalves, and fishes, including migrants from neighbouring biogeographical provinces. The evidence points to a heterogeneous landscape of peatlands, open woodlands, and sustained water bodies, developed under a moderately humid climate with well-defined, though not extreme, seasonality.

Middle Permian, Wordian–Capitanian coal-bearing succession (Leninsk Fm): The climate was temperate, moderately humid, with equal wet and dry seasons. Peat accumulation slightly declined, suggesting localised peatlands. For most of the phase, fusinite content in coal remained low (<10%), indicating persistently high humidity, at least seasonally. A sharp increase in fusinite content (up to 25–30 wt%) in the upper part of the Leninsk Fm indicates increasing dryness and more frequent wildfires.

The near disappearance of Rufloria [89] and the spread of xerophytic forms—including callipers and small-leaved cordaitoids—reflect the trend toward degrading peatland ecosystems. Nevertheless, well-defined annual growth rings in fossil wood [81,90] confirm the persistence of a seasonal climate. Fish migrations from Euramerica and the appearance of new genera may be linked to forming water bodies that connected adjacent biogeographical provinces. Overall, the Wordian–Capitanian climate represents a transitional phase towards the increasingly dry semi-humid conditions of the Late Permian.

Late Permian, Wuchiapingian–Changhsingian coal-bearing succession (Gramoteino and Tailugan Fms): Climatic conditions were strongly seasonal, with a clear trend toward progressive dryness. These changes favoured increasing ecological instability and widespread ecosystem degradation [91]. Biotic communities became structurally simplified, r-selection intensified, specialist taxa declined, polymorphism was reduced, and opportunistic species capable of rapidly colonising unstable or vacant ecological niches became increasingly dominant.

High fusinite contents (25–30 wt%) in coal seams—particularly in seams 78 and 88 of the Tailugan Fm—indicate frequent wildfire activity [92], driven by alternating wet and dry seasons and the proximity of forest areas to peatlands. Although the coal-forming regime persisted, peat accumulation was spatially restricted to relatively stable local environments.

The vegetation structure differed markedly between the beginning and the end of this phase. The earlier interval is characterised by the widespread occurrence and gigantism with both sparsely veined and ‘sulcial’ leaves of cordaitoids, sphenopsids with large stems and leaf whorls of the Annularia. Leaf mosses, diverse pteridosperms (Comia, Permocallipteris), proto-ginkgophytes (Psygmophyllum [syn. Iniopteris], Rhipidopsis), and cycadophyte foliage (Yavorskyia) were also common [93]. In contrast, the latter part of the phase is dominated by xerophytic, small-leaved ‘sulcial’ cordaitoids and Mesozoic-type pteridosperms. Leaf mosses disappear, indicating a shortened growing season and increasingly unstable, seasonally dry conditions.

At the beginning of this phase, an increase in the rate of new insect family origination is observed, leading to a peak in taxonomic diversity and the blurring of distinctions between regional entomofaunas [86]. In the latter part of the phase, a reduction in insect body size is documented, especially in Coleoptera, along with a declining role of xylophagous taxa. This trend is reflected in the dominance of primitive schizophoroid beetles over the increasingly rare cupedoids [88].

Freshwater ecosystems were dominated by monogeneric bivalve assemblages, indicating ecological degradation. The second half of the phase records a peak in fish diversity [94], probably associated with a short-lived episode of environmental stabilisation in aquatic habitats.

The final part of the phase records signals of a major biotic crisis (Figure 4 and Figure 6), marked by the disappearance of the most characteristic Permian floral and faunal groups. These patterns align with the global trends associated with the end-Permian mass extinction.

Comparative studies of other Permian coal basins, including those in South America [95], South Africa [96], and Australia [97], offer valuable insights into palaeoclimatic diversity. Future work may benefit from such multi-basin perspectives to contextualise the Siberian record.

7. Conclusions

High-precision U–Pb radioisotope ages obtained using CA-IDTIMS and LA-ICP-MS methods from five stratigraphic levels—including new results for the Ishanova and Kemerovo Fms—provide robust age constraints for key units within the Permian succession of the Kuznetsk Basin. Biostratigraphic data from previous and present studies independently support these results.

For the first time, the new radioisotopic dates allow for a direct correlation of regional units with the International Chronostratigraphic Chart, revealing significant discrepancies in previous correlations, particularly at the Cisuralian–Guadalupian boundary and within the Late Permian interval.

A comparison with the official 1982–1996 regional stratigraphic scheme reveals significant inconsistencies in the age and duration of several units. In response, a revised scheme is proposed, integrating both biostratigraphic and radioisotopic data.

Estimates of sedimentation rates and cumulative durations of coal-bearing intervals were used to assess the extent and timing of stratigraphic gaps. The results show that hiatuses’ most significant cumulative duration occurs within the Cisuralian, particularly in the late Kungurian.

Palaeobiological data reflect distinct climatic phases during the Permian: a relatively warm and humid Asselian–Sakmarian, a semi-humid late Kungurian, a temperate and moderately humid Guadalupian, and an increasingly dry Lopingian. These reconstructions are broadly consistent with the global climate record but also show regional variations, in particular, prevailing temperate humid or moderately humid conditions corresponding to the middle latitude position of the basin during the Permian.

U–Pb radioisotopic dating, combined with sedimentation rate analysis, has proven to be an effective tool for verifying and refining regional stratigraphic frameworks. This integrated approach allows for a more accurate definition of unit boundaries and regional data integration into a global chronostratigraphic context.