<<

https://doi.org/10.1130/G47365.1

Manuscript received 3 January 2020 Revised manuscript received 11 May 2020 Manuscript accepted 18 May 2020

© 2020 The Authors. Gold Open Access: This paper is published under the terms of the CC-BY license. Published online 12 June 2020

Field evidence for coal combustion links the 252 Ma Siberian Traps with global carbon disruption L.T. Elkins-Tanton1, S.E. Grasby2, B.A. Black3,4, R.V. Veselovskiy5,6, O.H. Ardakani2 and F. Goodarzi7 1School of Earth and Space Exploration, Arizona State University, 781 Terrace Mall, Tempe, Arizona 95287, USA 2Geological Survey of Canada, Natural Resources Canada, 3303 33rd Street NW, Calgary, Alberta T2L 2A7, Canada 3Department of Earth and Atmospheric Science, City College of New York, 160 Convent Avenue, New York, New York 10031, USA 4Earth and Environmental Science, City University of New York Graduate Center, 365 Fifth Avenue, New York, New York, USA 5Institute of Physics of the Earth, Russian Academy of Sciences, Moscow 123242, 6Geological Faculty, Lomonosov Moscow State University, Moscow 119991, Russia 7FG & Partners Ltd., 219 Hawkside Mews NW, Calgary, Alberta T3G 3J4, Canada

ABSTRACT and Czamanske, 1997). Near , the basal The - extinction was the most severe in Earth history. The Siberian Traps volcaniclastic sequences is typically only several eruptions are strongly implicated in the global atmospheric changes that likely drove the meters thick. extinction. A sharp negative carbon isotope excursion coincides within geochronological The presence of coal layers at the uncertainty with the oldest dated rocks from the Norilsk section of the Siberian flood . end-Permian boundary in Arctic Canada pro- We focused on the voluminous volcaniclastic rocks of the Siberian Traps, relatively unstud- vides tantalizing evidence for coal combustion ied as potential carriers of carbon-bearing gases. Over six field seasons we collected rocks at that time (Grasby et al., 2011). We examined from across the Siberian platform, and we show here the first direct evidence that the earli- the organic carbon content of the Siberian Traps est eruptions in the southern part of the province burned large volumes of a combination rocks with a particular focus on early volcanicla- of vegetation and coal. We demonstrate that the volume and composition of organic matter stic rocks spanning the interacting with magmas may explain the global carbon isotope signal and may have signifi- (Table 1), from earliest eruptions to latest (here cantly driven the extinction. we are referring to the late-stage rocks of the Maymecha-Kotuy region), to provide a compre- INTRODUCTION morphism and combustion of coal, carbonates, hensive assessment of organic-matter incorpo-

With loss of >90% of marine species, the and organic-rich shales produce significant CO2 ration during magmatism. Here we give further

Permian-Triassic extinction was the most severe and CH4, as well as carbonate metamorphism evidence that Siberian Traps magmas intruded

in Earth history (Erwin, 2006). High-precision producing CO2, in addition to the gases released into and incorporated coal and organic material, geochronology implicates Siberian Traps erup- by volcanics, all of which would have contrib- and, for the first time, give direct evidence that tions in the global environmental changes that uted to global warming (Retallack and Jahren, the magmas also combusted large quantities of caused the extinction (Wignall, 2001; Grasby 2008; Svensen et al., 2009; Iacono-Marziano coal and organic matter during eruption. et al., 2011; Burgess and Bowring, 2015; Bur- et al., 2012). However, the magnitude, tempo, gess et al., 2017) and perturbation, and origin of carbon emissions during Siberian METHODS AND RESULTS including a sharp negative carbon isotope excur- Traps magmatism have remained in question Field Sampling sion that is a key feature of the mass-extinction despite their critical atmospheric importance We sampled along a traverse north from Ust- interval (e.g., Payne and Clapham, 2012). This (Cui and Kump, 2015; Black et al., 2018). Ilimsk along ∼200 km of the Angara River, and carbon isotope excursion coincides within geo- The earliest volcanic deposits of the Siberian a similar distance along the Nizhnyaya Tun- chronological uncertainty with the oldest dated Traps include volcaniclastic rocks that overlie guska River centering on Tura (Fig. 1). Almost rocks from the Norilsk section of the Siberian sedimentary rocks and underlie the every outcrop on these rivers consists of thick flood basalts (Burgess and Bowring, 2015). main pile in the southern regions of the sequences of volcaniclastic rocks, which have Siberian Traps magmas were chambered province (Naumov and Ankudimova, 1995). been mapped in direct contact with upper Perm- within, and intruded through, the Tunguska sedi- The thickest volcaniclastic rocks are near the ian sedimentary rocks (Malich et al., 1974). Car- mentary sequence (Il’yukhina and Verbitskaya, town of Tura and farther south (Fig. 1). Near bonized woody fragments as much as 10 cm in 1976). The Tunguska Basin varies between 3 Tura, drill cores reveal >600 m of volcaniclastic length were embedded in a number of outcrops and 12 km thick, and includes carbonates, evap- rocks, grading directly into the earliest of on both the Angara and Nizhnyaya Tunguska orites, oil and gas, and coal (e.g., Svensen et al., the flood basalts (Levitan and Zastoina, 1985). Rivers (Black et al., 2015). No exposures of 2018). Coal strata range in age from Carbonifer- In the Maymecha-Kotuy region, the Pravoboyar- Permian and older coal layers were observed ous to Permian, with a cumulative coal thickness sky Suite basal volcaniclastic sequence reaches along either river. However, dolerite in a coal of ∼100 m (Ryabov et al., 2014). Thermal meta- a maximum thickness of 200–300 m (Fedorenko quarry near the Angara River in Ust-Ilimsk

CITATION: Elkins-Tanton, L.T., et al., 2020, Field evidence for coal combustion links the 252 Ma Siberian Traps with global carbon disruption: , v. 48, p. 986–991, https://doi.org/10.1130/G47365.1

986 www.gsapubs.org | Volume 48 | Number 10 | GEOLOGY | Geological Society of America

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/48/10/986/5146678/986.pdf by guest on 02 October 2021 Figure 1. Map showing early southern volcanicla- stics of the Siberian Traps that display abundant evi- dence for coal burning. Black dots mark sam- pling areas from the larger Siberian flood basalts and end-Permian extinction research project (funded under U.S. National Sci- ence Foundation grant EAR-0807585). Samples analyzed but with no coal found are labeled in gray with blue circles; those with red circles and black labels contain combusted, coked, or thermally altered coal. The majority of southern volcaniclastics analyzed contain coal, but only one from the north- ern Kotuy and Norilsk regions; in the intermedi- ate Nizhnyaya Tunguska region, three out of eight samples contained coal. Map after Svensen et al. (2009) and Malich et al. (1974); stratigraphic column after Permyakov et al. (2012). On the strati-

graphic column: P1br1 and P1br2—lower and upper parts of the Burguklins- kaya Formation (Lower Permian), respectively;

P2in1—lower part of the Inganbinskaya Forma- tion (Middle Permian);

T1tt—Tutonchanskaya For- mation (Lower Triassic);

T1uc1 and T1uc2—lower and upper parts of the the Uchamskaya For- mation (Lower Triassic), respectively.

­contains vesicles filled with carbon-rich mate- ­whole-rock bulk carbon content of as much as that span five geographically separated regions rial, malleable with a fingernail Fig. 2( ). 2.7 wt%, and total organic carbon (TOC) con- had visible large organic fragments enclosed in We examined 16 samples of volcaniclastic tent from 0.01 to 1.16 wt% (Table 1; see the the rock matrix (Fig. 2) as well as organic mac- rocks from the Angara, Nizhnyaya Tunguska, Supplemental Material). As context, TOC values erals visible under the microscope (Figs. 3A– and Podkamennaya Tunguska Rivers for carbon in shales of >0.5 wt% have potential as a petro- 3N; Table S1). High values of random vitrinite

content, along with six samples from northern leum source rock (Peters and Cassa, 1994); the reflectance (Ror) are indicative of higher ther- regions (Table 1; Fig. 1; detailed localities are volcaniclastic rocks studied here may exceed mal maturation of organic matter. The thermal provided in Figs. S3–S7 in the Supplemen- the carbon threshold for an economically viable maturity of the particles ranged from marginally 1 tal Material ). These rocks have a range of petrochemical source. mature to mature (Ror = 0.56%–0.83%), indi- cating the varying degree that organic matter 1Supplemental Material. Methods and detailed Characteristics of Burnt Coal and Organic was thermally altered by incorporation into the location maps for samples. Please visit https://doi​ Matter in Siberian Volcaniclastic Rocks magma. We divided the organic particles into .org/10.1130/GEOL.S.12425381 to access the supple- Samples were prepared as crushed-rock pol- three general maceral types based on morphol- mental material, and contact [email protected] with any questions. Additional sample material is avail- ished pellets and examined under reflected light ogy and thermal maturation (Table 1). able from the corresponding author (L.T. Elkins-Tanton) (Table 1; Table S1 in the Supplemental Mate- Type 1 macerals are coal fragments within at Arizona State University, Tempe, Arizona, USA. rial). Of the 22 samples examined, 11 samples the volcaniclastic host rock that predominantly

Geological Society of America | GEOLOGY | Volume 48 | Number 10 | www.gsapubs.org 987

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/48/10/986/5146678/986.pdf by guest on 02 October 2021

(°E) 102.673 99.9726 98.23712 98.23712 98.23712 98.23712 88.33128 88.33128 88.33128 Longitude 101.31985 102.37122 102.72170 100.86677 100.86677 102.30627 102.53425 102.56322 102.65560 102.68647 102.68647 102.58963 102.58999

(°N) Latitude 71.18120 71.15095 71.18231 64.11388 64.11388 58.76710 58.76710 64.03910 58.81039 64.16063 64.16932 64.16932 64.16932 64.16932 69.61982 69.61982 69.61982 58.87805 58.82252 58.79238 58.73786 60.35750 M, K; 3 , 3F §§ , , 3L, 3 3H , 3D – Figure Fig. S2v Fig. ; Figs. S2w, S2x S2w, Figs. ; Figs. 3E Figs. S2p–S2u Figs. Figs. 3G 3N; Figs. S2a–S2o Figs. 3N; Figs. 3A Fig. 3I †† T N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. Eruption temperature temperature sequence, is sequence, Possibly high Possibly Possibly high Possibly Sample K08-7.11, K08-7.11, Sample higher in the same pyroclastic, 270 °C pyroclastic, km 35 m ∼ outcrop. + m cm in diameter. clasts. A10-13. pebbles. reworked. Lapilli tuff. Reddish lapilli tuff. Reddish Sample description Sample 35-m-height outcrop. 1997, their section 3). 1997, (1997, their section 3). (1997, Organic chunk from a tuff. chunk from Organic Tuff from Mokulay stream outcrop. stream Mokulay from Tuff outcrop. stream Mokulay from Tuff Tuff from Mokulay stream outcrop. stream Mokulay from Tuff unit in the Arydzhangsky Formation. unit in the Lapilli tuff with abundant lithic clasts. with abundant Lapilli tuff Tuff breccia with abundant lithic clasts. with abundant breccia Tuff Lapilli tuff with primarily juvenile clasts. with primarily juvenile Lapilli tuff Accretionary lapilli 0.1–3 Accretionary km continuous river cliff on right side of Kata River, River, on right side of Kata cliff river km continuous Bedded accretionary lapilli–bearing lapilli tuff. km continuous river cliff of accretionary lapilli–bearing cliff river km continuous at base of 100 lapilli tuff Yellow-hued clasts both rounded and angular, lithic and juvenile. lithic and juvenile. and angular, clasts both rounded sparse accretionary lapilli. Apparent charcoal pieces. charcoal Apparent accretionary lapilli. sparse Lapilli tuff with predominantly juvenile clasts. Possibly Possibly clasts. juvenile with predominantly Lapilli tuff Volcaniclastic dike with abundant charcoal fragments. charcoal with abundant dike Volcaniclastic Tuff breccia with stringers of glass and angular juvenile of glass and angular juvenile with stringers breccia Tuff clast in bedded lapilli tuff; sampled at waterline of waterline sampled at Mafic clast in bedded lapilli tuff; lapilli tuff and lapilli tuff with abundant small lithic clasts. small lithic clasts. with abundant and lapilli tuff lapilli tuff from mouth on Angara River; organic and basaltic clasts; and basaltic clasts; organic River; Angara mouth on from 1–1.5 above Tunguska contact (cf. Fedorenko and Czamanske, and Czamanske, Fedorenko (cf. contact Tunguska above Tuff breccia with large stromatolitic blocks and small river and small river blocks stromatolitic with large breccia Tuff Lapilli tuff with juvenile clasts from the top of the first lava lava of the first the top clasts from with juvenile Lapilli tuff Located on right side of Kata River, further than site west River, on right side of Kata Located further west than site A10-13. Faintly bedded lapilli tuff with bedded lapilli tuff Faintly A10-13. further than site west 8–10 River, bank of Kata on left of bedded lapilli tuff Cliff 1–1.5 Lapilli tuff with juvenile clasts in Fedorenko and Czamanske and Czamanske clasts in Fedorenko with juvenile Lapilli tuff Lapilli tuff with juvenile clasts from crack in top of flow of flow in top crack clasts from with juvenile Lapilli tuff 5% 0% 4% 5% 0% 5% 0% N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. T3** 80% 40% 70% 70% 1% N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. T2** 15% 19% 15% 99% 60% 25% 20% 30% 30% 100% 1% 1% 0% 0% 0% 0% N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. T1** 95% 80% 70% 80% 80% # 1. 16 47.8 0.11 0.01 0.01 0.39 0.44 0.08 0.04 0.35 0.35 0.30 0.02 0.38 0.57 0.27 0.62 0.36 0.21 0.02 0.06 0.06 VOLCANICLASTIC SAMPLES FROM THE SIBERIAN TRAPS, WITH THEIR CARBON CONTENT AND COAL INTERACTION AND COAL CONTENT THEIR CARBON WITH TRAPS, THE SIBERIAN FROM SAMPLES VOLCANICLASTIC

Total organic organic Total carbon (wt%) § TABLE 1. TABLE 1.52 1.38 1.52 1.08 1.84 N.D. N.D. N.D. N.D. 0.18 0.32 0.83 0.35 0.55 2.70 0.51 0.71 0.07 0.48 0.21 0.75 0.37 (pre-lava volcaniclastics, possibly reworked in explosion pipe, southernmost pipe, region) in explosion reworked possibly volcaniclastics, (pre-lava content (wt%) content (pre-lava volcaniclastics, southern region) volcaniclastics, (pre-lava Bulk-rock carbon Bulk-rock ) estimated from paleomagnetism (Black et al., 2015). paleomagnetism (Black from ) estimated T † C# in text, and Figure S2 in the Supplemental Material (see text footnote 1). footnote (see text Material S2 in the Supplemental and Figure in text, (volcaniclastics early in the lava sequence, northern sequence, in the lava region) early (volcaniclastics C-597434 C-597435 C-597436 C-597437 C-597438 C-597439 C-580294 C-580299 C-580293 C-580292 C-580287 C-580286 C-580291 C-580298 C-580290 C-580297 C-580284 C-580285 C-580283 C-580282 C-580289 C-596833 (earliest volcaniclastics, below lavas, southernmost region) lavas, below (earliest volcaniclastics, (volcaniclastics early in the lava sequence, northern sequence, in the lava region) early (volcaniclastics Figure 3 Eruption temperature ( Eruption temperature See Analysis number in the lab of Stephen Grasby, Geological Survey of Canada (Calgary, Alberta, Canada). (Calgary, of Canada Geological Survey Grasby, in the lab of Stephen number Analysis using IR. Canada) Ontario, (Ancaster, ActLabs determined by carbon content Bulk rock 1]). footnote [see text Material (see the Supplemental Rock-Eval6 carbon determined by organic Total Note: N.D.—no detection, or in the case of eruption temperature, not determined. or in the case of eruption temperature, detection, N.D.—no Note: USA). Arizona, (Tempe, University Arizona State at Elkins-Tanton L.T. in main collection, held by number *Sample † § # assemblage. all macerals to group of each maceral percentage as a proportion of the total expressed highest (T3); (T1) to thermal lowest alteration from type, as described maceral in text, **Coal †† §§ Sample* Tunguska River Nizhnyaya NT12-4.2 NT12-4.3 NT12-6.2 NT12-7 NT12-15.1 NT12-15.2 NT12-15.3 NT12-15.4 Region Norilsk N12-8.1 N12-8.2 N12-8.3 River Angara A10-7.1 A10-8.2 A10-9.1 A10-11.1 A10-13.1 A10-14.1 A10-14.2 Tunguska River Podkamennaya R06-12A River Kotuy K08-5.12 K08-7.1 K08-7.4

988 www.gsapubs.org | Volume 48 | Number 10 | GEOLOGY | Geological Society of America

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/48/10/986/5146678/986.pdf by guest on 02 October 2021 A B C Light-Carbon Release from Magma-Coal Figure 2. (A,B) Coaly and Interactions other organic pieces in The observations presented here are inter- volcaniclastics at sam- preted as evidence that coal and organic-matter pling site NT12, Tura area combustion, along with forest fires, occurred in (see Fig. 1 for location). response to volcanism. Moreover, we infer that D (C) Liquefied coal injected into a displaced boulder these interactions were widespread, based on the from the Ivakinsky unit presence of thermally altered and/or burnt coal at the Kaerkan open-pit and organics in volcaniclastic rocks spanning mine near the town of the southern and central Siberian Traps prov- Talnakh, in the Norilsk region. (D) Small bitumi- ince (Fig. 1). nous inclusions, soft to The onset of the Permo-Triassic mass extinc- the fingernail, in coarse tion is marked by a major carbon isotope excur- dolerite in the mine near sion, which coincides within geochronological Ust-Ilimsk (see Fig. 1 for uncertainty with the oldest dated rocks from the location). Photos by L.T. Elkins-Tanton. Norilsk section (Burgess and Bowring, 2015). Emplacement of the organic matter–bearing southern volcaniclastic rocks preceded the main lava sequence (Levitan and Zastoina, 1985),

consist of vitrinite, with a mean Ror of 0.56% have been interpreted as evidence for incorpora- permitting alignment with the carbon isotope (Fig. S1A). These bituminous coal fragments tion of hydrocarbons in this area (e.g., Ryabov excursion. Coincident with the isotope excur- likely reflect the level of thermal maturity prior et al., 2012). sion, Siberian Traps magmatism was character- to eruption and show devolatilization features Near Norilsk, in the town of Kaerkan, a large ized by emplacement of laterally extensive sill such as small vacuoles and desiccation cracks open-pit coal mine contains outcrops where the complexes that could have facilitated significant (Figs. 3A–3D). Ivakinsky, the earliest lava flow, is in contact interaction with coal and organic matter–bearing Type 2 macerals have bright high-tempera- with coal. The coal appears to have been liq- units (Burgess et al., 2017) and production of ture char rims surrounding a less-altered interior uefied and injected into cracks in the cooling the volcaniclastic rocks. 13 (Figs. 3G and 3H). These outer chars show Ror lava, leading to more-reducing conditions in Mantle carbon (δ C ≈ −5‰) (Javoy values as high as 4%, along with contraction the magma (e.g., Ryabov et al., 2014) (Fig. 2). et al., 1986) and carbon in marine limestones cracks and combustion rims, and likely reflect Melenevsky et al. (2008) reported that coal in the (δ13C ≈ −2‰ to +2‰) are too isotopically combusted wood. broader aureole has been converted to anthracite, heavy to have caused the end-Permian car- Type 3 macerals are cenospheres and char indicating heating to ∼200 °C and release of bon isotope excursion (e.g., Cui and Kump, particles embedded in the volcaniclastic matrix ∼260 kg HC/t organic matter (mass of hydrocar- 2015; Gales et al., 2020; Payne and Kump, (Figs. 3I–3N). Cenospheres are formed by bon [HC] per unit mass of rock in metric tons). 2007). Consequently, coal or organic matter explosive devolatilization of organic matter that The southern volcaniclastics were also capable (δ13C ≈ −25‰) (Cui et al., 2013), was heated rapidly to high temperatures (∼1300 of burning or coking coal: paleomagnetic data clathrates (δ13C ≈ −56‰) (Krull and Retallack, °C; Goodarzi et al., 2008). We recognized two demonstrating unidirectional remnant magne- 2000), or petroleum (δ13C ≈ −30‰ to −25‰) types: isotropic particles (Figs. 3I–3L) with tization among some Angara rocks imply that (Svensen et al., 2009) represent the most plau- plastic deformation and bright oxidation rims temperatures exceeded 600 °C during magma sible sources of light-carbon injection. Assum- indicative of rapid heating in the presence of air emplacement (Black et al., 2015). ing equilibrium with an end-Permian dissolved (∼30% of sampled cenospheres), and anisotropic A major question is whether these samples inorganic carbon (DIC) reservoir of 38,000 Gt particles (Figs. 3M and 3N) with a fine-grained record coal heated by magma, or incorpora- C with an initial DIC δ13C = 0, and a C isotope 13 13 optical texture typical of combustion byproducts tion of charcoal formed previously in wildfires mass balance in which Δ C = ΔM*δ Ccoal/ of coal (Goodarzi and Hower, 2008) (Fig. S2). (Grasby et al., 2015; Hudspith et al., 2014). (ΔM + DIC), where Δ13C denotes the magni- Maceral texture distinguishes these options. tude of the isotope excursion and ΔM denotes DISCUSSION We identify both isotropic high-reflectance C release from coal, we infer that each 1000

Origins of Carbon-Rich Material within organic matter (Ror >2% and as high as 11%), Mt C released from coal or organic matter with 13 Siberian Volcanic Rocks which may have resulted from forest fires, as δ Ccoal ≈ −25‰ would translate to an ∼−0.64‰ Of the eight volcaniclastic samples from the well as high-reflectance chars with anisotropic shift in ocean-atmosphere δ13C (Cui and Kump, most southerly regions (the Podkamennaya Tun- cenospheres characteristic of coal combustion 2015). Mass-balance calculations indicate that guska and Angara Rivers), seven contained coal (Figs. 3M and 3N). 6000–10,000 Gt C with δ13C ≈ −25‰ could and combusted organic-matter fragments. Abun- Cenospheres form in present-day coal- yield a global carbon isotope perturbation with dant charcoal was also found in an end-Permian burning power plants (Hudspith et al., 2014). the observed magnitude of −3‰ to −6‰ (Cui crater-lake deposit near Bratsk (Fristad et al., These particles are rarely reported in the pre- and Kump, 2015). 2017). Only three of the eight samples from the industrial sedimentary record, but have been The primary uncertainties for estimating Nizhnyaya Tunguska River, in central observed in sedimentary rocks at the Permian- the magnitude of light-carbon release are (1) where the southernmost lavas appear, contain Triassic boundary in the Sverdrup Basin of Arc- the total mass of coal and organic matter that coal and organic matter. Farther north, only one tic Canada and are interpreted as a signature of interacted with Siberian Traps magmas and (2) of our six analyzed samples from the Norilsk magmatic coal combustion (Grasby et al., 2011), the efficiency of carbon release to the atmo- and Kotuy regions contain coal and organic- consistent with models of Ogden and Sleep sphere during these interactions. The cumu- matter fragments. However, previous work has (2012). In contrast, char and inertinite particles lative thickness of Carboniferous to Permian identified graphite, bitumen, and carbonaceous observed in our samples may be products of for- coal layers in the Tunguska Basin has been esti- material within Norilsk lavas and sills, which est fires (e.g., Fristad et al., 2017) Figs. 3I( –3L). mated as ∼100 m (Retallack and Jahren, 2008),

Geological Society of America | GEOLOGY | Volume 48 | Number 10 | www.gsapubs.org 989

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/48/10/986/5146678/986.pdf by guest on 02 October 2021 i hosted in organic-rich shales and in the A B Tunguska Basin is much larger than that of coal proper (Svensen et al., 2009), suggesting that v cu 4 5 i ∼10 –10 Gt C represents a lower estimate of Figure 3. Photomicro- v carbon available, sufficient to drive the observed graphs of macerals taken

v n carbon isotope excursion based on the mass-

io under reflected white light v and oil immersion; 50× balance calculations discussed above. lterat

la objective unless speci- 50 µm 50 µm

rma fied. (A–C) Low thermal CONCLUSIONS he alteration (random vitrin- C D wt Our findings provide direct field evidence v lo ite reflectanceR or = 0.56%) 1– of coal fragments of that Siberian Traps magmas incorporated and ld pe combusted coal and organic-rich material. This Ty vitrinite (v), semi fusinite (sf), inertinite (i), cutin- combustion may have been linked to the for- sp sp v ite (cu), sporinite (sp), mation of breccia pipes in the region (Jerram sp sp and liptodetrinite (ld). All et al., 2016; Ogden and Sleep, 2012; Svensen vitrinite particles show et al., 2009). The presence of cenospheres and 50 µm 50 µm features of low thermal sf alteration such as minor char also provides evidence for ejection of com- E F v dehydration fracturing. busted coal ash into the atmosphere, support- v (D) Sample from C under ing previous suggestions of significant coal fly ultraviolet light. Orange- fluorescing macerals ash formation at this time in Siberia, carried on n

vc io (e.g., sp and ld) indicate global air currents and deposited in the Arctic vc erat low thermal maturity coal

lt Canada Sverdrup Basin (Grasby et al., 2011). la fragments. (E–F) Vitrin- In addition to carbon released from the mantle rma ite particles (Ror = 0.83%) he (Sobolev et al., 2011) and thermally metamor- 50 µm 50 µm ht with desiccation cracks ig and vacuoles suggesting phosed country rocks, our results show that coal dh

G H an higher temperatures and combustion also liberated light carbon, contrib-

um devolatilization. (G–H)

di uting to the global warming and carbon-cycle

me Highly thermally altered disruption that characterized the Permo-Triassic 2– particles from plants pe (ch—char). Bright rim mass extinction (e.g., Cui and Kump, 2015). ch Ty and internal structures ch are evidence of high- ACKNOWLEDGMENTS temperature combustion. We thank reviewers Paul Wignall, Henrik Svensen, 50 µm 50 µm (I–J) High-temperature and Ying Cui, and our team members. Funding was provided by U.S. National Science Foundation (NSF) I J char (ch) particles with v fine devolatilization vacu- Continental Dynamics grant EAR-0807585 to Elkins- v oles and granular texture. Tanton, a grant of the Russian Foundation for Basic

Mean Ror = ∼4%. (K–L) Iso- Research (18-35-20058) to Veselovskiy, and NSF Inte- grated Earth Systems grant EAR-1615147 to Black. ch tropic cenospheric (c) and char (ch) particles with ch devolatilization vacuoles REFERENCES CITED and fine granular texture Black, B.A., Weiss, B.P., Elkins-Tanton, L.T., Vesel- and high reflectance (6%– ovskiy, R.V., and Latyshev, A.V., 2015, Sibe- 50 µm 50 µm 10%). (M–N) Cenospheres rian Traps volcaniclastic rocks and the role of K L with anisotropy due to ele- magma-water interactions: Geological Society of rs vated temperatures. Right America Bulletin, v. 127, p. 1437–1452, https:// ha

dc images in M and N were doi​.org/10.1130/B31108.1. c an taken with cross polarized Black, B.A., Neely, R.R., Lamarque, J.-F., Elkins rs

c he light (XPL); all others were Tanton, L.T., Kiehl, J.T., Shields, C.A., Mills, sp

no taken in plane polarized M.J., and Bardeen, C., 2018, Systemic swings in

ce light (PPL). Images A–D, end-Permian climate from Siberian Traps carbon 3– M, and N are from sample and sulfur degassing: Nature Geoscience, v. 11, pe

25 µm 25 µm Ty NT12-6.2; E and F are from p. 949–954, https://doi.org/10.1038/s41561-018-​ PPL XPL PPL XPL sample NT12-4.3; G and H 0261-y. M N are from sample R06-12A; Burgess, S.D., and Bowring, S.A., 2015, High-preci- I is from sample A10-7.1; J sion geochronology confirms voluminous magma- ch is from sample A10-14.2; K tism before, during, and after Earth’s most severe c is from sample R06-12A; L extinction: Science Advances, v. 1, e1500470, c is from sample NT12-6.2. https://doi​.org/10.1126/sciadv.1500470. Burgess, S.D., Muirhead, J.D., and Bowring, S.A., 2017, Initial pulse of Siberian Traps sills as the trigger of the end-Permian mass extinction: 50 µm 25 µm Nature Communications, v. 8, 164, https://doi​ .org/10.1038/s41467-017-00083-9. Cui, Y., and Kump, L.R., 2015, Global warming ­comprising ∼104–105 Gt C. Thermodynamic magmatism, Permian coal measures were indeed and the end-Permian : Proxy and experimental data (Iacono-Marziano et al., located near the surface (Retallack and Jahren, and modeling perspectives: Earth-Science Reviews, v. 149, p. 5–22, https://doi.org/10.1016/​ ​ 2012) suggest that coal combustion and/or cok- 2008). We interpret some of the combusted par- j.earscirev.2014.04.007. ing takes place only at pressures of several hun- ticles as originating from organic matter rather Cui, Y., Kump, L.R., and Ridgwell, A., 2013, Ini- dred bars or less, and at the onset of Siberian than mature coal, and the mass of organic matter tial assessment of the carbon emission rate and

990 www.gsapubs.org | Volume 48 | Number 10 | GEOLOGY | Geological Society of America

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/48/10/986/5146678/986.pdf by guest on 02 October 2021 ­climatic consequences during the end-Permian v. 357, p. 308–318, https://doi​.org/10.1016/​ v. 256, p. 264–277, https://doi​.org/10.1016/​ mass extinction: Palaeogeography, Palaeoclima- j.epsl.2012.09.051. j.epsl.2007.01.034. tology, Palaeoecology, v. 389, p. 128–136, https:// Il’yukhina, N.P., and Verbitskaya, N.G., 1976, Forma- Permyakov, S.A., et al., 2012, State Geological Map of doi​.org/10.1016/​j.palaeo.2013.09.001. tion conditions and stratigraphy of Carboniferous the Russian Federation, Scale 1:1000000, (third Erwin, D.H., 2006, Extinction: How Life Nearly Died coal deposits of the Siberian platform: Interna- generation): Angaro-Yeniseyskaya Series (O-48 250 Million Years Ago: Princeton, New Jersey, tional Geology Review, v. 19, p. 429–440, https:// - Ust’-Ilimsk), Explanatory Notes: Saint Peters- Princeton University Press, 296 p. doi​.org/10.1080/00206817709471037. burg, VSEGEI, 433 p. Fedorenko, V.A., and Czamanske, G.K., 1997, Results Javoy, M., Pineau, F., and Delorme, H., 1986, Peters, K.E., and Cassa, M.R., 1994, Applied source of new field and geochemical studies of the volca- Carbon and nitrogen isotopes in the mantle: rock geochemistry, in Magoon, L.B., and Dow, nic and intrusive rocks of the Maymecha-Kotuy Chemical Geology, v. 57, p. 41–62, https://doi​ W.G., eds., The Petroleum System—From Source area, Siberian flood- province, Russia: .org/10.1016/0009-2541(86)90093-8. to Trap: American Association of Petroleum International Geology Review, v. 39, p. 479–531, Jerram, D.A., Svensen, H.H., Planke, S., Polozov, Geologists Memoir 60, p. 93–120, https://doi​ https://doi​.org/10.1080/00206819709465286. A.G., and Torsvik, T.H., 2016, The onset of flood .org/10.1306/M60585C5. Fristad, K.E., Svensen, H.H., Polozov, A.G., and volcanism in the north-western part of the Sibe- Retallack, G.J., and Jahren, A.H., 2008, Methane Planke, S., 2017, Formation and evolution of rian Traps: Explosive volcanism versus effusive release from of coal dur- the end-Permian Oktyabrsk volcanic crater in lava flows: Palaeogeography, Palaeoclimatol- ing Late Permian extinction events: The Jour- the Tunguska Basin, Eastern Siberia: Palaeo- ogy, Palaeoecology, v. 441, p. 38–50, https://doi​ nal of Geology, v. 116, p. 1–20, https://doi​ geography, Palaeoclimatology, Palaeoecol- .org/10.1016/​j.palaeo.2015.04.022. .org/10.1086/524120. ogy, v. 468, p. 76–87, https://doi​.org/10.1016/​ Krull, E.S., and Retallack, G.J., 2000, δ13C depth Ryabov, V.V., Ponomarchuk, V.A., Titov, A.T., and j.palaeo.2016.11.025. profiles from paleosols across the Permian-Tri- Semenova, D.V., 2012, Micro- and nanostruc- Gales, E., Black, B., and Elkins-Tanton, L.T., 2020, assic boundary: Evidence for methane release: tures of carbon in Pt-low-sulfide ores of the Carbonatites as a record of the carbon iso- Geological Society of America Bulletin, v. 112, Talnakh deposit (Siberian platform): Doklady tope composition of large igneous province p. 1459–1472, https://doi​.org/10.1130/0016- Earth Sciences, v. 446, p. 1193–1196, https:// outgassing: Earth and Planetary Science Let- 7606(2000)112<1459:CDPFPA>2.0.CO;2. doi​.org/10.1134/S1028334X12100121. ters, v. 535, 116076, https://doi​.org/10.1016/​ Levitan, M.M., and Zastoina, A.N., 1985, Federal Ryabov, V.V., Shevko, A.Y., and Gora, M.P., 2014, j.epsl.2020.116076. Geological Map of USSR: Leningrad, VSEGEI, Trap Magmatism and Ore Formations in the Sibe- Goodarzi, F., and Hower, J.C., 2008, Classifica- scale 1:100,000. rian Noril’sk Region, Volume 1: Trap Petrology: tion of carbon in Canadian fly ashes and their Malich, N.S., et al., 1974, Map of geological forma- Dordrecht, Springer, Modern Approaches in Solid implications in the capture of mercury: Fuel, tions of the Siberian platform cover: Leningrad, Earth Sciences, v. 3, 390 p. v. 87, p. 1949–1957, https://doi​.org/10.1016/​ VSEGEI, scale 1:1,500,000. Sobolev, S.V., Sobolev, A.V., Kuzmin, D.V., Krivoluts- j.fuel.2007.11.018. Melenevsky, V.N., Fomin, A.N., Konyshev, A.S., and kaya, N.A., Petrunin, A.G., Arndt, N.T., Radko, Goodarzi, F., Huggins, F.E., and Sanei, H., 2008, Talibova, O.G., 2008, Contact coal transforma- V.A., and Vasiliev, Y.R., 2011, Linking mantle Assessment of elements, speciation of As, Cr, Ni tion under the influence of dolerite dike (Kaier- plumes, large igneous provinces and environ- and emitted Hg for a Canadian power plant burn- kan deposit, Noril’sk district): Russian Geology mental catastrophes: Nature, v. 477, p. 312–316, ing bituminous coal: International Journal of Coal and Geophysics, v. 49, p. 667–672, https://doi​ https://doi​.org/10.1038/nature10385. Geology, v. 74, p. 1–12, https://doi​.org/10.1016/​ .org/10.1016/​j.rgg.2008.01.007. Svensen, H., Planke, S., Polozov, A.G., Schmidbauer, j.coal.2007.09.002. Naumov, V.A., and Ankudimova, L.A., 1995, Paly- N., Corfu, F., Podladchikov, Y.Y., and Jamtveit, Grasby, S.E., Sanei, H., and Beauchamp, B., 2011, Cata- nocomplexes and age of volcanogenic deposits B., 2009, Siberian gas venting and the end-Perm- strophic dispersion of coal fly ash into oceans during of the Angara-Katanga area (Middle Angara ian environmental crisis: Earth and Planetary the latest Permian extinction: Nature Geoscience, Region): Geologiya Geofizica, v. 36, p. 39–45. Science Letters, v. 277, p. 490–500, https://doi​ v. 4, p. 104–107, https://doi.org/10.1038/ngeo1069​ . Ogden, D.E., and Sleep, N.H., 2012, Explosive erup- .org/10.1016/​j.epsl.2008.11.015. Grasby, S.E., Sanei, H., and Beauchamp, B., 2015, tion of coal and basalt and the end-Permian mass Svensen, H.H., Frolov, S., Akhmanov, G.G., Polo- Latest Permian chars may derive from wildfires, extinction: Proceedings of the National Academy zov, A.G., Jerram, D.A., Shiganova, O.V., Mel- not coal combustion: Comment: Geology, v. 43, of Sciences of the United States of America, v. 109, nikov, N.V., Iyer, K., and Planke, S., 2018, Sills p. e358, https://doi​.org/10.1130/G36539C.1. p. 59–62, https://doi.org/10.1073/pnas.1118675109​ . and gas generation in the Siberian Traps: Philo- Hudspith, V.A., Rimmer, S.M., and Belcher, C.M., Payne, J.L., and Clapham, M.E., 2012, End-Perm- sophical Transactions of the Royal Society A: 2014, Latest Permian chars may derive from ian mass extinction in the oceans: An ancient Mathematical, Physical, and Engineering Sci- wildfires, not coal combustion: Geology, v. 42, analog for the twenty-first century?: Annual ences, v. 376, 20170080, https://doi.org/10.1098/​ p. 879–882, https://doi​.org/10.1130/G35920.1. Review of Earth and Planetary Sciences, v. 40, rsta.2017.0080. Iacono-Marziano, G., Marecal, V., Pirre, M., Gaillard, p. 89–111, https://doi​.org/10.1146/annurev- Wignall, P.B., 2001, Large igneous provinces and F., Arteta, J., Scaillet, B., and Arndt, N.T., 2012, earth-042711-105329. mass extinctions: Earth-Science Reviews, Gas emissions due to magma-sediment interac- Payne, J.L., and Kump, L.R., 2007, Evidence for v. 53, p. 1–33, https://doi​.org/10.1016/S0012- tions during flood magmatism at the Siberian recurrent massive volcanism from 8252(00)00037-4. Traps: Gas dispersion and environmental con- quantitative interpretation of carbon isotope fluc- sequences: Earth and Planetary Science Letters, tuations: Earth and Planetary Science ­Letters, Printed in USA

Geological Society of America | GEOLOGY | Volume 48 | Number 10 | www.gsapubs.org 991

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/48/10/986/5146678/986.pdf by guest on 02 October 2021