Wildfire evidence from the Middle and Late Permian Hanxing Coalfield, North China Basin

DOI: 10.1344/GeologicaActa2020.18.12  L. Xiao, Q. Zhao, J. Wang, V. Mishra, S.I. Arbuzov, M. Zhang, 2020 CC BY-SA L . X i a o e t a l . G e o l o g i c a A c t a , 1 8 . 1 2 , 1 1 1 ( 2 0 2 0 ) D O I : 1 0 . 1 3 4 4 / G e o l o g i c a A c t a 2 0 2 0 . 1 8 . 1 2 Permian wildfires in North China 2 b). Most of the studies of palaeo-wildfires are related to the wildfires occurred in the Late Paleozoic era (Arzadún et al., 2017; Jasper et al., 2013; Kauffmann et al., 2016; Manfroi et al., 2015; Shen et al., 2011; Singh and Shukla, 2004; Sun et al., 2017). These palaeo-wildfires consumed O2 and released CO, CO2 to the atmosphere which modified the atmospheric temperatures and ecosystems (Berner, 2006; Lenton, 2013; Scott and Glasspool, 2006; Scott et al., 2016). Sun et al. (2017) have reported also that the palaeowildfires of the Middle Permian could have discharged massive quantities of pollutants and increased the ambient temperature, causing the floral extinction that occurred at the Middle Permian. Shen et al. (2011) have reported also that the wildfires of the Late Permian may have acted as a catalyst to the Permo-Triassic Boundary (PTB) mass extinction events. Although the palaeo-wildfires of the Late Permian in southern China have been studied by Shen et al. (2011), northern China is devoid of this kind of study as coal seams of this age are not present there and wildfire evidence are difficult to find in sedimentary rocks. The main purpose of this study is to examine the available sedimentary rocks from the northern China basin to search for pieces of evidence of the palaeo-wildfires in the Late Permian is. Findings of this investigation may be of relevance for the comprehension of the causes of the Permo-Triassic mass extinction. With this aim, 14 rock samples were collected from a drill core of Hanxing Coalfield and they were investigated through petrography (macro and micro), SEM, GC and GC-MS to characterize their organic structures and compounds.


INTRODUCTION
. Most of the studies of palaeo-wildfires are related to the wildfires occurred in the Late Paleozoic era (Arzadún et al., 2017;Jasper et al., 2013;Kauffmann et al., 2016;Manfroi et al., 2015;Shen et al., 2011;Singh and Shukla, 2004;Sun et al., 2017). These palaeo-wildfires consumed O 2 and released CO, CO 2 to the atmosphere which modified the atmospheric temperatures and ecosystems (Berner, 2006;Lenton, 2013;Scott and Glasspool, 2006;Scott et al., 2016). Sun et al. (2017) have reported also that the palaeowildfires of the Middle Permian could have discharged massive quantities of pollutants and increased the ambient temperature, causing the floral extinction that occurred at the Middle Permian. Shen et al. (2011) have reported also that the wildfires of the Late Permian may have acted as a catalyst to the Permo-Triassic Boundary (PTB) mass extinction events.
Although the palaeo-wildfires of the Late Permian in southern China have been studied by Shen et al. (2011), northern China is devoid of this kind of study as coal seams of this age are not present there and wildfire evidence are difficult to find in sedimentary rocks. The main purpose of this study is to examine the available sedimentary rocks from the northern China basin to search for pieces of evidence of the palaeo-wildfires in the Late Permian is. Findings of this investigation may be of relevance for the comprehension of the causes of the Permo-Triassic mass extinction. With this aim, 14 rock samples were collected from a drill core of Hanxing Coalfield and they were investigated through petrography (macro and micro), SEM, GC and GC-MS to characterize their organic structures and compounds.

GEOLOGICAL SETTING
Hanxing Coalfield was formed and developed in the center of the North China basin (Fig. 1), which belongs to the northern part of the Sino-Korean Craton (Zhao et al., 2006). The Late Paleozoic stratigraphic records geological setting of the North China basin has been reviewed by Stevens et al. (2011), Sun et al. (2002 and Wang and Pfefferkorn (2013) Sun et al., 2017).   (Li et al., 2010;Stevens et al., 2011;Yang et al., 2017). The Taiyuan and Shanxi formations are coal-bearing units of Early Permian and earlier Middle Permian age respectively (Fig. 2). The Taiyuan Formation was deposited in a predominantly paralic environment and the sediments mainly consist of siltstone and coal seams. The Shanxi Formation was deposited in a fluvial-dominated deltaic depositional environment and the sediments mainly consist of mudstone, siltstone and coal seams (Sun et al., 2017;Wang et al., 2019).
The Upper Permian consists of the Lower and Upper Shihhotse formations, and the overlying Shiqianfeng Formation. These formations were dominated by alluvial sandstones, siltstones and clays, and a certain amount of coal with abundant plant fossils. The Lower Shihhotse Formation are dominated by grey sandstones, brown-green homogeneous siltstones and clays. The Upper Shihhotse Formation is of similar composition but with a higher proportion of yellow laminated clays. The Shiqianfeng Formation lies unconformably above the Upper Shihhotse Formation (Hu et al., 1990;Wang, 2010), and is characterized by red to brown claystones interpreted as evidence of arid conditions (Stevens et al., 2011).

SAMPLES AND METHODS
All the analyses of the present study were carried out in the Key Laboratory of Resource

Petrographic analysis
For reflected light microscopy or organic petrography analysis, the samples were crushed to a size less than 20mesh. These fractions were embedded in epoxy resins, and the epoxy pellets were grounded and polished to obtain a smooth surface (Petersen and Lindström, 2012). Charcoal particles were investigated under a Leica DM2500P polarised light microscope, equipped with a Craic QDI 302 TM spectrophotometer. Vitrinite reflectance (R o ) was also measured under Leica DM2500P reflected light microscope that was fitted with a halogen lamp. The measurement was calibrated using a Leitz glass standard (R oil =0.89%).

FE-SEM analysis
A Field-Emission-Scanning Electron Microscope (HITACHI UHR FE-SEM, SU8220), in conjunction with an EDAX energy-dispersive X-ray spectrometer (Genesis Apex 4), was used to study the morphology of the minerals and determine the distribution of some elements. The coal samples were coated with conductive carbon materials and were analyzed at a working distance of 10mm, a beam voltage of 20.0kV, an aperture of 6, and a spot size of grade 5.0. The images were captured via a retractable solid-state backscatter electron detector. The experimental methods were performed following Sun et al. (2017).

Solvent extraction and liquid chromatography
For organic geochemical analyses, samples were solventextracted for 48 h using chloroform as solvent. Extract yields were determined gravimetrically after removal of the solvent. The extracts were separated into three fractions (saturated hydrocarbons, aromatic hydrocarbons and NSOcompounds) by column chromatography over pre-washed silica gel (70-230mesh, 50-1cm). The alkanes were eluted with n-hexane, the aromatic hydrocarbons and aromatic sulfur compounds with dichloromethane, and the polar compounds (hetero-compounds) with methanol (40ml for each fraction).

Gas Chromatography-Mass Spectrometry
GC and GC-MS analyses of the aromatic fractions were performed on a Hewlett-Packard model 6890 GC coupled to a Hewlett-Packard model 5973 quadrupole Mass Spectrometric Detection (MSD). Squalane has been added to the aromatic hydrocarbon fractions as internal standards prior to analysis. GC separation was achieved on a fused silica capillary column coated with DB5 (30m×0.25mm i.d., 0.25μm film thickness). The GC operating conditions were performed through three steps: temperature held at 60º for 5min, increased from 60 to 300º at a rate of 4ºC min -1 with final isothermal hold at 300ºC for 15min. Helium was used as a carrier gas. The sample was injected at a split ratio of 30:1 with the injector temperature at 290ºC. The mass spectrometer was operated in an electron impact mode at 70eV and scanned from 50 to 650Da. Data were acquired and processed with Chemstation software. Individual compounds were identified through comparisons of mass spectra with available pieces of literature and library data, along with interpretations of mass spectrometric fragmentation patterns.

Charcoal particles
Many paleoecologists, palaeobotanists and palaeoclimatologists defined most wildfire products as charcoal (including inertinite and natural char (sensu International Committee for Coal and Organic Petrology (ICCP), see Kwiecińska and Petersen, 2004), which is referred as inertinite in the petrography study of coals (Robson et al., 2015;Scott, 2010;Sýkorová et al., 2005). In this study, the term inertinite is used following the definition of ICCP (2001), and the term charcoal following the definition of Scott (2010).
Abundant charcoal particles were observed on the drill core samples (Fig. 3). Through naked eyes these particles were dark in color, irregular shaped and distributed throughout the sediments. The particle sizes vary from several millimeters to several centimeters. Under the microscope, these charcoal particles show a prominent cellular structure with very high reflectivity (Fig. 4) and belong to the distinctive fusinite maceral (sensu ICCP). Kwiecińska and Petersen (2004) found that some particles have higher reflectance than the associated inertinite macerals which are referred as natural char on the basis of the distribution of pores (Fig. 4).
About their genetic mechanisms, Kwiecińska and Petersen (2004) believed that "natural char is formed by the influence of heat from the fire on (1) coal or (2) gelified organic matter in peat". Glasspool and Scott (2013) stated that inertinite (fusinite) is synonymous of charcoal and is considered to be formed almost exclusively as a product of wildfires. Therefore, charcoal has been considered as the strongest evidence of palaeo-wildfires (Jasper et al., 2013;Kauffmann et al., 2016;Scott and Glasspool, 2006;Shen et al., 2011;Sun et al., 2017;Yan et al., 2019). The presence of abundant charcoal particles in this study evidence that the palaeo-wildfires occurred in North China in the Late Permian.
Charcoal structure and their reflectivity were studied here because the wildfire type and its burning temperature can be deduced from the charcoal shape and reflectivity (Scott, 2010). Under the electron microscope, wood cell walls displayed both lamellar and homogenous aspect (Fig. 5) that confirmed the material as charcoal (Degani-Schmidt et al., 2015). According to Scott (2010), the cell walls of charcoal would homogenize at >300-325ºC and fragmented at >600ºC. In this study, some cell walls of charcoal observed under SEM show a homogeneous aspect indicating that the combustion temperature of wildfire was >300-325ºC. Crashed cell walls were also observed indicating that these fragmented charcoal particles could have formed at a temperature above 600ºC (Fig. 5).
The reflectance of charcoal is also related to the burning temperature showing a positive relationship between them (Ascough et al., 2010;Guo and Bustin, 1998;Scott, 2010;Scott and Glasspool, 2005). According to Petersen and Lindström (2012), the burning temperature of wildfire can be roughly calculated by the following equation: T= 184.10+117.76×%Ro (r 2 = 0.91) where T is the burning temperature and %R o is the measured inertinite reflectance.
The reflectance of the charcoal particles from the Hanxing coal here analyzed varies from 3.12% to 4.84% and, according to the above equation, it formed in a temperature range from 552ºC to 754ºC.
Thus, here we get two sets of temperatures derived from the samples' morphology (>300-325ºC) and reflectance (552-754ºC). Wildfires can be characterized by different burning temperatures and according to that, they can be grouped into ground fires, surface fires, and crown fires. The temperature of a ground fire is generally lower than 350ºC, while the temperature of a surface fire is generally between 300ºC and 700ºC. The temperature of a crown fire is higher and can reach 1000ºC (Petersen and Lindström, 2012;Scott and Jones, 1994). Therefore, it can be speculated that wildfires occurred in the Late Permian belong to surface and crown fires.

Polycyclic Aromatic Hydrocarbons (PAHs)
A total number of 16 PAHs with 4-6 rings were detected in most of the samples ( Figure 6; perylene (BghiP). The typical distribution of PAHs in the samples is presented in Figure 6.
There are two main sources of PAHs in sedimentary strata i.e. incomplete combustion of vegetation or fossil fuels and degradation of organic matter by microorganisms during diagenesis (Hossain et al., 2013;Meng et al., 2019). PAHs derived from combustion mainly produce >4 ring PAHs, while diagenetic PAHs often have 2-4 rings (Denis et al., 2012;Sun et al., 2017). The PAHs Sample S3 Sample S3 Sample S9 Sample S12 Sample S13 Sample S13 originally formed by combustion may be alkylated during diagenesis, and those susceptible PAHs may disappear during this process (Denis et al., 2012;Hossain et al., 2013;Jiang et al., 1998). The higher the number of aromatic rings, the stronger the antioxidant ability of PAHs. Therefore, PAHs with 5-6 rings are less likely to be affected during diagenesis (Denis et al., 2012). Studies show that BePy is the most stable PAH among the 5-ring PAHs (Jiang et al., 1998;Sullivan et al., 1989). BghiP (5-ring) generally originated through combustion and had been reported from high intensity paleo vegetation fires (Denis et al., 2012;Hossain et al., 2013;Jiang et al., 1998). Therefore, identified PAHs with 5-6 rings in this study should be formed by pyrogenic activity. Because there is no volcanic activity in the area, the occurrence of pyrogenic PAHs indicates that a wide range of ancient wildfires occurred in North China basin during the Middle and Late Permian, and the abundance of >4 ring PAHs in the top three samples (S1-S3) indicate that wildfires could have occurred in the P-T boundary. Samples S4 and S5 are devoid of >4 ring PAHs indicating absence of wildfire or absence of vegetation at that time.
Compounds produced at different combustion temperatures are different. The higher the combustion temperature is, the likelier are PAHs with more than 5 rings to be produced (Finkelstein et al., 2005;Hossain et al., 2013). Therefore, PAHs with low molecular weight (Fla, Chry, etc.) mostly indicate low temperature fire like in samples S10, S11, S12 and S14, while PAHs with high molecular weight (BePy, BaPy, InPy and Bpery) generally indicate high temperature fire like in Samples S2, S3, S6-S9 and S13 (Denis et al., 2012;Hossain et al., 2013).

Possible influences of the wildfire
It is generally accepted that the release of pollutants such as sulphur dioxide and toxic compounds such as the polycyclic aromatic hydrocarbons by wildfire could have influenced paleoclimate and palaeoecosystem (Scott and Jones, 1994;Van de Schootbrugge et al., 2009;Xu et al., 2020a, b). Shen et al. (2011) studied widespread wildfires close to PTB in South China and believed that wildfire, along with fire-derived products, could have been one of the reasons for the PTB mass extinction. The charcoal particles of this study indicate wildfires occurred not
Permian wildfires in North China 9 only in South China but also in North China basin during the Late Permian, with the other global wildfire events (Arzadún et al., 2017;Grasby et al., 2011;Kauffmann et al., 2016). These wildfires could have consumed enormous amounts of O 2 from the atmosphere and also discharge massive quantities of pollutants causing the increase of the environmental temperature.

CONCLUSIONS
This is the first report of the occurrence of wildfire close to PTB in the North China Basin. Abundant charcoal particles were observed in the sedimentary rock samples from the Hanxing Coalfield, indicating that widely prevalent wildfires occurred in the Middle and Late Permian. The reflectance values of the charcoal reveal that the wildfires belong to surface and crown type fires. The presence of high molecular PAHs is another evidence of wildfires in the study area. The occurrence of 5 and 6 ring PAHs in the studied samples indicates that the wildfire events had high temperature and most likely were crown fires. These wildfires could have consumed O 2 and discharged pollutants to the atmosphere being one of the factors that influenced the paleoclimate and palaeoecology in the Late Permian. Compound name m/z Sample No. Molecular structure S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14