Investigation on the biodegradation levels of super heavy oils by parameter-striping method and refined Manco scale: a case study from the Chepaizi Uplift of Junggar Basin

The Carboniferous volcanic reservoir in the Chepaizi Uplift became an exploration hot target in recent years for its substantial amount of oils discovered. However, most of the Carboniferous heavy oils were biodegraded to PM7 or higher with orders of magnitude variation in oil viscosities. Two oil groups (I and II) exactly corresponding to the western and eastern Chepaizi Uplift were distinguished according to their source diagnose. Furthermore, three oil families (II1, II2 and II3), with the biodegradation level of PM7, PM8–8+, PM9+, respectively, were classified based on molecular compositions and parameter-stripping method of strongly bioresistant parameters. Allowing for this extremely high biodegradation case, more biodegradation refractory compound class were added to establish a refined Manco scale to quantitatively evaluate the biodegradation extent. Refined Manco number (RMN2) positively correlated with the oil density, NSO contents, and absolute concentrations of diasteranes and gammacerane, negatively correlated with the absolute concentrations of diahopane, summed tricyclic terpanes and pentacyclic terpanes. This refined scale showed higher resolution than the PM one to differentiate the biodegradation extent of Carboniferous heavy oils from the Chepaizi Uplift, especially those with same PM values but different oil viscosities.


Introduction
Heavy oil and oil sand bitumen dominate the world's oil inventory (Marcano et al. 2013), even exceed the normal oil in quantity (Wang et al. 2016;Li and Huang 2020). Biodegradation may be responsible for these unconventional resources, which primarily formed from the microbial degradation of the conventional oils over geological timescales . With the proceeding of biodegradation, the oil will exhibit decreases in mass and net volume, increases in oil density, oil viscosity, oil acidity and sulfur content, enrichments in nitrogen, sulfur and oxygen-containing organic compounds, and trace mental elements, which heavily affect the oil's chemical compositions and physical properties (Larter et al. 2006;Zhou et al. 2008;Forsythe et al. 2019). Even within a single petroleum reservoir, biodegradation may create orders of magnitude variation in oil viscosity and fluid property laterally and vertically . Allowing for their great impact on the economic value and predictability, knowledge of the petroleum biodegradation is critical in selecting exploitation strategies (Marcano et al. 2013;Yin et al. 2013), evaluating the altering extent (Bautista et al. 2015) and forecasting the oil compositional characteristics and viscosity Zhang et al. 2014a, b).
Experimental examinations and case studies indicate that relative sensitivity to microbial attack exist among different compound class, even among the individual compound Edited by Jie Hao and Chun-Yan Tang 1 3 in each class. Therefore, biodegradation is a quasi-stepwise process with consume of biomarkers in a preferential order (Peters et al. 2005). Based on the alteration/removal of saturated biomarkers, i.e., normal alkanes, acyclic isoprenoids, terpanes and steranes, and selected aromatic compounds (mainly aromatic steroids), Peters and Moldowan (1993) developed a classical scale to assess the extent of oil biodegradation, which is typically abbreviated as PM scale. Five terms, i.e., light (PM1-3), moderate (PM4-5), heavy (PM6-7), very heavy (PM8-9) and severe (PM10) are assigned to ranges of the PM scale. Another scheme was established by  fundamentally based on the alteration extent with a compound class and the presence or depletion of single key compound class, only with the terms of heavy and severe to describe the PM 4-10 range.
The normal alkanes are generally more susceptible to biodegradation than the branched alkanes and isoprenoids (Chen et al. 2017), thus ratios of i-C 5 /n-C 5 , pristane/n-C 17 and phytane/n-C 18 are widely used to assess the alteration content of light to moderate biodegraded oils (Peters and Moldowan 1993). However, some case studies showed no evident correlations between Ph/n-C 18 and oil API, and then the parameter "mean degradative loss" was proposed to assess the biodegradation extent by means of quantifying the depletion in volumetrically important individual oil constituents, especially for the light to moderate level of alteration (Elias et al. 2007). Additionally, the production and biodegradation of oxygen-containing compounds also follow a preferential order, thereafter the acyclic (DBE 1)/ cyclic  can reveal the biodegradation level, which well correlated with the PM scale when it less than PM < 6 (Angolini et al. 2015).
Many case investigations suggest that the sequential microbial degradation of oil constitutes does not occur in a true stepwise fashion and strictly follow the schemes previously developed (Bennett and Larter 2008;Wang et al. 2013;Chang et al. 2018). All these scales are unsuitable for heavy oil or super heavy oil, and for the mixtures of oils Zhang et al. 2014a, b). Selecting 8 compound classes (alkyl toluenes, C 0-1 naphthalenes, C 2 naphthalenes, C 3 naphthalenes, methyl dibenzothiophenes, C 4 naphthalenes, C 0-2 phenanthrenes and steranes) to reflect the increasing resistance to biodegradation, assigning 0-4 scores to describe the removal extent, Larter et al. (2012) proposed a Manco scale to assess the biodegradation level (PM4-8), with its MN2 value positively correlates with oil viscosity, and effectively used (López 2014).
Till now, published studies mainly focused on the geochemical behaviors of molecular biomarkers at the biodegradation level of PM8 or less, for those biodegradation level higher than PM8, rare studies were reported. With the reduction in conventional resources, extremely biodegraded oils continuously discovered in the petroleum exploration attracted much more attention. Using semi-quantitative parameter-stripping method and refined Manco scale, this paper aims to deeply understand and quantitatively determine the biodegradation extent of heavy oils in the Chepaizi Uplift, Junggar Basin, NW China.

Geological setting
The Chepaizi Uplift, covering an area of 10,500 km 2 , is a petroliferous target of the northwestern Junggar Basin (Zhao et al. 2019), which is surrounded by two hydrocarbon generative source kitchens, i.e., the Changji Sag to the east and Sikeshu sag to the south (Fig. 1). The early stage of late Hercynian tectonic movement witnessed the formation of prototype of the Chepaizi Uplift, which subsequently experienced the intense uplift during the Indosinian and Yanshan movements, slow subsidence during the Himalayan tectonic movement. As an inherited paleo-uplift, Chepaizi Uplift was characterized by the seriously lack of deposition succession in the structurally high parts, with the Cretaceous, Paleogene, Neogene and Quaternary sedimentary rocks overly Carboniferous volcanics (Song et al. 2007).
Due to the multi-staged transgression and uplifting, three sets of favorable reservoir rocks were developed in the Chepaizi Uplift, i.e., medium-fine sandstone and glutenite occurred in the Neogene Shawan Formation, fine sandstones and siltstone occurred in the Cretaceous interval, and basalt, andesite, tuff and volcanic breccia in the Carboniferous interval.
The Hongche Fault, an active fault separating the Chepaizi Uplift from the Changji Sag (Dong et al. 2017;Ni et al. 2019), cut through sandbodies developed in the Cretaceous, Neogene Shawan Formation and volcanics in the Carboniferous, providing a favorable vertical conduit for the migration of hydrocarbon generated in the Changji Sag to the Chepaizi Uplift and yielding three vertical oil-bearing intervals (Miao et al. 2015;Meng et al. 2016).

Sample preparation
Eleven DST oil samples from the Carboniferous volcanic interval were collected from wells in the Chepaizi Uplift for geochemical analysis. Asphaltenes were removed from the oil samples and source rocks using n-hexane precipitation, and the deasphaltened oil was divided into two aliquots. The first of these underwent column chromatography using a routine silica gel and alumina column from which aliphatic and aromatic fractions were obtained using n-hexane and dichloromethane (DCM).

Gas chromatography-mass spectrometry
Gas chromatography-mass spectrometry (GC-MS) analysis of the aliphatic and aromatic fractions was performed with a Finnigan Model SSQ-710 quadrupole analytical system coupled to a DB-5 fused silica column (30 m × 0.32 mm i.d.) and linked to an IAIS data processing system. GC temperature operating conditions for the aliphatic fraction were as follows: 100 °C (1 min) to 220 °C at 4 °C/min and, then to 300 °C (held 5 min) at 2 °C/min; for the aromatic fraction: 80 °C (1 min) to 300 °C (held 15 min) at 3 °C/min. MS conditions were as follows: electron impact (EI) ionization mode; 70-eV electron energy; 300-mA emission current; and 50-550 amu/s scan range.

Oil bulk compositions and oil families
The Carboniferous oils from the eastern Chepaizi Uplift are characterized by higher oil density (0.9285-0.9590 g/ cm 3 ) and viscosity (154-8968 mPa·s) than those from the western Chepaizi Uplift, which can be classified as heavy crude oils (Table 1). The Carboniferous oils are predominantly aliphatic, as indicated by their saturate/aromatic (ST/AR) ratio (1.92-3.56) and saturate fraction abundance (44.45%-64.38%). The oil density showed roughly positive correlation with the viscosity and the NSO (resin + asphaltene) fraction content, and negative correlation with the burial depth. Progressive biodegradation of crude oils may be responsible, as it decreases the content of saturate and aromatic hydrocarbons and enriches the resins and asphaltenes, resulting in an increase in oil density (López et al. 2015;. As for the study of oil origins, oil family is a widely used term to distinguish oils with different genetic affinity. An oil family, a group of oils that derived from a same source rock, possibly experienced similar reservoir-forming history, belonged to a same oil system, and possessed same or similar chemical compositions. Combing the ratios of C 27 diasterane 20S/20R (C 27 DS 20S/20R), C 27 diasterane/C 27 regular sterane (C 27 DS/C 27 RS), C 26 /C 28 triaromatic steroid (20S) (C 26 /C 28 TAS(20S)), and C 27 /C 28 TAS(20R) with the stable carbon isotope distribution, previous studies concluded that the Carboniferous oils in the eastern Chepaizi Uplift were mainly derived from the mudstone of the Middle Permian Wuerhe Formation (P 2 w) in the Changji Sag, whereas the ones in the western Chepaizi Uplift were essentially originated from the Jurassic mudstone in the Sikeshu Sag (Zhang et al. 2012;Xu et al. 2018;Mao et al. 2020). Although two oil charging episodes were defined in the Carboniferous reservoirs based on the oil geochemistry, fluid inclusions and basin modeling (Chang et al. 2019;Shi et al. 2020), the later charge actually was the remigration of early reservoired oils due to the tectonic adjustment (Cao et al. 2010;Song et al. 2016;Chang et al. 2019). Although the Carboniferous oils all exhibited the characteristics of lacustrine source facies, source-diagnostic and redox potential of depositional environment-related molecular biomarkers showed marked distinction between the eastern and western parts (Xu et al. 2018), implying at least two oil groups.
Cluster analysis, a statistical method, is an effective tool to classify studied samples into different groups by their similarity distances which were calculated from the different variables investigated. Allowing for the severe oil alteration, eight biomarker parameters strongly resistant to biodegradation were selected to act as the variables in the clustering analysis, i.e., C 26 /C 28 TAS (20S), C 27 /C 28 TAS (20R), C 24 tetracyclic terpane/C 30 hopane (C 24 Tet/C 30 H), gammacerane/ C 30 hopane (G/C 30 H), C 35 H (22S)/C 34 H (22S), C 29 18α(H)-30-norneohopane/C 29 Hopane (C 29 Ts/C 29 H), C 24 tricyclic terpane (TT)/C 23 TT, C 22 TT/C 21 TT. In the clustering tree graph (Fig. 2), the Carboniferous oils can be clearly divided into two groups, that is, Group I for the western Chepaizi Uplift, Group II for the eastern Chepaizi Uplift. In addition, within the Groups II, three oils families can be further subdivided (II 1 , II 2 and II 3 ) according to their similarity distances.

Oil Group I
Group I (wells P70 and 702), Carboniferous oils from the western Chepaizi Uplift, was least biodegraded among the investigated oils. These oils showed faintly "UCM" and relatively intact n-alkanes on the TIC chromatogram ( Fig. 3a). Hopanes were slightly higher than the tricyclic terpanes (TTs) in abundance with C 30 hopane as the peak compound and reversed "L" -shaped distribution of C 20 TT-C 21 TT-C 23 TT (Fig. 3b). Gammacerane (G) was low in content as evidenced by the quietly low G/C 30 H values (0.08-0.09). Fully developed 25-norhopanes were detected indicating heavy biodegradation. Pregnane, homopregnane, and diasteranes (DS) were lower than the regular steranes (RS) in abundance with "V"-shaped distribution of ααα20R C 27 -C 28 -C 29 steranes (Fig. 3c). Stable carbon isotope varied from −27.5 ‰ ~ −27.8 ‰ (Table 1). Naphthalenes and phenanthrenes were nearly intact and triaromatic steroids (TAS) were essentially unchanged (Fig. 3d). Thereby, Group I can be assigned to biodegradation level of PM6.

Sequential biodegradation revealed by biomarkers
Although many studies reported that the microbial degradation of oil did not appear to be strictly consistent with the stepwise fashion in established schemes, relative variations of biomarkers in oil families II 1 , II 2 and II 3 can be prominently distinguished (Fig. 7). The values of C 22 T/C 21 TT, C 24 TT/C 23 TT, C 24 C 24 Tet/C 26 TT and (C 20 + C 21 )/C 26 TT ratios increased with the biodegradation extent increased from PM7 to PM8 and finally to PM9+ (Fig. 7a-d), suggesting the alterations of tricyclic terpanes, preferential removal of lower molecular weight homologues and more bioresistance of tetracyclic terpane (Huang and Li 2017). Hopane/sterane ratio varied with biodegradation in a zigzag-shaped fashion (Fig. 7e), i.e., essentially unchanged from the level of PM6 + to PM7, sharply increased from the level of PM7 to PM8, and progressively decreased from the level of PM8 to PM9+, confirming the different susceptibility to biodegradation for hopanes and steranes at different biodegradation level (Reed 1977;). Ratios of C 29 /C 30 H, G/ 30 H and C 30 diahopane/C 30 hopane (C 30 D/C 30 H) kept relatively unchanged from level PM7 to PM8, then sharply increased to PM9 + ( Fig. 7f- the same bioresistance of C 30 hopane to C 29 hopane, gammacerane and C 30 diahopane before PM8 level, but more susceptibility at level PM9 + . Gradually increased 18α(H)trisnorneohopane/17α(H)-trisnorhopane (Ts/Tm) ratio with biodegradation indicated the relatively more susceptible of Tm than Ts to microbial alteration (Fig. 7i). Increasing C 29 25-norhopane/gammacerane (C 29 NH/G) value with biodegradation confirmed the formation of 25-norhopane (Fig. 7j) at level above PM6, which was consistent with the increasing C 28 NH/C 29 H and C 29 NH/C 30 H (Fig. 7k, 7l). However, its decrease from PM8 to PM9 + revealed the degradation of 25-norhopanes occurred under extreme biodegradation level (Huang and Li 2017;Chang et al. 2018;Killops et al. 2019), as validated by the increasing C 29 NH/C 28 NH (Fig. 7m). 25-norhopane ratio (∑C 30 -C 34 25-norhopanes/ (∑C 30 -C 34 25-norhopanes + ∑C 31 -C 35 homohopanes) generally exhibited a tendency to be increased with biodegradation, which was consistent with the formation of 25-norhopanes and reduction in homohopanes (Fig. 7n). Increasing C 27 DS/C 27 RS with biodegradation displayed the faster degradation of regular steranes than the diasteranes after PM7 level (Fig. 7o). Unchanged C 26 /C 28 TAS(20S) and C 27 /C 28 TAS(20R) well documented the strongest bioresistance of triaromatic steroids (Fig. 7q, 7r), yet the decreasing (C 20 + C 21 )/C 26-28 TAS demonstrated that the short-chained counterparts were slightly destroyed above PM7 (Fig. 7p).

Semi-quantitative evaluation by parameter-stripping method
Ideally, biomarkers were sequentially removed by biodegradation in a stepwise fashion and can be described by the tenpoint scales (Peters and Moldowan 1993). To further proof the extent of biodegradation four biodegradation stages were defined following the general alteration tendency in PM scheme (Fig. 8). Stage 1 covers approximately from PM1 to PM5, stage 2 from PM5 to PM7, stage 3 from PM7 to PM8, and stage 4 involves PM8 and higher. Biodegradation levels of different oil families were semi-quantitatively defined by sequentially stripping the samples in the cross-plots of biodegradation-resistant parameters. 2005). Pregnane (P) and homopregnane have high resistance to biodegradation, comparable to the diasterane (PM9). Tricyclic terpanes can be altered at about the same level as the diasteranes (Lin et al. 1989). Non-hopanoid triterpanes, such as gammacerane, diahopanes, oleanane, are highly resistant to biodegradation (Peters et al. 2005), and even beyond the point (PM ≥ 9) where the tricyclic terpanes have been removed . Although aromatized steroids keep unaltered in all but the most severely biodegraded oils (PM10) and can be effectively used to determine the origin and thermal maturity for extremely biodegraded oils (Peters and Moldowan 1993), low molecular weight triaromatic steroids (C 20 -C 21 ) are among the first aromatic steroids to be depleted during biodegradation (Wardroper et al., 1984). From the cross-plots of gammacerane/C 30 diahopane (G/C 30 D) vs. C 27 DS(20S)/C 27 DS(20R), C 26 /C 28 TAS(20S) vs. C 27 /C 28 TAS(20R), Carboniferous oils in the western and eastern Chepaizi Uplift were clearly distinguished exactly corresponding to the Group I and II, respectively (Fig. 9a,  9b).
Using triaromatic steroids ratios, previous studies concluded that the Carboniferous oils from the eastern Chepaizi Uplift had a common origin and similar maturities (Xi et al. 2014;Xiao et al. 2014;Xu et al. 2018;Chang et al. 2019;Mao et al. 2020); therefore, the biodegradation extent may be responsible for the observed differences of oil compositions. The tricyclic terpanes feature similarly high bioresistance to the diasteranes (PM > 8; Lin et al. 1989), slightly lower than the non-hopanoid terpanes. In the extreme biodegradation cases, tricyclic terpanes with lower molecular weight would be preferentially degraded (Huang and Li 2017;Chang et al. 2018). Oil family II 3 exhibited prominently higher C 24 / C 23 TT and C 22 /C 21 TT ratios than other oil families, implying a biodegradation level at least up to PM 9-PM9+ (Fig. 9c).
Obviously, by contrast, the biodegradation rank of family II 1 and II 2 were lower than PM9.

Stage 3
Hopanes are removed before or after steranes, 25-norhopanes occur in oils where the hopanes are preferentially removed (Reed 1977;Peters et al. 2005), as supported by this investigation (Fig. 7e). C 31 and C 32 homohopanes were more susceptible to biodegradation than C 30 hopane in the asphalts from Madagascar (Rullkotter and Wendisch 1982), while C 28 -C 30 17α-hopanes are typically biodegraded in the same manner and at approximately the same rate as the C 31 -C 35 extended hopanes (Williams et al. 1986). The 25-norhopane ratio, that is ratio of the total C 30 -C 34 25-norhopanes to the sum of these compounds plus the C 31 -C 35 homohopanes, could be used to evaluate the alternation extent among severely biodegraded oils of PM ~ 6-9 (Peters et al. 1996). Oil family II 2 , featured medium ratios of C 21 P/∑C 27-29 RS and C 22 P/∑C 27-29 RS (Fig. 9d), which clearly distinguished from those of the family II 3 (quite high) and family II 1 (rather low), and further verified by the varying range of C 29 25-NH/C 30 H and C 30 25-NH/C 31 H ratios (Fig. 9e). Hence, the biodegradation level of the family II 2 can be evaluated at PM8-PM8+ , while the family II 1 less than PM8.

Stage 2
Normal alkanes are preferentially removed at the biodegradation level of PM1-2; however, selective biodegradation of the isoprenoid over more bioresistant steranes or hopanes is used to determine the level of PM3-4 (Peters et al. 2005). Long-chained alkylated cyclopentanes and cyclohexanes are about as susceptible as branched alkanes and monocyclic alkanes will be depleted or in trace quantities at PM3-4 level (Peters and Moldowan 1993). C 14 -C 16 bicyclic terpanes are less susceptible to biodegradation than isoprenoid and will be completely eliminated before the start of sterane and hopane biodegradation (Peters et al. 2005). Regular steranes are removed after the complete removal of C 15 -C 20 isoprenoids and before or after the hopanes (Peters et al. 2005). Comparatively, slight alteration of methyl and dimethylnapthalenes occurs during the removal of n-alkanes, trimethylnaphthalenes are altered during the removal of the isoprenoids, and tetramethylnaphthalenes persist until steranes are largely depleted (Fisher et al. 1998). Phenanthrenes are generally more resistant to biodegradation than alkylnaphthalenes, and methylbiphenyls, dimethylbiphenyls, and methyldiphenylmethanes lacks in the biodegraded oils at PM7 (Peters and Moldowan 1993). Higher-hopane homologues, particularly the C 35 pentahomohopanes, are preferentially bioresistant, and the alteration of C 35 homhopanes are possibly at the rank of PM ≥ 7 (Seifert et al. 1984;Moldowan et al. 1995). The G/C 30 H and C 29 hopane/ C 30 hopane (C 29 H/C 30 H) varied from roughly unchanged to rapidly increased with the increasing biodegradation level, confirming the faster degradation of C 30 H than C 29 H, especially at level higher than PM8 (Bennett et al. 2006;Chang et al. 2018); however, the substantial constancy of C 29 18α-30-norneohopane/C 29 hopane (C 29 Ts/C 29 H) implies the similar susceptibility of C 29 Ts and C 29 H to biodegradation ). Oil family II 3 was biodegraded to PM7 by its lower C 35 /C 34 hopane (22S) and HHI values (Fig. 9f,  g). Furthermore, the tetramethylnaphthalene ratio (TeBR = 1,3,6,7-TeMN/1,3,5,7TeMN, TeMN = tetramethylnaphthalene; Peters et al. 2005) begins to be altered significantly at levels PM5-6 (Fisher et al. 1998). Therefore, oil family II 1 distinctly distinguished from other families and showed biodegradation level of PM7 as evidenced by the higher TeBR values (Fig. 9h).

Stage 1
No biodegradation occurred.

Refined Manco scale
The present biodegradation scales are essentially established according to the presence or absence of single compound classes and the alteration extent within a compound class, which encountered issues for super heavy oils . Manco scale emerged as the requirement, integrating the extent of degradation of various sets of compound classes. This quantitative Manco scale provided a higher resolution, however, only effective for the biodegradation of PM4-8.   Fig. 7 Correlations of biomarker parameters among three oil families of Group II showing the alterations of biomarkers with different susceptibility to biodegradation. TT, tricyclic terpane; Tet, tetracyclic terpane; H, hopane; G, gammacerane; Ts, 18α(H)-trisnorneohopane; Tm, 17α(H)trisnorhopane; NH, 25-norhopane; C 30 D, C 30 diahopane; ∑25-NHs, sum of C 30 -C 34 25-norhopanes; ∑HHs, sum of C 31 -C 35 homohopanes; DS, diasterane; RS, regular sterane; TAS, triaromatic steroid. All the ratio data were cited after  More biodegradation refractory compound class, i.e., 25-norhopane, tricyclic terpanes and triaromatic steroids, were added to refine the Manco scale to meet the situation of very heavy to severe biodegraded oils from Chepaizi Uplift. Followed the academic principles of Manco scale, eight compound classes that covered the whole range of biodegradation rank (PM0-10) and possessed increasing resistance to biodegradation were selected as vector elements. The eight compound classes and their GC-MS detection m/z values are listed in Table 2.
Five levels of refined Manco score (RMS) from 0-4 were assigned to the 8 compound classes, referencing after the Manco scale (Table 3). The refined Manco number (RMN 1 and RMN 2 ) can be calculated by the following formulas.
where m refers to the refined Manco score for each out of the eight vector elements (0-4), i means the class number (0-7), and n is the number of compound classes. S max , maximum for RMN 2 , is designate to be 1000 to avoid confusion with the currently existing scales, and to ensure enough resolution at different levels of biodegradation when using integer values.
Generally, RMN 2 ranged from 546.04 to 909.63, which approximately distinguished the oils at the level of PM6 to PM9 + (Table 4). Exceptionally, by contrast, the oils from well P60 exhibited higher oil density but lower biodegradation extent than that from well P685. Factually, there is not a simple relationship between the oil viscosity and the Manco Number . Processes other than biodegradation, i.e., secondary oil charge, water washing, mixing of multiple maturity oil charges, and loss of light ends from heavy oils could produce variations in oil viscosity and API gravity (López 2014). The two episodes of oil charging, early biodegraded oils mixed with the later remigration of preexisting oils due to the structural adjustment, yet the same oil origin in the Chepaizi Uplift maybe responsible for this case (Chang et al. 2019;Shi et al., 2020).
Notably, although the RMN 2 calculated by the refined Manco scale showed roughly positive correlation with oil density, some samples, especially those three biodegraded at PM8, exhibited much the same RMN 2 but with different oil densities. This indicated that biodegradation refractory compound class added in this refined scale could not perfectly differentiate the appearances among oils biodegraded at about PM8. More assemblages of compound classes need to be established to refine the Manco scales in further trials.

Geochemical implications
Although the Carboniferous oils in the western and eastern Chepaizi Uplift were derived from different source kitchens, oils within the individual groups had common origin and similar maturities (Zhang et al. 2014a, b;Xi et al. 2014;Xiao et al. 2014;Xu et al. 2018;Mao et al. 2020). The differences observed in PM and refined Manco scales were mainly attributed to the biodegradation extent. By contrast, the refined Manco scale showed higher resolution than the PM scale, which can differentiate the biodegradation extent of super heavy oils with same PM values but different oil viscosities (Table 4).
For the eastern Chepaizi Uplift, Carboniferous oils were mostly biodegraded above PM8; however, their oil viscosity displayed orders of magnitude variations. Known to all, the physical properties, particularly the oil viscosity, were critical to the choice of exploration strategies. Subtle differences revealed by the refined Manco scale could be helpful. Besides the practical application, refined Manco scale also allow geochemists conducting more basic study to understand the relative differences in the extent of the biodegradation process among related samples.

Conclusions
The heavy to severe biodegradation was responsible for the Carboniferous heavy oils in the Chepaizi Uplift, with PM6 in the western and PM7-PM9+ in the eastern part, respectively. According the oil-source correlation, the oils in the western Chepaizi Uplift were derived from the Jurassic source rock in the Sikeshu sag and those in the eastern Chepaizi Uplift were originated from the Permian source rocks in the Changji sag. Therefore, two oil groups (I and II) were distinguished. Three oil families (II 1 , II 2 and II 3 ) were subdivided with the biodegradation level of PM7, PM8-8+ , PM9+ , respectively, based on molecular  n-alkanes + C 1-4 naphthalenes 0: intact n-alkanes 1: slightly degraded n-alkanes, C 3 naphthalene as the peak compound of alkyl naphthalenes 2: substantially degraded n-alkane, C 3 naphthalene as the peak compound of alkyl naphthalenes 3: essentially depleted n-alkanes, C 4 naphthalene as the peak compound of alkyl naphthalenes 4: partially remained C 4 naphthalene in the alkyl naphthalenes 1 Alkyl dibenzothiophenes 0: non-degraded 1: only slightly degraded 2: mid-way between the extremes 3: not quite fully degraded 4: typically absent 2 C 0-3 phenanthrenes 0: complete C 0-3 phenanthrenes with the C 0 phenanthrene as peak compound 1: slightly degraded phenanthrenes with the C 1-2 phenanthrene as peak compounds 2: dimethyl and ethyl phenanthrene begun to be altered with the C 2 phenanthrene as peak compound 3: substantially degraded dimethyl and ethyl phenanthrene, C 3 phenanthrene begun to be altered with the C 3 phenanthrene as peak compound 4: wholly removed dimethyl and ethyl phenanthrene, only trace C 3 phenanthrene remained 3 C 4 phenanthrenes 0: non-degraded 1: only slightly degraded 2: mid-way between the extremes 3: not quite fully degraded 4: typically absent 4 Hopanes +methylhopanes + 25-norhopanes 0: intact hopanes with C 30 hopane as the peak compound 1: hopanes begun to be degraded still with C 30 hopane as the peak compound, and 25-norhopane was present 2: hopanes were substantially degraded with C 30 or C 29 hopane as the peak compound, 25-norhopane was prominently enriched in abundance 3: C 29 -25norhopane was the peak compound in m/z 191 chromatograms, 25,28-binorhopane begun to come into existing. 4: completely depleted hopanes with C 29 25-norhopane as the peak compound, 25,28-binorhopane increased prominently in abundance. 5 Regular steranes + diasterane + short-chain sterane 0: completed regular steranes with higher contents compared to diasteranes 1: slightly degraded regular steranes, slightly lower contents than diasteranes 2: substantially degraded regular steranes, which were evidently lower than the slightly degraded diasteranes 3: essentially depleted regular steranes, substantially degraded diasteranes 6 Tricyclic terpane + 17-nor tricyclic terpane 0: intact TTs 1: slightly degraded TTs with the presence of 17-nor tricyclic terpane 2: substantially degraded TTs with relatively increased 17-nortricyclic terpane 3: essentially degraded TTs, with far higher 1contents of 17-nortricyclic terpane than TTs 7 Triaromatic steroid + methyl triaromatic steroid 0: intact triaromatic steroid and methyl triaromatic steroid 1: slightly degraded C 19 -C 20 triaromatic steroids, and unaltered C 26 -C 28 triaromatic steroids and methyl triaromatic steroids 2: substantially removed low-carbon-number triaromatic steroids and methyl triaromatic steroids, high-carbon-number triaromatic steroids still unchanged compositions and parameter-stripping method of strongly bioresistant parameters. Three biodegradation refractory compound class (25-norhopane, tricyclic terpanes and triaromatic steroids) were added to establish a refined Manco scale to quantify the biodegradation extent of Carboniferous oils from Chepaizi Uplift. The evaluation results clearly differentiate the biodegradation extent of heavy oils with same PM values but different oil viscosities, indicating a high resolution and potential prospect. . Associated editor Jie Hao and seven anonymous reviewers were deeply acknowledged for their critical comments and helpful suggestions, which greatly improved the early version of this manuscript.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. Correlations of the refined Manco number (MN2) with oil viscosity (a), NSO content (b), summed pentacyclic terpane (c) and tricyclic terpane concentrations (d), and absolute concentrations of diahopane (e) and diasteranes (f). All the concentration data were cited after