Science of the Total Environment 791 (2021) 148056

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Environmental fate of petroleum biomarkers in Deepwater Horizon oil

spill residues over the past 10 years
Marieh Arekhi a, Leigh G. Terry a, Gerald F. John b, T. Prabhakar Clement a,⁎
a
Department of Civil, Construction and Environmental Engineering, The University of Alabama, Tuscaloosa, AL, USA.
b
Department of Civil and Environmental Engineering, Auburn University, Auburn, AL, USA.

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• The fate of petroleum biomarkers in

DWH residues was monitored for 10
years.
• Low molecular weight tricyclic trepanes
and steranes weathered over time.
• Heavy tricyclic terpanes and steranes,
and all pentacyclic terpanes remained
stable.
• All the triaromatic steranes experienced
a similar level of weathering.
• Despite weathering, the diagnostic ra-
tios of all the biomarkers remained
stable.

a r t i c l e i n f o a b s t r a c t

Article history: The long-term fate of three groups of petroleum biomarker compounds (terpanes, steranes, and triaromatic
Received 26 February 2021 steranes) was investigated in the Deepwater Horizon (DWH) oil spill residues collected from Alabama (USA)
Received in revised form 19 May 2021 beaches over the past 10 years. This is the first study to investigate the long-term fate of these three groups of
Accepted 23 May 2021
petroleum biomarkers in DWH oil spill samples over 10 years. We employed the highly recalcitrant C30 αβ-
Available online 28 May 2021
hopane as an internal biomarker to quantify the degradation levels of different biomarker compounds, and
Editor: Damia Barcelo also to estimate the overall weathering levels of DWH oil spill residues. The data show that four lower molecular
weight tricyclic terpanes (TR21, TR22, TR23, and TR24), three lower molecular weight steranes (S21, S22, and C27),
and all triaromatic steranes degraded over the 10-year study period. All other terpanes (including hopanes)
Keywords: and steranes remained recalcitrant. There have been contradicting literature data on the degradation levels of
Biomarkers homohopanes, and this field study demonstrates that all the homohopanes remained recalcitrant after 10
Terpanes years of natural weathering. Our data also show that despite some degradation, the relative diagnostic ratios of
Hopanes the biomarkers remained stable for all three groups of biomarkers over the 10-year period.
Steranes
© 2021 Elsevier B.V. All rights reserved.
Triaromatic steranes
Deepwater Horizon oil spill

1. Introduction (MC252) located about 66 km off the coast of Louisiana (Beyer et al.,
2016; Crone and Tolstoy, 2010; Dubansky et al., 2013). Approximately
The explosion of the Deepwater Horizon (DWH) oil platform, which 10% of the leaked oil from the DWH spill formed surface oil slicks,
began on April 20, 2010, in the Gulf of Mexico (GOM), released more which eventually oiled various GOM beaches (Aeppli et al., 2014). This
than 700 million liters of crude oil from the Macondo Prospect catastrophe is the largest oil spill in the United States history and one
of the largest in the world that resulted in over 1000 km of oiled shore-
⁎ Corresponding author. line across four different states (Powers et al., 2017). Years after the
E-mail address: [email protected] (T.P. Clement). spill, oil-soaked sand agglomerates, also known as surface residual

https://doi.org/10.1016/j.scitotenv.2021.148056
0048-9697/© 2021 Elsevier B.V. All rights reserved.
M. Arekhi, L.G. Terry, G.F. John et al. Science of the Total Environment 791 (2021) 148056

balls (or tarballs) continue to break away from submerged oil mats (or degraded within two years in the DWH oil spill residues, and the au-
tarmats) and wash ashore onto GOM beaches (Clement et al., 2017; thors hypothesized that photo-oxidation was the main degradation
Gustitus and Clement, 2017; White et al., 2016; Yin et al., 2015b). pathway. However, other studies have shown that triaromatic steranes
Chemical fingerprinting methods are needed to distinguish the can be highly resistant to degradation (Connan, 1984; Douglas et al.,
DWH residues from other oil residues that are formed from natural 2012; Williams et al., 1986; Wu et al., 2013).
seeps, accidental releases from oil exploration, production of crude oil, Our literature review has indicated that there are gaps and contra-
and petroleum transportation activities (Aeppli et al., 2014). Currently, dicting data in describing the long-term fate of different types of petro-
the most common method used for identifying the oil spill source is to leum biomarkers in natural environments (see Table S1 of the
study the biomarker fingerprints. Petroleum biomarkers are geochemi- Supplementary Material for a detailed summary). Interestingly, our re-
cal organic compounds present in crude oils that can be related to their cent field studies have shown that Alabama beaches continue to be con-
unique biological precursors (Wang et al., 2014). Since spilled oil is taminated with DWH oil spill residues (Clement et al., 2017). These oil
subjected to a variety of weathering processes such as evaporation, dis- spill residues are ideal samples to investigate the long-term fate of pe-
solution, dispersion, photochemical oxidation, and microbial biodegra- troleum biomarkers under natural conditions. In this study, the environ-
dation (Wang et al., 2014), it is important to understand degradation mental fate of three common groups of biomarker compounds in DWH
levels of the petroleum biomarkers in the oil spill samples to validate oil spill residues was investigated by analyzing terpanes (including tricy-
their use for fingerprinting purposes. clic and pentacyclic terpanes) detected using a gas chromatogram at m/z
Petroleum biomarkers, such as terpanes, steranes, and triaromatic of 191, steranes (including diasteranes, coloastanes, ergostane, and
steranes, are commonly used to identify the source of oil spills, and stigmastane) detected at m/z of 217, and triaromatic steranes detected
some of these biomarkers are also used as recalcitrant internal tracers at m/z of 231. The DWH oil spill was unique since it occurred at a
for quantifying weathering levels (Aeppli et al., 2014; Munoz et al., lower latitude and higher temperature environment. The warmer
1997; Prince et al., 1994; Venosa et al., 1997). 17α(H), 21β(H)-hopane environmental setting allowed us to study how the higher temperatures
(known as C30 αβ-hopane), which belongs to a general class of bio- and intense solar irradiation would have affected biomarker weathering.
marker compounds known as terpanes, is one of the most commonly Specifically, the objective of this study is to test the following two
used biomarker compounds. Previous studies have shown that C30 hypotheses: 1) under natural weathering conditions, the lower molecu-
αβ-hopane is a highly stable biomarker and hence it is routinely used lar weight biomarker compounds (e.g., low carbon terpanes and
as a conservative internal standard for assessing the weathering levels steranes and triaromatic steranes) will weather, and the relatively
of oil spill residues (Aeppli et al., 2014; Mulabagal et al., 2013; Prince heavier molecular weight biomarker compounds (e.g., high carbon
et al., 1994; Venosa et al., 1997; Wang et al., 2001). terpanes and steranes) will be mostly conserved, and 2) despite some
Terpane compounds include tricyclic and pentacyclic terpanes, in natural weathering, the relative diagnostic ratios of the biomarkers
which hopanes are a class of pentacyclic terpanes that originate from will be preserved. The biomarker compounds in the DWH tarball sam-
hopanoids in bacterial membranes (Bost et al., 2001; Peters and ples collected over the past 10 years from Alabama beaches (samples
Moldowan, 1991). Numerous studies have shown that most pentacyclic from 2010, 2011, 2015, and 2020 field surveys) were analyzed to test
terpanes are highly stable compounds (Bost et al., 2001; Frontera-Suau these two hypotheses.
et al., 2002; Wang et al., 2001; Wang et al., 1994b, 1995). For example,
Frontera-Suau et al. (2002) showed that C35 homohopane 2. Materials and methods
(homohopanes are C31-C35 hopanes) was conserved in microcosm ex-
periments after 21 days of incubation of Bonny Light crude oil, which 2.1. Field sampling surveys
was degraded at 30 °C using mixed cultures of microorganisms enriched
from surface soils. In addition, tricyclic terpanes in crude oils can also be The study region included the sandy beaches of the Alabama Gulf
fairly resistant to biodegradation (Lin et al., 1989; Munoz et al., 1997; Coast region, which is approximately 45 km (28 miles) in length, ex-
Williams et al., 1986). However, some studies have shown that tricyclic tending from Fort Morgan, AL in the west to Orange Beach, AL in the
terpanes can be degraded by microbial processes under both field east (Fig. 1a). During each sampling event, numerous tarball samples
(Cheng et al., 2016; Howell et al., 1984; Prince et al., 2002; Wang of different sizes were collected from these beaches. Our most recent
et al., 2001; Wang et al., 1994b, 1995) and laboratory (Bost et al., survey was completed on March 17, 2020, where we observed a highly
2001) conditions. Cheng et al. (2016) showed that C19–C21 tricyclic contaminated zone facing the GOM side of Fort Morgan (see Fig. 1b).
terpanes were the most readily degraded compounds, followed by C22 Within a 50 m × 50 m contaminated area, we recovered over 150
and C23 terpanes, while the C24+ tricyclic terpanes were more resistant tarballs with sizes ranging from 2 cm to 10 cm. The total weight of
to biodegradation. tarballs collected from this region was 1250 g. These tarballs must
Another class of biomarkers abundant in petroleum is steranes, have originated from buried DWH tarmats that must have been
which are derived from steroids or sterols of living systems (Peake suspended by various nearshore transport processes. Fig. 1b shows
and Hodgson, 1973; Peters and Moldowan, 1993). Similar to terpanes, this field site and all the tarballs collected from this site during our
steranes are rarely affected by degradation processes (Shirneshan March 2020 survey. We also sampled the Mobile Bay side of the Fort
et al., 2016). Wang et al. (2001) showed that C29 18α(H), 21β(H)-30- Morgan site where the tarball contamination was spread over a larger
norneohopane and C29 αββ-stigmastanes (20R and 20S) were the region, and we collected about 100 tarballs with a total weight of
most biodegradation resistant terpane and sterane compounds in their 875 g over a kilometer. Early studies have claimed that the Alabama
24-year old field samples collected from the 1974 Metula oil spill site. beaches are clean and the coastal system has recovered to background
The authors also found that sterane degradation occurred in the follow- conditions after about three years of active cleanup (Smith, 2013). Our
ing order: C27 > C28 > C29. Overall, the different types of steranes and March 2020 field survey has indicated that the Alabama coastline con-
terpanes degraded in the following order: diasteranes > C27 steranes tinue to be contaminated by DWH tarballs, and the beaches have not
> tricyclic terpanes > pentacyclic terpanes > norhopanes ≈ C29 αββ yet recovered to the pre-oil spill background condition (which should
steranes (Wang et al., 2001). be about 2 g/km/year, according to Clement et al., 2017) even after 10
Triaromatic steranes are another category of steranes, which are years.
formed by diagenesis and maturation of sterols (Peters and All the DWH oil spill samples collected from the Alabama beaches
Moldowan, 1993; Wang et al., 2014). However, very few studies have over the past 10 years look similar to the tarballs shown in Fig. 1b. The
examined the effects of degradation on these compounds (Wang et al., field samples were transferred to the laboratory and stored in the refrig-
2014). Aeppli et al. (2014) study reported that triaromatic steranes erator at 4 °C. Seven tarball samples, which were collected during field

2
M. Arekhi, L.G. Terry, G.F. John et al. Science of the Total Environment 791 (2021) 148056

Fig. 1. (a) Field sampling locations; (b) photograph of the oil contaminated field site in Fort Morgan, AL (facing the Gulf side) and the tarballs collected during the March 2020 survey.

surveys completed from June 2010 through March 2020, were selected and the visible organic and inorganic debris on the tarball surfaces
for this study (note, multiple tarballs collected during these surveys were first removed. Then the tarball samples were broken down
were also analyzed, and the results were similar to the results of the into smaller fragments and thoroughly homogenized. The samples
seven tarball samples reported in this study). The description of these were then placed in a fume hood for 3–5 days to allow the trapped
seven tarball samples, collection date, location, and GPS coordinates moisture to evaporate. The fume hood lights were turned off to min-
are given in Table S2. During our field survey, the tarballs recovered imize direct photo-degradation. Next, about 3 g of subsample was
were initially identified as potential DWH or non-DWH samples using taken from the dried homogenized sample and was extracted using
the Tier-1 and Tier-2 screening methodology (Han and Clement, 5 ml of dichloromethane. The extraction step was repeated 3–4
2018). The samples were later confirmed using advanced biomarker times or until the dichloromethane extract was clear and colorless.
fingerprinting methods (data provided in Section 3.3.1). A reference The remaining solid residues were then placed in the fume hood
DWH crude oil sample (referred to as MC252) supplied by the British for 24 h and then weighed to estimate the oil content. The extraction
Petroleum (BP) was also analyzed. process was repeated three times for every sample with three differ-
ent subsamples to estimate the average oil content of the homoge-
2.2. Materials nized tarball samples. The average oil contents were estimated to
be 10%, 15%, 11%, 18%, 14%, and 11% for OB-2010, BS1–2011,
All organic solvents used in this study were HPLC grade or higher. BS2–2011, FMB1–2015, FMB2–2015, and FMG-2020, respectively.
The solvents (hexane, dichloromethane, methanol, and acetone), silica Then, the sample clean-up and fractionation procedures were per-
gel (60–200 μm), and anhydrous sodium sulfate (ACS grade) were pur- formed for each of the samples based on the oil content.
chased from VWR International Company (Suwanee, GA, USA).
Deactivated borosilicate glass wool was purchased from Restek 2.3.2. Sample clean-up and column fractionation procedures
Company (Bellefonte, PA, USA). Hopane standards, C30 ββ-hopane Sample clean-up and chromatographic column fractionation steps
(17β(H), 21β(H)-hopane, > 98% by GC/MS) as an internal standard were conducted using a previously well-established method (Arekhi
and C30 αβ-hopane (17α(H), 21β(H)-hopane, > 98% by GC/MS) as a et al., 2020; Wang et al., 1994a; Yin et al., 2015a). A 10 mm ID glass chro-
calibration standard, were purchased from Chiron (Trondheim, matographic column was plugged with glass wool. The column was
Norway). An internal standard p-terphenyl-d14 (purity >98.5%) was then filled with 3 g activated silica gel and topped with 1 g anhydrous
purchased from AccuStandard (New Haven, CT, USA). A mixture of deu- sodium sulfate. The column was then charged with hexane. A tarball
terated compounds consisting of acenaphthene-d10, phenanthrene-d10, sample containing about 25 mg of oil equivalent, calculated based on
chrysene-d12, and perylene-d12 was used as a surrogate standard and oil content, was weighed in a glass vial and spiked with 20 μl of 50 μg/
was purchased from Agilent Technologies (Wilmington, DE, USA). ml surrogate standards. The sample was then extracted with 3 × 1 ml
MC252 reference crude oil was supplied by the British Petroleum (BP) of hexane and transferred to the chromatographic column. Aliphatic
Company. GC capillary column (J&W DB-EUPAH, 60 m × 0.250 mm × (F1) and aromatic (F2) hydrocarbon fractions were obtained by
0.25 μm, p/n 122-96 L2) and deactivated GC liners (splitless tapered successively eluting with hexane and hexane:dichloromethane
glass wool) were purchased from Agilent Technologies (Wilmington, (50%, v/v) solvent mixture, respectively. The F1 and F2 fractions
DE USA). were concentrated under a gentle stream of nitrogen and the
required amount of solvent was added to adjust the final volumes
2.3. Sample extraction and clean-up procedures to 10 ml. Exactly 1 ml of the adjusted F1 and F2 fractions were trans-
ferred into 2 ml GC vials and then were spiked with 10 μl of 10 μg/ml
2.3.1. Estimation of oil percentage levels C 30 ββ-hopane and 10 μl of 50 μg/ml p-terphenyl-d 14 as internal
The oil content was estimated using a previously published standards, respectively, before chemical analysis. All the samples
method (Mulabagal et al., 2013; Yin et al., 2015b). The outer sides were prepared in duplicate.

3
M. Arekhi, L.G. Terry, G.F. John et al. Science of the Total Environment 791 (2021) 148056

2.4. Instrumental analysis and 200 ng/ml and spiked with an internal standard (C30 ββ-hopane,
100 ng/ml). The calibration response was linear across the selected an-
Analysis of biomarker compounds was accomplished using an alytical range, yielding correlation coefficient (R2) values of at least
Agilent 7890B gas chromatograph (GC) fitted with an Agilent 7000C tri- 0.995.
ple quadrupole mass spectrometer (MS). The separation of the various
biomarker compounds was achieved using an Agilent J&W DB-EUPAH 2.6.3. Determination of weathering percentages
column, and helium as the carrier gas. The GC conditions and MS param- A method using the C30 αβ-hopane as an internal conservative bio-
eters are given in Table S3. marker was used to estimate the degree of weathering of the DWH sam-
Biomarkers (terpanes, steranes, and triaromatic steranes) were ana- ples by applying the following equation (Eq. (2)) (Mulabagal et al.,
lyzed similar to the previously published GC/MS procedure performed 2013; Wang et al., 2001; Wang et al., 1995):
in the single ion monitoring (SIM) mode (Arekhi et al., 2020; Han
et al., 2018; Mulabagal et al., 2013). The list of the terpanes, steranes, WP ð%Þ ¼ ð1−C S =C W Þ  100 ð2Þ
and triaromatic steranes analyzed in this study using m/z values of
191, 217, and 231, respectively, are summarized in Tables S4-S6. Note,
terpanes and steranes were analyzed using the F1 fraction and where WP is the weathered percentages of the weathered samples, and
triaromatic steranes were analyzed using the F2 fraction. CS and CW are the concentrations of C30 αβ-hopane in the DWH source
oil and weathered samples, respectively.
2.5. Statistical analysis
2.7. Quality assurance and quality control
The two-sample student's t-test was used to test whether there is a
statistical difference between the calculated mean values of peak areas The samples were spiked with the internal standards (C30 ββ-
for the MC252 crude oil and the DWH samples. The t-test is a robust ap- hopane for F1 fractions and p-terphenyl-d14 for F2 fractions) prior to in-
proach to determine whether the means of two sets of data are signifi- strumental analysis, and the peak areas of all the analytes of interest
cantly different from each other (Devore, 2015; Sun et al., 2011), and were normalized to the internal standards' peak areas to compensate
the p-value is the evidence against the null hypothesis. The smaller for instrumental variations. Prior to sample clean-up and fractionation,
the p-value, the stronger the evidence that we should reject the null hy- the samples were spiked with the surrogate standard mixture to moni-
pothesis (Devore, 2015). The null hypothesis in this study is that the tor net recovery levels. The measured recovery levels were within the
mean values of peak areas for different biomarker compounds (normal- acceptable range (60–150%) for the four surrogate standards
ized to C30 αβ-hopane) for the MC252 crude oil are equal to the mean (acenaphthene-d10, phenanthrene-d10, chrysene-d12, and perylene-
values in the DWH samples, indicating the samples are not statistically d12). A midpoint calibration standard (50 ng/ml or 100 ng/ml) was
different. A 95% confidence level (level of significance of the test α = checked before starting a sample sequence to validate the instrument.
0.05) was used to test our hypothesis, therefore, p-values below 0.05 in-
dicated that we should reject the null hypothesis with 95% confidence 3. Results and discussion
concluding that the two means are significantly different. The
Minitab-19 software was used to calculate all the p-values. 3.1. Fate of biomarkers in DWH tarballs over the 10-year period

2.6. Quantification methods To investigate the potential degradation patterns of the three groups
of biomarker compounds (terpanes, steranes, and triaromatic steranes),
The ratios of the biomarkers and the concentrations of the C30 αβ- the peak areas of all the analytes of interest (see Tables S4-S6 for the
hopane were calculated by computing various peak areas using Agilent analytes) were normalized to C30 αβ-hopane peak areas in the GC/MS
Technologies MassHunter MS quantification software (version B.09.00). chromatograms of the DWH samples, and these results are discussed
Biomarker chromatograms were prepared using Agilent Technologies in the following sections.
MassHunter qualitative software (version B.08.00).
3.1.1. Fate of terpanes
2.6.1. Determination of fractional losses The values of the C30 αβ-hopane normalized peak areas of the tricy-
The fractional loss of a compound “X” in a sample was calculated by clic and pentacyclic terpanes for the DWH samples compared to the
normalizing the compound peak area to C30 αβ-hopane and comparing MC252 reference crude oil are shown in Fig. 2 (see Table S7 for the tab-
it to the corresponding value in the DWH source oil, according to the fol- ular values). The extracted ion chromatograms of terpanes (at m/z of
lowing equation (Eq. (1)) (Aeppli et al., 2014; Prince et al., 2002; Yin 191) for the MC252 reference crude oil (Fig. S1a) and two DWH tarball
et al., 2015b): samples collected in 2010 and 2020 (Figs. S1b&c) show that the MC252
  reference oil and the DWH samples contain terpanes ranging from C21
X W =H W tricyclic terpane (TR21) to C35 pentakishomohopanes (H35) with C30
LossX ¼ 1−  100 ð1Þ
X S =H S αβ-hopane (H30) being the most prominent peak. Four lower molecular
weight tricyclic terpanes, including TR21, TR22, TR23, and TR24, degraded
where XW and XS are the peak areas of the compound “X” in a weathered in the DWH samples as compared to the MC252 reference oil (Fig. 2a,
sample and DWH source oil, respectively, normalized to the peak areas and see Table S8 for the calculated p-values (< 0.05) suggesting signif-
of their proper internal standards (p-terphenyl-d14 for triaromatic icantly different mean values). However, five heavier molecular weight
steranes and C30 ββ-hopane for terpanes and steranes), and HW and tricyclic terpanes (TR25-TR29) in each of the DWH samples are statisti-
HS are the peak areas of the C30 αβ-hopane in a weathered sample cally similar to the MC252 reference oil (see Table S8 for the calculated
and DWH source oil, respectively, normalized to the peak areas of their p-values (> 0.05)) suggesting that degradation did not occur in the
proper internal standards (C30 ββ-hopane) in the GC/MS chromato- heavier molecular weight tricyclic terpanes (Fig. 2a). The fractional
gram of a specific sample. losses of the four lower molecular weight tricyclic terpane compounds,
which were calculated relative to the MC252 reference oil using Eq. (1),
2.6.2. Calibration curve show that the tricyclic trepanes (TR21, TR22, TR23, and TR24) experienced
The calibration curve was developed using the C30 αβ-hopane stan- continuous degradation for 11 months post-spill but remained stable
dard with five calibration points at concentration levels of 1, 10, 50, 100, henceforth (Table 1). Among these four tricyclic terpanes, the lower

4
M. Arekhi, L.G. Terry, G.F. John et al. Science of the Total Environment 791 (2021) 148056

Fig. 2. Peak areas of (a) tricyclic terpanes and (b) pentacyclic terpanes normalized to C30 αβ-hopane in the DWH tarball samples and MC252 crude oil.

molecular weight compounds degraded more than the heavier ones

(Table 1) in the suggested sequence of TR21 ≥ TR22 ≥ TR23 ≥ TR24.
The values of the pentacyclic terpanes remained statistically stable
over the 10-year period (see Table S8 for p-values >0.05) and are simi-
lar to the MC252 reference oil (Fig. 2b). These data indicate that all
pentacyclic terpanes including all hopanes and homohopanes (H31 to
H35) are relatively stable compounds that could resist weathering over
the 10-year period.

3.1.2. Fate of steranes

The sterane values (peak areas) normalized to C30 αβ-hopane for
the DWH samples compared to the MC252 reference crude oil are
shown in Fig. 3 (see Table S9 for the values). The extracted ion chro-
matograms of steranes (at m/z of 217) for the MC252 reference crude
oil (Fig. S2a) and two DWH tarball samples collected in 2010 and
2020 (Figs. S2b&c) show that steranes are characterized by the distribu-
tion from C21 sterane (S21) to C29 stigmastanes (C29) with C27 20S and
20R-diasteranes (DIA27S and DIA27R) and C27 20S and 20R-cholestanes
(C27) comprising the most prominent peaks. The values of three lower
molecular weight steranes including S21, S22, and C27 cholestanes in
the DWH tarballs show a statistical difference (p-values <0.05) from Fig. 3. Peak areas of steranes normalized to C30 αβ-hopane in the DWH tarball samples
and MC252 crude oil.
the crude oil mean values (Table S10). The fractional losses for these
three sterane compounds range from 26 ± 1 to 43 ± 1%, 19 ± 1 to 34
± 1%, and 3 ± 1 to 12 ± 3%, respectively (Table 2). All other sterane degraded than the heavier ones in all the DWH samples (Table 2) in the
compounds (C27 and C28 diasteranes, C28 ergostanes, and C29 following sequence of S21 ≥ S22 ≥ C27(S+R) cholestanes.
stigmastanes) remained statistically unchanged (p-values >0.05) over
the 10-year study period when compared to the MC252 reference
crude oil (Fig. 3 and Table S10). Similar to tricyclic terpanes, the frac- 3.1.3. Fate of triaromatic steranes
tional losses of the three lower molecular weight steranes (S21, S22, The C30 αβ-hopane normalized values (peak areas) of the
and C27 cholestanes) show that these steranes experienced ongoing triaromatic steranes for the DWH samples compared to the MC252 ref-
degradation for 11 months and remained stable henceforth (Table 2). erence oil indicate that these compounds are relatively less stable com-
The lower molecular weight steranes were observed to be more readily pounds (Fig. 4 and Table S11). The values of all triaromatic sterane

Table 1 Table 2
Fractional losses for the lower molecular weight tricyclic terpanes in the DWH samples Fractional losses for the lower molecular weight steranes in the DWH samples with re-
with respect to the MC252 reference crude oil. spect to the MC252 reference crude oil.

Samples Tricyclic terpane losses (%) Samples Steranes losses (%)

TR21 TR22 TR23 TR24 S21 S22 C27

OB-2010 33 ± 1 30 ± 3 16 ± 1 14 ± 1 OB-2010 26 ± 1 19 ± 1 3±1

BS1–2011 46 ± 1 44 ± 2 25 ± 1 21 ± 1 BS1–2011 38 ± 1 27 ± 1 10 ± 1
BS2–2011 45 ± 1 49 ± 5 23 ± 1 17 ± 2 BS2–2011 38 ± 1 27 ± 1 7±1
FMB1–2015 45 ± 3 44 ± 1 23 ± 1 14 ± 5 FMB1–2015 38 ± 1 27 ± 1 5±1
FMB2–2015 49 ± 3 47 ± 5 28 ± 4 24 ± 2 FMB2–2015 41 ± 3 30 ± 1 12 ± 3
FMG-2020 55 ± 1 46 ± 4 26 ± 1 23 ± 1 FMG-2020 43 ± 1 34 ± 1 11 ± 1

5
M. Arekhi, L.G. Terry, G.F. John et al. Science of the Total Environment 791 (2021) 148056

using Eq. (1) (see Fig. S3 for the individual peaks in the chromatograms
of the MC252 reference oil and the samples). The average loss of
SC28TA, for example, is estimated to be 31 ± 2%, 66 ± 2%, 91 ± 6%,
75 ± 8%, 69 ± 1%, and 55 ± 2% for OB-2010, BS1–2011, BS2–2011,
FMB1–2015, FMB2–2015, and FMG-2020, respectively (see Table S13
for all the values). The calculated losses of each triaromatic sterane com-
pound compared to the losses of SC28TA (set as a reference compound)
demonstrate that all compounds degraded at a similar rate, resulting in
a slope ≈ 1, as shown in Fig. 5, which shows that preferential removal of
individual compounds is not observed. Also, similar to the lower molec-
ular weight tricyclic terpanes and steranes, all triaromatic steranes ex-
perienced the highest degradation during the first year and then
remained almost stable. BS2–2011 sample was observed to be the
most degraded sample showing the lowest C30 αβ-hopane normalized
values (Fig. 4) and the highest losses (Table S13).

3.2. Discussion of terpanes, steranes, and triaromatic steranes degradation

data

Fig. 4. Peak areas of triaromatic steranes (TAS) normalized to C30 αβ-hopane in the DWH The data presented in Section 3.1 show that pentacyclic terpanes
tarball samples and MC252 crude oil. were not degraded in the DWH samples collected over the 10-year in-
vestigated time scale. These observations are consistent with several
compounds are statistically lower (p-values <0.05) in all DWH samples previous studies that characterized and reported stability levels of
when compared to the values in MC252 reference crude oil (Fig. 4 and hopanes and homohopanes in different types of oil spill samples,
Table S12). Fractional losses were calculated for each of the compounds e.g. the 6-month old samples collected from DWH oil spill sites

Fig. 5. Calculated losses for each triaromatic sterane compound versus the losses of the SC28TA in the DWH samples. Data points include the MC252 reference oil (the origin of the
coordinates) and six weathered DWH samples. SC28TA was set as a comparator relative to other compounds, and slope ≈ 1 for each graph represents similar loss values for the
compound of interest compared to SC28TA.

6
M. Arekhi, L.G. Terry, G.F. John et al. Science of the Total Environment 791 (2021) 148056

(White et al., 2012), the 12-year old (Wang et al., 1995), and 20-year old 3.3. Application of biomarkers in DWH tarballs over the 10-year period
(Prince et al., 2002) samples collected from the experimental Arctic oil
spill in Baffin Island, Canada, and the 22-year old samples collected 3.3.1. Assessment of diagnostic ratios for fingerprinting
from the Arrow oil spill in Canada (Wang et al., 1994b) (see Table S1 The diagnostic ratios for different pairs of hopane, sterane, and
for more details). However, these results are different from other obser- triaromatic sterane compounds were calculated to test the second hy-
vations, which all reported that heavy homohopanes (H32-H35) (bio) pothesis (see Tables S4-S6 for the list of the diagnostic ratios). These ra-
degraded faster than lighter homohopanes and C30 αβ-hopane; the ex- tios are commonly used in oil fingerprinting and are estimated based on
amples include the 28-month old samples collected from DWH oil spill the area of the individual peaks (Yang et al., 2015).
sites (Aeppli et al., 2014), field observations made on salt marshes on The radar plots of the diagnostic ratios of hopanes and steranes
the Strait of Magellan 24 years after the Metula oil spill in Chile for the DWH samples and a non-DWH sample are compared to the
(Wang et al., 2001), and an 8-year experimental oil spill study on man- MC252 reference oil (Fig. 6) to test oil origin (calculated values of di-
groves located near Guadeloupe, France (Munoz et al., 1997). Yet, other agnostic ratios can be found in Table S14). The radar plots of the
published studies have shown that the lower homologs of hopanes (Fig. 6a) and steranes (Fig. 6b) for all the DWH samples
homohopanes biodegraded faster than the higher counterparts (Lin and the MC252 reference crude oil are identical. For sterane, minor
et al., 1989; Peters et al., 1996). Therefore, the answer to the question variations are observed for S21/S22 and C27/C29 ratios due to the pref-
of whether the heavier or the lighter homohopanes degrade first is cur- erential degradation of lighter steranes (refer to Section 3.1.2). The
rently unclear. Our field data show that none of the homohopanes (H31 radar plots of both hopanes and steranes for the non-DWH sample
to H35) in the DWH oil spill samples degraded after 10 years of natural are different, clearly indicating that the sample originated from a dif-
weathering. ferent source (Figs. 6a&b).
In our field samples, tricyclic terpanes, steranes, and triaromatic The use of biomarker ratios for triaromatic sterane compounds was
steranes degraded to some extent. The degradation of tricyclic terpanes also explored, despite the high level of degradation relative to C30 αβ-
and steranes only occurred for the lower molecular weight compounds, hopane. The radar plots of the calculated triaromatic sterane ratios for
while their heavier molecular weight compounds were recalcitrant. The the DWH samples compared to the MC252 reference crude oil and the
observed degradation sequence of TR21 ≥ TR22 ≥ TR23 ≥ TR24 for the four non-DWH sample show that despite the high level of degradation, the
lower molecular weight tricyclic terpanes is consistent with the litera- diagnostic ratios remain fairly stable (Fig. 7 and Table S14). The radar
ture data (Wang et al., 1994b, 1995), and the degradation sequence of plots of the DWH samples match well with the MC252 reference oil
S21 ≥ S22 ≥ C27(S+R) cholestanes for the three lower molecular weight radar plot indicating that the ratios are preserved and the triaromatic
steranes is consistent with other published studies (Lin et al., 1989; sterane compounds should have degraded at a similar rate. For compar-
Munoz et al., 1997; Murray et al., 2016; Prince et al., 2002; Wang ison, the radar plot of the non-DWH sample is different from the DWH
et al., 2014; Wang et al., 2013; Wang et al., 2001; Wang et al., 1994b, samples and the MC252 crude oil.
1995; Watson et al., 2002). However, all the triaromatic sterane com- Our data suggest that the relative diagnostic ratios of the biomarkers
pounds degraded at the same level, which is consistent with the results are preserved in the DWH samples. These data support our second hy-
observed by Aeppli et al. (2014). Different degradation pathways have pothesis that despite the considerable natural weathering, the relative
been suggested in different studies for triaromatic steranes such as bio- diagnostic ratios of several key biomarkers will be preserved making
degradation (Douglas et al., 2012; Lin et al., 1989), photo-oxidation them useful for source identification. These observations are consistent
(Aeppli et al., 2014), and a combination of both (Wang et al., 2014) as with the published literature including the 22-year old Arrow oil spill
can be seen in Table S1. Although biodegradation is a possible pathway, study (Wang et al., 1994b), a laboratory-scale biodegradation experi-
some studies have also shown that triaromatic steranes are highly resis- ment (Wang et al., 2013), and a crude oil biodegradation/photo-
tant to biodegradation (Watson et al., 2002; Williams et al., 1986; Wu oxidation study (Wang et al., 2014), which all showed that the ratios
et al., 2013), and only small changes have been reported for the lower of several pairs of terpanes, steranes, and triaromatic steranes remained
molecular weight compounds (Douglas et al., 2012; Lin et al., 1989; reasonably stable even when some of the biomarkers degraded. A re-
Wardroper et al., 1984). Therefore, further studies are needed to better cent study also showed that even under severe burning conditions, sev-
understand the types of natural processes that could degrade these eral diagnostic ratios of hopanes remained stable in DWH crude oil
compounds. (John et al., 2018).

Fig. 6. Radar plots comparing the diagnostic ratios of (a) hopanes and (b) steranes for both DWH and non-DWH samples and the MC252 reference crude oil.

7
M. Arekhi, L.G. Terry, G.F. John et al. Science of the Total Environment 791 (2021) 148056

weather, but at a slower rate when compared to the initial rates. The
tarballs collected five to ten years after the spill (FMB1–2015,
FMB2–2015, FMG-2020) have an average weathering level of 62%,
which is only about a 3% increase from the 2011 value. These results in-
dicate that major weathering occurred initially during the early months,
and then the weathering process slowed down as the oil was mixed
with sand and buried in the nearshore environment. These observations
are consistent with Mulabagal et al. (2013) study, which also reported
rapid initial weathering during the early days, followed by relatively
slow weathering in the subsequent two years.

4. Conclusions

Petroleum contamination that resulted from the DWH oil spill con-
tinues to impact Alabama's beaches 10 years after the spill. During our
March 2020 survey, which was completed along Fort Morgan beaches,
we collected over 1250 g of tarballs from a highly contaminated 50-m
long zone facing the GOM; and we also recovered over 875 g of tarballs
from a kilometer-long shoreline facing the Mobile Bay. These contami-
nation levels are several orders of magnitude larger than the estimated
Fig. 7. Radar plots comparing the diagnostic ratios of triaromatic steranes for both DWH
historic background oil level of 2 g/km/year (Clement et al., 2017) that
and non-DWH samples and the MC252 reference crude oil.
existed before the DWH oil spill. These field observations show that
the DWH oil spill residues continue to linger along the Alabama shore-
3.3.2. Quantification of weathering levels line; therefore, it is necessary to have reliable fingerprinting methods
In order to estimate the overall weathering percentages of the DWH that can identify these residues. The reliability of these methods
samples, the concentration changes of any one of the stable biomarker depends on the stability of petroleum biomarkers. Previous studies
compounds can be quantified. In this study, the concentration of C30 have reported contradicting data for biomarker degradation levels in
αβ-hopane was quantified in all samples, and the results are shown in different types of environments. This is the first long-term field study
Fig. 8 (see Table S15 for the values). The concentration of C30 αβ- that provides data to document the fate of three groups of biomarkers
hopane measured in the MC252 reference crude oil was verified and including terpanes, steranes, and triaromatic steranes in the DWH oil
was found to be consistent with values reported in a previous study spill residues that have weathered in a coastal environment for over
(Mulabagal et al., 2013) and a multi-laboratory study completed by 10 years.
the National Institute of Standards and Technology (NIST) (Murray Our results show that higher molecular weight terpanes (heavy tri-
et al., 2016). The data presented in Fig. 8 show that C30 αβ-hopane con- cyclic terpanes (>TR24) and all pentacyclic terpanes) and higher molec-
centrated in all DWH samples over the 10-year time period. ular weight steranes (diasteranes, ergostanes, and stigmastanes) can
The concentration levels were then used to estimate the overall remain recalcitrant for a long period. Our data also show that all the
weathering percentages of the DWH samples using Eq. (2). The homohopanes (H31 to H35) remained stable after 10 years of natural
weathering (or mass loss) percentages of the DWH samples (Fig. 8) in- weathering in the coastal environment. Furthermore, our data indicate
dicate that the tarballs collected in June 2010 (OB-2010), two months that several lower molecular weight tricyclic trepanes (e.g., TR21, TR22,
after the spill, weathered by 41%, which indicates the oil weathered sig- TR23, TR24), lower molecular weight steranes (e.g., S21, S22, and C27
nificantly while floating over the ocean, mediated by various physico- cholestanes), and all the triaromatic steranes weathered rather rapidly
chemical processes (such as evaporation, dissolution, dispersion, and during the initial months when the oil was floating over the ocean.
photo-oxidation) and possibly some biological processes. The tarballs The reason for this early degradation is currently unclear. Further stud-
collected in March 2011 (BS1–2011 and BS2–2011), 11 months after ies are needed to better understand the types of ocean-scale transport
the spill, weathered by 59%, which suggests the oil continued to processes that could have facilitated these weathering processes. How-
ever, despite some degradation, the diagnostic ratios of different pairs of
hopanes, steranes, and also triaromatic steranes remained stable in the
DWH tarballs.

CRediT authorship contribution statement

Marieh Arekhi: Conceptualization, Methodology, Investigation,

Writing–original draft, Writing – review & editing, Formal analysis, Vi-
sualization. Leigh G. Terry: Investigation, Writing – review & editing,
Resources, Supervision, Funding acquisition. Gerald F. John: Conceptu-
alization, Methodology, Investigation, Formal analysis, Writing – review
& editing. T. Prabhakar Clement: Conceptualization, Methodology, In-
vestigation, Writing–original draft, Writing – review & editing, Visuali-
zation, Supervision, Resources, Project administration, Funding
acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial

Fig. 8. C30 αβ-hopane concentrations and weathering percentage (WP) levels for the interests or personal relationships that could have appeared to influ-
DWH tarball samples. ence the work reported in this paper.

8
M. Arekhi, L.G. Terry, G.F. John et al. Science of the Total Environment 791 (2021) 148056

Acknowledgment Lin, L., Michael, G., Kovachev, G., Zhu, H., Philp, R., Lewis, C., 1989. Biodegradation of tar-
sand bitumens from the Ardmore and Anadarko Basins, Carter County, Oklahoma.
Org. Geochem. 14 (5), 511–523.
This work was, in part, funded by a field project (GR27012) sup- Mulabagal, V., Yin, F., John, G., Hayworth, J., Clement, T., 2013. Chemical fingerprinting of
ported by the City of Orange Beach, Alabama, and by a research grant petroleum biomarkers in Deepwater Horizon oil spill samples collected from Ala-
bama shoreline. Mar. Pollut. Bull. 70 (1–2), 147–154.
awarded by the National Science Foundation (Award no: 2019561). Munoz, D., Doumenq, P., Jacquot, F., Scherrer, P., Mille, G., 1997. Long term evolution of pe-
TPC and his research team completed multiple field surveys for this troleum biomarkers in mangrove soil (Guadeloupe). Mar. Pollut. Bull. 34 (11), 868–874.
10-year study. We like to thank Mr. Roy Collins at the Alabama Murray, J.A., Murray, J.A., Reddy, C.M., Sander, L.C., Wise, S.A., 2016. Gulf of Mexico Re-
search Initiative 2014/2015 Hydrocarbon Intercalibration Experiment: Description
Department of Environmental Management and Mr. Phillip West at and Results for SRM 2779 Gulf of Mexico Crude Oil and Candidate SRM 2777 Weath-
the City of Orange Beach, Alabama, for helping us with several field sur- ered Gulf of Mexico Crude Oil. US Department of Commerce, National Institute of
veys. The lab experiments were completed by MA as part of her Ph.D. Standards and Technology.
Peake, E. and Hodgson, G.W. 1973. Origin of petroleum–steranes as products of early dia-
dissertation research. TPC, MA, and GFJ jointly conceived the ideas for
genesis in recent marine and freshwater sediments. AAPG Bull. 57(4), 799–799.
this work, MA conducted experimental work, prepared figures. GFJ Peters, K., Moldowan, J., 1991. Effects of source, thermal maturity, and biodegradation on the
helped with methods development efforts, reviewed the experimental distribution and isomerization of homohopanes in petroleum. Org. Geochem. 17 (1),
data. MA and TPC co-wrote the manuscript. LT and GFJ reviewed multi- 47–61.
Peters, K.E., Moldowan, J.M., 1993. The biomarker Guide: Interpreting Molecular Fossils in
ple versions of this manuscript and provided helpful comments. LT and Petroleum and Ancient Sediments.
TPC jointly supervised the work and developed the funding support and Peters, K.E., Moldowan, J.M., McCaffrey, M.A., Fago, F.J., 1996. Selective biodegradation of
laboratory facilities. All four authors read and approved the final manu- extended hopanes to 25-norhopanes in petroleum reservoirs. Insights from molecu-
lar mechanics. Org. Geochem. 24 (8–9), 765–783.
script and their overall responsibilities, estimated using the Clement Powers, S.P., Peterson, C.H., Cebrian, J., Heck Jr., K.L., 2017. Response of nearshore ecosys-
(2014) approach, are MA (50%), LT (15%), GFJ (10%), and TPC (25%). tems to the Deepwater Horizon oil spill. Mar. Ecol. Prog. Ser. 576, 107–110.
We also like to thank STOTEN reviewers for providing several valuable Prince, R.C., Elmendorf, D.L., Lute, J.R., Hsu, C.S., Haith, C.E., Senius, J.D., Dechert, G.J.,
Douglas, G.S., Butler, E.L., 1994. 17. alpha.(H)-21. beta.(H)-hopane as a conserved in-
suggestions. ternal marker for estimating the biodegradation of crude oil. Environ. Sci. Technol. 28
(1), 142–145.
Appendix A. Supplementary data Prince, R.C., Owens, E.H., Sergy, G.A., 2002. Weathering of an Arctic oil spill over 20 years:
the BIOS experiment revisited. Mar. Pollut. Bull. 44 (11), 1236–1242.
Shirneshan, G., Bakhtiari, A.R., Memariani, M., 2016. Distribution and origins of n-alkanes,
Supplementary data to this article can be found online at https://doi. hopanes, and steranes in rivers and marine sediments from Southwest Caspian coast,
org/10.1016/j.scitotenv.2021.148056. Iran: implications for identifying petroleum hydrocarbon inputs. Environ. Sci. Pollut.
Res. 23 (17), 17,484–17,495.
Smith, M. 2013. Coast Guard, BP end Gulf cleanup in 3 states http://www.cnn.com/2013/
References 06/10/us/gulf-oil-spill/ (accessed on Feb 21, 2021). CNN.
Sun, P., Gao, Z., Cao, L., Wang, X., Zhou, Q., Zhao, Y., Li, G., 2011. Application of a step-by-
Aeppli, C., Nelson, R.K., Radovic, J.R., Carmichael, C.A., Valentine, D.L., Reddy, C.M., 2014. step fingerprinting identification method on a spilled oil accident in the Bohai Sea
Recalcitrance and degradation of petroleum biomarkers upon abiotic and biotic nat- area. J. Ocean Univ. China 10 (1), 35–41.
ural weathering of Deepwater Horizon oil. Environ. Sci. Technol. 48 (12), 6726–6734. Venosa, A., Suidan, M., King, D., Wrenn, B., 1997. Use of hopane as a conservative bio-
Arekhi, M., Terry, L.G., John, G.F., Al-Khayat, J.A., Castillo, A.B., Vethamony, P., Clement, T.P., marker for monitoring the bioremediation effectiveness of crude oil contaminating
2020. Field and laboratory investigation of tarmat deposits found on Ras Rakan Island a sandy beach. J. Ind. Microbiol. Biotechnol. 18 (2–3), 131–139.
and northern beaches of Qatar. Sci. Total Environ. 139, 516. Wang, C., Gao, X., Sun, Z., Qin, Z., Yin, X., He, S., 2013. Evaluation of the diagnostic ratios for
Beyer, J., Trannum, H.C., Bakke, T., Hodson, P.V., Collier, T.K., 2016. Environmental effects the identification of spilled oils after biodegradation. Environ. Earth Sci. 68 (4), 917–926.
of the Deepwater Horizon oil spill: a review. Mar. Pollut. Bull. 110 (1), 28–51. Wang, C., Chen, B., Zhang, B., Guo, P., Zhao, M., 2014. Study of weathering effects on the
Bost, F., Frontera-Suau, R., McDonald, T., Peters, K., Morris, P., 2001. Aerobic biodegradation of distribution of aromatic steroid hydrocarbons in crude oils and oil residues. Environ.
hopanes and norhopanes in Venezuelan crude oils. Org. Geochem. 32 (1), 105–114. Sci.: Processes Impacts 16 (10), 2408–2414.
Cheng, X., Hou, D., Xu, C., Wang, F., 2016. Biodegradation of tricyclic terpanes in crude oils Wang, Z., Fingas, M., Li, K., 1994a. Fractionation of a light crude oil and identification and
from the Bohai Bay Basin. Org. Geochem. 101, 11–21. quantitation of aliphatic, aromatic, and biomarker compounds by GC-FID and GC–MS,
Clement, T., John, G. and Yin, F. 2017. Chapter 16—Assessing the Increase in Background part II. J. Chromatogr. Sci. 32 (9), 367–382.
Oil–Contamination Levels Along Alabama's Beaches Resulting From the Deepwater Wang, Z., Fingas, M., Sergy, G., 1994b. Study of 22-year-old Arrow oil samples using bio-
Horizon Oil Spill. Oil Spill Science and Technology (Second Edition). Boston: Elsevier marker compounds by GC/MS. Environ. Sci. Technol. 28 (9), 1733–1746.
Inc., 851–888. Wang, Z., Fingas, M., Sergy, G., 1995. Chemical characterization of crude oil residues from
an arctic beach by GC/MS and GC/FID. Environ. Sci. Technol. 29 (10), 2622–2631.
Clement, T.P., 2014. Authorship matrix: a rational approach to quantify individual contri-
Wang, Z., Fingas, M., Owens, E., Sigouin, L., Brown, C., 2001. Long-term fate and persis-
butions and responsibilities in multi-author scientific articles. Sci. Eng. Ethics 20 (2),
tence of the spilled Metula oil in a marine salt marsh environment: degradation of
345–361.
petroleum biomarkers. J. Chromatogr. A 926 (2), 275–290.
Connan, J., 1984. Biodegradation of crude oils in reservoirs. Adv. Pet. Geochem. 1,
Wardroper, A., Hoffmann, C., Maxwell, J., Barwise, A., Goodwin, N., Park, P., 1984. Crude oil
229–335.
biodegradation under simulated and natural conditions—II. Aromatic steroid hydro-
Crone, T.J. and Tolstoy, M. 2010. Magnitude of the 2010 Gulf of Mexico oil leak. Science carbons. Org. Geochem. 6, 605–617.
330(6004), 634–634. Watson, J., Jones, D., Swannell, R., Van Duin, A., 2002. Formation of carboxylic acids during
Devore, J.L., 2015. Probability and Statistics for Engineering and the Sciences Cengage aerobic biodegradation of crude oil and evidence of microbial oxidation of hopanes.
Learning. Org. Geochem. 33 (10), 1153–1169.
Douglas, G.S., Hardenstine, J.H., Liu, B., Uhler, A.D., 2012. Laboratory and field verification White, H.K., Hsing, P.-Y., Cho, W., Shank, T.M., Cordes, E.E., Quattrini, A.M., Nelson, R.K.,
of a method to estimate the extent of petroleum biodegradation in soil. Environ. Sci. Camilli, R., Demopoulos, A.W., German, C.R., 2012. Impact of the Deepwater Horizon
Technol. 46 (15), 8279–8287. oil spill on a deep-water coral community in the Gulf of Mexico. Proc. Natl. Acad. Sci.
Dubansky, B., Whitehead, A., Miller, J.T., Rice, C.D., Galvez, F., 2013. Multitissue molecular, 109 (50), 20,303–20,308.
genomic, and developmental effects of the Deepwater Horizon oil spill on resident White, H.K., Wang, C.H., Williams, P.L., Findley, D.M., Thurston, A.M., Simister, R.L., Aeppli,
Gulf killifish (Fundulus grandis). Environ. Sci. Technol. 47 (10), 5074–5082. C., Nelson, R.K., Reddy, C.M., 2016. Long-term weathering and continued oxidation of
Frontera-Suau, R., Bost, F.D., McDonald, T.J., Morris, P.J., 2002. Aerobic biodegradation of oil residues from the Deepwater Horizon spill. Mar. Pollut. Bull. 113 (1–2), 380–386.
hopanes and other biomarkers by crude oil-degrading enrichment cultures. Environ. Williams, J., Bjorøy, M., Dolcater, D., Winters, J., 1986. Biodegradation in South Texas Eo-
Sci. Technol. 36 (21), 4585–4592. cene oils—effects on aromatics and biomarkers. Org. Geochem. 10 (1–3), 451–461.
Gustitus, S.A., Clement, T.P., 2017. Formation, fate, and impacts of microscopic and macro- Wu, Y., Xia, Y., Wang, Y., Lei, T., Chang, J., Wang, Y., 2013. Distribution and properties of
scopic oil-sediment residues in nearshore marine environments: a critical review. biomarkers in severely biodegraded crude oil of Gudao reservoir, China. J. Pet. Sci.
Rev. Geophys. 55 (4), 1130–1157. Eng. 103, 97–105.
Han, Y., Clement, T.P., 2018. Development of a field testing protocol for identifying Deep- Yang, C., Wang, Z., Hollebone, B.P., Brown, C.E., Yang, Z., Landriault, M., 2015. Chromato-
water Horizon oil spill residues trapped near Gulf of Mexico beaches. PLoS One 13 graphic fingerprinting analysis of crude oils and petroleum products. Handb. Oil
(1). Spill Sci. Technol. 93–163.
Han, Y., Nambi, I.M., Clement, T.P., 2018. Environmental impacts of the Chennai oil spill Yin, F., Hayworth, J.S., Clement, T.P., 2015a. Correction: a tale of two recent spills—com-
accident–a case study. Sci. Total Environ. 626, 795–806. parison of 2014 Galveston Bay and 2010 Deepwater Horizon oil spill residues. PLoS
One 10 (4), e0124645.
Howell, V., Connan, J., Aldridge, A., 1984. Tentative identification of demethylated tricyclic
Yin, F., John, G.F., Hayworth, J.S., Clement, T.P., 2015b. Long-term monitoring data to de-
terpanes in nonbiodegraded and slightly biodegraded crude oils from the Los Llanos
Basin, Colombia. Org. Geochem. 6, 83–92. scribe the fate of polycyclic aromatic hydrocarbons in Deepwater Horizon oil sub-
merged off Alabama's beaches. Sci. Total Environ. 508, 46–56.
John, G.F., Han, Y., Clement, T.P., 2018. Fate of hopane biomarkers during in-situ burning
of crude oil—a laboratory-scale study. Mar. Pollut. Bull. 133, 756–761.