Insights about levels and source appointment of petroleum hydrocarbons in Brazilian semi-arid coastal: baseline status assessment for ocean decade targets

The Ocean decade (2021–2030) for sustainable development proclaimed in 2017 by the UN, seeks to promote and conserve the sustainable use of oceans, seas, and marine resources. For this, the distribution of n-alkanes, polycyclic aromatic hydrocarbons (PAHs), and biomarkers, in sediments from the Fortaleza coastal zone (Mucuripe harbor (MH) and Inner Continental shelf (ICS)) were used to assess the impacts of anthropogenic activities in the area. The concentrations of total n-alkanes (Σ16 n-alkanes) in MH and ICS sediments varied from 35.9 to 94.9 and 17.9 to 197.3 μg g−1, respectively, while the isoprenoids phytane and pristane in MH and ICS sediments ranged from 0.1 to 1.69 ug g−1 and from 0.14 and 1.20 μg g−1, respectively. Most of the sediment samples presented carbon preference index (CPI) values close to unity, indicating that the area is submitted to petroleum-related sources. The concentrations of Σ16 PAHs in MH and ICS sediments varied from 87.0 to 562.0 and 98 to 288.0 ng g−1. This work presents the first investigation of the petroleum biomarkers hopanes and steranes in the Fortaleza coastal zone, in which ΣBiomarkers varied from 0.10 to 1.79 and 0.02 to 0.24 ug g−1 in MH and ICS sediments, respectively. The presence at stations of biomarkers also indicates petrogenic input. The diagnosis of the distribution of pollutants in the investigated zones of the Fortaleza coast suggests contamination from urban areas and oil spills and vessel traffic.


Introduction
Oceans and coastal environments have received extensive human pressure for decades, compromising the environmental health of these ecosystems (Harden-Davies et al. 2022). At the end of 2017, the Decade of Ocean Science for Sustainable Development (2021Development ( -2030 was proclaimed by the United Nations to generate knowledge about ocean sciences and support countries in sustainably managing the marine environment. One of the challenges of the ocean decade will be to understand and characterize terrestrial and marine sources of pollutants and contaminants, as well as their impacts on human health and marine ecosystems (Kocak and Hotaling 2021).
Coastal environments (e.g., coastlines and estuaries) are considered large environmental reservoirs due to the input of organic matter (OM) of natural and anthropogenic origin, as well as the biogeochemical processes acting in these places (Cao et al. 2020). The OM deposition into systems is governed by many processes, including fluvial discharge, urban runoff, and atmospheric deposition (Liu et al. 2012). The Brazilian Northeast region is characterized such as oligotrophic and benthic environments, evidenced by sandy deposits, rich in carbonates of biogenic origin, and low contents of organic matter (Knoppers et al. 1999). These features may result in distinct conditions of sediment contamination (Moreira et al. 2021).
Fortaleza is the fifth largest capital of Brazil with more than 2.3 million inhabitants spread over an area of 313 km 2 . Its coastline is 34 km long, with a total of 15 main beaches. Its limits are the mouth of the Ceará River, to the north, and the Pacoti River, to the south. Fortaleza is home to three industrial parks in the steel, food, and refinery petrochemical sectors. Due to its privileged location concerning access to the USA and the European continent, the region receives major anthropogenic impacts from these activities (Moreira et al. 2017a, b).
Given the development of capitalist relations, harbor zones have played an important role as a point of intersection between built and natural environments, in which about 90% of all commercial products pass since half of the world's population lives 100 km from the coast (O'Neil et al. 2020). With the intense traffic in Harbor regions, the presence of contaminants tends to increase, bringing damage to the ecosystem.
Aliphatic hydrocarbons (AH) are compounds used in the coastal environment with various biogenic sources, originating from terrestrial sources (e.g., plant waxes) or algae, zooplankton, phytoplankton, bacteria, and other biogenic precursors (Volkman et al. 1992;Garcia et al. 2019). They are considered an important geochemical tool to assess OM sources and pollution levels in coastal systems (Tarozo et al. 2010). In addition, the hopanes and steranes are recalcitrant aliphatic hydrocarbons commonly found in crude oil, being systemic approached as petroleum biomarkers coming from prokaryotes and eukaryotic cells (Peters et al. 2004a).
PAH are widespread contaminants present throughout the marine environment and they also have been used to assess oil or pyrogenic contamination in marine sediments. Soares-Gomes et al. (2010) attributed toxic effects in biota to high PAH levels in Guanabara Bay (Brazil). Moreover, the news neoplasias occurrences in fishes associated with PAH levels in local sediment showed that there is a strong causal link between these factors (Hoffman et al. 2002;Beyer et al. 2010).
Some authors have already reported the presence of PAHs in sediments (Cavalcante et al. 2009;Moreira et al. 2017aMoreira et al. , b, 2021, rainwater (Cavalcante et al. 2007), particulate matter (Rocha et al. 2020), clams (Moreira et al. 2020a) in estuarine and harbor zones of the Fortaleza City, which may originate from runoff of surface water, discharge domestic, industrial sewage, or petroleum sources (Benson et al. 2020;Han et al. 2021). However, detailed information on petroleum and aliphatic hydrocarbons, mainly about the distribution of the hopanes and steranes in the Inner Continental Shelf adjacent remains scarcely documented.
Petroleum biomarkers, such as hopanes and steranes, have been found in surface sediments, indicating contamination of petroleum-derived products in several environments (Garcia et al. 2019;de Melo et al. 2020). Diagnostic ratios based on the distribution of these compounds have been applied to investigate the source of these biomarkers, mostly based on specific stereochemical changes to assess evidence of petrogenic pollution (Tarozo et al. 2010). For instance, values higher than 1.0 of the C30 αβ/ββ ratio and ranging from 0.51 to 0.66 for the C31 S/(R + S) ratio indicates the presence of petroleum-derived products in the environment (de Melo et al. 2020). The explanation is that during petroleum formation C30 17β(H)21β(H) hopane is diagenetically transformed in C30 17α(H)21β(H)-hopane, which is more stable, and the epimers C31 22R are converted in the 22S (Bouloubassi and Saliot 1993;da Silva and Bícego 2010).
In this context, the objective of this study was to investigate for the first time the levels and the distribution of hopanes and steranes, as well as n-alkanes, isoprenoids, and PAHs along an Urban Harbor Zone (Mucuripe Harbor) and Inner Continental Shelf (adjacent in Fortaleza coastal zone). The main goals of the study were (1) to characterize the natural and anthropogenic sources of hydrocarbons inputs in surface sediments from Fortaleza coastal zone; (2) to assess sedimentary diagenesis of pentacyclic triterpenoids.
The main purpose of this study is to monitor levels of different classes of hydrocarbons, with special attention to petroleum biomarkers, and investigate for the first time, along an Urban Harbor Zone (Mucuripe Harbor) and Inner Continental Shelf (adjacent to Fortaleza coastal zone). The sites are located in Fortaleza city, locally affected by continuous anthropogenic pressure. Therefore, based on an integrated approach of multivariate statistics, classes of hydrocarbons, and diagnostic ratios, we hypothesized that chronic hydrocarbon contamination may affect the Fortaleza coastal zone in distinct ways.

Study area and sample collection
Surface sediment samples (0-2 cm) were collected in an oceanographic mission in 2013 along Fortaleza city coastal zone, state of Ceará, Brazil: at six stations in the Mucuripe Harbor (MH1 to MH6); and 14 stations in the Inner Continental Shelf (ICS1to ICS14). The samples were stored in aluminum containers and conditioned at low temperatures (< 4° C) until laboratory analyses.

Analytical methods
The F1 and F2 fractions were analyzed by GC-FID (Agilent 6890) configured with a split/splitless auto-injector (7683B series) with a capillary Restek Rxi-1 ms (30 m length, 0.25 mm I.D., 0.25 μm df). Hydrogen was used as the carrier gas. For F1, the following oven temperature program was employed: 70 °C for 7 min, 70-320 °C at a ramp rate of 6 °C min −1 , and 320 °C for 15 min (Aeppli et al. 2012). The GC-MS (Agilent) was configured with a split/splitless auto-injector (7683B series) equipped with GC Real-Time Analysis software for confirmation of analytes. A capillary column of fused silica DB-5 (60 m × 0.25 mm id) (0.25 μm film) was used. The temperature program was followed: 50 °C for 1 min; 50-115 °C at a ramp 20 °C min −1 ; 115-320 °C min −1 at a ramp 5 °C min −1 to reach the temperature of 320 °C for 20 min. Helium was used as the carrier gas at a flow of 1.3 mL min −1 . The injector temperature was 300 °C, and the injection volume was 2 μL. The temperature of the interface and ion source was 300 °C.
All injection mode used was splitless. The compounds were identified according to their elution time on the chromatographic column and comparison spectral library. All data were subject to rigorous quality control procedures. Analysis of the reagent blanks demonstrated that the analytical system used (glassware, solvents, and materials) was free of contamination.
The analytes were quantified using a calibration curve with seven points (injected in triplicate) by the internal standard method. The calibration curve showed that the linearity varied from 0.52 to 37.00 µg L −1 and R 2 ranged from 0.9956 to 0.9999 for n-alkanes; pristane and phytane ranged from 0.51 to 33.04 µg L −1 and R 2 ranged from 0.9979 to 0.9986; biomarkers ranged from 0.01 to 8.49 µg L −1 and R 2 ranged from 0.7953 to 0.9999; PAHs ranged from 0.37 to 33.46 µg L −1 and R 2 ranged from 0.9984 to 0.9999.
The limit of detection (LOD) and limit of quantification (LOQ) were calculated as the concentration of analyzed blank injections (n = 7) that gave rise to a peak with a signalto-noise ratio (S/N) of 3.3 and 10, respectively. The Determination Coefficient, LOD, and LOQ are shown in Table 4 for n-alkanes, PAHs, hopanes, and steranes respectively.
All the obtained results were subject to rigorous quality control procedures. Blank analysis indicated that the analytical system used (glassware, solvents, and materials) was contamination free. The extraction efficiency was assessed through the deuterated surrogate recovery for each markers class (Cavalcante et al. 2008(Cavalcante et al. , 2009Aeppli et al. 2012). The surrogates recovery are displayed in the Supplementary information. All the statistical analyses were performed using the software Statistica v 12.0.

Distribution and sources of alkanes and isoprenoids
Concentrations in unpolluted intertidal and estuarine sediments may range from sub-ppm to 10 µg g −1 . Concentrations maybe two to three times higher where there are significant inputs of n-alkanes derived from higher plants. Organicrich marine sediments may contain up to 100 µg g −1 total aliphatic hydrocarbons, but concentrations higher than this are usually due to petroleum inputs (Volkman et al. 1992;Readman et al. 2002). Table 5 displays the total concentrations of n-alkanes ranging from C10 to C38 and the concentration of the isoprenoids pristane and phytane, in addition to diagnostic ratios to identify the natural and anthropogenic origins of sediment based on the aliphatic hydrocarbons: carbon preference index (CPI); terrigenous/aquatic ratio (TAR); ∑odd/∑Even ratios; and Pri/Phy ratio. The total concentration of n-alkanes (∑n-alkanes) ranges from 17.9 and 197.3 µg g −1 , which are comparable to areas affected by intense urbanization process, acute harbor activities, and the disposal of industrial and domestic effluents, such as on the west coast Peninsular Malaysia (27.9-254.4 μg g −1 ; Vaezzadeh et al. 2015), in the Santos Estuary (3.91-114.l μg g −1 ; Bícego et al. 2006) and the majority of the southeast Brazilian harbors (22.9-113 μg g −1 , Pinheiro et al. 2017). Furthermore, the concentrations found in this study were two to three times higher than the values reported by Moreira et al. (2017a, b) inside the Mucuripe Harbor (0.3-25.7 μg g −1 ).
The distribution observed for the resolved n-alkanes in the samples from the study areas evidenced a uni-modal pattern with a predominance of short-chain n-alkanes (< nC20).  This distribution occurred in the coastal areas and followed toward the ocean and near Mucuripe Harbor stations (e.g., MH2, ICS11, and MH4; Figs. 1, and 2) The n-alkanes also presented a regular distribution with an equivalent distribution pattern of odd and even-carbon numbered in the majority of stations, especially for short-chain (< nC23) n-alkanes, as showed by ∑odd/∑Even ratios, with stations presenting higher amounts of even n-alkanes. Alkanes with petrogenic provenance generated through high-temperature processes are depleted in odd carbon atoms dominating with C15 to C25 n-alkanes (Amijaya et al. 2006;Peters et al. 2004b). These results and the presence of uni-modal distributions, varying from nC11 to nC35, are characteristics of petroleum origin. The concentrations obtained in this study can be also related to the shipwreck of the Seawind ship that occurred in June 2012, where it released 85,000 L of bunker oil at the coastline. A Bi-modal distribution is observed for certain oil products, which could also be due to the superposition of two distinct types of petroleum product sources, as well as the presence of fresh and degraded oil (Yang et al. 2009;Ranjbar Jafarabadi et al. 2017).
The considerable concentrations of odd-numbered longchain n-alkanes (especially nC27, nC29, and nC31), were observed in stations inside Mucuripe Harbor and Shelf stations under its riverine influence (MH5, MH6, ICS1, ICS2, ICSC3, and ICS4), indicates an allochthonous source characteristic of biogenic contribution, such as terrigenous input derived from higher plant waxes (Volkman et al. 1992). The presence of odd-numbered short-chain n-alkanes (nC15, nC17, and nC19) as the major hydrocarbon in Shelf stations suggests an association with marine and autochthonous contributions such as planktonic and algae sources (Meyers 2003;Commendatore and Esteves 2004;Gao and Chen 2008). Despite the indications of petroleum pollution, the influence of marine bacterial sources with even carbon numbered predominance, no noticeable even/odd carbon preference, or degraded n-alkanes from terrestrial and biogenic sources cannot be ruled out (Nishimura and Baker 1986;Grimalt and Albaigés 1987;Hu et al. 2009;de Almeida et al. 2018).
The carbon preference index (CPI), a measure of biologically synthesized n-alkanes, indicates the relative contributions of natural (biogenic/terrestrial; CPI > 3) and anthropogenic (petroleum pollution; CPI ~ 1) sources Aboul-Kassim and Simoneit 1996). The CPI values ranged from 0.68 to 4.47 with a mean value of 1.7 ± 1.2 (Table 5). Stations located near the Mucuripe Harbor and the Shelf showed CPI ≈ 1 indicating the presence of petrogenic sources, except the stations MH5, which are under the influence of the Maceio stream, and the petroleum signature may be diluted by the biogenic terrestrial input. The terrigenous/ aquatic ratio (TAR) ranged from 0.02 to 29.0. As expected, the estuarine and adjacent stations presented higher values, as well as the station MH5, confirming the terrestrial organic matter contributions to these areas. The isoprenoids pristane (Pri) and phytane (Phy) are mostly products of the geologic processing of phytol and are not primary constituents of most terrestrial biota (Didyk et al. 1978;Peters & Moldowan 1993). Both Pri and Phy were found in all stations and ranged from 0.10 to 1.69 and 0.14 to 1.20 μg g −1 (dry weight) for the Mucuripe Harbor and Inner Continental Shelf, respectively ( Table 5). The ratio Pri/Phy ranged from 0.60 to 2.08, with a mean value of 1.18 ± 0.28, indicating an input of petrogenic sources at most of the stations (Pri/Phy ~ 1) (Colombo et al. 2002). The Pri/Phy ratio values may also be influenced by reducing environment formation contributing to the formation of Phy in estuarine areas and transported to coastal areas (Cripps 1989), although, the other calculated indices and the chain length agree with the presence of petroleum in the stations.
The ∑16PAHs values registered in Mucuripe Harbor were compared to other areas, and their ∑16PAHs values were lower than Beirut Harbor (3772.6 ng g −1 ; Merhaby et al. 2020) and Santos estuary (Bícego et al. 2006). However, it is interesting to note that when compared to prior studies in the same area done in 2007 and 2011, ∑16PAHs  Fig. 2 The typical distribution patterns of resolved n-alkanes by carbon number in the samples of different region in study area; MH2, MH4 and ICS11 representing stations values have increased (Buruaem et al. 2016;Moreira et al. 2020b). Fortaleza inner shelf sediment samples have similar levels of PAHs to Todos os Santos Bay (de Almeida et al. 2018), Brazilian south and southeast continental margin (Santos et al. 2020), and Jiaozhou Bay, China (Cao et al. 2020) (Table 6).
Low molecular weight PAHs (2-3 rings; LMW) are related to petrogenic inputs (e.g., fossil fuel leaking, natural seep, and oil spill) and are usually discharged straight to a water body where they can associate with particulate matter and sink, while high molecular weight PAHs (4-6 rings; HMW) are related to pyrolytic sources of PAHs (e.g., vehicular emissions, biomass burning, natural fires, and open-air burning of waste) and go first into the atmosphere, before getting to water body through atmospheric deposition, then to sediments (Tobiszewski and Namieśnik 2012). Analysis of 16 USEPA PAHs distribution showed a 4-6-ring PAHs predominance over stations (Fig. 4), indicating pyrolytic sources as primary PAHs input into the Fortaleza coastal sediments. Additionally, it is interesting to note a greater contribution of 5 rings of PAH at stations MH2, MH3, MH4, ICS8, ICS9, and ICS14, which slightly changes the distribution pattern of PAHs. This can be explained by Dibenzo[a,h]anthracene (DBah) concentrations, which represent up to 82.6 ± 11.3% of 5-ring PAHs in these stations.
The diagnostic ratios LMW/HMW, BaA/ΣPAH228, Ant/ΣPAH178, Fl/ ΣPAH202, Ind/ΣPAH276, and %Per/ Σ5rings (Fig. 6) were applied to estimate PAHs sources in the Fortaleza coastal zone and differentiate not only between pyrolytic and petrogenic source but also among different combustion sources (Yunker et al. 2002;Tobiszewski and Namieśnik 2012). The LMW/HMW is an index related to the pyrogenic-petrogenic differentiation and divides the 16 USEPA PAHs into the sum of 2-3 rings which are called low molecular weight (LMW) to the sum of 4-6 rings which are called high molecular weight (HMW). LMW/HMW values < 1 indicate pyrolytic pollution and values > 1 indicate petrogenic contamination (Lima et al. 2019). The LMW/ HMW values ranged from 0.02 to 0.57, with a mean value of 0.13 ± 0.1. This result shows a predominance of pyrolytic over petrogenic sources in all stations in the Fortaleza coastal zone. There was no significant difference among the four regions studied (p > 0.05, ANOVA). The low ratios were obtained due to high concentrations of 5-6-ring PAHs in these points, as discussed above, which can be a result of pyrolytic processes in the area. Nevertheless, HMW PAHs are usually associated with small soot-rich particles (Yunker et al. 2002) due to their properties. As a result, HMW PAHs are more often associated with sediments and become more resistant to microbial degradation than LMW PAHs (De Luca et al. 2005;Santos et al. 2020).
Usually, the 0.4 < Fl/ΣPAH202 < 0.5 are assumed to be from fossil fuel combustion, while Fl/ΣPAH202 > 0.5 are from grass, wood or coal combustion. In addition, Ant/ΣPAH178 > 0.1, Ind/ΣPAH276 > 0.2, and BaA/ ΣPAH228 > 0.35 ratio values are also considered to be from combustion sources. In accordance with LMW/HMW ratio results, the cross plots of ratios BaA/ΣPAH228, Ant/ ΣPAH178, and Ind/ΣPAH276 vs Fl/ΣPAH202 showed a distinct pattern of combustion source along the study area, with a contribution of fossil fuel and biomass burning (Fig. 5).
The herein studied stations are located near the large center of Fortaleza, which has a fleet of approximately one million vehicles (IBGE 2020). A similar scenario was observed by de Almeida et al. (2018) in sediment samples from Todos os Santos Bay, Brazil, and PAHs source was mainly attributed to both vehicles and ships' activities in the area. Previous studies in Fortaleza attribute the levels of PAHs detected to vehicle emissions as well as to burning wood as an energy source in commercial activities (Cavalcante et al. 2007). Most pyrolytic hydrocarbons are transported to marine sediment through the atmosphere (Peters et al. 2010). Another way is the road runoff that may contribute with asphalt, motor oil, gasoline, and exhaust (Peters et al. 2010). Thus, anthropogenic activities (e.g., vehicular emissions, urban runoff, and shipping activities) are some of the main factors responsible for the high levels of HMW PAHs detected in the Mucuripe Harbor and inner shelf sediments of Fortaleza-CE.
Perylene origin in superficial sediments of Fortaleza was also investigated. The index used consists of the relative abundance (%) of perylene over five ringed PAH (BkF, BbF, BeP, BaP, and DBA) and is useful to distinguish between natural or anthropogenic PAHs (Wang et al. 2014;Lima et al. 2019). This %Per/∑5ringsPAHs ratio ranged from 7.6 to 84.7% along the Fortaleza coastal zone (Fig. 6), with most stations greater than 10%, suggesting a predominance of diagenetic input in these locations (Wang et al. 2014;Lima et al. 2019). Some stations (MH3, MH4, ICS8, and ICS14) showed %Per/∑5ringsPAHs < 10% which suggests a predominance of pyrolytic natural sources of PAHs.   (Table 7). High individual biomarker concentrations were observed in the Inner Continental Shelf (ICS), showing C31 17α(H),21β(H)-22R-homohopane levels of 0.071 µg g −1 in the ICS4 station. Notably higher concentrations of the sum of hopanes were observed in sediments from the Mucuripe Harbor (0.06 to 1.48 µg g −1 ; mean = 0.68 µg g −1 ) than from the Inner Continental Shelf (ND to 0.18 µg g −1 ; mean = 0.04 µg g −1 ). The same behavior was observed for the steranes, with the sum of them ranging from 0.04 to 0.31 µg g −1 (mean = 0.17 µg g −1 ) in sediments from the Mucuripe Harbor and ranging from ND to 0.06 µg g −1 (mean = 0.02 µg g −1 ) in sediments from Inner Continental Shelf (Table 7). The higher concentration of the petrogenic compounds hopanes and steranes in the Mucuripe Harbor is consistent with ship traffic in the area, which can release oils and fuels into the environment, as occurred during the shipwreck of the Seawind ship at the Mucuripe Harbor in June 2012.
C30 17α(H),21β(H)-hopane was detected in the whole Mucuripe Harbor area with the maximum value observed in the MH6 station (0.56 µg g −1 ) (Supplementary information). This point is near Maceio stream discharge, responsible for chronic inputs (urban runoff) to the area (Fig. 1). The C30 17α(H),21β(H)-hopane is commonly attributed in geochemical forensics studies as sourced from crude oil due to its high thermodynamic stability (Peters et al. 2004a;Stout & Wang 2016  to the environment can be through diagenetic processes in acidic environments, such as peatlands (Inglis et al. 2018). Some diagnostic ratios based on the distribution of hopanes and steranes were also calculated for all sediment samples (Table 7). The C29 αββ/(αββ + ααα) ratio shows very low values for the samples, varying from ND to 0.12, presenting higher values for samples from Mucuripe Harbor (ND to 0.12) than for samples from Inner Continental Shelf (ND to 0.03). In addition, the C31 S/(R + S) ratio shows values varying from ND to 0.70 in the MH zone and from ND to 7.39 in the ICS zone. These results could indicate a distinct source of these biomarkers or some preferential biodegradation in environmental conditions of the C29 ααα epimers in the MH zone since the ααα isomers are removed faster than the βββ isomers (Bayona and Albaigés 2006). According to Melo et al. (2020), values for this ratio between 0.51 and 0.66 would confirm the presence of products derived from petroleum. Following these lines, the values of the C31 S/(R + S) ratio for most of the samples from the MH zone are in this range or near, which corroborates the petrogenic origin of the hopanes detected in this area, a then petroleum input in these sediments.

Conclusions
This study provides data on the pollutants concentrations in the surface sediments from Mucuripe Harbor (MH) and Inner Continental Shelf adjacent (ICS) in the Fortaleza coastal zone. The high total concentration of n-alkanes and the isoprenoids pristane and phytane indicate that the area is affected by the intense urbanization processes of Fortaleza, the fifth largest city in Brazil. Analysis of 16 USEPA PAHs distribution showed a 4-6-ring PAHs predominance over stations, indicating pyrolytic sources as main PAHs input into the Fortaleza coastal sediments, classified as low to moderate contamination. In addition, the presence of the petroleum biomarkers hopanes and steranes, widely used as molecular markers for oil pollution, in addition to some biomarker diagnostic ratios, indicated also petrogenic inputs in the coastal studied zones, mostly at the Mucuripe Harbor, consistent with the ship traffic in the area. This research was significant to highlight the multiple sources of contamination on the coast of Fortaleza, which encompasses touristic areas and marine ecosystems.