Phototoxic effects of two common marine fuels on the settlement success of the coral Acropora tenuis

Phototoxic effects of two common marine fuels on the settlement success of the coral Acropora tenuis www.nature.com/scientificreports OPEN Phototoxic effects of two common marine fuels on the settlement success of the coral Acropora tenuis Received: 12 December 2017 1,2 1 1 3 F. Mikaela Nordborg , Florita Flores , Diane L. Brinkman , Susana Agustí & 1,4 Accepted: 23 May 2018 Andrew P. Negri Published: xx xx xxxx Coral reefs are at risk of exposure to petroleum hydrocarbons from shipping spills and uncontrolled discharges during extraction. The toxicity of petroleum hydrocarbons can substantially increase in the presence of ultraviolet radiation (UVR), therefore spills in shallow coral reef environments may be particularly hazardous to reef species. Here we investigated the sensitivity of coral larvae (Acropora tenuis) to dissolved hydrocarbons from heavy fuel oil (HFO) and diesel in the absence and presence of UVR. Larval settlement success decreased with increasing concentrations of dissolved HFO, and co- exposure to UVR doubled the toxicity: 50% effect concentrations (EC ) decreased from 96 (−UVR) to 51 (+UVR) total petroleum aromatic hydrocarbons (TPAH). Toxic thresholds for HFO were similar −1 to concentrations reported during marine spills: EC s of 24 (−UVR) and 15 (+UVR) µg l . While less toxic, diesel also reduced settlement and exhibited phototoxicity: EC s of 122 (+UVR) and 302 (−UVR) −1 µg l . This study demonstrates that the presence of UVR increases the hazard posed by oil pollution to tropical, shallow-water coral reefs. Further research on the effects of oils in the presence of UVR is needed to improve the environmental relevance of risk assessments and ensure appropriate protection for shallow reef environments against oil pollution. Petroleum hydrocarbons in marine environments. Petroleum hydrocarbons are considered among the most prominent pollution threats to marine environments ; however, the risks they pose to coral reef ecosys- 2,3 tems remain poorly understood . The environmental ee ff cts resulting from oil spills and uncontrolled discharges from extraction vary widely and are dependent on a large number of factors . Hydrocarbon concentrations in 5–7 marine environments have been measured following large scale spills , with dissolved hydrocarbon concentra- −1 5,8,9 −1 tions ranging between 22 and 189 µg l total polycyclic aromatic hydrocarbons (PAH) and up to 10,600 µg l total recoverable hydrocarbons (TRH) . Hydrocarbons can be retained within an ecosystem during spill events and, despite natural dilution and degradation, can remain detectable long aer t ft he spill has ended . e t Th oxicity of petroleum hydrocarbons to most marine species is predominantly related to the water soluble components, which largely consist of the monoaromatic hydrocarbons (MAH) and PAHs . PAHs, in particular, 12,13 are considered acutely toxic to aquatic biota , with toxic threshold concentrations for PAHs often orders of magnitude lower than those of MAHs . However, the overall toxicity of dissolved petroleum hydrocarbon from spills is also dependent on the relative concentrations of each component . Aromatic hydrocarbons in petroleum oils are classified as type I narcotic chemicals . Assuming the same mode of toxic action, the total toxicity of the water soluble components of an oil can be predicted using the narcotic target lipid model (NTLM), which sums the expected toxicity and concentration of each aromatic component (see Methods). Phototoxicity of PAHs. The toxicity of the dissolved aromatic mixtures that result from oil spills may increase in the presence of UVR due to the phototoxic contribution of some PAHs . PAH phototoxicity occurs through the formation of radical oxygen species and/or transformation of PAHs into more toxic photoprod- 12,13 16 ucts . Co-exposure of PAHs and UVR can increase the toxicity of individual PAHs 1000-fold ; however, the 1 2 Australian Institute of Marine Science, Townsville, 4810, Queensland, Australia. Department of Marine Sciences, University of Gothenburg, Göteborg, 40530, Sweden. Red Sea Research Centre, King Abdullah University of Science and Technology, Biological Environmental Science and Engineering Division, Thuwal, 23955, Saudi Arabia. AIMS@ JCU, Division of Research & Innovation, James Cook University and Australian Institute of Marine Science, Townsville, 4810, Queensland, Australia. Correspondence and requests for materials should be addressed to F.M.N. (email: m.nordborg@aims.gov.au) ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 1 www.nature.com/scientificreports/ potential for phototoxicity depends on the compound’s stability, radiation absorbance properties , the type of 12,17 17 UVR exposure and the exposed organism’s sensitivity . The potential for harmful effects to marine organisms is higher for UVA (320–400 nm) than UVB (280–320 nm) as the absorption maxima (hence, photoactivation) of 12,18 19 many PAHs fall within the UVA range , and UVB is more strongly attenuated in seawater . The penetration of UVR in marine environments is also dependent on a range of physical and biological factors , and tropical oligotrophic coral reefs may be at a particularly high risk of PAH phototoxicity as reef organisms are frequently 13,18 exposed to high solar radiation, including UVR . PAH phototoxicity is not always taken into consideration for risk assessments and management unless the ecological relevance, including UVR exposure and spectral profiles, 13,20,21 in the ecosystem have been characterised . However, it is increasingly recognised that a considerable propor- 18,22–24 tion of UVR penetrates to ecologically relevant depths in some marine environments, including coral reefs and that exposure to cumulative pressures, such as pollution and UVR, can result in increased environmental 13,18,25 impacts . Petroleum hydrocarbon and PAH toxicity to corals. Despite a renewed demand for marine hydrocar- bon toxicity research following the Deepwater Horizon spill , signic fi ant knowledge gaps on the potential ee ff cts 18,27–29 of hydrocarbon exposure to corals remain and the sensitivity of tropical marine species to hydrocarbons is 14,30 relatively understudied . Investigations into the effects of petroleum hydrocarbons, including PAHs, on coral −1 31 indicate that negative impacts can occur at concentrations as low as 2–20 µ g l total hydrocarbons (THC) . However, inconsistencies in exposure methodologies, toxic endpoints and reported toxicity values make compar- 2,15,18,32 isons between studies problematic . This issue is further compounded by the failure of many studies to pres - ent the chemical composition of treatment solutions, in particular the more soluble and toxic MAHs and PAHs . However, the studies that have been conducted show that hydrocarbons can be toxic to all life history stages of 33 7,33,34 coral and that larval settlement is generally more sensitive than fertilization , larval survival or the health and survival of adult corals . The sensitive larval settlement process is recognised as an ecologically relevant end- 29,36 point due to its importance in the recruitment process and subsequent maintenance of adult populations . The larval life stages of aquatic animals may also be at higher risk than adults to phototoxic effects due to their small size, oen t ft ransparent bodies, and time spent in shallow waters . Four studies have investigated phototoxic effects of petroleum hydrocarbons on corals and each indicated that their sensitivity to dissolved aromatics may increase with co-exposure to UVR (summarised in Table 1). However, the majority of laboratory studies exposing corals to hydrocarbons have not included co-exposure to UVR , so the impacts of hydrocarbon pollution on coral reefs may be significantly underestimated in the context of likely high UVR exposure in situ. To assess the potential for UVR to increase the sensitivity of coral larvae to spills of heavy fuel oil (HFO) and diesel we: (i) assessed UVR irradiance on one inshore and one mid-shelf reef on the Great Barrier Reef (GBR; Australia); (ii) characterised the chemical composition of the two fuels and their water accommodated fractions (WAFs); and (iii) predicted their narcotic toxicity to marine species using the NTLM. We then (iv) exposed larvae of the reef building coral Acropora tenuis (Dana, 1846) to HFO and diesel WAFs in the absence and presence of UVR (±UVR), at intensities similar to those encountered on the GBR, and assessed the ability of exposed larvae to successfully complete settlement and metamorphosis into sessile polyps following each treatment. Results Chemical analysis. Neat HFO consisted primarily of higher molecular weight hydrocarbons while neat diesel contained a higher proportion of BTEX (benzene, toluene, ethylbenzene and xylene) and other lower molecular weight hydrocarbons (Figs S-1 and S-2, Supplementary information). ∑PAH in HFO (constituting 99.6% of TPAH) was almost 10-fold higher than in diesel (79% of TPAH), whereas ∑BTEX was ~8-fold lower in HFO compared to diesel (see summary in Table 2 and detailed results in Table S-1, Supplementary informa- tion). Phenols were below the limit of quantitation in both oils (Table S-1, Supplementary information). Freshly −1 prepared 100% HFO and diesel WAFs contained similar TPAH concentrations (930 and 913 µg l , respectively), but the proportions of ∑BTEX and ∑PAH varied (Table S-2, Supplementary information). While HFO WAF contained almost equal concentrations of ∑BTEX and ∑PAH (52% and 48% of TPAH, respectively), ∑BTEX dominated the diesel WAF (98% of TPAH). The most abundant PAHs in the WAFs were naphthalene, alkylnaph- thalenes, fluorene and phenanthrene; the HFO WAF also contained acenaphthene and dibenzothiophene (Table S-2, Supplementary information). The concentration of TPAH in fuel WAFs decreased by up to 34% over the 48 h exposure (Tables 2 and S-2, Supplementary information). Comparisons of the observed and predicted concentra- tions of 1-, 2- and 3-ring compounds in the freshly prepared, undiluted WAFs (Tables S-2 and S-3, Supplementary information) as per Redman et al. indicated that no oil droplets were present in either HFO or diesel WAFs. Ultraviolet radiation intensities. e Th spectral profiles for both the Trunk Reef (mid-shelf ) and Esk Reef (inshore) sites showed that the largest decrease in irradiance occurred when light passed from air to water (Fig. 1a and b). Attenuation of UVR at Esk Reef was somewhat lower than at Trunk Reef for shallow measurement depths, despite the higher turbidity (0.1 and 0.8 nephelometric turbidity units, NTU, respectively). However, irradiance was attenuated less with depth at Trunk Reef and the irradiance for the deepest measurements was much higher at Trunk Reef than at Esk Reef (Fig. 1a and b). The depth which 10% of UVR irradiance penetrated to (Z10%) for 305 and 340 nm at Trunk Reef was ~6 and 7 m, respectively. On Esk Reef, Z10% was ~2.7 and 2.4 m for 305 and 340 nm, respectively. Experimental lights, when positioned 170 mm from the sensor, emitted UVA radiation similar to the irradi- ance observed at 1 m depth on Trunk Reef and Esk Reef (Fig. 1c). Average total UVA and UVB (280–400 nm) −2 radiation was 0.9 mW cm (SE = 0.16). Attenuation by the glass scintillation vials used in the experiment reduced the average absolute irradiance by ~17% for wavelengths between 300–400 nm (Fig. 1c; measured using ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 2 www.nature.com/scientificreports/ Chemical Phototoxic −1 Species Hydrocarbon UVR source analysis Endpoint Toxicity values (µg l ) effects Reference LC : 44 (−UVR) and 18 (+UVR) Anthracene Survival Yes: Anthracene Artificial UVR −1 µg l (A); no effect (±UVR) (P) (A), −2 Acropora tenuis 0.68 mW cm Overmans et Phenanthrene PAH (GC-MS) EC : 45 (−UVR) and 6.3 (+UVR) (larvae)* (10 h per 24 h, co- 50 al. (P) (48 h −1 Metamorphosis µg l (A); 91 (−UVR) and 66 No: Phenathrene exposure) −1 exposure) (+UVR) µg l (P) Australian North Ambient solar Acropora tenuis West Shelf BTEX, PAH, IC : 339 (−UVR) and 132 (+UVR) −2 50 29 4.5–6.8 mW cm Metamorphosis Yes Negri et al. −1 (larvae) condensate TRH (GC-MS) µg TPAH l (2 h, co-exposure) (24 h exposure) Carmen Fluoranthene Ambient solar LC : 435.2 (−UVR; lower side of Porites divaricata Mortality/ Guzmán −1 (4.5 h (ambient L:D cycle Nil branches) and 31.4 µg l (+UVR; Yes (adult) Bleaching Martínez et exposure) for 6 d, seq. exposure) upper side of branches) al. Fungia scutaria F. scutaria (+UVR): 100% mortality Yes: −1 (larvae) ≤1 h aer exp ft osure to 48 µg l . F. scutaria Mortality M. verrucosa Montipora M. verrucosa (+UVR): bleached −1 P. damicornis verrucosa (adult) 24 h aer exp ft osure to ≥16 µg l . Ambient solar 0.41– Pyrene (2 h Peachey & −2 1.4 mW cm (up to Nil Pocillopora exposure) P. damicornis (+UVR): bleached Crosby 8 h, seq. exposure) −1 damicornis (adult) 24 h aer exp ft osure to 48 µg l . No: Bleaching P. compressa (+UVR): no effect. P. compressa Porites compressa (adult) All species (−UVR): no effect. Montipora Artificial UVR 0.98– verrucosa (adult) Pyrene (2 h −2 1.0 mW cm (up to Nil Bleaching Not reported No (both species) exposure) Pocillopora 8 h, seq. exposure) damicornis (adult) Table 1. Summary of previous studies of the phototoxic effects of petroleum hydrocarbons on scleractinian corals. Study methodology, species tested, chemical analysis performed (if applicable), toxic endpoint and main results shown for each study. If no threshold values or concentrations are presented no effect was observed. Seq = sequential, organisms first exposed to pollutant followed by exposure to UVR while kept in clean FSW. GC-MS = gas chromatography-mass spectrometry. Nil = no analysis reported. *Stress response-related gene expression and enzyme activity were also investigated (see reference for further details). Jaz spectrometer calibrated according to manufacturer’s recommendations) giving a calculated total irradiance −2 between 280–400 nm of approximately 0.75 mW cm inside the scintillation vials. Larval settlement assays. Temperature was maintained at 27.6 ± 1.4 °C (mean ± SD) in the experiments while photosynthetically active radiation (PAR) in the +UVR and −UVR treatments averaged 0.95 ± 0.10 and −2 −1 1.23 ± 0.10 mW cm , respectively. Dissolved oxygen concentration averaged 7.9 ± 0.33 mg l with all replicates −1 maintaining concentrations >7.0 mg l for the duration of the exposure period while pH and salinity averaged 8.1 ± 0.06 and 37.0 ± 0.32 psu, respectively. A. tenuis larvae in control treatments were observed to actively swim throughout the exposure (Fig. 2a), but swimming behaviour was not assessed in the fuel exposure treatments. In the control treatments, an average of 73% (SE = 4) larvae underwent settlement in the presence of CCA chips within 24 h (Fig. 2c). This level of settlement success did not change in the presence of UVR with 77% (SE = 3.5) of larvae successfully undergoing settlement. Average larval settlement ≥70% in control treatments was consid- ered indicative of a normal response to settlement inducers based upon several previous studies using CCA or 29,37,38 extracts of CCA to initiate settlement of Acropora spp . Heavy fuel oil toxicity. HFO WAF inhibited larval settlement in both the absence and presence of UVR −1 (Fig. 3a,c and Table 3). Little or no effect on settlement success was observed at low concentrations (< 10 µ g l TPAH); above which, settlement decreased with increasing TPAH concentration (Fig. 3). The toxicity of HFO −1 WAF was enhanced in the presence of UVR resulting in a ~50% decrease of the EC from 96 to 51 µ g l TPAH (95% confidence intervals did not overlap; Table  3). The toxic threshold value (EC ) also reduced in the presence of UVR (Table 3). Obvious mortality (disintegrating cell membranes) was only observed in the highest TPAH −1 treatment (890 µg l ), in the presence of UVR, at the end of the 48 h exposure. No other visible effects on larvae, in either control or HFO WAF treatments, were observed and surviving larvae in the highest HFO treatments exhibited normal morphology (Fig. 2a and b). At 72 h (following the 24 h settlement period) unattached larvae and attached juvenile polyps in control treatments retained normal morphology (Fig. 2c). At this point the frequency and severity of abnormalities increased with increasing TPAH concentration in attached juvenile polyps (asym- metrical or underdeveloped recruits; Fig. 2d) and non-settled larvae (bumps, deformities and necrosis; Fig. 2e) for both ±UVR treatments. Severe deformities were observed in unattached larvae at TPAH concentrations as low as −1 28 µ g l (+UVR) and a substantial proportion of larvae were immobile and/or dead in the highest concentration −1 treatment (890 µ g l ). Most of the successfully attached juvenile polyps underwent complete metamorphosis in the absence of UVR; however, larvae exposed to HFO WAFs appeared to develop more slowly than expected and some had only undergone partial metamorphosis in higher concentration treatments at the time of assessment. ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 3 www.nature.com/scientificreports/ −1 −1 [WAF] (µg l ) [Neat fuel] (mg kg ) TU TU WAF Neat fuel ∑BTEX ∑PAH TPAH ∑BTEX ∑PAH TPAH ∑BTEX ∑PAH TPAH ∑BTEX ∑PAH TPAH 498 440 938 −UVR 190 50494 50684 0.03 0.43 0.46 0.01 0.91 0.92 (+4%) (−1%) (+2%) HFO 478 414 892 +UVR 0.03 0.40 0.43 (−4%) (−13%) (−8%) 745 14 759 −UVR 1491 5723 7214 0.04 0.01 0.05 0.07 0.12 0.19 (−33%) (−57%) (−34%) Diesel 767 16 783 +UVR 0.05 0.02 0.06 (−28%) (−41%) (−29%) Table 2. Time-averaged concentrations of ∑BTEX, ∑PAH and TPAH in undiluted fuel WAFs, ∑BTEX, ∑PAH and TRH in neat fuels, and toxic units (narcosis) calculated from predicted and observed fuel WAF concentrations. Time-averaged concentrations per light treatment calculated from concentrations measured at t and t for HFO and diesel WAFs; % change in concentrations aer 48 ft h indicated in brackets. TUs 0h 48h −1 calculated using an average CTLBB value (86.8 µmol g octanol; n = 15) and aqueous concentrations of BTEX and PAH observed in fuel WAFs (TU ) or predicted from neat fuel oil concentrations (TU ). WAF Neat fuel TPAH = total petroleum aromatic hydrocarbons, ∑PAH = sum of individual PAH concentrations, ∑BTEX = sum of benzene, toluene, ethylbenzene and xylene concentrations. For full analytical results see Tables S-1 and S-2, Supplementary information. Diesel toxicity. Diesel WAF also inhibited larval settlement, but little effect was observed at low to mod- −1 erate concentrations (<100 µ g l TPAH) regardless of UVR treatment (Fig. 3). In the absence of UVR, only −1 the highest treatment concentration (759 µ g l TPAH) inhibited settlement success. An approximate EC of −1 ~1300 µ g l TPAH (without confidence intervals), was estimated by extrapolation (Fig.  3b and Table 3). UVR −1 greatly enhanced the toxiciy of diesel WAF and decreased the EC to 494 µ g l TPAH (Fig. 3d and Table 3). At −1 moderate to high TPAH concentrations (≥200 µ g l ) underdeveloped juvenile polyps were observed in the pres- −1 ence of UVR, and the few attached juvenile polyps observed in the highest concentration (782 µg l TPAH) were either underdeveloped or abnormal. In the absence of UVR, some underdeveloped and malformed attached juve- nile polyps were also observed; however, fully metamorphosed polyps were still present at the highest concen- −1 tration tested (758 µg l TPAH). Some minor abnormalities (e.g. bends and lumps) were observed in unattached larvae from ± UVR treatments but with no apparent relationship to TPAH concentration. Modelled narcotic toxicity. The narcotic toxicities of the undiluted diesel and HFO WAFs used in the present study were estimated using the NTLM . This modelling yielded toxic units (TU) based on both the predicted and measured concentrations of aromatics in undiluted WAFs. TUs can be used to compare the tox- icity of complex mixtures; the greater the TU, the more toxic the solution (see Methods). The narcotic toxicity of HFO and diesel WAFs, calculated using WAF concentrations predicted from neat fuel compositions (TU ; Neat fuel Table 2) , was primarily attributable to PAHs (TU > TU ; Table 2). For undiluted HFO WAF, 0.92 TU ∑PAH ∑BTEX Neat indicates that close to 50% mortality would be expected, while TUs, calculated by applying measured WAF fuel concentrations to the NTLM (TU : 0.46 −UVR and 0.43 +UVR), were approximately half that value (Table 2). WAF The total narcotic toxicity for undiluted diesel WAF indicated that relatively low mortality would be expected following exposure, when calculated using predicted WAF concentrations (TU : 0.19), or close to zero mor- Neat fuel tality, when measured WAF concentrations were applied (TU : 0.05 −UVR to 0.06 +UVR). WAF Discussion Co-exposure to UVR considerably enhanced the negative impacts of the water-soluble fractions of two petroleum fuels on the settlement success of coral larvae. This phototoxicity was evident under UVR intensities equal to or lower than expected exposures in shallow reef habitats, and the toxic thresholds (EC ) for dissolved aromatics 7–9 occurred at concentrations below those detected aer m ft ajor accidental spills or releases . These results indicate that by ignoring phototoxicity, the hazards posed by oil spills to coral larvae may be substantially underestimated in shallow-water tropical reef systems. Coral larval settlement was very sensitive to HFO WAF with threshold concentrations (EC and EC ) lower 10 50 −1 5,8,9 than or similar to previously reported concentrations during and after oil spills (42–189 µ g l ∑PAH and −1 9 −1 22 µ g l TPAH ). The EC and EC values (24 and 96 µ g l TPAH, respectively; −UVR) were also lower than 10 50 those reported for inhibition of larval settlement in the same species in 24 h exposures to natural gas condensate −1 29 in the absence of UVR (103 and 339 µ g l TPAH, respectively) . Differences in sensitivity are likely to be due to the longer duration of the current exposure and the higher proportion of more toxic PAHs in the HFO WAF. Conversely, the sensitivity of A. tenuis larvae to diesel WAF in the absence of UVR was less than that reported for −1 either HFO or condensate, potentially due to the lower concentrations of total dissolved PAHs (14 µ g l ) in the −1 −1 29 undiluted diesel WAF compared to HFO (440 µg l ; Table 2) and condensate (107 µg l ) WAFs. The sensitivity of coral larvae observed here is difficult to compare with other studies, which reported die ff rent exposure types, endpoints and measures of petroleum hydrocarbon concentration in WAFs (summarised in Negri, et al. and Turner, et al. ). Nevertheless, studies exposing coral larvae to hydrocarbons in the absence of UVR reported neg- −1 7,33,40 ative effects on coral settlement at THC concentrations between 82 and 620 µ g l and the sensitivity of coral larvae is likely to be greater than that of other life stages . ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 4 www.nature.com/scientificreports/ Figure 1. Penetration of ultraviolet radiation (UVR) on Trunk Reef (a) and Esk Reef (b) as well as a comparison (c) of exposure intensity and spectrum of artificial UVR used during settlement toxicity assays and UVR observed in situ on the Great Barrier Reef (GBR; Australia) during spring. Full spectrum measurements of UVR in air and at 0.1, 1, 2, 3, and 3.8 m depth at a mid-shelf (Trunk reef ) and inshore (Esk reef ) reef on the GBR in October 2016. Total irradiance values calculated using the percentage reductions in light intensity recorded for each wavelength at each depth (in relation to measurements in air) and total irradiance measurements in air on a clear day (cloud coverage <5%). Comparison of UVR intensity and spectrum emitted from fluorescent tubes used in settlement toxicity assays, calculated UVR exposure inside scintillation vials and UVR exposures observed at 1 m depth on the GBR during spring. ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 5 www.nature.com/scientificreports/ Figure 2. Photomicrographs of A. tenuis planulae larvae exposed to water accommodated fractions (WAFs) of heavy fuel oil (HFO) in the absence (−UVR) or presence (+UVR) of ultraviolet radiation as well as juvenile −1 polyps (following settlement). Larvae exposed to (a) filtered seawater (0 µg TPAH l ), (b) approximately 900 µg −1 TPAH l aer 48 ft h of exposure as well as juvenile polyps, attached and unattached larvae treated with (c) FSW −1 −1 −1 (0 µg TPAH l ), (d) 115 µg TPAH l and (e) approximately 900 µg TPAH l HFO WAF aer 48 ft h of exposure, introduction of settlement inducer and a 24 h settlement period (a total ~72 h aer exp ft eriment start). All concentrations in µg TPAH l-1. TPAH = total petroleum aromatic hydrocarbons. Figure 3. Concentration-response curves for coral larval settlement following exposure to heavy fuel oil (a and c) and diesel (b and d) water accommodated fractions (WAF) in the presence (blue) and absence (green) −1 of ultraviolet radiation (µ g TPAH l ). Model mean (solid line) and 95% confidence intervals (shaded area) for quasibinomial GLMs fitted for the settlement success data of each treatment combination as well as observed −1 settlement success for each replicate (open ring) used in model fitting. All concentrations in µ g l . n = 63, HFO+/−UVR n = 65, n = 64. TPAH = total petroleum aromatic hydrocarbons. diesel−UVR diesel+UVR −1 The toxicity of both HFO and diesel WAFs doubled (EC : 95 to 51 and 1300 to 494 µ g l TPAH ±UVR, respectively) in the presence of ecologically relevant intensities of UVR. This observed phototoxicity is consistent with the 2.5-fold increase in toxicity of natural gas condensate WAF in the presence of UVR to the same lar- val species in 24 h WAF exposures, based on measured TPAH concentrations . This toxicity increase is similar between studies despite differences in UVR exposure profiles, where Negri, et al . applied a single, higher inten- −2 sity UVR dose for a shorter period of time (2 h, summing to 39 W cm ) compared to this study (see Methods). Overmans et al. recently reported that the inhibition of coral larval metamorphosis by the PAH anthracene ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 6 www.nature.com/scientificreports/ EC EC Approximate r Deviance df 10 50 24 96 −UVR 0.748 146.12 61 (9–39) (75–116) HFO 15 51 +UVR 0.767 137.03 61 (6–23) (41–62) 302 ~1300* −UVR 0.062 196.07 63 (51–552) (253–2410) Diesel 122 494 +UVR 0.423 165.97 62 (50–194) (371–617) Table 3. Concentrations of TPAH in fuel WAFs which inhibited 50% (EC ) and 10% (EC ) of larval settlement 50 10 in the absence (−UVR) and presence (+UVR) of ultraviolet radiation. Effect concentrations calculated from settlement success data fitted with quasibinomial GLMs with approximate r , deviance and degrees of freedom (df ) shown for each treatment combination. 95% confidence intervals in brackets where applicable. All concentration −1 values in µ g TPAH l . TPAH = total petroleum aromatic hydrocarbons. *Extrapolated from fitted model. −1 increased 7-fold in the presence of UVR (EC : 45 −UVR to 6.3 +UVR µg l ), while there was little influence of UVR exposure on the toxicity of phenanthrene. The impacts of PAH and UVR on larval metamorphosis were also detected at lower PAH concentrations than other sub-lethal biomarkers investigated following the 48 h exposures 18 17 (including 10 h UVR exposure per day) . Peachey and Crosby observed 100% mortality of F. scutaria larvae −1 aer 1 ft h exposures to 48 µ g l pyrene followed by exposure to ambient solar radiation (including UVR). Strong phototoxic ee ff cts have also been demonstrated for adult P . varicata corals where UVR increased the toxicity of uo fl ranthene by approximately 14-fold within 72 h following a 4.5 h exposure to this PAH . It is not clear whether the main influence of UVR is to generate more toxic transformation products, or whether the additional toxicity is caused by elevated oxidative stress within the larvae as PAHs, excited by UVR, decay back to their ground states 12,13 (photosensitisation) . However, phototoxicity in aquatic organisms is generally considered to be caused by photosensitisation . While UVR increase the toxicity of some individual PAHs by more than 10 times, fuel WAFs are comprised of many aromatics and several of these, including BTEX, the naphthalenes and phenanthrene, are not considered phototoxic . Consequently, very large increases in oil or fuel WAF toxicity in the presence of UVR are unlikely. In addition to reductions in settlement success, exposure to fuel WAFs also caused changes to larval morphol- ogy and development. Slowed onset and completion of metamorphosis, as well as increased instances of morpho- logical abnormalities, were observed in larvae exposed to WAFs. The process of attachment and metamorphosis following addition of settlement inducers was slower at moderate-high TPAH concentrations, with only partial metamorphosis achieved in 24 h, similar to what would generally be expected 6–12 h ae ft r settlement induction (A.J. Heyward, personal communication, Dec 2016). Delayed onset of metamorphosis, abnormal development or deformations following exposures to petroleum hydrocarbons have also been reported for larvae and juvenile 28,29,42 polyps from other coral species . The development of abnormal morphologies in coral larvae may indicate narcosis or more specific toxic effects of petroleum hydrocarbons on cellular developmental processes. The results of this study clearly demonstrate the strong phototoxic effects of the water-soluble fractions of common marine fuels on coral larval settlement in the presence of environmentally realistic UVR exposures. The current study applied UVR intensities slightly lower than the intensities measured at 1 m depth on a mid-shelf (Trunk Reef ) and inshore reef (Esk reef ) on the Central GBR. These intensities are also consistent with previous 23 22,24 measurements on the GBR , and other coastal and oceanic locations . The penetration of UVR in marine environments is primarily ae ff cted by chromophoric dissolved organic matter (CDOM) and, for some systems, 22,24 particulate matter such as plankton and detritus . Attenuation is therefore likely to be lowest in clear-water oli- gotrophic coral reef environments, as measured at the mid-shelf site at Trunk Reef, compared to the more turbid waters of the inshore Esk Reef. Barron et al. found that at depths greater than 0.5 m attenuation of light varies between habitats, emphasising the need to accurately estimate the intensity and wavelengths experienced by each species or ecosystem when investigating the influence of UVR on the toxicity of pollutants. Coral gametes, embryos and larvae developing at the water surface may also be exposed to substantially higher UVR intensities in situ than applied in the present study, potentially reducing the toxic threshold values further. Moreover, these results suggest that the hazard (hence risk) posed by aqueous petroleum hydrocarbons to shallow-water tropical coral reefs will be underestimated if phototoxic activation by UVR is not taken into consideration. The NTLM for predicting narcotic mortality underestimated the impacts of HFO WAF in comparison to the observed effects on larval settlement. 50% larval mortality was predicted for HFO WAFs by the NTLM at −1 ~900 µ g l TPAH (TU ); however, coral larval settlement was reduced by 50% at far lower concentrations Neat fuel −1 (EC s 96 to 51 µg l TPAH ±UVR). It has been reported that the NTLM can underestimate the toxicity of some hydrocarbon mixtures , and these underestimations may result from the assumption of a simple narcotic mode of action for all aromatic components. The observed discrepancies could partially result from the contribution of other toxic mechanisms, especially those that specifically affect coral metamorphosis from pelagic planula to sessile juveniles, as suggested by Negri et al. . Furthermore, the sub-lethal endpoint of settlement inhibition (while ecologically relevant ) is by definition more sensitive than mortality, which is typically used to generate the CTLBB values used in the NTLM. Determining CTLBB values for both larvae and adults of key coral reef spe- cies would improve species-specific toxicity predictions at different life stages and assist in ranking the potential risks posed by crude oils and petroleum products . It is also possible that other components (e.g. photoproducts) that were not measured may have influenced the toxicity. Although a phototoxic target lipid model was recently ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 7 www.nature.com/scientificreports/ presented , its use is currently limited to estimation of the acute toxicity of individual PAHs, and further devel- opment is necessary before it can be applied to complex hydrocarbon mixtures or chronic exposures. Even in well managed and protected areas, groundings of large vessels and petroleum hydrocarbon releases 43–45 from offshore extraction facilities have occurred in the last decade . Large spills as well as chronic hydrocarbon 7,46,47 pollution can lead to degradation of adult coral health and changes to reef composition with recovery taking more than 10–20 years . The slow recovery of these reefs is likely to be at least partially due to long term effects of hydrocarbon contamination on recruitment processes, including larval settlement ; used as a sensitive toxicity endpoint in this study. The overlap between a large oil spill and the coral recruitment window can be substantial, with water from some spill sites remaining phototoxic to invertebrate embryos for up to 13 d and larvae of some coral species remaining in the water column for up to 3 months (as reviewed by Jones, et al. ). The increase in toxicity of dissolved aromatics from HFO by UVR exposure resulted in low toxic thresholds, underscoring the potential hazard to corals posed by phototoxic compounds found in petroleum oils and fuels. Previous assess- ments may therefore have substantially underestimated the risks posed by oil and petroleum product spills on shallow-water, tropical coral reefs by not accounting for interactions with environmental factors such as UVR. Further research into the effects of petroleum hydrocarbons on more tropical reef organisms, including potential interactions with UVR and other stressors, is needed to more effectively quantify these risks. Methods Coral collection and larval cultures. Gravid colonies of Acropora tenuis, a reef-building coral common throughout the Pacific Ocean, were collected by hand on SCUBA from Magnetic Island (October 2016, 19.157°S, 146.861°E), GBR, under Great Barrier Reef Marine Park Authority Permit G12/35236.1. Colonies were placed in flow-through seawater of ambient temperature and transported to the National Sea Simulator, Townsville, within 24 hours. On arrival colonies were transferred to 70% shaded flow-through outdoor holding tanks and kept at temperatures equivalent to the collection site (27 °C) until spawning. When showing signs of setting colonies were isolated and gametes collected by gentle scooping. th Larval cultures (6 parental colonies, 95% fertilisation) were initiated on the 19 October 2016. Cultures were maintained at 27 °C and densities <500 larvae per l at ~27 °C in round, 500 l fibreglass rearing tanks with cone-shaped bases. Flow-through seawater (1.5 turnovers per day) was 1 µ m-filtered and a round air stone at the base of each tank provided aeration and created a gentle curtain of bubbles to keep larvae from a submerged cylindrical mesh filter (15 h × 6 d cm, 100 µm) at the outflow. Preparation of fuel water accommodated fractions. Heavy residual fuel oil (HFO, International Bunker Supplies Pty Ltd, Gladstone, Australia) and automotive diesel (Puma Energy Australia, Fortitude Valley, Australia) WAFs were prepared in capped, solvent rinsed aspirator bottles (5 l) using 0.45 µ m filtered seawater −1 51 (FSW; pH 8.1, salinity 37.0 psu) loaded at 20 g fuel l with a 20% headspace . Solutions were protected from light, stirred at 50 rpm for 18 h without a vortex and allowed to settle for 30 min prior to use . Ten dilutions (0, 0.4, 0.8, 1.6, 3.125, 6.25, 12.5, 25, 50 and 100% WAF) were prepared from the fresh undiluted (100%) WAF using 29,52 0.45 µm FSW and used within 3 h in the assays below. Fuel phototoxicity assays. Acropora spp. larvae reach competency to settle and undergo metamorphosis aer ~4 d ft . HFO and diesel static non-renewal exposure experiments over 48 h were performed on 7- and 8-d old A. tenuis larvae, respectively. 10–14 larvae and 20 ml of WAF were gently added to 6 replicate glass vials for each WAF concentration and ±UVR treatment combination (12 replicate vials for 0% WAF controls). Vials were tightly capped with approximately 10% headspace to allow for oxygen exchange. Vials were randomly placed on their side in trays inside temperature controlled, orbital shaker incubators (Thermoline Scientific, Australia) at 80 rpm to ensure larvae did not settle during the exposure period. One set of vials (−UVR), n = 66, was placed in −2 an incubator fitted with actinic LEDs emitting 1.23 mW cm photosynthetically active radiation (PAR; Aqualina Blue 450 nm, 10,000 K and 420 nm Actinic LED strips, The Aquatic Life Product Company, Willawong, Australia). e s Th econd set of vials (n = 66) was placed in a second incubator in the presence of both actinic light and UVR −2 (+UVR). This incubator (+ UVR) was t fi ted with identical actinic LEDs as the first incubator (PAR: 0.95 mW cm ) as well as three sets of UV-emitting fluorescent tubes (each set consisting of one Deluxlite Blacklight Blue 18 W −2 and one Reptile One UVB 5.0 18 W T8 fluorescent tube) emitting 0.75 mW cm UVR. Fluorescent tubes emitting predominantly in the UVA spectrum were chosen due to the high UVB attenuation of seawater (See Ultraviolet radiation intensities section below on how fluorescent tube irradiance was characterised, and Fig.  1c for compari- son of experimental lights, light attenuation of vials used for experimental exposures and in situ intensities of UVR −2 on the GBR). PAR was provided on a 12:12 h L:D cycle and UVR on a 6:18 h L:D cycle (total irradiance 16.1 W cm −1 d ). An additional 30 vials containing undiluted WAF, but no larvae, were also placed in each incubator to allow for the collection of chemical samples at the end of the 48 h exposure. The positions of vials within each incubator were exchanged randomly throughout each experiment to minimize variation in light exposure. Temperature was continuously logged (Onset HOBO temperature logger, Onset Computer Corp., Massachusetts, USA) while pH, salinity and dissolved oxygen concentrations were measured at the beginning and end of each experiment. At the end of the 48 h exposure approximately 600 ml of undiluted WAF was pooled, from the 30 additional vials (not containing larvae), for chemical analysis (see Chemical analysis below). Following exposure to fuel WAFs in the presence or absence of UVR, larvae were transferred with 10 ml treatment solution directly to individual wells in 6-well cell culture plates (Nalge Nunc Int., Denmark). Larvae were presented with a settlement inducer consisting of 5 × 5 mm live chips of the crustose coralline algae (CCA) Porolithon onkodes . Settlement success was assessed as percentage of larvae which attached and underwent at least early metamorphosis (i.e. became firmly attached and flattened into a disc shape ) aer ~24 ft h incubation at 27 °C. Average settlement success ≥70% in controls was considered indicative of a normal response to settlement ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 8 www.nature.com/scientificreports/ 29,33,37,54,55 inducers based on several previous studies using CCA or extracts of CCA to initiate settlement . Additional notes regarding deformities, ratio of fully (as outlined by Heyward & Negri ) to partially metamor- phosed recruits as well as failure to attach were also made for each treatment combination. Settlement data analysis. Settlement data was fitted with binomial generalized linear models (GLMs) with a logit link function using the R stats-package (R version 3.4.1 ) to model the settlement success of A. tenuis larvae in response to treatment concentration (fixed numerical factor) for each light treatment and fuel type com- bination. The fitted models were validated by plotting the simulated residuals against fitted values and ensuring no individual values influenced the model fits’ disproportionally. Quasibinomial GLMs were fitted where diag- nostics indicated overdispersion. An r analogue was calculated using the deviance of the fitted models with and without the fixed numerical factor (equation (1)) to approximate the goodness-of-fit for each model .   deviance 2 Full model   Approximater =− 1   deviance   Null model (1) EC and EC values with 95% confidence intervals (CI) were interpolated from model mean values and 95% 10 50 CI (adapted from Venables & Ripley ). Predicted model mean values and 95% CI were exported and graphical outputs produced using GraphPad Prism (version 7.02, GraphPad Software Inc., CA, USA). The high levels of replication used allowed the identification of outliers in the dataset which were excluded. These comprised three HFO FSW controls and one diesel low concentration replicate where CCA chips induced 0% and 7% settlement, respectively, likely due to misidentification of a few of the live CCA chips. Chemical analysis. Samples of freshly prepared, undiluted WAF (“100% WAF”) were collected for chemical analysis at the beginning of each experiment (t ). Undiluted WAF was also added to 30 vials per UVR treatment 0h and incubated simultaneously with vials containing larvae to ensure that the undiluted WAF was exposed to the same experimental conditions as the test solutions. At the end of the exposure period (t ), the undiluted WAF, 48h in vials containing no larvae, was pooled and sampled for chemical analysis. For BTEX analysis, samples (40 ml) were collected in amber glass vials with open hole caps and PTFE septa. For all other analyses, single samples (500 ml) were collected in amber glass bottles with PTFE-lined caps. All samples were acidified to pH 2 using 6 M hydrochloric acid and stored at 4 °C until shipped to ChemCentre (Perth, Australia) for analysis as previ- ously described by Negri et al. . Briefly, WAFs were analysed directly for BTEX using Purge and Trap GC-MS in full scan mode (USEPA method 8260). The 500 ml WAF samples were extracted three times with dichlo- romethane (DCM) and the combined extracts analysed for PAH and alkylated PAH using GC-MS in SIM or scan mode, and TRH using GC-FID (USEPA method 8270). Neat HFO and diesel were diluted in DCM and ana- lysed for TRH, BTEX, PAH/alkylated PAH and phenols using the same methodology, except additional surrogate (2-fluorophenol, phenol-d5 and 2,4,6-tribromophenol) and internal (1,4-dichlorobenzene-d4) standards were added to the samples prior to analysis. To profile their general composition, the neat oils (1 µl) were subjected to whole oil analysis using GC-MS and hydrocarbons were identified through comparison with a pre-characterised reference oil. Ultraviolet radiation intensities. In situ UVR irradiance on the Central GBR during spring was assessed through full spectrum measurements of UVR in air and at five depths at two reefs. Esk Reef (18.775°S 146.522°E), located in the Palm Islands-group (QLD), was selected as a representative inshore reef site while Trunk Reef (18.329°S, 146.846°E) was selected as a representative clear-water, mid-shelf reef site. Three replicate light intensity measurements for wavelengths between 300–400 nm were performed on SCUBA using a Jaz handheld spectrome- ter (Ocean Optics, Inc., Florida, USA) and a 5 m fibre optic cable (CPATCH-5074768, Ocean Optics, Inc., Florida, USA) with a planar irradiance collector for underwater use (HOBI Labs, Inc., USA) following Shick et al. (1996). Measurements were made with the sensor positioned vertically at 0, 0.1, 1, 2, 3 and 3.8 m depth. Measurements were performed close to solar noon on the mid-shelf (14:17–14:20 on 12 October 2016) and inshore reefs (13:55– 14:04 on 14 October 2016), respectively. Intensity data collected was used to calculate the average relative decrease in radiation for wavelengths between 300–400 nm with depth. Total UVA and UVB radiation above the surface was recorded (Solarmeter Model 5.0 UVA + UVB meter, Solartech Inc., Pennsylvania, USA) and turbidity meas- urements were performed (90FL-T, TPS Pty Ltd, Brendale, QLD, Australia). Cloud coverage was low (<5%) with 17 km/h E winds (BOM, 2016) and turbidity of 0.1 NTU during mid-shelf measurements with medium-high cloud coverage (~80%), 33 km/h ENE winds (maximum gusts 46 km/h ENE; BOM, 2016b) and 0.8 NTU turbidity during inshore reef measurements. Theoretical irradiance at each depth was calculated using full spectrum measurements of natural sunlight close to solar noon on a low cloud coverage day (<5%). Measurements were made using a calibrated Jaz spectrometer and a 250 mm UVR compatible fibre optic cable (QP600-025-UV, Ocean Optics, Inc., Florida, USA) aimed directly towards the sun. Z10% values at 305 and 340 nm were estimated by calculating the irradiance corresponding to 10% of surface irradiance (in air) for each reef site and the negative linear relationship between irradiance and measurement depth of measurements made at Trunk Reef and Esk Reef. Full spectrum measurements of radiation emitted by the UVR fluorescent tubes (Deluxlite Black Light Blue 18W; Reptile One UVB 5.0 18W), used in +UVR larval settlement assays, were performed using the same cali- brated Jaz spectrometer and 250 mm UVR compatible fibre optic cable used to quantify the UV radiation of nat- ural sunlight. Full spectrum measurements were made at approximately the same distance as sample vials during experimental exposures (170 mm) in v fi e separate positions relative to the three sets of uo fl rescent tubes. Total UVA and B radiation measurements were also performed (Solarmeter Model 5.0 UVA + UVB meter, Solartech Inc., Pennsylvania, USA) for the five replicates. Additionally, the attenuation of UVR, between 300–400 nm, emit- ted from fluorescent tubes by scintillation vial glass was estimated (calibrated Jaz spectrometer). Measurements ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 9 www.nature.com/scientificreports/ were made 200 mm from UVR fluorescent tubes (Deluxlite Black Light Blue 18W; Reptile One UVB 5.0 18W) through the base of a 20 ml scintillation vial. The average total UVA and UVB exposure of larvae inside scintilla- tion vials was calculated using measurements performed with the Solarmeter model 5.0 and the average attenua- tion of scintillation vial glass between 300–400 nm. Toxic unit calculations for narcotic toxicity. e t Th oxicity of aromatics to aquatic organisms is depend- ent on the partitioning of dissolved compounds between water and lipids. The narcotic toxicity of hydrocarbon mixtures can be estimated using the narcotic target lipid model (NTLM) which combines the octanol-water partitioning coefficients (K ) of all dissolved components in the solution, their measured (or expected) aque- OW ous concentrations and the critical lipid body burden (CTLBB) of the organism of interest (where the CTLBB is the amount of the compound dissolved in the organism’s tissues which causes a specific toxic effect; e.g. 50% mortality) . The NTLM is useful for assessing the relative risks posed by different oil and fuel types to aquatic organisms, but may be less useful if the CTLBB is unknown or may underestimate toxicity if other toxic modes of action are important for a given species . e Th dissolved composition of petroleum hydrocarbons in a WAF can be measured (Table S-2, Supplementary information) or modelled by applying an oil solubility model (e.g. PETROTOX) to an oil of known composition . Toxic units (TU) for each constituent in a WAF are defined as the ratio between the concentration (C ) and the WAF, i 50% critical effect level (LC , EC or IC ) of each constituent (i) in a solution, and can be used to enable com- 50,i 50,i 50,i parisons of studies using differing experimental designs and hydrocarbon compositions . Assuming the same mode of action (narcosis) for all MAHs and PAHs, the TUs for each component of a WAF containing a mixture of PAHs and MAHs, are considered additive and can be summed to estimate the TU of the WAF according to equation (2). C C C C WAFW ,1 AF ,2 WAFW ,3 AF ,i TU = +++  + WAF ∑ EC EC EC EC (2) 50,1 50,2 50,3 50,i A TU of 1 indicates that the mixture is predicted to be toxic and will ae ff ct 50% of exposed organisms . In this study, expected TUs (TU ) were calculated as per Redman and Parkerton , by applying the observed Neat fuel −1 concentrations of individual compounds in neat HFO and diesel and the average LC CTLBB (86.8 µ mol g 14,61 octanol) of 15 marine and estuarine organisms . An average marine CTLBB was used as no CTLBB currently exist for acroporid corals. Additionally, observed TUs (TU ) for each light treatment were calculated using WAF measured concentrations of individual compounds in undiluted HFO and diesel WAFs and the same CTLBB value. Data availability. Data are available upon request. References 1. Haapkylä, J., Ramade, F. & Salvat, B. Oil pollution on coral reefs: a review of the state of knowledge and management needs. Vie Milieu 57, 95–111 (2007). 2. Turner, N. R. & Renegar, D. A. Petroleum hydrocarbon toxicity to corals: A review. Mar. Pollut. Bull. 119, 1–16, https://doi. org/10.1016/j.marpolbul.2017.04.050 (2017). 3. Kroon, F. et al. Identification, impacts, and prioritisation of emerging contaminants present in the GBR and Torres Strait marine environments. Report to the National Environmental Science Programme. Reef and Rainforest Research Centre Limited, Cairns (138 pp.) (2015). 4. Neff, J. M., Ostazeski, S., Gardiner, W. & Stejskal, I. Effects of weathering on the toxicity of three offshore Australian crude oils and a diesel fuel to marine animals. Environ. Toxicol. Chem. 19, 1809–1821 (2000). 5. Boehm, P. D., Neff, J. M. & Page, D. S. Assessment of polycyclic aromatic hydrocarbon exposure in the waters of Prince William Sound aer t ft he Exxon Valdez oil spill: 1989–2005. Mar. Pollut. Bull. 54, 339–356 (2007). 6. AMSA. Montara well release - Operational monitoring study O2: Monitoring of oil character, fate and effects. (Australian Maritime Safety Authority, Australian Government, https://www.amsa.gov.au Accessed on 13.04.2017, 2010). 7. Hartmann, A. C. et al. Crude oil contamination interrupts settlement of coral larvae aer dir ft ect exposure ends. Mar. Ecol. Prog. Ser. 536, 163–173 (2015). 8. Diercks, A. R. et al. Characterization of subsurface polycyclic aromatic hydrocarbons at the Deepwater Horizon site. Geophys. Res. Lett. 37 (2010). 9. Reddy, C. & Quinn, J. The North Cape oil spill: hydrocarbons in Rhode Island coastal waters and Point Judith pond. Mar. Environ. Res. 52, 445–461 (2001). 10. Boehm, P. D. & Fiest, D. L. Subsurface distributions of petroleum from an offshore well blowout. The Ixtoc I blowout, Bay of Campeche. Environ. Sci. Technol. 16, 67–74 (1982). 11. Redman, A. D. Role of entrained droplet oil on the bioavailability of petroleum substances in aqueous exposures. Mar. Pollut. Bull. 97, 342–348, https://doi.org/10.1016/j.marpolbul.2015.05.068 (2015). 12. Arfsten, D. P., Schaeffer, D. J. & Mulveny, D. C. The effects of near ultraviolet radiation on the toxic effects of polycyclic aromatic hydrocarbons in animals and plants: a review. Ecotoxicol. Environ. Saf. 33, 1–24 (1996). 13. Roberts, A. P., Alloy, M. M. & Oris, J. T. Review of the photo-induced toxicity of environmental contaminants. Comp. Biochem. Physiol. C 191, 160–167, https://doi.org/10.1016/j.cbpc.2016.10.005 (2017). 14. McGrath, J. A. & Di Toro, D. M. Validation of the target lipid model for toxicity assessment of residual petroleum constituents: monocyclic and polycyclic aromatic hydrocarbons. Environ. Toxicol. Chem. 28, 1130–1148 (2009). 15. Redman, A. D. & Parkerton, T. F. Guidance for improving comparability and relevance of oil toxicity tests. Mar. Pollut. Bull. 98, 156–170 (2015). 16. Pelletier, M. C. et al. Phototoxicity of individual polycyclic aromatic hydrocarbons and petroleum to marine invertebrate larvae and juveniles. Environ. Toxicol. Chem. 16, 2190–2199 (1997). 17. Peachey, R. L. & Crosby, D. G. Phototoxicity in a coral reef flat community. UV Radiation and Coral Reefs, HIMB Technol. Report 41, 193–200 (1995). 18. Overmans, S. et al. Phototoxic ee ff cts of PAH and UVA exposure on molecular responses and developmental success in coral larvae. Aquat. Toxicol. 198, 165–174, https://doi.org/10.1016/j.aquatox.2018.03.008 (2018). ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 10 www.nature.com/scientificreports/ 19. Vantrepotte, V. & Melin, F. UV penetration in the water column. European Commission D-GJRC, Institue for Environment and Sustainability. Luxembourg: European Communities (2006). 20. Barron, M. G., Little, E. E., Calfee, R. & Diamond, S. Quantifying solar spectral irradiance in aquatic habitats for the assessment of photoenhanced toxicity. Environ. Toxicol. Chem. 19, 920–925 (2000). 21. McDonald, B. G. & Chapman, P. M. PAH phototoxicity—an ecologically irrelevant phenomenon? Mar. Pollut. Bull. 44, 1321–1326 (2002). 22. Tedetti, M. & Sempere, R. Penetration of ultraviolet radiation in the marine environment. A review. Photochem. Photobiol. 82, 389–397, https://doi.org/10.1562/2005-11-09-IR-733 (2006). 23. Michael, K. J., Veal, C. J. & Nunez, M. Attenuation coec ffi ients of ultraviolet and photosynthetically active wavelengths in the waters of Heron Reef, Great Barrier Reef, Australia. Mar. Freshw. Res. 63, 142–149 (2012). 24. Barron, M. G., Vivian, D. N., Yee, S. H. & Santavy, D. L. Methods to estimate solar radiation dosimetry in coral reefs using remote sensed, modeled, and in situ data. Environmental Monitoring and Assessment 151, 445–455, https://doi.org/10.1007/s10661-008- 0288-4 (2009). 25. Yee, S. H., Santavy, D. L. & Barron, M. G. Comparing environmental influences on coral bleaching across and within species using clustered binomial regression. Ecological Modelling 218, 162–174, https://doi.org/10.1016/j.ecolmodel.2008.06.037 (2008). 26. McNutt, M. K. et al. Review of o fl w rate estimates of the Deepwater Horizon oil spill. Proc. Acad. Nat. Sci. 109, 20260–20267 (2012). 27. Carmen Guzmán Martínez, M. D., Romero, P. R. & Banaszak, A. T. Photoinduced toxicity of the polycyclic aromatic hydrocarbon, uo fl ranthene, on the coral, Porites divaricata. J. Environ. Sci. Health. A 42, 1495–1502 (2007). 28. Villanueva, R., Montaño, M. & Yap, H. Effects of natural gas condensate–water accommodated fraction on coral larvae. Mar. Pollut. Bull. 56, 1422–1428 (2008). 29. Negri, A. P. et al. Acute ecotoxicology of natural oil and gas condensate to coral reef larvae. Sci. Rep. 6, 21153, https://doi.org/10.1038/ srep21153 (2016). 30. Marzooghi, S. et al. Phototoxic target lipid model of single polycyclic aromatic hydrocarbons. Environ. Toxicol. Chem (2016). 31. Harrison, P. The effects of oil pollutants on fertilisation rates in the scleractinian coral Acropora tenuis. Proceedings of the Joint Scientific Conference on Science, Management and Sustainability of Marine Habitats in the 21st Century. Conference Abstracts 30 (1994). 32. Di Toro, D. M., McGrath, J. A. & Stubblefield, W. A. Predicting the toxicity of neat and weathered crude oil: toxic potential and the toxicity of saturated mixtures. Environ. Toxicol. Chem. 26, 24–36 (2007). 33. Negri, A. P. & Heyward, A. J. Inhibition of fertilization and larval metamorphosis of the coral Acropora millepora (Ehrenberg, 1834) by petroleum products. Mar. Pollut. Bull. 41, 420–427, https://doi.org/10.1016/S0025-326x(00)00139-9 (2000). 34. Kushmaro, A., Henning, G., Hofmann, D. & Benayahu, Y. Metamorphosis of Heteroxenia fuscescens planulae (Cnidaria: Octocorallia) is inhibited by crude oil: a novel short term toxicity bioassay. Mar. Environ. Res. 43, 295–302 (1997). 35. Mercurio, P., Negri, A. P., Burns, K. A. & Heyward, A. J. The ecotoxicology of vegetable versus mineral based lubricating oils: 3. Coral fertilization and adult corals. Environ. Pollut. 129, 183–194 (2004). 36. Warne, M. et al. Revised Method for Deriving Australian and New ZealandWater Quality Guideline Values for Toxicants. Prepared for the Council of Australian Government’s Standing Council on Environment and Water (SCEW). Department of Science. Information Technology and Innovation, Brisbane, Queensland, https://doi.org/10.13140/RG.2.2.20368.20483 (2015). 37. Negri, A. P. & Hoogenboom, M. O. Water contamination reduces the tolerance of coral larvae to thermal stress. Plos One 6, e19703 (2011). 38. Negri, A. P., Harford, A. J., Parry, D. L. & van Dam, R. A. Effects of alumina refinery wastewater and signature metal constituents at the upper thermal tolerance of: 2. The early life stages of the coral Acropora tenuis. Mar. Pollut. Bull. 62, 474–482, https://doi. org/10.1016/j.marpolbul.2011.01.011 (2011). 39. Redman, A. D., Parkerton, T. F., McGrath, J. A. & Di Toro, D. M. PETROTOX: An aquatic toxicity model for petroleum substances. Environ. Toxicol. Chem. 31, 2498–2506, https://doi.org/10.1002/etc.1982 (2012). 40. Goodbody-Gringley, G. et al. Toxicity of Deepwater Horizon source oil and the chemical dispersant, Corexit 9500, to coral larvae. Plos One 8, e45574 (2013). 41. Mekenyan, O. G., Ankley, G. T., Veith, G. D. & Call, D. J. QSARs for photoinduced toxicity: I. Acute lethality of polycyclic aromatic hydrocarbons to Daphnia magna. Chemosphere 28, 567–582 (1994). 42. Te, F. T. Effects of two petroleum products on Pocillopora damicornis planulae. Pac . Sci. 45 (1991). 43. GBRMPA. Great Barrier Reef Marine Park Authority. Grounding of the Shen Neng 1 on Douglas Shoal, April 2010: Impact Assessment Report, http://elibrary.gbrmpa.gov.au Accessed on 23 April 2017 (2011). 44. Storrie, J. Montara Wellhead Platform Oil Spill – A Remote Area Response. International Oil Spill Conference Proceedings (IOSC) 2011, Abstract 159 (2011). 45. NOPSEMA. Annual offshore performance report - to 31 December 2016. (National Offshore Petroleum Safety and Environmental Management Authority, Commonwealth of Australia, www.nopsema.gov.au Accessed on 07.03.2018 2017). 46. Rinkevich, B. & Loya, Y. Harmful effects of chronic oil pollution on a Red sea scleractinian coral population. i Th rd International Coral Reef Symposium Miami, Florida, pp. 585–591 (1977). 47. Jackson, J. B., Cubit, J. D., Keller, B. D., Batista, V. & Burns, K. Ecological ee ff cts of a major oil spill on Panamanian coastal marine communities. Science 243, 37 (1989). 48. Loya, Y. & Rinkevich, B. Effects of oil pollution on coral reef communities. Mar. Ecol. Prog. Ser. 3, 180 (1980). 49. Ho, K. et al. The chemistry and toxicity of sediment ae ff cted by oil from the North Cape spilled into Rhode Island Sound. Mar. Pollut. Bull. 38, 314–323 (1999). 50. Jones, R., Ricardo, G. & Negri, A. Effects of sediments on the reproductive cycle of corals. Mar. Pollut. Bull. 100, 13–33 (2015). 51. Aurand, D. & Coelho, G. Cooperative aquatic toxicity testing of dispersed oil and the chemical response to oil spills: Ecological Effects ResearchForum (CROSERF). Inc . Lusby, MD. Tech. Report, 07–03 (2005). 52. Barron, M. G. & Ka’aihue, L. Critical evaluation of CROSERF test methods for oil dispersant toxicity testing under subarctic conditions. Mar. Pollut. Bull. 46, 1191–1199, https://doi.org/10.1016/S0025-326X(03)00125-5 (2003). 53. Heyward, A. J. & Negri, A. P. Natural inducers for coral larval metamorphosis. Coral Reefs 18, 273–279 (1999). 54. Negri, A. P. & Heyward, A. J. Inhibition of coral fertilisation and larval metamorphosis by tributyltin and copper. Mar. Environ. Res. 51, 17–27, https://doi.org/10.1016/S0141-1136(00)00029-5 (2001). 55. Negri, A. et al. Effects of the herbicide diuron on the early life history stages of coral. Mar. Pollut. Bull. 51, 370–383, https://doi. org/10.1016/j.marpolbul.2004.10.053 (2005). 56. R: A language and evironment for statistical computing (R Foundation for Statistical Computing, Vienna, Austria, 2017). 57. DHARMa: Residual Diagnostics for Hierarchical (Multi-Level/Mixed) Regression Models (Hartig, F., 2017). 58. Logan, M. In Biostatistical Design and Analysis Using R 483–530 (Wiley-Blackwell, 2010). 59. Venables, W. N. & Ripley, B. D. Modern applied statistics with S. 4 edn, 193–194 (Springer-Verlag New York, 2002). 60. Billiard, S. M., Meyer, J. N., Wassenberg, D. M., Hodson, P. V. & Di Giulio, R. T. Nonadditive ee ff cts of PAHs on Early Vertebrate Development: mechanisms and implications for risk assessment. Toxicol. Sci. 105, 5–23, https://doi.org/10.1093/toxsci/kfm303 (2008). 61. Ahrens, M. J., Nieuwenhuis, R. & Hickey, C. W. Sensitivity of juvenile Macomona liliana (bivalvia) to UV‐photoactivated uo fl ranthene toxicity. Environ. Toxicol. 17, 567–577 (2002). ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 11 www.nature.com/scientificreports/ Acknowledgements e a Th uthors would like to thank the Australian Institute of Marine Science Townsville (AIMS) and the National Sea Simulator (SeaSim), Tristan Lever and Edith Strecker for logistical support as well as Gerard Ricardo and Patricia Menendez for their advice regarding statistical analysis. This work was supported by a research grant set up as part of the collaborative research project between the Red Sea Research Center (RSRC) at King Abdullah University of Science and Technology (KAUST) and the AIMS. Author Contributions F.M.N. and A.P.N. designed the experiment with input from F.F., D.L.B. and S.A. F.M.N., F.F., D.L.B. and A.P.N. performed experiments. F.M.N. analysed the data. F.M.N. wrote the manuscript with input from A.P.N. as well as F.F., D.L.B. and S.A. Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-018-26972-7. Competing Interests: The authors declare no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 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 Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted 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 license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2018 ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 12 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Scientific Reports Springer Journals

Phototoxic effects of two common marine fuels on the settlement success of the coral Acropora tenuis

Free
12 pages

Loading next page...
 
/lp/springer_journal/phototoxic-effects-of-two-common-marine-fuels-on-the-settlement-MNGUOwkAAK
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2045-2322
D.O.I.
10.1038/s41598-018-26972-7
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/scientificreports OPEN Phototoxic effects of two common marine fuels on the settlement success of the coral Acropora tenuis Received: 12 December 2017 1,2 1 1 3 F. Mikaela Nordborg , Florita Flores , Diane L. Brinkman , Susana Agustí & 1,4 Accepted: 23 May 2018 Andrew P. Negri Published: xx xx xxxx Coral reefs are at risk of exposure to petroleum hydrocarbons from shipping spills and uncontrolled discharges during extraction. The toxicity of petroleum hydrocarbons can substantially increase in the presence of ultraviolet radiation (UVR), therefore spills in shallow coral reef environments may be particularly hazardous to reef species. Here we investigated the sensitivity of coral larvae (Acropora tenuis) to dissolved hydrocarbons from heavy fuel oil (HFO) and diesel in the absence and presence of UVR. Larval settlement success decreased with increasing concentrations of dissolved HFO, and co- exposure to UVR doubled the toxicity: 50% effect concentrations (EC ) decreased from 96 (−UVR) to 51 (+UVR) total petroleum aromatic hydrocarbons (TPAH). Toxic thresholds for HFO were similar −1 to concentrations reported during marine spills: EC s of 24 (−UVR) and 15 (+UVR) µg l . While less toxic, diesel also reduced settlement and exhibited phototoxicity: EC s of 122 (+UVR) and 302 (−UVR) −1 µg l . This study demonstrates that the presence of UVR increases the hazard posed by oil pollution to tropical, shallow-water coral reefs. Further research on the effects of oils in the presence of UVR is needed to improve the environmental relevance of risk assessments and ensure appropriate protection for shallow reef environments against oil pollution. Petroleum hydrocarbons in marine environments. Petroleum hydrocarbons are considered among the most prominent pollution threats to marine environments ; however, the risks they pose to coral reef ecosys- 2,3 tems remain poorly understood . The environmental ee ff cts resulting from oil spills and uncontrolled discharges from extraction vary widely and are dependent on a large number of factors . Hydrocarbon concentrations in 5–7 marine environments have been measured following large scale spills , with dissolved hydrocarbon concentra- −1 5,8,9 −1 tions ranging between 22 and 189 µg l total polycyclic aromatic hydrocarbons (PAH) and up to 10,600 µg l total recoverable hydrocarbons (TRH) . Hydrocarbons can be retained within an ecosystem during spill events and, despite natural dilution and degradation, can remain detectable long aer t ft he spill has ended . e t Th oxicity of petroleum hydrocarbons to most marine species is predominantly related to the water soluble components, which largely consist of the monoaromatic hydrocarbons (MAH) and PAHs . PAHs, in particular, 12,13 are considered acutely toxic to aquatic biota , with toxic threshold concentrations for PAHs often orders of magnitude lower than those of MAHs . However, the overall toxicity of dissolved petroleum hydrocarbon from spills is also dependent on the relative concentrations of each component . Aromatic hydrocarbons in petroleum oils are classified as type I narcotic chemicals . Assuming the same mode of toxic action, the total toxicity of the water soluble components of an oil can be predicted using the narcotic target lipid model (NTLM), which sums the expected toxicity and concentration of each aromatic component (see Methods). Phototoxicity of PAHs. The toxicity of the dissolved aromatic mixtures that result from oil spills may increase in the presence of UVR due to the phototoxic contribution of some PAHs . PAH phototoxicity occurs through the formation of radical oxygen species and/or transformation of PAHs into more toxic photoprod- 12,13 16 ucts . Co-exposure of PAHs and UVR can increase the toxicity of individual PAHs 1000-fold ; however, the 1 2 Australian Institute of Marine Science, Townsville, 4810, Queensland, Australia. Department of Marine Sciences, University of Gothenburg, Göteborg, 40530, Sweden. Red Sea Research Centre, King Abdullah University of Science and Technology, Biological Environmental Science and Engineering Division, Thuwal, 23955, Saudi Arabia. AIMS@ JCU, Division of Research & Innovation, James Cook University and Australian Institute of Marine Science, Townsville, 4810, Queensland, Australia. Correspondence and requests for materials should be addressed to F.M.N. (email: m.nordborg@aims.gov.au) ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 1 www.nature.com/scientificreports/ potential for phototoxicity depends on the compound’s stability, radiation absorbance properties , the type of 12,17 17 UVR exposure and the exposed organism’s sensitivity . The potential for harmful effects to marine organisms is higher for UVA (320–400 nm) than UVB (280–320 nm) as the absorption maxima (hence, photoactivation) of 12,18 19 many PAHs fall within the UVA range , and UVB is more strongly attenuated in seawater . The penetration of UVR in marine environments is also dependent on a range of physical and biological factors , and tropical oligotrophic coral reefs may be at a particularly high risk of PAH phototoxicity as reef organisms are frequently 13,18 exposed to high solar radiation, including UVR . PAH phototoxicity is not always taken into consideration for risk assessments and management unless the ecological relevance, including UVR exposure and spectral profiles, 13,20,21 in the ecosystem have been characterised . However, it is increasingly recognised that a considerable propor- 18,22–24 tion of UVR penetrates to ecologically relevant depths in some marine environments, including coral reefs and that exposure to cumulative pressures, such as pollution and UVR, can result in increased environmental 13,18,25 impacts . Petroleum hydrocarbon and PAH toxicity to corals. Despite a renewed demand for marine hydrocar- bon toxicity research following the Deepwater Horizon spill , signic fi ant knowledge gaps on the potential ee ff cts 18,27–29 of hydrocarbon exposure to corals remain and the sensitivity of tropical marine species to hydrocarbons is 14,30 relatively understudied . Investigations into the effects of petroleum hydrocarbons, including PAHs, on coral −1 31 indicate that negative impacts can occur at concentrations as low as 2–20 µ g l total hydrocarbons (THC) . However, inconsistencies in exposure methodologies, toxic endpoints and reported toxicity values make compar- 2,15,18,32 isons between studies problematic . This issue is further compounded by the failure of many studies to pres - ent the chemical composition of treatment solutions, in particular the more soluble and toxic MAHs and PAHs . However, the studies that have been conducted show that hydrocarbons can be toxic to all life history stages of 33 7,33,34 coral and that larval settlement is generally more sensitive than fertilization , larval survival or the health and survival of adult corals . The sensitive larval settlement process is recognised as an ecologically relevant end- 29,36 point due to its importance in the recruitment process and subsequent maintenance of adult populations . The larval life stages of aquatic animals may also be at higher risk than adults to phototoxic effects due to their small size, oen t ft ransparent bodies, and time spent in shallow waters . Four studies have investigated phototoxic effects of petroleum hydrocarbons on corals and each indicated that their sensitivity to dissolved aromatics may increase with co-exposure to UVR (summarised in Table 1). However, the majority of laboratory studies exposing corals to hydrocarbons have not included co-exposure to UVR , so the impacts of hydrocarbon pollution on coral reefs may be significantly underestimated in the context of likely high UVR exposure in situ. To assess the potential for UVR to increase the sensitivity of coral larvae to spills of heavy fuel oil (HFO) and diesel we: (i) assessed UVR irradiance on one inshore and one mid-shelf reef on the Great Barrier Reef (GBR; Australia); (ii) characterised the chemical composition of the two fuels and their water accommodated fractions (WAFs); and (iii) predicted their narcotic toxicity to marine species using the NTLM. We then (iv) exposed larvae of the reef building coral Acropora tenuis (Dana, 1846) to HFO and diesel WAFs in the absence and presence of UVR (±UVR), at intensities similar to those encountered on the GBR, and assessed the ability of exposed larvae to successfully complete settlement and metamorphosis into sessile polyps following each treatment. Results Chemical analysis. Neat HFO consisted primarily of higher molecular weight hydrocarbons while neat diesel contained a higher proportion of BTEX (benzene, toluene, ethylbenzene and xylene) and other lower molecular weight hydrocarbons (Figs S-1 and S-2, Supplementary information). ∑PAH in HFO (constituting 99.6% of TPAH) was almost 10-fold higher than in diesel (79% of TPAH), whereas ∑BTEX was ~8-fold lower in HFO compared to diesel (see summary in Table 2 and detailed results in Table S-1, Supplementary informa- tion). Phenols were below the limit of quantitation in both oils (Table S-1, Supplementary information). Freshly −1 prepared 100% HFO and diesel WAFs contained similar TPAH concentrations (930 and 913 µg l , respectively), but the proportions of ∑BTEX and ∑PAH varied (Table S-2, Supplementary information). While HFO WAF contained almost equal concentrations of ∑BTEX and ∑PAH (52% and 48% of TPAH, respectively), ∑BTEX dominated the diesel WAF (98% of TPAH). The most abundant PAHs in the WAFs were naphthalene, alkylnaph- thalenes, fluorene and phenanthrene; the HFO WAF also contained acenaphthene and dibenzothiophene (Table S-2, Supplementary information). The concentration of TPAH in fuel WAFs decreased by up to 34% over the 48 h exposure (Tables 2 and S-2, Supplementary information). Comparisons of the observed and predicted concentra- tions of 1-, 2- and 3-ring compounds in the freshly prepared, undiluted WAFs (Tables S-2 and S-3, Supplementary information) as per Redman et al. indicated that no oil droplets were present in either HFO or diesel WAFs. Ultraviolet radiation intensities. e Th spectral profiles for both the Trunk Reef (mid-shelf ) and Esk Reef (inshore) sites showed that the largest decrease in irradiance occurred when light passed from air to water (Fig. 1a and b). Attenuation of UVR at Esk Reef was somewhat lower than at Trunk Reef for shallow measurement depths, despite the higher turbidity (0.1 and 0.8 nephelometric turbidity units, NTU, respectively). However, irradiance was attenuated less with depth at Trunk Reef and the irradiance for the deepest measurements was much higher at Trunk Reef than at Esk Reef (Fig. 1a and b). The depth which 10% of UVR irradiance penetrated to (Z10%) for 305 and 340 nm at Trunk Reef was ~6 and 7 m, respectively. On Esk Reef, Z10% was ~2.7 and 2.4 m for 305 and 340 nm, respectively. Experimental lights, when positioned 170 mm from the sensor, emitted UVA radiation similar to the irradi- ance observed at 1 m depth on Trunk Reef and Esk Reef (Fig. 1c). Average total UVA and UVB (280–400 nm) −2 radiation was 0.9 mW cm (SE = 0.16). Attenuation by the glass scintillation vials used in the experiment reduced the average absolute irradiance by ~17% for wavelengths between 300–400 nm (Fig. 1c; measured using ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 2 www.nature.com/scientificreports/ Chemical Phototoxic −1 Species Hydrocarbon UVR source analysis Endpoint Toxicity values (µg l ) effects Reference LC : 44 (−UVR) and 18 (+UVR) Anthracene Survival Yes: Anthracene Artificial UVR −1 µg l (A); no effect (±UVR) (P) (A), −2 Acropora tenuis 0.68 mW cm Overmans et Phenanthrene PAH (GC-MS) EC : 45 (−UVR) and 6.3 (+UVR) (larvae)* (10 h per 24 h, co- 50 al. (P) (48 h −1 Metamorphosis µg l (A); 91 (−UVR) and 66 No: Phenathrene exposure) −1 exposure) (+UVR) µg l (P) Australian North Ambient solar Acropora tenuis West Shelf BTEX, PAH, IC : 339 (−UVR) and 132 (+UVR) −2 50 29 4.5–6.8 mW cm Metamorphosis Yes Negri et al. −1 (larvae) condensate TRH (GC-MS) µg TPAH l (2 h, co-exposure) (24 h exposure) Carmen Fluoranthene Ambient solar LC : 435.2 (−UVR; lower side of Porites divaricata Mortality/ Guzmán −1 (4.5 h (ambient L:D cycle Nil branches) and 31.4 µg l (+UVR; Yes (adult) Bleaching Martínez et exposure) for 6 d, seq. exposure) upper side of branches) al. Fungia scutaria F. scutaria (+UVR): 100% mortality Yes: −1 (larvae) ≤1 h aer exp ft osure to 48 µg l . F. scutaria Mortality M. verrucosa Montipora M. verrucosa (+UVR): bleached −1 P. damicornis verrucosa (adult) 24 h aer exp ft osure to ≥16 µg l . Ambient solar 0.41– Pyrene (2 h Peachey & −2 1.4 mW cm (up to Nil Pocillopora exposure) P. damicornis (+UVR): bleached Crosby 8 h, seq. exposure) −1 damicornis (adult) 24 h aer exp ft osure to 48 µg l . No: Bleaching P. compressa (+UVR): no effect. P. compressa Porites compressa (adult) All species (−UVR): no effect. Montipora Artificial UVR 0.98– verrucosa (adult) Pyrene (2 h −2 1.0 mW cm (up to Nil Bleaching Not reported No (both species) exposure) Pocillopora 8 h, seq. exposure) damicornis (adult) Table 1. Summary of previous studies of the phototoxic effects of petroleum hydrocarbons on scleractinian corals. Study methodology, species tested, chemical analysis performed (if applicable), toxic endpoint and main results shown for each study. If no threshold values or concentrations are presented no effect was observed. Seq = sequential, organisms first exposed to pollutant followed by exposure to UVR while kept in clean FSW. GC-MS = gas chromatography-mass spectrometry. Nil = no analysis reported. *Stress response-related gene expression and enzyme activity were also investigated (see reference for further details). Jaz spectrometer calibrated according to manufacturer’s recommendations) giving a calculated total irradiance −2 between 280–400 nm of approximately 0.75 mW cm inside the scintillation vials. Larval settlement assays. Temperature was maintained at 27.6 ± 1.4 °C (mean ± SD) in the experiments while photosynthetically active radiation (PAR) in the +UVR and −UVR treatments averaged 0.95 ± 0.10 and −2 −1 1.23 ± 0.10 mW cm , respectively. Dissolved oxygen concentration averaged 7.9 ± 0.33 mg l with all replicates −1 maintaining concentrations >7.0 mg l for the duration of the exposure period while pH and salinity averaged 8.1 ± 0.06 and 37.0 ± 0.32 psu, respectively. A. tenuis larvae in control treatments were observed to actively swim throughout the exposure (Fig. 2a), but swimming behaviour was not assessed in the fuel exposure treatments. In the control treatments, an average of 73% (SE = 4) larvae underwent settlement in the presence of CCA chips within 24 h (Fig. 2c). This level of settlement success did not change in the presence of UVR with 77% (SE = 3.5) of larvae successfully undergoing settlement. Average larval settlement ≥70% in control treatments was consid- ered indicative of a normal response to settlement inducers based upon several previous studies using CCA or 29,37,38 extracts of CCA to initiate settlement of Acropora spp . Heavy fuel oil toxicity. HFO WAF inhibited larval settlement in both the absence and presence of UVR −1 (Fig. 3a,c and Table 3). Little or no effect on settlement success was observed at low concentrations (< 10 µ g l TPAH); above which, settlement decreased with increasing TPAH concentration (Fig. 3). The toxicity of HFO −1 WAF was enhanced in the presence of UVR resulting in a ~50% decrease of the EC from 96 to 51 µ g l TPAH (95% confidence intervals did not overlap; Table  3). The toxic threshold value (EC ) also reduced in the presence of UVR (Table 3). Obvious mortality (disintegrating cell membranes) was only observed in the highest TPAH −1 treatment (890 µg l ), in the presence of UVR, at the end of the 48 h exposure. No other visible effects on larvae, in either control or HFO WAF treatments, were observed and surviving larvae in the highest HFO treatments exhibited normal morphology (Fig. 2a and b). At 72 h (following the 24 h settlement period) unattached larvae and attached juvenile polyps in control treatments retained normal morphology (Fig. 2c). At this point the frequency and severity of abnormalities increased with increasing TPAH concentration in attached juvenile polyps (asym- metrical or underdeveloped recruits; Fig. 2d) and non-settled larvae (bumps, deformities and necrosis; Fig. 2e) for both ±UVR treatments. Severe deformities were observed in unattached larvae at TPAH concentrations as low as −1 28 µ g l (+UVR) and a substantial proportion of larvae were immobile and/or dead in the highest concentration −1 treatment (890 µ g l ). Most of the successfully attached juvenile polyps underwent complete metamorphosis in the absence of UVR; however, larvae exposed to HFO WAFs appeared to develop more slowly than expected and some had only undergone partial metamorphosis in higher concentration treatments at the time of assessment. ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 3 www.nature.com/scientificreports/ −1 −1 [WAF] (µg l ) [Neat fuel] (mg kg ) TU TU WAF Neat fuel ∑BTEX ∑PAH TPAH ∑BTEX ∑PAH TPAH ∑BTEX ∑PAH TPAH ∑BTEX ∑PAH TPAH 498 440 938 −UVR 190 50494 50684 0.03 0.43 0.46 0.01 0.91 0.92 (+4%) (−1%) (+2%) HFO 478 414 892 +UVR 0.03 0.40 0.43 (−4%) (−13%) (−8%) 745 14 759 −UVR 1491 5723 7214 0.04 0.01 0.05 0.07 0.12 0.19 (−33%) (−57%) (−34%) Diesel 767 16 783 +UVR 0.05 0.02 0.06 (−28%) (−41%) (−29%) Table 2. Time-averaged concentrations of ∑BTEX, ∑PAH and TPAH in undiluted fuel WAFs, ∑BTEX, ∑PAH and TRH in neat fuels, and toxic units (narcosis) calculated from predicted and observed fuel WAF concentrations. Time-averaged concentrations per light treatment calculated from concentrations measured at t and t for HFO and diesel WAFs; % change in concentrations aer 48 ft h indicated in brackets. TUs 0h 48h −1 calculated using an average CTLBB value (86.8 µmol g octanol; n = 15) and aqueous concentrations of BTEX and PAH observed in fuel WAFs (TU ) or predicted from neat fuel oil concentrations (TU ). WAF Neat fuel TPAH = total petroleum aromatic hydrocarbons, ∑PAH = sum of individual PAH concentrations, ∑BTEX = sum of benzene, toluene, ethylbenzene and xylene concentrations. For full analytical results see Tables S-1 and S-2, Supplementary information. Diesel toxicity. Diesel WAF also inhibited larval settlement, but little effect was observed at low to mod- −1 erate concentrations (<100 µ g l TPAH) regardless of UVR treatment (Fig. 3). In the absence of UVR, only −1 the highest treatment concentration (759 µ g l TPAH) inhibited settlement success. An approximate EC of −1 ~1300 µ g l TPAH (without confidence intervals), was estimated by extrapolation (Fig.  3b and Table 3). UVR −1 greatly enhanced the toxiciy of diesel WAF and decreased the EC to 494 µ g l TPAH (Fig. 3d and Table 3). At −1 moderate to high TPAH concentrations (≥200 µ g l ) underdeveloped juvenile polyps were observed in the pres- −1 ence of UVR, and the few attached juvenile polyps observed in the highest concentration (782 µg l TPAH) were either underdeveloped or abnormal. In the absence of UVR, some underdeveloped and malformed attached juve- nile polyps were also observed; however, fully metamorphosed polyps were still present at the highest concen- −1 tration tested (758 µg l TPAH). Some minor abnormalities (e.g. bends and lumps) were observed in unattached larvae from ± UVR treatments but with no apparent relationship to TPAH concentration. Modelled narcotic toxicity. The narcotic toxicities of the undiluted diesel and HFO WAFs used in the present study were estimated using the NTLM . This modelling yielded toxic units (TU) based on both the predicted and measured concentrations of aromatics in undiluted WAFs. TUs can be used to compare the tox- icity of complex mixtures; the greater the TU, the more toxic the solution (see Methods). The narcotic toxicity of HFO and diesel WAFs, calculated using WAF concentrations predicted from neat fuel compositions (TU ; Neat fuel Table 2) , was primarily attributable to PAHs (TU > TU ; Table 2). For undiluted HFO WAF, 0.92 TU ∑PAH ∑BTEX Neat indicates that close to 50% mortality would be expected, while TUs, calculated by applying measured WAF fuel concentrations to the NTLM (TU : 0.46 −UVR and 0.43 +UVR), were approximately half that value (Table 2). WAF The total narcotic toxicity for undiluted diesel WAF indicated that relatively low mortality would be expected following exposure, when calculated using predicted WAF concentrations (TU : 0.19), or close to zero mor- Neat fuel tality, when measured WAF concentrations were applied (TU : 0.05 −UVR to 0.06 +UVR). WAF Discussion Co-exposure to UVR considerably enhanced the negative impacts of the water-soluble fractions of two petroleum fuels on the settlement success of coral larvae. This phototoxicity was evident under UVR intensities equal to or lower than expected exposures in shallow reef habitats, and the toxic thresholds (EC ) for dissolved aromatics 7–9 occurred at concentrations below those detected aer m ft ajor accidental spills or releases . These results indicate that by ignoring phototoxicity, the hazards posed by oil spills to coral larvae may be substantially underestimated in shallow-water tropical reef systems. Coral larval settlement was very sensitive to HFO WAF with threshold concentrations (EC and EC ) lower 10 50 −1 5,8,9 than or similar to previously reported concentrations during and after oil spills (42–189 µ g l ∑PAH and −1 9 −1 22 µ g l TPAH ). The EC and EC values (24 and 96 µ g l TPAH, respectively; −UVR) were also lower than 10 50 those reported for inhibition of larval settlement in the same species in 24 h exposures to natural gas condensate −1 29 in the absence of UVR (103 and 339 µ g l TPAH, respectively) . Differences in sensitivity are likely to be due to the longer duration of the current exposure and the higher proportion of more toxic PAHs in the HFO WAF. Conversely, the sensitivity of A. tenuis larvae to diesel WAF in the absence of UVR was less than that reported for −1 either HFO or condensate, potentially due to the lower concentrations of total dissolved PAHs (14 µ g l ) in the −1 −1 29 undiluted diesel WAF compared to HFO (440 µg l ; Table 2) and condensate (107 µg l ) WAFs. The sensitivity of coral larvae observed here is difficult to compare with other studies, which reported die ff rent exposure types, endpoints and measures of petroleum hydrocarbon concentration in WAFs (summarised in Negri, et al. and Turner, et al. ). Nevertheless, studies exposing coral larvae to hydrocarbons in the absence of UVR reported neg- −1 7,33,40 ative effects on coral settlement at THC concentrations between 82 and 620 µ g l and the sensitivity of coral larvae is likely to be greater than that of other life stages . ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 4 www.nature.com/scientificreports/ Figure 1. Penetration of ultraviolet radiation (UVR) on Trunk Reef (a) and Esk Reef (b) as well as a comparison (c) of exposure intensity and spectrum of artificial UVR used during settlement toxicity assays and UVR observed in situ on the Great Barrier Reef (GBR; Australia) during spring. Full spectrum measurements of UVR in air and at 0.1, 1, 2, 3, and 3.8 m depth at a mid-shelf (Trunk reef ) and inshore (Esk reef ) reef on the GBR in October 2016. Total irradiance values calculated using the percentage reductions in light intensity recorded for each wavelength at each depth (in relation to measurements in air) and total irradiance measurements in air on a clear day (cloud coverage <5%). Comparison of UVR intensity and spectrum emitted from fluorescent tubes used in settlement toxicity assays, calculated UVR exposure inside scintillation vials and UVR exposures observed at 1 m depth on the GBR during spring. ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 5 www.nature.com/scientificreports/ Figure 2. Photomicrographs of A. tenuis planulae larvae exposed to water accommodated fractions (WAFs) of heavy fuel oil (HFO) in the absence (−UVR) or presence (+UVR) of ultraviolet radiation as well as juvenile −1 polyps (following settlement). Larvae exposed to (a) filtered seawater (0 µg TPAH l ), (b) approximately 900 µg −1 TPAH l aer 48 ft h of exposure as well as juvenile polyps, attached and unattached larvae treated with (c) FSW −1 −1 −1 (0 µg TPAH l ), (d) 115 µg TPAH l and (e) approximately 900 µg TPAH l HFO WAF aer 48 ft h of exposure, introduction of settlement inducer and a 24 h settlement period (a total ~72 h aer exp ft eriment start). All concentrations in µg TPAH l-1. TPAH = total petroleum aromatic hydrocarbons. Figure 3. Concentration-response curves for coral larval settlement following exposure to heavy fuel oil (a and c) and diesel (b and d) water accommodated fractions (WAF) in the presence (blue) and absence (green) −1 of ultraviolet radiation (µ g TPAH l ). Model mean (solid line) and 95% confidence intervals (shaded area) for quasibinomial GLMs fitted for the settlement success data of each treatment combination as well as observed −1 settlement success for each replicate (open ring) used in model fitting. All concentrations in µ g l . n = 63, HFO+/−UVR n = 65, n = 64. TPAH = total petroleum aromatic hydrocarbons. diesel−UVR diesel+UVR −1 The toxicity of both HFO and diesel WAFs doubled (EC : 95 to 51 and 1300 to 494 µ g l TPAH ±UVR, respectively) in the presence of ecologically relevant intensities of UVR. This observed phototoxicity is consistent with the 2.5-fold increase in toxicity of natural gas condensate WAF in the presence of UVR to the same lar- val species in 24 h WAF exposures, based on measured TPAH concentrations . This toxicity increase is similar between studies despite differences in UVR exposure profiles, where Negri, et al . applied a single, higher inten- −2 sity UVR dose for a shorter period of time (2 h, summing to 39 W cm ) compared to this study (see Methods). Overmans et al. recently reported that the inhibition of coral larval metamorphosis by the PAH anthracene ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 6 www.nature.com/scientificreports/ EC EC Approximate r Deviance df 10 50 24 96 −UVR 0.748 146.12 61 (9–39) (75–116) HFO 15 51 +UVR 0.767 137.03 61 (6–23) (41–62) 302 ~1300* −UVR 0.062 196.07 63 (51–552) (253–2410) Diesel 122 494 +UVR 0.423 165.97 62 (50–194) (371–617) Table 3. Concentrations of TPAH in fuel WAFs which inhibited 50% (EC ) and 10% (EC ) of larval settlement 50 10 in the absence (−UVR) and presence (+UVR) of ultraviolet radiation. Effect concentrations calculated from settlement success data fitted with quasibinomial GLMs with approximate r , deviance and degrees of freedom (df ) shown for each treatment combination. 95% confidence intervals in brackets where applicable. All concentration −1 values in µ g TPAH l . TPAH = total petroleum aromatic hydrocarbons. *Extrapolated from fitted model. −1 increased 7-fold in the presence of UVR (EC : 45 −UVR to 6.3 +UVR µg l ), while there was little influence of UVR exposure on the toxicity of phenanthrene. The impacts of PAH and UVR on larval metamorphosis were also detected at lower PAH concentrations than other sub-lethal biomarkers investigated following the 48 h exposures 18 17 (including 10 h UVR exposure per day) . Peachey and Crosby observed 100% mortality of F. scutaria larvae −1 aer 1 ft h exposures to 48 µ g l pyrene followed by exposure to ambient solar radiation (including UVR). Strong phototoxic ee ff cts have also been demonstrated for adult P . varicata corals where UVR increased the toxicity of uo fl ranthene by approximately 14-fold within 72 h following a 4.5 h exposure to this PAH . It is not clear whether the main influence of UVR is to generate more toxic transformation products, or whether the additional toxicity is caused by elevated oxidative stress within the larvae as PAHs, excited by UVR, decay back to their ground states 12,13 (photosensitisation) . However, phototoxicity in aquatic organisms is generally considered to be caused by photosensitisation . While UVR increase the toxicity of some individual PAHs by more than 10 times, fuel WAFs are comprised of many aromatics and several of these, including BTEX, the naphthalenes and phenanthrene, are not considered phototoxic . Consequently, very large increases in oil or fuel WAF toxicity in the presence of UVR are unlikely. In addition to reductions in settlement success, exposure to fuel WAFs also caused changes to larval morphol- ogy and development. Slowed onset and completion of metamorphosis, as well as increased instances of morpho- logical abnormalities, were observed in larvae exposed to WAFs. The process of attachment and metamorphosis following addition of settlement inducers was slower at moderate-high TPAH concentrations, with only partial metamorphosis achieved in 24 h, similar to what would generally be expected 6–12 h ae ft r settlement induction (A.J. Heyward, personal communication, Dec 2016). Delayed onset of metamorphosis, abnormal development or deformations following exposures to petroleum hydrocarbons have also been reported for larvae and juvenile 28,29,42 polyps from other coral species . The development of abnormal morphologies in coral larvae may indicate narcosis or more specific toxic effects of petroleum hydrocarbons on cellular developmental processes. The results of this study clearly demonstrate the strong phototoxic effects of the water-soluble fractions of common marine fuels on coral larval settlement in the presence of environmentally realistic UVR exposures. The current study applied UVR intensities slightly lower than the intensities measured at 1 m depth on a mid-shelf (Trunk Reef ) and inshore reef (Esk reef ) on the Central GBR. These intensities are also consistent with previous 23 22,24 measurements on the GBR , and other coastal and oceanic locations . The penetration of UVR in marine environments is primarily ae ff cted by chromophoric dissolved organic matter (CDOM) and, for some systems, 22,24 particulate matter such as plankton and detritus . Attenuation is therefore likely to be lowest in clear-water oli- gotrophic coral reef environments, as measured at the mid-shelf site at Trunk Reef, compared to the more turbid waters of the inshore Esk Reef. Barron et al. found that at depths greater than 0.5 m attenuation of light varies between habitats, emphasising the need to accurately estimate the intensity and wavelengths experienced by each species or ecosystem when investigating the influence of UVR on the toxicity of pollutants. Coral gametes, embryos and larvae developing at the water surface may also be exposed to substantially higher UVR intensities in situ than applied in the present study, potentially reducing the toxic threshold values further. Moreover, these results suggest that the hazard (hence risk) posed by aqueous petroleum hydrocarbons to shallow-water tropical coral reefs will be underestimated if phototoxic activation by UVR is not taken into consideration. The NTLM for predicting narcotic mortality underestimated the impacts of HFO WAF in comparison to the observed effects on larval settlement. 50% larval mortality was predicted for HFO WAFs by the NTLM at −1 ~900 µ g l TPAH (TU ); however, coral larval settlement was reduced by 50% at far lower concentrations Neat fuel −1 (EC s 96 to 51 µg l TPAH ±UVR). It has been reported that the NTLM can underestimate the toxicity of some hydrocarbon mixtures , and these underestimations may result from the assumption of a simple narcotic mode of action for all aromatic components. The observed discrepancies could partially result from the contribution of other toxic mechanisms, especially those that specifically affect coral metamorphosis from pelagic planula to sessile juveniles, as suggested by Negri et al. . Furthermore, the sub-lethal endpoint of settlement inhibition (while ecologically relevant ) is by definition more sensitive than mortality, which is typically used to generate the CTLBB values used in the NTLM. Determining CTLBB values for both larvae and adults of key coral reef spe- cies would improve species-specific toxicity predictions at different life stages and assist in ranking the potential risks posed by crude oils and petroleum products . It is also possible that other components (e.g. photoproducts) that were not measured may have influenced the toxicity. Although a phototoxic target lipid model was recently ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 7 www.nature.com/scientificreports/ presented , its use is currently limited to estimation of the acute toxicity of individual PAHs, and further devel- opment is necessary before it can be applied to complex hydrocarbon mixtures or chronic exposures. Even in well managed and protected areas, groundings of large vessels and petroleum hydrocarbon releases 43–45 from offshore extraction facilities have occurred in the last decade . Large spills as well as chronic hydrocarbon 7,46,47 pollution can lead to degradation of adult coral health and changes to reef composition with recovery taking more than 10–20 years . The slow recovery of these reefs is likely to be at least partially due to long term effects of hydrocarbon contamination on recruitment processes, including larval settlement ; used as a sensitive toxicity endpoint in this study. The overlap between a large oil spill and the coral recruitment window can be substantial, with water from some spill sites remaining phototoxic to invertebrate embryos for up to 13 d and larvae of some coral species remaining in the water column for up to 3 months (as reviewed by Jones, et al. ). The increase in toxicity of dissolved aromatics from HFO by UVR exposure resulted in low toxic thresholds, underscoring the potential hazard to corals posed by phototoxic compounds found in petroleum oils and fuels. Previous assess- ments may therefore have substantially underestimated the risks posed by oil and petroleum product spills on shallow-water, tropical coral reefs by not accounting for interactions with environmental factors such as UVR. Further research into the effects of petroleum hydrocarbons on more tropical reef organisms, including potential interactions with UVR and other stressors, is needed to more effectively quantify these risks. Methods Coral collection and larval cultures. Gravid colonies of Acropora tenuis, a reef-building coral common throughout the Pacific Ocean, were collected by hand on SCUBA from Magnetic Island (October 2016, 19.157°S, 146.861°E), GBR, under Great Barrier Reef Marine Park Authority Permit G12/35236.1. Colonies were placed in flow-through seawater of ambient temperature and transported to the National Sea Simulator, Townsville, within 24 hours. On arrival colonies were transferred to 70% shaded flow-through outdoor holding tanks and kept at temperatures equivalent to the collection site (27 °C) until spawning. When showing signs of setting colonies were isolated and gametes collected by gentle scooping. th Larval cultures (6 parental colonies, 95% fertilisation) were initiated on the 19 October 2016. Cultures were maintained at 27 °C and densities <500 larvae per l at ~27 °C in round, 500 l fibreglass rearing tanks with cone-shaped bases. Flow-through seawater (1.5 turnovers per day) was 1 µ m-filtered and a round air stone at the base of each tank provided aeration and created a gentle curtain of bubbles to keep larvae from a submerged cylindrical mesh filter (15 h × 6 d cm, 100 µm) at the outflow. Preparation of fuel water accommodated fractions. Heavy residual fuel oil (HFO, International Bunker Supplies Pty Ltd, Gladstone, Australia) and automotive diesel (Puma Energy Australia, Fortitude Valley, Australia) WAFs were prepared in capped, solvent rinsed aspirator bottles (5 l) using 0.45 µ m filtered seawater −1 51 (FSW; pH 8.1, salinity 37.0 psu) loaded at 20 g fuel l with a 20% headspace . Solutions were protected from light, stirred at 50 rpm for 18 h without a vortex and allowed to settle for 30 min prior to use . Ten dilutions (0, 0.4, 0.8, 1.6, 3.125, 6.25, 12.5, 25, 50 and 100% WAF) were prepared from the fresh undiluted (100%) WAF using 29,52 0.45 µm FSW and used within 3 h in the assays below. Fuel phototoxicity assays. Acropora spp. larvae reach competency to settle and undergo metamorphosis aer ~4 d ft . HFO and diesel static non-renewal exposure experiments over 48 h were performed on 7- and 8-d old A. tenuis larvae, respectively. 10–14 larvae and 20 ml of WAF were gently added to 6 replicate glass vials for each WAF concentration and ±UVR treatment combination (12 replicate vials for 0% WAF controls). Vials were tightly capped with approximately 10% headspace to allow for oxygen exchange. Vials were randomly placed on their side in trays inside temperature controlled, orbital shaker incubators (Thermoline Scientific, Australia) at 80 rpm to ensure larvae did not settle during the exposure period. One set of vials (−UVR), n = 66, was placed in −2 an incubator fitted with actinic LEDs emitting 1.23 mW cm photosynthetically active radiation (PAR; Aqualina Blue 450 nm, 10,000 K and 420 nm Actinic LED strips, The Aquatic Life Product Company, Willawong, Australia). e s Th econd set of vials (n = 66) was placed in a second incubator in the presence of both actinic light and UVR −2 (+UVR). This incubator (+ UVR) was t fi ted with identical actinic LEDs as the first incubator (PAR: 0.95 mW cm ) as well as three sets of UV-emitting fluorescent tubes (each set consisting of one Deluxlite Blacklight Blue 18 W −2 and one Reptile One UVB 5.0 18 W T8 fluorescent tube) emitting 0.75 mW cm UVR. Fluorescent tubes emitting predominantly in the UVA spectrum were chosen due to the high UVB attenuation of seawater (See Ultraviolet radiation intensities section below on how fluorescent tube irradiance was characterised, and Fig.  1c for compari- son of experimental lights, light attenuation of vials used for experimental exposures and in situ intensities of UVR −2 on the GBR). PAR was provided on a 12:12 h L:D cycle and UVR on a 6:18 h L:D cycle (total irradiance 16.1 W cm −1 d ). An additional 30 vials containing undiluted WAF, but no larvae, were also placed in each incubator to allow for the collection of chemical samples at the end of the 48 h exposure. The positions of vials within each incubator were exchanged randomly throughout each experiment to minimize variation in light exposure. Temperature was continuously logged (Onset HOBO temperature logger, Onset Computer Corp., Massachusetts, USA) while pH, salinity and dissolved oxygen concentrations were measured at the beginning and end of each experiment. At the end of the 48 h exposure approximately 600 ml of undiluted WAF was pooled, from the 30 additional vials (not containing larvae), for chemical analysis (see Chemical analysis below). Following exposure to fuel WAFs in the presence or absence of UVR, larvae were transferred with 10 ml treatment solution directly to individual wells in 6-well cell culture plates (Nalge Nunc Int., Denmark). Larvae were presented with a settlement inducer consisting of 5 × 5 mm live chips of the crustose coralline algae (CCA) Porolithon onkodes . Settlement success was assessed as percentage of larvae which attached and underwent at least early metamorphosis (i.e. became firmly attached and flattened into a disc shape ) aer ~24 ft h incubation at 27 °C. Average settlement success ≥70% in controls was considered indicative of a normal response to settlement ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 8 www.nature.com/scientificreports/ 29,33,37,54,55 inducers based on several previous studies using CCA or extracts of CCA to initiate settlement . Additional notes regarding deformities, ratio of fully (as outlined by Heyward & Negri ) to partially metamor- phosed recruits as well as failure to attach were also made for each treatment combination. Settlement data analysis. Settlement data was fitted with binomial generalized linear models (GLMs) with a logit link function using the R stats-package (R version 3.4.1 ) to model the settlement success of A. tenuis larvae in response to treatment concentration (fixed numerical factor) for each light treatment and fuel type com- bination. The fitted models were validated by plotting the simulated residuals against fitted values and ensuring no individual values influenced the model fits’ disproportionally. Quasibinomial GLMs were fitted where diag- nostics indicated overdispersion. An r analogue was calculated using the deviance of the fitted models with and without the fixed numerical factor (equation (1)) to approximate the goodness-of-fit for each model .   deviance 2 Full model   Approximater =− 1   deviance   Null model (1) EC and EC values with 95% confidence intervals (CI) were interpolated from model mean values and 95% 10 50 CI (adapted from Venables & Ripley ). Predicted model mean values and 95% CI were exported and graphical outputs produced using GraphPad Prism (version 7.02, GraphPad Software Inc., CA, USA). The high levels of replication used allowed the identification of outliers in the dataset which were excluded. These comprised three HFO FSW controls and one diesel low concentration replicate where CCA chips induced 0% and 7% settlement, respectively, likely due to misidentification of a few of the live CCA chips. Chemical analysis. Samples of freshly prepared, undiluted WAF (“100% WAF”) were collected for chemical analysis at the beginning of each experiment (t ). Undiluted WAF was also added to 30 vials per UVR treatment 0h and incubated simultaneously with vials containing larvae to ensure that the undiluted WAF was exposed to the same experimental conditions as the test solutions. At the end of the exposure period (t ), the undiluted WAF, 48h in vials containing no larvae, was pooled and sampled for chemical analysis. For BTEX analysis, samples (40 ml) were collected in amber glass vials with open hole caps and PTFE septa. For all other analyses, single samples (500 ml) were collected in amber glass bottles with PTFE-lined caps. All samples were acidified to pH 2 using 6 M hydrochloric acid and stored at 4 °C until shipped to ChemCentre (Perth, Australia) for analysis as previ- ously described by Negri et al. . Briefly, WAFs were analysed directly for BTEX using Purge and Trap GC-MS in full scan mode (USEPA method 8260). The 500 ml WAF samples were extracted three times with dichlo- romethane (DCM) and the combined extracts analysed for PAH and alkylated PAH using GC-MS in SIM or scan mode, and TRH using GC-FID (USEPA method 8270). Neat HFO and diesel were diluted in DCM and ana- lysed for TRH, BTEX, PAH/alkylated PAH and phenols using the same methodology, except additional surrogate (2-fluorophenol, phenol-d5 and 2,4,6-tribromophenol) and internal (1,4-dichlorobenzene-d4) standards were added to the samples prior to analysis. To profile their general composition, the neat oils (1 µl) were subjected to whole oil analysis using GC-MS and hydrocarbons were identified through comparison with a pre-characterised reference oil. Ultraviolet radiation intensities. In situ UVR irradiance on the Central GBR during spring was assessed through full spectrum measurements of UVR in air and at five depths at two reefs. Esk Reef (18.775°S 146.522°E), located in the Palm Islands-group (QLD), was selected as a representative inshore reef site while Trunk Reef (18.329°S, 146.846°E) was selected as a representative clear-water, mid-shelf reef site. Three replicate light intensity measurements for wavelengths between 300–400 nm were performed on SCUBA using a Jaz handheld spectrome- ter (Ocean Optics, Inc., Florida, USA) and a 5 m fibre optic cable (CPATCH-5074768, Ocean Optics, Inc., Florida, USA) with a planar irradiance collector for underwater use (HOBI Labs, Inc., USA) following Shick et al. (1996). Measurements were made with the sensor positioned vertically at 0, 0.1, 1, 2, 3 and 3.8 m depth. Measurements were performed close to solar noon on the mid-shelf (14:17–14:20 on 12 October 2016) and inshore reefs (13:55– 14:04 on 14 October 2016), respectively. Intensity data collected was used to calculate the average relative decrease in radiation for wavelengths between 300–400 nm with depth. Total UVA and UVB radiation above the surface was recorded (Solarmeter Model 5.0 UVA + UVB meter, Solartech Inc., Pennsylvania, USA) and turbidity meas- urements were performed (90FL-T, TPS Pty Ltd, Brendale, QLD, Australia). Cloud coverage was low (<5%) with 17 km/h E winds (BOM, 2016) and turbidity of 0.1 NTU during mid-shelf measurements with medium-high cloud coverage (~80%), 33 km/h ENE winds (maximum gusts 46 km/h ENE; BOM, 2016b) and 0.8 NTU turbidity during inshore reef measurements. Theoretical irradiance at each depth was calculated using full spectrum measurements of natural sunlight close to solar noon on a low cloud coverage day (<5%). Measurements were made using a calibrated Jaz spectrometer and a 250 mm UVR compatible fibre optic cable (QP600-025-UV, Ocean Optics, Inc., Florida, USA) aimed directly towards the sun. Z10% values at 305 and 340 nm were estimated by calculating the irradiance corresponding to 10% of surface irradiance (in air) for each reef site and the negative linear relationship between irradiance and measurement depth of measurements made at Trunk Reef and Esk Reef. Full spectrum measurements of radiation emitted by the UVR fluorescent tubes (Deluxlite Black Light Blue 18W; Reptile One UVB 5.0 18W), used in +UVR larval settlement assays, were performed using the same cali- brated Jaz spectrometer and 250 mm UVR compatible fibre optic cable used to quantify the UV radiation of nat- ural sunlight. Full spectrum measurements were made at approximately the same distance as sample vials during experimental exposures (170 mm) in v fi e separate positions relative to the three sets of uo fl rescent tubes. Total UVA and B radiation measurements were also performed (Solarmeter Model 5.0 UVA + UVB meter, Solartech Inc., Pennsylvania, USA) for the five replicates. Additionally, the attenuation of UVR, between 300–400 nm, emit- ted from fluorescent tubes by scintillation vial glass was estimated (calibrated Jaz spectrometer). Measurements ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 9 www.nature.com/scientificreports/ were made 200 mm from UVR fluorescent tubes (Deluxlite Black Light Blue 18W; Reptile One UVB 5.0 18W) through the base of a 20 ml scintillation vial. The average total UVA and UVB exposure of larvae inside scintilla- tion vials was calculated using measurements performed with the Solarmeter model 5.0 and the average attenua- tion of scintillation vial glass between 300–400 nm. Toxic unit calculations for narcotic toxicity. e t Th oxicity of aromatics to aquatic organisms is depend- ent on the partitioning of dissolved compounds between water and lipids. The narcotic toxicity of hydrocarbon mixtures can be estimated using the narcotic target lipid model (NTLM) which combines the octanol-water partitioning coefficients (K ) of all dissolved components in the solution, their measured (or expected) aque- OW ous concentrations and the critical lipid body burden (CTLBB) of the organism of interest (where the CTLBB is the amount of the compound dissolved in the organism’s tissues which causes a specific toxic effect; e.g. 50% mortality) . The NTLM is useful for assessing the relative risks posed by different oil and fuel types to aquatic organisms, but may be less useful if the CTLBB is unknown or may underestimate toxicity if other toxic modes of action are important for a given species . e Th dissolved composition of petroleum hydrocarbons in a WAF can be measured (Table S-2, Supplementary information) or modelled by applying an oil solubility model (e.g. PETROTOX) to an oil of known composition . Toxic units (TU) for each constituent in a WAF are defined as the ratio between the concentration (C ) and the WAF, i 50% critical effect level (LC , EC or IC ) of each constituent (i) in a solution, and can be used to enable com- 50,i 50,i 50,i parisons of studies using differing experimental designs and hydrocarbon compositions . Assuming the same mode of action (narcosis) for all MAHs and PAHs, the TUs for each component of a WAF containing a mixture of PAHs and MAHs, are considered additive and can be summed to estimate the TU of the WAF according to equation (2). C C C C WAFW ,1 AF ,2 WAFW ,3 AF ,i TU = +++  + WAF ∑ EC EC EC EC (2) 50,1 50,2 50,3 50,i A TU of 1 indicates that the mixture is predicted to be toxic and will ae ff ct 50% of exposed organisms . In this study, expected TUs (TU ) were calculated as per Redman and Parkerton , by applying the observed Neat fuel −1 concentrations of individual compounds in neat HFO and diesel and the average LC CTLBB (86.8 µ mol g 14,61 octanol) of 15 marine and estuarine organisms . An average marine CTLBB was used as no CTLBB currently exist for acroporid corals. Additionally, observed TUs (TU ) for each light treatment were calculated using WAF measured concentrations of individual compounds in undiluted HFO and diesel WAFs and the same CTLBB value. Data availability. Data are available upon request. References 1. Haapkylä, J., Ramade, F. & Salvat, B. Oil pollution on coral reefs: a review of the state of knowledge and management needs. Vie Milieu 57, 95–111 (2007). 2. Turner, N. R. & Renegar, D. A. Petroleum hydrocarbon toxicity to corals: A review. Mar. Pollut. Bull. 119, 1–16, https://doi. org/10.1016/j.marpolbul.2017.04.050 (2017). 3. Kroon, F. et al. Identification, impacts, and prioritisation of emerging contaminants present in the GBR and Torres Strait marine environments. Report to the National Environmental Science Programme. Reef and Rainforest Research Centre Limited, Cairns (138 pp.) (2015). 4. Neff, J. M., Ostazeski, S., Gardiner, W. & Stejskal, I. Effects of weathering on the toxicity of three offshore Australian crude oils and a diesel fuel to marine animals. Environ. Toxicol. Chem. 19, 1809–1821 (2000). 5. Boehm, P. D., Neff, J. M. & Page, D. S. Assessment of polycyclic aromatic hydrocarbon exposure in the waters of Prince William Sound aer t ft he Exxon Valdez oil spill: 1989–2005. Mar. Pollut. Bull. 54, 339–356 (2007). 6. AMSA. Montara well release - Operational monitoring study O2: Monitoring of oil character, fate and effects. (Australian Maritime Safety Authority, Australian Government, https://www.amsa.gov.au Accessed on 13.04.2017, 2010). 7. Hartmann, A. C. et al. Crude oil contamination interrupts settlement of coral larvae aer dir ft ect exposure ends. Mar. Ecol. Prog. Ser. 536, 163–173 (2015). 8. Diercks, A. R. et al. Characterization of subsurface polycyclic aromatic hydrocarbons at the Deepwater Horizon site. Geophys. Res. Lett. 37 (2010). 9. Reddy, C. & Quinn, J. The North Cape oil spill: hydrocarbons in Rhode Island coastal waters and Point Judith pond. Mar. Environ. Res. 52, 445–461 (2001). 10. Boehm, P. D. & Fiest, D. L. Subsurface distributions of petroleum from an offshore well blowout. The Ixtoc I blowout, Bay of Campeche. Environ. Sci. Technol. 16, 67–74 (1982). 11. Redman, A. D. Role of entrained droplet oil on the bioavailability of petroleum substances in aqueous exposures. Mar. Pollut. Bull. 97, 342–348, https://doi.org/10.1016/j.marpolbul.2015.05.068 (2015). 12. Arfsten, D. P., Schaeffer, D. J. & Mulveny, D. C. The effects of near ultraviolet radiation on the toxic effects of polycyclic aromatic hydrocarbons in animals and plants: a review. Ecotoxicol. Environ. Saf. 33, 1–24 (1996). 13. Roberts, A. P., Alloy, M. M. & Oris, J. T. Review of the photo-induced toxicity of environmental contaminants. Comp. Biochem. Physiol. C 191, 160–167, https://doi.org/10.1016/j.cbpc.2016.10.005 (2017). 14. McGrath, J. A. & Di Toro, D. M. Validation of the target lipid model for toxicity assessment of residual petroleum constituents: monocyclic and polycyclic aromatic hydrocarbons. Environ. Toxicol. Chem. 28, 1130–1148 (2009). 15. Redman, A. D. & Parkerton, T. F. Guidance for improving comparability and relevance of oil toxicity tests. Mar. Pollut. Bull. 98, 156–170 (2015). 16. Pelletier, M. C. et al. Phototoxicity of individual polycyclic aromatic hydrocarbons and petroleum to marine invertebrate larvae and juveniles. Environ. Toxicol. Chem. 16, 2190–2199 (1997). 17. Peachey, R. L. & Crosby, D. G. Phototoxicity in a coral reef flat community. UV Radiation and Coral Reefs, HIMB Technol. Report 41, 193–200 (1995). 18. Overmans, S. et al. Phototoxic ee ff cts of PAH and UVA exposure on molecular responses and developmental success in coral larvae. Aquat. Toxicol. 198, 165–174, https://doi.org/10.1016/j.aquatox.2018.03.008 (2018). ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 10 www.nature.com/scientificreports/ 19. Vantrepotte, V. & Melin, F. UV penetration in the water column. European Commission D-GJRC, Institue for Environment and Sustainability. Luxembourg: European Communities (2006). 20. Barron, M. G., Little, E. E., Calfee, R. & Diamond, S. Quantifying solar spectral irradiance in aquatic habitats for the assessment of photoenhanced toxicity. Environ. Toxicol. Chem. 19, 920–925 (2000). 21. McDonald, B. G. & Chapman, P. M. PAH phototoxicity—an ecologically irrelevant phenomenon? Mar. Pollut. Bull. 44, 1321–1326 (2002). 22. Tedetti, M. & Sempere, R. Penetration of ultraviolet radiation in the marine environment. A review. Photochem. Photobiol. 82, 389–397, https://doi.org/10.1562/2005-11-09-IR-733 (2006). 23. Michael, K. J., Veal, C. J. & Nunez, M. Attenuation coec ffi ients of ultraviolet and photosynthetically active wavelengths in the waters of Heron Reef, Great Barrier Reef, Australia. Mar. Freshw. Res. 63, 142–149 (2012). 24. Barron, M. G., Vivian, D. N., Yee, S. H. & Santavy, D. L. Methods to estimate solar radiation dosimetry in coral reefs using remote sensed, modeled, and in situ data. Environmental Monitoring and Assessment 151, 445–455, https://doi.org/10.1007/s10661-008- 0288-4 (2009). 25. Yee, S. H., Santavy, D. L. & Barron, M. G. Comparing environmental influences on coral bleaching across and within species using clustered binomial regression. Ecological Modelling 218, 162–174, https://doi.org/10.1016/j.ecolmodel.2008.06.037 (2008). 26. McNutt, M. K. et al. Review of o fl w rate estimates of the Deepwater Horizon oil spill. Proc. Acad. Nat. Sci. 109, 20260–20267 (2012). 27. Carmen Guzmán Martínez, M. D., Romero, P. R. & Banaszak, A. T. Photoinduced toxicity of the polycyclic aromatic hydrocarbon, uo fl ranthene, on the coral, Porites divaricata. J. Environ. Sci. Health. A 42, 1495–1502 (2007). 28. Villanueva, R., Montaño, M. & Yap, H. Effects of natural gas condensate–water accommodated fraction on coral larvae. Mar. Pollut. Bull. 56, 1422–1428 (2008). 29. Negri, A. P. et al. Acute ecotoxicology of natural oil and gas condensate to coral reef larvae. Sci. Rep. 6, 21153, https://doi.org/10.1038/ srep21153 (2016). 30. Marzooghi, S. et al. Phototoxic target lipid model of single polycyclic aromatic hydrocarbons. Environ. Toxicol. Chem (2016). 31. Harrison, P. The effects of oil pollutants on fertilisation rates in the scleractinian coral Acropora tenuis. Proceedings of the Joint Scientific Conference on Science, Management and Sustainability of Marine Habitats in the 21st Century. Conference Abstracts 30 (1994). 32. Di Toro, D. M., McGrath, J. A. & Stubblefield, W. A. Predicting the toxicity of neat and weathered crude oil: toxic potential and the toxicity of saturated mixtures. Environ. Toxicol. Chem. 26, 24–36 (2007). 33. Negri, A. P. & Heyward, A. J. Inhibition of fertilization and larval metamorphosis of the coral Acropora millepora (Ehrenberg, 1834) by petroleum products. Mar. Pollut. Bull. 41, 420–427, https://doi.org/10.1016/S0025-326x(00)00139-9 (2000). 34. Kushmaro, A., Henning, G., Hofmann, D. & Benayahu, Y. Metamorphosis of Heteroxenia fuscescens planulae (Cnidaria: Octocorallia) is inhibited by crude oil: a novel short term toxicity bioassay. Mar. Environ. Res. 43, 295–302 (1997). 35. Mercurio, P., Negri, A. P., Burns, K. A. & Heyward, A. J. The ecotoxicology of vegetable versus mineral based lubricating oils: 3. Coral fertilization and adult corals. Environ. Pollut. 129, 183–194 (2004). 36. Warne, M. et al. Revised Method for Deriving Australian and New ZealandWater Quality Guideline Values for Toxicants. Prepared for the Council of Australian Government’s Standing Council on Environment and Water (SCEW). Department of Science. Information Technology and Innovation, Brisbane, Queensland, https://doi.org/10.13140/RG.2.2.20368.20483 (2015). 37. Negri, A. P. & Hoogenboom, M. O. Water contamination reduces the tolerance of coral larvae to thermal stress. Plos One 6, e19703 (2011). 38. Negri, A. P., Harford, A. J., Parry, D. L. & van Dam, R. A. Effects of alumina refinery wastewater and signature metal constituents at the upper thermal tolerance of: 2. The early life stages of the coral Acropora tenuis. Mar. Pollut. Bull. 62, 474–482, https://doi. org/10.1016/j.marpolbul.2011.01.011 (2011). 39. Redman, A. D., Parkerton, T. F., McGrath, J. A. & Di Toro, D. M. PETROTOX: An aquatic toxicity model for petroleum substances. Environ. Toxicol. Chem. 31, 2498–2506, https://doi.org/10.1002/etc.1982 (2012). 40. Goodbody-Gringley, G. et al. Toxicity of Deepwater Horizon source oil and the chemical dispersant, Corexit 9500, to coral larvae. Plos One 8, e45574 (2013). 41. Mekenyan, O. G., Ankley, G. T., Veith, G. D. & Call, D. J. QSARs for photoinduced toxicity: I. Acute lethality of polycyclic aromatic hydrocarbons to Daphnia magna. Chemosphere 28, 567–582 (1994). 42. Te, F. T. Effects of two petroleum products on Pocillopora damicornis planulae. Pac . Sci. 45 (1991). 43. GBRMPA. Great Barrier Reef Marine Park Authority. Grounding of the Shen Neng 1 on Douglas Shoal, April 2010: Impact Assessment Report, http://elibrary.gbrmpa.gov.au Accessed on 23 April 2017 (2011). 44. Storrie, J. Montara Wellhead Platform Oil Spill – A Remote Area Response. International Oil Spill Conference Proceedings (IOSC) 2011, Abstract 159 (2011). 45. NOPSEMA. Annual offshore performance report - to 31 December 2016. (National Offshore Petroleum Safety and Environmental Management Authority, Commonwealth of Australia, www.nopsema.gov.au Accessed on 07.03.2018 2017). 46. Rinkevich, B. & Loya, Y. Harmful effects of chronic oil pollution on a Red sea scleractinian coral population. i Th rd International Coral Reef Symposium Miami, Florida, pp. 585–591 (1977). 47. Jackson, J. B., Cubit, J. D., Keller, B. D., Batista, V. & Burns, K. Ecological ee ff cts of a major oil spill on Panamanian coastal marine communities. Science 243, 37 (1989). 48. Loya, Y. & Rinkevich, B. Effects of oil pollution on coral reef communities. Mar. Ecol. Prog. Ser. 3, 180 (1980). 49. Ho, K. et al. The chemistry and toxicity of sediment ae ff cted by oil from the North Cape spilled into Rhode Island Sound. Mar. Pollut. Bull. 38, 314–323 (1999). 50. Jones, R., Ricardo, G. & Negri, A. Effects of sediments on the reproductive cycle of corals. Mar. Pollut. Bull. 100, 13–33 (2015). 51. Aurand, D. & Coelho, G. Cooperative aquatic toxicity testing of dispersed oil and the chemical response to oil spills: Ecological Effects ResearchForum (CROSERF). Inc . Lusby, MD. Tech. Report, 07–03 (2005). 52. Barron, M. G. & Ka’aihue, L. Critical evaluation of CROSERF test methods for oil dispersant toxicity testing under subarctic conditions. Mar. Pollut. Bull. 46, 1191–1199, https://doi.org/10.1016/S0025-326X(03)00125-5 (2003). 53. Heyward, A. J. & Negri, A. P. Natural inducers for coral larval metamorphosis. Coral Reefs 18, 273–279 (1999). 54. Negri, A. P. & Heyward, A. J. Inhibition of coral fertilisation and larval metamorphosis by tributyltin and copper. Mar. Environ. Res. 51, 17–27, https://doi.org/10.1016/S0141-1136(00)00029-5 (2001). 55. Negri, A. et al. Effects of the herbicide diuron on the early life history stages of coral. Mar. Pollut. Bull. 51, 370–383, https://doi. org/10.1016/j.marpolbul.2004.10.053 (2005). 56. R: A language and evironment for statistical computing (R Foundation for Statistical Computing, Vienna, Austria, 2017). 57. DHARMa: Residual Diagnostics for Hierarchical (Multi-Level/Mixed) Regression Models (Hartig, F., 2017). 58. Logan, M. In Biostatistical Design and Analysis Using R 483–530 (Wiley-Blackwell, 2010). 59. Venables, W. N. & Ripley, B. D. Modern applied statistics with S. 4 edn, 193–194 (Springer-Verlag New York, 2002). 60. Billiard, S. M., Meyer, J. N., Wassenberg, D. M., Hodson, P. V. & Di Giulio, R. T. Nonadditive ee ff cts of PAHs on Early Vertebrate Development: mechanisms and implications for risk assessment. Toxicol. Sci. 105, 5–23, https://doi.org/10.1093/toxsci/kfm303 (2008). 61. Ahrens, M. J., Nieuwenhuis, R. & Hickey, C. W. Sensitivity of juvenile Macomona liliana (bivalvia) to UV‐photoactivated uo fl ranthene toxicity. Environ. Toxicol. 17, 567–577 (2002). ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 11 www.nature.com/scientificreports/ Acknowledgements e a Th uthors would like to thank the Australian Institute of Marine Science Townsville (AIMS) and the National Sea Simulator (SeaSim), Tristan Lever and Edith Strecker for logistical support as well as Gerard Ricardo and Patricia Menendez for their advice regarding statistical analysis. This work was supported by a research grant set up as part of the collaborative research project between the Red Sea Research Center (RSRC) at King Abdullah University of Science and Technology (KAUST) and the AIMS. Author Contributions F.M.N. and A.P.N. designed the experiment with input from F.F., D.L.B. and S.A. F.M.N., F.F., D.L.B. and A.P.N. performed experiments. F.M.N. analysed the data. F.M.N. wrote the manuscript with input from A.P.N. as well as F.F., D.L.B. and S.A. Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-018-26972-7. Competing Interests: The authors declare no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 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 Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted 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 license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2018 ScIentIfIc Repo R ts | (2018) 8:8635 | DOI:10.1038/s41598-018-26972-7 12

Journal

Scientific ReportsSpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off