Water Column Monitoring 2017
Environmental monitoring of petroleum activities on the Norwegian continental shelf 2017
Report Reference: Pampanin DM, Brooks S, Grøsvik BE, Sanni S 2019. Water Column Monitoring 2017. Environmental monitoring of petroleum activities on the Norwegian continental shelf 2017. NORCE-Environment REPORT 007 – 2019, pp 92.
1
Project title: Water Column Monitoring 2017 Project number: 100399 Institutions: NORCE, NIVA, IMR, SINTEF Client/s: Norsk Olje og Gass
Classification: Confidential Report no.: NORCE Environment 007-2019 Number of pages: 92
Stavanger, 28.01.2020
Daniela M. Pampanin Shaw Bamber Hans Kleivdal Project manager Quality assurance Executive Vice President - Environment
2
Preface Companies operating on the Norwegian continental shelf are required to carry out environmental monitoring in order to obtain information on the actual and potential environmental impacts of their activities and to give environmental authorities a better basis for regulating releases of pollutants. The general purpose is to provide an overview of the environmental status and of trends over time seen in relation to offshore oil and gas activities. Monitoring is intended to indicate whether the environmental status on the Norwegian continental shelf is stable, deteriorating or improving, due to operators’ activities. In addition to identifying trends, the results should as far as possible provide a basis for projections for future developments. Operators and authorities use monitoring results as a source of information and as grounds for decision making regarding new measures to be implemented offshore. The results are also used to develop and report on national environmental indicators for the offshore oil and gas industry.
The WCM programme has been performed through collaboration between NORCE (which now include the previous International Research Institute of Stavanger, IRIS), the Norwegian Institute for Water Research (NIVA), the Institute of Marine Research (IMR) and SINTEF.
The work participants from these institutions include:
NORCE: Daniela M. Pampanin, Mark Berry, Steinar Sanni, Emily Lyng, Stig Westerlund, Kjell Birger Øysæd, Eivind Larssen, Sophia Mehdipour, Frederike Keitel-Groner, Ingrid Caroline Vaaland.
NIVA: Steven Brooks, Christopher Harman, Bjørnar Andre Beylich, Jarle Håvardstun, Sigurd Øxnevad, Tania Cristina Gomes, Lene Fredriksen, You Song, Maria Therese Hultman.
IMR: Bjørn Einar Grøsvik, Sonnich Meier, Guri Nesje, Grethe Tveit, Anna Ersland, Therese Aase, Stig Mæhle, Kai Ove Skaftnesmo.
SINTEF: Bjørn Henrik Hansen, Dag Altin, Lisbet Sørensen, Tone Haugen, Jørgen Skancke, Marianne Aas, Marianne Rønsberg, Roman Netzer, Trond R. Størseth.
Stavanger, September 2019
Dr Daniela M. Pampanin Project Manager
3
The Water Column Monitoring 2017 Final report Summary
Introduction The programme commonly referred to as water column monitoring (WCM) is presented as set out in the Norwegian Environment Agency Guidelines M-408. The WCM involves the mapping of pollutants or biological effects of pollutants, using caged and wild caught organisms in the water column. Detailed requirements for environmental monitoring on the Norwegian continental shelf have earlier been compiled within regulations relating to conducting petroleum activities (Aktivitetsforskriften).
Aim of the project The WCM aims to document whether and to what extent organisms in Norwegian waters are affected by pollution generated by oil and gas activities. The requirement to conduct monitoring of the water column is not solely related to produced water (PW) discharges, but include relevant chemical additives and seeps or leaks from the seabed, as well as any other relevant discharges. Therefore, it was important to ensure that the scope of the monitoring was proportional to the expected risk (section 1.3).
Methods The WCM 2017 program was designed to evaluate effects of oil and gas activities near field (i.e. around selected platforms) and at regional scale, combining two programs previously called “effect monitoring” and “condition monitoring”. It was carried out in three different regions of the North Sea: Tampen area, Central North Sea, and Egersund Bank (reference area) (section 2.1 and 2.2). Additionally, focus has been placed on studying the water column conditions around the Statfjord A and B platforms (Tampen area). The monitoring priorities were to determine: • potential effects of oil and gas related discharges in mussels caged around the Statfjord A and B platforms (WP1), • potential effects of oil and gas related discharges in wild fish caught around the Statfjord A and B platforms (WP2), • potential effects of oil and gas related discharges in wild fish caught in three regions of the North Sea (WP3), • the potential use of zooplankton as a monitoring tool for studying the effect of oil and gas related discharges (zooplankton-based monitoring) (WP4). In addition, priority was given to a research study on method development for DNA adducts in fish (WP5). WP 5 results will be presented as a separate report. Descriptions of WPs and references to report sections are presented in Table A to provide a quick overview of the project. Chemical and biological results are described in this report and discussed in relation to the stations distance from a PW discharge point (i.e. Statfjord A and B), or in relation to an area (i.e. Tampen, Central North Sea, Egersund Bank). In addition, biological marker results were
4 integrated using the Integrative Biological Response (IBR) index (section 3.4.1). This data treatment was developed to integrate biochemical, genotoxic and histochemical biomarkers by Beliaeff and Burgeot (2002). The method is based on the relative differences between biomarkers in each given data set and it has been applied in previous WCM programs. For the first time the ‘biomarker bridges’ approach (Sanni et al., 2017a, b, c) was applied to make assessments of interest, such as indications about the obtained biomarker responses compared to discharges and to expected impacts and risks (section 3.4.2 and appendix 9). Table A. Quick overview of the project results, including work package aim and descriptions and references to report sections. Work package Aim Activity description Report section
WP 1 Potential effects in mussels Chemical and biological analyses in Methods caged around the Statfjord A caged mussel 2.3.1/2.3.2/2.3.3 and B platforms Results 3.2/ 3.4.1 WP 2 and 3 Potential effects in wild fish Chemical and biological analyses in Methods caught in three regions of the wild caught fish 2.3.1/2.3.2/2.3.4 North Sea, with a focus around Results 3.3, 3.4.1 the Statfjord A and B platforms
WP 4 Potential use of zooplankton as Sampling and analyses (PAH Appendix 1 a monitoring tool exposure parameters, zooplankton community composition) of copepods from the Tampen area
WP 5 Method development for DNA Identification of DNA adducts in Separate report adducts in fish laboratory exposed haddock and selected field samples by LC-MS/MS.
Results and discussion With regard to the caged mussel study, it is important to note the shift in mussel species distribution among Norwegian populations. In comparison with previous surveys, there has been a reduction (-6%) in M. edulis and an increase in hybrid species. Mussels were purchased from a farm in Trondheimsfjorden (Norway), this source has been used previously for the WCM program. Since different genera may have differing capabilities to accumulate and respond to pollutants, this issue will need to be addressed in future WCM programs. The sum of PAH concentrations in mussel tissues showed higher body burden concentrations in organisms that were caged closest to both Statfjord A and B platforms. However, these values were lower than those from previous surveys. For PAH accumulation, naphthalenes were the most abundant followed by phenanthrenes and dibenzothiophenes. The source of the PAHs has been identified as petroleum, as in previous surveys at Gullfaks C, Troll C and Ekofisk. Regarding the biological effect parameters, some highlighted a stress condition present in mussels caged at Statfjord A and B. When comparing results against ICES assessment criteria, mussels appear to be in stress conditions, but compensating, as confirmed by the physiological level measurements (stress on stress and condition index) and their ability to maintain reproductive development (e.g spawning status). However, signs of more severe stress conditions were recorded in mussels caged 500 m from Statfjord A, by means of lysosomal membrane stability and micronucleus (MN) frequency in haemocytes. In particular, in 4 stations close to the platform, MN frequency values were above the elevated response (ER)
5 limit suggested by ICES, showing a clear sign of the presence of contaminants with genotoxicity potential. Data integration using IBR/n confirmed that stressed organisms were compensating, by showing IBR/n values within the range of the two reference stations (since mussels from both stations are considered to originate from a known clean area, it can be considered that all the stations with IBR/n within the range of the reference value are not impacted). Two stations located at 500 m from Stafjord A had the highest IBR/n values, indicating a more severe level of stress. Results also highlights the importance of having more than one reference group in the monitoring. Regarding analyses in wild caught fish, four areas were considered: Statfjord, Tampen, Central North Sea and Egersund bank (as a reference). Unfortunately, liver samples from Statfjord, planned to be held at -80 °C prior to analysis, appeared to have suffered a thawing incident, which may have occurred during the transport to the laboratory. DNA adduct analysis was still possible. However, the EROD, Cyp1A and gene expression results were below detection limits and could not be considered for further discussion in this study. PAH metabolites were significantly higher in cod collected at Statfjord (i.e. 2,3-ring PAHs and 5-ring PAHs), and in whiting sampled at Tampen area (i.e. 4-ring PAHs) compared to the reference site. The increased levels of PAHs in the water column at Tampen and Stafjord shown in soft tissues of deployed mussel was less evident in fish. Nevertheless, genotoxic effects were clear in fish, as revealed by both DNA adduct and comet assay results. DNA adduct levels were higher at Tampen and Statfjord compared to the reference area. In addition, DNA adduct levels were significantly different between species at Tampen, with higher levels for haddock and saithe, compared to the refence area. DNA adduct levels were also significantly different among fish species collected from Statfjord A, with higher levels for haddock, whiting and ling. DNA adducts were also analysed in the intestines of a subset of samples and higher levels of DNA adducts compared with fish from the reference area were found. Part of the difference could, however, be explained by age differences among the fish. Levels of DNA adducts from haddock livers at Tampen were higher than those reported in the condition monitoring programs from 2005-2011 and levels were above environmental assessment criteria (EAC) suggested by ICES at Tampen and at the Statfjord A field in all investigated species. In addition, adduct levels in intestine tissue from these species were also above EAC in samples taken from the Statfjord A field. A significant inhibition in acetylcholine esterase (AChE) activity were observed for in ling from Statfjord and cod from Statfjord and Tampen, compared with the Egersund Bank, although not found for whiting and saithe. Inhibition in AChE activity may indicate possible exposure to neurotoxic contaminants. ICES assessment criteria have been developed for AChE in a few marine fish, including dab, flounder, red mullet and eelpout, but not for the fish species used in this study. Highest IBR/n’s were found in fish from the Tampen and Statfjord areas compared to Egersund bank and Central North Sea. PAH metabolites gave the highest contribution in all species collected at the Tampen area. While for fish collected in Statfjord, different biomarkers contributed in the different species. For the first time, a complete WCM dataset was assessed using the biomarker bridge assessment methodology. It is important to note that this evaluation also highlighted an increase in genotoxicity in sample tissues.
6
Conclusions The WCM 2017 represents an extensive set of data providing information on environmental conditions in close proximity to oil and gas platforms (Statfjord A and B) and the wider North Sea. Mussels caged around two platforms, which showed negative biological effects, were compensating and therefore not considered stressed. However, in 2 stations (out of 17) at 500 m from the discharge point in the plume direction at Statfjord A, mussels showed evidence of stress at physiological and cellular level (as highlighted by the genotoxicity data). Wild fish collected at Tampen presented genotoxic damages, especially in haddock specimens where DNA adduct levels were higher than in previous years and above the environmental assessment criteria recommended by ICES, suggesting the presence in the water of compounds with genotoxic potential. In addition, the potential presence of compounds with neurotoxic effect was revealed by the AchE analysis.
7
Contents
Summary 4 1. Introduction 9 1.1 Discharge history of the Statfjord field 10 1.2 Goals 11 1.2.1 General purpose of the WCM program 11 1.2.2 Objective of the WCM 2017 12 2 Methods 12 2.1 Choice of sampling design, areas and stations 12 2.2 Field work 13 2.3 Laboratory procedure 15 2.3.1 The organisms 15 2.3.2 Chemical analyses 17 2.3.3 Biological analyses in mussels 19 2.3.4 Biological analyses in fish 24 2.4 Quality Assurance 32 2.5 Statistical Methods 32 2.6 Samples and data storage 33 3 Results and discussion 34 3.1 Current and temperature data 34 3.2 Chemical and biological analyses in caged mussels 39 3.2.1 Identification of Mytilus spp. 39 3.2.2 PAH and NPD concentrations 39 3.2.3 Metal body burden 45 3.2.4 Condition index 45 3.2.5 Stress on stress test (General health status) 46 3.2.6 Lysosomal membrane stability 48 3.2.7 Acetylcholine esterase inhibition 49 3.2.8 Lipid content 51 3.2.9 Gonad histological evaluation (spawning) 51 3.2.10 Micronucleus assay (DNA damage) 53 3.3 Chemical and biological analyses in wild caught fish 55 3.3.1 PAH body burden 55 3.3.2 Perfluorinated Compounds 57 3.3.3 Radioactive compound concentrations 57 3.3.4 Condition Indices and age determination 58 3.3.5 PAH metabolites 61 3.3.6 Alkylphenol metabolites 65 3.3.7 Tissue changes in liver 65 3.3.8 DNA damage in liver and gut, DNA adducts 70 3.3.9 DNA damage in lymphocytes, Comet assay 75 3.3.10 Acetylcholineesterase inhibition 76 3.3.11 Gene expression I, qPCR to select gene transcription 77 3.3.12 EROD 79 3.3.13 CYP 1A (ELISA analysis) 80 3.4 Data treatment 81 3.4.1 Integrative assessment using the IBR index 81 3.4.2 Biomarker bridge 84 4 Conclusions 85
8
5 Appendices 86 6 References 87
9
1 Introduction 1.1 Programme background and earlier surveys The programme commonly referred to as Water column monitoring (WCM) is presented as set out in the Norwegian Environment Agency Guidelines M-408 (Iversen et al. 2015). The WCM involves the mapping of pollutants or biological effects of pollutants, using caged and wild caught organisms in the water column. Detailed requirements for environmental monitoring on the Norwegian Continental Shelf (NCS) have earlier been compiled within the Regulations relating to conducting petroleum activities (the Activities Regulations). These detailed requirements were removed from the Activities Regulations with effect from 1 January 2010 and incorporated into the present guidelines, which were revised in January 2013 (TA 2849/2011) and June 2015 (M-408|2015) (op cit.). The early water column monitoring/effect monitoring effort developed during the mid-1990s was based on concerns related to increasing amounts of produced water (PW) discharges expected on the NCS. The programme evolved from measurements of oil constituent bioaccumulation in semi-permeable membrane devices (SPMDs) and caged mussels (Johnsen et al. 1998) to inclusion of different biological exposure and effect markers (biomarkers) in caged mussels and fish. The biomarker approach was motivated by the practical understanding of the ”0-discharge” regulation (Stortingsmelding nr. 58, 1996-97) to mean “no harmful effect discharges”, for which biomarker methods were considered to provide the most relevant information in this context. The 2001 and 2002 BECPELAG workshop was instrumental in the selection of core biomarkers for the program (Hylland et al. 2002). Since then opportunities have been given for testing and evaluation of different biomarkers, and the method selection has evolved over time. The aim has been to provide early warning of different kinds of toxicity with relevance to constituents of PW. In the regulation guidelines before 2015, the term “Effect Monitoring” was used associated with the WCM, though this term is no longer in use (Iversen et al. 2015). In parallel to the WCM program, methods for the Environmental Risk Assessment (ERA) of PW discharges were developed (Reed and Rye, 2011). The DREAM-EIF system (Dose Related Risk and Effect Assessment Model for chronic discharges – Environmental Impact Factor) became a “North Sea standard” for use on the NCS. However, there has been a lack of methods to connect the results of the WCM monitoring to the DREAM-EIF based ERA. This has recently been established theoretically and the first attempt to apply this approach as an interpretation tool was performed with the WCM 2013 data (Sanni et al. 2017a-c; 2018). The WCM 2017 was carried out in three different regions of the North Sea: Tampen area, Central North Sea, and Egersund Bank (reference area). In addition, focus has been placed on a study of the water column conditions around the Statfjord A and B platforms. Previous studies that included monitoring of the Tampen region are shown in Table 1. Table 1. Previous water column monitoring studies in the Tampen area. Year Effect Monitoring Condition Monitoring Reference 1995 Statfjord C Nilssen and Bakke, 2011 1997 Tampen Nilssen and Bakke, 2011, Røe, 1998. 2001 Tampen (Becpelag project) Hylland et al., 2002 and 2006 2002 Tampen (Becpelag project) Tampen Hylland et al., 2002 and 2006, Balk et al., 2011 2004 Statfjord B Hylland et al., 2005 2005 Tampen Grøsvik et al., 2007 2008 Tampen Grøsvik et al., 2009 2011 Gullfaks C Brooks et al 2011 2012 Tampen Grøsvik et al., 2012 10
1.2 Discharge history of the Statfjord field The Statfjord field is an enormous oil and gas field located close to the U.K.-Norwegian boundary of the North Sea (Figure 1). The average water depth is around 145 m. Equinor is the main operator of the field. At its peak, oil production of the field was almost 700,000 barrels (110,000 m3) per day. The produced oil is transported to refineries, whereas the produced gas is piped directly to mainland Norway (Statoil, 2017). The environmental risk related to PW discharges at Statfjord A and B, and in the Tampen area in general is reported as PEC/PNEC (predicted environmental concentration/predicted no effect concentration, Figure 2). The risk is calculated as Environmental Impact Factor (EIF) using DREAM.
A)
B)
Figure 1 Statfjord oil field relative location in comparison with Norway and UK coastlines (A), Statfjord oil field relative position in comparison with the Norwegian coastline (B).
11
Figure 2 Example of the total risk of the area around Statfjord A and B and in the Tampen area in general reported as PEC/PNEC, calculated by DREAM in April 2017.
1.3 Goals 1.3.1 General purpose of the WCM program Companies operating on the Norwegian continental shelf are required to carry out environmental monitoring to obtain information on the actual and potential environmental impacts of their activities, and to give environmental authorities a better basis for regulating releases of pollutants. The general purpose is to provide an overview of the environmental status and of trends over time seen in relation to offshore oil and gas activities. Monitoring is intended to indicate whether the environmental status on the NCS is stable, deteriorating or improving, due to operators’ activities. Operators and authorities use monitoring results as a source of information and as grounds for decision making regarding new measures to be implemented offshore. The results are also used to develop and report on national environmental indicators for the offshore oil and gas industry. In addition, monitoring data are reported to the Norwegian Ministry of Climate and Environment and to international bodies such as OSPAR. The monitoring aims to document whether and to what extent organisms in Norwegian waters, are affected by pollution generated by oil and gas activities. The requirement to conduct monitoring of the water column is not solely related to PW discharges but may include relevant 12 chemical additives and seeps or leaks from the seabed, as well as any other relevant discharges. Therefore, it is important to ensure that the scope of the monitoring is proportional to the expected risk (Iversen et al, 2015). 1.3.2 Objective of the WCM 2017 The purpose of the WCM 2017 was to provide a detailed picture of the situation near offshore oil and gas installations (PW discharge and cuttings piles) (Statfjord A and B, the near field study) and in the North Sea region as a whole (Tampen, Central North Sea, and Egersund Bank, the regional study). Special attention was given to investigate the origin of previously detected genotoxic effects at the Tampen region. The WCM monitoring priorities in 2017 were to determine the potential effects of oil and gas related chemicals in caged mussels (WP1), wild fish (WP2 and WP3), and zooplankton (WP4). In addition to the field monitoring survey, priority was given to a research study on method development for DNA adducts in fish (WP5). A two-tier approach was applied, screening several stations, and focusing analytical efforts on stations where the first tier showed a positive response signal (Viarengo et al., 2007). The priority for interpretation of the results in WP1-3 was to evaluate the findings as significantly different from background through comparisons to ICES background assessment criteria (BAC) and comparisons with reference stations, and to environmental assessment criteria (EAC) in relation to ICES defined tolerance levels (ICES, 2011). In addition, a first experience of the practical use of the Species Sensitivity Distribution based link to environmental risk assessment in relation to a PW discharge situation (´Biomarker Bridges´; Sanni et al. 2017a-c) was included within the study. The use of zooplankton as a tool for environmental monitoring of the water column was applied for the first time in the program. Calanus sp. were sampled from a reduced number of stations and data were compared with data obtained in a laboratory study. The results of this first attempt are presented as a separate report in Appendix 1 (Zooplankton monitoring WP4). The method development for DNA adducts in fish (WP5) is still an ongoing activity and the results will be presented as a separate report. In brief, this activity aims to: a) provide additional data and a publication of the project: comparative DNA damage and long-term health effects in juvenile haddock exposed to sediment/produced water associated PAHs, b) identify DNA adducts in laboratory exposed haddock and selected field samples using liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS), c) develop a new method to substitute the currently used DNA adduct assay (i.e. 32P-postlabelling) for evaluating the genotoxicity of oil related PAHs. 2 Methods 2.1 Choice of sampling design, areas, and stations The sampling program in 2017 was designed to evaluate potential effects of oil and gas activities in organisms living in the vicinity of a platform (near field study) and in the North Sea (regional study) (Iversen et al. 2015). The WCM 2017 focused on using caged mussels and wild fish in the near field around targeted offshore installations (Statfjord A and B) and it looked at the regional condition of the North Sea using wild fish in three areas: Tampen, Egersund Bank, and Central North Sea. Statfjord, a mature field with significant PW discharges, was selected as the target for the near field study. Considering the potential for a cessation plan, it is of interest to document environmental conditions before field abandonment. Moreover, important local fish populations are present in this area, with spawning zones of saithe, pollock, hake, haddock, and cod. In addition, the presence and chemical composition of cuttings piles are well documented (DNV GL, 2012).
13
The mussel cage stations were chosen based on a pre-simulation of most probable PW plume directions and distances from the Statfjord A and B installations (Figure 3). The Tampen area has significant PW discharges in the region (Figure 1-3), with additional contributions from U.K. waters (Figure 1). Here, the oil fields are old with significant cuttings piles on the seabed, including old cuttings with oil-based mud (discharged before 1993) and elevated levels of DNA adducts have previously been observed in haddock from this area (Balk et al., 2011; Grøsvik et al., 2012). Finally, it has been a long time since the last WCM survey was performed (i.e. 2004), when the conditions at Statfjord B were assessed. The Central North Sea was selected to evaluate the potential contribution of the discharges from the UK sector. The Egersund Bank served as the reference station for both the near field and the regional studies.
Figure 3 Maps showing overview of the Tampen region, the locations of Statfjord A and B (red squares in middle diagram) and the mussel rig stations. The mussel station numbers denote the order of field deployment. Current meters were deployed at stations 4, 8 and 12. The map on the right side shows stations within 500 m from the produced water discharge point at Statfjord A.
2.2 Field work Time frame Samples were collected during three offshore cruises: 19th – 23rd April 2017 (Vessel: M/V Olympic Delta) 12th – 19th May 2017 (Vessel: R/V Johan Hjort) 29th May – 3rd June 2017 (Vessel: M/V Olympic Triton) Sampling areas, stations, and instrument rigs During the first cruise, 19 mussel rigs were deployed in the areas of Statfjord A and B and in a reference area (Figure 3 and Table 2), with the following distribution: • Rigs no. 1 & 2 in far field Statfjord B and A, respectively
14
• Rigs no. 3 in intermediate distance (1000 m) SE-SW of Statfjord B • Rig no. 4 in near field Statfjord B • Rigs no. 5-9 at intermediate distance (1000 - 2000 m) SE-SW of Statfjord A • Rigs no. 10-14 in near field Statfjord A • Rigs no. 15-17 at intermediate distance NE-NW of Statfjord A • Rigs no. 18 & 19 in the reference area north of the Statfjord fields. For a better understanding of the results the stations where mussels were caged are named in figures as reported in the second column Table 2. The station names are coded as location, distance from the discharge point and direction, as for example SFB 10000 SE = 10000 m from the discharge point of Statfjord B in South East direction. During the first and the second cruises, fish were sampled in the near field area of Statfjord A within 2000 meters from the platform, in the wider Tampen area for the far field study, and in the Central North Sea and Egersund Bank for the regional study. The targeted fish samples included both pelagic and demersal species, with the aim to collect a minimum of 25 individuals from 3 species at each location (see Table 4).
Table 2. Mussel rig information: station name, deployment order number, map coordinates, water depth, heading, and distance from the platform (meters and direction), SFA = Statfjord A, SFB = Statfjord B. Station name Deploym Longitude Latitude Dept Distance from Location ent h (m) platform/ order direction SFB 10000 SE 1 1° 57.752’E 61° 8.640’ N 143 10000/135° Statfjord B SFA 10000 SE 2 1° 59.802’E 61° 11.917’ N 143 10000/130° Statfjord A SFB 1000 SE 3 1° 50.536’E 61° 12.010’ N 146 1000/135° Statfjord B SFB 500 SE 4 1° 50.099’E 61° 12.181’ N 147 500/140° Statfjord B SFA 2000 SE 5 1° 52.593’E 61° 14.498’ N 147 2000/140° Statfjord A SFA 2000 E 6 1° 53.082’E 61° 14.788’ N 147 2000/120° Statfjord A SFA 1000 E 7 1° 52.105’E 61° 15.049’ N 148 1000/120° Statfjord A SFA 1000 SE 8 1° 51.860’E 61° 14.904’ N 148 1000/140° Statfjord A SFA 1000 SW 9 1° 50.350’E 61° 14.923’ N 149 1000/225° Statfjord A SFA 500 SW 10 1° 50.738’E 61° 15.117’ N 149 500/225° Statfjord A SFA 500 SE 2 11 1° 51.415’E 61° 15.080’ N 149 500/150° Statfjord A SFA 500 SE 1 12 1° 51.493’E 61° 15.107’ N 149 500/140° Statfjord A SFA 500 E 13 1° 51.616’E 61° 15.180’ N 148 500/120° Statfjord A SFA 500 NE 14 1° 51.515’E 61° 15.504’ N 149 500/45° Statfjord A SFA 1000 NE 15 1° 51.903’E 61° 15.698’ N 148 1000/45° Statfjord A SFA 1000 NW 16 1° 50.322’E 61° 15.685’ N 150 1000/315° Statfjord A SFA 2000 NW 17 1° 49.517’E 61° 16.059’ N 151 2000/315° Statfjord A Ref 1 18 1° 44.893’E 61° 36.970’ N 314 - Reference area Ref 2 19 1° 44.893’E 61° 37.240’ N 316 - Reference area
Deployment and sampling programme Sampling cruise overview: 19th – 23rd April 2017: Collection of fish samples at Statfjord A for analyses and deployment of mussel rigs near Statfjord A and Statfjord B, in addition to two reference stations north of these fields. 12th – 19th May 2017: Collection of fish samples from 3 regions: Tampen area, Central North Sea, and Egersund Bank (reference area). 29th May – 3rd June 2017: Retrieval of mussel rigs and collection of mussel samples from Statfjord A, Statfjord B, and the two reference stations. Collection of zooplankton at Statfjord A and from a reference site.
15
2.3 Laboratory procedures 2.3.1 The organisms Mussels (Mytilus spp.) were provided by Biotrix from a clean location in Trondheims fjord (Snadder and Snaskum), the same source as used in previous WCM programmes. Mussels were transported to NORCE Environment facility in Stavanger and kept in clean seawater (continuous seawater flow pumped from Byfjord, adjacent to the laboratory, from 80 m depth) for a week prior to field deployment. During transportation from Stavanger to offshore locations, mussels were kept in polystyrene boxes in the dark with cooling elements until they were tied on to the line and deployed. Mussels of similar size were used throughout the study (typical length 55 mm ± 2 mm). A set of mussels were sampled after arrival at the laboratory in Stavanger and analysed for chemical and biological endpoints to establish pre-field exposure information (T0). Samples from individual mussels were taken for chemical and biological analyses and morphometric measurements were recorded. Biological samples were preserved as required by the specific analytical measurement. For analysis of lysosomal membrane stability (LMS), acetylcholine esterase (AChE) activity and gene expression, samples were snap frozen in liquid nitrogen and stored at -80°C. Tissues for histological evaluation were stored in formalin solution (10% neutral buffered, Chemi teknikk) and stored at +4 °C. Samples for chemical analysis were frozen with dry ice and stored below -20°C. Slides for the MN assay were prepared on board and stored at 4°C. Analyses were performed using a two-tier approach (Viarengo et al., 2007); parameters included in the first tier were analysed in all stations, while the second tier were applied in 10 selected stations. The analysis plan for the caged mussels with parameters, analytical methods, tissue, and the two-tier groups is shown in Table 3. The second-tier analyses were carried out on stations where the first tier showed relevant signals. Table 3 Overview of chemical and biological marker analyses in caged mussels, the two-tier approach is highlighted. 1st TIER - all stations Parameter Method Type of tissue Reference Speciation PCR gills Inoue et al., 1995 Size/Condition dry weight soft tissue whole organism Lucas and Beninger, 1985 Index mass/dry weight shell General health status stress on stress whole organism Pampanin et al., 2005 PAH exposure GC-MS whole organism EN-EN ISO/IEC 17025 Harman et al., 2008 Metals exposure ICP whole organism EN ISO 17294-2 Lysosomal histochemistry digestive gland Moore et al., 1976 membrane stability Acetylcholine enzymatic gills Bocquené & Galgani, esterase inhibition 1998; Ellman et al., 1961 Lipid content included in the body whole organism Marsh and Weinstein, burden analysis 1966.
2st TIER - 10 selected stations Parameter Method Type of tissue Gonad histological evaluation gonad Gosling 2003 development- spawning DNA damage micronuclei assay haemolymph Venier et al., 1997 Gene expression I qPCR (e.g. oxidative gill Lacroix et al., 2014 stress, DNA repair)
16
Fish were collected using a supply vessel (Olympic Delta) within the 500 m safety zone of both Statfjord A and B installations and following a gradient out to 2000 m from Statfjord A discharge point and 10000 m from the Statfjord B discharge point (Figure 2). The near platform collection was focused within 100 meters of the rigs. Due to subsea installations near the platform only “passive” fishing equipment could be used. Rod and line fishing, using mackerel as bait, proved extremely effective and a steady supply of fish were caught enabling time for processing of the fish on board within a maximum of 1 h and typically within 20 min after capture. The fish species caught at the platform are summarised in Table 4. Samples from individual fish were taken for chemical and biological analyses and morphometric measurements were recorded. Biological samples were preserved as required by the specific analytical measurement. For analysis of PAH and AP metabolites, DNA adducts, AChE, EROD, Cyp1A and gene expression, samples were snap frozen in liquid nitrogen and stored at -80 °C. Tissues for histological evaluation were stored in formalin solution (10% neutral buffered, Chemi teknikk) and stored at +4 °C. Samples for chemical analysis were frozen with dry ice and stored below -20°C. Slides for the comet assay were prepared on board and stored at +4 °C. Fish samples for biological analysis were prepared on board the vessels and stored for further analysis according to the selected method. Samples that required frozen preservation were collected in cryotubes and snap frozen in liquid nitrogen. For the first cruise, when fish samples were collected from close to the Statfjord A platform, the samples that were initially snap frozen in liquid nitrogen were placed in a large chest freezer containing dry ice. Before departure, the frozen samples were transferred to several polystyrene boxes that were packed with dry ice and used for the transport to the laboratory. The samples were driven to the laboratory within 5 hours where they were stored in a freezer at -80°C until analysis. This procedure has been performed on previous offshore monitoring programmes without any incident. However, during the analysis of EROD, Cyp1A, gene expression and DNA adducts, it was noticed that the liver samples from Statfjord appeared to have suffered a thawing incident, which may have occurred during the transport to the laboratory. DNA adduct analysis was still possible, however the EROD and Cyp1A results were all below detection limits and could not be considered for further discussion in this study. Table 4 Number of fish species sampled in the different areas
Fish analyses were also performed using a two-tier approach. The first tier included parameters that were applied to 4 fish species from all stations, whilst those included in the second tier were only applied to 2 species at all stations. The selected parameters for the two tiers are shown in Table 5. In addition to the analyses suggested in the WCM guidelines, perfluorinated compounds (PFCs) were analysed in the whole blood from five individual cod and saithe collected from Statfjord A and regional stations Egersund bank and Tampen (30 samples in total). Results are reported in Appendix 2 (PFAs in blood samples of offshore fish).
17
Table 5 Overview of chemical and biological marker analyses in sampled fish, the two-tier approach is highlighted. 1st TIER - all stations Parameter Method Type of tissue Reference Condition Index weight and length whole Hansson et al., 2017 organism Liver Somatic liver weight, fish weight whole Hansson et al., 2017 Index organism Gonad Somatic gonad weight, fish weight whole Hansson et al., 2017 Index organism Age microscopy otholith Mjanger et al., 2017 PAH/NDP GC-MS fillet EN-EN ISO/IEC 17025 Harman et al., 2008 PAH metabolites FF screening method (as previous bile Aas et al., 1998 and 2000 WCM) GC-MS (as previous WCM) or bile Jonsson et al., 2003 and LC-MS/MS 2004 Tissue changes histological evaluation liver Bernet et al., 1999; Feist et in liver al., 2004 DNA damage in DNA adducts liver Reichert and French, 1994; liver Le Goff et al., 2006 DNA damage in comet assay blood cells Brunborg et al., 2014 lymphocytes Acetylcholine enzymatic fillet Bocquené and Galgani, esterase 1998; Ellman et al., 1961 inhibition Alkylphenol GC-MS bile Jonsson et al., 2003 and metabolites 2004
2st TIER - 10 selected stations Parameter Method Type of tissue DNA damage in DNA adducts gut Reichert and French, 1994; gut Le Goff et al., 2006 Gene expression qPCR in AH depending genes liver Olsvik et al., 2012 I Gene expression RNAseq liver/gill Yadetie et al., 2018 II EROD Enzymatic liver Burke and Mayer, 1974; Eggens and Galgani, 1992 Cyp1A ELISA gut Nilsen et al., 1998
2.3.2 Chemical analyses PAH content in mussels At each station, 25 mussels were opened by cutting through the posterior adductor muscle with a clean scalpel. The mussels were left for approximately 2 minutes to drain excess liquid from the mantle cavity before removing the whole soft tissue and placing in a high temperature treated (560 °C) glass container. Mussels were pooled into 5 replicates of 5 mussels per replicate. Mussel samples were frozen and transported to NIVA on dry-ice. The samples were stored at -20 °C until analysis. The PAH content was analysed following the procedure described below (EN-EN ISO/IEC 17025; Harman et al., 2008). A 5g sub-sample of the mussel homogenate was taken and internal standards added (naphthalene d8, biphenyl d10, acenaphthene d8, phenanthrene d10, anthracene d10, pyrene d10, chrysene d12 and perylene d12) before extraction by saponification. Analytes were then extracted twice with 40 mL cyclohexane and dried over sodium sulphate. The extracts were
18 reduced by a gentle stream of nitrogen and cleaned by size exclusion chromatography. Analysis proceeded by gas chromatography-mass spectrometry (GC-MS) with the MS detector operating in SIM. The GC was equipped with a 30 m column with a stationary phase of 5% phenyl polysiloxane (0.25 mm i.d. and 0.25 μm film thickness), and the injector operated in ‘splitless’ mode. The initial column temperature was 60°C, which after 2 mins was raised stepwise to 310°C. The carrier gas was helium and the column flow rate was 1.2 mL / min. The quantification of individual components was performed by using the internal standard method. The alkylated homologues were quantified by baseline integration of the established chromatographic pattern and the response factors were assumed equal within each group of homologues. PAH in fish fillet (NIVA) Fish fillet, absent of skin, was sampled from the anterior dorsal surface, placed in a heat treated (560 °C) glass container, frozen and transported on dry ice to the NIVA laboratory for analysis. Five individual fish were analysed for each species (except Tusk, n=3). In the laboratory, fish fillet samples were defrosted, homogenised and a sub-sample of approximate 5 g was taken. PAH concentrations were determined using the same procedure as described above for mussels (EN-EN ISO/IEC 17025; Harman et al., 2008). PAH/NPD in fish fillet (IMR) The wet muscle tissue was extracted by saponification with 0.5N alcoholic KOH for 2 h, followed by liquid/liquid extraction with hexane. Extracts were volume reduced and cleaned on silica/alumina column using Powerprep© automated clean up system prior to injection on an Agilent N-5975 GC/MS (EI) in SIM mode. The GC/MS system was equipped with a HP- 6890 GC, a 30m x 0,25mm, 0.25µm DB-17ms capillary column from Agilent. Other conditions were: injector temperature 300ºC; column temperature, 50ºC for 2 min, 50-110ºC at 10ºC/min, 110-290ºC at 6ºC/min, 21 min at final temperature, carrier gas He at 36 cm/s. Samples were injected by auto sampler, 1 µl splitless injection. The method is validated to analyse PAH compounds in concentrations of 0.2 ng/g. Levels of detection (LOD) are defined as LOD: Y = YB + 3SDB, and levels of quantification (LOQ) is LOQ= Y = YB + 10SDB, where YB is the response of blank sample signal and SDB is the standard deviation of the blank samples. This method is accredited by Norsk Akkreditering and it is named O1 and 11 of the 50 reported PAH components are included in the yearly certification. The accredited components are: phenanthrene, anthracene, 2-methylphenanthrene, 3,6-dimethylphenanthrene, fluoranthene, pyrene, benz(a)anthracene, chrysene, benzo(a)pyrene, indeno(1,2,3-cd) pyrene, benzo(ghi)perylene. The laboratory participates in ring tests organized by QUASIMEME. Metals in mussels Metal concentrations were determined in homogenised whole mussel samples following the ISO 17294-2 international standard using microwave digestion and inductively coupled plasma - mass spectrometer (ICP-MS) (EN ISO 17294-2). Mercury concentrations were determined with an atomic absorption spectrometer. Perfluorinated Compounds A suite of 23 PFCs were measured in fish blood based on the method from Verreault et.al. (2005). Internal standards were added to 0.5 mL of sample and extracted twice with acetonitrile using an ultrasonic bath. The extract was mixed with ammonium acetate buffer, acetic acid and EnviCarb and then filtered (0.45 µm) before analysed by LC/MS-QToF (ESI negative mode). The PFCs measured in fish blood included the following compounds: perfluoro-n-pentanoic acid (PFPA); perfluoron-hexanoic acid (PFHxA); perfluoro-n-heptanoic acid (PFHpA); perfluoro-n-octanoic acid (PFOA); perfluoro-n-nonanoic acid (PFNA); perfluoro-n-decanoic acid (PFDA); perfluoro-n-undecanoic acid (PFUnDA); perfluoro-n-dodecanoic acid 19
(PFDoDA); perfluoro-n-tridecanoic acid (PFTrDA); perfluoron-tetradecanoic acid (PFTeDA); perfluoro-n-hexadecanoic acid (PFHxDA); perfluoro-n-octadecanoic acid (PFODA); perfluoro-1-butanesulfonate (PFBS); perfluoro-1-hexanesulfonate (PFHxS); perfluoro-1- octanesulfonate (PFOS); perfluoro-1-decanesulfonate (PFDS); perfluoro-1-dodecansulfonate (PFDoDS); perfluoro-1-octanesulfonamide (PFOSA); N-methylperfluoro-1- octanesulfonamide (N-MeFOSA); Nethylperfluoro-1-octanesulfonamide (N-EtFOSA); 2-(N- methylperfluoro-1-octanesulfonamido)-ethanol (N-MeFOSE); 2-(N-ethylperfluoro-1- octanesulfonamido)-ethanol (N-EtFOSE); 1H,2H-perfluorooctane sulfonate (6:2); (6:2FTS). 2.3.3 Biological analyses in mussels Speciation Mussel speciation was determined in gill tissue based on the method of Inoue et al., (1995). Total DNA was extracted from 20-40 mg of sample from frozen mussels using DNAzol reagent (Invitrogen, Madison, Wisconsin, USA), following the manufacturer’s recommended protocol. The tissue was homogenised in 1 mL DNAzol using Precellys 24 bead mill (Bertin, Montigny- le-Bretonneux, France), using ceramic CK14 beads at 5000 rpm for 10 sec. Cell debris were then removed by centrifugation at 10,000 g for 10 min (4ºC), before DNA was precipitated from the supernatant by addition of 500 µl of 100% ethanol. Following two wash steps with 75% ethanol, the DNA was pelleted by centrifugation at 4,000 g for 2 min, air dried and dissolved in 8 mM NaOH. The resulting DNA was quantified and quality controlled on a nanodrop spectrophotometer (Thermo Fisher Scientific, Waltham, Massachusetts, USA), and all samples had OD 260/280 > 1.8, which is indicative of a pure DNA. For species identification, the polymerase chain reaction (PCR) was used to amplify a specific 180 base pair (bp) segment for M. edulis, 168 bp segment for M. trossulus or 126 bp segment for M. galloprovincialis, as described by Inoue et al. (1995). The 50 µl PCR reactions contained 10 µl of DNA template, 300 µM forward and reverse primers, VWR 2x Taq mastermix (VWR, Radnor, Pennsylvania, USA), and were subjected to a 5 min pre-heating stage at 95ºC followed by 35 cycles of 30 sec at 95ºC, 30 sec at 55ºC, 30 sec at 72ºC, and final extension step of 10 min at 72ºC. One µl of the PCR product was loaded onto a DNA 1000 chip (Agilent technologies, Santa Clara, California, USA) and run in a Bioanalyzer instrument (Agilent technologies, Santa Clara, California, USA) for visualisation of amplicon size. Condition index Thirty animals from each station were used for calculating the condition index (CI) (Figure 4). The Mettler AT200 sensitive scale was used for measurements (Figure 4a). Mussel shells were carefully cleaned; shell length was measured with Vernier callipers (Figure 4b). Mussels were opened gently, soft tissue entirely removed, and placed in aluminium dishes (Figure 4c). Shells were also dried with a paper tissue and labelled. Thereafter, dishes containing shells and soft tissues were dried in an oven at 90 °C for 48 h (Figure 4d). After 48 h, soft tissues and shells were weighed, and values recorded. The CI was calculated according to the following formula (Lucas and Beninger, 1985): CI= dry weight of the organism/dry weight of shell
20
Figure 4 Instruments and procedure steps for the condition index measurement in mussels. a) Mettler AT200 scale; b) Shell measurement by Vernier calliper; c) Shell and soft tissue samples; d) Incubator set at 90 °C. Stress on stress test (General health status) Survival in air tests were performed on 30 animals per station. All the mussels were kept at 10 °C, with 100% humidity in a small incubator. Mussels were checked daily. When mussels’ valves gapped and failed to close by physical stimulation, the mussel was considered dead (Pampanin et al., 2005). Median survival time (LT50) was calculated for each station by using the Kaplan-Meier test in the SPSS software (Kaplan and Meier, 1958). Survival curves were compared using the Wilcoxon and Gehan test by using SPSS software. The Kaplan-Meier procedure is a method of estimating time-to-event models in the presence of censored cases. Overall comparison among all stations was done by using Wilcoxon test. The survival curves were considered statistically different if the significance value of the test was less than 0.05. Lysosomal Membrane Stability The LMS analysis was performed according to Moore (1976). After dissection of mussels (15 mussels per station), digestive glands were removed immediately and placed into cryovial tubes and kept frozen at -80 °C until analysis. Five digestive glands were attached by glue to an aluminium chuck (Figure 5a). Ten slices from each chuck were cut in 10 μm thick sections in Bright’s Cryostat machine (with cabinet temperature at -23°C with its knife cooled at -20°C) (Figure 5 b-d). Sections were then transferred to warm microscope slides (room temperature), which were pre-labelled in time series 0, 3, 6, 10, 20, 30, 40, and 50 min. The microscope slides were stored again in the freezer at -40°C prior to LMS analysis. The determination of LMS was based on the time of acid labialization treatment required to produce the maximum staining intensity according to UNEP/RAMOGE (1999), after demonstration of hexosaminidase (Hex) activity in digestive cell lysosomes. Serial cryostat
21 sections were exposed to acid labialization in intervals of 0, 3, 6, 10, 20, 30, 40, and 50 min in citrate buffer in a shacking bath at 37 °C (Figure 5e), in order to find out the range of pre- treatment time needed to complementary labilise the lysosomes. After each time interval reached, all slides were removed from citrate buffer.
Figure 5 Lysosomal membrane stability test steps; attachment of digestive gland to the aluminium chuck (a); Cryostat machine(b); attached digestive gland on cryostat chucks (c); Slides cut with the thickness of 10 μm (d); shacking bath (e) Slides were dipped in reaction medium in jars (f); Section mounting with mounted medium (g); taking picture with microscope (h).
Sections were incubated in the incubation medium (Component 1: 7 gr low viscosity polypeptide in 100 ml Citrate buffer (pH=4.5); Component2: 40 mg Naphthol AS BI N-Acetyl- β-D-glocosamid in 5 ml Dimethyl sulfoxide (DMSO) for 20 min in shacking bath at 37 °C for demonstration of Hex activity. Sections were then washed in 3 % NaCl for 2 min at room temperature, before embedding in the reaction medium (0.2 g of the diazonium salt Fast Violet B in 200 ml of phosphate buffer) for 10 min in the dark at room temperature (Figure 5f). The visualization of the enzyme-substrate complex was achieved by a post-coupling reaction. Finally, sections were rinsed three times with distilled water, left to dry out at room temperature, and then were mounted with mounting medium (Figure 5g) (Kaizer glycerine gelatin). Stained slides were viewed under optical microscope dividing each section into four approximately equal areas for statistical interpretation (Figure 6). Lysosomes appears reddish- purple due to the reactivity of the substrate with N-acetyl-β-hexosaminidase.
22
The average labialization period for each digestive gland corresponded to the average incubation time in the acid buffer that produced the maximum staining reactivity. A mean value was then derived for each station. The image analysis was performed by comparing intensity of staining using an image processing software called ImageJ (Ref: https://imagej.nih.gov/ij/index.html).
Figure 6 Schematic illustration of digestive gland under microscope. Acetylcholine esterase inhibition The method from Bocquené and Galgani (1998), adapted from Ellman et al (1961), to determine the AChE activity in biota tissue was followed with minor modifications to assess the activity in extracts from fish fillet samples and mussel gill tissue. Fish and mussel samples were homogenized in a potassium phosphate buffer (0.02 M, pH 7) and a Tris-HCl buffer (100 mM, pH 8.0, 0.1% Triton), respectively, using a Precellys tissue homogenizer (Bertin Technologies, Montigny-le-Bretonneux, France). The resulting homogenate was centrifuged at 10,000 × g for 20 min at 4°C and the supernatant used for protein determination and AChE analysis. Experiments were performed in 96-well microplates (Sarstedt, Nürnbecht, Germany) and every sample was run in quadruplicate. For this analysis, 220 μL of 0.02 M phosphate buffer (at pH 7) or Tris-HCl buffer were mixed together with 20 μL of 7.89 mM DTNB and 50 μL of supernatant. After 5 min of incubation at room temperature, 10 μL of 78.9 mM ATC were added to start the reaction. The enzyme activity was then followed by an absorbance plate reader at 405 nm at room temperature for a total of 10 min (VersaMax microplate reader from Molecular Devices, California, USA). The enzyme activity was followed by the production of the yellow coloured 5-thio-2- nitrobenzoic acid (TNB) anion. The production of TNB is based on coupling of the following reactions:
Acetylthiocholine (ATC) → thiocholine + acetate Thiocholine + dithiobisnitrobenzoate (DTNB) → 5-thio-2-nitrobenzoic acid (TNB)
The conversion of DTNB to TNB can be used as a measure of the hydrolysis of ATC into thiocholine. ATC is produced from hydrolysation of the neurotransmitter acetylcholine by AChE. AChE inhibitors will induce a decrease in the production of ATC and therefore a decrease in the production of TNB will be observed. The change in absorbance per min was used to calculate the AChE activity: AChE activity (µmol ATC/ min / mg protein)= [∆A × Volt × 1000] ε × light path × Vols × [protein], where ΔA = change in absorbance (OD) per min at 405 nm, corrected for spontaneous hydrolysis, Volt = total assay volume (0.300 mL), ε = extinction coefficient of TNB (1.36 x 104 M/cm), light path = microplate well depth (1 cm), Vols = sample volume (in mL), and [protein] = concentration of protein in the enzymatic extract (mg/ mL). Total protein concentrations were determined according to Lowry et al. (1951), adapted to measurement by plate reader, using Immunoglobulin G (IgG) as protein standard.
23
Lipid content Mussel tissues were extracted twice with a mixture of cyclohexane and acetone. The acetone was washed out with 0.5% NaCl in water, before the cyclohexane was transferred to a clean glass container. The cyclohexane extracts where evaporated to dryness and the amount of lipid determined gravimetrically (Marsh and Weinstein, 1966). Gonad histological evaluation (spawning) Sampled mussels were kept in ice after collections and dissections were performed immediately on board. Gonadal tissues were dissected, placed in pre-labelled histological cassettes, and transferred into histological fixative for the wax section preparation. Tissue samples were no thicker than 1 cm to ensure proper fixation. Samples were then stored at 4˚C until embedding. Mussel sections were dehydrated in alcohols. Tissues were cleared in methyl benzoate, rinsed in benzene, and embedded in paraffin. Histological sections (5 µm) were cut using a microtome, mounted on slides, dried at 37 ºC for 24 h and stained with haematoxylin and eosin. Tissues were examined for gonad maturation and health parameters related to reproductive conditions, based on the scheme proposed by Seed (1969). Gonadal status evaluation scheme: • Stage 0) no developed gonad/post spawn; • Stage 1) no gametes present/no evidence of recent spawn; • Stage 2) early gametogenesis/gonad with primary oocyte or spermatocytes; • Stage 3) 50% of gonad mature/half of gonad occupied by secondary oocytes or spermatids; • Stage 4) gonad fully mature; • Stage 5) gonad fully mature spawning capable with ova compacted in polygonal confirmation or spermatozoa (functional tail). The spawning progression was also assessed as followed: o Stage 1) evidence of spawning with some polygonal ova or spermatozoa remaining within gonad, gamete nuclei intensely stained purple; o Stage 2) evidence of some spawning with gonad containing <50% mature gametes; o Stage 3) evidence of spawning and gonad with some gametogenesis, similar to gamete developing Stage 3; o Stage 4) active spawning evident with fully mature gametes and few developing gametes. The following histological abnormalities were also examined: atresia of gonadal tissue; granulocytomas; presence of haemocytes (nodules); presence of parasites. Samples for these analyses were prepared on board the vessels and coded by numbers, which did not directly identify the location or the group. Slides were prepared at Helse Stavanger by the hospital personnel and were analysed blind by an operator not involved in the sample collection or preparation. The data treatment, after revealing the samples codes, was carried out by the project manager. Micronucleus assay (DNA damage) DNA damage in haemolymph cells was determined as a biomarker of genotoxicity in mussels (Venier et al., 1997). MN assay was performed on 20 individuals per station, for 10 stations: 6 stations in the platform safety zone (stations: 4, 10, 11, 12, 13, and 14). Stations 16 and 17 were outside. Two reference stations were numbered as 18 and 19. Mussel haemolymph was extracted from the posterior adductor mussel (Figure 7a). From each mussel, 800 μl of haemolymph was extracted with a syringe filled with 400 μl of seawater solution (seawater + EDTA) (Figure 7a). Samples were centrifugation at 800 rpm and haemocytes were applied to glass slides. For cell fixation, slides were dipped in cold methanol for 15 min. Slides were left to air dry at room temperature and stored in a microscope slide box
24 afterwards. Slides were stained with 3 % (v/v) Giemsa solution for 10 min. Thereafter, cover slips were glued to the slides using DPX Mounting Media. The slides were left at room temperature for 24 hours in order to dry and then were stored for scoring (Figure 7 b and c). The MN frequency scoring was performed using a blind reader. Slides for this analysis were prepared on board the vessels, and coded by numbers, which did not directly identify the location or the group. Slides were prepared stained and analysed blind by an operator which was not involved in the sample collection. The data treatment, after revealing the sample codes, was carried out by the project manager. Samples were scored for MN frequency under a light microscope (Olympus IX71 inverted microscope). About 2500 cells per mussel were analysed at 1000× microscope magnification (Figure 7d).
Figure 7 Extraction of haemolymph (a); Stained slides (b); Stored slides in box (c); Olympus IX71 microscope used for scoring slides under the microscope and taking pictures (d)
Only unharmed cells with obvious cellular and nuclear membranes were scored in accordance with the MN definition of Schmid (1975). In brief, spots in cells can be considered as MN if they agree with the following criteria: - They must be rounded and oval in shape; - Their diameters are less than one-third of the main nuclei (a cell with two same sized nuclei in one cell are considered as bi-nucleated cells); - Their staining intensity is the same or a little lower than the nucleus; - They must be completely separated from the nucleus (those who link to the nucleus are considered as extruded nuclear material, i,e, nuclear buds). 2.3.4 Biological analyses in fish Condition indices With the aid of a measuring board and a digital fish scale (Berkley® model BTDFS50-1) the length and total weight of each fish was measured on board the survey vessel. Fish were sexed by visual examination of their gonad. A motion compensated balance (Marel M2000 series) was used to measure the total liver and gonad weights on board the vessel. The Fulton’s condition index was determined by calculating the ratio between total weight and the cube of the fork length of the fish (Hansson et al., 2017). The liver somatic index (LSI) reflects the animal nourishment status. The gonadosomatic index (GSI) reflects the animals’ reproductive status (Hansson et al., 2017). 25
Condition index (CI) = [weight (g) /(length)3 (cm)] ×100. Liver somatic Index (LSI) = [liver weight (g) / fish weight (g)] × 100. Gonadosomatic Index (GSI) = [gonad weight (g) / fish weight (g)] ×100. Age determination Otolith pairs were removed from the cranium of individual fish and stored dry in labelled paper envelopes. Prior to reading, otoliths were snapped cleanly through the nucleus and soaked in water to reveal the annuli on the transverse surface. The annuli were scored from the transverse surface submerged in water with side light illumination and with the aid of a binocular microscope at 10-32x magnification (Mjanger et al., 2017). PAH metabolites Fixed fluorescence PAH metabolites were analysed using the fixed fluorescence wavelength screening method (Aas et al., 1998 and 2000). Bile samples were diluted 1:1600 in methanol: water (1:1). Slit widths were set at 2.5 nm for both excitation and emission wavelengths, and samples were analysed in a quartz cuvette. All bile samples were analysed by FF at the wavelength pairs 290/335, 341/383 and 380/430 nm, optimised for the detection of 2-3 ring, 4-ring and 5-ring PAH metabolites, respectively. The fluorescence signal was transformed into pyrene fluorescence equivalents (PFE) through a standard curve made by pyrene (Sigma St Louis, USA). Pyrene was measured at the same fluorimeter, with the same cuvette, same solvent, and with the same slit settings as the bile samples. It was, however, measured at the optimal wavelength pair of pyrene, 332/374 nm (excitation/emission). The concentration of PAH metabolites in bile samples was expressed as µg PFE/ mL bile. GC-MS: Fish bile samples were prepared for analysis as described by Jonsson et al. (2003; 2004). Briefly, 25–30 µL of bile was weighed accurately into a micro centrifuge vial. Internal standards (2,6-dibromophenol, 3-fluorophenanthrene and 1-fluoropyrene) and β-glucuronidase (3000 units) in sodium acetate buffer (0.4 M, pH = 5) were added and the solution left at 40°C for 2 h. The OH-PAHs were extracted with ethylacetate (4 times 0.5 mL), the combined extract dried with anhydrous sodium sulphate and concentrated to 0.5 mL. Trimethylsilyl (TMS) ethers of OH-PAHs were prepared by addition of 0.2 mL BSTFA and heating for 2 h at 60°C. TPA was added as a GC-MS performance standard before transferring the prepared samples to capped vials. Trimethylsilyl ethers of OH-PAHs (TMS-OH-PAHs) in fish bile samples were analysed by a GC-MS system consisting of a HP5890 series II Gas chromatograph, Shimudadzu QP2010 GCMS. Helium was used as carrier gas and the applied column was CP-Sil 8 CB-MS, 50 m x 0.25 mm and 0.25 µm film thickness (Varian). Samples and calibration standards (1 µL) were injected on a split / splitless injector with splitless mode on for one minute. The temperatures for the injector, transfer-line and ion source were held at 250oC, 300oC and 240oC, respectively, and the GC oven temperature programme was as follows: 80oC to 120oC at 15oC/ min, 120oC to 300oC at 6oC/ min and held at 300oC for 30 min. Mass spectra were obtained at 70 eV in selected ion mode (SIM). Based on the fragmentation pattern of non-alkylated TMS- O-PAHs (Jonsson et al., 2003), the molecular ions were selected for determination of both alkylated and non-alkylated TMS-O-PAHs. Tissue changes in liver Sampled fish were dissected on the vessel immediately after capture. Liver were dissected, placed in histological fixative solution and kept at 4˚C until further processing. Samples were dehydrated in alcohols and cleared in xylene and embedded in paraffin. Histological sections (3 μm thick) were cut using a microtome HM 355s (Microm, Bergman), mounted on slides, air dried and stained with haematoxylin and eosin.
26
Tissues were examined for health parameters related to physiological conditions, inflammatory and non-specific pathologies, and those associated with pathogen and parasite infections. All micrographs were captured using an AxioCam MRc5 (Zeiss) digital camera mounted on a Zeiss Axioplan 2 light microscope (Göttingen, Germany). All slides were analysed blind with no reference to its sample location. Detected histopathological liver lesions were assigned to one of the following groups: steatosis; circulatory disturbance; inflammatory changes; melanomacrophage aggregates; parasites and other pathological changes, according to a developed and adopted scoring system (Bernet et al., 1999; Feist et al., 2004). Vacuolation condition, macrovesicular and microvesicular steatosis were distinguished based on the size and the pattern of vacuoles present. Circulatory disturbances included various changes in normal structure of blood vessels (congestion, dilation, peliosis). Non-specific lesions were presented as: inflammatory changes (lymphocyte infiltration and granulomatosis); melano-macrophage aggregates, parasites, other pathological changes (degenerative – necrosis, proliferative – fibrosis, cirrhotic changes). According to the affected area or prevalence of each disorder within a specimen, all of the parameters were scaled using an established scoring system (Table 6).
Table 6. Categories for histological liver lesions and the scoring system used for their quantification. Category Score range Score description 0-absent Steatosis (normal cyclical, non- 1-area affected 0-3 pathological status of the liver) 2-some areas affected 3–distributed through the entire sampled tissue 0-absent Circulatory disturbances 0-2 1-sporadic/small area affected 2-some areas affected 0-absent Inflammatory changes 0-2 1-sporadic 2-multiple/widespread 0-absent 1-area affected (1-2 cases) Melano-macrophage aggregates 0-3 2-some areas affected/more than 2 in a sample 3–distributed through the entire sampled tissue 0-absent Other pathological changes 0-2 1-sporadic 2-multiple/widespread 0-absent Parasites 0-1 1-present
DNA damage in liver and gut, DNA adducts Analysis of DNA adduct patterns in liver and intestine was performed using the 32P postlabelling method (Reichert and French, 1994; Le Goff et al., 2006). The protocol used by AdnTox was suitable for the detection of so-called "bulky" DNA adducts, which are additional compounds in DNA associated to complex molecules such as PAHs. Preparation of DNA Fish liver and intestine samples were air freighted on dry ice to the AdnTox laboratory, France, for DNA adduct analysis. After receipt, samples were stored at -80°C until required for DNA extraction. Small pieces of tissue (70 to 120 mg of liver; 150-200 mg of intestine) were cut on ice. 1.5 ml of sucrose (0.32 M) was added and mixed thoroughly to lyse tissue (Tissue lyser, Qiagen: 20 Hz) for 2 min (liver) or 3 min (intestine). Samples were centrifuged at 800 g for 10 min, at 4°C.
27
Pellets were dissolved with 1.2 ml of 1 mM EDTA and 20 mM Tris, pH 7.4. 100 µl 10% SDS was added and vortexed for 1 min. Vials were incubated for 30 min at 37°C with 0.2 mg / ml RNase A and 33.4 U RNase T1. Then incubate for 2.5 h at 37°C with 0.50 mg/ml proteinase K until complete digestion. 0.5 volume (0.7 ml) of saturated phenol was added and vortexed for 1 min, before centrifugation for 5 min at 5000 rpm. The upper phase (aqueous phase) was removed and transferred to a clean tube. 0.5 volume (0.7 ml) of CIP (phenol + Sevag 1/1) was added and vials vortexed for 1 min and centrifuged for 5 min at 5000 rpm (+4°C). The upper phase was removed and transferred to a clean tube. 0.5 volume of Sevag (chloroform + isoamyl alcohol (1/24)) was added and vortexed for 1 min before centrifugation for 5 min at 5000 rpm (+4 °C). The upper phase was removed. Precipitation of DNA: 0.1 volumes of a solution of 5 M NaCl and 2 volumes of cold ethanol (stored at -20°C) were added to the aqueous phase. Vials were gently shaken and vortexed. The DNA was air dried before addition of 150 µl ultra-pure water. Spectrophotometric quantification of DNA solutions (Nanodrop, Thermo Scientific): 1 unit of absorbance at 260 nm corresponds to a double-stranded DNA solution concentration equal to 50 µg/ml. Quality criteria selected: 1.85
Hydrolysis - Prepare 5 µg of DNA/analysis - Dry sample (Speed Vac SV, 15 minutes) - Hydrolyse of DNA : MN : 0.7 µg / 5 µg DNA SPDE : 10 mU / 5 µg DNA 3.5 h / 37°C
28
+ Buffer solutions
MN= micrococcal nuclease (Sigma); SPDE: spleen phosphodiesterase (Calbiochem)
Enzymatic enrichment with NP1 - Dry sample (SV) after hydrolysis - NP1: 5 µg / 5 µg DNA + Buffer solutions 30 minutes / 37°C - Stop incubation with a tris base solution (1.8 µl/sample) NP1= Nuclease P1 (Sigma)
32P radioactive labelling -Add to sample: - PNK : 10U/5µg DNA - PNK buffer A 1X 30 minutes / 37°C - 32P-ATP : 25 µCi / 5 µg DNA PNK : polynucleotide kinase (+ buffer A 10X ; Fermentas)
Chromatographic separation Separation of radiolabelled adducts in the previous step was performed by bidirectional thin layer chromatography on polyethyleneimine (PEI) cellulose sheet (12 x 10 cm) (Macherey Nagel), by using D1 to D4 successive migrations (D1 and D4 being “clean-up” migrations). Solvent (mobile phase) composition was provided for each migration.
D1: - Mobile phase: Na Phosphate 1 M. pH 6 - Wash sheet in deionized H2O after D1 - Dry sheet - Cut up PEI Cellulose Sheet (transfer step)
D2: - Mobile phase: Li formate 4.5 M Urea 8.5 M pH 3.5 - Wash sheet in deionized H2O - Dry sheet
D3: - Mobile phase: Li chloride 1.6 M Tris 0.5 M pH 8 Urea 8.5 M - Wash sheet in deionized H2O - Dry sheet
D4: - Mobile phase: Na Phosphate 1 M. pH 6.8 - Dry sheet
29
DNA adduct patterns were detected by autoradiography (Kodak X-OMAT / BIOMAX). The optimum exposure time is a function of radioactive signal strength (exposure time at -80°C: from 12 to 72 h).
Quantification / results analysis Quantification was performed using the scintillation counting of spots cut on chromatographic sheets, by Cerenkov mode, and on the basis of the radioactive signal associated to the labeling of a known quantity of DNA adducts (positive control: 5 µg of a DNA which contained 110.7 adducts for 108 normal nucleotides).
DNA damage in lymphocytes, Comet assay: Microscope slides were prepared on board the survey vessel with blood samples taken from the caudal vein of freshly caught fish. A 10 μL volume of the blood was diluted 1000 fold in ice cold PBS buffer. A 15 μL volume of this diluted blood solution was added to 85 μL of warmed low melting point agarose. Then 7 μL of this agarose/blood solution was placed on an agarose pre-coated slide. Each fish had one spot on 6 pre-coated slides, 2 slides were labelled as LYS, 2 as FPG and 2 as BUF, relating to the treatment of these slides once back in the laboratory. The slides were kept in cool (4°C) lysis buffer, whilst stored on the survey vessel and transported back to the laboratory in Oslo. The slides were processed by NorGenoTech AS (Oslo) between 5 and 10 days of being taken. A high-through put method as described in Brunborg et al. (2014) was followed. Briefly, cells embedded in agarose on a microscope slide were lysed, leaving the DNA as a nucleoid, attached to the nuclear matrix. After brief incubation in alkali, gels were electrophoresed at high pH. DNA is attracted to the anode, but moves appreciably only if breaks are present. After neutralisation and staining, the nucleoids (visualised by fluorescence microscopy) resemble comets; the relative intensity of the comet tail reflects the frequency of DNA strand breaks (SBs). Base alterations (e.g. oxidation) were measured by digesting nucleoids with lesion- specific enzymes; formamidopyrimidine DNA glycosylase (FPG), to detect 8-oxoguanine and other purine oxidation products. To increase the number of samples that could be handled simultaneously, the high throughput version of the comet assay with 12 mini gels on one microscope slide coated with polycarbonate film substrate was adopted. The test was performed using three different treatments: 1) Lysis only (to measure SBs); 2) Incubation with FPG buffer after lysis; and 3) Incubation with FPG after lysis (to measure oxidised guanine, oxidised bases). Results are expressed as % DNA in tail (median of, in general, 50 comets per sample). % DNA in tail is linearly related to break frequency over the range of damage levels expected. Net FPG-sensitive sites are calculated as the difference between scores for 3 and 2. Acetylcholineesterase inhibition (AChE) The same protocol used for analysing AChE activity in mussels was used for fish. Details are reported above in section 2.3.3. Alkylphenol metabolites Fish bile samples were prepared for analysis as described by Jonsson et al. (2003; 2004). Briefly, 25– 30 µL of bile was weighed accurately into a micro centrifuge vial. Internal standards (2,6dibromophenol, 3-fluorophenanthrene and 1-fluoropyrene) and β-glucuronidase (3000 units) in sodium acetate buffer (0.4 M, pH = 5) were added and the solution left at 40°C for 2 hours. The OH-PAHs were extracted with Solid Phase Analytical Derivatisation (SPAD). Trimethylsilyl ethers of OH-APs (TMS-OH-APs) in fish bile samples were analysed by a GC- MS system, Shimudadzu QP2010 GC-MS. Helium was used as carrier gas and the applied column was CP-Sil 8 CB-MS, 50 m x 0.25 mm and 0.25 μm film-thickness (Varian). Samples
30
and calibration standards (1 μL) were injected on a split/splitless injector with splitless mode on for one minute. The temperatures for the injector, transfer-line and ion source were held at 280°C, 300°C and 240°C, respectively, and the GC oven temperature programme was as follows: 80°C to 120°C at 15°C min-1, 120°C to 300°C at 6°C min-1 and held at 300°C for 30 min. Mass spectra were obtained at 70 eV in SIM. Based on the fragmentation pattern of non- alkylated TMS-O-APs (Jonsson et al., 2003); the molecular ions were selected for determination of both alkylated and non-alkylated TMS-O-APs.
Gene expression I – qPCR to select gene transcripts The gene expression was done by means of qPCR analysis (Lacroix et al., 2014). RNA isolation The total RNA was extracted from cod liver tissue with Promega Reliaprep simply RNA HT 384, art nr X9601(Nerliens) on a Biomek 4000 Laboratory Automated Workstation (Beckman Coulter) according to the manufacturer’s instructions and quantitated using a NanoDrop™- 1000 spectrophotometer (Thermo Scientific). The RNA samples were normalised to the concentration of 100 ng/µL using the Biomek 4000 Laboratory Automated Workstation (Beckman Coulter). Reverse transcription was carried out using SuperScript® VILO™ cDNA Synthesis kit, art nr 11754050 (Life technologies) according to the manufacturer’s instructions, and the total RNA input was 500 ng in each reaction in a total volume of 10 µL. Quantitative real-time RT-qPCR PCR primers used to quantify the selected genes in cod were designed based on genome sequence for the respective species and shown in Table 7. The qPCR assay was run using Brilliant III Ultra-Fast SYBR® Green QPCR Master Mix, Catalog # 600882 (Life Technologies AS) according to the manufacturer’s instructions, with 2 µL of cDNA diluted 1 to 10 in a reaction mix containing 400 nM of forward primer, 400 nM of reverse primer in a total volume of 7 µl on a 384 well-plate. For the qPCR assay, amplification and fluorescence detection were performed by a QuantStudio™ 5 Real-Time PCR System (Applied Biosystems) for 40 cycles. Quality controls “no template controls” (ntc) and “no amplification controls” (nac) were run for quality assessment for each PCR assay. Mean normalized expression (MNE) of the target genes was determined using a normalization factor based upon EEF1A, UBA52 and RPL37 for cod, calculated by the QuantStudio™ Design & Analysis Software. Selected gene transcripts were: CYP1A, AHR2, AHRR, GADD45A, GADD45G.
Table 7. PCR primers, accession or contig numbers and amplicon sizes.
Accession no./contig Amplicon Gene name Marker for Forward primer Reverse primer Gene ID name size (bp)
Cytochrome P450, family 1, subfamily CYP1A1 A1 Detoxification >GmE090818c9407 AAGGTCACGAAGCGTTCGTT ACGAGTTCGGGAAGGTTGTG 118 AHR2 Aryl hydrocarbon receptor 2 Detoxification EX728781 CAACCGGCGGTCCACAT GCACCATGCAGTTGCCAGTA 132 AHRR Aryl-hydrocarbon receptor repressor Detoxification >F2Z8C0H01CDEIU GAGGTGTTTGTCTCCCGTCTCT TTCACCAGGTCCACGCAGTA 77 growth arrest and DNA-damage- DNA GADD45A inducible, alpha damage/apoptosis >GmE100127i27876 CAGCTCCGCCGAGTATCTGT GACGTTGCCCTTCAGATTCAC 115 growth arrest and DNA-damage- DNA GADD45G inducible, gamma damage/apoptosis >GmE090818c21550 AAGACCTGCTGGTAGCTGCAA ACTCCTCATCGGTGGCAAGA 123 CCGAGAAGCGCAAGAGAAAG GGTGGTACCTTCCCGGAATC RPL37 Ribosmal protein L13 Reference gene EX738140 131 Ubiquitin A-52 residue ribosomal UBA52 protein fusion product 1 Reference gene EX735613 GGCCGCAAAGATGCAGAT CTGGGCTCGACCTCAAGAGT 69 Eukaryotic translation elongation EEF1A factor 1 alpha 1 Reference gene EX722124 CGGTATCCTCAAGCCCAACA GTCAGAGACTCGTGGTGCATCT 93
31
EROD (2 Ethoxyresorufin O-deethylase) EROD activity was measured in the microsomal fraction of fish livers based on the method of Burke and Mayer (1974), modified to a fluorescence plate reader by Eggens & Galgani (1992). Microsomes were prepared on ice with pre-cooled equipment and solutions. Cryo-preserved liver samples were homogenized in a potassium phosphate buffer (0.1 M, pH 7.8) containing KCl (0.15 M), dithiothreitol (DTT) (1 mM), and glycerol (5% v/v), using a Potter-Elvehjem Teflon–glass homogenizer. The homogenate was centrifuged (10,000 × g; 30 min, 4°C) before the supernatant was re-centrifuged (50,000 × g; 120 min, 4°C). The microsomal fraction was obtained by resuspending the resulting pellet in potassium phosphate buffer (0.1 M, pH 7.8) containing KCl (0.15 M), DTT (1 mM), EDTA (1 mM), and glycerol (20% v/v). Microsome samples were diluted to ~2 mg/ mL in buffer and pipetted (50 μL) in 6 technical replicates onto a 96-well microplate. Pre-prepared resorufin standards (duplicates) were then added to subsequent wells. Reaction mixture (200 μL, containing 0.1 M potassium phosphate buffer, pH 8, and 3 μM 7 ethoxyresorufin) was added to the sample wells, before NADPH solution (2.4 mM in final well volume of 275 μL) was added to initiate the reaction. Transformation of 7- ethoxyresorufin to resorufin was read in 8 steps on the plate reader. Excitation was at 530 nm and fluorescence emission was measured at 590 nm. EROD activity values were normalized to the protein content in the microsomal fraction and expressed as pmol/ min/ mg microsomal protein. Protein concentrations were determined according to Lowry et al. (1951), adapted to measurement by plate reader. The protein standard was bovine gamma globulin. CYP1A (ELISA analysis) Cyp1A analysis was done using an ELISA method (Nilsen et al., 1998). Buffer for homogenising 0,1 M sodiumphosphate (NaH2PO4·H2O), 0,15 M potassiumchloride, 1 mM EDTA, 1 mM DTT, 10% v/v glycerol, pH 7,4. Homogenising of liver and preparation of post-mitochondrial supernatant (PMS) Approx. 0,5 g liver was added to the homogenising buffer (2 ml / 0.5 g liver) and homogenised with use of a Precellys 24 homogenizer and ceramic beads (PhastPrep Lysing Matrix D (MP Biomedicals) for 20 sec. The homogenate was transferred to Eppendorf vials and centrifuged for 20 min at 12000 xg, 4˚C. Samples were stored at -80˚C. Measurements of protein content The total protein content quantification was performed according to Bradford (1976). PMS- fraction of fish liver was diluted 1:1000 in dH2O. 50 µl of sample (in triplicate) was added to an ELISA-plate (Nunc 96 wells, flat bottom). 300 µl Coomassie G-250/17% phosphoric acid (1:1) was added to the samples and incubated for 5 min. Absorbance was measured at 595 nm by plate reader (Tecan SPECTRA Fluor). Protein concentration was determined by standard curve with bovine serum albumin. ELISA Performed as described in Nilsen et al. (1998). 1 µg total protein added per well, 4 parallels per sample, divided on two plates. For measurements of CYP1A1 in cod liver we used monoclonal mouse anti-cod CYP1A (NP-7, Biosense, Norway), diluted 1:1000. For CYP1A measurements in saithe, monoclonal rabbit anti-fish CYP1A (C10-7, Biosense, Norway), diluted 1:500 was used. For secondary antibodies we used polyclonal goat anti-mouse from DacoCytomation, Denmark, diluted 1:2000. Plates were incubated with TMB substrate for 22.5 minutes before addition of 0.5 M H2SO4 and absorbance read at 450 nm.
32
2.4 Quality Assurance Field and laboratory work at International Research Institute of Stavanger (now part of NORCE) performed in 2017 and 2018 were in accordance with the quality assurance systems NS-EN-ISO 9001:2000 and NS-EN ISO 14001:2004. To limit analysis bias, it is highly recommended that environmental samples are analysed without the knowledge and identification of where they were collected from. This approach is known commonly as blind sample analysis, where it is understood that the potential bias is unintentional and/or subconscious and no dishonesty is implied. Where possible, chemical and biological effects analysis was performed blind on coded samples. Coded samples were used for comet, EROD, AChE and PAH, metal and lipid body burden with the analysts unaware of the significance of the coding system. NIVA's environmental system was certified in accordance with ISO14001:2004 in January 2014. The certificate is awarded for delivery of services and products based on research and development in environment-related topics. NIVA chemistry are accredited according to the international standard NS-EN ISO / IEC 17025. General requirements for test- and calibration laboratories competence are regularly quality assured through annual accreditation visits and international SLP participation. IMR chemistry are accredited according to the international standard NS-EN ISO / IEC 17025. General requirements for test- and calibration laboratories competence are regularly quality assured through annual accreditation visits and international SLP participation (Quasimeme). 2.5 Statistical methods Statistical differences between the groups of biological data were assessed with analysis of variance (ANOVA). Homogeneity of variance for the different groups was checked using the Levene’s test and data were log transformed when necessary to obtain homogeneity. Where homogeneity could not be achieved, non-parametric analysis was performed, either as a Kruskal-Wallis test or Mann-Whitney U test. Where homogeneity of variance was achieved a Dunnett’s or Tukey post-hoc test was performed with the parametric ANOVA to compare significant differences from the reference group. The level of significance was set at p<0.05. For the lysosomal membrane stability assay, P‐values were calculated by using Scheffé F‐test within One‐Way ANOVA test in SPSS software. For the mussel condition index, normal distribution and homogeneity of variance were verified by Shapiro-Wilk test (common test for normality) before statistical treatment. According to Shapiro-Wilks test of normality, CI data in all stations were normally distributed (P- value>0.05). Hence CI results from exposed station were compared with reference stations (18 and 19) by using One-Way ANOVA test in SPSS software.
Integrative assessment The Integrative Biological Response (IBR) index was applied in this study, as in previous WCM studies. This data treatment was developed to integrate biochemical, genotoxic and histochemical biomarkers (Beliaeff and Burgeot, 2002). The method is based on the relative differences between the biomarkers in each given data set. Thus, the IBR index is calculated by summing-up triangular star plot areas (a simple multivariate graphic method) for each two neighbouring biomarkers in a given data set. The procedure is as follows: 1) calculate the mean and standard deviation for each sample group; (2) standardise the data for each sample group: xi'=(xi-x)/s; where, xi'=standardised value of the biomarker; xi=mean value of a biomarker from each sample; x=general mean value of xi calculated from all compared sample groups; s=standard deviation of xi calculated from all samples; (3) add the standardised value obtained for each sample group to the absolute standardised value of the minimum value in the data set (i.e. yi=xi'+|xmin'|); (4) calculate the Star Plot triangular areas by multiplication of the obtained 33 standardised value of each biomarker (yi) with the value of the next standardised biomarker value (yi+1), dividing each calculation by 2 (Ai=(yi x yi+1)/2); and (5) calculate the IBR index which is the summing-up of all the Star Plot triangular areas (IBR=ΣAi) (Beliaeff and Burgeot, 2002). Since the IBR value is directly dependent on the number of biomarkers in the data set, the IBR value was divided by the number of biomarkers used in each case (n) to calculate IBR/n, according to Broeg and Lehtonen (2006).
2.6 Samples and data storage The processed material used for analyses is not available for reanalysis. All results are provided as raw data in an electronic form to the client.
34
3. Results and discussion This chapter presents the results of the survey. As specified in the Norwegian Environment Agency guidelines, the report is targeted towards a specified readership including, oil and gas companies, environmental authorities, research institutions, and consultancy firms. 3.1 Current and temperature data Three mussel rigs, including two positioned near Statfjord A (SFA 500/SE2 and SFA 1000/SE) and one positioned at Statfjord B (SFB 500/SE), were equipped with current meters. Data are reported in Figure 8 and 9 and in Appendix 3 (Current meter reports). One of the current meters (SFB 500/SE) was equipped with a depth sensor that was able to provide current data at different depths. Data regarding the dominant current direction was predicted by the DREAM model simulation, which was performed during the design of the monitoring programme prior to the field study. As an example, figure 8 shows the current direction at SFB 500/SE during the 6 weeks of exposure as total flow per day (m3/m2/day) for the top (30 m), middle (50 m) and bottom (70 m) of the water column. Station SFA 500/SE2 and SFA 1000/SE data are shown in figure 9 and indicate current direction at a depth of 20 m in the water column. The flow recorded at stations SFA 500/SE2 and SFA 1000/SE were similar to each other, but a little narrower than that measured at station SFB 500/SE. This may indicate a potentially narrower distribution of the PW plume at Statfjord A compared to Statfjord B
Figure 8 Data regarding the current at station SFB 500/SE during the 6 weeks of exposure as total flow per day (m3/m2/day).
35
Station SFA 500/SE2 Station SFA 1000/SE
Figure 9 Data regarding the current at station SFA 500/SE2 and SFA 1000/SE, around Statfjord A, during the 6 weeks of exposure as flow per day (m3/m2/day).
Regarding the temperature in the water during the exposure study with caged mussels, at station SFB 500/SE (Statfjord B, 500 m from the discharge point), the temperature varied between 8.1 °C (value recorded after 1 week of exposure) and 9.8 °C (value recorded after 6 weeks of exposure). The temperature increased slowly from the deployment day for 5 weeks, with an increase from 9°C to 9.8°C in the last week of exposure (Figure 10). At station SFA 500/SE2 (Statfjord A, 500 m from the discharge point), the temperature varied from 7.4°C (value recorded after 1 week of exposure) to 9.4°C (value recorded after 6 weeks of exposure) (Figure 11). At station SFA 1000/SE (Statfjord A, 1000 m from the discharge point), the temperature varied from 7.5°C (value recorded after 1 week of exposure) to 9.5°C (value recorded after 6 weeks of exposure) (Figure 12). The temperature profile during the 6 weeks of exposure was almost identical at all stations around the two platforms Statfjord A and B.
36
Figure 10 Temperature profile during the 6 weeks of exposure at station SFB 500/SE, which was located at the Statfjord B platform area, 500 m from the discharge point in a South-East direction.
Figure 11 Temperature profile during the 6 weeks of exposure at station SFA 500/SE2, which was located at the Statfjord A platform area, 500 m from the discharge point in a South-East direction.
Figure 12 Temperature profile during the 6 weeks of exposure at station SFA 1000/SE, which was located at the Statfjord A platform area, 1000 m from the discharge point in a South-East direction.
37
The pressure sensors revealed that the mussel cages were at the designed depth for the 6 weeks of deployment, confirming the continuous exposure of the organisms to the predicted PW plume. As an example, the data from station SFB 500/SE are show in Figure 13. The mussel cages were maintained at the following depth, according to the rig log data: around 23 m at station SFB 500/SE, around 20 m at station SFA 1000/SE and around 18 m at station SFA 500/SE2.
Figure 13 Pressure data (in bar) from station SFB 500/SE, representing the position of the mussel cage in the water column.
The CTD unit (ST 200W CTD, Saiv Norway) was equipped with a turbidity sensor (Seapoint INC, USA), and was placed at station SFB 500/SE in order to continuously monitor the density of particles, such as algae during the exposure period of mussels. The chlorophyll distribution showed that the caged mussels had available food during the exposure duration (Figure 14).
38
Figure 14 Chlorophyll distribution at station SFB 500/SE, x-axis unit µg/L, y-axis unit m (depth)
39
3.2 Chemical and biological analyses in caged mussel 3.2.1 Identification of Mytilus spp. Mytilus species were identified from a sub sample of 96 randomly selected mussels from the same population that was used in the monitoring programme (Figure 15). The results showed that 82% of the mussels were Mytilus edulis. Compared to the last WCM study in 2012, which used mussels collected from the same location, a decrease of 6% in M. edulis dominance was observed. The data from WCM2012 are reported in figure 15 for comparison. Since different species may have a different capability to accumulate and respond to pollutants (Beyer et al., 2017), it will be important to consider this decrease in the proportion of M. edulis individuals in Norwegian populations in future WCM surveys when using caged mussels.
Figure 15 Mussel speciation in 96 individuals that are representative of the caged mussels in the present study (2017), and the previous WCM when mussels were used (2012). For 2012 n=66, for 2017 n=96.
3.2.2 PAH and NPD concentrations The sum of PAH includes the following compounds: naphthalene, C1-naphthalenes, C2- naphthalenes, C3-naphthalenes, acenaphthylene, acenaphthene, fluorene, dibenzothiophene, phenanthrene, anthracene, C1-phenanthrenes, C2-phenanthrenes, C3-phenanthrenes, C1- dibenzothiophenes, C2-dibenzothiophenes, C3-dibenzothiophenes, fluoranthene, pyrene, benzo(a)anthracene, chrysene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(e)pyrene, benzo(a)pyrene, perylene, indeno(1,2,3-cd)pyrene, dibenz(a,h)anthracene, and benzo(g,h,i)perylene. The LOQ for the single PAH compounds was 0.5 µg/kg (w.w.), except for naphthalene (1 µg/kg w.w.), whilst 2 µg/ kg (w.w.) was the LOQ for the alkylated NPD compounds. The sum of PAH concentration in mussel tissues showed higher body burden concentrations in mussels located closest to both Statfjord A and B platforms (Figure 16). Highest Sum PAH concentrations were found in mussels from station SFB 500/SE (409 μg/kg) and SFA 500/SE1 (402 μg/kg), 500 m and approximately 140° and 150° from Statfjord B and A platforms respectively. Sum of PAH concentrations reduced with distance from the platforms. Lowest concentrations were found in both reference groups and T0 groups. 40
In comparison with other WCM programmes, where mussel sum PAH concentrations were measured, the maximum concentrations reported 500 m from Statfjord A and B, were lower than those reported from Troll C in 2012 (Pampanin et al., 2013). Sum PAH body burden in mussels positioned 500 m from the Troll C platform ranged between approximately 1500 to 900 μg/ kg (w.w.), 2 to 3 times higher than sum PAH measured in the mussels 500 m from the Statfjord platforms. The sum PAH concentrations in mussels 500 m from Statfjord were comparable to the concentrations detected in mussels 1000 m from Troll C. At Gullfaks C in 2011, the sum PAH concentrations in mussels at 500 m from the platform (1500 to 850 μg/ kg w.w.) were similar to those reported in mussels 500 m for Troll C (Brooks et al., 2011).
Figure 16 Sum of PAH concentration in mussels from the different stations around the Statfjord A and B platforms, including reference and day 0 (t0) mussels. Median, quartiles (box), 10/90 percentiles (bar), n=5.
The NPD compounds make up most of the sum PAH concentration and it is therefore not surprising that almost identical profiles with distance from the Statfjord A and B platforms were observed for total naphthalenes, total phenanthrenes and total dibenzothiophenes (Figures 17- 19). Overall, naphthalenes were the most abundant followed by phenanthrenes and dibenzothiophenes. At Troll C, naphthalenes and phenanthrenes were almost equally contributing to the overall PAH concentration with lower contributions from dibenzothiophenes (Pampanin et al., 2013). This was also seen in the PAH body burden of mussels from the Ekofisk platform (Brooks et al., 2011). At Gullfaks C, dibenzothiophenes contributed almost equally to the Sum PAH together with naphthalenes and dibenzothiophenes (Brooks et al., 2011). This indicates the slight differences between the produced water composition with respect to PAH for the different oil and gas platforms.
41
Figure 17 Total naphthalene concentrations in mussels from the different stations around the Statfjord A and B platforms, including reference and day 0 (t0) mussels. Median, quartiles (box), 10/90 percentiles (bar), n=5.
Figure 18 Total phenanthrene concentrations in mussels from the different stations around the Statfjord A and B platforms, including reference and day 0 (t0) mussels. Median, quartiles (box), 10/90 percentiles (bar), n=5.
42
Figure 19 Total dibenzothiophene concentrations in mussels from the different stations around the Statfjord A and B platforms, including reference and day 0 (t0) mussels. Median, quartiles (box), 10/90 percentiles (bar), n=5.
The relative concentration of alkylated PAH to parent compound
PAHs found in coal and petroleum often contain one or more methyl (C1), ethyl (C2), propyl (C3), butyl (C4), or (occasionally) higher alkyl substituents on one or more of the aromatic carbons. The alkylated PAHs are generally more abundant than the parent PAHs in petroleum, however they are less abundant than the parent PAHs in pyrogenic PAH mixtures. Therefore, the likely source of PAHs in the environment (either petrogenic or pyrogenic) can be obtained by calculating the ratio of alkylated PAH to parent PAH. The ratio of alkylated: parent PAH in mussels, with respect to the NPD compounds are displayed (Figure 20-22). As would be expected for mussels located close to offshore oil and gas platforms, higher ratios of alkylated to parent compound were found in mussels near the Statfjord platforms with much lower ratios in the day 0 (t0) and reference groups. Similar ratios, indicating a petroleum source, have been reported in previous WCM programmes were mussels have been used (e.g. Gullfaks C, Troll C and Ekofisk).
43
Figure 20 Ratio of alkyl-naphthalenes/ naphthalene in mussels from the different stations around the Statfjord A and B platforms, including reference and day 0 (t0) mussels. Median, quartiles (box), 10/90 percentiles (bar), n=5. Alkylated groups include C1 to C3.
Figure 21 Ratio of alkyl-phenanthrenes/ phenanthrene in mussels from the different stations around the Statfjord A and B platforms, including reference and day 0 (t0) mussels. Median, quartiles (box), 10/90 percentiles (bar), n=5. Alkylated groups include C1 to C3.
44
Figure 22 Ratio of alkyl-dibenzothiophenes/ dibenzothiophene in mussels from the different stations around the Statfjord A and B platforms, including reference and day 0 (t0) mussels. Median, quartiles (box), 10/90 percentiles (bar), n=5. Alkylated groups include C1 to C3.
45
3.2.3 Metal body burden A total of 11 metals were measured in the whole mussel homogenates of pooled mussels from the different exposure and reference groups. There was no apparent relationship between metal body burden concentration and distance from the platforms for any of the metals measured (Table 8). One elevated concentration of aluminium (5000 μg/kg w.w.) was reported in one replicate. This was unlikely to reflect the actual accumulation of this metal in the mussels, but rather an unknown cross-contamination. This value was therefore treated as an outlier and not included in the summary table.
Table 8 Metal concentrations in whole mussel homogenates from the different stations. Data expressed as μg/ kg (w.w.) (minimum and maximum values, n=5). Values in parenthesis denote limits of quantification for the specific metal.
As Pb Cd Cu Hg Ni Zn Al Ba Fe Co Station (0.05) (0.03) (0.001) (0.02) (0.005) (0.04) (0.5) (15) (0.5) (0.5) (0.003) min max min max min max min max min max min max min max min max min max min max min max SFB 500 SE 1.90 2.20 0.07 0.09 1.10 1.40 1.10 1.20 0.012 0.016 0.26 0.48 12.00 17.00 16.00 18.00 0.54 0.61 16.00 22.00 0.09 0.09 SFB 1000 SE 1.70 2.00 0.06 0.08 1.10 1.30 0.79 1.00 0.008 0.013 0.40 1.20 11.00 13.00 15.00 18.00 0.51 0.65 12.00 17.00 0.07 0.08 SFB 10000 SE 1.70 2.30 0.07 0.09 0.93 1.30 1.10 1.30 0.012 0.014 0.23 0.82 11.00 13.00 16.00 19.00 0.00 0.00 13.00 21.00 0.09 0.10
SFA 500 E 2.00 2.30 0.08 0.10 1.40 1.80 1.10 1.40 0.008 0.013 1.10 2.20 12.00 14.00 0.00 0.00 0.00 0.00 16.00 27.00 0.10 0.14 SFA 500 SE1 2.00 2.20 0.08 0.14 1.30 1.50 0.91 1.10 0.008 0.008 0.59 3.30 11.00 13.00 0.00 0.00 0.00 0.00 12.00 72.00 0.08 0.15 SFA 500 SE2 2.10 2.20 0.08 0.11 1.60 1.70 1.10 1.20 0.008 0.009 1.80 5.10 12.00 16.00 16.00 19.00 0.00 0.00 20.00 39.00 0.12 0.21 SFA 500 SW 1.90 2.20 0.08 0.10 1.50 1.70 1.00 1.20 0.009 0.012 0.37 1.40 12.00 14.00 18.00 22.00 0.00 0.00 15.00 21.00 0.09 0.11 SFA 500 NE 2.20 2.30 0.08 0.09 1.50 1.80 1.10 1.30 0.008 0.009 0.78 4.50 12.00 14.00 17.00 20.00 0.00 0.00 13.00 35.00 0.09 0.20 SFA 1000 E 1.80 2.10 0.07 0.09 1.30 1.50 0.98 1.20 0.011 0.014 0.34 1.10 13.00 14.00 16.00 17.00 0.00 0.00 16.00 21.00 0.09 0.11 SFA 1000 SE 2.10 2.40 0.09 0.28 1.30 1.60 0.95 1.10 0.008 0.009 0.55 1.40 12.00 15.00 15.00 52.00 0.69 0.69 12.00 18.00 0.08 0.09 SFA 1000 SW 2.00 2.40 0.09 0.10 1.50 1.80 0.95 1.10 0.007 0.010 0.41 1.10 12.00 16.00 17.00 34.00 0.00 0.00 11.00 21.00 0.08 0.09 SFA 1000 NW 2.10 2.50 0.09 0.13 1.70 2.20 1.10 1.40 0.010 0.012 1.60 2.80 14.00 17.00 16.00 21.00 0.00 0.00 21.00 27.00 0.12 0.17 SFA 1000 NE 2.20 2.30 0.09 0.09 1.50 2.20 1.00 1.30 0.008 0.008 1.50 2.60 13.00 15.00 20.00 20.00 0.00 0.00 18.00 24.00 0.12 0.15 SFA 2000 E 2.10 2.20 0.08 0.10 1.20 1.80 0.91 1.10 0.007 0.009 0.34 3.00 11.00 15.00 18.00 18.00 0.00 0.00 11.00 68.00 0.07 0.13 SFA 2000 SE 1.90 2.00 0.07 0.09 1.30 1.50 0.97 1.30 0.009 0.011 0.22 0.94 12.00 14.00 15.00 22.00 0.00 0.00 12.00 17.00 0.08 0.10 SFA 2000 NW 1.80 2.00 0.08 0.09 1.40 1.70 1.10 1.30 0.012 0.014 1.50 3.40 12.00 15.00 16.00 19.00 0.00 0.00 21.00 32.00 0.13 0.18 SFA 10000 SE 1.70 1.90 0.07 0.09 1.20 1.50 1.00 1.10 0.011 0.012 0.34 0.85 11.00 14.00 17.00 22.00 0.00 0.00 16.00 20.00 0.09 0.10
Ref 1 2.10 2.40 0.08 0.10 1.40 1.90 0.96 1.30 0.009 0.010 1.10 3.60 12.00 15.00 0.00 0.00 0.00 0.00 15.00 28.00 0.10 0.16 Ref 2 1.90 2.00 0.07 0.09 1.30 1.60 1.10 1.40 0.013 0.015 1.10 3.40 12.00 16.00 16.00 16.00 0.00 0.00 23.00 32.00 0.12 0.18 t0 1.80 2.30 0.08 0.10 0.11 0.15 0.50 0.70 0.011 0.015 2.50 7.20 9.90 13.00 0.00 0.00 0.00 0.00 22.00 50.00 0.08 0.21
3.2.4 Condition index CI data are reported as mean ± standard error, and the statistical comparison is reported in Figure 23. CI values were significantly different from the ones recorded in the reference group (Ref 1) in stations SFA 500/NW, SFA 1000/NE and SFA 2000/E. Stations SFA 500/NW, SFA 500/SW, SFA 1000/NE, SFA 1000/SW and SFA 2000/E were significantly different from the reference station 19. The CI comparison between the two reference stations, Ref 1 and Ref 2, showed no differences. In general, the CI values from mussels in stations SFA 500/NW, SFA 1000/ NE and 2000/E were significantly higher compared to the reference stations, Ref 1 and 2. The CI is an indicator of an animal general physiological health condition and of the energy reserve of the organism (Pampanin et al., 2005). Various studies showed that the CI is also related to food availability (Hickman et al., 1991; Helson et al., 2007), temperature (Amiard et al., 2004), salinity (Hickman et al., 1991; Marsden, 2004) and seasonal variation (Peharda et al., 2007) as well as chemical contaminants. In this study, the lowest CI values were observed in station T0 (before the offshore exposure), probably linked to the reproductive state of the organisms. According to the other biological analyses this group of organisms were in good condition (i.e. the highest survival time and LT50 in the stress on stress test, the lowest frequency in MN and reduced occurrence of histological alteration in gills). This lower value is most likely linked to the reproductive state of the organisms. These mussels did not have developed gonads, as confirmed by the histological evaluation of gonads, and their body mass was smaller
46 compared to the one in the other groups. This highlights the importance of having a battery of biomarkers, including markers at tissue levels. The integrative assessment, using the IBR index, also confirmed these findings, showing that the T0 groups had a very low IBR value (for details see section 3.2.12 Integrative assessment in mussels). According to the gonad development assessment, mussels were capable of continuing gonad maturation during the caging period and this contributed to an increase in the CI in the exposed stations at different levels. The fact that mussels were able to develop gonads during the 6 weeks of caging shows that they had sufficient energy to continue the reproductive development. It is known that mussels under stress conditions do not have the necessary metabolic energy to continue gonad development. Compared to the results obtained in the WCM 2012 at Troll C platform, the last WCM survey that used caged mussels, the CI values were in the same range.
0.3 ° ° ° * * ° ° * 0.25
0.2
0.15 *
0.1
0.05
0
Figure 23 Mean condition index values ± standard error. Statistical comparisons were done using One- Way ANOVA; differences compared to Ref 1 are reported as *, differences compared to Ref 2 are reported as °, p<0.05.
3.2.5 Stress on stress test (general health status) Survival curves of all stations were compared by using the Wilcoxon and Gehan test (Gehan, 1965) in Kaplan–Meier estimator in SPSS. Mean LT50 values for survival in air test are presented in Table 9. Time zero (T0) sampling and reference stations, Ref 1 and Ref 2, were compared with each other and their survival curves are shown in Figure 24. In addition, obtained data from survival in air test of other stations were compared with Ref 1 and 2 separately and graphs were summarized in Appendix 4 (stress on stress test, all data). Various studies have shown that mussel tolerance to anoxia stress is decreased when they are exposed to various contaminants (Eertman et al 1993; Viarengo et al., 1995; Thomas et al., 1999). Exposure to contamination and subsequent stress of tissues can affect the mussels’ adaptive mechanisms that permit extended tolerance to low oxygen conditions (Bayne et al.,
47
1989; de Zwaan et al., 1995). The majority of energy available in pollutant exposed mussels is used in detoxification processes and to compensate body growth and gonad production. According to the obtained LT50 values, tolerance of mussels from T0 is significantly higher (LT50=14) than all the other stations, including the two reference stations, Ref 1 (LT50 = 11) and Ref 2 (LT50= 8). Survival curves from Ref 1 and 2 were statistically different. This highlights the importance of having more than one reference station in this type of environmental monitoring. It is noteworthy that the lowest survival times are recorded in mussels caged at 500 m from Statfjord A and B (SFB 500/SE , SFA 500/E, SFA 500/SE 1, SFA 500/SE 2, SFA 500/SW, SFA 500/NE) and a 1000 m from Statfjord A (SFA 1000/E, SFA 100/SE). Mussels from six stations located in the safety zone (500 m radius from the discharge point) were less tolerant to the applied stress and had the shortest survival time. In addition, mussels from two out of three stations at 1000 m from Statfjord A had also significantly lower survival time. The lowest value was observed in the group of mussels from station SFA 500/E, which was located at 500 m distance from Statfjord A in the middle of the PW plume. In general, LT50 values followed the station distribution gradient, with the exception of station SFA 2000/SE.
Table 9 Means, standard error and LT50 values (days) for the stress on stress test in mussels. Mean Median LT50 Standard 95% Condifence interval LT50 Standard error Lower Upper error bound bound SFB 500 SE 9.3 0.8 7.7 10.8 9 0.9 SFB 1000 SE 11.1 0.6 9.9 12.3 11 0.6 SFB 10000 SE 13.2 0.7 11.8 14.6 13 0.7
SFA 500 E 6.8 0.7 5.5 8.1 5 1.2 SFA 500 SE 1 9.4 0.5 8.4 10.5 10 0.4 SFA 500 SE 2 8.8 0.5 7.5 10.2 8 0.4 SFA 500 SW 8.4 0.7 7.4 9.4 8 0.6 SFA 500 NE 8.9 0.5 7.9 9.8 8 0.4 SFA 1000 E 9.9 0.6 8.8 11.0 10 0.7 SFA 1000 SE 8.4 0.7 7.1 9.7 8 0.7 SFA 1000 SW 13.8 0.5 12.8 14.8 14 0.6 SFA 1000 NW 11.3 0.5 11.3 13.4 13 0.3 SFA 1000 NE 10.7 0.7 9.2 12.1 11 0.8 SFA 2000 E 12.4 0.6 11.1 13.7 12 0.6 SFA 2000 SE 9.3 0.4 8.5 10.1 10 0.3 SFA 2000 NW 11.8 0.9 10.1 13.6 11 2.0 SFA 10000 SE 12.3 0.6 11.1 13.5 12 0.3
Ref 1 11.6 0.6 10.4 12.8 11 0.6 Ref 2 8.0 0.6 6.9 9.1 8 0.7
T0 14.1 0.5 13.1 15.1 14 0.5
Recorded values were also in the same range of previous studies (Viarengo and Canesi, 1991; Eertman et al., 1993; Viarengo et al., 1995; Thomas et al., 1999; Pampanin et al., 2005). ICES assessment criteria have provisional values reported for this assay in Mytilus spp., BAC is 10 days and EAC is 5 days (Davies and Vethaak, 2012). Organisms are considered healthy if the LT50 is more than 10 days, and in this study this would include 10 groups (SFB 1000/SE, SFB
48
10000/SE, SFA 1000/SW,SFA 1000/SW, SFA 1000/NE, SFA 2000/E, SFA 2000/NW, SFA 10000/SE, Ref 1 and T0). Mussels are considered stressed, but compensating, if the value is between 5 days and 10 days, as found in stations SFB 500/SE, SFA 500/SE 1, SFA 500/SE 2, SFA 500/SW, SFA 500/NE, SFA 1000/E, SFA 1000/SE, SFA 2000/SE and Ref 2. Finally, mussels are considered severely stressed if the value is less than 5 days and none of the groups in this study had a LT50 value lower than 5.
T0 Ref 1 Ref 2
Figure 24 Survival in air curves from the stress on stress test, comparison between reference stations: Ref 1, Ref 2 and T0.
3.2.6 Lysosomal Membrane Stability The LMS assay is one of the most common biomarkers in monitoring activities and laboratory studies and has been proposed as a screening test for field surveys, in particular for PW monitoring (Sundt et al., 2011; Davies and Vethaak, 2012). Results from LMS assay are summarized in Figure 25. Data are presented in as means and standard errors. T0 samples are not reported here as a technical problem led to a poor quality of slides that could not subsequently be analysed accurately. Significant differences were found between stations SFA 500/E, SFA 500/SE 1, SFA 500/SE 2, SFA 1000/SW, SFA 1000/NE, SFA 2000/E, SFA 2000/SE and the reference stations (Ref 1 and 2). LMS data followed the PW plume gradient, with the lowest value at station SFA 500/SE 1, 500 m from the discharge point at Statfjord A and in the PW plume, and significant lower values at other 6 stations along the PW plume (SFA 500/E and SFA 500/SE 2, at 500 m from the discharge point; SFA 1000/ SW and SFA 1000/NE at 1000 m from the discharge point; SFA 2000/E and SFA 2000/SE at 2000 m from the discharge point). According to ICES assessment criteria, mussels are considered to be healthy if the lysosomal stability is > 20 min; stressed, but compensating if < 20 but > 10 min and severely stressed and probably exhibiting pathology if < 10 min (Moore et al., 2006). Based on these assessment criteria, values recorded for mussels from stations SFA 500/E, SFA 500/SE 2, SFA 1000/SW, SFA 1000/NE, SFA 2000 E and SFA 2000/SE were indicative of healthy organisms. Only organisms caged at 500 m from the discharge point at Statfjord A in the plume direction, SFA 500/SE 1, showed severe stress according to this parameter.
49
In the North Sea, close to Troll B platform, decreased values of LMS were observed in studies performed back in 2001 and 2003 (Hylland et al., 2008). Values were also generally higher than those assessed in mussels caged at Trolls C in 2012, where relatively good condition in mussels caged around the platform area was reported and no statistical difference was observed for any station compared to the reference site (Pampanin et al., 2013).
45
40
35
30
25 * * * * * * 20
15 * 10 LYSOSOYSOSOMAL MEMBRANE STABILITY (MIN) 5
0
Figure 25 Lysosomal membrane stability, measures in mussel digestive gland using the histochemical method, results reported as min of the labilisation period, mean ± s.e., statistical differences compared to both reference stations, Ref 1 and 2, are reports as *, p<0,05.
3.2.7 Acetylcholine esterase inhibition AChE activity in mussel gill tissue at the different locations around the Statfjord A and B platforms, compared to the reference and T0 groups, is displayed in Figure 26. The day 0 (T0) mussels showed a significant reduction in AChE activity compared to all field exposure groups (ANOVA, Tukey p<0.05). There were no statistically significant differences in AChE activity between the field exposed mussel groups and AChE activity appeared to be non-responsive with respect to distance from the Statfjord platforms. ICES assessment criteria have been reported for gill tissue of M. edulis from the French and Portuguese Atlantic waters (Davies and Vethaak, 2012). BAC and EAC of 30-26 nmol ATC/min/mg protein and 21-19 nmol ATC/min/mg protein have been suggested. Applying these values to our data would indicate that all of the mussel groups were either on or below the suggested EAC and therefore showing a stress response. However, these assessment criteria have been specified for Portuguese and French Atlantic waters, and their comparability to mussels from the North Sea should be considered.
50
Figure 26 Acetylcholine esterase activity in gill tissue of mussels from the different stations. Median, quartiles (box) and 10/90 percentiles (bar) n=15.
51
3.2.8 Lipid content The percentage lipid content of whole mussel homogenates from the different groups is shown in figure 27. Significantly lower lipid content was found in the day 0 (T0) mussels compared to all other groups. No significant differences in percentage lipid content were observed between field exposed mussels. The mussel lipid content in field deployed mussels ranged between 0.5 and 1.1%. These lipid values are slightly lower than measured in previous WCM programmes when caged mussels were used (WCM2011, WCM2009). In these earlier studies the mussel lipid content was typically between 1 and 2%. The reproductive stage of the mussels has probably the greatest influence on lipid content and likely to be the reason for the differences reported. Interestingly, the low lipid content found in T0 group corresponds with the low condition index and the reproductive status. As discussed earlier in this report, this is not considered a sign of poor condition of the population, since other biomarkers showed that the organisms were in good health (stress on stress data and MN frequency), in addition to the IBR index.
Figure 27 The percentage lipid content in whole mussel homogenates from the different groups. Median, quartiles (box), 10/90 percentiles (bar) n= 5. * ANOVA, Tukey p<0.05.
3.2.9 Gonad histological evaluation (spawning) Evaluation of the gonad development in mussels is a very important parameter when working in field conditions. It is a good support parameter to interpret other biological responses, as for example the CI. However, it is known that some biological effect measurements are influenced by seasonal variation of abiotic and biotic factors, including the gonad development and the spawning status of the organisms (Beyer et al., 2017). For example, Schmidt et al. (2013) examined seasonal variations for various biomarkers (i.e. glutathione S-transferase, vitelline- like proteins, lipid peroxidation and DNA damage) in blue mussels collected in a pristine area
52 and showed that most parameters presented a seasonal variation linked to the organism reproductive cycle. Results regarding the gonad development in mussels are reported in Table 10. In male mussels, 50% of individuals in all stations had gonad status showing evidence of spawning. In female mussels, the gonad was fully mature in individuals after retrieval at all stations. At time T0 (mussels from the farm) gonads were in early gametogenesis (primary oocytes), with evidence of spawning. Compared to pre-exposure (T0), mussels caged offshore for 6 weeks were capable of developing gonads, which is a sign of general good environmental conditions, e.g. food availability, and good health of the organisms. Regarding the spawning status, even if most of the female mussels were fully mature and ready to spawn (gonadal status stage 4 and 5), and the males had at least 50% mature gonads (gonadal status stage 3), only a few male organisms showed evidence of spawning activity (spawning stage 3). There were no mussels recognised as actively spawning (spawning stage 4). The WCM survey has been historically performed in the May-June period to increase the chance of good sea state for carrying out ship-board activities. It has been observed that compared to previous years, mussel populations are spawning earlier in Norway, starting to develop gonads already in April, and being ready to spawn in May. Due to the possibility of reproductive status of organisms influencing biological effects responses, the possibility of commencing the WCM survey outside the spawning season is recommended when using caged mussels.
Table 10 Gonadal and spawning status of mussels. Male Female Station Gonadal Spawning Total n of Gonadal Spawning Total n of status individuals status individuals SFB 500 SE 3 3 9 4 2 6 SFB 1000 SE 3 3 7 5 2 8 SFB 10000 SE 3 3 10 4 3 5
SFA 500 E 4 3 9 5 1 6 SFA 500 SE 1 3 2 5 5 1 10 SFA 500 SE 2 3 3 8 4 2 7 SFA 500 SW 3 2 4 4 2 11 SFA 500 NE 3 3 9 4 1 6 SFA 1000 E 3 3 8 4 1 7 SFA 1000 SE 3 2 8 4 1 7 SFA 1000 SW 3 2 9 4 3 6 SFA 1000 3 3 12 5 1 3 NW SFA 1000 NE 3 3 11 5 1 4 SFA 2000 E 3 2 5 4 1 10 SFA 2000 SE 3 3 9 4 3 6 SFA 2000 3 3 6 4 3 9 NW SFA 10000 3 3 11 4 2 4 SE
Ref 1 3 3 8 4 2 7 Ref 2 4 3 9 5 3 6
T0 3 2 8 2 1 6
53
3.2.10 Micronucleus assay (DNA damage) The MN frequency in haemocytes of bivalves is widely used as biomarker of genotoxicity (Fernandez et al., 2011; Bolognesi and Hayashi, 2011; Sundt et al., 2011, Bolognesi and Fenech, 2012; Davies and Vethaak, 2012). The presence of MN is characterized by having a chromosome or a fragment of it not incorporated into the cell nucleus. A MN can be formed during mitosis telophase due to a leggy acentric chromosome, existence of a chromosome fragment, and/or exclusion of a whole chromosome. Enclosing these chromosomes by nuclear membranes creates a MN. MNs as small secondary structures of chromatin are located in the cytoplasm and have no detectable link to the cell nucleus (Heddle et al., 1983; Zoll-Moreux and Ferrier, 1999). Examples of cells during evaluation of the MN assay are shown in Figures 28-30. Figure 28 shows two examples of observed normal cells. Several observations of MN in haemolymph cells are shown in Figure 29. Besides MN, other types of structural abnormalities were also observed such as binuclear, nuclear buds and binucleated cells with a nucleoplasmic bridge between nuclei (illustrated in Figure 30). A new method compared to previous WCM surveys was used, based on Venier et al. (1997), and slides were of good quality and allowed a cost- efficient quantification of the MN frequency. In previous years, the quality of the slides has been a challenge, and sometimes it was not possible to assess this parameter due to the poor quality of the material, such as in the WCM 2012 (Pampanin et al., 2013). The highest mean number of MN in 1000 counted cells were recorded from station SFB 500/SE (10.2 ‰) (Figure 31). Organisms from stations SFB 500/E and SFA 500/E (500 m distance from the discharge point at Statfjord B and A) exhibit significantly higher values compared to the reference sites (Ref 1 and Ref 2, 2.1 ‰ and 2.2 ‰ respectively) and T0 (2 ‰). Mussels from stations SFA 1000/NW had a significantly higher frequency of MN compared to the reference station Ref 2. According to the ICES assessment criteria, a MN frequency below 4.1 ‰ is considered a background response (BR), while values above this are considered an elevated response (ER) in mussels caged for 6 weeks in the North Sea. The two reference stations (Ref 1 and 2), T0 and SFA 500/SE 1, SFA 500/SE 2 and SFA 500/NE had values below the BR. In 4 stations (SFB 500/SE, SFA 500/E, SFA 1000 NW and SFA 2000 NW), MN frequency values were above the BR, mean value ranging from 5.6 ‰ to 10.2 ‰. This is clearly a sign of the presence of contaminants with genotoxicity potential and further investigation should be done to assess if the source of contamination is PW or other oil and gas related activities (e.g. old cutting piles) or alternatively a contribution from other discharges outside the Norwegian sector.
Figure 28 Examples of normal cells in the mussel haemolymph.
54
Figure 29 Micronuclei observed in the micronucleus assay.
Figure 30 Nuclear abnormalities observed in haemolymph of mussels.
14 * 13
12
11
10
9 * 8
7 ° 6
5 Number of MN in 1000 cells in MN of Number
4
3
2
1
0 T0 SE Ref 1 Ref 2 E NE SW SE1 SE2 NW NW SFB 500 SFB SFA 500 SFA 500 SFA 500 SFA 500 SFA SFA 500 SFA SFA 1000 SFA 2000 SFA
Figure 31 Micronuclei (MN) frequency in 1000 cell (‰) (n=20). Statistical comparisons were performed using One-Way ANOVA and Independent T-Test, * p<0.05 compared to Ref 1 and 2; ° p<0.05 compared to Ref 2.
55
3.3 Chemical and biological analyses in wild caught fish 3.3.1 PAH body burden PAH-NPD concentrations were measured in fillet samples of selected fish species collected from within the 500 m safety zone of the Statfjord A platform (Figure 37). PAH concentrations were either undetected or measured just above their limits of quantification. Alkylated naphthalenes and phenanthrenes were the main PAH detected.
Figure 37 PAH concentrations in individual fish fillets in selected species from the Statfjord A platform. Data includes only the measured values above limits of quantification (LOQ, see first column in figure).
In the regional study, analyses of PAH compounds in fillet of cod and haddock were performed to document that oil and gas activities in the North and Norwegian Sea do not affect sea food quality. Analyses of PAH including alkylated homologs were carried out using GC-MS. The compounds included in the analysis are shown in Table 13. Fifty different PAH compounds have been analysed in cod fillet from Tampen and Egersund Bank. All were below their limits of quantification, except 1,4 dimethylnaphthalene. Regarding the cod species from Tampen, 4 of 6 cod had levels of 0.39 ± 0.07 ng/g ww. About cod from the Egersund Bank, 3 of 15 had a mean level of 0.27 ± 0.08 ng/g ww (Table 13). Of the 25 haddock caught at Tampen, 4 had a mean level of 0.27 ± 0.05 ng/g ww 1,4 dimethylnaphthalene. Concerning haddock at the Egersund Bank, 7 of 25 fish had mean levels of 0.29 ± 0.06 ng/g ww, 1,4 dimethylnaphthalene was undetected in haddock from the Central North Sea (Table 14, only value above LOQ are reported, all compounds listes in Table 13 were analysed in haddock as well). Only 1 out of 25 haddock from Tampen had levels of fluoranthene, pyrene and benz(a)anthracene above LOQ. These results do not indicate that PAH content in fish fillet are different in fish from Tampen compared with the reference region Egersund Bank. The observed quantifiable levels are very close to LOQ and no contamination effecting sea food quality could be observed. Levels of PAH compounds below LOQ are as
56 expected, as cod, haddock and saithe are lean fish species, and due to documentation from earlier measurements of PAH in muscle of cod and haddock below LOQ in Norwegian Seas (Grøsvik et al., 2012; Grøsvik et al., 2015).
Table 13 Level of PAH compounds in cod fillet, LOQ = limit of quantification. Compound Tampen (n=6) Egersund Bank (n=15) LOQ= 0,02 ng/g ww LOQ= 0,02 ng/g ww Naphthalene < LOQ < LOQ Benzothiophene < LOQ < LOQ 2-Methylnaphthalene < LOQ < LOQ 1-methylnaphtalene < LOQ < LOQ Biphenyl < LOQ < LOQ 2,6 - Dimethylnaphthalene < LOQ < LOQ 1,3-Dimethylnaphthalene < LOQ < LOQ 2,3 Dimethylnaphthalene < LOQ < LOQ n= 4: 0.39 ± 0.07 n= 12: < LOQ 1,4 Dimethylnaphthalene n= 2: < LOQ n= 3: 0,27 ± 0,08 Acenaphthylene < LOQ < LOQ Acenaphthene < LOQ < LOQ Dibenzofuran < LOQ < LOQ 1,3,7-Trimethylnaphthalene < LOQ < LOQ 2,3,5-Trimethylnaphtalene < LOQ < LOQ 1,2,3-Trimethylnaphthalene < LOQ < LOQ 1,4,6,7-Tetramethylnaphthalene < LOQ < LOQ 1,2,5,6-Tetramethylnaphthalene < LOQ < LOQ Fluorene < LOQ < LOQ 1-Methylfluorene < LOQ < LOQ 9-Ethylfluorene < LOQ < LOQ Dibenzothiofen < LOQ < LOQ Phenanthrene < LOQ < LOQ Anthracene < LOQ < LOQ 4-methyldibenzotiofen < LOQ < LOQ 3-Methylphenanthrene < LOQ < LOQ 2-Methylphenanthrene < LOQ < LOQ 9-Methylphenanthrene < LOQ < LOQ 1-Methylphenanthrene < LOQ < LOQ 4-ethyldibenzotiofen < LOQ < LOQ 3,6-Dimethylphenantrene < LOQ < LOQ 4-propyldibenzotiofen < LOQ < LOQ 1,7-Dimethylphenantrene < LOQ < LOQ 1,2-Dimethylphenanthrene < LOQ < LOQ 2,6,9-Triimethylphenantrene < LOQ < LOQ 1,2,6-Trimethylphenanthrene < LOQ < LOQ 1,2,7-Trimethylphenantrene < LOQ < LOQ 1,2,6,9-Tetramethylphenantrene < LOQ < LOQ
57
Fluoranthene < LOQ < LOQ Pyrene < LOQ < LOQ Benz(a)anthracene < LOQ < LOQ Chrysene < LOQ < LOQ Benzo(b)fluoranthene < LOQ < LOQ Benzo(k)fluoranthene < LOQ < LOQ Benzo(j)fluoranthene < LOQ < LOQ Benzo(e)pyrene < LOQ < LOQ Benzo(a)pyrene < LOQ < LOQ Perylene < LOQ < LOQ Indeno(1,2,3-cd)pyrene < LOQ < LOQ Dibenz(a,h)anthracene < LOQ < LOQ Benzo(ghi)perylene < LOQ < LOQ
Table 14 Level of PAH compounds in haddock fillet, LOQ = limit of quantification. Compound Tampen (n=25) Central North Sea (n=25) Egersund Bank (n=25) LOQ= 0,02 LOQ= 0,02 LOQ= 0,02 n= 21: < LOQ n= 18: < LOQ 1,4 Dimethylnaphthalene n= 4: 0,27 ± 0,05 < LOQ n= 7: 0,29 ± 0,06 n= 24: < LOQ Fluoranthene n= 1: 0,45 < LOQ < LOQ n= 24: < LOQ Pyrene n= 1: 0,29 < LOQ < LOQ n= 24: < LOQ Benz(a)anthracene n= 1: 0,25 < LOQ < LOQ
3.3.2. Perfluorinated Compounds Results of the perfluorinated compound analysis in fish blood are reported in Appendix 2, as this analysis in not included within the guidelines as a mandatory parameter. Briefly, of the 23 PFAS analysed in fish blood, only three (PFOSA, PFOS and PFUdA) were measured above detection limits. However, these three PFAS were not measured above detection limits in all fish samples. PFOSA was measured in 11 of the 29 samples ranging between 6 and 15 ng/ml. PFUdA was measured in 6 samples ranging between 6 and 12 ng/ml. Whilst PFOS was measured in 4 samples ranging between 6 and 10 ng/ml. There was no statistical difference between blood samples from Statfjord, Egersund Bank or Tampen for PFOSA, PFOS or PFUdA.
3.3.3 Radioactive compound concentrations IFE (Institutt for energiteknikk) has provided the analysis of a subset of sample of fish from the Statfjord area. The following samples were analysed: 8 ling (sample n 37, 41, 42, 43, 45, 46, 70 and 72), 1 cod (sample 44) and 1 tusk (sample 68) (Figure 38). It is considered unlikely that naturally occurring radioactivity associated with PW discharges represents a significant health risk to marine life or humans, as described in the Water Column monitoring 2010 report (Sundt et al., 2012b). For example, oxidative stress responses were indicated in rag worms (Hediste diversicolor) exposed to levels of 226Ra several orders of magnitude higher than levels measured in organisms caged close to a PW discharge. Radioactive compounds were measures in a few individuals to confirm previous findings. Levels were similar to the those previously measured in the North Sea, around the Ekofisk platform (Sundt et al., 2012b).
58
Fish species Fish n 226Ra 228Ra 228Th 210Pb 40K Ling 37 ≤ 15 ≤ 14 ≤ 7 ≤ 50 450 ± 60 Ling 41 ≤ 8 ≤ 7 ≤ 2,7 ≤ 13 510 ± 40 Ling 42 ≤ 11 ≤ 18 ≤ 5 ≤ 30 530 ± 60 Ling 43 ≤ 6 ≤ 8 ≤ 2,7 ≤ 19 560 ± 50 Ling 45 ≤ 6 ≤ 14 ≤ 4 ≤ 15 510 ± 50 Ling 46 ≤ 6 ≤ 6 ≤ 2,9 ≤ 14 560 ± 50 Ling 70 ≤ 4 ≤ 5 ≤ 2,4 ≤ 14 510 ± 40 Ling 72 ≤ 13 ≤ 40 ≤ 11 ≤ 29 580 ± 100 Cod 44 ≤ 9 ≤ 17 ≤ 5 ≤ 23 510 ± 60 Tusk 68 ≤ 14 ≤ 25 ≤ 10 ≤ 40 440 ± 80 Figure 38 Radioactive compound analysis performed by IFE (Institutt for energiteknikk), results are reported as Bq/kg of dry weight.
3.3.4 Condition Indices and age determination Fish from the Statfjord A platform were caught by rod within the security zone (< 500 m) of the platform from M/V Olympic Delta. Harsh weather conditions made weighing of liver and gonads difficult. Fish from the Egersund Bank, Central North Sea and the Tampen region were fished by bottom trawling using the R/V Johan Hjort as normally used during the ICES international bottom trawl (IBTS) survey in the North Sea. Biological data, age, and condition indices of the 5 fish species included (cod, haddock, saithe, ling and whiting) are listed in Tables 15-19. Biological parameters are included for the purpose of support, in assessing the overall biological condition and status of the sampled fish. Cod and whiting from Statfjord A were older compared with cod and whiting from the Egersunds Bank, while saithe from Statfjord A were younger than saithe from the Egersund Bank (Table 16). Only 2-4 ling were taken from the regional sampling undertaken in May 2017, compared with 25 from the Statfjord A survey in April 2017. Differences in GSI females for ling is explained by differences in spawning status.
Table 15 Biological data of cod (Gadus morhua) given as mean ± sd. Values in bold are significantly different from cod caught at the Egersund Bank (reference area), p< 0.05, nd = not determined. Region or oil platform Egersund Bank Tampen Statfjord A
Total number 15 8 10
Females/males 8/7 8/0 4/6
Length (cm) 52 ± 19 72 ± 15 86 ± 14
Weight (g) 1750 ± 2210 4182 ± 1299 7256 ± 4106
Liver weight (g) 57 ± 102 157 ± 150 725 ± 219
Age (year) 2.7 ± 1.3 4.0 ± 1.3 6.7 ± 1.7
LSI (%) 1.9 ± 1.2 3.1 ± 1.7 1.0 ± 2.2
Fulton’s condition index 0.85 ± 0.07 0.93 ± 0.22 1.02 ± 0.07
GSI females 0.34 ± 0.06 0.33 ± 0.16 1.00 ± 0.74
GSI males 0.12 ± 0.03 nd 0*
59
Table 16 Biological data of haddock (Melanogrammus aeglefinus), given as mean ± sd. Values in bold are significantly different from haddock caught at the Egersund Bank (reference area), p< 0.05. Region or oil platform Egersund Bank Central North Sea Tampen
Total number 25 25 25 Females/males 15/10 22/3 10/15
Length (cm) 40 ± 6 38 ± 6 48 ± 9
Weight (g) 579 ± 262 541 ± 262 1205 ± 929
Liver weight (g) 14 ± 4 22 ± 18 23 ± 19
Age (year) 3.5 ± 1.4 2.5 ± 0.6 4.9 ± 2.6
LSI (%) 2.7 ± 0.9 3.8 ± 1.2 1.9 ± 0.9
Fulton’s condition index 0.85 ± 0.10 0.94 ± 0.07 0.88 ± 0.11
GSI females 0.93±0.53 0.52 ± 0.13 0.69 ± 0.23
GSI males 0.46±0.19 0.26 ± 0.0.13 0.61 ± 0.56
Table 17 Biological data of saithe (Pollachius virens), given as mean ± sd. Values in bold are significantly different from saithe caught at the Egersund Bank (reference area), p< 0.05, nd = not determined. Region or oil platform Egersund Bank Central North Sea Tampen Statfjord A Total number 25 25 25 25
Females/males 10/15 11/14 8/17 13/12
Length (cm) 61 ± 12 70 ± 12 60 ± 13 49 ± 13
Weight (g) 2208 ± 1178 2955 ± 1388 1973 ± 1758 1220 ± 1384
Liver weight (g) 78 ± 59 146 ± 128 71 ± 89 nd*
Age (year) 6.0 ± 1.4 6.7 ± 1.5 5.8 ± 2.0 4.7 ± 1.6
LSI (%) 3.2 ± 1.3 4.3 ± 2.1 3.3 ± 3.4 nd
Fulton’s condition index 0.87 ± 0.20 0.81 ± 0.07 0.79 ± 0.07 0.85 ± 0.16
GSI females 0.62 ± 0.29 0.62 ± 0.20 0.66 ± 0.44 0.54 ± 0.28
GSI males 0.15 ± 0.07 0.16 ± 0.10 0.14 ± 0.15 0.9*
• Liver weight of saithe from Statfjord A is missing due to harsh weather condition.
60
Table 18 Biological data of ling (Molva molva), given as mean ± sd. Values in bold are significantly different from ling caught at the Egersund Bank (reference area), p< 0.05, nd = not determined. Region or oil platform Egersund Central North Sea Tampen Statfjord A Bank Total number 2 2 4 25 Females/males 2/0 2/0 4/0 20/5
Length (cm) 77 ± 7 64 ± 13 79 ± 8 94 ± 17
Weight (g) 2488 ± 449 1402 ± 831 2425 ± 738 5366 ± 3449
Liver weight (g) 92 ± 49 44 ± 42 128 ± 82 482 ± 420
Age (year) 5.5 ± 0.7 5.0 ± 0.0 5.5 ± 0.6 7.7 ± 2.5
LSI (%) 3.6 ± 1.3 2.7 ± 1.4 4.9 ± 2.2 6.6 ± 3.4
Fulton’s condition index 0.55 ± 0.04 0.50 ± 0.01 0.49 ± 0.03 0.58 ± 0.09
GSI females 0.62 ± 0.09 0.55 ± 0.08 0.86 ± 0.37 7.31 ± 7.10
GSI males nd nd nd nd
Table 19 Biological data of whiting (Merlangius merlangus), given as mean ± sd. Values in bold are significantly different from ling caught at the Egersund Bank (reference area), p< 0.05. Region or oil platform Egersund Bank Central North Sea Tampen Statfjord A Total number 14 25 25 25 Females/males 12/2 16/9 12/13 16/9
Length (cm) 37 ± 4 36 ± 4 41 ± 4 45 ± 6
Weight (g) 358 ± 118 352 ± 120 491 ± 136 707 ± 339
Liver weight (g) 12 ± 8 10 ± 5 15 ± 7 15 ± 7
Age (year) 4.1 ± 1.0 4.0 ± 1.0 4.8 ± 1.2 6.1 ± 1.9
LSI (%) 3.3 ± 1.6 2.8 ± 0.9 2,9 ± 1.0 2.8 ± 0.7
Fulton’s condition index 0.71 ± 0.07 0.71 ± 0.07 0.71 ± 0.07 0.73 ± 0.08
GSI females 4.6 ± 2.3 4.8 ± 2.3 3.3 ± 1.1 3.01 ± 2.24
GSI males 0.8 ± 0.1 0.6 ± 0.3 0.9 ± 0.4 1.70 ± 1.57
61
3.3.5 PAH metabolites Results of PAH metabolites analysed by the FF screening method are reported in Figure 39. Levels were in the same range as those reported in previous WCM surveys (Brooks et al., 2013 and 2014). The only statistical differences were found in cod between individuals collected around the Statfjord platform and the reference area (Egersund Bank), for the 2,3-ring PAHs and the 5-ring PAHs. In addition, whiting samples in the Tampen area had a significantly higher levels of 4-ring PAHs compared to the reference area. A subset of samples was also analysed for the quantification of some PAH metabolites using the GC-MS method. This method allowed the quantification of a limited number of PAHs, those with an available standard (Sundt et al., 2012a). Values were in the same range of previous surveys (Hylland et al., 2008; Sundt et al., 2012a, Brooks et al., 2013 and 2014). Only a few individuals had values above the quantification limit, and only these results are reported in Figure 40. A statistical evaluation was done considering the PAH metabolites as a total sum, using the Mann-Whitney U test. PAH metabolites were significantly higher only in whiting and haddock from the Tampen area compared to the reference area. In general, PAH metabolite values in fish collected in this study, measured by the FF method, were generally low, with no significant differences between areas, except in few cases. These results were confirmed by the GC-MS analyses, where most of the reported values were below the limit of quantification.
62
Figure 39 PAH metabolites analysed by the fixed florescence wavelength method. Median, quartiles (box) and 25/75 percentiles (bar), the number of analysed samples per station is reported in parenthesis in the graphs. Data reported as pyrene fluorescence equivalents (PFE) µg/ml, Egersund Bak = Reference area. 63
64
Figure 40 PAH metabolites in bile (ng/g bile). All the data above the quantification limit are reported in the graph, one species per graph, Reference = Egersund Bank. 65
3.3.6 Alkylphenol metabolites The high relevance of fish bile metabolites in relation to PW exposure has been observed in previous environmental monitoring campaigns in the North Sea, as well as in a pilot laboratory study (Sundt et al., 2012a). However, AP metabolites have not been analysed in the latest WCM surveys. A subset of samples was processed and data for AP metabolites in fish collected at Statfjord are reported in Figure 41. Most of the values were below the quantification limit and in general they were lower than previous findings (Sundt et al., 2012a).
Cod (n = 7) Ling (n = 7) Saithe (n = Whiting (n = 12) 5) mean s.d. mean s.d. mean s.d. mean s.d. 2-methylphenol 20 0 20 0 20 0 20 0 3-methylphenol 20 0 20 0 20 0 20 0 4-methylphenol 969 1530 395 644 311 368 37 34 3,5-dimethylphenol 20 0 20 0 20 0 20 0 2,4-dimethylphenol 20 0 20 0 20 0 20 0 4-ethylphenol 20 0 20 0 20 0 20 0 4-n-propylphenol 20 0 20 0 33 45 20 0 2,4,6-trimethylphenol 20 0 20 0 34 44 20 0 4-tert-butylphenol 20 0 20 0 20 0 20 0 4-isopropyl-3- 119 100 20 0 31 21 20 0 methylphenol 2-tert-butyl-4- 141 126 100 179 31 33 20 0 methylphenol 4-tert-butyl-2- 153 115 98 173 61 93 20 0 methylphenol 2,5-diisopropylphenol 20 0 20 0 20 0 20 0 4-n-butylphenol 23 7 33 16 46 34 28 12 4-n-pentylphenol 20 0 20 0 20 0 20 0 4-n-hexylphenol 20 0 20 0 32 41 20 0 4-tert-octylphenol 429 695 155 134 99 127 185 257 4,6-di-tert-butyl-2- 20 0 20 0 20 0 20 0 methylphenol 4-n-heptylphenol 20 0 20 0 20 0 20 0 4-n-octylphenol 23 4 28 11 25 7 27 11 4-n-nonylphenol 24 5 45 30 37 15 38 20 Figure 41 Alkylphenol metabolites in bile of fish collected in the Statfjord area, expressed in ng/g, mean values and standard deviations are reported.
3.3.7 Tissue changes in liver Histopathological evaluation of the fish liver data is reported as mean score, number of analysed individuals and cumulative score, per each station in each fish species (Figures 42-44). All the selected histopathological alterations were found in all samples with different levels of distribution. Raw data and statistics are reported in Appendix 5 (Histopathology of fish liver). A statistical evaluation was made comparing the Tampen, Staftjord and Central North Sea areas with the reference one (Egersund Bank). Differences are indicated as values in red in Figures 42-44.
66
Regarding the Tampen area, cod and haddock had a significantly higher score in steatosis, haddock and saithe had significantly higher score in circulatory disturbance, and whiting had a significantly higher score for the presence of parasites. Regarding other pathological changes, cod had a significantly lower score, and haddock had a higher score, when compared to the reference samples. Regarding Statfjord area, all species, except ling, had significantly higher scores for steatosis, and saithe had a significantly higher score for the presence of parasites. In samples from the Central North Sea, only two differences were found, haddock had a significantly higher score in steatosis and whiting had a significantly higher score for parasite presence. Steatosis lesions showed the major significant difference between areas. These parameters are suggested to be part of the normal cyclic (and non-pathological) status of liver. Similar occurrence of steatosis lesions was found in previous WCM surveys (Brooks et al., 2013 and 2014). There were differences in the presence of circulatory disturbance in two species. And in general, other minor pathologies were found only in haddock. As regards parasite presence, there was no indication of liver parasitic invasion. Furthermore, this parameter is less likely to be influenced by exposure to contaminant stress. All of the analysed species from all areas, including the reference, showed various occurrences of different histological abnormalities and pathologies. However, fish from Tampen and Stafjord presented significantly higher scores in more than one species, in steatosis and circulatory disturbance. Fish from the Central North Sea area appeared to have similar liver condition as the reference area (Egersund Bank). Previous studies conducted at Njord A, reported significant increased prevalence of melanomacrophage aggregates in fish suggesting a response to contaminant exposure. No difference was found in this study regarding this parameter. ICES assessment for non-specific and contaminant-specific liver histopathology is under development, and suggestions have been made (Davies and Vethaak, 2012). The present data together with the recently acquired results from previous WCM surveys could contribute to the elaboration of assessment criteria for fish species commonly caught in the North Sea.
67
Steatosis Egersund Bank Tampen number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 1.6 24 8 2.75 * 22 Haddock 25 2.36 59 25 1.56 * 39 Ling 2 3 6 4 3 12 Saithe 25 2.84 71 25 2.68 67 Whiting 14 2.78 39 25 2.96 74 Circulatory disturbance Egersund Bank Tampen number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.06 1 8 0.125 1 Haddock 25 0.08 2 25 0.32 * 8 Ling 2 0 0 4 0.24 1 Saithe 25 0.16 4 25 1 * 25 Whiting 14 0 0 25 0.12 3
melanomacrophage aggregates Egersund Bank Tampen number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.06 1 8 0 0 Haddock 25 0.88 22 25 1.04 26 Ling 2 0 0 4 0 0 Saithe 25 0.68 17 25 0.36 9 Whiting 14 0.357 5 25 0.4 10
inflammatory changes Egersund Bank Tampen number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.267 4 8 0.125 1 Haddock 25 0.2 5 25 0.12 3 Ling 2 0 0 4 0.25 1 Saithe 25 0.4 10 25 0.24 6 Whiting 14 0 0 25 0.2 5
other pathooplogical changes Egersund Bank Tampen number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.6 9 8 0 * 0 Haddock 25 0.24 6 25 0.52 * 13 Ling 2 0 0 4 0 0 Saithe 25 0.2 5 25 0.08 2 Whiting 14 0.071 1 25 0.16 4
parassites Egersund Bank Tampen number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.467 7 8 0.5 4 Haddock 25 0.16 4 25 0.16 4 Ling 2 0.5 1 4 0.25 1 Saithe 25 0.375 9 25 0.56 14 Whiting 14 0 0 25 0.4 * 10
Figure 42 Fish liver histopathology results, comparison between Tampen area and the reference area (Egersund Bank), statistically significant values compared to the reference area are reported in red, *p<0.05.
68
Figure 43 Fish liver histopathology results, comparison between Statfjord area and the reference area (Egersund Bank), statistically significant values compared to the reference area are reported in red, *p<0.05.
69
Steatosis Egersund Bank Central North Sea number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 1.6 24 _ _ _ Haddock 25 2.36 59 25 2.96 * 74 Ling 2 3 6 2 2.5 5 Saithe 25 2.84 71 25 3 75 Whiting 14 2.78 39 25 2.76 69
Circulatory disturbance Egersund Bank Central North Sea number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.06 1 _ _ _ Haddock 25 0.08 2 25 0 0 Ling 2 0 0 2 0.5 1 Saithe 25 0.16 4 25 0.32 8 Whiting 14 0 0 25 0.12 3
melanomacrophage aggregates Egersund Bank Central North Sea number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.06 1 _ _ _ Haddock 25 0.88 22 25 0.16 4 Ling 2 0 0 2 0.5 1 Saithe 25 0.68 17 25 0.48 12 Whiting 14 0.357 5 25 0.6 15
inflammatory changes Egersund Bank Central North Sea number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.267 4 _ _ _ Haddock 25 0.2 5 25 0.12 3 Ling 2 0 0 2 0 0 Saithe 25 0.4 10 25 0.52 13 Whiting 14 0 0 25 0.08 2 other pathooplogical changes Egersund Bank Central North Sea number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.6 9 _ _ _ Haddock 25 0.24 6 25 0.12 3 Ling 2 0 0 2 0.5 1 Saithe 25 0.2 5 25 0.12 3 Whiting 14 0.071 1 25 0.16 4
parassites Egersund Bank Central North Sea number of individuals mean score Cumulative score number of individuals mean score Cumulative score Cod 15 0.467 7 _ _ _ Haddock 25 0.16 4 25 0.12 3 Ling 2 0.5 1 2 0.5 1 Saithe 25 0.375 9 25 0.6 15 Whiting 14 0 0 25 0.48 * 12
Figure 44 Fish liver histopathology results, comparison between Central North Sea area and the reference area (Egersund Bank), statistically significant values compared to the reference area are reported in red, *p<0.05.
70
3.3.8 DNA damage in liver and gut, DNA adducts Liver DNA adduct assessments were performed by the ADnTox laboratory in France. It was noted by the analysing laboratory that liver samples from Statfjord A were of poor quality and that DNA extraction was not satisfactory. Since the amount of extracted DNA was lower than expected, samples were run in single replicate. For details, see Appendix 6 (DNA adduct report – liver). DNA adduct results are reported in Figure 45. Results were also reported as qualitative data (Appendix 6). When assessing the results from DNA adducts, they were evaluated to be of sufficient quality to be included in the report. The results of DNA adduct levels from Statfjord A were consistent with differences in exposure levels and other biomarker responses between the different regions/stations. An overview of DNA adducts in fish liver is presented in Figure 45. DNA adduct levels are expressed as relative adduct levels (RAL) x 10-8. When comparing levels with EAC levels used by ICES (ICES, 2012), DNA adduct levels have been multiplied by 10 to be expressed as nmol adduct per mol normal nucleotides.
Figure 45 DNA adduct quantitative data, expressed as RAL x 10-8: DNA adduct concentration for 108 normal nucleotides (RAL= Relative Adduct Level), 1 nmol adducts / mol DNA = 0.1 adducts x 10-8 normal nucleotides.
Statistical analyses for all species showed that DNA adduct levels per sample were different from one site to another (p = 8.712 x 10-12). Two-by-two comparison of all species showed that levels at the Egersund Bank were different from levels at Tampen (p = 1.01 x 10-9). Levels at the Egersund Bank were different from Statfjord A (p = 2.54 x 10-7), while adduct levels were not significantly different between Statfjord A and Tampen (p = 0.50). DNA adduct levels at the Egersund Bank were not significantly different between species (p = 0.13). DNA adduct levels were significantly different between species at Tampen, with higher levels for haddock and saithe (p = 3.19 x 10-3). For Statfjord A, DNA adduct levels were
71 significantly different between species, with higher levels for haddock, whiting and ling (p=0.0351). It is important to note that these samples had a high level of DNA adducts, even if they had suffered a potential defrosting event. Statistical analyses of data from haddock showed that DNA adduct levels in liver samples were significantly different between sites, resulting in the following order: Tampen > Statfjord A > Egersund Bank, p = 0.0029. For Atlantic cod, adduct levels in liver samples were significantly different between sites, resulting in the following order: Statfjord A > Tampen > Egersund bank, p = 0.0005. For whiting, DNA adduct levels in liver samples were significantly different between sites, resulting in the following order: Statfjord A > Tampen > Egersund bank, p = 0.0004. DNA adduct levels in saithe liver samples were significantly different between sites, resulting in the following order: Tampen > Statfjord A > Egersund bank, p = 0.0156. This was also the case for ling liver samples where DNA adduct levels were significantly different between sites, resulting in the following order: Statfjord A > Tampen > Egersund bank, p = 0.0316.
Intestine In addition to measurements in liver, fish intestine samples were also analysed for DNA adducts. Comparison of DNA adducts in liver with levels in intestine could provide information on exposure routes and whether PAH metabolism and adduct formation can have different patterns in these two tissues. Seven individuals of cod, saithe, and whiting from the Egersund Bank and the Statfjord A field were analysed to compare fish living close to the oil platform with the reference area. In addition, 7 ling from Statfjord A were included. In general, intestine from the four demersal fish of the cod family from the Statfjord A platform had higher levels of DNA adducts compared with fish from the Egersund Bank (reference area) (Figure 46). However, only DNA adduct levels in cod intestine were significantly higher at the Statfjord A platform compared with the Egersund Bank (n=7), for details see Appendix 7 (DNA adduct report – intestine).
Figure 46 Comparison of DNA adduct levels in intestine of cod, saithe, whiting and ling from Statfjord A and the Egersund Bank as box plot, n = 7 individuals per group.
72
Fish may be exposed to compounds leading to DNA adduct formation during their lifetime and age can be an explanatory factor. The relationship of DNA adducts with age is shown in Figure 47. Fish from the Egersund Bank in the adduct analysis had an average age of 4.3 ± 1.8 years, fish from Tampen 5.1 ± 2.0 years and fish from Statfjord were 6.2 ± 2.3 years. This shows that part of the difference from Egersund Bank vs Tampen and Statfjord A could be explained by the age of the analysed fish.
A B Figure 47 Plot of fish age against DNA adduct levels in fish intestine (A) and fish liver (B). Adduct data from cod, ling, saithe, and whiting are included in A and adducts of cod, haddock, ling, saithe, and whiting in B, n = 41 for intestine and n = 123 for liver.
Figure 48. Linear fit of DNA adducts in liver by intestine from analysed fish. Spearmans ρ = 0.36, n = 21.
Correlation analyses of fish where DNA adducts have been analysed from both liver and intestine show a positive relationship with Spearmans ρ of 0.36 (Figure 48). The DNA adducts measurements in liver from Statfjord were high although these samples had suffered the quality issue during transport. Comparisons of adduct levels in intestine versus liver show differences between species and regions (Figures 49-53). The differences observed may be affected by PAH exposure, age, choice of prey, and living pattern, e.g. contact with bottom sediments. These differences could also be affected by the relatively low number of individuals included in these comparisons and the migration pattern of the adducts. Mean adduct levels in intestine of fish from Statfjord A (23.5 ± 15.5 nmol adducts per mol DNA) were significantly higher compared with mean levels of intestine of fish from Egersund Bank (6.3 ± 4.4 nmol adducts per mol DNA). Mean adduct levels in liver of fish from Statfjord A (20.4 ± 24.1 nmol adducts per mol DNA) were higher compared with mean levels of liver of
73 fish from Tampen ((18.4 ± 35.5 nmol adducts per mol DNA).) and liver of fish from the Egersund Bank (2.2 ± 4.2 nmol adducts per mol DNA). DNA adduct levels in liver of haddock from Tampen in 2017 are higher than what was reported by Balk et al (2011), with measurement from haddock caught in 2002 and from the condition monitoring in 2005, 2008 and 2011 (Grøsvik et al., 2007, 2009 and 2012). DNA adduct levels in fish from Statfjord A and Tampen in 2017 are surprisingly high as EAC levels in cod and haddock are set to 6.7 nmol per mol normal nucleotides. Levels above EAC are considered high and cause for concern (ICES, 2011 and 2012).
70
60
50
40 Intestine 30 liver nucleotides 20
nmol adducts per mol mol per adducts nmol 10
0 Cod Egersund Bank Cod Tampen Cod Statfjord A
Figure 49 Comparisons of DNA adducts from intestine and liver of cod, n = 7 for each group with intestine and 10 for each group with liver.
60
50
40
30 Intestine liver nucleotides 20
nmol adducts per mol 10
0 Ling Egersund Ling Tampen Ling Statfjord A Bank
Figure 50 Comparisons of DNA adducts from intestine and liver of ling, n = 7 for each group with intestine and 10 for each group with liver.
74
35
30
25
20 Intestine 15 liver 10
5
nmol adductsper mol nucleotides 0 Saithe Egersund Saithe Tampen Saithe Statfjord A Bank
Figure 51 Comparisons of DNA adducts from intestine and liver of saithe, n = 7 for each group with intestine and 10 for each group with liver.
70 60 50 40 Intestine 30 liver nucleotides 20
nmol adducts per mol 10 0 Whiting Egersund Whiting Tampen Whiting Statfjord A Bank
Figure 52 Comparisons of DNA adducts from intestine and liver of whiting, n = 7 for each group with intestine and 10 for each group with liver.
140 120 100 80
60 liver
nucleotides 40 20 nmol adducts per mol 0 Haddock Egersund Haddock Tampen Bank
Figure 53 DNA adducts from liver of haddock, n = 10.
75
3.3.9 DNA damage in lymphocytes, Comet assay The percentage DNA damage assessed through the size of the comet-like tail showed a significant increase in the lymphocytes of cod collected from the Statfjord A platform compared to cod from the Egersund Bank (Figure 54). Whiting, ling, and saithe showed no significant difference in DNA damage between the different sampling locations, with generally low levels of DNA damage in all groups.
Figure 54 DNA strand breakage in fish blood samples from four fish species from the different locations. Median, quartiles (box), 10/90 percentiles. Letters denote significant differences between locations for the same species Values in parenthesis indicate specific n value for each group.
76
Figure 55 DNA strand breakage and oxidation of base pairs in fish blood samples from four fish species at the different locations. Median, quartiles (box), 10/90 percentiles. Letters denote significant difference between the groups for each fish species. Values in parenthesis indicate specific n value for each group.
The % DNA damage together with the oxidation of base pairs measured through enzyme treatment is shown in Figure 55. When considering DNA strand breakages including the oxidation of base pairs, significant differences between the locations for three of the four fish species were found. In whiting, significantly higher damage plus oxidation was found in the Statfjord A and Central North Sea fish compared to Egersundbank (ANOVA, Tukey p < 0.05). In saithe, significantly higher DNA damage plus oxidation was found in Tampen fish compared to Central North Sea and Egersund Bank (ANOVA, Tukey p < 0.05). In cod, very high levels of DNA damage plus oxidation were observed in fish from Statfjord A, which was significantly higher than Egersundbank and Tampen fish (ANOVA, Tukey p < 0.05).
3.3.10 Acetylcholine esterase inhibition (AChE) AChE activity in the muscle of four fish species from the different locations is shown in Figure 56. For ling, a significant inhibition of AChE activity was found in fish from Statfjord A compared to all other locations. Cod from Statfjord A were also significantly inhibited, along with Tampen cod, compared to those from Egersundbank (ANOVA, Tukey p<0.05). No significant differences were found for whiting and significantly higher AChE activity were observed in saithe from the Tampen region compared to saithe from the Central North Sea (ANOVA, Tukey, p < 0.05). The species differences in AChE activity observed may be described with respect to the different habitats that would influence exposure pathways. For example, whiting and saithe are more pelagic, whilst ling and cod are more demersal, often feeding on or just over the sediment
77 surface. The lower AChE activity in ling and cod from Statfjord A and Tampen may indicate possible exposure to neurotoxic contaminants from the seafloor, such as from drill cuttings and muds, and leakages from well deposits. ICES assessment criteria have been developed for AChE in a few marine fish including dab (Limanda limanda), flounder (Platichthys flesus), red mullet (Mullus surmuletus) and eelpout (Zoarces viviparus) with BACs ranging from 235 to 124, and EACs ranging from 165 to 87 nmol/min/mg protein in fish fillet. It is important to remember that since the test measures the inhibition of the AChE enzyme, the EACs are lower than the respective BACs. In the present survey, the median AChE activity values for the four fish species were between 5 and 28 nmol/min/mg protein. These values were 10 to 15-fold lower than the assessment criteria of the other fish species listed above; therefore, the ICES assessment criteria are not suitable as reference in this study.
60
50 Whiting Ling
40 b b 30 b
20 a 10
0 CNS Egersundbank Statfjord Tampen CNS Egersundbank Statfjord Tampen (14) (15) (15) (2) (2) (15) (4) (15) 60
50 Saithe Cod
nmol ATC/ min/ mg protein 40 ab b
30 ab b
20 a a a
10
0 CNS Egersundbank Statfjord Tampen CNS Egersundbank Statfjord Tampen (15) (15) (15) (15) (0) (15) (10) (7) Figure 56 The inhibition of acetylcholine esterase in fish fillet samples of four fish species from the different locations. Median, quartiles (box), 10/90 percentiles. Letters denote significant differences between groups within each fish species. Values in parenthesis indicate specific n value for each group.
3.3.11 Gene expression I – qPCR to select gene transcripts Gene transcript levels of cod liver from Ah-related genes are shown Figure 57. The Ah-receptor AHR2 transcript and the Ah receptor repressor transcript did not significantly change at the Tampen region compared with the Egersund Bank. Levels of CYP1A also did not show significant differences. Unfortunately, liver of cod caught at the Statfjord A platform was possibly degraded and not of sufficient quality to be included in the qPCR analyses.
78
A
B
C Figure 57 Relative concentration of gene transcripts from the Ah-receptor related genes AHR2, AHRR and CYP1A from cod liver from the Egersund Bank and the Tampen region.
Relative concentration of the gene transcripts GADD45A and GADD45G from cod liver from the Egersund Bank and the Tampen region is shown in Figure 58. These genes were selected as they belong to the DNA repair pathway. They did not show significant changes between the two regions.
79
A B Figure 58 Relative concentration of the gene transcripts GADD45A and GADD45G from cod liver from the Egersund Bank and the Tampen region.
3.3.12 EROD EROD activity in fish liver was measured in saithe and whiting from the different locations Figure 59. Due to the poor integrity of the liver sample from the Statfjord A region, which resulted in negative EROD values, the data for Statfjord A is not shown. Overall, EROD activity was higher in saithe compared to whiting at all locations. For both fish species, there were no significant differences in EROD activity between three locations. The EROD activity measured in saithe in the present study was similar to the values measured in saithe for WCM2014 and WCM2013. Furthermore, EROD activity in whiting from the present study was similar to the activity levels measured in WCM2013. In all cases, EROD activity in fish showed no relationship between exposure to oil and gas related activity.
Figure 59 EROD activity in liver samples from saithe and whiting sampled at the different locations. Median, quartiles (box), 10/90 percentiles, n=10. No significant differences between the groups within fish species (Kruskal-Wallis ANOVA).
80
3.3.13 CYP 1A (ELISA analysis) Levels of CYP1A in liver of cod (A) and saithe (B) from the regional survey are shown in Figure 60A. No significant differences between cod in CYP1A levels at the Egersund Bank compared with the Tampen region were observed. For saithe, we did not observe significant differences between fish caught at the three regions; Central North Sea, Egersund Bank and Tampen (Figure 60B). Cod and saithe caught close to the Statfjord A platform were not considered to be of sufficient quality to be included in the analyses. These samples were probably thawed during transport to the laboratory. The same conclusion was drawn on qPCR of cod from the Statfjord A platform based on DNA quality assessment. Non-significant differences in cod from the Egersund Bank vs the Tampen region were also observed in the Condition monitoring in 2011 (Grøsvik et al., 2012).
A B Figure 60 Levels of CYP1A in liver of cod (A) and saithe (B). For measurements of CYP1A1 in cod liver we used monoclonal mouse anti-cod CYP1A (NP-7, Biosense, Norway), diluted 1:1000. For CYP1A measurements in saithe, we used monoclonal rabbit anti-fish CYP1A (C10-7, Biosense, Norway), diluted 1:500.
81
3.4 Data treatment 3.4.1 Integrative assessment using the IBR index The IBR index was calculated from star plots of normalised biomarker data from both mussels and fish. Due to the nature of the assessment only those biomarkers with complete data sets could be used for the analysis, thereby enabling comparisons between groups. With this in mind the following biomarkers were used in the IBR/n calculation in mussels: CI, SOS, AChE, LMS and MN. Since some variables are expected to decrease (i.e. CI, SOS, AChE, LMS), the inverse of these results were applied. In fish, the following parameters were included in the analysis: PAH metabolites, AChE, liver histopathology, comet assay including oxidation and DNA adducts. Since some variables are expected to increase (i.e. PAH metabolites, comet, oxidation, DNA adducts, liver histopathology) and others to decrease (AChE), the inverse of the latter test was applied.
Mussels The IBR/n for 19 stations and T0 is reported in Figure 61, this calculation was based on four biomarker responses (CI, SOS, AChE and LMS). All star plots are included in Appendix 8 (IBR index all data). Highest IBR/n’s were reported in mussel cage at stations SFA 500/E and SFA 500 SE1, which were positioned 500 m from the Statfjord A platform in the direction of the PW plume. The highest contribution is different; for station SFA 500/SE1, LMS contributed the most, and for station SFA 500/E the major contribution was from LT50 results (from the stress on stress test). The field reference groups (Ref 1 and 2) showed very different responses, with mussels from Ref 2 having a high IBR/n. In this case, the contribution was equal for 3 out of 4 biomarkers. This highlights the importance of having more than one reference group in biological effect monitoring studies. In addition, since mussels from both stations are considered to originate from a known clean area, it can be considered that all the stations with IBR/n within the range of the reference value are not impacted. As a further assessment, the biomarker MN was included in the calculation of the IBR/n in 10 mussel groups (Figure 62). In this case, mussel groups caged at SFB 500/E, SFA 500/E and SFA 500/SE1 were the most elevated indicating a stress response. All three of these mussel groups were positioned closest (500 m) to either Statfjord A and Statfjord B platform. Highest concentrations of PAH (measured as sum of PAHs) were measured in mussels from station SFB 500/E, showing an exposure / effect relationship. Elevated PAH concentrations were also found in mussels caged at stations SFA 500/E and SFA 500/SE1.
82
2 IBR/n 1
0
Figure 61 Integrated biological response (IBR/n) calculated from the star plots of mean normalised biomarker data in mussels (19 stations and T0). The biomarkers included in the analysis were: CI, SOS, AChE and LMS.
1.5
1 IBR/n 0.5
0 SFB 500 SFA 500 E SFA 500 SFA 500 SFA 500 SFA 500 SFA 1000 SFA 2000 Ref 1 Ref 2 T0 SE SE1 SE2 SW NE NW NW
Figure 62 Integrated biological response (IBR/n) calculated from the star plots of mean normalised biomarker data in mussels (10 stations and T0). The biomarkers included in the analysis were: CI, SOS, AChE, LMS and MN. 83
Fish The IBR/n calculations for all fish locations were performed using five biomarker responses (PAH metabolites, AChE, liver histopathology, comet and oxidation, Figure 63). All star plots are reported in Appendix 8. Highest IBR/n’s were found in fish from the Tampen and Statfjord areas. PAH metabolites gave the highest contribution in all species collected at the Tampen area. While for fish collected in Statfjord, different biomarkers contributed in the different species. Lower IBR/n’s were calculated in fish from the reference area Egersund bank. Fish sampled in Central North Sea had IBR/n similar to the reference area. Since DNA adducts were not measured in fish from the Central North Sea, an additional calculation of IBR/n was made using DNA adduct data from the 3 locations (Figure 64). Results were confirmed, fish from the Tampen and Statfjord A locations showed the highest IBR/n and indicated a clear biological effect response. The nature of the IBR/n calculation restricts comparisons with other similar studies, but it is a useful tool for highlighting differences between groups within the same study.
Figure 63 Integrated biological response (IBR/n) calculated from the star plots of mean normalised biomarker data in fish. The biomarkers included in the analysis were: PAH metabolites, AChE, liver histopathology, comet assay including oxidation. CNS = Central North Sea; Egersund = Egersund bank.
Figure 64 Integrated biological response (IBR/n) calculated from the star plots of mean normalised biomarker data in fish. The biomarkers included in the analysis were: PAH metabolites, AChE, liver histopathology, comet assay including oxidation, and DNA adducts.
84
3.4.2 Biomarker Bridge This first attempt to apply the ‘biomarker bridges’ approach (Rye et al. 2006, Smit et al. 2009, and Sanni et al. 2017a, b) to enhance the interpretation of biomarker based offshore surveys related to produced water on the NCS, has been able to utilize a considerable part of the survey data. It has been possible to indicate relationships between obtained biomarker responses and discharges and link these with expected impacts and risks. A brief introduction to the approach and a detailed description of the results are given in Appendix 9 (Interpretation of environmental data using the ‘Biomarker Bridges’approach). The main indications that could be made by the ‘biomarker bridges’’ assessments in this specific survey were: • There was a reasonably good correspondence between PAH metabolites and exposure concentrations of PAH components, as calculated by DREAM based on currents measured during the survey. This was the case both in the ‘near field’ and ‘far field’ areas, and it was also supported by the chemical measurements of PAHs in mussels. • PAH metabolites also seemed to be in reasonable accordance with expected impact and risk values as evaluated by DREAM calculations and ‘biomarker bridges’ curves. If reliable within the variability boundaries of this approach, this is a noteworthy result, as comparisons to the discharges and validations of the calculated risk are otherwise difficult and highly uncertain to accomplish. • There seemed to be considerably higher genotoxicity responses in comparison to the PAH metabolite response, when applying the ‘biomarker bridge’ curves. There is limited experience using this approach to determine with certainty if these genotoxic responses were truly higher; however, a high genotoxicity response is in agreement with previous surveys carried out at the Tampen area (Balk et al., 2011).
85
4. Conclusions and recommendations 4.1. Conclusions Mussels • The PAH concentration measured in mussel tissues was higher in organisms that were caged closest to Statfjord A and B platforms compared to reference areas, but values were lower than in previous surveys. The source of the PAHs has been identified as petrogenic. Naphthalenes were the most abundant followed by phenanthrenes and dibenzothiophenes. • Some biological effect parameters (CI, stress on stress, LMS, MN), highlighted a stress condition present in mussels caged at Statfjord, compared to reference areas. However, according to the ICES assessment criteria, mussels are categorised as stressed but compensating, as confirmed by the physiological level measurements (stress on stress and CI) and their ability to maintain reproductive development (e.g spawning status). • Data integration using IBR/n confirmed that stressed organisms were compensating, by showing IBR/n values within the range of the two reference stations. • Signs of severe stress (LMS and MN) were recorded only in mussels caged 500 m from Statfjord A, as also confirmed by higher IBR/n values. • In 4 stations at the Statfjord platform, MN frequency values were above the elevated response limit suggested by ICES, showing a clear sign of the presence of contaminants with genotoxicity potential. Fish • PAH metabolites were significantly higher in cod collected at Statfjord (i.e. 2,3-ring PAHs and 5-ring PAHs), and in whiting sampled at Tampen area (i.e. 4-ring PAHs) compared to the reference sites. • The increased levels of PAHs at Tampen and Stafjord shown in soft tissues of deployed mussels was less evident in fish. • Genotoxic effects were clear in fish, as revealed by both DNA adduct and comet assay results. • DNA adduct levels were significantly different between species at Tampen, with higher levels in liver for haddock and saithe, compared to the reference area. Levels of DNA adducts in fish liver of all investigated species (cod, haddock, saithe and whiting) were above EAC at Tampen and at the Statfjord A field. • Levels of DNA adducts from haddock livers at Tampen were higher than those reported in the condition monitoring programs from 2005-2011. • DNA adducts from intestine samples from fish caught at Statfjord had levels above EAC threshold values and were statistically significantly higher than values from fish collected from the reference area. Part of the difference could, however, be explained by age differences among the fish. • AChE activity in muscle from fish collected at Statfjord (ling and cod) and Tampen (cod) may indicate possible exposure to neurotoxic contaminants.
4.2 Knowledge gaps and future development This extensive survey has been a successful integration of the previous Effect Monitoring Program and the Condition Monitoring Program, supporting the use of this merged approach for future WCM. The WCM has gained from the synergy of the activities between the involved institutes and lead to a better basis for the evaluations. In general, the selected biological parameters and chemical analyses were capable of identifying the health status of the organisms. However, the sensitivity of the single parameters and the ecological relevance of some of them could be revised. In this context, the guidelines provided
86 in 2015 (Iversen et al., 2015) could be re-evaluated as well, trying to address even more the impact of oil and gas activities and in particular the discharge of PW at the ecosystem level. Additional biological parameters at higher level of organisations could improve the evaluation of the effects at population and ecosystem level. The presence of contaminants with genotoxic and neurotoxic effects should be addressed in future surveys, supporting the existing biomarkers (DNA adducts, micronuclei, AchE) with additional ones and also specific laboratory studies. Due to the possibility of the reproductive status of organisms influencing biological effects responses, the opportunity of commencing the WCM survey outside the mussels and fish spawning season is recommended. The species selection is a very important factor of the entire study and the use of both benthic and demersal species is highly recommended. The use of multiple species is clearly a key element of the evaluation at ecosystem level. More data on PAH levels in benthic invertebrates and from sedimentation traps would be interesting to better model and understand contribution from ongoing discharges of produced water versus old discharges of oil based mud in cutting piles. Method development that better identify and quantify bile metabolites and which PAHs that contribute most to DNA adducts would be beneficial, as well as better understanding of effects through gene transcription studies. The biomarker data were also integrated using the Species Sensitivity Distributions based on biomarkers. The overall data evaluation based on this approach has been valuable and it is suggested to develop the tool even further to be applied in future WCM surveys.
5. Appendices Appendix 1 (Zooplankton monitoring WP4) Appendix 2 (PFAs in blood samples of offshore fish) Appendix 3 (Current meter reports) Appendix 4 (Stress on stress all data) Appendix 5 (Histopathology of fish liver) Appendix 6 (DNA adduct report – liver) Appendix 7 (DNA adduct report – intestine) Appendix 8 (IBR index all data) Appendix 9 (Interpretation of environmental data using the ‘Biomarker Bridges’ approach)
87
6 References
Aas et al. 1998. PAH in fish bile detected by fixed wavelength fluorescence. Marine Environmental Research 46, 225-228. Aas et al. 2000. PAH metabolites in bile, cytochrome P4501A and DNA adducts as environmental risk parameters for chronic oil exposure: a laboratory experiment with Atlantic cod. Aquatic Toxicology 51, 241-258. Amiard et al. 2004. Temporal changes in nickel and vanadium concentrations and in condition index and metallothionein levels in three species of molluscs following the “Erika” oil spill. Aquatic Living Resources 17(3), 281-288. Balk et al. 2011. Biomarkers in Natural Fish Populations Indicate Adverse Biological Effects of Offshore Oil Production. PlosOne 6, e19735, 10pp. Bayne 1989. Measuring the Biological Effects of Pollution: The Mussel Watch Approach. Water Science and Technology 21(10-11), 1089-1100. Beyer et al. 2017. Blue mussels (Mytilus edulis spp.) as sentinel organisms in coastal pollution monitoring: A review. Marine Environmental Research 130, 338-365. Beliaeff and Burgeot 2002. Integrated biomarker response: a useful tool for ecological risk assessment. Environmental Toxicology and Chemistry 21, 1316-1322. Benly et al. 2008. Sublethal ammonia exposure of Nile tilapia (Oreochromis niloticus L.): effects on gill, liver and kidney histology. Chemosphere 72, 1355-1358. Bernet et al. 1999. Histopathology in fish: proposal for a protocol to assess aquatic pollution. Journal of Fish Diseases 22(1), 25-34. Bigas et al. 2001. Cytological effects of experimental exposure to Hg on the gill epithelium of the European flat oyster Ostrea edulis: ultrastructural and quantitative changes related to bioaccumulation. Tissue Cell 33:178–188. Bignell et al. 2011. Histopathology of mussels (Mytilus sp.) from the Tamar estuary, UK. Marine Environmental Research 72, 25-32. Bocquené and Galgani 1998. Biological effects of contaminants: Cholineesterase inhibition by organophosphate and carbamate compounds. ICES Techniques in Marine Environmental Sciences, 22 pg. 19. Bolognesi and Fenech 2012. Mussel micronucleus cytome assay. Nature Protocols 7, 1125- 1137. Bolognesi and Hayashi 2011. Micronucleus assay in aquatic animals. Mutagenesis 26, 205- 213. Bradford 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analical Biochemistry 72, 248-254. Broeg and Lehtonen 2006. Indices for the assessment of environmental pollution of the Baltic Sea coasts: integrated assessment of the multi-biomarker approach. Marine Pollution Bulletin 53, 508-522. Brooks et al. 2011. Water column monitoring 2011. NIVA report 6237-2012. Brooks et al. 2013. The Water Column Monitoring Programme 2013: Determining the biological effects of two offshore platforms on local fish populations. NIVA report 6595- 2013, pp. 62. Brooks et al. 2014. Water column monitoring 2014. Determining the biological effects of an offshore platform on local fish populations. NIVA report 6735-2014, pp. 75. Brooks et al. 2011. Water Column Monitoring of the Biological Effects of Produced Water from the Ekofisk Offshore Oil Installation from 2006 to 2009. Journal of Toxicology and Environmental Health A 74, 582-604. Brunborg et al. 2014. High throughput sample processing and automated scoring. Frontiers in Genetics 28, 373, pp.6.
88
Burke and Mayer 1974. Ethoxyresorufin – direct fluorimetric assay of a microsomal O- dealkylation which is preferentially inducible by 3-methylcholanthrene. Drug Metabolism and Disposition 2, 583-588. Davies and Vethaak 2012. Integrated monitoring of chemicals and their effects. ICES Cooperative Research Report No. 315. 277 pp. de Zwaan et al. 1995. Resistance of bivalves to anoxia as a response to pollution-induced environmental stress. Science of The Total Environment 171(1), 121-125. Divi et al. 2002. Highly sensitive chemiluminescence immunoassay for benzo[a]pyrene-DNA adducts: validation by comparison with other methods and use in human biomonitoring. Carcinogenesis 23, 2043-2049. Dytham 2003. Choosing and Using Statistics: a Biologist's Guide. Oxford: Blackwell Publishing. DNV 2012. Characterization of drill cuttings piles at Statfjord A. Report No./DNV Reg No.: 2012-4016/ 13I64UL-10. Eertman et al. 1993. “Survival in air” of the blue mussel Mytilus edulis L. as a sensitive response to pollution-induced environmental stress. Journal of Experimental Marine Biology and Ecology 170, 179-195. Eggens and Galgani 1992. Ethoxyresorufin-O-deethylase (EROD) activity in flatfish: Fast determination with a fluorescence plate-reader. Marine Environmental Research 33, 213- 221. Ellman et al. 1961. A new and rapid colorimetric determination of acetylcholinesterase activity. Biochemical Pharmacology 7, 88-90. Feist et al. 2004. Use of liver pathology of the European flatfish dab (Limanda limanda L.) and flounder (Platichthys flesus L.) for monitoring. ICES Techniques in Marine Environmental Sciences 38, ICES, Copenhagen. Fernández et al. 2011. Micronuclei and other nuclear abnormalities in mussels (Mytilus galloprovincialis) as biomarkers of cyto-genotoxic pollution in Mediterranean waters. Environmental Molecular Mutagenesis 52, 479-91. Gehan 1965. A generalized Wilcoxon test for comparing arbitrarily singly-censored samples. Biometrika 52(1-2), 203-224. Gomz-Mendikute et al. 2005. Characterization of mussel gill cells in vivo and in vitro. Cell Tissue Research 321, 131-140. Gosling 2003. Reproduction, settlement and recruitment. In Bivalve Molluscs. Biology, Ecology and Culture. Blackwell Publishing, pp. 131-168. Grøsvik et al. 2012. Condition monitoring in the water column 2011: Oil hydrocarbons in fish from Norwegian waters. IMR Report No. 19-2012. Grøsvik et al. 2015. Condition monitoring 2014- with focus on the Halten Bank area. IMR report No 23-2015. pp 46. Grøsvik et al. 2009. Condition monitoring in the water column 2008: Oil hydrocarbons in fish from Norwegian waters. IMR Report No 2-2009. pp 61. Grøsvik et al. 2007. Condition monitoring in the water column 2005: Oil hydrocarbons in fish from Norwegian waters. IMR Report No. 2-2007. pp 33. Hansson et al. 2017. Supporting variables for biological effects measurements in fish and blue mussel. ICES Techniques in Marine Environmental Sciences. No. 60. 22 pp. http://doi.org/10.17895/ ices.pub.2903 Harman et al. 2008. Uptake of some selected aquatic pollutants in semipermeable membrane devices (SPMDs) and the polar organic chemical integrative sampler (POCIS). Journal Environmental Monitoring 10, 239–47.
89
Heddle 1983. The induction of micronuclei as a measure of genotoxicity: A report of the U.S. environmental protection agency Gene-Tox program. Mutation Research/Reviews in Genetic Toxicology 123, 61-118. Helson et al. 2007. Does differential particulate food supply explain the presence of mussels in Wellington Harbour (New Zealand) and their absence on neighbouring Cook Strait shores? Vol. 72. 223-234. Hickman et al. 1991. The relationship between farmed mussels,Perna canaliculus, and available food in Pelorus-Kenepuru Sound, New Zealand, 1983–1985. Aquaculture, 99(1): p. 49-68. Hylland et al. 2002 An ICES Workshop on Biological Effects in Pelagic Ecosystems (BECPELAG): Summary of Results and Recommendations. Hylland et al. 2006. Biological effects of contaminants in marine pelagic ecosystems. SETAC (Society of Environmental Toxicology and Chemistry) publications, pp. 474. Hylland et al. 2005. Water column monitoring 2004. Summary report. NIVA-report 4993- 2005, pp. 228. Hylland et al. 2008. Water column monitoring near oil installations in the North Sea 2001– 2004. Marine Pollution Bulletin, 56(3): p. 414-429. Kaplan and Meier 1958. Nonparametric Estimation from Incomplete Observations. Am. state. Association 53, 203-223. ICES 2011. Report of the Study Group on Integrated Monitoring of Contaminants and Biological Effects (SGIMC), 14–18 March 2011, Copenhagen, Denmark. ICES CM 2011/ACOM:30. 265 pp. ICES 2012. Report of the Working Group on Biological Effects of Contaminants (WGBEC), 12–16 March 2012, Porto, Portugal. ICES CM 2012/SSGHIE:04. 131 pp. Inoue et al. 1995. Interspecific variations in adhesive protein sequences of Mytilus edulis, M. galloprovincialis and M. trossulus. Biology Bulletin 189, 370–375. Iversen et al. 2015. Retningslinjer for miljøovervåking av petrolemsvirksomheten til havs - Guidelines for environmental monitoring of petroleum activities on the Norwegian continental shelf. Norwegan Environment Agency M-408, pp 63. Johnsen et al. 1998. Dilution and Bioavailability of Produced Water Compounds in the Northern North Sea. A Combined Modelling and Field Study. Society of Petroleum Engineers pp. 11 doi:10.2118/46578-MS. Jonsson et al. 2003. The application of HPLC-F and GC-MS to the analysis of selected hydroxy polycyclic hydrocarbons in two certified fish bile reference materials. Journal of Environmental Monitoring 5, 513-520. Jonsson et al. 2004. Quantitative determination of de-conjugated chrysene metabolites in fish bile by HPLC-fluorescence and GC-MS. Chemosphere 54, 1085-1097. Lacroix et al. 2014. A selection of reference genes and early-warning mRNA biomarkers for environmental monitoring using Mytilus spp. as sentinel species. Marine Pollution Bullettin 86, 304-313. Le Goff et al. 2006. DNA adduct measurements in zebra mussels. Dreissena polymorpha. Pallas. Potential use for genotoxicant biomonitoring of fresh water ecosystems. Aquatic Toxicology 79, 55-64. Lohrmann et al. 2019. Histopathological assessment of the health status of Mytilus chilensis (Hupé 1854) in southern Chile. Aquaculture 503, 40-50. Lowry et al. 1951. Protein measurement with the Folin Phenol Reagent, Journal of Biological Chemistry 192, 265-275. Lucas and Beninger 1985. The use of physiological condition indices in marine bivalve aquaculture. Aquaculture 44, 187– 200. Marsh and Weinstein 1966. Simple charring method for determination of lipids. Journal of Lipid Research 7, 574-576.
90
Marsden 2004. Effects of reduced salinity and seston availability on growth of the New Zealand little-neck clam Austrovenus stutchburyi. Marine Ecology Progress Series 266, 157-171. Matozzo et al. 2018. Assessing the health status of farmed mussels (Mytilus galloprovincialis) through histological, microbiological and biomarker analyses. Journal of Invertebrate Pathology 153, 165-179. Moore 1976. Cytochemical demonstration of latency of lysosomal hydrolases in digestive cells of the common mussel Mytilus edulis, and changes induced by thermal stress. Cell and Tissue Research 175, 279–287. Moore et al. 2006. Environmental prognostics: an integrated model supporting lysosomal stress responses as predictive biomarkers of animal health status. Marine Environmental Research 61, 278–304. Mjanger et al. 2017. Håndbok for prøvetaking av fisk og krepsdyr. Versjon 4.0 (SPD). Pp 194. Nilsen BM, Berg K, Goksøyr A. 1998. Induction of Cytochrome P450 1A (CYP1A) experimental integrated aquaculture. Aquaculture Research 38, 1714-1720. Nilssen and Bakke 2011. Water Column Monitoring of Offshore Oil and Gas Activities on the Norwegian Continental Shelf: Past, Present and Future. In K. Lee, J. Neff (eds.), Produced Water, DOI 10.1007/978-1-4614-0046-2_23,Springer Science Business Media, 431-439. Nogarol et al. 2016. Histopathological effects of the herbicide atrazine on gills of the Brazilian endemic bivalve Diplodon expansus. International Journal of Environmental Analytical Chemistry 96, 387-403. Olsvik et al. 2012. Is chemically dispersed oil more toxic to Atlantic cod (Gadus morhua) larvae than mechanically dispersed oil? A transcriptional evaluation. BMC Genomics 13,702. Pampanin et al. 2005. Physiological measurements of native and transplanted mussel (Mytilus galloprovincialis) in the canals of Venice. Survival in air and condition index. Comparative Biochemistry and Physiology A 140, 41-52. Pampanin et al. 2013. Water Column Monitoring 2012 Troll C platform. IRIS Final report 2013-252, pp. 95. Peharda et al. 2007. Growth and condition index of mussel Mytilus galloprovincialis in petroleum activities on the Norwegian continental shelf. Norwegian Environment Agency. M-408. 63 pp. Phillips and Castegnaro 1999. Standardization and validation of DNA adduct postlabelling methods: report of interlaboratory trials and production of recommended protocols. Mutagenesis. 14, 301-15. Reed and Rye 2011. The DREAM Model and the Environmental Impact Factor: Decision Support for Environmental Risk Management. In: Lee K., Neff J. (eds) Produced Water. Springer, New York, NY. DOI https://doi.org/10.1007/978-1-4614-0046-2_9. Reichert and French 1994. The 32P-postolabeling protocol for assaying levels of hydrophobic DNA adducts in fish. US Dep Commer, NOAA Technical Memo. NMFS-NWFSC-14. Rye et al. 2006. Comparing biomarker responses with risk estimates for decision analysis. In: Water Pollution VIII: Modelling, Monitoring and Management. WIT Transactions on Ecology and the Environment 95, 323-335. http://dx.doi.org/10.2495/WP060331. Røe 1998. Produced water discharges to the North Sea: a study of bioavailability of organic produced water compounds to marine organisms. Ph.D. Thesis, Faculty of Chemistry and Biology, Norwegian University of Science and Technology. Sanni et al. 2017a. I: Biomarker quantification in fish exposed to crude oil as input to Species Sensitivity Distributions. Marine Environmental Research 125, 10-24. Sanni et al. 2017c. III: Use of biomarkers as risk indicators in environmental risk assessment of oil based discharges offshore. Marine Environmental Research 127, 1-10.
91
Sanni et al. 2017 b: II: Species Sensitivity Distributions based on Biomarkers and Whole Organism Responses for integrated impact and risk assessment criteria. Marine Environmental Research 127, 11-23. Sanni et al. 2018. Experience with the use of Biomarkers as Risk Indicators in Environmental Risk Assessment of oil based discharges offshore. Journal of Chemical Engineering And Bioanalytical Chemistry 2(1). doi:10.25177/JCEBC.2.1.4. Schmid 1975. The micronucleus test. Mutation Research/Environmental Mutagenesis and Related Subjects 31, 9-15. Schmidt et al. 2013. Seasonal variations of biomarker responses in the marine blue mussel (Mytilus spp.). Marine Pollution Bulletin 74, 50-55. Seed 1969. The ecology of Mytilus edulis L. (Lamellibranchiata) on exposed rocky shores – I. Breeding and settlement. Oecologia 3, 277–316. Sensini et al. 2008. First observations of histopathological effects of 2,4,6-trinitrotoluene (TNT) in gills of European eel Anguilla anguilla (Linnaeus, 1758). Cellular Biology and Toxicology 24, 621-628. Smit et al. 2009. Relating biomarkers to whole-organism effects using species sensitivity distributions: a pilot study for marine species exposed to oil. Environmental Toxicology and Chemistry 28, 1104-1109. St.meld. nr. 58 (1996-97). Miljøvernpolitikk for en bærekraftig utvikling - Dugnad for framtida. Miljøvernpolitikk for en bærekraftig utvikling. Klima- og Miljødepartementet. https://www.regjeringen.no/no/dokumenter/st-meld-nr-58_1996-97/id191317/ Statoil. The Statfjord area. 2017; Available from: https://www.statoil.com/en/what- wedo/norwegian-continental-shelf-platforms/statfjord.html 37. petroleum, N., Statfjord. 2017. Sundt et al. 2011. PAH body burden and biomarker responses in mussels (Mytilus edulis) exposed to Produced Water from a North Sea oil field: laboratory and field assessments. Marine Pollution Bulletin 62, 1498-1505. Sundt et al. 2012a. Biomarker responses in Atlantic cod (Gadus morhua) exposed to produced water from a North Sea oil field: laboratory and field assessment. Marine Pollution Bulletin 64, 144-152. Sundt et al. 2012b. Water column monitoring of offshore produced water discharges. Compilation of previous experience and suggestions for future survey design. IRIS report, 7911854. Thomas et al. 1999. Lack of physiological responses to hydrocarbon accumulation by Mytilus trossulus after 3–4 years chronic exposure to spilled Exxon Valdez crude oil in Prince William Sound. Comparative Biochemistry and Physiology Part C: Pharmacology, Toxicology and Endocrinology 122, 153-163. UNEP/RAMOGE 1999. Manual on the biomarkers recommended for the MED POL biomonitoring programme. UNEP, Athens. Venier et al. 1997. Detection of micronuclei in gill cells and haemocytes of mussels exposed to benzo(a)pyrene. Mutation Research 390, 33-44. Verreault et al. 2005. Chlorinated hydrocarbon contaminants and metabolites in polar bears (Ursus maritimus) from Alaska, Canada, East Greenland, and Svalbard: 1996-2002. Science of the total Environment 351-352(1), 369-390. Viarengo and Canesi 1991. Mussels as biological indicators of pollution. Aquaculture 94(2), 225-243 Viarengo et al. 1995. Stress on stress response: A simple monitoring tool in the assessment of a general stress syndrome in mussels. Marine Environmental Research 39, 245-248.
92
Viarengo et al. 2007. The use of biomarkers in biomonitoring: a 2-tier approach assessing the level of pollutant-induced stress syndrome in sentinel organisms. Comparative Biochemistry and Physiology C 146, 281-300. Yadetie et al. 2018. RNA-Seq analysis of transcriptome responses in Atlantic cod (Gadus morhua) precision-cut liver slices exposed to benzo[a]pyrene and 17α-ethynylestradiol. Aquatic Toxicology 201, 174–186. Zhan et al. 1995. Separation of 32P-labeled 3′,5′-bisphosphate nucleotides of polycyclic aromatic hydrocarbon anti-diol-epoxides and derivatives Journal of Chromatography A. 710, 149-157. Zoll-Moreux and Ferrier 1999. The Jaylet test (newt micronucleus test) and the micronucleus test in xenopus: two in vivo tests on amphibia evaluation of the genotoxicity of five environmental pollutants and of five effluents. Water Research,33, 2301-2314.
93