Widespread persistent contaminants are a global environmental problem. In the Baltic Sea, wildlife contamination was first noticed in the 1960s, prompting the Swedish Environmental Protection Agency to establish a comprehensive Swedish National Monitoring Programme for Contaminants in Marine Biota (MCoM) in 1978 run by the Swedish Museum of Natural History. Eight species have been analysed, four fish species (Atlantic herring, Atlantic cod, European perch, viviparous eelpout), one bivalve species (blue mussel), and egg from three bird species (common guillemot, common tern, Eurasian oystercatcher). Here, we present a dataset containing MCoM data from its start until 2021. It includes 36 sets of time-series, each analysed for more than 100 contaminants. The longest time-series is for common guillemot and starts in 1968. We describe the structure of MCoM including historic changes to the number of stations, sample treatment, analytical methods, instruments, and laboratories. The MCoM data is available at the Bolin Centre repository and on GitHub through our R package mcomDb. The latter will be updated yearly with new MCoM records.

The Water Framework Directive (WFD) requires member states to implement a strategy to meet the environmental quality standards (EQS) set by the WFD for a number of priority substances. Since the EQSs listed in the WFD are derived to protect the most sensitive species in the ecosystem, often top predators or humans, it has been decided that for certain priority substances the quality standards should be compared to a monitoring species with a trophic position (TP) similar to the diet of the most sensitive species. To enable such adjustment require both knowledge about the monitoring species TP in the food web as well as the relationship between contaminant concentration and trophic position. In this study, we focus on finding a suitable method for TP estimates of perch in Swedish lakes by evaluating both traditional stable isotope analysis in bulk samples (BSIA) using different baseline matrices and the more recent development of compound-specific stable isotope analysis in amino acids (CSIA-AA). For this, three representative monitoring lakes were selected in which perch together with potential baseline matrices (bivalves, gastropods and sediment) were sampled. We applied triple-isotope analyses, d¹⁵N, d¹³C and d³⁴S, of bulk material of all sampled matrices, and in addition d¹⁵N in perch using CSIA-AA. Results showed that TP estimates derived from CSIA-AA were significantly (p<0.001) lower compared to all the BSIA-derived methods and further that the BSIA-derived TP estimates using gastropods as a baseline were significantly higher (p<0.001) than all other TP estimates. Since no statistical differences could be detected between TP estimates based on bivalves, sediment or a ‘mixture’ baseline these were assumed to produce similar results and therefore all valid baseline matrices for TP estimates of perch in these three lakes. In the present study we also attempted to adjust mercury contaminant data to a specific TP of 3.5 according to the WFD. The adjustment resulted in significantly different concentration for one of the two tested lakes but did not influence the chemical status classification as all lakes were well above the threshold for mercury in freshwater lakes.

Within the Swedish National Monitoring Programme for Contaminants in marine biota, a selection of the wide array of contaminants that can be found in the environment is analysed. Analysing the samples for all possible contaminants would hardly be feasible, however, screening for different substance groups is a way to investigate if and where new substances arise and may pose a threat to wildlife and humans. In this report, data from a spatial screening study is presented, which aimed to densify the ongoing Swedish National Monitoring Programme for Contaminants in marine biota with regard to pharmaceuticals, phthalates and polycyclic aromatic hydrocarbons (PAHs). The study includes 16 sampling sites along the Swedish coast, from where two different species of bivalves, Limecola balthica and Mytilus edulis, were collected. All applied sampling material in this screening study originates from the Swedish Environmental Specimen Bank of the Swedish Museum of Natural History. The screening included a total of 100 pharmaceuticals, out of which 17 were detected and quantified in at least one sampling site. Risperidone was the pharmaceutical detected at most sites (10 of 13). The only detected phthalate was di(2-ethylhexyl) phthalate (DEHP), which was found in the samples from 3 of 13 sampling sites. Among the PAHs, Benzo(a)pyrene was the substance quantified at most sites (14 out of 16). No geographical patterns could be identified for the detected contaminants, besides for the PAHs. However, this pattern could also be due to a difference in species rather than due to location. PAHs could be detected in the Bothnian Sea and the Sea of Åland, where to date no mussel sampling sites exist within the Swedish National Monitoring Programme for Contaminants in marine biota. The Baltic clam might be a good additional monitoring species besides the Blue mussel, due to their difference in feeding strategy and the potential higher PAH uptake from contaminated sediments rather than the water phase.

Unexpected increasing trends in the concentration of contaminants in European perch (Perca fluviatilis) and inactivity of ethoxyresorufin-O-deethylase (EROD) in European perch and eelpout (Zoarces viviparus) have beenobserved at a Swedish coastal reference site. This study uses data from different sources to investigate plausibleexplanations. The results showed that a change in diet and an improved overall condition coincide with anincrease in mercury in European perch. Furthermore, an increase in several organic contaminants in Europeanperch coincided with the introduction of an invasive deep-burrowing polychaete, which likely contributed to therelease of contaminants through bioturbation. The increase in EROD-activity in both species seems to be relatedto contaminants that reach the fish through the water rather than the diet. The results show that for contaminantsthat are taken up via the diet, trends in contamination can be opposite for different species of fish inthe same area.

Within the Swedish National Monitoring Programme for Contaminants in Freshwater Biota, perch, pike and Arctic char are being analysed from 32 lakes distributed all over Sweden. One of the main objectives of the monitoring has been to detect temporal trends in different areas and for different contaminants. However, the measured concentrations in biota can be affected by biological factors such as age, length, trophic position (TP), fat or dry weight content. A change in one or several of these variables over time, or even among specimens taken within the same year, is likely to result in an increased variation of the measured concentrations and hence in a lower chance of detecting a trend. The purpose of this study was to evaluate if an adjustment of contaminant time trends by primarily stable isotopes, but also other confounding factors can:  A) reduce the individual and between-year variation and hence B) improve the likelihood of detecting temporal trends in time series for metals and organic contaminants. The study revealed an improvement in the likelihood of detecting a contaminant trend after adjustment of the data for almost all lakes and for all tested substances. For some lakes and substances, this improvement was substantial. Age seems to be an important confounding factor for metals and per- and polyfluoroalkyl substances. For lipophilic substances, fat content also appears to be an important factor in some lakes together with adjustments for carbon and nitrogen isotope ratios (d¹³C and d¹⁵N), which all reduced the variation in contaminant data. This suggests that there is a benefit of adjusting data on a regular basis in contaminant monitoring programmes in the future.

This report present a graphic and statistical overview of temporal trends and spatial variations within the Swedish National Monitoring Programme for Contaminants in Marine Biota.

The aim of the report is three-fold. Firstly, to evaluate to what extent phase-out of certain per- and polyfluoroalkyl substances (PFAS) first initiated in the early 2000s are reflected in the biota concentration within the Baltic Sea. Secondly, to investigate the spatial differences of PFAS across the Baltic Sea, which has not previously been evaluated, and couple these results to PFAS observations in other matrices to better understand the flow of PFAS through the Baltic Sea ecosystem. Lastly, to investigate the implication of a proposed PFAS EQS dossier on the evaluation of Good Environmental Status in the Baltic Sea.

We use data on PFAS from the Swedish Monitoring Program for Contaminants in Marine Biota covering 40 years for the longest time series (four Time Trend Stations) and 26 stations at present day. The program covers stations from the Bothnian Bay in the north to Skagerrak at the Swedish west coast (referred to as the Greater Baltic Sea) and PFAS is analyzed in four fish species and three bird species. The target compounds are perfluoroalkyl sulfonic acid (PFSA; C4-C8), perfluoroalkyl carboxylic acid (PFCA; C6-C15) and FOSA. In the data evaluation, observations from other monitoring programmes and research campaigns are included (river, marine surface water, sediment, and top predators) to set the observations from the Swedish Monitoring Program for Contaminants in Marine Biota into context. 

After an initial exponential increase in PFAS concentrations, we found that PFAS displayed a rapid response to phase-out and regulations in the early 2000s. As a result, PFAS concentrations stabilized (rather than displaying an immediate decrease). This is linked to the few removal pathways of the long lived PFAS homologues in the Baltic Sea water column. However, within the last decade PFAS has started to show significant declines at many stations, the exception being a few PFAS homologues (PFOA (C8) and PFNA (C9)). 

Two distinct water masses are present in the Greater Baltic Sea, North Sea water and Baltic Sea water, the latter with origin in the Baltic Sea drainage basin. Differences in the PFAS loads of these water masses likely drives geographical differences seen in the PFAS concentrations and homologue distribution between fish in Kattegat and the Baltic Sea. Different water mass lifetimes in the Baltic Sea basins further affect the concentration and response time with regards to changes in external sources for individual homologues.

A proposed PFAS EQS dossier presents an EQS (with the PFAS sum expressed as PFOA-equivalents) a factor 100 lower than the current PFOS EQS. We find that PFOS, PFNA and PFUnDA contribute more than 80% to the sumPFAS-equivalent. PFNA is the only homologue that is currently increasing in some parts of the Baltic Sea and special attention should be on this homologue in the future. Despite the slow recovery with regards to PFAS concentrations over the past decade the biota PFAS concentrations are still 5-230 times higher than the threshold proposed in the PFAS EQS dossier. Screening studies has further identified a range of PFAS in the Greater Baltic Sea not currently part of the Swedish Monitoring Program for Contaminants in Marine Biota. As an example the cyclic PFECHS has been found at various trophic levels in the food web at concentrations indicating biomagnification potential. These findings indicate that PFAS is still affecting the Greater Baltic Sea environment negatively and the development in concentrations of individual homologues should be followed closely over the coming decade both for those PFAS included in the current program but also the emergence of novel PFAS through screening program.

Fish tissue levels have to comply with environmental quality standards (EQSs) within the European Water Framework Directive. However, within monitoring, contaminants are sometimes measured in a different tissue than the tissue for which the environmental (whole fish) or human (fillet (equivalent to muscle tissue)) quality standard is set. Tissue conversion factors (k), describing the relationship between concentrations in different tissues, can be used to obtain a quality standard for the appropriate tissue. Several different approaches have been suggested for the calculation of k. For monitoring purposes, we propose the use of a simple, easy reproducible approach that assumes proportionality between two tissue, or tissue and whole fish, concentrations. This allows for an easy comparison of studies and adoption of ks into independent monitoring programs. Here, we determined ks for three metals (mercury (Hg), lead (Pb), cadmium (Cd)) and nine per- and polyfluoroalkyl substances (PFAS) including perfluorooctanesulfonic acid (PFOS) across six marine and freshwater fish species from Northern European lakes and the Baltic Sea. We found significant species differences for Hg for k_muscle/whole fish, for Cd and Pb for k_{liver/whole fish} and for Cd for kliver/muscle. For perfluoroalkyl carboxylic acids (PFCA), we found a chain length dependence with lowest kliver/muscle at low and high chain lengths (C₈, C₁₃) and highest for median chain lengths (C₉-C₁₂). Further, there were differences between fish species with k_liver/muscle for PFOS almost doubling from eelpout (10.3) to herring (19.2) and increasing up to a factor 4 between eelpout and herring for other PFASs. FOSA had two distinctive groups, herring with a k_liver/muscle of 48.7 and a second group with ks of 2.3 to 5.9 for all other fish species. Our results suggest that differences in the tissue somatic index, and contaminant uptake, tissue transfer and metabolism result in the need for species-specific ks within monitoring

The aim of the report is to produce improved estimates of tissue conversion factors (k) to use within environmental monitoring. The contaminant distribution across a range of marine and freshwater fish species from Northern Europe freshwater and the Baltic Sea was investigated. New tissue conversion factors were established and converted threshold limits (C-EQS or C-QS) for tissues of relevance for monitoring (liver and muscle) presented. We further explored inter-species variability in contaminant distribution, which can result in the need to create species-specific conversion factors. 

                      In this study, we use conversion factors that assumes a proportional relation between two tissue concentrations. We recommend that this method is always used when the aim is to convert between tissue concentrations for monitoring purposes. This will allow for better transparency and allow for easy comparison between studies, something that is made difficult at the moment due to the use of different alternative methods for conversion.

                      Based on the analysis we recommend four new threshold value estimates that can be used within environmental monitoring of fish. For mercury (Hg) we only looked at the kmuscle/whole fish and did not consider liver measurements. We recommend the use of a C-EQSmuscle of 24 ng g^-1 ww across all fish species. For cadmium (Cd) and lead (Pb) we calculated C-QSliver estimates based on conversion from both the QSmuscle-human health and QSwhole fish-sec pois. For Cd, we found significant differences between the two species investigated (herring and perch) as well as ten times higher C-QSliver based on the QSmuscle-human health than the QSwhole fish-sec pois. From a precautionary principle we recommend to use the herring specific k and the QS for secondary poising set for whole fish concentration to set the C-QSliver, which lead to a C-QSliver for Cd of 2.6 μg g^-1 ww. For lead (Pb), the liver:muscle dataset was considered less robust than the liver:whole fish dataset. Also, for Pb significant differences were found between k’s in herring and perch. We recommend using a C-QSliver of 0.3 μg g^-1 ww for Pb based on perch data. For PFOS we only looked at the kliver/muscle and did not consider whole fish measurements. We recommend the use of a C-EQSliver of 153 ng g^-1 ww across all fish species despite some differences in k between the six species included in the estimate. For all contaminants, more data on additional fish species and data on the same species but better distributed between marine and freshwater environments (for species living in both like perch) is needed to elucidate the need for species specific and environment specific thresholds. In addition, more observations with concentrations close to the thresholds are needed for Cd and Pb to bring down the uncertainty on the QSliver estimates, which are currently, based on extrapolation of values more than a factor ten below the thresholds.