4a-2: Informing the implementation of the Minamata Convention: the role of scientific assessments

The Minamata Convention on Mercury is a multilateral environmental agreement that obligates Parties to reduce or control sources of mercury pollution in order to protect human health and the environment. Evaluating the effectiveness of the Convention (as required by Article 22) is a crucial component to ensure that it is meeting this objective. The purpose of this talk is to describe human biomonitoring of mercury, and how this represents a scientifically robust, straightforward, and potentially cost-effective approach to track the effectiveness of the Convention over temporal and spatial scales. First, I will discuss the pros and cons associated with using accepted biomarkers of Hg exposure (hair, blood, and urine) and associated exposure assessment strategies (e.g., surveys). Second, I will describe national biomonitoring programs that exist in countries including the U.S. (via NHANES), Canada (via CHMS), Germany (via HBM), and Korea (via KNHANES), and describe how these programs have enabled countries to establish baselines and reference ranges (at the national and regional level), help set priorities and take action, and track changes over time as part of surveillance efforts. Finally, I will discuss approaches to gauge exposures in vulnerable populations (e.g., dental professionals, small-island nations, ASGM sector). The talk will attempt to articulate challenges and opportunities concerning the human biomonitoring of mercury in the context of the Convention, and portray global patterns.

The UNEP Minamata Convention on Mercury, signed in 2013 and to date ratified by 35 countries with 128 signatories, recognizes that mercury must be managed at a global scale. To develop regional and local bases for evidence-based management decision-making requires the ability to capture, synthesize, analyze and visualize data at varying spatial scales and time frames. It is essential to facilitate understanding among stakeholders and regulators of how emissions reductions and site-specific projects associated with implementing provisions of the Convention affect the broader exposure and risk profiles of human populations and ecosystems.

This project, which started in 2015, attempts to design and populate a centralized platform, deployed via UNEP Live, for data upload and download as well as knowledge sharing through information synthesis and facilitation of Communities of Practice comprising scientific expert and critical stakeholder groups. The term knowledge sharing is a deliberate attempt to move beyond traditional web-based delivery of quantitative but uninterpreted information. This will be done through user-mediated requests for data synthesis and visualization as well as implementation of data protocols and delivery of meta-data as a component of shared output.

A demonstration pilot project, using a selected set of data on total mercury in billfish and food consumption advisories, was recently completed. In order to understand and anticipate user needs for the platform, a global survey was conducted. Essential findings of the survey will be described. A key element of knowledge creation is incorporation of pre-set or user-defined ways to provide insight and answers, not just data. The foreseen critical elements of a larger production scale deployment of the platform will also be presented. A particular focus will be on connecting prospective users with the information and data providers through a mediated Communities of Practice. A brief vision of future planned work will be provided.

The Minamata Convention was established to protect humans and the environment from exposure to mercury. Implementation of the Convention will rely on signatory countries adopting a series of measures to limit the amount of Hg used in commercial and industrial practices, reduce emissions and releases of Hg, and establish legislation and trade policies to further reduce and restrict the amount of Hg available on the global market (UNEP 2013).

The Conventions success at achieving its goal of protecting human health and the environment can be assessed using a variety of short-, medium- and long-term metrics that monitor compliance with stipulations set forth in the Convention (Evers et al. 2016). The Convention also recognizes the important role that monitoring environmental Hg concentrations will have in assessing its long-term effectiveness. Articles 19 and 22 of the Convention specifically address the importance of biomonitoring to evaluate spatial and temporal patterns and trends of Hg in the environment at geographically relevant scales.

Here we present data on fish mercury concentrations from South America to identify patterns and trends in environmental Hg concentrations at the continental scale. The Web of Science was used to search for published literature using the search terms mercury AND fish AND [country name]. Additional searches were conducted using Google Scholar. All literature published prior to April 2016 were collected for inclusion in the database resulting in more than 200 peer-reviewed sources. At least one peer-reviewed source was identified from each country with the exception of Guyana and Paraguay.

To-date, our database includes results from 110 peer-reviewed sources representing 24,550 individual fish tissue samples. Samples reflect environmental Hg concentrations in more than 240 sites within 100 different major waterbodies across the continent. Mercury concentrations ranging from below detection limit to 4.4 parts per million (wet weight) are reported in 403 different species representing 227 different genera. The most commonly sampled taxa include species within the Serrasalmus (piranhas), Hoplias (tigerfishes) Pseudoplatystoma (sorubim catfishes) and Odontesthes (silversides) genera. Subsequent analysis will incorporate spatial data on known point sources of mercury, ecosystem types, geology and hydrology to better understand abiotic variables that influence environmental mercury concentrations. Ultimately, these data provide a baseline for identifying biological mercury hotspots of concern for human and ecological health at the continental scale. This baseline will help prioritize future biomonitoring efforts to best evaluate the long-term effectiveness of policy decisions related to the implementation of the Minamata Convention.

The wreck site of the World War II submarine U-864 is located at 160m depth near the island of Fedje on the west coast of Norway. The submarine carried 67 tons of metallic mercury. The sediment surrounding the wreck is heavily contaminated, and steel containers holding the cargo is still considered to be present inside the wreck. In 2015 it was decided to place a counter fill in front of the bow section which has been found to rest on a slope with unstable sediments. The Norwegian institute for water research (NIVA) was assigned to be responsible for implementation of the monitoring program required during the operations in May-June 2016, where objectives were to minimize spreading of mercury during installation of the counter fill material and to estimate the total amount of mercury transported out of the working area.

The counter fill covered a 200m by 60m area immediately north of the bow section of the wreck. Five landers measuring turbidity at 1,5m and 10m above sea floor surrounded the area. Data values were transmitted every five minutes through a connected buoy and Iridium satellite, presented visually on a website, and used to generate automatic alarms (when turbidity was >10 NTU). Alarms enforced water sampling from the designated Environmental Monitoring Vessel (EMV, Siddis Mariner), and the content of mercury was analyzed at the onboard laboratory within 30 minutes to a detection limit of 0.5 ng/L. Mercury levels above the threshold would trigger construction stop and mitigation actions. Daily routine sampling was done using hydrographic instruments for salinity, temperature, and turbidity (CTD) and LISST which measures particle size distribution. Analyses of mercury and grain size was performed on particles collected in sediment traps and vibrocore samples.

Mercury concentrations in the water samples never exceeded 1.3 ng/L, much less than the predetermined limit of 50 ng/l for initiation of mitigating actions. Operations were never stopped due to risk of mercury spreading. Estimated total flux of 34 g of mercury for the entire operation was two orders of magnitude less than the predetermined acceptance criterion of 3 kg. All the data collected through mercury analyses, CTD, LISST and the online monitoring provided information contributing to a more complete understanding of the mechanisms controlling the spreading of mercury during such operations. All relevant data consistently confirmed that the particle clouds observed was dust from the counter fill material with negligible contributions from resuspended sediments.

The key objective of the Minamata Convention, a UN treaty designed to protect human health and the environment from adverse effects of mercury, is the reduction of mercury emissions. To assess the effectiveness of the convention in the future, better and independent constraints about the current mercury emissions is a premise. Existing inventories for anthropogenic mercury emissions use a bottom-up approach, where the releases are estimated using activity data and activity-specific release factors. In this study an alternative top-down approach, based on measured atmospheric mercury concentrations, is applied to quantify European mercury emissions. Gaseous elemental mercury (GEM) measurements were conducted at the remote high-altitude monitoring station Jungfraujoch, situated at 3580 m.a.s.l. in the Swiss Alps. Using a Lagrangian Particle Dispersion Model we backtracked the air masses arriving at the measurement site and established the source-receptor relationship between our measurement location and the source domain. By the means of Bayesian inversion we then were able to quantify spatially resolved European mercury emissions. We found mercury emissions to be higher by 17% compared to existing inventories. While for some areas we obtained similar or even lower estimates other regions in Greece, Spain, and Romania have shown much stronger source strength. Overall, stationary combustion, in particular in coal-fired power plants, is found to be the main responsible sector for increased emission estimates. Our top-down study, based on measurements, provides a novel and independent approach to capture mercury emissions. Our investigations help to improve and refine reported emission inventories, and can serve for continued assessment of future changes in emissions independent from bottom-up inventories.

Seabirds are powerful biomonitors of Hg contamination, and analyses of their feathers constitutes an appropriate and non-lethal method to evaluate the levels of Hg exposure in marine habitats. Albatrosses range widely within the Southern Hemiphere where they feed at high trophic levels. Therefore, they appear suitable for monitoring Hg contamination over large spatial scales. Albatrosses are known to present high but variable Hg concentrations. Such variability could be due to the combined effects of their breeding frequency, moulting patterns, and trophic ecology. To determine the significance of various factors (taxonomy, breeding frequency, feeding habitats [δ13C] and trophic level [δ15N]) on Hg concentrations, we used an exceptional data set of 455 individuals from all the species and subspecies of albatrosses breeding in the Southern Ocean and adjacent waters (n = 22). In some species, we analysed Hg, δ13C and δ15N values in different body feathers from the same individuals. As predicted, mean Hg concentrations varied widely (ten-fold) among albatrosses, from means ± SD of 3.8±1.5 to 34.6±12.5 μg.g-1 in the Indian yellow-nosed and Amsterdam albatrosses, respectively. Considering the taxonomy, Diomedea (7 taxa) had the highest Hg levels, followed by Phoebetria (2 taxa) and then Thalassarche (11 taxa). Consequently, biennial breeders (Diomedea and Phoebetria) are more contaminated by Hg than annual breeders (Thalassarche). A general decrease of Hg concentrations was found with increasing latitude (using δ13C as a proxy), with the lowest Hg concentrations in feathers moulted in Antarctic waters, and the highest in those moulted in subtropical waters. The latitudinal effect was verified at the individual level by measuring different feathers from the same wandering, light-mantled and sooty albatrosses. The study shows that albatrosses (and especially Diomedea taxa) are the seabirds with the highest Hg burdens worldwide, and that Hg exposure to predators is lower in antarctic than subantarctic and subtropical latitudes.

Regulatory assessments of the impacts of mercury (Hg) emissions reductions generally assume constant environmental conditions when evaluating and quantifying benefits of policy actions. By contrast, biomonitoring programs in the field are affected by changing biogeochemical conditions that influence temporal changes in methylmercury (MeHg) concentrations. To develop effective monitoring programs, such effects would ideally be quantified separately from the benefits associated with reductions in anthropogenic Hg emissions. This presentation will discuss and, when possible, quantify the influence of changing ecosystem conditions on MeHg production and bioaccumulation in marine food webs. Changing ocean productivity leads to shifts in phytoplankton community structure and MeHg uptake at the base of marine food webs. Global ecological modeling indicates such effects are most pronounced in the polar oceans. Intra-species variability in MeHg generally increases with age of aquatic biota. Food-web bioaccumulation modeling shows increasing seawater temperatures increases energetic requirements of different fish species leading to elevated MeHg concentrations. Changes in trophic structure by contrast can either increase of decrease fish MeHg concentrations, depending on the initial diet of the species in question. We find these shifts can easily exceed a 20-30% change in most fish species, which is within the range of declines in ecosystem Hg inputs achieved by controls on global anthropogenic emissions. We find sample sizes on the order of hundreds of fish each year are needed to have sufficient power to detect statistically significant changes in MeHg concentrations and such considerations would assist in the design of effective biomonitoring strategies supporting the Minamata Convention.

The Mercury Air Transport and Fate Research Area, a Global Mercury Partnership of the United Nations Environment Programme (UNEP FT), has the overarching objective to increase the understanding of worldwide sources of mercury (Hg) emissions and releases, their transport and fate, and impacts to humans and the environment. As part of the UNEP FT Business Plan, an important objective is to monitor, assess and report on information that can be used as the basis for assessing the environmental and public health benefits and effectiveness of global Hg reductions pursuant to the Minamata Convention on Mercury.

One policy-relevant contribution of the UNEP FT, for use by all Parties, is to develop a standardized and replicable approach for Hg biomonitoring. Our approach is an extension of the currently used UNEP Toolkit for Identification and Quantification of Mercury Releases used for the Minamata Initial Assessment process (currently conducted by over 90 countries). We now propose the development of a Toolkit for the Identification and Quantification of Biomonitoring Needs to be used by Parties for prioritizing next steps for biomonitoring related to national interests and contributing toward evaluating the effectiveness of the Minamata Convention.

The biomonitoring toolkit will use an Excel platform. Such an evidence-based approach is simple, straightforward, replicable and scalable. The premise for the toolkit on Hg biomonitoring is to respond to applicable, policy-relevant questions that will contribute to the needs of Article 22 through methods outlined in Article 19, to evaluate the effectiveness of the Minamata Convention, as well as to help Parties identify and prioritize areas of greatest concern to human health and the environment (i.e., biological Hg hotspots).

Multiple input variables that use existing data from a countrys National Mercury Profile, generated from the Minamata Initial Assessment and other publicly available sources (e.g., national and international), will provide the basis for generating policy-driven outputs, such as mapped models of air, water and biotic Hg concentrations, and when combined with landscape sensitivity variables (e.g., wetlands) biological Hg hotspots of concern can be identified and used to prioritize biomonitoring. Hazard-exposure-receptor models, based on seven input variables and the concept of their combined attributes, will allow biological Hg hotspots of concern for human and ecological health to be mapped and prioritized according to criteria set by a country. Understanding the interaction of biological Hg hotspots with sources of wild food items that are important for human communities will be of highest priority.