|Abstract Title:||Mercury risk to the environment, humans, and stakeholders in the Mackenzie River|
|Presenter Name:||Una Jermilova|
|Session:||Risk Assessment of Hg exposure to wildlife, birds and fish|
|Day and Session:||Monday 25th July - Session Four|
|Start Time:||15:30 UTC|
|Co-Authors:||Una Jermilova,Holger Hintelmann|
Abstract Information :
This study is the Canadian contribution to ARCRisk (Risk evaluation, risk reduction and risk management action plans for mercury in the Arctic ? a circumpolar management approach), a multilateral collaborative project between Canada, Russia, and Norway. The project will use Bayesian-Network Relative Risk models (BN-RRM) to identify targeted risk reduction measures for mercury (Hg) releases from key sources to land and water in the Arctic. The Mackenzie River Basin (MRB) is the Canadian study site. The MRB covers 20% of Canada?s landmass and is the fourth largest freshwater contributor to the Arctic Ocean.
Taking into account the geological diversity of the MRB, and the spatial distribution of numerous mercury sources, we intend to use an integrated BN-RRM to analyze this multi-stressor, multi-endpoint system. BN-RRM is a deterministic model that uses probability distributions to estimate risk, and by nature will incorporate uncertainty. A BN-RRM can integrate data and observations from environmental models and expert opinion into a single framework. They are visual representations of cause-and-effect relationships that can facilitate communication of complex interactions to stakeholders.
For the MRB, risk is determined for four endpoints: 1) the health and cultural impact to the Northwest Territories Indigenous Canadians; 2) the risk to health of aquatic and terrestrial organisms of the MRB; 3) the threat to recreational and commercial fisheries; and 4) risk of increased Hg load into the Arctic Ocean. Included is a review of sources and identification of key measures, aimed at recommending concrete reduction measures with high relevance for a larger part of the Arctic. Once developed, the model can use diagnostic analysis methods to predict what variable is driving the risk to our endpoints. When an influential stressor is identified, management nodes can be implemented. The model may be applied to compare management strategies and the associated uncertainties.