Abstract:Since climate change extends the ice-free period in high-latitude areas, the Arctic faces increasing maritime traffic and petroleum extraction activity. A growing number of oil drilling and shipping operations in areas experiencing harsh weather and extensive sea ice raises concerns about oil-related accidents. CEARCTIC and CEPOLAR consortiums (2013-2021) have developed solutions for safer arctic shipping and methods to assess the risks related to these activities. The role of my team has been mainly in the latter (https://www2.helsinki.fi/en/researchgroups/environmental-and-ecological-statistics/oil-spill-risk-analysis-in-the-arctic).
Since Arctic regions are remote, data on Arctic ecosystems and environment are often totally missing and when they exist, they are scarse, messy and distributed in several locations ranging from freely open to closed. This poses challenges to estimating the risks of human operations. Hence, we had to get the most out of heteregeneous and partly unreliable data sets, expert knowledge and prior information. In this talk I will summarize the key methods, results and lessons learned from our work. I will also discuss and draw links to statistics and machine learning research and how they can help in these tasks.
Speakers: Jarno Vanhatalo
Jarno Vanhatalo is an assistant professor of statistics at the University of Helsinki working both in the Department of Mathematics and Statistics (Faculty of Science) and Organismal and Evolutionary Biology Research Program (Faculty of Biological and Environmental Sciences). He leads Environmental and Ecological Statistics group and is one of the PIs in the Research Center for Ecological Change. His research interests span from Bayesian statistics and machine learning to environmental and ecological sciences.
Affiliation: University of Helsinki