Integrating Geophysical and Inupiat Knowledge on Alaska Shorefast Ice Stability Using Fault Tree Analysis
Matthew L. Druckenmiller1, Hajo Eicken2
1Geophysical Institute, University of Alaska Fairbanks, 903 Koyukuk Drive, Fairbanks, AK, 99775-7320, USA, Phone 907-474-1156, mldruckenmiller [at] alaska [dot] edu
2University of Alaska Fairbanks, Fairbanks, AK
As significant changes are observed in the seasonal ice zone north of Alaska, such as later freeze-up and less abundant multi-year ice over the last 30 years, understanding the state of shorefast ice as a safe and quasi-persistent platform for human activity requires further investigation and synthesis of past and present knowledge. By explicitly defining specific modes of failure for shorefast ice in the Chukchi Sea, we employ fault tree analysis (FTA), which relies on deductive and Boolean logic, to map causal relationships for both quantitative (probability) and qualitative analysis of failure. Here, deterministic paths through the fault tree to eventual 'failure' represent different types of behavior of the local shorefast ice environment as the result of destabilizing forces acting on the ice. FTA allows us to investigate the interaction of subsystems across scales, integrate different types of knowledge and observations, and organize prior qualitative and quantitative studies. Traditionally, FTA, an approach based in complex systems engineering, is used to determine the most likely cause of failure and, accordingly, pinpoint system components that need to be improved or replaced. However, as we analyze a natural system, over which man has little or no control, FTA identifies areas requiring improved environmental monitoring to better assess failure risks and thus shorefast ice stability. As a subjective methodology, the analysis relies on the knowledge of those constructing the fault tree. Collaborating with local hunters in Barrow, Alaska, we attempt to integrate geophysical and Inupiat expert knowledge to model diverse perspectives on stability and identify important variables (e.g., ice thickness, type, and morphology) and processes (e.g., atmospheric and oceanic) that lead to failure. This integrated approach may inform observing systems that aim to provide useful information to a range of ice users and stakeholders, such as indigenous marine mammal hunting communities and industrial operations.