Impact of Climate Change on Hurricane Hazards and the Resilience and Sustainability of Infrastructures

点击翻译按钮获取中文摘要

Many parts of Canada have experienced hurricanes in the past and will again. Hurricane Hazel caused 81 fatalities in Toronto in 1954 and Typhoon Freda claimed six lives in southwest British Columbia in 1962. One of the most destructive hurricanes to hit a Canadian city was Juan in 2003. It struck Nova Scotia and Prince Edward Island as a Category 2 hurricane, causing major damage, especially within the urban core of Halifax. The storm killed 8 people and caused more than $200 million in damage. It was the region’s most powerful hurricane since 1873. Hurricane Sandy was also unusual in more ways than its size. One factor contributing to the storm strength was abnormally warm sea surface temperatures off the east coast, which is probably linked to global warming. “Hurricane Sandy’s 4m storm surge may be an example of what will, by midcentury, be the “new norm on the Eastern Seaboard.” The long-term objective of the DG is to identify the underlying mechanisms and define the factors that affect hurricane and extra-tropical storm hazards along the East Coast to improve design and resilience for a wide range of infrastructure (e.g. offshore and coastal facilities). Oil exploration is predicted to increase significantly on the East Coast and is poised to become an important future major economic zone for Canada. The short-term objectives (next 5 years) are to: 1: develop a model that can represent site specific hazards for coastal and offshore locations along the East Coast of Canada for the recurrence and severity of storms; 2: predict effects associated with these storms in terms of wind, current, wave and storm surge, 3: estimate the impacts of climate change relative to impacts on storms and water level, and 4: develop procedures for the propagation uncertainty in the models and obtain estimates of risks for onshore and offshore facilities. First hurricane hazard models will be formulated for site specific hazards as is currently done for the building codes. These models will be based on non-homogenous Poisson processes for track location and frequency and spatially dependent probability distribution functions for intensity. The models will then be modified to account for meteorological and oceanographic features known to affect the intensity and trajectories of hurricanes. These historical and predicted characteristics of these features will be analyzed to incorporate potential effects of climate change. Surrogate models for hurricane effects (wave, currents and surge) will be developed using Joint Probability Methods and Support Vector Machines to allow the analysis of several scenarios as well to propagate uncertainty in the models for the prediction of hazards and risks. The DG proposal will directly support 2 PhD, 3 MEng thesis and 4 undergraduate students in an area of high demand and of great importance for the future of our community within a multidisciplinary team and engineers and scientists.