Providing the Framework for Earthquake and Tsunami Early Warning in British Columbia, Canada: WARN, the Web-enabled Awareness Research Network

Tuesday, 15 December 2015
Poster Hall (Moscone South)
Benoit Pirenne, Andreas Rosenberger, Robert Crosby, Melissa MacArthur, Nick Allen and Mehrnaz Bayaki, Ocean Networks Canada, Victoria, BC, Canada
The main seismic hazard in western Canada is associated with the subduction of the Juan de Fuca plate under the North American continent. It threatens the major population centres of Vancouver and Victoria but also communities along the west coast of Vancouver Island, which face an additional threat from any earthquake-generated tsunami.

WARN is a project of Ocean Networks Canada (ONC), a not-for-profit organization that manages several ocean observatories on behalf of the University of Victoria.

WARN integrates an array of off-shore and on-shore sensors, both strong motion seismometers and ocean bottom pressure recorders, into a real-time network that is capable of detecting and classifying the early phases of an earthquake rupture as well as waves generated by a local or distant tsunami source. All of WARN's instruments perform complex signal processing tasks on site, on-line and in real time.

For earthquakes, WARN’s software receives event reports and waveform parameters from off-shore and on-shore strong motion seismometers and associates them with an epicentre and an average magnitude based on their respective empirical relationships. A client application computes the local impact time and expected severity in terms of Modified Mercalli Intensity from its own location in relation the reported epicentre, origin time and magnitude.

Event reports from ocean bottom pressure recorders together with earthquake parameters are used to select precomputed scenarios from a tsunami propagation model and to forecast arrival times and inundation at specific points along the West coast of Vancouver Island.

WARN's architecture is extremely flexible and can incorporate high rate GNSS based observations.