Contributions to Tsunami Detection by High Frequency Radar

This thesis is comprised of three separate manuscripts, detailing recent work into the detection of Tsunamis in coastal waters using High Frequency (HF) coastal radar systems. The overarching focus of the three manuscripts is the development and performance analysis of a tsunami detection technique from using synthetic radar data and synthetic, simulated tsunamis, to using recorded raw radar data combined with synthetic tsunamis, and the comparison of this work against another published method. The proposed detection algorithm tests for fluctuations in correlations of HF measurements of the sea surface, and has been named the “Time Correlation Algorithm" or TCA. The first two manuscripts are published journal papers, with the third currently being edited for submittal. Tsunami detection by High Frequency Radar beyond the continental shelf II: Extension of time correlation algorithm and validation on realistic case studies The first manuscript covers work based on a realistic case study, a simulated radar signal and two simulated tsunamis are used in order to validate the operation of the TCA modeled on an existing radar system in Tofino Canada and realistic tsunami threats to the area. The TCA is defined as correlations between radar cells connected along an intersecting wave ray, shifted in time by the long wave propagation time along the ray, c equal to √gh. The correlation is taken over a long time window, on the order of 10-15 minutes, in order to capture a meaningful portion of the tsunami wave, and average out the correlation values of smaller period waves. The correlation in the radar signal between cells is expected to be a near uniform value of one when no tsunami is present, this is due to the general lock of other naturally occurring oceanic waves or patterns with the same period of tsunamis. The first synthetic tsunami was modeled on a Mw 9.1 far-field source based in the Semidi Subduction Zone (SSZ). It was demonstrated that the TCA is able to detect extremely small currents, less than 10 cm/s, in ...