Measurement and magnitude-based system identification of tornado-associated infrasound

Abstract

Recent evidence indicates that acoustic waves at frequencies below human hearing (infrasound) are produced during tornadogenesis and continue through the life of a tornado. The currently available tornadic infrasound data remains sparse, which has prevented the identification of the fluid mechanism responsible for its production. To increase the probability of detection, this thesis presents an adaptation of fixed infrasound sensing technology to equip storm chasers that are in regular proximity to tornadoes with mobile infrasound measurement capabilities. This approach has and continues to increase the quantity of samples while also reducing the long-range propagation-related uncertainties in the measurement analysis. This thesis describes the design and deployment of the Ground-based Local INfrasound Data Acquisition (GLINDA) system - a system which includes a specialized infrasound microphone, GPS receiver, and an IMU package for data remote collection. This thesis additionally presents analyses of measurements during an EFU tornado and a significant hail event as collected by the GLINDA system. The measured signal from the EFU tornado event notably produced an elevated broadband signal between 10 and 15 Hz consistent with past observations from small tornadoes. Frequency peak identification approaches are discussed in context of the EFU tornado. Utilizing the tornado event's frequency response magnitude as the output and a selection of wind noise pressure and energy models considered for acoustic response forcing inputs, frequency response system identification approaches are utilized to develop a set of transfer function models for the tornado acoustic production mechanism.