An Analysis of Tornado-Emitted Infrasound during the VORTEX-SE Field Campaign

Throughout year two of the Verifications of the Origins of Rotation in Tornadoes Experiment – Southeast (VORTEX-SE) field campaign, The University of Alabama in Huntsville and The University of Mississippi deployed multiple sensor arrays across Northern Alabama capable of observing acoustic signatures in the infrasound (0-20 Hz) acoustic range. These arrays were placed in an effort to detect infrasonic emissions hypothesized to be the result of tornadic activity, and use the data collected to determine the efficacy of such a network in the identification and tracking of tornadic storms. In addition to the presence of the aforementioned infrasound observing arrays, a network of meteorological instrumentation was also present (including atmospheric profiling systems, radiosondes, polarimetric Doppler radars, and airborne Doppler radars) in order to characterize the environment and the storms in which these emissions may occur.

On April 22, 2017, a quasi-linear convective system passed through North Alabama producing three confirmed tornadoes in addition to numerous damaging wind reports within the VORTEX-SE domain. Of the tornadoes to impact the region, two were rated as EF0 and one was rated as EF1. Processing of the data recorded by the infrasound collection arrays revealed that infrasonic emissions were present throughout the event, with fluctuations in intensity seeming to be correlated with tornadic activity. In order to spatially correlate these emissions with the storms present, sensor elements within the arrays were placed in geometries such that bearing estimates of the acoustic sources were able to be calculated at ranges of up to 60 km. Analysis of these bearing estimates revealed a trend in which storms often depicted infrasonic emissions during the time preceding tornadogenesis, followed by a rapid increase in emission intensity as tornadic activity commenced, and a subsequent weakening of the acoustic signal after tornado dissipation. Furthermore, the track of the EF1 tornado brought it through two arrays simultaneously, which provided data that was able to be used to derive a tornado track derived purely from the location of the location of the infrasonic emission.

This presentation aims to showcase the results from initial analyses of the datasets collected, with an emphasis placed on evaluating the correlation between tornado damage and intensity (of both the tornado and parent mesoscale circulation) and the observed infrasonic signatures by utilizing ancillary data collected by ground-truth post event surveys and polarimetric Doppler radar. Additionally, the utility of infrasound observations as a tornado detecting tool will be also discussed, as well as addressing future avenues of research for this topic.