Quantifying the Influence of Short Timescale Atmospheric Dynamics on Infrasound Propagation

Current atmospheric specification systems are restricted in their ability to resolve short timescale atmospheric dynamics. This undocumented variability has the potential to impact infrasound propagation and diminish the efficacy of signal processing algorithms. Previous studies have revealed the effectiveness of utilizing infrasound observations as a means of improving our knowledge of the atmospheric state in real-time. This study is the first to explore the effect of these fluctuations on infrasound propagation. During the summer of 2016 and the winter of 2021, repetitive signals recurrent over 20-second time intervals were detected by local and regional infrasound sensors in Oklahoma. It was determined that the signals emanated from discrete pit locations at two distinct blast sites on McAlester Army Ammunitions Plant in McAlester, Oklahoma. Subsequent analysis revealed that these detections constitute a viable dataset to quantify time-varying effects on infrasound where atmospheric models lack the capability to do so. In this study, we evaluate the effectiveness of array processing algorithms and cross-correlation techniques to quantify the effect of short timescale atmospheric variability on tropospheric and stratospheric infrasound captured at the local and regional arrays, respectively. Our objective is to reveal trends in the atmospheric state that fall below current model temporal resolutions and assess the impact of this variability on the effectiveness of infrasound as a component of explosion monitoring for international treaty verification. Propagation modeling exhibited greater complexity in the day-to-day variability in the troposphere than the stratosphere. Frequency wavenumber analysis is applied to the data in order to examine potential systematic differences between primary arrivals from both blast sites. No systematic deviations are observed, however daily variability is evident. A sliding window is stepped through the time-series data to investigate the stability of back azimuth estimates as a function of time. Estimates remain consistently stable for both tropospheric and stratospheric infrasound, highlighting the efficacy of infrasound monitoring. Cross-correlation analysis revealed disparate trends in the pattern of quantified similarity for tropospheric and stratospheric observations. Results indicate a more complex tropospheric propagation environment within sub-hour time intervals than the stratosphere