Assessing the Repeatability of Acoustic Signals Originating from Closely Spaced Events

Abstract:

The Earth's atmosphere is dynamic over time scales ranging from seconds to centuries. This leads to changes in acoustic propagation, implying that identical sources can have very different signal manifestations if they occur over a sufficiently long time frame. Recent studies have shown that even 90 minutes is enough to result in major amplitude and waveform changes if the source and the sensors are separated by about 10 km. We sought to understand the magnitude of changes that occurred if the sources occurred tens of seconds, not tens of minutes apart. We present acoustic signals from two ground chemical explosion campaigns in Socorro, New Mexico, observed at local (<25 km) and regional (>150 km) ranges. We show that minute scale atmospheric dynamics exert a major impact on acoustic waveforms, especially at ranges greater than 10 km. Using full waveform modeling, we account for local topography that explains some of the differences in the observed signals. Our results suggest that deterministic acoustic signal prediction is unreliable beyond 10 km, and that a stochastic approach may be more appropriate. SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.