Stephen J. Arrowsmith

Automatic infrasound detection and location of sources in the western United States

A catalog that characterizes sources of regional infrasound observed in the western U.S. (1 November 2010 to 31 October 2012) is produced. Data from nine University of Utah Seismograph Stations infrasonic arrays are supplemented by three additional arrays in Nevada, operated by Southern Methodist University. The detection procedure using an adaptive F-detector provides input into the Bayesian Infrasonic Source Location procedure. The catalog consists of 1510 events with indication of repeated events from many locations such as Dugway Testing Ground, Utah Test and Training Range, and New Bomb. We analyzed the relationship between seasonal variations in the event locations and wind conditions using the Ground-to-Space specifications based on publicly available operational numerical weather prediction data analysis products supplement by empirical models above 80 km. There is significant commonality between this studys bulletin and the Western United States Infrasonic Catalog published by Walker et al. (2011). A previous study utilized infrasound signals detected on the USArray Transportable Array seismic stations (2007–2008). Both results document the vast majority of events that occur during working hours, suggesting a human cause. To illustrate the utility of the event bulletin for exploring atmospheric dynamics, propagation paths of an event detected during the equinox period, when the stratospheric wind is low, were generated using a ray-tracing algorithm. We found that the observations contain stratospheric arrivals, not predicted by ray theory, possibly due to gravity waves increasing the effective jet speed.

On infrasound generated by wind farms and its propagation in low‐altitude tropospheric waveguides

Infrasound from a 60-turbine wind farm was found to propagate to distances up to 90 km under nighttime atmospheric conditions. Four infrasound sensor arrays were deployed in central New Mexico in February 2014; three of these arrays captured infrasound from a large wind farm. The arrays were in a linear configuration oriented southeast with 13, 54, 90, and 126 km radial distance and azimuths of 166°, 119°, 113°, and 111° from the 60 1.6-MW turbine Red Mesa Wind Farm (RMWF), Laguna Pueblo, New Mexico, USA. Peaks at a fundamental frequency slightly below 0.9 Hz and its harmonics characterize the spectrum of the detected infrasound. The generation of this signal is linked to the interaction of the blades, flow gradients, and the supporting tower. The production of wind-farm sound, its propagation, and detection at long distances can be related to the characteristics of the atmospheric boundary layer. First, under stable conditions, mostly occurring at night, winds are highly stratified, which enhances the production of thickness sound (TS) and the modulation of other higher-frequency wind turbine sounds. Second, nocturnal atmospheric conditions can create low-altitude waveguides (with altitudes on the order of 100s of meters) allowing long distance propagation. Third, night and early morning hours are characterized by reduced background atmospheric noise that enhances signal detectability. This work describes the characteristics of the infrasound from a quasi-continuous source with the potential for long-range propagation that could be used to monitor the lower part of the atmospheric boundary layer.

Detection of regional infrasound signals using array data: Testing, tuning, and physical interpretation

This work quantifies the physical characteristics of infrasound signal and noise, assesses their temporal variations, and determines the degree to which these effects can be predicted by time-varying atmospheric models to estimate array and network performance. An automated detector that accounts for both correlated and uncorrelated noise is applied to infrasound data from three seismo-acoustic arrays in South Korea (BRDAR, CHNAR, and KSGAR), cooperatively operated by Korea Institute of Geoscience and Mineral Resources (KIGAM) and Southern Methodist University (SMU). Arrays located on an island and near the coast have higher noise power, consistent with both higher wind speeds and seasonably variable ocean wave contributions. On the basis of the adaptive F-detector quantification of time variable environmental effects, the time-dependent scaling variable is shown to be dependent on both weather conditions and local site effects. Significant seasonal variations in infrasound detections including daily time of occurrence, detection numbers, and phase velocity/azimuth estimates are documented. These time-dependent effects are strongly correlated with atmospheric winds and temperatures and are predicted by available atmospheric specifications. This suggests that commonly available atmospheric specifications can be used to predict both station and network detection performance, and an appropriate forward model improves location capabilities as a function of time.

Multivariate acoustic detection of small explosions using Fisher's combined probability test

A methodology for the combined acoustic detection and discrimination of explosions, which uses three discriminants, is developed for the purpose of identifying weak explosion signals embedded in complex background noise. By utilizing physical models for simple explosions that are formulated as statistical hypothesis tests, the detection/discrimination approach does not require a model for the background noise, which can be highly complex and variable in practice. Fishers Combined Probability Test is used to combine the p-values from all multivariate discriminants. This framework is applied to acoustic data from a 400 g explosion conducted at Los Alamos National Laboratory.

Analysis and modeling of infrasound from a four-stage rocket launch

Infrasound from a four-stage sounding rocket was recorded by several arrays within 100 km of the launch pad. Propagation modeling methods have been applied to the known trajectory to predict infrasonic signals at the ground in order to identify what information might be obtained from such observations. There is good agreement between modeled and observed back azimuths, and predicted arrival times for motor ignition signals match those observed. The signal due to the high-altitude stage ignition is found to be low amplitude, despite predictions of weak attenuation. This lack of signal is possibly due to inefficient aeroacoustic coupling in the rarefied upper atmosphere.

Development of a matched filter detector for acoustic signals at local distances from small explosions

A method for acoustic detection of small explosions at local distances is presented combining a matched filter with a p-value representing the conditional probability of detection. Because the physics of signal generation and propagation for small, locally recorded acoustic signals from small explosions is well understood, the single hypothesis to be tested is a signal corrupted by additive noise. A simple analytical signal representation is used where a known signal is assumed with parameters to be determined. The advantage of the approach is that the detector can be combined with other detectors that measure different signal characteristics all under the same unifying hypothesis.

Geoacoustic Observations on Drifting Balloon-Borne Sensors

Infrasound microphones on free flying balloons experience very little wind noise, can cross regions that lack ground station coverage, and may capture signals that seldom reach the Earth’s surface. Despite the promise of this technique, until recently very few studies had been performed on balloon-borne acoustic sensors. We summarize the history of free flying infrasound stations from the late 1940s to 2014 and report on results from a series of studies spanning 2014–2016. These include the first efforts to record infrasound in the stratosphere in half a century, the presence of a persistent ocean microbarom peak that is not always visible on the ground, and the detection of distant ground explosions. We discuss the unique operational aspects of deploying infrasound sensors on free flying balloons, the types of signals detected at altitude, and the changes to sensor response with height. Finally, we outline the applications of free flying infrasound sensing systems, including treaty verification, bolide detection, upper atmosphere monitoring, and seismoacoustic exploration of the planet Venus.

Pickless event detection and location: The waveform correlation event detection system (WCEDS) revisited

The standard seismic explosion‐monitoring paradigm is based on a sparse, spatially aliased network of stations to monitor either the whole Earth or a region of interest. Under this paradigm, state‐of‐the‐art event‐detection methods are based on seismic phase picks, which are associated at multiple stations and located using 3D Earth models. Here, we revisit a concept for event‐detection that does not require phase picks or 3D models and fuses detection and association into a single algorithm. Our pickless event detector exploits existing catalog and waveform data to build an empirical stack of the full regional seismic wavefield, which is subsequently used to detect and locate events at a network level using correlation techniques. We apply our detector to seismic data from Utah and evaluate our results by comparing them with the earthquake catalog published by the University of Utah Seismograph Stations. The results demonstrate that our pickless detector is a viable alternative technique for detecting events that likely requires less analyst overhead than do the existing methods.

Extracting changes in air temperature using acoustic coda phase delays

Blast waves produced by 60 high-explosive detonations were recorded at short distances (few hundreds of meters); the corresponding waveforms show charge-configuration independent coda-like features (i.e., similar shapes, amplitudes, and phases) lasting several seconds. These features are modeled as reflected and/or scattered waves by acoustic reflectors/scatters surrounding the explosions. Using explosion pairs, relative coda phase delays are extracted and modeled as changes in sound speed due to changes in air temperature. Measurements from nearby weather towers are used for validation.

Infrasound Signal Detection: Re-examining the Component Parts that Makeup Detection Algorithms

Detecting a Signal Of Interest (SOI) is the first step in many applications of infrasound monitoring. This intuitively simple task is defined as separating out signals from background noise on the basis of the characteristics of observed data; it is, however, deceptively complex. The problem of detecting signals requires multiple processes that are divisible at their highest level into several fundamental tasks. These tasks include (1) defining models for SOIs and noise that properly fit the observations, (2) finding SOIs amongst noise, and (3) estimating parameters of the SOI (e.g., Direction Of Arrival (DOA), Signal-to-Noise Ratio (SNR) and confidence intervals) that can be used for signal characterization. Each of these components involves multiple subcomponents. Here, we explore these three components by examining current infrasound detection algorithms and the assumptions that are made for their operation and exploring and discussing alternative approaches to advance the performance and efficiency of detection operations. This chapter does not address new statistical methods but does offer some insights into the detection problem that may motivate further research.