NET-VISA from Cradle to Adulthood. A Machine-Learning Tool for Seismo-Acoustic Automatic Association

Abstract

A research concept that was first presented on a poster at a Comprehensive Nuclear-Test-Ban Treaty (CTBT) conference in 2009 has resulted in a fully-fledged operational software product named NET-VISA (Network Processing Vertically Integrated Seismic Analysis), which performs the Network Processing step of the automatic processing at the International Data Centre (IDC). It has become one of the tools of the waveform analysts to review and improve the IDC standard automatic third Standard Events List (SEL3) bulletin and produce the Reviewed Events Bulletin (REB), the only global seismo-acoustic bulletin. The basic scientific concepts imbedded into NET-VISA are briefly summarized in this paper but the emphasis is on the process of adopting, developing, adapting, testing, and operationalizing the initial prototype. Extensive off-line testing has shown that one of the expected benefits of NET-VISA was the significant reduction in missed events rate compared to the current Network Processing software, Global Association (GA). NET-VISA also finds some events previously missed by manual review. Starting in July 2017, and as of January 2020, NET-VISA generates an automatic bulletin, called VSEL3, in parallel to SEL3, which is also used by analysts since January 2018. Tracing the origin of the REB events has confirmed the significant reduction in missed events when complementary VSEL3 events are reviewed in addition to the SEL3 events. If sufficient confidence is established, NET-VISA has the potential to replace GA in producing the standard automatic SEL3 bulletin.

Updated Global Reference Models of Broadband Coherent Infrasound Signals for Atmospheric Studies and Civilian Applications

“…This ANS factor is determined from the monthly reference curves and therefore should account for seasonal variability (at least on a monthly scale). There are many infrasound data quality/event identification tools, such as the Modular Utility for STatistical kNowledge Gathering (MUSTANG) (Casey et al., 2018) and the Network Processing—Vertically Integrated Seismic Analysis (NET‐VISA) (Bras et al., 2020; Mialle et al., 2019). However, the ANS factors, which quantify the deviation from the reference curves, provide an additional means to identify anomalous events or times of poor coherent signal quality.…”

“…In addition, the ANS factor could be an additional metric for event identification by the aforementioned data quality tools (e.g., MUSTANG and NET‐VISA). NET‐VISA is currently considered to be fully integrated in the processing environment of the International Data Centre (IDC) of the CTBTO (Bras et al., 2020). The ANS factors could additionally support the discrimination between infrasound clutter and events of interest in the IDC workflow with the potential to reduce the number of false event hypotheses, resulting from clutter at different stations (Mialle et al., 2019).…”

CTBT seismic monitoring using coherent and incoherent array processing

“…Methods for processing general seismic processing pipelines are changing. In the coming years, we are likely to see a replacement of the Global Association system for generating event hypotheses with the NET-VISA Machine Learning approach (Le Bras et al 2020). Elsewhere in seismology, machine learning and deep learning are making steady progress into operational signal detection and characterization (Ross et al 2018) and reliable real-time signal classification (Meier et al 2019).…”

GAN-LSTM Joint Network Applied to Seismic Array Noise Signal Recognition

“…For the global IMS network, Nimar proposed a network association method called NET-VISA based on machine learning to achieve the station's phases association and events detection. NET-VISA mainly uses a probabilistic model and the Bayesian method to find the best set of events explaining the parametric detection observations [26][27][28].…”