Urban sensing using existing fiber-optic networks

Nature Communications, Год журнала: 2025, Номер 16(1)

Опубликована: Март 31, 2025

The analysis of urban seismic signals offers valuable insights into environments and society. Yet, accurate detection localization sources on a city-wide scale with conventional seismographic network is unavailable due to the prohibitive costs ultra-dense arrays required for imaging high-frequency anthropogenic sources. Here, we leverage existing fiber-optic networks as distributed acoustic sensing system accurately locate estimate how their intensity varies over time. By repurposing 50-kilometer telecommunication fiber an array, generate spatiotemporal maps source power (SSP) across San Jose, California. Our approach overcomes proximity limitations sensing, enabling remote generated by activities, such traffic, construction, school operations. We also show strong correlations between SSP values environmental noise levels, well various persistent features, including land use patterns demographics. This study leverages sensing. mapping Seismic Source Power, it reveals patterns, demographic trends, scalable monitoring without additional sensor deployment.

Язык: Английский

---

Characterizing Vehicle-Induced Distributed Acoustic Sensing Signals for Accurate Urban Near-Surface Imaging

Seismological Research Letters, Год журнала: 2025, Номер unknown

Опубликована: Апрель 21, 2025

Abstract Continuous seismic monitoring of the near-surface structure is crucial for urban infrastructure safety, aiding in detection sinkholes, subsidence, and other hazards. Utilizing existing telecommunication optical fibers as distributed acoustic sensing (DAS) systems offers a cost-effective method creating dense arrays areas. DAS leverages roadside fiber-optic cables to record vehicle-induced surface waves imaging. However, influence roadway vehicle characteristics, such weight, size, driving speed, on their induced resulting imaging structures poorly understood. In this study, we investigate generated by vehicles varying weights speeds provide insights into accurate efficient characterization. We first classify light, midweight, heavy based maximum amplitudes quasi-static records. Vehicles are also classified traveling speed (slow, medium, fast) using arrival times at channels. To how characteristics waves, extract phase velocity dispersion invert subsurface each class retrieving virtual shot gathers. Our results reveal that produce higher signal-to-noise ratio sevenfold increase weight can reduce uncertainties measurements from spectra up 3 × , particularly lower frequencies. Thus, data better constrain greater depths. addition, with ranging 5 30 m per second our differences curves due less pronounced than those weight. suggest judiciously selecting processing surface-wave signals certain types improve quality environments.

Язык: Английский

---

Forensic Seismic Evidence for Precursory Mobilization in Gaza Leading to the October 7 Hamas Attack

The Seismic Record, Год журнала: 2024, Номер 4(4), С. 288 - 298

Опубликована: Окт. 1, 2024

Abstract This study analyzes the seismic noise wavefield recorded before October 7 Hamas attack on Israel. Preattack activity involved large-scale mobilization, as was documented by various media sources. Opportune conditions stemming from a temporary reduction in Israeli anthropogenic enabled identification of signals generated vehicular movement Gaza at three regional stations. Seismogram analysis reveals widespread signal that abruptly emerged above nighttime levels about 20 min began. Previous Saturday mornings did not exhibit interstation correlations and amplitudes high ones observed stations minutes Statistical suggests preattack is highly anomalous unlikely to emerge chance. Tripartite array used detect surface waves, locate their sources, demarcate extent within Strip. The signal’s amplitude, frequency, spatiotemporal distribution appear be aligned with traffic emanating south-central region Strip extending toward its peripheries half-hour window preceding invasion. underscores potential utility measurements identifying terrorist mobilizations advance.

Язык: Английский

---

Deep learning based earthquake and vehicle detection algorithm

Journal of Seismology, Год журнала: 2024, Номер unknown

Опубликована: Дек. 14, 2024

Seismic recorders register vibrations from all possible sources. Even though the purpose of seismic instrument is, usually, to record ground motions coming tectonic sources, other sources such as vehicles can be recorded. In this study, a machine learning model is developed by using convolutional neural network (CNN) separate three different classes which are earthquakes, vehicles, and noises. To do that vehicle signals various accelerometric stations Italy visually detected. Together with noise earthquake information used. Inputs database 10 s long traces along their frequency content channels recorder. CNN has an accuracy rate more than 99 % for classes. understand capabilities model, earthquakes given input successfully separates case superposition vehicle, prediction in favor earthquake. Moreover, databases predicted 90 accuracy.

Язык: Английский

---

Enhancing traffic monitoring with noise-robust distributed acoustic sensing and deep learning

Journal of Applied Geophysics, Год журнала: 2024, Номер unknown, С. 105616 - 105616

Опубликована: Дек. 1, 2024

Язык: Английский

---