Potential of deep predictive coding networks for spatiotemporal tsunami wavefield prediction

Fauzi, Ardiansyah; Mizutani, Norimi

doi:10.1186/s40562-020-00169-1

Cited by 3 publications

(1 citation statement)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, deep learning has recently been introduced for tsunami data assimilation. Fauzi and Mizutani (2020) used a deep predictive coding network along with data assimilation to forecast the tsunami wavefield in real time. They first developed a tsunami propagation database from predefined scenarios to train a predictive coding network.…”

Section: Improvement On Assimilation Speedmentioning

confidence: 99%

Review on Recent Progress in Near-Field Tsunami Forecasting Using Offshore Tsunami Measurements: Source Inversion and Data Assimilation

Wang

Tsushima

Satake

et al. 2021

Pure Appl. Geophys.

View full text Add to dashboard Cite

Traditional tsunami forecasting methods are based on seismic observations. The development of offshore observational devices makes it possible to utilize offshore tsunami measurements for providing early warnings. This paper reviews the tsunami data-oriented forecasting methods of two main types: tsunami source inversion and tsunami data assimilation. In the first type, offshore tsunami measurements are used to enhance and speed up the traditional tsunami source inversion methods. The effectiveness and uncertainties of the tsunami source estimation are discussed. In the second type, tsunami data assimilation, the wavefield is directly reconstructed and forecasted without considering the source. We review the techniques that are adopted to improve the speed and accuracy of data assimilation and to determine the optimal design for an ocean bottom pressure gauge (OBPG) network. It is important for tsunami data-oriented forecasting methods to accurately extract tsunami signal in real time, particularly when the OBPG is located inside the source area.

show abstract

Section: Improvement On Assimilation Speedmentioning

confidence: 99%

Review on Recent Progress in Near-Field Tsunami Forecasting Using Offshore Tsunami Measurements: Source Inversion and Data Assimilation

Wang

Tsushima

Satake

et al. 2021

Pure Appl. Geophys.

View full text Add to dashboard Cite

show abstract

Geophysical model generation with generative adversarial networks

et al. 2022

View full text Add to dashboard Cite

With the rapid development of deep learning technologies, data-driven methods have become one of the main research focuses in geophysical inversion. Applications of various neural network architectures to the inversion of seismic, electromagnetic, gravity and other types of data confirm the potential of these methods in real-time parameter estimation without dependence on the starting subsurface model. At the same time, deep learning methods require large training datasets which are often difficult to acquire. In this paper, we present a generator of 2D subsurface models based on deep generative adversarial networks. Several networks are trained separately on realistic density and stratigraphy models to reach a sufficient degree of accuracy in generation of new highly detailed and varied models in real-time. This allows for creation of large synthetic training datasets in a cost-effective manner, thus facilitating the development of better deep learning algorithms for real-time inversion and interpretation.

show abstract

Synthetic analysis of the efficacy of the S-net system in tsunami forecasting

Mulia

Satake

2021

Earth Planets Space

View full text Add to dashboard Cite

The Seafloor Observation Network for Earthquakes and Tsunamis along the Japan Trench (S-net) is presently the world’s largest network of ocean bottom pressure sensors for real-time tsunami monitoring. This paper analyzes the efficacy of such a vast system in tsunami forecasting through exhaustive synthetic experiments. We consider 1500 hypothetical tsunami scenarios from megathrust earthquakes with magnitudes ranging from Mw 7.7–9.1. We employ a stochastic slip model to emulate heterogeneous slip patterns on specified 240 subfaults over the plate interface of the Japan Trench subduction zone and its vicinity. Subsequently, the associated tsunamis in terms of maximum coastal tsunami heights are evaluated along the 50-m isobath by means of a Green’s function summation. To produce tsunami forecasts, we utilize a tsunami inversion from virtually observed waveforms at the S-net stations. Remarkably, forecasts accuracy of approximately 99% can be achieved using tsunami data within an interval of 3 to 5 min after the earthquake (2-min length), owing to the exceedingly dense observation points. Additionally, we apply an optimization technique to determine the optimal combination of stations with respect to earthquake magnitudes. The results show that the minimum requisite number of stations to maintain the accuracy attained by the existing network configuration decreases from 130 to 90 when the earthquake size increases from Mw 7.7 to 9.1.

show abstract

Potential of deep predictive coding networks for spatiotemporal tsunami wavefield prediction

Cited by 3 publications

References 35 publications

Review on Recent Progress in Near-Field Tsunami Forecasting Using Offshore Tsunami Measurements: Source Inversion and Data Assimilation

Review on Recent Progress in Near-Field Tsunami Forecasting Using Offshore Tsunami Measurements: Source Inversion and Data Assimilation

Geophysical model generation with generative adversarial networks

Synthetic analysis of the efficacy of the S-net system in tsunami forecasting

Contact Info

Product

Resources

About