Tsunami Database Development in the Sunda Arc Indonesia to Support Early Warning through Artificial Intelligence Technology

Abstract

The Sunda Arc-Indonesia is very vulnerable to tsunamis. There have been at least 55 tsunamis from 416–2018. Tsunami in the Sunda Arc is classified as a near-field tsunami with an arrival time of < 30 minutes after the earthquake. Meanwhile, the BMKG issued a warning within 5 minutes after the earthquake; therefore, speed in giving warnings is very vital. Artificial intelligence is an alternative technology that can quickly predict a tsunami's height and arrival time. For developing this technology, adequate quality and quantity of data and information on tsunamis are needed. Therefore, this study was conducted to build a tsunami database based on the results of simulations and numerical modeling of multiple scenarios from hypothetical and historical earthquake sources. This study used the open-source TUNAMI F1 model. This model simulates the propagation of tsunami waves using a linear equation. This study obtained 465 hypothetical earthquake sources, 534 historical earthquake sources, and 9,990 datasets from tsunami model simulation results. Each dataset contains ten information. Based on the 8.2 magnitude earthquake scenario, the potential tsunami hazard is 3–47 m with an estimated arrival time of < 30 minutes. An earthquake < 7 Mw can trigger a tsunami, especially an earthquake that is shallow and close to the coast, even though the tsunami height is < 0.5m. This data will be used to train an artificial intelligence-based tsunami prediction system. The artificial intelligence-based tsunami prediction system is expected to be used to strengthen the Indonesia tsunami early warning system (InaTEWS).

An insight into the Application of AI in maritime and Logistics toward Sustainable Transportation

“…This can help decrease downtime and lower the chances of mechanical breakdowns while at sea. In addition, machine learning algorithms can be used to optimize vessel routing and speed profiles by utilizing real-time data on fuel [68]- [71], weather forecasts, and traffic patterns [72]- [76], leading to a decrease in fuel consumption, greenhouse gas emissions, and operating costs. A typical application spectrum is depicted in Fig.…”