Item

Abstract

Early warning systems should be available for volcanic tsunamis. Present tsunami warning systems have been specialized for earthquake-generated tsunamis, and it remains challenging to rapidly evaluate the tsunami potential accompanied by a volcanic eruption and/or volcanic sector collapse. By targeting the volcanic sector collapse of Anak Krakatau in 2018, which generated a large tsunami as a case event, we develop a real-time forecasting method for tsunamis accompanied by such eruptions in Indonesia. The proposed forecasting method is based on the utilization of observation stations near the source area and a pre-computed database. In this study, six virtual observation stations were placed in the vicinity of the volcano. Then, a pre-computed waveforms database was constructed from various collapse scenarios using a numerical simulation of sector collapse and tsunami propagation. In the real application, tsunami forecasting computation will be conducted using the initial conditions which are selected from the best combination of scenarios inside the database through waveforms fitting at six observation stations. The reliability of the method was examined using three hypothetical collapse scenarios of Anak Krakatau assuming three different sliding directions. Based on numerical experiments, the forecasted tsunami along the coast of Java and Sumatra by our method resulted in satisfactory performance. Our results indicate that the combination of a pre-computed database and the existence of observation stations near the source area was able to produce appropriate tsunami forecasting for the coastal area. We conclude that our method has a reasonable predictive skill for volcanic-induced tsunamis in Anak Krakatau.