Three-dimensional (3D) modeling of tsunami events is intended to promote tsunami safety. However, the developed 3D modeling methods based on Computational Fluid Dynamics and photorealistic particle visualization have some weaknesses, such as not being similar to the original environment, not measuring the wave’s end point, and low image accuracy. The method for 3D modeling of tsunamis that results from this research can fulfil those weaknesses because it has advantages, such as being able to predict the end point of waves, similar to the original environment, and the height and area of inundation. In addition, the method produces more detailed and sharper spatial data. Modeling in this research is conducted using Agisoft Metashape Professional software to a produce 3D orthomosaic from pictures taken with Unmanned Aerial Vehicle (UAV) technique or drone (photogrammetry), and 3ds max software is used for wave simulation. We take a sample of an area in Cilacap, Indonesia that was impacted by the 2006 southwest coast tsunamis and may be vulnerable to future big megathrust earthquakes and tsunamis. The results could be used to provide several benefits, such as the creation of evacuation routes and the determination of appropriate locations for building shelters.

Mapping methods for loosened rock mass in mountainous areas are useful for risk management of landslide disasters. Depending on the type of aircraft and sensor, there are several different aerial electromagnetic measurement methods for estimating subsurface structures. Helicopter-borne electromagnetic methods are commonly used. Recently, unmanned aerial vehicles (drones) have been used. By understanding the characteristics of each method, it is possible to choose a suitable method for the target of observation. In this study, resistivity from the frequency-domain helicopter-borne electromagnetic (HEM) method and resistivity from the time-domain drone-grounded electrical-source airborne transient electromagnetic (D-GREATEM) method were compared to estimate loosened zones in mountainous areas. The resistivity cross-sectional profiles were largely similar, but differences were observed near the surface in some zones. The comparative analysis of both methods with outcrop observations revealed that D-GREATEM resistivity data can detect both loosened rock mass from the surface to an approximately 30 m depth located above the groundwater and saturated rock mass. It is because D-GREATEM resistivity was obtained by assuming five layers from the surface to a depth of 40 m. This indicates that D-GREATEM is suitable for estimating near-surface loosened rock mass distribution in the valleys. However, D-GREATEM has a limited observation range. Therefore, it was concluded that the D-GREATEM method is suitable for a detailed and localized estimation of landslide susceptibility near the surface, whereas the HEM method is suitable for wide-area analysis.

Catastrophic earthquake-induced landslides can occur on slopes composed of loosened and fractured rock masses. Although it is impossible to prevent such landslides, estimation of the susceptibility of slopes to earthquake-induced landslides is useful for risk management. An index of susceptibility to earthquake-induced landslides (ISEL) was developed by using helicopter-borne electromagnetic resistivity data. However, the ISEL has not yet been validated through the analysis of pre-earthquake data. In this study, ISEL values were estimated from resistivity and digital elevation data obtained in 2013 around Mt. Aso, Kyushu, before the 2016 Kumamoto earthquake. Although most of the landslides around Mt. Aso during the 2016 Kumamoto earthquake were mass movements of volcanic tephra layers, some of them occurred on loosened rock masses. Landslide susceptible areas at loosened rock masses are the target for ISEL value estimation. Our results validated the effectiveness of the ISEL as a predictor of earthquake-induced rock mass landslides.

To decrease human and economic damage owing to earthquakes, it is necessary to discover signals preceding earthquakes. We focus on the concept of “early warning signals” developed in bifurcation analysis, in which an increase in the variances of variables precedes its transition. If we can treat earthquakes as one of the transition phenomena that moves from one state to the other state, this concept is useful for detecting earthquakes before they start. We develop a covariance matrix from multi-channel time series data observed by an observatory on the seafloor and calculate the first eigenvalue and corresponding eigenstate of the matrix. By comparing the time dependence of the eigenstate to some past earthquakes, it is shown that the contribution from specific observational channels to the eigenstate increases before earthquakes, and there is a case in which the eigenvalue increases as predicted in early warning signals. This result suggests the first eigenvalue and eigenstate of multi-channel data are useful to identify signals preceding earthquakes.