Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Kae Tsunematsu has a total of 62 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Volcanic ballistic projectiles in supersonic jets: Model testing against experimental data

    by Nils Steinau, Kae Tsunematsu, Kiyonobu Ohtani, Kazuya Seo, Tsubasa Okaze, Akuto Kaneko

  • Open Access

    Volcanic ballistic projectiles in supersonic jets: shock tube experiments with glass beads

    by Nils Steinau, Kae Tsunematsu, Kiyonobu Ohtani, Kazuya Seo, Akuto Kaneko

    To investigate volcanic ballistic projectile (VBP) interactions with supersonic volcanic jets, shock tube experiments using spherical glass beads as VBP surrogates were conducted. Trajectory data and Schlieren/shadowgraph images of the jets were obtained. Comparisons of experimental and volcanic data indicate that the experiments best represent Strombolian eruptions, with some overlap with Vulcanian and Plinian eruptions. Analysis of the jet images and trajectories revealed the following key findings: (1) the VBP exit velocity at the vent increases with the reservoir-to-ambient pressure ratio and initial particle depth but decreases with the particle size and (2) the VBPs are accelerated by the supersonic jet over a distance exceeding the Mach disk distance before a constant maximum velocity is reached. No trend between the acceleration distance and the pressure ratio was observed. The maximum VBP velocity increases with increasing pressure ratio and decreases with increasing particle size. While the literature provides data from field observations and numerical modeling for VBP acceleration above the vent, experimental studies have thus far predicted that maximum VBP velocities occur at the vent. In contrast, the results of this experimental study reveal that the VBP velocity can increase after the VBPs exit the vent because of acceleration by the flow field of supersonic volcanic jets. The maximum velocity that the VBPs can reach affects their travel distance, making it an essential parameter for VBP hazard assessment. Graphical Abstract

  • Open Access

    Magma feeding system of the 1893 CE Meiji eruption of Azumayama Volcano, NE Japan

    by Masao Ban, Shun Kanno, Motohiro Sato, Takumi Imura, Kae Tsunematsu, Takeshi Hasegawa, Tsukasa Ohba

    Azumayama Volcano is one of the most active stratovolcanoes in northeastern Japan, with a ca. 1.2 million-year eruption history. Its hazard status was raised to volcanic alert level 2 (restricted access to the crater) in 2018 and 2019 due to an increase in volcanic activity, including the number of volcanic earthquakes. The latest magmatic eruption of Azumayama was the 1893 CE. This was preceded by phreatic eruptions, and was followed by Vulcanian eruptions. The products of the magmatic eruptions are the most suitable candidates for revealing the magma feeding system and the magmatic processes beneath the volcano. This information is crucial in assessing the hazard of future eruptions, and the recent unrest makes it an urgent issue. The samples were collected from juvenile bombs in the vulcanian fall deposits of the 1893 CE Meiji eruption. The bombs were divided into darker gray and lighter gray groups (D-group; 58–59 and L-group; 60–63 wt% SiO 2 ), both of which have olivine-orthopyroxene-clinopyroxene-plagioclase phenocrysts. Whole rock compositions of all samples plot along the same linear trend on SiO 2 variation diagrams. The phenocrysts showed compositional variations (e.g., 60–83 Mg# for pyroxenes and An 51–95 for plagioclases). Based on the compositional and textural features, it was deduced that low-Mg pyroxenes and low-An plagioclases, Fo-rich olivine and Cr-spinel, and intermediate-Mg pyroxenes and intermediate-An plagioclase might be from felsic end-member, mafic end-member, and mixed magmas, respectively. Using pyroxene and olivine-spinel geothermometers, and the rhyolite-MELTS program, the felsic magma was estimated to be dacitic (64–66 wt% SiO 2 ), 860–900 °C, and stored at ca. 5–10 km, and the mafic magma was estimated to be basaltic (ca. 49 wt% SiO 2 and 1138–1166 °C), precipitating Fo-rich olivine and spinel at ca. 14–20 km in depth. Based on the petrological data combined with the geophysical data, the mafic magmas might have ascended from a deeper reservoir (> 20 km in depth). The shallow dacitic crystal mush body might have been repeatedly injected by the mafic magma, which resulted in the remobilization of the crystal mush body and the formation of mixed magmas. The intermediate magmas formed by the mixing would have been vertically heterogeneous. These three magmas were incompletely mixed during the eruption. The timescale from the injection to the eruption, estimated by diffusion modelling in the pyroxene phenocrysts, was several months to years. These results are important for assessing the hazard of future eruptions; for example, they may provide important information for geophysical monitoring and the development of evacuation plans. Graphical abstract


  • Open Access

    Quantitative Evaluation of Probabilistic Hazard Mapping With Polynomial Chaos Quadrature and Its Practical Application

    by T. Tanabe, K. Tsunematsu, K. Nishimura

    Snow avalanches pose a significant threat to settlements and their inhabitants. Consequently, hazard maps that delineate avalanche runout areas serve as valuable tools for mitigating their destructive impact. Dynamic models have been used to visualize areas affected by avalanches; however, these models require uncertain inputs. This study develops probabilistic hazard maps by quantifying uncertain input variables through probability density functions. These maps represent the probability of model outputs, such as maximum flow thickness, exceeding specific thresholds, allowing for more quantitative hazard assessments. Three uncertainty quantification methods—Monte Carlo, Latin hypercube sampling, and polynomial chaos quadrature (PCQ)—are employed to generate probabilistic hazard maps for snow avalanches. These maps are compared with a reference hazard map created using parameter sets that cover the entire parameter space. Among the three methods, PCQ yields the most accurate results for a given number of simulations, assuming a uniform distribution for each input. The optimal PCQ settings, which deliver superior results with fewer simulations, are then determined. Additionally, a PCQ application is proposed to generate hazard maps based on non‐uniform input distributions without requiring extra simulations. This approach reduces the computational cost associated with creating hazard maps for non‐uniform distributions if PCQ has already been applied to a uniform case. The application generates two types of probabilistic hazard maps: one considering all potential parameter ranges during the snow season using uniform distributions, and another reflecting non‐uniform distributions that account for uncertainty in near‐term current snow cover conditions.


  • Open Access

    Quantitative evaluation of probabilistic hazard mapping with polynomial chaos quadrature and its practical application

    by Takahiro Tanabe, Kae Tsunematsu, Kouichi Nishimura

    Snow avalanches pose a significant threat to settlements and their inhabitants. Consequently, hazard maps are a valuable tool for mitigating their impact. Dynamic models have been used to visualize areas affected by avalanches; however, these models require uncertain inputs. This study develops probabilistic hazard maps by quantifying uncertain input variables through probability density functions. These maps represent the probability of model outputs, such as maximum flow thickness, exceeding specific thresholds, allowing for more quantitative hazard assessments. Three uncertainty quantification methods—Monte Carlo, Latin hypercube sampling, and polynomial chaos quadrature (PCQ)—are employed to generate probabilistic hazard maps for snow avalanches. These maps are compared with a reference hazard map created using parameter sets that cover the entire parameter space. Among the three methods, PCQ yields the most accurate results for a given number of simulations, assuming a uniform distribution for each input. The optimal PCQ settings, which deliver superior results with fewer simulations, are then determined. Additionally, a PCQ application is proposed to generate hazard maps based on non-uniform input distributions without requiring extra simulations. This approach reduces the computational cost associated with creating hazard maps for non-uniform distributions if PCQ has already been applied to a uniform case. The application generates two types of probabilistic hazard maps: one considering all potential parameter ranges during the snow season using uniform distributions, and another reflecting non-uniform distributions that account for uncertainty near-term real-world snow cover conditions.

  • Open Access

    Seismic location and tracking of snow avalanches and slush flows on Mt. Fuji, Japan

    by Cristina Perez-Guillen, Kae Tsunematsu, Kouichi Nishimura

  • Experimental Multiblast Craters and Ejecta—Seismo‐Acoustics, Jet Characteristics, Craters, and Ejecta Deposits and Implications for Volcanic Explosions

    by Ingo Sonder, Alison Graettinger, Tracianne B. Neilsen, Robin S. Matoza, Jacopo Taddeucci, Julie Oppenheimer, Einat Lev, Kae Tsunematsu, Greg Waite, Greg A. Valentine, et al.

    Blasting experiments were performed that investigate multiple explosions that occur in quick succession in unconsolidated ground and their effects on host material and atmosphere. Such processes are known to occur during phreatomagmatic eruptions at various depths, lateral locations, and energies. The experiments follow a multi‐instrument approach in order to observe phenomena in the atmosphere and in the ground, and measure the respective energy partitioning. The experiments show significant coupling of atmospheric (acoustic)‐ and ground (seismic) signal over a large range of (scaled) distances (30–330 m, 1–10 m J−1/3). The distribution of ejected material strongly depends on the sequence of how the explosions occur. The overall crater sizes are in the expected range of a maximum size for many explosions and a minimum for one explosion at a given lateral location. As previous research showed before, peak atmospheric over‐pressure decays exponentially with scaled depth. An exponential decay rate of was measured. At a scaled explosion depth of 4 × 10−3 m J−1/3 ca. 1% of the blast energy is responsible for the formation of the atmospheric pressure pulse; at a more shallow scaled depth of 2.75 × 10−3 m J−1/3 this ratio lies at ca. 5.5%–7.5%. A first order consideration of seismic energy estimates the sum of radiated airborne and seismic energy to be up to 20% of blast energy. Finally, the transient cavity formation during a blast leads to an effectively reduced explosion depth that was determined. Depth reductions of up to 65% were measured.


  • Open Access

    Experimental multiblast craters and ejecta - seismo-acoustics, jet characteristics, craters, and ejecta deposits and implications for volcanic explosions

    by Ingo Sonder, Allison Graettinger, Tracianne B. Neilsen, Robin S. Matoza, Jacopo Taddeucci, Julie Oppenheimer, Einat Lev, Kae Tsunematsu, Gregory P. Waite, Greg A. Valentine

    Blasting experiments were performed that investigate multiple explosions that occur in quick succession in the ground and their effects on host material and atmosphere. Such processes are known to occur during volcanic eruptions at various depths, lateral locations, and energies. The experiments follow a multi-instrument approach in order to observe phenomena in the atmosphere and in the ground, and measure the respective energy partitioning. The experiments show significant coupling of atmospheric (acoustic)- and ground (seismic) signal over a large range of (scaled) distances (30–330 m, 1–10 mJ^-1/3). The distribution of ejected material strongly depends on the sequence of how the explosions occur. The overall crater sizes are in the expected range of a maximum size for many explosions and a minimum for one explosion at a given lateral location. The experiments also show that peak atmospheric over-pressure decays exponentially with scaled depth at a rate of d0 = 6.47×10-4 mJ-1/3; at a scaled explosion depth of 4×10-3 mJ-1/3 ca. 1% of the blast energy is responsible for the formation of the atmospheric pressure pulse; at a more shallow scaled depth of 2.75×10-3 mJ-1/3 this ratio lies at ca. 5.5–7.5%. A first order consideration of seismic energy estimates the sum of radiated airborne and seismic energy to be up to 20% of blast energy.

  • Open Access

    Crisis hazard assessment for snow-related lahars from an unforeseen new vent eruption: the 2018 eruption of Kusatsu-Shirane volcano, Japan

    by Kyoko S. Kataoka, Kae Tsunematsu, Takane Matsumoto, Atsushi Urabe, Katsuhisa Kawashima

    Two-thirds of the 111 active volcanoes in Japan are covered with snow for several months during winter and demonstrate high hazard and risk potentials associated with snow-related lahars during and after eruptions. On 23 January 2018, a sudden phreatic eruption occurred at the ski field on Kusatsu-Shirane (Mt. Motoshirane) volcano, Japan. This new vent eruption from the snow-clad pyroclastic cone required forecasting of future snow-related lahars and crisis hazards zonation of downslope areas including Kusatsu town, a popular tourist site for skiing and hot springs. In order to achieve a prompt hazard assessment for snow-related lahars, a multidisciplinary approach was carried out involving characterization of proximal tephra deposits, snow surveys, and numerical lahar flow simulations using the Titan2D model. To determine the input parameters for the flow model, the consideration of snow water equivalent (SWE) immediately after the eruption (on 29 January) and in the post-eruptive period (on 12 March), was significant. In the case of Kusatsu-Shirane volcano during the winter of 2018, linear relationships between altitude and SWE, obtained at different elevations, were used to estimate the snow volume around the new vents. Several scenarios incorporating snow and snowmelt (water), with or without the occurrence of a new eruption, were simulated for the prediction of future lahars. Three lahar scenarios were simulated, including A) rain-on-snow triggered, B) ice/snow slurry, and C) full snowmelt triggered by a new eruption, and indicated the flow paths (inundation areas) and travel distances. These were useful for lahar hazard zonation and identification of potential high-risk areas. Since the input parameters required for the Titan2D flow model can be relatively easily determined, the model was suitable for the 2018 eruption at Motoshirane where historical and geological lahar records are not available for calibration. The procedure used in the study will enable rapid lahar prediction and hazard zonation at snow-clad volcanoes. Further consideration for simulating a cohesive-type flow, which was predicted by the primary deposits containing large amounts of clay minerals and could not be expressed in the Titan2D flow model, is necessary.


  • Open Access

    Application of an Inertia Dependent Flow Friction Model to Snow Avalanches: Exploration of the Model Using a Ping-Pong Ball Experiment

    by Kae Tsunematsu, Fukashi Maeno, Kouichi Nishimura

    Snow avalanches are catastrophic phenomena because of their destructive power. Therefore, it is very important to forecast the affected area of snow avalanches using numerical simulations. In our study, we focus on applying a numerical model to snow avalanches. The inertia-dependent flow friction model, which we call the “I-dependent” model, is a promising numerical model based on granular flow experiments and includes the local inertial effect. This model was introduced in previous studies as it predicts the shape and velocity of the granular flow accurately. We numerically investigated the particle diameter effect of the I-dependent model, and found that the smaller the particle diameter is, the faster the flow front velocity becomes. The final flow shape is similar to a crescent shape when the particle diameter is small. We applied this model to the ping-pong ball flow experiment, which imitated a snow avalanche on a ski jump slope. Comparing between the experimental and simulated results, the flow shape is better reproduced when the particle diameter is small, while the numerical simulation using a real ping-pong ball diameter did not show the clear crescent shape. Moreover, the relative error analysis shows that the best fit between experimental and simulated flow front velocity occurs when the particle diameter is larger than the actual size of a ping-pong ball. We conjecture that this discrepancy is mainly caused by aerodynamic effects, which, in this case, are large due to the low density of ping-pong balls. Therefore, it is necessary to explore the granular features of ping-pong balls or snow avalanches by conducting experiments, as done in previous experimental studies. Through such efforts, it may be possible to apply this I-dependent model to snow avalanches in the future.