An official website of the United States government
Abstract On Dec. 22, 2018, at approximately 20:55–57 local time, Anak Krakatau volcano, located in the Sunda Straits of Indonesia, experienced a major lateral collapse during a period of eruptive activity that began in June. The collapse discharged volcaniclastic material into the 250 m deep caldera southwest of the volcano, which generated a tsunami with runups of up to 13 m on the adjacent coasts of Sumatra and Java. The tsunami caused at least 437 fatalities, the greatest number from a volcanically-induced tsunami since the catastrophic explosive eruption of Krakatau in 1883 and the sector collapse of Ritter Island in 1888. For the first time in over 100 years, the 2018 Anak Krakatau event provides an opportunity to study a major volcanically-generated tsunami that caused widespread loss of life and significant damage. Here, we present numerical simulations of the tsunami, with state-of the-art numerical models, based on a combined landslide-source and bathymetric dataset. We constrain the geometry and magnitude of the landslide source through analyses of pre- and post-event satellite images and aerial photography, which demonstrate that the primary landslide scar bisected the Anak Krakatau volcano, cutting behind the central vent and removing 50% of its subaerial extent. Estimated submarine collapse geometries result in a primary landslide volume range of 0.22–0.30 km3, which is used to initialize a tsunami generation and propagation model with two different landslide rheologies (granular and fluid). Observations of a single tsunami, with no subsequent waves, are consistent with our interpretation of landslide failure in a rapid, single phase of movement rather than a more piecemeal process, generating a tsunami which reached nearby coastlines within ~30 minutes. Both modelled rheologies successfully reproduce observed tsunami characteristics from post-event field survey results, tide gauge records, and eyewitness reports, suggesting our estimated landslide volume range is appropriate. This event highlights the significant hazard posed by relatively small-scale lateral volcanic collapses, which can occuren-masse, without any precursory signals, and are an efficient and unpredictable tsunami source. Our successful simulations demonstrate that current numerical models can accurately forecast tsunami hazards from these events. In cases such as Anak Krakatau’s, the absence of precursory warning signals together with the short travel time following tsunami initiation present a major challenge for mitigating tsunami coastal impact.
more »
« less
Submarine landslide megablocks show half of Anak Krakatau island failed on December 22nd, 2018
Abstract As demonstrated at Anak Krakatau on December 22 nd , 2018, tsunamis generated by volcanic flank collapse are incompletely understood and can be devastating. Here, we present the first high-resolution characterisation of both subaerial and submarine components of the collapse. Combined Synthetic Aperture Radar data and aerial photographs reveal an extensive subaerial failure that bounds pre-event deformation and volcanic products. To the southwest of the volcano, bathymetric and seismic reflection data reveal a blocky landslide deposit (0.214 ± 0.036 km 3 ) emplaced over 1.5 km into the adjacent basin. Our findings are consistent with en-masse lateral collapse with a volume ≥0.175 km 3 , resolving several ambiguities in previous reconstructions. Post-collapse eruptions produced an additional ~0.3 km 3 of tephra, burying the scar and landslide deposit. The event provides a model for lateral collapse scenarios at other arc-volcanic islands showing that rapid island growth can lead to large-scale failure and that even faster rebuilding can obscure pre-existing collapse.
more »
« less
- Award ID(s):
- 1756665
- PAR ID:
- 10311365
- Date Published:
- Journal Name:
- Nature Communications
- Volume:
- 12
- Issue:
- 1
- ISSN:
- 2041-1723
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The Chaos Canyon landslide, which collapsed on the afternoon of 28 June 2022 in Rocky Mountain National Park, presents an opportunity to evaluate instabilities within alpine regions faced with a warming and dynamic climate. Video documentation of the landslide was captured by several eyewitnesses and motivated a rapid field campaign. Initial estimates put the failure area at 66 630 m2, with an average elevation of 3555 m above sea level. We undertook an investigation of previous movement of this landslide, measured the volume of material involved, evaluated the potential presence of interstitial ice and snow within the failed deposit, and examined potential climatological impacts on the collapse of the slope. Satellite radar and optical measurements were used to calculate deformation of the landslide in the 5 years leading up to collapse. From 2017 to 2019, the landslide moved ∼5 m yr−1, accelerating to 17 m yr−1 in 2019. Movement took place through both internal deformation and basal sliding. Climate analysis reveals that the collapse took place during peak snowmelt, and 2022 followed 10 years of higher than average positive degree day sums. We also made use of slope stability modeling to test what factors controlled the stability of the area. Models indicate that even a small increase in the water table reduces the factor of safety to <1, leading to failure. We posit that a combination of permafrost thaw from increasing average temperatures, progressive weakening of the basal shear zone from several years of movement, and an increase in pore-fluid pressure from snowmelt led to the 28 June collapse. Material volumes were estimated using structure from motion (SfM) models incorporating photographs from two field expeditions on 8 July 2022 – 10 d after the slide. Detailed mapping and SfM models indicate that ∼1 258 000 ± 150 000 m3 of material was deposited at the slide toe and ∼1 340 000 ± 133 000 m3 of material was evacuated from the source area. The Chaos Canyon landslide may be representative of future dynamic alpine topography, wherein slope failures become more common in a warming climate.more » « less
-
Abstract During the 3 month long eruption of Kı̄lauea volcano, Hawaii in 2018, the pre‐existing summit caldera collapsed in over 60 quasi‐periodic failure events. The last 40 of these events, which generated Mw > 5 very long period (VLP) earthquakes, had inter‐event times between 0.8 and 2.2 days. These failure events offer a unique data set for testing methods for predicting earthquake recurrence based on locally recorded GPS, tilt, and seismicity data. In this work, we train a deep learning graph neural network (GNN) to predict the time‐to‐failure of the caldera collapse events using only a fraction of the data recorded at the start of each cycle. We find that the GNN generalizes to unseen data and can predict the time‐to‐failure to within a few hours using only 0.5 days of data, substantially improving upon a null model based only on inter‐event statistics. Predictions improve with increasing input data length, and are most accurate when using high‐SNR tilt‐meter data. Applying the trained GNN to synthetic data with different magma‐chamber pressure decay times predicts failure at a nearly constant stress threshold, revealing that the GNN is sensing the underling physics of caldera collapse. These findings demonstrate the predictability of caldera collapse sequences under well monitored conditions, and highlight the potential of machine learning methods for forecasting real world catastrophic events with limited training data.more » « less
-
Abstract In early December 2020, an atmospheric river (AR) and rain-on-snow (ROS) event impacted the Haines, Alaska area, resulting in record-breaking rainfall and snowmelt that caused flooding and dozens of mass movement events. We consider the AR—a one-in-500-year event—as the trigger for the devastating Beach Road Landslide (BRLS), which destroyed or damaged four residences and took the lives of two people. The BRLS started as a debris avalanche and transitioned into a debris flow, with a total approximate landslide volume of 187,100 m3. Geomorphic analysis using lidar data identified evidence of paleo-landslides and displaced masses of rock, one of which served as the source area for the BRLS. Significant structural features in the weak ultramafic bedrock defined the head scarp area and formed the failure plane. This study illustrates the importance of identifying pre-existing landslide features and source areas likely to produce future landslides. As an increase in ROS events is projected for Southeast Alaska with warmer and wetter winters, we recommend the development of an AR scale coupled with geological information for the region, to enhance warnings to residents in landslide-prone areas.more » « less
-
Abstract Landslides commonly occur in areas with steep topography and abundant precipitation and pose a significant hazard to local communities. Some of the largest known landslides occur in Alaska, including several that caused local tsunamis. Many landslides may have gone undetected in remote areas due to lack of observations. Here, we develop a semiautomated workflow using both seismic and geodetic observations to detect, locate, validate, and characterize landslides in Alaska. Seismic observations have shown promise in continuously monitoring landslide occurrence, while remote sensing techniques are well suited for verification and high‐resolution imaging of landslides. We validate our procedure using the 28 June 2016, Lamplugh Glacier landslide. We also present observations of a previously unknown landslide occurred on 22 September 2017 in the Wrangell Mountains region. The Wrangell Mountains landslide generated a coherent surface wavefield recorded across Alaska and the contiguous United States. We used Sentinel‐1 Synthetic Aperture Radar and Sentinel‐2 optical imagery to map the respective mass deposit. To investigate the landslide dynamics, we inverted regional seismic surface wave data for a centroid single force failure model. Our model suggests that the Wrangell Mountains landslide lasted for about 140 s and had two subevents involving at least five distinct stages. We estimate that the landslide had displaced 3.1–13.4 million tons of rocks over a distance of ∼2 km. Our results suggest that combining seismic and geodetic observations can vastly improve the detection and characterization of landslides in remote areas in Alaska and elsewhere, providing new insights into the landslide dynamics.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1038/S41467-021-22610-5
