Debris flow initiates from the shallow landslide can develop by eroding surface materials. If numerous debris flows occur concurrently in a single catchment, each debris flow can join each other and increase its effect on the downstream area. Under such conditions, the topographic change affects the inundation of the debris flow itself and the following floods. The debris flow's initiation location may have a high effect on the damage distribution in the catchment. However, the predicting method of such points has not been established. In this study, we introduced statistical landslide prediction to generate such point data artificially. Cell-by-cell logistic regression model estimated the possibility to be the debris-flow initiation point from the terrain and precipitation data. This possibility distribution and pseud-random number sets generated artificial landslide location. On the other hand, we have also developed the 2D simulation based on an existing stony debris flow model. The code has been parallelized by introducing the OpenMP and MPI hybrid parallelization to enable application on supercomputers. This simulation only requires the location of debris flow sources (i.e., landslides) and topographical conditions (DEM), and several physical parameters. By using the numerical model, firstly, we have conducted the simulation employing actual location data. The results showed high reproducibility of the actual affected area. Artificial landslide data are also used instead of the actual initiation points To evaluate source locations' effect. Multiple simulation results revealed that the downstream area has low variability compared to the upstream area under the landslides. This result also indicated that accurate landslide location data is not necessary to predict the downstream area's damage.

The slope at Huafan University is recorded as a dip slope on the Dalun Mountain in the northern Western Foothills of Taiwan. The lithology of the dip slope is mainly composed of intercalation of sandstone and shale and the thickness of the sandstone varies from thin to massive interbedded with shale of Miocene age. By interpolating the thickness of colluvium as the cover material derived from the borehole data and analysing the contouring of the interpolation result in the surfer software, it is revealed that debris accumulates at the slope foot toward the southwest in the direction of movement. Because of the influences of tectonic activities, especially, according to previous investigations, two faults pass through the study area: the one (Fault A) extending perpenticular to the Nanshihkeng Fault plays an important role to change the orientation of the strata. For detailed analysis of the subsurface geological structure of this slope, this study focuses on the development of a 3D geological model by using a polynomial surface fitting equation which is aimed to compute the regressive orientation of bedding plane derived from the boreholes. In the first stage, the orientation of the dip slope is determined by calculating the regression plane passing through the elevations of the boundary between the sandstone and overlying shale. The results show that the regression plane dips southwestward with an inclination angle approximating 16°. In the second stage, several cross-sectional profiles are made to visualize and clarify the 3D geological model. The geological model will contribute significantly to the phenomenon of the slope failure and can be guidelines to minimize the disasters. Finally, numerical simulations could be performed in the future to define the best fit to the observed landslide behaviour by Finite Element Method (FEM).

Landslides were triggered in Mon State of Myanmar in August, 2019 during the monsoon season. Total 90 rainfall-triggered landslides are registered from Google Earth imageries and most of them are relatively shallow and small in dimension (500-15000 sq. m). Among these a massive one, hit Thae Phye Kone village, 30 buildings were damaged and 75 people were killed by the sliding debris. Weathered granites are major lithology of this area. A mountain extended from northwest to southeast provides orographic effect results significant variation in distribution of landslides. 77 out of 90 landslides located on the southwest side of the ridge due to the direction of monsoon wind is from the southwest to the northeast. The topographic features of 35 landslides with area >4000 sq.m were identified by using SRTM and Google eath images. The results shows that these rainfall-triggered are mainly located on the lower slope but approximately 20% of  are located on the higher slope, which is not common and sperated that with thick weathered regolith accumulated on gentle topography near the ridge, enhances the erosion. Landslides mobility is decreased (angle of reach increased) with increasing slope angle of the landslide source area reflects  on slope angle of the landslide source area correlating the friction angle of sliding debris positively. The angle of reach is controlled by other factors such as strength heterogeneity, local topography aside of the valley rather than slope gradient could obstacle the mobility.

The purpose of this study is to establish the basis for improving the skill score of early warning of landslides by identifying the characteristics of soil moisture variation based on the Noah Land Surface Model (LSM) as well as rainfall and for analyzing the soil moisture thresholds that cause the rainfall-induced landslides. Soil moisture is more directly related to slope stability than rainfall because it affects pore water pressure that causes landslides, whereas rainfall thresholds indirectly determine the likelihood of landslides through cumulative rainfall or rainfall intensity or rainfall duration. The commonly-used antecedent precipitation index (API) has a limit because it calculates the recession coefficient by reflecting only the daily average temperature to estimate the soil moisture by considering ground evaporation. Although an attempt to improve API has been made via correlation analysis with soil moisture observations, it does not reflect the fact that the maximum retentive capacity, i.e., porosity, is different for different soil texture. The Noah LSM estimates soil moisture content considering the evapotranspiration from the ground, based on various atmospheric and ground conditions, including temperature, relative humidity, vegetation cover, potential evaporation, soil texture, soil water flow, etc. Therefore, the utilization of Noah LSM has advantages of estimating the soil moisture variation for each soil texture type and for various environmental conditions and identifying integrated soil moisture and rainfall threshold information for each rainfall event, which are essentially required for the landslides early warning system. Furthermore, it is useful to recognize shallow landslides by considering the rainfall conditions, e.g., the recent 3-day cumulative rainfall, and the soil moisture's critical condition.

The 20 September 2018 Naga Landslide in Cebu Island represents a classic deep-seated, low-angle, high velocity translational slope failure in a suburban open-pit mining environment. With a total of 134 casualties, the Naga landslide is considered as the largest landslide in terms of volume and risk in the Philippines’ recorded history. Drone photo interpretation, field investigation, analysis of video footage, and back analyses gave light to the initiation, transport, depositional mechanisms and pre-failure slope stability. The Naga landslide has a peculiar low-angle slip surface, was spread over an area of 9.5 × 10⁵ m², a volume of 27 M m³ and a distal reach of 1.3 km. This complex landslide, moved predominantly as an extremely rapid translational block slide, with minor dry avalanches, and rockfalls. Slide blocks were detached along tensional fractures that formed and grew at least three weeks prior to failure. There was no apparent direct trigger (e.g., rainfall, earthquake) but preliminary back analyses, using limit equilibrium method, yielded unstable to marginally stable failure surfaces even under dry conditions. The 2018 massive failure was interpreted to be caused by progressive weakening of the cut slopes and exposure of potential failure surfaces from slope excavation. The Naga landslide event highlights the need to re-examine policies and scientific protocols in the country to avert a similar disaster in the future.

We present the occurrence tendency of rainfall prone to landslide hazards in the whole Japanese archipelago under climate change influence in the finest resolution of 1 by 1 km. In this study the proxy tool to identify the hazardous rainfall is the well-calibrated critical line method, which is currently being applied for official early warning practice by the Japanese government. The analysis output is based on the grid system in the very fine resolution of 1 by 1 km. For future climate, we analyzed the future projections under the scenarios of RCP2.6 and RCP8.5 from the famous 2-km and 5-km Nonhydrostatic Regional Climate Models, developed and published by Meteorological Research Institute of Japan Meteorological Agency (JMA). Using 15-year reanalyzed precipitation data published by JMA as the observation dataset, a bias-correction method is applied to adjust the extracted simulated precipitation at each 1-km grid. Then, with the critical line method the future changes of hazardous rainfall are successfully obtained in the nationwide and geographically regional scales. The seasonal change is also revealed. The results indicate a clear increasing trend of landslide risk in the period from July to September in the whole Japan, and mostly in the regions of Pacific Ocean side. Our analysis provides a new methodology to quantify valuable and high-resolution hazard map of future landslide risk.

Sufficient information on slope characteristics is key in detailed mapping of landslides. On-site investigations are mainly conducted to closely identify landslide features and gather relevant geomorphological data. Remote sensing is also used as a supplementary tool as it offers a different perspective in landslide investigations, allowing additional insights that are not commonly observed with conventional in situ methods. We used unmanned aerial vehicles (UAVs) to survey deep-seated landslide sites in the Philippines, and generated digital elevation models (DEMs) from the acquired aerial images. Different morphometric parameters were extracted from the DEMs producing nine derivatives. These derivatives were consolidated through principal component analysis into a composite raster image that emphasized morphologically-distinct surface features. Five morphometric parameters, namely, slope, aspect, multiple shaded relief, roughness, and surface relief ratio, were found to be the most descriptive of the surface morphology of two deep-seated landslide sites. In the village of Parasanon, Pinabacdao, Samar Island, most of the previously mapped landslide scarps and cracks were highlighted. In addition, potential new features such as the upper boundary of the landslide, extensions of cracks, and a linear feature at the right flank were identified. In Manghulyawon village in La Libertad, Negros Island, the extent of the head scarp was recognized, and depletion and accumulation zones were revealed. These features require further field verification, as would be the case with every other remotely-sourced data. In addition, it should be noted that analysis from this method is only limited to the ground surface without significant vegetative cover and aboveground structures. The study demonstrated how remote sensing through UAV-derived DEMs supplemented findings from conventional approaches, providing a better understanding of landslides.

Taiwan has risk of natural disaster due to young and complex geological condition, leading high vulnerability to landslide. Knowing more about the mechanism of the natural disaster is critical for mitigation, hence the monitoring implementation will be essential accordingly. The landslide monitoring methods would be divided into the underground, surface to the space monitoring one. Among remote sensing technologies, Synthetic Aperture Radar (SAR) is considered as a powerful tool in this recent decade, and SAR can monitor the land surface movement clearly without the distraction of the cloud or night effect. Therefore, the Interferometry Synthetic Aperture Radar (InSAR) method was proposed to accurately retrieve small displacement. Recently, some techniques of InSAR method have been further improved, such as Persistent Scatter (PS)-InSAR for long-term and robust monitoring. However, there is few rule or guideline especially for the landslide monitoring in practice. Thus, the parametric study was proposed to examine the appropriate ranges of each parameter during PS-InSAR processes, and Ali-Mt. was chosen as the verified case with in-field GNSS observations. First, SNAP software was suggested to be employed at the beginning and combined with STAMPS method for PS-InSAR. Then one significant parameter in SNAP was revealed as the degree of flat earth polynomial which depends on the site location, and in STAMPS step we found the capable range for: (1) filter_grid_size is 15-50; and (2) unwrap_time_win is 20-100 in landslide monitoring case. Further discussion will be proceeded with different case study in near future.

Speleothem d¹⁸O records have extensively been used to study Asian monsoon changes. However, less attention has been paid to the spatial distribution of speleothem d¹⁸O values. Despite some caveats, we advocate an approach to reconstruct spatial and temporal transects (“maps”) of speleothem d¹⁸O, thus time series of precipitation d¹⁸O distribution over the region. We obtained three speleothem d¹⁸O records from caves along a SW-NE transect from coastal Myanmar to southwestern China. All the three records cover the whole or a major portion of the past 40,000 years, particularly the last glacial maximum and present day. The comparisons between the records show a broadly decreasing trend in speleothem d¹⁸O values along the transect, consistent with an overall continental rainout effect of water isotopes when surface moisture is transported further inland. A much larger d¹⁸O gradient however exists during the last glacial maximum than in the late Holocene. A stronger water isotope fractionation during the glacial period is likely caused by a larger temperature gradient and suppressed plant transpiration along the transport pathway. Caution therefore is needed when interpreting the speleothem d¹⁸O records from a monsoon downwind region.

Since the late 19th century, it has been understood that lakes of the western United States reached their high stands after the last maximum of mountain glaciers (e.g., Russell, 1889). This interpretation is consistent with absolute age data obtained since the mid-20th century (e.g., Broecker and Orr, 1958). More recent studies of the last three decades show evidence to support lake high stands occurred during the last deglacial interval (~18 to 12 ka), with Heinrich Stadial 1 (~18 to 15 ka) encompassing a window when nearly all lakes had reached their maximum sizes (Munroe and Laabs, 2013). Efforts to understand the mechanisms that enhanced regional wetness depend strongly on age determinations for lake high stands. However, significant intra- and inter-lake basin disagreements on high stand age permit different interpretations of the climate dynamics that led to greater regional wetness. Our extensive stratigraphic, geomorphic, and geochronologic data from the Mono Basin leads us to suggest that the deglacial high stand of Mono Lake occurred in a very brief interval at about 16 ka. The time spanning the lake rise to its high stand cannot be exactly determined with the available data because the age constraints at hand across a 45 m elevational range are overlapping (16.07 ± 0.08 a to 15.94 ± 0.05 a). These observations require that the rise was sudden and achieved in less than 300 years. The timing of this sudden rise of the level of Mono Lake correlates with changes in tropical climate proxy records that are interpreted to reflect an abrupt southward diversion of the tropical rain belt. The correlation between tropical rainfall records and Mono Lake level is consistent with the hypothesis that changes in the position of the tropical rain belt result in modifications to the hydroclimate of the western US.

In the Fiji Islands, late Holocene environmental change is generally better understood for coastal and lowland situations than for the interior uplands of high volcanic islands. Lake Tagimaucia, a montane volcanic crater lake on Taveuni Island, is Fiji’s only high-elevation lake. It therefore potentially represents a very important site for preserving an environmental record in the highlands through the late Holocene. Physiographic factors expected to promote fast sedimentation rates are the large annual rainfall (>9000 mm), frequent passage of intense tropical cyclones, young geomorphology with erodible volcanic soils, steep catchment slopes, and rapid organic productivity in the surrounding cloud forest. The lake’s remote location means that its catchment is unlikely to have been anthropogenically affected during early human occupation in Fiji after 3000 BP. At present, however, it is not known whether Lake Tagimaucia’s deposits are able to provide a detailed, interpretable, environmental archive through the late Holocene. How fast is sedimentation occurring? Can any disturbances be identified, such as fire events, strong episodes of ENSO-driven drought or exceptional storms? Is there any evidence of the posited significant regional climatic shift called the ‘AD 1300 event’? Results are presented from the analysis of a shallow lake core, charcoal fragments, age-dating, stable isotopes δ¹³C and δ¹⁵N, and observed sedimentary anomalies. Possible influences of volcanic activity, catchment fire and the growth of unusual floating lake islands of sedge peat are discussed.

Differentiating between natural earthquakes and quarry blasts is a labor-intensive task. Recent studies show that artificial neural network (ANN) can achieve good accuracy for these two types of seismic events based on a variety of engineered waveform features. Here, we use convolutional neural network (CNN) to achieve automatic classification with waveform inputs. The training and test datasets consist of 25100 blast recordings from 2010 to 2020 and 31542 earthquake recordings from 2019 to 2020, recorded by the Southern California Seismic Network (SCSN). We input the normalized waveforms to the CNN model and apply the grid search method to determine the optimal model configurations and hyperparameters. Our preliminary results reveal that the CNN can differentiate between these two event types. Next, we will focus on enhancing the accuracy of the CNN model and compare it with that of other methods. Updated results will be presented in the meeting.

Earthquakes generated from different mechanisms could have different rupture processes and thus potentially different moment release histories. While the generation of deep earthquakes (origin depth 100-700 km) is supposed to be different from shallow ones (0-100 km), it remains enigmatic whether intermediate-depth (100~300 km) and deep-focus earthquakes (400~700 km) have different mechanisms. Here, we train machine learning models to test the possibility of distinguishing shallow, intermediate-depth, and deep-focus earthquakes with their source time functions (STFs). Our results show that deep and shallow earthquakes are distinguishable with STFs at an accuracy of 87%±1.1%, and intermediate-depth and deep-focus earthquakes are distinguishable at an accuracy of 68%±2.7%. We find that two STF features, i.e., scaled duration and scaled variance, are critical in both classification tasks, which is consistent with previous findings that rupture duration of same-size earthquakes is generally shorter at deeper depth. Our results indicate that at least part of intermediate-depth and deep-focus earthquakes have different mechanisms; the indistinguishable portion of events could either have the same mechanisms, or have different mechanisms but no recognizable STF differences.

Central Taiwan is a seismic active region that also experiences landslides and typhoons. 5 devastating earthquakes, the 2009 M = 6.2 Nantou, 2010 M = 6.4 Jiashian, 2013 M = 6.2 Nantou in March, 2013 M = 6.5 Nantou in June and 2016 M = 6.6 Meinong occurred in central mountainous area after (3 - 78 months) the 2009 Typhoon Morokat hit the same area. The identified close proximity in space and time between the 2009 Typhoon Morokat and earthquakes suggest the occurrence of cascading hazards. Triggering relations between typhoor or landslide induced surface erosion and earthquakes remain poorly understood. We propose to test the hypothesis that typhoon or landslide induced surface erosion resulted in surface unloading which may promote the rupture of earthquakes. We also probe the contribution of other sources of erosions, including interseismic erosion and large earthquake induced erosion to earthquake triggering. In this study, we present a comprehensive analysis of stress perturbations induced by four sources of erosional unloading (background, a large earthquake, and two typhoons). Our results indicate that surface erosion promoted the initiation and rupture propagation of the 2016 Meinong event, rupture propagation of the 1999 Chi-Chi mainshock, and future failure on the Chukou fault, Chelungpu fault, and flat decollement of Central Taiwan at ~ 4 – 15 km in depth. The cumulative stress perturbations due to surface erosion suggest a dominant role of background erosion with an order of 0.1 bar and secondary effect from earthquakes or typhoons with an order of 0.01 bar. Our study demonstrates a prominent earthquake triggering mechanism related to short-term background, large earthquakes, or wet typhoons induced surface erosion.

Delineation of physical factors that contribute to earthquake triggering is a challenging issue in seismology. We analyze hydrological modulation of seismicity in Taiwan using groundwater level data and GNSS time series. In western Taiwan, the seismicity rate reaches peak levels in February-April and drops to its lowest values in July-September, exhibiting a direct correlation with annual water unloading. The elastic hydrological load cycle may be the primary driving mechanism for the observed synchronized modulation of earthquakes, as also evidenced by deep earthquakes in eastern Taiwan. However, shallow earthquakes in eastern Taiwan (<18 km) are anti-correlated with water unloading, which is not well explained by either hydrological loading, fluid transport or pore pressure changes, and suggests other time-dependent processes. The moderate correlation between stacked monthly trends of large historic earthquakes and present-day seismicity implies a modestly higher seismic hazard during the time of low annual hydrological loading. Our analysis of interannual time series of earthquakes and water loading in 1994-2018 also shows more frequent small earthquakes in the dry years.

Coseismic landslides represent a major cascading hazard associated with high-magnitude earthquakes in mountainous regions. Although recent evidence suggests that landslide hazard can persist and even increase in the years immediately after an earthquake, our understanding of how patterns of landsliding and specifically how runout from rainfall-triggered reworking via debris flows, evolves through time and space remains limited. To address this, we established a systematic multi-temporal landslide inventory covering the full rupture area of the 2015 M_w 7.8 Gorkha, Nepal earthquake, extending across pre-, co- and post-seismic periods (2014-2020). We map landslides bi-annually from satellite imagery to record changes to existing landslides and new failures. We use the mapped landslides as source locations for a spatially distributed empirical runout model, Flow-R, that routes debris downslope simulating plausible ‘worst case’ scenarios. Our results show that even 4.5 years after the earthquake, the number and area of mapped landslides remains higher than on the day of the earthquake. Whilst coseismic landslides persist through time, we also document spatial shifts in the density of landsliding that develop from new but dispersed post-seismic landslides. Comparing modelled runout predictions to our sequential mapping of landslides, we find that 14% (>80 km²) of the total modelled coseismic runout hazard area is realised within 4.5 years of the earthquake, and so represents a portion of post-earthquake landslide hazard that can be foreseen. More broadly, we demonstrate that the persistence of coseismic landslides combined with the occurrence of new post-seismic landslides means that the potential runout hazard also remains well above coseismic levels by the end of the time series. These findings are particularly important for informing a comprehensive and precautionary approach to building into earthquake preparedness planning the likely nature of post-earthquake landslide hazard and risk.

Detailed image of subsurface structure is needed to understand the dynamics of tectonic plates and the processes that occur in the Earth's interior such as convection currents. Seismic tomographic techniques can be used to produce images that describe the structure of the 3D seismic wave velocity in the Earth's mantle globally. The phase data such as P, pwP, pP, S, and sP can be used to image the structure of crust, lithosphere, and Earth's mantle. One of the complex subduction zones resulting from plate tectonic dynamics is the Banda Arc in the eastern part of Indonesia. The Banda Arc is the result of the collision between the volcanic arc and the Australian continent. The Banda Arc region is marked by active seismicity in the eastern part of Indonesia.  This study used the catalog data of the International Seismological Center-Engdahl, van der Hilst, Buland (ISC-EHB) from 1964 to 2015. During this period, 5,243 earthquake events were obtained from 314 global seismic stations with a total of 19.533 phases ( P (16.941 phases), pP (30 phases), pwP(26 phases), S (2.516 phases), and SP (30 phases)). The preliminary result of our travel time tomographic imaging provides a good description of subsurface structure in the crust and mantle down to a depth of 650 km beneath the Banda Arc. More detailed subsurface structure information is being used as input for mapping disaster-prone areas and will be presented during the meeting.

Earthquakes in oceanic subduction zones are closely related to dehydration of subducting slabs. Here we present an integrated study on retrograded eclogites in Yezhai village of the Dabie ultra-high pressure (UHP) metamorphic belt, China. The eclogites have experienced amphibolite facies retrograde metamorphism and partial melting, and developed recumbent folds, tectonic boudins and fault breccias. Quartz veins were accumulated at the fold hinge zone, in which migmatitic eclogites appear as breccias. Zircons from host migmatitic eclogites, eclogite breccias and quartz veins are euhedral and have a core-rim structure. Their rare earth element (REE) distribution is characterized by light REE depletion, heavy REE enrichment, positive Ce anomaly and negative Eu anomaly. The U-Pb ages of 755–722 Ma for magmatic core of zircons indicate their origin of Neoproterozoic protoliths, whereas the ages of 242–221 Ma of metamorphic zircon rims from quartz veins yield timing of fluid flux and earthquakes. The Ti-in-zircon thermometer suggest equilibrium temperature from ~763 ° C and 814 ° C. In addition, EBSD analyses reveal that quartz in both eclogite migmatite and quartz veins show prism <c> slip and prism <a> slip, representing deformation temperature at 650–750℃ and 400–450℃. After high-temperature deformation, rapid healing of the fractures by crystallization of quartz veins will allow overlapping of mid-temperature ductile deformation in both migmatitic eclogites and quartz veins. Breakdown of phengite during exhumation of UHP metamorphic rocks could release water and facilitate partial melting and folding of eclogites. We propose that Si-rich fluids precipitated during melt crystallization and accumulated at the fold hinge, leading to a rapid increase in pore pressure. Therefore, the pulsed fluid flux of UHP rocks during exhumation can trigger shallow to intermediate-focus earthquakes in subduction channels. Such melting-associated hydraulic embrittlement process could also occur during subduction and dehydration of slabs.