Real-time landslide warning during heavy rainfall

Landslide hazard cascades can trigger earthquakes

While earthquakes are well-known to trigger surface hazards and initiate hazard cascades, whether surface hazards can instead trigger earthquakes remains underexplored. In 2018, two landslides on the Tibetan plateau created landslide-dammed lakes which subsequently breached and caused catastrophic outburst floods. Here we build an earthquake catalog using machine-learning and cross-correlation-based methods which shows there was a statistically significant increase in earthquake activity (local magnitude ≤ 2.6) as the landslide-dammed lake approached peak water level which returned to the background level after dam breach. We further find that ~90% of the seismicity occurred where Coulomb stress increased due to the combined effect of direct loading and pore pressure diffusion. The close spatial and temporal correlation between the calculated Coulomb stress increase and earthquake activity suggests that the earthquakes were triggered by these landslide hazard cascades. Finally, our Coulomb stress modeling considering the properties of landslide-dammed lakes and reservoir-induced earthquakes globally suggests that earthquake triggering by landslide-dammed lakes and similar structures may be a ubiquitous phenomenon. Therefore, we propose that earthquake-surface hazard interaction can include bidirectional triggering which should be properly accounted for during geological hazard assessment and management in mountainous regions.

Deep learning forecast of rainfall-induced shallow landslides

Rainfall triggered landslides occur in all mountain ranges posing threats to people and the environment. Given the projected climate changes, the risk posed by landslides is expected to increase, and the ability to anticipate their occurrence is key for effective risk reduction. Empirical thresholds and physically-based models are used to anticipate the short-term occurrence of rainfall-induced shallow landslides. But, evidence suggests that they may not be effective for operational forecasting over large areas. We propose a deep-learning based strategy to link rainfall to landslide occurrence. We inform and test the system with rainfall and landslide data available for the last 20 years in Italy. Our results indicate that it is possible to anticipate effectively the occurrence of rainfall-induced landslides over large areas, and that their location and timing are controlled primarily by the precipitation, opening to the possibility of operational landslide forecasting based on rainfall measurements and quantitative meteorological forecasts.

Identification and deformation analysis of potential landslides after the Jiuzhaigou earthquake by SBAS-InSAR

A World Natural Heritage Site, Jiuzhaigou, is the first nature reserve in China whose primary purpose is to protect natural scenery. On August 8, 2017, a M_s 7.0 earthquake caused many unstable slopes in Jiuzhaigou, Sichuan Province, China. In the extreme storm conditions that follow, the unstable slopes tend to develop into potential landslides, which can cause many casualties and property losses in scenic areas. Sentinel-1A ascending orbit data were obtained in this paper to establish a SAR database. The large-scale deformation rate map of the study area was obtained using a small baseline set InSAR technology. The potential landslides in the deformation area are preliminarily confirmed with remote sensing interpretation. The field verification is further carried out by studying the deformation information of the characteristic points on the potential landslides. The results show that 13 deformation zones were preliminarily identified, and three typical deformation zones were selected for coupling verification and identified as potential landslides. At the same time, further analysis shows that the four potential landslides have been in continuous linear deformation for a long time since the earthquake, posing a severe threat to the safety of local people's lives and property. The research results provide a reference for the early identification and warning of potential landslides in earthquake-prone regions.

The impact of environmental change on landslides, fatal landslides, and their triggers in Pakistan (2003-2019)

The actual impact of landslides in Pakistan is highly underestimated and has not been addressed to its full extent. This study focuses on the impact which landslides had in the last 17 years, with focus on mortality, gender of deceased, main triggers (landslides and fatal landslides), and regional identification of the hotspots in Pakistan. Our study identified 1089 landslides (including rockfalls, rockslides, mudslides, mudflows, debris flows) out of which 180 landslides were fatal and claimed lives of 1072 people. We found that rain (rainfall and heavy rainfall)-related landslides were the deadliest over the entire study period. The main trigger of landslides in Pakistan is heavy rainfall which comprises over 50% of the triggers for the landslide, and combined with normal rainfall, this rate climbs to over 63%. The second main reason for landslide occurrence is spontaneous (due to rock instability, erosion, climate change, and other geological elements) with landslides accounting for 22.3% of all the landslides. Landslides caused by rain-related events amounted to 41.67% of the fatalities, whereas spontaneous landslides caused 29.44% of the deaths and the human induced events accounted for 25.5% of the fatalities. The fatal landslides accounted for 19.53% deaths of the children. Our study also found that more than 48% of the deadly landslides occurred between the months of January to April, whereas the least fatal landslides occurred in the month of June which accounted for only 3% of all the fatal landslides in Pakistan.

The Impact of Landslides, Fatal Landslides and their Triggers in Pakistan (2003-2019).. [DOI: 10.21203/rs.3.rs-1993614/v1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

The actual impact of landslides in Pakistan is highly underestimated and has not been addressed to its full extent. This study focuses on the impact which landslides had in the last 17 years, with focus on mortality, gender of deceased, main triggers (landslides and fatal landslides) and regional identification of the hot spots in Pakistan. Our study identified 1089 landslides (including rockfalls, rockslides, mudslides, mudflows, debris flows) out of which 180 landslides were fatal and claimed lives of 1072 people. We found that rain (rainfall and heavy rainfall) related landslides were deadliest over the entire study period. The main trigger of landslides in Pakistan is heavy rainfall which comprises over 50% of the triggers for the landslide and combined with normal rainfall this rate climbs to over 63%. The second main reason for landslide occurrence is spontaneous (due to rock instability, erosion, climate change and other geological elements) with landslides accounting for 22.3% of all the landslides. Landslides caused by rain related events amounted to 41.67% of the fatalities whereas spontaneous landslides caused 29.44% of the deaths and the human induced events accounted for 25.5% of the fatalities. The fatal landslides accounted for 19.53% deaths of the children. Our study also found that more than 48% of the deadly landslides occurred between the months of January to April whereas the least fatal landslides occurred in the month of June which accounted for only 3% of all the fatal landslides in Pakistan.

The Post-Failure Spatiotemporal Deformation of Certain Translational Landslides May Follow the Pre-Failure Pattern

Investigating landslide deformation patterns in different evolution stages is important for understanding landslide movement. Translational landslides generally slide along a relatively straight surface of rupture. Whether the post-failure spatiotemporal deformation for certain translational landslides follows the pre-failure pattern remains untested. Here, the pre- and post-failure spatiotemporal deformations of the Simencun landslide along the Yellow River in 2018 were analyzed through multi-temporal remote sensing image analysis, Interferometric Synthetic Aperture Radar (InSAR) deformation monitoring and intensive field investigations. The results show that the pre- and post-failure spatial deformations both follow a retrogressive failure pattern. The long time series of the displacement before and after failure is characterized by obvious seasonal and periodic stage acceleration movements. Effective rainfall played an important role in the increase of the displacement acceleration, and the change in temperature might have accelerated the displacement. Finally, there is a possibility that the post-failure spatiotemporal deformation pattern of translational landslides does follow the pre-failure pattern when certain conditions are satisfied. The results are of great significance to improving our understanding of the spatiotemporal deformation pattern of landslides and to post-failure risk prevention and control.

“Satark”: Landslide Prediction System over Western Ghats of India

Mountains on the west coast of India are known as the Western Ghat (WG). The WG region has a landslide (LS) susceptibility index of four and is prone to LSs in the monsoon season due to rainfall activity. The LS study focuses on the area between 15.5–20.5° N, 72.5–77.0° E in the Maharashtra state. A catalog of 115 LS events in the study area has been prepared by collecting LS data for 17 years (2000–2016) from various sources. The area from the windward to the leeward side of the WG mountains is divided into three regions: (1) the windward region (72.5–73.4° E) (90 km width), (2) the immediate lee side (ILS) (73.40–74.20° E) (80 km width), and (3) distant lee side (DLS) (74.2–77.0° E) (280 km width). The Center for Citizen Science (CCS), Pune, India, developed the LS-predicting model “Satark” using data from satellites, the India Meteorological Department weather forecasts, radar products, synoptic conditions, and atmospheric sounding data from the Wyoming site for inferring conditions for a hydraulic jump on the WG. The model validation for the 5 years (2017–2021) showed a reasonably good Heidke skill score of 0.44. The model showed 76.5% success in LS prediction 1 day in advance. It is the first attempt of this kind in the Indian region.

Warning Models for Landslide and Channelized Debris Flow under Climate Change Conditions in Taiwan

Climate change has caused numerous disasters around the world. It has also influenced the climate of Taiwan, with urban areas exhibiting a temperature increase by 1 °C between 1998 and 2020. In this study, climate change and landslides in Taiwan were statistically analyzed. Cumulative annual precipitation in mountain watersheds in central Taiwan exhibit a declining trend and is lower than that in urban areas. The relatively few typhoons reduced the distribution of rainfall in mountain watersheds and fewer landslides. From 2017 to 2020, typhoon-induced rains caused fewer landslides than did other climate events such as the meiyu front, tropical low pressure, and southwesterly flow events. Three rainfall characteristics of landslide initiation were identified: high rainfall intensity over a short duration (<12 h), high-intensity and prolonged rainfall, and high cumulative rainfall over a long duration (>36 h). Combinations of warning models for landslides in cumulative rainfall–duration plots with rainfall intensity classification and mean rainfall intensity–duration plots with cumulative rainfall classification were presented. In recent (2018–2020) years, climate change has resulted in higher temperatures, less rainfall in mountain watersheds, and a lower rainfall threshold at which landslides are initiated by non-typhoon climate events.

Improved Method of Defining Rainfall Intensity and Duration Thresholds for Shallow Landslides Based on TRIGRS

The Transient Rainfall Infiltration and Grid-Based Regional Slope-Stability (TRIGRS) model has been widely used to define rainfall thresholds for triggering shallow landslides. In this study, the rainfall intensity(I)-duration(D) thresholds for multiple slope units of an area in Pu’an County, Guizhou Province, China were defined based on TRIGRS. Given that TRIGRS is used to simulate the slope stability under the conditions of a given increasing sequence of I-D data, if the slope reaches instability at I = a, D = b, it will also become unstable in the case of I = a, D > b or I > a, D = b. To explore the effect of these I-D data with the same I or D values on the definition of I-D thresholds and the best method to exclude these data, two screening methods were used to exclude the I-D data that caused instability in the TRIGTS simulation. First, I-D data with the same I values when D values are greater than a certain limit value were excluded. Second, several D values were selected to exclude I-D data with the same I values for a slope unit. Then, an I value was selected to exclude I-D data with the same D values. After screening, two different I-D thresholds were defined. The comparison with the thresholds defined without screening shows that the I-D data with the same I or D values will reduce the accuracy of thresholds. Moreover, the second screening method can entirely exclude these data.

Numerical analysis of the precursory information of slope instability process with constant resistance bolt

The instability of slope has already threatened life and property safety of the people, and improving the monitoring method of slope stability has important theoretical and practical significance for disaster prevention and reduction. According to the idea of “Newton force sudden drop and catastrophic occurrence” proposed by M.C. He in the landslide monitoring, a numerical model with constant resistance bolt has been established. Through numerical simulation research, it is found that the maximum principal stress, minimum principal stress and shear stress of the intersection point P of landslide surface and constant resistance bolt are sudden growth and sudden decrease, the vertical displacement and lateral displacement of this point P appear rise and fall before three kinds of stress. When loading to the next step of the step where three stress have reduced to a minimum value the slope is unstable and destroyed. At this time, the constant resistance bolt has undergone larger plastic deformation and damaged. Finally, comparing the stress curves and the acoustic emission (AE) curves, it can be seen that stress decreases from the maximum value and the AE curves begin to show a significant rise, the two curves display opposite law. It can be seen from the AE diagram that the failure mode of the slope is a combined tension and shear failure. The numerical test results provide a new idea for real-time monitoring and forecasting of slope.

Multihazards Scenario Generator: A Network-Based Simulation of Natural Disasters

The impact of natural disasters has been increasing in recent years. Despite the developing international interest in multihazard events, few studies quantify the dynamic interactions that characterize these phenomena. It is argued that without considering the dynamic complexity of natural catastrophes, impact assessments will underestimate risk and misinform emergency management priorities. The ability to generate multihazard scenarios with impacts at a desired level is important for emergency planning and resilience assessment. This article demonstrates a framework for using graph theory and networks to generate and model the complex impacts of multihazard scenarios. First, the combination of maximal hazard footprints and exposed nodes (e.g., infrastructure) is used to create the hazard network. Iterative simulation of the network, defined by actual hazard magnitudes, is then used to provide the overall compounded impact from a sequence of hazards. Outputs of the method are used to study distributional ranges of multihazards impact. The 2016 Kaikōura earthquake is used as a calibrating event to demonstrate that the method can reproduce the same scale of impacts as a real event. The cascading hazards included numerous landslides, allowing us to show that the scenario set generated includes the actual impacts that occurred during the 2016 events.

Experimental Study on the Relationship between the Velocity of Surface Movements and Tilting Rate in Pre-Failure Stage of Rainfall-Induced Landslides

With the development of deformation measuring technology at slope surfaces, prediction methods for rainfall-induced landslides based on the surface movements and tilting of slopes in the pre-failure stage have been recognized as a promising technique for risk reduction of slope failure triggered by rainfall. However, the correlation and possible mechanism for these prediction methods were rarely discussed. In this study, the comparison between the prediction methods of slope failure based on the time history of surface displacements and tilting in the acceleration stage was carried out by conducting a series of laboratory tests and field tests under rainfall, in which the movements and tilting behaviors at the slope surface were measured. The results show that the predicted failure time of tested slopes obtained by different prediction methods is consistent, and the correlation between these landslide prediction methods were also detected. A proportional relationship between the velocity of surface displacements and tilting rate was observed, and a possible mechanism for the sliding behavior was discussed based on this linear relationship as well. In addition, an expression for the linear relationship between the rate of the surface tilting and displacement was also established in this study, and the results could have significance for the understanding of the sliding behavior in the failure process in rainfall-induced landslides.

Worldwide Research Trends in Landslide Science

Landslides are generated by natural causes and by human action, causing various geomorphological changes as well as physical and socioeconomic loss of the environment and human life. The study, characterization and implementation of techniques are essential to reduce land vulnerability, different socioeconomic sector susceptibility and actions to guarantee better slope stability with a significant positive impact on society. The aim of this work is the bibliometric analysis of the different types of landslides that the United States Geological Survey (USGS) emphasizes, through the SCOPUS database and the VOSviewer software version 1.6.17, for the analysis of their structure, scientific production, and the close relationship with several scientific fields and its trends. The methodology focuses on: (i) search criteria; (ii) data extraction and cleaning; (iii) generation of graphs and bibliometric mapping; and (iv) analysis of results and possible trends. The study and analysis of landslides are in a period of exponential growth, focusing mainly on techniques and solutions for the stabilization, prevention, and categorization of the most susceptible hillslope sectors. Therefore, this research field has the full collaboration of various authors and places a significant focus on the conceptual evolution of the landslide science.

HydroMet: A New Code for Automated Objective Optimization of Hydrometeorological Thresholds for Landslide Initiation

Landslide detection and warning systems are important tools for mitigation of potential hazards in landslide prone areas. Traditionally, warning systems for shallow landslides have been informed by rainfall intensity-duration thresholds. More recent advances have introduced the concept of hydrometeorological thresholds that are informed not only by rainfall, but also by subsurface hydrological measurements. Previously, hydrometeorological thresholds have been shown to improve capabilities for forecasting shallow landslides, and they may ultimately be adapted to more generalized landslide forecasting. We present HydroMet, a code developed in Python by the U.S. Geological Survey, which allows users to guide the automated estimation of hydrometeorological thresholds for a site or area of interest, with the flexibility to select preferred threshold variables for the antecedent hydrologic conditions and the triggering meteorological conditions. Users can import hydrologic time-series data, including rainfall, soil-water content, and pore-water pressure, along with the times of known landslide occurrences, and then conduct objective optimization of warning thresholds using receiver operating characteristics. HydroMet presents many additional options, including selecting the threshold formula, the timescale of possible threshold variables, and the skill statistics used for optimization. Users can develop dual-stage thresholds for watch and warning alerts, with a lower, risk-averse threshold to avoid missed alarms and a less conservative threshold to minimize false alarms. Users may also choose to split their inventory data into calibration and evaluation subsets to independently evaluate the performance of optimized thresholds. We present output and applications of HydroMet using monitoring data from landslide-prone areas in the U.S. to demonstrate its utility and ability to produce thresholds with limited missed and false alarms for informing the next generation of reliable landslide warning systems.

Developing Real-Time Nowcasting System for Regional Landslide Hazard Assessment under Extreme Rainfall Events

In this research, a real-time nowcasting system for regional landslide-hazard assessment under extreme-rainfall conditions was established by integrating a real-time rainfall data retrieving system, a landslide-susceptibility analysis program (TRISHAL), and a real-time display system to show the stability of regional slopes in real time and provide an alert index under rainstorm conditions for disaster prevention and mitigation. The regional hydrogeological parameters were calibrated using a reverse-optimization analysis based on an RGA (Real-coded Genetic Algorithm) of the optimization techniques and an improved version of the TRIGRS (Transient Rainfall Infiltration and Grid-based Regional Slope-Stability) model. The 2009 landslide event in the Xiaolin area of Taiwan, associated with Typhoon Morakot, was used to test the real-time regional landslide-susceptibility system. The system-testing results showed that the system configuration was feasible for practical applications concerning disaster prevention and mitigation.

Application of High-Resolution Radar Rain Data to the Predictive Analysis of Landslide Susceptibility under Climate Change in the Laonong Watershed, Taiwan

Extreme rainfall has caused severe road damage and landslide disasters in mountainous areas. Rainfall forecasting derived from remote sensing data has been widely adopted for disaster prevention and early warning as a trend in recent years. By integrating high-resolution radar rain data, for example, the QPESUMS (quantitative precipitation estimation and segregation using multiple sensors) system provides a great opportunity to establish the extreme climate-based landslide susceptibility model, which would be helpful in the prevention of hillslope disasters under climate change. QPESUMS was adopted to obtain spatio-temporal rainfall patterns, and further, multi-temporal landslide inventories (2003–2018) would integrate with other explanatory factors and therefore, we can establish the logistic regression method for prediction of landslide susceptibility sites in the Laonong River watershed, which was devastated by Typhoon Morakot in 2009. Simulations of landslide susceptibility under the critical rainfall (300, 600, and 900 mm) were designed to verify the model’s sensitivity. Due to the orographic effect, rainfall was concentrated at the low mountainous and middle elevation areas in the southern Laonong River watershed. Landslide change analysis indicates that the landslide ratio increased from 1.5% to 7.0% after Typhoon Morakot in 2009. Subsequently, the landslide ratio fluctuated between 3.5% and 4.5% after 2012, which indicates that the recovery of landslide areas is still in progress. The validation results showed that the calibrated model of 2005 is preferred in the general period, with an accuracy of 78%. For extreme rainfall typhoons, the calibrated model of 2009 would perform better (72%). This study presented that the integration of multi-temporal landslide inventories in a logistic regression model is capable of predicting rainfall-triggered landslide risk under climate change.

Landslide Displacement Prediction Combining LSTM and SVR Algorithms: A Case Study of Shengjibao Landslide from the Three Gorges Reservoir Area

Displacement predictions are essential to landslide early warning systems establishment. Most existing prediction methods are focused on finding an individual model that provides a better result. However, the limitation of generalization that is inherent in all models makes it difficult for an individual model to predict different cases accurately. In this study, a novel coupled method was proposed, combining the long short-term memory (LSTM) neural networks and support vector regression (SVR) algorithm with optimal weight. The Shengjibao landslide in the Three Gorges Reservoir area was taken as a case study. At first, the moving average method was used to decompose the cumulative displacement into two components: trend and periodic terms. Single-factor models based on LSTM neural networks and SVR algorithms were used to predict the trend terms of displacement, respectively. Multi-factors LSTM and SVR models were used to predict the periodic terms of displacement. Precipitation, reservoir water level, and previous displacement are considered as the candidate factors for inputs in the models. Additionally, ensemble models based on the SVR algorithm are used to predict the optimal weight to combine the results of the LSTM and SVR models. The results show that the LSTM models display better performance than SVR models; the ensemble model with optimal weight outperforms other models. The prediction accuracy can be further improved by also considering results from multiple models.

Investigation on Surface Tilting in the Failure Process of Shallow Landslides

In recent decades, early warning systems to predict the occurrence of landslides using tilt sensors have been developed and employed in slope monitoring due to their low cost and simple installation. Although many studies have been carried out to validate the efficiency of these early warning systems, few studies have been carried out to investigate the tilting direction of tilt sensors at the slope surface, which have revealed controversial results in field monitoring. In this paper, the tilting direction and the pre-failure tilting behavior of slopes were studied by performing a series of model tests as well as two field tests. These tests were conducted under various testing conditions. Tilt sensors with different rod lengths were employed to investigate the mechanism of surface tilting. The test results show that the surface tilting measured by the tilt sensors with no rods and those with short rods located above the slip surface are consistent, while the tilting monitored by the tilt sensors with long rods implies an opposite rotational direction. These results are important references to understand the controversial surface tilting behavior in in situ landslide monitoring cases and imply the correlation between the depth of the slip surface of the slope and the surface tilting in in situ landslide monitoring cases, which can be used as the standard for tilt sensor installation in field monitoring.

Application of a Three-Dimensional Deterministic Model to Assess Potential Landslides, a Case Study: Antong Hot Spring Area in Hualien, Taiwan

This study proposes a landslide disaster assessment model combining a fully three-dimensional, physically-based landslide model with high precision of in situ survey data such as surface slip signs, geologic drilling results, underground water observation, and displacement monitoring results over time to perform distribution of potential landslide zones and the size of landslides (area and volume) in the Antong hot spring area in Hualien, Taiwan. The distribution of potential landslide zones in the study area was represented by slope stability safety factors. The results of the analysis showed that the toe of the slope and two upward slopes in the study area were potential landslide areas with safety factors of 1.37, 0.92, and 1.19, respectively. The 3D model analysis results indicated that a landslide could occur at a depth of 20 m at the toe of the slope. Monitoring results for 2015 and 2016 showed that the sliding depth at the toe of the slope was approximately 22.5 m; consequently, the error of landslide depth was only 2.5 m. The simulated results and in situ monitoring results were in good agreement. In addition, the simulated landslide volume was also compared with the results of an empirical equation commonly used in Taiwan to determine their differences. The landslide volumes estimated using the empirical equation were only approximately 38.5% in zone 1, 42.9% in zone 2, and 21.7% in zone 3 of that generated by the proposed model. The empirical equation was used to calculate the landslide volume according to the landslide area, which was subsequently converted into landslide depth. However, the obtained landslide depth was considerably lower than that derived from the in situ monitoring, implying that an empirical estimation approach may result in serious underestimation. Thus, the proposed model could predict landslide area and volume in advance to assist authorities in minimizing loss of life and property damage during a heavy rainfall event.

Application of Kalman Filter Model in the Landslide Deformation Forecast

Nonlinear exponential trend model is linearized into the linear model, then linearized model parameters are regarded as the state vector containing the dynamic noise to erect Kalman filter model based on exponential trend model to predict the deformation of the rock landslide. Deformation observation values of the landslide are regarded as a time series to erect AR(1) model, then model parameters of AR(1) model are regarded as the state vector containing the dynamic noise to erect Kalman filter model based on AR(1) model to predict the deformation of the rock landslide. The deformation of the landslide is regarded as the function of the time, then Taylor series is used to determine the functional relationship between the deformation of the landslide and the time, and Taylor series is spread to erect Kalman filter model based on Taylor series to predict the deformation of the earthy landslide. The deformation of landslides relates to many factors, the rainfall and the temperature influence the deformation of landslides specially, thus Kalman filter model based on multiple factors is erect to predict the deformation of the earthy landslide on the basis of Taylor series. Numerical examples show that the fitting errors and the forecast errors of the four Kalman filter models are little.

Daytime Rainy Cloud Detection and Convective Precipitation Delineation Based on a Deep Neural Network Method Using GOES-16 ABI Images

Precipitation, especially convective precipitation, is highly associated with hydrological disasters (e.g., floods and drought) that have negative impacts on agricultural productivity, society, and the environment. To mitigate these negative impacts, it is crucial to monitor the precipitation status in real time. The new Advanced Baseline Imager (ABI) onboard the GOES-16 satellite provides such a precipitation product in higher spatiotemporal and spectral resolutions, especially during the daytime. This research proposes a deep neural network (DNN) method to classify rainy and non-rainy clouds based on the brightness temperature differences (BTDs) and reflectances (Ref) derived from ABI. Convective and stratiform rain clouds are also separated using similar spectral parameters expressing the characteristics of cloud properties. The precipitation events used for training and validation are obtained from the IMERG V05B data, covering the southeastern coast of the U.S. during the 2018 rainy season. The performance of the proposed method is compared with traditional machine learning methods, including support vector machines (SVMs) and random forest (RF). For rainy area detection, the DNN method outperformed the other methods, with a critical success index (CSI) of 0.71 and a probability of detection (POD) of 0.86. For convective precipitation delineation, the DNN models also show a better performance, with a CSI of 0.58 and POD of 0.72. This automatic cloud classification system could be deployed for extreme rainfall event detection, real-time forecasting, and decision-making support in rainfall-related disasters.

Global CO2 Emission-Related Geotechnical Engineering Hazards and the Mission for Sustainable Geotechnical Engineering

Global warming and climate change caused by greenhouse gas (GHG) emissions have rapidly increased the occurrence of abnormal climate events, and both the scale and frequency of geotechnical engineering hazards (GEHs) accordingly. In response, geotechnical engineers have a responsibility to provide countermeasures to mitigate GEHs through various ground improvement techniques. Thus, this study provides a comprehensive review of the possible correlation between GHG emissions and GEHs using statistical data, a review of ground improvement methods that have been studied to reduce the carbon footprint of geotechnical engineering, and a discussion of the direction in which geotechnical engineering should proceed in the future.

Spatiotemporal Variation of Sediment Export from Multiple Taiwan Watersheds

Soil erosion and landslide triggered by heavy rainfall are serious problems that have threatened water resources in Taiwan watersheds. This study investigated the relationship among streamflow, sediment load, sediment concentration and typhoon characteristics (path and rainfall amount) during 2000–2017 for nine gauging stations in five basins (Tamshui River basin, Zhuoshui River basin, Zengwen River basin, Gaoping River basin, and Hualien River basin) representing the diverse geomorphologic conditions in Taiwan. The results showed that streamflow and sediment load were positively correlated, and the correlation was improved when the sediment load data were grouped by sediment concentration. Among these basins, the Zhuoshui River basin has the highest unit-discharge sediment load and unit-area sediment load. The soil in the upstream was more erodible than the downstream soil during the normal discharge conditions, indicating its unique geological characteristics and how typhoons magnified sediment export. The spatiotemporal variation in sediment loads from different watersheds was further categorized by typhoons of different paths. Although typhoon path types matter, the Zhuoshui and Hualien River basin were usually impacted by typhoons of any path type. The results indicated that sediment concentration, the watershed soil characteristics, and typhoons paths were the key factors for sediment loads. This study can be useful for developing strategies of soil and water conservation implementation for sustainable watershed management.

Satellite-based assessment of rainfall-triggered landslide hazard for situational awareness

Determining the time, location, and severity of natural disaster impacts is fundamental to formulating mitigation strategies, appropriate and timely responses, and robust recovery plans. A Landslide Hazard Assessment for Situational Awareness (LHASA) model was developed to indicate potential landslide activity in near real-time. LHASA combines satellite-based precipitation estimates with a landslide susceptibility map derived from information on slope, geology, road networks, fault zones, and forest loss. Precipitation data from the Global Precipitation Measurement (GPM) mission are used to identify rainfall conditions from the past seven days. When rainfall is considered to be extreme and susceptibility values are moderate to very high, a "nowcast" is issued to indicate the times and places where landslides are more probable. When LHASA nowcasts were evaluated with a Global Landslide Catalog, the probability of detection (POD) ranged from 8 to 60%, depending on the evaluation period, precipitation product used, and the size of the spatial and temporal window considered around each landslide point. Applications of the LHASA system are also discussed, including how LHASA is used to estimate long-term trends in potential landslide activity at a nearly global scale and how it can be used as a tool to support disaster risk assessment. LHASA is intended to provide situational awareness of landslide hazards in near real-time, providing a flexible, open source framework that can be adapted to other spatial and temporal scales based on data availability.

Fiber Optic Thermo-Hygrometers for Soil Moisture Monitoring

This work deals with the fabrication, prototyping, and experimental validation of a fiber optic thermo-hygrometer-based soil moisture sensor, useful for rainfall-induced landslide prevention applications. In particular, we recently proposed a new generation of fiber Bragg grating (FBGs)-based soil moisture sensors for irrigation purposes. This device was realized by integrating, inside a customized aluminum protection package, a FBG thermo-hygrometer with a polymer micro-porous membrane. Here, we first verify the limitations, in terms of the volumetric water content (VWC) measuring range, of this first version of the soil moisture sensor for its exploitation in landslide prevention applications. Successively, we present the development, prototyping, and experimental validation of a novel, optimized version of a soil VWC sensor, still based on a FBG thermo-hygrometer, but able to reliably monitor, continuously and in real-time, VWC values up to 37% when buried in the soil.

The Loma Prieta, California, Earthquake: An Anticipated Event

The first major earthquake on the San Andreas fault since 1906 fulfilled a long-term forecast for its rupture in the southern Santa Cruz Mountains. Severe damage occurred at distances of up to 100 kilometers from the epicenter in areas underlain by ground known to be hazardous in strong earthquakes. Stronger earthquakes will someday strike closer to urban centers in the United States, most of which also contain hazardous ground. The Loma Prieta earthquake demonstrated that meaningful predictions can be made of potential damage patterns and that, at least in well-studied areas, long-term forecasts can be made of future earthquake locations and magnitudes. Such forecasts can serve as a basis for action to reduce the threat major earthquakes pose to the United States.