Landslide Hazard Assessment Research Articles

Research on landslide hazard assessment in data-deficient areas: a case study of Tumen City, China

Research on landslide hazard assessment in data-deficient areas: a case study of Tumen City, China

Downhill progressive landslide hazard assessment: a simple framework for prediction of triggering thresholds and failure modes

The pronounced discrepancies in triggering intensities and failure modes observed in downhill progressive landslides pose significant challenges to their risk prediction and mitigation. Instead of focusing on an individual triggering event and/or failure mode as done in previous studies, this study systematically explores the instability issues of downhill progressive failure in long slopes. A unified analytical model is proposed to understand these observed discrepancies, in which criteria for the initiation of slope failure corresponding to all possible failure modes are formulated. Accordingly, a simple framework is established that is capable of predicting possible triggering thresholds and failure modes in advance. This would improve the accuracy of landslide hazard assessment, which is often reduced by the unknowns of the failure mechanisms and inaccurate assumptions of stress distribution along the failure surface. Furthermore, three practical cases are analyzed based on the established framework, and the findings are compared with field observations and existing studies, showing good consistency. This work gives an insight into a systematic understanding of the various failure phenomena in downhill progressive landslides and provides a reference for their risk assessment and subsequent decision-making.

Landslide hazard assessment and their application in land management in Kendari, Southeast Sulawesi Province, Indonesia

Kendari is the capital of Southeast Sulawesi Province which is prone to landslides. Good land management needs to be done to minimize the impact of landslides. This study aimed to map the Kendari landslide hazard that can be used as an input into land management strategy, especially in vulnerable to the threat of landslides. The primary data used in this study were DEMNAS and Sentinel-2. Landslide detection was carried out using a Process Hierarchy Analysis (AHP) approach and validated by field surveys. Land capability analysis was based on landform analysis using land system data. Land management directions were carried out based on the integration of landslide hazard analysis with the ability of the land to be calibrated with actual land cover. The analysis showed that areas with high and very high landslide hazards reached 2654.09 ha (9.64%) and 4354.78 ha (15.82%). Capability class of VII is spread over structural hills to the north and south of Kendari with an area of 7,215.81 ha (26.21%). Land management in areas with very high landslide hazards and land capability class VII is to add cover crops on land that is not protected by a canopy. Cover crops that can be added are the grass type to minimize the danger of erosion that can trigger landslides.

A Hybrid Multi-Hazard Susceptibility Assessment Model for a Basin in Elazig Province, Türkiye

Preparation of accurate and up-to-date susceptibility maps at the regional scale is mandatory for disaster mitigation, site selection, and planning in areas prone to multiple natural hazards. In this study, we proposed a novel multi-hazard susceptibility assessment approach that combines expert-based and supervised machine learning methods for landslide, flood, and earthquake hazard assessments for a basin in Elazig Province, Türkiye. To produce the landslide susceptibility map, an ensemble machine learning algorithm, random forest, was chosen because of its known performance in similar studies. The modified analytical hierarchical process method was used to produce the flood susceptibility map by using factor scores that were defined specifically for the area in the study. The seismic hazard was assessed using ground motion parameters based on Arias intensity values. The univariate maps were synthesized with a Mamdani fuzzy inference system using membership functions designated by expert. The results show that the random forest provided an overall accuracy of 92.3% for landslide susceptibility mapping. Of the study area, 41.24% were found prone to multi-hazards (probability value > 50%), but the southern parts of the study area are more susceptible. The proposed model is suitable for multi-hazard susceptibility assessment at a regional scale although expert intervention may be required for optimizing the algorithms.

Hazard assessment for regional typhoon-triggered landslides by using physically-based model – a case study from southeastern China

ABSTRACT Landslide hazard assessment is an important component of risk management and land-use planning. This study aims to investigate the application of a physically-based model named after the fast shallow landslide assessment model (FSLAM) to rainfall-triggered landslide hazard assessment. In August 2015, a total of 123 landslides induced by Typhoon Soudelor in Wenzhou City, southeastern China, was taken as an example. Five input raster files (elevation, soil types, vegetation, antecedent rainfall, event rainfall) and two parameter files regarding soil properties and vegetation were determined. Considering the randomness and uncertainty of soil and vegetation parameters on the regional scale, FSLAM model computes the probability of failure (PoF) by using random parameters inputs. Finally, the landslide hazard map was generated for the study area to reflect the landslide risk. The results showed that FSLAM could accurately capture the effect of rainfall on PoF of slopes, and more than 70% of the landslide were identified in very high/high hazard zones. The accuracy of the receiver operating characteristic (ROC) reached 0.720, which was higher than that of the Transient Rainfall Infiltration and Grid-Based Regional Slope-Stability (TRIGRS) model (0.620). Regarding the computational time, FSLAM had better efficiency, and the consuming time was 1/25 compared with TRIGRS.

The Role of Historical Data to Investigate Slow-Moving Landslides by Long-Term Monitoring Systems in Lower Austria

Landslides are one of the most significant natural hazards worldwide. They can have far-reaching negative impacts on societies in different socio-economic sectors as well as on the landscape. Among the different types and processes that can also affect infrastructure and land use planning, slow-moving landslides are often underestimated. Therefore, studying areas affected by slow movements provide an opportunity to better understand the spatial and temporal patterns of these processes, their forcings, mechanisms, and potential risks. This study aims to investigate the importance of historical data for improving landslide hazard assessment in Lower Austria (Austria), which is particularly prone to landslides. This paper focuses on how historical information formed the basis for the establishment of three long-term landslide monitoring observatories in this region. The analysis conducted highlights the importance of using historical data to better assess the frequency and magnitude relationships and phases of landslide activity. In particular, they can extend the temporal window and provide relevant information on past events and accelerations to improve knowledge of landslide dynamics and the resulting socio-economic impacts. In order to better assess the landslide hazard associated, it is necessary to integrate historical data and monitoring datasets obtained by surface and subsurface methods. Both components allow for the characterization of the spatio-temporal evolution of slow movements and the analysis of the hazard over time. Based on a variety of historical sources, it was possible to install the instruments constituting the long-term landslide monitoring observatories in a meaningful manner. The results demonstrate the influential role of human impact on the stability conditions, which may also contribute to landslide occurrence. In this regard, the attempt to combine historical data and long-term, continuous monitoring systems in the presented landslide observatories can improve landslide risk reduction measures in the region. The integration of different techniques and tools, along with ongoing research and collaboration with local authorities, will further improve our understanding of these slow-moving processes and the development of effective management strategies.

Earthquake and rainfall-induced landslide hazard assessment of Kutupalong Rohingya camp using meteorological and geological information

Earthquake and rainfall-induced landslide hazard assessment of Kutupalong Rohingya camp using meteorological and geological information

Integrated Approach for Landslide Risk Assessment Using Geoinformation Tools and Field Data in Hindukush Mountain Ranges, Northern Pakistan

Landslides are one of the most recurring and damaging natural hazards worldwide, with rising impacts on communities, infrastructure, and the environment. Landslide hazard, vulnerability, and risk assessments are critical for landslide mitigation, land use and developmental planning. They are, however, often lacking in complex and data-poor regions. This study proposes an integrated approach to evaluate landslide hazard, vulnerability, and risk using a range of freely available geospatial data and semi-quantitative techniques for one of the most landslide-prone areas in the Hindukush mountain ranges of northern Pakistan. Very high-resolution satellite images and their spectral characteristics are utilized to develop a comprehensive landslide inventory and predisposing factors using bi-variate models to develop a landslide susceptibility map. This is subsequently integrated with landslide-triggering factors to derive a Landslide Hazard Index map. A geospatial database of the element-at-risk data is developed from the acquired remote sensing data and extensive field surveys. It contains the building’s footprints, accompanied by typological data, road network, population, and land cover. Subsequently, it is analyzed using a spatial multi-criteria evaluation technique for vulnerability assessment and further applied in a semi-quantitative technique for risk assessment in relative risk classes. The landslide risk assessment map is classified into five classes, i.e., very low (18%), low (39.4%), moderate (26.3%), high (13.3%), and very high (3%). The developed landslide risk index map shall assist in highlighting the landslide risk hotspots and their subsequent mitigation and risk reduction.

Landslide Hazard Assessment Method Considering the Deformation Factor: A Case Study of Zhouqu, Gansu Province, Northwest China

Landslides are geological disasters that can cause great damage to natural and social environments. Landslide hazard assessments are crucial for disaster prevention and mitigation. Conventional regional landslide hazard assessment results are static and do not take into account the dynamic changes in landslides; thus, areas with landslides that have been treated and stabilized are often still identified as high-risk areas. Therefore, a new hazard assessment method is proposed in this paper that combines the deformation rate results obtained by interferometric synthetic aperture radar (InSAR) with the results of conventional hazard assessments to obtain the hazard assessment level while considering the deformation factor of the study area, with Zhouqu, Gansu Province, selected as the case study. First, to obtain the latest landslide inventory map of Zhouqu, the hazard assessment results of the study area were obtained based on a neural network and statistical analysis, and an innovative combination of the deformation rate results of the steepest slope direction from the ascending and descending data were obtained by InSAR technology. Finally, the hazard assessment level considering the deformation factor of Zhouqu was obtained. The method proposed in this paper allows for a near-term hazard assessment of the study area, which in turn enables dynamic regional landslide hazard assessments and improves the efficiency of authorities when conducting high-risk-area identification and management.

Simple method for probabilistic seismic landslide hazard analysis based on seismic hazard curve and incorporating uncertainty of strength parameters

Seismic landslide hazard assessment is typically based on deterministic analysis to calculate slope displacements during an earthquake. However, deterministic analysis is unable to account for various uncertainties in the analysis process. This study focuses on utilizing the established results from probabilistic seismic hazard analysis (PSHA) and extending probabilistic seismic landslide hazard analysis (PSLHA) to account for the uncertainties of ground motions and soil strengths. Two simplified methods, being flexible and fast, are proposed and evaluated by comparing with the fully but complex probabilistic method. One of the methods that yields a better performance is applied to evaluate the probabilistic seismic landslide hazard in the Tsao-Ling area of Taiwan using information regarding the seismic hazard curve and it's deaggregation results. Lastly, seismic landslide hazard maps are constructed for different return periods and compared with the deterministic map for the 1999 Chi-Chi earthquake.

ПРОГНОЗУВАННЯ ЗСУВНОЇ НЕБЕЗПЕКИ ТА ОЦІНКА ВПЛИВУ НА СУСПІЛЬСТВО: ГОЛОВНІ ПІДХОДИ, ПРИНЦИПИ, МЕТОДИ

The problem of predicting the landslide hazard is a priority area of research in the field of assessment of risks and natural disasters, which requires a comprehensive in-depth analysis of the factors of landslide formation, as well as the synthesis of existed theorethical and empiric data for a full understanding of the problem of landslide hazard and comprehensive assessment of its impact on community. The presented research is aimed at the development, implementation, and application of a comprehensive methodology for predicting landslide hazards and assessing their impact on the infrustructure. The research was carried out within the framework of national and international projects with the partiсipation of international partners from universities in France, Austria and Great Britain. The methodology of regional landslide hazard prediction for different structural regions of Ukraine is based on the methods of spatial modelling and aims at the landslide susseptibility mapping, creating multifactorial spatial models. As a result of a comprehensive analysis of landslide factors and spatial modelling integrated landslide hazard maps were created. These maps provide an opportunity to comprehensively assess the landslide hazard for different regions. Methods of local prediction of landslide hazard based on the application of a rational complex of geological, physical, remote, thermographic studies, and deterministic modelling enable to identify the main features and potential activity of landslide processes within landslide-prone areas and suggest preventive measures for risk mitigation. Examples of the integrated methodology applications for landslide hazard prediction within model sites in Kaniv and Kyiv regions are given. The concept of informing people about the potential geohazards was given.

Effect landslide hazard mitigation using an integrated of Analytical Hierarchy Process and Multi Criteria Evaluation: A case study the Jeneberang watershed

Jeneberang watershed has an important role in maintaining environmental stability, especially for the surrounding area. In 2004 a landslide occurred in the Jeneberang watershed which caused casualties and material losses. This study aims to determine the distribution and level of vulnerability to landslides in the Jeneberang watershed and to formulate an effective scenario to reduce the level of vulnerability. This research is located the Jeneberang watershed, South Sulawesi Province. The data used include data on the weight of each parameter and criteria for determining landslide susceptibility obtained from expert interviews, land use maps obtained from analysis of Landsat 8 OLI satellite imagery in 2020, slope maps, rainfall maps, forest area status maps, and map type. soil. The method used is the integration of AHP and MCE for landslide hazard assessment, GIS analysis and spatial decision-making for scenario simulation. The results of the analysis show that the rainfall parameter has the highest weight in determining the level of landslide susceptibility with a value of 0.327, the results of the MCE analysis for assessing the level of landslide susceptibility in the Jeneberang watershed show an area of 18.829 ha. or 18% of the Jeneberang watershed area which is classified as high hazard. The results of the analysis show that the use of an effective scenario reduces the landslide hazard at a high level of hazard. In the optimistic scenario, it can reduce the high vulnerability level by 9.14%, while in the moderate scenario it is effective in reducing the high vulnerability level by 8.72%.

Landslide Hazard Assessments of A Potential Earthquake-triggered in Central Taiwan Using Newmark's Model with the Stochastic Semi-Empirical Technique

This research aims to evaluate the probability distribution of landslides triggered by a potential earthquake. Based on the evaluations, successive disaster prevention and mitigation can be well-prepared in advance. Deriving the critical acceleration value depends on the geometry and strength of the material on the slope. Thus, the Newmark's displacement of slope can be determined using the critical acceleration and strong motion recording of an earthquake. A threshold value of Newmark's displacement is assumed as 5cm, which means a landslide may occur if Newmark's displacement is greater than it. To identify the feasibility, we synthesize two waveforms of the 1999 Chi-Chi earthquake (M_w 7.6) and one for the 2013 Nantou earthquake (M_w 6.2) to compute the Newmark's displacements at Jiou-Jiou Peaks and Shinyi town in Nantou County, Taiwan by using the stochastic semi-empirical technique. The Newmark's displacements are 9.75 averaged and 6cm respectively, which are coincident with the earthquake-triggered landslides occurring actually. By counting accumulations of Coulomb's stress changes of Nantou earthquakes in 2013 in central Taiwan, we predict the epicenter of a potential earthquake. The regression relation between critical acceleration with Arias intensity of the seismic waveform is employed to calculate Newmark's displacements in the Nontou County. Accompanied by Newmark's displacement, the empirical probabilities of earthquake-induced landslides are derived. Accordingly, there are four populated densely areas with higher Newmark's displacements and probabilities, indicating that high landslide risk is existing potentially in these areas and the corresponding disaster prevention management and mitigation must be carefully regulated.

Integrating multiple information sources for landslide hazard assessment: the case of Italy

Landslides are the most frequent and diffuse natural hazards in Italy causing the greatest number of fatalities and damage to urban areas. The integration of natural hazard information and social media data could improve warning systems to enhance the awareness of disaster managers and citizens about emergency events. The news about landslide events in newspapers or crowdsourcing platforms allows fast observation, surveying and classification. Currently, few studies have been produced on the combination of social media data and traditional sensors. This gap indicates that it is unclear how their integration can effectively provide emergency managers with appropriate knowledge. In this work, rainfall, human lives, and earmarked fund data sources were correlated to “landslide news”. Analysis was applied to obtain information about temporal (2010–2019) and spatial (regional and warning hydrological zone scale) distribution. The temporal distribution of the data shows a continuous increase from 2015 until 2019 for both landslide and rainfall events. The number of people involved and the amount of earmarked funds do not exhibit any clear trend. The spatial distribution displays good correlation between “landslide news”, traditional sensors (e.g., pluviometers) and possible effects in term of fatalities. In addition, the cost of soil protection, in monetary terms, indicates the effects of events.

Earthquake-Induced Landslide Susceptibility Assessment Using a Novel Model Based on Gradient Boosting Machine Learning and Class Balancing Methods

Predicting the susceptibility of a specific part of a landslide (SSPL) involves predicting the likelihood that the part of the landslide (e.g., the entire landslide, the source area, or the scarp) will form in a given area. When predicting SSPL, the landslide samples are far less than the non-landslide samples. This class imbalance makes it difficult to predict the SSPL. This paper proposes an advanced artificial intelligence (AI) model based on the dice-cross entropy (DCE) loss function and XGBoost (XGBDCE) or Light Gradient Boosting Machine (LGBDCE) to ameliorate the class imbalance in the SSPL prediction. We select the earthquake-induced landslides from the 2018 Hokkaido earthquake as a case study to evaluate our proposed method. First, six different datasets with 24 landslide influencing factors and 10,422 samples of a specific part of the landslides are established using remote sensing and geographic information system technologies. Then, based on each of the six datasets, four landslide susceptibility algorithms (XGB, LGB, random-forest (RF) and linear discriminant analysis (LDA)) and four class balancing methods (non-balance (NB), equal-quantity sampling (EQS), inverse landslide-frequency weighting (ILW), and DCE loss) are applied to predict the SSPL. The results show that the non-balanced method underestimates landslide susceptibility, and the ILW or EQS methods overestimate the landslide susceptibility, while the DCE loss method produces more balanced results. The prediction performance of the XGBDCE (average area under the receiver operating characteristic curve (0.970) surpasses that of RF (0.956), LGB (0.962), and LDA (0.921). Our proposed methods produce more unbiased and precise results than the existing models, and have a great potential to produce accurate general (e.g., predicting the entire landslide) and detailed (e.g., combining the prediction of the landslide source area with the landslide run-out modeling) landslide susceptibility assessments, which can be further applied to landslide hazard and risk assessments.

The risk assessment of landslide hazards in the Badong section of Three Gorges using the variable fuzzy set theory

The risk assessment of landslide hazards has a tremendous influence on people’s lives and property safety; therefore, its investigation is significant. The stratigraphic lithology, degree of weathering, relationship between the structural plane and slope direction, cohesive force, angle of internal friction, severity, average slope degree, height of slope, and type of landslide are adopted as the evaluation factors first. Second, an assessment model is developed based on the variable fuzzy set theory. In addition, the proposed model is utilized to assess the landslide hazards in the Badong section of Three Georges in China. Finally, the results demonstrate that the results derived from the proposed model are consistent with the current specifications; the accuracy rate reaches 83%. The method can determine the risk level of landslide hazards and provide an alternative scheme. Hence, this study can accurately present a new approach for assessing landslide hazards in the future.

Geology, geomorphology and geochronology of the coseismic? Emad Deh rock avalanche associated with a growing anticline and a rising salt diapir, Zagros Mountains, Iran

This work documents for the first time the prehistoric, but morphologically pristine, Emad Deh rock avalanche of the Zagros Mountains. The ca. 420 Mm3 rock avalanche was initiated as a rockslide on a dip slope affecting a weak unit of marls with interbedded limestones overlain by a thick and competent limestone formation. The slope is located on the northern limb of the growing Gavbast Anticline and at the edge of the inflating Gavbast Dome, related to a buried salt diapir of Hormuz salt. The unconfined rock avalanche deposits, covering 32 km2, were accumulated on coalescing alluvial fans and a floodplain environment. The depositional lobe shows sectors with distinctive morphological features attributable to the variable flow-depositional behavior of the debris stream, controlled by the nature of the substrate: (1) proximal continuous breccia flanked by levees; (2) intermediate depression; and (3) flat-topped polygonal hills and conical hills. A morphometric and spatial distribution analysis has been performed with the 550 conical hills (hummocks) mapped in the distal sector. The rock avalanche, with 914 m of maximal height drop (H) and 9280 m of runout (L), displays an extraordinarily high mobility (H/L index of 0.09), that can be attributed to the combined effect of dynamic rock fragmentation and basal lubrication by the soft substrate in the floodplain. Relief rejuvenation and slope over-steepening related to the growth of the anticline and the inflation of the Gavbast Dome are considered the main long-term preparatory factors that shifted the slope to a state of marginal stability. The slope failure was likely triggered by a large M ≥ 6.5–7 earthquake at ca. 5.4 ka, as indicated by OSL ages obtained from folded alluvium situated just beneath the rock avalanche deposit. The source of the earthquake was probably a blind reverse fault situated beneath the asymmetric Gavbast Anticline, as support the structural and geomorphic features of the fold. The identification and dating of large coseismic landslides in the Zagros Mountains, where large earthquakes are rarely accompanied by primary surface ruptures, could help to improve both landslide and seismic hazard assessments.

Hazard Assessment of Potential Large-Scale Landslides in the Watershed of the Chenyulan River

Taiwan’s mountains are steep, geologically dispersed, and their hillsides are over-utilized. Serious debris flow disasters are prone to happen whenever monsoons or typhoons bring a lot of rain. Typhoon Morakot’s large-scale landslide in 2009, which devastated half of Xiaolin Village and claimed around 500 lives, is one such instance. Large-scale landslides have the potential to seriously harm people, property, and the economy. Large-scale landslides were less frequent in the past, but they have recently increased in frequency. An important topic is how to avert disasters through efficient risk management. Eight influencing factors were chosen after multivariate analysis screening, including three terrestrial factors (total catchment area size, average catchment area slope, and total curvature), three material factors (stratum type, distance from fault, and average normalized-difference vegetation index prior to the event), and two trigger factors (maximum daily rainfall and maximum hourly rainfall). Our overall analysis’ accuracy rankings were as follows: neural network, 93.7%; logistic regression analysis, 92.2%; and discriminant analysis, 89%. The area under curve (AUC) values for the neural network (0.819), discriminant analysis (0.824), and logistic regression analysis (0.732) all demonstrated high discriminative abilities based on the receiver operating characteristic curve, indicating that all three of our models are adequate for this purpose. Large-scale debris flow disasters are more likely to happen when the catchment region’s total area, total curvature, and average slope are higher, according to a cluster study. These findings can help to decrease the effects of disasters of this kind by helping to create early warning systems.

A dynamic landslide hazard monitoring framework for the Lower Mekong Region

The Lower Mekong region is one of the most landslide-prone areas of the world. Despite the need for dynamic characterization of landslide hazard zones within the region, it is largely understudied for several reasons. Dynamic and integrated understanding of landslide processes requires landslide inventories across the region, which have not been available previously. Computational limitations also hamper regional landslide hazard assessment, including accessing and processing remotely sensed information. Finally, open-source software and modelling packages are required to address regional landslide hazard analysis. Leveraging an open-source data-driven global Landslide Hazard Assessment for Situational Awareness model framework, this study develops a region-specific dynamic landslide hazard system leveraging satellite-based Earth observation data to assess landslide hazards across the lower Mekong region. A set of landslide inventories were prepared from high-resolution optical imagery using advanced image-processing techniques. Several static and dynamic explanatory variables (i.e., rainfall, soil moisture, slope, relief, distance to roads, distance to faults, distance to rivers) were considered during the model development phase. An extreme gradient boosting decision tree model was trained for the monsoon period of 2015–2019 and the model was evaluated with independent inventory information for the 2020 monsoon period. The model performance demonstrated considerable skill using receiver operating characteristic curve statistics, with Area Under the Curve values exceeding 0.95. The model architecture was designed to use near-real-time data, and it can be implemented in a cloud computing environment (i.e., Google Cloud Platform) for the routine assessment of landslide hazards in the Lower Mekong region. This work was developed in collaboration with scientists at the Asian Disaster Preparedness Center as part of the NASA SERVIR Program’s Mekong hub. The goal of this work is to develop a suite of tools and services on accessible open-source platforms that support and enable stakeholder communities to better assess landslide hazard and exposure at local to regional scales for decision making and planning.

Potential of satellite-derived hydro-meteorological information for landslide initiation thresholds in Rwanda

Abstract. Satellite and hydrological model-based technologies provide estimates of rainfall and soil moisture over larger spatial scales and now cover multiple decades, sufficient to explore their value for the development of landslide early warning systems in data-scarce regions. In this study, we used statistical metrics to compare gauge-based and satellite-based precipitation products and assess their performance in landslide hazard assessment and warning in Rwanda. Similarly, the value of high-resolution satellite and hydrological model-derived soil moisture was compared to in situ soil moisture observations at Rwandan weather station sites. Based on statistical indicators, rainfall data from Integrated Multi-Satellite Retrievals for Global Precipitation Measurement (GPM_IMERG) showed the highest skill in reproducing the main spatiotemporal precipitation patterns at the study sites in Rwanda. Similarly, the satellite- and model-derived soil moisture time series broadly reproduce the most important trends of in situ soil moisture observations. We evaluated two categories of landslide meteorological triggering conditions from IMERG satellite precipitation: first, the maximum rainfall amount during a multi-day rainfall event, and second, the cumulative rainfall over the past few day(s). For each category, the antecedent soil moisture recorded at three levels of soil depth, the top 5 cm by satellite-based technologies as well as the top 50 cm and 2 m by modelling approaches, was included in the statistical models to assess its potential for landslide hazard assessment and warning capabilities. The results reveal the cumulative 3 d rainfall RD3 to be the most effective predictor for landslide triggering. This was indicated not only by its highest discriminatory power to distinguish landslide from no-landslide conditions (AUC ∼ 0.72), but also the resulting true positive alarms (TPRs) of ∼80 %. The modelled antecedent soil moisture in the 50 cm root zone Seroot(t−3) was the most informative hydrological variable for landslide hazard assessment (AUC ∼ 0.74 and TPR 84 %). The hydro-meteorological threshold models that incorporate the Seroot(t−3) and RD3 following the cause–trigger concept in a bilinear framework reveal promising results with improved landslide warning capabilities in terms of reduced rate of false alarms by ∼20 % at the expense of a minor reduction in true alarms by ∼8 %.