Mountain torrents and debris flows are widely distributed in the mountainous region, threatening the urban development and infrastructure in mountain areas. The adverse effects of these hazards may increase due to the continued socio-economic development and influence of climate change on the frequency and magnitude of the hazards. This lecture introduces an early warning system of mountain hazards based on hazards process simulation and associated risk forecasting. The system identifies the watershed with high susceptibility to mountain hazard occurrences by monitoring the hazard-fostering conditions and real-time meteorological data. Focusing on those watersheds, the formation and movement of the hazards were simulated while different characteristics were captured, such as debris flow scale amplification and flash flood erosion. The risk of the mountain hazards was assessed based on the whole process of disaster formation-movement-deposition/disaster-causing. Compared with traditional early warning systems, which largely rely on rainfall thresholds and expert judgment, this proposed system is fully data-driven and process-based, while little human intervention is required. This system provides more accurate early warning information, and risk forecasting can better support disaster response planning for the government agency. This system is currently under trial in Liangshan Prefecture, Sichuan Province of China. Just in 2022, 15 debris flow and 52 flash flood events were captured and the early warning information was delivered to the residents and government. The accuracy is more than 79% and significantly improved the disaster resilience of the mountainous region.

About the Presenter

 Prof. CUI Peng has long been engaged in research on the formation mechanism, risk assessment, monitoring and early warning, prevention and control technology of debris flows and other mountain hazards. He has given a strong pulse to several topics of major relevance for disaster risk reduction and management, including (1) deepening the understanding of debris flow formation, scale amplification, and disaster-causing mechanisms; (2) providing rigorous insights concerning the formation and evolution of earthquake-induced hazards and multi-hazard chaining effect; (3) development of multi-scale disaster risk assessment model; (4) building of risk-level-based monitoring and early system to support efficient disaster reduction; and (5) creating the mass control and energy-based disaster mitigation theory and technology. He has published more than 400 papers with over 12000 citations and is the world's most published scholar in the field of debris flow.

How to cite: Cui, P.: A data-driven and process-based system for mountain torrent and debris flow early warning and risk forecasting, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-17044, https://doi.org/10.5194/egusphere-egu23-17044, 2023.

In the context of increasing number of landslide disasters, it is important to have efficient Landslide Early Warning Systems (LEWS). LEWS can reduce the risk with sufficient warning time and understanding the hazard and forecasting landslides is an important component of any LEWS. On local or slope scales, an early warning can be achieved with continuous monitoring, but on a regional scale, precise monitoring is still a question due to economical and practical concerns. Regional scale LEWS often relies on data-driven approaches such as rainfall thresholds, while process-based approaches are applied to smaller areas like single basins or watersheds due to complexities associated with precise data collection. The process-based approaches consider both spatial and temporal rainfall triggering factors as inputs, and hence they provide deterministic indices for the stability of a slope, based on both spatial and temporal conditions. In this study, a data-driven approach integrating probabilistic hydro-meteorological thresholds and landslide susceptibility maps (LSM) is used to develop a spatio-temporal landslide forecasting framework for a district in the southern part of India, Idukki. The method is then compared with two process-based approaches (Transient Rainfall Infiltration and Grid-based Regional Slope Stability (TRIGRS) and SHALSTAB) using a receiver operating characteristic curve (ROC) approach, using the landslide data of August 2018. From the analysis, it was observed that the data-driven approach has an efficiency of 81.21 %, while for TRIGRS and SHALTAB, the efficiencies are 72.15 % and 70.10 % respectively. The corresponding area under curve (AUC) values for all three models are 0.92, 0.80, and 0.76 respectively. The results indicate that the proposed data-driven model can perform better than both the process-based approaches, bypassing the complexities associated with physics-based modeling.

How to cite: Abraham, M. T. and Satyam, N.: Process-based and data-driven approaches for landslide forecasting: A quantitative comparison on regional scale, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1591, https://doi.org/10.5194/egusphere-egu23-1591, 2023.

With the increase of frequency and intensity of heavy precipitation in the future, rainfall triggered landslides (RTL) can be one of the major threat to human life and property security. Early warning systems of natural hazards are one of the most effective measure for reducing disaster losses and risks. However, the forecast of RTL in near-real-time (NRT) is extremely difficult since the quality of NRT precipitation data is relatively poor. Quantile regression forest (QRF), a state-of-the-art statistical postprocessing method, has been proved to reduce the difference existing between NRT satellite precipitation estimates and ground-based rainfall data. When predicted rainfall maps are put side by side with raw NRT satellite product, the pattern of the first matches much more closely the locations where landslide events have been mapped in a test site in North-Eastern Turkey. This leave an optimistic perspective on the application of statistical postprocessing techniques in the field of weather science and in general for natural hazard assessment. Ideally, by correcting the continuous information in space and time provided by satellite rainfall estimates, one could create a new operational tool for landslide early warning system, not bound to the financial and deployment requirement typical of rain gauge and terrestrial radar stations.

How to cite: Wang, X., Lombardo, L., and Tanyaş, H.: The potential application of statistical post processing techniques on landslide early warning system, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-3899, https://doi.org/10.5194/egusphere-egu23-3899, 2023.

Analysis and prediction of climate-driven geohazards, such as rainfall-induced landslides and slope failures, are becoming more challenging given the changing climate where extreme events are inevitable. Therefore, there is a need to move beyond conventional sources of data and consider multiple types of data for more accurate analysis and prediction of landslides. In recent years, Data Fusion and Machine Learning techniques have played an important role in paving the path towards a better understanding of the problem and finding more accurate models at regional and local levels that incorporate several contributing factors for slope failures. The purpose of the study is thus to evaluate the capacities of machine learning models in landslide susceptibility prediction and analyze their model performance in comparison of a numerical method, Transient Rainfall Infiltration and Grid-Based Regional Slope-Stability Model (TRIGRS). Classic machine learning models, namely Multi-Layer Perceptron Neural Network (MLP), Random Forest (RF), Gradient Boosted Regression Tree (GBRT) and Extreme Gradient Boosting (XGBoost) are selected and developed respectively. The study is carried out based on a preliminary field survey of rainfall-induced landslides near Kvam village, Norway, in June 2011. A methodology workflow of landslide susceptibility modeling is proposed, in which effective data processing approaches including feature selection, data resampling, data splitting, and feature scaling are discussed and summarized. The optimal hyperparameter optimization method is determined by performing a comparative time efficiency analysis of Bayesian and Grid Search methods. It is concluded that GBRT is the optimal method for landslide susceptibility mapping in the study case of Kvam based on seven popular model evaluation metrics. Other tree-based machine learning algorithms (RF and XGBoost) also show an overall outstanding performance and computational efficiency in comparison to MLP and TRIGRS models. The landslide susceptibility maps developed by prediction results from five models are also presented and statistically analyzed. Corresponding model performance ranks are found with results from model evaluation metrics.

How to cite: Luo, H., Liu, Z., Pan, Y., and Rocchi, I.: GIS-based rainfall-induced landslide susceptibility mapping: a comparative analysis of machine learning algorithms and a numerical method in Kvam, Norway, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-17051, https://doi.org/10.5194/egusphere-egu23-17051, 2023.

When and where shallow landslides occur depends on an interplay of predisposing, preparatory, and triggering factors. At a regional scale, data-driven analyses are extensively used to assess landslide susceptibility based on “static” maps of predisposing conditions. In contrast, data-driven analyses focusing on landslide triggering factors often rely on non-spatially explicit approaches to derive empirical rainfall thresholds. So far, few attempts have been made to integrate the spatial and temporal analysis domains beyond a posterior combination of separately derived susceptibility models and rainfall thresholds.

This work focuses on the mountainous Italian province of South Tyrol (7400 km²) and proposes a novel data-driven landslide prediction model that jointly considers landslide predisposition and dynamic preparatory and triggering factors. The approach builds on a hierarchical generalized additive model, multi-temporal shallow landslide data from 2000 to 2020 and a range of environmental variables (e.g., daily rainfall, topography, lithology, forest cover). The model produces maps that portray the relative probability of landslide occurrence. These spatially explicit predictions change dynamically as a function of local predisposition, seasonality, and observed (or hypothesized) dynamic preparatory and triggering rainfall (i.e. cumulative rainfall amounts based on varying day-windows). Linking the model output to known measures of model performance, such as hit rate and false alarm rate, enables the creation of dynamic classified maps that can be interpreted in analogy to commonly used empirical rainfall thresholds. The approach also accounts for potential spatial and temporal biases in the landslide inventory by restricting the underlying data sampling to effectively surveyed areas and time periods and by including (and averaging out) bias-describing random effect variables. Our validation confirms the model's high generalizability and predictive power while providing insights into the interplay of predisposing, preparatory and triggering factors for shallow landslide occurrence in South Tyrol. Application possibilities of this novel approach are discussed.

The research leading to these results is related to the PROSLIDE project that received funding from the research program Research Südtirol/Alto Adige 2019 of the Autonomous Province of Bozen/Bolzano – Südtirol/Alto Adige.

How to cite: Steger, S., Moreno, M., Crespi, A., Gariano, S. L., Brunetti, M. T., Melillo, M., Peruccacci, S., Marra, F., Borga, M., de Vugt, L., Zieher, T., Rutzinger, M., Mair, V., Campalani, P., and Pittore, M.: A data-driven approach to derive spatially explicit dynamic "thresholds" for shallow landslide occurrence in South Tyrol (Italy), EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1353, https://doi.org/10.5194/egusphere-egu23-1353, 2023.

A key component for Landslide early warning systems (LEWS) is constituted by thresholds that provide the conditions under which landslide events can be potentially triggered. Traditionally, thresholds based on rainfall characteristics have been proposed, but recently, the hydro-meteorological approach, combining rainfall with soil moisture or catchment storage information, is increasingly gaining attention. Usually, the hydro-meteorological thresholds proposed in the literature rely on soil moisture information relating to a single layer (i.e., depth or depth range). Nevertheless, multi-layered soil moisture information can be readily provided by in-situ observations, reanalysis projects, or hydrological models. Approaches based on this multi-layered information are lacking, probably because simpler thresholds, e.g., two-dimensional, are preferred and better understood by decision makers. This study, thus, proposes a methodology, based on principal component analysis (PCA), to derive two-dimensional hydro-meteorological thresholds that use multi-layer soil moisture information. Furthermore, a piece-wise linear equation is also suggested as threshold’s shape, which can be more flexible than the traditional power-law or bi-linear thresholds. Overall, results for Sicily Island (Italy), obtained using reanalysis soil moisture data at four different depths, corroborate the advantages of the hydro-meteorological approach with respect to the traditional rainfall thresholds. Specifically, a True Skill Statistic Index (TSS) equal to 0.5 is obtained for the traditional precipitation intensity-duration threshold, while a significantly higher one is obtained for the proposed hydro-meteorological thresholds using multi-layer information condensed in one variable by PCA (TSS = 0.71). Furthermore, comparing single- vs. multi-layer threshold performances provides insights on whether shallow or deep soil depth hydrological processes are more or less influent on landslide triggering. In this regard, for the analyzed study area, the multi-layer approach provides performances in terms of TSS are similar to those obtained with single-layer soil moisture at the upper depths, 0-7 cm and 7-28 cm, pointing out that landslide occurrences in Sicily are mostly driven by surface soil moisture. 

How to cite: Palazzolo, N., Peres, D. J., Creaco, E., and Cancelliere, A.: Deriving hydro-meteorological thresholds for landslide early warning using multi-layer soil moisture information , EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-6651, https://doi.org/10.5194/egusphere-egu23-6651, 2023.

Landslides are one of the most dangerous natural hazards, causing every year fatalities, considerable damage and relevant economic losses. Early warning systems (EWS) for rainfall-induced landslides represent an useful tool for mitigating the impact of such hazard. Traditionally, EWS are based on physically-based models or on empirical relationships between rainfall and landslide occurrence.

With the aim of taking into account the hydrological settings within the slope, the Umbria Regional Civil Protection Service started in considering also the soil moisture conditions as triggering factor during the daily analysis of shallow landslide hazard. The historical analysis of landslide events leaded to the definition of a set of soil moisture-based thresholds.

By analyzing the soil saturation conditions before and after the rainfall event (by using a hydrological model), it has been seen that most of the activations occurred when the soil reached saturation. This hypothesis has been validated by performing a historical analysis on more than 500 landslides occurred during the period 1990-2022. In this work, we took advantages of this finding and proposed an improvement of the current thresholds that considers the amount of rainfall needed by the soil to reach saturation, and hence, the slope instability. The amount of rainfall needed to reach saturation has been calculated through the definition of soil hydraulic parameters and the saturation degree at the start of the rainfall event. Then, if the fallen rainfall is higher than the critical value needed to reach saturation, an alarm is issued. The obtained threshold is based on soil characteristics and it is independent by the input data (no need for recalibration or threshold adjustment). The proposed methodology is able to identify correctly most of the proposed events (>70%) with a very limited amount of false alarms (4%) considering all the rainfall events occurred during the 1990-2022 period.

Further analyses are required for a better definition of the soil hydraulic parameters and the rainfall events but the obtained results confirmed the added value of using soil moisture conditions as triggering factor for shallow landslides activation.

How to cite: Ciabatta, L., Galeazzi, S., Ponziani, F., Berni, N., Camici, S., and Brocca, L.: Definition of physically-based regional soil moisture-rainfall thresholds for the assessment of landslide hazard, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-6688, https://doi.org/10.5194/egusphere-egu23-6688, 2023.

The Arctic plays a key role in understanding and mitigating the challenge of climate adaptation. Indeed, the observed Arctic warming is more than twice the global mean, implying that the Arctic may serve as an ‘early warning region’ in terms of climate change impacts. Longyearbyen, the Arctic capital settlement situated at 78°N in the archipelago of Svalbard, is located in a geographical hotspot affected by extreme Arctic climate change. Reports show that both experienced changes from 1971-2000 and projected 2071-2100 changes for Svalbard include increased air temperature, increased precipitation (especially in summer and autumn) and more frequent and intense events with heavy rainfall. Immediate and potentially detrimental impacts of these climate changes are seen in the widespread permafrost of the Arctic, which is particularly sensitive to climate change. These impacts include increases in active-layer thickness and melting of ground ice, resulting in increased risk of landslides.

Based in Longyearbyen, Svalbard, the PermaMeteoCommunity project develops a permafrost and meteorological response system that consists of (1) instrumented boreholes for direct observations of ground temperature and pore water pressure in the active layer and top meters of permafrost, (2) a network of meteorological stations, which records key standard parameters such as air temperature and precipitation with high spatial and temporal resolution. Using IoT technology, the observations are to be connected with an open online platform that receives and displays all data in near real-time. The data can thereby be used for local authorities and decision makers, during operational evaluations and extreme weather events such as large amounts of rain, potentially inducing permafrost-related landslides. The platform will also give access to historical data and the system will be highly relevant for use in research, for education, and in outreach as well as for long-term societal infrastructure and overall land area planning. Furthermore, work is being done to include more elements in the response system, among others (1) remote sensing data for monitoring of ground movement (2) high-resolution numerical weather simulations to be employed in preparedness situations on an on-demand basis and (3) a machine learning component for enhanced predictions of landslides.

How to cite: Jonassen, M. O., Christiansen, H. H., Shestov, A., and Tveit, K. I. L.: Improved monitoring and prediction of permafrost climate change related landsliding in the Arctic, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-13214, https://doi.org/10.5194/egusphere-egu23-13214, 2023.

Growing urbanization is pushing communities further into areas of known landslide hazard, elevating the risk posed to these communities. Hence, there is an increasing need to develop approaches that can characterize and monitor landslide hazards in urban areas. Here, we present recent developments in the rapid characterization of the landslide hazard using geophysics and remote sensing to parametrize hydromechanical models to assess probability of failure across a site in the highly populated Berkeley Hills, California. Calculating slope gradient from LiDAR, and estimating soil thickness from ambient seismic noise measurements, and total cohesion from vegetation distribution, we include the spatial variability of some of the most critical soil parameters in our hazard assessment. The results highlight various areas of elevated landslide hazard. Focusing on one such area, we used geophysical monitoring data to link changes in subsurface properties with slope instabilities, and found that rainfall induced increases in pore pressure drive slope deformation. Changes in seismic properties occurred up to 5h before actual soil displacements commenced. To monitor the hazard across the entire study site and to further increase our understanding of their triggering factors, we developed and  installed a dense wireless network of deformation, soil moisture, and pore pressure sensors. Using machine learning, we use this data to predict subsurface conditions critical to slope failure. We show that short-term predictions are comparably accurate, while long-term forecasts fail to predict sudden changes, mostly due to a lack of training data. The data obtained from these studies is starting to be incorporated into site management with the aim of mitigating the landslide risk. 

How to cite: Uhlemann, S., Fiolleau, S., Wielandt, S., and Dafflon, B.: Assessing and monitoring urban landslide hazards - integrating geophysics, remote sensing, and wireless sensor networks, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-8179, https://doi.org/10.5194/egusphere-egu23-8179, 2023.

Many mountainous areas of Campania, Southern Italy, are characterized by steep slopes covered by layered granular unsaturated pyroclastic deposits, mainly consisting of ashes and pumices, laying upon fractured limestone bedrock. The total thickness of the soil is quite variable with a few meters (1.0 m–1.5 m) in the steepest part of the slopes, and larger at the foot. Shallow landslides are often triggered after large and intense precipitations, turning into destructive debris flows that cause heavy damage and victims. The slope of Cervinara, located around 40 km Northeast of Naples (Campania, Italy), was involved in a catastrophic debris flow between 15-16 December 1999, triggered by a rainfall event of 325 mm in about 48 h. Since 2001, hydrological monitoring activities have been carried out at the slope, by measuring precipitation depth, soil volumetric water content and capillary tension. Moreover, in December 2017 an automatic hydro-meteorological station has been installed at the elevation of 575 m a.s.l., near the scarp of the 1999 landslide. It allows the assessment of slope hydrological balance, by identifying the major hydrological processes involving the cover and the perched aquifer, which develops in the upper part of the fractured bedrock during the rainy season. Lately, since 1 December 2022, new monitoring activities started. A remotely accessible low-cost network has been installed moving away from the landslide scarp of 1999, for expanding the area interested by soil moisture monitoring. The tested prototype network is based on the use of capacitive sensors placed at nodes located 20 m apart from each other with a communication system within the domain of Internet of Things (IoT) technology. Specifically, the low-cost sensors network allowed measurements of soil water content, communicating through short-range wireless IoT system (i.e., Wi-Fi) thanks to ESP32 boards. The field data can be visualized remotely on ThingSpeak_TM IoT platform on laptops and smartphones.

The tested IoT-based low-cost network shows the potential to enhance the amount of monitored hydrological data at affordable cost, so to improve risk management in landslide-prone areas. The same IoT network architecture with diffuse measurements can be replicated with long-distance radio communication between nodes, which allows extending the mutual distance up to few kilometers.

How to cite: Marino, P., Roman-Quintero, D. C., Santonastaso, G. F., and Greco, R.: Field hydrological monitoring with IoT-based low-cost sensor network on slopes subjected to rainfall-induced landslides, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-2708, https://doi.org/10.5194/egusphere-egu23-2708, 2023.

Abstract

Shallow landslides induced by heavy rainfall are a worldwide widespread phenomena and their related hazard is expected to increase due to more intense rainfall as a consequence of climate change (EEA Report No 15/2017). Since 2017, a decision-making tool based on Multi-Criteria Analysis (MCA) has been proposed as an objective approach to obtain landslide susceptibility maps and plan proper remedial works along linear infrastructure corridors (Tamburini et al., 2017). The study of low-cost sensors for Landslides Early Warning Systems (LEWS) as a risk mitigation tool to these phenomena along highways, railways and pipelines is here presented.

Soil hydrological conditions before a rainfall event for the estimation of trigger moments (Bordoni et al., 2019) are the starting point of a LEWS. Different sensors for the measure of these parameters, particularly soil volumetric water content, exist with different pros and cons. The aim of the research is to compare seven low-cost sensors selected by IMAGEO Srl company together with HORTUS Srl. The sensors have been engineered with a datalogging system and an automatic in-cloud transmission of the data and in June 2022 have been located on field at 2 different depths (-0.6 m and -1.2 m) at the test-site of Montuè in the Northern Apennines (Italy) where an Hydrometeorological Monitoring Station (Andromeda Project) is operating since 2012 with high-cost TDR probes present at the same depths. In November 2022 a volumetric water content profiler with nine measurement depths up to 1 m deep has been added to the new monitoring system.

The comparison between the hydrological data acquired by different sensors allows to evaluate the quality and reliability of the low-cost system before its final installation along the infrastructures lines. Monitored data together with rainfall parameters provided by both in situ rain gauges and ERA5-LAND satellite-derived data are used as input for the reconstruction of soil moisture values physically based thresholds.

The near real-time access to the monitored data allows to send warning alert when the established thresholds are exceeded, resulting in a LEWS able to identify periods of imminent landslide danger and to assess security along the lines.

How to cite: Pavanello, M., Bordoni, M., Vivaldi, V., Reguzzoni, M., Tamburini, A., Villa, F., and Meisina, C.: Low-cost Hydrological Monitoring System For Assessing Shallow Landslide Occurrence Along Linear Infrastructures, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-15170, https://doi.org/10.5194/egusphere-egu23-15170, 2023.

Landslide monitoring is essential to a better understanding of their dynamics and to the reduction of human casualties by detecting precursors before failures. In general, observations on the surface must be supplemented by sub-surface observations, in investigating the material in depth. Ten years ago, seismic ambient noise interferometry method has been applied to monitor the relative variations in surface seismic wave velocity (dV/V). As seismic wave velocities are directly related to material stiffness, any reduction in seismic velocity can be associated with a loss of stiffness with high probability (i.e. ground liquefaction or strong fracturation). This technique has led to the detection of a decrease in wave velocity several days before the rupture of a clay landslide [1], opening the way to a new precursor signal that could be used for alerts or early warning systems. Since then, several landslides have been monitored to this end [2].

In addition, by analysing the spectral content of seismic data, the natural resonance frequencies of rock instabilities (rocks columns, rock glaciers) can be monitored [3]. Their relative variation (dF/F) over time depends on the elastic properties and the geometry of the vibrating structure, which makes it possible to monitor its mechanical state, and to deduce precursor signals to significant failure.

In order to make these technologies operational, we have built a web-service, Soilstab, which allows for the processing of an existing dataset with the seismological methods described previously. This service is associated with Evorisk, a web-platform that displays the temporal evolution (updated daily or over a fixed period) of the results (dV/V and/or dF/F). This platform also integrates other available observations, such as environmental parameters (temperature, rainfall, snow, …) or surface observations (photogrammetry, GNSS/GPS-based displacement measurements, extensometers, etc..). Correlating all these observations is thus made easier to better understand and quantify the effect of environmental forcings (temperature, rain, freezing, etc.) on the dynamics of landslides and rock instabilities.

[1] G. Mainsant, E. Larose, C. Brönnimann, D. Jongmans, C. Michoud, M. Jaboyedoff, Ambient seismic noise monitoring of a clay landslide : toward failure prediction, J. Geophys. Res. 117, F01030 (2012).

[2] M. Le Breton, N. Bontemps, A. Guillemot, L. Baillet, E. Larose, Landslide Monitoring Using Seismic Ambient Noise Interferometry:: Challenges and Applications, Earth Science Review (2020)

[3] Colombero, C., Jongmans, D., Fiolleau, S., Valentin, J., Baillet, L., & Bièvre, G. (2021). Seismic noise parameters as indicators of reversible modifications in slope stability: a review. Surveys in Geophysics, 42(2), 339-375.

How to cite: Royer, A., Le Breton, M., Guillemot, A., Larose, E., Baillet, L., Guyoton, F., and Mayoraz, R.: Soilstab and Evorisk: a web-service and web-platform for landslide and rock-fall hazards monitoring using ambient seismic noise methods, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-3187, https://doi.org/10.5194/egusphere-egu23-3187, 2023.

The 5G RuralDorset project (https://5gruraldorset.org/) was a large (£9M; 2020-2022), multi-disciplinary project funded by the UK Department for Culture, Media and Sport that aimed to understand how 5G mobile network technologies could address some specific challenges in rural communities in Dorset, UK: public safety, economic growth, food production and environmental.  Work Package X aimed to develop and trial a novel landslide monitoring system for coastal cliffs using 5G/NB-IoT (Narrow Band - Internet of Things) technologies.  The system comprised a set of small, fully autonomous, highly integrated and power efficient sensing devices that were able to collect sensory data to identify landslide activity and landslide movement. These data were transmitted wirelessly using 5G/NB-IoT to a cloud-based Data Management Platform, where they were presented to the end user over a web interface for processing by Machine Learning algorithms. It is important to note that the term ‘Internet of Things’ has been used widely in recent years in application to landslide monitoring to in fact describe real-time telemetry of data. However, the true added value of IoT-enabled systems lies in their ability to extract knowledge from collected data, make decisions and take actions based on ambient conditions and evolution of physical processes.  This talk will describe the lessons learned from this work and highlight some of the obstacles to overcome when attempting to develop and commission such a system.

How to cite: Pennington, C., Angelopoulos, M., Arnhardt, C., Dray, M., Banks, V., Holyoake, S., and Christopher, J.: Lessons learned from the demonstrator 5G RuralDorset project: a pilot landslide monitoring system using Internet of Things, Machine Learning and 5G/NB-IoT mobile networks, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-17343, https://doi.org/10.5194/egusphere-egu23-17343, 2023.

Rainfall is considered the most important triggering factor for landslide initiation. It is expected that changes in the precipitation regimes, as a direct consequence of climate change, will influence slope stability at different temporal and geographical scales altering the frequency and the distribution of rainfall-induced landslides. Therefore, there is a need to develop and implement efficient landslide risk management to deal with the increasing landslide risks. In this context, territorial landslide early warning systems (Te-LEWS) can be valuable tools to warn authorities, civil protection personnel and the population about the occurrence of rainfall-induced landslides over wide areas, typically through the prediction and measurement of meteorological variables with a limited consideration of soil behaviour. Currently, widespread deployments of Te-LEWS integrating monitoring data collected at local scale have been inhibited by the high cost of sensors, the requirement of frequent maintenance and the inflexibility of cable-based systems.

The use of advanced monitoring and communication technologies could provide the means to solve these challenges.

This study proposes a four-phase approach to set up an IoT-based early warning system at municipal scale. The territory of a municipality has been chosen as the reference spatial unit of assessment because it has an extension that is intermediate between slope units and regional warning zones. The framework is based on the following four phases: monitoring, modelling, forecasting, and warning. The study focuses on the first phase of the proposed approach, i.e., combination of widespread meteorological data and local real-time measurements coming from monitoring networks installed at specific locations of great geomorphological interest within the study area. The measurements—specifically soil water content, pore water pressure and suction— are used to provide additional information to be used for enhancing the performance of the warning model. It is important to highlight that, within the proposed framework, an important role for the warning system will be played by community members and other people working or living in the municipality, herein called human sentinels, which will be involved, for instance, in the proper maintenance of sensors and for documenting the impacts of extreme climate events (e.g., photos and reports uploaded in local data platforms).

The monitored sites are located within the municipality of Amalfi, southern Italy, and the implementation will be addressed within the activities of the Horizon Europe project “The HuT: The Human-Tech Nexus - Building a Safe Haven to cope with Climate Extremes”. The territory of Amalfi consists of a steep mountain front that rises abruptly from the Tyrrhenian Sea. Steep topographic gradients forced human settlements to develop along the coast at the mouth of the main streams. The town is a densely populated area with high touristic impact. The town is located in a morphologically complex zone of southern Italy frequently affected by dangerous and calamitous landslides.

This study aims at highlighting importance of considering both climate forcing factors and in-situ geotechnical parameters within a warning model operational at municipal scale.

How to cite: Pecoraro, G., Calvello, M., and Rianna, G.: A framework based on IoT and human sentinels for a municipal landslide early warning system: a case study in southern Italy of the project “The HuT”, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-818, https://doi.org/10.5194/egusphere-egu23-818, 2023.

The Himalayas being an actively deforming terrain with steep hillslopes and significant rainfall is highly susceptible to landslide hazards. The fractured nature of the rocks, steep river bank and moderate to steep road cut-slope provide added risk to the slope failure vulnerability.  Landslides are generally triggered after substantial downpours during monsoon resulting in significant economic loss and casualties. Thus, an Early Warning System (EWS) is an absolute necessity. Our study explores the potential of geogenic Fracture Induced Electromagnetic Radiation (FEMR) technique for landslide forecasting. The FEMR technique is getting increasingly popular amongst geoscientists due to its ability to determine the zones of enhanced stress accumulation enabling it to be an effective precursor to a mass failure episode. This method is cost-effective and quick compared to other conventional rock mechanical studies. In the Eastern Himalayas, slopes get reactivated causing recurrent landslide episodes. The slope failure is generally guided by tensile rapture followed by shear sliding (TRSS) mechanism.  We acquired high-resolution FEMR linear profiles along the landslide planes with a portable instrument called ANGEL-M. Additionally, soil strength tests and numerical modelling were carried out to complement FEMR results. We concluded that the most severe deep landslides could be correlated to very high FEMR amplitudes whereas very low FEMR amplitude often corresponds to a lack of failure. Moderate FEMR amplitudes, however, are related to shallow-intermediate landslide occurrences. We further recommend that the FEMR technique can be utilized by moderately skilled surveyors from local municipalities as a pre-monsoon landslide forecasting methodology and mitigation strategies can be planned in advance.

How to cite: Das, D. and Mallik, J.: Landslide forecasting in the Eastern Himalayas by Fracture Induced Electromagnetic Radiation (FEMR) Technique, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-744, https://doi.org/10.5194/egusphere-egu23-744, 2023.

Rainfall-induced sediment disasters are catastrophic events that occur compounded during extreme precipitation. Territorial early warning systems (Te-LEWS) are necessary to predict these disasters. The warning information is disseminated based on thresholds derived from the correlation between rainfall magnitude and disaster occurrences. Nations that established successful Te-LEWS have maintained historical rainfall records and corresponding landslide occurrences that result in the precise derivation of early warning thresholds. In contrast, countries newly establishing Te-LEWS face difficulties setting the thresholds due to a lack of precise information on rainfall magnitude and historical landslide occurrences. In India, the India Meteorological Department (IMD) provides impact-based forecasts of rainfall that may induce landslides based on daily, 3-day cumulative and longer antecedent thresholds. However, thresholds correlating landslides with continuous monitoring through hourly/sub-hourly rainfall observations, which are the basis of the nowcast in real time, still need to be developed. Here, we present a framework for predicting landslide occurrences, i.e., shallow landslides, debris slides, and debris flows, using hourly rainfall. Using the framework, we analyse case studies of extreme precipitation-induced landslides in the Himalayas and Western Ghats, India. Through this, catchment-wise early warning thresholds are derived. This study opens avenues to improve the precision of impact-based rainfall forecasts for landslides. 

How to cite: Siva Subramanian, S., Srivastava, P., Sen, S., and Yunus, Ali. P.: Physically-based model derived thresholds of sediment disasters for impact-based rainfall forecasts, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-8674, https://doi.org/10.5194/egusphere-egu23-8674, 2023.