Abstract Accurate early warning systems for landslides are a reliable risk-reduction strategy that may significantly reduce fatalities and economic losses. Several machine learning methods have been examined this purpose, underlying deep (DL) models’ remarkable prediction capabilities. The long short-term memory (LSTM) gated recurrent unit (GRU) algorithms the sole DL model studied in extant comparisons. However, several other suitable time series forecasting tasks. In paper, we assess,...

Abstract Mapping of landslides over space has seen an increasing attention and good results in the last decade. While current methods are chiefly applied to generate event-inventories, whereas multi-temporal (MT) inventories rare, even using manual landslide mapping. Here, we present innovative deep learning strategy which employs transfer that allows for Attention Deep Supervision Multi-Scale U-Net model be adapted detection tasks new areas. The method also provides flexibility re-training...

Abstract. In the domain of landslide risk science, susceptibility mapping (LSM) is very important, as it helps spatially identify potential landslide-prone regions. This study used a statistical ensemble model (frequency ratio and evidence belief function) two machine learning (ML) models (random forest XGBoost; eXtreme Gradient Boosting) for LSM in province Belluno (region Veneto, northeastern Italy). The investigated importance conditioning factors predicting occurrences using mentioned...

Abstract. Multiple landslide events occur often across the world which have potential to cause significant harm both human life and property. Although a substantial amount of research has been conducted address mapping landslides using Earth observation (EO) data, several gaps uncertainties remain with developing models be operational at global scale. The lack high-resolution globally distributed event-diverse dataset for segmentation poses challenge in machine learning that can accurately...

Abstract. Multiple landslide events occur often across the world which has potential to cause significant harm both human life and property. Although a substantial amount of research been conducted address mapping landslides using Earth Observation (EO) data, several gaps uncertainties remain when developing models be operational at global scale. To this issue, we present HR-GLDD, high resolution (HR) dataset for composed instances from ten different physiographical regions globally: South...

Landslides represent a significant geological hazard, particularly in mountainous regions where ground deformations can lead to devastating impacts on infrastructure, ecosystems, and communities. The Belluno Province, situated the Veneto region of northeastern Italy, is characterized by its complex topography features, rendering it susceptible landslide occurrences. To mitigate risks associated with these natural phenomena, effective hazards mapping essential. This study explores integration...

Structural Health Monitoring (SHM) represents a very powerful tool to assess the health condition of buildings. In recent years, growing availability high-resolution SAR satellite images has made possible application multi-temporal Interferometric Synthetic Aperture Radar (MT-InSAR) techniques for structural monitoring purposes, with high precision, low costs, timesaving, and possibility investigate wide areas. However, comprehensive validation effectiveness MT-InSAR in this field not been...

Mapping landslides in space has gained a lot of attention over the past decade with good results. Current methods are primarily used to generate event inventories, but multi-temporal (MT) inventories rare, even manual landslide mapping. Here, we present an innovative deep learning strategy employing transfer learning. This allows our Attention Deep Supervision multi-scale U-Net model be adapted detection tasks new regions. method also provides flexibility retrain pretrained detect both rain...

Accurate landslide early warning systems are a trustworthy risk-reduction method that may greatly minimize human and economic losses. Several machine learning algorithms have been investigated for this goal, underlying the impressive potential in prediction capability of Deep Learning (DL) models. Despite this, only DL models evaluated so far long short-term memory (LSTM) Gated Recurrent Unit (GRU) algorithms. alternative algorithms, however, appropriate time series forecasting problems. In...

In recent years, Multi-Temporal InSAR (MT-InSAR) techniques have become popular as an effective way of monitoring volcanic areas and the ground deformations caused by their activity. This work used MT-InSAR to investigate Lipari, Salina, Vulcano islands belonging Aeolian archipelago (southern Tyrrhenian Sea, Italy). particular, three distinct approaches were applied compared: PS, SBAS, IPTA. These compared with in-situ measurements from GNSS network managed INGV-OE private operators...

<p>In the last decades, extreme meteorological events, such as wind disturbances, have increased their frequency and strength due to effects of climate changes are expected further intensify in future. The strong winds combined with heavy rain modify water-soil interaction soil mechanics raising landslides hazard. An example damages caused by this atmospheric phenomenon is windstorm Vaia, that affected north-eastern part Italy October...

The activity of Aso and Sakurajima volcanoes is monitored through interferometric analysis SAR data, ALOS Palsar-2 Sentinel-1, to obtain caldera displacements over the observed period, from 2014 2018 investigate deformations correlation with inflation deflation cycles magma plumbing system, eruptions diking. time series, calibrated global navigation satellite System (GNSS) measurements, are used identify pre post seismic effects Mw 7.0 Kumamoto earthquake, occurred in April 2016, on volcano...

Following extreme climate events, a timely and detailed landslide mapping is necessary to determine which areas have been most affected support civil protection in rescue operations. Moreover, the monitoring of slope instabilities can lead an appropriate hazard risk assessment effective design remediation works. The integration optical SAR remote sensing data acquired by spaceborne sensors plays key role these types evaluations. Optical perform better terms spatial temporal resolution;...

Multiple landslide events occur often across the world which have potential to cause significant harm both human life and property. Although a substantial amount of research has been conducted address mapping landslides using Earth Observation (EO) data, several gaps uncertainties remain when developing models be operational at global scale. To this issue, we present HR-GLDD, high-resolution (HR) dataset for composed instances from ten different physiographical regions globally: South...

Sakurajima volcano is one of the most active volcanoes in Japan and caused powerful eruption last century (1914). Volcanic eruptions are often preceded by ground deformations. The activity monitored through interferometric analysis SAR data, ALOS Palsar-2 Sentinel-1, to obtain caldera displacements over observed period, from 2015 2019 investigate deformations correlated with inflation deflation cycles magma plumbing system, diking. Time series calibrated validated GEONET stations Geospatial...

<p>Multi-temporal landslide inventories are crucial for understanding the changing dynamics and states of activity masses. However, mapping landslides over space time is challenging as it requires lots resources to delineate bodies affected areas. With current advances in artificial intelligence models acquisition very high-resolution satellite imageries, need map not just spatially, but also temporally, has become evident. Generating multi-spatiotemporal can allow improve our...

<p>Landslide susceptibility maps are often not validated after significant landslide events. In this work, we analyse the impact of Vaia windstorm on activity in Belluno province (Veneto Region, NE, Italy). The storm hit area October 27-30, 2018, causing 8,679 ha damaged forests and widespread landslides. As shown case Vivian (1990) Lothar (1999) (Switzerland), extreme meteorological events can influence slope stability three to ten years (Bebi et al 2019). Through...

<p>Frequent and extreme meteorological events can lead to an increase in landslide hazard. A multi-temporal inventory plays essential role monitoring slope processes over time forecasting future evolution. In recent years, the province of Belluno (Veneto Region, NE Italy) was affected by two relevant intense phenomena that occurred on October 27-30, 2018 (i.e. windstorm Vaia) December 4-6, 2020. Both were characterized heavy rainfall up 600 mm 72 hours, triggering widespread...

Abstract. In the domain of landslide risk science, susceptibility mapping (LSM) is very important as it helps spatially identify potential landslide-prone regions. This study used a statistical ensemble model (Frequency Ratio and Evidence Belief Function) two machine learning (ML) models (Random Forest XG-Boost) for LSM in Belluno province (Veneto Region, NE Italy). The investigated importance conditioning factors predicting occurrences using mentioned models. this paper, we evaluated...