Search

Local governments are faced with increasing levels of risk from extreme hydro-meteorological events such as (tropical) storms, flooding, landslides, drought, heatwaves, wildfires, etc. The frequency and interaction of these events, also in combination with other events that do not have a direct climate driver, makes that it is likely that many areas are faced with higher impacts from compounding events. Global trends such as population growth, urbanization, increased dependency on technology also contributed to larger exposure and vulnerability. In order to plan for future developments, and for reducing the increasing levels of risk, local governments require to plan ahead and evaluate the options available for reducing the risk under future scenarios. For this task Spatial Decision Support Systems are required that allow local governments to make informed decisions, considering the current and future levels of risk. RiskChanges is a Spatial Decision Support System for the analysis of current and future multi-hazard risk at a local level, in order to analyze optimal risk reduction alternatives. The system is developed by the University of Twente in collaboration with the Asian Institute of Technology, GeoInformatics Centre. RiskChanges ( http://www.riskchanges.org/ ) is an Open-Source, web-based tool, based on a series of python scripts, which are integrated into a Graphical User Interface. The tool includes several major features: multi-hazard, multiple assets, a vulnerability curve database, multi-user approach, comparison of risk, and spatial analysis. Users can upload their own datasets (in the form of hazard maps, elements-at-risk maps, administrative unit maps, and vulnerability curves). The tool contains an open-source vulnerability curve database, allowing to sharing of physical vulnerability curves among users. Multiple users can collaborate on the same project, and provide different input data. The multi-hazard feature allows performing the risk assessment for multiple natural and manmade hazard interactions. Exposure and vulnerability are combined in a loss calculation for each combination of element-at-risk and hazard. Loss maps are integrated into a risk map, where the user indicates the interaction between the hazard types. The system allows to analyze the risk of multiple asset types with different spatial characteristics. Users can compare the risk for the current situation and future scenarios and/or planning alternatives.

Frequent and extreme meteorological events can lead to an increase in landslide hazard. A multi-temporal inventory plays an essential role in monitoring slope processes over time and forecasting future evolution. In recent years, the province of Belluno (Veneto Region, NE Italy) was affected by two relevant and intense meteorological phenomena that occurred on October 27-30, 2018 (i.e. windstorm Vaia) and on December 4-6, 2020. Both events were characterized by heavy rainfall of up to 600 mm in 72 hours, triggering widespread landslides throughout the area. The analyses conducted on some local rain gauges in the sectors most affected by each storm show very high return periods (over 100 years) for both events, even though they occurred in a two-year time frame. This study aims to evaluate whether these strong meteorological phenomena are characterized by an increase in their frequency in the province of Belluno and to see what influence they have on slope instabilities, which are important for assessing landslide risk. The rainfall data available since 1950 have been investigated through statistical analysis to achieve these goals. The spatial and temporal evolution of slope instabilities has been examined through remote sensing techniques to compare landslides triggered in 2018 and 2020 with past instability phenomena in the same area. The results show the importance of multi-temporal databases for landslide hazard assessment after extreme meteorological events at the regional scale.

The evolution of the Sicily Channel Rift Zone (SCRZ), located south of the Central Mediterranean, is thought to accommodate the regional tectonic stresses of the Calabrian subduction system. It is unclear whether the rifting of the SCRZ is passive from far-field extensional stresses or active from mantle upwelling beneath. To map the structure and dynamics of the region, we measure Rayleigh- and Love-wave phase velocities from ambient seismic noise and invert for an isotropic 3-D shear-velocity and radial anisotropic model. Variations of crustal S-velocities coincide with topographic and tectonic features: slow under high elevation, fast beneath deep sea. The Tyrrhenian Sea has a VSV); areas experiencing compression and subduction-related volcanism have negative anisotropy (VSH

Multi-temporal landslide inventories are crucial for understanding the changing dynamics and states of activity of landslide masses. However, mapping landslides over space and time is challenging as it requires lots of time and resources to delineate landslide bodies for affected areas. With the current advances in artificial intelligence models and acquisition of very high-resolution satellite imageries, the need to map landslides not just spatially, but also temporally, has become evident. Generating multi-spatiotemporal landslide inventories can allow to improve our understanding of evolving landslides and landslide re-activations, addressing the changing susceptibilities, and the associated risks to elements-at-risk. Furthermore, as a result of having multi-temporal inventories, the temporal probability of occurrence of landslides can also be investigated with the help of envelop curves based on variables like rainfall duration, intensity, cumulative rainfall, antecedent rainfall. Therefore, in this endeavour, we have developed a model that generates multi-temporal landslide inventories for some of the most affected landslide regions by using several inventories around the world, for example, in Nepal (Gorkha earthquake of 2015). This study is the first attempt to map landslides over space and time using the state-of-the-art artificial intelligence models and gives a new perspective at mapping landslides through a temporal lens. Subsequently, the modelled outputs are utilised to assess and understand the changing dynamic behaviour of past landslides.

Multiple landslide events are common around the globe. They can cause severe damage to both human lives and infrastructures. Although a huge quantity of research has been shaped to address rapid mapping of landslides by optical Earth Observation (EO) data, various gaps and uncertainties are still present when dealing with cloud obscuration and 24/7 operativity. To address the issue, we explore the usage of SAR data over the eastern Iburi sub-prefecture of Hokkaido, Japan. In the area, about 8000 co-seismic landslides were triggered by an Mw 6.6 earthquake on 6 September 2018, at 03.08 local time (JST). In the following study, we modify a Deep Learning (DL) convolutional neural network (CNN) architecture suited for pixel-based classification purposes, the so-called Attention U-Net (Attn-U-Net) and we employ it to evaluate the potential of bi- and tri-temporal SAR amplitude data from the Sentinel-1 satellite and slope angle to map landslides even under thick cloud cover. Four different datasets, composed of two different band combinations per two satellite orbits (ascending and descending) are analyzed. Moreover, the impact of augmentations is evaluated independently for each dataset. The models’ predictions are compared against an accurate landslide inventory obtained by manual mapping on pre-and post-event PlanetScope imagery through F1-score and other common metrics. The best result was yielded by the augmented ascending tri-temporal SAR composite image (61% F1-score). Augmentations have a positive impact on the ascending Sentinel-1 orbit, while metrics decrease when augmentations are applied on descending path. Our findings demonstrate that combining SAR data with other data sources may help to map landslides quickly, even during storms and under deep cloud cover. However, further investigations and improvements are still needed, this being one of the first attempts in which the combination of SAR data and DL algorithms are employed for landslide mapping purposes.

To understand the factors that make certain areas especially prone to landslides, statistical approaches are typically used. The interpretation of statistical results in areas characterised by complex geological and geomorphological patterns can be challenging, and this makes the understanding of the causes of landslides more difficult. In some cases, landslide inventories report information on the state of activity of landslides, adding a temporal dimension that can be beneficial in the analysis. Here, we used an inventory covering a portion of Northwestern Turkey to demonstrate that active and relict landslides (that is, landslides that occurred in the past and are now stabilised) could be related to different triggers. To do so, we built two landslide susceptibility models and observed that the spatial patterns of susceptibility were completely distinct. We found that these patterns were correlated with specific controlling factors, suggesting that active landslides are regulated by current rainfalls while relict landslides may represent a signature of past earthquakes on the landscape. The importance of this result resides in that we obtained it with a purely data-driven approach, and this was possible because the active/relict landslide classification in the inventory was accurate.

In this study, a new tool for quantitative, data-driven susceptibility zoning (SZ) is presented. The SZ plugin has been implemented as a QGIS plugin to maximize its operational use within the geoscientific community. QGIS is in fact a commonly used open-source geographic information system. We have scripted the plugin in Python, and developed it as a collection of functions that allow one to pre-process the input data, calculate the susceptibility, and then estimate the quality of the classification results. The susceptibility zoning can be carried out via a number of classifiers including weight of evidence, frequency ratio, logistic regression, random forest, support vector machine, and decision tree. The plugin allows one to use any kind of mapping units, to fit the model, to test it via a k-fold cross-validation, and to visualize the relative receiving operating characteristic (ROC) curves. Moreover, a new classification method of the susceptibility index (SI) has been implemented in the SZ plugin. A typical workflow of the SZ plugin is described, and its application for landslide susceptibility zoning in Northeast India is reported. The data of the predisposing factors used are open, and the analysis has been carried out using a logistic regression and weight of evidence models. The corresponding area under the curve of the relative ROC curves reflects an optimal model prediction capacity. The user-friendly graphical interface of the plugin has allowed us to perform the analysis efficiently in few steps.

Event-based landslide inventories are important for analyzing the relationship between the intensity of the trigger (e.g., rainfall, earthquake) and the density of the landslides in a particular area as a basis for the estimation of the landslide probability and the conversion of susceptibility maps into hazard maps required for risk assessment. They are also crucial for the establishment of local rainfall thresholds that are the basis of early warning systems and for evaluating which land use and land cover changes are related to landslide occurrence. The completeness and accuracy of event-based landslide inventories are crucial aspects to derive reliable results or the above types of analyses. In this study, we generated a relatively complete landslide inventory for the 2018 monsoon landslide event in the state of Kerala, India, based on two inventories that were generated using different methods: one based on an object-based image analysis (OBIA) and the other on field surveys of damaging landslides. We used a collaborative mapping approach based on the visual interpretation of pre- and post-event high-resolution satellite images (HRSIs) available from Google Earth, adjusted the two inventories, and digitized landslides that were missed in the two inventories. The reconstructed landslide inventory database contains 4728 landslides consisting of 2477 landslides mapped by the OBIA method, 973 landslides mapped by field survey, 422 landslides mapped both by OBIA and field methods, and an additional 856 landslides mapped using the visual image (Google Earth) interpretation. The dataset is available at https://doi.org/10.17026/dans-x6c-y7x2 (van Westen, 2020). Also, the location of the landslides was adjusted, based on the image interpretation, and the initiation points were used to evaluate the land use and land cover changes as a causal factor for the 2018 monsoon landslides. A total of 45 % of the landslides that damaged buildings occurred due to cut-slope failures, while 34 % of those having an impact on roads were due to road cut-slope failures. The resulting landslide inventory is made available for further studies.

Kathmandu Kyirong Highway (KKH) is one of the most strategic Sino-Nepal highways. Low-cost mitigation measures are common in Nepalese highways, however, they are not even applied sufficiently to control slope instability since the major part of this highway falls still under the category of feeder road, and thus less resources are made available for its maintenance. It is subjected to frequent landslide events in an annual basis, especially during monsoon season. The Gorkha earthquake, 2015 further mobilized substantial hillslope materials and damaged the road in several locations. The aim of this research is to access the dynamic landslide susceptibility considering pre, co and post seismic mass failures. We mapped 5,349 multi-temporal landslides of 15 years (2004–2018), using high resolution satellite images and field data, and grouped them in aforementioned three time periods. Landslide susceptibility was assessed with the application of ‘certainty factor’ (CF). Seventy percent landslides were used for susceptibility modelling and 30% for validation. The obtained results were evaluated by plotting ‘receiver operative characteristic’ (ROC) curves. The CF performed well with the ‘area under curve’ (AUC) 0.820, 0.875 and 0.817 for the success rates, and 0.809, 0.890 and 0.760 for the prediction rates for respective pre, co and post seismic landslide susceptibility. The accuracy for seismic landslide susceptibility was better than pre and post-quake ones. It might be because of the differences on completeness of the landslide inventory, which might have been possibly done better for the single event based co-seismic landslide mapping in comparison with multitemporal inventories in pre and post-quake situations. The results obtained in this study provide insights on dynamic spatial probability of landslide occurrences in the changing condition of triggering agents. This work can be a good contribution to the methodologies for the evaluation of the dynamic landslide hazard and risk, which will further help to design the efficient mitigation measures along the mountain highways.

The Intergovernmental Panel on Climate Change (IPCC) reports provide policy-relevant insights about climate impacts, vulnerabilities and adaptation through a process of peer-reviewed literature assessments underpinned by expert judgement. An iconic output from these assessments is the burning embers diagram, first used in the Third Assessment Report to visualize reasons for concern, which aggregate climate-change-related impacts and risks to various systems and sectors. These burning embers use colour transitions to show changes in the assessed level of risk to humans and ecosystems as a function of global mean temperature. In this Review, we outline the history and evolution of the burning embers and associated reasons for concern framework, focusing on the methodological approaches and advances. While the assessment framework and figure design have been broadly retained over time, refinements in methodology have occurred, including the consideration of different risks, use of confidence statements, more formalized protocols and standardized metrics. Comparison across reports reveals that the risk level at a given temperature has generally increased with each assessment cycle, reflecting accumulating scientific evidence. For future assessments, an explicit, transparent and systematic process of expert elicitation is needed to enhance comparability, quality and credibility of burning embers.

For many regions on Mars, the surface composition and its geological history have been debated in the literature. Because of the limited surface coverage of in situ measurements, either new data or new processing methodologies are required to get a better understanding of the Martian geology. This paper presents the results of a multivariate, unsupervised, analysis on underutilized CRISM (Compact Reconnaissance Imaging Spectrometer) multispectral mapping mode data set for surface type analysis. The devised summary products of Pelkey et al. (2007) and Viviano-Beck et al. (2014) are averaged for the CRISM ~5° × 5° mosaic grids and used to analyze the variability in this data set. The averaged summary product values are studied using correlation coefficients calculated between summary products, as well as the correlation with dust coverage and elevation. The degree of correlation is used to interpret the summary products for the global distribution of mafic and secondary minerals and the effect of external factors such as ice, atmosphere, and dust coverage. With unsupervised clustering, all grid pixels are classified based on the spectral variability. These clusters are plotted as global maps and interpreted for geological variations in the CRISM data. Some clusters spatially correspond with previously recognized compositional distinct regions Northern Lowlands, Southern Highlands, Meridiani Planum, Syrtis Major, and Nili Fossae. Several other clusters are described here for the first time such as Solis Planum, Ophir Planum, and Hellas Basin. These regions cover known geological units, but the interpretation of the spectral variability is uncertain whether it relates to geological or external factors.

Frequent landslide events affect the Kathmandu Kyirong Highway (KKH), one of the most strategic Sino-Nepal highways, with multiple social effects. Amongst them, the impacts on local tourism, although being substantial, have not been studied so far. The aim of this research is to analyze the characteristics of such landslides and their influence on road damages and/or blockages as well as on local tourism industry. We analyzed the co-seismic landslides triggered by the Gorkha Earthquake, 2015 (7.8 Mw), the post-seismic landslides that occurred during the monsoons following the earthquake, as well as landslides which occurred or reactivated in 2018, with relation to the damage that they caused to the highway. High resolution satellite images from 2015 to 2018, and field data were used for the analysis. The Langtang avalanche that locates off the highway was also mapped due to its high impacts on tourism. Between 2015 and 2018, the number of road damaging landslides in the Betrawati-Rasuwagadhi section of KKH (where Dhunche and Syafrubesi towns are located) was 101 in the main track (MT) and 103 in the new track (NT), with respective average density of 1.46/km and 3.63/km. The dominant observed landslide types were debris slides and rock falls. Landslides were mostly concentrated in the locations with the following characteristics: 1) having higher elevated area, 2) being located with the ‘main central thrust’ and other lineaments’ belts, 3) belonging to the Proterozoic lesser Himalayan rocks, 4) having a slope gradient of 25°–45°, 5) having northern, western and southern slope aspect, 6) being subjected to average annual rainfall of higher than 1,000 mm, and 7) having less than 4 km distance from the past earthquake epicenters. The results further indicated that 7 rain-induced and 4 co- and post-seismic landslides have great impact on tourist flows. An impact analysis was also assessed through a door to door questionnaire survey with local hotel operators from Dhunche and Syafrubesi towns (n = 29 + 31). The results reveal that out of six rigorously affected sectors by landslides leading to road blockage, tourism business is the most impacted livelihood sector in these towns. The reduction of visitors in different hotels ranged from 50%–100% in Dhunche and 70%–100% in Syafrubesi for the first year aftermath of the tremor. This is higher than the respective 5%–50% tourist reduction due to rain-induced landslides. Using as a reference the base year 2014, the income loss of hotels in both towns was found to be 50%–100% in 2015, 20%–100% in 2016, 5%–75% in 2017, and similar to 35% in 2018. These results provide insights on the synergic effect of contributing factors for cut slope as well as down slope instability along mountainous motorways and their impact on income sources for local communities.

Botswana experienced a Mw 6.5 earthquake on 3rd April 2017, the second largest earthquake event in Botswana's recorded history. This earthquake occurred within the Limpopo-Shashe Belt, ~350 km southeast of the seismically active Okavango Rift Zone. The region has no historical record of large magnitude earthquakes or active faults. The occurrence of this earthquake was unexpected and underscores our limited understanding of the crustal configuration of Botswana and highlight that neotectonic activity is not only confined to the Okavango Rift Zone. To address this knowledge gap, we applied a regularized inversion algorithm to the Bouguer gravity data to construct a high-resolution crustal thickness map of Botswana. The produced crustal thickness map shows a thinner crust (35–40 km) underlying the Okavango Rift Zone and sedimentary basins, whereas thicker crust (41–46 km) underlies the cratonic regions and orogenic belts. Our results also show localized zone of relatively thinner crust (~40 km), one of which is located along the edge of the Kaapvaal Craton within the MW 6.5 Botswana earthquake region. Based on our result, we propose a mechanism of the Botswana Earthquake that integrates crustal thickness information with elevated heat flow as the result of the thermal fluid from East African Rift System, and extensional forces predicted by the local stress regime. The epicentral region is therefore suggested to be a possible area of tectonic reactivation, which is caused by multiple factors that could lead to future intraplate earthquakes in this region.

Landslide susceptibility map (LSM) provide useful tool for decision makers in hazard mitigation concerns. In the present paper, a novel hybrid block-based neural network model (HBNN) for the purpose of producing high-resolution LSM was developed. This hybrid approach was found through the mixture of expert modular structures and divide-and-conquer strategy incorporated with genetic algorithm (GA). The introduced HBNN then was applied on southern part of Guilan province (north of Iran) using 14 causative factors covering topographic and geomorphologic features, and geological and tectonical factors as well as hydrology, land data, and climate conditions. The landslide inventory map was provided using a synergy work from monitored events, interpretation of aerial photographs, and carried out geotechnical investigations in the area as well as field surveys. To insight, the predictability of proposed HBNN was compared with two developed models using multilayer perceptrons (MLPs) and generalized feed forward neural network (GFFN). The generated LSM was validated using receiver operating characteristic (ROC) curves, statistical error indices, and sensitivity and weight analyses as well as monitored landslides. Based on the compared metrics, HBNN with 86.52% and 90.15% in prediction and success rate as well as 89.36% for precision-recall curve demonstrated more consistent tool for future landslide susceptibility zonation. This implies on capability of developed HBNN in producing higher resolution and more reliable LSM for urban and land-use planners.

Remote sensing images have long been recognized as useful for the detection of building damages, mainly due to their wide coverage, revisit capabilities and high spatial resolution. The majority of contributions aimed at identifying debris and rubble piles, as the main focus is to assess collapsed and partially collapsed structures. However, these approaches might not be optimal for the image classification of façade damages, where damages might appear in the form of spalling, cracks and collapse of small segments of the façade. A few studies focused their damage detection on the façades using only post-event images. Nonetheless, several studies achieved better performances in damage detection approaches when considering multi-temporal image data. Hence, in this work a multi-temporal façade damage detection is tested. The first objective is to optimally merge pre- and post-event aerial oblique imagery within a supervised classification approach using convolutional neural networks to detect façade damages. The second objective is related to the fact that façades are normally depicted in several views in aerial manned photogrammetric surveys; hence, different procedures combining these multi-view image data are also proposed and embedded in the image classification approach. Six multi-temporal approaches are compared against 3 mono-temporal ones. The results indicate the superiority of multi-temporal approaches (up to ~25% in f1-score) when compared to the mono-temporal ones. The best performing multi-temporal approach takes as input sextuples (3 views per epoch, per façade) within a late fusion approach to perform the image classification of façade damages. However, the detection of small damages, such as smaller cracks or smaller areas of spalling, remains challenging in this approach, mainly due to the low resolution (~0.14 m ground sampling distance) of the dataset used.

Recovering from major earthquakes is a challenge, especially in mountainous environments where postearthquake hazards may cause substantial impacts for prolonged periods of time. Although such impacts were reported in the 1923 Great Kantō earthquake and the 1999 Chi-Chi earthquake, careless reconstruction in hazard-prone areas and consequently huge losses were witnessed following the 2008 Wenchuan earthquake in the Sichuan province of China, as several reconstructed settlements were severely damaged by mass movements and floods. In order to summarize experiences and identify problems in the reconstruction planning, a monitoring of one of the settlements, the town of Longchi, was carried out by image interpretation and field investigation. Seven inventories containing buildings, farmlands, roads and mitigation measures were made to study the dynamics of elements at risk and exposure over a period of 11 years. It was found that the total economic value of the newly reconstructed buildings was several times more than in the preearthquake situation in 2007, because of enormous governmental investment. Postseismic hazards were not sufficiently taken into consideration in the recovery planning before the catastrophic debris flow disaster in 2010. As a result, the direct economic loss from postseismic disasters was slightly more than the loss caused by the Wenchuan earthquake itself. The society showed an impact-adapt pattern, experiencing losses from disasters and then gaining resistance by abandoning buildings in hazard-prone areas and installing mitigation measures. The locations potentially exposed to postearthquake hazards were summarized, and a possible timetable for reconstruction was proposed. Problems might be encountered in hazard assessment, and possible solutions were discussed.

There is considerable interest in optimizing geothermal exploration techniques via the mapping of alteration and evaporate mineralisation, as well as of thermal emissions associated with geothermally active areas on the Earth’s surface. Optical and thermal satellite sensor technologies, improvements in processing algorithms and the means for large scale (e.g. 1:250,000) spatial data distribution are required for detecting both these attributes. The extensive visible, -near, -shortwave and thermal infrared (VNIR-SWIR-TIR) data archive acquired by the multi-spectral Advanced Spaceborne Thermal Emission Reflectance Radiometer (ASTER) provides a rich source of geoscience related imagery for geothermal exploration. Examples of generating large scale mosaicked ASTER imagery to provide province to continental mineral mapping have been undertaken in areas including such as Australia, western USA, Namibia and Zagros Mountains Iran. In addition, ASTER’s thermal infrared imagery also provides night time land surface temperature (LST) estimates relevant for detecting possible geothermal related anomalies. This study outlines existing methods for the application of ASTER data for geothermal exploration in East Africa. The study area encompasses a section of the East African Rift System across the Tanzanian and Kenyan border. The area includes rugged volcanic terrain which has had geological mapping of limited coverage at detailed scales, from various heritages and mapping agencies. This study summarizes the technology, the processing methodology and initial results in applying ASTER imagery for such compositional and thermal anomaly mapping related to geothermal activity. Fields observations have been used from the geothermal springs of Lake Natron, Tanzania, and compared with ASTER derived spectral composition and land surface temperature results. Published geothermal fields within the Kenyan portion of the study area have also been incorporated into this study.

Ground surface topography influence the spatial distribution of earthquake induced ground shaking. This study shows the influence of topography on seismic amplification during the 2005 Kashmir earthquake. Earth surface topography scatters and reflects seismic waves, which causes spatial variation in seismic response. We perform a 3D simulation of the 2005 Kashmir earthquake in Muzaffarabad with spectral finite element method. The moment tensor solution of the 2005 Kashmir earthquake is used as the seismic source. Our results show amplification of seismic response on ridges and de-amplification in valleys. It is found that slopes facing away from the source receive an amplified seismic response, and that 98 % of the highly damaged areas are located in the topographically amplified seismic response zone.

Geospatial information acquired with Unmanned Aerial Vehicles (UAV) provides valuable decision-making support in many different domains, and technological advances coincide with a demand for ever more sophisticated data products. One consequence is a research and development focus on more accurately referenced images and derivatives, which has long been a weakness especially of low to medium cost UAV systems equipped with relatively inexpensive inertial measurement unit (IMU) and Global Navigation Satellite System (GNSS) receivers. This research evaluates the positional accuracy of the real-time kinematics (RTK) GNSS on the DJI Matrice 600 Pro, one of the first available and widely used UAVs with potentially surveying-grade performance. Although a very high positional accuracy of the drone itself of 2 to 3 cm is claimed by DJI, the actual accuracy of the drone RTK for positioning the images and for using it for mapping purposes without additional ground control is not known. To begin with, the actual GNSS RTK position of reference center (the physical point on the antenna) on the drone is not indicated, and uncertainty regarding this also exists among the professional user community. In this study the reference center was determined through a set of experiments using the dual frequency static Leica GNSS with RTK capability. The RTK positioning data from the drone were then used for direct georeferencing, and its results were evaluated. Test flights were carried out over a 70 x 70 m area with an altitude of 40 m above the ground, with a ground sampling distance of 1.3 cm. Evaluated against ground control points, the planimetric accuracy of direct georeferencing for the photogrammetric product ranged between 30 and 60 cm. Analysis of direct georeferencing results showed a time delay of up to 0.28 seconds between the drone GNSS RTK and camera image acquisition affecting direct georeferencing results.

A landslide inventory is indispensable for determination of landslide susceptibility, hazard, risk assessment and disaster mitigation strategies. These inventories were traditionally developed using manual digitization of remote sensing images and aerial photographs, and pixel-based image classification. Recently, Object-Based Image Analysis (OBIA) supersedes visual interpretation and pixel-based methods. OBIA utilizes spectral, textural, contextual, morphological and topographical information in remote sensing images. However, OBIA-based landslide detection methods are often designed for specific areas and remote sensing dataset. The aim of this study is to evaluate the transferability of three published OBIA landslide detection methods for semi-automated landslide detection in the Himalaya mountainous region of northern Pakistan. A SPOT-6 multi-spectral image with Advanced Land Observing Satellite (ALOS) Phased Array type L-band Synthetic Aperture Radar (PALSAR) Digital Elevation Model (DEM) derivatives, i.e. slope, aspect, hill-shade, relief, elevation and stream network are used for landslide detection using eCognition developer software. The three published methods scale parameters for image segmentation and parameter thresholds are evaluated first. It is observed that the aforementioned methods are not directly applicable to our study area and remote sensing datasets. Therefore, an alternate (proposed) method is developed for semi-automated landslide detection. Accuracy assessment of the selected methods and proposed method is assessed by Precision, Recall and F1 measures. Using the proposed method, a total of 357 landslides are detected with 91.46% Precision, 93.31% Recall and 92.38% F1 measure accuracy.

Successive major landslides during October and November 2018 in Baige village, eastern Tibet, dammed the Jinsha River on two occasions, and the subsequent dam breaches instigated a multi-hazard chain that flooded many towns downstream. Analysis of high-resolution aerial images and field investigations unveiled three potentially unstable rock mass clusters in the source area of the landslides, suggesting possible future failures with potential for river-damming and flooding. In order to evaluate and understand the disaster chain effect linked to the potentially unstable rock mass, we systematically studied the multi-hazard scenarios through an integrated numerical modelling approach. Our model begins with an evaluation of the probability of landslide failure, including runout and river damming, and then addresses the dam breach and resultant flood—hence simulating and visualising an entire disaster chain. The model parameters were calibrated using empirical data from the two Baige landslides. Then, we predict the future cascading hazards via seven scenarios according to all possible combinations of potential rock mass failure. For each scenario, the landslide runouts, dam-breaching, and flooding are numerically simulated with full consideration of uncertainties among the model input parameters. The maximum dam breach flood extent, depth, velocity, and peak arrival time are predicted at sequential sites downstream. As a first attempt to simulate the full spectrum of a landslide-induced multi-hazard chain, our study provides insights and substantiates the value provided by multi-hazard modelling. The integrated approach described here can be applied to similar landslide-induced chains of hazards in other regions.

This thesis focuses on changes in the precipitation regime in the Maritime Continent. The Maritime Continent is the region between the Indian and Pacific oceans consisting of the Indonesian and Philippine archipelagos and countless other smaller islands. The Maritime Continent connects the Indian and Pacific oceans and plays a crucial role in the world’s climate. Land-sea interactions dominate precipitation in the Maritime Continent. The precipitation is heavily influenced by seasonal monsoons and oceanic oscillations such as the El Niño Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO). Climate change influences many aspects of the climate in South-East Asia. While changes in air and sea temperatures are well documented, the Maritime Continent precipitation changes are understudied. Lack of data and high temporal and spatial variability, among other factors, lead to difficulties in forming a conceptual model of historical and future change in precipitation regime in the Maritime Continent. This study contributes to filling this gap. This study applies the precipitation regime shift – a rapid change from one stable regime to another, to explain historical variations in precipitation in the Maritime Continent. The study suggests that the interaction between the PDO and ENSO creates alternating precipitation regimes in the Maritime Continent and Indochina peninsula. The precipitation regime changes manifest differently in different parts of the Maritime Continent but can be observed almost everywhere in this region. This thesis presents three separate lines of evidence for the regime shift's existence and its connection to the PDO. Chapters 1 and 2 present evidence of the regime shift in historical datasets. Chapter 3 looks for evidence of the regime shift in historical and future climate model simulations. Chapter 4 investigates the use of machine learning to predict long-term oscillations in the atmospheric and the ocean that lead to a precipitation regime shift. Chapters 1 and 2 present statistical evidence for the regime shift in different datasets and using different methods. Chapter 1 uses long-period monthly data for precipitation, sea surface temperature, Pacific Decadal Oscillation, and El Niño Southern Oscillation. By separating precipitation and sea surface temperature data by different combinations of PDO and ENSO phases, the results demonstrate that statistically different regimes are formed. This part of the study also shows how changes in the PDO phase can cause rapid precipitation regime changes in the Maritime Continent. A key piece of evidence for existence of the regime shift comes from Chapter 2, where a novel Bayesian method for detecting precipitation regime shifts is presented. The Bayesian regime detection method demonstrates that precipitation regime shift models explain historical changes better than linear change models. Moreover, the algorithm, using only precipitation data as input, can detect a regime shift at the precise time we expect it to occur – coinciding with the PDO phase change. The regime shift detection method requires a large amount of data. Hereto we used the state-of-the-art precipitation dataset of the South Asian Climate Assessment and Dataset (SACA&D). SACA&D is a daily observational dataset developed at KNMI in collaboration with Indonesian National Meteorological Service BMKG and other national meteorological services. SACA&D offers the most extensive collection of precipitation observations in the Maritime Continent available today. Chapter 3 presents a separate line of evidence for precipitation regime shifts based on the model simulations. This study uses a simplified version of Bayesian regime shift detection to analyse precipitation regimes in the Maritime Continent simulated by CMIP6 climate models. The results show that the models that better reproduce the historical precipitation regime also show regime shifts occurring in the future simulations. The final Chapter 4 investigates the use of machine learning in predicting long-period oscillations in the ocean and the atmosphere that govern the precipitation regime changes in the Maritime Continent. Different neural network architectures are tested in predicting Madden-Julian Oscillation and results are compared to traditional approaches. Both models trained to base prediction on spatial patters and models that use temporal patters demonstrate decent performance. The results demonstrate how modern machine learning models can improve seasonal prediction, including predicting the potential shift in precipitation regimes. In conclusion, this study presents a coherent and compelling conceptual model of historical changes in the Maritime Continent precipitation regime. The Maritime Continent's precipitation regime is governed by the Pacific Decadal Oscillation and goes through alternating drier and wetter regimes following the PDO cycle. The presence of the regime shift is demonstrated in several independent datasets and using different methods. While natural variability still plays a large role in determining day-to-day weather, the current precipitation regime shifts the precipitation probability distribution towards wetter or drier conditions. The rapid changes in precipitation regime have far-reaching consequences for adaptation planning and understanding climate change processes in general. The rate of change often plays a more prominent role in determining impacts than the magnitude of change. The regime shift creates a period with a high rate of change followed by perceived stability. Long stable periods create the illusion of a lack of climate changes in precipitation. The periods of fast changes can lead to overestimations of the real rate of change over a more extended period. It is critical to understand the mechanism of precipitation regime shifts in the Maritime Continent and its effect on society to develop an effective adaptation strategy for future regime shifts. This study contributes to a growing body of evidence of non-linear changes in the climate system and opens a discussion about our perception of climate change in general.

Mafic magma intrusions into silica-rich magmatic reservoirs are commonly proposed to drive the eruptions of andesitic hybrid magmas that characterise many arc volcanoes. However, interactions between contrasting magmas involve large gradients of physical and chemical properties that change over time, and the details of such processes have proven difficult to constrain. In this paper we investigate the dynamics of magma mingling and mixing using chemical and textural zoning patterns recorded in plagioclase crystals from the February 2010 eruption at Soufrière Hills Volcano, Montserrat. This eruption is considered a classic example of interactions between the andesite magma that constitutes the bulk of the erupted volume and the basaltic andesite that occurs as enclaves. We find that plagioclase crystals are characterised by two well-defined zones that record mafic–silicic magma interaction: a crystal interior, often identified by a patchy, dusty and oscillatory zoning, and an overgrowth zone (rim) of a different composition. We use the anorthite and Mg contents to track the thermal and compositional changes experienced by the crystals over time. Our results reveal that the crystal rims formed a few hours to days prior to eruption, during co-eruptive magmatic interactions. The interaction between the two magmas with contrasting rheology is likely increased by a narrowing conduit geometry towards the surface, which facilitates convection and additional interface contact of the two magmas. Our findings shed new light on the nature and timing of magmatic interactions driving the final eruptive phase at Soufrière Hills Volcano and help to propose an interpretative framework of the monitoring signals. National Research Foundation (NRF) Published version This work was funded by Macquarie University Research Excellence Scholarship (MQRES) (2015081) associated with an Australian Research Council Discovery Project (DP150100328) awarded to HH, ST and JB. F. Costa acknowledges a National Research Foundation Investigatorship Award (Grant number NRFNRFI2017-06).

The uniaxial compressive strength (UCS) and elasticity modulus (E) are two of the most quoted rock strength parameters in engineering application. Due to approved technical difficulties indirect measurements, the tendency for determining these parameters through predictive models using simpler and cheaper tests in practical oriented applications have widely been highlighted. In this paper, a new hybridized multi-objective support vector regression (MSVR) model integrated with the firefly metaheuristic algorithm (FMA) was developed to touch upon a computational method in rock engineering purposes. The optimum internal parameters were adjusted through parametric investigation using 222 physical and mechanical rock properties corresponding to a variety of quarried stones from all over Iran. The accuracy and robustness of models were evaluated using different error indices, the area under the curve for receiver operation characteristics (AUCROC), and F1-score criteria. Comparing to MSVR, the predictability level of UCS and E showed 8.35% and 5.47% improvement in hybrid MSVR-FMA. The superior and more promising results imply that hybrid MSVR-FMA as a flexible alternative can be applied for rock strength prediction in designing of construction projects. Using tow sensitivity analyses, the point load index and P-wave velocity were distinguished as the main effective factors on predicted UCS and E.

Acid stimulation to enhance the permeability of geological reservoirs is the oldest technique still in use. Acid stimulation sometimes fails due to either poor stability of the acid or due to an undesirable influence on clay minerals in the reservoir formation. Among others, clay minerals in a reservoir formation govern the permeability of the reservoir. The interaction is investigated of biodegradable chelating agents (BCA1, BCA2 and BCA3) and clay minerals kaolinite-natural (KN) and montmorillonite-K10 (MM) in a hydrothermal reactor in which the conditions of a geothermal reservoir are simulated under a CO2 environment with temperature of 230 οC and pressure of 85 bars. The porisimetry, ATR-FTIR and infrared reflectance spectroscopy show changes on surface properties and structure of MM and KN which in turn suggest chemical interaction to take place. However, XRD results show insignificant changes on structure of MM and KN. The ICP-OES, pH, conductivity and salinity results show the chemical interactions (dissolution reactions) to take place on both KN and MM and therefore strongly support the results of the earlier techniques. The results show that BCA1 and BCA3 cause greater dissolution with lower precipitation as compared to BCA2 and no-BCA. Further, the R -values from multiple linear regression show that pH and BCAs’ function groups have strong influence on the normalized concentration of dissolved ions though the p-values are insignificant. The coefficient values show that the BCAs’ functional groups have more influence than pH. BCA1 and BCA 3 would therefore be useful acids for reservoir stimulation in improving permeability especially in geothermal reservoir formations which contain clay minerals.

Events characterized by widespread landslides provide rare but valuable opportunities to investigate the spatial and size distributions of landslides in relation to seismic, climatic, geological and morphological factors. This study presents a unique event inventory for the co-seismic landslides induced by the February 25, 2018 Mw 7.5 Papua New Guinea earthquake. The mainshock rupture was dominated by reverse fault motion, and this was also the case for the aftershocks. The latter also triggered widespread landslides in combination with rainfall during the period between February 26 and March 19. We mapped approximately 11,600 landslides of which, more than 10,000 were triggered by the mainshock, with a total failed planimetric area of about 145 km². Such a large area makes this inventory the world's second-largest recorded landslide event after the 2008 Mw 7.9 Wenchuan earthquake, where the motion changed from predominantly thrust to strike-slip. Large landslides are abundant throughout the study area located within the remote Papua New Guinea Highlands. Specifically, more than half of the landslide population is larger than 50,000 m² and overall, post-seismic landslides are even larger than their co-seismic counterparts. To understand the factors controlling the distribution of landslides' occurrence and size, we combine descriptive statistics as well as more rigorous bivariate and multivariate analyses. We statistically show that the 15-day antecedent precipitation plays a role in explaining the spatial distribution of co-seismic landslides. Also, we examine four strong aftershocks (Mw ≥ 6.0) within 9 days after the mainshock and statistically demonstrate that the cumulative effect of aftershocks is the main factor disturbing steep hillslopes and causing the initiation of very large landslides, up to ~5 km². Overall, the dataset and the findings presented in this paper represent a step towards a holistic understanding of the seismic landslide hazard assessment of the entire Papua New Guinea mainland.

Strategies to cope with floods, for instance, integrated flood management, require proper flood hazard assessment. Such flood hazard assessment relies on the flood modelling approaches, which require high-quality and quantity data to produce realistic flood hazard maps. In many cities in the developing countries, also known as data scarce-areas, flood modelling is challenging as there is sparse or no observation on rainfall data and soil information used for proper model development, model calibration, and validation. However, open-source geospatial data and the NWP model rainfall product can overcome the data scarcity problem. Therefore, this Ph.D. thesis aimed to explore publicly available geospatial datasets and their integration with hydro-meteorological modelling systems to overcome the data-scarcity challenges, specifically, to explore the data of extreme rainfall, soil, and land-cover information for flash flood modelling in the urbanized catchment. The findings are summarized into three phases, as discussed in the following. The first phase of this Ph.D. study focuses on exploring soil information used for flash flood modelling in the urbanized catchment. Accordingly, the soil information that is determining the infiltration processes (i.e., Ksat, porosity, initial condition, soil matric suction, and soil depth) is derived following three different soil databases: (1) FAO soil map (SMFAO); (2) soil map derived based on the soil-landscape relationships (SMLS), and (3) soil map derived from the SoilGrids database (SMSG)). The soil information derived from these data sources is believed to be overcome the data limitation problem. However, the open-source soil databases cannot correctly consider the local features (e.g., wetlands, fragmented vegetation cover, and soil compaction), which would lead to the data quality problem. Therefore, in this study, the local features' influence on the derived soil information is numerically adjusted by incorporating the land cover data derived from the satellite image. The derived soil information is then used as the input to openLISEM integrated flood modelling system to assess their impact on flood dynamics in the urbanized catchment. The impact analysis is evaluated as the compacted and uncompacted soil condition. The results indicated that the flood dynamics are highly sensitive to different soil databases. The incorporation of soil compaction into the soil information has the largest impact on the flood dynamics in the catchment. This study showed that open-source data choice strongly influences both the simulated quantity and spatial variability of the infiltration, which directly affects runoff and flooding. On top land information in the context of Rwanda: 1) the hilly terrain, 2) UAV regulations, 3) ground-truthing in urban areas, 4) data processing capacities at the national land administration authority. In sum, although UAVs can play a key role in satisfying stakeholder needs, structural and capacity conditions currently present major challenges to exploit the potential of UAV data acquisition workflows. Triggered by tremendous regulatory obstacles to fly UAVs in Eastern Africa, the second part of this thesis analyses the interface between UAV technology and stakeholders in terms of legal and regulatory frameworks. To get a better understanding of past, present and future developments of UAV regulations, a thorough literature review and a comparative analysis of national UAV regulations was conducted. Although all UAV regulations have a similar target – minimizing the risk for people and property on the ground – a distinct heterogeneity of national regulatory frameworks was investigated. However, commonalities such as mandatory platform registration, obligatory insurance coverage and standard pilot licensing procedures indicate trends towards mature national UAV regulation. Desirable trends are seen in the coexistence of hard and soft regulations and the successful international dialogue that may eventually provide a legal framework for harmonized regulatory standards. The third part of this thesis explores the technical and operational aspect of UAV workflows and assesses different data quality measures with the aim to recommend optimal UAV data acquisition workflows. Imagery from six study areas across Europe and Africa, representing overall more than 100 different flight configurations, provided the basis for an integrated data quality assessment including various aspects of the photogrammetric processing chain. The results suggest that scene context, flight configuration, and GCP setup significantly impact the final data quality and subsequent automatic delineation of visual cadastral boundaries. Moreover, even though the root mean square error of checkpoint residuals as a commonly accepted error measure lies within a range of few centimeters in all datasets, our results show large discrepancies of the accuracy and the completeness of automatically detected cadastral features for orthophotos generated from different flight plans. With its unique combination of methods and integration of various study sites, these results and recommendations can help land professionals and bottom-up initiatives alike to optimize existing and future UAV data collection workflows. Despite the prospects and market opportunities in land administration, there is a gap between experimentation and widespread diffusion of UAV technology. To understand the dynamics of the diffusion process of UAV technology as a tool for effective land administration, the fourth part of this thesis integrates the Framework for Effective Land Administration (FELA) and the Hype Cycle concept. The majority of experts estimated UAV technology to be in a phase in which the innovation needs to overcome initial unmet expectations to foster market development and increased adoption. The results indicate a changing importance of different FELA pathways during this process, led by alternating top-down and bottom-up dynamics. Enabling laws and policies in addition to supporting governance, accountability and institutions are found to be crucial to create a UAV-friendly national ecosystem early on in the process of technology adoption and allay exaggerated expectations. Once this ecosystem has been created, market demand is expected to surge driven by partnerships, adapted standards, tech advocacy and awareness-raising campaigns, highlighting the superiority of highresolution data amongst other benefits of UAV technology. These insights can be used as a baseline to direct national strategic decisions towards the increased adoption of UAVs in land administration. In conclusion, the integrated view on the socio-technical system, including stakeholders, land administration processes and UAV technology as system elements, allows identifying interactions and positive and negative dependencies. Whereas the aspects of data quality and regulations both influence but are also impacted by other elements, land data needs were found to have a critical role in the system and predominantly affect other elements as a starting point for any land tool intervention. Representing the overlap of all three system elements, the diffusion of UAV technology in land administration takes a central place. Here, the superior quality of UAV data is a key driver throughout the diffusion process. In contrast, regulations may positively or negatively affect the emergence of UAV technology, as appropriate UAV legislation presents a vital pre-condition for technology uptake. Ultimately and potentially most importantly, land data needs stimulate the scaled implementation creating a demand for up-to-date high-resolution land data. Overall, the findings of this thesis suggest that UAV technology is a highly flexible, widely available, easy-to-use and end-user responsive data acquisition tool that complements the range of mapping tools for land administration processes side by side with satellite data and GNSS surveying. Yet, regulatory constraints, missing awareness and weak governance support currently hamper exploiting the full potential of UAV technology for land administration in many countries. of that, the effect of sealing and compaction is equally essential and nearly outweighs the differences caused by the use of different soil databases. The second part of the Ph.D. thesis is to model and analyze the high-intensity rainfall product using the WRF model, which combined two main research objectives (chap.3 & 4 in the document). The first objective is to evaluate the satellite-driven urban fraction appropriateness in the WRF model for simulating high-intensity rainfall events in urbanized areas. Three different simulations are performed in order to distill the impact of changing urban fractions and adjusted urban parameters on the simulated rainfall: The first simulation (1) StandardWPS, which is carried out using the default urban fraction with the default urban parameters; (2) UFD_Parameter, which is using the default urban fraction, but with adjusted urban parameters; and (3) Updated, which is with the updated urban fraction based on the Landsat 2016 image and the adjusted urban parameters. All model simulations are configured at high spatial (1 km) and temporal (10-minute) resolutions forced with the latest ERA5 global reanalysis dataset. The model result was validated using the rainfall observation from the gauging station and CHIRPS data. The results showed that the simulated rainfall performs better with a relatively lower error when using the updated urban fraction. The satellite-derived urban map represents a more realistic extent and intensity of the urban fraction with a heterogeneous urban fraction, which results in more realistic rainfall simulations. The second objective of the second part of the Ph.D. is to evaluate the suitability of the WRF model in simulating high-intensity rainfall events. Here, we evaluated the procedure to select the appropriate WRF parameterization combination for proper high-intensity rainfall simulation through the sensitivity analysis. The WRF model set up with the updated urban fraction is used for the WRF model simulation as the combination of microphysics, cumulus, and planetary boundary layer (i.e., MP-CP-PBL procedure). The result showed that the WRF model's ability to simulate the HIRE that can be used for flash flood modelling is highly determined by the appropriate selection of the parametrization combinations. The last phase of this Ph.D. study focuses on examining the WRF rainfall product's applicability for urban flood hazard assessment by proposing a new methodology to select the representative gridcell-rainfall events from three known WRF simulated rainfall events (HIREs). The two-step procedure is followed. Firstly, the potential gridcell-rainfall events from the WRF simulated HIREs are selected based on the given criteria. Secondly, the representative gridcell-rainfall events as a design storm of a given return period are defined using the quantile function where the quantile function is applied to the cumulative rainfall amount of each of the selected potential gridcell-rainfall events. Finally, three different gridcell-rainfall events representing the design storms varying between T = 2 and 10-year return periods are extracted for each three HIRE. The developed design storms are then compared with the design storms from the pre-established Intensity-Duration-Frequency (IDF) curves in terms of their 24-hour total rainfall amount (mm), peak intensity (mm/hr), and the time to peak intensity (minute). The constructed design storms are then applied to the openLISEM model for flood hazard modelling in Kampala's upper Lubigi catchment. The derived design storm can give an insight into the applicability and usability of the numerical weather prediction model outputs for flood modelling in the data-scarce areas. In general, the results of this study indicates that open-source database such as SoilGrids and their combination with satellite-driven land-cover data can provide soil information needed for flood modelling in the data-scarce area. Moreover, the high-intensity rainfall that has the potential to trigger the localized flood can be produced using the mesoscale WRF model. However, the procedure to improve the performance of the NWP model in simulating high-intensity rainfall must be taken into consideration. The MP-CP-PBl procedure followed in this study and updating the urban fraction certainly improved the performance of the WRF model to simulate high-intensity rainfall. The WRF data-assimilation and model coupling system can further improve the model's performance in simulating the events.

Self-supervised monocular depth estimation that does not require ground truth for training has attracted attention in recent years. It is of high interest to design lightweight but effective models so that they can be deployed on edge devices. Many existing architectures benefit from using heavier backbones at the expense of model sizes. This paper achieves comparable results with a lightweight architecture. Specifically, the efficient combination of CNNs and Transformers is investigated, and a hybrid architecture called Lite-Mono is presented. A Consecutive Dilated Convolutions (CDC) module and a Local-Global Features Interaction (LGFI) module are proposed. The former is used to extract rich multi-scale local features, and the latter takes advantage of the self-attention mechanism to encode long-range global information into the features. Experiments demonstrate that Lite-Mono outperforms Monodepth2 by a large margin in accuracy, with about 80% fewer trainable parameters., Comment: Accepted to CVPR2023

Wildfires drive deforestation that causes various losses. Although many studies have used spatial approaches, a multi-dimensional analysis is required to determine priority areas for mitigation. This study identified priority areas for wildfire mitigation in Indonesia using a multi-dimensional approach including disaster, environmental, historical, and administrative parameters by integrating 20 types of multi-source spatial data. Spatial data were combined to produce susceptibility, carbon stock, and carbon emission models that form the basis for prioritization modelling. The developed priority model was compared with historical deforestation data. Legal aspects were evaluated for oil-palm plantations and mining with respect to their impact on wildfire mitigation. Results showed that 379,516 km2 of forests in Indonesia belong to the high-priority category and most of these are located in Sumatra, Kalimantan, and North Maluku. Historical data suggest that 19.50% of priority areas for wildfire mitigation have experienced deforestation caused by wildfires over the last ten years. Based on legal aspects of land use, 5.2% and 3.9% of high-priority areas for wildfire mitigation are in oil palm and mining areas, respectively. These results can be used to support the determination of high-priority areas for the REDD+ program and the evaluation of land use policies.

Deep gullies in the Dashtiari Region prompted us to couple different morphometric indices obtained from a UAV-derived DEM to automatically extract gully signatures. The extraction of gully signatures is commonly undertaken via pattern recognition techniques, whose recent advancements seem to require more data and rather cumbersome modeling processes, making it even more difficult for those who are not well-versed in such contexts. Among these methods, object-based image analysis (OBIA), machine learning, and deep learning techniques are the most common. Conversely, here we took advantage of simple morphometric indices and their combinations for gully extraction, including valley depth (VD), topographic position index (TPI), positive openness (PO), red relief image map (RRIM), elevation, slope degree, and the coupled PO-DEM. Furthermore, we compared the automatically derived gully patterns to the manually extracted ones (treated as the ground truth), and their spatial autocorrelation was investigated. Additionally, the application of the classification tree (CT) as a powerful machine learning model was comparatively assessed for morphometric indices. The performance of the adopted pattern extraction techniques was estimated using four different metrics: precision index, true skill statistics (TSS), Cohen's kappa, and Matthews correlation coefficient (MCC). The results revealed that the single use of PO, TPI, and RRIM indices failed to reliably capture the gullies’ pattern, leading to partial success. Notably, combinations of indices showed that the coupled PO-DEM could successfully classify the gully presence locations from the absences and outperform the CT model in terms of both goodness-of-fit and generalization capacity (prediction power), considering all four-performance metrics. Hence, comparing the amount of time spent for manual delineation of gullies, the application of simple morphometric indices, and machine learning models is beyond comparison.

The mapping and characterisation of building footprints is a challenging task due to inaccessibility and incompleteness of the required data, thus hindering the estimation of loss caused by natural and anthropogenic hazards. Major advancements have been made in the collaborative mapping of buildings with platforms like OpenStreetMap, however, many parts of the world still lack this information or the information is outdated. We created a semi-automated workflow for the development of elements-at-risk (EaR) databases of buildings by detecting building footprints using deep learning and characterising the footprints with building occupancy information using building morphological metrics and open-source auxiliary data. The deep learning model was used to detect building EaR footprints in a city in Kerala (India) with an F1 score of over 76%. The footprints were classified into 13 building occupancy types along with information such as average number of floors, total floor space area, building density, and percentage of built-up area. We analysed the transferability of the approach to a different city in Kerala and obtained an almost similar F1 score of 74%. We also examined the exposure of the buildings and the associated occupancies to floods using the 2018 flood susceptibility map of the respective cities. We notice certain shortcomings in our research particularly, the need for a local expert and good quality auxiliary data to obtain reasonable building occupancy information, however, our research contributes to developing a rapid method for generating a building EaR database in data-scarce regions with attributes of occupancy types, thus supporting regional risk assessment, disaster risk mitigation, risk reduction initiatives, and policy developments.

Several territorial landslide early warning systems in different parts of the world are based on empirical rainfall thresholds for landslide triggering. The calculation of such thresholds, using rainfall measurements gathered from rain gauges, has been examined frequently, especially considering uncertainties, modeling complexity, spatial assumptions, and analytical tools. Installed rain gauge networks that are spatially clustered in crowded areas have different spatial and attribute settings based on landslide occurrence conditions, such as rainfall record accessibility, processing, and usability, as well as specific locational, morphological, and hydrological settings.In this research work, we introduce an automatic tool called DEWS (Distance, Elevation, Watershed, and Slope unit) for rainfall-induced landslide spatial reference rain gauge selection. DEWS can be considered supplementary and complementary to the CTRL-T tool (Calculation of Thresholds for Rainfall-induced Landslides Tool) developed earlier, and works on a macro-to-micro scale of the spatial components of CTRL-T rain gauge selections. The output information, i.e. the list of selected reference rain gauges, can be used as input for CTRL-T to calculate frequentist rainfall thresholds at different non-exceedance probabilities. The DEWS tool fills the gap of the current literature, where the selection of reference rain gauges is mostly based on the nearest distance location and on statistical or manual procedures, without considering the morphological and hydrological settings of the area in which landslides occurred.The tool allows extracting rain gauges referring to landslide locations by employing four spatial filters: F1 (Distance), F2 (Elevation), F3 (Watershed), F4 (Slope unit), needing only a DEM, the coordinates of landslide and rain gauge locations and the parameters of the filter’s algorithms as inputs. More in detail, F1 selects rain gauges within a specified buffer distance from the landslide locations using the setting parameters and the coordinates of the landslides and rain gauges. Then, F2 uses the DEM to extract the elevation of the rain gauges and the landslides and then calculates the differences within each buffer circle; therefore, the filter keeps only the rain gauge with closest elevation values to each landslide (within F1 results) using the recommended/preferred/ or allowable elevation difference defined by the parameter’s settings. In F3, the rain gauges falling in the watershed that contains the landslide locations are extracted (within F1 and F2 results). F4, which is the smallest and most focused filter, uses a previously developed tool pack (within F1, F2, and F3 results) to extract the slope units associated with each landslide. Consequently, only the rain gauges falling within these slope units are selected.DEWS was implemented in a free tool pack in QGIS software, with default parameter values for non-expert users. The tool pack is divided into three main blocks following the filter structure (F1 and F2 are kept together). The reliability of DEWS was tested at a territorial scale in South Korea, using 223 landslides and 328 rain gauges. As a second step, frequentist rainfall thresholds were calculated in the study area.

Landslides are one of the most disastrous natural hazards that frequently occur in Indonesia. In 2017, Balai Sabo developed an Indonesia Landslide Early Warning System (ILEWS) by utilizing a single rainfall threshold for an entire nation, leading to inaccuracy in landslide predictions. The study aimed to improve the accuracy of the system by updating the rainfall threshold. We analyzed 420 landslide events in Java with the 1-day and 3-day effective antecedent rainfall for each landslide event. Rainfall data were obtained from the Global Precipitation Measurement (GPM), which is also used in the ILEWS. We propose four methods to derive the thresholds: the first is the existing threshold applied in the Balai Sabo ILEWS, the second and third use the average and minimum values of rainfall that trigger landslides, respectively, and the fourth uses the minimum value of rainfall that induces major landslides. We used receiver operating characteristic (ROC) analysis to evaluate the predictability of the rainfall thresholds. The fourth method showed the best results compared with the others, and this method provided a good prediction of landslide events with a low error value. The chosen threshold was then applied in the Balai Sabo-ILEWS.

The patterns and controls of the transient enhanced landsliding that follows strong earthquakes remain elusive. Geostatistical models can provide clues on the underlying processes by identifying relationships with a number of physical variables. These models do not typically consider thermal information, even though temperature is known to affect the hydro-mechanical behavior of geomaterials, which, in turn, controls slope stability. Here, we develop a slope unit-based multitemporal susceptibility model for the epicentral region of the 2008 Wenchuan earthquake to explore how land surface temperature (LST) relates to landslide patterns over time. We find that LST can explain post-earthquake landsliding while it has no visible effect on the coseismic scene, which is dominated by the strong shaking. Specifically, as the landscape progressively recovers and landslide rates decay to pre-earthquake levels, a positive relationship between LST and landslide persistence emerges. This seems consistent with the action of healing processes, capable of restoring the thermal sensitivity of the slope material after the seismic disturbance. Although analyses in other contexts (not necessarily seismic) are warranted, we advocate for the inclusion of thermal information in geostatistical modeling as it can help form a more physically consistent picture of slope stability controls.

The complex tectonic setting of the central-western Mediterranean has interested geoscientists for decades, but its geodynamic evolution remains a matter of debate. We rely on 807 seismometers from southern Europe and northern Africa to measure Rayleigh and Love phase velocities in the period range ∼5–200 s, based on teleseismic earthquakes and seismic ambient noise. By nonlinear joint inversion of the phase-velocity maps, we obtain a 3-D shear-wave velocity (VS) model of the study area. At shallow depths, our model correlates with surface geology and reveals the presence of a sedimentary cover in the Liguro-Provençal basin, as opposed to the Tyrrhenian basin where this is either very thin or absent. At ∼5-km depth, high velocities below the Magnaghi, Vavilov, and Marsili seamounts point to an exhumed, scarcely serpentinized mantle. These are replaced by lower velocities at larger depths, likely connected to the presence of partial melt. At 50–60-km depth, a very heterogeneous structure characterizes the Tyrrhenian basin, with low velocities pointing to the presence of fluids due to the lateral mantle inflow from the Ionian slab edges, and higher velocities associated with a relatively dry upper mantle. Such heterogeneity disappears at depths ≳75 km, replaced by more uniform velocities which are ∼2% lower than those found in the Liguro-Provençal basin. We infer that, at the same depths, the Tyrrhenian basin is characterized by a larger concentration of fluids and possibly higher temperatures.

The thermochemical structure of the subcontinental mantle holds information on its origin and evolution that can inform energy and mineral exploration strategies, natural hazard mitigation and evolutionary models of the Earth. However, imaging the fine-scale thermochemical structure of continental lithosphere remains a major challenge. Here we combine multiple land and satellite datasets via thermodynamically-constrained inversions to obtain a high-resolution thermochemical model of central and southern Africa. Results reveal diverse structures and compositions for cratons, indicating distinct evolutions and responses to geodynamic processes. While much of the Kaapvaal lithosphere retained its cratonic features, the western Angolan-Kasai shield and the Rehoboth block have lost their cratonic keels. The lithosphere of the Congo Craton has been affected by metasomatism, increasing its density and inducing its conspicuous low-topography, geoid and magnetic anomalies. Our results reconcile mantle structure with the causes and location of volcanism within and around the Tanzanian Craton, whereas the absence of volcanism towards the north is the result of local asthenospheric downwellings, not to a previously-proposed lithospheric root connecting with the Congo Craton. Our study offers improved integration of mantle structure, magmatism and the evolution and destruction of cratonic lithosphere and lays the groundwork for futurelithospheric evolutionary models and exploration frameworks for the Earth and otherterrestrial planets.

To perform landslide susceptibility prediction (LSP), it is important to select appropriate mapping unit and landslide-related conditioning factors. The efficient and automatic multi-scale segmentation (MSS) method proposed by the authors promotes the application of slope units. However, LSP modeling based on these slope units has not been performed. Moreover, the heterogeneity of conditioning factors in slope units is neglected, leading to incomplete input variables of LSP modeling. In this study, the slope units extracted by the MSS method are used to construct LSP modeling, and the heterogeneity of conditioning factors is represented by the internal variations of conditioning factors within slope unit using the descriptive statistics features of mean, standard deviation and range. Thus, slope units-based machine learning models considering internal variations of conditioning factors (variant slope-machine learning) are proposed. The Chongyi County is selected as the case study and is divided into 53,055 slope units. Fifteen original slope unit-based conditioning factors are expanded to 38 slope unit-based conditioning factors through considering their internal variations. Random forest (RF) and multi-layer perceptron (MLP) machine learning models are used to construct variant Slope-RF and Slope-MLP models. Meanwhile, the Slope-RF and Slope-MLP models without considering the internal variations of conditioning factors, and conventional grid units-based machine learning (Grid-RF and MLP) models are built for comparisons through the LSP performance assessments. Results show that the variant Slope-machine learning models have higher LSP performances than Slope-machine learning models; LSP results of variant Slope-machine learning models have stronger directivity and practical application than Grid-machine learning models. It is concluded that slope units extracted by MSS method can be appropriate for LSP modeling, and the heterogeneity of conditioning factors within slope units can more comprehensively reflect the relationships between conditioning factors and landslides. The research results have important reference significance for land use and landslide prevention.

Structural disaster damage detection and characterization is one of the oldest remote sensing challenges, and the utility of virtually every type of active and passive sensor deployed on various air- and spaceborne platforms has been assessed. The proliferation and growing sophistication of unmanned aerial vehicles (UAVs) in recent years has opened up many new opportunities for damage mapping, due to the high spatial resolution, the resulting stereo images and derivatives, and the flexibility of the platform. This study provides a comprehensive review of how UAV-based damage mapping has evolved from providing simple descriptive overviews of a disaster science, to more sophisticated texture and segmentation-based approaches, and finally to studies using advanced deep learning approaches, as well as multi-temporal and multi-perspective imagery to provide comprehensive damage descriptions. The paper further reviews studies on the utility of the developed mapping strategies and image processing pipelines for first responders, focusing especially on outcomes of two recent European research projects, RECONASS (Reconstruction and Recovery Planning: Rapid and Continuously Updated Construction Damage, and Related Needs Assessment) and INACHUS (Technological and Methodological Solutions for Integrated Wide Area Situation Awareness and Survivor Localization to Support Search and Rescue Teams). Finally, recent and emerging developments are reviewed, such as recent improvements in machine learning, increasing mapping autonomy, damage mapping in interior, GPS-denied environments, the utility of UAVs for infrastructure mapping and maintenance, as well as the emergence of UAVs with robotic abilities.

Lava flows are one of the main hazards related to effusive basaltic volcanism. To minimize their impact during emplacement, we use lava flow potential distance-to-run predicted by propagation models. These models are partly based on infrared (IR) measurements of lava radiative heat fluxes by remote sensing (RS) methods (ground-based or satellite-based detectors) [1]. These results are however subjected to important errors related to the poor knowledge of spectral emissivity (ε), commonly considered constant by these well-established techniques[2, 3]. This oversimplification is an important source of uncertainties in derived temperatures, which restrain our capacity to accurately model active lava flows. In this study, we developed new algorithms that take into account the effect of spectral emissivity for calculating radiative heat fluxes. We describe the temperature-emissivity relationship with equations established at two wavelengths of interest for RS (10.9 μm and 1.6 μm) that are retrieved from in situ measurements of spectral emissivity for basaltic magma from the 2014–2015 Holuhraun eruption. Spectral emissivity data were systematically acquired over a wide spectral range (400–8000 cm−1) covering TIR, MIR and SWIR, and up to 1473 K [4]. Our results show that spectral emissivity varies linearly with temperature in TIR (10.9 μm), and nonlinearly in SWIR (1.6 μm). We confronted our lab-based results to the field IR data retrieved by [5] and found that temperature precision increases compared to data using constant emissivity value. These new insights will ultimately improve the thermo-rheological models of lava flows [6] in order to support hazard assessment in volcanic systems.

In October 2020, Central Vietnam was struck by heavy rain resulting from a sequence of 5 tropical depressions and typhoons. The immense amount of water led to extensive flooding and landslides that killed more than 200 people, injured more than 500 people, and caused direct damages valued at approximately 1.2 billion USD. Here, we quantify how the intensity of the precipitation leading to such exceptional impacts is attributable to anthropogenic climate change. First, we define the event as the regional maximum of annual maximum 15-day average rainfall (Rx15day). We then analyse the trend in Rx15day over Central Vietnam from the observations and simulations in the PRIMAVERA and CORDEX-CORE ensembles, which pass our evaluation tests, by applying the generalised extreme value (GEV) distribution in which location and scale parameters exponentially covary with increasing global temperatures. Combining these observations and model results, we find that the 2020 event, occurring about once every 80 years (at least 17 years), has not changed in either probability of occurrence (a factor 1.0, ranging from 0.4 to 2.4) or intensity (0%, ranging from −8 to +8%) in the present climate in comparison with early-industrial climate. This implies that the effect of human-induced climate change contributing to this persistent extreme rainfall event is small compared to natural variability. However, given the scale of damage of this hazard, our results underline that more investment in disaster risk reduction for this type of rainfall-induced flood hazard is of importance, even independent of the effect of anthropogenic climate change. Moreover, as both observations and model simulations will be extended with the passage of time, we encourage more climate change impact investigations on the extreme in the future that help adaptation and mitigation plans and raise awareness in the country.

Climatically-induced natural hazards are a threat to communities. They can cause life losses and heavy damage to infrastructure, and due to climate change, they have become increasingly frequent. This is especially in tropical regions, where major hurricanes have consistently appeared in recent history. Such events induce damage due to the high wind speed they carry, and the high intensity/duration rainfall they discharge can further induce a chain of hydro-morphological hazards in the form of widespread debris slides/flows. The way the scientific community has developed preparatory steps to mitigate the potential damage of these hydro-morphological threats includes assessing where they are likely to manifest across a given landscape. This concept is referred to as susceptibility, and it is commonly achieved by implementing binary classifiers to estimate probabilities of landslide occurrences. However, predicting where landslides can occur may not be sufficient information, for it fails to convey how large landslides may be. This work proposes using a flexible Bernoulli-log-Gaussian hurdle model to simultaneously model landslide occurrence and size per areal unit. Covariate and spatial information are introduced using a generalised additive modelling framework. To cope with the high spatial resolution of the data, our model uses a Markovian representation of the Matérn covariance function based on the stochastic partial differential equation (SPDE) approach. Assuming Gaussian priors, our model can be integrated into the class of latent Gaussian models, for which inference is conveniently performed based on the integrated nested Laplace approximation method. We use our modelling approach in Dominica, where Hurricane Maria (September 2017) induced thousands of shallow flow-like landslides passing over the island. Our results show that we can not only estimate where landslides may occur and how large they may be, but we can also combine this information in a unified landslide hazard model, which is the first of its kind.

Mapping existing landslides is a fundamental prerequisite to build any reliable susceptibility model. From a series of landslide presence/absence conditions and associated landscape characteristics, a binary classifier learns how to distinguish potentially stable and unstable slopes. In data rich areas where landslide inventories are available, addressing the collection of these can already be a challenging task. However, in data scarce contexts, where geoscientists do not get access to pre-existing inventories, the only solution is to map landslides from scratch. This operation can be extremely time-consuming if manually performed or prone to type I errors if done automatically. This is even more exacerbated if done over large geographic regions. In this manuscript we examine the issue of mapping requirements for west Tajikistan where no complete landslide inventory is available. The key question is: How many landslides should be required to develop reliable landslide susceptibility models based on statistical modeling? In fact, for such a wide and extremely complex territory, the collection of an inventory that is sufficiently detailed requires a large investment in time and human resources. However, at which point of the mapping procedure, would the resulting susceptibility model produce significantly better results as compared to a model built with less information? We addressed this question by implementing a binomial Generalized Additive Model trained and validated with different landslide proportions and measured the induced variability in the resulting susceptibility model. The results of this study are very site-specific but we proposed a very functional protocol to investigate a problem which is underestimated in the literature.

Surface heat flow provides essential information on the thermal state and thickness of the lithosphere. Southern Africa is a mosaic of the best‐preserved and exposed crustal blocks, assembled in the early late Archean and then modified by a series of major tectono‐thermal events, both of Precambrian and Phanerozoic. Understanding the thermal and compositional structure of the southern African lithosphere provides crucial information for the actual causes, processes of lithospheric stability, and modification. Curie depth, interpreted as the depth to 580°C, provides a valuable constraint on the thermal structure of the lithosphere. Due to the sparse distribution of surface heat flow data, we examine the degree to which the thermal structure of the crust can be constrained from Curie depth estimates in southern Africa. We provide a Curie depth map for southern Africa (obtained from the inversion of magnetic anomaly data via power spectral methods and wavelet analysis) alongside with a heat flow map (based on the previous Curie depth estimates), both equipped with uncertainties via a Bayesian approach. Opposed to other cratonic regions, the observation of a shallow Curie depths and low heat flow over the Kaapvaal Craton suggests a thermochemical reworking of the cratonic lithosphere in this region. Furthermore, a comparison with a model for the Moho depth reveals significant variations of the Curie depth, which may be located below or above the Moho in large regions. Both observations indicate that in certain regions magnetic anomaly based Curie depth estimates may reflect a compositional rather than a temperature controlled constraint., Plain Language Summary: The thermal state and thickness of the lithosphere are reflected, among other quantities, in the surface heat flow. While heat flow data are rather sparse, magnetic anomaly maps are widely available and allow, under certain conditions, the estimation of the bottom of the magnetized layer within the lithosphere. Latter can be associated with the 580°C‐isotherm (frequently called the Curie depth), therefore allowing inferences on the thermal state of the lithosphere from magnetic data. Here, we use classical power spectral methods in combination with wavelet analysis and Bayesian methods to estimate the Curie depth and its uncertainty from magnetic anomaly maps, and subsequently, use it to estimate the surface heat flow over southern Africa. Comparison with the sparsely available measured heat flow allows us to assess the quality of the estimation and to interpret it with respect to the lithospheric structure. Southern Africa is particularly suitable for such a study due to its well‐preserved and well‐studied crustal blocks. The shallow Curie depths and low heat flow observed over the Kaapvaal Craton suggest a thermal reworking of the old lithospheric structures in this region. Comparison with a Moho depth model further suggests that in some regions the obtained Curie depth estimates do not necessarily reflect a temperature controlled but rather a compositional boundary., Key Points: Curie depth is estimated over southern Africa via power spectral methods and wavelet transforms within a Bayesian framework. Heat flow and its uncertainty is estimated based on Curie depth. Shallow Curie depth and low heat flow suggest thermochemical reworking of the Kaapvaal Craton., Bundesministerium für Wirtschaft und Energie (BMWi) http://dx.doi.org/10.13039/501100006360, Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO) http://dx.doi.org/10.13039/501100003246

Laser Return Intensity (LRI) holds excellent opportunity in helping to map lithology and discriminating rock units based on their reflectivity. To date, many rock types have been examined under a controlled environment using Terrestrial Lidar Scanning (TLS). However, the utilization of LRI for detection of rocks properties in a geothermal environment still has received less attention especially from Airborne Lidar Scanning (ALS). This study proposes a novel approach to map and characterize geothermal surface manifestations using airborne lidar return intensity. We use an airborne lidar dataset acquired by a Leica ALS70 above the geothermal field near Bajawa, Flores, Indonesia. The effects of scan angle and atmosphere on the resulting LRI images were investigated and where possible corrected for optimizing the positional and spectral information. Characterization of geothermal surface manifestation signatures based on altered volcanic rock reflectivity through LRI were validated by the ground truth observation and airborne thermal infrared imagery in Wawomuda and Mataloko, Bajawa. We will present an overview of the data corrections needed to evaluate the lidar return intensity values for detecting geothermal surface manifestations in the tropical mountainous environment.