Fig 1 - uploaded by Biljana Abolmasov
Content may be subject to copyright.
Geographical position of the Republic of Serbia and the position of the included municipalities in the research
Source publication
This paper focuses on a specific event-based landslide inventory compiled after the May 2014 heavy rainfall episode in Serbia as a part of the post-disaster recovery actions. The inventory was completed for a total of 23 affected municipalities, and the municipality of Krupanj was selected as the location for a more detailed study. Three sources of...
Contexts in source publication
Context 1
... study area covered 11.840 km 2 , i.e. 23 of the 27 municipal- ities included in the UNDP post-disaster BEWARE Project ac- tivities in the western and central parts of the Republic of Serbia (Fig. 1). Four municipalities were excluded from the analysis because there were no landslides that occurred during the May 2014 rainfall event, only floods and flash floods. In the 23 municipalities that were included, the total population was approximately 1,000,000 people. It is also important to mention that the majority of these ...
Context 2
... municipality of Krupanj, one of the most affected areas in May 2014, was selected as a test site for more detailed research. This area is located in western Serbia and covers 341 km 2 , with a population of 17,295 inhabitants ( Fig. 1). The majority of the territory consists of Devonian-Carboniferous weathered low-crystalline metamorphic rocks, including mud and clay shales, phyllites and argillaceous schist (approximately 80%). Permian meta-sediments (clay shales, sandstone and flysch), Triassic limestone and sandstone and Cretaceous lime- stone make up the ...
Context 3
... in BData processing^ section were analysed with a semi-automatic supervised classification in or- der to determine the target classes. The following classes were selected: areas that included clouds and/or shadows-class 0; areas of cultivated land-class 1; areas under water and vegetation-class 2; and areas with bare land (eroded areas)-3 (Fig. 10). The process of automatic classification misclassified alluvial plains and river terraces as eroded areas (that poten- tially contain landslides); therefore, they were manually re- moved from class 3. It was found that instabilities that represent a single body were disaggregated into multiple poly- gons. All such cases were identified ...
Context 4
... largest landslide/debris flow was recognized in the municipality of Ljubovija, near the village of Selanac, with an area of 85,222.23 m 2 (Table 4). This landslide/debris flow was 1.5 km long and, at its wider part, was 220 m wide (Figs. 11 and ...
Context 5
... of the visual interpretation of the satellite images in the Krupanj test area are shown in Fig. 13. A total of 507 landslides (Table 5) were assigned certainty labels of B1^ and ...
Context 6
... semi-automatic analysis was applied in order to highlight the differences in the relief for the period after a heavy rainfall by comparing the pre-event and post-event images (Fig. 14). The red/orange colour represents a deficit of mass, while the blue colour represents a surplus of mass (accumulation of transported material). Since the resolution of the Landsat 8 images was 30 m, only instabilities longer/wider than 30 m were recognized. White zones in the image represent a mask for clouds and shadows. These results ...
Context 7
... 8 images was 30 m, only instabilities longer/wider than 30 m were recognized. White zones in the image represent a mask for clouds and shadows. These results were burdened with unde- sired interferences, such as alluvial fans and agricultural zones, which were shown as surplus/deficit masses. Given the spectral signature of the target classes (Fig. 10), the minimum distance algorithm had difficulties coping with the classifications in spectral overlap domains, especially for cultivated land, vegeta- tion and bare ...
Context 8
... visual satellite image analysis, supervised classifications and field observations were cross-compared (Fig. 15). Since the lowest common resolution for these three sources was 30 m (Landsat 8 image resolution), only landslides larger than 900 m 2 in the area were considered. The field data were loosely considered to be the ground truth but were mostly used as a reference. The reason for this lies in the fact that the flows were recognized with ...
Similar publications
The Oligocene-Miocene turbidite successions of the Emilia-Tuscany Apennines represent one of the most intriguing topic of the Northern Apennines (NA), due to the complex relationships between tectonics and sedimentation in a migrating thrust wedge – foredeep system, to the not-well constrained differences between orogenic landslides (olistostrome)...
This paper presents a landslide inventory map for the Chittagong Hilly Areas of Bangladesh based on Google Earth and field mapping. We developed a set of criteria to identify landslides in Google Earth and introduced a method to assess the accuracy of mapped landslides when they are recorded as points rather than polygons in the field. In total, 23...
Citations
... Under these conditions, traditional surveying methods were challenging to implement, making it nearly impossible to manually obtain on-site information and accurately assess the landslide's movement, which posed significant challenges to the rescue efforts. Consequently, UAV photogrammetry provided a rapid and accurate means of acquiring comprehensive image data of the slope (Denis et al. 2016;Duric et al. 2017). Utilizing UAV-based photogrammetry and 3D reconstruction technology, whole-process survey and analysis of the landslide area were conducted. ...
Timely and accurate mastery of the spatial and temporal development process of landslides is crucial for emergency rescue work. In the early morning of July 5, 2022, under the influence of continuous heavy rainfall, the left embankment slope of a highway in Guangdong Province, slipped as a whole, and a truck fell and was buried, resulting in the interruption of highway traffic. In order to accurately carry out landslide analysis and management, and to provide information support for emergency rescue and relief work, whole-process survey and analysis of the landslide area was carried out based on unmanned aerial vehicle tilt photography, 3D reconstruction and spatial analysis technology. In situations where traditional surveying and mapping work is difficult to carry out, it is possible to accurately grasp the whole process of the development of landslides in space and time under the circumstances and construct corresponding realistic three-dimensional models to provide effective reference information for on-site rescue and relief work. At the same time, through the comprehensive analysis of the acquired information, it is inferred that the landslide is a traction landslide under the effect of continuous rainfall.
... In Serbia, numerous such events occurred in 2014 due to the Tamara cyclone effect. Specifically, over 200 flow-type landslides were recorded in the western part of Serbia, with many others identified indirectly through satellite image analysis [3,4]. Many of these incidents were unexpected, even in areas designated as relatively stable according to hazard maps. ...
... The Pleiades satellite image for Selanac from June 2014 was scrutinized, taken immediately after the landslide activation in May 2014. This satellite image served visualization purposes and facilitated the determination of the location and extent of the broader investigation area [4]. Subsequently, other remote sensing methods were utilized to primarily gather more precise data, such as generating a high-resolution Digital Terrain Model (DTM). ...
Flow-type landslides are not typical in this region of the Balkans. However, after the Tamara cyclone event in 2014, numerous such occurrences have been observed in Serbia. This paper presents the initial results of a detailed investigation into debris flows in Serbia, comparing findings from two programs: RAMMS DBF and Geoflow SPH. Located in Western Serbia, the Selanac debris flow is a complex event characterized by significant depths in the initial block and entrainment zone. Previous field investigations utilized ERT surveys, supplemented by laboratory tests, to characterize material behavior. Approximately 450,000 m3 of material began to flow following an extreme precipitation period, ultimately traveling 1.2 km to the deposition zone. For validation purposes, ERT profiles from both the deposition zone and the source area were utilized, with particular attention given to areas where entrainment was substantial, as this had a significant impact on the final models. The first objective of this research is to conduct a detailed investigation of debris flow using field investigations: geophysical (ERT) and aerial photogrammetry. The second objective is to evaluate the capacity of two debris flow propagation models to simulate the reality of these phenomena. The GeoFlow-SPH code overestimated the maximum propagation thickness in comparison to the RAMMS model. The numerical results regarding final depths closely align, especially when considering the estimated average depth in the deposition zone. The results confirm the necessity of using multiple simulation codes to more accurately predict specific events.
... After the 2005 earthquake, many researchers have studied earthquake-induced landslides using various data and techniques (Das et al., 2007, Bulmer et al., 2007, Owen et al., 2008, Ray et al., 2009, Saba et al., 2010, Basharat et al., 2012, Basharat et al., 2014, Shafique et al., 2016, Shafique, 2020. Different studies show the impact of spatial resolution and its importance for the mapping of landslides in visual interpretation and also the semi-automatic techniques using different spatial resolution of data i.e. (Pléiades, WorldView-2, and SPOT 6) (Shafique et al., 2011, Ðurić et al., 2017. The reliability of the landslide inventory will lead to an accurate landslide susceptibility map. ...
In mountainous terrains, the frequent landslides and their associated impacts on human lives and the economy is increasing globally. Development of landslide inventory and afterward landslide susceptibility mapping are the main prerequisites for implementing landslide mitigation measures and protection in mountainous regions. The 2005 Kashmir earthquake induced different small and large landslides and some were active for the long term. So far many studies have used medium and high-resolution data to develop landslide inventory. This study aims to develop a 1st detailed, comprehensive and accurate landslide inventory using a very high-resolution image using a semi-automatic technique. The precise landslide inventory is employed to develop an accurate and comprehensive landslide susceptibility map considering the landslide inventory data logistic regression training model. Furthermore, the landslide’s temporal recovery from the earthquake and its reactivation due to rainfall in spare vegetation areas have been evaluated. Fine-resolution satellite images of Worldview-2 are applied to develop a detailed landslide inventory using a Support Vector Machine (SVM) classifier. A total of 63630 landslides were identified using a semi-automatic technique within a study area of 265 km2. From regression modeling, the results show that geology, topography, and road networks have a significant impact on the spatial distribution of landslides. Model performance was evaluated based on the testing data, the model gives an AUC of 0.93 and the kappa value of 0.9353. The spatiotemporal NDVI has been and assessed to identify the landslide recovery and its reactivation due to extreme rainfall. The results show that 72.1 % of the landslides occurred in Muzaffarabad formation in the study area. The developed landslide susceptibility map can be further used for land-use planning and implementing mitigation measures for the safety of roads and other infrastructure in the area.
... However, this approach is limited by the limited ability of two-dimensional images to accurately represent landscape spatial complexities (Dou et al. 2020;Lu et al. 2019). In response, researchers have enhanced this method by superimposing remote sensing images onto terrain data, leveraging the three-dimensional visualization strengths of DEMs to better detect landslides through spatial scale, rotation, and scaling adjustments (Đurić et al. 2017;Fan et al. 2018). Despite these advancements, detecting older landslides, which often blend into the surrounding terrain, remains challenging. ...
Current research on deep learning-based intelligent landslide detection modeling has focused primarily on improving and innovating model structures. However, the impact of terrain factors and data fusion methods on the prediction accuracy of models remains underexplored. To clarify the contribution of terrain information to landslide detection modeling, 1022 landslide samples compiled from Planet remote sensing images and DEM data in the Sichuan–Tibet area. We investigate the impact of digital elevation models (DEMs), remote sensing image fusion, and feature fusion techniques on the landslide prediction accuracy of models. First, we analyze the role of DEM data in landslide modeling using models such as Fast_SCNN, the SegFormer, and the Swin Transformer. Next, we use a dual-branch network for feature fusion to assess different data fusion methods. We then conduct both quantitative and qualitative analyses of the modeling uncertainty, including examining the validation set accuracy, test set confusion matrices, prediction probability distributions, segmentation results, and Grad-CAM results. The findings indicate the following: (1) model predictions become more reliable when fusing DEM data with remote sensing images, enhancing the robustness of intelligent landslide detection modeling; (2) the results obtained through dual-branch network data feature fusion lead to slightly greater accuracy than those from data channel fusion; and (3) under consistent data conditions, deep convolutional neural network models and attention mechanism models show comparable capabilities in predicting landslides. These research outcomes provide valuable references and insights for deep learning-based intelligent landslide detection.
... Unfortunately, these products are rarely free-cost and require an acquisition plan and/or only cover some of the area affected. Moreover, detailed mapping needs a timeconsuming expert manual approach (Donnini et al., 2023;Guzzetti et al., 2012) or the development of a machine learning algorithm that requires a high level of expertise (Đurić et al., 2017;Galli et al., 2008;Lu et al., 2019;Meena et al., 2023;Murillo-García et al., 2015). The advent of remote sensing technologies has opened opportunities for the development of methodologies and techniques for automating landslides mapping through the utilization of various high-resolution optical sensor products, significantly reducing the timing for the mapping (Casagli et al., 2016;Guzzetti et al., 2012). ...
Shallow landslides, frequently triggered by extreme events such as heavy rainfall, snowmelt, or earthquakes, affect vast areas with remarkable density. In the immediate aftermath of such events, it becomes crucial to rapidly assess landslides distribution and pinpoint the most severely affected areas to prioritize damage assessments and guide field survey operations effectively. Once the emergency phase subsides, the attention can shift to enhancing the accuracy of landslide inventory. In this work, we introduce the two-phase methodology "PANDA", the unsuPervised shAllow laNdslide rapiD mApping, for the low-cost mapping of the potential landslides, firstly in the emergency phase and then, with an improved version, in the post-emergency one. This approach utilizes variations in NDVI derived from Sentinel-2 satellite imagery and geomorphological filters. We applied PANDA to rainfall events in the northeastern Apennine range, Italy, occurred in May 2023, causing dramatic social and economic consequences for this mountain territory. Within just five days of obtaining Sentinel-2 post-event imagery, we produced a reliable, ready-to-use map covering a vast area (~4000 km 2). The map tested during emergency field mapping shows positive feedback. In the post-emergency phase, accuracy was enhanced using completely cloud-free imagery, a filter to identify false positives associated with land use changes, a higher resolution digital terrain model (DTM), and an iterative approach to optimize NDVI and slope thresholds. Potential landslide density related with rainfall, indicating that the most severely affected region attained a density of approximately 50 landslides/km². Validation against an independent manual inventory based on high-resolution imagery demonstrated encouraging accuracy results from both inventories, with a noticeable increase in the F1 score for the post-emergency version.
... Landslides in Serbia represent a significant environmental and economic problem [36][37][38]. In urban areas, many landslides are influenced by anthropogenic activities [36], but they can also be triggered by extreme climatic events [39]. In our study, we investigated one of the most catastrophic landslide events in recent Serbian history. ...
In February 1963, a huge landslide (ca. 1,950,000 m3) blocked the Visočica River and, thus, formed Zavoj Lake. The primary objective of this research was to investigate the importance of snowmelt in relation to landslide occurrence and to define the critical climatic conditions that may trigger massive winter landslides. We used monthly precipitation and average monthly maximum temperature data from meteorological and precipitation stations in the Visočica River basin (Dojkinci) and in the immediate proximity of Lake Zavoj (Pirot, Dimitrovgrad and Topli Do) as data inputs to the Snow-Melt Landslide (SML) index. It considers the summed monthly precipitation for previous months that continuously have an average maximum temperature below 0 °C. According to this method, the event at Zavoj Lake stands out among all other precipitation and snowmelt values for the past 72 years. After applying the SML index, all stations showed values of >300 mm for February 1963, which we consider as the threshold value for potential landslides appearance. In addition to meteorological data, we applied the SML index to data from the Coordinated Regional Downscaling Experiment (CORDEX) regional climate model outputs for the region from 2022 to 2100. As expected, climate change will have influenced the temperature values, especially during the winter. Conversely, the study area is experiencing drastic changes in land use caused by depopulation, leading to a reduced risk of winter landslides in the Visočica basin. We suggest that future climatic conditions in the area will make it more likely to experience extreme summer precipitation events, which might trigger large landslides. The SML method can be implemented for all landscapes that experience snowy winters, providing information in a timely manner so that local residents can react properly when the probability of landslide occurrence rises. The SML index, grounded in essential meteorological principles, provides a tailor-made, data-driven methodology applicable across varied geographical settings. Its utility extends to mitigating hydro-meteorological hazards on scales ranging from local to national scales, offering diverse and effective early warning solutions.
... Landslides in Serbia represent a significant environmental and economic problem [36][37][38]. In urban areas, many landslides are influenced by anthropogenic activities [36], but they can also be triggered by extreme climatic events [39]. In our study, we investigated one of the most catastrophic landslide events in recent Serbian history. ...
In February 1963, a huge landslide (ca. 1,950,000 m3) blocked the Visoˇcica River and, thus, formed Zavoj Lake. The primary objective of this research was to investigate the importance of snowmelt in relation to landslide occurrence and to define the critical climatic conditions that may trigger massive winter landslides. We used monthly precipitation and average monthly maximum temperature data from meteorological and precipitation stations in the Visoˇcica River basin (Dojkinci) and in the immediate proximity of Lake Zavoj (Pirot, Dimitrovgrad and Topli Do) as data inputs to the Snow-Melt Landslide (SML) index. It considers the summed monthly precipitation for previous months that continuously have an average maximum temperature below 0 ◦C. According to this method, the event at Zavoj Lake stands out among all other precipitation and snowmelt values for the past 72 years. After applying the SML index, all stations showed values of >300 mm for February 1963, which we consider as the threshold value for potential landslides appearance. In addition to meteorological data, we applied the SML index to data from the Coordinated Regional Downscaling Experiment (CORDEX) regional climate model outputs for the region from 2022 to 2100. As expected, climate change will have influenced the temperature values, especially during the winter. Conversely, the study area is experiencing drastic changes in land use caused by depopulation, leading to a reduced risk of winter landslides in the Visoˇcica basin. We suggest that future climatic conditions in the area will make it more likely to experience extreme summer precipitation events, which might trigger large landslides. The SML method can be implemented for all landscapes that experience snowy winters, providing information in a timely manner so that local residents can react properly when the probability of landslide occurrence rises. The SML index, grounded in essential meteorological principles, provides a tailor-made, data-driven methodology applicable across varied geographical settings. Its utility extends to mitigating hydro-meteorological hazards on scales ranging from local to national scales, offering diverse and effective early warning solutions.
... AbOLMASOv et al. (2017) made the first assessments of the terrain`s susceptibility to land-slides in the municipality of Krupanj using AHP and Woe method. ĐuRIć et al. (2017) represent a cadaster of landslides that occurred after heavy rainfall in 2014 using satellite images. MARIjANOvIć et al. (2018) did a landslide risk assessment on the road network of the municipality of Krupanj. ...
The aim of this paper is to show the susceptibility to sliding in the Krupanj area. Intense rainfall in May 2014 triggered many landslides in western Serbia. The Krupanj area was particularly affected by this event. The material damage occurred affected the awareness of the importance of knowing the locations that are prone to sliding. Therefore, from 2014 to 2021, field research was carried out in the Krupanj area. During this period a large amount of landslide data were collected by engineering geological mapping. These data were used for susceptibility analysis. For present study the analytical hierarchy process (AHP) and weight overlay (WO) tool were used. Nine factors were used for susceptibility analysis: slope, aspect, curvature, elevation, lithology, distance from rivers, faults, boundary?s and land cover. In order to be able to make a comparison, all factors were evaluated using the Saaty scale, so that the weights of the individual factors were obtained. The weight values obtained by AHP were used in GIS tool. Final map was validated by ROC analysis. The validation results show accuracy of 77,1% (good) for model.
... Since 2016, The World Bank Transport Team has supported the Government of Serbia to mainstream climate resilience in the road transportation sector. The general idea was to connect transportation and natural disaster management after devasting floods and landslides in 2014 (Abolmasov 2019;Đurić et al. 2017;Marjanović et al. 2018aMarjanović et al. , 2018b. The total negative effect on the transport sector was EUR 166.5 million (RNA Report 2014). ...
In the framework of the project “Mainstreaming Climate Resilience in the Road Transportation Management in Serbia (CliRtheRoads)”, a complex mapping tool was developed to support the Government of Serbia and Public Enterprise Roads of Serbia in climate change adaptation planning and management. The software solution comprises of: a web portal for data entry and management for authorised users; a publicly available web-GIS application; a mobile GIS application; and a back-end database. This paper briefly overviews the system, focusing on the landslide data model. Therein, the main system and software solution breakthrough is automatizing the estimation of investment costs of engineering and non-engineering measures recommended at affected locations along the road network. A retrospective on its applicability and user feedback is also included. The objective was to facilitate seamless road management by providing necessary data in a simple, understandable fashion, indicating which locations on the road network have higher priority. The introduced level of automation allows easier decision-making and investment planning.
... Table 1 reviews recent efforts that have attempted to use airborne platforms for landslide studies. Majority of the studies have exploited the optical data (red, green, blue bands) for landslide detection through image classification [27][28][29][30][31][32][33][34][35] or the aerial photogrammetry for landslide analysis [36][37][38][39]. However, as the photogrammetry method measures the reflectance of sunlight from an object, it hardly captures the topographic information if the ground is obscured by trees and canopy in densely wooded or vegetated environments. ...
... The trees were captured with dense points well. However, there were only minimal number of points capturing the ground (Fig. 8a) because the area was densely wooded and it limited sunlight transmission and reflection from the ground [27][28][29][30][31][32][33][34][35]. Fig. 8b highlights that the UAV-LiDAR system provided the richer point data from the ground, compared to the photogrammetry result [43,[45][46][47][48]. Therefore, the UAV-LiDAR system proves its effectiveness in acquiring the terrain-relevant information, such as topographic information and relative ground displacement. ...
High-resolution topographic information of landslide-prone areas plays an important role in accurate prediction and characterization of potential landslides and mitigation of landslides-associated hazards. This study presents an advanced geomorphological surveying system that integrates the light detection and ranging (LiDAR) with an unmanned aerial vehicle (UAV), a multi-rotor aerial vehicle in specific, for prediction, monitoring and forensic analysis of landslides, and for maintenance of debris-flow barriers. The test-flight over a vegetated area demonstrates that the integrated UAV-LiDAR system can provide high-resolution, three-dimensional
(3D) LiDAR point clouds below canopy and vegetation in forest environments, overcoming the limitation of aerial photogrammetry and terrestrial LiDAR platforms. An algorithm is suggested to delineate the topographic information from the acquired 3D LiDAR point clouds, and the accuracy and performance of the developed UAV-LiDAR system are examined through field demonstration. Finally, two field demonstrations are presented: the forensic analysis of the recent Gokseong landslide event, and the sediment deposition monitoring for debris-flow barrier maintenance in South Korea. The developed surveying system is expected to contribute to geomorphological field surveys in vegetated, forest environments, particularly in a site that is not
easily accessible.
