31 research outputs found
Comparison and evaluation of landslide susceptibility maps obtained from weight of evidence, logistic regression, and artificial neural network models
No full textEvaluation of landslide susceptibility assessment methods by using geoinformatics
No full textThe knowledge about the spatial occurrence probability of a natural hazard is particularly useful for its mitigation, and estimation of the potential amount of damages and losses that it may cause. However, the acquisition of this knowledge requires dealing with important issues such as the existence of complex relationships among factors related to the occurrence of such a hazard, the lack of relevant data, and the integration of the dynamic changes taking place in the environment. The modern technological advances in the field of Geosciences allow the simulation of natural hazards. For this reason, by using the capabilities of Geoinformatics-based technologies (such as Geographical Information Systems, Remote Sensing, etc.), the present PhD thesis aims to the figuration of spatial predictions of landslide occurrence. The target predictions were derived from the landslide susceptibility assessment through empirical analyses in Greece. In terms of this assessment, the impact of the change of analysis scale on the performance of selected models was examined. A set of qualitative, quantitative and integrated models was applied in two different analysis scales (regional and more detailed), and by extension, in two areas with different size in Greece (system of catchments in northern Peloponnese, and Selinous river catchment, respectively). Due to the difference in the scale analysis, similar datasets of landslides and causal factors were collected which were characterized by different spatial resolution (with cell size 90 and 20 meters, respectively). The performance of the models was evaluated comparatively through specific validation methods. Furthermore, for each of the two analysis scales, the existence of spatial non-stationarity in the relationships between the landslide occurrence and the analyzed factors was investigated. Finally, the impact expected to have the “optimal” susceptibility result on the socio-economic elements of the study area in the more detailed analysis scale was examined through the landslide vulnerability assessment.Η γνώση σχετικά με τη χωρική πιθανότητα εκδήλωσης ενός φυσικού κινδύνου είναι ιδιαιτέρως χρήσιμη για τον μετριασμό του, και την εκτίμηση του πλήθους των ενδεχόμενων ζημιών και απωλειών που μπορεί να προκαλέσει. Ωστόσο, η απόκτηση αυτής της γνώσης απαιτεί την αντιμετώπιση σημαντικών ζητημάτων, όπως είναι η ύπαρξη των πολύπλοκων σχέσεων μεταξύ των παραγόντων που σχετίζονται με την εκδήλωση ενός τέτοιου κινδύνου, η έλλειψη σχετικών δεδομένων, και η ενσωμάτωση των δυναμικών αλλαγών που λαμβάνουν χώρα στο περιβάλλον. Οι σύγχρονες τεχνολογικές εξελίξεις στον επιστημονικό κλάδο των Γεωεπιστημών επιτρέπουν την προσομοίωση των φυσικών κινδύνων. Γι’ αυτό το λόγο, αξιοποιώντας τις δυνατότητες των βασισμένων στη Γεωπληροφορική τεχνολογιών (όπως Συστήματα Γεωγραφικών Πληροφοριών, Τηλεπισκόπηση, κ.ά.), η παρούσα διδακτορική διατριβή στοχεύει στη διαμόρφωση χωρικών προβλέψεων εκδήλωσης κατολισθήσεων. Οι επιδιωκόμενες προβλέψεις προέκυψαν από την εκτίμηση της επιδεκτικότητας σε εκδήλωση κατολίσθησης μέσω εμπειρικών αναλύσεων στον Ελληνικό χώρο. Στα πλαίσια αυτής της εκτίμησης, εξετάστηκε η επίδραση της μεταβολής της κλίμακας ανάλυσης στην απόδοση επιλεγμένων μοντέλων. Ένα σύνολο ποιοτικών, ποσοτικών και ενοποιημένων μοντέλων εφαρμόστηκε σε δύο διαφορετικές κλίμακες ανάλυσης (περιφερειακή και λεπτομερέστερη), και κατ’ επέκταση, σε δύο διαφορετικού μεγέθους περιοχές του Ελληνικού χώρου (σύστημα λεκανών απορροής της βόρειας Πελοποννήσου, και λεκάνη απορροής του ποταμού Σελινούντα, αντιστοίχως). Λόγω της διαφοράς της κλίμακας ανάλυσης, παρόμοια σύνολα δεδομένων κατολισθήσεων και αιτιολογικών παραγόντων συλλέχθηκαν τα οποία χαρακτηρίζονταν από διαφορετική χωρική ανάλυση (με μέγεθος ψηφίδας 90 και 20 μέτρα, αντιστοίχως). Η απόδοση των μοντέλων αξιολογήθηκε συγκριτικώς μέσω εξειδικευμένων μεθόδων επικύρωσης. Επιπλέον, για κάθε μια από τις δύο κλίμακες ανάλυσης, διερευνήθηκε η ύπαρξη χωρικής μη-στασιμότητας στις σχέσεις μεταξύ της εκδήλωσης κατολισθήσεων και των αναλυθέντων παραγόντων. Τέλος, η επίδραση που αναμένεται να έχει η επαλήθευση του «βέλτιστου» αποτελέσματος επιδεκτικότητας στα κοινωνικο-οικονομικά στοιχεία της περιοχής μελέτης στη λεπτομερέστερη κλίμακα ανάλυσης, εξετάστηκε μέσω της εκτίμησης της τρωτότητας σε εκδήλωση κατολίσθησης
GIS-Based Landslide Susceptibility Mapping on the Peloponnese Peninsula, Greece
No full text: In this paper, bivariate statistical analysis modeling was applied and validated to derive a landslide susceptibility map of Peloponnese (Greece) at a regional scale. For this purpose, landslide-conditioning factors such as elevation, slope, aspect, lithology, land cover, mean annual precipitation (MAP) and peak ground acceleration (PGA), and a landslide inventory were analyzed within a GIS environment. A landslide dataset was realized using two main landslide inventories. The landslide statistical index method (LSI) produced a susceptibility map of the study area and the probability level of landslide occurrence was classified in five categories according to the best classification method from three different methods tested. Model performance was checked by an independent validation set of landslide events. The accuracy of the final result was evaluated by receiver operating characteristics (ROC) analysis. The prediction ability was found to be 75.2% indicating an acceptable susceptibility map obtained from the GIS-based bivariate statistical model
GIS Supported Landslide Susceptibility Modeling at Regional Scale: An Expert-Based Fuzzy Weighting Method
No full textThe main aim of this paper is landslide susceptibility assessment using fuzzy expert-based modeling. Factors that influence landslide occurrence, such as elevation, slope, aspect, lithology, land cover, precipitation and seismicity were considered. Expert-based fuzzy weighting (EFW) approach was used to combine these factors for landslide susceptibility mapping (Peloponnese, Greece). This method produced a landslide susceptibility map of the investigated area. The landslides under investigation have more or less same characteristics: lateral based and downslope shallow movement of soils or rocks. The validation of the model reveals, that predicted susceptibility levels are found to be in good agreement with the past landslide occurrences. Hence, the obtained landslide susceptibility map could be acceptable, for landslide hazard prevention and mitigation at regional scale
Exploring the Impact of Analysis Scale on Landslide Susceptibility Modeling: Empirical Assessment in Northern Peloponnese, Greece
No full textThe main purpose of this study is to explore the impact of analysis scale on the performance of a quantitative model for landslide susceptibility assessment through empirical analyses in the northern Peloponnese, Greece. A multivariate statistical model like logistic regression (LR) was applied at two different scales (a regional and a more detailed scale). Due to this scale difference, the implementation of the model was based on two landslide inventories representing in a different way the landslide occurrence (as point and polygon features), and two datasets of similar geo-environmental factors characterized by a different size of grid cells (90 m and 20 m). Model performance was tested by a standard validation method like receiver operating characteristics (ROC) analysis. The validation results in terms of accuracy (about 76%) and prediction ability (Area under the Curve (AUC) = 0.84) of the model revealed that the more detailed scale analysis is more appropriate for landslide susceptibility assessment and mapping in the catchment under investigation than the regional scale analysis
Immune response against viral infections and nucleic acid-based vaccines
No full textTo the editor,We read with interest the letter by Polykretis,1 which sum-marized some basic immunological principles in relation to COVID- 19 vaccines. However, the terminologies the author uses to describe the types of vaccines and targets of immune response are in our view misinforming and have basic errors
Adaptive neuro-fuzzy inference system (ANFIS) modeling for landslide susceptibility assessment in a Mediterranean hilly area
No full textStudying Land Use and Land Cover Spatial Patterns Distribution in Crete, Greece with Means of Satellite Remote Sensing
No full textMulti-temporal Land use and Land cover (LULC) monitoring is a crucial parameter for assessing an area’s landscape ecology regime. LULC changes can be effectively used to describe dynamics of both urban or rural environments and vegetation patterns as an important indicator of ecological environments. In this context, spatial land use properties can be quantified by using a set of landscape metrics. Landscape metrics capture inherent spatial structure of the environment and are used to enhance interpretation of spatial pattern of the landscape. This study aims to monitor diachronically the LULC regime of the island of Crete, Greece with the use of Landsat satellite imageries (Landsat 5, Landsat-7 and Landsat-8) in terms of soil erosion. For this reason, radiometric and atmospheric corrections are applied to all satellite products and unsupervised classification algorithms are used to develop detail LULC maps of the island. The LULC classes are developed by generalizing basic CORINE classes. Following, various landscape metrics are applied to estimate the temporal changes in LULC patterns of the island. The results denote that the diachronic research of spatial patterns evolution can effectively assist to the investigation of the structure, function and landscape pattern changes
Assessment and Mapping of Spatio-Temporal Variations in Human Mortality-Related Parameters at European Scale
No full textResearch efforts focusing on better understanding and capture of mortality progression over the time are considered to be of significant interest in the field of demography. On a demographic basis, mortality can be expressed by different physical parameters. The main objective of this study is the assessment and mapping of four such parameters at the European scale, during the time period 1993–2013. Infant mortality (parameter θ), population aging (parameter ξ), and individual and population mortality due to unexpected exogenous factors/events (parameter κ and λ, respectively) are represented from these parameters. Given that their estimation is based on demographics by age and cause of death, and in order to be examined and visualized by gender, time-specific mortality and population demographic data with respect to gender, age, and cause of death was used. The resulting maps present the spatial patterns of the estimated parameters as well as their variations over the examined period for both male and female populations of 22 European countries in all
Exploring the Impact of Various Spectral Indices on Land Cover Change Detection Using Change Vector Analysis: A Case Study of Crete Island, Greece
No full textThe main objective of this study was to explore the impact of various spectral indices on the performance of change vector analysis (CVA) for detecting the land cover changes on the island of Crete, Greece, between the last two decades (1999–2009 and 2009–2019). A set of such indices, namely, normalized difference vegetation index (NDVI), soil adjusted vegetation index (SAVI), albedo, bare soil index (BSI), tasseled cap greenness (TCG), and tasseled cap brightness (TCB), representing both the vegetation and soil conditions of the study area, were estimated on Landsat satellite images captured in 1999, 2009, and 2019. Change vector analysis was then applied for five different index combinations resulting to the relative change outputs. The evaluation of these outputs was performed towards detailed land cover maps produced by supervised classification of the aforementioned images. The results from the two examined periods revealed that the five index combinations provided promising performance results in terms of kappa index (with a range of 0.60–0.69) and overall accuracy (with a range of 0.86–0.96). Moreover, among the different combinations, the use of NDVI and albedo were found to provide superior results against the other combinations
