Assessing the quality of landslide susceptibility maps – case study Lower Austria

Abstract. Landslide susceptibility maps are helpful tools to identify areas which might be prone to future landslide occurrence. As more and more national and provincial authorities demand for these maps to be computed and implemented in spatial planning strategies, the quality of the landslide susceptibility map and of the model applied to compute them is of high interest. In this study we focus on the analysis of the model performance by a repeated k-fold cross-validation with spatial and random subsampling. Furthermore, the focus is on the analysis of the implications of uncertainties expressed by confidence intervals of model predictions. The cross-validation performance assessments reflects the variability of performance estimates compared to single hold-out validation approaches that produce only a single estimate. The analysis of the confidence intervals shows that in 85% of the study area, the 95% confidence limits fall within the same susceptibility class. However, there are cases where confidence intervals overlap with all classes from the lowest to the highest class of susceptibility to landsliding. Locations whose confidence intervals intersect with more than one susceptibility class are of high interest because this uncertainty may affect spatial planning processes that are based on the susceptibility level.

Download Full-text

Assessing the quality of landslide susceptibility maps – case study Lower Austria

Natural Hazards and Earth System Science ◽

10.5194/nhess-14-95-2014 ◽

2014 ◽

Vol 14 (1) ◽

pp. 95-118 ◽

Cited By ~ 101

Author(s):

H. Petschko ◽

A. Brenning ◽

R. Bell ◽

J. Goetz ◽

T. Glade

Keyword(s):

Landslide Susceptibility ◽

Additive Model ◽

Susceptibility Map ◽

Model Quality ◽

Open Communication ◽

Entire Study ◽

Landslide Occurrence ◽

Susceptibility Maps ◽

Landslide Susceptibility Maps

Abstract. Landslide susceptibility maps are helpful tools to identify areas potentially prone to future landslide occurrence. As more and more national and provincial authorities demand for these maps to be computed and implemented in spatial planning strategies, several aspects of the quality of the landslide susceptibility model and the resulting classified map are of high interest. In this study of landslides in Lower Austria, we focus on the model form uncertainty to assess the quality of a flexible statistical modelling technique, the generalized additive model (GAM). The study area (15 850 km2) is divided into 16 modelling domains based on lithology classes. A model representing the entire study area is constructed by combining these models. The performances of the models are assessed using repeated k-fold cross-validation with spatial and random subsampling. This reflects the variability of performance estimates arising from sampling variation. Measures of spatial transferability and thematic consistency are applied to empirically assess model quality. We also analyse and visualize the implications of spatially varying prediction uncertainties regarding the susceptibility map classes by taking into account the confidence intervals of model predictions. The 95% confidence limits fall within the same susceptibility class in 85% of the study area. Overall, this study contributes to advancing open communication and assessment of model quality related to statistical landslide susceptibility models.

Download Full-text

Integrating landslide susceptibility maps into warning models at regional scale in Italy

10.5194/egusphere-egu21-2005 ◽

2021 ◽

Author(s):

Gaetano Pecoraro ◽

Michele Calvello

Keyword(s):

Spatial Resolution ◽

Land Use Planning ◽

Landslide Susceptibility ◽

Spatial Information ◽

Regional Scale ◽

Susceptibility Map ◽

Rainfall Thresholds ◽

Landslide Occurrence ◽

Susceptibility Maps ◽

Landslide Susceptibility Maps

The importance of susceptibility maps in the initial phase of landslide hazard and risk assessment is widely recognized in the literature, since they provide to stakeholders a general overview of the location of landslide prone areas. Usually, the use of these maps is limited to support land use planning. However, many researchers have recently recognized that susceptibility maps may also be used to improve the performance and spatial resolution of landslide warning at regional scale and provide a better updating of hazard assessment over time. Indeed, landslides prediction may be difficult at regional scale only considering rainfall condition, due to the difference of the spatial and temporal distribution of rainfall and the complex diversity of the disaster-prone environment (topography, geology, and lithology). As a result, a critical issue of models solely based on rainfall thresholds may be the issuing of warnings in areas that are not prone to landslide occurrence, resulting in an excessive number of false positives. In this work, we propose a methodology aimed at combining a susceptibility map and a set of rainfall thresholds by using a matrix approach to refine the performance of an early warning model at regional scale. The main aim is the combination of rainfall thresholds (typically used to accomplish a dynamic temporal forecasting with good temporal resolution but very coarse spatial resolution), with landslide susceptibility maps (providing static spatial information about the probability of landslide occurrence with a finer resolution). The methodology presented herein could allow a better prediction of &#8220;where&#8221; and &#8220;when&#8221; landslides may occur, thus: i) allowing to define a time-dependent level of hazard associated to their possible occurrence, and ii) markedly refining the spatial resolution of warning models employed at regional scale, given that areas susceptible to landslides typically represent only a fraction of territorial warning zones.

Download Full-text

Significance of landslide susceptibility maps in creation of spatial planning documentation

Proceedings of the 4th Regional Symposium on Landslides in the Adriatic - Balkan Region ◽

10.35123/resylab_2019_22 ◽

2019 ◽

Author(s):

Kenan Mandžić ◽

Nedreta Kikanović ◽

Adnan Ibrahimović

Keyword(s):

Landslide Susceptibility ◽

Spatial Planning ◽

Susceptibility Maps ◽

Landslide Susceptibility Maps

Download Full-text

Spatio-temporal landslide forecasting based on combination of rainfall thresholds and landslide susceptibility maps: a test in the Northern Apennines (Italy).

10.5194/egusphere-egu2020-9124 ◽

2020 ◽

Author(s):

Ascanio Rosi ◽

Samuele Segoni ◽

Veronica Tofani ◽

Filippo Catani

Keyword(s):

Spatial Resolution ◽

Landslide Susceptibility ◽

End Users ◽

Rainfall Thresholds ◽

Landslide Occurrence ◽

Susceptibility Maps ◽

Coarse Spatial Resolution ◽

Landslide Forecasting ◽

Spatio Temporal ◽

Landslide Susceptibility Maps

Landslide forecasting and early warning at regional scale are difficult task and they are usually accomplished by the mean of statistical approaches aimed to define rainfall thresholds and landslide susceptibility maps. &#160;Landslide susceptibility maps are based on the analysis of predisposing factors to assess the spatial probability of landslide occurrence, while rainfall thresholds are based on the correlation, valid on a wide area, between landslide occurrence and triggering factors, which usually are a couple of rainfall parameters, such as rainfall duration and intensity. Susceptibility maps are static map that can be used for the spatial prediction of the most landslide prone areas, nut cannot be used to predict the temporal occurrence of a landslide triggering. Rainfall thresholds can be used for temporal prediction, but with a coarse spatial resolution (usually some hundreds or thousands of km2), and the reference areas could contains both plains and hillslopes, so the alerts could involves both areas, even if landslides are improbable in river plains; this means that rainfall thresholds are not very suitable to identify the most probable triggering sites. Rainfall thresholds and susceptibility maps can be therefore conveniently combined into dynamic hazard matrixes to obtain spatio-temporal forecasts of landslide hazard. To combine these inputs, they are combined in a purposely-built hazard matrix, where each parameter is classified into 3 classes: landslide susceptibility map has been classified in S1 (low susceptibility), S2 (medium susceptibility) and S3 (high susceptibility), while rainfall rate has been classified in the classes R1, R2 and R3, by the definition of 2 rainfall thresholds. The combination of the aforementioned classes allowed to define a matrix with 5 hazard classes, from H0 (null hazard) to H4 (high hazard), which was calibrated so that there was not any landslide in the H0 class and that the 90% of the landslide were in H2-H4 classes. The result of this procedure is a dynamic hazard map, where the hazard, which is calculated for each pixel, can change over the time, based on rainfall rate variations. For operational purposes, such a map cannot be used, since the pixel based resolution is too fine to be used during an emergency or to plan any activity in the planification phase, so the results have been aggregated at municipality scale, which is more easily readable for the end-users as local administrators and decision makers. In this way it is possible to overcome the issues due to the stillness of susceptibility maps and to the coarse spatial resolution of rainfall thresholds, also avoiding results which could be hardly understandable outside of the scientific community. This procedure was tested in a test site located in Northern Tuscany (Italy) and the work showed the possibility of obtaining results which are balanced between the scientific soundness and the needs of end-users like mayors, local administrators and civil protection personnel.&#160;

Download Full-text

Retrospective evaluation of landslide susceptibility maps and review of validation practice

Environmental Earth Sciences ◽

10.1007/s12665-021-09770-9 ◽

2021 ◽

Vol 80 (15) ◽

Author(s):

Paul Fleuchaus ◽

Philipp Blum ◽

Martina Wilde ◽

Birgit Terhorst ◽

Christoph Butscher

Keyword(s):

Landslide Susceptibility ◽

Predictive Accuracy ◽

Predictive Performance ◽

Susceptibility Map ◽

Retrospective Evaluation ◽

Widespread Application ◽

Evaluation Approach ◽

Susceptibility Maps ◽

Landslide Event ◽

Landslide Susceptibility Maps

AbstractDespite the widespread application of landslide susceptibility analyses, there is hardly any information about whether or not the occurrence of recent landslide events was correctly predicted by the relevant susceptibility maps. Hence, the objective of this study is to evaluate four landslide susceptibility maps retrospectively in a landslide-prone area of the Swabian Alb (Germany). The predictive performance of each susceptibility map is evaluated based on a landslide event triggered by heavy rainfalls in the year 2013. The retrospective evaluation revealed significant variations in the predictive accuracy of the analyzed studies. Both completely erroneous as well as very precise predictions were observed. These differences are less attributed to the applied statistical method and more to the quality and comprehensiveness of the used input data. Furthermore, a literature review of 50 peer-reviewed articles showed that most landslide susceptibility analyses achieve very high validation scores. 73% of the analyzed studies achieved an area under curve (AUC) value of at least 80%. These high validation scores, however, do not reflect the high uncertainty in statistical susceptibility analysis. Thus, the quality assessment of landslide susceptibility maps should not only comprise an index-based, quantitative validation, but also an additional qualitative plausibility check considering local geomorphological characteristics and local landslide mechanisms. Finally, the proposed retrospective evaluation approach cannot only help to assess the quality of susceptibility maps and demonstrate the reliability of such statistical methods, but also identify issues that will enable the susceptibility maps to be improved in the future.

Download Full-text

GIS-Based Logistic Regression for Landslide Susceptibility Analysis in Western Washington State

International Journal of Applied Geospatial Research ◽

10.4018/ijagr.2017040101 ◽

2017 ◽

Vol 8 (2) ◽

pp. 1-19 ◽

Cited By ~ 1

Author(s):

Lucas A. Dailey ◽

Sven Fuhrmann

Keyword(s):

Logistic Regression ◽

Landslide Susceptibility ◽

Puget Sound ◽

Washington State ◽

Susceptibility Map ◽

The North ◽

Susceptibility Maps ◽

Western Washington ◽

Landslide Susceptibility Maps ◽

Puget Sound Region

The Oso landslide, one of the most recent disasters, occurred on March 22nd, 2014 in western Washington State. It caused significant property damage and killed over 40 people. As a result, a renewed interest has emerged for creating more accurate landslide susceptibility maps for this region. Research addressing landslide susceptibility within the north Puget Sound region of western Washington is lacking; therefore, this study develops a probabilistic GIS-based landslide susceptibility model for the north Puget Sound region. Multivariate logistic regression was utilized to create a landslide susceptibility map of Whatcom, Skagit, Snohomish, and King Counties. To predict probable areas of landslide occurrence, a landslide inventory map was prepared and fourteen topographic, geologic, environmental, and climatic predictor variables were considered. This research aims to assist in restructuring western Washington's landslide policies, and could serve as the first step in producing more accurate landslide susceptibility maps for the region.

Download Full-text

Landslide Susceptibility in Majalengka Regency, West Java Province

IOP Conference Series Earth and Environmental Science ◽

10.1088/1755-1315/884/1/012053 ◽

2021 ◽

Vol 884 (1) ◽

pp. 012053

Author(s):

S Selaby ◽

E Kusratmoko ◽

A Rustanto

Keyword(s):

Landslide Susceptibility ◽

Public Works ◽

Disaster Mitigation ◽

Susceptibility Map ◽

Landslide Susceptibility Map ◽

Multi Criteria Evaluation ◽

Susceptibility Maps ◽

The Government ◽

The Impact ◽

Landslide Susceptibility Maps

Abstract Majalengka is one of districts in Indonesia which is susceptible to landslides. Landslides in Majalengka caused enormous losses such as damage to infrastructure, loss of property, and even human fatalities. Seeing of the impact, mitigation efforts are needed to reduce risks and losses by making landslide susceptibility maps. This study aims to map areas landslide susceptibility and as a reference for the government and related agencies to reduce losses. The method used overlay using Spatial Multi-Criteria Evaluation (SMCE), using weighting values from the Minister Public Works Regulation NO.22/PRT/M/2007, Puslittanak Bogor (2014) and Directorate Volcanology and Disaster Mitigation (DVMBG) (2004). Then comparison of these sources is carried out to determine weighting value with the highest accuracy. The variables are slope, rainfall, soil type, lithology, and land use. The results of this study indicate that landslide susceptibility areas are divided into non-susceptible, low, moderate, and high areas. Where areas Majalengka Regency is dominated by moderate susceptibility level. For the accuracy value of the landslide susceptibility map produced by the weighted value source from the Minister of Public Works Regulation NO.22/PRT/M/2007 has the highest accuracy value of 76%. For weighting from the Bogor Puslittanak is 73%, while weighting source from DVMBG is 68%.

Download Full-text

Application of Ensemble-Based Machine Learning Models to Landslide Susceptibility Mapping

Remote Sensing ◽

10.3390/rs10081252 ◽

2018 ◽

Vol 10 (8) ◽

pp. 1252 ◽

Cited By ~ 50

Author(s):

Prima Kadavi ◽

Chang-Wook Lee ◽

Saro Lee

Keyword(s):

Machine Learning ◽

Landslide Susceptibility ◽

Susceptibility Mapping ◽

Landslide Susceptibility Mapping ◽

Learning Models ◽

Landslide Occurrence ◽

Susceptibility Maps ◽

Multiclass Classifier ◽

Landslide Susceptibility Maps ◽

Machine Learning Models

The main purpose of this study was to produce landslide susceptibility maps using various ensemble-based machine learning models (i.e., the AdaBoost, LogitBoost, Multiclass Classifier, and Bagging models) for the Sacheon-myeon area of South Korea. A landslide inventory map including a total of 762 landslides was compiled based on reports and aerial photograph interpretations. The landslides were randomly separated into two datasets: 70% of landslides were selected for the model establishment and 30% were used for validation purposes. Additionally, 20 landslide condition factors divided into five categories (topographic factors, hydrological factors, soil map, geological map, and forest map) were considered in the landslide susceptibility mapping. The relationships among landslide occurrence and landslide conditioning factors were analyzed and the landslide susceptibility maps were calculated and drawn using the AdaBoost, LogitBoost, Multiclass Classifier, and Bagging models. Finally, the maps were validated using the area under the curve (AUC) method. The Multiclass Classifier method had higher prediction accuracy (85.9%) than the Bagging (AUC = 85.4%), LogitBoost (AUC = 84.8%), and AdaBoost (84.0%) methods.

Download Full-text

Comparing multi-criteria methods for landslide susceptibility mapping in Chania Prefecture, Crete Island, Greece

Natural Hazards and Earth System Sciences Discussions ◽

10.5194/nhessd-1-73-2013 ◽

2013 ◽

Vol 1 (1) ◽

pp. 73-109 ◽

Cited By ~ 6

Author(s):

M. Kouli ◽

C. Loupasakis ◽

P. Soupios ◽

D. Rozos ◽

F. Vallianatos

Keyword(s):

Landslide Susceptibility ◽

Landslide Susceptibility Mapping ◽

Subjective Knowledge ◽

Susceptibility Maps ◽

Land Use Types ◽

Crete Island ◽

Gis Environment ◽

Geological Formations ◽

Landslide Susceptibility Maps

Abstract. In this work, two multi-criteria methods, an expert-based, semi-quantitative, relative weighting – rating approach, the weighted linear combination (WLC) and a quantitative, statistical method, the weights of evidence (WoE) approach were applied for landslide susceptibility zonation mapping in the Chania Prefecture of Crete Island, Greece. Several thematic maps representing various landslide casual factors, such as geological formations, faults proximity, elevation, slope gradient, aspect and curvature, rivers proximity, precipitation, roads proximity and land use types; have been generated in a GIS environment. Two landslide susceptibility maps were created; one for each method. The maps were compared and validated using the success rate curve (SRC) analysis. The resulting landslide susceptibility maps have uncertainties introduced due to the subjective knowledge of experts in the case of WLC method and to the quality of the recorded landslides sample in the case of the WoE method. Both approaches produced almost equally accurate maps with the WoE method to produce slightly superior predictions.

Download Full-text