Retrospective evaluation of landslide susceptibility maps and review of validation practice

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

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

Natural Hazards and Earth System Sciences Discussions ◽

10.5194/nhessd-1-1001-2013 ◽

2013 ◽

Vol 1 (2) ◽

pp. 1001-1050 ◽

Cited By ~ 2

Author(s):

H. Petschko ◽

A. Brenning ◽

R. Bell ◽

J. Goetz ◽

T. Glade

Keyword(s):

Confidence Intervals ◽

Landslide Susceptibility ◽

Spatial Planning ◽

Cross Validation ◽

Susceptibility Map ◽

Landslide Occurrence ◽

High Interest ◽

Susceptibility Maps ◽

Landslide Susceptibility Maps

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

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

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

<p>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.</p>

Download Full-text

Stochastic procedure to extract and to integrate landslide susceptibility maps: an example of mountainous watershed in Taiwan

Natural Hazards and Earth System Science ◽

10.5194/nhess-6-803-2006 ◽

2006 ◽

Vol 6 (5) ◽

pp. 803-815 ◽

Cited By ~ 34

Author(s):

◽

Keyword(s):

Landslide Susceptibility ◽

Full Range ◽

Model Performance ◽

Susceptibility Map ◽

Risk Area ◽

High Risk Area ◽

Susceptibility Maps ◽

Mountainous Watershed ◽

Generalized Likelihood Uncertainty Estimation ◽

Landslide Susceptibility Maps

Abstract. The Generalized Likelihood Uncertainty Estimation (GLUE) is here incorporated into a deterministic landslide model (SHALSTAB) to generate 4000 landslide susceptibility maps which enclose various combinations of full range parameters. Furthermore, an improved index is adopted into GLUE as a criterion to measure model performance, and through that, 200 maps holding top 5% performance are retrieved. Proper ranges for parameters are obtained through GLUE yet they only perform well if combined appropriately. The 200 better maps are overlapped to construct an integrated landslide susceptibility map. Instead of giving a single parameter set or a single susceptibility map, the merit of extracting and integrating procedure is to envelope uncertainties inherited in model structure and input parameters. Bias due to subjective parameter input is potentially reduced. The entire procedure is applied to the Chi-Jia-Wan, a mountainous watershed in Taiwan. The integrated map shows high-risk area (>50% predicted landslide probability) only occupies 16.4% of the entire watershed while able to correctly identify 60% of the actual landslides. For areas above 2100 m height the map is even more successful (projects 77 of the 98 actual landslides). Interactions among parameters are discussed to highlight the unsolvable equifinality problem and improperness of presenting a single model result.

Download Full-text

On the influence of temporal change on the validity of landslide susceptibility maps

Natural Hazards and Earth System Science ◽

10.5194/nhess-9-1495-2009 ◽

2009 ◽

Vol 9 (4) ◽

pp. 1495-1507 ◽

Cited By ~ 29

Author(s):

K. Meusburger ◽

C. Alewell

Keyword(s):

Landslide Susceptibility ◽

Dynamic Change ◽

Swiss Alps ◽

Validation Dataset ◽

Susceptibility Map ◽

Causal Factors ◽

Logistic Regression Models ◽

Susceptibility Maps ◽

Time Element ◽

Landslide Susceptibility Maps

Abstract. The consideration of non-stationary landslide causal factors in statistical landslide susceptibility assessments is still problematic. The latter may lead to erroneous model predictions, especially in times of dramatic environmental change. In this case study in the Central Swiss Alps, we aim to evaluate the effect of dynamic change of landslide causal factors on the validity of landslide susceptibility maps. Logistic regression models were produced for two points in time, 1959 and 2000. Both models could correctly classify >70% of the independent spatial validation dataset. By subtracting the 1959 susceptibility map from the 2000 susceptibility map a deviation susceptibility map was obtained. Our interpretation was that these susceptibility deviations indicate the effect of the change of dynamic causal factors on the landslide probability. The deviation map explained 85% of new landslides occurring after 2000. We believe it to be a suitable tool to add a time element to the susceptibility map pointing to areas with changing susceptibility due to recently changing environmental conditions.

Download Full-text