Functional soil-landscape modelling to estimate slope stability in a steep Andean mountain forest region

Ließ, Mareike; Glaser, Bruno; Huwe, Bernd

doi:10.1016/j.geomorph.2011.05.015

Cited by 18 publications

(6 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This wide range has to be attributed to the very small dataset. Ließ et al (2011) reach an equally high model uncertainty when predicting bulk density using the same dataset. In contrast to this, Ließ (2011) achieved a lower variability in r xy by using a bigger dataset of 315 auger points to predict soil horizon thickness and occurrence probability.…”

Section: Regression Tree and Random Forest Model Performancementioning

confidence: 79%

Uncertainty in the spatial prediction of soil texture

2012

Self Cite

View full text Add to dashboard Cite

Section: Regression Tree and Random Forest Model Performancementioning

confidence: 79%

Uncertainty in the spatial prediction of soil texture

2012

Self Cite

View full text Add to dashboard Cite

“…In a second phase, a statistical classification model, for example logistic regression (LR) (Atkinson and Massari 1998;Ayalew and Yamagishi 2005;Ohlmacher and Davis 2003), discriminant analysis (Adrizzone et al 2002;Carrara 1983) or generalized additive models (Goetz et al 2011;Park and Chi 2008), is applied to discriminate landslide from non-landslide pixels/objects in the multidimensional space of topographic, geological and land use features. Due to the complexity of topographic and geological conditions associated with landslide occurrence, more flexible nonlinear methods such as machine learning algorithms have also been considered, in particular artificial neural networks (Ermini et al 2005;Lee et al 2004;Melchiorre et al 2008;Neaupane and Achet 2004;Yilmaz 2010b), support vector machines (SVM) (Ballabio and Sterlacchini 2012;Yao et al 2008;Yilmaz 2010a), Gaussian processes (Gallus 2010) and random forests (RF) (Liess et al 2011;Stumpf and Kerle 2011). Great effort has been placed upon comparing these different approaches both for LS mapping (Brenning 2005) and in a broader context for geomorphological and landform mapping (Brenning 2009;Marmion et al 2008Marmion et al , 2009Nefeslioglu et al 2008;Pradhan and Lee 2010;Yesilnacar and Topal 2005).…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Feature Selection Methods for Landslide Susceptibility Mapping

et al. 2013

View full text Add to dashboard Cite

This paper explores the use of adaptive support vector machines, random forests and AdaBoost for landslide susceptibility mapping in three separated regions of Canton Vaud, Switzerland, based on a set of geological, hydrological and morphological features. The feature selection properties of the three algorithms are studied to analyze the relevance of features in controlling the spatial distribution of landslides. The elimination of irrelevant features gives simpler, lower dimensional models while keeping the classification performance high. An object-based sampling procedure is considered to reduce the spatial autocorrelation of data and to estimate more reliably generalization skills when applying the model to predict the occurrence of new unknown landslides. The accuracy of the models, the relevance of features and the quality of landslide susceptibility maps were found to be high in the regions characterized by shallow landslides and low in the ones with deep-seated landslides. Despite providing similar skill, random forests and AdaBoost were found to be more efficient in performing feature selection than adaptive support vector machines. The results of this study reveal the strengths of the classification algorithms, but evidence: (1) the need for relying on more than one method for the identification of relevant variables;

show abstract

“…The lower probability below 2100 m a.s.l. must be attributed to the higher inclination that supports a higher discharge of surface and subsurface flow and the higher bulk soil density [52]. In contrast, particularly the flat platform-like areas above 2100 m show a much higher probability.…”

Section: Resultsmentioning

confidence: 98%

The Sloping Mire Soil-Landscape of Southern Ecuador: Influence of Predictor Resolution and Model Tuning on Random Forest Predictions

Ließ

Huwe

2014

Applied and Environmental Soil Science

Self Cite

View full text Add to dashboard Cite

The sloping mire landscape of the investigation area, in the southern Andes of Ecuador, is dominated by stagnic soils with thick organic layers. The recursive partitioning algorithm Random Forest was used to predict the spatial water stagnation pattern and the thickness of the organic layer from terrain attributes. Terrain smoothing from 10 to 30 m raster resolution was applied in order to obtain the best possible model. For the same purpose, several model tuning parameters were tested and a prepredictor selection with the R-package Boruta was applied. Model versions were evaluated and compared by 100 repetitions of the calculation of the residual mean square error of a five-fold cross-validation. Position specific density functions of the predicted soil parameters were then used to display prediction uncertainty. Prepredictor selection and tuning of the Random Forest algorithm in some cases resulted in an improved model performance. We therefore recommend testing prepredictor selection and tuning to make sure that the best possible model is chosen. This needs particular emphasis in complex tropical mountain soil-landscapes which provide a real challenge to any soil mapping approach but where Random Forest has proven to be successful due to the testing of model tuning and prepredictor selection.

show abstract

Functional soil-landscape modelling to estimate slope stability in a steep Andean mountain forest region

Cited by 18 publications

References 25 publications

Uncertainty in the spatial prediction of soil texture

Uncertainty in the spatial prediction of soil texture

Machine Learning Feature Selection Methods for Landslide Susceptibility Mapping

The Sloping Mire Soil-Landscape of Southern Ecuador: Influence of Predictor Resolution and Model Tuning on Random Forest Predictions

Contact Info

Product

Resources

About