&lt;b&gt;Variability in soil physical properties in landslide-prone areas

Fonseca, Letícia d’Agosto Miguel; Lani, João Luiz; Filho, Elpídio Inácio Fernandes; Santos, Gérson Rodrigues dos; Ferreira, Williams Pinto Marques; Santos, Adriana Maria Rocha Trancoso

doi:10.4025/actasciagron.v39i1.30561

Cited by 10 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But the value of the physical properties of the soil in the form of soil porosity, permeability, and bulk density of the soil is considered a soil factor that can affect the occurrence of landslides if it is associated with land use cultivated by the community. (Reichenbach et al 2014;Fonseca et al 2017a). Statistical test results indicated signi cant collaboration of the type of land use, slope, and physical properties of the soil (soil porosity, bulk density, and permeability) on the occurrence of landslides at the study site (Table 2).…”

Section: Discussionmentioning

confidence: 99%

“…Soil parameters have been used by several researchers in assessing landslide susceptibility, especially soil texture, macro-micro porosity, and erodibility (Fonseca et al 2017b;Conforti and Ietto 2021). The addition of soil micromorphology data in the form of plane voids has signi cantly affected the physical occurrence of landslides at the study site (Ahmad et al 2022a).…”

Section: Speci C Parameters For Landslide Susceptibility Mappingmentioning

confidence: 99%

See 1 more Smart Citation

Soil Micromorphology for Modeling Spatial on Landslide Susceptibility Mapping A Case Study in Kelara Subwatershed, Jeneponto Regency of South Sulawesi, Indonesia

Ahmad

Farida

Juita

et al. 2022

Preprint

View full text Add to dashboard Cite

Most of the results of classifying the level of susceptibility show different results, where landslides are more common in areas with a relatively high to moderate susceptibility class compared to those with a high susceptibility class. Differences in methods result in differences in the susceptibility maps resulting from the parameters that cause the tested landslides. The Spatial Regression Model can precisely interpret the relationship between several landslide parameters and events and shows better data accuracy than other methods. Utilization of soil micromorphological parameter data in mapping the level of susceptibility of the soil that triggers landslides with a Spatial Regression model so that the resulting susceptibility map can be more accurate. The soil parameter test method was carried out using a split-plot design with land use as the main plot, slope as a sub-plot, and soil physics (permeability, bulk density, and porosity) as a sub-sub-plot with three replications. Spatial modeling is done through regression analysis using ordinary least squares. The first test analysis was carried out with general parameters: lithology, rainfall, slope, land cover/land use, and population, while the second test was with parameters: lithology, rainfall, slope, land cover/land use, population, soil organic carbon, texture, erodibility and soil micromorphology. Classification of vulnerable classes using the natural breaks method. The interaction between the type of land use, slope, and physical properties of the soil on the occurrence of landslides at the study site shows a strong relationship with a significant p-value = 0.043 less than the α 5% level. Increased land use by the community has triggered the formation of soil micromorphology in the form of plane voids, cross-striated and grano-striated, which can trigger internal shifts (micro-shifts) in the soil body. The landslide susceptibility map at the study site is divided into seven spatial susceptibility classes: extremely low, very low, low, moderate, high, very high, and extremely high. Spatial modeling with OLS shows that the independent factors in the form of lithology, rainfall, slope, land cover/land use, and population only get an R2 value of 30.8%. Adding landslide independent parameter data in the form of soil organic carbon factor, texture, erodibility, and soil micromorphology produces a spatial model of landslide susceptibility with an increase in the accuracy value of R2 by 66.66%. The spatial model shows a high level of consistency with very significant soil micromorphology at a p-value < 0.01. The resulting spatial model is more accurate, where the high susceptibility class has a more significant number of landslide events, and landslides decrease according to the class.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Speci C Parameters For Landslide Susceptibility Mappingmentioning

confidence: 99%

Soil Micromorphology for Modeling Spatial on Landslide Susceptibility Mapping A Case Study in Kelara Subwatershed, Jeneponto Regency of South Sulawesi, Indonesia

Ahmad

Farida

Juita

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Lots of studies are conducted on the role of soil texture in occurrence of landslides (Lee et al 2017;Rahim et al 2018). The studies show that soils with higher silt content are more likely to erode compared to clay and sandy soils (Fonseca et al 2017). This factor has 4 classes incuding Silty Clay Loam, Loam, Silty Loam, and Clay Loam.…”

Section: Soil Texturementioning

confidence: 99%

Novel Ensemble-Based Machine Learning Models Based on The Bagging, Boosting and Random Subspace Methods for Landslide Susceptibility Mapping

Al-Fugara

Ahmadlou

Pradhan

2021

Preprint

View full text Add to dashboard Cite

Indivisual machine learning models show different limitations such as low generalization power for modeling nonlinear phenomena with complex behavior. In recent years, one of the best approaches to this issue is to use ensemble models. The purpose of this paper is to investigate the predictive power and modeling of three novel ensemble models constructed with four machine learning models: Decision Tree (DT), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naive Bayes (NB) models based on three approaches of Bagging, boosting and Random Subspace (RS) in landslide susceptibility mapping (LSM) in the Province of Ajloun in Jordan. A total number of 91 landslide locations along with 16 conditioning factors in LSM were identified and used. Also, before modeling, the selection of effective conditioning factors in LSM was done using genetic algorithm and four single models including DT, KNN, NB and SVM. The selected factors were used in modeling with individual and ensemble models. The results show that the area under the receiver operating characteristic curve (AUROC) for ensemble models is significantly higher than the individual models and the AUC for ensemble models was on average 14% higher than individual models. Based on the results, the most accurate models were RS ensemble model (AUROC = 0.850), Boosting (AUROC = 0.848) and Bagging (AUROC = 0.814), respectively. This study showed that by combining the results of simple machine learning models and making ensemble models, models with the desired accuracy can be achieved.

show abstract

“…Comparing the AHP and RF maps, with some limiting factors for micro-dam construction, it was evident that in the multicriteria analysis, there were many areas with high potential (levels 4 and 5) in steep slope areas (Figure 5a). Overestimating the high potential levels for micro-dam in hillslope can be a critical problem, since these zones coincide with the dominance of the Inceptisols, soils with higher silt content and more susceptible to erosion (FU et al, 2011;FONSECA et al, 2017). On the other hand, the RF did not overestimate the high and very high potential levels in in hillslopes.…”

Section: Prediction Of Potential Areas For Dams (Ahp and Rf)mentioning

confidence: 99%

Multicriteria Analysis and Machine Learning Algorithm for Definition of Areas for Micro-Dam, Southeastern Brazil

Souza

Veloso

Faria

et al. 2021

RCG

Self Cite

View full text Add to dashboard Cite

As barraginhas são eficientes no controle da erosão e aumentam a taxa de infiltração de água no solo. Este estudo visa prever áreas potenciais para alocação de barraginhas. Preparamos o mapa de áreas potenciais para alocação de barraginhas a partir da análise multicritério (AHP) usando dados topográficos e uso do solo. Além disso, desenvolvemos uma estrutura metodológica com algoritmo Random Forest (RF) para prever áreas potenciais de barraginhas a partir de pontos extraídos do mapa gerado por AHP e auxiliados pela inserção de covariáveis; e também aplicamos a sobreposição dos dois mapas (AHP + RF). O método AHP superestima as áreas de alto potencial em zonas sem potencialidade. O modelo de RF usou sete covariáveis topográficas, e elas estão relacionadas aos fluxos hidrológicos. O desempenho de R2 no RF foi de 0,43, estaticamente satisfatório, mas espacialmente subestimou as classes de potencial muito baixo e muito alto. A superposição da classe de baixo potencial do método AHP no mapa de RF favoreceu uma redução de 99% das áreas de muito alto potencial em zonas não adequadas para alocação de micro-barragens. Portanto, a combinação do método AHP e ML gera resultados espaciais mais consistentes.

show abstract

<b>Variability in soil physical properties in landslide-prone areas

Cited by 10 publications

References 18 publications

Soil Micromorphology for Modeling Spatial on Landslide Susceptibility Mapping A Case Study in Kelara Subwatershed, Jeneponto Regency of South Sulawesi, Indonesia

Soil Micromorphology for Modeling Spatial on Landslide Susceptibility Mapping A Case Study in Kelara Subwatershed, Jeneponto Regency of South Sulawesi, Indonesia

Novel Ensemble-Based Machine Learning Models Based on The Bagging, Boosting and Random Subspace Methods for Landslide Susceptibility Mapping

Multicriteria Analysis and Machine Learning Algorithm for Definition of Areas for Micro-Dam, Southeastern Brazil

Contact Info

Product

Resources

About