Using and comparing two nonparametric methods (CART and MARS) to model the potential distribution of gullies

Gutiérrez, Álvaro Gómez; Schnabel, Susanne; Contador, Joaquín Francisco Lavado

doi:10.1016/j.ecolmodel.2009.06.020

Cited by 104 publications

(52 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although models developed for gully erosion commonly deal with the estimation of erosion rates (Knisel, 1980;Merkel et al, 1988;Knisel, 1993;Flanagan and Nearing, 1995;Sidorchuk, 1999), only a few authors focused on gully erosion susceptibility (Meyers and Martínez-Casasnovas, 1999;Bou Kheir et al, 2007;Conoscenti et al, 2008;Fernández et al, 2008;Gómez Gutiérrez et al, 2009;Krishna Bahadur, 2009;Pike et al, 2009;Conforti et al, 2011). Conversely, it is possible to find many studies in literature for landslide susceptibility and hazard mapping.…”

Section: Introductionmentioning

confidence: 94%

“…After the estimation of the contribution of each predisposing factors, the area is divided in zones characterized by different susceptibilities (Soeters and van Westen, 1996;Aleotti and Chowdhury, 1999;Guzzetti et al, 1999). These statistical techniques, not common in gully erosion research (Gómez Gutiérrez et al, 2009), can be grouped into bivariate and multivariate methods. In bivariate statistics, each parameter is analyzed individually and the calculation and application are straightforward and efficient (Suzen and Doyuran, 2004).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of GIS-based gullying susceptibility mapping using bivariate and multivariate statistics: Northern Calabria, South Italy

2011

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 94%

Section: Introductionmentioning

confidence: 99%

Comparison of GIS-based gullying susceptibility mapping using bivariate and multivariate statistics: Northern Calabria, South Italy

2011

View full text Add to dashboard Cite

“…Meanwhile, according to the literature, the BRT, CART, and RF models haven't been used in the groundwater potential mapping, but several studies have been applied to assess accuracy of the mentioned machine learning models in different cases such as landslide susceptibility and hazard mapping (Stumpf and Kernel 2011;Vorpahl et al 2012;Lee et al 2013;Trigila et al 2013), ground subsidence hazard mapping (Oh and Lee 2010), wildfire (Oliveira et al 2012;Leuenberger et al 2013), gully susceptibility mapping (Gutiérrez et al 2009a(Gutiérrez et al , 2009b, ecology (Elith et al 2008;Aertsen et al 2010Aertsen et al , 2011, environmental modeling (Bachmair and Weiler 2012;Catani et al 2013). According to the aforementioned literature, machine learning models had better performance than bivariate and multivariate models in different studies.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Assessment Between Three Machine Learning Models and Their Performance Comparison by Bivariate and Multivariate Statistical Methods in Groundwater Potential Mapping

Naghibi

Pourghasemi

2015

Water Resour Manage

230

View full text Add to dashboard Cite

As demand for fresh groundwater in the worldwide is increasing, delineation of groundwater spring potential zones become an increasingly important tool for implementing a successful groundwater determination, protection, and management programs. Therefore, the objective of current study is to evaluate the capability of three machine learning models such as boosted regression tree (BRT), classification and regression tree (CART), and random forest (RF), and comparison of their performance by bivariate (evidential belief function (EBF)), and multivariate (general linear model (GLM)) statistical methods in the groundwater potential mapping. This study was carried out in the Beheshtabad Watershed, Chaharmahal-e-Bakhtiari Province, Iran. In total, 1425 spring locations were detected in the study area. Seventy percent of the spring locations were used for model training, and 30 % for validation purposes. Fourteen conditioning-factors were considered in this investigation, including slope angle, slope aspect, altitude, plan curvature, profile curvature, slope length (LS), stream power index (SPI), topographic wetness index (TWI), distance from rivers, distance from faults, river density, fault density, lithology, and land use. Using the above conditioning factors and different algorithms, groundwater potential maps were generated, and the results were plotted in ArcGIS 9.3. According to the results of success rate curves (SRC), values of area under the curve (AUC) for the five models vary from 0.692 to 0.975. In contrast, the AUC for prediction rate curves (PRC) ranges from 77.26 to 86.39 %. The CART, BRT, and RF machine learning techniques showed very good performance in groundwater potential mapping with the AUC Water Resour Manage values of 86.39, 86.12, and 86.05 %, respectively. By the way, The GLM and EBF models in comparison by machine learning models showed weaker performance in spring groundwater potential mapping by the AUC values of 77.26, and 67.72 %, respectively. The proposed methods provided rapid, accurate, and cost effective results. Furthermore, the analysis may be transferable to other watersheds with similar topographic and hydro-geological characteristics.

show abstract

“…This method is able to describe complex data structures and is useful in data exploration (Breiman et al 1984). It has been used, for instance, to predict tree species distribution (Thuiller et al 2003) and map potential areas of gullies (Gómez-Gutiérrez et al 2009). Moreover, the results can be converted into probabilities (Breiman 1996;Provost and Domingos 2003), which may be used to estimate the effects of different forest management options on surface erosion.…”

Section: Introductionmentioning

confidence: 99%

Site and stand characteristics related to surface erosion occurrence in forests of Catalonia (Spain)

2011

View full text Add to dashboard Cite

This study aims at identifying forest areas affected by surface erosion in Catalonia. It analyses the characteristics of forests that are related to erosion occurrence. The data on erosion observations and stand variables were obtained from the Third Spanish National Forest Inventory (2000)(2001). We used the classification tree method to study the presence-absence of surface erosion in four different forest types, pure coniferous, pure broadleaf and mixed stands as well in forested semiarid areas. The method provided a description of the site and stand variables, which are associated with the occurrence of surface erosion. The results revealed that forest type and stand structure are strongly related to the probability of surface erosion occurrence. In pure broadleaf forests, surface erosion occurrence was greater in dense stands, whereas in pure coniferous pine forest, the erosion occurrence was greater in sparse stands. Surface erosion occurrence was the highest in stands dominated by Fagus sylvatica and Abies alba. In order to estimate the erosion probabilities of stands, we converted the results of the classification tree analysis into erosion probabilities and mapped forest areas as low, moderate or high surface erosion risk. The accuracy of the erosion probabilities was assessed using the ROC curve, which gave a fair level (0.75) of accuracy for the total classification tree, the best (0.78) for pure broadleaf and the lowest (0.66) for mixed forest. The results of this study can be applied in future forest management planning aiming to reduce erosion risk in forests.

show abstract

Using and comparing two nonparametric methods (CART and MARS) to model the potential distribution of gullies

Cited by 104 publications

References 36 publications

Comparison of GIS-based gullying susceptibility mapping using bivariate and multivariate statistics: Northern Calabria, South Italy

Comparison of GIS-based gullying susceptibility mapping using bivariate and multivariate statistics: Northern Calabria, South Italy

A Comparative Assessment Between Three Machine Learning Models and Their Performance Comparison by Bivariate and Multivariate Statistical Methods in Groundwater Potential Mapping

Site and stand characteristics related to surface erosion occurrence in forests of Catalonia (Spain)

Contact Info

Product

Resources

About