How can statistical and artificial intelligence approaches predict piping erosion susceptibility?

Mohsen Hosseinalizadeh*, Narges Kariminejad, Omid Rahmati, Saskia Keesstra, Mohammad Alinejad, Ali Mohammadian Behbahani

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

12 Citations (Scopus)

Abstract

It is of fundamental importance to model the relationship between geo-environmental factors and piping erosion because of the environmental degradation attributed to soil loss. Methods that identify areas prone to piping erosion at the regional scale are limited. The main objective of this research is to develop a novel modeling approach by using three machine learning algorithms—mixture discriminant analysis (MDA), flexible discriminant analysis (FDA), and support vector machine (SVM) in addition to an unmanned aerial vehicle (UAV) images to map susceptibility to piping erosion in the loess-covered hilly region of Golestan Province, Northeast Iran. In this research, we have used 22 geo-environmental indices/factors and 345 identified pipes as predictors and dependent variables. The piping susceptibility maps were assessed by the area under the ROC curve (AUC). Validation of the results showed that the AUC for the three mentioned algorithms varied from 90.32% to 92.45%. We concluded that the proposed approach could efficiently produce a piping susceptibility map.

Original languageEnglish
Pages (from-to)1554-1566
JournalScience of the Total Environment
Volume646
DOIs
Publication statusPublished - 1 Jan 2019

Fingerprint

artificial intelligence
piping
Artificial intelligence
Erosion
Discriminant analysis
erosion
discriminant analysis
Weathering
Unmanned aerial vehicles (UAV)
Support vector machines
Learning systems
Pipe
Soils
environmental degradation
loess
environmental factor
pipe
modeling
soil

Keywords

  • Loess plateau
  • Machine learning algorithms
  • Piping collapse
  • Susceptibility map
  • Unmanned aerial vehicle (UAV)

Cite this

Hosseinalizadeh, Mohsen ; Kariminejad, Narges ; Rahmati, Omid ; Keesstra, Saskia ; Alinejad, Mohammad ; Mohammadian Behbahani, Ali. / How can statistical and artificial intelligence approaches predict piping erosion susceptibility?. In: Science of the Total Environment. 2019 ; Vol. 646. pp. 1554-1566.
@article{63d980c0bbef4c8c848e6e16019cec44,
title = "How can statistical and artificial intelligence approaches predict piping erosion susceptibility?",
abstract = "It is of fundamental importance to model the relationship between geo-environmental factors and piping erosion because of the environmental degradation attributed to soil loss. Methods that identify areas prone to piping erosion at the regional scale are limited. The main objective of this research is to develop a novel modeling approach by using three machine learning algorithms—mixture discriminant analysis (MDA), flexible discriminant analysis (FDA), and support vector machine (SVM) in addition to an unmanned aerial vehicle (UAV) images to map susceptibility to piping erosion in the loess-covered hilly region of Golestan Province, Northeast Iran. In this research, we have used 22 geo-environmental indices/factors and 345 identified pipes as predictors and dependent variables. The piping susceptibility maps were assessed by the area under the ROC curve (AUC). Validation of the results showed that the AUC for the three mentioned algorithms varied from 90.32{\%} to 92.45{\%}. We concluded that the proposed approach could efficiently produce a piping susceptibility map.",
keywords = "Loess plateau, Machine learning algorithms, Piping collapse, Susceptibility map, Unmanned aerial vehicle (UAV)",
author = "Mohsen Hosseinalizadeh and Narges Kariminejad and Omid Rahmati and Saskia Keesstra and Mohammad Alinejad and {Mohammadian Behbahani}, Ali",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.scitotenv.2018.07.396",
language = "English",
volume = "646",
pages = "1554--1566",
journal = "Science of the Total Environment",
issn = "0048-9697",
publisher = "Elsevier",

}

How can statistical and artificial intelligence approaches predict piping erosion susceptibility? / Hosseinalizadeh, Mohsen; Kariminejad, Narges; Rahmati, Omid; Keesstra, Saskia; Alinejad, Mohammad; Mohammadian Behbahani, Ali.

In: Science of the Total Environment, Vol. 646, 01.01.2019, p. 1554-1566.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - How can statistical and artificial intelligence approaches predict piping erosion susceptibility?

AU - Hosseinalizadeh, Mohsen

AU - Kariminejad, Narges

AU - Rahmati, Omid

AU - Keesstra, Saskia

AU - Alinejad, Mohammad

AU - Mohammadian Behbahani, Ali

PY - 2019/1/1

Y1 - 2019/1/1

N2 - It is of fundamental importance to model the relationship between geo-environmental factors and piping erosion because of the environmental degradation attributed to soil loss. Methods that identify areas prone to piping erosion at the regional scale are limited. The main objective of this research is to develop a novel modeling approach by using three machine learning algorithms—mixture discriminant analysis (MDA), flexible discriminant analysis (FDA), and support vector machine (SVM) in addition to an unmanned aerial vehicle (UAV) images to map susceptibility to piping erosion in the loess-covered hilly region of Golestan Province, Northeast Iran. In this research, we have used 22 geo-environmental indices/factors and 345 identified pipes as predictors and dependent variables. The piping susceptibility maps were assessed by the area under the ROC curve (AUC). Validation of the results showed that the AUC for the three mentioned algorithms varied from 90.32% to 92.45%. We concluded that the proposed approach could efficiently produce a piping susceptibility map.

AB - It is of fundamental importance to model the relationship between geo-environmental factors and piping erosion because of the environmental degradation attributed to soil loss. Methods that identify areas prone to piping erosion at the regional scale are limited. The main objective of this research is to develop a novel modeling approach by using three machine learning algorithms—mixture discriminant analysis (MDA), flexible discriminant analysis (FDA), and support vector machine (SVM) in addition to an unmanned aerial vehicle (UAV) images to map susceptibility to piping erosion in the loess-covered hilly region of Golestan Province, Northeast Iran. In this research, we have used 22 geo-environmental indices/factors and 345 identified pipes as predictors and dependent variables. The piping susceptibility maps were assessed by the area under the ROC curve (AUC). Validation of the results showed that the AUC for the three mentioned algorithms varied from 90.32% to 92.45%. We concluded that the proposed approach could efficiently produce a piping susceptibility map.

KW - Loess plateau

KW - Machine learning algorithms

KW - Piping collapse

KW - Susceptibility map

KW - Unmanned aerial vehicle (UAV)

U2 - 10.1016/j.scitotenv.2018.07.396

DO - 10.1016/j.scitotenv.2018.07.396

M3 - Article

VL - 646

SP - 1554

EP - 1566

JO - Science of the Total Environment

JF - Science of the Total Environment

SN - 0048-9697

ER -