Gully erosion susceptibility assessment and management of hazard-prone areas in India using different machine learning algorithms

Amiya Gayen, Hamid Reza Pourghasemi*, Sunil Saha, Saskia Keesstra, Shibiao Bai

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

11 Citations (Scopus)

Abstract

Gully erosion is one of the most effective drivers of sediment removal and runoff from highland areas to valley floors and stable channels, where continued off-site effects of water erosion occur. Gully initiation and development is a natural process that greatly impacts natural resources, agricultural activities, and environmental quality as it promotes land and water degradation, ecosystem disruption, and intensification of hazards. In this research, an attempt is made to produce gully erosion susceptibility maps for the management of hazard-prone areas in the Pathro River Basin of India using four well-known machine learning models, namely, multivariate additive regression splines (MARS), flexible discriminant analysis (FDA), random forest (RF), and support vector machine (SVM). To support this effort, observations from 174 gully erosion sites were made using field surveys. Of the 174 observations, 70% were randomly split into a training data set to build susceptibility models and the remaining 30% were used to validate the newly built models. Twelve gully erosion conditioning factors were assessed to find the areas most susceptible to gully erosion. The predisposing factors were slope gradient, altitude, plan curvature, slope aspect, land use, slope length (LS), topographical wetness index (TWI), drainage density, soil type, distance from the river, distance from the lineament, and distance from the road. Finally, the results from the four applied models were validated with the help of ROC (Receiver Operating Characteristics) curves. The AUC value for the RF model was calculated to be 96.2%, whereas for those for the FDA, MARS, and SVM models were 84.2%, 91.4%, and 88.3%, respectively. The AUC results indicated that the random forest model had the highest prediction accuracy, followed by the MARS, SVM, and FDA models. However, it could be concluded that all the machine learning models performed well according to their prediction accuracy. The produced GESMs can be very useful for land managers and policy makers as they can be used to initiate remedial measures and erosion hazard mitigation in prioritized areas.

Original languageEnglish
Pages (from-to)124-138
JournalScience of the Total Environment
Volume668
DOIs
Publication statusPublished - 10 Jun 2019

Fingerprint

gully erosion
Learning algorithms
Learning systems
Erosion
Hazards
Discriminant analysis
discriminant analysis
Splines
Support vector machines
Rivers
hazard
hazard area
machine learning
Water
site effect
water erosion
Natural resources
gully
prediction
lineament

Keywords

  • Flexible discriminant analysis
  • Geospatial modelling
  • Gully erosion
  • Multivariate additive regression splines
  • Random forest
  • Support vector machine

Cite this

@article{de3f5525aba747e29732fa9cd33651e5,
title = "Gully erosion susceptibility assessment and management of hazard-prone areas in India using different machine learning algorithms",
abstract = "Gully erosion is one of the most effective drivers of sediment removal and runoff from highland areas to valley floors and stable channels, where continued off-site effects of water erosion occur. Gully initiation and development is a natural process that greatly impacts natural resources, agricultural activities, and environmental quality as it promotes land and water degradation, ecosystem disruption, and intensification of hazards. In this research, an attempt is made to produce gully erosion susceptibility maps for the management of hazard-prone areas in the Pathro River Basin of India using four well-known machine learning models, namely, multivariate additive regression splines (MARS), flexible discriminant analysis (FDA), random forest (RF), and support vector machine (SVM). To support this effort, observations from 174 gully erosion sites were made using field surveys. Of the 174 observations, 70{\%} were randomly split into a training data set to build susceptibility models and the remaining 30{\%} were used to validate the newly built models. Twelve gully erosion conditioning factors were assessed to find the areas most susceptible to gully erosion. The predisposing factors were slope gradient, altitude, plan curvature, slope aspect, land use, slope length (LS), topographical wetness index (TWI), drainage density, soil type, distance from the river, distance from the lineament, and distance from the road. Finally, the results from the four applied models were validated with the help of ROC (Receiver Operating Characteristics) curves. The AUC value for the RF model was calculated to be 96.2{\%}, whereas for those for the FDA, MARS, and SVM models were 84.2{\%}, 91.4{\%}, and 88.3{\%}, respectively. The AUC results indicated that the random forest model had the highest prediction accuracy, followed by the MARS, SVM, and FDA models. However, it could be concluded that all the machine learning models performed well according to their prediction accuracy. The produced GESMs can be very useful for land managers and policy makers as they can be used to initiate remedial measures and erosion hazard mitigation in prioritized areas.",
keywords = "Flexible discriminant analysis, Geospatial modelling, Gully erosion, Multivariate additive regression splines, Random forest, Support vector machine",
author = "Amiya Gayen and Pourghasemi, {Hamid Reza} and Sunil Saha and Saskia Keesstra and Shibiao Bai",
year = "2019",
month = "6",
day = "10",
doi = "10.1016/j.scitotenv.2019.02.436",
language = "English",
volume = "668",
pages = "124--138",
journal = "Science of the Total Environment",
issn = "0048-9697",
publisher = "Elsevier",

}

Gully erosion susceptibility assessment and management of hazard-prone areas in India using different machine learning algorithms. / Gayen, Amiya; Pourghasemi, Hamid Reza; Saha, Sunil; Keesstra, Saskia; Bai, Shibiao.

In: Science of the Total Environment, Vol. 668, 10.06.2019, p. 124-138.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Gully erosion susceptibility assessment and management of hazard-prone areas in India using different machine learning algorithms

AU - Gayen, Amiya

AU - Pourghasemi, Hamid Reza

AU - Saha, Sunil

AU - Keesstra, Saskia

AU - Bai, Shibiao

PY - 2019/6/10

Y1 - 2019/6/10

N2 - Gully erosion is one of the most effective drivers of sediment removal and runoff from highland areas to valley floors and stable channels, where continued off-site effects of water erosion occur. Gully initiation and development is a natural process that greatly impacts natural resources, agricultural activities, and environmental quality as it promotes land and water degradation, ecosystem disruption, and intensification of hazards. In this research, an attempt is made to produce gully erosion susceptibility maps for the management of hazard-prone areas in the Pathro River Basin of India using four well-known machine learning models, namely, multivariate additive regression splines (MARS), flexible discriminant analysis (FDA), random forest (RF), and support vector machine (SVM). To support this effort, observations from 174 gully erosion sites were made using field surveys. Of the 174 observations, 70% were randomly split into a training data set to build susceptibility models and the remaining 30% were used to validate the newly built models. Twelve gully erosion conditioning factors were assessed to find the areas most susceptible to gully erosion. The predisposing factors were slope gradient, altitude, plan curvature, slope aspect, land use, slope length (LS), topographical wetness index (TWI), drainage density, soil type, distance from the river, distance from the lineament, and distance from the road. Finally, the results from the four applied models were validated with the help of ROC (Receiver Operating Characteristics) curves. The AUC value for the RF model was calculated to be 96.2%, whereas for those for the FDA, MARS, and SVM models were 84.2%, 91.4%, and 88.3%, respectively. The AUC results indicated that the random forest model had the highest prediction accuracy, followed by the MARS, SVM, and FDA models. However, it could be concluded that all the machine learning models performed well according to their prediction accuracy. The produced GESMs can be very useful for land managers and policy makers as they can be used to initiate remedial measures and erosion hazard mitigation in prioritized areas.

AB - Gully erosion is one of the most effective drivers of sediment removal and runoff from highland areas to valley floors and stable channels, where continued off-site effects of water erosion occur. Gully initiation and development is a natural process that greatly impacts natural resources, agricultural activities, and environmental quality as it promotes land and water degradation, ecosystem disruption, and intensification of hazards. In this research, an attempt is made to produce gully erosion susceptibility maps for the management of hazard-prone areas in the Pathro River Basin of India using four well-known machine learning models, namely, multivariate additive regression splines (MARS), flexible discriminant analysis (FDA), random forest (RF), and support vector machine (SVM). To support this effort, observations from 174 gully erosion sites were made using field surveys. Of the 174 observations, 70% were randomly split into a training data set to build susceptibility models and the remaining 30% were used to validate the newly built models. Twelve gully erosion conditioning factors were assessed to find the areas most susceptible to gully erosion. The predisposing factors were slope gradient, altitude, plan curvature, slope aspect, land use, slope length (LS), topographical wetness index (TWI), drainage density, soil type, distance from the river, distance from the lineament, and distance from the road. Finally, the results from the four applied models were validated with the help of ROC (Receiver Operating Characteristics) curves. The AUC value for the RF model was calculated to be 96.2%, whereas for those for the FDA, MARS, and SVM models were 84.2%, 91.4%, and 88.3%, respectively. The AUC results indicated that the random forest model had the highest prediction accuracy, followed by the MARS, SVM, and FDA models. However, it could be concluded that all the machine learning models performed well according to their prediction accuracy. The produced GESMs can be very useful for land managers and policy makers as they can be used to initiate remedial measures and erosion hazard mitigation in prioritized areas.

KW - Flexible discriminant analysis

KW - Geospatial modelling

KW - Gully erosion

KW - Multivariate additive regression splines

KW - Random forest

KW - Support vector machine

U2 - 10.1016/j.scitotenv.2019.02.436

DO - 10.1016/j.scitotenv.2019.02.436

M3 - Article

VL - 668

SP - 124

EP - 138

JO - Science of the Total Environment

JF - Science of the Total Environment

SN - 0048-9697

ER -