Machine Learning-Based Gully Erosion Susceptibility Mapping: A Case Study of Eastern India.

Sunil Saha,Jagabandhu Roy,Alireza Arabameri,Dieu Tien Bui,Thomas Blaschke

doi:10.3390/s20051313

Abstract

Gully erosion is a form of natural disaster and one of the land loss mechanisms causing severe problems worldwide. This study aims to delineate the areas with the most severe gully erosion susceptibility (GES) using the machine learning techniques Random Forest (RF), Gradient Boosted Regression Tree (GBRT), Naïve Bayes Tree (NBT), and Tree Ensemble (TE). The gully inventory map (GIM) consists of 120 gullies. Of the 120 gullies, 84 gullies (70%) were used for training and 36 gullies (30%) were used to validate the models. Fourteen gully conditioning factors (GCFs) were used for GES modeling and the relationships between the GCFs and gully erosion was assessed using the weight-of-evidence (WofE) model. The GES maps were prepared using RF, GBRT, NBT, and TE and were validated using area under the receiver operating characteristic (AUROC) curve, the seed cell area index (SCAI) and five statistical measures including precision (PPV), false discovery rate (FDR), accuracy, mean absolute error (MAE), and root mean squared error (RMSE). Nearly 7% of the basin has high to very high susceptibility for gully erosion. Validation results proved the excellent ability of these models to predict the GES. Of the analyzed models, the RF (AUROC = 0.96, PPV = 1.00, FDR = 0.00, accuracy = 0.87, MAE = 0.11, RMSE = 0.19 for validation dataset) is accurate enough for modeling and better suited for GES modeling than the other models. Therefore, the RF model can be used to model the GES areas not only in this river basin but also in other areas with the same geo-environmental conditions.

Highlights

One of the major problems in modern societies in the last decade is the degradation of natural resources, especially soil and water [1]
The results of the multi-collinearity show that the tolerance and variance inflation factor (VIF) values of gully conditioning factors are less than 0.1 and 4.5, indicating no multi-collinearity problems among the gully conditioning factors, which means that they can be used for predicting the gully erosion
Based on the results of Gradient Boosting Regression Tree (GBRT) (Figure 7b), the research area has 315.85 km2 (71.31%) that falls into the low susceptibility class, followed by 78.47 km2 (17.71%) in the medium susceptibility class, 35.60 km2 (8.04%) in the high susceptibility class and 13.03 km2 (2.94%) in the very high gully erosion susceptibility (GES) classes (Table 4) In case of the TN model (Figure 7c), the results show that the study area has 338.66 km2 (76.46%)

Summary

Introduction

One of the major problems in modern societies in the last decade is the degradation of natural resources, especially soil and water [1]. The rapid population growth and careless use of natural resources lead to soil and water degradation, which in turn threatens human lives and property [2,3]. Soil erosion by water, such as in the form of gully erosion, is one of the most common soil degradation processes worldwide [4,5]. Geo-environmental factors such as precipitation, altitude, slope, aspect, curvature of the plane, lithology [11], soil physio-chemical properties [12], and land use/land cover (LULC) [13] have a strong influence on gully erosion

Objectives

Methods

Results

Discussion

Conclusion