Implementation of Artificial Intelligence Based Ensemble Models for Gully Erosion Susceptibility Assessment

The Rarh Bengal region in West Bengal, particularly the eastern fringe area of the Chotanagpur plateau, is highly prone to water-induced gully erosion. In this study, we analyzed the spatial patterns of a potential gully erosion in the Gandheswari watershed. This area is highly affected by monsoon rainfall and ongoing land-use changes. This combination causes intensive gully erosion and land degradation. Therefore, we developed gully erosion susceptibility maps (GESMs) using the machine learning (ML) algorithms boosted regression tree (BRT), Bayesian additive regression tree (BART), support vector regression (SVR), and the ensemble of the SVR-Bee algorithm. The gully erosion inventory maps are based on a total of 178 gully head-cutting points, taken as the dependent factor, and gully erosion conditioning factors, which serve as the independent factors. We validated the ML model results using the area under the curve (AUC), accuracy (ACC), true skill statistic (TSS), and Kappa coefficient index. The AUC result of the BRT, BART, SVR, and SVR-Bee models are 0.895, 0.902, 0.927, and 0.960, respectively, which show very good GESM accuracies. The ensemble model provides more accurate prediction results than any single ML model used in this study.

Butterfly diversity in heterogeneous habitat of Bankura, West Bengal, India

Butterfly diversity was observed in different habitats of Bankura District, West Bengal, India. This district is located at the junction of Chotanagpur plateau and Gangetic plain; it contains a variety of transitional habitats. We found 117 butterfly species from our covered survey area. The highest species recorded in the present study belonged to family Lycaenidae (30.76%) and Nymphalidae (29.91%) followed by Hesperiidae (16.23%), Pieridae (13.67%), Papilionidae (8.54%), and Riodinidae (0.85%), respectively. Based on sighting we found that 12.82% of all the butterflies recorded were abundant in nature while 21.36% were very common, 41.88% were frequent, and 23.93% were rare. Cluster analysis and other diversity indices gives us an overall idea about environmental health. The pattern of diversity change from plain to plateau gradient gives important insight about ecological edge effect. High species number in relation with low individual numbers were found in forest habitat. This preliminary study showed that heterogeneous habitats could harbour many butterflies and need proper conservation efforts to sustain it.

Responses of chemical weathering and gully erosion causing land degradation in the extended part of Chotanagpur plateau in India

<p>Whether the hot and humid subtropical plateau region could leads to land degradation in the form of weathering and gully erosion. In this study, chemical weathering, gully erosion and cohesiveness are investigated together to bring out a new comprehensive idea with a view to understand their controlling factors. This study aimed to address potential land degradation in the extended part of Chotanagpur plateau region. The layers of controlling factors of gully erosion were developed and prioritized considering advanced decision tree, decision tree and random forest algorithms in the R software and the results of these methods were also validated using receiver operating characteristic (ROC) curves. Degree of chemical weathering and cohesiveness were measured through the chemical, physical and spectroscopic analysis of the randomly collected 412 soil samples. Apart from this, the climatic elements like temperature and rainfall were considered for estimating the chemical weathering. The results of the gully erosion models have superb accuracy, i.e. ROC values were 0.970, 0.960 and 0.955 respectively. Therefore, advanced decision tree model has been integrated with the results of degree of chemical weathering and cohesiveness in GIS platform end eventually the land degradation map has been developed. The land degradation map shown that 15% of the study area is highly affected by land degradation whereas 18% area is moderately affected by land degradation and rest of the 67% area is less affected by land degradation. This study provides essential information to the policy makers in order to taking decision for minimizing and controlling the land degradation. This innovative comprehensive approach is significant to assess degradation of existing land to a large scale.</p><p><strong>Keywords: </strong>Land degradation; weathering; cohesiveness; gully erosion; spectroscopic analysis</p>