Climate Change Impact on Soil Erosion in Sub-tropical Environment: Application of Empirical and Semi-empirical Models (Geography of the Physical Environment) 3031157206, 9783031157202

Table of contents :
Preface
Acknowledgements
Contents
1 Introduction to Soil Erosion Study
Abstract
1.1 Basic Ideas
1.2 Important Concept
1.3 Threats of Soil Erosion to Agriculture and Environment
1.4 Classification of Erosion Based on Forms
1.4.1 Splash Erosion
1.4.2 Sheet Erosion
1.4.3 Channel Erosion
1.4.4 Rill Erosion
1.4.5 Gully Erosion
1.4.6 Ravine Erosion
1.5 Importance of Soil Erosion Risk Assessment
1.6 Impact of Climate Change on Soil Erosion Risk
1.7 Application of Remote Sensing and GIS in Soil Erosion and Risk Assessment
References
2 Geophysical Settings of South Bengal
Abstract
2.1 Study Area
2.2 Physical Setup
2.2.1 Geology
2.2.2 Topography
2.2.3 Geomorphology
2.2.4 Drainage System
2.2.5 Soil
2.2.6 Climate
2.2.7 Vegetation
References
3 Morphotectonics Characteristics and Its Control on Soil Erosion
Abstract
3.1 Introduction
3.2 Materials and Methods
3.2.1 Selection of the Causal Parameters
3.2.2 Preparation of the Causal Parameters
3.2.3 Erosion Potentiality Assessment
3.2.3.1 Application of “Evidential Belief Function (EBF)” for “Erosion Potentiality”
3.2.3.2 Application of Logistic Regression for Erosion Potentiality
3.2.3.3 Application of Ensemble “Evidential Belief Function-Logistic Regression” for “Erosion Potentiality”
3.2.4 Validation of the Models
3.2.5 Multi-collinearity Assessment
3.3 Results
3.3.1 Multi-collinearity Assessment
3.3.2 Erosion Potentiality Assessment
3.3.3 Validation of the Models
3.4 Discussion
3.5 Conclusion
References
4 Estimation of Surface Runoff
Abstract
4.1 Introduction
4.2 Methodology
4.3 Result
4.3.1 Runoff Estimation
4.4 Validation
4.5 Discussion
4.6 Conclusion
References
5 Soil Loss Estimation Using Different Empirical and Semi-empirical Models
Abstract
5.1 Introduction
5.2 Database
5.3 Methodology
5.3.1 Factor for USLE, RUSLE and MUSLE for Estimation of Soil Loss
5.3.1.1 Rainfall and Runoff Erosivity Factor
5.3.1.2 Soil Erodibility Factor
5.3.1.3 Slope Length
5.3.1.4 Steepness Factor
5.3.1.5 Slope Length and Steepness Factor
5.3.1.6 Cover and Management Factor
5.3.1.7 Support Practice Factor
5.4 Application of USLE
5.5 Application of RUSLE
5.6 Application of MUSLE
5.7 Results and Discussion
5.7.1 Average Annual Soil Loss Using USLE
5.7.2 Average Annual Soil Loss Using RUSLE
5.7.3 Average Annual Soil Loss and Its Associated Sediment Yield Using MUSLE
5.8 Sensitivity Analysis
5.9 Discussion
5.10 Conclusion
References
6 Potential Sediment Yield Estimation Using Machine Learning, Artificial Intelligence Techniques and GIS
Abstract
6.1 Introduction
6.2 Materials and Methods
6.3 Selection of the Causative Factors
6.4 Preparation of Causal Parameters
6.4.1 Rainfall and Runoff Erosivity
6.4.2 Drainage Density
6.4.3 Land Use and Land Cover (LULC)
6.4.4 Slope
6.4.5 Soil Texture
6.4.6 Elevation
6.4.7 Geology
6.4.8 Stream Power Index
6.4.9 Topographical Wetness Index
6.4.10 Soil Erodibility Factor
6.5 Sediment Yield Estimation
6.5.1 Fuzzy Logic
6.5.2 Analytical Neural Network
6.6 Results
6.6.1 Application of Fuzzy Logic
6.6.2 Application of Analytical Neural Network
6.7 Sensitivity Analysis
6.8 Discussion
6.9 Conclusion
References
7 Impact of Climate and LULC Change on Soil Erosion
Abstract
7.1 Introduction
7.2 Materials and Methods
7.2.1 Soil Erosion Factors
7.2.2 Selection of the Suitable GCM Model for Simulating the Future Rainfall
7.2.3 Statistical Downscaling
7.2.4 Estimating the Rainfall and Runoff Erosivity Factor in Future Period
7.2.5 LULC Prediction
7.2.6 Estimation of Soil Erosion
7.3 Results and Discussion
7.3.1 Soil Erosion in Base Period
7.3.2 Estimation of Future Climate
7.3.3 LULC Prediction
7.3.4 Soil Erosion in Projected Period
7.4 Discussion
7.5 Conclusion
References
8 Sociopolitical Policy Implication
Abstract
8.1 Introduction
8.2 Strategic Choices
8.2.1 Environmental Measures to Controlling Erosion
8.2.1.1 Afforestation with Indigenous Plant Species
8.2.2 Potential Economic and Institutional Strategies for Soil Erosion Management
8.2.2.1 Reclamation of Land
8.2.2.2 Management of Irrigation
8.2.2.3 Managing Agricultural Intensification
8.2.2.4 Alternative Agricultural Practices
8.2.2.5 Minimizing the Overgrazing
8.3 Conclusion
References

Citation preview

Geography of the Physical Environment

Subodh Chandra Pal Rabin Chakrabortty

Climate Change Impact on Soil Erosion in Sub-tropical Environment Application of Empirical and Semi-empirical Models

Geography of the Physical Environment

The Geography of the Physical Environment book series provides a platform for scientific contributions in the field of Physical Geography and its subdisciplines. It publishes a broad portfolio of scientific books covering case studies, theoretical and applied approaches as well as novel developments and techniques in the field. The scope is not limited to a certain spatial scale and can cover local and regional to continental and global facets. Books with strong regional focus should be well illustrated including significant maps and meaningful figures to be potentially used as field guides and standard references for the respective area. The series appeals to scientists and students in the field of geography as well as regional scientists, landscape planners, policy makers, and everyone interested in wide-ranging aspects of modern Physical Geography. Peer-reviewed research monographs, edited volumes, advance and undergraduate level textbooks, and conference proceedings covering the major topics in Physical Geography are included in the series. Submissions to the Book Series are also invited on the theme ‘The Physical Geography of…’, with a relevant subtitle of the author’s/editor’s choice. Please contact the Publisher for further information and to receive a Book Proposal Form.

Subodh Chandra Pal Rabin Chakrabortty

•

Climate Change Impact on Soil Erosion in Sub-tropical Environment Application of Empirical and Semi-empirical Models

123

Subodh Chandra Pal Department of Geography The University of Burdwan Bardhaman, West Bengal, India

Rabin Chakrabortty Department of Geography The University of Burdwan Bardhaman, West Bengal, India

ISSN 2366-8865 ISSN 2366-8873 (electronic) Geography of the Physical Environment ISBN 978-3-031-15720-2 ISBN 978-3-031-15721-9 (eBook) https://doi.org/10.1007/978-3-031-15721-9 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover image by Sonja Weber, München This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

Climate change is now a major concern related to the vulnerability of living beings all over the world, and as a result, it has gained attention in the field of scientific research. Water-induced soil erosion due to extreme rainfall during the monsoon period is one of the major problems that have a global impact. The monsoon-dominated Bengal basin is facing the storm rainfall event in wet monsoon period which is capable to large scale erosion. Changes in intense precipitation have the major direct impact on soil erosion under climate change. A warmer atmosphere’s increased ability to store moisture is expected to lead to an increase in extreme precipitation, which will result in a more active hydrological cycle. Bengal basin is a fertile agricultural belt that produces a significant amount of production and contributes to the country’s GDP. However, the rate of agricultural production has decreased rapidly in recent times due to climate change and soil erosion. This book is divided into eight chapters. Each chapter begins with an abstract that briefly describes the chapter's theme. Chapter 1 provides an overview of the study, including basic ideas and definitions of soil erosion and important terms related to soil erosion study, important concept, statement of research problem, its regional and global issues, types of erosion based on forms, importance of study, and impact assessment with reference to climate change, application of modern tools and techniques in soil erosion, and risk assessment. Chapter 2 depicts the investigation of the area and its distinct geophysical settings. Chapter 3 describes morphotectonics and its impact on soil erosion. Chapter 4 estimates surface runoff in the Bengal basin. In situ ground observations of various soil and land use land cover types are investigated. Chapter 5 discusses soil loss using various empirical and semi-empirical models. In situ field observation, measurement and model validation are also included. Chapter 6 explores potential sediment yield estimation using machine learning, artificial intelligence techniques and GIS. Field observations and measurements of sediment yield using modern methods were presented. Chapter 7 illustrates the impact of climate and land use and land cover change on soil erosion. Chapter 8 discusses the policy implication for restoring this unique landscape. Bardhaman, India

Subodh Chandra Pal Rabin Chakrabortty

v

Acknowledgements

The authors would like to thank the India Meteorological Department (IMD), Pune, for providing climate-related datasets used in this study for long-term climate change analysis. We are grateful to the following organizations for providing data, reports and relevant maps: National Bureau of Soil Survey and Land Use Planning (Kolkata), Geological Survey of India (Kolkata), Survey of India (Kolkata), National Atlas and Thematic Mapping Organization (Kolkata) and National Remote Sensing Centre (Hyderabad). We are grateful to officials from West Bengal Meteorology Department, Agri-Irrigation Department and Forest Department for providing relevant data for the study. We are also grateful to the Bengal basin residents for their assistance during the field survey. We are also grateful to the University of Burdwan, Purba Bardhaman, West Bengal, for providing various research facilities such as library access, instruments, software, maps and so on, which are used at various stages of the study. We are also grateful to Doris Bleier (Publishing Editor, Springer Nature), Ambrose Berkumans, Carmen Spelbos and Catalina Sava (Springer Nature) for their constant support in completing the work on time. We would also like to thank the anonymous reviewers for their time and comments. Finally, we want to thank our family members from the bottom of our hearts for their endless support and encouragement. Subodh Chandra Pal Rabin Chakrabortty

vii

Contents

1 Introduction to Soil Erosion Study . . . . . . . . . . . . . . . . . . . . . . . 1.1 Basic Ideas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Important Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Threats of Soil Erosion to Agriculture and Environment . . . 1.4 Classification of Erosion Based on Forms . . . . . . . . . . . . . . 1.4.1 Splash Erosion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.2 Sheet Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.3 Channel Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.4 Rill Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.5 Gully Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.6 Ravine Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Importance of Soil Erosion Risk Assessment . . . . . . . . . . . 1.6 Impact of Climate Change on Soil Erosion Risk. . . . . . . . . 1.7 Application of Remote Sensing and GIS in Soil Erosion and Risk Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 3 3 4 4 5 5 5 5 7 7 10

2 Geophysical Settings of South Bengal . . . 2.1 Study Area . . . . . . . . . . . . . . . . . . . . 2.2 Physical Setup . . . . . . . . . . . . . . . . . 2.2.1 Geology . . . . . . . . . . . . . . . . . 2.2.2 Topography . . . . . . . . . . . . . . 2.2.3 Geomorphology . . . . . . . . . . . 2.2.4 Drainage System . . . . . . . . . . 2.2.5 Soil . . . . . . . . . . . . . . . . . . . . 2.2.6 Climate . . . . . . . . . . . . . . . . . 2.2.7 Vegetation . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . .

12 13

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

15 15 17 17 19 19 21 23 25 25 31

3 Morphotectonics Characteristics and Its Control on Soil Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1 Selection of the Causal Parameters . . . . . . . . . 3.2.2 Preparation of the Causal Parameters . . . . . . . 3.2.3 Erosion Potentiality Assessment . . . . . . . . . . . 3.2.4 Validation of the Models . . . . . . . . . . . . . . . . . 3.2.5 Multi-collinearity Assessment . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

33 33 35 35 35 39 41 42

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

ix

x

Contents

3.3

Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.1 Multi-collinearity Assessment . . . . 3.3.2 Erosion Potentiality Assessment . . 3.3.3 Validation of the Models . . . . . . . . 3.4 Discussion. . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4 Estimation of Surface Runoff . . 4.1 Introduction . . . . . . . . . . . 4.2 Methodology . . . . . . . . . . . 4.3 Result . . . . . . . . . . . . . . . . 4.3.1 Runoff Estimation . 4.4 Validation . . . . . . . . . . . . . 4.5 Discussion. . . . . . . . . . . . . 4.6 Conclusion . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

42 42 42 42 45 46 46

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

51 51 53 54 54 60 61 63 63

. . . .

. . . .

. . . .

. . . .

67 67 69 69

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

69 77 77 79 79 79 79

. . . . .

. . . . .

. . . . .

. . . . .

80 80 82 84 85

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

89 89 91 91 91 91 92 93 93 95

5 Soil Loss Estimation Using Different Empirical and Semi-empirical Models . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.1 Factor for USLE, RUSLE and MUSLE for Estimation of Soil Loss . . . . . . . . . . . . . . . . . . 5.4 Application of USLE . . . . . . . . . . . . . . . . . . . . . . . . . 5.5 Application of RUSLE . . . . . . . . . . . . . . . . . . . . . . . . 5.6 Application of MUSLE . . . . . . . . . . . . . . . . . . . . . . . . 5.7 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 5.7.1 Average Annual Soil Loss Using USLE . . . . . 5.7.2 Average Annual Soil Loss Using RUSLE . . . . 5.7.3 Average Annual Soil Loss and Its Associated Sediment Yield Using MUSLE . . . . . . . . . . . . 5.8 Sensitivity Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . 5.9 Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.10 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Potential Sediment Yield Estimation Using Machine Learning, Artificial Intelligence Techniques and GIS . . . . 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Selection of the Causative Factors . . . . . . . . . . . . . . . 6.4 Preparation of Causal Parameters . . . . . . . . . . . . . . . . 6.4.1 Rainfall and Runoff Erosivity . . . . . . . . . . . . . 6.4.2 Drainage Density . . . . . . . . . . . . . . . . . . . . . . . 6.4.3 Land Use and Land Cover (LULC) . . . . . . . . . 6.4.4 Slope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.5 Soil Texture. . . . . . . . . . . . . . . . . . . . . . . . . . .

Contents

xi

6.4.6 Elevation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.7 Geology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.8 Stream Power Index . . . . . . . . . . . . . . . . . . . . 6.4.9 Topographical Wetness Index . . . . . . . . . . . . . 6.4.10 Soil Erodibility Factor . . . . . . . . . . . . . . . . . . . 6.5 Sediment Yield Estimation . . . . . . . . . . . . . . . . . . . . . 6.5.1 Fuzzy Logic . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5.2 Analytical Neural Network . . . . . . . . . . . . . . . 6.6 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6.1 Application of Fuzzy Logic . . . . . . . . . . . . . . . 6.6.2 Application of Analytical Neural Network . . . 6.7 Sensitivity Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . 6.8 Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

96 96 97 97 97 98 98 98 99 99 101 101 103 105 106

7 Impact of Climate and LULC Change on Soil Erosion . . 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1 Soil Erosion Factors . . . . . . . . . . . . . . . . . . . . 7.2.2 Selection of the Suitable GCM Model for Simulating the Future Rainfall . . . . . . . . . . 7.2.3 Statistical Downscaling . . . . . . . . . . . . . . . . . . 7.2.4 Estimating the Rainfall and Runoff Erosivity Factor in Future Period . . . . . . . . . . . . . . . . . . 7.2.5 LULC Prediction . . . . . . . . . . . . . . . . . . . . . . . 7.2.6 Estimation of Soil Erosion. . . . . . . . . . . . . . . . 7.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 7.3.1 Soil Erosion in Base Period. . . . . . . . . . . . . . . 7.3.2 Estimation of Future Climate. . . . . . . . . . . . . . 7.3.3 LULC Prediction . . . . . . . . . . . . . . . . . . . . . . . 7.3.4 Soil Erosion in Projected Period . . . . . . . . . . . 7.4 Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

109 109 111 111

.... ....

114 115

. . . . . . . . . . .

. . . . . . . . . . .

115 116 116 117 117 118 118 119 120 122 123

8 Sociopolitical Policy Implication . . . . . . . . . . . . . . . . . . . . . . . . . 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 Strategic Choices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.1 Environmental Measures to Controlling Erosion . . . . 8.2.2 Potential Economic and Institutional Strategies for Soil Erosion Management . . . . . . . . . . . . . . . . . 8.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

127 127 128 128

. . . . . . . . . . .

. . . . . . . . . . .

130 131 133

1

Introduction to Soil Erosion Study

Abstract

Keywords

“Soil erosion” is a global problem to natural resources and is particularly effective in reducing crop yield because of a loss in soil fertility, while multifunctional reserve storage is depleted due to continuous sedimentation. Accelerated erosion of the soil has detrimental economic and environmental effects. The “soil erosion” risk assessment can be useful in the area where “soil erosion” is the biggest challenge to sustainable agriculture, as the soil is the base of agricultural development. “Soil erosion” models can take into account all of the complex interrelationships that impact erosion rates by modelling hydrologic erosion processes. In order to predict “soil erosion”, most of these models include climate, topography, soil condition, land use and land cover. The generation of input data, too spatial and traditional approaches have proven too expensive and time consuming to generate these input data are the most difficult problems for evaluating these models. The advancement of remote sensing technology has improved the accessibility and cost-effectiveness of spatial information regarding the input parameters. The spatial data processing capabilities of “geographic information systems (GIS)” have contributed to the advancement of robust approaches to “soil erosion modelling”.

Soil erosion Natural resources Climate change GIS Soil erosion modelling

.

1.1

.

.

.

Basic Ideas

In comparison to Asia and Africa, North America has less soil deterioration (Lal 1998). Since around 1800, “soil erosion” has been ascribed to land settlement and vegetation clearing (Miller et al. 1985). This generally starts with soil degradation, which obstructs water and air flow, and progresses to desertification. According to studies, soil deterioration affects around a third of the world’s land area (Lal 1994). “Soil erosion” is a two-step process in which soil particles are separated from the soil bulk and moved by erosive agents such as water and wind. A third phase, deposition, occurs when there is insufficient energy to transport the particles (Morgan 2009). The most essential detaching ingredient is rain splash. Soil particles can be blown several cm through the air as an outcome of rainfall impacting a bare soil surface. Continued exposure to heavy rainfall causes the soil to deteriorate significantly. Weathering activities, both physical and biological, break up the soil via alternating wetting and drying, freeze–thaw and frost activity

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 S. C. Pal and R. Chakrabortty, Climate Change Impact on Soil Erosion in Sub-tropical Environment, Geography of the Physical Environment, https://doi.org/10.1007/978-3-031-15721-9_1

1

2

(Trudgill et al. 2022). Tillage activities, as well as the trampling of people and cattle, disrupt the soil. Separation of soil particles is also aided by running water and the wind. All of these activities loosen the soil so that transport agents can readily remove it. Those that operate are ally and contribute to the removal of a rather uniform layer of soil, as well as those that limit their activity in channels make up the carrying agents. Rain splash, surface runoff in the form of infinite-width shallow flows, often dubbed sheet flow but more accurately named overland flow, and wind make up the first category. Water in smaller channels called as rills, that can be erased by weathering and ploughing, or in larger, greater permanent structures such as gullies and rivers, is covered by the second category (Blanco-Canqui and Lal 2010). When it comes to water erosion, the distinction between “rill erosion” and erosion on the ground induced by a mix of “raindrop impact” and overland flow is frequently made. Transport by mass movements including such soil flows, slides and creep, wherein water influences the soil internally and changes its strength, must be included to these exterior forces that pick up materials from and transfer it across the surface of the ground (Mitchell and Soga 2005). The intensity of erosion is determined by the amount of material given by detachment over time and the capacity of the eroding agents to transport it (Toy et al. 2002). Detachment restricted erosion occurs when the agents have the ability to move more material than detachment can provide (Foster et al. 1985). The erosion is transport constrained when more material is given than can be transferred. Soil is the source for agricultural and forestry activity, the nurturer of the human population and a major component of the human surroundings. The soil should therefore be paid much greater consideration and greater control should be exerted in relation to its fair usage, preservation and enhancement. The rapid growth of the world population and subsequent requirements for the “food production” on the one hand and the increasing technological progress leading to further industrialization and urbanization include erosion, degradation, contamination and

1

Introduction to Soil Erosion Study

environmental pollution from industrial waste and others are challenging issues for soil conservation, on the other. “Soil erosion” is a method of detachment, transportation and deposition into a new deposition zone of the surface soil particles from the original site (Nearing et al. 2017). The processes of diverse human activities damage the soil and thereby cause a substantial change in natural erosion rates (Pimentel 2006). The most severe soil conservation problem has previously been identified as erosion produced by water that restricts the production of soil by depleting the fertile topsoil (Blanco and Lal 2008). The soil is a non-renewable resource, and sediment is largely responsible for reducing the amount and purity of water, as well as silting up water bodies. In addition to the significant loss of soil degradation and land depletion caused by human activities or disturbance, the extreme and premature siltation of several reservoirs often responds to this problem immediately (Dotterweich 2013). “Soil erosion” is an irreversible process that degrades land and degrades the quality of surface water (Eswaran et al. 2019). Unsustainable resource use and inadequate management are too responsible. Every year, soil deterioration renders 0.3–0.8% of the world’s “arable land” unfit for “agricultural production”, and an additional 200 million ha of cultivated land will be necessary to feed the growing population over the next 30 years (Singh et al. 2011). As a result, this valuable finite resource should be protected from all forms of degradation and deterioration in order to ensure the long-term viability of agricultural output and environmental conservation (Parr et al. 1992). Because of its influence on both the protection of poorly degraded land and its many resources, as well as the development of productivity on such areas, soil and water conservation for watersheds is presently a top priority in India (Narayana and Babu 1983). Furthermore, due to increased “soil erosion” in their catchments, reservoirs’ usable life and volume are now being lost quicker than expected. According to Singh et al. (1992), the “erosion potential maps” of the both periods were then divided into four erosion

1.3 Threats of Soil Erosion to Agriculture and Environment

prospective classifications: low, medium, high and extremely high erosion potential. As a result, “soil erosion” risk maps were compared at the microwatershed level (a basic management level for adopting soil conservation policies) to give clear insights to watershed managers and policymakers on changes in the geographic extent and severity of “soil erosion”.

1.2

Important Concept

The word erosion comes from the “Latin verb erodere—to carry away (rodere—to rub)”, to excavate. The word erosion was used first in geology to describe holes made of water and solids flushed by river water (Penck 1894), whereas erosion by rain and surface lava is called “ablation (Latin ablatio—to carry away)”. By the end of the nineteenth century, the challenges of river erosion and its relation to modelling the surface of the earth became well known. A few additional words were used in addition to the terms “erosion and ablation” as a basis for “geomorphological processes” induced by water and wind. The terms “Corrasion (latin corradere)”, “Corrosion (latin corrodere—to rip piece)”, “Abrasion (latin abradere—to rip off)”, and “Denudation (latin denudere—to strip)”, among other terms, were created. The term mechanical, side-washing (by rivers) means, in this case, corrosion, which means the chemical destruction of rocks easily soluble, abrasion means the sampling of seawater by wind, and denudation, indicating the process of washing sheets. This is an indication of a process of washing. This term is used to interpret the washing procedure generally downstream. Different scientists have used these terms differently. Many scholars are still using the word “erosion” to include all forms of deterioration of soil or the crust of the Earth by water. On-site impacts are especially critical on agricultural land, as soil distribution within a field, soil loss from a field, soil structure disintegration, and organic compounds and nutrient decrease lead to a reduction of productive “agricultural soil depth and soil fertility”. Erosion

3

also limits accessible soil moisture, making dry conditions more likely. The total outcome is a decrease of productivity, which limits what may be produced and necessitates higher fertilizer expenditures to sustain yields. The loss of soil fertility due to erosion eventually leads to land abandonment, with negative repercussions for food supply and security, as well as a significant drop in land value. Sedimentation downstream causes off-site issues by reducing the capacity of rivers and drainage ditches, increasing the danger of floods, blocking irrigation systems, and shortening the life span of reservoirs. The landowner must bear the on-site costs of erosion, however they may be transferred on to the community in the form of increased food prices if yields decrease or land goes out of operation. Off-site costs, which depend on local governments for road clearing and repair, insurance firms, and all land owners in the local community impacted by sedimentation and floods, are borne largely by the farmer. Although “soil erosion” is a mechanical fact involving wide variations in intensity and frequency around the globe, social, economic, political and institutional variables all have a role for where and when erosion happens. The avoidance of “soil erosion”, that entails limiting the rate of soil loss to that which might occur under natural conditions, is dependent on the selection of appropriate “soil conservation practices”, which, in turn, necessitates a comprehensive perception of erosion processes. Energy, resistance and protection are the three categories of elements that determine the rate of erosion. The energy category comprises the erosioncausing capacity of rainfall, runoff and wind.

1.3

Threats of Soil Erosion to Agriculture and Environment

“Soil erosion” is among the world’s significant environmental issues today, since agriculture and the natural environment have been severely threatened (Pimentel and Burgess 2013). As an example, 15 t of “soil erosion” in a single storm removes just about 1 mm of soil from the earth

4

(Brandt 1988). It takes at least 500 years to shape 2.5 cm of topsoil under tropical and temperate agricultural environments—the best rate estimate of renewals is about 1 t/ha/year (Lal and Moldenhauer 1987). In most of the agricultural regions around the world are experiencing severe “soil erosion”, which is worsening as more marginalized land is converted to cultivation and less crop waste returns to the soil to be preserved and improved. Annual average soil losses in Europe range between 10 and 20 t/ha/year. In the USA, cropland “soil erosion” averages 15–30 t/ha/ year. In Asia, Africa, and South America, cropland annual average “soil erosion” rates range from 20 to 40 t/ha/year. “Soil erosion” negatively impacts plant growth by reducing the supply of water, nutrients and organic material and reducing the depth of the root as the topsoil is diluted (Bakker et al. 2004). The most detrimental consequence of deforestation is the reduction of water needed for plants. Most water is lost due to the rapid fluctuation during water erosion. Furthermore, wind and water erosion decreases the soil’s water retaining potential by selectively extracting organic matter and smaller soil particles. Water can be limited to up to 90% in soils eroded by erosion. Runoffs could lead to water shortages for plants even in years with strong precipitation (Pilgrim et al. 1988). Sufficient food supply depends on fertile soil. More than one thousand million people now are undernourished—in the history of human civilization more than ever (Pimentel et al. 1993). Currently, 7.9 billion people (May, 2021) still live on earth and about 370,000 children are born and must be fed every single day. More than 97% of the world’s food originates from land instead of lakes and other water bodies, with steep terrain in tropical climates as a result (Archer et al. 2010). For all initiatives on food security and climate protection, “soil erosion” controls are therefore important to sustainable agriculture and the ecosystem (Powlson et al. 2011). A significant limiting factor in crop production is soil nitrogen shortages removed from erosion. “A total of 4 kg of nitrogen, 1 kg of phosphorus, 20 kg of potassium and 2 kg of calcium” could be included in one tonne of

1

Introduction to Soil Erosion Study

weighty agricultural soil. In the finer soil particles and organic matters selectively eliminated by erosions, soil nutrients are plentiful. In addition to providing energy and nutrients directly for the production of organic matter, it is also necessary through better penetration, water retention, soil structure and cation exchange capability. In the preservation of soil health and regeneration of minerals, organic matter and soil biota are interdependent. Erosion impacts are more extreme than in temperate environments in the tropics, where “typic” agriculture cannot allow for the optimal substitution of missing soil qualities.

1.4

Classification of Erosion Based on Forms

Rain splash, sheet, channel and stream erosion are all examples of water-induced “soil erosion”. A brief description of each form of erosion is provided below.

1.4.1 Splash Erosion The soil particles start losing and splashing owing to their impact force when rain droplets contact the surface of the ground. At the moment, the pressure gradients develop up to the margins of the drop disintegrates the soil and releases some particles. The dropping rain drops may produce a force approximately 14 times their own weight at an average speed of 75 cm/s. Particles less than 20 µm in diameter are usually separated from the surface due to this splash erosion. Splash erosion was involved for both the disintegration of the soil surface and the movement of material over small distances; wash erosion was accountable for the flushing of loosened soil into “rills”, but the impact of wash erosion was only equivalent to that of rill erosion under severe “sheet flow”. As a result, rill erosion contributed the most to erosion losses in this scenario as well. “Splash erosion” was a major issue in areas with cultivating rills and short strips of depressions, especially on loosened, recently cultivated soil.

1.4 Classification of Erosion Based on Forms

5

1.4.2 Sheet Erosion

1.4.3 Channel Erosion

Water flows along a sheet of more or less uniform depth as it flows over a gentle and smooth sloped soil. Under such conditions, soil is removed in a relatively uniform manner from all parts of the area with a similar degree of slope. Sheet erosion refers to the moderately uniform removal of surface soil caused by rainfall and runoff water. “Sheet erosion” is defined by the somewhat consistent “erosion of soil” over the whole surface of the ground or over a specific portion of a slope (Smith and Wischmeier 1957). As the slope’s surface gets more equal, the chances of water accumulating reduce, and sheet erosion is becoming more consistent (Hussein et al. 2007). Even on the smoothest of slopes, water may accumulate. As a result, separating sheet erosion, which has virtually unnoticeable external symptoms of erosion, from rill erosion, which is a surface phenomenon if the rills do not deepen, is challenging (Zachar 2011; BernatekJakiel and Poesen 2018). The amount of water runoff that accumulates is determined by the height of the “water stream”, the “coarseness of the ground”, as well as other parameters (LópezVicente et al. 2013). The soil crust thins as a result of sheet erosion, and the subsurface rock and mineral sequences are eventually exposed across a vast region. Sheet erosion is the process of removing two types of particles: (1) weathered particles, and (2) readily “dissoluble matter (stuff rendered soluble by weak acids in rainwater)” (Osman 2014). Therefore, sheet erosion is microerosion in the proper sense, i.e. the degrading and washing of soil to generate limited formations such as “raindrop erosion, laminar erosion, rill erosion, and layer erosion” (Goudie 2003). Soil loss by raindrop action—raindrop erosion—is the initial phase of sheet erosion, which is unique in terms of shape (Wu et al. 2018). Soil may be fragmented and degraded in this fashion, but the displaced of soil particles stays minimal (as long as no “surface runoff” occurs), despite the fact that displacement is persistent (Culling 1965). This is a significant erosion factor on ridges, furrows and erosion remains, among other places.

The final kind of “water erosion” is “channel erosion (Latin canalis—channel)”, which works similarly to river erosion in natural beds. Zachar (2011) mentions drain erosion in relation to drainage systems in particular. This sort of erosion is also classified as anthropogenic erosion. Channel erosion takes place where surface water has accumulated, and a significant amount of water provides the energy for both detaching and carrying the soil. There are three types of “channel erosion: rill erosion, gully erosion and ravine erosion”.

1.4.4 Rill Erosion “Rill erosion” occurs when water flowing over the soil surface follows preferred channels, producing a visible channel. These rills are often small erosion structures that Loch and Silburn (1996) characterize as “flow channels which can be eliminated by ploughing”. “Rill initiation” is influenced by the soil’s “cohesive strength” and the shear forces acting on it (Merritt et al. 2003). While rill flow could remove substantial amounts of dirt if the shear tension in the rill is high enough, rill flow works as a transportation mechanism for the removal of material downslope between “rill and interill sources” (Kinnell 2005). Rill erosion is the loss of soil by runoff water which may be fully eased by regular erosion, with forming shallow channel (Elliot 1988). Rill occurs because of the concentration of runoff, when the runoff water begins to flow via little finger channels down the hillsides. The silt can be removed from a hillside by rills, even if this relies on the distance and the extent of the region concerned. This can lead to loss of soil fertility and production if erosion occurs more quickly than the creation of soil (Fig. 1.1).

1.4.5 Gully Erosion Gullies of various sizes and shapes emerge as a result of the gathering of increasing amounts of

6

Fig. 1.1 Rill erosion in the western part of the study region

1

Introduction to Soil Erosion Study

1.5 Importance of Soil Erosion Risk Assessment

water or the progressive deepening of rills (Zachar 2011). Because it has the broadest definition, the term gully is selected. When it comes to gully erosion, there are a few different types to look for. Any erosion of “gully” with a depth of 30 cm to 2–3 m falls into the first category. The eroding curve is adjusted by waterfall erosion in this manner, with a “pronounced backward or retrograde (Latin retrogradare 5—to travel back)” erosion and “vertical (Latin verticalis—perpendicular)” or depth erosion. Gullies are greater in size and have a more intricate growth (Harvey 1992). Aside from retrograde and vertical erosion, lateral erosion, as well as auxiliary landslides, soil movement and other occurrences, may be found here. Gullies can develop into gorges and canyons, which are frequently part of the hydrogeological system and patterned by fluvial erosion (Stokes et al. 2008). Gullies are relatively constant steep watersides, with ephemeral flows during rainstorms. Gullies are defined by a headcut and several steps or points along the way contrasted to constant banks of the river that have a relatively flat concave upward contour. Gullies are often comparatively more deep and wider than stable channels, have higher sediment loads and show very irregular behaviour such that the connection between sediment discharge and runoff are also weak (Fig. 1.2). Gullies are more than 1 m2 (929 cm2) in cross sectional area (Poesen 1993). Gullies have nearly mostly been linked to increased erosion, hence landscape instability.

1.4.6 Ravine Erosion A ravine is a shape that is narrower than a canyon and often causes erosion on streams. Ravines are typically categorized as larger than gullies, but much less than valleys (Fig. 1.3). Surface runoff has a role in the formation of ravines, but the primary cause is the saturation of the soil’s external surface with water, which reduces internal friction and leads to solifluction, which leads to the formation of gullies (Hobbs et al. 2017). Despite the fact that ravine development,

7

also known as “soil erosion”, is not classed as a natural disaster, it is a plague-like sickness that engulfs precious agricultural land every year. However, when taken as a whole, it has a comparable impact to disasters in terms of undermining a region’s socioeconomic framework (Pani and Mohapatra 2011). As a hazard to livelihood, ravine development should be treated similarly to a disaster in order to minimize harmful effect. A total of 1.965 million ha of land are predicted to be degraded globally. 1.094 million ha are prone to water-induced “soil erosion”, whereas 549 million ha are impacted by salinity, sodicity or even both. In recent decades, the subject of ravines and the resulting loss of agricultural land has been examined with attention in tropical and mid-latitude nations. There are no historical accounts in India that can be used to trace the origins of ravine development. Research on the Indian subcontinent suggest to a crucial magnitude of ravine erosion along the Yamuna and its major tributaries, as well as the Himalayan, Siwalik, Hazaribagh and Chotanagpur Plateaus (Sharma 1979). According to projections from the “Planning Commission” (1965), India’s ravines harm around 3 million ha of agricultural land, with 0.5 million ha near the Chambal (Verma and Singh 2017).

1.5

Importance of Soil Erosion Risk Assessment

Evaluating the risk of “soil erosion” can be useful in the assessment of soil in the area where the main challenge to sustainable agriculture is “soil erosion”. The estimation of “soil erosion” and mapping of erosion-prone regions was used to conserve soil and to protect watersheds (Chakrabortty et al. 2020b). There is also a need to measure the erosion rates as they are crucial for policymakers. Therefore, surveillance, particularly those concerning agricultural practices, is necessary to control the degradation of natural resources. The soil degradation risk of the soil management programme is therefore critical. Different methods are available to evaluate “soil erosion” rates (Pal and Shit 2017). Conventional

8

1

Introduction to Soil Erosion Study

Fig. 1.2 Gully erosion in the western part of this region that has been collected in the time of filed visit

1.5 Importance of Soil Erosion Risk Assessment

9

Fig. 1.2 (continued)

experiments are costly and time consuming methods to assess “soil erosion” and degradation (Pal and Chakrabortty 2019a). In addition, due to the complexities of relationships and the difficulties of generalizing outcomes, these methods are limited. Precision is probably low, and where the soil came from and when is not known. The outcome of the drainage basin discharge and sediment is determined by calculating suspended sediment in order to provide empirical proof of drainage basin soil loss. The eroded soil accumulated elsewhere may not be calculated likewise, without touching the measuring station. It is therefore debatable to use these experimental results to measure the soils depletion from a larger region with a variety of situational variations. About the volume of erosion by calculating the adjustment of surface

level, terrestrial surveying using reconnaissance techniques for soil losses estimates considers. It does not, however, affect arable land because agriculture and occupation are affecting the surface level. The geomorphological parameters reflect virtually all watershed-based causal variables that impact the rate of “runoff and sediment loss” directly or indirectly. The surface characteristics are the essential analytical units prior to the adoption of any advanced instruments to monitor the reactions of hydrological processes in conjunction with these. Over the last several decades, the demand for exact information concerning runoff and sediment production and an acceleration in the watershed management programmes to conserve, develop and exploit those resources have risen quickly.

10

1

Introduction to Soil Erosion Study

Fig. 1.3 Ravine erosion

1.6

Impact of Climate Change on Soil Erosion Risk

Water-induced “soil erosion” is an important global issue for the loss of fertile land and its transformation to unproductive, unproductive land in the tropical and subtropical areas. The subtropical region influenced by the monsoon has distinct characteristics regarding seasonal changes in temperature and the availability of precipitation (Yihui and Chan 2005). The monsoon season is linked to a greater level of precipitation, both having a direct and an indirect effect on soil loss (Chakrabortty et al. 2020a). Short-term “soil erosion” and its repercussions cannot be detected appropriately, while the longterm effects of “soil erosion” may be plainly seen (Houben 2008). According to the reports, the erosion of the soil may ruin the whole civilize

and its lives. The subtropical region is very crowded and the bulk of the population depends on farming systems for rain fed, which means that the quantity of production is decreasing every day (Lal 2000). Nevertheless the region experienced serious “soil erosion” and, as a result, agricultural productivity decreases on a daily basis. Many scientists in many disciplines attempt to measure the volume and spatial variations of earth erosion. Every year, “soil erosion” destroys 3975 million ha of land in India. Many researchers have shown that “soil erosion” is the result of the fast growing trend in land degradation in this area (Mishra et al. 2022). A homosexual process and a divide between “soil erosion” and the proliferation of regolith production cannot replace “soil erosion”. The erosion of soil in a given region does not only diminish agricultural production but also causes the deterioration of land and related natural

1.6 Impact of Climate Change on Soil Erosion Risk

resources as many environmental issues to affect the condition of life (Lal 2015). Soil erosion in any location is not only linked to degradation of land but also causes deterioration of the water quality and of the environmental harm involved (Lal 1998). Extreme changes in land use and land coverage, recent population trend increase and conventional agricultural techniques with no assistance and management measures are capable of speeding up “soil erosion” in most tropical and subtropical nations and its related sedimentation in reservoirs. In tropical and subtropical locations, climate change accelerates “soil erosion” due to severe precipitation. Evaluation of regions prone to “soil erosion” in this region should contribute to sustainable land management methods where “soil erosion” is a widespread concern, and this is an urgent problem that inhibits sustainable soil use practices. Quantitative information with the greatest possible precision may be taken as an essential element of a suitable and sustainable universal “soil and water conservation” strategy through validation with relation to soil loss (Beskow et al. 2009). So it is important to estimate regions prone to “soil erosion” in this location since this river basin is distinctive in terms of the flash, mountain and fluctuations. The GIS method and probability statistics for “soil erosion” sensitivity mapping are precision elements in current management systems (Vijith et al. 2012). Using the empirical model, most “soil erosion” studies or multivariate analysis validates the selection of the “soil erosion” causative variables. Regenerating and flush erosivity is one of the most important dissuasive components of “soil erosion”. The effect of storm rainfall on soil erosions susceptible is great in the subtropical monsoon climate (Pal et al. 2021). The extended dry season is linked to short wet season with heavy rainfalls which is particularly susceptible to “soil erosion”. During this season, the rainfall is quite heavy and intensive. Here, historical long-term “GCM” data were examined with data gathered over the same time. A systematic method to statistical downscaling was used (Pal et al. 2021). Here, we distinguish between the records for the monsoon period and

11

observations of the downsized “GCM information”. In this work the different “GCM” models of various “RCP” scenarios were examined for the simulation of future precipitation. The GCM control and scenario bases of https://pcmdi.llnl. gov/projects/esgf-llnl/ have been gathered. A total of 49 GCM models with various RCP scenarios were examined with the information observed (Pal and Chakrabortty 2019b). Different “GCM” models best suited for this investigation are chosen. The set of best “GCM” models can forecast the exact precipitation scenario for the next term. In general, the “GCM” data are linked to errors that cannot estimate the real precipitation situation. The key job to simulate the scenario is therefore to rectify the distortion using acceptable approaches. Here, using the control or historical era, the observed datasets are evaluated to eliminate the bias and distribution in scenarios. The quantile-based technique in the distribution of probability is explored for this aim with the use of a and b as a form and scale function. Researchers and rainfall know the potential influence of climate change on soil loss, and a major element for the subtropical region is the runoff erosion, able to estimate the impact of the rainfall scenario on “soil erosion”. There has been an increasing tendency in the projected precipitation scenario in most regions of the planet. This rising tendency is essential to assess “soil erosion” in the simulated rainfall scenario, considering that many subtropical locations are faced with significant “soil erosion” owing to tempest precipitation. “General circulation models (GCMs)” are useful instruments for gaining a numerical knowledge of the climatic dynamics during the “last glacial maximum (LGM)”. Sustainable development, regional competition, inequality, fossil-fuelled development and middle-of-theroad development are all described in the SSPs (Riahi et al. 2017). Atmospheric “GCMs”, atmospheric “GCMs” connected to basic oceanmixed layer models, and coupled atmosphere– ocean “GCMs” are all examples of “GCMs” that have been employed for this objective. “Shared socioeconomic pathways (SSPs)” are projections of worldwide socioeconomic developments up to

12

1

2100. They highlight possible important worldwide trends that, taken collectively, might provide various problems for climate action in future. They have also been referred to as “future tales”, with the goal of examining how the progress can evolve given a constant set of assumptions. The SSPs are tales that describe various socioeconomic scenarios. These storylines offer a qualitative explanation of the logic that connects the various aspects of the stories. They give statistics to complement the projections on national population, urbanization and “GDP” in form of numerical factors (per capita). The “SSPs” may be used in conjunction with several “integrated assessment models (IAMs)” to investigate future socioeconomic and climatic scenarios. “Population and economic changes will have a significant impact on the predicted mitigating and adaptation problems”. A bigger, weaker population, for instance, will find it much more difficult to adjust to the negative consequences of climate change. Understanding how population and economic growth evolve in the SSPs provides first level of insight into the numerous issues.

1.7

Application of Remote Sensing and GIS in Soil Erosion and Risk Assessment

In light of the fact that many countries are focusing on developing ecological indicators to guarantee that their natural resources stay vital and sustainable, a probability map of “soil erosion” is shown utilizing a multivariate geostatistical approach and a geographic information system (GIS). The conventional approaches showed that generation of this input was highly costly and time consuming. With the introduction of “remote sensing (RS) technologies” it is becoming increasingly viable and cost efficient to draw up spatial information on input parameters. Moreover, “soil erosion” modelling techniques have grown more robust and comprehensive with the strong spatial data handling capabilities and compatibility with the geographic information

Introduction to Soil Erosion Study

system (GIS). RS can enable the analysis of elements that enhance the erosion process, such as “geology, soil type, slope, drainage, land use and land cover (LULC)” etc. A multi-temporal satellite image provides useful information on the seasonal use of the land. For investigating erosion characteristics, such as rainfall interception by vegetation cover, satellite data may be used. A stereoscopic optical and microwave RS analyses a “digital elevation model (DEM)” as one of the essential inputs necessary for “soil erosion” modelling. In real time, RS offers important and reliable data about soil and land. It permits equivalent information over wide regions and may therefore help substantially to the evaluation of regional erosion. The essential information relating to seasonal land use dynamics for mapping “LULC” may be obtained using multitemporal satellite imagery. It may be utilized for the generation of an erosion modelling decking and management factor. The elements related with the categorization of soil, such as “soil type, climate, vegetation, topography and lithology”, might possibly be mapped to account for “regional erodibility” variations, which are used as data input for erosion modelling. The visual delineation of the soil patterns may mainly employ the optical satellite images to map soil. The fundamental requirements of any hydrological or geomorphological trial are “DEMs” that enable us to extract different topographical properties, including elevation, inclination, and the aspect that is crucial for analysing the physical features of the water body. Now, GIS has evolved into a powerful tool for handling georeferenced spatial and non-spatial data, preparing and visualizing input and output, and interacting with “soil erosion” models. “GIS technology” may be used to help the land erosion stock in the modelling of “soil erosion” and erosion risk assessment. It has tremendous promise. The “GIS” may be used to calculate the variations in soil loss estimates generated by different scales of soil mapping utilized as a data layer in the model and to quantify them to regional levels. Integrated RS and GIS use might assist to evaluate a quantitative loss of soil at different scales and to detect potentially “soil erosion” danger

References

regions. Several researches have demonstrated the possible use of GIS methods as a quantitative “soil erosion” hazard assessment utilizing various models of “soil erosion”. In view of the inaccessibility of the hillside, RS is necessary to address spatial variability and information when it is an extended region. Spatial modelling comprises the use of GIS as a model and the performance of basic mathematical calculations of the saved GIS object characteristics for spatially presenting the results.

References Archer DR, Forsythe N, Fowler HJ, Shah SM (2010) Sustainability of water resources management in the Indus Basin under changing climatic and socio economic conditions. Hydrol Earth Syst Sci 14:1669–1680 Bakker MM, Govers G, Rounsevell MD (2004) The crop productivity–erosion relationship: an analysis based on experimental work. CATENA 57:55–76 Bernatek-Jakiel A, Poesen J (2018) Subsurface erosion by soil piping: significance and research needs. Earth Sci Rev 185:1107–1128 Beskow S, Mello CR, Norton LD et al (2009) Soil erosion prediction in the Grande River Basin, Brazil using distributed modeling. CATENA 79:49–59 Blanco H, Lal R (2008) Principles of soil conservation and management. Springer, New York Blanco-Canqui H, Lal R (2010) Soil and water conservation. In: Principles of soil conservation and management. Springer, Berlin, pp 1–19 Brandt J (1988) The transformation of rainfall energy by a tropical rain forest canopy in relation to soil erosion. J Biogeogr 15:41–48 Chakrabortty R, Pal SC, Sahana M et al (2020a) Soil erosion potential hotspot zone identification using machine learning and statistical approaches in eastern India. Nat Hazards 104:1259–1294 Chakrabortty R, Pradhan B, Mondal P, Pal SC (2020b) The use of RUSLE and GCMs to predict potential soil erosion associated with climate change in a monsoondominated region of eastern India. Arab J Geosci 13:1–20 Culling W (1965) Theory of erosion on soil-covered slopes. J Geol 73:230–254 Dotterweich M (2013) The history of human-induced soil erosion: geomorphic legacies, early descriptions and research, and the development of soil conservation—a global synopsis. Geomorphology 201:1–34 Elliot WJ (1988) A process based rill erosion model Eswaran H, Lal R, Reich P (2019) Land degradation: an overview. In: Response to land degradation, pp 20–35

13 Foster G, Young R, Römkens M, Onstad C (1985) Processes of soil erosion by water. In: Soil erosion and crop productivity, pp 137–162 Goudie A (2003) Geomorphological techniques. Routledge Harvey A (1992) Process interactions, temporal scales and the development of hillslope gully systems: Howgill Fells, Northwest England. Geomorphology 5:323–344 Hobbs S, Paull D, Clarke J (2017) Testing the water hypothesis: quantitative morphological analysis of terrestrial and martian mid-latitude gullies. Geomorphology 295:705–721 Houben P (2008) Scale linkage and contingency effects of field-scale and hillslope-scale controls of long-term soil erosion: anthropogeomorphic sediment flux in agricultural loess watersheds of Southern Germany. Geomorphology 101:172–191 Hussein MH, Kariem TH, Othman AK (2007) Predicting soil erodibility in northern Iraq using natural runoff plot data. Soil Tillage Res 94:220–228 Kinnell P (2005) Raindrop-impact-induced erosion processes and prediction: a review. Hydrol Proces Int J 19:2815–2844 Lal R (1994) Soil erosion research methods. CRC Press Lal R (1998) Soil erosion impact on agronomic productivity and environment quality. Crit Rev Plant Sci 17:319–464. https://doi.org/10.1080/07352689891304249 Lal R (2000) Soil management in the developing countries. Soil Sci 165:57–72 Lal R (2015) Restoring soil quality to mitigate soil degradation. Sustainability 7:5875–5895 Lal R, Moldenhauer WC (1987) Effects of soil erosion on crop productivity. Crit Rev Plant Sci 5:303–367 Loch R, Silburn D (1996) Constraints to sustainability— soil erosion. Sustainable crop production in the subtropics: an Australian perspective QDPI López-Vicente M, Poesen J, Navas A, Gaspar L (2013) Predicting runoff and sediment connectivity and soil erosion by water for different land use scenarios in the Spanish pre-pyrenees. CATENA 102:62–73 Merritt WS, Letcher RA, Jakeman AJ (2003) A review of erosion and sediment transport models. Environ Model Softw 18:761–799 Miller FP, Rasmussen WD, Donald Meyer L (1985) Historical perspective of soil erosion in the United States. In: Soil erosion and crop productivity, pp 23–48 Mishra PK, Rai A, Abdelrahman K et al (2022) Land degradation, overland flow, soil erosion, and nutrient loss in the Eastern Himalayas, India. Land 11:179 Mitchell JK, Soga K (2005) Fundamentals of soil behavior. Wiley, New York Morgan RPC (2009) Soil erosion and conservation. Wiley, New York Narayana DV, Babu R (1983) Estimation of soil erosion in India. J Irrig Drain Eng 109:419–434 Nearing M, Lane LJ, Lopes VL (2017) Modeling soil erosion. In: Soil erosion research methods. Routledge, Milton Park, pp 127–158 Osman KT (2014) Soil degradation, conservation and remediation. Springer

14 Pal SC, Shit M (2017) Application of RUSLE model for soil loss estimation of Jaipanda watershed, West Bengal. Spat Inf Res 25:399–409 Pal SC, Chakrabortty R (2019a) Modeling of water induced surface soil erosion and the potential risk zone prediction in a sub-tropical watershed of Eastern India. Model Earth Syst Environ 5:369–393 Pal SC, Chakrabortty R (2019b) Simulating the impact of climate change on soil erosion in sub-tropical monsoon dominated watershed based on RUSLE, SCS runoff and MIROC5 climatic model. Adv Space Res 64:352–377 Pal SC, Chakrabortty R, Roy P et al (2021) Changing climate and land use of 21st century influences soil erosion in India. Gondwana Res 94:164–185. https:// doi.org/10.1016/j.gr.2021.02.021 Pani SP, Mohapatra S (2011) Ravine erosion in India. Geogr You 68:1–4 Parr J, Papendick R, Hornick S, Meyer R (1992) Soil quality: attributes and relationship to alternative and sustainable agriculture. Am J Altern Agric 7:5–11 Penck A (1894) Morphologie der erdoberfläche. Engelhorn, Stuttgart Pilgrim D, Chapman T, Doran D (1988) Problems of rainfall-runoff modelling in arid and semiarid regions. Hydrol Sci J 33:379–400 Pimentel D (2006) Soil erosion: a food and environmental threat. Environ Dev Sustain 8:119–137. https://doi. org/10.1007/s10668-005-1262-8 Pimentel D, Burgess M (2013) Soil erosion threatens food production. Agriculture 3:443–463 Pimentel D, Allen J, Beers A, et al (1993) Soil erosion and agricultural productivity. In: World soil erosion and conservation, pp 277–292 Poesen J (1993) Gully typology and gully control measures in the European loess belt. In: Farm land erosion in temperate plains environments and hills proceedings, pp 221–239 Powlson DS, Gregory PJ, Whalley WR et al (2011) Soil management in relation to sustainable agriculture and ecosystem services. Food Policy 36:S72–S87

1

Introduction to Soil Erosion Study

Riahi K, Van Vuuren DP, Kriegler E et al (2017) The shared socioeconomic pathways and their energy, land use, and greenhouse gas emissions implications: an overview. Glob Environ Chang 42:153–168 Sharma HS (1979) The physiography of the lower Chambal valley and its agricultural development: a study in applied geomorphology. Concept Publishing Company Singh G, Babu R, Narain P et al (1992) Soil erosion rates in India. J Soil Water Conserv 47:97–99 Singh R, Panda R, Satapathy K, Ngachan S (2011) Simulation of runoff and sediment yield from a hilly watershed in the eastern Himalaya, India using the WEPP model. J Hydrol 405:261–276 Smith DD, Wischmeier WH (1957) Factors affecting sheet and rill erosion. EOS Trans Am Geophys Union 38:889–896 Stokes M, Mather AE, Belfoul A, Farik F (2008) Active and passive tectonic controls for transverse drainage and river gorge development in a collisional mountain belt (Dades Gorges, High Atlas Mountains, Morocco). Geomorphology 102:2–20 Toy TJ, Foster GR, Renard KG (2002) Soil erosion: processes, prediction, measurement, and control. Wiley, New York Trudgill ST, Goudie AS, Viles HA (2022) Weathering processes and forms. Geological Society, London, p 58 Verma GP, Singh YP (2017) Rainfed farming development in central India. Scientific Publishers Vijith H, Suma M, Rekha V et al (2012) An assessment of soil erosion probability and erosion rate in a tropical mountainous watershed using remote sensing and GIS. Arab J Geosci 5:797–805 Wu X, Wei Y, Wang J et al (2018) Effects of soil type and rainfall intensity on sheet erosion processes and sediment characteristics along the climatic gradient in central-south China. Sci Total Environ 621:54–66 Yihui D, Chan JC (2005) The East Asian summer monsoon: an overview. Meteorol Atmos Phys 89:117–142 Zachar D (2011) Soil erosion. Elsevier, Amsterdam

2

Geophysical Settings of South Bengal

Abstract

Keywords

The “Bengal basin” is located in the state of West Bengal, in the north-eastern region of the Indian subcontinent. The “Bengal basin” was formed by divergent subsidence during the “Middle-Upper Cretaceous period”. The subaerial fluvial classics of the Bolpur Formation, as well as its stratigraphic variation of the shell limestone and shale–sandstone of the Ghatal Formation, were deposited in the shelf region during this period. A thick succession of roughly 20 km of deposits in the “Bengal basin” was formed by the proximal accumulation of orogenic material from the eastern “Himalaya and the Indo-Burman Uplifts”. The “Bengal basin” experienced a polycyclic tectonic evolution. It was a diverge edge basin that formed as a result of the split of “Gondwanaland” along rifted margin of the “Indian plate” from the “Carboniferous” to the “upper Eocene”. It shows all of the evolutionary aspects of marginal sag basins. During the “lower Cretaceous”, the Indian plate eventually broke from “Gondwanaland” and migrated northwards, colliding for the first time with the “Eurasian Plate” in the north and the “Burmese Plate” in the northeast during the “upper Eocene”. This chapter is aimed to give the clear-cut idea about geophysical settings of the study region. In this perspective, the overall topographical, geological as well as geomorphological characteristics of this region can be identified.

Bengal basin Tectonic evolution Gondwanaland Evolutionary aspects

.

2.1

.

.

Study Area

The study area considers south Bengal basin where agriculture and fertile soil is present predominantly along with infertile soil of Bankura and Purulia. South portion of “Bengal basin” is the western fringe of the “Ganga–Brahmaputra– Meghna river delta”1 (Fig. 2.1). According to Alam et al. (2003) since the “Miocene”, this basin has seen marine regression and transgression, as well as “climate change and neotectonic activity”, and is considered the western stable shelf province of the “Bengal basin”. The laterite

1 With an area of over 150,000 km2, the Ganges– Brahmaputra–Meghna (GBM) delta is the largest in the world and poses a significant obstacle to adapting to the expected future climatic stress of roughly 200 million people. Two-thirds of Bangladesh (around 100,000 km2) and a portion of the Indian state of West Bengal are included in this deltaic area. One of the greatest population densities in the world, exceeding 1000 people per square kilometer, is found in the low-lying Bangladesh delta plain, where at least 10% of the land is below 1 m above mean sea level.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 S. C. Pal and R. Chakrabortty, Climate Change Impact on Soil Erosion in Sub-tropical Environment, Geography of the Physical Environment, https://doi.org/10.1007/978-3-031-15721-9_2

15

16

Fig. 2.1 Study area

2

Geophysical Settings of South Bengal

2.2 Physical Setup

and “ferruginous soils” of “Rarh” are found in limited to the eastern part of “Chota Nagpur Plateau fringe” which cover Murshidabad, Bankura, Purulia, Birbhum and West Midnapore (approximately 7700 km area) (Ghosh et al. 2015). Due to parallel underlying structure the slope flows from west to east. The peninsular drainage system flows parallel to west–east direction. The “lateritic Rarh region” into patches of badland and tropical deciduous forest of West Bengal was dissected through this drainage system namely Ajay, Damodar, Dwarakeswar, Brahmani, Dwarka, Mayurakshi, Silai, Kangsabati and Subarnarekha. The high elevation zone in Rajmahal basalt trap is found in western part where basalt rocks are found on “Gondwana rock beds” (Ghosh and Guchhait 2019). The alluvium zones are found within 0–50 m and laterite are mostly found in the elevation zone of 50–100 m. A notable zone of dislocation separates the Precambrian shield and Gondwana rocks from the Bengal basin in the west and an easterly inclines shelf in the south-west (Hossain et al. 2019). This region mainly shows the lateritic soil along with ferruginous sandstone, red shale, grit and gravels containing fossils. Acharyya et al. (2000) said that lateritic profiled on Quaternary alluvium has been recorded which contains lateritic pisolites. In previous literature it had said that lateritic soils are the oldest soils of “IndoGangetic plains” (Niyogi 1975; Singh et al. 1998). The Bengal basin’s Ganga–Brahmaputra delta is a structural depression between the Indian and Eurasian plates (Bandyopadhyay 2007). Fringing basalt of “early Cretaceous Rajmahal Basalt traps, sandstone of Gondwana formations, granite, gneiss of Archean formations, and late Quaternary older alluvium” were found in the western region of Bengal basin, whereas the lateritic zone of Bengal basin is located in the north-western portion (Ghosh et al. 2015). Thick regolith envelop of laterites is found in “Chota Nagpur Plateau” but it present as low level sedimentary area in the eastern sides of this plateau which is located in river valleys and interfluves region (Ghosh and Guchhait 2020).

17

2.2

Physical Setup

The state of “West Bengal” in the Indian union consists the south-western and north-western segments of the cultural–geographic entity regarded as Bengal; the other half, which was separated from it by formal and indigenous colonial events in 1947 as part of India’s and east Pakistan’s political independence, has become the sovereign state of Bangladesh (De 1990). The whole “Bengal basin” is that portion of the enormous “Indian shield” east of longitude 87° E that vanishes behind a covering of “Gangetic alluvium”. A number of “intracratonic Gondwana” basins in the “Damodar Valley2”, as well as a few earlier “Tertiary exposures” around “Durgapur and Baripada” and the “late Mesozoic volcanoes” of the “Rajmahal Hills”, are located west of it. In the “Shillong Plateau”, that confronts the “Garo- Rajmahal gap” through which the “Ganga” and its tributaries escape to the sea, the “Archaean Shield” reappears. The “Mahananda”, together with the “Punarbhaba” and Tangan in West Bengal and the “Atrai” in Bangladesh, characterized the Holocene. A path south showed the following spatial pattern: (a) “the boulder-strewn undulated plains of the Tarai getting stretched westward, (b) the sandy undulations of the Duars along rivers starting to come down from Bhutan (into Jalpaiguri and Kuch Bihar), (c) the clayey flat of the Tal, and (d) the slow rise eastward, of the ferallitic Barind from which the old regional name of overall south of the area as Barendrabhumi”.

2.2.1 Geology In West Bengal, there are no worthwhile large constructions in the on land region that need to be drilled into. Seismic studies have identified 2 The Damodar River is located in south central Bihar state, on the Chota Nagpur Plateau, in the northeastern region of India. The river travels through West Bengal for its whole 563 km length before joining the Hooghly River. Damodar is the main economic activity in the upper valley region. The most advanced industrial location is in Damodar Valley.

18

some of the small elements, which have already been drilled without economic success (Roybarman 1984). Nevertheless, it is impossible to believe that the basin, with its massive sediment thickness, diverse lithology and interaction of marine, intermediate and deltaic ecosystems, has not created and imprisoned vast amounts of hydrocarbons (Roybarman 1984). The vast Indian shield vanishes behind a covering of alluvium so beyond West Bengal’s western edge. A succession of intracratonic Gondwana basins, a series of thrust zones in Singhbhum and widespread exposure of basic volcanic rocks in the Rajmahal Hills characterize the exposed part of the shield abutting the Bengal basin3 (Supriya 1966). The Lower Gangetic Plains of India’s subcontinent are one of the world’s largest fluviordeltaic plains. The Bengal Sedimentary Basin is hidden behind them. Changing climate and neotectonism are key influences in the formation of landscapes and soils in the “Upper and Middle Gangetic plains”, according to regional investigations (Kumar et al. 1996; Singh et al. 1998). Ball (1877) classified the “Quaternary sediments” of the study region in West Bengal into “older and newer alluvium”, although the foundation for this categorization was unclear. Various researchers defined the morphology of West Bengal soils based on their distribution in the “Ganga riverine alluvium, Ganga flats, Ganga uplands, Ganga lowlands, Vindhyan riverine lands, Vindhyan flat lands, Vindhyan highlands, and coastal regions” (Singh et al. 1998). The “older alluvium” of the “Bengal basin, West Bengal, India and Bangladesh”, according to (Morgan and McIntire 1959), comprised of numerous Pleistocene terraces known as the “Barind” and the “Madhupur Jungle” Niyogi et al. (1970), Mallick (1971), and Bhattacharya and Banerjee (1979) investigated the geomor-

3

Bengal basin is a region in Bangladesh and India’s West Bengal state that is located at the northeastern tip of the Indian Peninsula. It is roughly located between Latitudes 25° and 20° 30′ and Longitudes 87° and 90° 30′. The basin reaches the Bay of Bengal’s offshore zone as it moves south.

2

Geophysical Settings of South Bengal

phology of the “Lower Gangetic Plain”, although these studies did not involve soil analysis. The “West Bengal basin” is a portion of the wider “Bengal pericratonic” environment, which extends southerly into the “Bay of Bengal” up to 400 m bathymetry and encompasses the West Bengal State of India and parts of western and southern Bangladesh (Prasad and Pundir 2020). The early entrance of the Eastern Indian Ocean is closely connected to the tectonic evolvement of the Bay of Bengal that also lies off the east coast of India. Although Norton and Sclater (1979) initially explained the as a whole development of the Indian Ocean, the specifics of the tectonic transformation of the “Bay of Bengal” have yet to be revealed (Talwani et al. 2016). Geomorphology is the study of landscape structure and function, as well as the interplay between rivers and their surroundings. The kind and quantity of sediment delivered to the system are determined by geology, which acts as a limit on the level of geomorphic alterations. Through fluvial processes streams construct flood plains and changes many other geological structures by creating natural leaves, oxbow lake, point bar and meandering channels. Inside this “late Quaternary series”, two main stratigraphic facies, the oxidized facies and the sand facies, may be seen at the bottom of boreholes (Goodbred and Kuehl 2000). The “oxidized and sand units” are overlain in the “southern Bengal basin” by “three facies: Lower delta mud, muddy sand, and thin mud” (Goodbred and Kuehl 2000). Wood bits were also found in the same sediments as “intertidal mollusk shells and planktonic sea diatoms”, according to Umitsu (1993). The finergrained stratigraphy in the far “western Bengal basin” is analogous to that in the eastern section, with a “numerous long stretches of interfingering muds and fine sands that thin out over a shoaling oxidized facies” at the basin’s boundary. The Thin Mud facies caps the western delta, and carbon dating from consecutive peat and clay layers show ages ranging from “6500 to 7500 cal year BP” at “6–12 m deep” and “2000– 6000 cal year BP” in the top 5 m (Banerjee 1987). A meandering structure of a river and lateral movement of rivers and its bank instability

2.2 Physical Setup

causes sometimes channel shifting. Erosion is caused by the current’s strength and consistency, and it can modify the route of the river. The geologic monitoring manually provides guidance for resources manages seeking to establish the status and trends of geologic resources and to further the understanding of how geologic process impact dynamic ecosystem. Geologic processes include the uplift of isostatic changes in land surfaces elevation and also form a large sedimentary basin where the surface of the earth drops and is filled with material eroded from the parts of the landscape. This research area is made up of various rocks from various geological periods. The study area’s Archean formation is largely made up of old gneiss, granite and shiest. The rock with various minerals is found in the form of metallic and non-metallic. The metamorphic granitite and basaltic materials with some amount of sediment composed this igneous type of rock (Fig. 2.2). Largest portion of the basin is associated with Barakar formation, Barren measure formation, Chota Nagpur gneissic complex, Dalma volcanos, Dubrajpur formation, Durgapur bed, Gabbro and Anthrasite complex, Kuilapal Grainite, laterite and bauxaite, Manbhum Granite, Panchet Formation, Rajmahal trap, Raniganj Formation, Singhbhum Group, Talchir, unclassified metamorphic and undefined fluvial sediments. The north-western part is the extended part of Chota Nagpur Plateau4 and associated with Gondwana formation. Western portion of this basin is riched with metamorphic old Gneiss rock and recent alluviam deposition is found in the eastern portion. The lateritic soils have also a role to grow the Sal plant in this dominated area (Table 2.1).

4

The majority of the state of Jharkhand, as well as nearby areas of Bihar, Odisha, West Bengal and Chhattisgarh, are covered by the Chota Nagpur Plateau, a plateau in eastern India. To the north and east of the plateau is the IndoGangetic plain, while to the south is the Mahanadi river basin. The Chota Nagpur Plateau has a total size of around 65,000 km2.

19

2.2.2 Topography The separation and assessment of the causal components is one of the key goals of scientific hydrology. The “climatic factor” and the “soilvegetation complex” are two variables that have a significant impact on runoff volume. Drainage basin topography is a substantially permanent feature that primarily effects the concentration or temporal distribution of flow from a drainage basin. The GIS framework was used to gather data on “drainage area, stream length, stream density, land slope, channel slope, area-altitude distribution and basin water body area”. These ranges have elevations ranging from − 9 to 699 m. The “Bay of Bengal” runs across the southern section of West Bengal. The Gangetic plain is alluviumrich and naturally productive. The salty soil of Sundarban makes the area unfit for farming.

2.2.3 Geomorphology Murthy (1982) investigated the micromorphology of reference soils from six Lower Gangetic Plains soil profiles, which hold an important interpretative position in the soil categorization and span a broad region. In the Lower Gangetic Plains, no extensive research on the development of features in connection to soils has been done (Kooistra 1982). Of operating to morphodynamics, the study of landscape changes due to erosion and sedimentation or operating to dynamic changes in morphology. Various geomorphological units are found in this particular region namely “water bodies, salt pan, quarry and mine dam, pediment pediplain complex, offshore island, moderately dissected hills and valleys, low dissected hills and valleys, highly dissected plateau, highly dissected hills and valleys, flood plain, deltaic plain, dam and reservoir, coastal plain, anthropogenic terrain and alluvial plain” (Fig. 2.3). Western most point of this south Bengal is associated with pediment, barren wasteland and large amount of mechanical weathering also found here. Water-induced surface erosion was also found on lateritic type of soil. In the western part of south Bengal

20

Fig. 2.2 Geology

2

Geophysical Settings of South Bengal

2.2 Physical Setup

21

Table 2.1 Geological formation Formation

Period

Area

Area in percentage (%)

Durgapur bed

Jurassic

7.00

0.01

Kuilapal granite

Mesoproterozoic

72.11

0.12

Undiff. fluvial/aeolian/coastal and glacial sediments

Quaternary

47,417.80

78.07

Laterite and bauxite

Cenozoic

1513.09

2.49

Gabbro and anorthosite complex

Proterozoic

290.77

0.48

Rajmahal trap

Cretaceous

224.98

0.37

Manbhum granite

Mesoproterozoic

553.11

0.91

Panchet\Pachmarhi fm

Triassic

366.52

0.60

Dalma volcanics

Archaean–Proterozoic

442.72

0.73

Unclassified metamorphics

Archaean–Proterozoic

571.52

0.94

Chota Nagpur gneissic complex

Proterozoic (undifferentiated)

6079.22

10.01

Singhbhum Gp

Palaeoproterozoic

2263.10

3.73

Barren measure Fm

Permian

134.59

0.22

Talchir Fm

Carboniferous

11.23

0.02

Raniganj Fm

Permian

607.23

1.00

Dubrajpur Fm

Jurassic–Cretaceous

47.29

0.08

Barakar Fm

Permian

138.48

0.23

60,740.74

100

Total

(Bankura, Purulia) where tillage practices is facing an acute problem. Maximum sedimentation deposit of river was seen in the lower portion where river meets to Bay of Bengal.5

2.2.4 Drainage System The science of hydrology is concerned with the processes of hydrological cycle, their three 5

The Andaman and Nicobar Islands of India and Myanmar border the Bay of Bengal on the east, which is the northeastern portion of the Indian Ocean. Bangladesh borders the Bay of Bengal on the north and northwest. Sangaman Kanda in Sri Lanka and the most northwestern point of Sumatra in Indonesia form its southern boundary. It is the world’s largest bay, a body of water. South Asian and Southeast Asian nations are reliant on the Bay of Bengal. It was given the name Bay of Bengal, while British India was still in existence as a result of the ancient Bengal area. The Port of Kolkata functioned as the entrance to Indian Crown control at the time. Along the bay lie Cox’s Bazar, the longest sea beach in the world, and Sundarbans, the largest mangrove forest and home to the Bengal tiger.

dimensional distribution across spatial scales ranging from regions, through catchment and rivers corridors to river channel and across temporal scale ranging from centuries to second. Within this wide range of time and space scale, hydrologists are concerned both with the average condition, seasonal variability and temporal trend in water resources with the magnitude and frequency of hydrological events such as extreme rainstorms, floods and droughts. Thus, hydrology has always been a truly four dimensional discipline focusing upon the complexities of water storage and movement across the landscape in space and time. Furthermore, hydrology has always been concerned with ecology to the extent that vegetation is an important control on hydrological processes. The Ganga–Brahmaputra drainage basin had a major role in the formation of the “Bengal fan” and is one of the world’s biggest contemporary deltas (Milliman and Syvitski 1992; Garzanti et al. 2004). This river system has conveyed vast amounts of “sediments from the Himalayan” region since the tertiary. As

22

Fig. 2.3 Geomorphology

2

Geophysical Settings of South Bengal

2.2 Physical Setup

a result, the Himalayan belt will very certainly be searched for a probable source. The “Ganga– Brahmaputra” river system conveys roughly 1800 tonnes/km2 of sediment per year on aggregate, with suspended loads ranging from 540 to 1157 million tonnes per year (Milliman and Syvitski 1992). The “West Bengal basin” is a “pericratonic polyhistory sedimentary basin” located along India’s East Coast borders, and its depositional evolution is remarkably similar to that of the “Mahanadi, Krishna–Godavari, Palar and Cauvery” basins (Veevers et al. 1996). The break-up of Eastern “Gondwanaland” and subsequent separation of the “Indian Plate” from the “Antarctica–Australian Plate” led in the entrance of the “Bay of Bengal”, as did adjacent Indian East Coast basins, during the “late Jurassic–early Cretaceous (Tithonian–Neocomian)” period (Prasad and Phor 2009). In such regions, “Precambrian basement” or eroding Coors of Gondwanas are succeeded by “late Cretaceous– Cenozoic passive-margin successions”. Nevertheless, in the “subsurface” parts of the “West Bengal basin”, “large basaltic traps” and “accompanying infra- and intertrappean layers” of the later “Jurassic–early Cretaceous Rajmahal Formation”, previously classified as “Upper Gondwana”, top the “Permo-Triassic Gondwana” sequences. Thick Gondwana segments have recently been discovered in a multitude of explorative “boreholes drilled” in the “West Bengal basin”, whose instances are usually regarded as the subsurface eastern extension of the “Damodar Gondwanic Graben” under its “pericratonic set-up”, as well-developed “Gondwana successions” are exposed in the adjoining “Raniganj basin” towards the basin’s western margins (Prasad and Pundir 2020). According to geomorphic history, the “Damodar river”, like other rivers, flowed east to join the “Bhagirathi” in the mid-eighteenth century, but has since relocated its mouth 120 km to the south (Deshmukh and Goswami 1973). The present study is concerned with soil erosion and associated problem of south Bengal. All of these rivers of the district not only act as the agent of fluctuating flow but also form of an important source of drinking water, irrigation and hydropower

23

project. The features of rivers are determined by their discharge pattern, silt content, and other factors. All of the rivers flow easterly and south easterly direction (Fig. 2.4). The whole river basin is characterized by fine drainage and texture density. Drainage in lower region indicating the basin is characterized by highly permeable material with low relief (Malik and Pal 2020). Mainly dendritic river pattern is found in this particular basin. Dendritic pattern, in general, is considered as a group of resequent streams within homogeneous sand and gently lithology.

2.2.5 Soil Erosion occurs when fluvial water discharge, a stream-related process, dissolves or removes surface material. Stream-produced fluvial erosion picks up weathered material for transport and transfer to new places, while sediment is laid down by another process, deposition. The clay, silt, and sand deposited by moving water are referred to as alluvium. Agricultural activity and urbanization are attracted to riverbed soils, which are soaked in new nutrients from floodwater. Despite our knowledge of flood devastation in the past, floodplains have been populated, raising questions about human risk perception. The fact that a low-lying region along a stream channel is liable to floodplain conditions which is ultimately influences the overall geomorphic characteristics in terms of sedimentation, channel width and inundation. The terms “soil erodibility” and “climate erosivity” are interchangeable terms for bank erosion. The flood in main hazards which causes severe river bank erosion in the lower catchment of the rivers in the Bengal Basin. Soil characteristics of river bank side determine the erosion. Soil stratification and particle size also cause for fluvial condition. Soil characteristics of this study area depend on geological structure. Laterite and alluvium cover the vast majority of the surface. The lateritic gravel and calcareous are common in surface layer. In the upper part, soil of Purulia district is reddish in colour, and it is said the iron content of this soil is high in comparison to the rest of the

24

Fig. 2.4 Drainage

2

Geophysical Settings of South Bengal

2.2 Physical Setup

state. This district’s soil is largely sedentary in character, with alluvial soil only present near the valley bottom. Undulated highland soils are shallow, gravelly, coarse and have a limited water retention capacity. This upper land’s soil texture is fine loamy in nature and associated with a fine texture (Fig. 2.5). The soil of the middle reach, which is part of the Bankura area, may be categorized into three main groups. i. Soil that is red alluvial soil is a type of soil that is found in rivers and streams. ii. Soil made of laterite in the south central, south-eastern and southern parts of the district, near Bishnupur, Kotulpur, and Raipur block, respectively, typical red soil has a restricted distribution. According to texture types, soil of the district can be classified following types: “Sandy loam”, “loam”, “sandy clay loam”, “clay loam” and “clay”. The lower region of this study area is rich in sediment and fertile soil. Clays in alluvial soil clays have very well-physiochemical features. X-ray basal spacings are generally weak, DTGA organisms are broad and SiO2/Al2O3 proportions are low in such soil clay particles. Dehydroxylation processes have lower activation energy. In tetrahedral and octahedral strata, the chemical structure of these clay minerals shows varied degrees of isomorphous substitution. In tetrahedral and octahedral strata, the chemical structure of these clay minerals shows varied degrees of isomorphous substitution. The different physiochemical property of soil in this study region is shown in Fig. 2.6.

2.2.6 Climate The climate of this region varies from western (Puruliya and western portion of Bankura district) to eastern portion, and the western portion is much drier comparison to the eastern portion which is situated in the south-eastern Bengal. Apart from this, there is an association of hot summer with extreme humidity and cold dry winter season. The maximum amount and intensity of rainfall is associated in the monsoon period. There is a direct impact of monsoon upon the variability and nature of rainfall.

25

The south-west monsoon usually takes place during the month of June to September and returns in between October to November. The average annual rainfall of Puruliya, Bankura, Hugli, Bardhaman and Medinipur districts is 1193.948, 1340.382, 1340.442, 1581.855 and 1629.859 mm, respectively (Fig. 2.7). Overall the whole area is under the monsoon climate. Frequently happening drought and floods in upperparts and lower part naturally change its geomorphic features. In flood time in Bankura district that is in under middle portion is high; it creates meandering channel that go to oxbow lake in future. Climate change by extreme weather events which is drought and floods mainly in March April and June–July finally changes river shape, and it has impact on fluvial condition. This region receives 111.76 mm of yearly rainfall (which is not evenly distributed throughout the basin). Between 1961 and 2010, the highest temperature was discovered in May, ranging from 27 to 32 °C, while the lowest temperature was recorded in January, ranging from 10 to 15 °C. The variability of temperature, rainfall and evapotranspiration is shown in Figs. 2.8, 2.9 and 2.10, respectively.

2.2.7 Vegetation Vegetation and the rate of the plant growth directly related soil erosion and fluvial aspect. Plants grew on newly deposited bars and braided plain regions that were occupied during periods of low flow. The pace of bank erosion was delayed by vegetation. In contrast to a braided system, where channel flipping occurs almost continuously, vegetation maintained a coherent channel until flow was diverted completely by cut-off and avulsion, at which time the prior channel tended to be highly unfavourable for flow. As a result, vegetation prohibits numerous channels from coexisting. The impacts of vegetation on “river behaviour” and “fluvial geomorphology” are examples of these processes. The fluvial process and topography can be influenced by vegetation through five processes: “flow resistance, bank strength, bar

26

Fig. 2.5 Soil

2

Geophysical Settings of South Bengal

2.2 Physical Setup

27

Fig. 2.6 Soil various soil properties of this study region; percentage of sand (a), percentage of silt (b), percentage of clay (c), pH (d), organic carbon (e), chemical exchange capacity (f), bulk density (g), C/N ratio (h), and nitrogen (i)

28 Fig. 2.7 Rainfall

Fig. 2.8 District-wise variation of annual temperature in the study area during 1901 and 2020

2

Geophysical Settings of South Bengal

2.2 Physical Setup

29

Fig. 2.9 District-wise variation of annual rainfall in the study area during 1901 and 2020

Fig. 2.10 District-wise variation of annual potential evapotranspiration in the study area during 1901 and 2020

sedimentation, log creation and concave bank beach deposition”. Here in this particular selected area, deciduous broadleaf forest, mixed forest, shrub land, plantation and grassland have developed through vegetation (Fig. 2.11).

According to the report from forest survey of India (FSI 2019), the study region has been divided into various forest types, i.e. very dense forest, moderate dense forest, open forest, etc. The spatial distribution of all forest categories is

30

2

Geophysical Settings of South Bengal

Fig. 2.11 Vegetation

shown in Table 2.2. The state’s reserved forest area (RFA) is 11,879 km2, or 13.38% of the total land area. Reserved, protected and unclassified forests account for 59.38, 31.76, and 8.86% of

the state’s total forest area, respectively. However, the state’s digitized forest boundary is 13,418.77 km2 (FSI 2019).

References

31

Table 2.2 District-wise forests cover area (km2) of the study region District

Geographical area (GA)

Very dense forest

Moderate dense forest

Open forest

Bankura

6882

222.33

395.27

667.98

Bardhaman

7024

57.53

91.78

190.00

Birbhum

4545

1.00

34.14

149.66

Haora

1467

0.00

5.00

253.77

Hugli

3149

0.00

14.00

146.00

Kolkata Murshidabad

185

0.00

0.00

1.00

5324

0.00

53.06

291.83

Nadia

3927

1.00

160.16

318.84

North 24 Parganas

4094

13.02

184.98

524.98

Paschim Medinipur

9368

256.21

591.64

1313.69

Purba Medinipur

4713

1.99

197.96

620.10

Puruliya

6259

37.36

306.94

571.58

South 24 Parganas

9960

983.10

745.03

1060.58

References Acharyya S, Lahiri S, Raymahashay B, Bhowmik A (2000) Arsenic toxicity of groundwater in parts of the Bengal basin in India and Bangladesh: the role of quaternary stratigraphy and holocene sea-level fluctuation. Environ Geol 39:1127–1137 Alam M, Alam MM, Curray JR et al (2003) An overview of the sedimentary geology of the Bengal Basin in relation to the regional tectonic framework and basinfill history. Sed Geol 155:179–208 Ball V (1877) Geology of the Rajmehal Hills Bandyopadhyay S (2007) Evolution of the Ganga Brahmaputra delta: a review. Geogr Rev India 69:235–268 Banerjee M (1987) Palaeobiology in understanding the changes of sea-level and coastline in Bengal Basin during Holocene Period. Indian J Earth Sci 14:307– 320 Bhattacharya A, Banerjee S (1979) Quaternary geology and geomorphology of the Ajay-Bhagirathi valley, Birbhum and Murshidabad districts, West Bengal. Indian J Earth Sci 6:91–102 De B (1990) West Bengal: a geographical introduction. Econ Pol Wkly 25:995–1000 Deshmukh DS, Goswami AB (1973) Geology and groundwater resources of the alluvial areas of West Bengal FSI (2019) Forest survey of India. https://fsi.nic.in/. Accessed 22 Apr 2022 Garzanti E, Vezzoli G, Andò S et al (2004) Sand petrology and focused erosion in collision orogens: the Brahmaputra case. Earth Planet Sci Lett 220:157–174 Ghosh S, Guchhait SK (2019) Modes of formation, palaeogene to early quaternary palaeogenesis and

geochronology of laterites in Rajmahal Basalt Traps and Rarh Bengal of Lower Ganga Basin. In: Quaternary geomorphology in India. Springer, Berlin, pp 25– 60 Ghosh S, Guchhait SK (2020) Laterites of the Bengal Basin: characterization, geochronology and evolution. Springer, Berlin Ghosh S, Guchhait SK, Hu X-F (2015) Characterization and evolution of primary and secondary laterites in northwestern Bengal Basin, West Bengal, India. J Palaeogeogr 4:203–230 Goodbred S Jr, Kuehl SA (2000) The significance of large sediment supply, active tectonism, and eustasy on margin sequence development: late quaternary stratigraphy and evolution of the Ganges-Brahmaputra delta. Sed Geol 133:227–248 Hossain MS, Khan MSH, Chowdhury KR, Abdullah R (2019) Synthesis of the tectonic and structural elements of the Bengal Basin and its surroundings. In: Tectonics and structural geology: Indian context. Springer, Berlin, pp 135–218 Kooistra MJ (1982) Micromorphological analysis and characterization of 70 benchmark soils of India: a basic reference set. NSSI Kumar S, Parkash B, Manchanda M et al (1996) Holocene landform and soil evolution of the western Gangetic Plains: implications of neotectonics and climate. J Geol 88:100–108 Malik S, Pal SC (2020) Decreasing downstream channel capacity of a low-lying ephemeral river of Bengal Basin, Eastern India. J Indian Soc Rem Sens 48:1057– 1081 Mallick S (1971) Geomorphology and quaternary geology of the area around Burdwan, West Bengal, India. Unpublished PhD Thesis, Indian Institute of Technology, Kharagpur, p 131

32 Milliman JD, Syvitski JP (1992) Geomorphic/tectonic control of sediment discharge to the ocean: the importance of small mountainous rivers. J Geol 100:525–544 Morgan JP, McIntire WG (1959) Quaternary geology of the Bengal basin, East Pakistan and India. Geol Soc Am Bull 70:319–342 Murthy R (1982) Benchmark soils of India: morphology, characteristics, and classification for resource management. National Bureau of Soil Survey and Land Use Planning, ICAR Niyogi D (1975) Quaternary geology of the coastal plain of West Bengal and Orissa. Indian J Earth Sci 2:51–61 Niyogi D, Sarkar S, Mallick S (1970) Geomorphic mapping in plains of West Bengal, India. Select Pap Phys Geogr 1:89–94 Norton I, Sclater J (1979) A model for the evolution of the Indian Ocean and the breakup of Gondwanaland. J Geophys Res Solid Earth 84:6803–6830 Prasad B, Phor L (2009) Palynostratigraphy of the subsurface Gondwana and post-Gondwana Mesozoics of the Cauvery basin, India. J Palaeontol Soc India 54:41

2

Geophysical Settings of South Bengal

Prasad B, Pundir B (2020) Gondwana biostratigraphy and geology of West Bengal Basin, and its correlation with adjoining Gondwana Basins of India and Western Bangladesh. J Earth Syst Sci 129:1–45 Roybarman A (1984) Geology and hydrocarbon prospects of West Bengal. Pet Asia J (India) 6 Singh LP, Parkash B, Singhvi A (1998) Evolution of the lower Gangetic plain landforms and soils in West Bengal, India. CATENA 33:75–104 Supriya S (1966) Geological and geophysical studies in western part of Bengal basin, India. AAPG Bull 50:1001–1017 Talwani M, Desa MA, Ismaiel M, Krishna KS (2016) The tectonic origin of the Bay of Bengal and Bangladesh. J Geophys Res Solid Earth 121:4836–4851 Umitsu M (1993) Late quaternary sedimentary environments and landforms in the Ganges Delta. Sed Geol 83:177–186 Veevers J, Tewari R, Mishra H (1996) Aspects of late Triassic to early Cretaceous disruption of the Gondwana coal-bearing fan of east-central Gondwanaland, pp 637–646

3

Morphotectonics Characteristics and Its Control on Soil Erosion

Abstract

3.1

The influence of the “morphotectonic characteristics” on “erosion potentiality” assessment has been estimated in this chapter. The related parameter has been selected with considering recent literatures related to this field. In this perspective, the “evidential belief function (EBF)”, “logistic regression (LR)” and ensemble of “EBF-LR” have been considered. Here, the efficiency of ensemble model is quite high then any single alone model, i.e. “EBF” and “LR”. The “area under curve (AUC)” values from “receiver operating characteristics (ROC)’ for ensemble of “EBF-LR” are 0.99 and 0.92, respectively. The western and central parts of this region are related with an erosion potential zone that ranges from very high to high. So, the special emphasis regarding the suitable management strategies has to be taken into consideration for this region to overcome this type of situation. This sort of data aids decision-makers in implementing the most appropriate development initiatives in vulnerable areas. The role of future researchers is to quantifying the erosion potentiality with maximum possible accuracy and considering maximum-related parameters. Keywords

Morphotectonic characteristics potentiality Ensemble model management strategies

.

. Erosion . Suitable

Introduction

Soil is a “non-renewable resource” in the sense that, whereas topsoil forms over ages, the world’s rising human population depletes it over decades (Pal 2016). “Soil erosion prevention techniques” are crucial for sustainable development since soil is a “non-renewable resource” that accounts for 97% of all “agricultural production” (Pimentel 1993). “Soil erosion” is a “natural process” driven by geomorphic factors such as flowing water, winds, coastal waves, and glaciers. As a result, it has been happening from the beginning of time (Leopold et al. 2020). “Soil erosion” due to water is thought to be responsible for around 80% of the present “degradation” on “agricultural land” across the world (Angima et al. 2003). “Soil erosion” is a major issue all around the world. Human-induced degradation affects 644.4 million ha of land worldwide (UNEP 1997). However, as anthropogenic involvement has risen over time, it has become a severe concern. “Soil erosion” is the process of soil particles being detached and transported by geomorphic processes (Chakrabortty et al. 2020b). Detaching “soil particles” from the “soil mass” is referred to as detaching, and transportation refers to the transfer of separated soil particles (“sediment”) from their original position (Lisle et al. 1998). In India, severe soil depletion has resulted in a high rate of “sedimentation” in “reservoirs” and reduced soil quality, posing

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 S. C. Pal and R. Chakrabortty, Climate Change Impact on Soil Erosion in Sub-tropical Environment, Geography of the Physical Environment, https://doi.org/10.1007/978-3-031-15721-9_3

33

34

3

Morphotectonics Characteristics and Its Control on Soil Erosion

environmental challenges (Srinivasarao et al. 2014). “Soil erosion” generally impacts “forest lands, agricultural fields, dry and semi-arid regions, surface mines, roadways, building sites and coastal areas”, among others. Furthermore, because “soil development” is a slow process, the loss of the top layer of soil renders it barren for a lengthy period of time, causing major “agricultural challenges”. In India, “soil erosion” is a major issue that poses a serious threat to people’s lives and wellbeing (Pal et al. 2021). Forest lands, dry and “semi-arid regions”, agricultural fields, building sites, highways, disturbed grounds, surface mining, glaciated and coastal regions and locations where natural or geologic instabilities occur are all examples of where it may be found. In the worst-case situation, it might result in the full loss of soil and the exposure of bed rock. Because soil development takes such a long time, if it is fully exhausted, new soil will take hundreds or millions of years to form, and the area would remain unproductive in the meanwhile. In India, water has a 90% role among the two primary drivers of erosion, notably water and air. In the event of water erosion, topsoil is removed by impacting raindrops or “runoff” water moving over the ground’s surface. High hitting speeds (up to 9 m/s) and a significant number of droplets combine to create a powerful “hydrodynamic pressure that detaches a significant number of soil particles (Saroha 2017). Raindrop detachment is extensive, whereas “runoff” is restricted to narrow, easily discernible channels. “Rainfall, runoff, soil properties, terrain and cover circumstances” all influence the rate of separation (Hudson 2015). As a result, the rates of erosion are determined by “climate, hydrology, structure, terrain, soil surface characteristics” and the interplay of all of these primary elements (Ravi et al. 2010). Water washes away a lot of top soil during heavy rains. When raindrops strike the surface, “sands and silts” are separated from the soil body, causing splash erosion (Morgan 2009). As the rain keeps falling, a considerable volume of water runs as “surface runoff”, removing top soil from a vast region, this is known as sheet erosion (Singh and Hartsch 2019). As a result of

the increased intensity of the “runoff” in regions of extreme gradient and soft parent material, many finger-shaped holes may occur all throughout the region. Where water rushes quickly, such as around the borders of roadways and embankments, such grooves or channels emerge on the surface, this is known as rill erosion. If the erosion persists, these rills may intensify and expand into gullies. “Gully formation” may influence large regions, transforming the entire area into badlands. The vertical and lateral eroding of rills causes “gully erosion”. On sandy soils, gullies are more likely to form. Soil erosion is caused by a mixture of human and animal disturbances, as well as the action of “water, winds and glaciers”. Soil loss is exacerbated by deforestation, overgrazing and poor agricultural practices (Osman 2014). Trees and plants protect soil from erosion by tying it together and continuously giving moisture to it. During rainstorms, vegetation and tree debris function as a buffer against splash erosion. As a result, deforestation always results in soil erosion and flooding. So, identification of erosion-prone regions with appropriate modelling is one of the important and initial tasks by the researchers in this field. In this work we applied the “evidential belief function (EBF)”, “logistic regression (LR)” and ensemble of “EBF-LR” for estimating the “erosion potentiality”. “Logistic regression” (also termed as logit model) is an advanced analytics and modelling technique that also has machine learning technology. The dependant parameters in this analytics technique seems to be either finite or categorical: “either A or B (binary regression)” or a “variety of finite possibilities A, B, C, or D (multiple regression) (multinomial regression)” (Strickland 2015). By estimating “probabilities” using a “logistic regression equation”, it is employed in “statistical packages” to comprehend the connection between the dependent parameters and one or even more independent variables (Al-Ghamdi 2002). Shafer (1976) created the “Dempster–Shafer theory of evidence”, which is a geographical integrated model with “mathematical expression” that is commonly used as a “knowledge-based

3.2 Materials and Methods

method” to mineral potential assessment. The “Dempster–Shafer” approach’s strength is its ability to deal with insufficient data coverage. Apart from this, it was clearly established by different researchers that the optimal capacity of the ensemble model is quite high than any standard alone method (Kocaguneli et al. 2011; Amozegar and Khorasani 2016; Başakın et al. 2021). In general, ensembles have higher prediction accuracy. The size of the ensembles has an effect on the test findings. As a result, ensembles are frequently challenge winners. Each approach has its own unique features, for instance, when it comes to data wrangling and tweaking possibilities. Models may be tweaked to make them fit properly.

3.2

Materials and Methods

3.2.1 Selection of the Causal Parameters The causal parameter was chosen after reviewing the most recent literature in this topic. Table 3.1 shows the rationale for the selection of causative parameters.

3.2.2 Preparation of the Causal Parameters The erosion potentiality of the study region was estimated using a variety of datasets. “Elevation, aspect, slope, drainage density, distance to stream, overland flow, SPI, TRI, TWI, STI” has been estimated from “ALOS PALSAR DEM (digital elevation model)” with 12.5 m spatial resolution. Rainfall raster has been prepared with considering GIS environment and sourced from “IMD (India meteorological department)”. “Soil texture map” has been prepared from “NBSS&LUP” soil texture data. “Geology, geomorphology and distance from lineament” were prepared from “Geological Survey of India” (Fig. 3.1). All the thematic layers are shown in Fig. 3.2.

35

The “drainage density (DD)” has been calculated using the following formula: DD ¼

Pn

i¼1

Si

ð3:1Þ

a

where “i = 1nSi represents total span of all drainages in km and “a” is the total region of drainage catchment in km2” (Arabameri et al. 2020). The overland flow of this region has been calculated using the equation below: Lof ¼

1 2Dd

ð3:2Þ

where “L is the length of the overland flow and Dd is the drainage density” (Horton 1945). The following method was used to calculate the “stream power index (SPI)”: SPI ¼ As × tan r

ð3:3Þ

where “As is the specific catchment area in metres and r is the slope gradient” (Roy et al. 2020). Under the premise that contributing area is directly associated to discharge, the sediment transport index (STI) includes upslope contributing region (As) and slope (B). The index is determined using the following method: (

As STI ¼ ðm þ 1Þ × 22:13

)m

(

B × sin 0:0896

)n

ð3:4Þ where “As is the particular catchment area calculated considering one of the hydrology toolbox’s flow accumulation methods; B is the local slope gradient in degrees; and the generating area exponent, m, is typically set to 0.4 and the slope exponent, n, to 1.4” (Moore and Wilson 1992). The “topographic ruggedness index (TRI)” was calculated using the equations below: TRI ¼

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi j X jðmax2 − min2 Þ

ð3:5Þ

36

3

Morphotectonics Characteristics and Its Control on Soil Erosion

Table 3.1 Explains why the erosion causality factor was chosen Erosion causality factor

Reason behind selection

Elevation

It may be claimed that the quantity and kind of rainfall, temperature, evaporation, and solar radiation intensity are all affected by the elevation above sea level due to changes in soil qualities at various elevation categories. It has a substantial impact on the development and evolution of soils, resulting in variances in soil characteristics across elevation divisions

Aspect

Temperature, moisture and water availability are all controlled by aspect, followed by plant and soil formation (Måren et al. 2015). The influence of slope aspect on soil erosion is extensively reported in the literature and is connected to the varying extents of insolation that occur on sunny and shaded slopes (Nadal‐Romero et al. 2014)

Slope

The steep slope will expand the amount and velocity of runoff, causing erosion to expedite as more particles are carried and dissolved (Siswanto and Sule 2019). The flow will be enhanced by a steeper slope, leading in more force and water to convey the soil. Increased erosion on steep slopes is a result of increased surface runoff, which is also the cause of decreased infiltration (Poesen 1986)

Drainage density

The density of the drainage system has a significant impact on the erosion process. As a result, its management can help to reduce erosion in the area. Drainage density is determined by the type of soil and the volume of water flowing through the channel (Moeini et al. 2015)

Distance to stream

Distance to stream is an important factor which is directly influences the erosion. The formation and development of rills and gullies are directly influenced by stream. More nearest to stream is more prone to erosion and vice versa (Arabameri et al. 2020)

Overland flow

Overland flow mainly happens in small rills or as medium sheets across vast areas; however, bigger gullies may erode, especially along dry valley bottoms where surface runoff accumulates (Bracken 2010). To cause erosion, the rate of rainfall must be high enough to cause runoff, and the shear stress created by flowing water must be greater than the resistance of the surface of the ground (Peng et al. 2014)

Stream power index (SPI)

Stream power index (SPI) is a measure of a stream’s ability to alter the geomorphology of a given region by gully erosion and transportation (Vijith and Dodge-Wan 2019). SPI is a measurement of running water’s erosional capacity concerning the relation across discharge and catchment area. SPI identifies places in the watershed where overland flow has a greater erosive potential (Ettazarini 2021)

Sediment transport index (STI)

Many methodology-related erosion estimates need sediment transport capacity, which is described as the greatest amount of sediment which a particular flow rate can move. When the real sediment load is less or over the transport capacity, it is known as net erosion or deposition, accordingly. As a result, quantifying the link among detaching frequency and sediment loading is critical (Xiao et al. 2017)

Topographical ruggedness index (TRI)

The topographical ruggedness index (TRI) is a metric that measures the difference in elevation among neighbouring cells in a DEM. This raster feature template creates a visual depiction of the TRI using the elevation data (Amatulli et al. 2018)

Topographical wetness index (TWI)

Runoff fields differed depending on the topography. Beven introduced the topographic wetness index (TWI) (Beven and Kirkby 1979), that also reflects both a local slope geometry and site location in the environment, trying to combine information on slope steepness and specific catchment area, followed by a full evaluation of the topographic impact on the creation and modifications of the watershed runoff

Rainfall

While heavy rainstorms can produce spectacular and severe erosion, most soil erosion occurs over time and is frequently difficult to detect without continual monitoring. Rainfall breaks down soil aggregates, producing detachment and movement of soil, whether immediately through raindrop impact or passively by huge bodies of water. To avoid erosion and conserve soil, you must first comprehend the many forms of erosion that might occur (Chakrabortty et al. 2021) (continued)

3.2 Materials and Methods

37

Table 3.1 (continued) Erosion causality factor

Reason behind selection

Soil texture

Storm water infiltration rates are influenced by soil texture, which is an important soil property. The amount of sand, silt and clay in a soil determines its textural class. The pace at which water drains through a saturated soil is determined by soil texture; water travels more readily through sandy soils than it does across clayey soils. Soil texture impacts how much water is accessible to the plant after field capacity is achieved; clay soils have a larger water holding capacity than sandy soils (Sasidharan et al. 2018)

Geology

Weathering occurs when rock is simply broken down by mechanical or chemical methods. Erosion occurs when the broken-down substance is transported by water, wind, or ice in any way (Mitchell and Soga 2005)

Geomorphology

Various geomorphological features have the dissimilar function in regard with erosion potentiality. These information and models are used sparingly in geomorphological study on long-term landscape evolution, possibly because they are regarded relevant to landscape evolution because to the short time period and small geographical scale that they typically cover (Bhattacharyya 2011)

Distance to lineament

Surface lineaments are porous, low-resistance zones that might impact slope stability and soil deterioration (Saha et al. 2019)

Fig. 3.1 Methodology flowchart

38

3

Fig. 3.2 Erosion potentiality causal parameters

Morphotectonics Characteristics and Its Control on Soil Erosion

3.2 Materials and Methods

39

Fig. 3.2 (continued)

where “X represents altitude of every neighbour cell to a definite cell, and max and min are the highest and smallest altitude among different neighbouring cell” (Arabameri et al. 2020). The following method was used to calculate the “topographical wetness index (TWI)”: TWI ¼ Ln

/ tan b þ C

ð3:6Þ

where “/ is the flow accumulation b is the slope and C is the constant (0.01)” (Chowdhuri et al. 2020).

3.2.3 Erosion Potentiality Assessment 3.2.3.1 Application of “Evidential Belief Function (EBF)” for “Erosion Potentiality” “Evidential belief” is a type of belief that is based on evidence. The “Dempster–Shafer theory rule” is based on the “Dempster–Shafer theory” rule, which was first created by Dempster (Dempster 2008), this is calculated using “Bayesian probability theory” (Feizizadeh and Blaschke 2014). The key benefit of this strategy is its adaptability

40

3

Morphotectonics Characteristics and Its Control on Soil Erosion

to many functions from a variety of sources. The key benefit of this strategy is its adaptability to many functions from a variety of sources (Pearl 1990) As a result, this approach is a trustworthy predictor that may be utilized in a variety of disciplines that employ “GIS” (Malpica et al. 2007). “Belief function, disbelief function, uncertainty function and plausibility function” are the four EBF related with this study (Feizizadeh et al. 2014). The lesser and higher bounds of probability are “belief” and “plausibility” (Althuwaynee et al. 2014). In a given distribution, “uncertainty” is defined as the variation between the “belief function” and the “plausibility function” (Tehrany et al. 2017). “Disbelief” is the task which is dependent on the “belief” to be false. Wcf ij D Bel cf ij ¼ Pn i¼1 Wcf ij D

ð3:7Þ

] [ ( ) N Cij \ D =N ðDÞ ( )} ] W cf ij D ¼ [{ ( ) N Cij − N Cij \ D =fN ðT Þ − N ðDÞg

ð3:8Þ W cf ij D Dis cf ij ¼ Pn i¼1 W cf ij D

ð3:9Þ

] ( )} N ðDÞ − N C ij \ D =NðDÞ ( ) ] N ðT Þ − N ðDÞ − N Cij þ NðCij \ DÞ =½N ðT Þ − NðDÞ]

Unc cf ij is the uncertainty value of jth class of the ith soil erosion causative factor and Pls cf ij is the plausibility value of the ith soil erosion causative variables’ jth class”. Belf 1f 2 ¼

Disf 1f 2 ¼

Belf 1 Belf 2 þ Belf 1 Uncf 2 þ Belf 2 Uncf 1 b ð3:13Þ Disf 1 Disf 2 þ Disf 1 Uncf 2 þ Disf 2 Uncf 1 b ð3:14Þ Uncf 1f 2 ¼

Uncf 1 Uncf 2 b

Plsf 1f 2 ¼ Belf 1f 2 þ Uncf 1f 2

ð3:15Þ ð3:16Þ

where “Bel is the belief function for each factor or range, Dis is the disbelief function for each factor or range, Unc is the degree of uncertainty for each factor or range and b =1 − Belf 1 Disf 2 − Disf 1 Belf 2 is a normalizing factor which fulfil that Bel þ Unc þ Dis ¼ 1. EBFs of maps f3 ... fn by applying the equations stated, all functions are integrated simultaneously and in a methodical manner” (Chakrabortty et al. 2020a).

[{

Wcf ij D ¼ [

ð3:10Þ Unc cf ij ¼ 1 − Bel cf ij − Dis cf ij Pls cf ij ¼ Bel cf ij þ Unc cf ij or Pls cf ij ¼ 1 − Dis cf ij

ð3:11Þ ð3:12Þ

where “Bel cf ij is the belief value of jth class of the ith soil erosion causative factor, Wcf ij D is the weight of cf ij that supports the idea that erosion is more active than landscape selfresistance or regolith formation. NðC ij \ DÞ is the number of g, N ðDÞ is the total number of ( ) gully points, N C ij is the respective area in domain, N ðT Þ is the total area, Dis cf ij is the disbelief value of the ith soil erosion causative factor causative factor, Wcf ij D is the weight of cf ij it emphasizes that the landscape’s selfresistance is more prevalent than soil erosion,

3.2.3.2 Application of Logistic Regression for Erosion Potentiality The regression coefficient was chosen for assessing the erosion potentiality in this study. “Logistic regression” is a “multivariate analytic tool” for estimating the presence or absence of an attribute or result depending on the principles of a collection of variables (Lee 2005). It’s a “multivariate regression” technique in which a “dependent variable” is connected to a large number of “independent variables” (Del Hoyo et al. 2011). After turning the variables (the presence of gullies) into a logit variable, the “logistic regression” technique is used to estimate maximum likelihood. The benefits of “logistic regression” include the fact that the components do not have to have a “normal distribution”; they can be continuous, discrete, or a combination of the two (Pradhan and Lee 2010). “Logistic regression”

3.2 Materials and Methods

41

coefficients may be used to predict proportions in each of the “independent variables” in a “multivariate analytic” model. In multi-regression modelling, the components must be “numerical”, and the data must have a normally distributed. The “dependent variable” in this study is the “presence or absence” of gullies, which should be input as “1” or “0”, and the model uses sound to calculate the risk of soil erosion. For assessing the soil erosion potentiality study, the following approach was considered: P¼

1 1 þ e−2

ð3:17Þ

where “P is the expected probability of erosion, which ranges from 0 to 1 on an S-shaped curve, and z is the linear combination, which is expressed in the following equation”. Z ¼ b0 þ b 1 x 1 þ b 2 x 2 þ . . . þ b n x n

ð3:18Þ

where “b0 is the model’s intercept, bi (i = 1, 2, …, n) are the logistic regression model’s slope coefficients, and xi (i = 1, 2, …, n) are the independent variables” (Chowdhuri et al. 2020).

3.2.3.3 Application of Ensemble “Evidential Belief Function-Logistic Regression” for “Erosion Potentiality” An ensemble of “EBF” and “LR” was utilized to assess the erosion potentiality of this location. An ensemble can make better estimates and achieve better outcomes than a single generating model. Using an ensemble reduces the spread or variation of estimates and model performance. On a predictive modelling challenge, ensembles are utilized to progress the prediction performance of a particular predictive model. This is done by adding bias to the model, which decreases the variance component of the prediction error. With the ensemble “EBF-LR” model in mind, the following technique was proposed for evaluating erosion potentiality:

EBF − LR ¼

expðpÞ 1 þ expðpÞ

ð3:19Þ

where “P is the expected probability of erosion that is mentioned earlier” (Chowdhuri et al. 2020). For assessing erosion potentiality with an ensemble of “EBF-LR”, the following technique was examined: P ¼ ½b0 þ ðElevationb ∗ ElevationEBF Þ þ ðAspectb ∗ AspectEBF Þþ ðSlopeb ∗ SlopeEBF Þ þ ðDDb ∗ DDEBF Þ þ ðDtSb ∗ DtSEBF Þ þ ðOFb ∗ OFEBF Þþ ðSPIb ∗ SPIEBF Þ þ ðSTIb ∗ STIEBF Þ þ ðTRIb ∗ TRIEBF Þ þ ðTWIb ∗ TWIEBF Þþ ðSoilb ∗ SoilEBF Þ þ ðRainfallb ∗ rainfallEBF Þ þ ðGeologyb ∗ GeologyEBF Þ þ ðGeomorphologyb ∗ GeomorphologyEBF Þ ð3:20Þ

þ ðDtLb ∗ DtLEBF Þ

where “b0 is the logistic regression model’s intercept value, b is the logistic regression coefficients for each 0 factor and EBF is the raster layers of factors weighted by the belief function value”.

3.2.4 Validation of the Models To confirm the results and to assess the performances of the models, several statistical measures are calculated. These are “sensitivity, specificity, PPV, NPV, F score and AUROC”. The equation for each follows: Sensitivity ¼

TP TP þ FN

ð3:21Þ

Specificity ¼

TN FP þ TN

ð3:22Þ

PPV ¼

TP FP þ TP

ð3:23Þ

42

3

TN TN þ FN P P ð TP þ TNÞ AUC ¼ ðP þ NÞ NPV ¼

F-Score ¼ 2 ∗

Morphotectonics Characteristics and Its Control on Soil Erosion

ð3:24Þ ð3:25Þ

Sensitivity ∗ Specificity ð3:26Þ Sensitivity þ Specificity

where “TP, TN, FP and FN are true positive, true negatively, false positive and false negative, respectively” (Band et al. 2020).

3.2.5 Multi-collinearity Assessment To test for “multi-collinearity”, the “VIF” and “TOL” are two often used statistics (Costache et al. 2020). When the values of one predictor are spatially correlated with the values of another predictor, this is known as “multi-collinearity”. Because of substantial collinearity, if a predictor’s “VIF” is greater than 10 and its “TOL” is less than 0.1, that variable should be eliminated from the modelling process (Costache et al. 2019). The “multi-collinearity” was investigated using the equations below: TOL ¼ 1 − R2j VIF ¼

1 TOL

ð3:27Þ ð3:28Þ

where “R2j indicates the regression value of j on further diverse variables. The threshold values of > 5 for VIF and < 0.1 for TOL indicate whether multi-collinearity problems exist or not” (Arabameri et al. 2021).

3.3

Results

3.3.1 Multi-collinearity Assessment The “multi-collinearity” of the selected parameters has been estimated with considering “TOL” and “VIF”. The ranges of “TOL” and “VIF” in this analysis are 0.33–0.84 and 1.26–3.01, respectively. All the considerable parameters are

in the ranges of acceptable “TOL” and “VIF” limit. So, there is no problem of “multicollinearity” of the selected parameters for estimating the “erosion potentiality” of this region. The value of “TOL” and “VIF” of all selected parameters is shown in Table 3.2.

3.3.2 Erosion Potentiality Assessment The erosion potentiality of the study region has been quantified with considering “EBF”, “LR” and ensemble of “EBF-LR” model. In “EBF” model, the spatial distribution of “very high and high” erosion potential zones is mainly found in western portion and rest of the portion of this region is associated with “moderate to very low” erosion potential zone. In “LR” model, the “very high and high” erosion potential zones mainly found in the western and middle part and rest of the portion is associated with “moderate to very low” erosion potential zones. In ensemble of “EBF-LR” model, the spatial distribution of “very high and high” erosion potential zones is mainly concentrated in the western and middle portion of this region. And rest of the portion of this region is associated with “moderate to very low” erosion potential zones (Fig. 3.3). Some erosion-prone areas of the study region are shown in Fig. 3.4.

3.3.3 Validation of the Models The validation of all models has been done with considering various indices. The values of “sensitivity, specificity, PPV, NPV, F score and AUC” in “EBF” model with considering “training datasets” are 0.97, 0.88, 0.89, 0.96, 0.92 and 0.98, respectively. The values of “sensitivity, specificity, PPV, NPV, F score and AUC” in “EBF” model with considering “validation datasets” are 0.85, 0.75, 0.73, 0.86, 0.80 and 0.82, respectively. The values of “sensitivity, specificity, PPV, NPV, F score and AUC” in “LR” model with considering “training datasets” are 0.96, 0.89, 0.91, 0.96, 0.92 and 0.97, respectively. The values of “sensitivity, specificity,

3.3 Results Table 3.2 Multicollinearity statistics of the selected parameters

43 Factors

Collinearity statistics Tolerance

VIF

Elevation

0.66

1.52

Aspect

0.65

1.55

Slope

0.75

1.33

Drainage density

0.79

1.26

Distance to stream

0.84

1.19

Overland flow

0.53

1.88

Stream power index

0.45

2.23

Sediment transport index

0.79

1.27

Topographic ruggedness index

0.79

1.26

Topographical wetness index

0.65

1.55

Soil texture

0.78

1.29

Rainfall

0.33

3.01

Geology

0.48

2.09

Geomorphology

0.60

1.67

Distance to lineament

0.79

1.26

Fig. 3.3 Erosion potentiality assessment using EBF (a), LR (b) and ensemble of EBF-LR (c)

44

3

Morphotectonics Characteristics and Its Control on Soil Erosion

Fig. 3.4 Some erosion-prone region of the study area Table 3.3 Validation of the models

Models

Stage

Parameters

EBF

Training Validation

0.85

0.75

0.73

0.86

0.80

0.82

LR

Training

0.96

0.89

0.91

0.96

0.92

0.97

Validation

0.92

0.63

0.67

0.91

0.75

0.89

EBF-LR

Training

0.98

0.92

0.93

0.97

0.95

0.99

Validation

0.92

0.75

0.75

0.92

0.83

0.92

Sensitivity

Specificity

PPV

NPV

F score

AUC

0.97

0.88

0.89

0.96

0.92

0.98

PPV, NPV, F score and AUC” in “LR” model with considering “validation datasets” are 0.92, 0.63, 0.67, 0.91, 0.75 and 0.89, respectively (Table 3.3). The values of “sensitivity, specificity, PPV, NPV, F score and AUC” in ensemble of “EBF” and “LR” model with considering

“training datasets” are 0.98, 0.92, 0.93, 0.97, 0.95 and 0.99, respectively. The values of “sensitivity, specificity, PPV, NPV, F score and AUC” in ensemble of “EBF” and “LR” model with considering “validation datasets” are 0.92, 0.75, 0.75, 0.92, 0.83 and 0.92, respectively.

3.4 Discussion

3.4

Discussion

“Soil erosion” is a severe environmental problem in subtropical countries like India. The country loses soil at a rate of 16 t/ha/year on average, and over three times the permitted limit of 4–5 t (Narayana and Babu 1983). Soil erosion is the loss of topsoil, which is mostly caused by rain. Water is 800 times higher than air and weighs roughly the same as loose dirt (Hudson 2015). When a drop of rain falls hard enough on the ground, it dislodges the soil, which is subsequently carried away by the running water (Foster et al. 1985). Wind is the primary cause of erosion in India, although primarily in the Thar Desert (Moharana et al. 2016). Soil structure, plant cover, slope and land and water usage patterns are all elements that influence erosion (García-Ruiz 2010). So, the identification of erosion-prone areas through probability analysis is one of the initial and primary works to identify the vulnerable region and also to implement the appropriate measures for escaping this type of situation. In this work, we incorporate the “EBF”, “LR” and ensemble of “EBF-LR” to estimate the erosion potentiality. Using the natural breaks approach in a GIS setting, the full final result was classed into multiple qualitative groups. “Natural breaks classification” (Jenks) creates classes based on “natural groupings” in the data (North 2009). Class divisions are constructed in such a way that like values are clustered together and class differences are maximized (Brewer and Pickle 2002). The traits are divided into groups, with lines drawn where the data values differ considerably. Natural breaks are data-specific categories that can’t be used to compare several maps with different original dataset (Ahmed 2015). A categorization approach for predicting the likelihood of an event’s success or failure is known as “logistic regression”. Due to limited samples, we applied the “logistic function” which considers the logit distribution instead of “probit model” which flows the “normal distribution”. When the dependent variable is binary,

45

it is utilized (0/1, “True/False”, “Yes/No”) (Birjais et al. 2019). It facilitates in the categorization of data into various categories by looking at the relationship between a set of labelled data (Bruce and Bruce 2017). It establishes a linear connection from the input dataset before introducing nonlinearity in the sort of the sigmoid function (Ranzato et al. 2007). As a result, the ideal combination of “EBF” and “LR” is one that emphasizes the erosion potentiality causal factor as well as subclasses of each component. Here the “optimal capacity” of the “ensemble model” is much higher than the “single alone model” for erosion potentiality assessment. Ensemble techniques are models that aggregate predictions from two or more different models to make a single forecast. This is a “machine learning model” that combines predictions from two or more separate models into a single forecast (Sagi and Rokach 2018). The “integrated models”, also known as team members, can be of the same or distinct types, and they can be trained using the same or different “training data”. The forecasts of “team members” can be combined using statistics like “median or average” or more complex methods that learn when and how to trust each individual. The western part of this region is very much potential for water-induced erosion in comparison to other part of this region. Significant water erosion characterizes the “Chota Nagpur Plateau” in eastern tropical India (Mahala 2018; Hembram et al. 2020). The states of “Odisha, West Bengal and Jharkhand” have deteriorated more than 30% of their geographical area (Mahala 2020). The “Chota Nagpur Plateau” is dominated by the “Gondwana geological formation”. Owing to the granite–gneiss geological structure of the terrain and the sloping topography of the plateau, land production is hampered (Mukherjee et al. 2017). The region’s soil is characterized by less developed. Owing to “gully extension and head retreat, badland topography produces the lateritic capping highlands”. Water deficit also develops as a result of increased aridity and excessive “drainage network”. These elements combine to create the “Chota Nagpur Plateau” a unique and

46

3

Morphotectonics Characteristics and Its Control on Soil Erosion

accurate representation of the tropical plateau, with most of its land deteriorated. A major portion of the lateritic terrain in the “Rahr Bengal” has been eroded by a network of “rills, gullies, and streams” (Jha and Kapat 2009). “Lateritic exposures”, on the other hand, are modest and dispersed in character and are generally susceptible to “rill erosion”, as well as minor or inconsequential gullies, throughout the rest of the study region (Jha and Kapat 2009). “Soil erosion” prevention is necessary in areas where water erosion is a problem. An important preventive strategy is “integrated water resource management”, which combines “soil and water conservation” with appropriate “crop management”. Installation of “check dams along gullies, conservation tillage, contoured bunding, landleveling, and grass plantings along contouring” are all part of the process. These should enhance water accessibility by increasing water percolation through the “soil surface, reducing surface runoff and increasing water access”. Managing “soil erosion” also entails ensuring that the affected areas have adequate plant cover. Controlling “soil erosion” or treating “sediments” at the depositing site are two ways to manage sedimentation. However, the latter is a time-consuming and costly operation.

researchers to suggest the most suitable development strategies. In this work, the “EBF”, “LR” and ensemble of “EBF-LR” have been applied to estimate the “erosion potentiality”. The ensemble “EBF-LR” model has a higher optimum capacity than the individual “EBF” and “LR” models. So it can be say that the ensemble approach can be applied by the future researchers to estimate the erosion potentiality. The western and middle part of this region is mainly associated with higher “erosion potential zone”. So it is directly indicate the impact of “morphotectonic parameters” on erosion potentiality. The “underlying geology”, as well as endogenetic and exogenetic processes occurring in the area, all influences the genesis and evolution of drainage systems. The various drainage network characteristics (drainage morphometry) provide information on the basin’s hydrologic and geological formation features. The “hydrological measurements”, together with “morphometric features”, provide important information about the basin’s rock formations. The western part of this region is already facing the acute problem of soil erosion. This sort of information can assist decision-makers in developing appropriate solutions for sensitive areas.

References 3.5

Conclusion

“Water-induced erosion” is one of the most pressing issues in India, as well as other “subtropical” parts of the world. It disrupts the balances between the rate of regolith formation and “soil erosion”. The overall condition of the environment can be hampered by this event and gradually hampered the ecological balances. Land is one of the most essential natural resources in the worlds that keeps people alive and develop. As a result, it must be managed by the identifying of degraded land surfaces, particularly in developing nations like India where the monsoon dominates the climate. So, identifying the vulnerable regions with considering the “probability modelling” is the initial task of the

Ahmed B (2015) Landslide susceptibility mapping using multi-criteria evaluation techniques in Chittagong Metropolitan Area, Bangladesh. Landslides 12:1077– 1095 Al-Ghamdi AS (2002) Using logistic regression to estimate the influence of accident factors on accident severity. Accid Anal Prev 34:729–741 Althuwaynee OF, Pradhan B, Park H-J, Lee JH (2014) A novel ensemble bivariate statistical evidential belief function with knowledge-based analytical hierarchy process and multivariate statistical logistic regression for landslide susceptibility mapping. CATENA 114:21–36. https://doi.org/10.1016/j.catena.2013.10. 011 Amatulli G, Domisch S, Tuanmu M-N et al (2018) A suite of global, cross-scale topographic variables for environmental and biodiversity modeling. Sci Data 5:1–15 Amozegar M, Khorasani K (2016) An ensemble of dynamic neural network identifiers for fault detection and isolation of gas turbine engines. Neural Netw 76:106–121

References Angima S, Stott D, O’neill M et al (2003) Soil erosion prediction using RUSLE for central Kenyan highland conditions. Agr Ecosyst Environ 97:295–308 Arabameri A, Asadi Nalivan O, Chandra Pal S et al (2020) Novel machine learning approaches for modelling the gully erosion susceptibility. Rem Sens 12:2833. https://doi.org/10.3390/rs12172833 Arabameri A, Pal SC, Rezaie F et al (2021) Comparison of multi-criteria and artificial intelligence models for land-subsidence susceptibility zonation. J Environ Manage 284:112067 Band SS, Janizadeh S, Chandra Pal S et al (2020) Novel ensemble approach of deep learning neural network (DLNN) model and particle swarm optimization (PSO) algorithm for prediction of gully erosion susceptibility. Sensors 20:5609 Başakın EE, Ekmekcioğlu Ö, Çıtakoğlu H, Özger M (2021) A new insight to the wind speed forecasting: robust multi-stage ensemble soft computing approach based on pre-processing uncertainty assessment. Neural Comput Appl 1–30 Beven KJ, Kirkby MJ (1979) A physically based, variable contributing area model of basin hydrology/Un modèle à base physique de zone d’appel variable de l’hydrologie du bassin versant. Hydrol Sci J 24:43–69 Bhattacharyya K (2011) The Lower Damodar River, India: understanding the human role in changing fluvial environment. Springer Science & Business Media Birjais R, Mourya AK, Chauhan R, Kaur H (2019) Prediction and diagnosis of future diabetes risk: a machine learning approach. SN Appl Sci 1:1–8 Bracken LJ (2010) Overland flow and soil erosion. In: Sediment cascades. Wiley, Chichester, pp 181–216 Brewer CA, Pickle L (2002) Evaluation of methods for classifying epidemiological data on choropleth maps in series. Ann Assoc Am Geogr 92:662–681 Bruce P, Bruce A (2017) Practical statistics for data scientists: 50 essential concepts. O’Reilly Media, Inc. Chakrabortty R, Pal SC, Chowdhuri I et al (2020a) Assessing the importance of static and dynamic causative factors on erosion potentiality using SWAT, EBF with uncertainty and plausibility, logistic regression and novel ensemble model in a sub-tropical environment. J Indian Soc Remote Sens 48:765–789. https://doi.org/10.1007/s12524-020-01110-x Chakrabortty R, Pal SC, Sahana M et al (2020b) Soil erosion potential hotspot zone identification using machine learning and statistical approaches in eastern India. Nat Hazards 104:1259–1294 Chakrabortty R, Pal SC, Santosh M et al (2021) Gully erosion and climate induced chemical weathering for vulnerability assessment in sub-tropical environment. Geomorphology 108027 Chowdhuri I, Pal SC, Chakrabortty R (2020) Flood susceptibility mapping by ensemble evidential belief function and binomial logistic regression model on river basin of eastern India. Adv Space Res 65:1466– 1489

47 Costache R, Hong H, Wang Y (2019) Identification of torrential valleys using GIS and a novel hybrid integration of artificial intelligence, machine learning and bivariate statistics. CATENA 183:104179 Costache R, Hong H, Pham QB (2020) Comparative assessment of the flash-flood potential within small mountain catchments using bivariate statistics and their novel hybrid integration with machine learning models. Sci Total Environ 711:134514. https://doi. org/10.1016/j.scitotenv.2019.134514 Del Hoyo LV, Isabel MPM, Vega FJM (2011) Logistic regression models for human-caused wildfire risk estimation: analysing the effect of the spatial accuracy in fire occurrence data. Eur J Forest Res 130:983–996 Dempster AP (2008) Upper and lower probabilities induced by a multivalued mapping. In: Classic works of the Dempster-Shafer theory of belief functions. Springer, Berlin, pp 57–72 Ettazarini S (2021) GIS-based land suitability assessment for check dam site location, using topography and drainage information: a case study from Morocco. Environ Earth Sci 80:1–17 Feizizadeh B, Blaschke T (2014) An uncertainty and sensitivity analysis approach for GIS-based multicriteria landslide susceptibility mapping. Int J Geogr Inf Sci 28:610–638. https://doi.org/10.1080/13658816. 2013.869821 Feizizadeh B, Blaschke T, Nazmfar H (2014) GIS-based ordered weighted averaging and Dempster-Shafer methods for landslide susceptibility mapping in the Urmia Lake Basin, Iran. Int J Digit Earth 7:688–708 Foster G, Young R, Römkens M, Onstad C (1985) Processes of soil erosion by water. In: Soil erosion and crop productivity, pp 137–162 García-Ruiz JM (2010) The effects of land uses on soil erosion in Spain: a review. CATENA 81:1–11 Hembram TK, Paul GC, Saha S (2020) Modelling of gully erosion risk using new ensemble of conditional probability and index of entropy in Jainti River basin of Chotanagpur Plateau Fringe Area, India. Appl Geomat 12:337–360 Horton RE (1945) Erosional development of streams and their drainage basins; hydrophysical approach to quantitative morphology. Geol Soc Am Bull 56:275. https:// doi.org/10.1130/0016-7606(1945)56[275:EDOSAT]2. 0.CO;2 Hudson NW (2015) Soil conservation. Scientific Publishers Jha V, Kapat S (2009) Rill and gully erosion risk of lateritic terrain in South-Western Birbhum District, West Bengal, India. Sociedade & Natureza 21:141– 158 Kocaguneli E, Menzies T, Keung JW (2011) On the value of ensemble effort estimation. IEEE Trans Softw Eng 38:1403–1416 Lee S (2005) Application of logistic regression model and its validation for landslide susceptibility mapping using GIS and remote sensing data. Int J Remote Sens 26:1477–1491

48

3

Morphotectonics Characteristics and Its Control on Soil Erosion

Leopold LB, Wolman MG, Miller JP, Wohl E (2020) Fluvial processes in geomorphology. Courier Dover Publications Lisle I, Rose C, Hogarth W et al (1998) Stochastic sediment transport in soil erosion. J Hydrol 204:217– 230 Mahala A (2018) Soil erosion estimation using RUSLE and GIS techniques—a study of a plateau fringe region of tropical environment. Arab J Geosci 11:1–18 Mahala A (2020) Land degradation processes of Silabati river basin, West Bengal, India: a physical perspective. In: Gully erosion studies from India and surrounding regions. Springer, Berlin, pp 265–278 Malpica JA, Alonso MC, Sanz MA (2007) DempsterShafer theory in geographic information systems: a survey. Expert Syst Appl 32:47–55 Måren IE, Karki S, Prajapati C et al (2015) Facing north or south: does slope aspect impact forest stand characteristics and soil properties in a semiarid transHimalayan valley? J Arid Environ 121:112–123 Mitchell JK, Soga K (2005) Fundamentals of soil behavior. Wiley, New York Moeini A, Zarandi N, Pazira E, Badiollahi Y (2015) The relationship between drainage density and soil erosion rate: a study of five watersheds in Ardebil Province, Iran. WIT Trans Ecol Environ 1:129–138 Moharana P, Santra P, Singh D, et al (2016) ICAR-Central Arid Zone Research Institute, Jodhpur: erosion processes and desertification in the Thar Desert of India Moore ID, Wilson JP (1992) Length-slope factors for the revised universal soil loss equation: simplified method of estimation. J Soil Water Conserv 47:423–428 Morgan RPC (2009) Soil erosion and conservation. Wiley, New York Mukherjee S, Dey A, Sanyal S et al (2017) Petrology and U-Pb geochronology of zircon in a suite of charnockitic gneisses from parts of the Chotanagpur Granite Gneiss Complex (CGGC): evidence for the reworking of a Mesoproterozoic basement during the formation of the Rodinia supercontinent. Geol Soc Lond Spec Publ 457:197–231 Nadal-Romero E, Petrlic K, Verachtert E et al (2014) Effects of slope angle and aspect on plant cover and species richness in a humid Mediterranean badland. Earth Surf Proc Land 39:1705–1716 Narayana DV, Babu R (1983) Estimation of soil erosion in India. J Irrig Drain Eng 109:419–434 North MA (2009) A method for implementing a statistically significant number of data classes in the Jenks algorithm. IEEE, pp 35–38 Osman KT (2014) Soil erosion by water. In: Soil degradation, conservation and remediation. Springer, Berlin, pp 69–101 Pal S (2016) Identification of soil erosion vulnerable areas in Chandrabhaga river basin: a multi-criteria decision approach. Model Earth Syst Environ 2:5. https://doi. org/10.1007/s40808-015-0052-z Pal SC, Chakrabortty R, Roy P et al (2021) Changing climate and land use of 21st century influences soil

erosion in India. Gondwana Res 94:164–185. https:// doi.org/10.1016/j.gr.2021.02.021 Pearl J (1990) Reasoning under uncertainty. Ann Rev Comput Sci 4:37–72 Peng X, Shi D, Jiang D et al (2014) Runoff erosion process on different underlying surfaces from disturbed soils in the Three Gorges Reservoir Area, China. CATENA 123:215–224 Pimentel D (1993) World soil erosion and conservation. Cambridge University Press Poesen J (1986) Surface sealing as influenced by slope angle and position of simulated stones in the top layer of loose sediments. Earth Surf Proc Land 11:1–10 Pradhan B, Lee S (2010) Delineation of landslide hazard areas on Penang Island, Malaysia, by using frequency ratio, logistic regression, and artificial neural network models. Environ Earth Sci 60:1037–1054 Ranzato M, Huang FJ, Boureau Y-L, LeCun Y (2007) Unsupervised learning of invariant feature hierarchies with applications to object recognition. IEEE, pp 1–8 Ravi S, Breshears DD, Huxman TE, D’Odorico P (2010) Land degradation in drylands: interactions among hydrologic–aeolian erosion and vegetation dynamics. Geomorphology 116:236–245 Roy P, Chandra Pal S, Chakrabortty R et al (2020) Threats of climate and land use change on future flood susceptibility. J Clean Prod 272:122757. https://doi. org/10.1016/j.jclepro.2020.122757 Sagi O, Rokach L (2018) Ensemble learning: a survey. Wires Data Min Knowl Discov 8:e1249 Saha S, Gayen A, Pourghasemi HR, Tiefenbacher JP (2019) Identification of soil erosion-susceptible areas using fuzzy logic and analytical hierarchy process modeling in an agricultural watershed of Burdwan district, India. Environ Earth Sci 78:649 Saroha J (2017) Soil erosion: causes, extent, and management in India. Int J Creat Res Thoughts 5:1321–1330 Sasidharan S, Bradford SA, Šimůnek J et al (2018) Evaluating drywells for stormwater management and enhanced aquifer recharge. Adv Water Resour 116:167–177 Shafer G (1976) A mathematical theory of evidence. Princeton University Press Singh M, Hartsch K (2019) Basics of soil erosion. In: Watershed hydrology, management and modeling. CRC Press, Boca Raton, pp 1–61 Siswanto S, Sule M (2019) The impact of slope steepness and land use type on soil properties in Cirandu subsub catchment, Citarum Watershed. IOP Publishing, p 012059 Srinivasarao C, Lal R, Kundu S et al (2014) Soil carbon sequestration in rainfed production systems in the semiarid tropics of India. Sci Total Environ 487:587– 603 Strickland J (2015) Predictive analytics using R. Lulu.com Tehrany MS, Shabani F, Javier DN, Kumar L (2017) Soil erosion susceptibility mapping for current and 2100

References climate conditions using evidential belief function and frequency ratio. Geomat Nat Haz Risk 8:1695–1714 UNEP (1997) World atlas of desertification 2ED Vijith H, Dodge-Wan D (2019) Modelling terrain erosion susceptibility of logged and regenerated forested region in northern Borneo through the analytical

49 hierarchy process (AHP) and GIS techniques. Geoenviron Disasters 6:1–18 Xiao H, Liu G, Liu P et al (2017) Sediment transport capacity of concentrated flows on steep loessial slope with erodible beds. Sci Rep 7:1–9

4

Estimation of Surface Runoff

Abstract

The land degradation from the different forms of erosion due to water action is very much familiar in the subtropical environments. The “Soil Conservation Service-Curve Number (SCS-CN)” is considered as a reliable technique for estimating the potential surface runoff. The application of GIS technique is not only capable to estimate the scenario with less effort and money but also deals with the maximum possible accuracy. In the subtropical environments, the “land use and land cover (LULC)” changes rapidly for the expansion of the agricultural area as well as settlement area from the forest and fallow land. The character and amount of surface runoff are very much dependent regarding the nature of “LULC”. This scenario leads to play an influential function regarding the erosion susceptibility.

The most of the areas of this region are facing the very high surface runoff which leads to large-scale “land degradation” in the different forms of erosion. Due to high “surface runoff”, the “infiltration capacity” also significantly decreases which also decreases the storage of the subsurface water. The weighted curve number of all features is 15,248,412.7, and the feature-wise “weighted CN” ranges from 78,690.44 to 9,977,959. The water discharge of this region is 719,641.88 m3 which indicates the larger erosion potentiality. Keywords

.

4.1 Numerous factors, such as extreme meteorological conditions, particularly drought, contribute to land degradation. Additionally, it is brought on by human activities that deteriorate or impair soil quality and land usability. Production of food, livelihoods, as well as the creation and delivery of other ecosystem products and services are all negatively impacted. Due to increased and coupled pressures from agricultural and livestock production (over-cultivation, over-grazing and forest conversion), urbanization, deforestation and extreme weather events like droughts and coastal surges that salinate land, land degradation has increased during the twentieth and twenty-first centuries.

.

.

Land degradation SCS-CN GIS technique LULC Erosion potentiality

.

Introduction

The soil surface is vulnerable to water erosion, or erosion caused by flowing water, whenever “runoff” happens (Kinnell 2020). “Detachment, transport and deposition” are the three basic steps of erosion (Nearing et al. 2005). Raindrop contact, the dissolution of soil aggregates when wet, and the combing power of runoff water can all cause separation of soil particles from the bulk of the soil bodies at the surface of the ground (Bryan 2000). During a runoff occurrence, the transfer of water and “sediment”, i.e. detached soil particles,

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 S. C. Pal and R. Chakrabortty, Climate Change Impact on Soil Erosion in Sub-tropical Environment, Geography of the Physical Environment, https://doi.org/10.1007/978-3-031-15721-9_4

51

52

across the soil surface generally follows discrete patterns that indicate various forms of “water erosion”. “Sheet erosion” is the first kind, which is defined by water flow and “soil erosion” that occurs in a generally regular pattern over the soil surface (Bartley et al. 2006). This sort of erosion is very sneaky, often staying unreported for years while producing significant “soil erosion”. “Rill erosion” occurs when water and sediment moving around across surface initially congregate in narrow, shallow channels. “Rills” may develop and extend as “water flows downhill, forming gullies, culminating” in “gully erosion” (Zachar 2011). “Sediment deposition” is the third phase of the erosion process. Sedimentation is a significant problem that affects streams, reservoirs and coastal regions (Smith et al. 2001). It’s also one of the main contributors to alluvial soils’ extreme spatial variability. In farming, construction and engineering, sediment management is a critical managerial challenge (Piman and Shrestha 2017). The “process-based runoff erosion model” relies heavily on rill eroding mechanisms. A better knowledge, representation and modelling of the “rill network’s development” can help us better comprehend the slope-scale erosion process to improve the prediction of erosion simulations (Ou et al. 2021). Furthermore, the rill patterns on the degraded slope and the river convergence structure have long been noted for their resemblance. Through laboratory computations, Sofia et al. (2017) demonstrated the resemblance between the river and rill networks. When “Horton’s law” and fractal feature techniques were used, Gessesse et al. (2015) discovered that the small-scale drainage system of degraded slope “runoff” exhibited comparable features to the river system (Fang et al. 2018). If the parallels exist, Wu and Chen (2020) think that the information gained at the river size may be considered to analyse and replicate the rill network mechanism. Surface runoff could damage the Earth’s surface, and the degraded particles can be transported rather far away. “Splash erosion, sheet erosion, rill erosion and gully erosion” are the four basic forms of soil erosion caused by water (Poesen 2018). “Splash erosion” occurs when raindrops collide mechanically with the “soil surface, dislodging soil

4

Estimation of Surface Runoff

particles”, which subsequently travel with the “surface runoff” (Dunkerley 2020). There are different methods have been introduced for estimating the surface runoff which deals with numerous database and empirical evidences. To simplify this matter regarding the estimation of the volume of runoff, the “Soil Conservation Service-Curve Number (SCS-CN)” method is considered which can represent the more accurate result with incorporating simplified equation and table value (Al-Juaidi 2018). Initially, this method is developed by the “US Natural Resources Conservation Service” regarding the estimation of surface runoff with considering the effects of all environmental parameters (Hosseini and Mahjouri 2018). This method is very useful to calculate the volume of runoff in catchment scale in various parts of the world (De Winnaar et al. 2007). There is direct impact of infiltration on the volume of runoff, generally the area of settlement and concrete area are not favourable for large amount of infiltration and the volume of runoff is very high (Mohammad and Adam 2010). But otherwise the barren or uncovered land is very much optimistic on infiltration, and the rate of surface runoff is very marginal. The estimation of CN is based on the infiltration rate, “Hydrologic Soil Group (HSG)”, “land use and land cover (LULC)” and “Antecedent Moisture Content (AMC)” (Deshmukh et al. 2013). HSG is based on the infiltration rate of the soil which is estimated from the percentage of sand, silt, clay and granule. The runoff characteristic of the surface is very much influential with the changes of LULC. Various researches of the different discipline used the “SCS-CN” method as reliable method for estimating the volume of the runoff with considering the CN (Kim and Lee 2008; Adham et al. 2016; Mishra et al. 2018; Ross et al. 2018). Melesse and Graham (2004) used the “SCS-CN” method using RS and GIS data for predicting the storm runoff in Simms Creek watershed and found that there is a very much similarities are associated between the estimated runoff and observed runoff. Rao et al. (2010) used the GIS techniques for surface runoff estimation; they estimated the runoff with considering the “HSG”, “AMC” and the features of “LULC”.

4.2 Methodology

From this analysed, they found that there is an increasing tendency of runoff and vice versa for infiltration rate. Most of the areas of the world are facing similar kind of problem regarding the decreasing tendency of “infiltration”. The changes “LULC” play a crucial role in regard with the increasing of impervious areas like settlement and other concrete area. The conversion of the barren land into agricultural and settlement area is most favourable for increasing the surface runoff. In the developing countries, the growth of the urban area is practised without considering the environmental factors that could be harmful for the environment. The conversion of forest and fallow land into agricultural land, as well as the expansion of impermeable area, are the most important factors in enhancing runoff propensity. Due to variability of rainfall of subtropical region, the number of rainy days significantly decreases but the quantity of rainfall always same of increases which is favourable for large amount of runoff potential. The main objective of this work is to estimate the surface runoff with considering the impact of “LULC” and associated “HSG” for “runoff” potentiality.

4.2

Methodology

The empirical investigation regarding the surface runoff is very time-consuming and expensive. For this purpose, the application of geospatial technologies is very much significant for estimating this scenario in accurate way. The database and its sources are shown in Table 4.1. The overall

53

precise methodology for estimating the weighted curve number is shown in Fig. 4.1. Runoff is an important element because it influences the subsurface water storage in form of infiltration from the surface. The amount of erosion can be influenced by the large surface runoff and which obviously is deposited in the lower catchment in the form of sediment. From the erosion and its associated sedimentation, the rate of agricultural production is declining day by day. So the estimation of runoff is essential for adopting the suitable remedies for reducing the land degradation from soil loss and sedimentation. From this purpose, different kind of model has been incorporated by the different researchers. But the “SCS-CN” method is very reliable for estimating the runoff with incorporating the different environmental factors and summative impact in considering simple equation and curve number. The “United State SCS” was formulated the equation to measure the rate of “surface runoff”: Q¼

ðP - IaÞ2 ðP - Ia þ SÞ

ð4:1Þ

where “Q is the runoff in mm, P is the total amount of precipitation, S is the highest infiltration rate, Ia is the specific parameter which is largely concentrated with soil and LULC, another observation that Ia is the eventual rate of infiltration (Ia ¼ js); j is the abstraction rate in the initial stage (Mishra et al. 2006). The j values range between 0 and 0.3” (Shrestha 2003).

Table 4.1 Database and its reason for selection Data base

Purpose

Reason for selection

Sentinel 2a

Land use and land cover

There is direct impact of LULC on amount of infiltration and surface runoff. With the passage of time, the impervious area and the agricultural areas are increasing from vegetation cover and barren land (Lei and Zhu 2018; Patra et al. 2018)

Soil sample

For preparing the soil texture map, hydrological soil group and infiltration map

The percentage of sand, silt, clay and granule is a dominating function for the amount of infiltration. From the basis of this, the HSG and its associated LULC have a dominating function in regard with the surface runoff (Berhanu et al. 2013; Pal and Chakrabortty 2019)

54

4

Estimation of Surface Runoff

Fig. 4.1 Methodology flowchart

S¼

25;400 - 254 CN

ð4:2Þ

where “S = Soil retention rate, CN = curve number” (Pal and Chakrabortty 2019). Q¼

ðP - 0:2SÞ2 ðP þ 0:8SÞ

ð4:3Þ

where “20% (0.2S) of the potential maximum retention is the initial abstraction before runoff begins and other 80% (0.8S)” (Pal and Chakrabortty 2019). But in Indian condition, the modification of this specific equation has been done for better accuracy and authentic outcomes (Kumar et al. 1991; Mishra et al. 2006): Q¼

ðP - 0:3SÞ2 ðP - 0:7SÞ

ð4:4Þ

where “Q is the actual direct runoff from precipitation, P is the total rainfall, S is the possible maximum retention”.

Weighted CN ¼

CN1 × A1 × CN2 × A2 . . .CNn × AN A1 þ A2 þ . . . An

ð4:5Þ where “CN is the curve number of the individual land use and A1 … An is the percentage (%) area under individual land use” (Pal and Chakrabortty 2019).

4.3

Result

4.3.1 Runoff Estimation The soils in this region are generally seven types, these are fine (1265.74 km2), fine loamy (1618.19 km2), fine loamy–course loamy (282.40 km2), fine–loamy sandy (376.70 km2), fine–fine loamy (608.27 km2), fine–very fine (49.18 km2) and gravelly loam (140.22 km2) (Table 4.2). From the textural analysis of the collected samples, the “NBSS&LUP (ICAR)” methods are adopted for classifying it into different textural classes. Here the impervious urban

4.3 Result Table 4.2 Spatial coverage of soil texture classes

55 Soil texture

Area in km2

Area in percentage (%)

Clay

11,434.68

18.22

Clay loam

0.17

0.00

Loam

23,273.54

37.08

Sandy clay loam

18,613.39

29.66

Sandy loam

9440.45

15.04

Total

62,762.24

100

area has been considered as a separate unit because the inclusion into textural classes is very difficult. Most of the areas including the upper portion are associated with fine loamy soil, and the lower portion is associated with fine, fine loamy–sandy and fine–fine loamy textural soil. Estimating the infiltration rate of respective textural classes, percentage of sand, silt, clay and granule has been considered. The infiltration rate Fig. 4.2 Infiltration rate

of this region has been reclassified into different qualitative classes with considering its cutting threshold, i.e. low, moderate, high and very high (Fig. 4.2). The very high infiltration rate is mainly concentrated into the south-western, middle, north-eastern and south-eastern portions. The high infiltration rate is found in only the middle portion. The moderate infiltration rate is found in the south-western portion, and the low infiltration

56

rate is found in most of the part of this river basin. There is a negative relationship found between infiltration rate and surface runoff. So this scenario indicates the peak surface runoff in most of the portion of the river basin which is very much optimistic regarding the erosion potentiality as well as potential for sedimentation. There is direct impact of hydrologic soil group on surface runoff which is estimated on the basis of the textural characteristics and amount of infiltration. It directly indicates the runoff potentiality of the soil. “HSG A” indicates the smallest runoff potentiality where “HSG” plays the vice versa role for “runoff potentiality” of the soil. “HSG A” and “HSG B” are found in the south-western and middle portions of the river basin. The “HSG C” and “HSG D” are situated in most of the portion of this region (Fig. 4.3). This scenario indicates the

Fig. 4.3 Hydrologic soil group

4

Estimation of Surface Runoff

much surface runoff potentiality and its associated eroding capacity of this river basin. The “LULC” is one of the most important dominating elements which is very much influential regarding the runoff characteristics at both the catchment and regional levels. The amount and intensity of runoff are changeable with the changing character with the changing nature of “LULC”, and the catchment area is also facing the hydrologically active character (Abdulkareem et al. 2018). Here there are several types of LULC features and are associated with this region; these are vegetation (dense and scattered), agricultural land, fallow land, water body, settlement, etc. (Fig. 4.4). Then the overall “LULC map” has been reclassified according to the different HSG (i.e. A, B, C and D) (Fig. 4.5). The areas of agricultural land in “HSG A, B, C

4.3 Result

57

Fig. 4.4 Land use and land cover

and D” are 7748.54, 13,178.78, 6522.77 and 17,794.39 km2, respectively. The areas of fallow land in “HSG A, B, C and D” are 60.61, 1351.06, 45.27 and 262.12 km2, respectively. The areas of settlement in “HSG A, B, C and D” are 342.81, 2255.05, 2189.81 and 3525.03 km2, respectively. The areas of vegetation in “HSG A, B, C and D” are 947.63, 1287.14, 1728.97 and 1016.87 km2, respectively. The areas of water body in “HSG A, B, C and D” are 339.50, 9778.60, 851.48 and 810.01 km2, respectively. The curve numbers of each “LULC” unit and its respective “HSG” area have been considered for estimating the weighted curve number of this

region. The curve number of water body in all “HSG A, B, C and D” is 100. The curve number of settlement in “HSG A, B, C and D” is 77, 86, 91and 92, respectively. The “curve number” of the vegetation cover in “HSG A, B, C and D” is 30, 55, 70 and 77, respectively (Table 4.3). The curve number of the fallow land in “HSG A, B, C and D” is 77, 86, 91 and 94, respectively. The “weighted curve number” of the “agricultural land, fallow land, settlement, vegetation cover and water body” are 9,977,959, 743,920.70, 78,690.44, 149,616.98 and 4,298,225.60, respectively (Table 4.4). The “overall curve number” of all features in this region is

58

Fig. 4.5 Hydrologic soil group-wise land use and land cover

4

Estimation of Surface Runoff

4.3 Result

59

Table 4.3 Distribution of curve number of particular LULC LULC

A soil group

B soil group

C soil group

D soil group

Water body

100

100

100

100

Settlement

77

86

91

92

Vegetation cover

30

55

70

77

Fallow land

77

86

91

94

Agricultural land

95

95

95

95

15,248,412.7. There are some variations associated with feature-wise weighted “CN” of this region with considering HSG and its associated variations of “LULC” (Fig. 4.6). The featureswise weighted “CN” of this region varies from 78,690.44 to 9,977,959. Weighted CN value of Water Body ¼ ð339:50 × 100Þ þ ð9778:60 × 100Þ þ ð851:48 × 100Þ

Weighted CN value of Agricultural Land ¼ ð7748:54 × 95Þ þ ð13;178:78 × 95Þ þ ð6522:77 × 95Þ þ ð17;794:39 × 95Þ ¼ 4;298;225:6

ð4:10Þ The Weighted CN value of all features ¼ 9; 977; 959 þ 743;920:7 þ 78;690:441 þ 149;616:98 þ 4;298;225:6 ¼ 15;248;412:7

ð4:11Þ

þ ð810:01 × 100Þ ¼ 9;977;959

ð4:6Þ Weighted CN value of Settlement ¼ ð342:81 × 77Þ þ ð2255:05 × 86Þ þ ð2189:97 × 91Þ þ ð3525:03 × 92Þ ¼ 743;920:7

The total area of the watershed = 63,476.038 km2 CN ¼

15;248;412:7 ¼ 240:22 63;476:038

ð4:12Þ

ð4:7Þ Weighted CN value of Vegetation Cover ¼ ð947:63 × 30Þ þ ð1287:14 × 55Þ þ ð1728:97 × 70Þ þ ð1016:87 × 77Þ ¼ 78;690:441

ð4:8Þ Weighted CN value of Fallow Land ¼ ð60:61 × 77Þ þ ð1351:06 × 86Þ þ ð45:27 × 91Þ

S¼ Q¼

25;400 - 254 ¼ 148:26 240:22

ð4:13Þ

ð1236 - 0:3 × 148:26Þ2 ¼ 183;264:205 ð1236 - 0:7 × 148:26Þ ð4:14Þ 183;264:205 × 392:68 m3 100 ¼ 719;641:88 m3

Q¼

þ ð262:12 × 94Þ ¼ 149;616:98

ð4:9Þ

ð4:15Þ

Table 4.4 Areal distribution of respective LULC LULC

A soil group

B soil group

C soil group

D soil group

7748.54

13,178.78

6522.77

17,794.39

Fallow land

60.61

1351.06

45.27

262.12

743,920.70

Settlement

342.81

2255.05

2189.81

3525.03

78,690.44

Agricultural land

Vegetation cover

947.63

1287.14

1728.97

1016.87

Water body

339.50

9778.60

851.48

810.01

CN 9,977,959

149,616.98 4,298,225.60

60

4

Estimation of Surface Runoff

Fig. 4.6 Weighted curve number of the study area

The water discharge of the research area is estimated to be around 719,641.88 m3 based on the aforementioned analysis and the final equation of “surface runoff (Q)”.

4.4

Validation

The overall work has been validated with the help of the primary observation obtained from the field study (Fig. 4.7). Some station selected purposively as a training site for collecting the information regarding the surface runoff. The “Area Under Curve (AUC)” of “Receiver Operating Curve (ROC)” have been incorporated within this study for model validation. Different

cutting threshold values of binary classifier have been incorporated within this study. For estimating the unbiased scenario of this model, the “AUC” of “ROC” has been accounted with the help of the following equation: SAUC ¼

n X ðX k þ 1 - X k ÞðSk þ 1 - Sk þ 1 - Sk =2Þ k¼1

ð4:16Þ where “SAUC is the area under curve, X k is the 1-Specificity and Sk is the sensitivity of the ROC” (Chakrabortty et al. 2020). The “AUC” value of “ROC” for this model is 0.724 which indicates the high accuracy and validity of this model (Fig. 4.8).

4.5 Discussion

61

Fig. 4.7 Field information in different LULC categories

4.5

Fig. 4.8 Sensitivity analysis

Discussion

The nature and amount of runoff is one of the most influential elements which are associated with hydrological cycle directly. There is a direct impact is associated between surface runoff and the amount of erosion of a particular region. Changes in land cover are thought to be the most important factor in influencing rainfall–runoff processes since modifications in soil and topography are inconsequential in the short term. In recent years, “LULC” change and any resulting hydrologic consequence have been major areas of investigation (Dutta and Sen 2018). Several studies have looked at the influence of land use change on storm runoff at the event level (Bronstert et al. 2002; Nearing et al. 2005; Chen et al. 2009). In “LULC” scenario-based research,

62

event-scale hydrological models have been used to explore the hydrological response of catchments in past and current “LULC” states, as well as extreme “LULC change” scenarios (Sanyal et al. 2014). The soil surface is vulnerable to “water erosion”, or “erosion” caused by flowing water, when “runoff” happens (Pierson et al. 2007). “Separation, transport and deposition” are the three basic steps of erosion. Raindrop contact, the dissolution of soil aggregates when wet, and the combing power of surface runoff can all cause detachment of soil particles from the majority of the soil mass at the surface of the ground (Bryan 2000). During a runoff event, the transfer of water and sediment, i.e. separated soil particles, from across soil surface generally follows discrete patterns that indicate various forms of water erosion (Wilson et al. 2018). “Sheet erosion” is the first kind, which is defined by water flow and “soil erosion” that occurs in a generally regular pattern over the soil surface (Vrieling et al. 2008). This sort of erosion is very sneaky, often staying unreported for years while producing significant soil damage. “Rill erosion” occurs when water and sediment moving throughout the surface initially concentrate in narrow, shallow channels. Rills may expand and broaden as water flows downhill, forming gullies, and leading in “gully erosion”. Rills differ from gullies in that rills may frequently be cleared with tillage and spanned with equipment, but gullies cannot be spanned with field equipment (Foster et al. 1985; Haile and Fetene 2012). The distribution of streamflow from a single land use is not proportionate to its area and is highly dependent on where that land use is located inside the basin (Warburton et al. 2012). The geographical distribution of diverse land uses prevalent in the whole catchment, as well as the trying to balance or neutralizing impact of those land uses, impacted the streamflow pattern at the basin outflow, according to this analysis. For example, if urbanization occurs in the upper subcatchments, the flood peak downstream rises significantly higher (Amini et al. 2011). Human action, in the form of improved channel management in urban areas, has been discovered to

4

Estimation of Surface Runoff

act as a balancing factor in reducing excess surface runoff generated by expanding urban areas or decreasing forest cover (Fox et al. 2012). The continual rainfall during the first rainfall event led to the increased initial roughness of the soil surface to gradually reduce (Römkens et al. 2002). The runoff pathway’s changing velocity was also increased. The rise in the slope of the low-level “runoff channel” was perhaps the most noticeable among these (Rivett et al. 2016). Because of the various growth levels and durability of crusts, as well as the creation and development of rills, the gradients of the three soils showed some differentiation during the 2nd and 3rd rainfall episodes (Puigdefábregas et al. 1999). The ability to accurately characterize the creation and evolution of the rill network is critical for fully comprehending the slope water erosion mechanism and simulating small-scale “runoff” “configurations” (Bryan 2000). The inclination produced by a given “runoff” and the vertical downward slope is equivalent to the value of positional accuracy, and its modification can describe the network configuration’s development trend (Zhang et al. 2021). Typically, runoff erosion models are geographically limited. The accuracy of the model might be improved by adjusting the parameters and adding new ones. Surface roughness is a changeable characteristic that influences “runoff” and “erosion” activities on the soil surface (Zhao et al. 2014). It has a significant impact on the “genesis, properties, infiltration, erosion and sediment formation processes” of “surface runoff” (Modeste et al. 2018). Geohydrological characteristics of erosive discharge must be conceptualized, and this has been necessary in contemporary soil erosion management. “Hydrological models” can be used to regulate and minimize the adverse consequences of surface runoff and “soil erosion” in watersheds. Furthermore, they may be used to model numerous mixture of the two land and water control situations in a watershed, making them valuable for comparing alternative choices and determining which “Best Management Practices (BMPs)” can be used to cut emissions from “point and non-point sources” (Shrestha et al. 2006).

References

4.6

Conclusion

The surface runoff is an important influential element which can influence on the erosion potentiality. The “LULC” changes from time to time as an anthropogenic impact on physical environment. The growth of the population is rapidly increasing in this region because this region is one of the fertile parts of the eastern India. Most of the people are dependent on the agricultural system and try to increase the possible maximum production by increasing the agricultural land. The increasing tendency of agricultural land is mainly done after converting the fallow into cultivated land. This scenario plays a leading role for increasing the amount of erosion of the topsoil in the form different erosion processes, i.e. rill, gully, etc. The water storage capacity also decreases which is the most abundant element of the life. So the estimation of the surface runoff with considering impact of “LULC” is very much essential for indemnify the probable vulnerable areas for erosion susceptibility. The application of “Geospatial technologies” is very much realistic regarding this matter and which can give the truthful result with less effort and time. For this purpose, the application of this type of techniques is used rapidly in various allied disciplines as a reliable technique. The most of the areas of this river basin are associated with very high “surface runoff”, and there is a direct impact of storm rainfall event in “peak monsoon” season. So “erosion” due to water action in various forms of “erosion” is common in this region. The major surface runoff-prone areas are mainly concentrated in the major channel and it’s adjoining areas. These regions have to focus special attention in regard in keeping with the view of local environment. The local stakeholders can be benefitted in acquiring knowledge about the amount and nature of surface runoff and its associated erosion. In this perspective, the rate of “sedimentation” can be reduced with considering appropriate measures. The “sediment trapping” and related measures can be one of the most dependable and cost-consuming measures to reduce the transportation and deposition of

63

sediment in lower catchment. The main limitation of this work is the inadequacy of database in related information. The main task of the future researcher is to estimate the “surface runoff” and directed this information in policy and managerial perspective. So the “planers, hydrologist, pedogeomorphologist” have to give the proper attention in most of the areas of this region for incorporating suitable remedies from escape this type of vulnerable situation.

References Abdulkareem JH, Sulaiman WNA, Pradhan B, Jamil NR (2018) Long-term hydrologic impact assessment of non-point source pollution measured through land use/land cover (LULC) changes in a tropical complex catchment. Earth Syst Environ 2:67–84. https://doi. org/10.1007/s41748-018-0042-1 Adham A, Riksen M, Ouessar M, Ritsema CJ (2016) A Methodology to assess and evaluate rainwater harvesting techniques in (semi-) arid regions. Water 8:198. https://doi.org/10.3390/w8050198 Al-Juaidi AE (2018) A simplified GIS-based SCS-CN method for the assessment of land-use change on runoff. Arab J Geosci 11:1–11 Amini A, Ali TM, Ghazali AHB et al (2011) Impacts of land-use change on streamflows in the Damansara Watershed, Malaysia. Arab J Sci Eng 36:713–720 Bartley R, Roth CH, Ludwig J et al (2006) Runoff and erosion from Australia’s tropical semi-arid rangelands: influence of ground cover for differing space and time scales. Hydrol Proces Int J 20:3317–3333 Berhanu B, Melesse AM, Seleshi Y (2013) GIS-based hydrological zones and soil geo-database of Ethiopia. CATENA 104:21–31. https://doi.org/10.1016/j.catena. 2012.12.007 Bronstert A, Niehoff D, Bürger G (2002) Effects of climate and land-use change on storm runoff generation: present knowledge and modelling capabilities. Hydrol Process 16:509–529 Bryan RB (2000) Soil erodibility and processes of water erosion on hillslope. Geomorphology 32:385–415 Chakrabortty R, Pal SC, Sahana M et al (2020) Soil erosion potential hotspot zone identification using machine learning and statistical approaches in eastern India. Nat Hazards 1–36 Chen Y, Xu Y, Yin Y (2009) Impacts of land use change scenarios on storm-runoff generation in Xitiaoxi basin, China. Quatern Int 208:121–128 De Winnaar G, Jewitt G, Horan M (2007) A GIS-based approach for identifying potential runoff harvesting sites in the Thukela River basin, South Africa. Phys Chem Earth Parts a/b/c 32:1058–1067

64 Deshmukh DS, Chaube UC, Ekube Hailu A et al (2013) Estimation and comparision of curve numbers based on dynamic land use land cover change, observed rainfall-runoff data and land slope. J Hydrol 492:89– 101. https://doi.org/10.1016/j.jhydrol.2013.04.001 Dunkerley D (2020) A review of the effects of throughfall and stemflow on soil properties and soil erosion. In: Precipitation partitioning by vegetation, pp 183–214 Dutta S, Sen D (2018) Application of SWAT model for predicting soil erosion and sediment yield. Sustain Water Resour Manage 4:447–468 Fang Y, Ceola S, Paik K et al (2018) Globally universal fractal pattern of human settlements in river networks. Earth’s Fut 6:1134–1145 Foster G, Young R, Römkens M, Onstad C (1985) Processes of soil erosion by water. In: Soil erosion and crop productivity, pp 137–162 Fox DM, Witz E, Blanc V et al (2012) A case study of land cover change (1950–2003) and runoff in a Mediterranean catchment. Appl Geogr 32:810–821 Gessesse GD, Mansberger R, Klik A (2015) Assessment of rill erosion development during erosive storms at Angereb watershed, Lake Tana sub-basin in Ethiopia. J Mt Sci 12:49–59 Haile G, Fetene M (2012) Assessment of soil erosion hazard in Kilie catchment, East Shoa, Ethiopia. Land Degrad Dev 23:293–306 Hosseini SM, Mahjouri N (2018) Sensitivity and fuzzy uncertainty analyses in the determination of SCS-CN parameters from rainfall–runoff data. Hydrol Sci J 63:457–473 Kim NW, Lee J (2008) Temporally weighted average curve number method for daily runoff simulation. Hydrol Process 22:4936–4948. https://doi.org/10. 1002/hyp.7116 Kinnell P (2020) The influence of time and other factors on soil loss produced by rain-impacted flow under artificial rainfall. J Hydrol 587:125004 Kumar P, Tiwart KN, Pal DK (1991) Establishing SCS runoff curve number from IRS digital data base. J Indian Soc Rem Sens 19:245–252. https://doi.org/10. 1007/BF03023971 Lei C, Zhu L (2018) Spatio-temporal variability of land use/land cover change (LULCC) within the Huron River: effects on stream flows. Clim Risk Manag 19:35–47. https://doi.org/10.1016/j.crm.2017.09.002 Melesse AM, Graham WD (2004) Storm runoff prediction based on a spatially distributed travel time method utilizing remote sensing and Gis1. JAWRA J Am Water Resour Assoc 40:863–879. https://doi.org/10. 1111/j.1752-1688.2004.tb01051.x Mishra SK, Sahu RK, Eldho TI, Jain MK (2006) An improved IaS relation incorporating antecedent moisture in SCS-CN methodology. Water Resour Manage https://doi.org/10.1007/s11269-00520:643–660. 9000-4 Mishra SK, Singh VP, Singh PK (2018) Revisiting the soil conservation service curve number method. In: Singh VP, Yadav S, Yadava RN (eds) Hydrologic modeling. Springer, Singapore, pp 667–693

4

Estimation of Surface Runoff

Modeste M, Abdellatif K, Nadia M, Mohamed S (2018) Effects of land use and cover type on the risks of runoff and water erosion: infiltration tests in the Ourika watershed (High Atlas, Morocco). EuroMediterr J Environ Integr 3:1–12 Mohammad AG, Adam MA (2010) The impact of vegetative cover type on runoff and soil erosion under different land uses. CATENA 81:97–103 Nearing M, Jetten V, Baffaut C et al (2005) Modeling response of soil erosion and runoff to changes in precipitation and cover. CATENA 61:131–154 Ou X, Hu Y, Li X et al (2021) Advancements and challenges in rill formation, morphology, measurement and modeling. CATENA 196:104932 Pal SC, Chakrabortty R (2019) Simulating the impact of climate change on soil erosion in sub-tropical monsoon dominated watershed based on RUSLE, SCS runoff and MIROC5 climatic model. Adv Space Res 64:352–377 Patra S, Sahoo S, Mishra P, Mahapatra SC (2018) Impacts of urbanization on land use/cover changes and its probable implications on local climate and groundwater level. J Urb Manage 7:70–84. https://doi.org/10. 1016/j.jum.2018.04.006 Pierson FB, Bates JD, Svejcar TJ, Hardegree SP (2007) Runoff and erosion after cutting western juniper. Rangel Ecol Manage 60:285–292 Piman T, Shrestha M (2017) Case study on sediment in the Mekong River Basin: current state and future trends. Sotckholm Environment Institute, Stockholm Poesen J (2018) Soil erosion in the anthropocene: research needs. Earth Surf Proc Land 43:64–84 Puigdefábregas J, Sole A, Gutierrez L et al (1999) Scales and processes of water and sediment redistribution in drylands: results from the Rambla Honda field site in Southeast Spain. Earth Sci Rev 48:39–70 Rao KN, Narendra K, Latha PS (2010) An integrated study of geospatial information technologies for surface runoff estimation in an agricultural watershed, India. J Indian Soc Remote Sens 38:255–267. https:// doi.org/10.1007/s12524-010-0032-8 Rivett MO, Cuthbert MO, Gamble R et al (2016) Highway deicing salt dynamic runoff to surface water and subsequent infiltration to groundwater during severe UK winters. Sci Total Environ 565:324–338 Römkens MJ, Helming K, Prasad S (2002) Soil erosion under different rainfall intensities, surface roughness, and soil water regimes. CATENA 46:103–123 Ross CW, Prihodko L, Anchang J et al (2018) HYSOGs250m, global gridded hydrologic soil groups for curve-number-based runoff modeling. Sci Data 5:180091. https://doi.org/10.1038/sdata.2018.91 Sanyal J, Densmore AL, Carbonneau P (2014) Analysing the effect of land-use/cover changes at sub-catchment levels on downstream flood peaks: a semi-distributed modelling approach with sparse data. CATENA 118:28–40 Shrestha MN (2003) Spatially distributed hydrological modelling considering land-use changes using remote sensing and GIS. Citeseer, pp 1–8

References Shrestha S, Babel MS, Das Gupta A, Kazama F (2006) Evaluation of annualized agricultural nonpoint source model for a watershed in the Siwalik Hills of Nepal. Environ Model Softw 21:961–975. https://doi.org/10. 1016/j.envsoft.2005.04.007 Smith S, Renwick W, Buddemeier R, Crossland C (2001) Budgets of soil erosion and deposition for sediments and sedimentary organic carbon across the conterminous United States. Global Biogeochem Cycles 15:697–707 Sofia G, Di Stefano C, Ferro V, Tarolli P (2017) Morphological similarity of channels: from linear erosional features (Rill, Gully) to Alpine rivers. Land Degrad Dev 28:1717–1728 Vrieling A, de Jong SM, Sterk G, Rodrigues SC (2008) Timing of erosion and satellite data: a multi-resolution approach to soil erosion risk mapping. Int J Appl Earth Obs Geoinf 10:267–281

65 Warburton ML, Schulze RE, Jewitt GP (2012) Hydrological impacts of land use change in three diverse South African catchments. J Hydrol 414:118–135 Wilson GV, Wells R, Kuhnle R et al (2018) Sediment detachment and transport processes associated with internal erosion of soil pipes. Earth Surf Proc Land 43:45–63 Wu S, Chen L (2020) Modeling soil erosion with evolving rills on hillslopes. Water Resour Res 56. https://doi.org/10.1029/e2020WR027768 Zachar D (2011) Soil erosion. Elsevier, Amsterdam Zhang L, Liu X, Song Y et al (2021) Characterization of surface runoff pathways and erosion using hydrological attributes under simulated rainfall. Front Earth Sci 636 Zhao L, Liang X, Wu F (2014) Soil surface roughness change and its effect on runoff and erosion on the Loess Plateau of China. J Arid Land 6:400–409

5

Soil Loss Estimation Using Different Empirical and Semi-empirical Models

Abstract

“Soil erosion” caused by water is one of the most severe aspects of land degradation processes. This type of problem is more acute in the “subtropical environment” than in other parts of the world. The erosion of the topsoil is influenced by rainfall variability and anthropogenic activities. Soil erosion is presently recognized as a separate field for large-scale land degradation caused by soil loss across the world. Many academics from several disciplines have emphasized the need of using multiple approaches to assess the quantity of soil erosion. The empirical model can estimate the amount of soil loss from ground-based observations of the field site, but it takes a long time and is expensive. Apart from this most of the areas of the world are ungauged in nature facing the challenges for incorporating the necessary relevant data. For this purpose, the use of this model in a “GIS” integrated platform is rapidly increasing. This integrated model can produce nearly identical results across a wide range of variables, making it a reliable predictor. In recent times, the use of multiple predictive models in identical work has increased significantly, with the aim of determining which one is more reliable in terms of incorporating primary data. In this study, the application of “Universal Soil Loss Equation (USLE)”, “Revised Universal Soil Loss Equation (RUSLE)” and “Modified

Universal Soil Loss Equation (MUSLE)” has been done to estimate the annul “soil loss” in Bengal Basin. From the analysis, the “RUSLE” gives the better outcome in comparison with “USLE” and “MUSLE”. Keywords

. .

.

Soil erosion Variability of rainfall Land degradation USLE RUSLE MUSLE

5.1

.

.

Introduction

“Soil erosion” due to water action is considered as a global problem, and this scenario is associated with various environmental problems (Peterson et al. 2003). “Soil erosion” is a complicated process that is influenced by a variety of factors including soil characteristics, ground slope, vegetation and the amount and intensity of rainfall (Cantón et al. 2011). Changes in “land use” are commonly described as having the ability to profoundly accelerate “soil erosion” because it has long been understood that excessive soil formation causes erosion, which ultimately leads to a reduction in “agricultural production” (Arnalds and Barkarson 2003). While faster erosion causes “soil fertility” to decrease, soil quality is a result of agricultural techniques and site factors such “soil type, nutrient availability and organic matter content”

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 S. C. Pal and R. Chakrabortty, Climate Change Impact on Soil Erosion in Sub-tropical Environment, Geography of the Physical Environment, https://doi.org/10.1007/978-3-031-15721-9_5

67

68

(Mairura et al. 2007). “Soil erosion” is a continuing mechanism in which soil particles are detached and removed by the influence of water or wind, causing the soil to degrade (Blanco and Lal 2008). The condition might worsen to the point where the land can no anymore be farmed and must be deserted. Due to incompetence of land and natural resources, several agricultural societies have fallen, and the experience of such societies serves as a valuable warning to safeguard our resources. Erosion is a critical issue for both productive agricultural land and worries about surface water (Pimentel et al. 1993). To enhance water and soil health, sediment prevention should be an inherent aspect of any soil management system (National Research Council 1993). Soil erosion is a complex process by which the particles of the surface are detached and transported by the various agents and accumulated in places as sediment (Toy et al. 2002). Soil erosion has a considerable influence on water quality, especially as soil “surface runoff” (El Kateb et al. 2013). “Soil erosion” and sediment generation are inextricably linked. As a result, stabilizing the sediment supply through reducing erosion is the most efficient strategy to reduce “sediment” generation (Zuazo and Pleguezuelo 2009). So it is clear that this is the combination of three several processes like detachment of the particles from surface, transportation of these materials by various active and deposition in a low lying place or reservoir area (Leopold et al. 2020). However, these three processes are all active at the same time, eroding the surface materials and producing “sedimentation”. Soil erosion is caused several problems, including deterioration of the quality of the water, reduce the fertility of the topsoil and affecting the local ecosystem services in a negative manner (Butzer 2005; Keesstra et al. 2018). “Water erosion” is caused by the hydrologic mechanisms of rainfall and runoff (RodríguezCaballero et al. 2013). Erosion and sediment movement can be influenced by the amount and pace of “surface runoff” (Nunes et al. 2006). As a result, soil conservation methods are critical for preventing “soil erosion” (Mekonnen et al.

5

Soil Loss Estimation Using Different Empirical …

2015). Soil erosion can be reduced by increasing soil infiltration rates, which results in much less surface runoff (Mohammad and Adam 2010). Preventing soil erosion can be accomplished by a variety of “agricultural, sociocultural or infrastructural” approaches. Physical alterations in the shape and elevation of the ground are part of structural activities. These techniques do not have to be mutually restrictive. When the topography is extremely complicated, some problems may necessitate both managerial and construction adjustments. In some cases, erosion management can be implemented by applying a single activity, such as the creation of vegetated waterways, where erosion is minimal (Vrieling et al. 2008). The subtropical environments like monsoon-dominated region are facing this problem in more acute level (Pal and Chakrabortty 2019). About 57% of India’s total land mass is degraded in some way including surface erosion, wind erosion, chemical and mechanical deterioration (Jena et al. 2018). According to the “National Bureau of Soil Survey and Land Use Planning (NBSS&LUP)”, a total of 120.72 Mha of land is degraded, including areas impacted by water erosion. 82.5 million ha, wind erosion 12.4 million ha, salinity/alkalinity 6.7 million ha, 17.9 million ha of soil acidity and 1.0 million ha are affected by additional difficult issues (Maji et al. 2010). In this region, short span of wet season and storm rainfall with high kinetic energy play a crucial role for eroding the surface material in drastic way (Jat et al. 2018). There is a direct impact off surface runoff for soil erosion susceptibility, larger amount of runoff is very much optimistic on erosion of the topsoil (Chen et al. 2019). However, the amount and direction of the slope gradient is also governed the amount and intensity of the runoff (Morbidelli et al. 2018). So quantify the soil loss and identify the vulnerable areas is necessary for adopting the suitable strategies in this region. But most of the empirical model is associated with various kinds of primary information related to soil erosion which is very much time-dependent and associated with huge expensive. In most of the countries of subtropical region, this type of

5.3 Methodology

information is unavailable due to infrastructural problem as well as less financial support regarding this matter. For this reason, the application of “USLE, RUSLE, MUSLE”, etc., is going on in various disciplines as dependable and reliable method for estimating the average annual soil loss in catchment scale (Pal and Shit 2017). The use of “RS and GIS” technology is increasing in alarming rate for estimating the “soil loss” with less effort and maximum possible accuracy (Prasannakumar et al. 2012). “Soil loss” in India considers as a great environmental damage which causes the reduction of the valuable components from the soil, siltation of the lower catchment as well as “reservoir”, etc. In India, the due to soil loss, the agricultural productivity is declining day by day (Sarvade et al. 2019). The government and local stakeholders have implemented several steps to mitigate this situation, including different structural measures, yet it keeps rising on a regular basis (Keesstra et al. 2018). In the study, in recent time, there is large-scale land degradation process due to soil are going on in various forms of erosion. The creation of spatial database in soil erosion of this region is necessary for identifying the very high erosion-prone areas. The main objective of this research is to estimate the quantity of soil erosion with incorporating the multiple empirical or semi-empirical models and suggest that model is best fitted with the help of primary observed information.

5.2

Database

For determining average annual soil loss and modelling it using primary observed data, the following database was used (Table 5.1).

5.3

Methodology

For assessing and quantifying the “soil erosion” and its vulnerable areas, the several empirical and semi-empirical methods are taken into consideration the detailed and precise methodology is as in Fig. 5.1.

69

5.3.1 Factor for USLE, RUSLE and MUSLE for Estimation of Soil Loss 5.3.1.1 Rainfall and Runoff Erosivity Factor In a given “storm rainfall event”, the volume and intensity of rainfall are linked to soil erosion. This element is incorporated in empirical models like “USLE and RUSLE” as the effect or strength of raindrops and their related runoff in a single storm rainfall event, and it is represented as “MJ mm ha−1 h−1 year−1”. The “rainfall and runoff erosivity factors” were calculated using primary observed data at each location in this investigation. R factor was determined using the algorithms by Zhang and Fu (2003), Yin et al. (2013) updated this model, which was ultimately employed by Liu et al. (2013). R¼

24 X

Rhmk

ð5:1Þ

k ¼1

Rhmk ¼

n X m ( ) 1X a . P1:7265 i;j;k n i¼1 j¼0

WRhmk ¼

Rhmk R

ð5:2Þ ð5:3Þ

Here, “R is the average annual rainfall erosivity (MJ mm ha−1 h−1), Rhmk is the rainfall erosivity of the kth half-month (MJ mm ha−1 h−1), pi,j,k is the average daily erosive rainfall occurring on the jth day, kth half-month and ith year. WRhmk is the proportion of the rainfall erosivity of the kth half-month to the average annual rainfall erosivity”. In this region, the “R factor” varies between 374.60 and 651.00 MJ mm ha−1 h−1 year−1. The southern and south-eastern part of this basin is associated with very high “rainfall and runoff erosivity factor” values, and rest of the portion of this region is associated with moderate to low R factor (Fig. 5.2).

5.3.1.2 Soil Erodibility Factor This component reveals the erosion’s intrinsic capability, which is linked to the soil’s physical

70

5

Soil Loss Estimation Using Different Empirical …

Table 5.1 Database Database

Source

Purpose

ALOS PALSER DEM

ALOS PALSAR DEM (Japan Aerospace Exploration Agency)

For estimating the slope length, slope steepness, slope length and steepness factor

Sentinel 2A

European Space Agency

For estimating the cover and management factor

Storm rainfall data

Primary observed information

For estimating the rainfall and runoff erosivity factor

Soil texture

Texture analysis from the collected samples

For estimating the soil erodibility factor

Soil chemical properties

Analysis from the collected samples

For estimating the soil erodibility factor

Support practice information

Primary information regarding the support practices that are adopted by the local stakeholders

For estimating the support practice factor

Amount of soil loss

Empirical observation regarding the actual amount of soil erosion as a validation purpose

As a validation of the output models

and chemical properties. For this study, a different sample has been considered for estimating various physical and chemical properties. The “K factor” is calculated in this study using the physical and chemical parameters of the soil gathered from the samples. The proportion of sand, silt, clay and organic content raster has been calculated using the “GIS” platform and the obtained sample. The “organic matter” of the soil was derived with considering the organic carbon. The percentage of sand, silt and clay was derived from the textural classification of the samples in pedology laboratory. The nature permeability of the soil was obtained from the textural characteristics. Then the overall information is integrated into “GIS” platform for estimating the “K factor” with consideration. The ranges of K factor in this region are ranged from 0.24 to 0.42. The western and south-western portion is associated with high K factor, and rest of the portion of this region is associated with moderate to low K values (Fig. 5.3). The details method for estimating the “K factor” are shown in Table 5.2. The following equation has been considered for estimating the “K factor” with considering different parameters:

( ) 1-Sil K ¼ 0:0137 × 0:2 þ 0:3 × e½-0:0256×San×ð 100 Þ] ( )0:3 Sil × Cla þ Sil ] [ 0:25 × TOC × 1TOC þ eð3:72-2:95×TOCÞ ] [ 0:7 × SN1 ð5:4Þ × 1SN1 þ eð22:9×SN1 -5:51Þ where “K is the soil erodibility, San is the percentage of sand, Sil is the percentage of silt, Cla is the percentage of clay and SN1 is the 1 − San/100” (Pal et al. 2021).

5.3.1.3 Slope Length The effect of slope length regarding the soil erosion is essential for determining the importance of L factor on erosion susceptibility assessment. This factor was estimated from the “digital elevation model (DEM)”. The amount of “soil loss” increases with rising tendency of “slope length”. The “slope length” values are accounted from “GIS” platform with considering pixel size of the raster, and the m values are obtained with considering the amount of slope of the respective

5.3 Methodology

Fig. 5.1 Methodology flowchart

71

72

5

Soil Loss Estimation Using Different Empirical …

Fig. 5.2 R factor

pixel. After filling the gap of the “DEM”, the “flow accumulation” raster has been created for estimating the slope length (McCool et al. 1989): ( L¼ m¼ F¼ L¼

k 22:13

)m

F 1 þ Fi

sin b=0:0896 3ðsin bÞ0:8 þ 0:56i

ð5:5Þ ð5:6Þ ð5:7Þ

ðflowacc þ 625Þðm þ 1Þ -flowaccðm þ 1Þ ð5:8Þ 25ðm þ 1Þ × 2:13m

where “L is the slope length factor, k is the slope length, m is the eroding potentiality in regard with the amount of slope in percentage, F is the

ratio of rill and inter rill erosion, b is the slope angle (in degree) in GIS environment and flowacc is the flow accumulation”.

5.3.1.4 Steepness Factor The effect of slope steepness is influential for soil loss and its related phenomenon like sedimentation. The steepness of the slope was computed from the slope grid that was obtained from the “DEM” from spatial analyst tool in “GIS”. The gradient can be play a vital role for increasing the erodibility function, though there are direct influences of vegetation roughness in erosion susceptibility. The angle of the slope in regard with the horizontal distance and vertical differences and this approach adopted from Renard et al. (1997):

5.3 Methodology

73

Fig. 5.3 Soil erodibility factor

S ¼ Conððtanðslope × 0:01745Þ\0:09Þ; ð10:8 × sinðslope × 0:01745Þ þ 0:3Þ; ð16:8 × sinðslope × 0:01745Þ - 0:5ÞÞ ð5:9Þ

5.3.1.5 Slope Length and Steepness Factor There is direct impact of “slope length and steepness” factor on the nature of runoff which is very much influential regarding the loss of surface soil by water action (Amsalu and Mengaw 2014). This factor is the combination of “slope length and steepness” factor, which is integrated into “GIS” platform. In “RUSLE”, this scenario is incorporated for getting the importance of slope on soil loss in a realistic way.

LS ¼ L × S

ð5:10Þ

where “L is the slope length, S is the slope steepness and LS is the integration of the slope length and steepness”. The LS factor of this region ranges from 0.00 to 3.21. The value of “LS factor” in western part is high in comparison with other portion of this region. In south-eastern portion, the value is considerably low in nature (Fig. 5.4).

5.3.1.6 Cover and Management Factor “C factor” is considered as an important factor which indicate the effect of specific cover and its relevant management strategies on soil erosion (Renard et al. 1997; Schmidt et al. 2018). The rainfall pattern plays a crucial role on crop system and its related management strategies. The main problem for estimating the “C factor” is the

74

5

Soil Loss Estimation Using Different Empirical …

Table 5.2 Parameters for estimating K factor Mean “K”-based % of organic material content

Textural class

Soil composition Sand

Silt

Clay

Clay

0–45

0–40

40–100

Unknown

< 2%

> 2%

0.22

0.24

0.21

Sandy clay

45–65

0–20

35–55

0.2

0.2

0.2

Silty clay

40–60

40–60

40–60

0.26

0.27

0.26

Sand

0–14

0–14

0–10

0.02

0.03

0.01

Sandy loam

0–50

0–50

0–20

0.13

0.14

0.12

Clay loam

15–22

15–22

27–40

0.3

0.33

0.28

Loam

28–50

28–50

7–27

0.3

0.34

0.26

Loamy sand

0–30

0–30

0–15

0.04

0.05

0.04

Sandy clay loam

0–28

0–28

20–35

0.2

0.2

0.2

Silty clay loam

40–73

40–73

27–40

0.32

0.35

0.3

Silt

88–100

88–100

0–12

0.38

0.41

0.37

Silty loam

74–88

74–88

0–27

0.38

0.41

0.37

Fig. 5.4 LS factor

5.3 Methodology

75

Fig. 5.5 NDVI

complexity of “LULC” and overlapping character. So the vegetation algorithm can estimate the appropriate result regarding the “C factor”. But there is temporal variation of the vegetation cover which is the main barrier for estimating the “C factor” in meaningful way. Here “Normalized Difference Vegetation Index (NDVI)” have been considered as reliable vegetation indices for estimating the C factor which is capable to cover the maximum possible characteristics of indices. Here the NDVI raster layer has been accounted with incorporating the following equation in GIS environment. NDVI ¼

ðNIR - RedÞ ðNIR þ RedÞ

ð5:11Þ

Then, in a “GIS” system, the exponential function of the “NDVI” algorithm is used to

estimate the C factor raster (Zhou et al. 2008; Kouli et al. 2009): [

NDVI C ¼ exp -a ðb - NDVIÞ

] ð5:12Þ

where “a and b are the unit less parameters, which is capable to estimate the curve between NDVI and its associated C factor”. In comparison with the linear connection, this scenario provides stronger exponential prediction (Van der Knijff et al. 2000). The “NDVI” of this region varies between − 0.23 and 0.50 (Fig. 5.5). This region’s C factor ranges from 0.40 to 1.30 (Fig. 5.6). The centre, southern and eastern areas of the region have extremely high and high “C factor” values, whereas the remainder of the country has moderate to low “C factor” values. There is a lot of overlap between the “NDVI” and the “C factor” (Fig. 5.7).

76 Fig. 5.6 C factor

Fig. 5.7 Correlation between NDVI and C factor

5

Soil Loss Estimation Using Different Empirical …

5.5 Application of RUSLE

5.3.1.7 Support Practice Factor The “support practice factor” (P) indicates the impact of support practice which is able to modify the amount and direction of slope. Which is capable to modify the runoff and its associated flow pattern, this scenario can help for reducing the amount of surface runoff, erosion and rate of sedimentation. The “P factor” was used to describe the relationship between erosion and support practice. In this region for the estimation of “P factor”, the p values were assigned from the support practice that was obtained from the primary field visit and associated percentage slope. The “P factor” of this region ranges from 0.27 to 0.50, the greater values indicate the “support practice” which are adopted in larger scale. The western section of this region is mostly related with a high “support practice factor”, while the remaining is associated with moderate to low support practices (Fig. 5.8). The details for assigning the values of “support practice factor” are shown in Table 5.3.

77

relative status of soil erosion (Morgan et al. 1998). “USLE” with integrating GIS, the “spatially distributed sediment delivery ratio (SDR)” model can be run for measuring the status of soil erosion and deposition (Hui et al. 2010). Besides this, the potential soil loss of an island can be identified with the help of this model (Dumas and Printemps 2010). There has been a wider application of the condition of soil loss in microwatershed level (Shinde et al. 2010). It is used for identifying the suitable areas for measuring and implementing the erosion control (Pham et al. 2018). So the suitable measures are applied for overcome from such monotonous situation. For analyse the variation of soil loss, there has been a statistical analysis which is possible with the help of zonal statistics tool (Csáfordi et al. 2012). The “USLE” equation is the outcome of five input data (raster) factors such as “soil erodibility; rainfall erosivity; slope length and steepness; cover management; and support practice” (Belasri and Lakhouili 2016). A¼R×K×L×S×C×P

5.4

Application of USLE

This model first developed by the “United States Department of Agriculture (USDA)”, and it has a global application. The initial application for this model was bounded for statistical analysis, and the datasets are obtained from 10,000 plots years’ natural runoff and 2000 plots years’ artificial rainfall stimulators in USA (Wischmeier and Smith 1978). The “sediment erosion” of a basin should be determined by the “USLE” model with the basis of the natural characteristics of the basin (Calhoun and Fletcher III 1999). For identifying the suitable conservation unit for management practices, the USLE and integrated “GIS” has a wide application (Srinivas et al. 2002). The field condition can be evaluated with the help of “USLE” model (Nearing et al. 2017). For identification the status of a particular ecosystem, the measured data can be applicable against “USLE” for the validation of the study (Renschler and Harbor 2002). There has been a co-relation between the dynamicity of land use and land cover and the

ð5:13Þ

where “A is the estimate gross soil erosion, t/ha/year, R is the rainfall erosivity factor, J/ (ha/year), (t m/ha) (mm/h) per year, K is the soil erodibility factor (t/ha)/erosivity factor (R), t/J, t/ha year, L is the slope length factor, S is the slope gradient factor, C is the crop cover or crop management factor and P is the supporting conservation practice factor”.

5.5

Application of RUSLE

The RUSLE is one of the dependable empirical models which is based on the character of “topography, surface canopy cover, support practices, rainfall and runoff characteristics” and so on (Renard et al. 1991; Wang et al. 2003; Pan and Wen 2014). The average annual soil erosion can be quantified with the help of “RUSLE” with incorporating the less number of information (Angima et al. 2003). This model is capable to estimate the potential soil loss with adequate accuracy in a single storm event. This model is fitted in all the

78

5

Soil Loss Estimation Using Different Empirical …

Fig. 5.8 P factor

Table 5.3 Assign values for P factor in various parts of the world

Slope (%)

Contouring

Strip cropping

Terracing

0–7.0

0.55

0.27

0.10

7.0–11.3

0.60

0.30

0.12

11.3–17.6

0.80

0.40

0.16

17.6–26.8

0.90

0.45

0.18

> 26.8

1.00

0.50

0.20

region of various environments with little modifications. This model is suitable in tropical, subtropical, arid, temperate region, rugged topography and even the forest area. To estimate the average annual soil loss, the following equation is taken into consideration (Renard et al. 1997): A ¼ R × K × LS × C × P

ð5:14Þ

where “A is the Estimated gross soil erosion (tonne/ha/year), R is the runoff intensity factor in (MJ mm/ha/h/year), K is the soil erodibility factors (tonne/ha), LS is the slope length and steepness factor (dimensionless), C is the Crop cover and management factor (dimensionless) and P is the supporting conservation practice factor (dimensionless)”.

5.7 Results and Discussion

Because high-quality quantitative data are scarce in most third-world countries, the use of such a model is restricted. A single model is insufficient to achieve all of the objectives since each empirical and semi-empirical model has a specific function. “RUSLE” variables must be determined, however obtaining and fitting data are time-consuming and costly.

5.6

Application of MUSLE

This is the modified part of the original “USLE” model, which is derived accounting for the potential sediment yield from soil loss. “MUSLE” has been applicable for estimate annual soil loss and its associated sedimentation in average scale (Williams and Berndt 1977). In this model, the rainfall factor has been replaced by runoff factor because the uncertainty of rainfall in different periods (Arekhi and Niazi 2010). The application of soil loss model is not very easy but the sediment yield model is comparatively easy and meaningful for its outlet data structure and thus determines the watershed outlet (Sadeghi and Mizuyama 2007). This model has been able to determine the sediment yield with the basis of runoff characteristics of the watershed (Williams and Berndt 1977; Nearing et al. 2005). An erosivity factor has been included for its validation for runoff characteristics (Foster et al. 1977). “Runoff volume” and “peak discharge” have been evaluated in account of soil particle detachment for a single storm event (Sadeghi and Mizuyama 2007). The “MUSLE” has been fitted in the following equation for estimating the sediment yield loss for storm event: ( )0:56 S ¼ 11:8 Qqp K × L × S × C × P ð5:15Þ where “S is sediment yield, Q is volume of runoff, qp is peak flow and K, L, S, C and P are, respectively, the soil erodibility, slope length,

79

slope steepness, crop management and soil erosion control practice factors as similar to the USLE model” (Williams and Berndt 1977).

5.7

Results and Discussion

5.7.1 Average Annual Soil Loss Using USLE The “Universal Soil Loss Equation” (USLE) is used to calculate the average annual “soil loss”, which ranges from < 5.00 to > 35.00 tonnes per hectare per year. This is now divided into five categories: very high, high, moderate, low and very low. “Soil erosion” that is very high (> 35.00) is primarily found in the western part of the region. The western, middle portion of the region has the most high “soil erosion” (25.00– 35.00). The moderate soil erosion (15.00–25.00) is mostly found in the middle part. Low “soil erosion” (5.00–15.00) is primarily located in the centre and eastern parts of the region. “Soil erosion” is predominantly found in the eastern and southern parts of this region, with a very low (< 5.00) rate (Fig. 5.9).

5.7.2 Average Annual Soil Loss Using RUSLE The “Revised Universal Soil Loss Equation” (RUSLE) is a revised version of the “USLE” that is used to forecast “average soil loss” in a more realistic manner by taking into account slope length and steepness as a whole. The “average annual soil loss” ranges between < 5.00 and > 35.00 (tonnes/ha/year). The areas with the very high “soil erosion” (> 35) are primarily found in the western and south-western parts of the region. The western, south-western and centre portions of the region have the high of “soil erosion” (25–35). The moderate (15–25) “soil erosion” zone is mostly located in the centre.

80

5

Soil Loss Estimation Using Different Empirical …

Fig. 5.9 Average annual soil loss using USLE

Low “soil erosion” regions (5–15) are mostly found in the centre and eastern parts of the region. The very low (< 5) “soil erosion” zones are primarily situated in the region’s central and eastern regions (Fig. 5.10).

5.7.3 Average Annual Soil Loss and Its Associated Sediment Yield Using MUSLE This is a modified version of the original “USLE” model that can predict prospective “sediment yield” while taking into account the amount of “soil erosion”. The potential sediment output from soil loss in this study varies from < 5.00 to > 35.00 (tonnes/ha/year). The locations with the very high

“soil erosion” (> 35) are mostly located in the western part of the region. The western and centre portions of the region have the high “soil erosion” (25–35). The moderate “soil erosion” regions (15–25) are mostly clustered in the centre. Low “soil erosion” regions (5–15) are mostly found in the centre and south-eastern parts of the region. The very low (< 5) “soil erosion” zones are mostly located in the region’s central and eastern parts (Fig. 5.11).

5.8

Sensitivity Analysis

Selection of the suitable model by judging its accuracy with incorporating the different statistical characteristics is essential for management activities. We can decide the homogeneity and

5.8 Sensitivity Analysis

81

Fig. 5.10 Average annual soil loss using RUSLE

heterogeneity of the models with reality. For this purpose, large dataset is required for identifying the similarities with considering the specificity. The “ROC” is not only for use the statistical models but it is applicable for the models that are obtained from the “semi-empirical outcome, machine learning, artificial intelligence” and so on. The “AUC” (“Area Under Curve”) of “ROC” is the combination of sensitivity (“true positive rate”) and 1 − specificity (“false positive rate”). The “AUC” of the “ROC” has been determined using the equation below: SAUC ¼

n X ðXk þ 1 - Xk ÞðSk þ 1 - Sk þ 1 - Sk =2Þ k¼1

ð5:16Þ

where “SAUC is the area under curve, Xk is the 1 − specificity and Sk is the sensitivity of the ROC” (Chakrabortty et al. 2020). Some major erosion-prone areas of this region are shown in Fig. 5.12. The “AUC” values of “ROC” of “USLE, RUSLE and MUSLE” are 0.891, 0.924 and 0.981, respectively (Fig. 5.13), which indicates the RUSLE model is quite better than the “USLE” and “MUSLE” with considering its similarity with primary observed information. This investigation has emphasized upon the acceptability of the “RUSLE” model in subtropical region.

82

5

Soil Loss Estimation Using Different Empirical …

Fig. 5.11 Average annual soil loss using MUSLE

5.9

Discussion

Researchers from various fields, i.e. geography, geology, agriculture and engineering, have extensively investigated land degradation and, in especially, soil erosion as a physical phenomenon (Hudson 2015). Several research works have been done over the last two decades to estimate risk of soil erosion in various ecosystems (Pandey et al. 2021). The findings have resulted in a big dataset that may be used to define individual accomplishments on a more theoretical level (Boix-Fayos et al. 2006). Several scientists working on various elements of soil erosion mechanisms and at different scales, as well as in longer evaluations of gathered soil erosion datasets, gave similar observations (Batista et al.

2019). All of these insights are crucial because they allow us to learn from our past experiences and define future research areas. Generally, there is a necessity for understanding of the soil erosion mechanisms that occur in plots of various sizes, as well as the elements that drive natural variability, as a foundation for acquiring high-quality soil datasets (Doetterl et al. 2016). Because of the disparities in precision and uniformity of how human hands operate, human interruption in data gathering has been documented to be a substantial source of variance. When it comes to the applicability and effectiveness of a practical experimental design, there are many factors to consider: (i) “temporal and geographical dimensions”, (ii) “natural situation representation”, (iii) “natural situation disruption” and (iv) “diversity of ecosystem connections”

5.9 Discussion

83

Fig. 5.12 Some major erosion-prone area of this study region

Fig. 5.13 Sensitivity analysis

(Boix-Fayos et al. 2006). The matter that is taken from the hillslopes, deposited to the channel and carried to other “waterways, lakes and seas” is quantified in erosion research using a number of

methodologies (Hoffmann et al. 2007). The approach used is determined by the research team’s features (number of participants and training capabilities), economic assistance, aims and the extent of the study region. The observed results, meanwhile, were not irrespective of the approach employed (De Vente et al. 2007), because every approach is associated with different scales or a range of different levels, and therefore, every technique is chosen to quantify certain erosion mechanisms. Therefore, experimental plots have been a primary source of knowledge on erosion under diverse “land uses/covers” for the last several decades, but their utility as a method of calculating erosion rates has lately become questioned (García-Ruiz et al. 2015). Erosion modelling isn’t constant: it’s just as important to look ahead to future models as it is to evaluate current models for their application in

84

earth system change analysis. The current possibility for soil erosion is assessed to be at 0.38 mm year−1, with Southeast Asia having the most acute erosion issue (Yang et al. 2003). Numerous flooding or other environmental issues are induced by and are being exacerbated by, this tremendous quantity of soil erosion (Dotterweich 2008). “Soil erosion” is a physical occurrence on the earth’s crust that is hastened by human activities, particularly land development (Larson et al. 1983). In this chapter, we have considered “Universal Soil Loss Equation (USLE)”, “Modified Universal Soil Loss Equation (MUSLE)” and “Revised Universal Soil Loss Equation (RUSLE)” for estimating the average annual soil erosion in this region. From the validation part, we found that the optimal capacity of “RUSLE” is comparatively high than the “USLE” and “MUSLE”. The “Revised Universal Soil Loss Equation (RUSLE)” replaces the commonly used “Universal Soil Loss Equation (USLE)” (Renard et al. 1997). “RUSLE” estimates long-term rate of soil erosion more precisely than “USLE” by using four indicators: “rainfall erosivity, soil erodibility, topography and vegetation” (Mbanzamihigo 2021). In addition to consider these erosion components, the “RUSLE” model may be scaled up to the “global level” (Borrelli et al. 2020). Despite the fact that the “RUSLE” approach is one of the most extensively utilized soil erosion models in the world, its complete and full application is confined to nations that already have the necessary full datasets (Tanyaş et al. 2015). Countries without such datasets are attempting to quantify the required variables using a variety of methods (Tanyaş et al. 2015). As a response, there could be a variety of methods for estimating the quantity of eroded material in a particular watershed, each yielding an opposite outcome (Nearing et al. 2005). As a result, eroded sediment quantity estimating procedures are completed with separate “RUSLE” applications that cannot be comparable owing to approach variations for determining input variables, and testing of the applied model is not possible in the majority of situations (Jetten et al.

5

Soil Loss Estimation Using Different Empirical …

2003). “RUSLE” has the drawback of ignoring sedimentation mechanisms in the computation (Ranzi et al. 2012). “RUSLE” estimates soil movement at a specific place rather than the volume of sediment leaving an area or catchment (Biesemans et al. 2000).

5.10

Conclusion

Our existence is dependent on land resources, which are the primary source of food, which is the most vital ingredient for livelihood. So reducing the amount of soil erosion is very much essential for maintaining the balance between demand of food for the population and the amount of agricultural production. The region dominated by the “subtropical monsoon” suffers the most from water-induced surface soil erosion. So, in order to adopt the most appropriate development strategies or measures for working out in this situation, assessment and quantification with appropriate methodologies are required. The empirical and semi-empirical models with an integrated “GIS platform” were considered in this work. When comparing primary data, the “RUSLE” is a far better predictor than the “USLE” and “MUSLE”. The western part of the study region is mainly experiencing the higher erosion in comparison with the rest of the part. This region is associated with rugged topography and high slope land. In vulnerable areas, appropriate structural and nonstructural measures must be implemented to prevent this type of monotonous situation. Traditional approaches for minimizing surface erosion have already been adopted by local stakeholders. The majority of individuals involved in the agricultural system are of a medium or marginal nature, and thus, they try to ensure the highest possible production by employing various labourintensive strategies. In order to reduce the rate of erosion and associated sedimentation, some tillage methods must be used in the upper half of the basin. Land use planning strategies that control runoff or run-on are known as water erosion risk controls. The mentioned approaches of soil and rock bunds can also be employed interchangeably

References

depending on availability of stones and manpower. In a subtropical monsoon environment like India, the “RUSLE” can be said to be the most reliable predictor of soil loss. The knowledge of the local stakeholders can be considered as the main component to identifying the nature of problem and adopting a holistic approach in regard to escaping from this hazardous situation. Agriculturists used to have a leg up on the competition when it came to soil erosion, management and productivity-boosting techniques. As a result, building on farmers’ experience via participatory improvements to address the roadblocks that prevent them from fully utilizing recognized soil erosion prevention and conservation techniques would have a positive impact on “soil and water conservation (SWC)” and “long-term land management”. The next initiative for future researchers is to emphasize the erosivity factor of rainfall and runoff on “RUSLE” and determine the rainfall and runoff threshold. If this scenario is identified, the risk associated with this matter can be reduced significantly.

References Amsalu T, Mengaw A (2014) GIS based soil loss estimation using RUSLE model: the case of Jabi Tehinan Woreda, ANRS, Ethiopia. Nat Resour 2014 Angima S, Stott D, O’Neill M et al (2003) Soil erosion prediction using RUSLE for central Kenyan highland conditions. Agric Ecosyst Environ 97: 295–308 Arekhi S, Niazi Y (2010) Assessment of GIS and RS applications to estimate soil erosion and sediment loading by using RUSLE model (case study: upstream basin of Ilam dam). J Soil Water Conserv 17:1–27 Arnalds O, Barkarson B (2003) Soil erosion and land use policy in Iceland in relation to sheep grazing and government subsidies. Environ Sci Policy 6:105–113 Batista PV, Davies J, Silva ML, Quinton JN (2019) On the evaluation of soil erosion models: are we doing enough? Earth Sci Rev 197:102898 Belasri A, Lakhouili A (2016) Estimation of soil erosion risk using the universal soil loss equation (USLE) and geo-information technology in Oued El Makhazine watershed, Morocco. J Geogr Inf Syst 8:98 Biesemans J, Van Meirvenne M, Gabriels D (2000) Extending the RUSLE with the Monte Carlo error propagation technique to predict long-term average off-site sediment accumulation. J Soil Water Conserv 55:35–42

85 Blanco H, Lal R (2008) Principles of soil conservation and management. Springer, New York Boix-Fayos C, Martínez-Mena M, Arnau-Rosalén E et al (2006) Measuring soil erosion by field plots: understanding the sources of variation. Earth Sci Rev 78:267–285 Borrelli P, Robinson DA, Panagos P et al (2020) Land use and climate change impacts on global soil erosion by water (2015–2070). Proc Natl Acad Sci 117:21994–22001 Butzer KW (2005) Environmental history in the Mediterranean world: cross-disciplinary investigation of cause-and-effect for degradation and soil erosion. J Archaeol Sci 32:1773–1800 Calhoun RS, Fletcher CH III (1999) Measured and predicted sediment yield from a subtropical, heavy rainfall, steep-sided river basin: Hanalei, Kauai, Hawaiian Islands. Geomorphology 30:213–226 Cantón Y, Solé-Benet A, De Vente J et al (2011) A review of runoff generation and soil erosion across scales in semiarid south-eastern Spain. J Arid Environ 75:1254–1261 Chakrabortty R, Pal SC, Chowdhuri I et al (2020) Assessing the importance of static and dynamic causative factors on erosion potentiality using SWAT, EBF with uncertainty and plausibility, logistic regression and novel ensemble model in a sub-tropical environment. J Indian Soc Remote Sens 48:765–789. https://doi.org/10.1007/s12524-020-01110-x Chen G, Zhang Z, Guo Q et al (2019) Quantitative assessment of soil erosion based on CSLE and the 2010 national soil erosion survey at regional scale in Yunnan Province of China. Sustainability 11:3252 Csáfordi P, Pődör A, Bug J, Gribovszki Z (2012) Soil erosion analysis in a small forested catchment supported by ArcGIS model builder. Acta Silv Lignaria Hung 8:39–55 De Vente J, Poesen J, Arabkhedri M, Verstraeten G (2007) The sediment delivery problem revisited. Prog Phys Geogr 31:155–178 Doetterl S, Berhe AA, Nadeu E et al (2016) Erosion, deposition and soil carbon: a review of process-level controls, experimental tools and models to address C cycling in dynamic landscapes. Earth Sci Rev 154:102–122 Dotterweich M (2008) The history of soil erosion and fluvial deposits in small catchments of central Europe: deciphering the long-term interaction between humans and the environment—a review. Geomorphology 101:192–208 Dumas P, Printemps J (2010) Assessment of soil erosion using USLE model and GIS for integrated watershed and coastal zone management in the South Pacific Islands, pp 856–866 El Kateb H, Zhang H, Zhang P, Mosandl R (2013) Soil erosion and surface runoff on different vegetation covers and slope gradients: a field experiment in Southern Shaanxi Province, China. CATENA 105:1–10 Foster GR, Meyer L, Onstad C (1977) A runoff erosivity factor and variable slope length exponents for soil loss estimates. Trans ASAE 20:683–687

86 García-Ruiz JM, Beguería S, Nadal-Romero E et al (2015) A meta-analysis of soil erosion rates across the world. Geomorphology 239:160–173 Hoffmann T, Erkens G, Cohen K et al (2007) Holocene floodplain sediment storage and hillslope erosion within the Rhine catchment. The Holocene 17:105–118 Hudson NW (2015) Soil conservation. Scientific Publishers Hui L, Xiaoling C, Lim KJ et al (2010) Assessment of soil erosion and sediment yield in Liao watershed, Jiangxi Province, China, using USLE, GIS, and RS. J Earth Sci 21:941–953 Jat ML, Stirling CM, Jat HS et al (2018) Soil processes and wheat cropping under emerging climate change scenarios in South Asia. Adv Agron 148:111–171 Jena R, Padua S, Ray P et al (2018) Assessment of soil erosion in sub tropical ecosystem of Meghalaya, India using remote sensing, GIS and RUSLE Jetten V, Govers G, Hessel R (2003) Erosion models: quality of spatial predictions. Hydrol Process 17:887– 900 Keesstra S, Mol G, de Leeuw J et al (2018) Soil-related sustainable development goals: four concepts to make land degradation neutrality and restoration work. Land 7:133. https://doi.org/10.3390/land7040133 Kouli M, Soupios P, Vallianatos F (2009) Soil erosion prediction using the revised universal soil loss equation (RUSLE) in a GIS framework, Chania, Northwestern Crete, Greece. Environ Geol 57:483–497. https://doi.org/10.1007/s00254-008-1318-9 Larson WE, Pierce FJ, Dowdy RH (1983) The threat of soil erosion to long-term crop production. Science 219:458–465 Leopold LB, Wolman MG, Miller JP, Wohl E (2020) Fluvial processes in geomorphology. Courier Dover Publications Liu B, Guo S, Li Z et al (2013) Sampling program of water erosion inventory in China. Soil Water Conserv China 10:26–34 Mairura FS, Mugendi DN, Mwanje J et al (2007) Integrating scientific and farmers’ evaluation of soil quality indicators in Central Kenya. Geoderma 139:134–143 Maji A, Reddy GO, Sarkar D (2010) Degraded and wastelands of India: status and spatial distribution Mbanzamihigo E (2021) Spatial-temporal relationships between crops production and soil erosion: case of Western Province of Rwanda McCool DK, Foster GR, Mutchler C, Meyer L (1989) Revised slope length factor for the universal soil loss equation. Trans ASAE 32:1571–1576 Mekonnen M, Keesstra SD, Stroosnijder L et al (2015) Soil conservation through sediment trapping: a review: individual, combined and integrated sediment trapping. Land Degrad Dev 26:544–556. https://doi. org/10.1002/ldr.2308 Mohammad AG, Adam MA (2010) The impact of vegetative cover type on runoff and soil erosion under different land uses. CATENA 81:97–103

5

Soil Loss Estimation Using Different Empirical …

Morbidelli R, Saltalippi C, Flammini A, Govindaraju RS (2018) Role of slope on infiltration: a review. J Hydrol 557:878–886 Morgan R, Quinton J, Smith R et al (1998) The European soil erosion model (EUROSEM): a dynamic approach for predicting sediment transport from fields and small catchments. Earth Surf Process Landf J Br Geomorphol Group 23:527–544 National Research Council (1993) Soil and water quality: an agenda for agriculture. National Academies Press Nearing M, Jetten V, Baffaut C et al (2005) Modeling response of soil erosion and runoff to changes in precipitation and cover. CATENA 61:131–154 Nearing M, Lane LJ, Lopes VL (2017) Modeling soil erosion. In: Soil erosion research methods. Routledge, pp 127–158 Nunes J, De Lima J, Singh V et al (2006) Numerical modeling of surface runoff and erosion due to moving rainstorms at the drainage basin scale. J Hydrol 330:709–720 Pal SC, Chakrabortty R (2019) Simulating the impact of climate change on soil erosion in sub-tropical monsoon dominated watershed based on RUSLE, SCS runoff and MIROC5 climatic model. Adv Space Res 64:352–377 Pal SC, Shit M (2017) Application of RUSLE model for soil loss estimation of Jaipanda watershed, West Bengal. Spat Inf Res 25:399–409 Pal SC, Chakrabortty R, Roy P et al (2021) Changing climate and land use of 21st century influences soil erosion in India. Gondwana Res 94:164–185. https:// doi.org/10.1016/j.gr.2021.02.021 Pan J, Wen Y (2014) Estimation of soil erosion using RUSLE in Caijiamiao watershed, China. Nat Hazards 71:2187–2205 Pandey S, Kumar P, Zlatic M et al (2021) Recent advances in assessment of soil erosion vulnerability in a watershed. Int Soil Water Conserv Res 9:305–318 Peterson GD, Cumming GS, Carpenter SR (2003) Scenario planning: a tool for conservation in an uncertain world. Conserv Biol 17:358–366 Pham TG, Degener J, Kappas M (2018) Integrated universal soil loss equation (USLE) and geographical information system (GIS) for soil erosion estimation in A Sap basin: Central Vietnam. Int Soil Water Conserv Res 6:99–110 Pimentel D, Allen J, Beers A et al (1993) Soil erosion and agricultural productivity. In: World soil erosion and conservation, pp 277–292 Prasannakumar V, Vijith H, Abinod S, Geetha N (2012) Estimation of soil erosion risk within a small mountainous sub-watershed in Kerala, India, using revised universal soil loss equation (RUSLE) and geoinformation technology. Geosci Front 3:209–215. https://doi.org/10.1016/j.gsf.2011.11.003 Ranzi R, Le TH, Rulli MC (2012) A RUSLE approach to model suspended sediment load in the Lo river (Vietnam): effects of reservoirs and land use changes. J Hydrol 422:17–29

References Renard KG, Foster GR, Weesies GA, Porter JP (1991) RUSLE: revised universal soil loss equation. J Soil Water Conserv 46:30–33 Renard KG, Foster GR, Weesies GA, McCool DK, Yoder DC (eds) (1997) Predicting soil erosion by water: a guide to conservation planning with the revised universal soil loss equation (RUSLE). Washington, DC Renschler CS, Harbor J (2002) Soil erosion assessment tools from point to regional scales—the role of geomorphologists in land management research and implementation. Geomorphology 47:189–209 Rodríguez-Caballero E, Cantón Y, Chamizo S et al (2013) Soil loss and runoff in semiarid ecosystems: a complex interaction between biological soil crusts, microtopography, and hydrological drivers. Ecosystems 16:529–546 Sadeghi S, Mizuyama T (2007) Applicability of the modified universal soil loss equation for prediction of sediment yield in Khanmirza watershed, Iran. Hydrol Sci J 52:1068–1075 Sarvade S, Upadhyay V, Kumar M, Khan MI (2019) Soil and water conservation techniques for sustainable agriculture. In: Sustainable agriculture, forest and environmental management. Springer, pp 133–188 Schmidt S, Alewell C, Meusburger K (2018) Mapping spatio-temporal dynamics of the cover and management factor (C-factor) for grasslands in Switzerland. Remote Sens Environ 211:89–104 Shinde V, Tiwari K, Singh M (2010) Prioritization of micro watersheds on the basis of soil erosion hazard using remote sensing and geographic information system. Int J Water Resour Environ Eng 5:130–136 Srinivas C, Maji A, Chary G (2002) Assessment of soil erosion using remote sensing and GIS in Nagpur district, Maharashtra for prioritisation and delineation of conservation units. J Indian Soc Remote Sens 30:197–212 Tanyaş H, Kolat Ç, Süzen ML (2015) A new approach to estimate cover-management factor of RUSLE and validation of RUSLE model in the watershed of Kartalkaya Dam. J Hydrol 528:584–598

87 Toy TJ, Foster GR, Renard KG (2002) Soil erosion: processes, prediction, measurement, and control. Wiley Van der Knijff J, Jones R, Montanarella L (2000) Soil erosion risk assessment in Europe, EUR 19044 EN. Office for Official Publications of the European Communities, Luxembourg, p 34 Vrieling A, de Jong SM, Sterk G, Rodrigues SC (2008) Timing of erosion and satellite data: a multi-resolution approach to soil erosion risk mapping. Int J Appl Earth Obs Geoinf 10:267–281 Wang G, Gertner G, Fang S, Anderson AB (2003) Mapping multiple variables for predicting soil loss by geostatistical methods with TM images and a slope map. Photogramm Eng Remote Sens 69:889–898 Williams J, Berndt H (1977) Sediment yield prediction based on watershed hydrology. Trans ASAE 20:1100– 1104 Wischmeier WH, Smith DD (1978) Predicting rainfall erosion losses: a guide to conservation planning. Department of Agriculture, Science and Education Administration Yang D, Kanae S, Oki T et al (2003) Global potential soil erosion with reference to land use and climate changes. Hydrol Process 17:2913–2928. https://doi. org/10.1002/hyp.1441 Yin S, Zhang W, Xie Y et al (2013) Spatial distribution of rainfall erosivity in China based on high-density station network. Soil Water Conserv 10:45–51 Zhang W, Fu J (2003) Rainfall erosivity estimation under different rainfall amount. Resour Sci 25:35–41 Zhou Z, Chu S, Wang Z, Chen Q (2008) Analysis of vegetation coverage change based on NDVI—a case study in Ganzhou area, Zhangye city, Gansu. Pratacult Sci 12 Zuazo VHD, Pleguezuelo CRR (2009) Soil-erosion and runoff prevention by plant covers: a review. Sustain Agric 785–811

6

Potential Sediment Yield Estimation Using Machine Learning, Artificial Intelligence Techniques and GIS

Abstract

Keywords

“Soil erosion”-induced land degradation is the most common issues in the subtropical region. When this eroded material transported in a different place and accumulate there, this is called the sediment yield. This scenario is very much influential harmful element which decreases the navigation power of water resources. The “fuzzy logic (FL)” and “analytical neural network (ANN)” have been incorporated with considering different influential factors of sediment yield. In the recent studies, the researchers are tried to estimate the sediment yield potentiality and its spatial difference then the estimating the point-specific information. The weight of the different influential factors and its associated subfactors has been assigned with considering the function of “fuzzy logic”. The “multi-layer perceptron classifier (MLPC)” approach in “ANN” has been considered to estimate the sediment yield potentiality in this region. The middle, southern and south-eastern part of this region is associated with moderate to very high sediment yield zones though it is unevenly distributed throughout this region. This sort of particular information aids decision-makers and future researchers in developing appropriate strategies and doing future study on this issue in more precise ways.

Soil erosion Sediment yield Fuzzy logic Analytical neural network Multi-layer perceptron classifier

.

6.1

.

.

.

Introduction

Soil erosion and sediment-related challenges jeopardize effective “land management and water resource development” in several developing nations (Tamene et al. 2006). Soil is a significant natural material (Lal 2015), it is necessary for human safety, it acts a key role in the ongoing ecosystem (Bardgett 2005), and it produces value commodities and services. Soil loss is a natural geomorphic mechanism and an environmental issue (Ravi et al. 2010) that is exacerbated by human activities such as intensive agriculture, deforestation, soil degradation and anthropogenic climate shifts (Jie et al. 2002). Soil erosion is one of the serious risks to the environment (Pimentel 2006), because it not only induces land degradation from upper reaches and accumulation in rivers and lakes throughout geologic time (Issaka and Ashraf 2017) but also transports nutrition, herbicides, chemical products and other contaminants, resulting in groundwater pollution. Low to extreme types of soil erosion induced by water have been estimated to destroy over 56% of worldwide soil. Increased types of soil loss by

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 S. C. Pal and R. Chakrabortty, Climate Change Impact on Soil Erosion in Sub-tropical Environment, Geography of the Physical Environment, https://doi.org/10.1007/978-3-031-15721-9_6

89

90

6

water are becoming a global concern (Jie et al. 2002), posing a threat not just to rivers’ catchment areas, but also with the “United Nations Sustainable Development Goals” (Keesstra et al. 2016). As a result, estimating soil erosion caused by river catchment water is critical, so that appropriate soil erosion prevention techniques may be targeted (Beskow et al. 2009). As a result, estimating soil erosion caused by river catchment water is critical, so that appropriate soil erosion prevention techniques may be targeted (Beskow et al. 2009). Early civilizations recognized soil erosion as an issue. It has hampered socioeconomic well-being for some time, resulting in a loss of nutrient and agricultural productivity, as well as an increase in hunger and starvation throughout the world (Keesstra et al. 2016). The least developed nations, on the other hand, were given special attention because their effort to feed an ever-increasing population, along with human meddling in natural forests and ineffective conservation measures, has hastened land degradation in the form of soil erosion. The most significant quantitative variables in rivers are suspended and dissolved loads, which are produced by the two important types of erosion, mechanical and chemical (Holland 1978). River transportation of substances has been included in geochemical budgeting for almost a year (Chakrapani and Subramanian 1993). It gives critical information to the volume and kind of materials moved to aquatic bodies as well as activities that influence the continental surface (weathering, plant biomass, pollution and so on). Rivers are estimated to convey roughly 37,000 km3 of water and 15–16 × 1012 g of sediment to the seas every year (Walling and Webb 1983). Although Indian rivers carry just 5% of world water runoff, they convey over 30% of the entire sediment burden delivered to the seas. The use of rivers for mass transportation in India has been documented on several occasions (Ray et al. 1984; Sarin and Krishnaswami 1984). The Bengal Basin is the world’s biggest delta basin, and it is still dynamic with a high flow velocity (Chamberlain et al. 2020). The river transports the majority of the silt to the continental margin. In fact, this river-transported silt is

Potential Sediment Yield Estimation Using Machine …

a primary determinant of coastal elevation and growth. The Bengal delta basin, on the other hand, is not a new basin. It began to evolve throughout the Holocene epoch (Bandyopadhyay 2007). The “Ganges–Brahmaputra River” delta was formed some 11,000 years ago when increasing sea levels swamped the Bengal basin, confining most of the river’s flow on the Bay of Bengal’s near shore/coast marine (Mukherjee et al. 2009). The rates of sediment storage on the edge are calculated using chronostratigraphic evidence from such deltaic sediments, which offers a minimum approximation of the river’s previous sediment load. Sedimentation surveys are difficult to do in developing and rural places (Wolman 1967). To maintain their potentiality, such regions, on the other hand, require earlier focus on resource management and monitoring (Tamene et al. 2006). As a consequence, obtained data at one area must be extrapolated to a location with a comparable scale and environment (De Vente and Poesen 2005). Most “sediment yield” projections are currently made using basic empirical approaches that link a river’s yearly sediment delivery to watershed parameters such as “drainage area, topography, land use and cover (LULC), geological properties and climatic parameters” (Rajbanshi and Bhattacharya 2020). If maps of adequate scale are obtainable, catchment area might be regarded one of the topographical qualities that can be consistently assessed. Usually, while attempting to develop microdams, an assessment of basin area is established since it is one of the prerequisites for determining quantity of runoff and reservoir capacity (Tamene et al. 2006). Predicting “specific sediment yield” focusing on catchment area or boosting its capacity depending on readily accessible data is thus critical, particularly in data-scarce locations where “processbased approaches” will not be practical (Grayson et al. 1992). The complexities and unpredictability of linked terrain features like “topography, LUC, soils and climate” affect the connection among yearly sediment output and catchment area (Verstraeten et al. 2003). This implies that the two watersheds of equal size have distinct climates and geomorphologies,

6.4 Preparation of Causal Parameters

their sediment delivery may varied (Hamel et al. 2017). If the erosion mechanisms involved and the related factors within consideration are varied, the connection among sediment output and area may very well change across neighbouring regions taking into account local circumstances and the spatial scales we’re working with (Schiefer et al. 2001). So, the estimation of sediment yield potentiality is one of the important research topics in recent time. Various scholars are attempting to estimate this kind of event using various methods. This will identify the primary important points of soil erosion in a given location, allowing conservation efforts to be focused for better outcomes. The “Universal Soil Loss Equation (USLE)” and the “Revised Universal Soil Loss Equation (RUSLE)”, despite being two of the most commonly recognized and frequently used soil erosion models, do not reveal evidences on sediment transport to river channels. To quantify the actual scenario of erosion, various models have been developed. In this research, the “fuzzy logic (FL)” and “analytical neural network (ANN)” have been considered for estimating the sediment yield potentiality of Bengal basin. “Fuzzy logic” is a computing approach which uses “degrees of truth” instead of the usual “true or false” (1 or 0) Boolean logic that is used in mainframe technology (Garrido 2012). “Artificial intelligence (AI) systems” use “fuzzy logic” to replicate human thinking and cognition. “Fuzzy logic” provides values of truth between 0 and 1, as well as a range of degrees of truth in between, rather than only binary instances of truth (Bělohlávek et al. 2017). In the nineteenth and early twentieth decades, a few of the groundwork for the area of “artificial neural networks (ANNs)” was done. This mostly consists of physics, psychology and neurophysiology multidisciplinary work (Kumar 2004). This early study focused on general learning, vision, conditioning and other ideas, rather than detailed scientific methods of neuron activity. The field of neural networks has been revitalized as a result of these new advancements. Many articles have been published in the previous two decades, and many different forms of ANNs have been studied.

91

6.2

Materials and Methods

The following precise approach has been considered for estimating the “sediment yield” potentiality using “fuzzy logic (FL)” and “analytical neural network (ANN)” (Fig. 6.1). In this perspective, different database has been considered for the estimation of the sediment yield potentiality. The “ALOS PALSER DEM (digital elevation model)” from “Japan Aerospace Exploration Agency” with 12.5 m spatial resolution, Sentinel 2a satellite imageries from “European Space Agency” with 10 m spatial resolution, geology from “Geological Survey of India” and different primary information has been considered for estimating the sediment yield potentiality.

6.3

Selection of the Causative Factors

The selection of the causal parameters related to the erosion and its associated sedimentation has been selected on the basis of different literatures related to this field (Table 6.1).

6.4

Preparation of Causal Parameters

The preparation of causal parameters has been done with the help of various datasets related to erosion and sedimentation (Fig. 6.2).

6.4.1 Rainfall and Runoff Erosivity R is represented as rainfall–runoff erosivity factor. It is the average yearly accumulation (EI) values for a typical rainy year. The erosion index is a measurement of the force of erosion caused by a certain amount of rainfall. Once all circumstances remain constant, storm losses from precipitation are proportion to the product of the storm’s total “kinetic energy (E)” times its greatest 30-min intensity (I). Storms that are smaller than 0.5 in. in diameter are excluded from the erosivity calculations since they add less

92

6

Potential Sediment Yield Estimation Using Machine …

Fig. 6.1 Methodology flowchart

to the total R value. Over a 22-year period, R factors indicate the mean storm EI values. The two most essential elements of a storm that determine its erosivity are the volume of precipitation and the intensity values maintained over time. R factor was determined using the algorithms by Zhang and Fu (2003), Yin et al. (2013) updated this model, which was ultimately employed by Liu et al. (2013): R¼

24 X

Rhmk

ð6:1Þ

k¼1

Rhmk ¼

n X m ( ) 1X a . P1:7265 i;j;k n i¼1 j¼0

WRhmk ¼

Rhmk R

ð6:2Þ

ð6:3Þ

Here, “R is the average annual rainfall erosivity (MJ mm ha−1 h−1), Rhmk is the rainfall erosivity

of the kth half-month (MJ mm ha−1 h−1), pi,j,k is the average daily erosive rainfall occurring on the jth day, kth half-month and ith year. WRhmk is the proportion of the rainfall erosivity of the kth half-month to the average annual rainfall erosivity”.

6.4.2 Drainage Density Drainage density of a specified area is determined by relationship of the total length of the stream with its associated area. The overall character of the basin is very much related with the nature of the drainage. The character of the hydrology is also dependent on the nature of drainage density. The high drainage density is more prone to erosion in comparison with other areas. Here the stream network is extracted from ALOS PALSAR DEM, and this output is validated with the help of satellite data and topographical maps. To prepare DEM, the raster

6.4 Preparation of Causal Parameters

93

Table 6.1 Reason for the selection of causal parameters Causal factors

Reason for selection

Rainfall and runoff erosivity

It is directly related to the amount and direction of sedimentation. The rate of surface erosion can be influenced with the amount of rainfall and runoff factor which is favourable for sedimentation (Liu et al. 2015)

Drainage density

There is direct relationship between the drainage density and rate of erosion (Mofidi et al. 2021)

LULC

Conversion of the forest land into agricultural land leads to increase in the rate of soil erosion and its associated sedimentation in tropical and subtropical environments (Chakrabortty et al. 2020a)

Slope

The most important determining element of erosion and its associated sediment yield is the amount and direction of slope (Jain and Das 2010)

Soil texture

A textural characteristic of the soil is responsible for large-scale erosion and deposition. The infiltration capacity of the soil is also influenced by the textural characteristics. The amount of surface runoff is the determining element of deposition which is also determined by the textural characteristics of the soil (Puigdefábregas et al. 1999)

Elevation

Surface elevation can influence the amount and direction of slope and which indicates the probable areas of erosion and sedimentation (Dietrich et al. 1995)

Geology

The lithological association and surface geology is one of the important determining elements of erosion. The amount and intensity of erosion are influenced by the geological association (Avanzi et al. 2004)

Stream power index

There is negative relationship found between stream power index and sedimentation. Apart from this, the strongly positive relationship found between stream power index and rate of erosion (Harel et al. 2016)

Topographical wetness index

The source of runoff can be estimated through the saturated point source with incorporating TWI. The effect of topography on the association of saturated point for creating the runoff is an important determining element for erosion susceptibility as well as sedimentation (Tilahun et al. 2013)

Soil erodibility

There is a direct positive relationship has been found between the probability of sediment yield and the erodibility of soil. Soil physical, chemical and biological association can favourable for eroding the surface materials (De Vente and Poesen 2005)

DEM is filled, then flow accumulation algorithm is incorporated for working out the stream network with considering the threshold unit. We found that the middle portion of the basin is characterized with the high and moderate drainage density, and surrounding areas are associated with low drainage density.

6.4.3 Land Use and Land Cover (LULC) Changes in “land use and land cover” and its associated environmental problems are considered as a global concern. For increasing the population, the forest area and grassland are converted into agricultural and settlement area

which phenomenon leads to large-scale environmental degradation. The rate of changes of LULC is not uniform in the overall area of this region. If this region is very much fertile in nature, so the people are trying to emphasize to increasing the production with expansion of agricultural area. There are so many LULC features in this area; these are dense vegetation, scattered vegetation, agricultural land, fallow land, water body, sand area and settlement area.

6.4.4 Slope Slope angle, particularly for bare soils, is significant in the dynamics of geomorphological events. Rainsplash, sheet and intense overland flow, as well

94

6

Potential Sediment Yield Estimation Using Machine …

Fig. 6.2 Sediment yield causal parameters; R factor (a), drainage density (b), LULC (c), slope (d), soil texture (e), elevation (f), geology (g), SPI (h), TWI (i) and K factor (j)

6.4 Preparation of Causal Parameters

95

Fig. 6.2 (continued)

as soil losses, are theoretically increased on high angle slopes. Because the most deteriorated soils are found on the highest slopes, the slope angle usually has an inverse connection with infiltration rates. Surfaces with a high inclination had a greater impact on soil composition than those with a gentle slope. Minerals and organic material in the soil flow down the slope as a result of heavy rainfall. On slopes with less vegetation, soil erosion is more likely. To keep the topsoil from eroding, shrubs and vegetation with deep roots can be introduced. Soil particles gather near the slope’s base as a result of heavy rainfall. Lighter materials on the topsoil, such as fine sand, flow readily with the water and wash off from the bottom.

6.4.5 Soil Texture Soil texture is an essential characteristic of soil that affects plant development. This task will give us a more quantification of soil characteristics. Various soil textures have the unique function on erosion and sedimentation. Various soil textures are associated with this region which has the differentiate role on erosion and sedimentation. The quantity of sediment generated and moved during the chemical and physical disintegration of surface rocks, and minerals demonstrate the relevance of natural and artificial processes in affecting Earth’s subaerial surface.

96

6

6.4.6 Elevation The elevation is one of the determining elements which directly influences the amount and direction of slope in a particular region. On the other side, the changes of elevation may lead to possible causes for erosion. The eroded materials transported by different means of element and eventually accumulated in the lower portion of the catchment as sediment. In this region, the ranges of elevation are between − 9 and 699 m.

6.4.7 Geology Among the most critical challenges, many soil and water conservation researches are comparing various geological formations in terms of sediment output (Amare et al. 2014). Furthermore,

Potential Sediment Yield Estimation Using Machine …

watershed management requires the assessment of runoff production rate and sediment generation (Shi et al. 2012). The erosion capacity of a watershed is determined by its “erosivity, erodibility, geological formations, slope gradient and land use patterns” (Ganasri and Ramesh 2016). As a result, runoff is one of the major contributors to water erosion problems. Various geological units have been associated with this region. These are “Barakar Formation, Barren Measure Formation, Chotanagpur Gneissic Complex, Dalma Volcanics, Dubrajpur Fm, Durgapur bed, Gabbro and Anorthosite Complex, Kuilapal Granite, Laterite and bauxite, Manbhum Granite, Manbhum Granite, Panchet\Pachmarhi Formation, Rajmahal Trap, Raniganj Formation, Singhbhum Gp, Talchir Formation, Unclassified metamorphic, Undefined fluvial and coastal glacial sediments” (Table 6.2).

Table 6.2 Geological formation of the study region Formation

Period

Durgapur bed

Jurassic

Kuilapal Granite

Mesoproterozoic

Undiff. fluvial/aeolian/coastal and glacial sediments

Quaternary

Laterite and bauxite

Cenozoic

Area

Area in percentage (%) 7.00

72.11 47,417.80 1513.09

0.01 0.12 78.07 2.49

Gabbro and Anorthosite Complex

Proterozoic

290.77

0.48

Rajmahal Trap

Cretaceous

224.98

0.37

Manbhum Granite

Mesoproterozoic

553.11

0.91

Panchet\Pachmarhi Fm

Triassic

366.52

0.60

Dalma Volcanics

Archaean–Proterozoic

442.72

0.73

Unclassified metamorphics

Archaean–Proterozoic

571.52

0.94

Chotanagpur Gneissic Complex

Proterozoic (undifferentiated)

6079.22

10.01

2263.10

3.73

134.59

0.22

11.23

0.02

607.23

1.00

Singhbhum Gp

Palaeoproterozoic

Barren Measure Fm

Permian

Talchir Fm

Carboniferous

Raniganj Fm

Permian

Dubrajpur Fm

Jurassic–Cretaceous

Barakar Fm

Permian

Total

47.29

0.08

138.48

0.23

60,740.74

100

6.4 Preparation of Causal Parameters

97

6.4.8 Stream Power Index The erosive power of the flowing water can be estimated through stream power index with considering the discharge as proportional scenario in respect to the catchment area. The available energy which is responsible for largescale erosion comes from stream power index. The entrain of the sediment is responsible for the amount of stream power index. D¼W ×D×V

ð6:4Þ

where “D is the amount of discharge, W is the width of the river, D is depth of the river and V is the velocity of the river”. USP ¼ / QS

ð6:6Þ

where “R is the hydraulic radius and S is the slope”. R¼

A WP

ð6:7Þ

where “R is the hydraulic radius, A is the area and WP is the wetted perimeter” (Chakrabortty et al. 2020b). The following equation has been used to estimate the stream power index in a GIS framework: SPI ¼ As × tan r

TWI ¼ ln

/ tan b þ C

ð6:9Þ

ð6:5Þ

where “USP is the unit stream power, / is the constant, Q is the discharge and S is the slope”. 1 3 1 V ¼ R 2 × S2 n

by the flow size and imitates upslope water supply. The total catchment area, flow width and slope gradient are the three basic components of the TWI. The total catchment area is the accumulated steep slope area that discharges through the cell, the flow depth is the span of a contour transverse, and the slope angle is either the slope of the centre point cell or the slope determined with flow routing algorithm in between focal cell and a cell further downward slope. There is a direct link between TWI and erosion susceptibility in a certain area (Roy et al. 2020).

ð6:8Þ

where “As is the specific catchment area in metres and r is the slope gradient”.

6.4.9 Topographical Wetness Index The “Topographic Wetness Index (TWI)” is presented as the supply of water from upslope watershed regions with downslope drainage system In the TWI, the specific catchment area is calculated by dividing the total catchment region

6.4.10 Soil Erodibility Factor “Soil erodibility (K)” can be used to represent the influence of soil features and soil factors on erosion and related sedimentation since it depicts the physiochemical assets of the soil utilizing equations connected to soil texture, soil organic matter, and proportions of sand, silt and clay. Moreover, the K factor is determined by the permeability of the soil as well as the particle size distribution. Through the soil loss rate per kinematic energy of rainfall erosivity index, the K factor is significantly connected to the R factor. With the use of the following equation, the K factor has been quantified: ) ( 1-Sil K ¼ 0:0137 × 0:2 þ 0:3 × e½-0:0256×San×ð 100 Þ] ( )0:3 Sil × Cla þ Sil [ ] 0:25 × TOC × 1TOC þ eð3:72-2:95×TOCÞ ] [ 0:7 × SN1 × 1SN1 þ eð22:9×SN1 -5:51Þ

ð6:10Þ where “K is the Soil Erodibility, San is the percentage of sand, Sil is the percentage of silt, Cla is the percentage of clay and SN1 is the 1 − San/100” (Chakrabortty et al. 2020c).

98

6.5

6

Sediment Yield Estimation

The sediment yield of this region has been estimated with the help of fuzzy logic and analytical neural network (ANN) model.

6.5.1 Fuzzy Logic One of the methods used to deal with complicated situations is Zadeh’s fuzzy set theory, which he proposed in 1965 (Zadeh 1996). As a result, the fuzzy set theory had been widely applied in variety range of scientific fields (Pradhan 2010). Fuzzy logic works by considering map spatial objects as members of a set. The value of membership of each member is 1, and non-membership value is 0. In this case, value varies between 0 and 1 and reflects the certainty of membership (Zadeh 1984). The membership function is used in fuzzy set theory to represent the grade of membership with regard to a certain characteristic of concern (de Oliveira 1999). The function may be defined as a table that ties map classes to membership grades, and the attribute of concern is frequently tested over discrete intervals with maps. The fuzzy and function are the same as a Boolean AND (logical intersection) function on a conventional set of values (1,). It has a precise definition: lcombinition ¼ MINðlA; lB; lC; . . .Þ ð6:11Þ where “combination is the computed fuzzy membership feature, lA is the membership grade for map A at a certain place, lB is the value for map B and so on. The output membership values of the fuzzy or are controlled by the maximum values of any of the input maps, similar to the Boolean OR (logical union)”. The term “fuzzy or” is defined as follows: lcombinition ¼ MAXðlA; lB; lC; . . .Þ ð6:12Þ The fuzzy algebraic product has the following definition:

Potential Sediment Yield Estimation Using Machine …

lcombinition ¼

n Y i¼1

l1 ;

ð6:13Þ

where “l1 is the ith map’s fuzzy membership criterion and i ¼ 1; 2; . . .; n maps are to be joined”. A parallel to the fuzzy algebraic products is the fuzzy algebraic sum, which would be described as: lcombinition ¼ 1 -

n Y ð1 - l1 Þ

ð6:14Þ

i¼1

The gamma operation is defined by the fuzzy algebraic product and the fuzzy algebraic sum: lcombinition ¼ ðfuzzy algebric sumÞk × ðfuzzy algebric productÞ1-k ð6:15Þ where “k is a factor in the range (0, 1), and Formulas (6.12) and (6.13) are used to determine the fuzzy algebraic summation and fuzzy algebraic item, accordingly” (Pradhan 2010).

6.5.2 Analytical Neural Network “Backpropagation artificial neural network (BPANN)” is the node of the input layer where it takes the data input, evaluate it and send the results to the nodes of the subsequent hidden layer(s), and from the last “hidden layer” goes to the “output layer” (Paola and Schowengerdt 1995). The input layer nodes receive the information in a feed-forward BPANN algorithm (input). A sigmoid activation function is used to produce the weighted sum pertaining to every junction of another layer, which is then sent to the next layer. Changing the weights of interconnections, the error (E) detected at the output which is propagated back to “hidden layer(s)” and then to the “input layer”. The word “error” (E) has the following definition:

6.6 Results

99

E¼

k [ ]2 1X dðkÞ - OðkÞ 2 1

ð6:16Þ

where “dðkÞ is the observed output at the output layer’s kth node OðkÞ is the predicted output at the output layer’s kth node”. The following expression is used to update weights during all iterations: WðijÞn þ 1 ¼ WðijÞn þ DWðijÞn

ð6:17Þ Fig. 6.3 Structure of the ANN model

The convergence speed is usually boosted by incorporating a momentum component b and the impact of previous weight changes, as seen below: WðijÞn þ 1

SeðijÞ ¼

[ ] ¼ WðijÞn þ DWðijÞn þ b WðijÞn - WðijÞn-1 ð6:18Þ

In the path of negative slope, the variation in weights (DW) is indicated by: DWðijÞ ¼ -a

@E @EðijÞ

SeðijÞ

WðiijÞ ¼0

Z

WðfijÞ

I

a

WðfijÞ WðfijÞ - WðiijÞ

ð6:21Þ

where I and f are the starting and end weight values, respectively. The generalized structure of the ANN is shown in Fig. 6.3. The structure of the ANN is consisting of input, output and hidden layers.

ð6:19Þ

where “a is the learning rate so that 0\a\1 and controls the rate at which weights vary”. An “A information criterion (AIC)” (Akaike 1974), a “B information criterion (BIC)” (Rissanen 1978) or network pruning also can be used to attain model simplicity (Karnin 1990). For determining the amount of free factors, AIC and BIC criteria use the “root mean square error (RMSE)” statistic, which penalizes having many free parameters (Xu and Li 2002). Karnin (1990) proposed that while pruning, the sensitivity of errors SeðijÞ with regard to weight WðijÞ be used to eliminate the relevant weight without doing extra computations. The following is the definition of SeðijÞ : @E ¼@EðijÞ

it DW 2 X ðijÞ

ð6:20Þ

WðfijÞ

Ultimately, the sensitivity of the error to weight is as follows:

6.6

Results

6.6.1 Application of Fuzzy Logic The sediment yield has been calculated with the help of fuzzy logic in GIS environment. The sediment yield raster has been reclassified into different classes, i.e. “very low, low, moderate, high and very high”. The considerable parameters in this model and its associated matrix are shown in Table 6.3. The geometric mean of the considerable parameters has been estimated and is shown in Table 6.4. The rank of different themes and associated normalized weight have been estimated for determining the final outcomes with regard to sediment yield estimation (Table 6.5). The spatial coverage of very high, high and moderate sediment yield is mainly concentrated in the eastern, middle and southern portion of this region. The remaining portion of this region is associated with very low and low sediment yield (Fig. 6.4).

0

4

1

0

5

5

0

2

0

3

1

0

4

4

0

1

Stream power index

Topographical wetness index

Geology

Slope

Rainfall and runoff erosivity

Elevation

Land use/land cover

3

1

6

6

1

0

0

2

3

0

0

0

1

0

1

3

4

0

0

0

2

1

1

4

5

0

1

1

3

1

0

1

0

4

3

0

0

0

1

0

0

1

1

5

4

0

0

0

1

1

0

Soil erodibility factor

1

1

6

5

0

0

1

1

1

1

0

0

4

2

0

0

1

2

2

0

0

0

5

3

0

0

1

3

3

0

1

0

6

4

0

0

1

4

4

0

Stream power index

2

1

4

5

0

1

4

3

2

1

3

1

5

6

1

1

5

4

3

1

4

1

6

7

1

0

5

4

1

1

1 1

3

4

4

2

2

1

6

5

1

2

4

5

5

3

Geology

6

5

4

1

Topographical wetness index

3

1

7

6

1

3

5

6

6

4

0

0

2

1

0

0

0

0

0

0

Slope

0

0

3

1

0

0

0

0

0

0

0

0

4

1

0

0

1

0

0

0

0

0

1

0

0

0

0

0

0

0

0

0

1

0

0

0

0

0

0

0

Rainfall and runoff erosivity

0

0

1

1

0

0

0

0

1

0

3

1

4

3

1

1

4

1

1

2

4

1

5

4

2

1

5

2

2

3

Elevation

5

1

6

5

3

1

6

3

3

4

1

0

3

4

0

0

2

1

2

0

1

0

0 1

5

6

1

1

4

1

4

1

4

5

1

0

3

1

3

1

Land use/land cover

6

1

5

1

1

0

Soil erodibility factor

1

0

4

3

Soil texture

5

1

1

Drainage density

1

Soil texture

Drainage density

Table 6.3 Considerable parameters for estimating the sediment yield

100 Potential Sediment Yield Estimation Using Machine …

6.7 Sensitivity Analysis

101

Table 6.4 Geometric mean of the parameters Geometric mean 1

Total

2

3

0.453799

0.572009

0.748984

1.048122

1.463259

1.987134

0.85134

1.148698

1.463259

0.971642

1.244184

1.603671

0.371151

0.453799

0.561675

0.2861

0.375381

0.524061

2.309662

2.938489

3.612012

3.011932

3.827634

4.612054

0.317127

0.399736

0.567624

0.645195

0.876187

1.141309

10.26607

13.29938

16.82178

Total−1

0.0974

0.075191

0.059447

Ascending order

0.059447

0.075191

0.0974

Table 6.5 Rank of different themes and associated normalized weight 1

2

3

Mi

Ni (weight)

Rank

0.026977

0.04301

0.072951

0.047646

0.04312371

7

Drainage density

0.062307

0.110024

0.193547

0.121959

0.11038379

3

Soil texture

0.050609

0.086372

0.142521

0.093168

0.08432466

5

Soil erodibility factor

0.057761

0.093551

0.156198

0.102503

0.09277435

4

Stream power index

0.022064

0.034122

0.054707

0.036964

0.03345575

8

Topographical wetness index

0.017008

0.028225

0.051044

0.032092

0.0290462

10

0.137302

0.220948

0.35181

0.236687

0.2142218

2

Slope

0.179049

0.287804

0.449214

0.305356

0.27637331

1

Rainfall and runoff erosivity

0.018852

0.030057

0.055287

0.034732

0.03143524

9

Relief

0.038355

0.065881

0.111163

0.09376

0.08486119

6

Land use/land cover

6.6.2 Application of Analytical Neural Network The sediment yield probability of this region has been quantified by considering ANN and final outcomes, and its spatial distribution has been prepared in GIS environment. The same procedure has been followed as mentioned previously to reclassify the final outcomes into different qualitative classes. The spatial distribution of very high and high sediment yield of this region

Parameters

Geology

is mainly found in the middle, eastern and southern part of this region. The very low and low sediment yield of this region is mainly found in rest of the part of this region (Fig. 6.5).

6.7

Sensitivity Analysis

The information about the nature of erosion and its associated sedimentation was collected during the field visit (Fig. 6.6). The validation of the

102

6

Potential Sediment Yield Estimation Using Machine …

Fig. 6.4 Spatial distribution of sediment yield using fuzzy logic

predicted outcomes has been done with considering “area under curve (AUC)” value of “receiver operating characteristics curve (ROC)”. The “ROC” curve method, a standard procedure for model evaluation, measures the accuracy of a model (Moayedi et al. 2019; Yuan and Moayedi 2019a; Wang et al. 2020). The ROC curve is plotted between FPR and TPR. The FPR or sensitivity is assigned to X-axis and TPR on Yaxis (Mallick et al. 2019). The “area under the curve (AUC) of ROC (AUROC)” determines the accuracy of the predictions generated by a model (Chao et al. 2018; Nguyen et al. 2019; Yuan and Moayedi 2019b). Fressard et al. (2014) have presented the ROC values into four categories with different levels of accuracy implications, viz. 0.6–0.7 stands for poor accuracy; 0.8–0.9 means fair accuracy; 0.8–0.9 refers to good level

of accuracy; and 0.9–1.0 means excellent accuracy. The FPR and TPR are calculated: FPR ¼ Sensitivity ¼

FP FP þ TN

TPR ¼ 1 - Specificity ¼

TP TP þ FN

ð6:22Þ ð6:23Þ

where FP and TP are the number of false positive and true positive cases and TN and FN are true negative and false negative cases, respectively (Mallick et al. 2019). The AUC values with considering the training datasets in FL and ANN model are 0.93 and 0.95, respectively. On the other side, the values of AUC in FL and validation datasets of ANN model are 0.92 and 0.94, respectively (Fig. 6.7).

6.8 Discussion

103

Fig. 6.5 Spatial distribution of sediment yield using ANN

In terms of predictive capacity, the ANN model is slight better than FL, though both of the models are presented with higher accuracy.

6.8

Discussion

Investigations of “reservoir sedimentation”, “river morphology”, “soil and water conservation was planning”, “water quality modelling” and the development of effective erosion frameworks all require assessments of sediment production (Jain and Das 2010). “Physically based models” and “lumped models” are the two types of models for estimating sediment yields that are present in the literature (Pandey et al. 2016). Balanced ecosystems, which include soil, water and plant habitats, are critical for humanity’s existence and

well-being (De Groot et al. 2002). However, ecosystems in many regions of the world, including certain portions of India, have been disrupted in the past owing to overexploitation (Kothyari 1996). The ecosystem’s ensuing imbalance manifests itself in a variety of negative consequences, such as soil depletion, high incidence of severe floods and so on. Due to rapid soil erosion driven by the aforementioned and other reasons, vast swaths of land across the country have been irrevocably turned into infertile surfaces. These deteriorated land surfaces also became a source of natural water contamination. Rivers provide drinking water as well as agriculture, hydropower generation and navigation (Sadoff and Grey 2002). Reservoirs are constructed to store water and so enhance the

104

6

Potential Sediment Yield Estimation Using Machine …

Fig. 6.6 Filed information about the erosion and its associated sedimentation

quantity of accessible water (Jackson et al. 2001). The building of reservoirs is a common feature of hydropower projects. The extreme erosion and its associated sedimentation is the most influential aspect in subtropical region (Chakrabortty et al. 2020a). The aggravation is due to the deposition of eroded material from upland regions in the downstream portions of rivers. As a consequence, the floodplain area of rivers has grown, clearance beneath bridges and culverts has decreased, and reservoirs have sedimented. Soil erosion, sediment transport and sediment deposition are also key issues for the Indian sub-Himalayan region’s various hydroelectric plants (Kothyari 2011). Soil eroded from the catchment’s upland parts is transported downwards in the rivers, aggravating the problem. The primary source of floods and sedimentation in the Gangetic plains’ lowlands is generally cited as deforestation in headwater

regions. The extension of river flood plains, bridges and culverts blocked, and reservoir storage is lost as a result of this deposition (Rudra et al. 2018). The sediment yield is calculated considering the sediment delivery ratio and the potential erosion (Kothyari and Jain 1997). Erosion differs from sediment load in that the earlier relates to the sediment determined in the waterbodies over a specified timeframe, while the latter relates to the prospective erosion, which is similar to the total of “sheet (upland) erosion” and “channel erosion”. The delivery ratio approach, according to Wolman (1977), covers a range of procedures that influence the periodic or continuous sediments deposited in an eroding watershed, since these activities are extremely variable, discontinuous and only quantitatively explainable. Before a catchment’s outflow, a portion of the eroded soil is placed inside of it. “Sediment delivery ratio” is defined

6.9 Conclusion

Fig. 6.7 ROC curve for training (a) and validation (b) datasets

as the proportion of sediment load to overall “surface erosion” (Van Rompaey et al. 2001). Understanding the relationship between erosion from the land mass and movement through streams will need a significant amount of research. The notion of sediment transportation is reasonable. However, as Glymph (1975) pointed out, deciding the position of sediment references in different regions, such as upland surface or gully erosion and bank erosion, necessitates inferences combined with models that can be considered to describe, at the very least, the inputs of the components evaluated in river

105

systems. The topic of time is maybe as crucial as the matter of origins and distribution. Arid or semi-arid environments are more prone to periodic transport mechanisms than humid environments (Hillier 1995). Small watersheds in humid areas act similarly to periodic or temporary streams in dry areas (Wolman 1977). A variety of effective mental models assume that the erosion and transportation processes on the environment are in some type of equilibrium. Such stability or open systems indicate that modifications take place throughout time (Renwick 1992). Different temporal constants apply to various dimensions or aspects of the environment, as well as the same characteristics in diverse environments (Maddock Jr 1970). In this work, we try to estimate the sediment yield estimation with considering FL and ANN models and its spatial variations have been quantified with the help of GIS environment. The natural breaks method in GIS environment has been considered for reclassifying it into different qualitative classes. The middle, eastern and south-eastern part of the region is mainly prone to very high, high and moderate sediment yield potential zones. According to predictive capacity, the ANN model is slight better than FL model. Because many of the relationships between inputs and outputs in real life are both nonlinear and intricate, ANNs can learn and represent nonlinear and complex connections (Kan et al. 2015). ANNs may infer unknown associations on unknown data after comprehending the original inputs and their relationships, allowing the model to generalize and predict on unknown data (Mendez et al. 2019).

6.9

Conclusion

The problem of erosion and land degradation is one of the major problems which is largely associated with disruption of the land resources. The subtropical monsoon-dominated region is mostly facing this type of scenario. One of the most likely causes of large-scale erosion and sedimentation is short-duration storm rainfall. So, the identification of the most vulnerable

106

6

regions in terms of sediment yields potentiality for adopting the suitable management strategies. In this research, the fuzzy logic (FL) and analytical neural network (ANN) have been considered for quantifying the sediment yield potentiality zones. The spatial variation of the output raster has been quantified with the help of GIS environment. The application of probability modelling in GIS environment is not only less time-consuming but efficient in terms of accuracy and optimal capacity. In terms of predictive ability, the ANN model is comparatively better than FL. The future researcher’s key objective will be to estimate this sort of scenario while taking into account the maximal causal factors and climatic uncertainty. This sort of data aids the decision-maker in deciding on the best course of action. A conceptual way of assessing the demands for sedimentation technical expertise may or not align with contemporary societal existing concerns and goals. It is undeniable that the greater the body of understanding, the better scientists will be able to serve society’s needs by giving more exact evaluations of policy possibilities. There is a pressing necessity research on the origins, modes of movement, and sinks of organic and other materials originating from “point and nonpoint sources”. In the predicted exercises of “environmental impact assessment”, the extremely variable temporal and geographical nature of eroding and sedimentation mechanisms should be recognized. Field observation is the sole way to validate or evaluate the prediction ability of our existing information.

References Akaike H (1974) A new look at the statistical model identification. IEEE Trans Autom Control 19:716–723 Amare T, Zegeye AD, Yitaferu B et al (2014) Combined effect of soil bund with biological soil and water conservation measures in the northwestern Ethiopian highlands. Ecohydrol Hydrobiol 14:192–199 Avanzi GD, Giannecchini R, Puccinelli A (2004) The influence of the geological and geomorphological settings on shallow landslides. An example in a temperate climate environment: the June 19, 1996 event in northwestern Tuscany (Italy). Eng Geol 73:215–228

Potential Sediment Yield Estimation Using Machine … Bandyopadhyay S (2007) Evolution of the Ganga Brahmaputra delta: a review. Geogr Rev India 69:235–268 Bardgett R (2005) The biology of soil: a community and ecosystem approach. Oxford University Press Bělohlávek R, Dauben JW, Klir GJ (2017) Fuzzy logic and mathematics: a historical perspective. Oxford University Press Beskow S, Mello CR, Norton LD et al (2009) Soil erosion prediction in the Grande River Basin, Brazil using distributed modeling. CATENA 79:49–59 Chakrabortty R, Pal SC, Chowdhuri I et al (2020a) Assessing the importance of static and dynamic causative factors on erosion potentiality using SWAT, EBF with uncertainty and plausibility, logistic regression and novel ensemble model in a sub-tropical environment. J Indian Soc Remote Sens 48:765–789. https://doi.org/10.1007/s12524-020-01110-x Chakrabortty R, Pal SC, Sahana M et al (2020b) Soil erosion potential hotspot zone identification using machine learning and statistical approaches in eastern India. Nat Hazards. https://doi.org/10.1007/s11069-020-04213-3 Chakrabortty R, Pradhan B, Mondal P, Pal SC (2020c) The use of RUSLE and GCMs to predict potential soil erosion associated with climate change in a monsoon-dominated region of eastern India. Arab J Geosci 13:1–20 Chakrapani G, Subramanian V (1993) Rates of erosion and sedimentation in the Mahanadi river basin, India. J Hydrol 149:39–48 Chamberlain EL, Goodbred SL, Hale R et al (2020) Integrating geochronologic and instrumental approaches across the Bengal Basin. Earth Surf Process Landf 45:56–74 Chao L, Zhang K, Li Z et al (2018) Geographically weighted regression based methods for merging satellite and gauge precipitation. J Hydrol 558:275–289 De Groot RS, Wilson MA, Boumans RM (2002) A typology for the classification, description and valuation of ecosystem functions, goods and services. Ecol Econ 41:393–408 de Oliveira JV (1999) Semantic constraints for membership function optimization. IEEE Trans Syst Man Cybern Part A Syst Hum 29:128–138 De Vente J, Poesen J (2005) Predicting soil erosion and sediment yield at the basin scale: scale issues and semi-quantitative models. Earth Sci Rev 71:95–125 Dietrich WE, Reiss R, Hsu M, Montgomery DR (1995) A process-based model for colluvial soil depth and shallow landsliding using digital elevation data. Hydrol Process 9:383–400 Fressard M, Thiery Y, Maquaire O (2014) Which data for quantitative landslide susceptibility mapping at operational scale? Case study of the Pays d’Auge plateau hillslopes (Normandy, France). Nat Hazards Earth Syst Sci 14:569–588. https://doi.org/10.5194/nhess14-569-2014 Ganasri B, Ramesh H (2016) Assessment of soil erosion by RUSLE model using remote sensing and GIS—a case study of Nethravathi Basin. Geosci Front 7: 953–961

References Garrido A (2012) A brief history of fuzzy logic. BRAIN Broad Res Artif Intell Neurosci 3:71–77 Glymph L (1975) Evolving emphases in sediment-yield predictions. In: Present and prospective technology for predicting sediment yields and sources. Agricultural Research Service Pub. ARS-S-40-61-65, Washington, DC, pp 1–4 Grayson RB, Moore ID, McMahon TA (1992) Physically based hydrologic modeling: 2. Is the concept realistic? Water Resour Res 28:2659–2666 Hamel P, Falinski K, Sharp R et al (2017) Sediment delivery modeling in practice: comparing the effects of watershed characteristics and data resolution across hydroclimatic regions. Sci Total Environ 580:1381– 1388 Harel M-A, Mudd S, Attal M (2016) Global analysis of the stream power law parameters based on worldwide 10 Be denudation rates. Geomorphology 268:184–196 Hillier S (1995) Erosion, sedimentation and sedimentary origin of clays. In: Origin and mineralogy of clays. Springer, pp 162–219 Holland HD (1978) The chemistry of the atmosphere and oceans Issaka S, Ashraf MA (2017) Impact of soil erosion and degradation on water quality: a review. Geol Ecol Landsc 1:1–11. https://doi.org/10.1080/24749508. 2017.1301053 Jackson RB, Carpenter SR, Dahm CN et al (2001) Water in a changing world. Ecol Appl 11:1027–1045 Jain MK, Das D (2010) Estimation of sediment yield and areas of soil erosion and deposition for watershed prioritization using GIS and remote sensing. Water Resour Manage 24:2091–2112 Jie C, Jing-Zhang C, Man-Zhi T, Zi-Tong G (2002) Soil degradation: a global problem endangering sustainable development. J Geogr Sci 12:243–252 Kan MS, Tan AC, Mathew J (2015) A review on prognostic techniques for non-stationary and non-linear rotating systems. Mech Syst Signal Process 62:1–20 Karnin ED (1990) A simple procedure for pruning backpropagation trained neural networks. IEEE Trans Neural Netw 1:239–242 Keesstra SD, Bouma J, Wallinga J et al (2016) The significance of soils and soil science towards realization of the United Nations Sustainable Development Goals. Soil Kothyari UC (1996) Erosion and sedimentation problems in India. IAHS Publ Ser Proc Rep Int Assoc Hydrol Sci 236:531–540 Kothyari UC (2011) Sediment problems and sediment management in the Indian sub-Himalayan region. In: Sediment problems and sediment management in Asian river basins, 349 Kothyari UC, Jain SK (1997) Sediment yield estimation using GIS. Hydrol Sci J 42:833–843 Kumar S (2004) Neural networks: a classroom approach. Tata McGraw-Hill Education Lal R (2015) Restoring soil quality to mitigate soil degradation. Sustainability 7:5875–5895

107 Liu B, Guo S, Li Z et al (2013) Sampling program of water erosion inventory in China. Soil Water Conserv China 10:26–34 Liu D, She D, Shao G, Chen D (2015) Rainfall intensity and slope gradient effects on sediment losses and splash from a saline–sodic soil under coastal reclamation. CATENA 128:54–62 Maddock T Jr (1970) Indeterminate hydraulics of alluvial channels. J Hydraul Div 96:2309–2323 Mallick J, Khan RA, Ahmed M et al (2019) Modeling groundwater potential zone in a semi-arid region of Aseer using fuzzy-AHP and geoinformation techniques. Water (Switzerland) 11. https://doi.org/10. 3390/W111226562656 Mendez KM, Broadhurst DI, Reinke SN (2019) The application of artificial neural networks in metabolomics: a historical perspective. Metabolomics 15:1–14 Moayedi H, Tien Bui D, Gör M et al (2019) The feasibility of three prediction techniques of the artificial neural network, adaptive neuro-fuzzy inference system, and hybrid particle swarm optimization for assessing the safety factor of cohesive slopes. ISPRS Int J Geo-Inf 8:391 Mofidi S, Mohammadi Torkashvand A, Moeini A et al (2021) Relationship between fractal dimensions of stream and morphometric characteristics of basin for the soil conservation from water erosion. Int J River Basin Manag 1–11 Mukherjee A, Fryar AE, Thomas WA (2009) Geologic, geomorphic and hydrologic framework and evolution of the Bengal basin, India and Bangladesh. J Asian Earth Sci 34:227–244 Nguyen H, Mehrabi M, Kalantar B et al (2019) Potential of hybrid evolutionary approaches for assessment of geo-hazard landslide susceptibility mapping. Geomat Nat Hazards Risk 10:1667–1693 Pandey A, Himanshu SK, Mishra SK, Singh VP (2016) Physically based soil erosion and sediment yield models revisited. CATENA 147:595–620 Paola JD, Schowengerdt RA (1995) A review and analysis of backpropagation neural networks for classification of remotely-sensed multi-spectral imagery. Int J Remote Sens 16:3033–3058 Pimentel D (2006) Soil erosion: a food and environmental threat. Environ Dev Sustain 8:119–137. https://doi. org/10.1007/s10668-005-1262-8 Pradhan B (2010) Application of an advanced fuzzy logic model for landslide susceptibility analysis. Int J Comput Intell Syst 3:370–381 Puigdefábregas J, Sole A, Gutierrez L et al (1999) Scales and processes of water and sediment redistribution in drylands: results from the Rambla Honda field site in Southeast Spain. Earth Sci Rev 48:39–70 Rajbanshi J, Bhattacharya S (2020) Assessment of soil erosion, sediment yield and basin specific controlling factors using RUSLE-SDR and PLSR approach in Konar river basin, India. J Hydrol 587:124935 Ravi S, Breshears DD, Huxman TE, D’Odorico P (2010) Land degradation in drylands: interactions among

108

6

hydrologic–aeolian erosion and vegetation dynamics. Geomorphology 116:236–245 Ray S, Mohanti M, Somayajulu B (1984) Suspended matter, major cations and dissolved silicon in the estuarine waters of the Mahanadi river, India. J Hydrol 69:183–196 Renwick WH (1992) Equilibrium, disequilibrium, and nonequilibrium landforms in the landscape. Geomorphology 5:265–276 Rissanen J (1978) Modeling by shortest data description. Automatica 14:465–471 Roy P, Chandra Pal S, Chakrabortty R et al (2020) Threats of climate and land use change on future flood susceptibility. J Clean Prod 272:122757. https://doi. org/10.1016/j.jclepro.2020.122757 Rudra K, Rudra L, Leuchner (2018) Rivers of the GangaBrahmaputra-Meghna delta. Springer Sadoff CW, Grey D (2002) Beyond the river: the benefits of cooperation on international rivers. Water Policy 4:389–403 Sarin M, Krishnaswami S (1984) Major ion chemistry of the Ganga-Brahmaputra river systems, India. Nature 312:538–541 Schiefer E, Slaymaker O, Klinkenberg B (2001) Physiographically controlled allometry of specific sediment yield in the Canadian Cordillera: a lake sedimentbased approach. Geogr Ann Ser A Phys Geogr 83: 55–65 Shi Z, Ai L, Fang N, Zhu H (2012) Modeling the impacts of integrated small watershed management on soil erosion and sediment delivery: a case study in the Three Gorges Area, China. J Hydrol 438:156–167 Tamene L, Park SJ, Dikau R, Vlek PLG (2006) Reservoir siltation in the semi-arid highlands of northern Ethiopia: sediment yield–catchment area relationship and a semi-quantitative approach for predicting sediment yield. Earth Surf Process Landf 31:1364–1383. https://doi.org/10.1002/esp.1338 Tilahun SA, Guzman C, Zegeye A et al (2013) An efficient semi-distributed hillslope erosion model for the subhumid Ethiopian Highlands. Hydrol Earth Syst Sci 17:1051–1063

Potential Sediment Yield Estimation Using Machine … Van Rompaey AJ, Verstraeten G, Van Oost K et al (2001) Modelling mean annual sediment yield using a distributed approach. Earth Surf Process Landf 26:1221–1236 Verstraeten G, Poesen J, de Vente J, Koninckx X (2003) Sediment yield variability in Spain: a quantitative and semiqualitative analysis using reservoir sedimentation rates. Geomorphology 50:327–348 Walling DE, Webb B (1983) Patterns of sediment yield, pp 69–100 Wang S, Zhang K, van Beek LP et al (2020) Physicallybased landslide prediction over a large region: scaling low-resolution hydrological model results for highresolution slope stability assessment. Environ Model Softw 124:104607 Wolman MG (1967) A cycle of sedimentation and erosion in urban river channels. Geogr Ann Ser A Phys Geogr 49:385–395 Wolman MG (1977) Changing needs and opportunities in the sediment field. Water Resour Res 13:50–54 Xu Z, Li J (2002) Short-term inflow forecasting using an artificial neural network model. Hydrol Process 16:2423–2439 Yin S, Zhang W, Xie Y et al (2013) Spatial distribution of rainfall erosivity in China based on high-density station network. Soil Water Conserv 10:45–51 Yuan C, Moayedi H (2019a) Evaluation and comparison of the advanced metaheuristic and conventional machine learning methods for the prediction of landslide occurrence. Eng Comput 1–11 Yuan C, Moayedi H (2019b) The performance of six neural-evolutionary classification techniques combined with multi-layer perception in two-layered cohesive slope stability analysis and failure recognition. Eng Comput 1–10 Zadeh LA (1984) Making computers think like people [fuzzy set theory]. IEEE Spectr 21:26–32 Zadeh LA (1996) Fuzzy sets. In: Fuzzy sets, fuzzy logic, and fuzzy systems: selected papers by Lotfi A. Zadeh. World Scientific, pp 394–432 Zhang W, Fu J (2003) Rainfall erosivity estimation under different rainfall amount. Resour Sci 25:35–41

7

Impact of Climate and LULC Change on Soil Erosion

Abstract

Keywords

In the present study, the potential impact of “climate” and “land use and land cover (LULC)” change on soil erosion has been estimated. In this perspective, the future rainfall scenario in the projected period and LULC dynamics in the same period has been estimated. The pattern, intensity and amount of rainfall are always changing which has been established by different researchers. The “average annual soil erosion” for the base period and projected period has been estimated with considering “Revised Universal Soil Loss Equation (RUSLE)”. The “cellular automata (CA”) and “Markov chain (MC)” model has been used for projecting the future LULC in projected period. The selection of the suitable “general circulation model (GCM)” model has been done with the help of statistical analysis. That is very confusing task in any type of climatic modelling with considering suitable “GCM”. Then the combine impact of changing climate and “LULC” has been considered for future soil erosion modelling. There is a rising pattern of “soil erosion” from base year to projected period has been confirmed in this research. This type of information is useful to the regional planner to take the suitable remedies in keeping in the view of sustainable and long term planning.

Future rainfall scenario Cellular automata Markov chain GCM Soil erosion modelling

.

7.1

. .

.

Introduction

“Soil erosion” is a worldwide issue that has severe consequences for food production, water security and ecosystems (Borrelli et al. 2020). “Soil erosion” will be influenced by climate change in two ways: primarily via variations in precipitation volume and intensity, and secondarily by alterations in land cover and soil organic content decomposition processes (Duulatov et al. 2019). If the total of mobilizing pressure exerted on particles is higher than the total of frictional resistance, erosion will happen. No particles are removed from of the soil surface if this is not the case (Routschek et al. 2014). The quantity of material that may be separated from the soil surface or the flow’s transit capacity are both factors that restrict erosion (Knapen et al. 2007). A turbulence, vertical flow element should fight the resting of the particles for accumulation in order to convey unattached fragments (Routschek et al. 2014). Agricultural land use, particularly extensive farmland usage, has resulted in substantial transformations in rural regions during the previous few decades. Water and wind erosion capability of soils has

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 S. C. Pal and R. Chakrabortty, Climate Change Impact on Soil Erosion in Sub-tropical Environment, Geography of the Physical Environment, https://doi.org/10.1007/978-3-031-15721-9_7

109

110

substantially risen caused by human activity and alterations in land use and crop types, through application of large farm equipment, and land consolidation (Klik and Eitzinger 2010). The global climate change has been experiencing a severe problem and associated with various negative impacts on natural environment. Human activity has heating the atmosphere, oceans, and land at such an astounding level, according to the latest “Intergovernmental Panel on Climate Change (IPCC)” climate assessment, authored by a various team of 234 authors from 66 nations (Nevitt 2020; Roös 2021). Unless major reduction in greenhouse gas emissions happen in the subsequent decades, global warming of 1.5–2 °C will be reached in the twenty-first century (DeConto et al. 2021). The changing pattern of rainfall can be impacted on soil erosion by influencing the rainfall and runoff erosivity. “Regional weather circulation” characteristics are influenced by an increase in “water vapour” in the “atmosphere” (Guan et al. 2019). As a result of the changing circulation patterns, the quantity, severity and timing of severe rainstorms will be altered throughout the year (Westra et al. 2014). The storm rainfall event with high kinetic energy can enhances the rate of erosion (Fornis et al. 2005). “Climate change” has an impact on a variety of “physical” and “chemical” characteristics of soil, as well as “infiltration and erosion mechanisms” (Li and Fang 2016). “Soil moisture regime, organic carbon percentage and vegetation cover” are the three most climate-sensitive components. Plants and plant wastes modify the soil surface canopy cover for a variety of reasons: Planting and harvest dates will be changed due to a shift in phenology (Bisbis et al. 2018). “New crops, agricultural rotations and management practices” will be used to adapt to the changing climate regime. Plant biomass output is expected to change during the same time period. Land use alterations, as well as changes in land use structure, are compelled by climate change. In terms of rainfall volume, intensity, and spatio-temporal patterns, alteration in rainfall patterns are prone to have a direct influence on “soil erosion” (Wei et al. 2009). Nearing (2001) explored the effects of “climate change” on

7

Impact of Climate and LULC Change on Soil Erosion

rainfall patterns in the “USA”, using the output of “global circulation models (GCMs)” and numerical erosivity connections from the “Revised Universal Soil Loss Equation (RUSLE)” to quantify climate change-induced changes in precipitation erosional power. The consequences of changing rainfall temporal distributions are frequently paired to land use impacts. Soil erosion is likely to be exacerbated if precipitation time shifts to coincide with growing seasons of plants, or if rainfall rises throughout dry season when vegetation cover is lesser, due to less protection and more precipitation (Zhang et al. 2012). Extended rainfall periods may also restrict sun radiation, reducing plant development and resulting in less coverage and greater soil erosion. Increasing air temperatures impact the soil water balances by affecting evaporation and transpiration (Hatfield and Dold 2019). Furthermore, climatic change alters growth circumstances, which has a wide assortment of positive and negative consequences on soil erosion (Beniston et al. 2007). Soil cover by vegetation structure and “crop residues”, as well as “root development” and the amount of “organic matter” in the soil, are all important parameters (Bot and Benites 2005). Higher “rainfall erosivity” alters “hydrological processes”, resulting in increased “surface runoff” and “soil erosion, even if land use does not alter” (Yang et al. 2003). Besides infiltration surplus runoff, another runoff generating process that can induce gully erosion is saturated surplus runoff. Surplus runoff due to saturation takes place once rain falls on already saturated soil. As a result, it is mostly found in humid locations with long duration and low precipitation. Indirectly, increased temperatures can cause soil erosion in a variety of ways. As atmospheric CO2 concentrations and temperatures rise, evapotranspiration rates rise and soil moisture falls, increasing the “soil’s infiltration capacity” and reducing “runoff” and “soil loss” (Pruski and Nearing 2002a). Furthermore, enhanced “CO2 concentration” in the atmosphere may lead to higher plant biomass production, which aids in canopy interception and reduces runoff and soil erosion. Rainfall erosivity is calculated using rainfall intensity,

7.2 Materials and Methods

which “global climate models (GCMs)” and even “regional climate models (RCMs)” struggle to represent (Panagos et al. 2017). “Climate models” are mostly not considered to determine “rainfall erosivity” because it depend on rainfall intensity, which “global climate models (GCMs)” and even “regional climate models (RCMs)” find it difficult to depict (Chapman et al. 2021). As a result, there is few research on how erosion may evolve in the future, and those that are accessible are limited in scope. Because generalized convection climate models do not include precipitation data at the temporal framework applied to determine prospective rainfall erosivity, these research findings generally assume that the connection between accumulated climate performance measures and erosivity in the current day will retain in the long term. Many scientists have assessed the effects of expected changes in “precipitation, temperature and CO2 on agricultural yield” (Ciais et al. 2005). Several researches looking at mean and variance differences in rainfall and temperature, and even some of the findings suggested that influenced by climatic variability (as assessed by variance) may have a significant impact on agricultural output (Zhang and Nearing 2005). Variations in rainfall amount or occurrence have been used to assess the effects of “global climate change” on “soil erosion” and “surface runoff” (Zhang et al. 2004). Changes in “land use and land cover (LULC)” have long been a focus of study, and they are among the most acute markers of human–environment interactions. It is linked to the area’s meteorological and geomorphologic characteristics, which has a faster influence on land degradation. Natural and human-induced LULC fluctuations in regional soils have a variety of consequences, including “soil erosion, acidification, nutrient leaching and organic matter loss” (Alkharabsheh et al. 2013). In recent years, several studies have been conducted to estimate the potential impact of “LULC change” on “soil erosion” at various temporal and spatial scales (Wen and Deng 2020). The major goal of this

111

research is to figure out how “climate” and “LULC change” could affect soil erosion. In this perspective, the future rainfall and future “LULC” has been projected.

7.2

Materials and Methods

For measuring “average annual soil erosion” for the base period as well as the future period, the following method was explored. The rainfall scenario and projected “LULC” were chosen for quantifying “soil erosion” in the future period. For determining “soil erosion” in the future, the following method has been proposed (Fig. 7.1).

7.2.1 Soil Erosion Factors The following factor has been considered for estimation of the “average annual soil erosion” in the study region (Fig. 7.2). Rainfall and Runoff Erosivity (R) Factor R factor was determined using the algorithms by Zhang and Fu (2003), Yin et al. (2013) updated this model, which was ultimately employed by Liu et al. (2013). The “R factor” for this region has been calculated using the equation below: R¼

24 X

Rhmk

ð7:1Þ

k¼1

Rhmk ¼

n X m ( ) 1X 1:7265 a:Pi;j;k n i¼1 j¼0

WRhmk ¼

Rhmk R

ð7:2Þ ð7:3Þ

Here, “R is the average annual rainfall erosivity (MJ mm ha−1 h−1), Rhmk is the rainfall erosivity of the kth half-month (MJ mm ha−1 h−1), pi,j,k is the average daily erosive rainfall occurring on the jth day, kth half-month and ith year. WRhmk is the proportion of the rainfall erosivity of the kth halfmonth to the average annual rainfall erosivity”.

112

7

Impact of Climate and LULC Change on Soil Erosion

Fig. 7.1 Methodology flowchart

Slope Length and Steepness (LS) Factor The “LS factor” for this location has been calculated using the equation below (McCool et al. 1989): ( )m ( ) k −0:0256×San×ð1−Sil ½ Þ] 100 L¼ ð7:5Þ K ¼ 0:0137 × 0:2 þ 0:3 × e 22:13 ( )0:3 Sil F × m¼ ð7:6Þ Cla þ Sil 1 þ Fi ] [ 0:25 × TOC sin b=0:0896 × 1− F¼ ð7:7Þ TOC þ eð3:72−2:95×TOCÞ [ ] 3ðsin bÞ0:8 þ 0:56i 0:7 × SN1 × 1− SN1 þ eð22:9×SN1 −5:51Þ ðflowacc þ 625Þðm þ 1Þ −flowaccðm þ 1Þ L¼ 25ðm þ 1Þ × 2:13m ð7:4Þ ð7:8Þ where “K is the soil erodibility, San is the percentage of sand, Sil is the percentage of silt, Cla where “L is the slope length factor, k is the slope is the percentage of clay and SN1 is the length, m is the eroding potentiality in regard 1 − San/100” (Pal et al. 2021; Roy et al. 2022; with the amount of slope in percentage, F is the ratio of rill and inter rill erosion, b is the slope Chakrabortty et al. 2022).

Soil Erodibility (K) Factor Different physical and chemical parameters have been considered for estimating the “K factor” with considering the following equation:

7.2 Materials and Methods

Fig. 7.2 Soil erosion factors; R factor (a), K factor (b), LS factor (c), C factor (d) and P factor (e)

113

114

7

angle (in degree) in GIS environment and flowacc is the flow accumulation”. S ¼ ConððTanðslope × 0:01745Þ\0:09Þ; × ð10:8 × sinðslope × 0:01745Þ þ 0:3Þ; × ð16:8 × sinðslope × 0:01745Þ − 0:5ÞÞ ð7:9Þ LS ¼ L × S

ð7:10Þ

where “L is the slope length, S is the slope steepness and LS is the integration of the slope length and steepness” Renard et al., (1997). Cover And Management (C) Factor The “C factor” is commonly determined in this approach just after three dimensions of canopy covering, surface, and harshness have been applied (Blanco-Canqui and Lal 2010). C is the most important dimensionless element, reflecting the form of soil erosion, which is partly caused by plant cover (Pal and Chakrabortty 2019). The “cover and management (C) factor” has been calculated based on LULC characteristics across the region. Support Practice (P) Factor The “support practice factor” has been quantified according to the slope and its associated support practice values. The percentage of slope in different categories in according to support practice values was considered for estimating the P factor in this region.

Impact of Climate and LULC Change on Soil Erosion

7.2.2 Selection of the Suitable GCM Model for Simulating the Future Rainfall The selection of the suitable GCM model has been done with considering trend (slope “b”), R2 (“observed corrected GCM”) and “Corl R Mod” in statistical analysis. The trend (slope “b”) values of “BCC-CSM2-MR”, “CNRM-CM6-1”, “CNRMESM2-1”, “CanESM5”, “GFDL-ESM4”, “IPSLCM6A-LR”, “MIROC-ES2L”, “MIROC6” and “MRI-ESM2-0” are 86.59, 71.86, 84.5, 66.58, 90.17, 76.32, 64.37, 90.39 and 79.07, respectively. In this perspective, the MIROC6 model has been selected with the help of mentioned approaches (Table 7.1). The extraordinary global warming has been linked to a significant rise in “greenhouse” elements in the atmosphere (Dai et al. 2016). This phenomenon is not only responsible for harsh climatic events of temperature increase, recurrent of floods and droughts but also influence the degradation of forest ecology, retreat of glacier, sea-level changes and more (Mal et al. 2018). We need scenarios of different socioeconomic futures to study climate change forecasts, ramifications and remedies since climate change is so dependent on socioeconomic growth (Riahi et al. 2017). Such socioeconomic scenarios offer a unified set of assumptions concerning societal, technological, cultural and economic changes in the twenty-first century. They frequently centre on a broad theme about the future of humanity to

Table 7.1 Selection of the suitable model GCM “BCC-CSM2-MR”

Trend (slope “b”)

R2 (“Observed corrected GCM”)

“Corl R Mod”

86.59

0.914

0.914

“CNRM-CM6-1”

71.86

0.923

0.923

“CNRM-ESM2-1”

84.5

0.895

0.905

“CanESM5”

66.58

0.954

0.954

“GFDL-ESM4”

90.17

0.961

0.961

“IPSL-CM6A-LR”

76.32

0.893

0.903

“MIROC-ES2L”

64.37

0.978

0.978

“MIROC6”

90.39

0.926

0.926

“MRI-ESM2-0”

79.07

0.939

0.939

7.2 Materials and Methods

115

represent the varied and necessarily unpredictable character of these occurrences. Quantification forecasts of aspects such as population, economic activity and urbanization, which are required as sources to energy, land use, climate effects, and integrated and holistic modelling, are woven into the store. The SSPs are made up of five various stories about the future of the globe, each accompanied by a growing collection of forecasts, such as population, economic activity, urbanization and income disparity. So they were first presented to the research as “pathways”, the SSPs fit within the category of “projections” as outlined in the beginning (Rao et al. 2019). While “pathway” is much more commonly used in the order to describe goal-oriented situations, it was also used to define a subgroup of paths in a multi-dimensional situation.

second. A statistical model that may be used to convert a global scale anomaly that became an anomaly of certain factors of local climate is represented with statistical downscaling concerning the relation among coarse-scale grid (predictor) and small-scale data (response) (Xue et al. 2014). In comparison to RCMs, which involve detailed modelling of physical processes, statistical downscaling approaches are computationally inexpensive. As a result, for institutions that lack the computing resources and technical competence necessary for “dynamical downscaling”, they offer a realistic and sometimes preferable option.

7.2.3 Statistical Downscaling

Extreme rainfall can hasten runoff, resulting in “large-scale erosion” in a short period of time (Cerdà and Rodrigo-Comino 2020). As a result, evaluating the rate of future soil erosion requires estimating the “rainfall and runoff erosivity factor” during the predicted time. As previously stated, the appropriate ensemble “GCM” was used to estimate future rainfall in this study. The rainfall in the “GCM model” has been downscaled using the appropriate downscaling procedures in the RStudio programme. To remove the bias and spread it in the scenario time, the observed information is compared to the control or historical period. The quantile-based approach is examined for this objective, with the given expression utilizing and as a “shape and scale function” in likelihood allocation (Mishra et al. 2018):

The word “downscaling” refers to a technique for producing predictions at smaller sizes using data acquired at larger scales (Ekström et al. 2015). The two most common strategies for downscaling climate data are dynamical and statistical downscaling. Using observational information or lesser climate simulation results as a boundary condition, dynamic downscaling involves running high-resolution climate change models on a regional subdomain (Tang et al. 2016). Such models are computationally costly, yet they employ physical principles to simulate local climates (Rummukainen 2010). “Statistical downscaling” is a two-step process that involves creating correlations between local climatic parameters and large-scale predictors and applying those connections to the results of global climate model try out to represent future local climate attributes (Tisseuil et al. 2010). Downscaling may be done in two ways: utilizing regional data (from a regional climate model, RCM) or using statistical information (obtained from the GCM) (Hosseinzadehtalaei et al. 2021). Statistical dynamic downscaling is the first way, whereas statistical downscaling (SD) is the

7.2.4 Estimating the Rainfall and Runoff Erosivity Factor in Future Period

−1 XGCMPresent−Corrected ¼ Fobs ðFGCMPresent ðXGCM20Present ÞÞ

ð7:11Þ XGCMFuture−Corrected ¼ XGCMFuture F −1 ðFGCMFuture ðXGCMFuture ÞÞ ¼ −1 Obs FGCMPresent ðFGCMFuture ðXGCMFuture ÞÞ ð7:12Þ

116

7

f ðXÞ ¼

1 X X a−1 exp b s ð aÞ a a

ð7:13Þ

Zx f ðtÞdt

FðX Þ ¼

ð7:14Þ

0

7.2.5 LULC Prediction The prevalence of land use categories and how they evolve over time is both a natural and a man-made phenomenon. As a result of changes in land use types, many geoenvironmental conditions have changed and been influenced (Saha et al. 2021). Several academics have previously determined the influence of climate and “LULC change” on future “soil erosion” in order to develop more effective and long-term management strategies (Pal and Chakrabortty 2019; Pal et al. 2021). Therefore, this type of particular field is now popularly investigated by different researchers to for sustainable management practices. The future “LULC prediction map” for the projected time has been created in this study to assess the possible influence of “LULC change” on “soil erosion”. Using ENVI GIS software, the maximum likelihood technique was consider creating a land use map for the base year. The maximum likelihood estimation technique is a parametric classification method which is dependent on the “Gaussian probability density” function’s second-order parameter (Paola and Schowengerdt 1995). The following process is used in the maximum likelihood technique: gi ð X Þ ¼ pðXjwi Þpðwi Þ P−1 pðwi Þ −ð12ÞðX−Ui ÞT ðX−Ui Þ i ∗ e ¼ | | n=2 |P |1=2 ð2pÞ

Impact of Climate and LULC Change on Soil Erosion

depending on spatio-temporal information and cellular automata (CA)-Markov simulation. The Markov technique is a theory-based random process system that is primarily used to analyse land use prediction performance. The “CAMarkov” model is a statistically based probability matrix that uses neighbourhood effects to predict outcomes (Myint and Wang 2006). The Markov model also faces challenges due to the absence of geographical dispersion of events (Nouri et al. 2014). To solve this difficulty, a “CA–Markov model” was created that uses the cellular automata mechanism to extract spatial sense. This cellular automaton functions by changing its location randomly based on neighbourhood influences (Clarke et al. 1997). The following equations have represented the mathematical function of a “Markov” model: Lðt þ 1Þ ¼ Pij ∗ LðtÞ

ð7:16Þ

where “Lðt þ 1Þ and LðtÞ is the land use status at time ðt þ 1Þ and t respectively. The transition probability matrix in a state represent through ( ) Pm 0 ≤ Pij \1 and ði; j ¼ 1; 2; . . .; mÞ ” j¼1 Pij ¼ 1; (Janizadeh et al. 2021; Pham et al. 2021). 2

P11 6 P21 6 . 6 Pij ¼ 6 .. 6 . 4 .. Pm1

P12 P22 .. . .. . Pm2

... ... .. . .. . ...

3 P1m P2m 7 .. 7 7 . 7 .. 7 . 5 Pmm

ð7:17Þ

7.2.6 Estimation of Soil Erosion The “average annual soil erosion” has been estimated with the help of the following equation

i

ð7:15Þ where “ n is the respective band numbers, X indicates the vector data, Ui is the mean vector of class P i and j i j is the covariance matrix of class i”. “TerraSet software” was used to create a “land use change and future land use map” (2100)

A ¼ R × K × LS × C × P

ð7:18Þ

where “A is the average annual soil erosion, R is the rainfall and runoff erosivity factor, K is the soil erodibility factor, LS is the slope length and steepness factor, C is the cover and management factor and P is the support practice factor”.

7.3 Results and Discussion

A ¼ RP × K × LS × CP × P

117

ð7:19Þ

where “A is the average annual soil erosion, R is the simulated rainfall and runoff erosivity factor, K is the soil erodibility factor, LS is the slope length and steepness factor, C is the projected cover and management factor, P is the support practice factor”.

7.3

Results and Discussion

7.3.1 Soil Erosion in Base Period The “soil erosion” raster of this region has been reclassified into several qualitative classifications,

Fig. 7.3 Average annual soil erosion in current year

with considering “natural breaks” lassifier in “GIS environment”. The average annual soil loss ranges from 5.00 to > 35.00 tonnes per hectare per year. The very high “soil erosion”zones of this region (> 35) are predominantly located in the west and south-west. The high soil erosion is mainly located in the western, south-western and central part of this region (25–35). The moderate “soil erosion” zone (15–25) is typically found in the middle part of this region. Low soil erosion zones (5–15) are primarily located in the middle and eastern part of this region. The rest of the part of this region is associated with very low (< 5) “soil” erosion zones (Fig. 7.3). In the time of filed visit, information regarding the erosion has been collected and the nature of erosion also identified (Fig. 7.4).

118

7

Impact of Climate and LULC Change on Soil Erosion

Fig. 7.4 Some major erosion-prone areas of the study region

7.3.2 Estimation of Future Climate Soil erosion is a common occurrence that is mostly influenced by rainfall intensity. Rainfall intensity has a power function-like effect on soil loss. Increased rainfall intensity, as predicted by the IPCC (2001), would likely result in increased soil loss. Extreme erosion episodes raise the risk of harm both on and off-site. In this perspective, the future rainfall scenario and its associated R factor has been estimated. The value of simulated “R factor” for the projected period with considering SSP 126 is 910.60–1700.00. The value of simulated “R factor” for the projected period with considering SSP 245 is 965.32– 1740.00. The value of simulated “R factor” for the projected period with considering SSP 370 is 1010.11–1790.00. The value of simulated “R factor” for the projected period with

considering SSP 585 is 1032.65–1820.00. Here, there is increasing tendency of rainfall and its associated “R factor” with higher SSP and vice versa. The higher “R factor” is associated in the eastern and south-eastern part of this region. Though there is no such differences is found in the overall region (Fig. 7.5).

7.3.3 LULC Prediction The future LULC prediction has been done with considering the CA–Markov model. With considering the previous historical records and 2020s as a base year, the future LULC has been predicted for the projected period, i.e., 2100s. Various LULC is associated in this region, these are, “deciduous broadleaf forest, cropland, builtup land, mixed forest, shrubland, barren land,

7.3 Results and Discussion

119

Fig. 7.5 Projected R factor with considering SSP 126 (a), SSP (245) (b), SSP 370 (c) and SSP 585 (d)

fallow land, wasteland, water bodies, plantations, aquaculture, grassland and permanent wetlands” (Fig. 7.6). In the projected period, there is a decline of forest cover land and increasing of settlement and some extent agricultural land.

7.3.4 Soil Erosion in Projected Period The “soil erosion” for the projected period has been estimated with considering the simulated rainfall and its associated “R factor and projected C factor”. From this analysis, it is found that there is a possibility of higher erosion for the future period. A different soil erosion outcome has been prepared for the projected period with considering “SSP 126, SSP 245, SSP 370 and SSP 585”,

respectively. There is some extent of similarity of “soil erosion” zones in regard with the different SSP scenario, but there is increasing tendency of “soil erosion” has been found with higher SSP scenario. The “average annual soil loss” ranges from 5.00 to > 35.00 tonnes per hectare per year. The the region with “very high levels of soil erosion” (> 35) are predominantly located in the west and south-west. The high soil erosion is mainly found in western, south-western, and central part of this region(25–35). The moderate soil erosion zone (15–25) is typically found in the middle part of this region. Low “soil erosion” zones (5–15) are primarily located in the middle and eastern part of this region. The very low (