A novel framework for sub-watershed prioritization: A multi-method approach
In: Environmental Earth Sciences. Springer: Heidelberg; Berlin. ISSN 1866-6280; e-ISSN 1866-6299, more
|  |
| Authors | | Top |
- Shekar, P.R.
- Prusty, J.K.
- Sahu, S.S.
- Mathew, A.
|
|
|
| Abstract |
Soil erosion, which is caused by both natural processes and human activities, still has serious effects on ecosystems, farming, and water quality in watersheds all over the world. Identifying areas most at risk of erosion and in need of prompt action requires understanding and measuring the physical features of a watershed. This study is about figuring out which sub-watersheds in the Dantewara watershed are most likely to erode. This study introduces a new way of doing things that combines morphometric analysis with advanced multi-criteria decision-making (MCDM) methods, including grey relational analysis (GRA) and the technique for order of preference by similarity to ideal solution (TOPSIS). Further strengthening the prioritization process, machine learning (ML)-primarily the support vector machine (SVM) method-is a new way to improve the process of ranking areas that are at risk of soil erosion. The keynovelty of this study lies in integrating morphometric parameters, land use and land cover (LULC), MCDM methods like TOPSIS and GRA, and machine learning to develop a unique model for sub-watershed prioritization. The sub-watersheds designated as high priority are SW1, SW6, and SW9, whilst SW2, SW3, and SW7 are categorized as medium priority. SW4, SW5, SW8, and SW10 are classified as low-priority sub-watersheds. The findings guide planners and policymakers in implementing sustainable watershed management practices. |
|
