Ellur et al. (2025) Prediction and Mapping of Soil Texture at High Spatial Resolution in a Canal Irrigated Region Using Machine Learning

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Journal of Geography Environment and Earth Science International
• Year: 2025
• Date: 2025-07-03
• Authors: Rajath Ellur, M. A. Ananthakumar, Krishna Desai
• DOI: 10.9734/jgeesi/2025/v29i6911

Research Groups

Not specified

Short Summary

This study mapped the spatial distribution of soil texture (sand, silt, and clay) in the Cauvery command area of southern Karnataka, India, using Random Forest and Sentinel-2 data, identifying clay loam as the predominant soil type.

Objective

• To estimate and map the spatial distribution of soil texture components (sand, silt, and clay) across the Cauvery command area using geospatial techniques and machine learning.

Study Configuration

• Spatial Scale: Cauvery command area, southern Karnataka, India (mapping resolution: 10 m).
• Temporal Scale: Not specified.

Methodology and Data

• Models used: Random Forest (RF) algorithm.
• Data sources: 289 surface soil samples (analyzed via the international pipette method), Sentinel-2 spectral indices (10 m), terrain attributes, and remote sensing data.

Main Results

• Model Performance: High calibration accuracy for sand ($R^2 = 0.958$) and silt ($R^2 = 0.930$), with moderate validation accuracy for clay ($R^2 = 0.282$).
• Soil Classification: The majority of the study area is classified as clay loam.
• Drivers: Spatial distribution of soil texture is primarily governed by topography and depositional processes.

Contributions

• Provides a high-resolution soil texture map for a canal-irrigated landscape, demonstrating the integration of Sentinel-2 data and machine learning to support precision agriculture and soil health monitoring.

Funding

Not specified

Citation

@article{Ellur2025Prediction,
  author = {Ellur, Rajath and Ananthakumar, M. A. and Desai, Krishna},
  title = {Prediction and Mapping of Soil Texture at High Spatial Resolution in a Canal Irrigated Region Using Machine Learning},
  journal = {Journal of Geography Environment and Earth Science International},
  year = {2025},
  doi = {10.9734/jgeesi/2025/v29i6911},
  url = {https://doi.org/10.9734/jgeesi/2025/v29i6911}
}