Mohana et al. (2026) Advancing Soil Assessment Quality for Crop Yield Prediction Using an Optimized Category Integrated Dual Task Graph Neural Network Approach

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

Identification

• Journal: Irrigation and Drainage
• Year: 2026
• Date: 2026-09-22
• Authors: J. Mohana, P. Dass, M. Sathesh, G Jayandhi, Raja Meganathan
• DOI: 10.1002/ird.70227

Research Groups

• Department of Agricultural Engineering, Punjab Agricultural University
• Remote Sensing Laboratory, Indian Space Research Organisation

Short Summary

This study proposes a new model for soil assessment quality (SAQ) and crop yield prediction (CYP) using a category integrated dual task graph neural network (CIDTGNN), achieving improved accuracy compared to existing machine learning methods.

Objective

• Develop an accurate and efficient method for estimating soil health parameters (SHP) and predicting crop yields in the Rupnagar district of Punjab, India

Study Configuration

• Spatial Scale: Local scale, focusing on the Rupnagar district of Punjab, India
• Temporal Scale: Long-term data analysis, with a focus on seasonal variations

Methodology and Data

• Models used: Category Integrated Dual Task Graph Neural Network (CIDTGNN)
• Data sources: Remotely sensed Sentinel‐1 and Sentinel‐2 satellite data, field observations

Main Results

• The SAQ-CIDTGNN model achieved an R 2 value of 0.78 in crop yield prediction with a significantly lower error rate compared to existing machine learning methods.
• The proposed method outperformed the ordinary least squares (OLS) regressor by 42% in terms of R 2 and reduced Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) by 38% and 39%, respectively.

Contributions

• This study provides a novel approach to soil assessment quality and crop yield prediction using graph neural networks, improving upon existing machine learning methods.
• The proposed method can be applied to other regions with similar climate and soil conditions, contributing to the development of precision agriculture.

Funding

• This research was funded by the Indian Space Research Organisation (ISRO) under the project code ISRO/SAGA/2020/01.

Citation

@article{Mohana2026Advancing,
  author = {Mohana, J. and Dass, P. and Sathesh, M. and Jayandhi, G and Meganathan, Raja},
  title = {Advancing Soil Assessment Quality for Crop Yield Prediction Using an Optimized Category Integrated Dual Task Graph Neural Network Approach},
  journal = {Irrigation and Drainage},
  year = {2026},
  doi = {10.1002/ird.70227},
  url = {https://doi.org/10.1002/ird.70227}
}