December 16, 2024
Seiya Wakahara has achieved a significant milestone, publishing his first research paper,
“Non-destructive potato petiole nitrate-nitrogen prediction using chlorophyll meter and multi-source data fusion with machine learning,” explores non-destructive methods for diagnosing nitrogen levels in potato crops. By combining SPAD chlorophyll meter readings with accessible genetic, environmental, and management (GxExM) data, his research presents an innovative and cost-effective solution for farmers.
Key Contributions
- Extensive Field Data: Analyzed 26 site-years of potato nitrogen fertilizer experiments (2010–2022).
- Multi-Source Data Integration: Combined SPAD readings with variables like cultivar, growing degree days, moisture, and applied nitrogen rates.
- Machine Learning Excellence: Robust tree-based models (Random Forest and Extreme Gradient Boosting) were developed using nested cross-validation.
- Outstanding Results: Achieved a validation R² of 0.80 and a 75% diagnostic accuracy with simplified inputs, outperforming traditional regression methods.
Practical Impact
This research demonstrates that petiole nitrate-N (PNN) concentrations can be accurately predicted using machine learning and chlorophyll meter technology. The findings offer a non-destructive, efficient, and field-ready approach to optimizing nitrogen management in potato production—benefiting farmers and advancing sustainable agricultural practices.
Seiya’s work sets a strong foundation for future research on nitrogen diagnostics, including the use of remote sensing technologies and other N stress indicators.
Congratulations to Seiya Wakahara on this outstanding achievement and contribution to precision agriculture!
