Evaluating Machine Learning Models for Corn Yield Prediction Using Vegetation Indices in a Sandy Agroecosystem
Katoch, Karun, Deep, Nikhil Raj, Birhan, Dereje A., Sá Leitão, Diego Arruda Huggins, Singh, Ravinder, Sriram, Sindhura, Babaeian, Ebrahim, Singh, Hardeep, Sandhu, Hardev S., Dukes, Michael, Sharma, Lakesh K. International Journal of Plant Production
· Springer Nature
· 2026-08-13
· DOI: 10.1007/s42106-026-00458-3
Abstract
Florida’s sandy soils have low nitrogen (N) retention and are prone to leaching. Corn ( Zea mays L.) grown under these conditions with fixed N fertilizer rates makes recommendations unreliable, highlighting the need for sensor-based tools to predict N status in the middle of the season. This study evaluated whether combining canopy optical reflectance-based indices with leaf chlorophyll proximal sensing improves the prediction of corn grain yield (CGY). A three-year irrigated field experiment (2022–2024) evaluated six N rates (0–392 kg N ha⁻¹ in 78.5 kg increments) and an additional seventh rate of 471 kg N ha⁻¹ in 2024 using a randomized complete block design with four replications. Proximal sensing measurements of NDVI, NDRE, and SPAD were collected at 20, 34, 41, and 62 days after planting (DAP), corresponding to growth stages V4 to V14. Six models [multiple linear regression (MLR), partial least squares regression (PLSR), random forest (RF), extreme gradient boosting (XGBoost), least absolute shrinkage and selection operator (LASSO), and ridge regression] were compared using a 5 × 5 nested cross-validation framework to ensure unbiased model comparisons. Model performance was comparable, with test R² values of 0.60–0.74 and RMSE of 2552–3157 kg ha⁻¹. PLSR, Lasso, and MLR achieved the highest and most stable predictive accuracy (R² = 0.73–0.74), whereas RF and XGBoost exhibited greater variability across validation folds. These results demonstrate the potential of proximal sensing-based prediction of CGY for improving in-season N management in sandy agroecosystems.
Katoch, Karun, Deep, Nikhil Raj, Birhan, Dereje A. et al. Evaluating Machine Learning Models for Corn Yield Prediction Using Vegetation Indices in a Sandy Agroecosystem. // International Journal of Plant Production. – 2026. – DOI: 10.1007/s42106-026-00458-3. – URL: http://link.springer.com/openurl/fulltext?id=doi:10.1007/s42106-026-00458-3
Katoch, Karun, Deep, Nikhil Raj, Birhan, Dereje A, Sá Leitão, Diego Arruda Huggins, Singh, Ravinder, Sriram, Sindhura, Babaeian, Ebrahim, Singh, Hardeep, Sandhu, Hardev S, Dukes, Michael, Sharma, & Lakesh K (2026). Evaluating Machine Learning Models for Corn Yield Prediction Using Vegetation Indices in a Sandy Agroecosystem. International Journal of Plant Production. https://doi.org/10.1007/s42106-026-00458-3
Katoch, et al. “Evaluating Machine Learning Models for Corn Yield Prediction Using Vegetation Indices in a Sandy Agroecosystem.” International Journal of Plant Production, 2026. doi:10.1007/s42106-026-00458-3.
@article{Katoch2026,
title = {Evaluating Machine Learning Models for Corn Yield Prediction Using Vegetation Indices in a Sandy Agroecosystem},
author = {Katoch and Karun and Deep and Raj, Nikhil and Birhan and A., Dereje and Leitão, Sá and Huggins, Diego Arruda and Singh and Ravinder and Sriram and Sindhura and Babaeian and Ebrahim and Singh and Hardeep and Sandhu and S., Hardev and Dukes and Michael and Sharma and K., Lakesh},
journal = {International Journal of Plant Production},
year = {2026},
doi = {10.1007/s42106-026-00458-3},
url = {http://link.springer.com/openurl/fulltext?id=doi:10.1007/s42106-026-00458-3}
}
Soil-related properties associated with potato yield and quality are implicitly linked to soil water storage (SWS) determined by soil texture (ST). This soil property also affects the potato canopy, which could be evaluated indirectly...
Soft rot and blackleg diseases caused by Soft Rot Pectobacteriaceae (SRP) continue to significantly hinder global potato production. However, their epidemiology is often attributed to a single pathogen, which oversimplifies their...
A pesar de la larga historia del cultivo de papa en Brasil, su dinámica territorial sigue siendo poco comprendida. Analizamos datos de producción municipal y macrorregional de 1994 a 2024 utilizando análisis de ruptura estructural,...
The experiments were conducted at the tissue culture laboratory of the National Commission for Biotechnology (NCBT) and the Faculty of Science, Damascus University, to evaluate the response of nine introduced potato cultivars (...