A Multi-crop Ensemble Framework for Yield Prediction
Indian Journal Of Agricultural Research · 13.04.2026 · (Of)
DOI: 10.18805/ijare.a-6496 ↗Kurz gefasst
Fachartikel in der Zeitschrift Indian Journal Of Agricultural Research, online erschienen am 13.04.2026. Verfasst von Anuj Mehla und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Background
Accurate global crop yield predictions are crucial for ensuring food security, effective agricultural planning and climate adaptation. However, existing machine learning and deep learning approaches lack crop-specific feature learning, uncertainty quantification and multiscale spatial contexts, which limits their application in precision agriculture.
Methods
This study presents a hybrid ensemble deep learning framework integrated with a crop-aware transformer encoder with heteroscedastic uncertainty for global multi-crop yield prediction (Maize, rice, wheat, soybean) at a 5-arcminute (~9 km) global resolution. The architecture integrates three key innovations: (1) conservative feature engineering using historical yields (4-year sequences with a 3-year temporal gap), geographic coordinates, climate zone indicators alongside temporal trends and stress indicators; (2) crop-aware multi-head attention different mechanisms with crop-specific Query/Key/Value projections enabling differential pattern learning per crop and (3) heteroscedastic output heads predicting both mean yield (μ) and uncertainty (σ) via negative log-likelihood optimization. We implemented rigorous validation using temporal splitting (training: years ≤2013; test: years greater than 2013) and geographic blocking (5-fold GroupKFold with 0.5° spatial blocks) to prevent both spatial and temporal data leakage.
Result
Evaluated on the GlobalCropYield5min dataset (1982-2015) across four crops with 60,000 samples in 34 years. The proposed model achieved an overall test R2 of 0.9281, RMSE of 0.585 t/ha and MAE of 0.379 t/ha, with statistical significance confirmed by a paired t-test (p=0.011). Five-fold geographic cross-validation yielded R2 of 0.9337±0.0074 and rice R²=0.9462 (best), maize R²=0.9268, wheat R²=0.9201 and soybean R²=0.8298. Uncertainty quantification achieved excellent calibration (Expected calibration error = 0.024), with empirical coverage matching theoretical values (68% intervals: 69.1% coverage; 95% intervals: 94.8% coverage). Regional analysis showed consistent performance across continents (R²=0.871-0.941), with data-scarce regions showing the expected performance reduction. Ablation studies confirmed that crop-aware attention contributed +3.4% to R2, multiscale spatial features contributed 5.8% and temporal sequence features contributed +3.66%.
Zitieren
APA
Mehla, A., Deora, S. S., & Dahiya, N. (2026). A Multi-crop Ensemble Framework for Yield Prediction. Indian Journal Of Agricultural Research. https://doi.org/10.18805/ijare.a-6496
Vancouver
Mehla A, Deora SS, Dahiya N. A Multi-crop Ensemble Framework for Yield Prediction. Indian Journal Of Agricultural Research. 2026. doi:10.18805/ijare.a-6496
BibTeX
@article{Mehla2026,
author = {Mehla, Anuj and Deora, Sukhvinder Singh and Dahiya, Neeraj},
title = {A Multi-crop Ensemble Framework for Yield Prediction},
journal = {Indian Journal Of Agricultural Research},
year = {2026},
number = {Of},
doi = {10.18805/ijare.a-6496}
}
EUVIMED