Accurate prediction of crop yield remains a critical research priority due to the increasing 19 vulnerability of agricultural systems to climate change and the growing need for food security. In 20 this study, we developed a hybrid modeling framework to predict irrigated wheat yield in Razavi 21 Khorasan Province, Iran, using long-term (2004–2023) climatic, edaphic, and nutritional datasets 22 comprising 47 variables collected across 17 counties. After preprocessing, feature selection was 23 performed using an integrated approach combining Mutual Information (MI), Recursive Feature 24 Elimination (RFE), and the advanced NSGA-III multi-objective optimization algorithm. Final 25 yield prediction was conducted with a Stacking Regressor meta-learner incorporating LightGBM 26 (LGBM) and a Deep Neural Network (DNN). The optimal subset of 10 features—Tmin, TS, K, 27 Silt, EC, HCO₃, Mg, Prec_OC, AI_Clay, and a regional indicator variable (county_te) — 28 achieved a test-set R² of 0.44, reflecting a moderate yet meaningful level of explained variance 29 given the multidimensional, nonlinear, and environmentally heterogeneous nature of the wheat 30 production system. SHAP (SHapley Additive Explanations) analysis further highlighted the 31 dominant influence of regional heterogeneity alongside complex interactions among climatic, 32 soil, and nutritional factors. While the model does not capture all sources of variability, the 33 results demonstrate that this hybrid optimization–learning pipeline reliably characterizes a substantial portion of wheat yield variation and offers a practical decision-support tool for site- 35 specific management and climate adaptation planning.

Keywords