Rapid and accurate monitoring of soil total nitrogen (STN) is fundamental to optimising nitrogen fertiliser management, improving nitrogen use efficiency, and reducing agricultural non-point source pollution; however, traditional methods struggle to provide timely, large-scale, and continuous monitoring. Based on the relationship between maize canopy spectral characteristics and soil nitrogen distribution, this study acquired multi-spectral drone imagery during the jointing, tasselling, and grain filling stages, whilst simultaneously collecting soil samples from the 0–20, 20–40, and 40–60 cm soil layers. Using canopy reflectance (Band), vegetation indices (VIs), textural features (TF), and a fusion of these three features (Band+VIs+TF) as input variables, machine learning models, including Random Forest Regression (RFR), Gradient Boosted Decision Trees (GBDT), K-Nearest Neighbours (KNN), and XGBoost, were constructed to estimate STN. The results indicate that (1) STN is significantly correlated with spectral reflectance, vegetation indices, and textural features, with absolute correlation coefficients ranging from 0.60 to 0.89; (2) the accuracy of STN estimation by the four models showed a decreasing trend with increasing soil depth, with the highest accuracy observed in the 0–20 cm soil layer; (3) the validation accuracy of STN estimation followed the order: grain filling stage (R² = 0.833) > tasselling stage (R² = 0.794) > jointing stage (R² = 0.761); (4) under feature fusion conditions, the STN estimation accuracy of all four models was superior to that of using a single feature, and the prediction accuracy of the XGBoost model (0.833) was higher than that of the other three models (0.701–0.734). In summary, during the grain filling stage of maize, using feature fusion as input variables and the XGBoost model can significantly improve the accuracy of STN estimation in maize fields, with the 0–20 cm soil layer yielding the best results.

Hydrological fluctuations affect cropping-structure optimization in arid inland river basins by altering agricultural water-supply boundaries and crop yields. This study focused on five counties and districts in the middle and lower reaches of the Shiyang River Basin. Based on the 1995–2024 runoff series, four hydrological scenarios—wet, normal, dry, and extremely dry—were constructed. Crop yield responses were incorporated into blue, green, and gray water footprint accounting, and NSGA-III coupled with entropy-weighted TOPSIS was applied for multi-objective optimization. The results showed that (1) under the current cropping structure, unit water footprints generally increased as hydrological conditions shifted from wet to extremely dry, with gray water footprints showing greater sensitivity to yield reductions; (2) the overall directions of cropping-structure adjustment were broadly consistent across scenarios, with the planted areas of wheat, maize, and tubers decreasing by nearly 30% and melon area increasing by nearly 30%, while vegetable and oilseed areas exhibited clear scenario-dependent adjustments; (3) relative to the corresponding 2024 cropping structure evaluated under the same hydrological conditions, the selected compromise schemes increased net economic return by 14.58–16.23 × 10⁸ CNY across the four scenarios. However, the attainable net economic return declined from 73.74 × 10⁸ CNY under the wet scenario to 17.95 × 10⁸ CNY under the extremely dry scenario. The blue water footprint decreased by 18.18–19.07%, whereas the gray water footprint decreased by approximately 17–18% across scenarios. Deteriorating hydrological conditions reduced the attainable economic return and highlighted clear county-level differences in adjustment pathways and economic vulnerability. These findings provide a quantitative basis for scenario- and county-specific cropping-structure optimization and water-resource regulation in arid inland river basins.