The rapid and accurate prediction of crop yield and the construction of optimal yield prediction models are important for guiding field-scale agronomic management practices in precision agriculture. This study selected the leaf area index (LAI) of winter wheat (Triticum aestivum L.) at four different stages, and collected canopy spectral information and extracted vegetation indexes through unmanned aerial vehicle (UAV) multi-spectral sensors to establish the yield prediction model under the condition of slow-release nitrogen fertilizer and proposed optimal fertilization strategies for sustainable yield increase in wheat. The prediction results were evaluated using random forest (RF), support vector machine (SVM) and back propagation neural network (BPNN) methods to select the optimal spectral index and establish yield prediction models. The results showed that LAI has a significantly positive correlation with yield across four growth stages of winter wheat, and the correlation coefficient at the anthesis stage reached 0.96 in 2018–2019 and 0.83 in 2019–2020. Therefore, yield prediction for winter wheat could be achieved through a remote sensing estimation of LAI at the anthesis stage. Six vegetation indexes calculated from UAV-derived reflectance data were modeled against LAI, demonstrating that the red-edge vegetation index (CI_{red edge}) achieved superior accuracy in estimating LAI for winter wheat yield prediction. RF, SVM and BPNN models were used to evaluate the accuracy and precision of CI_{red edge} in predicting yield, respectively. It was found that RF outperformed both SVM and BPNN in predicting yield accuracy. The CI_{red edge} of the anthesis stage was the best vegetation index and stage for estimating yield of winter wheat based on UAV remote sensing. Under different N application rates, both predicted and measured yields exhibited a consistent trend that followed the order of SRF (slow-release N fertilizer) > SRFU1 (mixed TU and SRF at a ratio of 2:8) > SRFU2 (mixed TU and SRF at a ratio of 3:7) > TU (traditional urea). The optimum N fertilizer rate and N fertilizer type for winter wheat in this study were 220 kg ha⁻¹ and SRF, respectively. The results of this study will provide significant technical support for regional crop growth monitoring and yield prediction.

Introduction: Blending urea (U) with slow-release N fertilizer (R) can improve the synchrony between N supply and crop demand, while mitigating N losses to the environment. However, cultivar-specific responses to varied U/R blending ratios under rainfed conditions remain poorly characterized. Methods: In this study, two dominant maize cultivars (ZD958 and QL14) widely planted in Northwest China were selected as experimental materials. Five fertilization treatments (U, R, and three U/R blending ratios: UR1, UR2, UR3) each at two N rates (N1: 180 kg·ha ⁻¹ ; N2: 240 kg·ha ⁻¹ ) were tested. Results: The results showed that UR treatments generally achieved higher grain yield and N use efficiency (NUE), while reducing cumulative NH ₃ volatilization and soil NO ₃ ^- –N residue compared with U and R. For both ZD958 and QL14, the maximum grain yield and NUE were achieved under the UR2N1 (U/R = 3/7, 180 kg·ha ⁻¹ ). Compared with other treatments, UR2N1 increased NUE by 3.51–19.07% (ZD958) and 4.01–20.41% (QL14), while reducing NH ₃ volatilization soil NO ₃ ^- –N residue. Notably, under UR2N1, QL14 showed 9.70% higher NUE than ZD958, along with 8.78% lower soil NO ₃ ^- –N residue and 14.12% lower NH ₃ volatilization. Conclusion: Under rainfed conditions, the UR2N1+QL14 combination effectively balances the trade-off between grain yield and environmental impact, and is therefore recommended as an optimal fertilization–cultivar strategy for rainfed maize production in Northwest China.