This study investigates a multimodal deep learning framework that integrates unmanned aerial vehicle (UAV) multispectral imagery with meteorological data to predict cotton yield. The study analyzes the impact of different neural network architectures, including the CNN feature extraction layer, the depth of the fully connected layer, and the method of integrating meteorological data, on model performance. Experimental results show that the model combining UAV multispectral imagery with weekly meteorological data achieved optimal yield prediction accuracy (RMSE = 0.27 t/ha; R² = 0.61). Specifically, models based on AlexNet (Model 9) and CNN2conv (Model 18) exhibited superior accuracy. ANOVA results revealed that deeper fully connected layers significantly reduced RMSE, while variations in CNN architectural complexity had no statistically significant effect. Furthermore, although the models exhibited comparable prediction accuracy (RMSE: 0.27–0.33 t/ha; R²: 0.61–0.69 across test datasets), their yield prediction spatial distributions varied significantly (e.g., Model 9 predicted a mean yield of 3.88 t/ha with a range of 2.51–4.89 t/ha, versus Model 18 at 3.74 t/ha and 2.33–4.76 t/ha), suggesting the need for further evaluation of spatial stability. This study underscores the potential of deep learning models integrating UAV and meteorological data for precision agriculture, offering valuable insights for optimizing spatiotemporal data integration strategies in future research.

Accurate and non-destructive monitoring of leaf chlorophyll content (LCC) is essential for assessing crop photosynthetic activity and nitrogen status in precision agriculture. This study introduces a phenology-aware machine learning framework that combines hyperspectral reflectance data with various regression models to estimate leaf chlorophyll content (LCC) in cotton at six key reproductive stages. Field experiments utilized synchronized spectral and SPAD measurements, incorporating spectral transformations—such as vegetation indices (VIs), first-order derivatives, and trilateration edge parameters (TEPs, a new set of geometric metrics for red-edge characterization)—for evaluation. Five regression approaches were evaluated, including univariate and multivariate linear models, along with three machine learning algorithms: Random Forest, K-Nearest Neighbor, and Support Vector Regression. Random Forest consistently outperformed the other models, achieving the highest R² (0.85) and the lowest RMSE (4.1) during the bud stage. Notably, the optimal prediction accuracy was achieved with fewer than five spectral features. The proposed framework demonstrates the potential for scalable, stage-specific monitoring of chlorophyll dynamics and offers valuable insights for large-scale crop management applications.

The development of arid-oasis agriculture and territorial spatial restructuring strongly influence regional eco-environmental quality. Clarifying production–living–ecological space (PLES) transitions and the nonlinear mechanisms driving eco-environmental quality is important for sustainable agricultural land use and ecological governance. Northern Xinjiang combines economic agglomeration with ecological sensitivity, making it a representative region for examining changes in eco-environmental quality and their underlying drivers. Land-use data for 1990, 2000, 2010, and 2020 were analyzed using land-use transition matrices, the eco-environmental quality index (EEQI), spatial autocorrelation, and hotspot analysis to characterize spatiotemporal changes under PLES transitions. XGBoost–SHAP was applied to identify the dominant factors, nonlinear responses, and interactions. The results showed that: (1) ecological space remained dominant from 1990 to 2020, but its share declined from 91.72% to 87.36%. Production and living spaces increased from 7.78% and 0.50% to 11.72% and 0.92%, respectively. Agricultural production space expanded most markedly, and spatial restructuring was most intense during 2000–2010. (2) EEQI decreased from 0.3100 to 0.2922 and showed an overall fluctuating decline. Significant spatial clustering was observed, with Moran’s I values consistently above 0.81 (p < 0.001). High-value areas were mainly distributed in the Ili River Valley and along the northern slope of the Tianshan Mountains, whereas low-value areas were concentrated in the Junggar Basin interior and the arid areas along its eastern margin. (3) During 2000–2020, LAI and NPP in agricultural production space increased by 45.7% and 48.2%, respectively. These increases indicated overall improvements in canopy condition and productivity. After 2010, NPP growth slowed markedly while LAI remained relatively stable, indicating a stage-specific divergence. (4) XGBoost–SHAP showed good robustness and spatial generalizability, with R² values of 0.868–0.918 and RMSE values of 0.059–0.075. NDVI remained the primary driver, while Elev contributed consistently. The dominant interaction shifted from NDVI ∩ PopDen in 1990 and 2000 to NDVI ∩ Elev in 2010 and NDVI ∩ GDP in 2020. This shift indicated fundamental natural constraints and stronger joint effects of population and economic activities during the later stages. These findings provide a scientific basis for optimizing PLES, regulating arid-oasis agriculture, and safeguarding regional ecological security.