Understanding the complex interplay between environmental factors and fruit quality development requires sophisticated analytical approaches linking cellular architecture to environmental conditions. This study introduces a novel application of dual-resolution X-ray computed tomography (CT) for the non-destructive characterization of apple internal tissue architecture in relation to fruit growth, thereby advancing beyond traditional methods that are primarily focused on postharvest analysis. By extracting detailed three-dimensional structural parameters, we reveal tissue porosity and heterogeneity influenced by crop load, maturity timing and canopy position, offering insights into internal quality attributes. Employing correlation analysis, Principal Component Analysis, Canonical Correlation Analysis, and Structural Equation Modeling, we identify temperature as the primary environmental driver, particularly during early developmental stages (45 Days After Full Bloom, DAFB), and uncover nonlinear, hierarchical effects of preharvest environmental factors such as vapor pressure deficit, relative humidity, and light on quality traits. Machine learning models (Multiple Linear Regression, Random Forest, XGBoost) achieve high predictive accuracy (R² > 0.99 for Multiple Linear Regression), with temperature as the key predictor. These baseline results represent findings from a single growing season and require validation across multiple seasons and cultivars before operational application. Temporal analysis highlights the importance of early-stage environmental conditions. Integrating structural and environmental data through innovative visualization tools, such as anatomy-based radar charts, facilitates comprehensive interpretation of complex interactions. This multidisciplinary framework enhances predictive precision and provides a baseline methodology to support precision orchard management under typical agricultural variability.

Temporal hyperspectral imaging enables non-destructive monitoring of agricultural stress through spectral signatures that evolve over extended observation periods, yet processing high-dimensional spatial-spectral-temporal sequences remains computationally demanding. Traditional machine learning methods compress temporal information through dimensionality reduction, while hybrid deep learning architectures combining convolutional and recurrent networks face optimization difficulties at module boundaries. We introduce PSUMamba, a triple-path bidirectional Mamba architecture that processes 204-band hyperspectral sequences across eight timepoints through linear-complexity state space models. The architecture separates spectral biochemical correlations, temporal physiological degradation, and multi-scale spatial structure into three dedicated pathways combined through cross-path self-attention, learnable softmax weights, and a logit ensemble. Evaluated under stratified source-based splitting across five random seeds, PSUMamba achieves 94.47% ± 2.17% accuracy, 99.36% ± 0.40% AUC-ROC, and 98.98% ± 1.25% sensitivity using 2.35 million parameters. Compared with Vision Transformer, PSUMamba uses approximately 2.5× fewer parameters and 10× fewer FLOPs while reaching accuracy that is at least comparable under the present evaluation, and achieves statistically significant improvements over PLS-DA, 3D CNN, 1D CNN-LSTM, and the recent MambaHSI state-space baseline with large effect sizes. Ablation experiments under matched block configurations show that the spectral, temporal, and spatial pathways encode complementary information, and that the full triple-path configuration improves over the strongest single pathway at the same parameter count. The learned fusion weights converge to a near-balanced distribution across the three pathways, indicating that each pathway contributes meaningfully to stress detection. These results position PSUMamba as a computationally efficient alternative to transformer-based models for temporal hyperspectral classification in agricultural monitoring.