Screening waterlogging-tolerant woody plants is essential for urban greening, and predictive models based on physiological indicators offer an efficient approach for rapid selection. In this study, 27 woody plant species were subjected to 10 and 15 days of waterlogging stress. Chlorophyll fluorescence parameters and biomass-related traits were measured and subsequently analyzed using linear mixed models, principal component analysis, clustering, correlation analysis, and random forest modeling. The results were as follows: (1) Waterlogging significantly suppressed most photosynthetic and biomass-related traits, whereas initial fluorescence (Fo) and dry matter content (DMC) increased; the stress response of maximum photochemical efficiency of PSII (Fv/Fm) intensified with prolonged waterlogging. (2) The first four principal components explained 83.2% of total variance, with stressed and control groups separating distinctly along PC1, reflecting photochemical inhibition. (3) Based on post-recovery phenotypes and survival rates, the 27 species were classified into strong and weak recovery capability categories; physiological clustering showed strong concordance with these recovery categories (Cramér’s V = 0.714, accuracy = 85.2%). Representative strong-recovery species included Fraxinus chinensis, Ulmus pumila, and Lagerstroemia indica, which maintained high survival (≥80%) and good ornamental quality after recovery. In contrast, representative weak-recovery species such as Kerria japonica, Syringa oblata, and Forsythia suspensa displayed marked post-recovery deterioration despite relatively intact stress-period phenotypes. (4) Among the measured parameters, maximum fluorescence (Fm) emerged as the most robust indicator, maintaining significant associations with recovery-phase phenotypes and survival in both contemporaneous and cross-period analyses (cross-period r = 0.46 with survival). Other fluorescence parameters exhibited distinct time-dependent indicative patterns. (5) Random forest models achieved the best predictive performance for 15-day survival (R² = 0.712), substantially outperforming 10-day models (R² = 0.153). Collectively, these findings demonstrate that chlorophyll fluorescence and biomass-related indicators effectively capture post-waterlogging recovery capability in woody plants, and the random forest framework provides technical support for rapid screening of waterlogging-tolerant germplasm.

Ginkgo biloba is a dioecious species valued for landscaping and medicine, but rapid sex identification outside the flowering and fruiting stages remains challenging. Hyperspectral imaging offers a potential solution, though whether spectral sex markers are stable across phenological stages and can support a single year-round model is unclear. Here, leaf hyperspectral reflectance (400–1000 nm) was acquired from mature G. biloba at flowering (n = 360), green-leaf (n = 1395), and yellow-leaf (n = 243) stages. Sex identification models were built using multiple machine learning classifiers, with leaf flavonoid content as biochemical validation. Stage-specific models achieved optimal test accuracies of 96.67% (flowering, MSC + PLS-DA), 95.77% (green-leaf, raw spectra + LDA), and 97.12% (yellow-leaf, MSC + LDA). However, discriminative bands shifted from the visible (520–690 nm) at the green-leaf stage to the near-infrared (700–1000 nm) at the yellow-leaf stage, with almost no bands shared across all stages. Furthermore, cross-stage prediction accuracy dropped to near-chance levels (~50%). As t-SNE analysis revealed, the universal model had learned phenological rather than sex-specific information. In addition, flavonoid measurements revealed a highly significant sex × stage interaction (p < 0.001) and a reversal of the sex difference between green-leaf (male > female) and yellow-leaf (female > male) stages. Thus, the spectral and biochemical sex markers examined in this study varied substantially with phenology, and stage-specific models are required for practical sex identification in G. biloba.