Global soil salinization is accelerating. As the dominant fast-growing plantation genus, Populus spp. largely determines the success of coastal and inland saline-land restoration and the associated carbon-sequestration benefits. Yet most current studies rely on single indicators and lack a multidimensional physiological framework for ranking salt tolerance. Here, six elite poplar cultivars were exposed to 0% (CK), 0.2% (ST1), 0.3% (ST2) and 0.4% NaCl (ST3) for 30 d. We quantified membrane-lipid peroxidation, photosynthetic performance, osmotic adjustment and antioxidant enzymatics, then integrated the data with a principal-component–membership-function model. With increasing NaCl, MDA and REC either rose continuously or peaked slightly below ST3. ‘YX2’ reached the highest MDA (71.3 µmol g⁻¹ FW) and REC (80.3%) under ST2. Pn and SPAD declined overall, but ‘YX3’ retained the greatest photosynthetic stability (6.1 µmol m⁻² s⁻¹ at ST3). Osmolytes accumulated differentially: soluble sugar in ‘PZ2’ rose 52% above CK at ST3; soluble protein in ‘YX2’ peaked at 12.7 mg g⁻¹ FW; proline exceeded 110 µg g⁻¹ FW in ‘YX2’, ‘PZ1’ and ‘PZ2’. Antioxidant enzymes were up-regulated with stress; ‘YX3’ CAT peaked at 69.7 U g⁻¹ FW under ST2, while SOD and POD remained highly active. Correlation analysis revealed that photosynthetic decline is tightly linked to membrane oxidative damage, while the coordinated enhancement of antioxidant enzymes and concurrent accumulation of osmolytes form a synergistic protection mechanism. PCA showed that PC1 (57.1%) integrated photosynthetic capacity, membrane integrity and antioxidant synergy, whereas PC2 (14.3%) represented osmotic and enzymatic protection. The combined D-value ranked cultivars as ‘YX2’ > ‘YX3’ > ‘PZ2’ > ‘PZ1’ > ‘ZX1’ > ‘YX1’. This multi-trait platform provides both a theoretical reference and a germplasm basis for saline-site afforestation and salt-tolerant poplar breeding.

As an important reference for assessing plant water consumption and estimating plant transpiration, it is of great significance to achieve accurate prediction of plant sap flow. A number of deep learning models were established and compared using approximately 3 years of continuous eucalyptus flow time series data collected from the SAPFLUXNET open dataset and 6 environmental factors, including shortwave solar incident radiation, air temperature, air relative humidity, net radiation, vapor pressure deficit, and photosynthetic photon flux density. The experimental results show that the improved Transformer model, with the introduction of a two-step self-attention mechanism and simplified design, maintains significant predictive performance advantages compared to the original Transformer model, long short-term memory, gated recurrent unit, and temporal convolutional neural network models. In the shorter 1-h forecast, the mean squared error and coefficient of determination ( R ² ) of the improved Transformer model are 0.0191 and 0.965, respectively. Compared to the suboptimal typical Transformer model, the MSE is reduced by 22.9%, and R ² is increased by 1.0%. Additionally, the improved model maintains stable predictive performance advantages in long-term plant flow prediction. In the longest 8-h advance prediction, the MSE is reduced by 14.9% compared to the suboptimal Transformer model, and R ² increases by 3.0% compared to the Transformer model. The comprehensive experimental results show that the improved Transformer model makes more effective use of environmental information to achieve more accurate and long-term plant flow prediction. This study emphasizes the basic principle and validity of the two-step self-attention network structure and provides a valuable basis for developing more effective methods for predicting plant sap flow.

Drought responses in poplar are genotype-dependent, yet standardized metrics for quantifying drought tolerance remain scarce. Here, we employed logistic modeling of relative electrolyte leakage (REC) for the first time in poplar to derive clone-specific semi-lethal polyethylene glycol (PEG) concentrations (LC₅₀), transforming a traditional descriptive assay into a quantitative, high-throughput drought-injury metric. Six elite Populus cultivars were exposed to increasing PEG concentrations, and their REC curves were fitted using a logistic function (R² = 0.885−0.981). The derived semi-lethal PEG concentration (LC₅₀) ranged from 7.99% in ‘PZ1’ (drought-sensitive, SS) to 13.44% in ‘YX2’ (drought-tolerant, ST), enabling clear classification. Under 10% PEG stress, ST maintained leaf water content (LWC) at 73%, while SS dropped to 63%. Malondialdehyde (MDA) content doubled in SS (44.7 nmol·g⁻¹ FW) but increased by only 25% in ST (33.5 nmol·g⁻¹ FW). Gas exchange analysis revealed that SS exhibited approximately twice the reduction in net photosynthetic rate (Pn), transpiration rate (Tr), and stomatal conductance (Gs) compared to ST, with intercellular CO₂ accumulation (Ci) occurring only in SS—indicating both stomatal and non-stomatal limitations. Osmolyte profiling showed that SS accumulated large amounts of soluble sugars (Ss) (+128%) and proline (Pro) (+230%), whereas ST maintained stable soluble protein (Sp) levels and only moderately increased proline (+120%). Antioxidant capacity differed markedly: catalase (CAT), superoxide dismutase (SOD), and peroxidase (POD) activities increased by 5.6-, 1.8-, and 2.0-fold in ST, respectively, compared to 3.4-, 1.3-, and 1.7-fold in SS. Principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) of ten physiological traits explained 89% of the total variance (R²X = 0.954, Q² = 0.973), identifying POD, SOD, CAT, and Pro as the most discriminative variables (VIP > 1). This four-marker signature converts a conventional dose–response assay into a rapid, low-cost screening module that can be deployed in robotic phenotyping platforms. Specifically, the high-ranking genotype ‘YX2’ is recommended for immediate use in water-limited plantations and as a robust parent for next-generation dryland breeding programs.

As a fast-growing timber tree species with a wide cultivation area, poplar is facing the problem of declining economic benefits under long-term monoculture. Intercropping provides an effective solution. Using Illumina Miseq sequencing, we analyzed soil microbiomes under four patterns: poplar monoculture, and intercropping with amorpha fruticosa, black locust, or cassia seed. The results showed that the Alpha diversity index of intercropping area was significantly higher than that of single planting poplar area under intercropping and monoculture conditions. In the intercropping area, the highest species richness was the intercropping of poplar and black locust, and the lowest was the intercropping of poplar and amorpha fruticosa. The dominant microorganisms in the intercropping mode were Vicinamibacterales, and the fungi were Alternaia and Enterocarpus. In the single planting poplar area, a large number of bacteria gathered in the soil were Dongia and Alphaproteobacteria, and fungi were Fusarium and Mortierella. Functional prediction results showed that the biosynthetic function of ansamycin was the highest in the bacterial community. In the intercropping area, the functional abundance of methanol oxidation, sulfate respiration, sulfate compound respiration, nitrate denitrification, nitrite denitrification, and nitrous oxide denitrification was higher than that in the single planting poplar area. On the contrary, the abundance of methanotrophy function is lower than that of single planting poplar area. In the fungal community, the functional abundance of animal pathogens and the animal pathogen–dung saprotroph–endophyte–plant saprotroph–soil saprotroph–wood saprotroph group in the monoculture poplar area was higher than that in the three intercropping areas. In summary, the intercropping mode of poplar is better than the monoculture mode, and the species richness is the highest when poplar and black locust are intercropped. Therefore, the intercropping pattern of poplar and other tree species improved microbial community. This provides some theoretical guidance for the subsequent solution of continuous cropping obstacles in poplar.