Atmospheric elevated CO₂ concentration (e[CO₂]) decreases plant nitrogen (N) concentration while increasing water use efficiency (WUE), fertigation increases crop nutrition and WUE in crop; yet the interactive effects of e[CO₂] coupled with two N-fertigation levels during deficit irrigation on plant gas exchange, root morphology and WUE remain largely elusive. The objective of this study was to explore the physiological and growth responses of ambient [CO₂] (a[CO₂], 400 ppm) and e[CO₂] (800 ppm) tomato plant exposed to two N-fertigation regimes: (1) full irrigation during N-fertigation (FIN); (2) deficit irrigation during N-fertigation (DIN) under two N fertilizer levels (reduced N (N1, 0.5 g pot⁻¹) and adequate N (N2, 1.0 g pot⁻¹). The results indicated that e[CO₂] associated with DIN regime induced the lower N2 plant water use (7.28 L plant⁻¹), maintained leaf water potential (−5.07 MPa) and hydraulic conductivity (0.49 mol m⁻² s⁻¹ MPa⁻¹), greater tomato growth in terms of leaf area (7152.75 cm²), specific leaf area (223.61 cm² g⁻¹), stem and total dry matter (19.54 g and 55.48 g). Specific root length and specific root surface area were increased under N1 fertilization, and root tissue density was promoted in both e[CO₂] and DIN environments. Moreover, a smaller and denser leaf stomata (4.96 µm² and 5.37 mm⁻²) of N1 plant was obtained at e[CO₂] integrated with DIN strategy. Meanwhile, this combination would simultaneously reduce stomatal conductance (0.13 mol m⁻² s⁻¹) and transpiration rate (1.91 mmol m⁻² s⁻¹), enhance leaf ABA concentration (133.05 ng g⁻¹ FW), contributing to an improvement in WUE from stomatal to whole-plant scale under each N level, especially for applying N1 fertilization (125.95 µmol mol⁻¹, 8.41 µmol mmol⁻¹ and 7.15 g L⁻¹). These findings provide valuable information to optimize water and nitrogen fertilizer management and improve plant water use efficiency, responding to the potential resource-limited and CO₂-enriched scenario.

The increasing CO₂ concentration ([CO₂]) in the atmosphere decreases mineral nutrients concentration in crops, whereas it increases water use efficiency (WUE). Partial root-zone irrigation (PRI) could not only increase WUE but also improve plant nutrient status. Yet the effect of PRI combined with elevated CO₂ concentration (e[CO₂]) on the element stoichiometry of tomato leaves remains unknown. This study sought to investigate the responses of leaf mineral nutrients status and element stoichiometric ratios in tomatoes to PRI combined with e[CO₂]. Tomato plants (cv. Ailsa Craig) were grown in pots in climate-controlled growth chambers with ambient [CO₂] (a[CO₂], 400ppm) and elevated [CO₂] (e[CO₂], 800ppm), respectively. Three irrigation regimes, i.e., full irrigation (FI), deficit irrigation (DI) and PRI, were applied to tomato plants at the flowering stage. The results showed that plants grown under DI and PRI had a similar biomass, enhanced root growth including greater root to shoot ratio, root length, surface area, volume and specific length, and an improved WUE in comparison with FI under e[CO₂]. Additionally, under e[CO₂], PRI showed an increase in leaf [C](+1.5%) and [N] (+9.3%), no decrease in leaf [K], [Ca], [Mg], [S] and [¹⁵N], but a decrease in leaf C/N (−6.6%) as compared with FI. Conclusively, PRI had the ability to improve leaf N concentration, maintain most leaf mineral nutrient concentrations, and optimize or maintain leaf element stoichiometric ratios under e[CO₂]. Therefore, PRI would be a practicable mode of irrigation for optimizing WUE and nutrient status in tomato leaves in a future freshwater-limited and higher-CO₂ environment.

Increasing CO2 concentration ([CO2]) in atmosphere decreases mineral nutrients concentration in crops whereas increasing water use efficiency (WUE) especially in drought conditions. Partial root-zone irrigation (PRI) could stimulate soil nutrient mineralization and improve crop nutrient status. Yet the effect of PRI combined with elevated CO2 concentration (e[CO2]) on the element stoichiometry of tomato leaves remains unknown. This study was to investigate the responses of leaf mineral nutrients status and element stoichiometric ratio in tomato grown at PRI exposed to e[CO2]. Tomato plants (cv. Ailsa Craig) were grown in pots in climate-controlled growth chambers with ambient [CO2] (a[CO2], 400ppm) and elevated [CO2] (e[CO2] 800ppm), respectively. Three irrigation regimes (IR) i.e. full irrigation (FI), deficit irrigation (DI) and partial root-zone irrigation (PRI) were applied to tomato plants at flowering stage. The result showed that plants exposed to both reduced irrigations had a similar biomass, an enhanced root growth including greater root to shoot ratio (R/S), root length (RL), surface area (RS), volume (RV) and specific length (RSL), and an improved WUE under e[CO2]. Compared to a[CO2], e[CO2] growth environment resulted in a similar leaf [C], [K] and [Mg], leaf C/K and C/Mg, increased leaf [Ca], [S] and leaf C/N, C/P and N/P, but decreased leaf [N], [P], leaf C/Ca, C/S and N/K at PRI. Additionally, under e[CO2], PRI showed an increase in leaf [C], [N] and [S], no decrease in leaf [K], Ca], [Mg], [S] and [15N], but a decrease in leaf C/N and C/S as compared with FI. Conclusively, PRI had the ability to alleviate the negative effects on mineral nutrient concentrations and maintain or improve most leaf element stoichiometric ratios under e[CO2]. Therefore PRI would be a practicable irrigation for optimizing WUE and nutrient status in tomato leaves in a future freshwater-limited and higher CO2 environment.

Instance-dependent label noise is realistic but rather challenging, where the label-corruption process depends on instances directly. It causes a severe distribution shift between the distributions of training and test data, which impairs the generalization of trained models. Prior works put great effort into tackling the issue. Unfortunately, these works always highly rely on strong assumptions or remain heuristic without theoretical guarantees. In this paper, to address the distribution shift in learning with instance-dependent label noise, a dynamic distribution-calibration strategy is adopted. Specifically, we hypothesize that, before training data are corrupted by label noise, each class conforms to a multivariate Gaussian distribution at the feature level. Label noise produces outliers to shift the Gaussian distribution. During training, to calibrate the shifted distribution, we propose two methods based on the mean and covariance of multivariate Gaussian distribution respectively. The mean-based method works in a recursive dimension-reduction manner for robust mean estimation, which is theoretically guaranteed to train a high-quality model against label noise. The covariance-based method works in a distribution disturbance manner, which is experimentally verified to improve the model robustness. We demonstrate the utility and effectiveness of our methods on datasets with synthetic label noise and real-world unknown noise.

This study aimed to explore the effect of doubled CO₂ concentration (d[CO₂]) on the modulation of root morphological structure, leaf potassium (K)/sodium (Na) ratio, and nutrient stoichiometry, as well as water use efficiency (WUE) of a C₄ maize (Zea mays L.) in response to soil drought and salinity. C₄ maize was grown in two atmospheric CO₂ concentrations of 400 and 800 ppm (a[CO₂] and d[CO₂]), subjected to two soil water regimes (well-watered and drought stress) and two soil salinity levels (0 and 100 mM NaCl pot⁻¹ (non-salt and salt stress)). The results indicated that soil drought increased maize root tissue density and specific root length. Both d[CO₂] and salt stress reduced leaf phosphorus (P) and K concentrations; conversely, drought stress enhanced leaf nitrogen (N) and K concentrations. The lower specific leaf area, but greater specific leaf N and N/K under soil drought, was amplified by salt stress. In contrast, d[CO₂] promoted leaf carbon (C)/N and C/K. Notably, d[CO₂] combined with soil drought enhanced leaf K/Na under salt stress. Moreover, d[CO₂] ameliorated the adverse impacts of soil drought and salinity on root morphology in terms of enlarged root length and root surface area, contributing to superior leaf C, N, and K use efficiency and consequently improved C₄ maize plant dry mass and WUE. These findings would provide essential knowledge to elevate salt tolerance and achieve optimal nutrient homeostasis and WUE in C₄ maize, adapting to future drier and more saline soils under a CO₂-enriched scenario.