Hydroponic Screening and Comprehensive Evaluation System for Salt Tolerance in Wheat Under Full-Fertility-Cycle Salt Stress Conditions

Abstract

Soil salinity is a major constraint to wheat production worldwide. Efficient screening of salt-tolerant cultivars is essential for breeding programs, yet a rapid and reliable evaluation system based on full-life-cycle salt stress treatment is lacking. To address this, we conducted a hydroponic experiment encompassing the entire growth cycle of 37 wheat cultivars under control and salt stress (85.5 mM NaCl). Using principal component and stepwise regression analyses on 15 agronomic and yield-related traits, we identified five key indicators—total dry weight, root dry weight, plant height, thousand-grain weight, and number of grains per spike—that effectively represent overall salt tolerance. Based on a comprehensive evaluation value (D-value), the cultivars were classified into five distinct categories: highly salt-tolerant, salt-tolerant, moderately salt-tolerant, weakly salt-tolerant, and salt-sensitive. Notably, the highly salt-tolerant cultivar ‘Yangfumai 8′ and the salt-sensitive cultivar ‘Yangmai 22’ were selected as representative extremes. A subsequent pot experiment confirmed significant physiological differences between them in antioxidant enzyme activities (SOD, POD, CAT) and proline accumulation under salt stress. This study establishes a practical and efficient screening framework, providing breeders with a simplified index set for high-throughput evaluation and offering ideal contrasting materials for in-depth physiological research on salt tolerance mechanisms in wheat.

1. Introduction

The salinization of soil has become a major global ecological and environmental challenge, affecting plant distribution, reducing their productivity, and endangering food security [1,2]. Wheat (Triticum aestivum L.) is one of the world’s most important food crops [3]. It serves as a staple food for approximately 40 percent of the global population, yet it is high sensitive to salt stress [4]. Coastal mudflats in Jiangsu Province are characterized by a high abundance of resources, yet soil salinization poses a substantial challenge, significantly impeding the enhancement of local wheat production and the effective utilization of arable land resources [5]. The development and deployment of salt-tolerant cultivars has been identified as a pivotal strategy for mitigating the impact of this threat. However, the efficiency of breeding programs is contingent upon the availability of reliable and practical methods for evaluating salt tolerance across extensive germplasm collections. Therefore, Developing reliable evaluation systems for wheat salt tolerance and identifying salt-tolerant germplasm is essential to safeguard wheat yields under saline conditions.

Existing approaches for evaluating salt tolerance in wheat frequently exhibit two principal shortcomings that limit their predictive power and practical utility. Firstly, it is important to note that screening is predominantly conducted at the seed germination and seedling stage [6,7]. While logistically convenient, tolerance mechanisms at early vegetative phases may not reliably translate to performance during later, yield-determining reproductive stages, such as flowering and grain filling [8,9]. Consequently, cultivars selected solely based on seedling vigor may be rendered ineffective under sustained, whole-life-cycle stress conditions in the field. Secondly, a plethora of evaluation indices have been proposed, ranging from complex physiological and ion-content measurements to simpler agronomic traits [10,11,12]. For example, Shiksha [13] demonstrated that biomass accumulation, along with the levels of aboveground K⁺ and root Ca²⁺, aboveground K⁺ and root Ca²⁺ levels, and the ratio of aboveground K⁺ to Na⁺ in plants, were the most effective indicators of salt tolerance in wheat seedlings. Ehab S. A. Moustafa et al. [14] found significant positive correlations between number of grains per spike, thousand kernel weight and plant height with yield through field experiments, indicating their importance in selecting salt-tolerant cultivars. While acknowledging the value of mechanistic indices for research purposes, it is important to recognize the limitations of such methods for rapid assessment of large breeding populations. The labor-intensiveness, cost and impracticality of mechanistic indices often renders them unviable for such applications. This engenders a significant discrepancy between scientific understanding and its practical application in breeding. Consequently, a substantial gap remains in the field: the absence of a comprehensive, life-cycle-based screening system that integrates scientific rigor with operational simplicity, ultimately delivering a streamlined set of indicators strongly correlated with final yield under stress.

To address these limitations, this study establishes an integrated framework for the evaluation and selection of salt-tolerant wheat. Our strategy innovatively combines controlled, full-life-cycle salt stress imposition in a hydroponic system with multivariate statistical analysis to distill complex phenotypic responses. We hypothesize that a holistic assessment across the entire growth cycle will enable more accurate cultivar classification and facilitate the identification of a minimal suite of key agronomic traits that can serve as reliable proxies for comprehensive salt tolerance. The specific objectives of this study are: (i) to implement a full-life-cycle salt stress treatment in a controlled hydroponic system to assess 37 diverse wheat cultivars uniformly; (ii) to establish a robust, multi-trait evaluation framework using multivariate statistics and to identify a minimal set of key diagnostic indicators for rapid screening; and (iii), to physiologically validate the contrasting tolerance of cultivars selected from this framework. The objective of this methodology is to furnish breeders with a theoretical framework for salt-tolerant screening and growers with definitive salt-tolerant germplasm.

2. Materials and Methods

2.1. Hydroponic Salt Tolerance Screening Test

The experiment was conducted at the hydroponic experimental base of the College of Agriculture, Yangzhou University, from November 2023 to June 2024. The experimental setup was a fully automatic circulating hydroponic system, with hydroponic tanks measuring 580 cm in length, 142 cm in width, and 45 cm in depth. Each board had 14 planting holes with a diameter of 4 cm and spacing of 10 cm, and the spacing between holes on two boards was 16 cm. Mechanized hard disk seedling raising was used, with seedlings raised on October 20 and transplanted on November 20. Two plants were transplanted per hole, and 21 holes were planted per cultivar. Based on the results of preliminary concentration screening tests, two salt concentration treatments (0 and 85.5 mM) were set up in this experiment, using industrial salt. The solution in the hydroponic tanks was a mixture of Epsino nutrient solution and Arnon micronutrient nutrient solution, with specific reference to IWMI formulations. The solution was full-strength at the time of transplantation, half-strength 20 days after transplantation, and quarter-strength after heading. The tops of the tanks were covered with a rain shelter to prevent rain throughout the entire growth period. The pH was adjusted daily with dilute sulfuric acid to maintain it at around 6.5. Pump aeration was used to maintain continuous flow of the nutrient solution, ensuring consistent nutrient concentration, salt concentration, and pH in all parts of the hydroponic tanks.

A total of 37 wheat cultivars were collected and tested, as shown in

Table 1. List of wheat cultivars for testing.

Number	Cultivar	Number	Cultivar	Number	Cultivar	Number	Cultivar
1	Yangmai 25	11	Yangmai 21	21	Ningmaizi 166	31	Ruihuamai 521
2	Yangmai 20	12	Suyanmai 017	22	Nongmai 152	32	Xumai 178
3	Yangmai 18	13	Sukenmai 1008	23	Yangmai 1	33	Ruihua 518
4	Zhenmai 12	14	Yangmai 4	24	Shanrong 3	34	Nongmai 158
5	Huamai 33	15	Yangfumai 5145	25	Huaimai 46	35	Huamai 1430
6	Huamai 8	16	Yangmai 5	26	Xumai 2023	36	Yangmai 23
7	Yangfumai 8	17	Yangmai 16	27	Qianmai 088	37	Huaihemai 16132
8	Luomai 163	18	Yangmai 22	28	Yangfumai 10
9	Yangfumai 9	19	Yangmai 28	29	Xumai 38
10	Zhongkenmai 212	20	Yangmai 15	30	Nongmai 126

.

2.2. Pot Experiment

2.2.1. Plant Materials, Treatments, and Experimental Design

Subsequent experiments were conducted in a pot setting for the purpose of validating the physiological differences between cultivars identified as highly salt-tolerant (Yangfumai 8) and salt-sensitive (Yangmai 22) from the hydroponic screening. The pot experiment was conducted on the campus of Yangzhou University (32°30′ N, 119°25′ E), located in Jiangsu Province, China. The experiment was conducted in cultivation pots with a diameter of 25 cm and a height of 30 cm. A total of 30 seeds were sown in each pot, and the seedlings were thinned to 16 seeds after the process of colonization. The experiment utilized a factorial design comprising two factors: cultivar (Yangfumai 8 and Yangmai 22) and salinity (0, 34.2, and 68.4 mM NaCl). Each combination was replicated 20 times (pots), resulting in a total of 120 pots (2 cultivars × 3 treatments × 20 replicates). The pots were arranged in blocks to account for potential environmental gradients, and all treatment combinations were randomly positioned within each block with sufficient spacing to minimize edge effects. The upper portion of the test site is shielded from precipitation by a transparent plastic sheet, thus ensuring that salt is not compromised by the effects of rainwater.

2.2.2. Soil Preparation and Salinity Maintenance

Saline soils were artificially prepared by mixing local desalted farmland soil with high-salinity soil collected from coastal beaches of Yancheng, Jiangsu Province. The initial physicochemical properties of the mixed soil were: pH 8.0 ± 0.2; organic matter content 20.8 ± 1.0 g kg⁻¹. The target initial soil EC₁:₅ values were 0.35 ± 0.05 (CK), 3.8 ± 0.3 (34.2 mM), and 7.5 ± 0.4 (68.4 mM) dS m⁻¹. Total nitrogen, 1.33 ± 0.2 g/kg. Available phosphorus, 50.12 ± 1.6 mg/kg. Available potassium, 315.23 ± 2.5 mg/kg. Each pot was filled with 15 kg of sieved soil. In order to maintain stable soil moisture and salinity levels, a weight-based method was utilized. The pots were weighed at three-day intervals, and water was added to restore the weight to a target corresponding to 70% ± 5% of field capacity. This approach facilitated the replenishment of water loss via evapotranspiration without the necessity for supplementary salts. Soil electrical conductivity (EC) was meticulously monitored on a monthly basis using a portable conductivity meter (CTS 50C, Spectrum Technologies, Inc. USA). It was observed that the variation remained within the ±10% margin of error of the initial values throughout the experimental period.

2.2.3. Fertilizer Application

The basal fertilizer application consisted of 1.305 g urea (nitrogen source), 8.3 g calcium superphosphate (phosphorus source), and 1.665 g potassium chloride (potassium source). Additional urea was applied at the tillering stage; during the first and second heading stages, 1.305 g urea and 1.665 g potassium chloride were applied, respectively, while only 1.305 g urea was applied during the third heading stage.

2.3. Determination of Traits and Methods

2.3.1. Morphological Characteristics

Relevant morphological traits were measured as follows: plant height (PH), spike length (SL), and the length of top 1st leaf (LT1). The leaf length of wheat flag leaves was determined using a ruler [15]. The length of top 2nd leaf (LT2): using a ruler to deter-mine the leaf length of a leaf below the flag leaf of wheat. Root length (RL): the longest root length of wheat plants was determined by removing the wheat plants from the hydroponic tank and draining the roots with absorbent paper [16].

2.3.2. The Dry Matter Weight of Each Part

The dry matter weight indicator was counted: Wheat straw dry weight (WDW), spike dry weight (SDW), root dry weight (RDW): After sampling at maturity, the plant materials were sorted by organ. The separated samples were then deactivated at 105 °C for 30 min and subsequently dried at 80 °C for 48 h until a constant weight was achieved, after which they were weighed individually. Ground dry weight (GDW): GDW = WDW + SDW + RDW. Root shoot ratio (RSR): RSR = RDW/GDW. Total dry weight (TDW): TDW = WDW + SDW + RDW.

2.3.3. Yied and Its Components

Yield and its composition: wheat yield and its constituent factors: the number of spikes (SN) refers to the number of wheat spikes per hole in a hydroponic cell and the number per pot for potted plants., the number of grains per spike (SGN) is for each wheat, the thousand-grain weight (TGW) is the weight of a thousand wheat seeds [17]. Theoretical Yield (TY) is the weight of rice per hole in a hydroponic tank in grams per hole or per bucket in a potting trial in grams per pot.

2.3.4. Saline Tolerance Index (STI)

Calculation of salt tolerance coefficient: Saline tolerance index (STI) = the average values of each index under salt treatment/the average values under the comparative treatment of each index

2.3.5. Measurement of Physiological Indicators

Determination of physiological indices on the morning of the 10th day after flowering, 20 pieces of sword leaves from each treatment were collected, immediately frozen in liquid nitrogen, and then stored at −80 °C for measurement. 1.5 g of the sample was cut and added to 15 mL of 150 mmol/L phosphate buffer (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China), pH 7.0, ground on an ice bath, and centrifuged at 15,000× g for 5 min, and the supernatant was used as crude enzyme extracts for the determination of malondialdehyde (MDA), superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT). The MDA content was determined by the TCA method [18]. The nitro blue tetrazolium (NBT) photo-reduction method was used to analyze the superoxide dismutase (SOD); POD activity was determined by the guaiacol colorimetric method; CAT activity was determined by the UV absorption method [19]; proline content was determined by the sulfosalicylic acid method [20].

2.4. Statistical Analysis

2.4.1. Salt Tolerance and High-Yield Index (STHYI)

Calculation of salt tolerance and high-yield index (STHYI):

STHYI = (Y_S/Y_CK + Y_CK/Y_MAX)/2

(1)

Y_S means salt stress yield; Y_CK means control treatment yield; Y_CK means control treatment yield; Y_MAX means Maximum yield of all tested cultivars under control conditions.

2.4.2. Principal Component and Membership Function Analysis

The membership function value of each comprehensive index of different wheat cultivars is as below [21]:

$$\mathsf{\mu }\left({\mathrm{X}}_{\mathrm{j}}\right) = ({\mathrm{X}}_j-X_{\mathit{min}})/(X_{\mathit{max}} {-\mathrm{X}}_{\mathit{min}}), \mathrm{j} = 1,2,\dots , \mathrm{n}$$

(2)

where X_j is the measured value of the index, X_min and X_max are the maximum and minimum values of an index for all the participating materials.

2.4.3. Weight of Each Comprehensive Indicator

Weight of composite indicators: ${\mathrm{W}}_j = {\mathrm{p}}_j/{\sum }_{j = 1}^n{ p}_j, \mathrm{j} = 1, 2, \dots , \mathrm{n}$.

In the formula, W_j represents the weight of the jth composite index, and P_j represents the contribution of the jth composite index of each wheat cultivar obtained by principal component analysis.

2.4.4. D Value Calculation

The overall salt resistance ability of each wheat cultivar is as below:

$$\mathrm{D} = \sum _{j = 1}^n\left[\mathrm{u}\left({\mathrm{X}}_j\right) \times W_j\right], \mathrm{j} = 1, 2, \dots , \mathrm{n}$$

(3)

D value is the comprehensive evaluation value of salt tolerance of each wheat cultivar evaluated by the comprehensive index under salt stress conditions.

Data were organized using Excel 2021, all the data used are the average values of three repetitions. The salt tolerance coefficients of agronomic trait parameters for each wheat cultivar were analyzed using principal component analysis (PCA) to obtain the value of the membership function [22]. The value of the membership function and weight were then used to obtain the D value of the comprehensive evaluation of salt tolerance of wheat [23,24]. This D value was taken as the dependent variable. The salt tolerance coefficients of the agronomic trait parameters of each wheat cultivar were designated as the independent variables. A series of parameter indexes and mathematical model equations with considerable influence on the salt tolerance ability of wheat cultivars were obtained by carrying out stepwise regression analysis [25,26]. These equations and indexes can expeditiously and efficiently ascertain the extent of salt tolerance in wheat. The application of this equation, in conjunction with the specified parameters and indicators, facilitates a rapid and efficient assessment of the salt tolerance level in wheat. Multivariate analyses such as principal component analysis, membership function analysis, multiple stepwise regression analysis, and cluster analysis were carried out using IBM SPSS Statistics 25 (http://ibm.com). Figures were drawn using The Origin 2024 Software (Origin 2024 Inc., Northampton, MA, USA). A weighted random forest analysis was carried out using randomForest package, and correlation analysis was performed with linkET package in R (R for windows 4.0.2) (https://www.r-project.org, accessed on 1 August 2025).

3. Results

3.1. Descriptive Statistics Traits of Tested Wheat Cultivars

This study employed descriptive statistical analyses on 16 traits of 37 wheat germplasm resources, as shown in

Table 2. Descriptive statistics and variation analysis of agronomic traits in 37 accessions.

Character	CK Trait					S Trait
	Mean	SD	Min	Max	CV (%)	Mean	SD	Min	Max	CV (%)
PH	66.5	7.09	88.5	55.3	10.7	62.3	7.66	80.3	44.3	12.3
NST	4.43	1.24	7.00	2.00	28.0	3.30	1.26	7.50	2.00	38.1
SL	8.47	1.13	10.0	5.50	13.3	8.86	1.11	10.8	6.50	12.5
LT1	23.1	2.63	27.8	18.0	11.4	19.9	3.09	26.3	13.3	15.5
LT2	23.0	2.66	29.3	16.0	12.0	21.2	2.49	26.0	16.00	11.8
RL	47.4	9.28	63.5	21.0	19.6	42.2	10.2	59.0	19.8	24.1
RDW	1.42	0.42	2.31	0.42	29.7	0.89	0.34	1.51	0.19	38.4
WDW	4.59	1.53	8.54	2.48	33.4	3.24	1.26	7.11	1.45	38.8
SDW	9.16	3.17	16.0	3.02	34.6	6.62	2.81	14.6	1.78	42.4
GDW	13.8	4.49	24.1	5.79	32.6	9.86	3.86	21.7	4.52	39.2
RSR	0.11	0.03	0.16	0.06	26.2	0.09	0.03	0.16	0.03	30.1
TDW	15.2	4.77	26.0	6.46	31.4	10.7	4.13	23.1	4.93	38.4
SN	4.26	1.25	7.00	2.00	29.4	3.14	1.21	7.00	2.00	38.6
SGN	36.6	6.66	55.5	22.8	18.2	36.5	7.28	49.8	20.3	20.0
TGW	45.0	5.10	52.1	28.1	11.3	41.5	8.33	53.0	20.1	20.1
TY	6.96	2.33	12.4	2.43	33.5	4.90	2.55	12.1	1.27	52.0

Note: S means Salt stress conditions (85.5 mM); CK means control condition; Min means Minimum; Max means Maximum; SD means Standard Deviation; CV means Coefficient of Variation. PH (Plant height); NST (Number of stem and tiller); SL (spike length); LT1 (The length of top 1st leaf); LT2 (The length of top 2nd leaf); RL (Root length); RDW (Root dry weight); WDW (Wheat straw dry weight); SDW (spike dry weight); GDW (Ground dry weight); RSR (Root shoot ratio); TDW (Total dry weight); SN (Spike number); SGN (Number of grains per spike); TGW (Thousand-grain weight); TY (Theoretical yield).

. Under control conditions, the coefficient of variation (CV) for the 16 trait indices ranged from 10.7% to 34.6%, with an average CV of 23.4%, indicating relatively low variability. Under salt stress treatment, the CV for the 16 trait indices ranged from 11.9% to 52.0%, with an average CV of 29.5%, indicating relatively high variability. This finding indicates that the 16 traits exhibit both a high degree of interdependence and independence. Under salt treatment, the traits with the highest CVs were theoretical yield, spike dry weight, aboveground dry weight, stem dry weight, spike number, root dry weight, and tiller number, with CVs ranging from 38.1% to 52.0%. This substantial variation indicates that the biomass and yield of different wheat cultivars are significantly affected by salt stress, and there are considerable differences in salt tolerance among the cultivars. Therefore, a single metric for evaluating salt tolerance in wheat is deemed inadequate.

3.2. Differential Analysis of Maturity Traits of Tested Wheat Materials Under Salt Stress

Independent-sample t-tests were performed on phenotypic traits of the tested wheat cultivars at the maturity stage to evaluate inter-treatment variations (Figure 1). The analysis revealed that nine traits, NST, LT1, LT2, RDW, WDW, SDW, GDW, TDW, and TGW, exhibited highly significant differences between treatments (p < 0.01), while no significant differences were observed for SL and SGN (p > 0.05). Additionally, PH, RL, RSR, SN, and TGW showed statistically significant differences at p < 0.05. Notably, under salt stress conditions, multiple growth and yield-related parameters, including NST, LT1, LT2, RDW, WDW, SDW, GDW, TDW, and TY, were significantly reduced compared to the control group (p < 0.05). These results demonstrate that salt stress severely impairs biomass accumulation and grain yield in the studied wheat materials.

3.3. Correlation Analysis

Correlation analysis of salt tolerance coefficients across traits under salt stress revealed 13 trait pairs with statistically significant associations (p < 0.05) and 30 pairs exhibiting highly significant correlations (p < 0.01). Among these (Figure 2), PH was positively correlated with RL and WDW at highly significant levels; NST was positively correlated with the biomass indicators (WDW, SDW, GDW, TDW, and SN) at highly significant levels; WDW was positively correlated with SDW, GDW, TDW, and SN at significant levels and negatively correlated with RSR at a significant level; SDW was positively correlated with GDW, TDW, and SN at highly significant levels, and negatively correlated with RSR at a highly significant level; TY was highly significantly positively correlated with PH, WDW, SDW, GDW, TDW, SN, and TGW. In conclusion, correlation analysis revealed a high degree of correlation between phenotypic indicators, indicating that wheat salt tolerance is a complex trait influenced by multiple factors. To address the limitations of evaluating salt tolerance using individual indicators, further application of principal component analysis and cluster analysis is necessary for comprehensive evaluation.

3.4. Principal Component Analysis

Principal component analysis (PCA) has been demonstrated to be an effective method for condensing data and simplifying indicators, thereby reducing information loss that might otherwise occur when evaluating salt tolerance using a single indicator. In this study, PCA was used to evaluate salt tolerance coefficients across 15 phenotypic parameters (

Table 3. Principal component eigenvectors and contributions.

Traits	Comprehensive Indexes
	F1	F2	F3	F4	F5
TDW	0.972	−0.081	0.022	0.016	0.170
GDW	0.966	−0.134	0.031	−0.059	0.161
SDW	0.850	−0.274	−0.038	−0.123	0.348
WDW	0.822	0.188	0.111	0.104	−0.293
SN	0.796	0.179	−0.372	0.050	−0.205
NST	0.697	0.385	−0.326	−0.159	−0.189
LT1	−0.152	0.659	−0.172	−0.411	0.355
LT2	−0.045	0.620	0.071	−0.532	0.371
SL	−0.229	0.514	0.419	−0.307	−0.317
PH	0.421	0.286	0.625	−0.018	−0.190
SGN	0.003	−0.470	0.616	−0.022	0.120
RL	0.264	0.286	0.584	0.109	−0.348
RDW	0.379	0.537	−0.080	0.690	0.133
RSR	−0.564	0.513	−0.106	0.590	0.049
TGW	0.087	0.142	0.435	0.292	0.761
EV	5.19	2.35	1.787	1.53	1.47
Weight (%)	42.1	19.1	14.5	12.4	12.0
CR (%)	34.6	15.7	11.9	10.2	9.82
CCR (%)	34.6	50.3	62.2	72.4	82.2

Note: EV means Eigen value; CR means Contributive ratio; CCR means Cumulative contributive ratio. PH (Plant height); NST (Number of stem and tiller); SL (spike length); LT1 (The length of top 1st leaf); LT2 (The length of top 2nd leaf); RL (Root length); RDW (Root dry weight); WDW (Wheat straw dry weight); SDW (spike dry weight); GDW (Ground dry weight); RSR (Root shoot ratio); TDW (Total dry weight); SN (Spike number); SGN (Number of grains per spike); TGW (Thousand-grain weight); TY (Theoretical yield).

), successfully transforming these variables into five mutually orthogonal composite indices (F1–F5). The resulting principal components had eigenvalues, weighting coefficients, and variance contributions as follows (Table 3): the first five principal components accounted for 34.6%, 15.7%, 11.9%, 10.2%, and 9.82% of the total variance, respectively, achieving a cumulative explanatory power of 82.2%. This transformation showed that over four-fifths of the information from the original 15 traits could be effectively captured by the five synthesized indices. Notably, the first principal component (F1) dominated the model with the highest variance contribution (34.6%) and weighting coefficient (42.1%). Its eigenvector was primarily loaded by biomass-related parameters, including dry matter weight and tiller number—indicating that enhanced biomass allocation is a key physiological determinant of salt tolerance in wheat. Consequently, F1 emerges as a scientifically validated composite metric for systematic salt tolerance phenotyping.

3.5. Evaluation of the Membership Function and Composite D-Value

Principal component analysis was assess to calculate the contribution rates and D-values (

Table 4. Principal component values, membership function values and D-values of each wheat cultivar.

Number	Comprehensive Index					Membership Function					D Value
	x1	x2	x3	x4	x5	μ1	μ2	μ3	μ4	μ5
1	0.14	1.38	−2.68	−0.21	0.77	0.70	0.85	0.00	0.46	0.71	0.60
2	0.60	−2.37	−1.90	−2.04	1.24	0.81	0.00	0.15	0.04	0.81	0.47
3	0.64	0.11	−0.94	−0.81	0.47	0.82	0.56	0.34	0.32	0.65	0.62
4	−0.08	1.46	1.90	−2.23	−2.28	0.65	0.86	0.88	0.00	0.09	0.58
5	0.35	2.07	−1.43	−0.51	0.36	0.75	1.00	0.24	0.39	0.63	0.67
6	1.35	0.65	−0.73	0.19	0.76	1.00	0.68	0.38	0.55	0.71	0.76
7	0.84	0.41	2.51	−0.51	1.05	0.87	0.63	1.00	0.39	0.77	0.77
8	0.82	0.51	0.60	−0.72	−0.81	0.87	0.65	0.63	0.34	0.39	0.67
9	0.55	−0.90	−0.18	−1.05	−0.89	0.80	0.33	0.48	0.27	0.37	0.55
10	1.16	0.93	0.35	−0.21	0.37	0.95	0.74	0.58	0.46	0.63	0.76
11	1.02	0.50	0.69	−0.74	0.24	0.92	0.65	0.65	0.34	0.60	0.72
12	0.67	−1.30	−0.31	−0.25	−0.98	0.83	0.24	0.46	0.45	0.36	0.56
13	0.57	−0.17	−0.10	0.05	−0.33	0.81	0.50	0.50	0.51	0.49	0.63
14	−0.57	−0.33	1.92	−0.72	2.19	0.53	0.46	0.89	0.34	1.00	0.60
15	1.18	−0.14	−0.46	0.40	0.06	0.96	0.50	0.43	0.59	0.57	0.70
16	0.59	0.95	−0.34	−0.13	0.84	0.81	0.75	0.45	0.47	0.73	0.70
17	1.22	−1.08	0.22	0.45	0.79	0.97	0.29	0.56	0.61	0.72	0.70
18	−2.71	−1.54	−0.72	−0.57	0.36	0.00	0.19	0.38	0.37	0.63	0.21
19	0.92	−0.64	0.11	0.90	−0.03	0.89	0.39	0.54	0.71	0.55	0.68
20	−1.36	1.20	0.35	−0.73	−0.55	0.33	0.80	0.58	0.34	0.44	0.47
21	−1.01	0.46	−0.82	−0.39	−2.01	0.42	0.64	0.36	0.42	0.15	0.42
22	0.10	−0.81	0.42	0.49	−0.10	0.69	0.35	0.60	0.61	0.53	0.58
23	−0.21	−0.59	−0.13	1.68	−0.36	0.61	0.40	0.49	0.88	0.48	0.57
24	−0.43	−0.03	−0.78	0.08	−2.73	0.56	0.53	0.37	0.52	0.00	0.45
25	0.30	−0.38	−0.08	2.19	−0.95	0.74	0.45	0.50	1.00	0.36	0.64
26	0.32	0.55	−0.17	0.84	−0.07	0.75	0.66	0.48	0.69	0.54	0.66
27	0.13	0.21	0.21	−0.04	−0.17	0.70	0.58	0.56	0.49	0.52	0.61
28	−1.77	1.66	0.08	0.53	1.49	0.23	0.91	0.53	0.62	0.86	0.53
29	−0.52	0.32	−1.20	0.83	0.35	0.54	0.61	0.29	0.69	0.63	0.54
30	−1.48	0.13	0.64	0.61	1.24	0.30	0.56	0.64	0.64	0.81	0.50
31	1.00	−0.15	0.41	1.10	−0.59	0.91	0.50	0.60	0.75	0.44	0.71
32	−0.81	−0.41	−0.26	−0.18	0.27	0.47	0.44	0.47	0.46	0.61	0.48
33	0.13	−1.09	0.66	1.75	−0.03	0.70	0.29	0.64	0.90	0.55	0.62
34	−2.15	−0.43	0.41	1.07	−0.47	0.14	0.44	0.60	0.75	0.46	0.37
35	−0.68	1.36	0.29	0.81	0.49	0.50	0.84	0.57	0.69	0.65	0.62
36	−0.98	−1.38	0.16	−2.14	0.21	0.42	0.22	0.55	0.02	0.60	0.37
37	0.19	−1.12	1.30	0.22	−0.20	0.71	0.28	0.77	0.55	0.51	0.60

). The membership function method was employed to determine the membership function values. As shown in the table, the principal components had weight values of 42.1, 19.1, 14.5, 12.4, and 12.0, respectively. Using these weights and membership function values, a comprehensive salt tolerance evaluation (D-value) was calculated for each wheat cultivar at maturity. The membership function values and D-values for each cultivar are presented in the table. A higher D-value indicates stronger salt tolerance, while a lower D-value indicates weaker salt tolerance. This method provides an objective approach to evaluating the salt tolerance of the tested wheat cultivars. Among the tested cultivars, Yangfumai 8 exhibited the strongest salt tolerance, followed by Zhongkenmai 212. Yangmai 22 showed the weakest salt tolerance, with Nongmai 158 ranking second to last.

3.6. Regression Analysis

In this study, the D-value was designated as the dependent variable, and the salt tolerance coefficients of 15 indicators under salt stress were used as independent variables in a stepwise regression analysis to establish the optimal regression equation: D = −0.381 + 0.336 × TDW + 0.222 × RDW + 0.392 × PH + 0.161 × TGW + 0.091 × NST (R² = 0.979).

Table 5. Analysis of the estimation accuracy of the regression equation.

Varieties	Primary Value	Regression	Difference	Evaluation Accuracy (%)
Yangmai 25	0.62	0.60	0.02	96.06
Yangmai 20	0.48	0.47	0.01	97.42
Yangmai 18	0.63	0.62	0.01	98.69
Zhenmai 12	0.55	0.58	−0.03	94.93
Huamai 33	0.67	0.67	0.00	99.28
Huamai 8	0.75	0.76	−0.01	98.50
Yangfumai 8	0.77	0.77	0.00	99.92
Luomai 163	0.65	0.67	−0.02	97.23
Yangfumai 9	0.54	0.55	−0.01	98.26
Zhongkenmai 212	0.73	0.76	−0.03	95.54
Yangmai 21	0.71	0.72	−0.01	98.44
Suyanmai 017	0.57	0.56	0.01	98.25
Sukenmai 1008	0.62	0.63	−0.01	98.67
Yangmai 4	0.59	0.60	−0.01	98.01
Yangfumai 5145	0.69	0.70	−0.01	98.01
Yangmai 5	0.67	0.70	−0.03	96.06
Yangmai 16	0.73	0.70	0.02	96.76
Yangmai 22	0.20	0.21	−0.01	94.05
Yangmai 28	0.69	0.68	0.01	98.94
Yangmai 15	0.47	0.47	0.00	99.82
Ningmaizi 166	0.42	0.42	0.00	99.47
Nongmai 152	0.56	0.58	−0.03	95.55
Yangmai 1	0.56	0.57	−0.01	98.37
Shanrong 3	0.45	0.45	0.00	99.75
Huaimai 46	0.65	0.64	0.02	97.58
Xumai 2023	0.69	0.66	0.03	95.90
Qianmai 088	0.59	0.61	−0.02	96.91
Yangfumai 10	0.53	0.53	0.00	99.84
Xumai 38	0.55	0.54	0.01	98.71
Nongmai 126	0.51	0.50	0.01	98.44
Ruihuamai 521	0.72	0.71	0.01	98.22
Xumai 178	0.51	0.48	0.03	93.36
Ruihua 518	0.65	0.62	0.03	95.77
Nongmai 158	0.37	0.37	0.00	99.54
Huamai 1430	0.59	0.62	−0.03	95.56
Yangmai 23	0.40	0.37	0.02	94.66
Huaihemai 16132	0.60	0.60	0.01	98.67

details the accuracy of the regression equation and its predictive performance for the wheat cultivars included in the analysis. The regression equation achieved an accuracy exceeding 90%, with a maximum prediction accuracy of 99.84% and a minimum of 93.36%. The mean percentage of correct predictions was 96.72%. The regression equation indicated that total dry weight (TDW) was the most sensitive to salt stress among the individual parameters, followed by root dry weight (RDW), plant height (PH), thousand-grain weight (TGW), and number of stem and tiller (NST). These results demonstrate that increases in these parameters correspond to an increase in the D-value, indicating enhanced salt tolerance in wheat cultivars. The use of these indicators provides a rapid and efficient method for assessing wheat salt tolerance, and the mathematical model can be employed for the expedited evaluation of salt tolerance in wheat at maturity.

3.7. Random Forest Model Prediction

The results of the random forest model analysis (Figure 3) indicated that spike dry weight, ground dry weight, and total dry weight are the strongest predictors of theoretical yield, with their feature importance proportions of 19.57%, 16.84%, and 15.43%, respectively, all reaching an extremely significant level (p < 0.001). The feature importance proportions of plant height and thousand-grain weight were 9.50% and 6.11%, respectively, both achieving a significant level (p < 0.05). Spike number, stem dry weight, and number of stems and tillers also showed predictive importance and made positive contributions to improving model prediction accuracy. In contrast, the feature importance proportions of morphological indicators such as spike length and root length were relatively low, at only 0.74% and 0.39%, respectively. This suggests that within the material types and treatment conditions set in this experiment, such indicators have limited marginal explanatory power for yield variation.

3.8. Systematic Cluster Analysis

Cluster analysis, using Ward’s method based on D-values, classified the 37 wheat cultivars into five categories of salt tolerance (Figure 4). These categories, along with the specific cultivars in each, are as follows: Strong salt tolerance: Six cultivars were classified in this group Yangfumai 8, Zhongkenmai 212, Huamai 8, Yangmai 21, Ruihuamai 521, and Yangmai 16. Salt tolerance: This category included eleven cultivars: Yangfumai 5145, Yangmai 5, Yangmai 28, Luomai 163, Huamai 33, Xumai 2023, Huaimai 46, Sukenmai 1008, Yangmai 18, Ruihua 518, and Yangmai 25. Moderate salt tolerance: Eleven cultivars exhibited moderate salt tolerance: These were Huamai 1430, Qianmai 088, Yangmai 4, Huaihemai 16132, Nongmai 152, Zhenmai 12, Yangmai 1, Suyanmai 017, Yangfumai 9, Xumai 38, and Yangfumai 10. Weak salt tolerance: Eight cultivars were found to have weak salt tolerance: Nongmai 126, Xumai 178, Yangmai 15, Yangmai 20, Shanrong 3, Ningmaizi 166, Yangmai 23, and Nongmai 158. Salt sensitivity: Only one cultivar, Yangmai 22, was identified as salt-sensitive.

Hierarchical cluster analysis was also conducted using Ward’s method based on the salt-tolerant high-yield index (STHYI), classifying 37 wheat genotypes into five distinct salt tolerance categories (Figure 5): highly salt-tolerant, salt-tolerant, moderately salt-tolerant, weakly salt-tolerant, and salt-sensitive. The classification results are as follows: Strong salt tolerance: This category included two cultivars, Yangfumai 8 and Zhongkenmai 212. Salt tolerance: Eight cultivars were classified as salt-tolerant. These were Yangmai 16, Huaimai 46, Huamai 8, Yangfumai 9, Huamai 33, Yangmai 1, Ruihua 518, and Qianmai 088. Moderate salt tolerance: Nine cultivars exhibited moderate salt tolerance. They were Ruihuamai 521, Nongmai 152, Huaihemai 16132, Yangmai 21, Suyanmai 017, Yangmai 28, Yangmai 4, Yangmai 20, and Luomai 163. Weak salt tolerance: Thirteen cultivars were found to have weak salt tolerance. These included Yangfumai 5145, Huamai 1430, Xumai 2023, Nongmai 158, Xumai 178, Yangmai 5, Nongmai 126, Yangfumai 10, Yangmai 22, Sukenmai 1008, Yangmai 18, Xumai 38, and Yangmai 23. Salt sensitivity: Five cultivars were identified as salt-sensitive: Ningmaizi 166, Shanrong 3, Yangmai 15, Zhenmai 12, and Yangmai 25.

3.9. Boston Matrix Analysis

A Boston Matrix plot was generated, with the D-value on the x-axis and the salt-tolerant high-yield index (STHYI) on the y-axis (Figure 6). As showed in the diagram, cultivars such as Yangfumai 8, Zhongkenmai 212, Yangmai 16, and Huamai 8 exhibited both high D-values and substantial salt tolerance, as well as high-yield indices. These cultivars were identified as the most salt-tolerant through a multifaceted evaluation approach. Some cultivars, including Yangfumai 9, Yangmai 1, Suyanmai 017, and Yangmai 20, showed strong salt tolerance and high-yield indices but lower D values. Other cultivars, such as Yangfumai 5145, Huamai 1430, Xumai 2023, Yangmai 5, Sukenmai 1008, Yangmai 18, and Yangmai 25, exhibited low salt tolerance and high-yield indices but relatively higher D-values. Cultivars with low salt tolerance, high-yield indices, and low D-values included Yangmai 23, Ningmaizi 166, Shanrong 3, and Yangmai 15. A comprehensive analysis was conducted to determine the salt tolerance indices of the specimens in the study. The results indicated that Yangmai 22 had the lowest salt tolerance index and the smallest D-value.

Meanwhile, A Boston Matrix plot was also generated with the D-value as the x-axis and the yield salt tolerance coefficient as the y-axis (Figure 7). Several cultivars, including Yangmai 20, Huamai 33, Sukenmai 1008, Yangfumai 5145, Yangmai 5, and Xumai 2023, showed significant discrepancies from the results in Figure 6. The remaining cultivars performed consistently across both analyses, indicating that the salt tolerance of the cultivars, as derived from these statistical results is reliable.

The horizontal and vertical lines represent the average values of the D value and the salt-tolerance and high-yield index. The numbers represent the variety codes.

The horizontal and vertical lines represent the average values of the D-value and the yield-salt-tolerance coefficient. The numbers represent the variety codes.

3.10. Verification of Salt Tolerance Based on Physiological Indicators of Wheat in Pot Experiments

As salt concentration increased, the net photosynthetic rate (Pn) and peroxidase (POD) activity of the salt-tolerant wheat cultivar Yangfumai 8 showed a downward trend, while superoxide dismutase (SOD) and catalase (CAT) activities first increased and then decreased. In contrast, malondialdehyde (MDA) and proline contents showed an upward trend. For the salt-sensitive cultivar Yangmai 22, higher salt concentrations led to decreases in net photosynthetic rate, SOD, POD, and CAT activities, along with increases in MDA and proline contents. Compared with the control group, Yangfumai 8 treated with 68.4 mM salt showed reductions in net photosynthetic rate, SOD, POD, and CAT activities by 36.0%, 22.0%, 38.7%, and 19.7%, respectively, while MDA and proline contents increased by 77.1% and 131.1%, respectively. In contrast, Yangmai 22 under 68.4 mM salt stress exhibited a 54.31% decrease in net photosynthetic rate, and 69.4%, 46.7%, and 54.3% reductions in SOD, POD, and CAT activities, respectively. At the same time, MDA and proline contents increased by 156.2% and 93.9%, respectively. Among the measured physiological indices, Yangfumai 8 and Yangmai 22 differed significantly in all indices except POD (

Table 6. Effects of salt stress on net photosynthetic rate (Pn), activities of SOD, POD, and CAT, and contents of MDA and proline in wheat cultivars Yangfumai 8 and Yangmai 22 under pot conditions.

Cultivar	Treatment (mM NaCl)	Pn mmol CO2 m−2 s−2	SOD U g−1 FW	POD U g−1 FW	CAT U g−1 FW	MDA nmol g−1 FW	Pro mg g−1 FW
Yangfumai 8	0	18.9 a	1616.0 a	472.7 a	63.3 a	25.9 c	152.3 c
	34.2	15.5 b	1811.7 b	385.7 b	68.7 b	33.1 b	279.7 b
	68.4	12.0 c	1259.3 c	290.0 c	43.6 c	45.9 a	352.0 a
Yangmai 22	0	16.8 a	1438.7 a	490.0 a	54.7 a	24.8 c	150.7 c
	34.2	13.5 b	1050.0 b	357.3 b	37.3 b	45.4 b	250.0 b
	68.4	7.80 c	440.7 c	261.3 c	25.0 c	63.1 a	291.3 a
Significance of factors
Cultivar		**	**	NS	**	**	**
Treatment		**	**	**	**	**	**
Cultivar × treatment		**	**	NS	**	**	**

Values within the same column followed by different lowercase letters indicate significant differences at the p = 0.05 level. ** denote significant differences at p < 0.01 levels; NS indicates non-significant differences.

).

3.11. Verification of Salt Tolerance Based on Yield Components of Wheat in Pot Experiments

Two wheat cultivars with contrasting salt tolerance were selected for the pot experiment. Under 34.2 mM and 68.4 mM salt stress, the number of spikes per pot, grains per spike, grain weight, and yield were all lower than those in the control group (

Table 7. Effects of salt stress on yield components of wheat cultivars Yangfumai 8 and Yangmai 22 under pot conditions.

Cultivar	Treatment (mM NaCl)	Spike Number (pot−1)	Number of Grains per Spike	Kernel Weight (mg)	Actual Yield (g pot−1)
Yangfumai 8	0	43.3 a	45.3 a	44.6 a	86.9 a
	34.2	41.9 a	42.1 b	41.0 b	76.6 b
	68.4	38.0 b	40.8 b	38.2 c	59.3 c
Yangmai 22	0	38.0 a	46.3 a	41.1 a	72.3 b
	34.2	24.7 b	34.7 b	32.7 b	28.0 d
	68.4	14.3 c	18.8 c	30.3 c	8.16 e
Significance of factors
Cultivar		**	**	**	**
Treatment		**	**	**	**
Cultivar × treatment		**	**	**	**

Values within the same column followed by different lowercase letters indicate significant differences at the p = 0.05 level. ** denote significant differences at p < 0.01 levels.

), with reductions increasing as salt concentration rose. For the salt-tolerant cultivar Yangfumai 8, under 34.2 mM salt stress, the number of spikes per pot, grains per spike, grain weight, and yield decreased by 3.22%, 7.05%, 7.93%, and 12.52%, respectively, compared with the control. At 68.4 mM, the number of spikes per pot, grains per spike, grain weight, and yield decreased by 12.3%, 14.3%, 9.92%, and 32.3%, respectively.

For Yangmai 22 (the salt-sensitive cultivar), under 34.2 mM salt stress, the four indices (spikes per pot, grains per spike, grain weight, yield) decreased by 35.1%, 25.2%, 20.4%, and 61.3%, respectively. At 68.4 mM, the reductions were more pronounced: spikes per pot, grains per spike, grain weight, and yield decreased by 62.3%, 59.4%, 26.2%, and 88.7%, respectively. Significant differences were observed between Yangfumai 8 and Yangmai 22 in the number of spikes per pot, the number of grains per spike, grain weight, and yield.

4. Discussion

The selection and breeding of salt-tolerant wheat germplasm constitutes a pivotal strategy for the utilization of China’s extensive saline-alkaline land resources, which are imperative for the preservation of global food security [27,28]. However, conventional screening methodologies, frequently confined to the germination or seedling phases, are inadequate in predicting yield under prolonged salt stress, as tolerance can exhibit substantial variation across these developmental stages [29,30]. While field or pot-based evaluations are ecologically relevant, they are hindered by uncontrolled environmental variability, which hinders the establishment of standardized phenotyping protocols [12,31]. In order to address this issue, a full-life-cycle hydroponic screening system was developed, which provides precise control over salt stress and enables high-throughput evaluation. This study integrates this innovative platform with multivariate statistical analysis to achieve a synergistic evaluation of salt tolerance and yield potential, offering new insights for breeding.

4.1. A Reliable Hydroponic System for Integrated Salt Tolerance Evaluation

The controlled hydroponic system utilized in this study incorporates homogeneous salt distribution and root-zone oxygenation, enabling the precise evaluation of 37 wheat cultivars across 16 maturity-stage traits. Multivariate analysis revealed that biomass accumulation (TDW, RDW) and its allocation (source-sink relationships) were the dominant factors governing yield under salt stress, aligning with the emphasis on biomass as a core tolerance indicator [32,33]. In order to overcome the redundancy present among the correlated traits, Principal Component Analysis (PCA) was employed to distil the fifteen phenotypic indicators into a total of five independent composite factors, which collectively captured 85.2% of the total variance. This dimensionality reduction formed the basis for a comprehensive evaluation (D-value) via the membership function method. Subsequent stepwise regression established a highly accurate predictive model (R² > 0.97) for the D-value using key traits such as TDW and PH, thereby confirming the robustness of the integrated evaluation approach. Most innovatively, the application of the Boston Matrix analysis, combining the D-value with a salt-tolerant high-yield index, successfully identified cultivars such as Yangfumai 8 and Zhongkenmai 212 that break the typical trade-off between tolerance and productivity. This finding is consistent with the “salt tolerance-high yield synergistic regulatory gene” mechanism identified by Li Jiayang’s team in rice [34].

4.2. Linking Phenotype to Physiology: Mechanisms and Field Predictivity

A critical consideration is the translatability of hydroponic screenings to field performance. While hydroponics controls confounding soil factors, key physiological differences exist. In soil, root architecture, soil porosity, and water dynamics create heterogeneous ion and oxygen gradients absent in hydroponics [35]. Our system mitigates this by simulating a key field stress: sustained ionic toxicity. The high correlation between our hydroponic D-value and subsequent pot trial performance (lower yield reduction in tolerant cultivars) suggests that our method captures constitutive physiological mechanisms critical for field success. The pot test results demonstrated that the yield of the salt-tolerant variety Yangfumai 8 was diminished by 11.8% and 31.8% at 34.2 mM and 68.4 mM salt concentrations, respectively, and the yield of the salt-sensitive variety Yangmai 22 was reduced by 61.3% and 88.7%, respectively. The identified tolerant cultivars have been shown to exhibit superior phenotypes in terms of several key physiological characteristics, including maintenance of biomass, plant height, and total grain weight. These observations suggest the presence of a coordinated physiological mechanism that underpins the observed agronomic benefits. Firstly, effective Na⁺ and K⁺ homeostasis and selectivity in roots, likely mediated by transporters such as SOS1 and HKT1, minimizes cytotoxic Na⁺ accumulation while preserving essential K⁺ for enzymatic functions [36]. Secondly, osmotic adjustment via compatible solutes (e.g., proline, glycine betaine) and ion compartmentalization into vacuoles has been demonstrated to help maintain turgor and cellular water content, thereby directly supporting the sustained growth that was observed. Thirdly, the mitigated reduction in photosynthetic parameters in tolerant lines suggests the presence of more robust mechanisms to manage reactive oxygen species (ROS) generated under salt stress, possibly through enhanced antioxidant enzyme activity (e.g., SOD, CAT) [37]. Finally, differential hormonal responses, particularly in ABA (stress signaling) and auxin (growth promotion), may orchestrate the observed balance between stress adaptation and yield component preservation (TGW, NST) [38]. Subsequent research will involve the profiling of these physiological and molecular traits in the cultivars identified in this study, with the objective of directly testing these hypotheses and establishing a correlation between our phenotypic screening and the causal genes.

4.3. Implications for Breeding and Practical Application

The integrated evaluation framework that has been developed provides a practical pipeline for the execution of breeding programs. The hydroponic platform facilitates the expeditious and economical screening of extensive germplasm collections at the mature plant stage, thereby signifying a substantial advancement over conventional single-stage assays. The PCA-Boston Matrix approach represents an advancement in the field of plant breeding by extending the conventional single-trait selection method. This approach identifies genotypes that combine multiple favorable attributes, including tolerance and yield, thereby enhancing selection efficiency. The specific cultivars identified, such as Yangfumai 8, serve as excellent donor parents for hybridization. Furthermore, the key predictive traits identified (e.g., TDW, PH) can serve as proxies for complex physiological tolerance in field-based selection. While final validation in target saline fields is imperative, the present method provides a powerful pre-screening tool that enriches the breeding population with high-probability candidates, thereby significantly accelerating the development of salt-tolerant, high-yielding wheat varieties.

5. Conclusions

This study establishes a novel, integrated framework for the efficient evaluation and selection of salt-tolerant wheat germplasm. By combining a controlled, full-life-cycle hydroponic screening system with multivariate statistical analysis, we successfully transformed complex phenotypic data into a robust composite evaluation index (D-value). The subsequent validation of contrasting cultivars (Yangfumai 8 vs. Yangmai 22) in a pot experiment not only confirmed the reliability of the hydroponic screening but also linked the selected agronomic traits to key physiological mechanisms-such as enhanced antioxidant enzyme activity (SOD, POD, CAT) and osmoregulation (proline accumulation)-underpinning salt tolerance. Crucially, the five simplified, yield-related indicators (TDW, RDW, PH, TGW and NST) identified through stepwise regression offer a practical tool for high-throughput preliminary screening in breeding programs. Although developed under controlled conditions, this system and the diagnostic indicators provide a scientifically rigorous platform for pre-field selection. The identification of highly tolerant (e.g., Yangfumai 8, Zhongkenmai 212) and sensitive (Yangmai 22) cultivars offers ideal, well-characterized materials for both applied breeding and in-depth physiological research. In summary, it delivers a transferable methodology that can accelerate the breeding of salt-tolerant wheat, contributing directly to strategies aimed at stabilizing yield and ensuring food security in salinization-affected regions.