Genotypic Specificity in Pea Under Increasing Salinity Stress

1. Introduction

Pea (Pisum sativum L.) is an important grain legume crop with high nutritional quality, agronomic importance and ecological benefits. Pea seeds represent a valuable source of protein, dietary fiber, starch, vitamins, minerals and bioactive compounds, which support peas’ growing significance in human nutrition [1]. In addition to its nutritional importance, pea plays a significant role in sustainable agriculture through symbiotic nitrogen fixation, which can improve soil fertility and reduce dependence on mineral nitrogen fertilizers [2]. Given its high protein content and relatively low environmental footprint, pea has also emerged as an important component of plant-based diets and climate-resilient agricultural production systems. However, pea productivity is increasingly threatened by abiotic stress, among soil salinity is considered one of the major constraints limiting crop growth and yield worldwide [3]. Climate change has accelerated the expansion of salt-affected soils through rising temperatures, irregular precipitation, prolonged drought conditions and high dependence from irrigation, particularly in arid and semi-arid regions [4]. Additionally, intensive agricultural practice with unappropriated irrigation and excessive fertilizer application have accelerated secondary salinization, reduced soil productivity and threatening global food security [5]. Apart from naturally soil salinity, secondary salinization resulting from irrigation practices and rising groundwater has become an increasing concern in agricultural production systems. Approximately 20% of irrigated agricultural soils worldwide are currently affected by salinity, resulting in significant reductions in crop productivity [6,31,32,33]. Future projections indicate that the area of salt-affected soils will continue to increase, emphasizing the urgent need to identify and develop crop genotypes with improved salinity tolerance. Salinity affects plant development through a two-phase process involving an initial osmotic component that limits water uptake and cell expansion, followed by progressive ionic toxicity caused by excessive accumulation of Na⁺ and Cl⁻ ions in plant tissues [6,29,34]. These processes reduce nutrient utilization, photosynthetic activity, enzyme function and cellular metabolism, ultimately leading to yield reduction [6,29,34]. Soil salinity negatively affects seed germination, vegetative growth, biomass accumulation and reproductive development [9,10,12]. Prolonged soil salinity also causes excessive production of reactive oxygen species, which damage membrane lipids, proteins and nucleic acids unless efficiently detoxified by antioxidant defiance systems [29,34]. Recently, controlled-environment agriculture (CEA) and vertical farming (VF) systems have attracted considerable attention as innovative approaches for sustainable crop production. These systems provide precise regulation of environmental conditions and efficient management of water, nutrients and light, while providing accurate replication of trial conditions [7,8,37]. As a result, vertical farming systems are increasingly being used for abiotic stress experiments, physiological studies and screening crop genotypes for stress tolerance. Compared with conventional field experiments, controlled environment significantly reduces the impact of environmental variation and enables assessment of plant responses to individual stress factors, with improved accuracy and repeatability of experimental trials. Besides genetic approaches, agronomic remediation strategies such as the cultivation of halophytic species have also been proposed as an alternative way for reclaiming and managing salt-affected soils [44].

Pea is considered one of the most adaptable legume crops, but sensitive to soil salinity, particularly during vegetative growth and reproductive development [10,11]. Salinity reduces numerous physiological processes, resulting in nutrient imbalance, reduced biological nitrogen fixation, flower abortion, impaired photosynthesis, oxidative stress and lower yield [12,13]. Despite the pea is recognized for its wide adaptability, it showed lower salinity tolerance compared with several other cultivated legumes, especially during the reproductive stage, when yield formation is highly sensitive to abiotic stress [14,15]. Since reproductive development largely determines final seed yield, understanding genotype-specific responses during this developmental stage is important for breeding programs targeting saline environments. Pea responses to salinity vary significantly among various genotypes, indicating significant genetic variability in tolerance mechanisms. Previous studies have reported that pea can tolerate moderate salinity levels, with limited reductions in plant growth, physiological traits and yield production [19,20]. Within this range of moderate salinity, plants may maintain metabolic activity through osmotic adjustment, ion homeostasis and activation of antioxidant defiance systems [21,29,30,36]. Moderate salinity may induce hormetic responses, in which mild stress activates adaptive physiological and metabolic mechanisms that enhance growth and development [22,23]. However, significant decreases in yield related traits, such as a pod number, seed production, seed weight and biomass production are frequently observed in conditions when salinity levels rise above approximately 150 mM NaCl [12,17,22]. The magnitude of these responses varies considerably between various genotypes, reflecting differences in their capacity to regulate ion transport, osmotic balance, antioxidant metabolism. Despite considerable progress in understanding salinity responses in pea, limited information is available regarding genotype-specific responses under controlled-environment conditions relevant to vertical farming systems. Moreover, only a small number of studies have simultaneously assessed multiple yield-related traits across a range of salinity levels under highly controlled environments, where the influence of confounding environmental factors is minimized. Furthermore, previous studies have focused primarily on initial stages, germination, seedling growth and physiological responses, while reproductive development and yield stability under controlled salinity conditions have been insufficiently investigated. This gap limits the identification of stable genotypes suitable for soil saline agricultural regions and future controlled-environment production. Since yield-related traits directly determine final yield productivity, their evaluation is important for identifying salt-tolerant genotypes, supporting breeding programs and production for soil salinity environments. Controlled vertical farming has previously been used to characterize genotype-specific responses of pea to different water regime and water-deficit stress [39]. The application of a comparable experimental approach to salinity enables an opportunity to estimate another major abiotic stress under standardized environmental conditions. Such experiments can contribute to the identification of genetic variability, which is useful for subsequent screening salinity tolerance and breeding, while providing a controlled basis for further physiological and field validation. Therefore, the aim of this study was to estimate genotype responses to increasing soil salinity levels by assessing plant growth, yield and yield-related traits of five pea genotypes cultivated under controlled-environment vertical farming conditions.

2. Materials and Methods

2.1. Plant Material and Experimental Conditions.

The experimental material used in this study comprised five pea (Pisum sativum L.) genotypes: Favorit (G1), Korvin (G2), Mali Provansalac (G3), Vilo (G4), and Virtus (G5). These genotypes were selected because of their contrasting agronomic performance and presumed differences in salinity tolerance, allowing the evaluation of genotype-dependent responses under controlled salinity conditions. The experiment was conducted during two consecutive growing cycles in a controlled-environment vertical farming system (Meta-Plast d.o.o., Pitomača, Croatia). The first growing cycle lasted from 9 January to 4 March 2025, whereas the second was conducted from 17 March to 12 May 2025. Seeds of each genotype were germinated in plastic trays under standardized conditions. Healthy and uniformly developed seedlings were subsequently transplanted into 2-L plastic pots when they reached approximately 5 cm in height. Plants were grown under a 14-h light/10-h dark photoperiod provided by LED lamps. The light regime remained constant throughout both experimental cycles to ensure uniform plant growth and minimize environmental variability among treatments. Environmental conditions were maintained at 22 ± 3 °C and 75 ± 5% relative humidity, and all environmental parameters were continuously monitored throughout both growing cycles to ensure identical growing conditions for all experimental treatments. Pots were arranged on cultivation shelves (4.0 × 0.6 m) in a completely randomized design to minimize positional effects and environmental variability within the vertical farming system. Plants were cultivated in a commercial substrate, Potgrond H (Klasmann-Deilmann GmbH, Germany), composed of frozen black sphagnum peat and finely structured white sphagnum peat, enriched with water-soluble NPK fertilizer (14:10:18) and essential micronutrients. The substrate had a fine structure and a pH ranging from 5.5 to 6.5. This substrate was selected because of its high uniformity, excellent water-holding capacity, favourable aeration properties, and suitability for controlled-environment crop production. No additional fertilizers were supplied during the experiment to avoid potential interactions between nutrient availability and salinity treatments.

2.2. Experimental Design and Salinity Treatments

The main experimental factor was the salinity of the irrigation solution, with three NaCl concentrations: 100 mM (T1), 150 mM (T2), 200 mM (T3), and non-saline control, 0 mM NaCl (C). Plants were irrigated every three days with the respective NaCl treatment solutions, while control plants received water without adding NaCl. All remaining cultivation conditions were maintained constant throughout the experiment, ensuring that salinity was the only experimental factor differing among treatments. For each genotype × salinity treatment combination, three replications were established, each consisting of three plants grown in one pot. Salinity treatments were initiated after seedling establishment to ensure uniform plant development before stress application. The NaCl concentrations remained constant throughout each growing cycle. The experiment was independently repeated during the second growing cycle using the same experimental design and cultivation conditions to verify the reproducibility of genotype responses. At physiological maturity, the following traits were evaluated: number of pods per plant, pod length (cm), number of seeds per plant, seed weight per plant (g), fresh biomass (g) and dry biomass (g). Fresh biomass was measured immediately after harvest, whereas dry biomass was determined after oven-drying the plant material at 65 °C to constant weight.

2.3. Statistical Analysis

Statistical analyses were performed using Python version 3.13.0 (Python Software Foundation, Wilmington, DE, USA) with the pandas, NumPy, SciPy, and stats models libraries. Descriptive statistics were calculated for all evaluated traits. Prior to the analysis of variance, the assumptions of normality and homogeneity of variance were evaluated using the Shapiro–Wilk and Levene tests, respectively. A two-way analysis of variance (ANOVA) was performed separately for each growing cycle to evaluate the effects of genotype, salinity treatment, and their interaction on the analysed traits. Whenever significant differences were detected, treatment means were separated using Tukey’s Honestly Significant Difference (HSD) post hoc test at a significance level of p < 0.05. Results are presented as mean ± standard deviation (SD). To quantify the magnitude of salinity-induced changes relative to the control treatment, the percentage reduction of each yield-related trait was calculated as:

Reduction (%) = [(C − T)/C] × 100

where C represents the mean value recorded under the control treatment and T represents the mean value recorded under the corresponding salinity treatment. Positive values indicate reductions relative to the control treatment and negative values indicate increases. This parameter was used to facilitate comparison of genotype tolerance across various NaCl concentrations and to assess the significance of salinity-induced reductions in yield-related traits. In addition to absolute trait values, relative changes (%) were calculated to provide a standardized comparison of genotype responses across different salinity levels, thereby facilitating the identification of tolerant and sensitive genotypes irrespective of their baseline performance under control conditions. Statistical significance was considered at p < 0.05 unless otherwise stated.

3. Results

The results revealed considerable variability among pea genotypes in their response to different salinity treatments across two growing cycles. Analysis of variance (ANOVA) showed significant (p < 0.05) and highly significant (p < 0.01) effects of genotype, salinity treatment and their interaction on several evaluated traits (Table 1). Overall, genotype showed the predominant source of variation for most evaluated traits in both growing cycles. Significant genotype effects were detected for all yield-related and biomass traits, whereas the effects of salinity treatment varied depending on the trait and growing cycle. Genotype × treatment interactions were non-significant for most evaluated traits, indicating that differences among genotypes in their responses to salinity treatments were not statistically significant for these traits. This suggests relatively stable genotype performance across salinity treatments. In the first growing cycle (GC1), salinity treatment significantly affected pod length, seed weight per plant, fresh biomass and dry biomass. Genotype effects were highly significant for all estimated traits. The highest F-values were recorded for fresh biomass (F = 19.37) and dry biomass (F = 23.65), indicating that these traits were the strongly influenced by genetic differences. The genotype × treatment interaction was significant only for pod length, while it remained non-significant for other evaluated traits. In the second growing cycle (GC2), genotype remained the dominant factor affecting trait variability. The strongest genotype effects were observed for fresh biomass (F = 18.96), dry biomass (F = 15.20) and seed weight per plant (F = 13.99). Salinity treatment significantly affected fresh biomass, whereas its effect on dry biomass and reproductive traits was generally non-significant. Genotype × treatment interactions were non-significant for all evaluated traits in GC2, indicating that differences among genotypes in their responses to salinity treatments were not statistically significant.

Tukey’s HSD test indicated that differences among salinity treatments were limited for most traits (Table 2). Significant differences were detected mainly for pod length, seed weight per plant, and biomass-related traits. Significant differences in pod length were observed between the control treatment and T3, whereas fresh biomass differed significantly between the control treatment and T2. No significant differences were detected among treatments for number of pods per plant, number of seeds per plant or dry biomass.

3.1. Number of Pods per Plant

ANOVA revealed that salinity treatment did not significantly affect the number of pods per plant in either growing cycle, while genotype effects were highly significant (p < 0.01) in both GC1 and GC2 (Table 1). The genotype × treatment interaction was non-significant in both growing cycles, indicating that differences among genotypes in their responses to salinity treatments were not statistically significant for pod number. Tukey’s HSD test showed no significant pairwise differences between salinity treatments and the control for the number of pods per plant (Table 2). Noticeable variation in mean pod number was observed among genotypes under different salinity treatments (Table 3).

In GC1, genotype Virtus produced the highest mean number of pods per plant under T2 (3.83 pods plant⁻¹), whereas Vilo recorded the lowest value under T3 (1.17 pods plant⁻¹). Favorit and Mali Provansalac maintained relatively stable pod numbers across treatments, with slight increases under moderate salinity (T1 and T2), suggesting that these salinity levels had little influence on pod production in these genotypes. The highest variability was observed for genotypes Virtus under T1 and T2 (SD = 2.34 and 2.56, respectively), whereas genotypes Korvin and Mali Provansalac exhibited comparatively lower variability across treatments. In GC2, Favorit achieved the highest pod number under T2 (2.50 pods plant⁻¹), while Korvin reached its maximum value under T1 (2.33 pods plant⁻¹). The lowest mean value was recorded for genotype Vilo under the control treatment (0.83 pods plant⁻¹). Compared with GC1, pod production was generally lower during GC2, although the relative ranking of genotypes remained largely unchanged. Standard deviation values were also generally lower in GC2, indicating greater consistency among replications.

The greatest variability was observed for genotype Vilo under T3 (SD = 1.52). Overall, lower and intermediate salinity levels (100 and 150 mM NaCl) had small effect on pod production, while reductions became more evident under the highest salinity levels (200 mM NaCl), particularly in genotype Vilo. These findings indicated that pod number was primarily determined by genotype rather than salinity treatment.

3.2. Pod Length

ANOVA indicated that pod length was significantly affected by salinity treatment and genotype during GC1, whereas genotype remained the dominant source of variation during GC2 (Table 1). In GC1, both salinity treatment (p < 0.01) and genotype (p < 0.01) significantly affected pod length. A significant genotype × treatment interaction (p < 0.05) was also observed, indicating that genotypes differed in their responses to increasing salinity levels. In GC2, genotype effects were highly significant (p < 0.01), while salinity treatment and genotype × treatment interaction did not have a significant effect on pod length. According to Tukey’s HSD test, significant pairwise differences were detected only between T3 and the control treatment, whereas remaining treatment comparisons were not statistically significant (Table 2). In GC1, Favorit exhibited the longest pods under control conditions (7.63 cm), followed by a progressive reduction with increasing salinity, reaching 4.75 cm under T3. A similar decline was recorded in Vilo, whose pod length fell from 6.08 cm (control) to 3.50 cm under severe salt stress.

Table 4. Mean pod length (cm) ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	7.63 ± 0.54	4.96 ± 0.84	5.00 ± 0.84	4.58 ± 0.66	6.08 ± 1.31
T1	6.54 ± 0.84	3.33 ± 2.63	4.67 ± 0.61	4.58 ± 0.49	6.29 ± 0.64
T2	6.25 ± 0.65	3.71 ± 1.97	4.63 ± 0.85	4.83 ± 0.52	4.33 ± 2.21
T3	4.75 ± 2.49	4.79 ± 0.43	4.42 ± 1.02	4.58 ± 0.66	3.50 ± 2.81
GC2
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	7.12 ± 0.89	4.92 ± 0.47	5.38 ± 0.80	4.82 ± 0.43	5.75 ± 2.88
T1	5.75 ± 0.92	4.79 ± 0.51	4.96 ± 0.77	4.25 ± 2.12	5.58 ± 2.89
T2	5.92 ± 1.11	4.65 ± 0.90	5.54 ± 0.40	4.88 ± 0.59	6.88 ± 0.63
T3	5.96 ± 0.70	4.99 ± 0.54	5.08 ± 0.58	5.04 ± 0.64	4.29 ± 3.36

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

Table 4. Mean pod length (cm) ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	7.63 ± 0.54	4.96 ± 0.84	5.00 ± 0.84	4.58 ± 0.66	6.08 ± 1.31
T1	6.54 ± 0.84	3.33 ± 2.63	4.67 ± 0.61	4.58 ± 0.49	6.29 ± 0.64
T2	6.25 ± 0.65	3.71 ± 1.97	4.63 ± 0.85	4.83 ± 0.52	4.33 ± 2.21
T3	4.75 ± 2.49	4.79 ± 0.43	4.42 ± 1.02	4.58 ± 0.66	3.50 ± 2.81
GC2
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	7.12 ± 0.89	4.92 ± 0.47	5.38 ± 0.80	4.82 ± 0.43	5.75 ± 2.88
T1	5.75 ± 0.92	4.79 ± 0.51	4.96 ± 0.77	4.25 ± 2.12	5.58 ± 2.89
T2	5.92 ± 1.11	4.65 ± 0.90	5.54 ± 0.40	4.88 ± 0.59	6.88 ± 0.63
T3	5.96 ± 0.70	4.99 ± 0.54	5.08 ± 0.58	5.04 ± 0.64	4.29 ± 3.36

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

In contrast, Mali Provansalac maintained relatively stable pod length across all salinity treatments, indicating lower sensitivity of this trait to salt stress. Although Virtus produced shorter pods under moderate salinity (T1 and T2), pod length partially recovered under T3, suggesting a genotype-specific response. Standard deviation values indicated greater variability in Favorit and Vilo under T3 (SD = 2.49 and 2.81, respectively), whereas Korvin and Mali Provansalac showed lower variability across treatments, reflecting more stable pod length expression. In GC2, pod length was generally more stable across salinity treatments than in GC1. Vilo achieved the highest pod length under T2 (6.88 cm), whereas Korvin and Mali Provansalac maintained relatively consistent values regardless of treatment. Standard deviation values were generally lower than values observed in GC1, indicating greater uniformity among replications. Overall, severe salinity (200 mM NaCl) reduced pod length primarily in genotypes Favorit and Vilo, whereas moderate salinity caused only minor changes in most genotypes. These findings suggest that pod elongation was mainly influenced by genotype than by salinity intensity, particularly during the second growing cycle.

3.3. Number of Seeds per Plant

Related to the trait the Number of seeds per plant, ANOVA showed that the number of seeds per plant was significantly influenced by genotype in both growing cycles, whereas salinity treatment and genotype × treatment interaction were not significant (Table 1). Genotype effects were highly significant (p < 0.01) in both GC1 and GC2, indicating significant genetic variation in seed production among the evaluated pea genotypes. In contrast, salinity treatment did not significantly affect the number of seeds per plant under the experimental conditions. Tukey’s HSD test confirmed the absence of significant pairwise differences among salinity treatments and the control (Table 2). In GC1, Virtus produced the highest number of seeds per plant under T2 (10.50), representing a marked increase compared with the control treatment (3.50). Mali Provansalac also exhibited slightly higher seed numbers under moderate salinity (T1 and T2) than under control conditions, whereas Vilo showed a clear reduction under severe salinity (T3). Favorit maintained consistently high seed numbers across all treatments, indicating stable reproductive performance despite increasing salinity. Although variation among genotypes was evident, the overall response of seed number to salinity remained relatively limited, supporting the ANOVA results that identified genotype as the primary source of variation. In GC2, Favorit recorded the highest number of seeds per plant under T2 (11.33), while Vilo also showed increased seed production under T2 and T3 compared with the control treatment. In contrast, Virtus exhibited significantly lower seed numbers in GC2 than in GC1, suggesting considerable variation between growing cycles. Overall, seed number was primarily determined by genotype more than salinity treatment. Moderate salinity occasionally promoted seed production in individual genotypes, whereas severe salinity generally resulted in genotype-dependent reductions without producing statistically significant overall treatment effects.

Table 5. Mean number of seeds per plant ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	7.67 ± 2.07	3.50 ± 3.27	2.83 ± 0.75	2.33 ± 0.82	6.33 ± 3.83
T1	7.17 ± 2.71	8.50 ± 7.94	5.17 ± 1.17	3.83 ± 1.17	6.50 ± 2.43
T2	7.67 ± 3.01	10.50 ± 5.89	3.83 ± 1.33	4.67 ± 2.42	5.83 ± 5.71
T3	7.67 ± 4.93	5.83 ± 3.13	3.50 ± 0.84	3.67 ± 1.03	2.67 ± 3.08
GC2
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	9.00 ± 4.82	4.67 ± 1.75	4.50 ± 1.87	1.33 ± 0.52	4.50 ± 4.55
T1	8.50 ± 3.89	5.00 ± 2.00	4.50 ± 2.43	3.50 ± 1.87	4.00 ± 3.16
T2	11.33 ± 5.20	1.67 ± 1.97	5.33 ± 1.63	4.17 ± 2.93	10.67 ± 7.06
T3	8.67 ± 1.97	1.67 ± 1.86	4.83 ± 2.14	4.50 ± 1.64	8.50 ± 9.33

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

Table 5. Mean number of seeds per plant ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	7.67 ± 2.07	3.50 ± 3.27	2.83 ± 0.75	2.33 ± 0.82	6.33 ± 3.83
T1	7.17 ± 2.71	8.50 ± 7.94	5.17 ± 1.17	3.83 ± 1.17	6.50 ± 2.43
T2	7.67 ± 3.01	10.50 ± 5.89	3.83 ± 1.33	4.67 ± 2.42	5.83 ± 5.71
T3	7.67 ± 4.93	5.83 ± 3.13	3.50 ± 0.84	3.67 ± 1.03	2.67 ± 3.08
GC2
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	9.00 ± 4.82	4.67 ± 1.75	4.50 ± 1.87	1.33 ± 0.52	4.50 ± 4.55
T1	8.50 ± 3.89	5.00 ± 2.00	4.50 ± 2.43	3.50 ± 1.87	4.00 ± 3.16
T2	11.33 ± 5.20	1.67 ± 1.97	5.33 ± 1.63	4.17 ± 2.93	10.67 ± 7.06
T3	8.67 ± 1.97	1.67 ± 1.86	4.83 ± 2.14	4.50 ± 1.64	8.50 ± 9.33

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

3.4. Seed Weight per Plant

ANOVA revealed significant effects of both salinity treatment and genotype on seed weight per plant during GC1, whereas genotype remained the dominant source of variation during GC2 (Table 1). In GC1, both salinity treatment (p < 0.01) and genotype (p < 0.01) significantly affected seed weight per plant, whereas the genotype × treatment interaction was not significant. In GC2, only genotype had a significant effect (p < 0.01), indicating that differences in seed weight were primarily determined by genetic background rather than salinity treatment during the second growing cycle. According to Tukey’s HSD test, a significant pairwise difference was detected only between T1 and T2 treatments, whereas all remaining treatment comparisons were not statistically significant (Table 2). In GC1, genotype Favorit consistently produced the highest seed weight across all salinity treatments, ranging from 2.33 to 2.70 g per plant, indicating high reproductive stability under salt stress. Pea genotype Korvin achieved its maximum seed weight under T1 (2.63 g), whereas Vilo exhibited a progressive decline from 1.22 g under control conditions to only 0.32 g under T3, reflecting a pronounced sensitivity to severe salinity. Virtus also reached its highest seed weight under T2 (2.63 g), approximately doubling the value observed under the control treatment (1.30 g), while Mali Provansalac displayed a similar but less pronounced response. These results indicate that moderate salinity occasionally stimulated seed production in specific genotypes, whereas severe salinity generally reduced seed weight, particularly in salinity-sensitive genotypes. In GC2, Favorit again recorded the highest seed weight under both the control and T2 treatments (4.22 and 4.29 g, respectively), confirming its superior productivity across growing cycles. Unlike GC1, Vilo exhibited substantially higher seed weight under T2 and T3, suggesting improved reproductive performance during the second growing cycle. In contrast, Virtus produced considerably lower seed weights than in GC1, particularly under T2 and T3 treatments. Overall, seed weight per plant was influenced by both genotype and salinity during GC1, but genotype remained the predominant factor governing variation across both growing cycles. These findings demonstrate that the reproductive response to salinity was strongly genotype dependent, with Favorit showing the greatest stability under increasing salt concentrations. Among all evaluated yield components, seed weight per plant exhibited the clearest genotype response to salinity, highlighting Favorit as the most stable genotype across increasing NaCl concentrations, whereas Vilo and Virtus displayed contrasting adaptive responses between growing cycles.

Table 6. Mean seed weight per plant (g) ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	2.55 ± 0.32	1.30 ± 0.67	1.61 ± 0.42	1.23 ± 0.42	1.22 ± 1.04
T1	2.70 ± 1.23	1.74 ± 1.68	2.63 ± 0.74	1.83 ± 0.80	1.20 ± 0.47
T2	2.41 ± 1.40	2.63 ± 1.50	1.13 ± 0.73	1.42 ± 0.83	0.87 ± 1.23
T3	2.33 ± 1.76	1.50 ± 1.36	0.88 ± 0.52	0.99 ± 0.45	0.32 ± 0.37
GC2
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	4.22 ± 2.70	1.07 ± 0.62	1.94 ± 0.74	0.57 ± 0.39	1.22 ± 1.01
T1	2.83 ± 1.40	1.31 ± 0.86	1.95 ± 1.00	1.34 ± 1.07	1.11 ± 0.89
T2	4.29 ± 3.34	0.50 ± 0.61	1.97 ± 0.55	0.62 ± 0.59	2.79 ± 2.07
T3	2.64 ± 0.39	0.55 ± 0.63	1.41 ± 0.65	0.94 ± 0.94	2.52 ± 2.54

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

Table 6. Mean seed weight per plant (g) ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	2.55 ± 0.32	1.30 ± 0.67	1.61 ± 0.42	1.23 ± 0.42	1.22 ± 1.04
T1	2.70 ± 1.23	1.74 ± 1.68	2.63 ± 0.74	1.83 ± 0.80	1.20 ± 0.47
T2	2.41 ± 1.40	2.63 ± 1.50	1.13 ± 0.73	1.42 ± 0.83	0.87 ± 1.23
T3	2.33 ± 1.76	1.50 ± 1.36	0.88 ± 0.52	0.99 ± 0.45	0.32 ± 0.37
GC2
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	4.22 ± 2.70	1.07 ± 0.62	1.94 ± 0.74	0.57 ± 0.39	1.22 ± 1.01
T1	2.83 ± 1.40	1.31 ± 0.86	1.95 ± 1.00	1.34 ± 1.07	1.11 ± 0.89
T2	4.29 ± 3.34	0.50 ± 0.61	1.97 ± 0.55	0.62 ± 0.59	2.79 ± 2.07
T3	2.64 ± 0.39	0.55 ± 0.63	1.41 ± 0.65	0.94 ± 0.94	2.52 ± 2.54

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

3.5. Fresh Biomass

ANOVA revealed significant effects of salinity treatment and genotype on fresh biomass during GC1, whereas genotype remained the dominant source of variation during GC2 (Table 1). In GC1, both salinity treatment (p < 0.01) and genotype (p < 0.01) significantly affected fresh biomass, whereas the genotype × treatment interaction was not significant. In GC2, genotype remained the only significant source of variation (p < 0.01), while neither salinity treatment nor the genotype × treatment interaction significantly influenced fresh biomass. According to Tukey’s HSD test, a significant pairwise difference was detected only between T2 and the control treatment, whereas all remaining treatment comparisons were not statistically significant (Table 2). Fresh biomass differed markedly among genotypes, indicating substantial variation in vegetative growth under saline conditions. In GC1, Favorit consistently maintained high fresh biomass across all treatments, reaching its maximum value under T2 (12.46 g). Vilo also exhibited relatively high fresh biomass throughout the experiment, with the highest value recorded under T1 (13.06 g). In contrast, Korvin and Mali Provansalac produced considerably lower fresh biomass across all treatments. Virtus maintained relatively high biomass under moderate salinity (T1 and T2), but biomass decreased sharply under T3, indicating greater sensitivity to severe salt stress. These results suggest that moderate salinity had little detrimental effect on vegetative growth in the more tolerant genotypes, whereas severe salinity substantially reduced biomass accumulation in salt-sensitive genotypes. In GC2, genotype-dependent differences remained evident. Favorit exhibited the highest fresh biomass under T3 (14.20 g), closely followed by T2 (13.61 g), confirming its high productivity under saline conditions. Vilo also maintained high biomass across all salinity treatments, whereas Mali Provansalac showed improved biomass accumulation under T2 compared with the remaining treatments. Korvin displayed relatively stable fresh biomass across treatments, while Virtus maintained intermediate values throughout GC2. Compared with GC1, biomass accumulation in GC2 was generally more stable across salinity treatments, indicating greater consistency of vegetative growth during the second growing cycle. Overall, fresh biomass was influenced by both genotype and salinity during GC1; however, genotype represented the principal source of variation across both growing cycles. The consistently high biomass production observed in Favorit and Vilo demonstrates superior vegetative performance under saline conditions, whereas Virtus exhibited greater sensitivity to severe salinity. The maintenance of high fresh biomass in Favorit under moderate and severe salinity corresponded with its stable reproductive performance, suggesting superior tolerance to salt stress, whereas the marked biomass reduction observed in Virtus under severe salinity reflected greater susceptibility of vegetative growth to increasing NaCl concentrations.

Table 7. Mean fresh biomass (g) ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	10.88 ± 4.08	9.67 ± 7.86	4.41 ± 1.47	3.64 ± 2.19	10.20 ± 5.44
T1	8.82 ± 2.48	8.98 ± 9.15	4.24 ± 1.22	4.50 ± 2.16	13.06 ± 4.20
T2	12.46 ± 2.88	8.08 ± 5.26	1.83 ± 1.05	3.85 ± 2.18	11.91 ± 4.56
T3	9.25 ± 4.01	2.72 ± 2.20	1.98 ± 0.76	2.56 ± 1.23	8.14 ± 3.21
GC2
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	10.29 ± 2.87	4.86 ± 3.21	3.58 ± 1.02	2.13 ± 0.45	5.26 ± 3.61
T1	10.37 ± 5.12	5.55 ± 4.84	4.51 ± 0.77	4.86 ± 2.84	13.76 ± 1.40
T2	13.61 ± 4.66	7.47 ± 5.64	4.41 ± 2.09	7.57 ± 3.27	11.69 ± 7.84
T3	14.20 ± 3.52	6.65 ± 5.78	3.79 ± 2.81	3.81 ± 2.13	12.59 ± 7.95

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

Table 7. Mean fresh biomass (g) ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	10.88 ± 4.08	9.67 ± 7.86	4.41 ± 1.47	3.64 ± 2.19	10.20 ± 5.44
T1	8.82 ± 2.48	8.98 ± 9.15	4.24 ± 1.22	4.50 ± 2.16	13.06 ± 4.20
T2	12.46 ± 2.88	8.08 ± 5.26	1.83 ± 1.05	3.85 ± 2.18	11.91 ± 4.56
T3	9.25 ± 4.01	2.72 ± 2.20	1.98 ± 0.76	2.56 ± 1.23	8.14 ± 3.21
GC2
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	10.29 ± 2.87	4.86 ± 3.21	3.58 ± 1.02	2.13 ± 0.45	5.26 ± 3.61
T1	10.37 ± 5.12	5.55 ± 4.84	4.51 ± 0.77	4.86 ± 2.84	13.76 ± 1.40
T2	13.61 ± 4.66	7.47 ± 5.64	4.41 ± 2.09	7.57 ± 3.27	11.69 ± 7.84
T3	14.20 ± 3.52	6.65 ± 5.78	3.79 ± 2.81	3.81 ± 2.13	12.59 ± 7.95

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

3.6. Dry Biomass

ANOVA demonstrated significant effects of salinity treatment and genotype on dry biomass during GC1, whereas genotype remained the dominant factor influencing dry biomass accumulation during GC2 (Table 1). In GC1, both salinity treatment (p < 0.01) and genotype (p < 0.01) significantly affected dry biomass, whereas the genotype × treatment interaction was not significant. In GC2, only genotype remained a significant source of variation (p < 0.01), while neither salinity treatment nor the genotype × treatment interaction significantly affected dry biomass accumulation. Tukey’s HSD test revealed no significant pairwise differences among salinity treatments, indicating that differences in dry biomass were primarily associated with genotype rather than salinity treatment (Table 2). Dry biomass significantly differed among the evaluated genotypes. In GC1, Favorit and Vilo consistently produced the highest dry biomass values across salinity treatments, reaching maximum values of 4.24 g under T2 and 4.95 g under T1, respectively. Virtus also maintained relatively high dry biomass under moderate salinity (T1 and T2), whereas a pronounced reduction was observed under T3 (1.71 g).

Table 8. Mean dry biomass (g) ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	4.05 ± 0.39	2.56 ± 1.61	1.03 ± 0.37	1.31 ± 0.90	3.01 ± 0.88
T1	3.37 ± 0.82	2.92 ± 0.51	2.21 ± 0.43	1.48 ± 0.16	4.95 ± 1.19
T2	4.24 ± 0.31	3.33 ± 1.83	1.08 ± 0.40	1.82 ± 0.50	3.91 ± 1.07
T3	3.31 ± 0.54	1.71 ± 0.65	1.10 ± 0.24	1.38 ± 0.20	2.57 ± 0.37
GC2
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	3.34 ± 0.22	1.57 ± 0.68	1.11 ± 0.44	0.74 ± 0.24	2.39 ± 0.80
T1	2.75 ± 0.15	2.21 ± 1.20	1.61 ± 0.11	1.90 ± 0.56	4.71 ± 0.78
T2	3.64 ± 0.96	3.04 ± 2.05	1.61 ± 0.53	2.35 ± 0.84	4.15 ± 1.92
T3	3.77 ± 0.65	2.61 ± 1.09	1.81 ± 0.93	1.46 ± 0.64	4.06 ± 1.29

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

Table 8. Mean dry biomass (g) ± standard deviation (SD) of five pea genotypes under different salinity treatments during the first (GC1) and second (GC2) growing cycles.

GC1
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	4.05 ± 0.39	2.56 ± 1.61	1.03 ± 0.37	1.31 ± 0.90	3.01 ± 0.88
T1	3.37 ± 0.82	2.92 ± 0.51	2.21 ± 0.43	1.48 ± 0.16	4.95 ± 1.19
T2	4.24 ± 0.31	3.33 ± 1.83	1.08 ± 0.40	1.82 ± 0.50	3.91 ± 1.07
T3	3.31 ± 0.54	1.71 ± 0.65	1.10 ± 0.24	1.38 ± 0.20	2.57 ± 0.37
GC2
	Genotypes
Treatment	Favorit	Virtus	Korvin	Mali Provansalac	Vilo
	X ± SD	X ± SD	X ± SD	X ± SD	X ± SD
C	3.34 ± 0.22	1.57 ± 0.68	1.11 ± 0.44	0.74 ± 0.24	2.39 ± 0.80
T1	2.75 ± 0.15	2.21 ± 1.20	1.61 ± 0.11	1.90 ± 0.56	4.71 ± 0.78
T2	3.64 ± 0.96	3.04 ± 2.05	1.61 ± 0.53	2.35 ± 0.84	4.15 ± 1.92
T3	3.77 ± 0.65	2.61 ± 1.09	1.81 ± 0.93	1.46 ± 0.64	4.06 ± 1.29

* GC1: The first growing cycles, GC2: The second growing cycles; C: control (0 mM NaCl); T1: 100 mM NaCl; T2: 150 mM NaCl; T3: 200 mM NaCl. x: Mean values; SD: standard deviation across treatment, presented as mean ± SD.

In contrast, Korvin and Mali Provansalac produced comparatively lower dry biomass throughout the experiment. These results indicate that moderate salinity had only a limited effect on dry matter accumulation in the more tolerant genotypes, whereas severe salinity reduced biomass production in salt-sensitive genotypes. In GC2, genotype-dependent differences remained evident. Favorit maintained high dry biomass across all salinity treatments, with the highest value recorded under T3 (3.77 g). Vilo also sustained high dry biomass under all treatments, whereas Mali Provansalac exhibited increased dry biomass under T2 compared with the remaining treatments. Korvin showed relatively stable dry biomass across treatments, while Virtus maintained intermediate values throughout the second growing cycle. Compared with GC1, dry biomass responses were generally more consistent among salinity treatments during GC2, indicating greater stability of biomass accumulation under the environmental conditions of the second growing cycle. Overall, dry biomass accumulation was primarily determined by genotype across both growing cycles. Although salinity significantly affected dry biomass during GC1, genotype exerted the strongest and most consistent influence on biomass production, with Favorit and Vilo exhibiting superior vegetative performance under saline conditions. Across both growing cycles, substantial genotype-dependent variation in salinity tolerance was observed. Favorit consistently maintained superior reproductive performance and biomass accumulation under both moderate and severe salinity, whereas Vilo exhibited contrasting responses depending on the evaluated trait. Although Vilo showed pronounced reductions in pod length and seed weight under 200 mM NaCl during GC1, it maintained relatively high fresh and dry biomass, particularly during GC2, indicating that vegetative growth was less affected than reproductive development. In contrast, Virtus showed the greatest reduction in fresh biomass under severe salinity, reflecting a higher sensitivity of vegetative growth to salt stress. The magnitude of the salinity response differed considerably among genotypes. Under 200 mM NaCl, reductions relative to the control reached 42.4% for pod length, 73.8% for seed weight per plant, and 71.9% for fresh biomass. The largest reductions in pod length and seed weight per plant were observed in Vilo, whereas the greatest reduction in fresh biomass occurred in Virtus, confirming substantial genetic variation in tolerance to salinity stress among the evaluated pea genotypes.

3.7. Relative Changes of Yield-Related Traits Under Salinity Stress

The analysis of relative changes highlighted pronounced genotype dependent responses to salinity stress, with both reductions and increases in the evaluated traits depending on salinity intensity and growing cycle (Table 9). Overall, the largest reductions were generally associated with the highest salinity level (200 mM NaCl), whereas the greatest positive responses were predominantly observed under lower and intermediate salinity levels (100 or 150 mM NaCl). During the first growing cycle (GC1), genotype Vilo exhibited the greatest reductions in most reproductive traits, including pod number (36.1%), pod length (42.4%), seed number per plant (57.8%), and seed weight per plant (73.8%), all under T3. By contrast, the greatest reductions in vegetative traits were recorded in genotype Virtus, with fresh biomass and dry biomass decreasing by 71.9% and 33.2%, respectively, under T3. These findings indicate that reproductive performance was most severely affected in Vilo, whereas vegetative growth was more sensitive in Virtus (Table 9). Despite the detrimental effects of severe salinity, several genotypes exhibited marked improvements under moderate salinity. In GC1, the greatest increases were recorded in Korvin for pod number (63.2%) and dry biomass (114.6%) under T1, Mali Provansalac for pod length (5.5%) under T2, Virtus for seed number per plant (200.0%) and seed weight per plant (102.3%) under T2, and Vilo for fresh biomass (28.0%) under T1 treatment. A contrasting response pattern was observed during the second growing cycle (GC2). Genotype Virtus exhibited the highest reductions in pod number (20.4%), seed number per plant (64.2%) and seed weight per plant (53.3%), all under T2, whereas Vilo showed the greatest reduction in pod length (25.4%) under T3. For dry biomass, the only notable reduction was observed in Favorit (17.7%) under T1, whereas no reduction in fresh biomass was recorded in any genotype during GC2. The most pronounced positive responses during GC2 were associated with genotype Mali Provansalac, which exhibited increases of 83.0% in pod number, 238.3% in seed number per plant, 135.1% in seed weight per plant, 255.4% in fresh biomass, and 217.6% in dry biomass. The largest increase in pod length (19.7%) was recorded in Vilo under T2 (Table 9). Generally, these results demonstrate significant genotypic variation in salinity tolerance under controlled-environment conditions. Lower and intermediate salinity levels frequently stimulated vegetative growth and reproductive traits in specific genotypes, whereas the highest salinity concentrations (200 mM NaCl) predominantly reduced reproductive yield traits, with the magnitude of the response varying considerably among genotypes. This response may indicate a possible hormetic effect, characterized by increased values of several growth and yield-related traits under lower and intermediate NaCl concentrations, followed by reductions at the highest NaCl concentration.

The overall variation in trait responses across growing cycles and salinity treatments, indicates the comparatively high overall stability of Favorit, the greater reproductive sensitivity of Vilo to the highest NaCl concentration and the more pronounced reduction in vegetative growth observed in Virtus (Table 9). At the levels from 100 and 150 mM NaCl, several genotypes maintained or increased specific traits compared with their respective controls, indicating a possible stimulatory response to lower salinity levels, although this effect was not statistically confirmed by the genotype × treatment interaction. However, these, observed changes should be interpreted as descriptive responses, as the statistical evidence was insufficient to confirm the occurrence of hormesis.

4. Discussion

The present study demonstrated that genotype was the primary determinant of variation in most yield-related traits under salinity treatments. Significant genotype effects were consistently observed across both growing cycles, whereas the influence of salinity treatment depended on the evaluated trait and was generally less pronounced. These findings indicate significant genetic variability in salinity tolerance among the evaluated pea genotypes and highlight the importance of genetic background in determining plant performance under saline conditions [6,24]. Comparable observations were reported by Munns and Tester [6], who emphasized that salinity tolerance is governed by multiple physiological processes at different developmental stages, resulting in considerable variation among genotypes. Similarly, Flowers and Colmer [29] concluded that salinity tolerance is a complex quantitative trait involving coordinated mechanisms of ion exclusion, osmotic adjustment and tissue tolerance rather than a single adaptive response. The predominance of genotype over salinity treatment observed in the present study is consistent with previous investigations in legumes, where substantial genotypic variability in salinity tolerance has been reported. Hernández et al. [40] similarly reported pronounced genotype dependent differences in antioxidant and water-relation responses to NaCl stress in pea, while Taïbi et al. [41] reported comparable genotypic variability in salt-stress responses in the related legume common bean (Phaseolus vulgaris L.), together confirming that genotype selection represents one of the most effective strategies for improving salinity tolerance in legume crops.

The predominantly non-significant genotype × treatment interactions indicate a relatively consistent expression of genotypic differences across salinity treatments, suggesting that the controlled vertical-farming conditions reduced environmental variation and enabled clearer discrimination among genotypes. Similar observations have been reported in previous studies indicating that genetic factors frequently explain a larger proportion of phenotypic variation under salinity stress than environmental factors alone [6,30]. The stability of genotype performance observed in the present study may also reflect the highly controlled environmental conditions provided by the vertical farming system, which minimized uncontrolled variation and enabled clearer identification of genotype-specific responses. Unlike field experiments, controlled-environment conditions significantly reduce spatial and seasonal variability, thereby increasing the accuracy and repeatability of genotype evaluation under abiotic stress conditions [7,8].

Among the evaluated reproductive traits, pod number, pod length, seed number and seed weight showed different levels of sensitivity to salinity. Pod number and seed number were relatively stable across treatments, indicating that reproductive initiation and seed set were less affected by the applied salinity levels. In contrast, pod length and seed weight exhibited greater sensitivity, particularly under the highest salinity treatment (200 mM NaCl). Similar reductions in reproductive traits under severe salinity stress have been reported previously in pea and other legumes [16,21,22]. Several authors have suggested that reproductive organs are particularly vulnerable to salinity because reproductive development depends on continuous assimilate supply and hormonal regulation. Consequently, reductions in seed weight often reflect limitations during seed filling rather than during flower initiation or fertilization [16,21,22]. This observation is consistent with the present study, where seed weight and pod length were more sensitive to severe salinity than pod number or seed number.

Notably, several genotypes maintained or even improved reproductive performance under lower and intermediate salinity treatments (100 and 150 mM NaCl). The relative change analysis further supported these observations, showing that several genotypes exhibited marked increases in reproductive traits under moderate salinity, whereas the greatest reductions were consistently recorded under the highest NaCl concentration (200 mM NaCl). Such responses may indicate a hormetic response, whereby low to lower and intermediate stress levels stimulate adaptive physiological and metabolic processes. Similar stimulatory responses under moderate stress have been described previously and are commonly associated with improved osmotic adjustment, enhanced antioxidant activity, and more efficient assimilation partitioning [22,25]. Hormetic responses have been widely documented in plants exposed to mild abiotic stress, where moderate stress activates defence pathways and enhances physiological performance before inhibitory effects become dominant under more severe stress conditions.

The maintenance of reproductive traits under lower and intermediate salinity concentrations (100 and 150 mM) may also indicate that pea plants were able to activate adaptive mechanisms that preserved reproductive functionality despite osmotic stress. Osmotic adjustment, ion compartmentalization, maintenance of cellular turgor, and activation of antioxidant defence systems are among the most important mechanisms contributing to salinity tolerance in legumes [6,29,30]. Maintenance of reproductive performance may additionally reflect efficient source–sink relationships, enabling continued transport of assimilates to developing pods despite osmotic stress. Such physiological regulation has previously been associated with improved carbon allocation and reproductive resilience under salinity [29]. The ability of certain genotypes to sustain pod formation and seed production under moderate salinity conditions suggests the presence of efficient physiological regulation that minimize stress-related reductions in reproductive performance.

Biomass-related traits were generally more responsive to salinity than reproductive traits. Fresh and dry biomass showed significant treatment effects, particularly during the first growing cycle. Reductions in biomass accumulation under elevated salinity levels are consistent with the well-documented effects of osmotic stress on water uptake, cell expansion, photosynthetic activity, and assimilate production [6,12,17,18]. Salinity-induced growth inhibition has been associated with reduced photosynthetic efficiency, membrane instability, ionic imbalance, and impaired nutrient uptake [29,34].

Nevertheless, biomass reductions were not uniform across genotypes. Certain genotypes maintained relatively stable biomass accumulation even under elevated salinity levels, indicating superior adaptive capacity. Such stability may reflect more efficient osmotic regulation, maintenance of photosynthetic performance, and improved ion homeostasis. Similar genotype-dependent responses have been reported in pea during salinity screening studies [11]. The consistently high fresh and dry biomass observed in Favorit, and to a lesser extent in Vilo, suggests that these genotypes were better able to maintain vegetative growth under saline conditions, possibly through improved Na⁺ exclusion, maintenance of K⁺/Na⁺ homeostasis, and enhanced water-use efficiency. The observed differences among genotypes further confirm that salinity tolerance is a complex quantitative trait governed by multiple physiological and genetic mechanisms.

Differences between the two growing cycles were also evident. In general, variability among replicates was lower during GC2, and several traits exhibited more stable responses across treatments compared with GC1. Because independent plant populations were evaluated in each growing cycle, these differences should not necessarily be interpreted as physiological acclimation. Rather, they may reflect improved experimental uniformity under controlled-environment conditions and differences in plant developmental dynamics between cultivation cycles. Nevertheless, previous studies have shown that prolonged exposure to moderate salinity can induce physiological adjustments, including enhanced antioxidant capacity and osmotic regulation, which improve stress tolerance [28,29].

An important aspect of this study is the use of a vertical farming system as an experimental platform for salinity research. Cultivation in controlled environment offers several advantages compared with field experiments, including precise regulation of temperature, humidity, light intensity, and irrigation management [7,8]. Such conditions reduce environmental variability and improve experimental repeatability, enabling more accurate assessment of genotype performance under stress conditions. These characteristics make vertical farming systems particularly valuable for pre-breeding programmes, physiological studies, and high-throughput phenotyping aimed at identifying salt-tolerant germplasm. The results obtained in this study therefore support the suitability of vertical farming systems for salinity tolerance screening and physiological investigations.

Overall, the findings indicate that lower and intermediate salinity levels (100–150 mM NaCl) can be tolerated by certain pea genotypes without significant reductions in yield-related traits, whereas severe salinity stress (200 mM NaCl) results in more pronounced reductions in growth and yield-related traits. The relative change analysis further demonstrated that Vilo and Virtus exhibited the greatest reductions under severe salinity, whereas Favorit, Korvin, and Mali Provansalac maintained or improved several reproductive and biomass-related traits under moderate salinity, indicating greater adaptive capacity. The stimulation of several traits under moderate salinity further supports the occurrence of a hormetic response, whereas the significant reductions observed at 200 mM NaCl indicate that this concentration exceeded the physiological tolerance threshold for most genotypes. The considerable variability observed among genotypes confirms the presence of valuable genetic resources for breeding programmes aimed at improving salinity tolerance.

The observed tolerance under moderate salinity could be associated with physiological mechanisms such as osmotic adjustment, ion homeostasis, antioxidant protection and accumulation of compatible solutes, which are frequently reported in salt-tolerant plant species [26,27,35]. These adaptive mechanisms are in accordance with previous reports describing the physiological basis of salt-stress adaptation [43]. These findings also indicate a possible relationship between salinity tolerance and yield potential under non-saline conditions, as genotypes with greater tolerance to salinity may not necessarily exhibit the highest yield potential under optimal conditions [42]. In the present study, Favorit maintained generally high performance across salinity treatments, although it did not consistently exhibit the highest values for all yield-related traits.

5. Conclusions

The present study revealed significant genotypic variation in salinity tolerance among the evaluated pea genotypes grown under controlled vertical farming conditions. In this study genotype was identified as the primary source of variation for the evaluated yield-related and biomass traits, while the effects of salinity depended on trait and stress intensity. The lower and intermediate NaCl concentrations and salinity levels (100 and 150 mM) generally resulted in low changes in yield reproductive traits and, in several genotype × treatment combinations, were associated with the maintenance or increase of trait values relative to the corresponding controls. In contrast, the highest NaCl concentration and severe salinity stress (200 mM NaCl) resulted in reductions in pod length, seed weight per plant and biomass traits, indicating that this concentration exceeded the tolerance capacity of several genotypes. Salinity levels from 200 mM NaCl caused reductions of up to 42.4% in pod length, 73.8% in seed weight per plant, and 71.9% in fresh biomass, depending on the genotype. The obtained results indicated that yield reproductive traits were generally more stable than biomass related traits under salinity stress, highlighting the ability of certain pea genotypes to maintain reproductive performance under lower and intermediate stress conditions. The relative change analysis confirmed the contrasting responses of the evaluated genotypes to salinity stress. While genotypes Vilo and Virtus exhibited the greatest reductions under the highest soil salinity levels, genotypes Favorit, Korvin and Mali Provansalac maintained or improved several reproductive and biomass-related traits under the lower and intermediate salinity levels, indicating greater adaptive capacity and their potential value as genetic resources for breeding salt-tolerant pea cultivars. The observed differences among genotypes confirm the existence of valuable genetic variability that can be exploited in future breeding programs aimed at improving salinity tolerance. The study also demonstrated that the controlled environment vertical farming conditions provide accurate and precise evaluation of genotype responses to salinity, making it particularly suitable for rapid evaluation and the identification of promising germplasm for breeding programs targeting improved salt tolerance, as well as for wide production. Although the present study was conducted under controlled conditions, the observed genotypic differences provide a valuable basis for subsequent validation under greenhouse and field conditions, where genotype × environment interactions may further influence salinity responses.

Furthermore, validation of the identified tolerant genotypes under controlled-environment vertical farming conditions, greenhouse conditions and field conditions will be essential to confirm their agronomic performance and facilitate their utilization in breeding programs and sustainable agricultural production under saline environments.