Effects of Salt Tolerance Training on Multidimensional Root Distribution and Root-Shoot Characteristics of Summer Maize under Brackish Water Irrigation

To investigate the impact of brackish water irrigation on the multidimensional root distribution and root-shoot characteristics of summer maize under different salt-tolerance-training modes, a micro-plot experiment was conducted from June to October in 2022 at the experimental station in Hohai University, China. Freshwater irrigation was used as the control (CK), and different concentrations of brackish water (S0: 0.08 g·L−1, S1: 2.0 g·L−1, S2: 4.0 g·L−1, S3: 6.0 g·L−1) were irrigated at six-leaf stage, ten-leaf stage, and tasseling stage, constituting different salt tolerance training modes, referred to as S0-2-3, S0-3-3, S1-2-3, S1-3-3, S2-2-3, and S2-3-3. The results showed that although their fine root length density (FRLD) increased, the S0-2-3 and S0-3-3 treatments reduced the limit of root extension in the horizontal direction, causing the roots to be mainly distributed near the plants. This resulted in decreased leaf area and biomass accumulation, ultimately leading to significant yield reduction. Additionally, the S2-2-3 and S2-3-3 treatments stimulated the adaptive mechanism of maize roots, resulting in boosted fine root growth to increase the FRLD and develop into deeper soil layers. However, due to the prolonged exposure to a high level of salinity, their roots below 30 cm depth senesced prematurely, leading to an inhibition in shoot growth and also resulting in yield reduction of 10.99% and 11.75%, compared to CK, respectively. Furthermore, the S1-2-3 and S1-3-3 treatments produced more reasonable distributions of FRLD, which did not boost fine root growth but established fewer weak areas (FLRD < 0.66 cm−3) in their root systems. Moreover, the S1-2-3 treatment contributed to increasing leaf development and biomass accumulation, compared to CK, whereas it allowed for minimizing yield reduction. Therefore, our study proposed the S1-2-3 treatment as the recommended training mode for summer maize while utilizing brackish water resources.


Introduction
Brackish water irrigation, as an important unconventional water resource utilization method, plays a crucial role in alleviating the contradiction between water supply and demand, which contributes to ensuring food production security [1,2].In China, the natural replenishment of brackish water reaches 24.6 billion cubic meters, mainly distributed in salt-affected areas with a total area of over 36.67 million ha, serving as crucial backup irrigation water sources for such areas [3,4].However, the utilization of brackish water for irrigation inevitably increases the soil salt content, which may lead to secondary salinization issues [5].Therefore, the core issue of using brackish water for agricultural irrigation

Biomass Accumulation in Summer Maize at Different Growth Stages 2.2.1. Shoot and Root Biomass
Biomass is an important indicator for measuring the accumulation of organic matter and nutrient content in plants.As shown in Figure 2, salt stress generally had a significant impact on SB and RB throughout the growth period.For SB, after the FSTT stage (DAS = 21~28), the SB of S 0-2-3 , S 0-3-3 , and CK treatments were obviously higher than the other treatments.After a recovery stage (DAS = 29~34), compared to CK, S 1-2-3 , S 1-3-3 , S 2-2-3 , and S 2-3-3 resulted in a reduction in SB by 13.38%, 9.99%, 19.69%, and 22.18%, respectively.After the SSTT stage (DAS = 35~45), the SB of all treatments irrigated with brackish water significantly decreased compared to CK (p < 0.05) (Figure 2d).Among them, the SB of S 0-2-3 increased by 9.76% compared to S 0-3-3 , the SB of S 1-2-3 increased by 9.17% compared to S 1-3-3 , and the SB of S 2-2-3 increased by 5.34% compared to S 2-3-3 .At the beginning of the SST stage (DAS = 52), compared to those at the start of the SSTT stage (DAS = 35), S 2-2-3 and S 2-3-3 exhibited greater magnitude of increase, while S 0-2-3 and S 0-3-3 showed smaller magnitude of increase than the other treatments (Figure 2c,e).At the ending of the SST stage (DAS = 66), S 1-2-3 displayed the highest SB, and compared to those at the beginning of the SST stage, both S 1-2-3 and S 1-3-3 had greater magnitude of increase than the other treatments (Figure 2e,f).At harvest (DAS = 99), the SB for S 1-2-3 and CK were higher than those of the other treatments, with all differences reaching a significant level (p < 0.05).S 0-2-3 and S 0-3-3 exhibited greater magnitude of increase after the SST stage, but their SB were lower than those of the other treatments at harvest (Figure 2f,g).
For RB, all treatments exhibited a trend of initial increase followed by decrease.Except for S 2-3-3 , which started to decrease before the ending of the SST stage, the RB values for all treatments increased with the number of days after sowing until DAS = 66, but with different magnitudes of increase at different stages.During the FSTT stage, S 0-2-3 , S 0-3-3 , and CK showed much higher increases in RB compared to the other treatments (Figure 2a,b).At the ending of this stage, their RB values significantly differed from S 2-2-3 and S 2-3-3 (p < 0.05).However, this pattern changed during the SSTT stage, while the S 2-2-3 , S 2-3-3 , and CK treatments exhibited increases of over 190%, greatly higher than the other treatments.S 0-3-3 only showed an increase of 67.72%, and its RB was the smallest at DAS = 45.During the SST stage, S 0-3-3 exhibited a higher magnitude of increase compared to the other treatments.However, at the end of this stage, its RB value remained the smallest (Figure 2f).At harvest, all treatments exhibited a decrease in RB to varying degrees, with the RB values in the following order from highest to lowest: S 0-2-3 , CK, S 1-2-3 , S 2-3-3 , S 2-2-3 , S 0-3-3 , S 1-3-3 .This differed obviously from the pattern observed for SB.For RB, all treatments exhibited a trend of initial increase followed by decrease.Except for S2-3-3, which started to decrease before the ending of the SST stage, the RB values for all treatments increased with the number of days after sowing until DAS = 66, but with different magnitudes of increase at different stages.During the FSTT stage, S0-2-3, S0-3-3, and CK showed much higher increases in RB compared to the other treatments (Figure 2a,b).At the ending of this stage, their RB values significantly differed from S2-2-3 and S2-3-3 (p < 0.05).However, this pattern changed during the SSTT stage, while the S2-2-3, S2-3-3, and CK treatments exhibited increases of over 190%, greatly higher than the other treatments.S0-3-3 only showed an increase of 67.72%, and its RB was the smallest at DAS = 45.During the SST stage, S0-3-3 exhibited a higher magnitude of increase compared to the other treatments.However, at the end of this stage, its RB value remained the smallest (Figure 2f).At harvest, all treatments exhibited a decrease in RB to varying degrees, with the RB values in the following order from highest to lowest: S0-2-3, CK, S1-2-3, S2-3-3, S2-2-3, S0-3-3, S1-3-3.This differed obviously from the pattern observed for SB.

Grain Yield
The grain yield of summer maize is a primary indicator for assessing its response to salt stress.As shown in Table 1, different treatments generally exhibited an inverse relationship between grain number per ear and hundred-grain weight.In terms of yield formation, all treatments irrigated with brackish water showed lower grain yields than the CK treatment.Compared to CK, S0-2-3 and S0-3-3 decreased by 18.74% and 20.17%, respectively; S1-2-3 and S1-3-3 decreased by 5.98% and 8.61%, respectively; S2-2-3 and S2-3-3 decreased by 10.99% and 11.75%, respectively.This indicated that for similar irrigation modes, the higher the mineralization degree of the irrigated brackish water, the lower the grain yield.Additionally, S0-3-3 and S0-2-3 showed the largest decrease, which also had significant differences from the CK treatment (p < 0.05).

Grain Yield
The grain yield of summer maize is a primary indicator for assessing its response to salt stress.As shown in Table 1, different treatments generally exhibited an inverse relationship between grain number per ear and hundred-grain weight.In terms of yield formation, all treatments irrigated with brackish water showed lower grain yields than the CK treatment.Compared to CK, S 0-2-3 and S 0-3-3 decreased by 18.74% and 20.17%, respectively; S 1-2-3 and S 1-3-3 decreased by 5.98% and 8.61%, respectively; S 2-2-3 and S 2-3-3 decreased by 10.99% and 11.75%, respectively.This indicated that for similar irrigation modes, the higher the mineralization degree of the irrigated brackish water, the lower the grain yield.Additionally, S 0-3-3 and S 0-2-3 showed the largest decrease, which also had significant differences from the CK treatment (p < 0.05).The data are the means of the ears produced by the remaining 24 maize plants in each treatment at harvest.The number after the plus or minus sign represents the standard deviation.Different lowercase letters represent significant differences in grain yield.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.The root length density is an important metric in root studies.In this study, the roots within 2 mm diameter were counted and then converted into the indicator of fine root length density (FRLD).Micro-plots plan and illustration of root sampling method are shown in Figure 3.A total of seven sampling events were conducted, covering the entire growth period of summer maize.The distributions of FRLD obtained from these seven sampling events are shown in Figure 4.It could be seen from the figure that throughout the entire growth period, the changes in FRLD of S 1-2-3 and S 1-3-3 treatments were relatively similar.Their fine roots grew rapidly in the 10~20 cm soil layer and maintained dense distributions in this region since the ending of the SSTT stage (DAS = 45).The other treatments, however, exhibited a different pattern, with their FRLD being highest in the 0-10 cm soil layer.Among them, S 0-3-3 , S 2-2-3 , and S 2-3-3 treatments even showed boosting trends in their fine root growth.The data are the means of the ears produced by the remaining 24 maize plants in each treatment at harvest.The number after the plus or minus sign represents the standard deviation.Different lowercase letters represent significant differences in grain yield.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.

Two-Dimensional Distribution of Fine Roots during the Nutritional Growth Stage
The distribution of FRLD in maize at the tasseling stage, sampled at DAS = 52, are shown in Figure 5.It can be seen from the figure that the FRLD of all treatments were concentrated in 0~20 cm depth.In the horizontal direction, except for the S2-3-3 treatment, the FRLD of other treatments were concentrated within a radius of about 10 cm from the plant.Vertically, the FRLD values gradually decreased with increasing depth.Among The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.DAS is the abbreviation for days after sowing.FRLD is the abbreviation for fine root length density.

Two-Dimensional Distribution of Fine Roots during the Nutritional Growth Stage
The distribution of FRLD in maize at the tasseling stage, sampled at DAS = 52, are shown in Figure 5.It can be seen from the figure that the FRLD of all treatments were concentrated in 0~20 cm depth.In the horizontal direction, except for the S 2-3-3 treatment, the FRLD of other treatments were concentrated within a radius of about 10 cm from the plant.Vertically, the FRLD values gradually decreased with increasing depth.Among them, the root extension range of S 2-2-3 and S 2-3-3 exceeded that of the other treatments, and their average FRLD in the soil profile increased by 45.40% and 75.48%, compared to the CK treatment, respectively.However, for the FRLD in the 0~20 cm soil layer directly below the plants (x = 0 cm), S 2-2-3 showed an increase of 34.42% compared to CK, while S 2-3-3 showed a decrease of 11.52% compared to CK. S 1-2-3 and S 1-3-3 exhibited reductions of 26.97% and 10.40%, respectively, in the FRLD within the 0~20 cm soil layer at x = 0 cm, compared to CK.However, both treatments showed relatively dense distributions of FRLD in the 10~20 cm soil layer at a distance of 10 cm from the plant (Figure 5c,d), which were consistent with the vertical distribution in Figure 4c,d.Moreover, compared to the CK treatment, the average FRLD in the soil profile decreased by 5.90% in S 1-2-3 , while it increased by 7.18% in S 1-3-3 .The S 0-2-3 and S 0-3-3 treatments exhibited reductions of 36.58% and 47.34%, compared to CK, respectively, in the FRLD within the 0~20 cm soil layer at x = 0 cm, which were much lower than the other treatments.However, both treatments showed elongated Plants 2023, 12, 3329 7 of 18 root systems, with S 0-2-3 showing a relatively dense distribution of FRLD in the 30~40 cm soil layer at a distance of 10 cm from the plant (Figure 5a), and S 0-3-3 showing a relatively dense distribution of FRLD in the 20~30 cm soil layer at a distance of 10 cm from the plant (Figure 5b).Furthermore, compared to the CK treatment, the average FRLD in the soil profile increased by 2.23% in S 0-2-3 , while it decreased by 5.64% in S 0-3-3 .
compared to CK.However, both treatments showed relatively dense distributions of FRLD in the 10~20 cm soil layer at a distance of 10 cm from the plant (Figure 5c,d), which were consistent with the vertical distribution in Figure 4c,d.Moreover, compared to the CK treatment, the average FRLD in the soil profile decreased by 5.90% in S1-2-3, while it increased by 7.18% in S1-3-3.The S0-2-3 and S0-3-3 treatments exhibited reductions of 36.58% and 47.34%, compared to CK, respectively, in the FRLD within the 0~20 cm soil layer at x = 0 cm, which were much lower than the other treatments.However, both treatments showed elongated root systems, with S0-2-3 showing a relatively dense distribution of FRLD in the 30~40 cm soil layer at a distance of 10 cm from the plant (Figure 5a), and S0-3-3 showing a relatively dense distribution of FRLD in the 20~30 cm soil layer at a distance of 10 cm from the plant (Figure 5b).Furthermore, compared to the CK treatment, the average FRLD in the soil profile increased by 2.23% in S0-2-3, while it decreased by 5.64% in S0-3-3.
Horizontally, in 0~10 cm depth, all treatments exhibited the highest FRLD directly below the plants (x = 0 cm), and FRLD decreased with increasing distance from the plants.Below 10 cm depth, FRLD might increase and then decrease with increasing distance from the plants, exhibiting dense distribution at certain positions.This trend was more pronounced in the treatments irrigated with brackish water, compared to CK, indicating that irrigation with brackish water promoted horizontal root extension in the soil below 10 cm depth.Vertically, as the soil depth increased, the FRLD of each treatment showed a decreasing trend.Except for a few positions, this top-down decreasing trend was more pronounced in 0~40 cm depth, but there might be a phenomenon of larger FRLD below 40 cm depth.The majority of weak areas in all treatments were located in the 30~60 cm soil layer, particularly in S 2-2-3 and S 2-3-3 , where the weak areas in the 0~30 cm soil layer were less than 1% (Figure 6e,f).Horizontally, in 0~10 cm depth, all treatments exhibited the highest FRLD directly below the plants (x = 0 cm), and FRLD decreased with increasing distance from the plants.Below 10 cm depth, FRLD might increase and then decrease with increasing distance from the plants, exhibiting dense distribution at certain positions.This trend was more pronounced in the treatments irrigated with brackish water, compared to CK, indicating that irrigation with brackish water promoted horizontal root extension in the soil below 10 cm depth.Vertically, as the soil depth increased, the FRLD of each treatment showed a decreasing trend.Except for a few positions, this top-down decreasing trend was more pronounced in 0~40 cm depth, but there might be a phenomenon of larger FRLD below 40 cm depth.The majority of weak areas in all treatments were located in the 30~60 cm soil layer, particularly in S2-2-3 and S2-3-3, where the weak areas in the 0~30 cm soil layer were less than 1% (Figure 6e,f).

Effects of Salt Tolerance Training on Biomass Allocation and Yield Formation of Summer Maize under Brackish Water Irrigation
Numerous studies indicate that soil salt stress affects crop canopy development, biomass accumulation, and allocation processes [17,27,28].Among these, changes in leaf area are closely associated with crop growth and yield.Typically, a reduction in leaf area index (LAI) diminished light interception by crops, leading to a decrease in biomass production [29].In this study, regardless of the extinction from growth stages, Plants 2023, 12, 3329 9 of 18 maize experienced a reduction in LAI upon the first occurrence of salt stress.Subsequent growth was then inhibited for a certain stage, and the extent of this inhibition correlated positively with the degree of brackish water mineralization.This was attributed to salt stress inducing ionic imbalance and nutrient disruption in crops, damaging root water uptake and leaf photosynthesis, ultimately leading to retarded growth and development [30][31][32].Furthermore, as the brackish water mineralization degree increased, the accumulation of salts in the soil also intensified [33], resulting in a more pronounced stress response.Nevertheless, treatments exposed to the S1 and S2 levels during the first salt-tolerancetraining (FSTT) stage (DAS = 21~28) exhibited less LAI suppression and greater biomass increase during the second salt-tolerance-training (SSTT) stage (DAS = 35~45), compared to untreated ones.This indicated that the FSTT enhanced maize salt tolerance, with the S2 level showing more effective conditioning.After the SSTT stage, the treatments subjected to the S1 level during the FSTT stage began to exhibit stronger growth ability.Particularly, the S 1-2-3 treatment showed higher peak value of LAI and biomass accumulation, compared to CK.In other studies, it has also been found that irrigating with slightly brackish water in a reasonable manner did not inhibit canopy growth or biomass accumulation in crops [34,35].
However, the adverse consequences of brackish water irrigation should not be underestimated.In this study, the grain yields of all treatments irrigated with brackish water were lower than that of the CK treatment.Furthermore, under the same initial salt level during the FSTT stage, a higher salinity during the SSTT stage resulted in even lower maize grain yield.This trend aligned with the findings of Zhu's experiment [36].The decrease in grain yield was primarily attributed to a reduction in grain number per ear (Table 1).This reduction could be due to the elevated Na + /K + ratio in summer maize leaves caused by salt tolerance training, accelerating leaf senescence and death during the reproductive growth stage [37].Thus, CO 2 absorption by maize was reduced, leading to decreased organic compound synthesis.This also indicated that salt tolerance training only mitigated the adverse impact of salt stress on summer maize growth to a certain extent.
In addition, the S 0-2-3 and S 0-3-3 treatments, which were irrigated with freshwater during the early growth stage and brackish water during the middle and later growth stages, exhibited slower rates of leaf senescence.This allowed them to undergo compensatory growth after DAS = 66.A similar phenomenon was observed in Ma's research [24].However, the biomass accumulation and grain yield of S 0-2-3 and S 0-3-3 were lower than those of the other treatments at harvest.This could be attributed to the delayed salt tolerance training, leading to lower level of leaf development when entering the tasseling stage (DAS = 52), thereby reducing photosynthetic efficiency during the critical stage of yield formation and resulting in substantial yield reduction.Consequently, conducting salt tolerance training during the early growth stage of maize is highly necessary.

Effects of Salt Tolerance Training on the Multidimensional Root Distribution of Summer Maize under Brackish Water Irrigation
The morphological structural characteristics of the plant root system in the threedimensional soil space play an important role in the absorption of water and nutrients by roots, as well as the growth of the above-ground parts [38].However, research on the response patterns of the three-dimensional root distribution under salt stress are relatively limited.Relevant studies have been conducted on small scales, focusing on plants like Arabidopsis in controlled indoor environments [39].In the real-world context of saltaffected fields, there is still a scarcity of reports regarding the multidimensional root distribution traits exhibited by vital salt-tolerant crops such as maize, cotton, and sunflower under varying saline conditions.Therefore, investigating the growth and distribution of maize root systems in the multidimensional soil zone is of great significance for conserving freshwater resources and ensuring high maize grain yields.
Wu et al. [40] found that under field conditions, 50~80% of maize roots were distributed in 0~20 cm depth.In this study, similar results were obtained, with the majority of FRLD for all treatments (over 65% at DAS = 66) concentrated in 0~20 cm depth throughout the entire growth period.The growth of plant roots is influenced by multiple factors, including gravitropism, hydrotropism, and oxytropism [41,42].In this study, based on Figure 4, the CK, S 0-2-3 , and S 0-3-3 treatments indicated that conventional freshwater irrigation and early-stage freshwater irrigation resulted in the concentration of fine root growth in the surface layer (0~10 cm depth) during the crop cycle, whereas the FSTT resulted in more fine roots growing in the deeper layer (10~20 cm depth).Considering Figures 7 and 8, this phenomenon was attributed to the fact that during the FSTT stage (DAS = 21~28), the salt content in the surface layer was excessively high, while the moisture content and salt content in the 10~20 cm soil layer were within a suitable range for root growth.Nevertheless, under S 2-2-3 and S 2-3-3 , the fine roots shifted their concentrated growth area back to the surface layer after the SSTT stage (DAS = 35~45), which might be due to the S2 level training during the FSTT stage which enhanced the adaptability of fine roots to severe salt stress and gradually encouraged their growth back into the regular area.
bution traits exhibited by vital salt-tolerant crops such as maize, cotton, and sunflower under varying saline conditions.Therefore, investigating the growth and distribution of maize root systems in the multidimensional soil zone is of great significance for conserving freshwater resources and ensuring high maize grain yields.
Wu et al. [40] found that under field conditions, 50~80% of maize roots were distributed in 0~20 cm depth.In this study, similar results were obtained, with the majority of FRLD for all treatments (over 65% at DAS = 66) concentrated in 0~20 cm depth throughout the entire growth period.The growth of plant roots is influenced by multiple factors, including gravitropism, hydrotropism, and oxytropism [41,42].In this study, based on Figure 4, the CK, S0-2-3, and S0-3-3 treatments indicated that conventional freshwater irrigation and early-stage freshwater irrigation resulted in the concentration of fine root growth in the surface layer (0~10 cm depth) during the crop cycle, whereas the FSTT resulted in more fine roots growing in the deeper layer (10~20 cm depth).Considering Figures 7 and 8, this phenomenon was attributed to the fact that during the FSTT stage (DAS = 21~28), the salt content in the surface layer was excessively high, while the moisture content and salt content in the 10~20 cm soil layer were within a suitable range for root growth.Nevertheless, under S2-2-3 and S2-3-3, the fine roots shifted their concentrated growth area back to the surface layer after the SSTT stage (DAS = 35~45), which might be due to the S2 level training during the FSTT stage which enhanced the adaptability of fine roots to severe salt stress and gradually encouraged their growth back into the regular area.Relevant research has shown that fine roots have plasticity and can rapidly respond to various stressors in the soil by altering their own length, growth direction and other characteristics [43,44].Yang et al. [45] conducted research on the two-dimensional root distribution of Jerusalem artichoke under different saline conditions.It was found that there were no significant differences in the root length density (RLD) of Jerusalem artichoke in both horizontal and vertical distributions under low saline condition.Particularly, their moderate salinity level (1.6~1.8 g salt/kg soil) even exhibited the potential to stimulate RLD growth, which was also found in our study on summer maize irrigated with brackish water.However, the two-dimensional distribution of RLD in Yang's study was greatly changed when the salt level further increased.For example, at high salinity level (2.3~3.0 g salt/kg soil), the horizontal RLD was observed to surpass the vertical RLD.In the present study, when summer maize entered the tasseling stage (DAS = 52), the FRLD of S 2-2-3 and S 2-3-3 showed denser distribution both horizontally and vertically compared to the other treatments.Thereafter, at silking (DAS = 66), these two treatments still maintained dense distributions in the 0~30 cm soil layer, but more weak areas (FRLD < 0.66 cm −3 ) were found below 30 cm depth (Figure 6e,f).Particularly, the FRLD of S 2-3-3 generally showed a decline trend compared to that at DAS = 52, indicating that the root system of summer maize had reached its maximum before the reproductive growth stage under this training mode.Referring to Figure 8, it could be observed that S 2-3-3 reached an average EC e of Plants 2023, 12, 3329 11 of 18 3.70 dS•m −1 in 0~60 cm depth at DAS = 45, which was higher than the other treatments.Thus, we speculated that the root system of S 2-3-3 , under continuous high salt stress, underwent extensive growth early to maintain water and nutrient absorption.After tasseling, its root system had already met the requirements for reproductive growth, leading to a slowdown in its development.Relevant research has shown that fine roots have plasticity and can rapidly respond to various stressors in the soil by altering their own length, growth direction and other characteristics [43,44].Yang et al. [45] conducted research on the two-dimensional root distribution of Jerusalem artichoke under different saline conditions.It was found that there were no significant differences in the root length density (RLD) of Jerusalem artichoke in both horizontal and vertical distributions under low saline condition.Particularly, their moderate salinity level (1.6~1.8 g salt/kg soil) even exhibited the potential to stimulate RLD growth, which was also found in our study on summer maize irrigated with brackish water.However, the two-dimensional distribution of RLD in Yang's study was greatly changed when the salt level further increased.For example, at high salinity level (2.3~3.0 g salt/kg soil), the horizontal RLD was observed to surpass the vertical RLD.In the present study, when summer maize entered the tasseling stage (DAS = 52), the FRLD of S2-2-3 and S2-3-3 showed denser distribution both horizontally and vertically compared to the other treatments.Thereafter, at silking (DAS = 66), these two treatments still maintained dense distributions in the 0~30 cm soil layer, but more weak areas (FRLD < 0.66 cm −3 ) were found below 30 cm depth (Figure 6e,f).Particularly, the FRLD of S2-3-3 generally showed a decline trend compared to that at DAS = 52, indicating that the root system of summer maize had reached its maximum before the reproductive growth stage under this training mode.Referring to Figure 8, it could be observed that S2-3-3 reached an average ECe of 3.70 dS•m −1 in 0~60 cm depth at DAS = 45, which was higher than the other treatments.Thus, we speculated that the root system of S2-3-3, under continuous high salt stress, underwent extensive growth early to maintain water and nutrient absorption.After tasseling, its root system had already met the requirements for reproductive growth, leading to a slowdown in its development.
On the other hand, among all treatments, only S0-2-3 and S0-3-3 generated horizontal weak areas in x = 25~35 cm which vertically spanned from 10 to 60 cm depth (Figure 6a,b), and after DAS = 66, their FRLD declining rates were lower than those of the other treatments.This indicated that experiencing salt stress for the first time at ten-leaf stage could delay fine root adaptation and reduce the limit of root extension in the horizontal direction, leading to the majority of roots accumulating in the soil close to the plants.For S1-2-3 On the other hand, among all treatments, only S 0-2-3 and S 0-3-3 generated horizontal weak areas in x = 25~35 cm which vertically spanned from 10 to 60 cm depth (Figure 6a,b), and after DAS = 66, their FRLD declining rates were lower than those of the other treatments.This indicated that experiencing salt stress for the first time at ten-leaf stage could delay fine root adaptation and reduce the limit of root extension in the horizontal direction, leading to the majority of roots accumulating in the soil close to the plants.For S 1-2-3 and S 1-3-3 , their average FRLD of the profile (0~60 cm) did not differ greatly from CK at DAS = 52 and 66, but their weak areas of FRLD took up less space than CK in the three-dimensional soil zone, especially for S 1-2-3 , which was the least among all treatments.Considering the above-ground growth indicators, it could be inferred that these two irrigation modes, especially the S 1-2-3 treatment, promoted a reasonable spatial distribution of FRLD under salt tolerance training.

The Regulation Strategy of Root and Shoot Growth and Salt Tolerance Enhancement in Summer Maize under Brackish Water Irrigation
The ability of plants to survive and produce harvestable yield under salt stress is known as salt tolerance [12].The salt tolerance varies among different plant species.Generally, plant growth is inhibited under salt stress, and the stronger the salt tolerance of a plant, the less growth inhibition it experiences under salt stress [46].Some studies have indicated that plant root growth in the early stages could be severely restricted by root zone salinity [47,48].Our present study verifies this conclusion, demonstrating that salt tolerance training at S1 and S2 levels during the early growth stage would reduce the FRLD of summer maize, with the reduction degree positively correlated with the degree of brackish water mineralization, consistent with the growth pattern of the above-ground parts.The results of S 0-2-3 and S 0-3-3 treatments provided evidence that this pattern was similarly applicable to summer maize when it entered ten-leaf stage.Additionally, our study observed that after the first exposure to salt stress, the above-ground parts of all treatments could experience rapid growth during a subsequent period.This could be explained as a result of increased salt tolerance, in which the irrigation with brackish water enhanced the salt tolerance of summer maize, thus alleviating mid-to-late term salt stress and promoting leaf development and biomass accumulation.For S 2-2-3 and S 2-3-3 , this alleviating effect became evident from the ten-leaf stage; for S 1-2-3 and S 1-3-3 , it became apparent from the tasseling stage.However, for S 0-2-3 and S 0-3-3 , the relief from salt stress in summer maize only became evident after the silking stage, while irreversible damages had already occurred.
The summer maize under S 0-2-3 , S 0-3-3 , S 2-2-3 , and S 2-3-3 experienced rapid belowground growth ahead of the above-ground parts, indicating that when crops were subjected to salt stress, new biomass would be preferentially allocated to the root system to ensure water and nutrient absorption.Similar conclusions have also been reached by Mound and Maghsoudi [22] and Meloni et al. [49].Thus, our study suggested that increasing fine root growth could clearly enhance the salt tolerance of summer maize.However, S 1-2-3 and S 1-3-3 exhibited a different pattern.Their FRLD did not have a rapid growth like the above-ground parts, yet grain yield and other growth indicators performed better compared to other treatments irrigated with brackish water.As mentioned in the previous section, the spatial distributions of FRLD in S 1-2-3 and S 1-3-3 were more reasonable, which might be the reason behind this phenomenon.Related research has pointed out that a well-structured distribution and spatial configuration of crop's fine roots contributed to the efficient uptake and utilization of water and nutrients by plants [50,51].Moreover, in the presence of stress factors such as drought and salinity, plants were able to respond to such stressors by adjusting the spatial distribution of roots throughout the soil profile using their inherent root plasticity [43,44].This enabled them to enhance their survival capability under stressful conditions.Another interpretation of this result was that appropriate salt tolerance training enhanced the root vitality of summer maize, leading to improved water uptake capacity of the root system at the same length.As a result, the plants were able to maintain water and nutrient absorption without the need for increasing soil-root contact areas by extensive fine root growth.The study by Wang [52] provided evidence for this viewpoint, showing that root vitality increased with higher salinity.Further research and investigation are necessary to explore the relationship between root vitality and brackish water irrigation under salt tolerance training.

Experimental Site Description
The experiment was conducted from June 2022 to October 2022 in the experimental field of Jiangning Water-saving Park, Jiangsu Province, China (31 • 54 N, 118 • 46 E).The experimental site belongs to a subtropical monsoon climate with distinct seasons, characterized by cold winters and hot summers.The annual average temperature is 15.3 • C, with an annual sunshine duration of 2213 h and an average annual rainfall of 1051 mm.During the experiment, the average max temperature was 33.61 • C, the average min temperature was 24.96 • C, and the average radiation was 13.74 MJ•m −2 •day −1 (Figure 9).Before the experiment commenced, the basic properties of the experimental soil at depths ranging from 0 to 60 cm were determined (Table 2).

Experimental Site Description
The experiment was conducted from June 2022 to October 2022 in the experimental field of Jiangning Water-saving Park, Jiangsu Province, China (31°54′ N, 118°46′ E).The experimental site belongs to a subtropical monsoon climate with distinct seasons, characterized by cold winters and hot summers.The annual average temperature is 15.3 °C, with an annual sunshine duration of 2213 h and an average annual rainfall of 1051 mm.During the experiment, the average max temperature was 33.61 °C, the average min temperature was 24.96 °C, and the average radiation was 13.74 MJ•m −2 •day −1 (Figure 9).Before the experiment commenced, the basic properties of the experimental soil at depths ranging from 0 to 60 cm were determined (Table 2).

Experimental Design
The experiment was conducted in seven micro-plots within the experimental site.Prior to the experiment, impermeable membranes were laid at the bottom and surrounding areas of each micro-plot to prevent water exchange between adjacent micro-plots and the groundwater.The dimension of each micro-plot is 2.0 m × 2.6 m × 0.7 m (length × width × depth), with partitions used in the middle to separate each micro-plot into two replicated

Experimental Design
The experiment was conducted in seven micro-plots within the experimental site.Prior to the experiment, impermeable membranes were laid at the bottom and surrounding areas of each micro-plot to prevent water exchange between adjacent micro-plots and the groundwater.The dimension of each micro-plot is 2.0 m × 2.6 m × 0.7 m (length × width × depth), with partitions used in the middle to separate each micro-plot into two replicated sub-plots.The tested summer maize variety was Su Yu 29.When the moisture conditions in the plow layer were suitable, manual hill-dropping was conducted with a row spacing of 50 cm and a plant spacing of 25 cm.The maize was sown on 28 June 2022, with 3 seeds per hill and 1 seedling retained per hill after thinning, resulting in a planting density of 83,000 plants•ha −1 .Additionally, maize harvesting took place on 5 October 2022, which was 99 days after sowing (DAS = 99).
The experiment involved the irrigation of maize at different growth stages with brackish water of varying salt levels.Four levels of salt concentration were set for irrigation water: non-stress (S0, 0.08 g•L −1 ), mild salt stress (S1, 2.0 g•L −1 ), moderate salt stress (S2, 4.0 g•L −1 ), and severe salt stress (S3, 6.0 g•L −1 ).The brackish water was prepared by mixing NaCl, Na 2 SO 4 , and NaHCO 3 in a mass ratio of 0.61:0.31:0.08,based on the ionic composition and concentration of local brackish water.The irrigation was performed while summer maize entered the six-leaf stage, the ten-leaf stage, and the tasseling stage, forming seven different salt tolerance training modes.As shown in Table 3, the first salt-tolerancetraining (FSTT) started on 21 DAS when summer maize entered the six-leaf stage and ended on 28 DAS; the second salt-tolerance-training (SSTT) started on 35 DAS when summer maize entered the ten-leaf stage and ended on 45 DAS; the severe stress test (SST) started Prior to sowing, each micro-plot was plowed and fertilized with compound fertilizer as the base fertilizer.The effective contents of the fertilizer included 135 kg•ha −1 of nitrogen (as pure N), 75 kg•ha −1 of phosphorus (as pure P 2 O 5 ), and 90 kg•ha −1 of potassium (as pure K 2 O), with consistent fertilization amounts across different treatment plots.After fertilization, each micro-plot was thoroughly irrigated.When summer maize reached the ten-leaf stage, an additional 135 kg•ha −1 of nitrogen (as pure N) was applied as topdressing, using urea as the fertilizer.In addition, various cultivation management practices such as pest control and weed control were consistent with actual production.As the experiment was conducted under rain-shelter conditions, a rain shelter was used to protect against rain on rainy days, and it was opened on non-rainy days.Therefore, the influence of rainfall was not considered.In this experiment, soil samples were collected using a stainless steel soil auger in six layers at depths of 0~10 cm, 10~20 cm, 20~30 cm, 30~40 cm, 40~50 cm, and 50~60 cm.Each replicate plot was sampled twice, and the soil samples were placed in aluminum boxes and oven-dried at 105 • C in the laboratory to measure the soil moisture content.Sampling was conducted on 21, 28, 35, 45, 52, 66 and 99 DAS.The results are shown in Figure 7.
(2) Soil Salt Content After the determination of soil moisture content, the soil samples were ground, and soil extracts were prepared at a soil-to-water ratio of 1:5 [53].The soil extracts were shaken for 1 h and then allowed to settle.After the solution was clarified, the electrical conductivity (EC 1:5 , dS•m −1 ) was measured using a conductivity meter.The measured EC 1:5 value was then converted to the electrical conductivity of a saturated-paste extract (EC e , dS•m −1 ) using the empirical formula [54].The results are shown in Figure 8.The empirical formula is given by: EC e = 7.4 × EC 1:5 (1)

Crop Growth Parameters
(1) Leaf Area Index In this experiment, the leaf area of summer maize was determined using the conventional method, which involved measuring the area of each leaf and summing up total leaf area of the whole plant [55].The formula for calculating the area of a single leaf is as follows: Lea f area = Maximum lea f length × Maximum lea f width × 0.75 (2) Four maize plants were marked in each micro-plot for each treatment, with two plants in each replicate plot.Leaf area measurements were conducted on 11,18,21,28,35,38,43,48,53,58,63,66,76,86, and 96 DAS, and the leaf area index (LAI) was calculated using the formula.The LAI represents the ratio of total leaf area to land area on a unit land area [56], and its calculation formula is as follows: (2) Biomass Accumulation At the beginning and ending of each brackish water irrigation stage, namely on 21, 28, 35, 45, 52, and 66 DAS, two representative maize plants with good growth were destructively sampled from each replicate plot.The various plant organs, including leaves, stems, and ears, were oven-dried at 75 • C until they reached constant weights in order to obtain their biomasses.Among them, two plants were selected for destructive sampling on 66 DAS from the four designated maize plants.At harvest, which was 99 DAS, the remaining designated maize plants were subjected to the same treatment.
(3) Grain Yield At harvest, grain yield measurements were conducted on all remaining maize plants within hundred-grain weight, grain number per ear, and other indicators.

Maize Root System Parameters
In this experiment, root sampling was conducted on 21, 28, 35, 45, 52, 66, and 99 DAS.As shown in Figure 3, a root auger of 6.4 cm diameter was used to drill in layers from 0 to 60 cm depth to mainly sample the crown, primary, and seminal roots, with the maize plant as the center.For lateral root sampling, another root auger of 4.6 cm diameter was used at different distances from the maize plants.The sampling depth was consistent for all samples: 0~10 cm, 10~20 cm, 20~30 cm, 30~40 cm, 40~50 cm, 50~60 cm.Meanwhile, the maize roots below 60 cm depth had been proven to be scarce [57].The obtained root and soil samples were separated and washed.The roots were scanned using the Epson Perfection 4990 Photo scanner, and the WinRHIZO software was used to analyze the scanned images to obtain root length and other data.Only the length of fine roots (diameter ≤ 2 mm) was taken into account in this study.After scanning, the maize roots were oven-dried at 75 • C until a constant weight was reached.The dried roots were weighed using an analytical balance with a resolution of 0.0001 g.

Data Analysis
The data obtained in this experiment were processed using Excel 2016 (Microsoft Corp., Redmond, WA, USA).The results of each treatment were analyzed using SPSS statistics 25 (SPSS Inc.IBM Corp., Armonk, NY, USA).Graphs were generated using Origin 2021 (OriginLab Corp., Northampton, MA, USA).The comparison of means for each treatment was conducted using the Duncan test at a significance level of p < 0.05.

Conclusions
Our study indicated that different salt tolerance training modes significantly affected summer maize growth and altered the temporal and spatial distribution of fine roots under brackish water irrigation.Irrigated with brackish water in the first or second salttolerance-training (FSTT or SSTT) stage, it could stimulate the adaptive mechanism of maize roots and promote rapid growth of above-ground parts during a subsequent period.The distinction was that appropriate salt tolerance training (S 1-2-3 and S 1-3-3 ) could promote a reasonable root spatial distribution while maintaining a relatively stable FRLD, improving leaf development and biomass accumulation, and minimizing the adverse effects of salt stress on grain yield.On the other hand, improper salt tolerance training could lead to accelerated root senescence (S 2-2-3 and S 2-3-3 ) or delayed fine root adaptation (S 0-2-3 and S 0-3-3 ), inhibit the development of above-ground parts, and reduce yield formation during the reproductive growth stage.Among all the training modes, the S 1-2-3 treatment showed a 5.02% increase in leaf area, a 2.88% increase in above-ground biomass accumulation, and only a 5.98% decrease in grain yield, compared to conventional freshwater irrigation (CK).This suggests that irrigating the summer maize according to S 1-2-3 training mode can utilize brackish water resources rationally with minimal yield reduction.Overall, our study provides a valuable reference for exploring the water-saving potential by more efficient utilization methods of brackish water in salt-affected areas.

Figure 1 .
Figure 1.Dynamic changes in leaf area index (LAI) of summer maize under brackish water irrigation during the entire growth period.The data are averaged measurements (n = 4), and vertical bars indicate standard deviation.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.DAS is the abbreviation for days after sowing.

Figure 1 .
Figure 1.Dynamic changes in leaf area index (LAI) of summer maize under brackish water irrigation during the entire growth period.The data are averaged measurements (n = 4), and vertical bars indicate standard deviation.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.DAS is the abbreviation for days after sowing.

Figure 3 .
Figure 3. Micro-plots plan and illustration of root sampling method.Positions A and B were designated as sampling locations for the crown, primary, and seminal roots, while positions C-J were designated as sampling locations for the lateral roots.Root sampling was conducted at positions C and G on 21, 28, 35, 45 and 99 days after sowing.On 52 days after sowing, root sampling was conducted at positions A, C, G, and H.Then, 66 days after sowing, root sampling was conducted at all designated positions.

Figure 3 .
Figure 3. Micro-plots plan and illustration of root sampling method.Positions A and B were designated as sampling locations for the crown, primary, and seminal roots, while positions C-J were designated as sampling locations for the lateral roots.Root sampling was conducted at positions C and G on 21, 28, 35, 45 and 99 days after sowing.On 52 days after sowing, root sampling was conducted at positions A, C, G, and H.Then, 66 days after sowing, root sampling was conducted at all designated positions.

Figure 4 .
Figure 4. Changes in FRLD at different soil depths during the entire growth period of summer maize under different treatment.The data are averaged measurements at positions C and G in Figure 3.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.DAS is the abbreviation for days after sowing.FRLD is the abbreviation for fine root length density.

Figure 4 .
Figure 4. Changes in FRLD at different soil depths during the entire growth period of summer maize under different treatment.The data are averaged measurements at positions C and G in Figure 3.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.DAS is the abbreviation for days after sowing.FRLD is the abbreviation for fine root length density.

Figure 5 .
Figure 5. 2D distribution of FRLD of summer maize in each treatment on 52 days after sowing.The data are derived from measurements at positions A, G, and H in Figure 3.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.FRLD is the abbreviation for fine root length density.

Figure 5 .
Figure 5. 2D distribution of FRLD of summer maize in each treatment on 52 days after sowing.The data are derived from measurements at positions A, G, and H in Figure 3.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.FRLD is the abbreviation for fine root length density.

Plants 2023 , 19 Figure 6 .
Figure 6.3D distribution of FRLD of summer maize in each treatment on 66 days after sowing.The data are derived from measurements at all designated positions in Figure 3.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.FRLD is the abbreviation for fine root length density.The values within the curly brackets on the upper half of the Z-axis indicate the proportion of weak FRLD areas (FRLD < 0.66 cm −3 ) in the 0~30 cm soil layer in the three-dimensional graph.The values within the curly brackets on the lower half of the Z-axis indicate the proportion of weak FRLD areas in the 30~60 cm soil layer in the three-dimensional graph.The values within the curly brackets above the X-axis represent the proportion of weak FRLD areas in the lateral view of the three-dimensional graph.

Figure 6 .
Figure 6.3D distribution of FRLD of summer maize in each treatment on 66 days after sowing.The data are derived from measurements at all designated positions in Figure 3.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.FRLD is the abbreviation for fine root length density.The values within the curly brackets on the upper half of the Z-axis indicate the proportion of weak FRLD areas (FRLD < 0.66 cm −3 ) in the 0~30 cm soil layer in the three-dimensional graph.The values within the curly brackets on the lower half of the Z-axis indicate the proportion of weak FRLD areas in the 30~60 cm soil layer in the three-dimensional graph.The values within the curly brackets above the X-axis represent the proportion of weak FRLD areas in the lateral view of the three-dimensional graph.

Figure 7 .
Figure 7. Changes in soil moisture content of each treatment during the entire growth period.The data are averaged measurements (n = 4), and horizontal bars indicate standard deviation.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.DAS is the abbreviation for days after sowing.

Figure 7 .
Figure 7. Changes in soil moisture content of each treatment during the entire growth period.The data are averaged measurements (n = 4), and horizontal bars indicate standard deviation.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.DAS is the abbreviation for days after sowing.

Figure 8 .
Figure 8. Changes in soil salt content (EC e ) of each treatment during the entire growth period.The data are averaged measurements (n = 4), and horizontal bars indicate standard deviation.The S0, S1, S2, and S3 represent different salt concentration levels of irrigation water, corresponding to none (0.08 g•L −1 ), mild (2.0 g•L −1 ), moderate (4.0 g•L −1 ), and severe (6.0 g•L −1 ) stress.DAS is the abbreviation for days after sowing.
LAI = lea f area per plant × Number o f plants per micro − plot micro − plot area

Table 1 .
Grain yield of summer maize under brackish irrigation.

Table 1 . Cont. Treatments Grain Number per Ear Hundred-Grain Weight (g) Single Plant Grain Yield (g•Plant −1 ) Grain Yield (kg•ha −1 )
Multidimensional Root Distribution of Summer Maize 2.3.1.Vertical Distribution of Fine Roots throughout the Entire Growth Period

Table 1 .
Grain yield of summer maize under brackish irrigation.

Table 2 .
Basic properties of the experimental soil at 0~60 cm depth.

Table 2 .
Basic properties of the experimental soil at 0~60 cm depth.