Effect of Water Conservation and Nitrogen Reduction on Root Growth and Yield in Spring Maize in Typical Sand Interlayered Soil

: Given the low water and fertiliser use efficiency and the extensive distribution of sand interlayered soil in the Hetao irrigation district (HID), this study aimed to investigate the effects of different irrigation and fertilisation regimes on root parameters and yield in spring maize grown in sand interlayered soil. A two-year field plot experiment was conducted using the spring maize “Ximeng 3358” under three irrigation and nitrogen levels. Root length (RL), surface area (RS), diameter (RD), volume (RV), and length density (RLD), grain yield, and water use efficiency (WUE) were examined. Root growth was inhibited at the sand layer, with approximately 72.46–87.37% of the roots concentrated in the 0–40 cm soil layer. Notably, the proportion of roots in the bottom layer was 24.61–87.37% higher than that in the sub-bottom layer. Moreover, RL, RS, RD, and RV peaked in the medium irrigation and nitrogen fertilisation (I2F2) treatment. Furthermore, correlation analysis showed that the root parameters were significantly positively correlated with yield and WUE, with RS being most correlated to yield and WUE. Roots at a narrow row spacing of 20 cm (NR20) and at a depth of 10–20 cm were strongly correlated with yield and WUE. Conclusively, the I2F2 treatment can be used as the optimal combination of water and nitrogen for sand interlayered soil farmlands.


Introduction
The Hetao irrigation district (HID) is one of the largest irrigation districts in China and an important commercial grain and oil production centre.Crop production in the HID is entirely irrigation-dependent due to its unique geographic environment (northwest inland arid zone), which is characterized by low rainfall and a high evapotranspiration-descent ratio > 10 [1].Recently, the water quota for HID has reduced to 4 × 10 9 m 3 due to the national unified regulation of the Yellow River's water resources, indicating a decrease of over 20% [1].This poses a significant challenge for agricultural production.Current irrigation methods are still primarily traditional surface irrigation, including drip irrigation and furrow irrigation [2].During the growing season, farmers usually adopt an empiric irrigation and fertilisation strategy of "more irrigation, more fertiliser" to increase yield.Notably, the irrigation depth is up to 400 mm, and there has been a yearly increase in fertiliser use [3], resulting in severe water resource wastage and non-point source pollution in farmlands [4,5].Therefore, it is imperative to develop an efficient water-and fertilisersaving farming strategy to maximize crop production in the region.
Layered soils are common in nature and result from wind transport and deposition, water flow, sedimentation, and long-term human cultivation and production [6,7].The HID, being a Yellow River alluvial plain, is characterised by soil stratification, with fields commonly alternating between sandy and clay layers [8], particularly with sandy layers at depths of 40-100 cm [8,9].Water and solute movement in sand interlayered soil differs considerably from that in homogeneous soils owing to the discontinuity of the soil [10], causing a sudden change in the soil water potential at the sand layer interface.Consequently, the distribution of water and solutes is also affected [11].Additionally, the sand interlayered soil structure in farmlands affects crop transpiration, soil evaporation, soil water and salt movement [12], groundwater use, and water use efficiency (WUE) [13].Several studies have been performed on the effects of different sand layer distributions and textures on soil water infiltration using soil columns [14,15].Water scarcity, soil salinization, diverse soil stratification, and low water-and fertiliser-use efficiencies are major challenges to sustainable agricultural development in the HID.Therefore, the efficient and rational use of water and nitrogen resources to promote sustainable agricultural production on sand interlayered soils in the HID has become an urgent issue.
Roots are not only important organs for plants to absorb water and nutrients, but also play a role in supporting the aboveground parts of plants and are crucial in regulating crop growth, development, and yield [16][17][18].Research evidence indicates that crop grain yield is closely related to root length, dry weight, surface area, and other factors [19][20][21][22][23]. Therefore, crop yield and resource utilisation efficiency can be improved by regulating root growth and distribution [16,24].Root growth at different growth stages can be significantly affected by irrigation and fertilisation, which significantly promoted maize yield and aboveground dry matter [24].Root in the 0-100 cm soil layer and yield of maize can be promoted by alternate furrow irrigation and nitrogen application under different irrigation and nitrogen application methods [16].Additionally, with increasing nitrogen input, the root parameters of summer maize at the grain filling stage showed an initial increase under-rainfed conditions, followed by a decrease [25].Deep fertilization at appropriate soil depths can also effect the root distribution, biomass water productivity, and yield formation, it adjusted root distribution and increased nitrogen and soil water use efficiency [26].Moreover, biodegradable films significantly improved root length density and root surface area compared with polyethylene films, resulting in increased yield [27].Furthermore, the proportion of shallow roots and soil water utilization in spring maize can be increased by ridge furrow film mulching on the Loess Plateau [28].Root growth and distribution can be effectively improved under full-film mulching in semiarid areas, thereby increasing crop yield [29].
Based on these findings, it could be concluded that previous research on roots have mainly focused on the effect of different irrigation and fertilisation methods on root development.However, root growth is influenced not only by the distribution of water and nutrients, but also by climatic conditions and soil texture [30].Currently, studies on the effects of sand interlayered soil structure on root growth and distribution are limited.Therefore, this study aimed to investigate the effects of water and nitrogen interaction on maize root growth in a typical sand interlayered soil.Overall, it is anticipated that this study will provide theoretical and technical information for the development of efficient cropping systems and water-fertiliser strategies for sand interlayered soil farmlands in the HID.

Experimental Site
The experiment was conducted at the Shuguang Irrigation Research Station in Linhe, Bayannur, Western Inner Mongolia Autonomous Region, China (107 • 26 ′ 02 ′′ E, 40 • 49 ′ 01 ′′ N) in 2016 and 2018.The plant material was the spring maize variety "Ximeng 3358" (Inner Mongolia Simon Seed Industry Co., Ltd., Bayannur, China, 2008).The experimental area has a typical continental arid climate with a dry atmosphere, high evaporation, low precipitation, and large diurnal temperature variation.The average annual precipitation is 143.5 mm, which is unevenly distributed throughout the year, with approximately 70% concentrated from July to September.There is significant interannual variability, and the annual evaporation is 2435.6 mm, being more than 10 times the precipitation.Spring weather is quite severe, with an average temperature of 7.5 • C and annual average relative humidity of about 51%.During the study period, the frost-free period lasted for 150 d, and the annual average sunshine duration was 3180 h.The tested soil was a silty loam with a sand interlayer, generally appearing at a depth of 50-90 cm.The physicochemical properties of the soils are listed in Table 1 (USDA).Rainfall and average temperature data from the spring maize growth period are shown in Figure 1.The experiment was conducted at the Shuguang Irrigation Research Station in Linhe, Bayannur, Western Inner Mongolia Autonomous Region, China (107°26′02″ E, 40°49′01″ N) in 2016 and 2018.The plant material was the spring maize variety "Ximeng 3358" (Inner Mongolia Simon Seed Industry Co., Ltd., Bayannur, China, 2008).The experimental area has a typical continental arid climate with a dry atmosphere, high evaporation, low precipitation, and large diurnal temperature variation.The average annual precipitation is 143.5 mm, which is unevenly distributed throughout the year, with approximately 70% concentrated from July to September.There is significant interannual variability, and the annual evaporation is 2435.6 mm, being more than 10 times the precipitation.Spring weather is quite severe, with an average temperature of 7.5 °C and annual average relative humidity of about 51%.During the study period, the frost-free period lasted for 150 d, and the annual average sunshine duration was 3180 h.The tested soil was a silty loam with a sand interlayer, generally appearing at a depth of 50-90 cm.The physicochemical properties of the soils are listed in Table 1 (USDA).Rainfall and average temperature data from the spring maize growth period are shown in Figure 1.

Experimental Design and Field Practices
The experiment was arranged in a split-plot design consisting of three irrigation levels (local conventional border irrigation) and three fertilisation levels and a control treatment.The experimental design for two years was exactly the same.The experiment was repeated three times on a sand interlayered soil land, with sand layers buried deeper than 40 cm (the average position of the sandy layer in this study area ranged from 50 to 90 cm).Based on the amount of local irrigation, a high-water treatment (indicated as I1) was set with a conventional irrigation depth of 270 mm, which was reduced by 20% and 40% for the medium-water (I2) and low-water (I3) treatments, respectively.Three ni-

Experimental Design and Field Practices
The experiment was arranged in a split-plot design consisting of three irrigation levels (local conventional border irrigation) and three fertilisation levels and a control treatment.The experimental design for two years was exactly the same.The experiment was repeated three times on a sand interlayered soil land, with sand layers buried deeper than 40 cm (the average position of the sandy layer in this study area ranged from 50 to 90 cm).Based on the amount of local irrigation, a high-water treatment (indicated as I1) was set with a conventional irrigation depth of 270 mm, which was reduced by 20% and 40% for the medium-water (I2) and low-water (I3) treatments, respectively.Three nitrogen levels were set based on a survey of nitrogen input by local farmers prior to the experiment.Conventional nitrogen application (375 kg N/ha) was set as the high-nitrogen treatment (F1) and was reduced by 20% and 40% for medium-nitrogen (F2) and low-nitrogen (F3) treatments, respectively.The control treatment (CK) consisted of conventional irrigation without nitrogen application.The experiment consisted of 10 treatments, with 3 replicates per treatment, making a total of 30 plots, distributed in blocks.Each plot was 9 m long and 5 m wide (area = 45 m 2 ), with the plots isolated using a 15 cm high ridge and a deep PVC plastic sheet to reduce the lateral movement of water and nutrients between plots.The spring maize variety Ximeng 3358 was planted in wide and narrow rows with a wide row spacing of 60 cm, a narrow row spacing of 40 cm, and plant spacing of 24 cm.At sowing, P 2 O 5 (150 kg/ha) and N (150 kg/ha) were applied as base fertilisers before mulching, with the phosphate fertiliser consisting of diammonium phosphate (containing 46% P 2 O 5 and 18% N) and urea (containing 46% N); the remaining nitrogen was applied in the form of urea at the jointing and tasselling stages.During the entire growth period, water was applied three times at critical growth stages, and the irrigation volume was recorded using a water metre.The specific experimental design scheme is shown in Table 2 (2016 and 2018), with the amount of applied nitrogen calculated as the quantity of pure nitrogen.Three maize root samples were collected at the maize grain filling stage using the soil profile grid method, and soil profiles were dug to obtain the maize root samples (Figure 2); I2F1, I2F2, I2F3, and CK were collected in 2016, and all treatments were collected in 2018.Starting from the centre of the wide row, a sampling space of 50 × 24 × 100 cm 3 perpendicular to the planting direction was established, and sampling continued down to 1 m or until no roots were present.Each root sampling space was 10 × 10 × 24 cm, with 28-32 root samples collected each time.After removing remnants, grass roots, and impurities, root samples were cleaned using a root washer, transferred to a transparent tray (10 cm × 20 cm) filled with 1-2 cm deep clear water, and separated using tweezers to reduce overlapping areas.A scanner (Epson Perfection V750 photo; Seiko Epson Corp., Nagano, Japan) was used to scan 5 million pixels and create black-and white-contrast TIF image files for analysis.The images were analysed using Win RHIZO 2008 software (Regent Instrument Inc., Québec, QC, Canada), and root length (RL), surface area (RS), and volume (RV) were determined.Root length density (RLD, cm/cm 3 ) was calculated as follows: where RLD is root length density, cm/cm 3 ; L is root length, cm; and V is soil volume, cm 3 .

Maize Yield Determination
Yield and yield traits were determined at the end of the maize wax-ripening stage.Four rows of maize (5 m in length from each harvested row) were harvested in each experimental plot, with three replicates per treatment.All maize samples were dried and threshed, and the total weight of the grains was measured.Yield per unit area was calculated based on the total grain weight of the sample plots.Ten maize plants were randomly selected to measure the ear length, grain rows per ear, grains per row, and 100-grain weight.Maize yield was measured at a grain moisture content of 14%, and the grain yield at maturity was calculated.

Maize Yield Determination
Yield and yield traits were determined at the end of the maize wax-ripening stage Four rows of maize (5 m in length from each harvested row) were harvested in each ex perimental plot, with three replicates per treatment.All maize samples were dried and threshed, and the total weight of the grains was measured.Yield per unit area was calcu lated based on the total grain weight of the sample plots.Ten maize plants were random ly selected to measure the ear length, grain rows per ear, grains per row, and 100-grain weight.Maize yield was measured at a grain moisture content of 14%, and the grain yield at maturity was calculated.

Water Use Efficiency
Water use efficiency (WUE) was calculated as previously described [31]: where WUE is water use efficiency, kg/ha•mm −1 ;  is maize yield, kg•ha −1 ; and ET is wate consumption, mm.

Statistical Analysis
Data were analysed using SPSS software 22 (IBM, Inc., Armonk, NY, USA).Multiple comparisons of mean values were performed using the least significant difference (LSD0.05).All figures were drawn using OriginLab 2021 (Northampton, MA, USA) and Surfer 13.Linear mixed-effects models were created using the Origin 2021 software.

Effect of Water and Fertilizer on Maize Root Parameters at the Grain Filling Stage
The root characteristics of spring maize grown in a typical sand interlayered soi under different water and fertiliser treatments are shown in Table 3.Based on the result of the 2-year experiment, the maximum root depth of the plants corresponded to the lo cation of the sand layer, indicating that the sand layer inhibited root growth.At the same irrigation level, each root parameter initially increased and then decreased with the in creasing N application rate (decreasing with N application in 2016 [24]) in both year

Water Use Efficiency
Water use efficiency (WUE) was calculated as previously described [31]: where WUE is water use efficiency, kg/ha•mm −1 ; Y is maize yield, kg•ha −1 ; and ET is water consumption, mm.

Statistical Analysis
Data were analysed using SPSS software 22 (IBM, Inc., Armonk, NY, USA).Multiple comparisons of mean values were performed using the least significant difference (LSD0.05).All figures were drawn using OriginLab 2021 (Northampton, MA, USA) and Surfer 13.Linear mixed-effects models were created using the Origin 2021 software.

Effect of Water and Fertilizer on Maize Root Parameters at the Grain Filling Stage
The root characteristics of spring maize grown in a typical sand interlayered soil under different water and fertiliser treatments are shown in Table 3.Based on the results of the 2-year experiment, the maximum root depth of the plants corresponded to the location of the sand layer, indicating that the sand layer inhibited root growth.At the same irrigation level, each root parameter initially increased and then decreased with the increasing N application rate (decreasing with N application in 2016 [24]) in both years (under the I2 treatment, RV increased with the N application rate (2018)).At the same nitrogen level, each root parameter initially increased and then decreased with increasing irrigation rate (under the F1 treatment, RL decreased with the irrigation rate; under F3, RV increased with the irrigation rate).Compared with that in the CK group, there was a significant improvement in root parameters in the treatment groups in both years.Irrigation and nitrogen significantly affected (p < 0.01) the root parameters, and there was a significant interaction (p < 0.01) between irrigation and nitrogen (2016 and 2018).Notably, the maximum RL and RS were achieved in the I2F2 treatment group (2016 and 2018), whereas the maximum RD was achieved in the I2F1 (2016), I2F1 and I1F3 (2018) treatment groups and RV was achieved in the I2F2 (2016) and I2F1 (2018) treatment groups.Compared with I2F2, RL and RS decreased by 9.94-73.72 and 4.75-68.10%.Compared with I2F2 (2016), I2F1 (2018)'s RV decreased by 6.16-77.07%,and compared with I2F1 (2016), I2F1 and I1F3 (2018)'s RD decreased by 4.17-25.00% in the various water-nitrogen treatments.Compared with that in the I1F1 group, RL, RS, RV, and RD increased by 101.61, 72.59, and 28.46%, respectively, in the I2F2 treatment.

Spatial Distribution of Roots at Different Horizontal Positions at the Grain Filling Stage
The spatial distribution characteristics of spring maize roots under different water and nitrogen conditions are shown in Table 4.At the grain filling stage, the total root length (TRL) of spring maize at four horizontal positions, 20 cm in the narrow row (NR20), 10 cm (WR10), 20 cm (WR20), and 30 cm (WR30) in the wide row, showed an initial increase with increasing irrigation and nitrogen application rates, followed by a decrease in both years.Notably, the highest TRL was recorded in the I2F2 treatment (2016 and 2018).At the same irrigation level, TRL at each horizontal position showed an initial increase with increasing N application rates, followed by a decrease in both years.In the I3 treatment group, TRL decreased at WR10 and WR20 and increased at WR30 (2018).At the same nitrogen application level, TRL at NR20 showed an initial increase with increasing irrigation rates, followed by a decrease; TRL showed an initial increase with increasing irrigation rates in the F2 treatment, followed by a decrease; TRL at WR10 decreased with increasing irrigation rates in the F3 treatment; TRL showed an initial decrease with increasing irrigation rates in the F1 treatment, followed by an increase; TRL showed an initial increase with increasing irrigation rate in the F2 treatment, followed by a decrease; TRL at WR20 decreased with increasing irrigation rates in the F3 treatment; In the F2 and F3 treatments, TRL at WR30 increased initially and then decreased with increasing irrigation rate (2018).
Compared with that in the I2F2 treatment (2016 and 2018), TRL decreased by 22.34-66.66,18.77-79.45,17.95-75.76,and 19.93-81.61% at NR20, WR10, WR20, and WR30, respectively, under other water-nitrogen treatments.Irrigation and fertilisation significantly affected root growth at different horizontal positions and had a significant interaction effect (2018) (p < 0.01).Compared with that in the I1F1 treatment, RL increased by 80.88-124.51% at each horizontal position in the I2F2 treatment (2018).The distribution of the roots varied at different horizontal positions, with approximately 37.16-52.50% of the RL occurring at NR20 (Figure 3).Additionally, the proportion of RL extending from the narrow row to the wide row decreased gradually, with 9.01-18.37% of the RL occurring at WR30.Collectively, these results indicated that irrigation and nitrogen application significant increased (p < 0.05) RL at each horizontal position (except for NR20) compared with that in the CK treatment group.The percentage of root length in the 0-100 cm soil layer under different waternitrogen treatments is shown in Figure 4.The proportion of roots in the 0-40 cm soil layer was approximately 63.01-91.00%,with the smallest proportion recorded in the CK treatment group.Compared with that in the CK group, the other treatments increased the proportion of roots in the 0-40 cm soil layer by 0.21-27.17%.Additionally, the percentage of root in the 40-100 cm was approximately 9.01-36.99%,peaking in the CK treatment.Compared with that in that CK group, the percentage of root in the 40-100 cm decreased by 1.16-46.28% in the other treatments.Notably, the proportion of RL in the bottom layer was 24.61-83.17%higher than that in the sub-bottom layer.In the CK treatment, the percentage of roots decreased gradually with increasing soil depth, with a decrease of 5.90-29.01%in the bottom layer compared with the sub-bottom layer.Compared with that in the I1F1 treatment, the percentage of roots in the 40-50 cm layer decreased by 23.24, 11.97, and 17.25% in the I1F1, I2F2, and I3F2 treatment groups, respectively.Overall, approximately 63.01-91.00% of spring maize roots were concentrated in the 0-40 cm soil layer in two years.Compared with that in the I1F1 treatment, RL increased by 7.55-10.52% in the 40-60 cm layer in the other treatments (decreasing by 4.33% in I2F3).
Taking the I2 treatment as an example, the cumulative percentage of roots increased gradually with an increase in soil depth in the vertical direction, and the vertical depth with a cumulative percentage of 100% varied slightly depending on the location of the sand layer (Figure 5).Overall, these results indicated the soil depths at which the cumulative root percentage reached 100% were 50, 60, 70, and 80 cm.When the cumulative The percentage of root length in the 0-100 cm soil layer under different water-nitrogen treatments is shown in Figure 4.
Agriculture 2024, 14, x FOR PEER REVIEW 8 of 18  The percentage of root length in the 0-100 cm soil layer under different waternitrogen treatments is shown in Figure 4.The proportion of roots in the 0-40 cm soil layer was approximately 63.01-91.00%,with the smallest proportion recorded in the CK treatment group.Compared with that in the CK group, the other treatments increased the proportion of roots in the 0-40 cm soil layer by 0.21-27.17%.Additionally, the percentage of root in the 40-100 cm was approximately 9.01-36.99%,peaking in the CK treatment.Compared with that in that CK group, the percentage of root in the 40-100 cm decreased by 1.16-46.28% in the other treatments.Notably, the proportion of RL in the bottom layer was 24.61-83.17%higher than that in the sub-bottom layer.In the CK treatment, the percentage of roots decreased gradually with increasing soil depth, with a decrease of 5.90-29.01%in the bottom layer compared with the sub-bottom layer.Compared with that in the I1F1 treatment, the percentage of roots in the 40-50 cm layer decreased by 23.24, 11.97, and 17.25% in the I1F1, I2F2, and I3F2 treatment groups, respectively.Overall, approximately 63.01-91.00% of spring maize roots were concentrated in the 0-40 cm soil layer in two years.Compared with that in the I1F1 treatment, RL increased by 7.55-10.52% in the 40-60 cm layer in the other treatments (decreasing by 4.33% in I2F3).
Taking the I2 treatment as an example, the cumulative percentage of roots increased gradually with an increase in soil depth in the vertical direction, and the vertical depth with a cumulative percentage of 100% varied slightly depending on the location of the sand layer (Figure 5).Overall, these results indicated the soil depths at which the cumulative root percentage reached 100% were 50, 60, 70, and 80 cm.When the cumulative The proportion of roots in the 0-40 cm soil layer was approximately 63.01-91.00%,with the smallest proportion recorded in the CK treatment group.Compared with that in the CK group, the other treatments increased the proportion of roots in the 0-40 cm soil layer by 0.21-27.17%.Additionally, the percentage of root in the 40-100 cm was approximately 9.01-36.99%,peaking in the CK treatment.Compared with that in that CK group, the percentage of root in the 40-100 cm decreased by 1.16-46.28% in the other treatments.Notably, the proportion of RL in the bottom layer was 24.61-83.17%higher than that in the sub-bottom layer.In the CK treatment, the percentage of roots decreased gradually with increasing soil depth, with a decrease of 5.90-29.01%in the bottom layer compared with the sub-bottom layer.Compared with that in the I1F1 treatment, the percentage of roots in the 40-50 cm layer decreased by 23.24, 11.97, and 17.25% in the I1F1, I2F2, and I3F2 treatment groups, respectively.Overall, approximately 63.01-91.00% of spring maize roots were concentrated in the 0-40 cm soil layer in two years.Compared with that in the I1F1 treatment, RL increased by 7.55-10.52% in the 40-60 cm layer in the other treatments (decreasing by 4.33% in I2F3).
Taking the I2 treatment as an example, the cumulative percentage of roots increased gradually with an increase in soil depth in the vertical direction, and the vertical depth with a cumulative percentage of 100% varied slightly depending on the location of the sand layer (Figure 5).Overall, these results indicated the soil depths at which the cumulative root percentage reached 100% were 50, 60, 70, and 80 cm.When the cumulative percentage reached 100%, the depth of the soil layer was equal to that of the sandy layer.Therefore, it can be concluded that the vertical growth of maize roots stopped at the sand interlayered soil, indicating the maximum root depth.
Agriculture 2024, 14, x FOR PEER REVIEW 9 of 18 percentage reached 100%, the depth of the soil layer was equal to that of the sandy layer.Therefore, it can be concluded that the vertical growth of maize roots stopped at the sand interlayered soil, indicating the maximum root depth.

Effects of Different Water-Saving and Nitrogen-Reducing Measures on Two-Dimensional RLD of Spring Maize
Irrigation and nitrogen application significantly affected root growth and distribution in 2018, which affected the distribution of RLD in different soil layers.The distribution of RLD under the I1F1, I1F2, I1F3, I2F1, I2F2, I2F3, I3F1, I3F2, I3F3, and CK treatments at the spring maize filling stage is shown in Figure 6.In the analysis of the spatial distribution of RLD, we defined 1 cm/cm 3 as the threshold of root length density, ≥1 cm/cm 3 as the dense area of RLD, and <1 cm/cm 3 as the dispersed area of RLD [32].As shown in Figure 6, the RLD at each horizontal position was >1 cm/cm 3 under different water and nitrogen treatments, which belonged to the dense area of the RLD.Moreover, the RLD in the wide-row position increased initially and then decreased with increasing irrigation and nitrogen application rates.
In the horizontal direction, the dense area of the RLD of spring maize was mainly distributed within a radius of 11.6-25 cm, approximately 5 cm from the wide row of the maize plant.The root was denser and had a higher RLD in the narrow row than in the wide row.As the horizontal position extended towards the wide row, the RLD of the roots gradually decreased.Notably, the horizontal width of the dense area of RLD was 36.6-50.0,39.1-50.0,and 37.5-49.6cm in the I1, I2, I3 treatment groups, respectively.Compared with that in the I1 and I3 groups, RLD increased by 10.06 and 6.47%, respectively, in the I2 treatment.Additionally, the horizontal width of the dense area of RLD was 37.5-39.1,49.6-50.0,and 36.6-50.0cm in the F1, F2, F3 treatment groups, respectively.Moreover, RLD increased by 23.06 and 13.36% in the F2 treatment compared with that in the F1 and F3 groups, respectively.In contrast, RLD was <1 cm/cm 3 in the CK treatment.Collectively, these results indicate that controlled reduction in irrigation and nitrogen fertilisation rates may increase the dense area in the horizontal direction and expand the horizontal distribution of roots, which is beneficial for water and nutrient absorption.

Effects of Different Water-Saving and Nitrogen-Reducing Measures on Two-Dimensional RLD of Spring Maize
Irrigation and nitrogen application significantly affected root growth and distribution in 2018, which affected the distribution of RLD in different soil layers.The distribution of RLD under the I1F1, I1F2, I1F3, I2F1, I2F2, I2F3, I3F1, I3F2, I3F3, and CK treatments at the spring maize filling stage is shown in Figure 6.In the analysis of the spatial distribution of RLD, we defined 1 cm/cm 3 as the threshold of root length density, ≥1 cm/cm 3 as the dense area of RLD, and <1 cm/cm 3 as the dispersed area of RLD [32].As shown in Figure 6, the RLD at each horizontal position was >1 cm/cm 3 under different water and nitrogen treatments, which belonged to the dense area of the RLD.Moreover, the RLD in the wide-row position increased initially and then decreased with increasing irrigation and nitrogen application rates.
In the horizontal direction, the dense area of the RLD of spring maize was mainly distributed within a radius of 11.6-25 cm, approximately 5 cm from the wide row of the maize plant.The root was denser and had a higher RLD in the narrow row than in the wide row.As the horizontal position extended towards the wide row, the RLD of the roots gradually decreased.Notably, the horizontal width of the dense area of RLD was 36.6-50.0,39.1-50.0,and 37.5-49.6cm in the I1, I2, I3 treatment groups, respectively.Compared with that in the I1 and I3 groups, RLD increased by 10.06 and 6.47%, respectively, in the I2 treatment.Additionally, the horizontal width of the dense area of RLD was 37.5-39.1,49.6-50.0,and 36.6-50.0cm in the F1, F2, F3 treatment groups, respectively.Moreover, RLD increased by 23.06 and 13.36% in the F2 treatment compared with that in the F1 and F3 groups, respectively.In contrast, RLD was <1 cm/cm 3 in the CK treatment.Collectively, these results indicate that controlled reduction in irrigation and nitrogen fertilisation rates may increase the dense area in the horizontal direction and expand the horizontal distribution of roots, which is beneficial for water and nutrient absorption.In the vertical direction, the roots were mainly concentrated in the 0-40 cm soil layer, with fewer roots in layers deeper than 40 cm.Notably, the vertical depth of the dense area of RLD was 13.9-26.9,19.In the vertical direction, the roots were mainly concentrated in the 0-40 cm soil layer, with fewer roots in layers deeper than 40 cm.Notably, the vertical depth of the dense area of RLD was 13.9-26.9,19.3-38.4,and 20.0-25.3cm in the I1, I2, and I3 treatment groups, respectively.Additionally, RLD increased by 23.43 and 21.51% in the I2 treatment compared with that in the I1 and I3 groups, respectively.Moreover, the vertical depth of the dense area of RLD was 19.3-22.9,20.0-38.4,and 13.9-25.5cm in the F1, F2, and F3 treatment groups, respectively.Furthermore, RLD increased by 27.08 and 24.15% in the F2 group compared with that in F1 and F3 groups, respectively.In contrast, RLD was <1 cm/cm 3 in the CK treatment, and there was no dense area of RLD.Overall, these results indicate that appropriate irrigation and nitrogen fertilisation rates not only promote the downward shift of the dense area of the root system but also reduce nitrogen wastage and the risk of on-point source pollution in farmlands.An optimal decrease in irrigation and nitrogen fertilisation may promote the formation of a 'wide and deep' root configuration in spring maize, which may improve water and nutrient absorption within a limited root depth to increase yield.

Response of Yield and WUE of Spring Maize to Varying Irrigation and Fertilisation Rates
Maize yield showed an initial increase with increasing irrigation and nitrogen application rates, followed by a decrease (Figure 7).At the same irrigation level, the yield showed an initial increase with increasing application rates, followed by a decrease, except for the I3 treatment (increasing with the nitrogen application rate).At the same nitrogen level, the yield showed an initial increase with increasing irrigation rates, followed by a decrease, except in the F1 treatment (increasing with the increasing irrigation rate).Irrigation and nitrogen application significantly affected (p < 0.01) yield, and there was a significant interaction (p < 0.01) between irrigation and fertilisation rates.Notably, the highest yield was obtained in the I2F2 treatment, and was significantly higher (p < 0.05) than that in the other treatments.Compared with that in the I2F2 and I1F1 treatment, yield decreased significantly by 6.90-38.83and 4.20-32.51%,respectively, in the other treatments.Compared with that in the I1F1 treatment, yield decreased slightly by 4.20-8.30% in the I1 and I2 groups and decreased significantly by 14.60-32.50% in the I3 and CK groups.Overall, these results indicate that an appropriate decrease in irrigation and fertilisation rates may increase maize yield; however, extremely low rates may inhibit the growth of spring maize.Notably, irrigation and fertilisation at 270 mm and 375 kg N/ha, respectively, can reduce water and N use by 20% without negatively affecting yield.
Agriculture 2024, 14, x FOR PEER REVIEW 11 of 18 ment compared with that in the I1 and I3 groups, respectively.Moreover, the vertical depth of the dense area of RLD was 19.3-22.9,20.0-38.4,and 13.9-25.5cm in the F1, F2, and F3 treatment groups, respectively.Furthermore, RLD increased by 27.08 and 24.15% in the F2 group compared with that in F1 and F3 groups, respectively.In contrast, RLD was <1 cm/cm 3 in the CK treatment, and there was no dense area of RLD.Overall, these results indicate that appropriate irrigation and nitrogen fertilisation rates not only promote the downward shift of the dense area of the root system but also reduce nitrogen wastage and the risk of on-point source pollution in farmlands.An optimal decrease in irrigation and nitrogen fertilisation may promote the formation of a 'wide and deep' root configuration in spring maize, which may improve water and nutrient absorption within a limited root depth to increase yield.

Response of Yield and WUE of Spring Maize to Varying Irrigation and Fertilisation Rates
Maize yield showed an initial increase with increasing irrigation and nitrogen application rates, followed by a decrease (Figure 7).At the same irrigation level, the yield showed an initial increase with increasing application rates, followed by a decrease, except for the I3 treatment (increasing with the nitrogen application rate).At the same nitrogen level, the yield showed an initial increase with increasing irrigation rates, followed by a decrease, except in the F1 treatment (increasing with the increasing irrigation rate).Irrigation and nitrogen application significantly affected (p < 0.01) yield, and there was a significant interaction (p < 0.01) between irrigation and fertilisation rates.Notably, the highest yield was obtained in the I2F2 treatment, and was significantly higher (p < 0.05) than that in the other treatments.Compared with that in the I2F2 and I1F1 treatment, yield decreased significantly by 6.90-38.83and 4.20-32.51%,respectively, in the other treatments.Compared with that in the I1F1 treatment, yield decreased slightly by 4.20-8.30% in the I1 and I2 groups and decreased significantly by 14.60-32.50% in the I3 and CK groups.Overall, these results indicate that an appropriate decrease in irrigation and fertilisation rates may increase maize yield; however, extremely low rates may inhibit the growth of spring maize.Notably, irrigation and fertilisation at 270 mm and 375 kg N/ha, respectively, can reduce water and N use by 20% without negatively affecting yield.The variation in WUE was similar to that in yield.At the same irrigation level, WUE increased initially and then decreased with increasing fertilisation rate (I1 and I2), or increased with increasing nitrogen application rates (I3).At the same nitrogen level, WUE initially increased and then decreased with increasing irrigation rates.Irrigation and nitrogen application significantly affected (p < 0.01) WUE, and there was a significant interaction (p < 0.01) between irrigation and fertilisation rates.Notably, the highest WUE was recorded in the I2F2 treatment.Compared with the I2F2 treatment, WUE decreased The variation in WUE was similar to that in yield.At the same irrigation level, WUE increased initially and then decreased with increasing fertilisation rate (I1 and I2), or increased with increasing nitrogen application rates (I3).At the same nitrogen level, WUE initially increased and then decreased with increasing irrigation rates.Irrigation and nitrogen application significantly affected (p < 0.01) WUE, and there was a significant interaction (p < 0.01) between irrigation and fertilisation rates.Notably, the highest WUE was recorded in the I2F2 treatment.Compared with the I2F2 treatment, WUE decreased by 8.88-33.58% in the other treatments.Additionally, WUE increased by 17.31% in the I2F2 treatment and decreased by 22.09% in the CK treatment compared with that in the I1F1 treatment.WUE increased by 25.2-17.31%(I1 and I2), decreased by 5.50-7.40%(I3), and decreased significantly by 22.09% (CK).Collectively, these results indicate that the WUE of maize can be improved by appropriately reducing irrigation and fertilisation rates.Based on these results, irrigation and fertilisation at 270 mm and 375 kg N/a, respectively, can decrease irrigation water and N use by 20% and increase WUE by 17.31%.

Correlation between Root Parameters of Spring Maize, Yield, and WUE
Root growth directly affects the absorption of water and nutrients by plants, which in turn affects yield.To study the correlation between spring maize yield, WUE, root parameters, horizontal root length, and root length in different soil layers under different water and fertiliser supply conditions, a linear correlation analysis was conducted (Figure 8).Maize yield was significantly correlated (p < 0.01) with root parameters, and the correlation coefficients were in the order of RS > RV > RD > RL.These results indicate that RS is a better indicator of spring maize yield in a sand interlayered soil under different water and fertilisation rates.Additionally, WUE was significantly correlated with root parameters (p < 0.01), and the correlation coefficients were in the order of RS > RL > RD > RV.This result indicates that RS is more closely related to WUE than to the other factors.by 8.88-33.58% in the other treatments.Additionally, WUE increased by 17.31% in the I2F2 treatment and decreased by 22.09% in the CK treatment compared with that in the I1F1 treatment.WUE increased by 25.2-17.31%(I1 and I2), decreased by 5.50-7.40%(I3), and decreased significantly by 22.09% (CK).Collectively, these results indicate that the WUE of maize can be improved by appropriately reducing irrigation and fertilisation rates.Based on these results, irrigation and fertilisation at 270 mm and 375 kg N/a, respectively, can decrease irrigation water and N use by 20% and increase WUE by 17.31%.

Correlation between Root Parameters of Spring Maize, Yield, and WUE
Root growth directly affects the absorption of water and nutrients by plants, which in turn affects yield.To study the correlation between spring maize yield, WUE, root parameters, horizontal root length, and root length in different soil layers under different water and fertiliser supply conditions, a linear correlation analysis was conducted (Figure 8).Maize yield was significantly correlated (p < 0.01) with root parameters, and the correlation coefficients were in the order of RS > RV > RD > RL.These results indicate that RS is a better indicator of spring maize yield in a sand interlayered soil under different water and fertilisation rates.Additionally, WUE was significantly correlated with root parameters (p < 0.01), and the correlation coefficients were in the order of RS > RL > RD > RV.This result indicates that RS is more closely related to WUE than to the other factors.Moreover, maize yield was significantly correlated (p < 0.01) with root length at different positions in the horizontal direction, and the correlation coefficient was in the order of NR20 > WR10 > WR20 > WR30.WUE was significantly correlated (p < 0.01) with Moreover, maize yield was significantly correlated (p < 0.01) with root length at different positions in the horizontal direction, and the correlation coefficient was in the order of NR20 > WR10 > WR20 > WR30.WUE was significantly correlated (p < 0.01) with horizontal root length, and the correlation coefficients were in the order of NR20 > WR20 > WR30 > WR10.Overall, this result indicates that increasing the root length at NR20 is beneficial for improving maize yield and WUE.
In the vertical direction, RL in the 10-20 cm soil layer was strongly correlated with yield and WUE (r = 0.847, p < 0.01).In contrast, RL in the 40-50 cm soil layer was not significantly correlated with yield and WUE.Collectively, these results suggest WUE and maize yield can be improved by increasing RL and RS.

Discussion
The root is an important part of crop growth and development, directly affecting the absorption of water and nutrients and indirectly impacting crop productivity [33].During soil science experiments, the depth of soil sampling depends on the ecosystem [34].Generally, the sampling depth in the field ranges from 50-200 cm for studies on the water absorption depth of plant roots [34,35].Root growth depends on soil, and is significantly affected by changes in soil properties [36].Notably, the maximum root depth of maize is approximately 1 m [28,37]; owing to the unique characteristics of sand interlayered soils, maize roots grow downward and stop at the sand layer.In the present study, the maximum depth of the root was equivalent to the buried depth of the sand layer.Irrigation and fertilisation not only regulate root growth and development but also affect the vertical distribution of roots in the soil layer [25].Root growth was inhibited at the sand layer, indicating that a sand layer thickness > 15 cm (the upper and lower layers boundaries of the sand layer) may suppress root growth.
In the present study, the characteristic parameters of spring maize roots decreased with increasing soil depth in a sand interlayered soil under different water and fertiliser conditions [28].During the 2-year growing season, the percentage of roots decreased gradually with increasing soil depth under the different irrigation and nitrogen application rates.Owing to the tendency of roots to grow towards water and nutrients, root growth ceased in the sand layer, with horizontal growth occurring in the bottom layer, leading to root accumulation.A controlled decrease in irrigation and nitrogen application rates may increase root development in the 40-100 cm soil layer, extending the root into deeper soil profiles, which is beneficial for the absorption and utilisation of deep soil water and nutrients.Additionally, the proportion of RL in the bottom layer was higher than that in the sub-bottom layer by 24.61-83.17%under the different water and nitrogen treatments.This is because the discontinuity of soil hydraulic properties in the sand layer at the soil-sand interface limits the downward movement of soil water and the migration of nutrients, resulting in an increase in soil water storage capacity and an improvement in ecosystem productivity in the upper sand layer [38].The direction of crop root growth is directly affected by environmental stress [39].Soil layers with high levels of water and fertiliser, and low levels of salt are favourable for root growth; therefore, the horizontal growth of plants roots at the soil-sand interface may result in root accumulation.For example, approximately 50-80% of maize roots occupy the 0-20 cm soil layer in farmlands [40].In the present study, 35.77-64.03% of the roots were distributed in the 0-20 cm soil layer.Soil nutrients and water uptake by crops is dependent on the temporal and spatial distribution of roots [41][42][43][44].
Normal root morphology is the basis for nutrient and water absorption by plants, which promotes a high crop yield [45]. Excessively high or extremely low soil moisture can change the size, number, and distribution of crop roots, thereby affecting the canopy growth and grain yield [46]. Notably, crop yield is closely related to the extent of root development [47].Therefore, the spatial distribution of roots affect maize yield considerably [48].Increasing the number of plant roots enhances the absorption of water and nutrients [49].Controlled reduction in irrigation and fertilisation rates may promote root growth, whereas excessively high irrigation inhibits root growth, leading to water wastage.Additionally, excessive nitrogen application leads to nutrient wastage and toxicity, which can suppress crop growth.In contrast, extremely low irrigation and nitrogen application levels may result in water and nutrient stress, inhibiting root growth.In the present study, the I2F2 treatment was the most suitable water-nitrogen combination for root growth, as it promoted vertical and horizontal root growth and distribution.Vertical and horizontal root growth is necessary for optimal soil moisture and nutrient absorption, resulting in improved WUE and yield.Therefore, appropriately reducing irrigation and nitrogen appli-cation can effectively increase root growth, facilitate efficient water and nutrient utilisation, and enhance maize yield.
Water and fertilisers are the main factors that affect maize yield [50]. Optimal water and nitrogen supply has a considerable effect on crop growth and yield [51][52][53].However, excessive fertilisation can cause crop overgrowth and even toxicity, which is detrimental to yield [54,55].In the present study, irrigation and fertilisation significantly promoted yield and WUE, with significant interactive effects.Notably, maize yield showed an initial increase with increasing irrigation and fertilization rates, followed by a decrease.The relationships among irrigation, nitrogen application, and yield were similar to those reported by Xu et al. [54][55][56].Contrary to a previous study that reported maximum yield in a high-irrigation and medium-nitrogen treatment, WUE and yield peaked in the I2F2 treatment (medium irrigation and fertilisation) in the present study.The discrepancies in results may be due to differences in soil conditions, irrigation volume, fertilisation rate, and rainfall distribution in the experimental area.Overall, these results indicate that appropriately reducing irrigation and nitrogen application effectively improves yield and WUE, whereas excessive irrigation and nitrogen lead to resource wastage.
Yin et al. [25] analysed the correlation between root distribution and the yield of summer maize under different nitrogen fertiliser regimes and found a significant linear relationship between RL, RS, and yield.Additionally, Zou et al. [24] reported that root characteristics during the grain filling stage had the highest correlation (p < 0.01) with yield and aboveground biomass.Moreover, Zhou et al. [24,57] reported a significant correlation between root parameters and maize yield at the mature stage under different planting patterns.Similarly, yield was significantly correlated (p < 0.01) with root parameters at the grain filling stage in the present study, and the correlation coefficients were in the order of RS > RV > RD > RL.Overall, this result indicates that RS is a better indicator of spring maize yield than other root parameters in sand interlayered soils under different water and fertiliser conditions.Zou et al. [24] found that RL was more closely related to yield than RS, which may be due to differences in soil texture.In the study by Zou, the maximum rooting depth of maize roots at the grain filling stage was 100 cm, whereas the maximum rooting depth was limited to the depth of the sand layer in the present study.Generally, roots increase their surface area to absorb water and nutrients to meet the normal growth needs of crops [58][59][60][61].In this study, we found that RL in the horizontal and vertical directions was significantly correlated with yield and WUE, with the highest correlation observed in NR20.Compared with that in the other vertical soil layers, RL in the 10-20 cm soil layer had a better correlation with yield and WUE (r = 0.847, p < 0.01); in contrast, RL in the 40-50 cm layer showed little or no correlation with yield and WUE.This may be because capillary water remained in the sand layer when the fine sand layer exceeded 35 cm, thus interrupting the hydraulic connection between the layers above and below the sand layer [62][63][64].Given that roots cannot effectively absorb soil moisture below the sand layer, the growth of deep roots may not significantly increase maize yield.
Therefore, ensuring appropriate water and nutrient supply is the key to ensuring optimal root growth.Considering the high adjustability of spring maize root distribution [65,66] under water-saving and nitrogen-limiting conditions, appropriately reducing irrigation and nitrogen input can effectively promote reasonable root distribution to improve yield and WUE.
Although we have studied the effect of water and fertilizer on the growth of maize roots on typical sand interlayered soil land and obtained preliminary results, the experiment only divided the depth of sand layer in the range.We found that root growth was inhibited at the sand layer; however, relevant research needs to be further refined.In subsequent experiments, we can set several levels of sand layers and use sand layer thickness treatments to further clarify the effect of sand interlayered soil on crop growth and irrigation schedules; in doing so, we may find an explanation for the threshold of sand layer thickness, which limits downward root growth.This will provide a theoretical basis for water and fertilizer use efficiency and the growth of crop roots in typical sand interlayered soil farmland.

Conclusions
In this study, we examined the effect of different irrigation and fertilisation rates on the yield and root parameters of spring maize grown in a sand interlayered soil.The root growth of spring maize stopped at the sand layer, and the maximum rooting depth coincided with the depth of the sand layer.Approximately 72.46-87.37% of the roots were concentrated in the 0-40 cm soil layer, and the percentage of roots in the bottom layer was 24.61-83.17%higher than that in the sub-bottom layer.Moreover, controlled reduction in irrigation and nitrogen fertilisation rates may promote the formation of 'wide and deep' root configurations.Furthermore, the root parameters of spring maize were significantly positively correlated with yield and WUE, with RS being the most correlated with yield and WUE.Roots at NR20 and a soil depth of 10-20 cm were most closely related to yield and WUE.Therefore, increasing the entire root surface area and roots in NR20 and 0-30 cm layer is necessary to increase yield and WUE.Notably, maximum yield and WUE were achieved in the I2F2 treatment.Considering factors such as root growth, yield, and water shortage, the optimal irrigation and nitrogen fertilisation rates in the study area were 216 mm and 300 kg/ha, respectively.

Figure 1 .
Figure 1.Precipitation and temperature at the experimental site in 2016 and 2018.

Figure 1 .
Figure 1.Precipitation and temperature at the experimental site in 2016 and 2018.

Figure 2 .
Figure 2. Spatial layout of root sampling using the quadrate monolith method.

Figure 2 .
Figure 2. Spatial layout of root sampling using the quadrate monolith method.

Figure 3 .
Figure 3. Root length ratio of spring maize at the filling stage in different horizontal positions undergoing different fertilisation treatments in 2016 and 2018.

Figure 4 .
Figure 4. Root length ratio of spring maize at filling stage at each sub-layer in 0-100 cm soil layer under different fertilisation treatments in 2016 and 2018.

Figure 3 .
Figure 3. Root length ratio of spring maize at the filling stage in different horizontal positions undergoing different fertilisation treatments in 2016 and 2018.

Figure 3 .
Figure 3. Root length ratio of spring maize at the filling stage in different horizontal positions undergoing different fertilisation treatments in 2016 and 2018.

Figure 4 .
Figure 4. Root length ratio of spring maize at filling stage at each sub-layer in 0-100 cm soil layer under different fertilisation treatments in 2016 and 2018.

Figure 4 .
Figure 4. Root length ratio of spring maize at filling stage at each sub-layer in 0-100 cm soil layer under different fertilisation treatments in 2016 and 2018.

Figure 5 .
Figure 5. Cumulative percentage of vertical root length of spring maize during the filling period under different nitrogen application and irrigation rates in 2016 and 2018.Colored dotted lines represent the soil depth at which the cumulative curve reaches its maximum.

Figure 5 .
Figure 5. Cumulative percentage of vertical root length of spring maize during the filling period under different nitrogen application and irrigation rates in 2016 and 2018.Colored dotted lines represent the soil depth at which the cumulative curve reaches its maximum.

Figure 6 .
Figure 6.Distribution of root length density in border irrigation grouting period of different treatments in 2018.

Figure 6 .
Figure 6.Distribution of root length density in border irrigation grouting period of different treatments in 2018.

1 Figure 7 .
Figure 7. Yield and WUE of spring maize under different irrigation and fertilization treatments in 2016 and 2018.Note: Different lowercase letters represent significant differences at the p = 5% level.

Figure 7 .
Figure 7. Yield and WUE of spring maize under different irrigation and fertilization treatments in 2016 and 2018.Note: Different lowercase letters represent significant differences at the p = 5% level.

Figure 8 .
Figure 8. Correlation coefficient of root characteristic parameters with yield and WUE in the spring maize filling stage.(a) Correlation coefficients of whole-root parameters with yield and WUE.(b) Correlation coefficients of root length with yield and WUE in different horizontal positions.(c) Correlation coefficients of root length with yield and WUE in the different soil layers.Note: The size and color of the image represent the magnitude of the correlation coefficient.

Figure 8 .
Figure 8. Correlation coefficient of root characteristic parameters with yield and WUE in the spring maize filling stage.(a) Correlation coefficients of whole-root parameters with yield and WUE.(b) Correlation coefficients of root length with yield and WUE in different horizontal positions.(c) Correlation coefficients of root length with yield and WUE in the different soil layers.Note: The size and color of the image represent the magnitude of the correlation coefficient.

Table 1 .
Physical properties of typical sand interlayered soil.
Note: The position of the sand layer in the study area varied from 50-90 cm, and the representative layer for soil physical property analysis was selected from layers indicated in the table.

Table 1 .
Physical properties of typical sand interlayered soil.The position of the sand layer in the study area varied from 50-90 cm, and the representative layer for soil physical property analysis was selected from layers indicated in the table.

Table 2 .
Experimental design of water and nitrogen in the sand interlayered soil in 2016 and 2018.

Table 3 .
Root characteristics of spring maize at the filling stage under different irrigation and fertilization treatments.

Table 4 .
Root length at horizontal position under different irrigation and fertilization conditions cm.