Differences in Water Consumption and Yield Characteristics among Winter Wheat (Triticum aestivum L.) Varieties under Different Irrigation Systems

Abstract

To discuss the water consumption patterns of winter wheat (Triticum aestivum L.) and the difference in yield traits among varieties under different irrigation systems, three field water treatments were established (jointing water, W1, CK; jointing water + flowering water, W2; and rising water + booting water + filling water, W3). Two winter wheat varieties (Hengmai4399 and Hengguan35) were selected in 2020–2021, and three (Hengmai4399, Hengguan35, and Hengmai28) were selected in 2021–2022 to investigate the impact of the irrigation amount on water consumption and its interaction with the varieties on yield traits of winter wheat. The results showed that there was a positive and significant correlation between soil water consumption and soil moisture reserve presowing; the correlation was the strongest in the 150–200 cm layer. The response of the yield of the three varieties to irrigation was Hengmai4399 > Hengmai28 > Hengguan35, and the drought resistance was Hengguan35 > Hengmai28 > Hengmai4399. When the soil water storage presowing was insufficient, Hengmai4399 combined with the W3 treatment achieved the highest yield and water use efficiency; when the soil water storage presowing was sufficient, Hengmai28 combined with the W2 treatment achieved a high-level yield and the highest WUE.

1. Introduction

At present, water shortage is one of the key concerns in China, and the uneven distribution of water resources is one of the main causes of regional water shortages in China. The current water resource situation in China is generally “flooding in the south and drought in the north”. Agriculture accounts for 61.50% of China’s total water consumption [1]. Grain produced in the North China Plain (NCP) accounts for a large proportion of China’s grain production [2]. Winter wheat (Triticum aestivum L.) is a main food crop in the NCP [3]. Due to the influence of monsoons, the coupling of precipitation distribution and water demand of winter wheat in this region is poor, and the precipitation during the reproductive period is less than 40% of its water demand [4]. Therefore, a large amount of groundwater is consumed for winter wheat planting in the area, and the groundwater has been seriously overexploited. When the irrigation amount reaches a certain level, the winter wheat yield and the water use efficiency (WUE) will decrease [5]. Thus, irrigation systems need to be optimized to reduce groundwater extraction, improve WUE, and maintain stable winter wheat yields.

Winter wheat yield does not increase linearly with the increasing amount of irrigation, and the yield is influenced by the amount of rainfall and growth stage distribution [6]. By optimizing the irrigation system, the water consumption of 40.00–50.00 mm during the whole winter wheat growing period can be reduced, thereby ensuring food security in the NCP, but the annual water consumption is still greater than the groundwater recharge [7,8]. On this basis, researchers supplemented soil water storage to the field capacity of the upper soil layer of root zone presowing; then, the annual water consumption was reduced by 112.00 mm while the yield was maintained at 86.00% without irrigation throughout the winter wheat growth period [9]. One of the main reasons why modern winter wheat varieties are more resistant to drought than earlier varieties is their greater ability to utilize soil water [10]. Thus, ensuring sufficient presowing soil water storage under water-limited irrigation is an important condition for maintaining high yield of winter wheat [11].

Winter wheat growth under dry land conditions is more dependent on deep soil water storage during the filling period [12], and it has been shown that sufficient water supply during the filling period can enhance the greenness of wheat leaves and thus promote the dry matter accumulation of winter wheat [13]. Roots of winter wheat show a very positive response to irrigation, with the near-surface root system playing a major role, and light high-frequency irrigation can reduce deep seepage and ensure that the wheat is not influenced by water deficit while promoting the development of shallow roots, thus improving the WUE of wheat and achieving efficient water use [14,15,16]. Li et al. [17] found that when the irrigation amount is 120 mm, 60 mm irrigation at the jointing stage and booting stage can improve the yield and WUE of wheat significantly. However, more frequent irrigation will increase soil evaporation; thus, balancing single irrigation volume and irrigation frequency under a certain amount of irrigation is important to improve the WUE and maintain winter wheat yield [18,19].

Considering the current situation of water resource security and food security in the NCP, the irrigation amount of winter wheat should be increased to more than the minimum irrigation amount for sustainable food production but less than the current irrigation amount that causes the rapid decline in groundwater level. Among the existing studies, individual studies on the water consumption characteristics of different wheat varieties and the effect of irrigation on yield and the WUE of wheat have been frequent, but few studies have been reported to combine winter wheat varieties with irrigation systems to investigate the response to irrigation of different winter wheat varieties under different irrigation systems.

Additionally, although researchers in previous studies have concluded that adequate presowing soil water storage is an important factor for high wheat yield [11], they have not explored in depth how presowing soil water storage affects the water consumption characteristics of winter wheat in different soil layers. Thus, for this study, three winter wheat varieties were selected, and three water treatments were set up to explore the differences in water consumption rules and yield traits among varieties under different irrigation systems. In addition, due to the influence of the previous year’s experiment, the presowing soil moisture storage of W3 treatment in 2020–2021 differed significantly from the other treatments [20]; therefore, this study was able to complement the effect of presowing soil moisture storage on the spatial water consumption characteristics in winter wheat.

2. Materials and Methods

2.1. Test Site Overview

The experiments were conducted at the dryland water-saving experimental station of the Hebei Academy of Agricultural and Forestry Sciences (37°44′ N, 115°47′ E) from October 2020 to June 2021 and from October 2021 to June 2022. In the two-year experiment, the rainfall and temperature during the winter wheat growth period are shown in Figure 1. The test soil was a clay loam with a soil bulk weight of 1.40 g cm⁻³. At 0–20 cm, the soil organic matter, total nitrogen, available nitrogen, available phosphorus, available potassium, and soil pH were 24.50 g kg⁻¹, 1.52 g kg⁻¹, 106.58 g kg⁻¹, 28.62 g kg⁻¹, 187.20 g kg⁻¹, and 8.01, respectively.

2.2. Test Design

The experiment used a split-zone design, and three irrigation treatments were set up for the main treatment (one irrigation event in spring, jointing water, W1, CK; two irrigation events in spring, jointing water + flowering water, W2; and three irrigation events in spring, rising water + booting water + filling water, W3, 75 mm per irrigation, where normal irrigation treatment W1 was used as a control). Two winter wheat varieties (Hengmai4399 and Hengguan35) were set up in 2020–2021, with six treatments and three replications for each treatment, with 18 test plots in total; the winter wheat variety Hengmai28 was added in 2021–2022, with nine treatments and three replications each. The area of each individual plot was 67.50 m², with ridging between the plots.

2.3. Field Management Measures

Straw was returned to the field after the harvest of the previous corn crop. The NPK compound fertilizer of 750.00 kg hm⁻² (N:P:K = 15:23:10) was used as a bottom fertilizer. Urea was applied in spring at 375.00 kg hm⁻², all in combination with the first spring irrigation. The two-year trial was sown on 15 October 2020 and on 24 October 2021 (presowing with deep water) at a seeding quantity of 210.00 kg hm⁻². Other management practices were completely consistent, with regular weeding during winter wheat growth, and there were no diseases or insect infestations. The winter wheat was harvested on 7 June 2021 and 9 June 2022, respectively.

2.4. Measurement Indexes and Methods

The experiment in 2020–2021 was affected by COVID-19 and lacked indicators such as the SPAD value of flag leaves, spike traits, effective spikes, and grains per spike.

2.4.1. Precipitation and Average Temperature

Precipitation and daily average temperature during the winter wheat growth period were provided by the dryland water-saving experimental station of the Hebei Academy of Agriculture and Forestry Sciences.

2.4.2. Soil Moisture Content

Soil samples were collected every 10 d during the key growth period of winter wheat using a soil auger in layers of 0–200 cm soil profile, with one layer every 10 cm; tests were added before sowing, before wintering, and after harvest. The soil moisture content was established using the drying method and checked with intelligent moisture monitoring data.

2.4.3. Water Consumption (ET) and the WUE

The water balance equation was used to calculate the water consumption of winter wheat during the growth period. The equation is as follows [21]:

ET = P + I + S + ΔW − R − D,

(1)

where ET is the crop water consumption (mm) during the period; P is the precipitation during the period (mm); I is the irrigation amount during the period (mm); S is the amount of groundwater recharge to crop roots during the period (mm); ΔW is the variation in soil water storage between 0–200 cm during the period; R is the surface runoff (mm) of the measured area during the period; and D is the amount of deep leakage in the root zone during the period (mm). The experimental area is flat and has ridges without surface runoff loss; the groundwater depth is greater than 5 m; the soil depth selected for calculating soil water storage is deep (200 cm), and there is no intense precipitation during the growth of winter wheat, so no deep seepage is generated. Thus, the equation can be simplified as:

ET = P + I + ΔW

(2)

The WUE was calculated according to the following equation [22]:

WUE = Y/10ET

(3)

where the WUE is the water use efficiency (kg m⁻³); ET is the crop water consumption (mm); and Y is the wheat grain yield (kg hm⁻²).

2.4.4. Flag Leaf SPAD Value

SPAD values were measured using a SPAD-502 (Minolta, Tokyo, Japan) chlorophyll meter at the jointing, booting, flowering, and filling stages, and 10 representative plants of uniform growth were selected from each plot for measurement.

2.4.5. Seed Testing and Yield Measurement

In the mature period, 40 wheat plants with the same growth vigor were randomly selected from each plot to determine their spike length, spikelet numbers, infertile spikelet numbers, and grains per spike. Three uniformly grown 1 m² samples were randomly cut from each plot. After threshing and air drying, 1000 grains (three portions) were randomly selected, and the grain moisture content was determined using the drying method. The thousand-grain weight and yield per unit area (kg hm⁻²) were converted according to 13% safe water content. The effective spikes were calculated by counting two one-meter double rows in each plot before harvesting.

2.5. Data Processing

Experimental data were processed using Excel 2019, and IBM SPSS Statistics for Windows, Version 26.0. (IBM Corp: Armonk, NY, USA) was used for ANOVA, correlation analysis, and regression analysis. Duncan’s multiple range tests were used to compare the differences between treatments with a significance level of p < 0.05.

3. Results

3.1. Dynamic Changes in the Soil Moisture Content of Winter Wheat under Different Irrigation Treatments

Dynamic changes in the soil moisture content in different growing periods of winter wheat during the two years experiments are shown in Figure 2. Overall, the water content of soil below 140 cm fluctuated less from 2020 to 2021 than that from 2021 to 2022. After irrigation, the change in soil water content in the 0–60 cm soil layer for each treatment in the two years trials was greater than that in the 60–200 cm soil layer. Before the jointing stage of each treatment, the soil profile had a high water content and uniform distribution; with the advancement of the growth process, the soil moisture content reduced gradually. The soil moisture content decreased faster after the jointing stage than before the jointing stage.

Due to the lack of soil water supplement after the booting stage of the W1 treatment, the soil water content decreased sharply in the late growing stage of winter wheat from 2021 to 2022 (Figure 2d). However, the soil water content under 140 cm was not significantly reduced from 2020 to 2021 (Figure 2a). W2 and W3 treatments were irrigated after the booting stage to supplement the soil moisture content, so the soil water content under the 100 cm soil layer changed little during the whole growth period. In the two-year experiment, the decrease of soil moisture content in the 0–100 cm soil layer was greater than that under the 100 cm soil layer.

3.2. Water Consumption of Different Soil Layers in the Winter Wheat Growth Period under Different Irrigation Systems

From

Table 1. Soil water consumption in different soil layers during the whole growth period of winter wheat.

Year	Treatments	Soil Depth (cm)
		0–50		50–100		100–150		150–200
		Values (mm)	Rate (%)	Values (mm)	Rate (%)	Values (mm)	Rate (%)	Values (mm)	Rate (%)
2020–2021	W1	102.61 ± 5.36 a	43.13 ± 0.78 b	92.27 ± 3.27 a	38.80 ± 0.59 a	37.38 ± 2.46 a	15.71 ± 0.66 a	5.64 ± 1.06 a	2.36 ± 0.37 a
	W2	108.15 ± 6.15 a	54.84 ± 0.49 a	65.99 ± 4.44 b	33.45 ± 0.47 b	18.18 ± 1.37 b	9.21 ± 0.14 b	4.94 ± 0.36 b	2.50 ± 0.03 a
	W3	52.72 ± 6.42 b	53.94 ± 1.67 a	38.34 ± 2.28 c	39.38 ± 1.37 b	4.19 ± 0.07 c	4.33 ± 0.48 c	2.31 ± 0.38 b	2.35 ± 0.18 a
2021–2022	W1	89.08 ± 0.96 a	29.51 ± 1.08 b	96.26 ± 1.64 a	31.87 ± 0.29 b	79.33 ± 2.55 a	26.26 ± 0.23 a	37.43 ± 4.81 a	12.37 ± 1.25 a
	W2	88.32 ± 10.14 a	42.57 ± 3.27 a	71.44 ± 4.36 b	34.49 ± 1.46 a	29.82 ± 2.40 b	14.38 ± 0.63 c	17.68 ± 2.30 b	8.55 ± 1.27 b
	W3	58.12 ± 14.38 b	32.74 ± 4.29 b	63.97 ± 5.93 b	36.56 ± 1.43 a	30.86 ± 1.00 b	17.75 ± 1.96 b	22.59 ± 1.42 b	12.95 ± 0.92 a

Notes: Different lowercase letters after the same column of data indicate significant differences at the p < 0.05 level between different treatments, the same as below.

, in the two-year experiment, the water consumption of the winter wheat growing period under the W2 and W3 treatments mainly came from the soil layer above 100 cm, accounting for 77.06–87.39% and 69.30–94.22% of the total soil water consumption, respectively. Compared with the W2 and W3 treatments, the utilization capacity of soil moisture in the 100–150 cm soil layer of the W1 treatment was remarkably improved, accounting for 15.71% and 26.36%, respectively.

Under the same irrigation system, the SWC in the 100–200 cm soil layer in 2020–2021 was lower than that in 2021–2022. In the two-year experiment, the soil water consumption decreased with the increase in the irrigation volume. Except for the soil water consumption above the 50 cm soil layer, the difference between the W1 and W2 treatments did not reach a significant level, but the soil water consumption in the other soil layers of the W1 treatment was found to be obviously higher than that of the W2 and W3 treatments. In 2020–2021, there was a notable difference in soil water consumption between the W2 and W3 treatments above the 50 cm soil layer, and the difference between the W2 and W3 treatments was not notable in 2021–2022.

3.3. Total Water Consumption and Water Consumption Structure of Winter Wheat under Different Irrigation Systems

As shown in

Table 2. Total water consumption and water consumption structure of winter wheat.

Year	Treatments	Soil Water Storage Presowing (mm)	Soil Water Consumption		Precipitation		Irrigation Rate		Total Water Consumption (mm)
			Values (mm)	Rate (%)	Values (mm)	Rate (%)	Values (mm)	Rate (%)
2020–2021	W1	569.26 ± 8.58 b	237.89 ± 11.02 a	64.05 ± 1.05 a	58.40	15.74 ± 0.46 a	75.00	20.21 ± 0.59 c	371.29 ± 11.02 b
	W2	588.10 ± 4.72 a	197.25 ± 12.19 b	48.60 ± 1.54 b	58.40	14.40 ± 0.43 b	150.00	37.00 ± 1.11 b	405.65 ± 12.19 a
	W3	489.28 ± 3.00 c	97.57 ± 9.01 c	25.58 ± 1.78 c	58.40	15.34 ± 0.37 a	225.00	59.08 ± 1.41 a	380.97 ± 9.01 b
2021–2022	W1	604.94 ± 4.85 a	302.10 ± 7.94 a	67.65 ± 0.57 a	69.40	15.55 ± 0.27 ab	75.00	16.80 ± 0.30 c	446.50 ± 7.94 ab
	W2	609.62 ± 7.01 a	207.26 ± 13.20 b	48.54 ± 1.57 b	69.40	16.28 ± 0.50 a	150.00	35.18 ± 1.07 b	426.66 ± 13.20 b
	W3	609.57 ± 0.99 a	175.55 ± 22.54 c	37.26 ± 3.09 c	69.40	14.79 ± 0.73 b	225.00	47.95 ± 2.36 a	469.95 ± 22.54 a

Notes: Different lowercase letters after the same column of data indicate significant differences at the p < 0.05 level between different treatments, the same as below.

, presowing soil moisture storage, soil moisture storage consumption, and total water consumption in 2020–2021 were lower than those in 2021–2022 under the same irrigation system. Under different irrigation systems, the total water consumption in 2020–2021 was W2 > W3 > W1, with the W2 treatment being remarkably higher than the W1 and W3 treatments. In 2021–2022, the total water consumption showed that W3 > W1 > W2, with a notable difference among the W2 and W3 treatments. The percentage of precipitation in 2020–2021 was W1 > W3 > W2, with the W2 treatment receiving a significantly lower amount than the other treatments. In 2021–2022, it was W2 > W1 > W3, and the variance in precipitation between the W2 and W3 treatments was significant. In the two-year experiment, the percentage of irrigation volume was W3 > W2 > W1, with significant differences between treatments; the soil water storage consumption and its percentage were W1 > W2 > W3, with significant differences between treatments.

To clarify the impact of soil moisture storage presowing on the ability of winter wheat to use stored soil water, the correlation between soil moisture storage presowing and soil water consumption was analyzed. The results are shown in

Table 3. Correlation analysis among soil water storage and soil water storage consumption presowing.

Index	Soil Depth (cm)
	0–50	50–100	100–150	150–200	0–200
Soil water content presowing (mm)	0.432	0.632 **	0.582 *	0.634 **	0.689 **

Notes: * indicates a significant correlation between indicators at the p < 0.05 level; ** indicates a significant correlation between indicators at the p < 0.01 level, the same as below.

. The relationship between soil moisture storage presowing and soil water consumption showed a positive and highly significant correlation in the 50–100 cm soil layer, 150–200 cm soil layer, and 0–200 cm soil layer; a notable positive correlation in the 100–150 cm soil layer; and no significant linear relationship in the 0–50 cm soil layer.

3.4. Effect of the Irrigation System and Variety on the SPAD Value of Winter Wheat Flag Leaves

Changes in the SPAD values of winter wheat flag leaves under different treatments are shown in Figure 3. The SPAD values of the W3 treatment of the three varieties gradually increased following the growth process. The SPAD values of the W1 and W2 treatments gradually decreased after reaching a peak at the flowering period. During the filling period, the SPAD values of the three varieties showed W3 > W2 > W1 and reached the highest at the filling stage of the W3 treatment, which were 62.23, 61.07, and 58.66, respectively. This result shows that irrigation can delay the aging of wheat flag leaves and enhance the stay-green of leaves, thus prolonging the late functional period of plants. The SPAD values of the three varieties were generally expressed as Hengmai28 > Hengguan35 > Hengmai4399 under the same irrigation system. The SPAD values of Hengmai4399, Hengguan35, and Hengmai28 at the filling stage under the W1, W2, and W3 treatments increased by 4.68–37.75%, 2.06–21.27%, and 5.17–15.81%, respectively, compared with the jointing stage. On comparing the mean SPAD values under different irrigation treatments, Hengmai4399 showed W1 > W2 > W3, while Hengguan35 and Hengmai28 showed W2 > W1 > W3.

3.5. Effects of the Irrigation System and Variety on the Spike Traits of Winter Wheat

As shown in

Table 4. Effects of different treatments on the spike traits of winter wheat in 2021–2022.

Treatments	Variety	Spike Length (cm)	Spikelet Number	Infertile Spikelet Number
W1	Hengmai4399	7.63 ± 0.13 aA	17.33 ± 0.09 aA	2.18 ± 0.46 aB
	Hengguan35	6.80 ± 0.20 bB	16.49 ± 0.38 bA	2.02 ± 0.32 aA
	Hengmai28	7.11 ± 0.25 bB	17.56 ± 0.15 aA	2.02 ± 0.09 aB
W2	Hengmai4399	7.80 ± 0.07 aA	17.65 ± 0.26 aA	2.37 ± 0.30 aB
	Hengguan35	6.91 ± 0.22 bB	16.28 ± 0.52 bA	1.97 ± 0.28 abA
	Hengmai28	7.63 ± 0.26 aA	17.45 ± 0.40 aA	1.77 ± 0.15 bC
W3	Hengmai4399	7.37 ± 0.34 aA	17.22 ± 0.30 aA	3.41 ± 0.59 aA
	Hengguan35	7.43 ± 0.32 aA	16.55 ± 0.48 aA	2.53 ± 0.24 bA
	Hengmai28	7.56 ± 0.17 aAB	17.47 ± 0.61 aA	2.51 ± 0.12 bA
V		**	**	**
W		*	ns	**
V × W		**	ns	ns

Notes: In the same column of data, different lowercase letters indicate that different varieties of the same irrigation system have significant differences at the p < 0.05 level and different uppercase letters indicate that different irrigation systems of the same variety have a significant difference at the p < 0.05 level, *: p < 0.05; **: p < 0.01; ns: not significant, V: variety; W: irrigation treatment; V × W: interaction between variety and irrigation system, the same as below.

, the effects of winter wheat varieties on the spike length, spikelet number, and infertile spikelet number of winter wheat reached very significant levels. Under the W1 treatment, the spike length of Hengmai4399 was remarkably longer than that of Hengguan35 and Hengmai28, while the spikelet number of Hengguan35 was notably less than that of the Hengmai4399 and Hengmai28. Under the W2 treatment, the spike length and spikelet number of Hengguan35 were obviously lower than those of Hengmai4399 and Hengmai28. The infertile spikelet number of Hengmai28 was notably lower than that of Hengmai4399, and there was a minor difference compared with Hengguan35. Under the W3 treatment, the difference in the spike length and spikelet number among the three varieties did not reach a remarkable level. The infertile spikelet number of Hengmai4399 was notably more than that of Hengguan35 and Hengmai28.

Irrigation had a very significant influence on the infertile spikelet number of winter wheat, an obvious effect on the spike length, and no notable influence on the spikelet number (Table 4). Under different irrigation systems, the spike length and spikelet number of Hengmai4399 showed W2 > W1 > W3. The infertile spikelet number showed W3 > W2 > W1, with the W3 treatment showing obviously more results than the W1 and W2 treatments. The spike length of Hengguan35 showed W3 > W2 > W1, and the spike length of the W3 treatment was significantly longer than that of the W1 and W2 treatments. The spikelet number and infertile spikelet number showed W3 > W1 > W2. The spike length of Hengmai28 showed W2 > W3 > W1, and the spike length of the W1 treatment was significantly shorter than that of the W2 and W3 treatments. The spikelet number showed W1 > W3 > W2. The infertile spikelet numbers of Hengmai4399, Hengguan35, and Hengmai28 were the highest in the W3 treatment, which were 43.88–56.42%, 25.25–28.43%, and 24.26–41.81%, higher than those in the W1 and W2 treatments, respectively. The interaction of variety and irrigation had a highly significant influence on the winter wheat spike length, but not on the spikelet number and infertile spikelet number (Table 4).

3.6. Effect of the Irrigation System and Variety on the Yield, Yield Components, and WUE of Winter Wheat

As shown in

Table 5. Effects of different treatments on winter wheat yield, yield components, and water use efficiency.

Year	Treatments	Variety	Thousand-Grain Weight (g)	Effective Spikes (104·hm−2)	Grains per Spike	Grain Yield (kg·hm−2)	Water Use Efficiency (kg·m−3)	Increasing Rate (%)
2020–2021	W1	Hemgmai4399	39.98 ± 3.21 A	-	-	6178.09 ± 674.41 A	1.66 ± 0.18 B	-
		Hengguan35	41.05 ± 1.94 A	-	-	6178.09 ± 1097.13 B	1.66 ± 0.30 B	-
	W2	Hemgmai4399	41.21 ± 1.59 A	-	-	6695.01 ± 326.44 A	1.65 ± 0.08 B	8.37
		Hengguan35	43.65 ± 3.50 A	-	-	7261.96 ± 1051.22 AB	1.79 ± 0.26 AB	17.54
	W3	Hemgmai4399	42.20 ± 2.87 A	-	-	9254.63 ± 319.34 B	2.43 ± 0.08 A	49.80
		Hengguan35	44.46 ± 2.07 A	-	-	8379.19 ± 452.30 A	2.20 ± 0.12 A	35.63
2021–2022	W1	Hengmai4399	42.75 ± 2.10 aA	559.72 ± 65.72 aA	36.83 ± 2.10 aA	7894.44 ± 200.23 aA	1.77 ± 0.04 aA	-
		Hengguan35	43.29 ± 2.97 aA	535.98 ± 67.57 aA	35.71 ± 1.34 aA	8511.11 ± 571.81 aA	1.91 ± 0.13 aA	-
		Hengmai28	44.95 ± 3.64 aA	629.69 ± 19.83 aA	33.09 ± 2.04 aA	8061.11 ± 192.45 aA	1.81 ± 0.04 aA	-
	W2	Hengmai4399	45.92 ± 3.78 aA	560.14 ± 25.68 aA	37.75 ± 0.96 aA	8938.89 ± 554.86 aB	2.10 ± 0.13 aB	13.23
		Hengguan35	47.72 ± 3.29 aAB	551.26 ± 78.61 aA	36.88 ± 3.34 aA	9016.67 ± 758.47 aA	2.11 ± 0.18 aA	5.94
		Hengmai28	50.41 ± 3.56 aAB	601.37 ± 23.25 aA	35.22 ± 1.75 aA	9188.89 ± 318.13 aB	2.15 ± 0.07 aB	13.99
	W3	Hengmai4399	45.97 ± 4.51 aA	585.54 ± 62.35 aA	32.67 ± 1.45 aB	9388.89 ± 358.37 aB	2.00 ± 0.08 aB	18.93
		Hengguan35	50.31 ± 0.82 abB	578.88 ± 16.64 aA	38.66 ± 1.61 bA	9061.11 ± 209.72 aA	1.93 ± 0.04 aA	6.46
		Hengmai28	53.72 ± 3.61 bB	630.10 ± 70.34 aA	33.49 ± 1.77 aA	9472.22 ± 685.65 aB	2.02 ± 0.15 aB	17.51

Notes: - represents that there is no data in this cell. In the same column of data, different lowercase letters indicate that different varieties of the same irrigation system have significant differences at the p < 0.05 level and different uppercase letters indicate that different irrigation systems of the same variety have a significant difference at the p < 0.05 level, the same as below.

, in general, the thousand-grain weight and yield under different treatments in 2021–2022 were higher than those in 2020–2021, but the WUE under the W3 treatment was higher in 2020–2021. In the two-year experiment, the thousand-grain weight of the three varieties increased gradually with the increasing irrigation amount, and the thousand-kernel weight of Hengguan35 was greater than that of Hengmai4399 under each irrigation level. From 2020 to 2021, for the same variety, the difference in the thousand-grain weight among different irrigation systems is not obvious. From 2021 to 2022, the thousand-grain weight and effective spikes of Hengmai28 were higher than those of Hengmai4399 and Hengguan35. The thousand-grain weight of Hengmai28 and Hengguan35 were remarkably greater under the W3 treatment than those under the W1 treatment, but there were no obvious differences among the different irrigation systems of Hengmai4399. The thousand-grain weight of Hengmai28 under the W3 treatment was obviously greater than that of Hengmai4399. The effective spikes of Hengmai4399 and Hengguan35 increased gradually with increasing irrigation, but that of Hengmai28 firstly reduced and then increased with increasing irrigation. The grain number per spike of Hengmai4399 and Hengmai28 increased first and then dropped with increasing irrigation. The grains per spike of Hengmai4399 under the W3 treatment were significantly less than those under the W1 and W3 treatments. The grains per spike of Hengguan35 increased gradually with the volume of irrigation. Among different irrigation systems, there was no obvious difference in the grain number per spike. The grains per spike of Hengguan35 under the W3 treatment were notably more than those of the other varieties under the W3 treatment; the difference between the different varieties in the other irrigation treatments did not reach a significant level for the grains per spike.

For the same variety, the WUE of the W1 and W2 treatments in 2020–2021 was lower than that in 2021–2022, while the WUE of the W3 treatment was higher in 2020–2021. From 2020 to 2021, the yield and WUE of Hengmai4399 under the W3 treatment were remarkably higher than those under the other two irrigation treatments. The yield and WUE of the W3 treatment of Hengguan35 were notably higher compared to the W1 treatment while the difference between the W2 treatment and the other two irrigation treatments was not found to be significant. The yields of Hengmai4399 and Hengguan35 under the W2 and W3 treatments increased by 8.37–49.80% and 17.54–35.63%, respectively, compared with those under the W1 treatment. The WUE under the W3 treatment increased by 46.39–47.27% and 22.91–32.53%, respectively, compared with the W1 and W2 treatments (Table 5).

From 2021 to 2022, except for Hengguan35, the yield of other varieties in the W1 treatment was significantly decreased compared to the W2 and W3 treatments. There was no notable difference in the yield among varieties under the same irrigation amount. The yields of Hengmai4399, Hengguan35, and Hengmai28 in the W2 and W3 treatments increased by 13.23–18.93%, 5.94–6.46%, and 13.99–17.51%, respectively, compared with those in the W1 treatment. Different from the previous year, the WUE of Hengmai4399, Hengguan35, and Hengmai28 showed W2 > W3 > W1. The WUE of the W2 treatment was 5.00–18.64%, 9.33–10.47%, and 6.44–18.78% higher than that of the W3 and W1 treatments, respectively (Table 5).

From 2020 to 2021, winter wheat varieties had no significant influence on the thousand-grain weight, yield, or WUE. The effect of irrigation treatment on the yield and WUE reached an extremely significant level, but had no remarkable effect on the thousand-grain weight. The interaction between irrigation and variety had no significant impact on the thousand-kernel weight, yield, and WUE. From 2021 to 2022, the thousand-grain weight, effective spikes, and grains per spike was significantly influenced by variety. Different from the previous year, the effect of irrigation treatment on the yield, thousand-grain weight, and WUE reached a highly remarkable level. The synergistic effect among irrigation and variety had an obvious influence on the grains per spike but had no significant effect on other traits (

Table 6. Effect of variety, irrigation, and the interaction of both on yield, yield components, and water use efficiency.

Year	Treatments	Thousand-Grain Weight (g)	Effective Spikes (104·hm−2)	Grains per Spike	Gain Yield (kg·hm−2)	WUE (kg·m−3)
2020–2021	V	ns	-	-	ns	ns
	W	ns	-	-	**	**
	V × W	ns	-	-	ns	ns
2021–2022	V	*	*	*	ns	ns
	W	**	ns	ns	**	**
	V × W	ns	ns	*	ns	ns

Notes: *: p < 0.05; **: p < 0.01; ns: not significant, V: variety; W: irrigation treatment; V × W: interaction between variety and irrigation system, the same as below.

).

In the two-year experiment, the average thousand-grain weight and the average yield of the three varieties increased with increasing irrigation amount, and the average thousand-grain weight and average yield in 2020–2021 were generally lower than those in 2021–2022. From 2020 to 2021, the average WUE increased gradually with increasing irrigation amount. The yield and WUE in the W3 treatment were obviously higher than those in the W1 and W2 treatments. In 2021–2022, the thousand-grain weight and yield of the W2 and W3 treatments were significantly increased compared to the W1 treatment. Moreover, the average WUE in the W2 treatment was obviously higher than that in the W1 and W3 treatments (

Table 7. Effects of different irrigation systems on the yield, yield components, and WUE of winter wheat.

Year	Treatments	Thousand-Grain Weight (g)	Effective Spikes (104·hm−2)	Grains per Spike	Gain Yield (kg·hm−2)	WUE (kg·m−3)
2020–2021 (N = 6)	W1	40.52 ± 2.44 a	-	-	6178.09 ± 814.50 b	1.66 ± 0.22 b
	W2	42.43 ± 2.77 a	-	-	6978.49 ± 762.29 b	1.72 ± 0.19 b
	W3	43.33 ± 2.55 a	-	-	8816.91 ± 593.75 a	2.31 ± 0.16 a
2021–2022 (N = 9)	W1	43.66 ± 2.76 b	575.13 ± 64.03 a	35.21 ± 2.31 a	8155.56 ± 421.14 b	1.83 ± 0.09 c
	W2	48.02 ± 3.65 a	570.93 ± 48.79 a	36.61 ± 2.24 a	9048.15 ± 508.30 a	2.12 ± 0.12 a
	W3	50.00 ± 4.45 a	598.18 ± 53.47 a	34.94 ± 3.14 a	9307.41 ± 443.17 a	1.98 ± 0.09 b

Notes: Different lowercase letters after the same column of data indicate significant differences at the p < 0.05 level between different treatments, the same as below.

).

Regression analysis was conducted between the yield increase rate and irrigation amount (Figure 4). In 2020–2021, there was a highly obvious linear correlation between the grain yield increase rate and irrigation amount of Hengmai4399 (p < 0.01; Figure 4a). The linear correlation between the yield increase rate and irrigation amount of Hengguan35 reached a significant level (p < 0.05; Figure 4b). In 2021–2022, a significant linear relationship between the grain yield increase rate and irrigation amount of Hengmai4399 and Hengmai28 was found (p < 0.05; Figure 4c,e), and there was no notable linear correlation between the yield increase rate and irrigation amount in Hengguan35 (p > 0.05; Figure 4d).

3.7. Correlation Analysis of the Yield, Yield Components, and WUE

To clarify the correlation between the yield, yield components, and WUE, correlation analysis was conducted on the indexes. The yield in 2020–2021 was highly remarkably and positively correlated with the WUE, obviously and positively correlated with the thousand-kernel weight. In 2021–2022, the yield was extremely obviously positively correlated with the WUE and thousand-grain weight. The WUE was significantly positively correlated with the thousand-grain weight, indicating that irrigation increased the grain yield by increasing the thousand-kernel weight of winter wheat, thereby increasing the WUE; the effective spikes were significantly negatively correlated with the grains per spike, indicating that the increase in spike number would lead to a significant decrease in the grains per spike (

Table 8. Correlation analysis of yield, yield components, and water use efficiency.

Index	Gain Yield (kg·hm−2)	Water Use Efficiency (kg·m−3)	Thousand-Grain Weight (g)	Grains per Spike	Year
Water use efficiency (kg·m−3)	0.982 **	-	-	-	2020–2021
Thousand-grain weight (g)	0.564 *	0.517 *	-	-
Water use efficiency (kg·m−3)	0.857 **	-	-	-	2021–2022
Thousand-grain weight (g)	0.562 **	0.431 *	-	-
Grains per spike	0.049	0.190	0.008	-
Effective spikes (104·hm−2)	0.039	−0.075	0.029	−0.473 *

Notes: *: p < 0.05; **: p < 0.01

).

Regression analysis were conducted for the yield and irrigation amount for each of the three varieties; the results are shown in Table 8. In 2020–2021, a very significant and positive correlation between yield and irrigation water was found for Hengmai4399 (p < 0.01), and the yield of Hengguan35 was notably and positively related with the irrigation amount (p < 0.05). In 2021–2022, the yields of Hengmai4399 and Hengmai28 were very significantly positively correlated with the irrigation amount (p < 0.01). There was no notable linear correlation in the yield and irrigation amount of Hengguan35. Due to the lack of soil water storage presowing, the influence of irrigation on the yield of winter wheat was highlighted. Therefore, the F values of Hengmai4399 and Hengguan35 in 2020–2021 was higher than that in 2021–2022 (

Table 9. Regression analysis of yield and irrigation amount.

Year	Variety	Relational Expression	R2	F Value
2020–2021	Hengmai4399	y = 20.510x + 4299.371	0.806 **	29.119 **
	Hengguan35	y = 14.674x + 5071.979	0.591 *	10.119 *
2021–2022	Hengmai4399	y = 9.963x + 7246.296	0.748 **	20.762 **
	Hengguan35	y = 3.667x + 8312.963	0.185	1.589
	Hengmai28	y = 9.407x + 7496.296	0.655 **	13.267 **

Notes: y: yield, x: irrigation rate, *: p < 0.05; **: p < 0.01

).

4. Discussion

In this research, the water content of the soil profile of each treatment was uniformly distributed in the pre-jointing stage. After irrigation, the increase in the soil moisture content above the 60 cm soil layer was significantly greater than that below the 60 cm soil layer, which was in agreement with the results of Cheng et al. [23] and Zhang et al. [24] The soil water consumption among the 0–100 cm soil layer during the growth period of winter wheat under the W2 and W3 treatments accounted for 77.06–87.39% and 69.30–94.22% of the total soil moisture consumption, respectively. However, the soil moisture consumption in the 100–150 cm soil layer under the W1 treatment was obviously greater than that under the W2 and W3 treatments, accounting for 15.71% and 26.36%, respectively. This result is because the soil moisture absorbed by the winter wheat mainly depends on the root, but there is a less-deep root system, so the consumption of soil moisture mainly comes from the 0–120 cm soil layer [24,25,26]. The W1 treatment did not supplement the shallow soil water after the jointing stage because the reduction in irrigation will promote downward root development of winter wheat, thereby enhancing the capacity of winter wheat to absorb deep soil water [27]. The water consumption in each soil layer is greatly influenced by the soil and climatic conditions for winter wheat [28]. Under the condition of limited water irrigation, sufficient soil water storage before sowing is an important condition for winter wheat to maintain a high yield [11]; it is because sufficient soil water storage before sowing can promote wheat to absorb deep soil moisture. The results of this study showed that there was an obvious and positive correlation between deep soil water consumption and soil moisture storage presowing (Table 3). Due to more autumn rainfall in 2021, the soil moisture before winter wheat sowing was sufficient from 2021 to 2022, which improved the ability of wheat to use the soil moisture storage. Therefore, there was a large difference in the change in soil water content under 140 cm in the two-year test W1 treatment, but irrigation in the late growing period of the W2 and W3 treatments reduced this difference (Figure 2).

Water has a regulatory effect on the photosynthetic characteristics of wheat, and high chlorophyll content in flag leaves is one of the foundations of efficient photosynthesis [29]. Because the photosynthetic products in the late growing period account for 80% of the grains [30], leaf senescence caused by the decrease in chlorophyll content in the late growth period will reduce the production of winter wheat. In this research, the SPAD value in the W3 treatment was the lowest among the three irrigation treatments, but it maintained a high SPAD value after the filling stage, which effectively enhanced the photosynthesis of winter wheat in the late growing stage and promoted its dry matter accumulation [13]; this was a main reason for the high yield of the W3 treatment. Dai et al. [31] showed that the SPAD value of the flag leaves of winter wheat first increased and then reduced from the heading stage to the maturity stage. The W1 and W2 treatments obtained the same results, while the reason for the different rules of the W3 treatment was that irrigation was carried out during the filling stage among this study (Figure 3).

Dang et al. [32] showed that deep tillage among the fallow period can increase soil moisture storage, promote the absorption of nitrogen and phosphorus by wheat, and promote wheat spike formation. Deng et al. [33] found that increasing soil moisture storage can increase spike number and yield as well as improve the WUE. In this experiment, since there was a lack of soil moisture storage before sowing in 2020–2021, the correlation between the yield increase rate and irrigation amount is higher. Moreover, the thousand-grain weight, effective spikes, grains per spike, and yield under different treatments in 2021–2022 are higher than those in 2020–2021. At that time, irrigation was needed to supplement the water demand for the winter wheat development. Under the W1 and W2 treatments, irrigation failed to supplement the water demanded by winter wheat growth and development, resulting in low yields. The W3 treatment supplemented the water required by winter wheat development at the later stage, and the yield increased significantly. However, when the irrigation amount is too high, although the wheat yield has increased, its ET will also increase significantly, and reducing the irrigation amount will lead to a decrease in wheat yield, but the WUE can be improved [34,35]. Yang et al. [5] found that when the irrigation times increased from 2 to 3, the yield increase was not significant. This study showed that the yield of the W3 treatment was the highest in 2021–2022, but the WUE of the W2 treatment was remarkably higher than that of the W1 and W3 treatments, which was consistent with the above research rules. However, from 2020 to 2021, the yields of the W1 and W2 treatments were too low and the water consumption under the W3 treatment was low, resulting in the highest WUE. Xu et al. [36] found that irrigation at the jointing stage and flowering stage could increase the yield and WUE by increasing the biomass, harvest index, and grain number per spike of winter wheat after anthesis. In this experiment, the average grain number per spike and WUE of the W2 treatment were higher than those of the other irrigation treatments in 2021–2022, and the yield was higher than the W1 treatment. In 2020–2021, the WUE of the W2 treatment was also higher than that of the W1 treatment, which was consistent with the previous pattern. Irrigation affects yield by affecting the thousand-grain weight [24]. In this study, there was a notable and positive correlation between the yield and thousand-grain weight from 2020 to 2021, and there was an extremely remarkable and positive correlation between the yield and thousand-grain weight from 2021 to 2022, indicating that irrigation increased the wheat yield by increasing the thousand-kernel weight, which was in agreement with the previous studies.

Due to different factors such as drought resistance, different varieties of winter wheat have different responses to irrigation [34,37]. The regression analysis between the yield and irrigation amount showed that the yields of Hengmai4399 and Hengmai28 were highly significantly positively correlated with the irrigation amount from 2021 to 2022, and the fitting degree of Hengmai4399 was higher than that of Hengmai4399, but there was no notable linear relationship between the yield and irrigation amount of Hengguan35. The correlation between the yield and irrigation amount of Hengguan35 from 2020 to 2021 was also lower than that of Hengmai4399 (Table 8). The regression analysis between the grain yield increase rate and irrigation amount showed similar results (Figure 4). Hengmai4399 had the most positive response to irrigation, followed by Hengmai28, and Hengguan35 had a minor response. From 2021 to 2022, under the W1 treatment, Hengguan35 had the highest yield, indicating that its drought resistance was the strongest, followed by that of Hengmai28, and that of Hengmai4399, which was the weakest.

The spike traits of winter wheat are susceptible to drought [38]. The varieties with strong drought resistance can still maintain a high spike rate and spike number per unit area under drought stress to ensure the higher yield and WUE. Many scholars use the spike shape of winter wheat as an index to evaluate drought resistance. Farkas et al. [39] found that drought stress on winter wheat in the booting period will lead to a decrease in spike number, thousand-grain weight, and yield. Semenov et al. [40] found that high temperature and drought stress at the flowering stage could lead to a reduction in the grain number and yield. In this study, Hengmai4399 had more infertile spikelet numbers than Hengmai28 and fewer spikes per unit area than Hengmai28 under the same irrigation condition. The grain number per spike was slightly higher in the W2 treatment than in the W1 treatment, but the yield was significantly higher than that in the W1 treatment. This was attributed to the stronger drought resistance of Hengmai28. There was a significant and negative correlation between effective spikes and grains per spike. The increase in spike number led to a reduction in grain number per spike (Table 8), which was in contrast to the results of Sun et al. [41]. This difference may be related to climate, precipitation, and soil conditions, and the specific reasons need to be further explored.

5. Conclusions

Reducing the number of irrigation times in spring was helpful for improving the absorption capacity of winter wheat to deep soil water storage, and sufficient soil water storage before sowing was helpful to improve the utilization ability of winter wheat for soil water, ultimately increasing the winter wheat yield. The spike number and thousand-grain weight of Hengmai28 were the highest among the three varieties. The response of the three varieties to irrigation was Hengmai4399 > Hengmai28 > Hengguan35, and the drought resistance was Hengguan35 > Hengmai28 > Hengmai4399.

There was a significant or extremely significant correlation among the thousand-grain weight, yield, and WUE in the two-year experiment, indicating that irrigation increased the yield of winter wheat by increasing the thousand-grain weight of winter wheat and ultimately increased the WUE of winter wheat.

In the NCP, when the soil water storage presowing is insufficient, it is recommended to use Hengmai4399 combined with the W3 treatment. When the soil water storage presowing is sufficient, it is recommended to use Hengmai28 combined with the W2 treatment.