Deficit Irrigation Provides a Trade-Off Between Water Use and Alfalfa Quality

Abstract

Currently, the world is facing a serious agricultural water crisis, which also affects grassland areas. Alfalfa, a key perennial forage legume, consumes about 10% of China’s pastoral irrigation water. Reducing irrigation generally results in a loss of hay yield, but the effects on alfalfa quality and its relationship to water use are less clear. In this study, we explore alfalfa quality under different irrigation deficits and its relationship to water use in the Hexi Corridor of China. Alfalfa water use, quality yield (relative feeding value yield (RFV_yield) and crude protein yield (CP_yield)), and quality water use efficiency (relative feeding value water use efficiency (WUE_RFV) and crude protein water use efficiency (WUE_CP)) were measured in a field experiment. Alfalfa quality showed a negative correlation with the irrigation quota (the determination coefficient for relative feeding value was 0.375 and for crude protein was 0.289). There was a positive correlation between quality yield and irrigation quota (the determination coefficient for RFV_yield was 0.570 and for CP_yield was 0.631). The higher irrigation quota increased quality yield, which compensated for its negative effects on alfalfa quality. The mild and moderate water deficit treatments showed lower WUE_RFV than both the severe and no water deficit treatments. Moderate or mild water deficit is recommended to be used for one-year-old alfalfa treatment. No water deficit is beneficial to improve the quality water use efficiency of two-year-old alfalfa.

1. Introduction

The world is facing a serious agricultural water crisis, driven by rapid population growth, changing climate patterns, and unsustainable water usage [1]. Agriculture is the largest consumer of freshwater, accounting for approximately 70% of global water use [2]. In fact, many regions are experiencing severe water scarcity, with droughts, reduced rainfall, and the overextraction of groundwater further exacerbating the issue [3]. From 1980 to 2015, agricultural irrigation water use increased in China (60%), India (71%), and the United States (27%), mainly for a few key crops such as alfalfa (Medicago sativa L.), corn, rice, and wheat [4]. Addressing this challenge requires innovative solutions in water management, efficient irrigation techniques, and the adoption of drought-tolerant crops to optimize water use while ensuring global food supply and environmental sustainability.

Deficit irrigation is a water-saving strategy that involves applying less water than a crop’s full evapotranspiration requirement, aiming to optimize water use efficiency without significantly compromising yield [5,6,7]. This approach has been extensively studied across various crops to balance water conservation while maintaining acceptable productivity levels. However, different crops have different water sensitivity, and their response to water stress can vary significantly depending on their physiological traits and growth stages. Some crops, like wheat and barley, exhibit a high degree of drought tolerance and can maintain relatively stable yields under moderate water deficit conditions [8]. On the other hand, crops such as alfalfa are more sensitive to water stress, with even small reductions in irrigation leading to significant declines in yield [9,10]. This differential water sensitivity underscores the need for crop-specific irrigation management strategies, particularly in regions where water resources are limited. One benefit of water deficit to crops is that it can improve their quality. When crops experience limited water availability, stress triggers physiological responses that often lead to enhanced concentrations of nutrients such as proteins and sugars [11,12]. However, its impact on quality-related water use efficiency—particularly in perennial forage crops like alfalfa—remains poorly understood. Elucidating how water deficit affects aboveground biomass quality and its associated water use efficiency is essential for developing science-based irrigation management in pasture systems [13,14].

In addition, the relationship between irrigation quota and alfalfa quality remains a topic of debate. Previous studies have suggested that reducing irrigation can improving its quality [15,16,17,18,19,20]. A recent meta-analysis supports this view, although it also found that the irrigation method and soil properties may influence the relationship [21]. Some research has proposed that the relationship between irrigation quota and quality follows a quadratic function [22,23,24]. Interestingly, there are studies that suggest that no definitive connection exists between irrigation quota and alfalfa quality [25]. This variability highlights the complex role of irrigation in balancing hay yield and its quality. Few studies have established this relationship using the mean crop water stress index [15], only relevant if the study had employed the same methodology in its investigation. Additionally, research exploring the relationship between irrigation quota and alfalfa quality, quality yield and quality water use efficiency are lacking. Bridging these gaps is crucial for developing water-efficient management strategies that optimize both hay yield and alfalfa quality, especially for areas with water scarcity.

The Hexi Corridor unique geographical location in the central subspecies zone of China means that warm, humid air from the Pacific Ocean seldom reaches the area [26]. Consequently, drought and water scarcity are the dominant climatic characteristics of this region. These challenges impose significant pressure on agricultural water usage, especially for pasture water management. Alfalfa, as an important forage grass in the area, also faces this serious water crisis [20,27]. Therefore, we speculate that reduced irrigation will increase alfalfa quality, but high water use will improve its quality yield. Thus, it may improve its quality water use efficiency.

2. Materials and Methods

2.1. Study Area

The experiment was conducted at the National Field Scientific Observation and Research Station for Efficient Water Use in Oasis Agriculture, located in Wuwei, Gansu Province of China (102°50′ E, 37°52′ N) (Supplementary Figure S1a). Situated 1580 m above sea level, the station is located in the eastern part of the Hexi Corridor, bordered by the Tengger Desert to the east, the Qilian Mountains to the south, and the Badain Jaran Desert to the north. The region experiences a continental temperate arid climate characterized by abundant sunlight, a large diurnal temperature range, intense evaporation, and limited rainfall. The average annual precipitation is 164 mm, while the average annual surface evaporation reaches 2000 mm. This area is a typical example of desert oasis irrigation agriculture. The predominant soil type is sandy loam (Siltigic-Orthic Anthrosols), with an average bulk density of 1.48 g/cm³ for the 0–1.6 m soil layer. The field water holding capacity is 0.35 cm³/cm³, and the permanent wilting point is 0.09 cm³/cm³.

2.2. Materials

Test materials: The alfalfa variety MF4020 was selected as the test material. This variety is a taproot perennial forage species with a fall dormancy level of 4 and a drought resistance index of 1.8. It was sown in rows on 25 May 2017, at 2 cm sowing depth and 20 cm row spacing. The sowing rate was 20 kg hm⁻².

Drip Irrigation Materials: The branch pipe used in the experiment had a diameter of 32 mm. The drip irrigation tape featured an inner diameter of 16 mm, a wall thickness of 0.4 mm, and a dripper flow rate of 3 L h⁻¹. All pipes and equipment are provided by Dayu Water Saving Group Co., Ltd., Jiuquan, China. The dripper spacing was set at 30 cm.

2.3. Experimental Design and Field Management

Plot Layout: The experimental plot area was 24 m² (6 m × 4 m). Each plot contained five drip irrigation tapes, with an 80 cm spacing between them. The tapes were buried at a depth of 20 cm. Each drip irrigation tape controlled four rows of alfalfa.

Irrigation regime: The experiment included four treatments with three subsurface drip irrigation treatments and a no-irrigation treatment. They represent no water deficit treatments, moderate water deficit, mild water deficit and severe water deficit respectively. Each treatment was replicated three times, and a completely randomized block design was used (Supplementary Figure S1b). Based on the actual crop evapotranspiration [19], we determined that 30 mm of irrigation per week is the irrigation water volume for this area without water deficit. The irrigation volumes for the three subsurface drip irrigation treatments were 10 mm, 20 mm, and 30 mm, respectively. The irrigation schedules for the two years are described in Supplementary Table S1.

Mowing Management: The growing season for alfalfa in 2017 lasted from May to October, and from April to October in 2018. Mowing was typically performed at the beginning of flowering, with 3 mowings in 2017 and 4 mowings in 2018 (

Table 1. Details of the various irrigation treatments in 2017–2018.

Year	Harvest	Treatment	Rainfall/mm	Irrigation Frequency/No.	Irrigation Quota/mm	Irrigation Time /(Day Month)	Annual Irrigation/mm
2017	1st harvest	Severe water deficit	12.4	5	0	17 June a, 24 June, 1 July, 8 July, 15 July	0
		Moderate water deficit	12.4	5	10		50
		Mild water deficit	12.4	5	20		100
		No water deficit	12.4	5	30		150
	2nd harvest	Severe water deficit	73.6	3	0	22 July, 5 August, 12 August	0
		Moderate water deficit	73.6	3	10		40
		Mild water deficit	73.6	3	20		80
		No water deficit	73.6	3	30		120
	3rd harvest	Severe water deficit	23.4	5	0	27 August, 3 September, 10 September, 16 September, 23 September	0
		Moderate water deficit	23.4	5	10		50
		Mild water deficit	23.4	5	20		100
		No water deficit	23.4	5	30		150
2018	1st harvest	Severe water deficit	31.6	8	0	15 April, 22 April, 29 April, 6 May, 13 May, 20 May, 27 May, 4 June	0
		Moderate water deficit	31.6	8	10		80
		Mild water deficit	31.6	8	20		160
		No water deficit	31.6	8	30		240
	2nd harvest	Severe water deficit	13.6	4	0	10 June, 17 June, 24 June, 2 July	0
		Moderate water deficit	13.6	4	10		40
		Mild water deficit	13.6	4	20		80
		No water deficit	13.6	4	30		120
	3rd harvest	Severe water deficit	65.2	4	0	8 July, 15 July, 22 July, 29 July	0
		Moderate water deficit	65.2	4	10		40
		Mild water deficit	65.2	4	20		80
		No water deficit	65.2	4	30		120
	4th harvest	Severe water deficit	81.8	4	0	12 August, 10 September, 17 September	0
		Moderate water deficit	81.8	4	10		40
		Mild water deficit	81.8	4	20		80
		No water deficit	81.8	4	30		120

^a The irrigation management of alfalfa seedlings using micro-spraying before 17 June in 2017 is not included. The irrigation quota on 27 May, 1 June, 6 June, and 11 June in 2017 is 10 mm, and the irrigation quota in the seedling stage is 40 mm.

). The mowing dates in 2017 were 15 July, 18 August and 30 September, while in 2018, they were 7 June, 5 July, 7 August and 20 September. Pest control was conducted according to the unified management protocol, and manual weeding was performed after mowing.

2.4. Measurements

Meteorological data (Figure 1), including temperature, relative humidity, and precipitation, were automatically recorded at a height of 2.0 m by a meteorological monitoring system (HOBO, Campbell Scientific Inc., North Logan, UT, USA). The meteorological station was located approximately 20 m from the experimental site. The average temperature and precipitation during the growing seasons of 2017 and 2018 are shown in Figure 1. These data were used to assess the climatic conditions that influenced alfalfa growth and water consumption during the experimental period.

Soil moisture was measured using a portable soil moisture profiler (Diviner 2000, Sentek Pty Ltd., Stepney, Australia). This device provided real-time measurements of the volumetric water content in the 1.6 m soil layer of the root zone. The changes in soil water content at different harvests in different treatments over the two years are presented in Supplementary Figure S2. Soil moisture was recorded every 2 to 3 days throughout the experiment. To ensure accurate moisture readings, measurements were cross verified using a soil drill after each mowing. This combination of methods allowed for precise monitoring of the soil water content, which is crucial for understanding the effects of irrigation on alfalfa growth.

The water use of alfalfa was estimated using the water balance formula, which accounts for precipitation, irrigation, and soil moisture changes during the growing season (Supplementary Table S1). The water balance equation is as follows:

$$E{T}_c=I+P+S-\Delta W-R-D$$

(1)

where, ETc—alfalfa water use, mm; I—irrigation volume, mm; P—precipitation, mm; S—groundwater recharge, mm; ΔW—change in soil water storage in the 0–1.6 m soil layer, mm; R—surface runoff, mm; D—deep seepage, mm. According to previous studies, the groundwater depth is 25 m [19], so S is usually ignored. In this experiment, since no surface runoff is generated under drip irrigation conditions, and the designed single irrigation volume is small and insufficient to form deep seepage, R and D are ignored.

Determination of alfalfa quality content: The alfalfa yield was determined by combining large plots (1 m × 1 m) with small plots (20 cm × 20 cm) [19,20]. Five samples were randomly selected from the experimental area, fresh grass was weighed, and the samples in the plot were placed in an oven at 105 °C for 1 h. The temperature was then adjusted to 65 °C, and the grass was dried to constant weight and the yield was calculated. The dried samples were crushed into fine powder and passed through a 0.5 mm sieve. Each sample was measured three times. The crude protein content (CP) was determined using a FOSS KjeltecTM 8400 nitrogen analyzer, and the acid detergent fiber (%ADF) and neutral detergent fiber (%NDF) were tested in a bag boom using an ANKOM2000 automatic fiber analyzer (≤±0.5%) (ANKOM Technology, Wayne County, NY, USA) [20]. The relative feeding value (RFV) was calculated using the method of Rohweder et al. [28].

$$RFV=\left(120/\%NDF\right)\times \left[88.9-\left(0.779\times \%ADF\right)\right]$$

(2)

Calculation of alfalfa quality yield: Alfalfa crude protein yield (CP_yield, kg hm⁻²) is calculated by Formula (3), and relative feeding value yield (RFV_yield, t hm⁻²) is calculated by Formula (4) [27,29].

$${\mathit{CP}}_{\mathrm{yield}}=\%\mathit{CP} \times \mathit{Hay} \mathit{yield}$$

(3)

$${\mathit{RFV}}_{\mathit{yield}}=\mathit{RFV} \times \mathit{Hay} \mathit{yield}$$

(4)

Calculation of alfalfa quality water use efficiency (QWUE) [13,14,27]: Crude protein water use efficiency (WUE_CP, kg hm⁻² mm⁻¹) and relative feeding value water use efficiency (WUE_RFV, t hm⁻² mm⁻¹) were calculated by Formula (5) and Formula (6), respectively.

$${\mathit{WUE}}_{\mathit{CP}}={\displaystyle \frac{{\mathit{CP}}_{\mathit{yield}}}{{\mathit{ET}}_{\mathit{C}}}}$$

(5)

$${\mathit{WUE}}_{\mathit{RFV}}={\displaystyle \frac{{\mathit{RFV}}_{\mathit{yield}}}{{\mathit{ET}}_{\mathit{C}}}}$$

(6)

2.5. Data Processing

IBM SPSS Statistics 20 software was used for data processing, and the variance homogeneity test is shown in Supplementary Table S2. Duncan (p = 0.05) method was used for univariate comparative analysis. Origin 2021 was used for drawing graphs and fitting analysis.

3. Results

3.1. Alfalfa Quality Content and Hay Yield

Except for the first harvest in both 2017 and 2018, there were significant differences (p < 0.05) in the RFV and CP between treatments with the least irrigation (severe water deficit or moderate water deficit) and those with higher irrigation (mild water deficit or no water deficit) (

Table 2. Effects of different irrigation amounts on alfalfa quality content, quality yield, and quality water use efficiency.

Year	Harvest	Treatment	Relative Feeding Value/%	RFVyield/ (t hm−2)	WUERFV/ (t hm−2 mm−1)	Crude Protein/%	CPyield/ (kg hm−2)	WUECP/ (kg hm−2 mm−1)
2017	1st harvest	Severe water deficit	135.39 ± 1.30 a	37.26 ± 0.62 b	0.43 ± 0.01 a	21.80 ± 0.11 ab	600.09 ± 2.96 c	6.87 ± 06.08 b
		Moderate water deficit	118.89 ± 1.25 ab	37.75 ± 0.40 b	0.24 ± 0.01 b	20.19 ± 0.89 b	641.00 ± 28.20 c	4.10 ± 0.26 d
		Mild water deficit	99.23 ± 6.53 b	43.07 ± 2.83 b	0.23 ± 0.02 b	23.13 ± 0.89 a	1003.93 ± 38.41 b	5.34 ± 0.33 c
		No water deficit	117.75 ± 4.57 ab	67.18 ± 5.31 a	0.39 ± 0.03 a	23.86 ± 0.41 a	1364.40 ± 23.38 a	7.84 ± 0.26 a
	2nd harvest	Severe water deficit	145.60 ± 0.59 a	30.36 ± 0.12 c	0.24 ± 0.01 b	24.12 ± 0.70 a	502.96 ± 14.52 c	4.05 ± 0.13 c
		Moderate water deficit	128.17 ± 2.42 b	37.77 ± 0.71 b	0.30 ± 0.01 a	22.57 ± 0.65 a	664.99 ± 19.00 b	5.31 ± 0.13 a
		Mild water deficit	137.52 ± 4.40 a	44.98 ± 1.44 a	0.32 ± 0.01 a	20.80 ± 0.30 b	680.33 ± 9.80 b	4.78 ± 0.07 b
		No water deficit	99.46 ± 1.62 c	39.74 ± 1.62 b	0.25 ± 0.01 b	20.45 ± 0.14 b	816.98 ± 5.47 a	5.16 ± 0.17 ab
	3rd harvest	Severe water deficit	153.29 ± 3.11 a	23.86 ± 0.48 d	0.23 ± 0.00 b	24.01 ± 0.12 b	373.71 ± 1.81 d	3.56 ± 0.11 c
		Moderate water deficit	136.73 ± 3.07 b	28.34 ± 0.64 c	0.25 ± 0.01 ab	27.61 ± 0.72 a	572.29 ± 14.90 c	5.05 ± 0.13 ab
		Mild water deficit	115.23 ± 3.32 c	37.37 ± 1.08 b	0.26 ± 0.01 a	23.53 ± 0.19 b	762.90 ± 6.14 b	5.24 ± 0.11 a
		No water deficit	110.93 ± 5.23 c	43.66 ± 2.06 a	0.23 ± 0.01 ab	23.33 ± 0.35 b	918.24 ± 13.82 a	4.89 ± 0.02 b
2018	1st harvest	Severe water deficit	144.56 ± 0.28 ab	77.47 ± 9.42 bc	0.61 ± 0.10 a	19.59 ± 0.68 a	1050.46 ± 36.51 c	8.25 ± 0.35 a
		Moderate water deficit	127.58 ± 6.18 c	56.23 ± 5.95 c	0.30 ± 0.03 b	19.26 ± 0.51 a	857.62 ± 22.63 d	4.53 ± 0.14 c
		Mild water deficit	154.94 ± 3.67 a	97.43 ± 8.41 b	0.39 ± 0.03 ab	18.64 ± 0.20 a	1169.02 ± 12.49 b	4.74 ± 0.07 c
		No water deficit	136.81 ± 2.90 bc	133.25 ± 12.40 a	0.47 ± 0.04 ab	18.18 ± 0.13 a	1766.36 ± 13.11 a	6.18 ± 0.04 b
	2nd harvest	Severe water deficit	200.00 ± 7.75 a	31.63 ± 2.90 d	0.73 ± 0.12 a	24.36 ± 0.83 a	386.32 ± 13.24 c	8.79 ± 1.02 ab
		Moderate water deficit	145.12 ± 4.98 c	47.87 ± 6.57 c	0.49 ± 0.07 a	22.18 ± 0.28 b	732.19 ± 9.14 b	7.55 ± 0.10 b
		Mild water deficit	175.82 ± 7.83 b	65.65 ± 1.18 b	0.65 ± 0.01 a	20.90 ± 0.30 bc	782.32 ± 11.17 b	7.72 ± 0.28 b
		No water deficit	119.87 ± 3.69 c	90.88 ± 0.73 a	0.61 ± 0.00 a	19.43 ± 0.45 c	1476.36 ± 34.17 a	9.90 ± 0.24 a
	3rd harvest	Severe water deficit	207.48 ± 3.08 a	26.25 ± 6.99 c	0.65 ± 0.18 a	23.08 ± 0.27 a	289.83 ± 3.34 d	7.15 ± 0.09 a
		Moderate water deficit	166.75 ± 13.28 bc	34.78 ± 3.99 c	0.32 ± 0.04 a	23.43 ± 0.49 a	490.40 ± 10.33 c	4.58 ± 0.12 c
		Mild water deficit	189.10 ± 12.55 ab	70.83 ± 4.65 b	0.50 ± 0.04 a	21.95 ± 0.26 b	822.26 ± 9.60 b	5.76 ± 0.15 b
		No water deficit	137.56 ± 3.53 c	94.72 ± 4.17 a	0.51 ± 0.02 a	20.03 ± 0.19 c	1377.83 ± 13.19 a	7.37 ± 0.12 a
	4th harvest	Severe water deficit	189.78 ± 15.84 a	73.60 ± 7.88 b	0.51 ± 0.05 a	20.82 ± 0.36 a	808.69 ± 14.15 c	5.66 ± 0.09 b
		Moderate water deficit	153.66 ± 15.12 ab	58.87 ± 6.22 b	0.38 ± 0.02 bc	20.59 ± 0.18 ab	787.88 ± 6.81 c	5.13 ± 0.08 c
		Mild water deficit	146.11 ± 4.86 b	67.84 ± 0.58 b	0.31 ± 0.00 c	19.63 ± 0.23 bc	913.46 ± 10.74 b	4.12 ± 0.03 d
		No water deficit	138.93 ± 2.34 b	106.54 ± 3.72 a	0.50 ± 0.02 ab	19.26 ± 0.47 c	1476.44 ± 36.18 a	6.93 ± 0.17 a

Lowercase letters in different rows indicate differences among different treatments in the same harvest, p < 0.05.

). In general, RFV of alfalfa treated with severe water deficit was the highest in both years, apart from the first harvest in 2018. Additionally, the CP of the second harvest in 2017, the second harvest in 2018, and the fourth harvest in 2018 were highest in the severe water deficit treatment, while the third harvest in both 2017 and 2018 had the highest CP under moderate water deficit. These results indicate that reducing the quota of subsurface drip irrigation is beneficial for increasing the accumulation of alfalfa quality content.

Except for the second harvest in 2017 and the first harvest in 2018, the alfalfa quality yield followed this pattern, from smallest to largest: severe water deficit < moderate water deficit < mild water deficit < no water deficit. This suggests that increasing the irrigation quota generally promoted the alfalfa quality yield of each harvests. For instance, in 2017, the RFV_yield of the mild water deficit treatment was higher than that of the no water deficit treatment. This difference was significant (p < 0.05), highlighting that mild water deficit may be more beneficial for RFV_yield accumulation than no water deficit.

When comparing the different harvests among the year, it was observed that the first harvest of each year generally exhibited the highest quality yield. Over the two years, the first harvest’s RFV_yield accounted for 34.34–44.61% of the total RFV_yield in 2017 and 2018, while the CP_yield of the first harvest represented 34.13–44.02% of the total CP_yield in 2017 and 28.97–41.43% in 2018. This suggests that the first harvest in both years contributed significantly to the alfalfa quality yield.

3.2. Relationship Between Alfalfa Quality and Quality Yield with Irrigation Quota

Figure 2 shows that the irrigation quota of each harvest of alfalfa over 2 years was negatively correlated with the quality (RFV: R² = 0.375, p < 0.01; %CP: R² = 0.289, p = 0.003); and negatively correlated with the quality yield (RFV_yield: R² = 0.570, p < 0.001; CP_yield: R² = 0.631, p < 0.001). The increased irrigation quota led to a trade-off between yield and quality, with the higher hay quality yield compensating for the reduction in alfalfa quality.

Analysis of variance revealed significant interactive effects between growing years, harvest, and irrigation quota (between-subject effect) on both alfalfa quality and its yield parameters (p < 0.05; Supplementary Table S3). Furthermore, we observed a particularly strong interaction between the harvest and irrigation quota on forage quality (p < 0.01), indicating that the impact of water management on quality varied significantly across different harvest periods.

3.3. Relationship Between Alfalfa Quality and Water Use

The relationship between alfalfa water use and quality content and quality yield is illustrated in Figure 3. A negative correlation was observed between alfalfa water use and RFV_yield, with a coefficient of determination value of 0.146. Similarly, there was a negative correlation between water use and CP_yield, with a coefficient of determination value of 0.514 (Figure 3). However, there was a significant positive correlation (p < 0.01) between alfalfa water use and both the quality content and quality yield. The relationship between water use and RFV_yield is represented by the following equation: y = 0.2651x + 0.6579 (R² = 0.277). For CP_yield, the relationship is expressed as follows: y = 4.6962x +22.3619 (R² = 0.615). These results suggest that the patterns observed between water use, quality content, and yield are consistent with the relationship between the irrigation quota and these factors. In other words, reducing the irrigation quota increases alfalfa quality content, but this comes at the expense of the overall quality yield.

Factorial analysis revealed significant interaction effects between growing years, harvest, and water use (main effect) on alfalfa quality and its yield parameters (Supplementary Table S4). RFV_yield was affected by the interaction of different years and water use (p < 0.001), CP_yield was affected by the interaction of harvest and water use (p = 0.017), and the alfalfa quality was affected by the interaction of harvest and water use (p < 0.01).

3.4. Alfalfa QWUE and the Relationship with Irrigation Quota

As shown in Table 2, for the second and third harvests of 2017, the mild water deficit treatment resulted in the highest WUE_RFV, which increased by 33.33% and 13.04%, respectively, compared to severe water deficit. For the first harvest in 2017 and all harvests in 2018, the highest WUE_RFV was observed in the mild water deficit treatment, followed by no water deficit.

Compared to severe water deficit, the WUE_CP for the second and third harvests in 2017 increased by 17.99% to 47.38%. However, for the first harvest in 2018, the WUE_CP for moderate water deficit, mild water deficit, and no water deficit were lower than that for severe water deficit, with decreases of 45.08%, 42.58%, and 25.13%, respectively. This might be due to the winter water supply, which helped maintain the yield of the 1st harvest of alfalfa, thereby enhancing the WUE_CP in the severe water deficit treatment. For the first harvest in 2017 and the second, third, and fourth harvests in 2018, the WUE_CP pattern was consistent: alfalfa in moderate water deficit and mild water deficit treatments showed lower WUE_CP than severe water deficit, while no water deficit exhibited a higher WUE_CP than severe water deficit (except for the third harvest in 2018). This is likely because the water consumption of alfalfa in the severe water deficit treatment was much lower than in moderate water deficit and mild water deficit treatments, while CP_yield in no water deficit was significantly higher than in severe water deficit (p < 0.05) (Table 2).

The preliminary analysis found a significant quadratic function effect between irrigation quota and WUE_RFV; the equation is y = 0.0004x² − 0.0052x + 0.2712 (R² = 0.357 p = 0.0015). This result indicates that a lower irrigation quota can improve the alfalfa WUE_RFV. This is likely because the reduced irrigation quota encourages the accumulation of alfalfa RFV. However, higher irrigation volumes also enhance WUE_RFV, primarily by increasing the alfalfa quality yield.

4. Discussion

The water use of alfalfa and the relationship with hay yield have been widely studied [30,31], but the relationship between its water use and quality parameters has rarely been explored. In addition, there is still a large gap in the relationship between irrigation quota and alfalfa quality, especially in the water-limited arid region of northwest China.

4.1. The Relationship Between Alfalfa Quality and Water Use

Research on the response of different plant qualities to deficit irrigation indicates that drought stress generally improves plant quality [32,33,34]. For example, the ratio of soluble solids, reducing sugars, and sugar-acid content increased as irrigation quota decreased, improving tomato quality to a certain extent [34]. Research on melons [32,33] also showed that reducing irrigation quota can improve quality. Our study found that decreased irrigation quota generally leads to an increase in alfalfa quality content (Figure 2), which is consistent with our previous prediction. In most cases, the highest CP and RFV were observed in the severe water deficit treatment, which experienced the no irrigation. This may be explained by the fact that under severe water deficit conditions, alfalfa growth is hindered, leading to larger leave to stem ratio [15]. This results in a higher proportion of CP in the stems and a lower fiber content in the stems [15]. Stem biomass usually accounts for more than half of the aboveground biomass, and its quality can largely represent the alfalfa quality [24]. Our findings align with previous studies by Holman et al. [29] and a meta-analysis [21], supporting the notion that deficit irrigation can improve alfalfa quality. However, contrasting results have been reported by others, who observed improved quality under irrigated compared to rainfed conditions [24]. The discrepancy may stem from variations in experimental designs, environmental conditions, or irrigation regimes, though the underlying mechanisms remain unclear. Further research, including molecular-level investigations [18], is needed to elucidate how drought modulates carbon and nitrogen metabolism in alfalfa leaves and stems. Notably, Staniak and Harasim [35] reported no significant differences in CP between alfalfa subjected to deficit (40% of full irrigation) and optimal irrigation (70% of full irrigation). Their comparison corresponds to the contrast between our mild (1/3 full irrigation) and moderate (2/3 full irrigation) treatments, where quality differences were also marginal. This suggests that weakly significant effects may arise from subtle variations in water deficit severity, highlighting the need for standardized drought stress classifications in future studies. Establishing uniform thresholds for water deficit levels would enhance the comparability of research findings and improve the accuracy of drought impact assessments.

Additionally, there is a significant positive correlation between water use and both alfalfa quality and its yield (p < 0.01), although a significant negative correlation exists between water use and the alfalfa quality of each harvest (p < 0.01) (Figure 3). This suggests that alfalfa yield significantly influences quality, with quality content being subject to a “dilution effect”. Wang et al. [20] also found that while water deficit treatment increased crude protein content in alfalfa, it also led to a reduction in hay yield. The yield loss outweighed the increase in protein content, resulting in a decrease in crude protein yield. This “dilution effect” was also found in tomatoes [36]. Their result showed that Lycopene, β-carotene, total phenols and total antioxidant activity of tomato under deficit irrigation were neutralized by the negative effect on total yield [36]. Thus, the deficit irrigation may provide a trade-off between yield and the nutritional quality of the crops. Our supplementary analysis revealed significant interannual variation in irrigation quota effects on alfalfa quality yield (year × irrigation quota interaction, p < 0.001; Supplementary Table S3). While quality content showed no analogous interannual pattern, we hypothesize that hay yield serves as the primary determinant. Notably, both alfalfa quality yield and quality content were influenced by harvest and irrigation quota interactions. As demonstrated in our results, first harvests contributed disproportionately to seasonal quality yield, underscoring the critical importance of first harvest management in commercial alfalfa production systems.

Although our results show that quality is simply negatively correlated with irrigation quota and water use, we cannot deny that the above indicators may have nonlinear relationships. This nonlinear relationship was found in an old paper [15]. Their results showed that the relationship between the NDF of alfalfa and the crop water stress index can be expressed as follows: y = 448x² − 303x + 448 (R² = 0.69, p < 0.001). Although their x-axis is the crop water stress index, this indicator is also converted based on the field water holding capacity and can reflect the irrigation quota indicator to a certain extent. Another meta-analysis showed that the relationship between drought stress level and CP content was y = 0.014x² + 1.18x − 12.93 (R² = 0.71, p < 0.001) with fertilization [21]. In addition, Machado and Oliveira [37] proposed that the relationship between soluble solid content in tomatoes and irrigation quota follows a quadratic function: y = 2.14x² − 6.45x + 9.16 (R² = 0.73, p < 0.001). These results highlight the complexity of the relationship between different crop qualities and irrigation quota and water use at different locations, motivating us to conduct more extensive and comprehensive studies to evaluate their relationships.

4.2. Effects of Different Water Deficits on Alfalfa QWUE and the Relationship with Irrigation Quota

The concept of forage “QWUE” was introduced, and the alfalfa WUE_CP and WUE_RFV in the moderate water deficit treatment were significantly lower than those in the severe water deficit treatment in most cases (Table 2 and Figure 4). This suggests that in the Hexi Corridor area, the alfalfa WUE_CP and WUE_RFV without irrigation are higher than those in the moderate water deficit treatment. Our results are inconsistent with the results of our prediction and another experiment that used rain shelters to simulate drought in the Hexi Corridor. They believe that simulated drought can significantly increase the WUE_CP of soybeans, oats, and vetch [14]. Another study on controlled deficit irrigation for rapeseeds also showed that no irrigation during the vegetative growth period can increase the QWUE of oilseed crops [38]. In addition, Zhang et al. [27] compared the QWUE of various annual forages, finding that leguminous forages exhibited higher WUE_CP, while the WUE_RFV of gramineous crops like corn was higher. In our study, both WUE_RFV and WUE_CP were higher than those reported for annual forages by Zhang et al. [27]. This can likely be attributed to the application of subsurface drip irrigation, which significantly increased QWUE. This irrigation method accurately delivers water to the root zone, creating an “egg-shaped” moistening front around the roots and reducing alfalfa’s transpiration water consumption [39], thus improving its QWUE. In addition, the alfalfa we studied is widely used as a high-quality forage, so it is reasonable that its QWUE is higher than that of other forages.

In addition, when comparing the relationship between the irrigation quota and QWUE, it was found that irrigation quota has a quadratic function relationship with WUE_RFV (R² = 0.357, p < 0.001). This consistent result shows that both severe water deficit and no water deficit can improve WUE_RFV, but for different reasons. In the case of severe water deficit, the main factor is the higher proportion of quality content in the total WUE_RFV. On the other hand, with no water deficit, the improvement in WUE_RFV comes from the significant increase in quality yield. Thus, deficit irrigation provides a trade-off between irrigation quota and alfalfa quality.

4.3. Limitations and Prospects

While the results of this study highlight a significant linear relationship between alfalfa quality and irrigation quota, as well as water consumption, several limitations must be acknowledged. One of the primary constraints of this study is its focus on water deficit conditions. The research mainly explored scenarios where water availability was restricted, but the relationship between irrigation volume and alfalfa quality under excessive irrigation remains largely unexplored. Recent studies on furrow irrigation suggest that excessive irrigation can negatively impact the nitrogen content in alfalfa leaves and stems [40], which may suggest a potential shift from a linear to a curvilinear relationship in such cases. However, this remains speculative, and further studies are needed to assess the impact of excessive irrigation on alfalfa quality in more detail.

The specific conditions at the study site in the Hexi Corridor impose additional limitations. Due to the region’s inherent water scarcity, only an appropriate level of irrigation can be provided to alfalfa. Excessive irrigation is unlikely to occur naturally in this region, as the sandy soil in the Hexi Corridor often leads to deep infiltration [19,26], which can further exacerbate water wastage. Given the region’s reliance on limited water resources, irrigation managers must prioritize water efficiency and avoid over-irrigation, which could lead to unnecessary water loss and increased costs.

It should be noted that this two-year study was conducted in the same experimental plot, and we recognize that drought may have residual effects on subsequent growing seasons. Under prolonged drought conditions, alfalfa residue accumulation in the second year was reduced. Additionally, root development and soil microbial processes were likely influenced by the varying drought intensities experienced in the previous year [41]. These combined effects may ultimately reduce soil mineralization rates [41], potentially creating a negative feedback loop that further impacts alfalfa yield and forage quality. However, the mechanisms underlying these legacies’ effects remain poorly understood, warranting further investigation.

Furthermore, while alfalfa quality content is an important consideration for forage production, it is equally critical to evaluate the combined effect of quality content and yield, referred to as quality yield. In this study, increasing the irrigation quota was found to improve quality yield (Figure 3), which, in turn, enhances the overall profitability of alfalfa production. However, the trade-off between cost and profit requires further investigation. High-quality alfalfa is often more expensive, but its low yield limits its profitability. In regions like the Hexi Corridor, where water prices are relatively high in China [42], this trade-off must be carefully considered when designing irrigation strategies.

5. Conclusions

This study found that the irrigation quota and water use were positively correlated with CP_yield and RFV_yield but negatively correlated with alfalfa quality content. This means that although more water boosts the overall yield and quality yield, the actual quality content tends to decrease as water inputs increase. Thus, deficit irrigation provided a trade-off between water use and irrigation quota with alfalfa quality content and yield. The relationship between alfalfa WUE_RFV and the irrigation quota can provide a theoretical basis for the formulation of alfalfa irrigation regimes. These strategies can optimize both quality and yield while maintaining water use efficiency, especially in water-scarce regions, where water resources must be carefully managed to meet agricultural needs.