Skip to content


BMC Plant Biology

Open Access

Influence of drought stress on afalfa yields and nutritional composition

BMC Plant BiologyBMC series – open, inclusive and trusted201818:13

Received: 17 July 2017

Accepted: 28 December 2017

Published: 15 January 2018



It is predicted that climate change may increase the risk of local droughts, with severe consequences for agricultural practices.


Here we report the influence of drought on alfalfa yields and nutritional composition, based on artificially induced drought conditions during two field experiments. Two types of alfalfa cultivars were compared, Gold Queen and Suntory. The severity and timing of drought periods were varied, and the crop was harvested either early during flowering, or late at full bloom.


The obtained dry mass yields of Gold Queen were higher than Suntory, and the first was also more resistant to drought. Early harvest resulted in higher yields. Decreases in yields due to water shortage were observed with both cultivars, and the fraction of crude protein (CP) decreased as a result of drought stress; this fraction was higher in Gold Queen than in Suntory and higher in early harvest compared to late harvest. Severe drought late in spring had the highest effect on CP content. The fraction of fibre, split up into neutral detergent fibre (NDF) and acid detergent fibre (ADF) increased as a result of drought and was lower in early compared to late harvested plants. Suntory alfalfa produced higher fibre fractions than Gold Queen. The fraction of water-soluble carbohydrates (WSC) was least affected by drought. It was consistently higher in Gold Queen compared to Suntory alfalfa, and late harvest resulted in higher WSC content.


In combination, these results suggest that the nutritive value of alfalfa will likely decrease after a period of drought. These effects can be partly overcome by choosing the Gold Queen cultivar over Suntory, by targeted irrigation, in particular in late spring, and by harvesting at an earlier time.


Alfalfa (Medicago sativa L.)Drought stressWater regimesNutritive value


Grassland remains an important feed source for ruminant nutrition, with its high productivity and good fodder quality [1], but alfalfa is often a necessary feed additive or alternative, especially suitable for feed production under nitrogen-limiting conditions, due to the plant’s ability to fix atmospheric N2 [2]. With the increase of energy costs, fertiliser (as an artificial source of soil N) has become more expensive, a trend that is expected to continue in the future, which will likely further increase the need of legume production, including alfalfa [3, 4]. Agricultural forage production depends on an adequate water supply [5], a dependence that can become problematic in semi-arid climates, especially where local effects due to climate change increase the probability of summer droughts [68]. Insufficient water supply can strongly affect the production of forage legumes [9], resulting in decreasing yields, depending on the severity and duration of drought stress [10, 11].

It is well known that forage legumes differ in drought stress sensitivity [12]. White clover is one of the most important legumes in agricultural production, but it is also relatively drought sensitive [13]. Drought stress can have variable effects on plant content. Under drought conditions, concentrations of acid detergent fibre (ADF) and neutral detergent fibre (NDF) were reduced in a range of forage legumes, while inconsistent changes were reported in crude protein (CP) concentrations [14]. Another study reported an increase in water-soluble carbohydrates (WSC) under water shortage in two cultivars of soybean [15], while for clover species only a small drought-induced effect on WSC was observed [16]. One study described an increase in ADF with a minor effect only for CP and NDF concentrations in red clover and alfalfa [17]. However, there is limited and inconsistent knowledge available about the influence of drought stress on the nutritive value of alfalfa. This legume is possibly less sensitive to drought than white clover, but more research is needed to predict the influence of drought stress on the nutritive value of alfalfa, which was one of the aims of this study.

Alfalfa was used here to examine the effects of drought stress on the concentrations of CP, WSC and the fibre components NDF and ADF, which were chosen as indicators for nutritive value. The CP concentration is an essential component for ruminant nutrition, and is typically high in alfalfa due to effective N fixation; WSC have a positive influence on fodder intake and are important for an efficient utilisation of dietary N; NDF content provides an estimate of the cellulose, hemicellulose and lignin content and is inversely related to voluntary fodder intake; finally ADF includes lignin and cellulose and is negatively correlated with cell wall digestibility [18]. In the present study the following questions were addressed: (1) do different alfalfa species differ in their response to drought stress; and (2) what is the effect of timing, duration, and intensity of water restriction on the nutritional parameters of alfalfa.


Field experiments

A field study was conducted in 2012–2014 in a vegetation area of Ar Horqin Banner near the Nei Monggol Autonomous Region, China (coordinates 37°43’N; 120°22′E), using a randomized block design with three variable parameters that were tested with four replications. Two types of alfalfa were compared: Gold Queen and Suntory alfalfa. Conditions resembling severe draught and moderate draught were compared with optimal water supply whereby the timing of water restriction was varied. The plants were either harvested early during initial flowering, or late during full bloom. Details about the applied water regimes are described in Table 1. Drought stress was imposed during three periods with variable severity. The trials were divided into two experiments. In Experiment I (2013, spring through summer, sowing date 23 July 2012, final harvest date 14 September 2013), three restricted water supply regimes were compared to the optimal water supply control: severe drought stress during spring, and moderate and severe drought stress during summer. Moderate stress corresponded with 20–40% usable water capacity of the soil and severe stress corresponded to only 10–15%. In Experiment II, conducted a year later (sowing date 16 August 2013, final harvest date 21 September 2014), severe water restriction (15% available water) was applied either in early spring or in late spring, and this was compared to rain fed and optimally watered plots.
Table 1

The applied water regimes during field experiments and the yields of early and late harvest


Water regime, condition nr.

Early yield DM (t/ha)

Late yield DM (t/ha)

Experiment I (2013)

Optimal, control



Moderate drought in spring



Severe drought in spring



Moderate drought in summer



Severe drought in summer



Experiment II (2014)

Optimal, control



Severe drought in early spring



Severe drought in late spring






Drought stress was implemented by restricting rain precipitation on individual plots, using a 12 m long, 6 m wide, and 5 m high foil cover (CASADO, Dou-ville, France). This stationary shelter was covered by 200-μm polythene foil, which was mounted over the plot. In order to attain good ventilation and to minimize microclimate effects of the shelter, the front and the sides were left open.

Alfalfa types

Gold Queen alfalfa is a novel American cultivar, marketed as a salt-tolerant type that is suitable for high saline-alkali soil. Suntory alfalfa is a French cultivar with a high yield of high quality and is known to be disease-resistant.

The seeds were kindly provided by the College of Agriculture and Animal Husbandry, ChiFeng, China. The seeding density was 0.5 kg seed per 0.667 hm2 and plot sizes ranged from 6.6 to 8.2 m2.

Climate conditions

Air temperature and precipitation were recorded at 2 m height with an iMETOS weather station (Pessl Instruments, Weiz, Austria) located on the experimental site. The agrometeorological advisory system from the China Weather Service (CWS, 2014) was used to plan the irrigation schedule.

Soil composition and water content

The soil was characterized as Haplic Luvisol with an available water capacity of 120 mm (0–90 cm), and a groundwater level 10 m below surface. The soil was composed of 36% corn soil, 27% sand, 12% chernozem and 5% of other components. The pH of the soil (in CaCl2 suspension) measured in summer 2013 was 7.3. The soil moisture was recorded during the experiments using a portable soil moisture probe Diviner 2000 (Santé Technologies, Stepney, Australia). Plastic tubes with a diameter of 5 cm were installed to a depth of up to 150 cm. Soil moisture readings were taken at 10 cm intervals from 5 to 125 cm three times per week from the beginning of vegetation to harvest. The soil water content was also determined gravimetrically on several occasions in order to obtain a site-specific calibration (R2 = 0.64). The soil moisture data over time are presented as precentage, calculated as ml/100 g soil.

Sampling and measurements

For each experiment the plots were harvested by hand. For harvests, over an area of 0.09 m2 in Experiment 1, and 0.18 m2 in Experiment 2 per plot, the plants were cut at a height of 3–4 cm above the soil surface and cuttings were separated from weeds immediately after harvest. Dry weight of alfalfa harvest was determined after drying at 60 °C for 72 h in a drying oven (ULM 800, Member GmbH, Schwa Bach, Germany).

For analysis of CP, NDF, ADF and WSC, dried samples were ground to 1 mm particles and these were analysed by near-infrared reflectance spectroscopy (NIRS). All findings are reported as % dry mass (%DM). The spectra were analysed using a large dataset of calibration samples from different kinds of grasslands available from the Institute VDLUFA Qualitätssicherung NIRS GmbH, Kassel, Germany.

Statistical analyses

Analyses of variance were carried out with the GLIMMIX procedure of SAS 9.3 (SAS Institute, Cary, NC, USA). We performed a three factorial analysis of variance (ANOVA) for CP, NDF, ADF and WSC concentrations of the two cultivars for the two harvests [19]. The three factors were legume cultivar (LS), flowering phase (FS) and drought stress (DS). Relationships between selected variables were examined with a linear regression model. Experiment I and II as well as individual years were analyzed separately. Correlations were calculated with the CORR procedure of SAS. Graphs were created with SigmaPlot 12 (Systat Software Inc., Chicago, IL, USA).


In the two field experiments, conducted in two consecutive years, the two cultivars Gold Queen and Suntory alfalfa were subjected to various levels of drought. The plants were harvested at two developmental stages. In Experiment I, conducted in 2013, two drought levels were simulated during spring and two during summer (moderate spring drought, severe spring drought, moderate summer drought and severe summer drought); optimally watered plots served as control. In Experiment II, conducted a year later, severe drought was implemented either in early spring or in late spring, while optimal watering (control) was also compared to natural rainfall without further irrigation (rainfed). In both experiments, for all conditions, half of the plots were harvested early, during the initial blooming phase, and the other half was harvested later, at full blooming stage. The conditions tested are summarized in Table 1. All plots within one experiment were harvested at the same time, and the harvested alfalfa was chemically analyzed for a number of nutritional variables.

Figure 1 shows the relative available water in the soil during the various treatments of experiment I (2013) and of the treatments during experiment II (2014).
Fig. 1

Soil moisture levels (ml water/100 mg soil) over time for optimally watered control (a) and experimental conditions during Experiment 1 of moderate spring drought (b), severe spring drought (c), moderate summer drought (d) and severe summer drought (e), and for the experimental conditions of Experiment 2 corresponding to early spring severe drought (f), late spring severe drought (g), rainfed (h) and control (i). In each plot, two average moisture levels of four individual plots are shown, for those harvested early (solid lines) and late (dotted lines). The periods of artificially induced drought are indicated in grey. The three arrows indicate the time of early harvest, late harvest and final harvest

A graph showing local precipitation and temperature during the experiments is available as Additional file 1: Figure S1. In 2013, spring was warm and dry while summer temperatures were moderate. Yearly total rainfall amounted to 224 mm with an average temperature of 16.9 °C (Additional file 1: Figure S1). The year 2014 started with low temperatures, followed by a cool spring, with total rainfall (327 mm) and average air temperature (15.7 °C) lower than in 2013. The beginning of 2014 was relatively cool, which delayed vegetation development, while May was unusually wet with over 100 mm rainfall (Additional file 1: Figure S1).

The alfalfa early and late yields in t/ha, after removal of weeds and after drying (dry mass, DM), are shown in Table 1.

Crude protein concentration

Crude protein concentrations from the harvested alfalfa were determined and expressed as percentage of dry mass (% DM) (Table 2). Whereas moderate drought applied during the spring of 2013 had no significant effect on CP concentration, severe spring drought resulted in significantly lower CP values compared to optimally watered controls, for both alfalfa types, and for both harvest times (Table 2). ANOVA statistical analysis indicated there was a significant difference (P < 0.001) for CP content between the compared cultivars, with Gold Queen alfalfa producing significantly higher CP contents under all conditions tested. Early harvest produced significantly higher yields than late harvests, for both cultivars. The overall range of CP content of alfalfa grown with drought stress during spring varied between 15.61 and 17.61% DM. When drought stress was applied during the summer of 2013, CP content varied between 14.68 and 18.73% (Table 2), a range that was not significantly different to that obtained following spring drought stress. The response of the two alfalfa cultivars to drought stress applied in summer was again significantly different, and CP content was lower when summer drought stress was severe compared to moderate stress. As before, CP content was higher in Gold Queen alfalfa than in Suntory alfalfa, for all conditions tested. The ANOVA analysis further indicated that as a result of summer drought stress, Gold Queen suffered significantly more than Suntory alfalfa, resulting in a significance for DSxCV (Table 2).
Table 2

Crude protein fraction (%DM) of alfalfa in Experiment I, grown under drought stress during spring (top) and during summer (middle), and in Experiment II (bottom). Means of 4 analyses per plot are shown

Crude Protein fraction (CP, %DM)



Gold Queen alfalfa

Suntory alfalfa

Early harvest

Late harvest

Early harvest

Late harvest

Spring 2013


19.60 ± 0.16a

18.19 ± 0.15a

18.45 ± 0.12a

17.17 ± 0.14a

Moderate spring drought

18.43 ± 0.17a

17.69 ± 0.13a

17.45 ± 0.62a

16.17 ± 0.24a

Severe spring drought

17.61 ± 0.17b

15.61 ± 0.21b

16.48 ± 0.16b

14.25 ± 0.19b


20.30 ± 0.23a

19.63 ± 0.27a

18.68 ± 0.10a

18.32 ± 0.15a

Moderate summer drought

18.73 ± 0.05b

18.32 ± 0.08b

17.76 ± 0.24b

17.31 ± 0.15b

Severe summer drought

17.33 ± 0.22c

15.84 ± 0.35c

16.43 ± 0.26c

14.68 ± 0.25c

Spring 2014


18.98 ± 0.13a

17.80 ± 0.19a

17.97 ± 0.13a

16.78 ± 0.22a

Early spring severe drought

20.02 ± 0.25b

16.17 ± 0.14b

16.46 ± 0.21b

14.56 ± 0.16b

Late spring severe drought

18.65 ± 0.30a

15.82 ± 0.23c

15.49 ± 0.19b

13.43 ± 0.24c


16.73 ± 0.25c

15.27 ± 0.28c

16.21 ± 0.19b

14.54 ± 0.16b

*Superscripts indicate significant differences (P < 0.01) per column. Identical superscripts per column indicate there was no significant difference. The same applies to all other tables

In 2014, the timing of drought stress during spring was varied. A dry period early in spring reduced CP content of alfalfa compared to the optimally watered control, with the exception of the Gold Queen cultivar harvested early, which actually contained a higher ratio of CP compared to the control (Table 2). A period of drought in late spring affected all plots, except for Gold Queen alfalfa harvested early. Natural rainfall without irrigation significantly reduced the CP content (Table 2). For this experiment ANOVA analysis indicated that the combination of drought stress and harvest time produced significantly different results (a late harvest after late-spring drought produced lowest yields). Likewise, the combined factors of drought stress and cultivar, of harvest time and cultivar, and the combination of all three parameters were all significant. The highest CP content was obtained with Gold Queen harvested early following a severe drought in early spring. The lowest CP content was obtained with Suntory alfalfa harvested late following a severe drought in late spring.

Concentration of neutral and acid detergent fiber

The concentration of neutral detergent fibre (NDF) was higher in Suntory than in Gold Queen alfalfa, but the difference was not significant (P = 0.0908). Early harvest resulted in a significantly lower fraction of NDF than late harvest (Table 3). Independently of the time of harvest, the content of NDF was increased by severe drought stress in spring or summer, though a moderate drought during spring or summer had no effect on NDF content. During spring, both an early and a late drought increased the NDF fraction equally (Table 3).
Table 3

Neutral detergent fibre fraction (NDF, expressed as % DM) of alfalfa in Experiments I and II, means (n = 4)

Neutral Detergent Fibre fraction (NDF, %DM)



Gold Queen alfalfa

Suntory alfalfa

Early harvest

Late harvest

Early harvest

Late harvest

Spring 2013


47.63 ± 0.27a

49.96 ± 0.42a

48.93 ± 0.37a

51.44 ± 0.17a

Moderate spring drought

45.13 ± 0.17a

46.16 ± 0.12a

46.93 ± 0.27a

50.44 ± 0.67a

Severe spring drought

50.54 ± 0.21b

55.14 ± 0.41b

50.39 ± 0.32b

54.37 ± 0.65b


47.03 ± 0.32a

50.25 ± 0.27a

47.58 ± 0.21a

50.93 ± 0.27a

Moderate summer drought

47.92 ± 0.44a

50.15 ± 0.52a

47.99 ± 0.63a

51.59 ± 0.51a

Severe summer drought

50.33 ± 0.24b

55.35 ± 0.30b

50.97 ± 0.65b

55.84 ± 0.64b

Spring 2014


44.99 ± 0.47a

47.74 ± 0.55a

45.51 ± 0.22a

49.67 ± 0.18a

Early spring severe drought

48.37 ± 0.34b

53.30 ± 0.48b

49.22 ± 0.25b

52.58 ± 0.54b

Late spring severe drought

48.89 ± 0.31b

53.46 ± 0.50b

49.84 ± 0.48b

53.28 ± 0.55b


44.43 ± 0.52a

49.09 ± 0.45c

45.74 ± 0.35a

49.70 ± 0.22a

The concentration of fibre extracted with acid detergent (ADF) in part followed the same trends as those observed for NDF. ADF was also higher in Suntory than in Gold Queen, and this difference was now significant (Table 4). As was observed for NDF, late harvest increased the fraction of ADF, and both an early and a late period of drought during spring significantly increased the ADF fraction, as shown in Table 4. However, the ADF fraction was reduced under severe drought in spring and after a moderate drought in summer, while a severe drought in summer increased the ADF fraction (Table 4).
Table 4

Acid detergent fibre fraction (ADF, expressed as % DM) of alfalfa in in Experiments I and II, means (n = 4)

Acid Detergent Fibre fraction (ADF, %DM)



Gold Queen alfalfa

Suntory alfalfa

Early harvest

Late harvest

Early harvest

Late harvest

Spring 2013


34.52 ± 0.33a

36.23 ± 0.29a

36.15 ± 0.47a

37.67 ± 0.40a

Moderate spring drought

33.12 ± 0.37a

35.23 ± 0.73a

35.15 ± 0.17a

36.67 ± 0.49a

Severe spring drought

32.25 ± 0.27b

33.73 ± 0.30b

34.23 ± 0.40b

34.09 ± 0.38b


35.36 ± 0.35a

38.66 ± 0.26a

36.69 ± 0.40a

39.51 ± 0.24a

Moderate summer drought

32.83 ± 0.18b

34.49 ± 0.22b

34.91 ± 0.34b

36.07 ± 0.43b

Severe summer drought

38.70 ± 0.31c

42.64 ± 0.32c

39.90 ± 0.35c

43.66 ± 0.17c

Spring 2014


36.51 ± 0.37a

39.66 ± 0.23a

37.93 ± 0.34a

40.41 ± 0.32a

Early spring severe drought

38.73 ± 0.23b

40.82 ± 0.24a

38.97 ± 0.26b

42.83 ± 0.32b

Late spring severe drought

39.82 ± 0.14c

43.93 ± 0.32b

40.14 ± 0.20c

43.64 ± 0.19c


36.24 ± 0.32a

38.90 ± 0.20a

37.27 ± 0.63a

40.00 ± 0.51a

Concentration of water-soluble carbohydrates

The fraction of water-soluble carbohydrates (WSC) was significantly higher in Gold Queen than in Suntory alfalfa under all tested conditions, and late harvest produced higher WSC fractions (Table 5). In Experiment I drought stress did not significantly affect these fractions, but in Experiment II, lower WSC fractions were obtained following a drought period in spring, with the exception of early harvested Suntory (Table 5).
Table 5

Water-soluble carbohydrate fraction (%DM) of alfalfa in in Experiments I and II, means (n = 4)

Water-soluble carbohydrate fraction (WSC, %DM)



Gold Queen alfalfa

Suntory alfalfa

Early harvest

Late harvest

Early harvest

Late harvest

Spring 2013


6.58 ± 0.04a

6.84 ± 0.02a

6.40 ± 0.04a

6.72 ± 0.48a

Moderate spring drought

6.52 ± 0.14a

6.74 ± 0.92a

6.40 ± 0.14a

6.69 ± 0.41a

Severe spring drought

6.53 ± 0.03a

6.83 ± 0.02a

6.38 ± 0.03a

6.70 ± 0.05a


6.52 ± 0.03a

6.83 ± 0.01a

6.38 ± 0.02a

6.66 ± 0.05a

Moderate summer drought

6.52 ± 0.03a

6.82 ± 0.02a

6.38 ± 0.02a

6.66 ± 0.05a

Severe summer drought

6.57 ± 0.04a

6.87 ± 0.02a

6.50 ± 0.06b

6.74 ± 0.05a

Spring 2014


6.5 ± 0.03a

6.82 ± 0.02a

6.36 ± 0.03a

6.70 ± 0.04a

Early spring severe drought

6.34 ± 0.02b

6.71 ± 0.02b

6.31 ± 0.04a

6.55 ± 0.05b,c

Late spring severe drought

6.28 ± 0.02b

6.62 ± 0.03c

6.30 ± 0.03a

6.46 ± 0.07c


6.48 ± 0.03a

6.80 ± 0.02a

6.33 ± 0.03a

6.66 ± 0.04a,b

Hay yield reduction

At the end of the experiment all plant material was cut 3 cm above the surface and dried. Hay yields in Gold Queen alfalfa and Suntory alfalfa were reduced by all water-restricted conditions (Table 6), with a reduction between 27% (rain fed) and 83% (Gold Queen alfalfa, draught in late spring).
Table 6

Hay yield reduction (%) as a result of drought stress at the end of the experiments



Yield of Gold Queen hay

Yield of Suntory hay

Hay yield (kg/ha)


Hay yield (kg/ha)


Experiment I






Moderate spring drought





Severe spring drought





Moderate summer drought





Severe summer drought





Experiment II






Early spring severe drought





Late spring severe drought










n.a not applicable


The field experiments described here were conducted to assess the effect of water restriction regimes on two alfalfa cultivars, by the use of fixed and mobile rain shelters. The validity of such an approach to study microclimate effects has been convincingly demonstrated before [20]. We are aware that the shelters may have increased the temperature above ground, especially during hot days in summer, which may have affected plant development [21], as it would add heat stress to the plants in addition to drought stress. Experiments conducted in winter would not suffer from this combined effect, possibly resulting in smaller changes, for instance like those reported for grain yield studied in winter wheat [22, 23]. However, since reduced rainfall as a result of local climate changes often coincides with higher than normal temperatures, we believe our experimental conditions sufficed to investigate their combined effect. Although temporarily increased day temperatures might have negatively affected the growth and yield of alfalfa, water limitation was most likely the main driver of the observed changes. Irrespective of the water supply treatment, the analytical data of the harvested alfalfa resulted in a predictive nutritive value comparable to data described in the literature [14, 24]. With the obtained values, the harvested alfalfa could be considered a moderate to high quality feed [25, 26]. Under optimal water conditions, the parameters for feed quality were better for Gold Queen alfalfa than for Suntory alfalfa, with higher protein and lower fiber contents, though the higher water-soluble carbohydrate fraction in Gold Queen could be considered less beneficial.

The fraction of crude protein was reduced as a result of drought, with a similar decrease in both cultivars, although CP content remained higher in Gold Queen than in Suntory. The CP fraction generally depends on the amount of available N [1, 27] and alfalfa is particularly effective in N-fixation. When the N fixation performance of alfalfa was determined, it produced 10 to 30% higher fixation levels than other legumes [28]. Thus, the degree of nitrogen fixation determines the availability of N for protein production, but it is not the only limiting factor for biomass production: obviously, this is also determined by water availability. Experiments with soybean identified N uptake as an important factor for biomass production under drought [15], and peanut plants decreased their N fixation under drought stress [29]. We interpret the decrease in CP fractions in drought-stressed alfalfa to be caused by a combined stress response to water limitation in addition to a decrease in N fixation.

The content of neutral fiber increased under strong drought stress, a change that was also observed for acid-extracted fiber under certain conditions, though mixed results were obtained for the latter. Fiber concentration is influenced by many interacting factors, such the phase of plant development, leaf-to-stem ratio, environmental conditions (water, temperature, available light etc.), and the availability of nutrients [2426]. The increase in NDF and ADF fractions under stress is not supported by findings in the literature, where a delayed maturity under drought was reported, associated with lower NDF and ADF concentrations [14, 25]. The major difference between the NDF and the ADF fraction is that the former included hemicellulose (the other main components are cellulose and lignin for both fractions), and the stronger and more consistent increase of the NDF as a result of drought stress in alfalfa suggests that production of hemicellulose is most affected by water restriction. However, results on the effects of drought on hernicellulose concentrations are inconsistent in the literature, as some authors have reported decreased hemicellulose concentration under drought, while other reported an increase [30]. We found that the ADF concentration was consistently lower than that of NDF, a finding that has been reported for other legumes and for most grasses as well [26]. A lower fiber concentration is generally considered beneficial, as it may lead to a higher herbage intake and to an increase in digestibility of forage. An early harvest resulted in lower fiber content and this, combined with a higher protein content, suggests that harvesting early in the season may improve the quality of the alfalfa, particularly after drought.

The fraction of WSC was least affected by drought stress, producing only a minor decrease as a result of drought stress in spring, although others have reported an increase as a result of drought in other plant species [15, 31]. Gold Queen alfalfa contained significantly higher fractions of WSC, which might explain why it was also generally more capable to cope with drought stress. A high WSC concentration in plants would result in a higher osmotic potential, which drives the uptake of soil water and is therefore of importance to minimize drought stress effects [32]. This osmotic adjustment is a physiological mechanism in response to drought [15], but in our experiments the WSC content changed marginally, only producing a significant decrease during spring drought.

Without irrigation, yields were low and nutritional parameters poor, as demonstrated by the samples grown under natural rainfall. When water supply is limited and continuous irrigation may not be possible, the timing of irrigation needs to be carefully considered. Our results indicate that the most beneficial effect can be expected if irrigation prevents a severe drought in late spring.

Digestibility of fodder may decrease under strong drought stress due to a tendency to lower WSC and higher fiber fractions, and combined with lower protein content this would reduce the nitritive value. However, the decreased protein-to-fiber ratio in alfalfa following a drought would result in a decrease of nitrogen secretion in the urine of ruminants [20], which can be considered beneficial for the environment. The choice of cultivar (Gold Queen) and an early harvest can minimize drought effects. Animal experiments need to be performed to further assess the feed-to-weight conversion and waste production of alfalfa grown under drought stress.


The production of alfalfa is a main agricultural activity in areas in China where relatively sandy and infertile soils limit other crops to be produced, but in particular these areas are expected to suffer from increased periods of drought as a result of climate change. It is therefore important to anticipate possible changes in the nutritive value of alfalfa as a result of drought stress. We have demonstrated that only severe drought stress has an impact on yield and composition of alfalfa. Strong drought led to a decrease in hay yield, a decrease in CP content, and an increase in fibre. These effects might in combination decrease the digestibility of the herbage. However, as the ratio of CP to WSC decreased under drought, this could reduce the N surplus in ruminates. We observed differences between the two tested alfalfa cultivars, both in their performance under optimal water supply and in their response to drought stress, with Gold Queen performing better than the Suntory cultivar. Finally, an early harvest could minimize the effects of drought. The reported findings may assist farmers in choosing the best cultivar, irrigation strategy and harvesting time, to mitigate the effect of decreased precipitation that can be expected in the future.



Acid detergent fibre


Crude protein


Dry mass


Neutral detergent fibre


Near infrared reflectance spectroscopy


Water soluble carbohydrates



We are most grateful to all community members in the study region for their time and their confidence in our research. We also sincerely thank the staff of the provincial health department in Chifeng. The support of Zhu GD, Wang Mingchao, Fan Wenqiang, and Cheng Qiming for their help with the field experiments is gratefully acknowledged.


This work was supported by the National Natural Science Foundation of China (Grant No. 31360585). Inner Mongolia Agricultural University receives strategic funding from National Natural Science Foundation of China. The funding body had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Authors’ contributions

LYH and JYS conceived the study. GGT and LYH designed the experiments. WQ performed the fieldwork. HGD supervised the fieldwork. WQ performed the quantitative data analysis. LYH wrote the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The seeds were kindly provided by the College of Agriculture and Animal Husbandry, ChiFeng, China. The experimental research on plants, including collection of plant material, complied with institutional, national, or international guidelines and complied with the Convention on the Trade in Endangered Species of Wild Fauna and Flora. These two materials have been deposited in a publicly available herbarium. All field experiments were conducted in accordance with Environmental Protection law of the People’s Republic of China and did not involve endangered or protected species.

Consent for publication

Not applicable

Competing interests

All authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Authors’ Affiliations

Key Laboratory of Forage Cultivation, Processing and High Efficient Utilization, Ministry of Agriculture, Hohhot, China
Laboratory of Grassland Resources, Ministry of Education, Hohhot, China
College of Grassland, Resources and Environment, Inner Mongolia Agricultural University, Hohhot, China


  1. Watson CA, Bengtsson H, Ebbesvik M, Løes A-K, Myrbeck A, Salomon E, et al. A review of farm-scale nutrient budgets for organic farms as a tool for management of soil fertility. Soil Use Manag. 2002;18:264–73.View ArticleGoogle Scholar
  2. Jensen ES, Hauggaard-Nielsen H. How can increased use of biological N2 fixation in agriculture benefit the environment? Plant Soil. 2003;252:177–86.View ArticleGoogle Scholar
  3. Crews TE, Peoples MB. Can the synchrony of nitrogen supply and crop demand be improved in legume and fertilizer-based agroecosystems? A review Nutr Cycl Agroecosys. 2005;72:101–20.View ArticleGoogle Scholar
  4. Heinrich J. The legumes expert forum. Science, economy and society-making ecosystem services from legumes competitive. Braunschweig: German Agricultural Research Alliance (DAFA); 2012. p. 12–47.Google Scholar
  5. Hopkins A, Del Prado A. Implications of climate change for grassland in Europe: impacts, adaptations and mitigation options: a review. Grass Forage Sci. 2007;62:118–26.View ArticleGoogle Scholar
  6. Alcamo J, Moreno JM, Novaky B. Contribution of working group II to the fourth assessment report of the intergovernmental panel on climate change. Cambridge: Cambridge University Press; 2007. p. 541–80.Google Scholar
  7. Schindler U, Steidl J, Muller L, Eulenstein F, Thiere J. Drought risk to agricultural land in northeast and Central Germany. J Plant Nutr Soil Sci. 2007;170:357–62.View ArticleGoogle Scholar
  8. Trenberth KE. Changes in precipitation with climate change. Clim Res. 2011;47:123–38.View ArticleGoogle Scholar
  9. Foulds W. Response to soil moisture supply in three leguminous species I. Growth, reproduction and mortality. New Phytol. 1978;80:535–45.View ArticleGoogle Scholar
  10. Farooq M, Wahid A, Kobayashi N, Fujita D, Basra SM. A. Plant drought stress: effects, mechanisms and management. Agron Sustain Dev. 2009;29:185–212.View ArticleGoogle Scholar
  11. Jaleel CA, Manivannan P, Wahid A, Farooq M, Al-Juburei HJ, Somasundaram R, et al. Drought stress in plants: a review on morphological characteristics and pigments composition. Int J Agric Biol. 2009;11:100–5.Google Scholar
  12. Dierschke H, Briemle G. Kulturgrasland. Wiesen, Weiden, und verwandte Staudenfluren. Stuttgart: Ulmer Verlag; 2002. p. 212–7.Google Scholar
  13. Marshall AH, Rascle C, Abberton MT, Michaelson-Yeates TPT, Rhodes I. Introgression as a route to improved drought tolerance in white clover (Trifolium repens L.). J Agron Crop Sci. 2001;187:11–8.View ArticleGoogle Scholar
  14. Peterson PR, Sheaffer CC, Hall MH. Drought effects on perennial forage legume yield and quality. Agron J. 1992;84:774–9.View ArticleGoogle Scholar
  15. Nakayama N, Saneoka H, Moghaieb REA, Premachandra GS. Response of growth, photosynthetic gas exchange, translocation of 13C-labelled photosynthate and N accumulation in two soybean (Glycine max L. Merrill) Cultivars to drought stress. Int J AgricBiol. 2007;9:669–74.Google Scholar
  16. Abberton MT, Marshall AH, Michaelson-Yeates TPT, Williams TA, Rhodes I. Quality characteristics of backcross hybrids between Trifolium repens and Trifolium ambiguum. Euphytica. 2002;127:75–80.View ArticleGoogle Scholar
  17. Seguin P, Mustafa AF, Sheaffer CC. Effects of soil moisture deficit on forage quality, digestibility, and protein fractionation of Kura clover. J Agron Crop Sci. 2002;188:260–6.View ArticleGoogle Scholar
  18. Sarwar N, Mahmood S, Nisa MU, Bilal MQ. Factors affecting digestibility of feeds in ruminants. Int J Agric Biol. 1999;1:366–72.Google Scholar
  19. Payne. Gene expression profiling in aspergillus flavus. Mycopathologia. 2002;155:13.Google Scholar
  20. Moorby JM, Evans RT, Scollan ND, MacRae JC, Theodorou MK. Increased concentration of water-soluble carbohydrate in perennial ryegrass (Lolium perenne L.). Evaluation in dairy cows in early lactation. Grass Forage Sci. 2006;61:52–9.View ArticleGoogle Scholar
  21. Brisson N, Casals ML. Leaf dynamics and crop water status throughout the growing cycle of durum wheat crops grown in two contrasted water budget conditions. Agron Sustain Dev. 2005;25:151–8.View ArticleGoogle Scholar
  22. Porter JR, Gawith M. Temperatures and the growth and development of wheat: a review. Eur J Agron. 1999;10:23–36.View ArticleGoogle Scholar
  23. Mu H, Jiang D, Wollenweber B, Dai T, Jing Q, Cao W. Long-term low radiation decreases leaf photosynthesis, photochemical efficiency and grain yield in winter wheat. J Agron Crop Sci. 2010;196:38–47.View ArticleGoogle Scholar
  24. Fulkerson WJ, Neal JS, Clark CF, Horadagoda A, Nandra KS, Barchia I. Nutritive value of forage species grown in the warm temperate climate of Australia for dairy cows: grasses and legumes. Livest Sci. 2007;107:253–64.View ArticleGoogle Scholar
  25. Buxton DR. Quality-related characteristics of forages as influenced by plant environment and agronomic factors. Anim Feed Sci Tech. 1996;59:37–49.View ArticleGoogle Scholar
  26. Schwarz FJ. Rinderfütterung. In: Kirchgeßner M, Roth FX, Schwarz FJ, Stangl GI, editors. Tierernährung. Frankfurt: DLG-Verlag, Frankfurt am Main; 2008. p. 351–498. Google Scholar
  27. Zahran HH. Rhizobium-legume symbiosis and nitrogen fixation under severe conditions and in an arid climate. Microbiol Mol Biol Rev. 1999;63:968–89.PubMedPubMed CentralGoogle Scholar
  28. Gierus M, Kleen J, Loges R, Taube F. Forage legume species determine the nutritional quality of binary mixtures with perennial ryegrass in the first production year. Anim Feed Sci Tech. 2013;172:150–61.View ArticleGoogle Scholar
  29. Pimratch S, Jogloy S, Vorasoot N, Toomsan B, Kesmala T, Patanothai A, et al. Association of nitrogen fixation to water uses efficiency and yield traits of peanut. Int J Plant Prod. 2013;7:225–41.Google Scholar
  30. Al-Hakimi AMA. Counteraction of drought stress on soybean plants by seed soaking salicylic acid. Int J Bot. 2006;2:421–6.View ArticleGoogle Scholar
  31. Da Costa M, Huang B. Osmotic adjustment associated with variation in bentgrass tolerance to drought stress. J Amer Soc Hort Sci. 2006;131:338–44.Google Scholar
  32. Morgan JM. Osmoregulation and water stress in higher plants. Ann Rev Plant Physio. 1984;35:299–319.View ArticleGoogle Scholar


© The Author(s). 2018