Jinpeng Li,Zhimin Wng,Chunsheng Yo,Zhen Zhng,Yng Liu,Yinghu Zhng,*

a Anhui Agricultural University,School of Agronomy,Hefei 230036,Anhui,China

b College of Agronomy and Biotechnology,China Agricultural University,Beijing 100193,China

c Engineering Technology Research Center for Agriculture in Low Plain Areas,Cangzhou 061000,Hebei,China

Keywords:

A B S T R A C T Increased grain yield(GY)and grain protein concentration(GPC)are the two main targets of efforts to improve wheat(Triticum aestivum L.)production in the North China Plain(NCP).We conducted a three-year field experiment in the 2014–2017 winter wheat growing seasons to compare the effects of conventional irrigation practice(CI)and micro-sprinkling irrigation combined with nitrogen(N)fertilizer(MSI)on GY,GPC,and protein yield(PY).Across the three years,GY,GPC,and PY increased by 10.5%–16.7%,5.4%–8.0%,and 18.8%–24.6%,respectively,under MSI relative to CI.The higher GY under MSI was due primarily to increased thousand-kernel weight(TKW).The chlorophyll content of leaves was higher under MSI during the mid–late grain filling period,increasing the contribution of post-anthesis dry matter accumulation to GY,with consequent increases in total dry matter accumulation and harvest index compared to CI.During the mid–late grain filling period,the canopy temperature was markedly lower and the relative humidity was higher under MSI than under CI.The duration and rate of filling during the mid–late grain filling period were also higher under MSI than CI,resulting in higher TKW.MSI increased the contribution of post-anthesis N accumulation to grain N but reduced the pre-anthesis remobilization of N in leaves,the primary site of photosynthetic activity,possibly helping maintain photosynthate production in leaves during grain filling.Total N at maturity was higher under MSI than CI,although there was little difference in N harvest index.The higher GPC under MSI than under CI was due to a larger increase in grain N accumulation than in GY.Overall,MSI simultaneously increased both GY and GPC in winter wheat grown in the NCP.

1.Introduction

Wheat is a widely planted cereal food crop globally.The North China Plain(NCP)is the most important agricultural production area in China,with a cultivated area of 0.35 billion hectares.It constitutes 25%of the total arable land and provides 71%of the wheat yield of the entire country but receives less than 7%of China’s total water resources[1].The mean annual rainfall in the NCP is 562 mm,distributed mainly over the summer season.Precipitation during the winter wheat growing season is much lower,ranging from less than 50 mm in dry years to 200 mm in wet years[2,3].Winter wheat production in the NCP depends mainly on irrigation to increase grain yield(GY),resulting in a growing shortage of groundwater resources.The development of water-saving,highyield agriculture is of great importance to wheat production in the NCP.Crop quality also attracts increasing attention.Methods for increasing both wheat yields and grain quality in the NCP are needed.

Irrigation schedules and technologies greatly influence crop growth and yield[4–6].Compared to conventional surface irrigation practices,advanced irrigation systems,such as drip or sprinkler irrigation technologies,simultaneously improve water use efficiency and grain production in winter wheat[7].Microsprinkling technology,which is based on surface drip irrigation and sprinkler irrigation,efficiently produces high yields.Its use in wheat production has increased gradually in recent years,and more efficient use of nitrogen(N)fertilizer can be achieved by combined application of water and fertilizer[5,8].The cost of micro-sprinkling is low compared to that of other microirrigation systems[9],and the economic benefit is increased in comparison with that of conventional irrigation practice[10].In one study[7],compared to conventional irrigation(CI),suitable soil moisture and high soil nitrate N levels were maintained in topsoil under micro-sprinkling irrigation combined with N fertilizer during the grain-filling period,significantly increasing dry matter accumulation during grain filling.Grain weight,a primary factor determining yield,is affected by many factors,including grainfilling duration,grain-filling rate,and canopy environment[11–13].In previous studies[8,14],optimization of irrigation delayed flag leaf senescence and increased the photosynthetic rate in leaves after anthesis,resulting in higher grain weight and crop yield.Multiple studies[13,15,16]have shown that canopy temperature during the grain filling period greatly affects grain yield and that a cooler plant canopy during the middle of the grain filling period tends to increase winter wheat yield.

N is of paramount importance to crop yield[17,18].N fertilizer is generally the most effective input for increasing GY and grain protein concentration(GPC)in wheat production[17–20],but excessive or insufficient N supply can reduce both GY and GPC[21].The optimal N application rate to achieve high and efficient production of winter wheat in the NCP is 180–240 kg ha-1[22,23].Although GY and GPC are the major targets of wheat breeding efforts,simultaneous improvement of both has been difficult because of the negative genetic relationship between GY and GPC[24].It is generally believed that GPC decreases with increases in GY under equivalent N application rates,mainly because of the dilution effects of N in grain.Kichey et al.[25]reported that 50%–95% of grain N at harvest comes from the remobilization of N stored before anthesis,possibly owing to the application of N fertilizer mainly during the early stage(before the booting stage)of wheat growth.However,foliar N fertilizer applied later in the growth period,in particular at anthesis or during grain filling,often increases GPC[26,27].In our previous studies[5,7],MSI integrated with N fertilizer during the grain filling period significantly increased N uptake and accumulation after anthesis,and N fertilizer use efficiency and grain N accumulation also increased significantly.The mechanism by which micro-sprinkling irrigation combined with N fertilizer increases grain weight and protein accumulation invites investigation.

We hypothesized that micro-sprinkling irrigation combined with N fertilizer would simultaneously increase GY and GPC by moderating the canopy environment and reducing the conflict between dry matter production and N remobilization during the grain-filling period.Such an effect would delay canopy senescence and increase the grain-filling rate during the mid–late grain-filling period,thus achieving simultaneous increases in GY and GPC.To test our hypothesis,we conducted a three-year field study to investigate the effects of conventional irrigation practice and microsprinkling irrigation on 1)grain yield and yield components of winter wheat,dry matter accumulation,harvest index,and grainfilling characteristics,2)N accumulation,nitrogen harvest index,and grain protein concentration,3)canopy temperature and relative humidity during grain filling and leaf senescence after anthesis;to characterize the relationships among GY,GPC,and the accumulation and remobilization of dry matter and nitrogen;and to determine whether MSI can simultaneously improve GY and GPC in winter wheat grown in the NCP.

2.Materials and methods

2.1.Field site and experimental design

The field experiment was conducted during the 2014–2017 winter wheat growing seasons at Wuqiao Experimental Station of China Agriculture University located in Wuqiao county(37°41′N,116°37′E),Hebei province,China.The soil type in the experimental field was light loam consisting of 11.8% clay,78.1%silt,and 10.1% sand.The maximum soil field capacity and wilting point of crops in the experimental field were 27.6%(g g-1)and 8.6%(g g-1),respectively.In every year of the trial,summer maize was the crop preceding wheat.The organic matter,total N,available potassium,and available phosphorus content in the 0–40 cm soil layer was 1.17%,0.95 g kg-1,104.4 mg kg-1,and 29.2 mg kg-1,respectively.Total precipitation during the three wheat growing seasons was 168.5 mm in 2014–2015,127.7 mm in 2015–2016,and 95.5 mm in 2016–2017.

Two irrigation methods were applied during the wheat growing seasons:conventional flood irrigation practice(CI,irrigation at the jointing and anthesis stages,60 mm at each stage)and microsprinkling irrigation(MSI,irrigation at the jointing,booting,anthesis,and filling stages,30 mm each stage),as described by Li et al.[5].Surface flooding was performed with PVC pipe.The microsprinkling hose design and details of the orifice arrangement followed Man et al.[28].The micro-sprinkling hose was 30 m long with a flow rate of 6.0 m3h-1,and the sprinkling angle of the hose was 80°.Each experimental plot consisted of 24 rows of wheat plants spaced 15 cm apart.The inter-row spaces between wheat rows were designated as L1–L23.The micro-sprinkling hose was laid between L6 and L18(so that the sprinkling range was 1 m on either side of the hose).Well water was used for irrigation.The total N fertilizer application rate was 195 kg ha-1,with 105 kg ha-1as base fertilizer and 90 kg ha-1as topdressing following Li et al.[5].Before sowing,105 kg N ha-1,120 kg P2O5ha-1,and 90 kg K2O ha-1were applied as base fertilizer.For each MSI treatment,urea(22.5 kg N ha-1)was completely dissolved in a fertilization device and applied as topdressing together with the irrigation water,whereas 90 kg N ha-1was spread over the field for CI at the jointing stage.Each experimental plot was 4 m×30 m,and the experimental design was a randomized complete block design with three replications.The location of each experimental plot in all three years was the same.The high-yielding wheat cultivar Jimai 22,widely planted in the NCP,was used.Wheat was sown on October 15,2014,October 18,2015,and October 14,2016 at a planting density of approximately 540 plants m-2after emergence and harvested on June 15,2015,June 10,2016,and June 14,2017,respectively.

2.2.Sampling and measurement

2.2.1.Canopy temperature and relative humidity

Starting at the anthesis stage(Z61),instruments that automatically record temperature and relative humidity(MicroLite USB,V24.0.0,Fourtec-Fourier Technologies Ltd.,Rosh Haayin,Israel)were placed in each experimental plot to measure temperature and relative humidity daily until maturity(Z91).Temperature and humidity data were acquired hourly and exported by a data suite software for MicroLite to a computer.Each recorder was placed on a steel bracket at the height of the wheat flag leaf and was located at the center of the experimental plot in a white plastic weather-shutter box with a diameter of 15 cm and height of 20 cm.

2.2.2.Leaf chlorophyll content

To determine leaf chlorophyll(aandb)content,20 flag leaves,second leaves,and third leaves were randomly collected from each experimental plot at 5-day intervals from anthesis to maturity(three replicates each treatment).Leaf samples were chopped on ice,quickly weighed(200 mg),and extracted with 50 mL 95%ethanol for 48 h in the dark.Optical density(OD)values of the extracts were measured at 649 and 665 nm with a spectrophotometer[7].

2.2.3.Grain-filling dynamics

Fifty wheat ears that flowered on the same day on plants with similar overall plant height,ear height,and stem diameter were selected and tagged for each plot.Ten tagged ears from each plot were sampled at 7-day intervals from the tenth day after anthesis to maturity.The grain was collected and dried at 75 °C to a constant weight to determine dry weight.The grain-filling process was fitted with a logistic growth equation following Gao et al.[29]:W=A/(1+Be-Ct),whereWis the theoretical thousandkernel weight(TKW)(g),A is the maximum grain weight(g),tis the time after anthesis in days,andBandCare coefficients determined by regression.The filling phase,which differed from the physiology phase,was calculated from the derivative of the equation.The endpoint for grain filling of the early stage was defined asT1=(lnB-1.317)/C,the endpoint for grain filling of the middle stage was defined asT2=(lnB+1.317)/C,and the endpoint of grain filling for the late stage was defined asT3=(lnB+4.59512)/C.The dried grain was retained after dry weight measurement to determine N concentrations and to calculate grain protein concentration during grain filling.

2.2.4.Dry matter accumulation(DM)and plant nitrogen concentration analyses

To measure the DM and nitrogen concentration of winter wheat plants,two representative and adjacent 50-cm segments of rows(avoiding border rows)were selected at anthesis and maturity in each experimental plot and cut off at ground level.Sample plants were separated into stem and sheath,leaf,and chaff(ears without grain)at anthesis and into stem and sheath,leaf,and chaff and grain at maturity.The dry matter weight of each component was recorded after drying at 75°C for 72 h to constant weight.The total DM values at anthesis and maturity were calculated from the sum of the DM values of each component from each growth stage.As described by Arduini et al.[30],the remobilization of DM preanthesis(DMR)in each organ was calculated from the difference between DM values at anthesis and maturity,and the contribution of DMR to GY(DMRC)was calculated as the ratio of DMR to GY.The pre-anthesis total DM contribution(DMRCT)was the sum of the DMRC values of all plant organs,and the contribution of postanthesis DM accumulation to GY(DMPCT)was calculated as 100-DMRCT.Harvest index(HI)was calculated as the ratio of GY to total aboveground DM accumulation at maturity.

Total nitrogen concentration at anthesis and maturity was determined by the Kjeldahl method[31].Nitrogen accumulation(NA)was calculated following Ruisi et al.[32]:

where DMC is the DM accumulation of the plant components and NC is the N concentration of the plant components.

Total NA at anthesis and maturity was calculated as the sum of NA in each organ of the plant.The nitrogen remobilization(NR)of plant organs was calculated as the difference in N accumulation between anthesis and maturity,and the contribution of NR to grain N(NRC)was calculated as the ratio of NR to grain N at maturity[30].The total contribution of NR(NRCT)was the sum of the NRC values of all plant organs,and the contribution rate of NA post-anthesis to grain N(NPCT)was calculated as 100-NRCT.The N harvest index(NHI)was calculated as the ratio of grain N to total aboveground NA at maturity[33].GPC was calculated as grain N concentration×5.7[34].

2.2.5.Grain yield and protein yield

Spike number was determined by counting the spikes on six 1-m row segments in each plot before harvest.Kernel number per spike was determined by counting the kernels in each spike from 60 randomly selected plants in each plot before harvest.At maturity,wheat plants from a 3-m2area in each plot,which included 10 rows with two meters line length,were harvested and threshed for GY determination.GY was corrected to a 13%moisture basis.TKW was calculated by weighing 1000 seeds from each sample and calculating the mean of three replicates.Grain protein yield(PY)was calculated as GY×GPC at maturity.

2.3.Data analyses

Analyses of variance(ANOVA)were fitted using the general linear model procedure in SPSS 19.0(SPSS,IBM,USA).Combined ANOVAs were also fitted across years,irrigation methods,and their interactions.Treatment means in each year or over the three years were compared using the least significant difference test(P=0.05).Pearson correlations with three years’data were calculated with SPSS.Figures were made with OriginPro 2016 9.3(OriginLab,Northampton,MA,USA).

3.Results

3.1.Weather conditions during the study

Precipitation varied greatly among the three winter wheat growing seasons,with total precipitation of 168.5,127.7,and 95.5 mm for 2014–2015,2015–2016,and 2016–2017,respectively(Fig.1).More than 50%(93.7 mm)of the precipitation was concentrated in April and May in 2014–2015,and there were fewer rainfall events at the jointing stage(beginning of April)in the two subsequent years.The precipitation in the first half of April was 31.4,6.3,and 17.0 mm in 2014–2015,2015–2016 and 2016–2017,respectively.Although the precipitation in 2016–2017 was the lowest of the three growing seasons,the occurrence of rainfall was relatively compatible with the water requirements of wheat,such that some precipitation fell during the seedling period(October and November,30.7 mm)and during the jointing to heading stage(April and May,16.5 mm).The total rainfall was 61.2,33.4,and 25.8 mm during the grain-filling period(May and June)in the three growing seasons.

3.2.Canopy temperature and relative humidity

As shown in Fig.2,canopy temperature and relative humidity during the grain filling period were markedly affected by year and irrigation method.The highest mean air temperature during the grain-filling period was 29.0 °C in 2014–2015,27.5 °C in 2015–2016,and 31.8°C in 2016–2017.The air temperature during the grain-filling period was almost always lower than the canopy temperature during the 2014–2015 and 2015–2016 growing seasons,whereas it was higher than the canopy temperature on some days of the 2016–2017 growing season.Although there was little difference in canopy temperature between the two irrigation methods during the early grain-filling period(before 15 days),the canopy temperature was markedly higher under CI than under MSI during the later grain-filling period.The relative humidity of air at 14:00 during the grain-filling period was 43.2%,51.2%,and 49.4% in 2014–2015,2015–2016,and 2016–2017,respectively.Canopy relative humidity and temperature were significantly(P＜0.001)negatively correlated(–0.733,–0.720,and–0.514 in 2014–2015,2015–2016,and 2016–2017,respectively),indicating that the higher the temperature,the lower was the humidity.MSI maintained a lower canopy temperature and higher relative humidity during the late grain-filling stage of winter wheat than did CI.

Fig.1.Daily precipitation and mean air temperature recorded during the three growing seasons of winter wheat at the trial site.

3.3.Chlorophyll content

The chlorophyll content of flag and second leaves initially increased after anthesis and then decreased under both CI and MSI in all three years,whereas the chlorophyll content of third leaves(except 5 days after anthesis[DAA]under MSI)decreased continuously after anthesis until the lowest content was reached at maturity(Fig.3).Although there were some differences in leaf chlorophyll content under CI and MSI in different years,the changes were essentially consistent.The chlorophyll content of flag leaves was higher under MSI than under CI from DAA 15 to DAA 30 during the three growing seasons,in particular during the later period of grain filling(DAA 20–DAA 30),whereas there was little difference from anthesis to DAA 10.The chlorophyll content of second and third leaves was also maintained at a higher level in MSI than in CI during the later grain-filling period,but the difference was smaller than in flag leaves.Overall,MSI combined with N fertilization more effectively delayed the senescence of winter wheat leaves and maintained high levels of chlorophyll content in leaves during grain filling.

3.4.Dry matter accumulation and remobilization

Dry matter accumulation in stems and sheaths was significantly higher than in leaves and chaff at anthesis and maturity under both CI and MSI,and chaff showed the lowest DM accumulation(Table 1).The DM of stems and sheaths at anthesis was significantly higher in CI than in MSI in the 2014–2015 and 2016–2017 growing seasons,but there were no significant differences between the two irrigation methods in stem or sheath DM in the 2015–2016 growing season or in leaf and chaff DM in any of the three growing seasons.At maturity,the DM of each organ was significantly higher under MSI than under CI.The pre-anthesis dry matter remobilization(DMR)of the different plant organs and the total contribution rate of DMR to grain yield(DMRCT)decreased significantly under MSI relative to CI,whereas the contribution rate of DM post-anthesis to grain yield(DMPCT),the total DM at harvest,and harvest index(HI)increased greatly.DMRCT,DMPCT,total DM,and HI were significantly affected by irrigation method(I),year(Y)and the interaction between them(I×Y).Micro-sprinkling irrigation significantly increased the contribution of post anthesis dry matter to yield but not the pre-anthesis contribution,and increased the total DM and HI.Total DM,DMRCT,and DMPRCT were lower in 2015–2016 than in 2014–2015 and 2016–2017.In summary,relative to CI,MSI significantly increased total dry matter accumulation at maturity,significantly decreased pre-anthesis dry matter remobilization,and increased the contribution of post-anthesis dry matter to grain yield.

Fig.2.Effects of micro-sprinkling irrigation on canopy temperature and relative humidity during grain-filling period of winter wheat.CI,conventional flood irrigation practice;MSI,micro-sprinkling irrigation.*indicates significant difference between CI and MSI(P＜0.05).Values are means±standard error(n=3).

Fig.3.Effects of micro-sprinkling irrigation on chlorophyll content of leaves after anthesis of winter wheat.CI,conventional flood irrigation;MSI,micro-sprinkling irrigation.Asterisks indicate significant difference between CI and MSI(P＜0.05).Values are means±standard error(n=3).

3.5.Grain filling

3.5.1.Grain dry matter accumulation dynamics

As shown in Fig.4,the curves of grain dry matter accumulation were S-shaped for both irrigation methods.Across the three growing seasons,compared with CI,the grain dry matter accumulation in MSI was significantly lower during the early grain-filling stage but higher during the late grain-filling stage.The grain weight at maturity was significantly higher under MSI than under CI.In summary,MSI increased the grain weight primarily during the middle and late grain-filling stages and increased the grain weight at maturity.

3.5.2.Grain-filling parameters

As shown in Table 2,the simulated theoretical TKW(W)and the maximum grain-filling rate(Vmax)were significantly higher under MSI than under CI.The time to maximum filling rate(Tmax)and the endpoint for grain filling of the early stage(T1)were significantly delayed under MSI,and the filling rates in the middle(V2)and late stages(V3)were increased relative to CI.Across the three years,under the same irrigation method,the highest W was observed in the 2015–2016 growing season,and in comparison with 2014–2015 and 2016–2017,the ending date for grain filling of the middle stage(T2)and the late stage(T3)were extended in 2015–2016.In summary,MSI prolonged the grain-filling duration,delayed the timing of maximum grain filling,and increased the filling rate in the middle and late stages relative to CI,resulting in greater final grain weight.

3.6.Nitrogen accumulation and remobilization

Nitrogen accumulation(NA)was higher in stems and sheaths than in leaves and chaff at anthesis and maturity under both CI and MSI,and chaff showed the lowest NA(Table 3).At anthesis,there was no significant difference in the NA of stems and sheaths between CI and MSI in 2014–2015,whereas the NA of leaves and chaff was significantly lower under CI than under MSI.In 2015–2016,there was little difference in NA in the same plant organ between CI and MSI.In 2016–2017,the NA in stems,sheaths,and leaves was significantly higher under CI than under MSI,but there was no significant difference in chaff NA between CI and MSI.In summary,except in the 2016–2017 growing season,MSI exerted no adverse effect on nitrogen uptake at anthesis,even though the nitrogen application was reduced by half before anthesis compared with CI.At maturity in all three years,however,the NA in each organ was significantly higher under MSI than under CI,whereas the pre-anthesis nitrogen remobilization(NR)of the same organ and their contribution to grain N(NRC)were significantly higher under CI than under MSI.For the total contribution rate of N remobilization pre-anthesis to grain N(NRCT),MSI was lower than CI;however,the contribution of nitrogen accumulation post-anthesis to grain N(NPCT)was higher in MSI than in CI.Total NA at harvest was significantly higher under MSI than under CI,whereas there was little difference in nitrogen harvest index(NHI)between the two irrigation methods.NRCT,NPCT,total N,and NHI were significantly affected by I,Y,and I×Y.In comparison with the 2014–2015 and 2016–2017 growing seasons,the total N and NHI in 2015–2016 were decreased.In summary,MSI significantly increased NA at maturity,decreased pre-anthesis nitrogen remobilization,and increased the contribution of nitrogen accumulation post-anthesis to grain N relative to CI.

Table 1Effects of micro-sprinkling irrigation on dry matter accumulation,remobilization,and harvest index of winter wheat.

Table 2Effects of micro-sprinkling irrigation on grain-filling parameters of winter wheat.

3.7.Changes in grain protein concentration during grain filling

GPC showed an initial decrease and subsequent increase with grain growth(Fig.5).GPC was higher in the 2015–2016 growing season than in the 2014–2015 and 2016–2017 growing seasons.GPC was significantly higher under MSI than under CI after DAA 10.MSI increased GPC during the mid–late filling stages compared to CI.

Fig.4.Effects of micro-sprinkling irrigation on grain dry accumulation characteristics during grain-filling period of winter wheat.CI,conventional flood irrigation;MSI,micro-sprinkling irrigation.Asterisks indicate significant difference between CI and MSI(P＜0.05).Values are means±standard error(n=3).

3.8.Grain yield and protein yield

Across the three winter wheat seasons,GY,TKW,GPC,and PY were significantly increased by MSI compared to CI,and there was little difference in spike number(SN)and kernel number per spike(KN)(Table 4).As shown in Table 4,annual maximum GY and TKW were observed in MSI.Across the three years,GY,TKW,GPC,and PY increased by 13.9%,9.7%,7.0%,and 21.4%,respectively,under MSI relative to CI.GY,TKW,GPC,and PY were significantly affected by irrigation method(I)and year(Y),and GY and TKW were significantly affected by I×Y.SN and KN were also affected by Y.Although the TKW in 2015–2016 was higher than 2014–2015 and 2016–2017 growing seasons,the lower SN in this year resulted in the lowest GY during the three seasons in this study.GPC in 2015–2016 was higher than other two years,possibly owing to the lower GY,and minimum PY was observed in 2015–2016.Correlation analysis showed that GY was significantly and positively correlated with PY(Table 5).Overall,GY and PY after harvest were simultaneously increased by MSI.

Table 3Effects of micro-sprinkling irrigation on nitrogen accumulation,remobilization,and harvest index of winter wheat.

Table 4Effects of micro-sprinkling irrigation on grain yield(GY),yield components,grain protein concentration(GPC)and protein yield(PY)at harvest of winter wheat.

Table 5Correlations between main agronomic traits of winter wheat in this study.

Fig.5.Effects of micro-sprinkling irrigation on grain protein concentration during the grain-filling period of winter wheat.CI,conventional flood irrigation;MSI,microsprinkling irrigation.Asterisks indicate significant difference between CI and MSI(P＜0.05).Values are means±standard error(n=3).

4.Discussion

Crop growth and GY are greatly influenced by irrigation method[8,35].Timing of N fertilizer application also affects GY and GPC[19,36].In the present study,we found that the GY and GPC were significantly higher at maturity under MSI than under CI(Table 4).

Previous studies[37,38]have suggested that further increases in wheat yield may be achieved by increasing grain weight.In this three-year study,MSI showed little influence on spike or kernel number but significantly increased TKW(Table 4),which was the primary cause of the increased GY under MSI.Grain weight is determined by the duration and rate of grain filling[39,40],and is closely related to crop yield[41].Prolonging filling duration and increasing filling rate simultaneously are the main ways to increase wheat grain weight.Grain filling in wheat is affected by the amount of photosynthates produced in the photosynthetic organs and the mobilization of stored carbohydrates within these organs[42],and increasing dry matter accumulation after anthesis is an important means of increasing grain weight and grain yield[43].In this study,the chlorophyll content of the upper three leaves during the mid–late grain-filling period was higher under MSI than under CI(Fig.3),indicating that senescence of the canopy leaves was delayed by MSI.The high chlorophyll content of leaves in MSI may be due to the water and N fertilizer applied at jointing,booting,anthesis and filling,especially given the water and N fertilizer applied during grain filling in MSI,while the topdressing of N was applied only at jointing stage and irrigation was applied at jointing and anthesis in CI.In a word,the higher chlorophyll content of leaves accounted for the increase in DM accumulation and grain weight during grain filling under MSI(Figs.3 and 4;Table 1).

Drought stress and high-temperature damage are often concurrent in the mid–late grain-filling stage in the NCP and lead to premature senescence,shortened grain-filling duration,and reduced grain weight[11].Moderate soil drought during the grain-filling period increases the filling rate and grain yield[44,45],but a severe water deficit reduces the final grain weight[46].The optimal temperature for the post-heading period in wheat is less than 30 °C[47,48];at higher temperatures,wheat yield and grain quality deteriorate[49].During the three years of this study,the weather was characterized by low precipitation and gradually increasing temperature after anthesis(Fig.1).The daily high temperature at 14:00 was greater than 35 °C during grain filling and exceeded 38 °C during the late grain-filling period(Fig.2).However,the canopy temperature was markedly decreased and the relative humidity was increased during the mid–late grain-filling period by MSI relative to CI,a condition beneficial to delaying leaf senescence.The high relative humidity in the wheat canopy in MSI may ensure sufficient water supply during the grain-filling period,and drought stress may occur in the grain-filling later stage of CI,resulting in impaired grain filling.

GPC is directly determined by grain nitrogen concentration.Because grain N concentration at wheat maturity results from the remobilization of N stored before anthesis and the assimilation after anthesis[25],the N status of vegetative organs before anthesis and during the grain-filling period greatly influences grain N concentration[50,51].GPC is strongly affected by the timing and method of N application[19,20,27].Under conventional management of irrigation and fertilizer application,remobilization of N accumulated before anthesis accounts for more than 80% of grain nitrogen accumulation[25,52],possibly because N fertilizer is supplied mainly at the early growth stage of wheat.In the present study,75.6%–81.3% of grain N came from the remobilization of N accumulated before anthesis under CI,but only 52.5%–60.5% of grain N came from pre-anthesis nitrogen remobilization under MSI.Compared to CI,MSI significantly decreased the translocation of N stored in stems,sheaths,and leaves;increased nitrogen accumulation(NA)after anthesis;and greatly increased total NA at maturity(Table 3).Protein yield was increased by 18.8%–24.6%under MSI compared to CI(Table 4),whereas grain yield increased by 10.5%–16.7%under MSI.Thus,the increase in protein accumulation in grain was greater than the increase in grain yield under MSI,leading to an increase in GPC;thus,MSI produced simultaneous improvement in both GY and GPC.

N transfer from leaves to grain often weakens the photosynthetic capacity of leaves,resulting in decreased dry matter and grain nitrogen absorption and accumulation in wheat,the primary reason that it is difficult to improve both GY and GPC simultaneously[34].However,GPC can be increased and higher photosynthate production capacity in leaves can be maintained by the application of N fertilizer after anthesis[26,27].The grain-filling process also plays an important role in grain formation and nutrient accumulation[39,40]and is closely associated with GY and GPC[41].In this study,the amount of N absorption and accumulation after anthesis was significantly increased,and the total DM and NA at maturity were significantly increased,by MSI relative to CI(Tables 1 and 3).TKW and GY were significantly improved by MSI,and there were positive correlations among TKW,GY,and PY but no significant correlation between GY and GPC at harvest(Table 5),a finding inconsistent with previous reports[24,53]showing a negative correlation between grain weight and grain N concentration.We propose that the conflict between yield and grain N concentration in winter wheat can be greatly reduced by adoption of the MSI and fertilization regime used in this study.In addition,GPC was significantly higher under MSI than under CI during the mid–late grain-filling period(Fig.5).Regulating water and N fertilizer application by MSI may better balance the water and N supply and demand of wheat than is achieved under CI.

The difficulty of simultaneously increasing GY and GPC in wheat is due to the negative correlation between them[24,54]and may be associated with the competition between carbon and N for energy[55]as well as the N dilution effect of carbon compounds in the grain[53].In this study,micro-sprinkling irrigation combined with N fertilization changed the soil water status and N conditions during the growing season[5,7]and moderated the canopy environment during the grain-filling period(Fig.2).MSI may weaken the competition between carbon and N for energy by providing sufficient water and N supply during grain filling,resulting in simultaneous increases in grain weight and grain N concentration.Foliar fertilization of wheat can also result in more effective N absorption and utilization and enhance grain quality[56,57].The effects of micro-sprinkling irrigation combined with N fertilizer on grain carbon and N metabolism as well as N absorption and use in winter wheat await further study aimed at identifying the physiological mechanisms that underlie the simultaneous increase in GY and GPC.

5.Conclusions

MSI increased GY,GPC,and PY in winter wheat compared to CI.Across a three-year field experiment,yield under MSI increased by 10.5%–16.7%,and GPC and PY increased by 5.4%–8.0% and 18.8%–24.6%,respectively.The increases in GY and PY of MSI occurred mainly because MSI moderated the canopy microenvironment during the mid–late grain-filling period and delayed senescence,prolonged the duration of grain filling,increased the grain-filling rate,and increased the contribution of dry matter and nitrogen accumulation post-anthesis to grain yield and grain nitrogen.MSI combined with N fertilization is proposed to be an effective water-and fertilizer-management method for simultaneously realizing high yield and GPC of winter wheat in the NCP.

CRediT authorship contribution statement

Jinpeng Li:Writing-original draft,Investigation.Zhimin Wang:Conceptualization,data curation and writing-review &editing.Chunsheng Yao:Investigation.Yang Liu:Investigation.Yinghua Zhang:Conceptualization,data curation and writingreview & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported by the National Key Research and Development Program of China(2016YFD0300401),the National Natural Science Foundation of China(32001474,31871563),and the China Agriculture Research System(CARS-3).