Sunusi Amin ABUBAKAR ,Abdoul Kader Mounkaila HAMANl ,WANG Guang-shuai ,LlU HaoFaisal MEHMOODAbubakar Sadiq ABDULLAHl,GAO YangDUAN Ai-wang

1 Key Laboratory of Crop Water Use and Regulation,Ministry of Agriculture and Rural Affairs/Farmland Irrigation Research Institute,Chinese Academy of Agricultural Sciences,Xinxiang 453002,P.R.China

2 Graduate School of Chinese Academy of Agricultural Sciences,Beijing 100081,P.R.China

3 Department of Agricultural and Bioresource Engineering,Abubakar Tafawa Balewa University,Bauchi 740272,Nigeria

Abstract Excessive application of nitrogen (N) fertilizer is the main cause of N loss and poor use efficiency in winter wheat (Triticum aestivum L.) production in the North China Plain (NCP). Drip fertigation is considered to be an effective method for improving N use efficiency and reducing losses,while the performance of drip fertigation in winter wheat is limited by poor N scheduling. A two-year field experiment was conducted to evaluate the growth,development and yield of drip-fertigated winter wheat under different split urea (46% N,240 kg ha-1) applications. The six treatments consisted of five fertigation N application scheduling programs and one slow-release fertilizer (SRF) application. The five N scheduling treatments were N0-100 (0% at sowing and 100% at jointing/booting),N25-75 (25% at sowing and 75% at jointing and booting),N50-50(50% at sowing and 50% at jointing/booting),N75-25 (75% at sowing and 25 at jointing/booting),and N100-0 (100% at sowing and 0% at jointing/booting). The SRF (43% N,240 kg ha-1) was only used as fertilizer at sowing. Split N application significantly (P＜0.05) affected wheat grain yield,yield components,aboveground biomass (ABM),water use efficiency(WUE) and nitrogen partial factor productivity (NPFP). The N50-50 and SRF treatments respectively had the highest yield(8.84 and 8.85 t ha-1),ABM (20.67 and 20.83 t ha-1),WUE (2.28 and 2.17 kg m-3) and NPFP (36.82 and 36.88 kg kg-1).This work provided substantial evidence that urea-N applied in equal splits between basal and topdressing doses compete economically with the highly expensive SRF for fertilization of winter wheat crops. Although the single-dose SRF could reduce labor costs involved with the traditional method of manual spreading,the drip fertigation system used in this study with the N50-50 treatment provides an option for farmers to maintain wheat production in the NCP.

Keywords: split nitrogen strategies,wheat yield,drip fertigation,water use efficiency,nitrogen use efficiency

1.lntroduction

With nearly two-thirds of China’s wheat output,the North China Plain (NCP) is the largest and most important wheat growing area in the country,so is very important to national food security (Siet al.2020). Rainfall in this area is mostly distributed from June to September (Mehmoodet al.2019). The winter wheat growing season is from October to June of the next year when the rainfall accounts for only 20-30% of the annual total,which cannot meet the production demand of the crop (around 400-450 mm). Therefore,supplementary irrigation is a necessary management strategy to ensure winter wheat production in the region. To ensure food security,largescale production of winter wheat is necessary,which consumes a large quantity of water,which in turn could seriously threaten the security of water resources in North China. Drip irrigation systems with high irrigation efficiency may be of great importance for environmentally sustainable production of winter wheat (Jhaet al.2019).

China has become the world’s largest fertilizer consumer and relies on chemical fertilizers to feed about 22% of the world’s population (Baiet al.2016).The affordability of urea fertilizer coupled with improper management by farmers has led to the application of N fertilizer exceeding the total amount required by the crop.While the quantity of N fertilizer applied rises in China,its utilization efficiency decreases (Chenet al.2021). It is reported that the N partial factor productivity (NPFP) is less than 35% in China,only 70% of the world average(Chenet al.2021). The indiscriminate use of ordinary urea results in low utilization efficiency,which causes high fertilization labor costs,and could damage the roots of crops and reduce production. Nitrogen could also get into groundwater and rivers in the form of nitrate,resulting in eutrophication of groundwater,rivers,and lakes,pollution of water resources,and ultimately affecting human health. However,as a necessary nutrient element for crop growth,crop yield may be affected if the amount of N fertilizer is greatly reduced. Therefore,managing N loss through designed scheduling of N application rates for ensuring crop yield is particularly important.

Over the years,some promising N management practices have been explored,of which some have been considered effective technologies (Yoseftabar 2012). The core of these approaches has been to adjust the time of the standard application of urea according to the critical period of crop N demand,to improve the synchronization between crop N demand and N supply (Zhanget al.2022). The practice of split N application has proved to be a technology applicable to crops such as wheat (Wuet al.2019;Donget al.2022). Through improvement in the relationship between supply and demand of N fertilizer,the split application of urea has improved crop yield,product quality,and NPFP of grain crops (Zhouet al.2018). Luet al.(2021) concluded that although the amount of N application decreased,grain crops’ NPFP and economic benefit were considerably increased. In addition,some studies have shown that split N application reduces nitrate leaching (Liuet al.2021). Even though split N application has achieved positive results in improving crop yield,NPFP and reducing N loss,it leads to many additional field operations,which increases labor input costs compared with single N fertilization (Liuet al.2022). Rural to urban migration of the labor force is rising in China,which will seriously affect agricultural production in the future. This problem could be mitigated through use of the drip-fertigation system for the combined application of water and fertilizer. Drip irrigation is one of the most important water-saving irrigation systems because of its high water and fertilizer use efficiency (Mehmoodet al.2019). So far,drip irrigation has been successfully applied in the production of cash crops and fruits (Jhaet al.2019). However,the system is rarely used for high plant density crops such as wheat (Mehmoodet al.2019).

Slow release fertilizers (SRFs) have been recognized worldwide (Al-Rawajfehet al.2021). SRFs release nutrients slowly and have low nutrient diversion loss,which is conducive to crops nutrient absorption and utilization (Shanet al.2022). SRFs can reduce ammonia volatilization in the field,improve N application efficiency,meet the overall nutritional needs of crops during the growth period and reduce environmental pollution (Shanet al.2022). SRFs can increase crop yields and improve their growth and development,as well as the quality of produce (Trenkel 2010;Chenet al.2018). However,SRFs are expensive and thus unaffordable to many farming communities. Fertilizer prices have risen sharply during the COVID pandemic,further increasing the cost of agricultural inputs. However,Zhanget al.(2022) considered that it was not conclusive that SRFs could replace split N application techniques under certain conditions in wheat. It is crucial that a management practice is developed that can replace the traditional mode of fertilization,either by replacing the urea with SRF or by adopting drip-fertigation techniques.Otherwise,food production may decline,endangering food security. This study hypothesized that the split application of ordinary urea could be sustained with the help of scheduled drip fertigation. The study aimed to provide the scientific basis for the management of drip fertigation techniques of winter wheat under a designed N fertilization schedule. The objective of this work was to compare the effects of SRF and split application of N (urea) on the growth,yield and NPFP of drip-irrigated winter wheat.

2.Materials and methods

2.1.Experimental site

The two-year (2019/2020 and 2020/2021) experiments were carried out at the experimental station of the Institute of Farmland Irrigation (35°08′N,113°45′E;81 m above sea level) of the Chinese Academy of Agricultural Sciences,Xinxiang City,Henan Province,located in the NCP. The region’s climate is warm temperate continental monsoon,with an average annual rainfall of 578 mm(70-80% of which occurs from June to October) and an average rainfall of 161 mm in the wheat season (Siet al.2020). The total seasonal rainfall in 2019/2020 and 2020/2021 was 113 and 87 mm,respectively. The monthly averages of maximum and minimum temperatures and rainfall during the two seasons are presented in Fig.1. The soil in the experimental area is a sandy loam.The bulk density and particle size distribution of the soil at the experimental site are shown in Table 1. Other physical and chemical properties are presented by Zainet al.(2021b).

Fig.1 Monthly averages of total rainfall and minimum and maximum temperatures in the 2019/2020 and 2020/2021 winter wheat years (the data were collected from a weather station located at the center of the experimental field).

2.2.Field practice and experimental design

The winter wheat cultivar sown was a high-yielding variety called Aikang 58. The sowing rate was 180 kg ha-1(giving 350-400 plants m-2at a normal germination rate). The experiment consisted of five nitrogen application scheduling treatments and a single dose of SRF as shown in Table 2. The five N scheduling treatments were N0-100 (0% at sowing and 100%at jointing/booting),N25-75 (25% at sowing and 75% at jointing/booting),N50-50 (50% at sowing and 50% at jointing/booting),N75-25 (75% at sowing and 25% at jointing/booting),and N100-0(100% at sowing and 0% at jointing/booting). The SRF (43% N,240 kg ha-1) was only applied as base(at sowing) fertilizer. These six treatments were replicated three times in a randomized complete block design. The urea application rate of 240 kg ha-1was selected based on the recommendations of Duanet al.(2019) and Siet al.(2020). The plot size was 15 m by 3 m,with the blocks separated by 0.5 m access lanes. The soil was cultivated to 20 cm depth with a tractor-drawn rotary cultivator and then levelled using a harrow. The sowing dates were October 23,2019,and October 24,2020. The wheat was harvested on June 1,2020,and June 2,2021. The N,P,and K fertilizers were applied using urea (46% N),calcium superphosphate (16% P2O5) and potassium sulfate (50%K2O),respectively. Phosphorous (P) and K fertilizers were applied at the rate of 120 kg ha-1at sowing. N fertilization was applied at sowing and at the jointing and booting stages of crop growth (Table 2).

Table 1 Bulk density,particle size distribution,texture and moisture characteristics of the soil at the experimental site1)

Table 2 Fertilizer application schedules of the experimental treatments

2.3.lrrigation and fertigation methods

A surface drip irrigation system was installed with a lateral irrigation line spacing of 60 cm. The drippers were spaced at 20 cm along the laterals. The discharge rate of the drippers was 2.2 L h-1under a working pressure of 0.10-0.15 MPa. Flow meters were installed in each plot to regulate the volume of irrigation water released.The crop evapotranspiration between two consecutive irrigation events was calculated according to eq.(1):

whereETais actual crop evapotranspiration (mm d-1),Kcis crop coefficient (early seasonKcis 0.36;mid-seasonKc=1.19;late seasonKcis 0.28 according to Gaoet al.(2009)). The reference evapotranspiration (ETo) was calculated according to Allenet al.(1998).

The irrigation requirement (i) was determined using eq.(2):

Irrigation events occurred when the totalireached 45 mm as recommended by Shenet al.(2020). The details of the irrigation exercise are shown in Table 3.

Table 3 Irrigation schedules for the drip-fertigated winter wheat during the 2019/2020 and 2020/2021 seasons

The topdressing fertilizer was applied at the winter wheat jointing and booting stages using a closed tank fertigation system (Phocaides 2007). The SRF used was produced by the polymer coating process. The coating material was polyolefin polymer resin with an additive made of talcum powder. The SRF coating was 5.6% of the SRF mass,the nitrogen content was 43%,and the release period was 30 days (the number of days required for 80% of N release by the SRF at 25°C).

2.4.Field sampling and measurements

Soil water content (SWC) was measured every 10 days at 20 cm depth intervals to 100 cm depth using the gravimetric method. The soil samples were dried in the oven at a temperature of 105°C until a constant weight was attained. The weight basis moisture content was then converted to volumetric soil water. The seasonal crop evapotranspiration (ETc) was estimated by using the soil water balance equation (eq.(3)).

whereIis the total irrigation applied during the winter wheat growth period (mm);Pis the seasonal precipitation(mm);Uis the upward capillary rise;RandDare surface runoff (mm) and downward drainage,respectively;and ΔS is the change in the soil water between the start and end of the season. Here,the upward capillary rise and downward drainage were assumed to be zero since the water table was deeper than 12 m and the irrigation water was gradually appliedviaa drip system. No intense rainfall events occurred during the two experimental periods.

Aboveground biomass was measured for plant samples from 20 cm long rows. These samples were dried in an oven at 75°C until a constant weight was attained. At grain maturity,one square meter of undisturbed wheat area in each plot was randomly selected and harvested.The grain yield (Y,kg ha-1) was obtained after the grain was extracted and air-dried and weighed with a digital balance.

Water use efficiency (WUE,kg m-3) was calculated as follows (eq.(4)):

whereETcis crop evapotranspiration (mm).

The nitrogen partial factor productivity (NPFP) was calculated as follows (eq.(5)):

whereNis nitrogen application rate (kg ha-1).

Leaf area index (LAI) and plant height of winter wheat were measured at intervals of 10 to 15 days using 10 randomly selected plant samples from each plot. LAI was determined as in Zainet al.(2021b). The leaf length and width were manually measured using a ruler,and LAI was then calculated using eqs.(6-8).

whereAiis the leaf area of ith plant,mis the number of leaves in theith plant,LjandWjare the length and width of thejth leaf of theith plant (both measured in cm);LAis leaf area,nis the number of plant samples used to determine the leaf area;Nis the total number of plants(including tillers) in a meter row of the wheat,andSis row spacing (0.2 m).

Plant height was taken as the length from the ground surface to the tip of the plant. In the later stages,the spikelet was included in the plant height. Similarly,for each plot,plant height,and yield components (number of grains per spike and kernels per spike) were measured from 10 plants samples selected randomly at maturity.Finally,the 1 m2area of plants was used to measure the number of spikes per unit area,grain yield (t ha-1),aboveground biomass (t ha-1) and 1 000-grain weight (g)from each experimental plot. The moisture content of the grain was maintained at 12%. The harvest index (HI) was determined using eq.(9):

2.5.Statistical analysis

A mixed-model ANOVA was conducted to analyze the effects of the treatments,wheat season,and their interaction on wheat grain yield,yield components,aboveground biomass,water use efficiency and NPFP.Fisher’s LSD was used for mean separation where there were significant differences. The software used was SPSS version 23.0 (IBM SPSS Statistics for Windows,Armonk,IBM Corp,NY,USA) and GraphPad Prism version 9.0 for Windows (GraphPad Software,La Jolla California,USA).

3.Results

3.1.Seasonal variation of soil moisture

The temporal and spatial variations of the soil volumetric water content (SWC) for the various irrigation scheduling treatments in 2019/2020 are shown in Fig.2. In particular,the moisture distribution pattern was almost similar between the treatments. The seasonal average of the SWC within the 0-60 cm soil layer (where most of the wheat roots were distributed) was 24.6% (77.9%FC). In March (returning green to the jointing stage),the SWC was at 23.17% (76.7% FC). Most of the SWC was from irrigation as the rainfall was only 4.2 mm. In April(booting to flowering),the SWC was 28.1% (88.9% of field capacity). Similarly,in May (flowering to maturity),the SWC,which was mainly from rainfall of 32.0 mm,was 72.7% of FC.

Fig.2 Seasonal dynamics of volumetric water content for various treatments in 2019/2020 seasson. A,slow release fertilizer(SRF). B,0% of basal N applied at sowing and 100% of topdressing N equally split at jointing and booting (N0-100). C,25% of basal N applied at sowing and 75% of topdressing N equally split at jointing and booting (N25-75). D,50% of basal N applied at sowing and 50% of topdressing N equally split at jointing and booting (N50-50). E,75% of basal N applied at sowing and 25%of topdressing N equally split at jointing and booting (N75-25). F,100% of basal N applied at sowing and 0% of topdressing N(N100-0). SWC,soil volumetric water content. Topdressing N was applied via a drip fertigation system.

The soil moisture distribution pattern of the treatments was similar in the 2020/2021 season (Fig.3). The seasonal average of the SWC in 2020/2021 was 23.6%(74.8% FC),the soil moisture mostly coming from irrigation. In March,the SWC was 23.5% (74.3% FC).Between the booting and flowering stages of the wheat growth,the SWC was kept at an average of 25.1% (81.0%FC) which was from both irrigation and rainfall (20.8 mm).At maturity,the SWC was a result of rainfall and was within an average of 19.1% (63.2% FC).

Fig.3 Seasonal dynamics of volumetric water content for various treatments in 2020/2021 season. A,slow release fertilizer(SRF). B,0% of basal N applied at sowing and 100% of topdressing N equally split at jointing and booting (N0-100). C,25% of basal N applied at sowing and 75% of topdressing N equally split at jointing and booting (N25-75). D,50% of basal N applied at sowing and 50% of topdressing N equally split at jointing and booting (N50-50). E,75% of basal N applied at sowing and 25%of topdressing N equally split at jointing and booting (N75-25). F,100% of basal N applied at sowing and 0% of topdressing N(N100-0). SWC,soil volumetric water content. Topdressing N was applied via a drip fertigation system.

3.2.Leaf area index and plant height

The curves for LAI and plant height indicated similar patterns under the different N-scheduling treatments(Figs.4 and 5). The N50-50 treatment showed higher LAI (P＜0.05) than other treatments for the greater portion of the crop growth period in the two years. The second highest LAI was recorded in the N75-25 and SRF treatments. However,the N0-100 treatment exhibited the lowest profile for LAI in both years. Plant height increased from sowing to the maximum at winter wheat maturity. The N0-100 treatment had the lowest plant height in both seasons. At maturity,the maximum LAI was observed in the N50-50 and SRF treatments(Fig.6). Compared with the N0-100 and N100-0 treatments,the N50-50 treatment increased (P＜0.05)LAI by an average of 25.2 and 29.9%,respectively.Similarly,the SRF treatment increased (P＜0.05) LAI by 18.1 and 23.1%,respectively.

Fig.4 Seasonal variation in leaf area index of winter wheat under different nitrogen scheduling treatments in the 2019/2020 and 2020/2021 seasons. SRF,slow release fertilizer;N0-100,0% of basal N applied at sowing and 100% of topdressing N equally split at jointing and booting;N25-75,25% of basal N applied at sowing and 75% of topdressing N equally split at jointing and booting;N50-50,50% of basal N applied at sowing and 50% of topdressing N equally split at jointing and booting;N75-25,75% of basal N applied at sowing and 25% of topdressing N equally split at jointing and booting;N100-0,100% of basal N applied at sowing and 0% of topdressing N. Topdressing N was applied via a drip fertigation system.

Fig.5 Seasonal variation in winter wheat plant height in response to different nitrogen scheduling treatments in the 2019/2020 and 2020/2021 seasons. SRF,slow release fertilizer;N0-100,0% of basal N applied at sowing and 100% of topdressing N equally split at jointing and booting;N25-75,25% of basal N applied at sowing and 75% of topdressing N equally split at jointing and booting;N50-50,50% of basal N applied at sowing and 50% of topdressing N equally split at jointing and booting;N75-25,75% of basal N applied at sowing and 25% of topdressing N equally split at jointing and booting;N100-0,100% of basal N applied at sowing and 0% of topdressing N. Topdressing N was applied via a drip fertigation system.

Fig.6 Leaf area index response to N scheduling and SRF treatments at maturity during the 2019/2020 and 2020/2021 seasons.SRF,slow release fertilizer;N0-100,0% of basal N applied at sowing and 100% of topdressing N equally split at jointing and booting;N25-75,25% of basal N applied at sowing and 75% of topdressing N equally split at jointing and booting;N50-50,50%of basal N applied at sowing and 50% of topdressing N equally split at jointing and booting;N75-25,75% of basal N applied at sowing and 25% of topdressing N equally split at jointing and booting;N100-0,100% of basal N applied at sowing and 0% of topdressing N. Topdressing N was applied via a drip fertigation system. The error bars represent the standard errors of the mean of three replicates. Different letters indicate significant among the treatment at P＜0.05.

3.3.Grain yield,yield components,and aboveground biomass

The grain yield was significantly (P＜0.05) affected by the N scheduling treatments in the two winter wheat growing years (Table 4). In both the years,the SRF and N50-50 treatments had the maximum grain yield and were statistically similar (P＞0.05). The N50-50 and SRF treatments had higher grain yield than other treatments.Grain yield was affected by the treatments as shown by the mixed model ANOVA (Table 5),but the effect of season and its interaction with the treatments were not significant (P＞0.05).

Compared with the N100-0 treatment,split applications of N significantly improved spike number per hectare(SNPH),kernels per spike (KPS) and 1 000-grain weight(TGW) in both years (Tables 4 and 5),the maximum values were measured in the N50-50 treatment. The SRF treatment had yield components statistically similar(P＞0.05) to the N50-50 treatment. The number of KPS was significantly (P＜0.05) affected by the N scheduling treatments. The highest number of KPS was recorded in the SRF treatment,followed by the N50-50 treatment(P＞0.05) in both years.

Table 4 Grain yield,spike number,kernel per spike as affected by different N-scheduling treatments

The N scheduling treatments affected the aboveground biomass (ABM) significantly (P＜0.01),but the effect of season and its interaction with the treatments (Table 5) were not significant (P＞0.05). The treatments significantly (P＜0.05) affected the ABM of the winter wheat in both years. The SRF and N50-50 treatments had the highest ABM in both years (Fig.7).The ABM ranged from 17.0 to 20.8 t ha-1and 17.7 to 21.4 t ha-1in the 2019/2020 and 2020/2021 years,respectively. The results showed that the SRF and N50-50 treatments were comparable.

Fig.7 Aboveground biomass of winter wheat in response to different nitrogen treatments at harvest in the 2019/2020 and 2020/2021 seasons. SRF,slow release fertilizer;N0-100,0% of basal N applied at sowing and 100% of topdressing N equally split at jointing and booting;N25-75,25% of basal N applied at sowing and 75% of topdressing N equally split at jointing and booting;N50-50,50% of basal N applied at sowing and 50% of topdressing N equally split at jointing and booting;N75-25,75% of basal N applied at sowing and 25% of topdressing N equally split at jointing and booting;N100-0,100% of basal N applied at sowing and 0% of topdressing N. Topdressing N was applied via a drip fertigation system. The error bars represent the standard errors of the mean of three replicates. Different letters indicate significant among the treatment at P＜0.05.

3.4.Water use efficiency and nitrogen partial factor productivity

There were similar trends forETcamong the treatments in the 2019/2020 and 2020/2021 years (Tables 5 and 6).ETcranged from 387.5 to 433.5 mm in 2019/2020 and 402.30 to 456.1 mm in 2020/2021. The WUE was significantly affected by the N-fertigation level (P＜0.05;Table 5). The WUE increased with an increase in grain yield and a decrease inETc. The NPFP was significantly(P＜0.05) affected by the N-fertigation level in the two study years (Table 5). The SRF and N50-50 treatments were statistically similar in WUE and NPFP in 2019/2020(P＞0.05),but SRF was second to N50-50 in 2020/2021(P＜0.05). The ranges NPFP obtained were 31.5 to 36.9 kg kg-1and 30.5 to 38.9 kg kg-1in the 2019/2020 and 2020/2021 years,respectively.

Table 5 ANOVA (F-values) of the effect of season,treatment and their interaction on various winter wheat growth and yield traits

3.5.Harvest index

The effects of the N-fertigation scheduling,wheat season and their interaction on harvest index were not significant(P＞0.05) as shown by the mixed model ANOVA (Table 5).The HI of the N50-50 and the SRF treatments was similar(Table 6). The difference between the SRF and N50-50 treatments and other treatments was significant (P＜0.05) in both 2019/2020 and 2020/2021 years. The highest harvest index values of 53.3 and 56.3% were recorded in the N50-50 treatment in both seasons. Across the two seasons,the SRF recorded the second-highest harvest index,which was 13.7% lower than the N50-50 treatment. The lowest harvest index recorded was in the N100-0 treatment,which was 25.4% lower than the N50-50 treatment.

Table 6 Means of evapotranspiration (ET),water use efficiency (WUE),nitrogen partial factor productivity (NPFP) and harvest index (HI) as affected by different N-scheduling treatments

4.Discussion

4.1.Distribution of water in the soil profile

Water distribution in a soil profile mainly depends on the gradient between the inflow (irrigation or rainfall) and outflow (such as evapotranspiration or infiltration). The drip irrigation system and the water application scheduling involved in this study were aimed at ensuring uniform moisture distribution across all N treatments. This is because soil water influences crop N uptake by promoting N transfer to the wheat vegetative organs (Wuet al.2021). The 20-60 cm layer maintained a lower SWC than the field capacity during the winter wheat growing seasons (Figs.3 and 4). This could be due to high root system activities within this layer (Jhaet al.2017).However,the high SWC observed in the 80-100 cm soil layer in 2020/2021 (Fig.3) was due to frequent irrigation triggered by low rainfall. Further,this layer could have been poorly drained. There was no significant variation observed between the N fertigation treatments.The amount of N applied has no significant effect on the distribution and variation in soil moisture (Suiet al.2015)although it might have a slight indirect effect (Behera and Panda 2009). The results of this study indicate that irrigation management for drip-irrigated winter wheat should be targeting the 0-60 cm layer to achieve sufficient water supply to the crop. To avoid excessive watering of the deeper soil layers (60-100 cm),an irrigation system that applied a low and frequent amount to match the crop water requirement should be considered.

4.2.Effect of N scheduling on wheat growth parameters and aboveground biomass

LAI and plant height are important crop growth and development indices. It is a fact that the crop growth parameters are enhanced by the availability of N in the soil (Shiraziet al.2014;Zainet al.2021b). Nitrogen sufficiency results in a considerable increase in cell processes like cell division,cell elongation and duration that cause enlargement of the leaf area (Farooqet al.2009). Experimental results have also shown that the rational distribution of N at different growth stages is very important for the proper growth of crops under a fixed fertilization rate (Zainet al.2021b). In the present study,compared with other N scheduling treatments,the N50-50 treatment had the highest LAI at all wheat growth stages. This suggests that application of a balanced N dose between sowing and later growth stages is important for winter wheat to avoid poor vegetative growth at both early and mid-season points. Siet al.(2020) found that LAI,aboveground biomass and grain yield were the maximum at the N rate of 240 kg ha-1applied in a 40:60 ratio split twice in a season. The results of our study showed that the 240 kg N ha-1could be split in an N50-50 ratio,applied two times at sowing during the growth period. The peak values of the LAI were strongly correlated with grain yield and ABM (Fig.8).

Aboveground biomass is used as the overall index of crop growth and development. It comprehensively reflects the total contribution of LAI,plant density and plant height,and is the material basis of grain yield. The highest ABM was obtained in the SRF and N50-50 treatments,indicating that the greatest amount of N was supplied for winter wheat growth across the two years under these treatments. The objective of the topdressing N application was to improve the post-anthesis biomass formation,which is a useful way of increasing crop yield (Wanget al.2016). In our study,the SRF and N50-50 treatments provided the N needed at critical stages of the winter wheat growth,hence the higher ABM in these treatments (Fig.7). Correlation analysis suggested a strong positive relation between ABM and GY(Fig.8),indicating that a sufficient N supply leads to optimal ABM and GY (Siet al.2020).

Fig.8 Spearman correlation coefficients of winter wheat yield(GY) response to spike number per ha (SNPH),kernel per spike(KPS),1 000-grain weight (TGW),aboveground biomass (ABM),and leaf area index (LAI) during the 2019/2020 to 2020/2021 seasons.

4.3.Effect of N scheduling on grain yield and yield components

The ultimate objective of this work was to develop a suitable N application program that enhanced yield when compared to SRF application. Split application of N during the winter wheat growth period achieved a crop yield similar to that of the SRF. Compared with other treatments,the N50-50 and SRF treatments improved the main indicators of yield formation,the aboveground biomass,number of kernels per spike and 1 000-grain weight. Compared with the treatments which applied either 100% of the N at sowing or 100% at the later stages (jointing and booting)the N50-50 treatment improved grain yield by 18.1 and 10.1%,respectively. Similarly,the SRF treatment improved grain yield by 16.6 and 8.5%,respectively. The reason for the yield improvement is that the treatments influenced the yield components (NSPH,KPS and TGW) as revealed by the correlation analysis (Fig.8). The N50-50 treatment improved the NSPH by 21.8 and 20.0%,the KPS by 19.6 and 12.4% and the TGW by 12.7 and 13.2% compared with the N100-0 and N0-100 treatments across the 2019/2020 to 2020/2021 seasons respectively. Similarly,the SRF improved NSPH,KPS and TGW parameters by 20.3 and 13.1%;10.7 and 11.1%;and 20.0 and 13.2% as compared to N100-0 and N0-100 treatments across the two years,respectively. Similar results were reported by Liuet al.(2019),who found that reducing the initial dose and increasing N application at the jointing and booting stages significantly increased yield components and ultimately the grain yield. However,excessive fertilization at the later growth stage,could lead to a decline in yield due to poor grain filling,prolonged growth and delayed maturity (Lianget al.2017;Zhanget al.2017). Maet al.(2021) concluded that appropriate N supply ensured the reasonable distribution of soil inorganic N,which met growth demands and reduced N losses at the early stage and promoted the accumulation of N and dry matter of wheat at the later growth periods.

Previous studies have shown that under sufficient water conditions,crop yield is strongly correlated to the number of kernels per spike and the number of spikes per unit area (Rivera-Amadoet al.2019) as shown in Fig.8. Currently,measures taken to improve grain yield have been mainly to increase the number of grains per spike or 1 000-grain weight,but it is difficult to improve the size of the grain sink (Vahamidiset al.2019). Among the yield component parameters,the 1000-grain weight is the most important factor affecting grain production (Zainet al.2021b). Our experiments showed that the higher grain yield in the N50-50 and SRF treatments could be attributed to the higher number of kernels per spike and 1 000-grain weight observed in those treatments. Wheat grain yield is strongly correlated with spike number and the number of kernels per spike (Xuet al.2018). Zainet al.(2021b) reported that topdressing N application significantly improved grain yield by increasing grain number per unit area. The strong correlations between GY and spike number and the number of kernels per spike as shown in Fig.8 indicate that these parameters should be targeted in work aiming to improve GY. The patterns of the yield components showed that a balancing approach to the spilt application of N is advisable.However,the SRF was competitive to split application of N.

4.4.Effect of N scheduling on evapotranspiration and water use efficiency

Previous studies have reported that drip irrigation methods are advantageous for the production of winter wheat in the NCP because of their water-saving ability (Mehmoodet al.2019). This study recorded an averageETcwithin the range obtainable in the area (Siet al.2021;Zainet al.2021a). However,the variation recorded inETcamong the treatments was not significant (P＞0.05). The average increase inETcobserved in the N0-100 treatment could be attributed to the high moisture content and density of the canopy,affected by the late application of the high N dose(Lianget al.2017). Daret al.(2017) reported that seasonal ET increased with an increase in irrigation and N amount.Our results are in line with the findings of Rathoreet al.(2017),who observed that higher N-fertilized plots had 8%moreETcthan non-fertilized plots. This also shows that the application of N fertilizer affects the consumption of soil water and improves its utilization efficiency by promoting absorption in wheat roots (Zhong and Shangguan 2014).

Nitrogen fertilizer applications have been found to significantly increase WUE of grain crops,by increasing grain yield. Nitrogen application can promote root growth and water absorption (Zhaoet al.2006). Meiet al.(2013) found that WUE of wheat increased by 20% in high fertilizer treatments in the NCP. They recommended that to properly utilize the resources,it is necessary to improve water and N use efficiency. In the present work,it was observed that the SRF and N50-50 urea treatments improved WUE as reported by other studies (Chilundoet al.2016;Ghafooret al.2021;Zainet al.2021a). Over the two experimental years,maximum WUE values of 2.28 and 2.17 kg m-3were obtained in the SRF and N50-50 treatments,respectively. These values were higher than the average obtained for winter wheat in the NCP (Siet al.2020). The results of our study indicated that the proper split of urea or controlled release of the required N fertilizer(as in SRF),rather than an arbitrary increase in fertilizer application rate,is a key to improving WUE efficiency.

4.5.Effect of N scheduling on nitrogen partial factor productivity

Rational fertilization practice promotes fertilizer and soil N uptake by wheat,increasing the NPFP (Kaiseret al.2010). Previous results indicate that an appropriate reduction in the N application rate under sufficient water conditions can decrease N losses without affecting crop N uptake (Siet al.2021). The NPFP increase observed in the SRF treatment in our study could be attributed to a longer duration of N release (Yanget al.2011). However,the equal split N application (N50-50) treatment produced similar results (Table 6) in line with recent studies (Zhenget al.2016;Zhanget al.2021). Higher efficiency recorded in the N50-50 and SRF treatments may be attributed to proper N uptake by plants,leading to greater yield production and minimized N losses (Sandhuet al.2019).The results of this study showed that under water-saving irrigation,a synchronized release of N required by the crop can increase NPFP by improving plant N absorption.However,other work has shown that NPFP increases with reduced N application under a water sufficient condition(Cuiet al.2010;Pradhanet al.2013).

The harvest index is expressed as the ratio of grain yield to aboveground biomass. The range reported in this study agreed with the results of Zainet al.(2021b).The harvest index was improved by N application in the N50-50 treatment by 24.9 and 25.4% as compared with the N100-0 and N0-100 treatments,respectively.

5.Conclusion

The two-year study revealed that N fertigation scheduling significantly affected winter wheat growth,GY,WUE and NPFP. The results showed that splitting the N application by fertigation of 240 kg ha-1as urea between sowing (50%) and the jointing and booting stages (50%)positively affected wheat production under the drip irrigation system used. The SRF and N50-50 treatments produced the highest GY,WUE and NPFP in the two growing years. The SRF treatment positively affected winter wheat growth,development,yield,and water and N use efficiency. The results of the study showed that the strategy of adopting SRF to replace normal urea application is promising as it could reduce labor costs since it is a one-off application. However,fertigation with the N50-50 application ratio could provide a reasonable option because of its low initial cost. The study provides evidence that split application of N could help maintain high wheat production. Future work is suggested to focus on evaluating the soil water and N distributions and quantifying N losses through leaching and emissions.

Acknowledgements

This research was funded by the earmarked fund for China Agriculture Research System (CARS-03-19),the National Natural Science Foundation of China (51879267 and 51709264),the Open Fund Projects of the Agricultural Environment Experimental Station of Minstry of Agriculture and Rural Affairs,China (FIRI2021040103),and the Agricultural Science and Technology Innovation Program(ASTIP) of Chinese Academy of Agricultural Sciences.

Declaration of competing interest

The authors declare that they have no conflict of interest.