Effects of tridimensional uniform sowing on water consumption, nitrogen use, and yield in winter wheat

Zhiqiang Tao, Shaokang Ma, Xuhong Chang,Demei Wang, Yanjie Wang,Yushuang Yang,Guangcai Zhao*, Jiancang Yang*

Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Key Laboratory of Crop Physiology and Ecology,Ministry of Agriculture and Rural Affairs,Beijing 100081,China

Keywords:Winter wheat Sowing pattern Nitrogen absorption and utilization Yield Grain protein content

A B S T R A C T Wheat is a staple crop worldwide, but yields may diminish as climate change causes increasingly unpredictable patterns of precipitation and soil nutrient availability. Farmers are thus challenged to maximize planting efficiency to increase yield,while also improving their resource use efficiency. In this study the effectiveness of tridimensional uniform sowing was tested across a range of planting densities for winter wheat crops on the North China Plain. Tridimensional uniform sowing was tested against conventional drilling at three planting densities (180 × 104, 270 × 104, and 360 × 104 plants ha-1) and assessed for water consumption, biomass, nitrogen uptake and allocation, and aspects of yield. The tridimensional uniform sowing treatment outperformed the conventional drilling treatment in most metrics and at most planting densities, while performing markedly better at higher planting densities. Water consumption decreased and nitrogen efficiency increased. Tiller number and percentage of productive tillers, leaf area index, dry weight,and yield increased without a significant decline in grain protein. Nitrogen allocation was more efficient under tridimensional uniform sowing than with conventional drilling, and also varied according to annual precipitation and planting density. Both yield and grain protein contents were significantly correlated with the amount of pre-anthesis accumulated nitrogen translocated from vegetative organs to kernels after anthesis. Overall, a density of 270 × 104 plants ha-1 provided the highest water use efficiency and grain yield.Tridimensional uniform sowing will benefit farmers by forming stronger overall crops,promoting the coordinated improvement of yield, nitrogen uptake and efficiency, and increasing grain protein content at higher planting densities.

1.Introduction

North China is a major winter wheat (Triticum aestivum L.)producing area, occupying a prominent position in wheat production owing to its large geographic area and suitable climate,and accounting for about one-fifth of food production in China [1,2]. Despite this productivity, precipitation can be unpredictable and insufficient for the winter wheat crop.Water is the most important limiting factor in wheat production [2]. For winter wheat, yield and biomass are dependent on water availability[3].Irrigation is often required for augmentation of rainfall,especially during the drier winter and spring months,though this practice often exerts harmful environmental effects including drops in the water table and an associated reduction of groundwater supplies[4].

The current shortage of water resources in North China has become the main problem restricting wheat production and is expected to worsen as a result of global climate change.Rising temperatures contribute to decreased wheat yields and will reduce wheat productivity at a global scale [5,6]. In particular, elevated seasonal temperatures are expected in northern China, where low and variable annual precipitation already impairs wheat crops [5]. With increased global population and food demand, it has become particularly urgent to develop technologies to improve water use efficiency in wheat. Irrigation techniques are especially important for improving the productivity of winter wheat in the North China Plain,considered the breadbasket of China[7].

Recent technological advances have focused on the simultaneous and synergistic improvement of several factors including water use, nitrogen efficiency, yield, and grain quality [8]. Nitrogen uptake and efficiency are widely studied as features of crop growth [9,10]. Photosynthesis in crops is changed by addition of water and nitrogen, and nitrogen nutrition is influenced by water status [11]. In durum wheat(Triticum durum Desf.), yield and dry matter accumulation were increased by nitrogen levels and reduced by water stress[12]. Increasing nitrogen fertilizer levels without addressing dry soil conditions results in lost grain yield and quality[13].

One factor that can be easily modified to improve yield and grain quality is the planting pattern, which influences the spatial distribution of individuals as well as their growth[14-16]. In comparison with broadcast sowing, conventional drilling (D) results in a more uniform seed distribution and planting depth,leading to higher germination,crop vigor,and yield [17]. Adjustment of the spacing of plants with D can increase canopy radiation use efficiency [16,18,19], and for this reason there have been numerous studies on the effects of row spacing on wheat yield[20-22].

A novel sowing pattern, tridimensional uniform sowing(U),is a modified form of conventional drilling in which seeds are distributed evenly and in the same plane,eliminating the ridges and rows that would otherwise be present after emergence in conventional patterns [18]. Tridimensional uniform refers to the even distribution of seed, fertilizer, and soil and provides a balanced growth environment for each seed. The presence of compacted soil above and loose soil below not only prevents air leakage but also reduces water evaporation. The tridimensional uniform sowing pattern is known [23] to result in improved light distribution, increased plant growth, and higher overall yield through an increase in the number of panicles. Uniform distribution of plants has also been shown [24] to result in greater leaf area index (LAI)in wheat. Although some previous our team work indicated that U reduced the water requirement,further study is needed to determine how changes in row spacing affect water availability as well as demand for irrigation and N fertilizer.

The effects of planting density on grain protein content in wheat vary.Kristensen et al.[25]found no difference in grain protein content between planting density treatments of 0.22 million and 0.42 million plants ha-1,or between 0.082 million and 0.25 million plants ha-1. Kristensen et al. [25] have also shown that increasing planting density in wheat can increase grain yield, though grain protein content is concurrently reduced by intraspecific competition for soil N among individual plants in the high-density population. In comparison with conventional drilling, tridimensional uniform sowing resulted in higher grain yield owing to increased plant density with an increased number of panicles [18]. Uniform spatial distribution of plants can decrease intraspecific competition for soil N [23]. Thus, grain protein content is not reduced in high-density populations that are nutritionally satisfied under tridimensional uniform sowing.

The present study tested the hypothesis that tridimensional uniform sowing (U) can coordinately improve wheat grain yield and quality. Its objectives were to 1) clarify the effects of planting density under U on plant and yield traits including water consumption, number of tillers, LAI, dry weight, nitrogen content and utilization in different plant organs, photosynthesis in the flag leaf, yield and its components, and grain protein content; 2) identify the optimum planting density for wheat; and 3) identify a combination of sowing mode and planting density that will concurrently improve yield, grain quality, and efficiency of water and nitrogen use.

2. Materials and methods

2.1. Experimental site

The study was conducted at the Xinxiang Experimental Station of the Institute of Crop Sciences, Chinese Academy of Agricultural Sciences (37°41′N, 116°37′E, altitude 270 m)during the wheat growing seasons from October 2015 to June 2016 and from October 2016 to June 2017. The study area experiences a temperate continental monsoon climate, with an annual mean temperature of 14.5 °C, annual cumulative temperature ≥10 °C of 4647.2 °C, and 2324 annual sunshine hours. Precipitation occurs primarily from July to September,when temperatures are highest, and the mean annual precipitation is 573.4 mm. Precipitation data for the different stages of wheat growth and development are displayed in Table 1. Growing degree days (GDD) and sunshine hours are displayed in Table 2. Daily GDD = (maximum temperature +minimum temperature)/2-10 °C, if the mean daily temperature is lower than 10 °C then GDD = 0. The study area is characterized by sandy loam soil, with approximately 50%sand, 35% clay, and 65% silt [26]. The top (0-20 cm) soil layer contained 13.2 g kg-1organic matter, 90.2 mg kg-1available nitrogen, 14.5 mg kg-1available phosphorus, and 105.5 mg kg-1available potassium,with a pH of 8.05.

Table 1-Precipitation in mm at the experimental site in Xinxiang.Data source:Meteorological Observation of Xinxiang County,Henan province,China.

2.2. Experimental design of field trials

Wheat variety Zhongmai 895 (ZM895), which produces a moderate number of tillers, was used. Seeds were selected before sowing and treated with using a mixture of 100 mL phoxim, 50% miscible oil (Huayu Pesticide Co., Ltd. Tianjin,China), 5 kg water, and 50 kg wheat seed. A split plot design was used, with sowing pattern as the main plot and planting density as the subplot, with three replicates of each treatment. Two sowing patterns were used: tridimensional uniform sowing(U)and conventional drilling(D).

The U treatment comprised fertilization, rotary tillage,sowing, compaction, coverage with a thin layer of soil, and a second surface compaction using a tridimensional uniform sowing machine(a joint development of the Institute of Crop Sciences, Chinese Academy of Agricultural Sciences and Deyou Nongfeng Mechanical Technology Co. Ltd., China).Fertilizers were distributed evenly on the soil surface,followed by rotary tillage. Seeds were uniformly distributed in the loose soil using a seed plate, so that the seeds,fertilizers,and soil were thoroughly combined and uniformly distributed in the compacted soil,and the seeds were covered with a thin layer of soil. A second surface compaction was applied to ensure that the soil below the seeds was loose,the soil around the seeds was compacted,and there was a layer of compacted soil above the seeds. For the D sowing pattern,sowing was accomplished with a Wintersteiger air suction precision plot seeder(Wintersteiger Trading Co.,Ltd.,Ried im Innkreis,AU).Row spacing under the D treatment was 15 cm,with three planting densities: 180 × 104, 270 × 104, and 360 × 104plants ha-1. The area of each test plot was 15 m2(10 m × 1.5 m). After sowing, conditions were held constant across the treatments.

N(210 kg per hectare)was applied at a ratio of 1:1 as basal fertilizer and again as a topdressing at the stem elongation stage. P (P2O5) and K (K2O) were applied as basal fertilizers at 150 kg of each per hectare. The nitrogen, phosphate, and potash fertilizers contained urea (N 46%), diammonium phosphate (P2O546%, N 18%), and potassium chloride (K2O 60%), respectively. Each treatment was irrigated twice, with 75 mm water at the stem elongation and anthesis stages.The study area was subjected to regular weeding,and insecticides were applied.Winter wheat was sown on October 15,2015 and October 12,2016,and harvested on June 15,2016 and June 10,2017,respectively.

2.3. Field and laboratory methods for assessment of trait quality

2.3.1.Percentage of productive tillers

Tiller numbers were recorded at the three-leaves unfolded,overwintering,stem elongation,booting,anthesis,and maturation stages by counting in a 0.125 m2circular area of each experimental plot. Proportion of productive tillers (%) was calculated based on the following formula:

2.3.2.Leaf area index

Leaf area index was measured at the overwintering, stem elongation, booting, and anthesis stages and at 25 days after anthesis,using a SUNSCAN Canopy Analysis System(Delta-T Devices Ltd., Cambridge, UK).

2.3.3.Nitrogen accumulation across organs

Twenty individuals were destructively sampled at the overwintering,stem elongation,booting,anthesis,and maturation stages. Samples for the overwintering stage were the whole plant; samples at the anthesis stage were divided into leaves, stem + sheath, and chaff + spike rachis tissues; and plant samples at the maturation stage were divided intoleaves,stem + sheath, chaff + spike rachis,and grain tissues.These samples were subjected to 105 °C for 30 min until the tissues were no longer green, and then dried at 70 °C to constant weight, which was recorded. Determination of nitrogen content and nitrogen accumulation in plant organs employed the H2SO4-H2O2-indigo blue colorimetric method[23]at the overwintering, anthesis,and maturation stages.

Table 2-Growing degree days(GDD)and sunshine hours(SH)at the experimental site in Xinxiang.Data source:Meteorological Observation of Xinxiang County,Henan province,China.

2.3.4. Gas exchange in the flag leaf

Net photosynthetic rate (Pn), transpiration rate (Tr), and stomatal conductance (gs) of the flag leaf were measured using the photosynthesis system LI-6400 (LI-COR, USA). For measurement of the gas exchange parameters, the environmental CO2concentration was 400 ± 12 μmol mol-1. Internal light intensity provided by a red and a blue light source was set to 1200 μmol m-2s-1.The measurements were taken at 5,15,and 25 days after anthesis (DAA).

2.3.5. Yield components

Spike number, kernel number per spike, and 1000-kernel weight at the maturation stage were determined and averaged across 50 plants per plot. These 50 plants were also weighed to determine biomass. Grain yield was directly measured by harvesting a 10 m2area from the experimental plot.

2.3.6. Soil bulk density and water content

After the wheat was harvested,a 1 m deep pit was dug in each experimental plot,resulting in a vertical soil profile.Every 20-cm layer was considered a soil layer, and sampling was performed from 0 to 20 cm to the bottom of the sample(80-100 cm). Soil bulk density was determined using the core sampling method [27]. Soil samples from 0 to 100 cm (in 20-cm layers) were collected, and determination of soil water content followed the drying method[28].During each stage of wheat growth and at the overwintering stage, three samples from each plot were selected within a circular area of 0.125 m2, and the population tillers generated before winter were counted.

2.4. Calculation methods for water and nitrogen usage

2.4.1. Soil water storage and plant population water use efficiency

Following Hou et al. [29], these values were calculated based on the following formulae:

where SWSidenotes soil water storage in soil layer i(mm),Withe mass water content of soil layer i (%), Dithe soil bulk density in soil layer i(g cm-3), and Hithe thickness of the soil layer(cm).

where E denotes water consumption during the entire crop growing stage (mm), R the precipitation during this growth period (mm), ΔW the amount of soil water storage of the 0-100 cm soil layers at the sowing stage minus that value at the maturation stage (mm), M the irrigation amount during the growth period (mm), and K the amount of groundwater recharge during the growth period (mm). When the depth of buried groundwater is ＞4 m,the K value can be ignored.In this study, the groundwater depth in the field was ＞5 m, so the K value was 0.

WUE=Y/E

WUE denotes water use efficiency (kg ha-1mm-1), Y denotes grain yield(kg ha-1), and E is as above[28].

2.4.2.Nitrogen calculation

Values for nitrogen were calculated following Zhao and Yu[30],using the following formulae:

Pre-anthesis accumulated nitrogen translocation(PANT) =nitrogen accumulation in vegetative organs at the anthesis stage - nitrogen accumulation in vegetative organs at the maturation stage

Contribution to N in kernels (% )of PANT

=PANT/Nitrogen accumulation in kernels×100%

Nitrogen accumulation after anthesis (NAAA) = nitrogen accumulation in the plant at the maturation stage - nitrogen accumulation in the plant at the anthesis stage

Contribution to N in kernels (% )of NAAA

=NAAA/nitrogen accumulation in kernels×100%

N uptake efficiency

=nitrogen accumulation in plant/applied amount of nitrogen

N use efficiency

=grain yield/nitrogen accumulation in the plant

N productive efficiency

=grain yield/applied amount of nitrogen

2.5. Statistical analysis

Data for each trait were analyzed with ANOVA using SPSS 21.0(SPSS Inc., Chicago, IL, USA). Treatment means were compared using Duncan's multiple range tests at P ＜0.05 [31].Correlation was assessed by Pearson correlation analysis.Figures were prepared using Microsoft Excel 2013 (Microsoft Corporation,Redmond,Washington,USA).

3. Results

3.1. Tridimensional uniform sowing reduced water consumption and increased ΔW with decreased soil water content after the wheat was harvested

Compared with conventional drilling D,the average soil water content (0-100 cm) after wheat harvest decreased by 1.7% in 2015-2016 and by 2.1% in 2016-2017 (Fig. 1). The tridimensional uniform sowing treatment U significantly increased ΔW, by 21.8% in 2015-2016 and by 13.6% in 2016-2017. Total water consumption during the growth stages decreased by 9.5%in 2015-2016 and by 8.1%in 2016-2017.Water consumption was minimized in the 270 × 104plants ha-1treatment under both U and D in 2015-2016 (432.8 mm and 483.1 mm)and in 2016-2017(412.9 mm and 455.1 mm)(Fig.2).

Fig.1- Soil water content(0-100 cm soil layer)after wheat harvest at three different densities(180 × 104,270 × 104,and 360 × 104 plants ha-1) with tridimensional uniform sowing(U)and conventional drilling(D)in 2015-2016 and 2016-2017.Horizontal bars represent the standard errors of the means.

3.2.Effects of sowing pattern and planting density on growth,biomass, leaf area index, and percentage of productive tillers

For all planting densities across both study years, production increased with density under U. The number of tillers produced, leaf area index, and accumulated dry weight increased significantly under U in comparison with D at each measured interval (Fig. 3, Tables 3-4). Under the U treatment, increases in planting density increased the number of tillers, leaf area index, and dry weight, though the increases in leaf area index and dry weight were not significantly different between the U270 and U360 treatments.Under the D treatment,increases in density did not show the same effect.Number of tillers,leaf area index,and dry weight were highest in the D270 treatment from the stem elongation to the maturation stage, with the number of tillers and dry weight declining substantially in the D360 treatment after the anthesis stage.The percentage of productive tillers decreased gradually with increasing density under U and D after the anthesis stage, though the decrease was not significant between the 270 × 104and 360 × 104plants ha-1treatments(Fig. 4). Percentage of productive tillers was lower under D than under U.

3.3. Effects of sowing pattern and planting density on gas exchange in the flag leaf after the anthesis stage

Fig.2- Water consumption and ΔW of winter wheat planted at three different densities(180 × 104,270 × 104, and 360 × 10 4 plants ha-1)per with tridimensional uniform sowing(U) and conventional drilling(D)in 2015-2016 and 2016-2017.ΔW indicates the value of soil water storage of the 0-100 cm soil layer at the sowing stage minus that at the maturation stage(mm).Vertical bars represent the standard errors of the means.Different letters indicate that means are significantly different(P ＜0.05).

The net photosynthetic rate (Pn), transpiration rate (Tr), and stomatal conductance(gs)of the flag leaf decreased gradually with increasing density under U and D after anthesis stage,and Pn, Tr, and gsdecreased rapidly under D in comparison with U(Fig.5).Pn,Tr,and gswere reduced under U in comparison with D at the densities of 180 × 104and 270 × 104plants ha-1but increased at the density of 360 × 104plants ha-1.

3.4.Effects of sowing pattern and planting density on nitrogen accumulation across organs and life stages

For all planting densities across both study years, nitrogen accumulation increased with density under U. Nitrogen accumulation was measured at the overwintering, anthesis,and maturation stages,and increased significantly under U in comparison with D (Fig. 6). Under the U treatment, increases in planting density increased nitrogen accumulation, though the increase was not significant between the U270 and U360 treatments. Under the D treatment, increases in density did not show the same effect.Nitrogen accumulation was highest under the D270 and lowest under the D360 treatment.

Table 3-Effects of sowing pattern and planting density on leaf area index in winter wheat.

The amount of nitrogen accumulated in all wheat organs measured at the anthesis stage increased significantly under the U treatments in comparison with the corresponding D treatments(Table 5).With increased planting density in the U treatment, nitrogen accumulation increased in all organs,although this difference was not significant for the U270 and U360 treatments. For the D treatment, which had lower nitrogen accumulation at the anthesis stage,nitrogen content of organs reached their highest values in the D270 treatment and their lowest in the D360 treatment.

Nitrogen accumulation and allocation at the maturation stage showed different patterns under the U and D treatments. Under the U treatment, nitrogen accumulated in leaves and in chaff + spike rachis decreased, while nitrogen of stem + sheath and kernels as well as overall nitrogen in a plant increased. In general, increasing planting density under the U treatment resulted in decreased nitrogen in leaves and chaff + spike rachis and increased nitrogen in the stem +sheath, kernels, and in plants overall, though the differences between the U270 and U360 treatments were not always significant. With changing planting density under D, nitrogen accumulation in the leaf and chaff + spike rachis showed the lowest value in the D270 treatment,while nitrogen accumulation in the stem + sheath and the grain was highest in the D270 treatment.These values were lowest in the D360 treatment.

Fig.3- Number of tillers of winter wheat planted at three different densities(180 × 104,270 × 104,and 360 × 104 plants ha-1)with tridimensional uniform sowing(U)and conventional drilling(D)in 2015-2016 and 2016-2017.Vertical bars represent the standard deviations of the means.TS,trefoil stage;OW,overwinter stage;SE,stem elongation stage;BS,booting stage;AS,anthesis stage;MS,maturation stage.

Table 4-Effects of sowing pattern and planting density on cumulative dry weight in winter wheat(kg ha-1).

3.5. Effects of sowing pattern and planting density on translocation of accumulated nitrogen

3.5.1.Accumulation and translocation of nitrogen pre-and postanthesis

Pre-anthesis accumulated nitrogen translocation (PANT) and nitrogen accumulation after anthesis (NAAA) increased significantly in the U treatment in comparison with the D treatment (Table 6). With increased planting density in the U treatment, both values increased, though there was no detectable difference between the U270 and U360 treatments.Increasing density with the D treatment did not show the same pattern; PANT and NAAA reached their highest values in the D270 treatment and were lowest in the D360 treatment.There was no significant difference in the contribution of PANT to N in kernels between the U and D treatments.Increased planting density had no significant effects on the N contribution of kernels for the U treatment,while effects were variable in the D treatment.

Fig.4- Proportion of productive tillers of winter wheat planted at three different densities(180 × 104,270 × 104,and 360 × 104 plants ha-1) with tridimensional uniform sowing(U)and conventional drilling(D)in 2015-2016 and 2016-2017.Vertical bars represent the standard errors of the means.Different letters indicate that means are significantly different(P ＜0.05).

3.5.2. Pre-anthesis accumulated nitrogen translocation amount from plant organs

The pre-anthesis accumulated nitrogen translocation (PANT)from plant organs increased under U in comparison with D,and the increase in leaf and chaff + spike rachis was significant (Table 7). Increasing planting density increased the PANT from plant organs under the U treatment, but this difference was no longer significant at higher densities.The D treatment showed the opposite effect, with the lowest PANT in sampled organs at the highest planting density.

The effect of sowing treatment and planting density on the translocation from sampled organs to wheat kernels was variable. The contribution from leaves did not differ detectably across treatments or planting density. The contribution from stem + sheath decreased with increasing density in the D treatment in the 2015-2016 experimental year, but otherwise did not vary across treatments. With increased planting density, the contribution from chaff + spike rachis increased for the U treatment but decreased for the D treatment. The U treatment at elevated planting density is thus beneficial for increasing the contribution to N in kernels of PANT from multiple organs. These same effects were not seen in the D treatment, where increased density limited PANT from the studied organs.

3.6. Effects of sowing pattern and planting density on grain yield and protein content

The U treatment substantially increased grain yield in comparison with the D treatment. For 2015-2016 in the U treatment,grain yield was increased by 575.43-1707.00 kg ha-1(7.0%-21.8%),while spike number,kernel number per spike,and 1000-kernel weight varied from 9.5%to 26.9%,-12.8%to-3.5%,-5.5%to 3.2%, respectively. In 2016-2017 and U, grain yield was increased by 575.50-1707.00 kg ha-1(6.9%-21.3%), and spike number, kernel number per spike, and 1000-kernel weight varied from 9.2%to 25.9%,-11.9%to-1.8%,and - 5.4%to 3.5%.Thus,the increase in spike number was the main driver of the increase in grain yield under U. With increasing planting density under U, spike number and grain yield increased and 1000-kernel weight decreased,though there was no significant difference between the U270 and U360 treatments.Under the D treatment,spike number and grain yield were the highest at a planting density of 270 × 104plants ha-1, and spike number,kernel number per spike, 1000-kernel weight, and grain yield were the lowest at the highest planting density, 360 × 10 4 plants ha-1(Table 8). Grain protein content was slightly reduced under the U treatment in comparison with D,but this difference was not always detectable,and planting density had similarly minimal effects on grain protein(Table 8).

Fig.5-Net photosynthetic rate(Pn),transpiration rate(Tr),and stomatal conductance(gs)in the flag leaf of winter wheat planted at three different densities(180 × 104,270 × 104,and 360 × 104 plants ha-1)with tridimensional uniform sowing(U)and conventional drilling(D),5(DAA5),15(DAA15), and 25(DAA25)days after anthesis in 2015-2016 and 2016-2017.Vertical bars represent the standard errors of the means.

Fig.6-Nitrogen accumulation of the overwinter stage of winter wheat planted at three different densities(180 × 104,270 × 104,and 360 × 104 plants ha-1)with tridimensional uniform sowing(U)and conventional drilling(D)in 2015-2016 and 2016-2017.Vertical bars represent the standard errors of the means.OW indicates overwintering stage,AS anthesis stage,and MS maturation stage.

3.7. Effects of sowing pattern and planting density on water use efficiency and nitrogen use efficiency

Tridimensional uniform sowing increased the water use efficiency of winter wheat in comparison with wheat sown by conventional drilling.Efficiency increased across the three planting densities by 18.4%-32.9%in 2015-2016 and by 16.7%-29.9% in 2016-2017. Water use efficiency was highest at the planting density of 270 × 104plants ha-1(Table 9). The U treatment also increased the efficiency of nitrogen uptake,use, and production. With increasing planting density under the U treatment, nitrogen uptake and nitrogen productive efficiencies also increased, but there was no significant difference in nitrogen production efficiency between the U270 and U360 treatments. Under D, nitrogen uptake and nitrogen productive efficiencies were highest at the planting density of 270 × 104plants ha-1and lowest with 360 × 104plants ha-1.

Table 5-Effects of sowing pattern and planting density on nitrogen accumulation in various organs of wheat at the anthesis and maturation stages(kg ha-1).

Table 6-Effects of sowing pattern and planting density on pre-anthesis accumulated nitrogen translocation and nitrogen accumulation after anthesis.

3.8. Correlations between components of pre-anthesis accumulated nitrogen translocation, nitrogen accumulation after anthesis (NAAA) and yield

Values for PANT, NAAA, and components of yield averaged across planting densities showed significant correlations(Table 10). In 2015-2016, there were significant positive correlations between yield and PANT, yield and NAAA, and grain protein and NAAA.Among individual plant organs,yield was most tightly correlated with the PANT value for chaff +spike rachis. Grain protein was highly correlated with the PANT value of stem + sheath rather than of leaf or chaff +spike rachis. In 2016-2017, both yield and grain protein showed a significant positive correlation with PANT, and yield showed a significant positive correlation with NAAA.For both yield and grain protein,the correlation was highest with the PANT value of the leaf, in comparison with all other organs. Across the two experimental years, grain protein content was not significantly correlated with PANT or NAAA.Yield and grain protein showed a higher correlation with PANT than with NAAA. In the wetter year (2016-2017) there were higher correlations between the yield and chaff + spike rachis value for PANT than in the year with less precipitation(2015-2016).The wetter year also showed a higher correlationbetween grain protein yield and stem + sheath value for PANT, and higher correlations of leaf PANT with yield and grain protein.

Table 7-Effects of sowing pattern and planting density on pre-anthesis accumulated nitrogen translocation from organs and their contribution to wheat kernels.

Table 8-Effects of sowing pattern and planting density on components of yield and grain protein.

4. Discussion

The challenge of maximizing grain production by optimization of water use,planting patterns,and fertilization has long been an important goal in agricultural research. Impending climate changes that will limit water,as well as the expenses associated with production of nitrogen fertilizer drive the cost, and thus productivity, of large-scale wheat producing farms. In this two-year study we examined a range of productivity metrics across three planting densities for comparison of the newer tridimensional uniform sowing technique with the more widely-used conventional drilling technique in order to lay the groundwork for broader acceptance and integration of this practice in winter wheat farming. Tridimensional uniform sowing can significantly increase the utilization of existing soil water (ΔW in Fig. 2) in the wheat growing period, especially during the winter and spring seasons when precipitation is rare. This can benefit growers by limiting the amount of irrigation necessary for the same crop yield, as well as protecting environmental resources by limiting the drawing of groundwater.Severe water stress is known to reduce grain yield and dry matter accumulation in wheat [12], whereas increased water efficiency can ameliorate these effects[32].

For all plant densities, tridimensional uniform sowing performed better than conventional drilling in the metrics of number of tillers produced, leaf area index, cumulative dry weight,and overall nitrogen,and these values increased with plant density.Pn,Tr,and gsof the flag leaf decreased gradually with increasing density under tridimensional uniform sowing and conventional drilling after the anthesis stage, likelybecause higher planting densities are associated with faster flag-leaf senescence [25]. Pn, Tr, and gsof the flag leaf were much lower under conventional drilling than under tridimensional uniform sowing. Pn, Tr, and gsdecreased under tridimensional uniform sowing as compared to conventional drilling at the densities of 180 × 104and 270 × 104plants ha-1but increased at the density of 360 × 104plants ha-1. These results indicate that the tridimensional uniform sowing mode is preferable for planting at high densities because it creates an enhanced canopy structure, improving canopy interception and use of light [18]. Although the photosynthetic capacity of any individual plant is decreased, the advantage of overall total productivity (measured by number of tillers,leaf area index, cumulative dry weight, and overall nitrogen)compensates for the reduction in individual photosynthesis.Thus, the tridimensional uniform sowing treatment contributed to the establishment of strong plants with extensive lateral growth and a higher percentage of productive tillers,even at high densities, resulting in elevated dry matter production.Under the conventional drilling treatment,planting densities could be increased only to 270 × 104plants ha-1before reduced biomass and nitrogen accumulation became apparent. This finding is consistent with previous studies[25,33],in which increased crop density and spatial uniformity resulted in higher crop yields owing to weed suppression and changes in nitrogen use.

Table 9-Effects of sowing pattern and planting density on water use efficiency and nitrogen use efficiency.

Table 10-Correlations between components of PANT,NAAA,and yield under tridimensional uniform sowing.

Sowing pattern markedly affected nitrogen accumulation across plant organs and life stages, with tridimensional uniform sowing outperforming conventional drilling, and this superior performance increased with plant density.Tridimensional uniform sowing promotes nitrogen accumulation in multiple organs of wheat at the anthesis stage and allows for favorable nitrogen translocation to the grain from chaff + spike rachis [23]. Even at the highest density of tridimensional uniform sowing (360 × 104plants ha-1), nitrogen accumulation was elevated, with no significant decline from the planting density of 270 × 104plants ha-1. This pattern did not hold true for the conventional drilling treatment, where increasing planting density from 270 × 104to 360 × 104plants ha-1reduced overall nitrogen accumulation and increased nitrogen residue in various plant organs,especially in the stem + sheath.

Pre-anthesis accumulated nitrogen translocation (PANT)and nitrogen accumulation after anthesis (NAAA) measurements of nitrogen content and mobility were significantly improved under the tridimensional uniform sowing treatment and with increased planting density.It can be concluded that tridimensional uniform sowing increased the amount of favorable nitrogen translocation and accumulation, though the contribution to kernels was unchanged. Increasing planting density under tridimensional uniform sowing did not show the same negative effects as in the conventional drilling treatment, under which the translocation of PANT decreased as density increased. The conventional drilling treatment also displayed reduced nitrogen accumulation overall. Nitrogen absorption and mobilization are highly correlated with grain protein content and yield[34].Mobilization of nitrogen and other nutrients from pre-anthesis vegetative reserves is important because it may stabilize yield under unfavorable climatic conditions during the grain filling period[35].

This study demonstrated that the tridimensional uniform sowing mode can increase grain yield and protein content at the higher planting densities. Because the tridimensional uniform sowing can increase spike number at high planting density, it can be used to maximize yield especially in farms where all arable land is planted. At the highest planting densities, grain protein content decreases owing to increased intraspecific competition for soil N [25]. However, tridimensional uniform sowing benefits plants by creating a uniform spatial distribution,reducing competition for soil N and other nutrients very effectively at our intermediate planting density[18,23].In the 270 × 104plants ha-1treatment,grain yield was higher and grain protein content was comparable to that of the conventional drilling treatment with the same planting density. Thus, tridimensional uniform sowing at 270 × 10 4 plants ha-1resulted in increased grain yield and protein content, achieving the goal of simultaneous improvement of grain yield and quality. Changes in timing of nitrogen fertilizer application should be investigated to determine the effect of fertilizer application schedule on grain protein content[13].

Tridimensional uniform sowing outperformed conventional drilling for each metric of water and nitrogen use efficiency,and these values improved or remained stable with increased planting density. Increased density and spatial uniformity has been demonstrated to increase yield in crops[33]. Further research is needed to determine whether these metrics of nitrogen use show the same patterns across different varieties of wheat, as they are known [36] to vary with genetic background. Additional research is also needed to examine the role of climatic patterns in the relationship between components of PANT, NAAA,and yield. The present study has shown that although certain correlations (e.g.between yield and NAAA) hold across dry and wet years,others may vary. More years of study that incorporate both wet and dry years will help to clarify this pattern.

5. Conclusions

Compared with conventional drilling,tridimensional uniform sowing increased the number of tillers, leaf area index, dry weight, and spike number, increasing grain yield. Nitrogen accumulation and translocation increased, leading to greater efficiency of nitrogen uptake and use efficiency,though grain protein content decreased. This new sowing method also maximized water use efficiency. The highest water use efficiency was observed with 270 × 104plants ha-1and tridimensional uniform sowing.Number of tillers,leaf area index,dry weight, nitrogen accumulation, translocation, nitrogen uptake efficiency,nitrogen productive efficiency,and nitrogen use efficiency increased with planting density under tridimensional uniform sowing. There was no detectable difference in grain protein content at the density of 270 × 104plants ha-1under tridimensional uniform sowing compared with conventional drilling.Tridimensional uniform sowing at 270 × 104plants ha-1resulted in production of the highest number of the healthiest plants and an increase in the number of grain-bearing spikes with commensurate improvement in yield, efficiency of water and nitrogen use, thus simultaneously maximizing both yield and grain protein.This optimal sowing pattern and planting density can be reliably employed to increase output of winter wheat in the North China Plain and other agricultural regions with similar environmental conditions, soil conditions, and available water.

Acknowledgments

This work was supported by the National Key Research and Development Program of China(2016YFD0300407),Earmarked Fund for China Agriculture Research System (CARS-03), and Agricultural Technology Test Demonstration and Service Support(118003).