Determination of critical nitrogen dilution curve based on leaf area index for winter wheat in the Guanzhong Plain, Northwest China

QIANG Sheng-cai , ZHANG Fu-cang, Miles Dyck, ZHANG Yan , XIANG You-zhen, FAN Jun-liang

1 College of Urban and Rural Construction, Shanxi Agricultural University, Taigu 030801, P.R.China

2 Key Laboratory of Agricultural Soil and Water Engineering in Arid and Semiarid Areas, Ministry of Education/Northwest A&F University, Yangling 712100, P.R.China

3 Department of Renewable Resources, University of Alberta, Edmonton T6G 2H1, Canada

Abstract Excessive use of nitrogen (N) fertilizers in agricultural systems increases the cost of production and risk of environmental pollution. Therefore, determination of optimum N requirements for plant growth is necessary. Previous studies mostly established critical N dilution curves based on aboveground dry matter (DM) or leaf dry matter (LDM) and stem dry matter(SDM), to diagnose the N nutrition status of the whole plant. As these methods are time consuming, we investigated the more rapidly determined leaf area index (LAI) method to establish the critical nitrogen (Nc) dilution curve, and the curve was used to diagnose plant N status for winter wheat in Guanzhong Plain in Northwest China. Field experiments were conducted using four N fertilization levels (0, 105, 210 and 315 kg ha-1) applied to six wheat cultivars in the 2013-2014 and 2014-2015 growing seasons. LAI, DM, plant N concentration (PNC) and grain yield were determined. Data points from four cultivars were used for establishing the Nc curve and data points from the remaining two cultivars were used for validating the curve. The Nc dilution curve was validated for N-limiting and non-N-limiting growth conditions and there was good agreement between estimated and observed values. The N nutrition index (NNI) ranged from 0.41 to 1.25 and the accumulated plant N deficit (Nand) ranged from 60.38 to -17.92 kg ha-1 during the growing season. The relative grain yield was significantly affected by NNI and was adequately described with a parabolic function. The Nc curve based on LAI can be adopted as an alternative and more rapid approach to diagnose plant N status to support N fertilization decisions during the vegetative growth of winter wheat in Guanzhong Plain in Northwest China.

Keywords: winter wheat, leaf area index, critical nitrogen concentration, nitrogen nutrition index, nitrogen diagnosis

1. Introduction

Population growth and decline in the area of available arable land, means maximizing crop yield per unit area has become a major aim of agricultural production in China(Yao et al. 2014a; Zheng et al. 2019). Nitrogen (N) is an essential nutrient for plant growth and is a key agricultural input to increase yield (Zhao et al. 2014; Wang et al. 2019).However, because of poor understanding of plant N nutrition diagnosis, excessive use of N is a common practice among farmers. High N rates may not produce economic yield increases, and may adversely impact the environment,causing soil acidification (Guo et al. 2010) and soil and water pollution via N leaching (Zhu et al. 2005). Therefore,crop N nutrition diagnosis is an important area of research for improvement of N-use efficiency and for environmental protection (Lemaire et al. 2008; Yao et al. 2014b)

Critical nitrogen concentrations can be used to estimate N requirements of crops under various field conditions for reliable fertilizer recommendations. The critical nitrogen (Nc)dilution curve can be determined using the power equation(Greenwood et al. 1990):

where Ncis the total N concentration in aerial parts expressed as dry matter; W is the total aerial biomass expressed in t ha-1; a and b are positive constants, where a represents the N concentration in the dry matter (DM) when W=1 t ha-1,and b is a statistical parameter governing the slope of the relationship.

The critical nitrogen dilution curves established by previous studies were mainly based on DM or leaf dry matter (LDM) and stem dry matter (SDM). Successful examples include summer maize (Yue et al. 2014), spring maize (Plenet and Lemaire 1999; Li et al. 2012), spring wheat (Ziadi et al. 2010), winter wheat (Justes et al. 1994;Yue et al. 2012; Yao et al. 2014a) and rice (Ata-Ul-Karim et al. 2014a; Yao et al. 2014b). Although Nccurves based on whole plant DM or LDM and SDM can be used for diagnosis of plant N status, in practice, these methods require destructive and time-consuming procedures; it may take 3-4 days to determine whole plant DM, LDM, SDM and plant N content (PNC). Remote sensing may be used to estimate plant biomass at different growth stages (Zeng and Chen 2018), however, the high cost of remote sensing has hindered the widespread adoption of this technology in agriculture.

Leaves are important organs for photosynthesis, and N uptake by plants is optimized to satisfy the growth demand of leaves (Yao et al. 2014a). Leaf area index (LAI) is a fundamental agronomic and environmental indicator that can be used as a reference tool for growth monitoring, yield forecasting and management optimization in crop production(Zheng et al. 2018b; Qiang et al. 2019). Nitrogen drives plant DM production through control of both LAI and the amount of N per unit of leaf area (Hirel et al. 2007). Nitrogen not only affects the cumulative growth of dry matter, but also the expansion of leaf blades. Lemaire et al. (2007, 2008)showed that allometric relationships between DM and LAI(LAI=kDMc) exist under non-N-limiting conditions. In view of this relationship, it may be possible to develop critical N dilution curves based on LAI. Related research (Zhao et al. 2014) in wheat and rice (Ata-Ul-Karim et al. 2014b)has proved the feasibility of modeling Nccurves based on LAI. Thus, modeling based on LAI might be a simpler and more rapid method than modeling based on whole plant DM,LDM or SDM. In practice, LAI can be easily and quickly measured by using instruments such as the LAI-2000 plant canopy analyzer (Li-Cor, Lincoln, USA).

The Guanzhong Plain is one of the most important areas for winter wheat and summer maize production in China (Zheng et al. 2018a; Yan et al. 2019). In this region,the problem of excessive N application is common among farmers (Zhao and Yu 2006; Tong et al. 2007), although progress has been made in N nutrition diagnosis. Li et al. (2015) developed a DM-Nccurve for wheat cultivar Xiaoyan 22, while Zhao et al. (2014) developed a LAI-Nccurve for winter wheat in the eastern China. However,Nccurves vary among different regions, species, and even genotypes within species (Justes et al. 1994; Yue et al. 2012). At present, with progress in crop-breeding technology, many wheat genotypes now exist. Due to differences in nitrogen absorption and utilization among different varieties, significant differences exist in N dilution curves (Zhao et al. 2014). Therefore, determination of a critical Ncdilution curve based on multiple varieties for the Guanzhong Plain is necessary.

The present study was intended to develop an appropriate Ncdilution curve based on LAI for winter wheat cultivars in the Guanzhong Plain, and compare this curve with existing whole plant DM- and LDM-based Ncdilution curves. The study also aimed to assess the reliability of this new Nccurve for estimating the N nutrition status of winter wheat.The present study may provide a methodology for crop N status diagnosis and may be used for guiding precision N management during the growth period of winter wheat.

2. Materials and methods

2.1. Site description

Field experiments were conducted in the same field from 2013 to 2015 at Northwest A&F University in Yangling,Shaanxi Province, China (34°18′N, 108°24′E, 521 m altitude). The site is located in a semi-arid to sub-humid climate zone, and has a mean annual rainfall of 550 mm.The soil is alluvial (Fluvisol) with a pH of 8.15, organic matter content of 16.9 g kg-1, total N content of 0.98 g kg-1,total P content of 0.76 g kg-1, total K content of 14.9 g kg-1,available N content of 69.8 mg kg-1(Subbiah and Asija 1956), available P content of 10.07 mg kg-1and available K content of 163.30 mg kg-1(Hatti et al. 2018).

2.2. Experimental design

A randomized block design with three replicates was used for the study. Four N application rates (0, 105, 210 and 315 kg N ha-1) were applied to six winter wheat cultivars (Table 1). Four varieties (XY22, XN979, XM23 and LX988) were randomly selected to develop the Nccurve (Experiment 1), while the remaining two varieties (WN148 and XN509) were used to validate the Nccurve (Experiment 2). All plots were 21 m2and received 140 kg P2O5ha-1as diammonium phosphate and 90 kg K2O ha-1as potassium chloride as basal fertilizer.The N fertilizer rates were applied before the sowing stage(50%) and at the jointing stage (50%). The trial was run over two growing seasons. No obvious water deficits, weeds, or pests were observed during these growing seasons.

2.3. Sampling and measurement

LAI was measured using a LI-COR LAI-2000 at early spring,jointing, booting and heading stages. Measurements were taken at 17:00 every sampling date, and the probe was located about 3 cm above the crop surface. At each stage,20 plants per plot were collected for determination of DM(t ha-1) and PNC (g kg-1). Plants were sampled from the inner/central rows of each plot to minimize the border effects.Samples for DM determination were placed in a forced-draft drying oven at 75°C until constant weight and the final weight was recorded. The dried plant material was ground to pass through a 1-mm screen and a sample was digested using H2SO4-H2O2, after which the plant N concentration was determined using the Kjeldahl method (Horowitz 1970).

2.4. Data analysis

Ncvalues were determined using the methodology described by Justes et al. (1994). Determination of the Ncdilution curve requires identification of critical data points at which N neither limits nor enhances plant growth. An N-limiting growth treatment was defined as a treatment where increasing N supply resulted in a significant response in DM,LAI and PNC, while a non-N-limiting growth treatment did not show an significant increase in LAI and DM in response to increased N, but did result in a significant increased PNC.Two years of data points were used either for construction of the Ncdilution curve, or to validate it. In order to calculate the critical values, the DM, LAI and PNC values in different N treatments were compared using ANOVA (SPSS Inc.,Chicago, IL, USA) with significant effects declared at the 5% probability level (P＜0.05). A power regression equation was fitted to these theoretical critical points to determine the equation of the Ncdilution curve using Model II regression(SAS Institute Inc., Cary, NC, USA).

Nitrogen nutrition indexThe NNI of wheat DM at each sampling date was determined as the PNC divided by the Ncvalue, as described for studies on potato (Bélanger et al.2001) and corn (Ziadi et al. 2010):

Accumulated nitrogen deficitThe accumulated N deficit(Nand) was determined as derived for rice crops (Ata-Ul-Karim et al. 2013) for each sampling date by subtracting the N accumulation under the Nccondition (Ncna) from actual N accumulation (Nna) under different N rates, as eqs. (3)and (4):

where DM is the aboveground dry matter under different N rates and N% is total N concentration of DM.

Model validationIn this study, the accuracy of the model was evaluated with root mean squared error (RMSE) and normalized root mean squared error (n-RMSE) in 1:1 plots,which is a common method used to identify the fitness of observed and estimated values (Fan et al. 2019b; Wu et al.2019). We considered n-RMSE≤15% as “good” agreement;15-30% as “moderate” agreement; and ≥30% as “poor”agreement (Yang et al. 2013; Fan et al. 2019a). The RMSE and n-RMSE were calculated using eqs. (5) and (6):

where n is the number of samples, siis an estimated value from the model, miis an observed value, andis theaveraged observed value.

Table 1 Treatments used for construct and validation of the critical nitrogen (Nc) dilution curve of winter wheat in the Guanzhong Plain, Northwest China

Relative grain yieldThe relative grain yield (RY) was calculated as the ratio of grain yield obtained for a given N rate with the highest grain yield among all N application rates(Ata-Ul-Karim et al. 2016). The RY was calculated as eq. (7):

where Ytis the total yield in the fertilized plot, Yhis the yield of the highest yield treatment (kg ha-1).

3. Results

3.1. Changes in LAI and aboveground DM

LAI increased from the early spring re-growth to heading stages (Fig. 1), two years of mean LAI ranging from 1.25 to 5.72, 1.29 to 6.15, 1.16 to 6.53 and 1.33 to 6.83 for varieties XY22, XN979, XM23 and LX988, respectively. N application rate exhibited a significant effect on LAI throughout the growing season. In general, the maximum LAI was obtained in the N0and N105treatment at early spring in all varieties.While, during the rest growing stages, the maximum LAI was attained in the N105and N210treatments for wheat variety.

Aboveground DM increased from the early spring to heading stages (Fig. 2), the mean value of aboveground DM ranging from 1.33 to 8.28, 1.49 to 8.92, 1.30 to 8.74 and 1.16 to 8.20 t ha-1for varieties XY22, XN979, XM23 and LX988, respectively. N application rate also exhibited a significant effect on DM throughout the growing seasons.At early spring stage, the maximum DM was attained in the N105treatment almost in all wheat varieties. While, during the rest growing stages, the maximum DM almost was attained in the N105and N210treatments for wheat varieties.

Fig. 1 Changes in Leaf area index as a function of growth stage for four winter wheat cultivars at four N fertilizer rates in two growing seasons. XY22, XN979, XM23, and LX988 are the abbreviations for Xiaoyan 22, Xinong 979, Xingmai 23, and Lunxuan 988 cultivars, respectively. The vertical bars indicate standard deviations (n=3).

3.2. Relationships between DM and N uptake and LAI and N uptake

Data collected from four cultivars (XY22, XN979, XM23 and LX988) in the 2013-2014 and 2014-2015 seasons were used to establish allometric relationships between DM and N uptake and LAI and N uptake during vegetative growth (early stage data points for DM less than 1 t ha-1and LAI less than 1 were excluded) under non-N-limiting growth conditions (Figs. 3 and 4). The relationships shown in eqs. (8) and (9) accounted for 94 and 83% of the variation between DM accumulation and N uptake and LAI and N uptake, respectively.

3.3. Relationship between LAI and aboveground DM

The relationships between LAI and DM accumulation during vegetative growth in the 2013-2014 and 2014-2015 seasons under N-limiting and non-N-limiting growth conditions are shown in Fig. 5. The relationships shown in eqs. (10) and (11) accounted for 82 and 95% of the variation between DM and LAI under N-limiting and non-N-limiting growth conditions, respectively.

Fig. 2 Changes in aboveground dry matter as a function of growth stage for four winter wheat cultivars at four N rates in two growing seasons. XY22, XN979, XM23, and LX988 are the abbreviations for Xiaoyan 22, Xinong 979, Xingmai 23, and Lunxuan 988 cultivars, respectively. The vertical bars indicate standard deviations (n=3).

3.4. Constructing an LAI Nc dilution curve for winter wheat

LAI based Nc dilution curveIn this study, data points of LAI＞1 were used to establish the Ncdilution curve. The theoretical Ncpoints of LAI were determined for each sampling date, from early spring to heading stage for four cultivars, using 14 data points in the 2013-2014 season and 16 data points in the 2014-2015 season. Only the sampling dates satisfying the statistical requirements described in Section 2.4 were used for the establishment of Nccurve (Fig. 6).

The equations fitted to the curves for the 2013-2014 and 2014-2015 seasons were not significantly different(Fcalculated=0.77＜Ftabulated(1-15)=4.54, P=0.05). Therefore, the data for both seasons were pooled, and a unique function was fitted as eq. (12):

Fig. 3 Relationship between N uptake and aboveground dry matter during vegetative growth under non-N-limiting growth conditions. Data (n=40) collected from four cultivars at four growth stages in 2013-2014 and 2014-2015 growing seasons.Data points for DM＜1 t ha-1 have been excluded.

Fig. 4 Relationship between N uptake and leaf area index during vegetative growth under non-N-limiting growth conditions. Data (n=45) collected from four cultivars at four growth stages in 2013-2014 and 2014-2015 seasons. Data points of LAI＜1 have been excluded.

Fig. 5 Relationship between leaf area index and aboveground dry matter during vegetative growth under N-limiting and non-Nlimiting growth conditions using data collected for four cultivars at four growth stages in 2013-2014 and 2014-2015 growing seasons.

Nc and PNC where LAI＜1When LAI values are ＜1, the LAI-Nccurve does not fit the measured data due to a relatively smaller decline in Ncwith increasing LAI at early compared to later growth stages. Therefore, 16 data points were used to calculate a constant Nc, resulting in a mean value of minimum PNC of 4.16% for the non-N-limiting treatments and a mean value of maximum PNC of 3.20%for the N-limiting treatments as reported by Justes et al.(1994). The average PNC of the minimum and maximum result in a constant LAI-Ncvalues was 3.68%.

3.5. Validation of the LAI critical nitrogen dilution curve

Data points (n=48) from Experiment 2 were used for validation of the curve, and results indicated the curve discriminated well between N-limiting and non-N-limiting treatments (Fig. 7). Growth stages and seasons did not significantly affect the LAI-Nccurve. All data points from the non-N-limiting treatments were close to or above the LAI-Ncdilution curve, whereas those from the N-limiting treatments were close to or below the LAI-Ncdilution curve.The accuracy of the model was evaluated with RMSE and n-RMSE, by using eqs. (5) and (6). Results showed that the RMSE of the model was 0.29 g kg-1and the n-RMSE was 11.4%, indicating “good” agreement between observed and estimated values (Fig. 8). Thus, the LAI-Ncmodel we established can be used for diagnosis plant nitrogen nutrition.

3.6. Effect of N treatment on N nutrition index

The N nutrition index (NNI) is useful for diagnosing the crop N nutrition status. Nitrogen nutrition is considered optimal when NNI=1, while NNI＞1 shows excess N absorption luxury and NNI＜1 indicates N deficiency. Substantial differences in NNI across the different N application rates and growth stages were apparent (Fig. 9). The NNI increased with increasing N rate, and ranged from 0.44 to 1.36 and 0.41 to 1.25 during 2013-2014 and 2014-2015, respectively. For both seasons, the NNI values were ＜1 for the N0and N105treatments, indicating that these application levels were insufficient for optimal N nutrition, while the values of NNI were ＞1 for the N210and N315treatments, indicating that N application levels provided N in excess of requirements.The optimal N application rate was therefore between 105 and 210 kg N ha-1.

3.7. Effect of growth stage and N treatment on accumulated N deficit

Nitrogen nutrition is considered adequate when Nand=0,deficient when Nand＞0 and excessive when Nand＜0.Significant differences in Nandbetween growth stages and N treatments were observed (Fig. 10). The deficit increased with increasing N rate, and the mean value of Nandof the four cultivars ranged from 56.93 to -21.90 kg ha-1and 63.84 to -13.95 from early spring to heading stage in 2013-2014 and 2014-2015, respectively. During each year, the mean values of Nandof the four cultivars were ＜0 for N0and N105,and ＞0 for N210and N315treatments.

3.8. Relationship between relative grain yield and NNI

Fig. 6 Critical nitrogen (Nc) data points used to define the critical N dilution curve. The solid line represents the critical N dilution curve, describing the relationship between the critical N concentration and leaf area index of winter wheat in Guanzhong Plain,Northwest China. The dotted lines represent the confidence intervals (P=0.95).

Fig. 7 Validation of critical nitrogen (Nc) dilution curve using the independent data set from the Experiment 2. The solid curved line represents the Nc curve based on leaf area index, when leaf area index (LAI)＞1.

Fig. 8 Relationship between simulated and observed critical N concentrations for four growth stages from 2013 to 2015.

Fig. 9 Mean nitrogen nutrition index (NNI) as a function of growth stage for four winter wheat cultivars and at four N rates in two growing seasons. The reference line at NNI=1 represents optimal N nutrition, while NNI＞1 indicates excess N fertilizer application and NNI＜1 indicates N deficiency. XY22, XN979, XM23, and LX988 are the abbreviations for Xiaoyan 22, Xinong 979, Xingmai 23,and Lunxuan 988 cultivars, respectively.

The relationship between relative RY and NNI was well described by a second-order polynomial equation(RY=-1.19NNI2+2.42NNI-0.25), which accounted for 94% of the total variation in RY (Fig. 11). Based on this relationship,for NNI=1, the relative grain yield was near 1.0, while for NNI＞1 or NNI＜1, the relative grain yield decreased. This suggests that both inadequate and excessive of use N lower the relative grain yield, while the optimal N application rate results in the maximum relative grain yield.

4. Discussion

4.1. Comparison with other critical nitrogen dilution curves

The Ncdilution curve based on DM or LDM has been previously determined for different crop species and regions(Table 2). The Ncdilution curve we established on LAI was Nc=4.36LAI-0.46, where parameter acrepresents the Ncwhen the LAI=1. Early in the season, our selected constant LAINcvalue was 3.68%, which is consistent with Lemaire et al.(2007) who indicated Ncis constant when the LAI＜1. This selection was justified because during the early stages of development, plants grew predominantly in isolation from one another and with limited competition for light. Under such circumstances, plant N concentration in DM does not decrease significantly as leaf area index increases.

A number of differences in curve parameters are evident between our study and the earlier studies. Firstly, the acparameter (4.36%) in the present study was lower than the acvalue (5.35%) reported by Justes et al. (1994). The reasons for this difference might be: (1) winter wheat in France has a higher grain N concentration (2.2-2.7%; Martre et al. 2003)than that of winter wheat in China (2.01-2.36%; Tong et al.2007); and/or (2) the heading stage in the Guanzhong Plain occurs earlier than that of in France (Brancourt-Hulmel et al. 2003), resulting in higher N uptake. Secondly, the parameter acin the present study (4.27%) was also slightly lower than that reported by Li et al. (2015) for wheat in the Guanzhong Plain (4.64%). This difference is likely due to a time difference based on our observation that when LAI Nchad already started to decline, the DM of 1 t ha-1was still maintained. Thirdly the parameter acin the present study was slightly higher than that reported by Zhao et al. (2014)for wheat in East China (4.06%). The main reason for this difference is likely that the growth period of winter wheat in East China is shorter than that in the Guanzhong Plain, so a greater LDM in the Guanzhong Plain was achieved due to a significant positive correlation between LDM and LAI.

Fig. 10 Mean accumulated N deficit (Nand) as a function of growth stage for four winter wheat cultivars at four N rates in two growing seasons. The reference line at Nand=0 represents optimum N nutrition, while Nand＞0 indicates N deficiency and Nand＜0 indicates luxury consumption. XY22, XN979, XM23, and LX988 are the abbreviations for Xiaoyan 22, Xinong 979, Xingmai 23,and Lunxuan 988 cultivars, respectively.

Fig. 11 Relationship between relative grain yield (RY) and nitrogen nutrition index (NNI) for four winter wheat cultivars.The NNI data were averaged over two years.

For the model based on N concentration in LDM, the acparameter (4.36%) in the present study was higher than the acparameter reported by Qiang et al. (2015) in the Guanzhong Plain (3.96%). The reason for this difference may be that LAI was greater than 1 under non-N-limiting growth conditions, so that LDM reached 1 t ha-1earlier than LAI reached 1.

The b parameter describes the decline in PNC with crop growth, and therefore depends on shoot N uptake relative to LAI increase. The decline in PNC during vegetative growth can be attributed to the decrease in N concentration per unit leaf area of shaded leaves (Sadras et al. 1993). Compared with LDM model, the parameter we established in this study is lower than that reported by Qiang et al. (2015), probably because the LAI increased faster than LDM under non-Nlimiting growth conditions, so the Ncdeclined quickly.

Differences existed between C3and C4crops for the parameters a and b (Table 2). The mean value of parameter a in C3species was greater than C4species, no matter whether the model was based on DM (Justes et al. 1994; Li et al. 2015 ) or LDM (Yao et al. 2014a, b; Qiang et al. 2015;Zhao et al. 2017). The parameter b was higher in C3species than C4species when the model was based on DM, but was lower in C3species when the model was based on LDM.

4.2. Diagnosis of N nutrition status and accumulated N deficit

The present Ncdilution curve can be used for diagnosing the status of crop N nutrition, and to make N management strategies for optimizing N rates. The NNI can effectively distinguish conditions of N deficiency, optimal levels or N surplus. Data showed that NNI varied from 0.41 to 1.25 in the two growing seasons and was affected by the N application rate, growth stage and growing season (Fig. 9). The optimal N application rate for winter wheat in the Guanzhong Plain is between 105 to 210 kg ha-1, which is consistent with Tong et al. (2007), Zhao et al. (2006), and Li et al. (2015). The main limitation of using these curves for N management within the growing season is the need to determine the LAI and plant N concentration, and these procedures are not within the expertise of farmers (Lemaire et al. 2008). With the acceleration of agricultural development, the problemmay be solved using non-destructive sampling means,including remote sensing (Delegido et al. 2013; Zhao et al.2014). Thus the NNI could be used to assess crop N status and model Nandto determine the requirement for additional N fertilization using established relationships between ΔN and ΔNandor ΔNNI (Ata-Ul-Karim et al. 2014 b; Yao et al. 2014 b).

Table 2 Comparison of the parameters of the critical nitrogen (Nc) dilution curve for different regions and measured parameters of wheat, maize and rice

5. Conclusion

The allometric relationships between DM increase and LAI expansion are strictly proportional, which allowed an Ncdilution curve based on LAI to be developed for winter wheat in the Guanzhong Plain of Northwest China. The curve for winter wheat was described by Nc=4.36LAI-0.46, when the LAI＞1, for LAI＜1, the constant critical value Nc=3.68%was used. The equation was derived from data obtained at four growth development stages from four cultivars over four N fertilizer application rates and two growing seasons.The equation was validated using data from two separate cultivars under the same growing seasons and fertilizer conditions. These tests confirmed that development stage and cultivar did not significantly affect the Nccurve. The values of the NNI on different sampling dates were generally＜1 under N-limiting conditions and ＞1 under non-N-limiting conditions. There was a significant positive relationship between NNI and the N application rate across the four development stages, which suggests that the Ncdilution curve can be used as a tool for diagnosing the N status of winter wheat. The model Nandcan be used to calculate requirements of further N fertilization during crop growth.

Acknowledgements

We are grateful for the financial support from the National Key Research and Development Program of China(2017YFC0403303) and the Shanxi Agricultural University of Science and Technology Innovation Fund, China (2016YJ07 and 2016007).