DU Ya-dan, CUI Bing-jing, ZHANG Qian, SUN Jun, WANG Zhen, NIU Wen-quan

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

Abstract Oilseed rape is one of the most important oil crops globally. Attaining the appropriate cultivation method (planting pattern and nitrogen level) is necessary to achieve high yield, quality and resource utilization efficiency. However, the optimal method for oilseed rape varies across countries and regions. The objective of the present study was to determine an appropriate cultivation method, including planting pattern and nitrogen application, for winter oilseed rape in northwestern China. Two planting patterns: ridge film mulching and furrow planting (RFMF) and flat planting (FP), and six nitrogen (N) amounts: 0(N0), 60 (N60), 120 (N120), 180 (N180), 240 (N240), and 300 (N300) kg N ha-1 were applied across three growing seasons(2014-2017). Three comprehensive decision analysis methods: principal component analysis, grey correlation degree analysis and the combined entropy weight and dynamic technique for order preference by similarity to ideal solution method were used to evaluate the growth and physiological indicators, nutrient uptake, yield, quality, evapotranspiration, and water use efficiency of winter oilseed rape. Planting pattern, nitrogen amount and their interaction significantly affected the indicators aforementioned. The RFMF pattern significantly increased all indicators over the FP pattern. Application of N also markedly increased all the indicators except for seed oil content, but the yield, oil production and water use efficiency were decreased when N fertilizer exceeded 180 kg N ha-1 under FP and 240 kg N ha-1 under RFFM. The evaluation results of the three comprehensive decision analysis methods indicated that RFMF planting pattern with 240 kg N ha-1 is an appropriate cultivation method for winter oilseed rape in northwestern China. These findings are of vital significance to maximize yield, optimize quality and improve resource use efficiencies of winter oilseed rape.

Keywords: appropriate cultivation method, comprehensive decision analysis method, northwestern China, winter oilseed rape

1. Introduction

Oilseed rape (Brassica napus L.) is an important oil crop worldwide, and the main uses for rapeseed are production of food, biodiesel and animal feed. Global rapeseed production has undergone sustained growth over the past 20 years,and it has become the second most common oil crop, after soybean (Glycine max) (Carré and Pouzet 2014; Gu et al.2017). In 2016, 33.7 million hectare of farmland yielded 68.9 million tonnes of rapeseed globally (FAOSTAT 2018).China, one of the most important countries for oilseed rape cultivation and production, devoted 7.6 million hectare to rapeseed cultivation in 2016, with production and yield of 15.3 million tonnes and approximately 2 000 kg ha-1,respectively (FAOSTAT 2018).

Cultivation method, including planting pattern and nitrogen(N) application, can be optimized in oilseed rape to obtain high yields, quality and resource utilization efficiencies.In China, winter oilseed rape is primarily cultivated in the Yangtze River basin, with a subtropical monsoon climate characterized by uneven rainfall. Suet al. (2014) found that straw mulching can enhance productivity in winter oilseed rape; however, some changes in topdressing N will be needed to reduce N loss if straw mulching is to become a practice in winter oilseed rape cultivation in the Yangtze River basin. In recent years, planting of winter oilseed rape has shown a northward and westward trend, and the production of winter oilseed rape in northwestern China has expanded year by year due to rising temperatures (Yinet al.2010; Zhang and Wang 2012; Guet al. 2016). However,because crops were subjected to frost and drought, the seed quality was poor, and yield and resource use efficiency were low (Jing and Dong 2004). In addition, the appropriate cultivation method for oilseed rape varies across countries and regions, and thus finding an appropriate cultivation method for winter oilseed rape in northwestern China is necessary to guarantee food and oil security.

Most of the previous studies just from the yield or/and quality aspects to evaluate whether a certain planting pattern or a certain N application amount is suitable for oilseed rape or not (Rathkeet al. 2005; Liet al. 2016; Sielinget al.2017). No study has yet evaluated a cultivation method for oilseed rape comprehensively from growth, physiology,nutrient uptake, yield, quality, and water use efficiency. The objective of the present study was to conduct an objective and comprehensive evaluation and to identify the optimal planting pattern and N application for winter oilseed rape to simultaneously achieve high yield, quality and resource use efficiency in the arid and semi-arid areas of northwestern China.

2. Materials and methods

2.1. Experimental design and data collection

This three-year field experiment was conducted at the Irrigation Station of Northwest A&F University, Yangling,Shaanxi, China. General climatic conditions and soil properties are shown in Table 1. Two planting patterns:ridge film mulching and furrow planting (RFMF; Fig. 1-A)and flat planting (FP; Fig. 1-B) were used, and each pattern had six N (used as urea) application levels: 0 (N0),60 (N60), 120 (N120), 180 (N180), 240 (N240), and 300(N300) kg N ha-1, applied as basal fertilizer. Treatments were arranged in a split-plot design with three replicates,in which the planting patterns were the main plots and the nitrogen levels were the subplots (subplots 4 m×5 m). Other fertilizers, including phosphorus (calcium super-phosphate),potassium (potassium sulphate) and borax (sodium borate)were applied as basal fertilizers in the amounts of 90 kg P2O5ha-1, 120 kg K2O ha-1and 15 kg B ha-1, respectively.Seeds of Shaanyou 107 were sown on 21 September 2014,16 September 2015 and 10 September 2016, and the crop density was determined as 120 000 plants ha-1at the fifthleaf stage. The crop was finally harvested on 23 May 2015,20 May 2016 and 23 May 2017.

Chlorophyll content and photosynthetic rate were determined at the flowering stage. For the chlorophyll content analysis, five leaf samples were randomly collected. Fresh samples were stored at -25°C until the biochemical assays were performed. Chlorophyll was first extracted with 96% ethanol, and then the chlorophyll content was determined by spectrophotometry (Gao 2006).The photosynthetic rate was measured with a portable photosynthetic apparatus Li-6400 (Li-cor, licoln, Nebraska,USA) at 10:00-12:00 a.m. on sunny days. The light intensity and flow rate were maintained at 1 400 μmol m-2s-1and 500 mL min-1. The mean value of photosynthetic rate in each treatment was calculated by the average of five leaves.At harvest, plants were harvested from a 1 m×1 m area to determine shoot dry matter and seed yield in each subplot,and the seed oil and protein contents were determined by near-infrared reflectance spectroscopy (NIR System5000, Foss, Denmark). Oil production was calculated by multiplying seed oil content by seed yield. Nitrogen,phosphorus and potassium uptake were determined using the Kjeldahl, vanadium molybdate yellow colorimetric and atomic absorption spectrometric methods, respectively,after the plant samples were digested in H2SO4-H2O2(Lu 2000). Evapotranspiration and water use efficiency were also measured with the method described in Gu et al. (2016).

Table 1 Brief description of the experimental field at the Key Laboratory of Agricultural Soil and Water Engineering in Arid and Semiarid Areas of the Ministry of Education, Northwest A&F University, Yangling, Shaanxi, China

Fig. 1 Schematic of the cultivation patterns of winter oilseed rape in this study.

2.2. Statistical analysis and evaluation methods

To determine the effects and interactions of planting patterns, nitrogen fertilization amount and growing season on the indicators of shoot dry matter, chlorophyll content,photosynthetic rate, nutrition (nitrogen, phosphorus, and potassium), seed yield, seed quality (oil and protein content),oil production, evapotranspiration, and water use efficiency,analysis of variance (ANOVA) was performed using SPSS 18.0 Software, and the significance of differences between treatments was determined with Fisher's least significant difference test. To better clarify the appropriate planting pattern and nitrogen application for winter oilseed rape, three methods: principal component analysis (PCA) method, grey correlation degree analysis (GCDA) method and DTOPSIS with entropy weight (EW) method were performed in this study.

PCA methodStep 1, standardize the original variable data; step 2, calculate the correlation coefficient matrix of the standardization matrix, and the eigenvalues and eigenvectors of the correlation coefficient matrix; step 3,define the contribution rate and cumulative contribution rate of components; step 4, calculate the scores of the principal components.

GCDA methodStep 1, determine the comparison and reference sequences. The comparison sequence (denoted as Xi) was formed by the indicators of winter oilseed rape cultivated under the specified treatment, and the reference sequence (denoted as X0) was composed by the maximum value of each indicator of winter oilseed rape under the 12 treatments; step 2, make the original variable data dimensionless; step 3, calculate the correlation coefficients first, subtract the dimensionless value of each indicator of the kthcomparison sequence (k=1, 2, 3, …, 12) by the reference sequence, and then process the value to an absolute value.

Minimum and maximum values are recorded as Δ(min)and Δ(max). The correlation coefficient is calculated as:

where ζi(k) is the correlation coefficient of the kth comparison sequence; ρ is the resolution ratio, selected as 0.5 in the present study.

Step 4, calculate the correlation degree.

The correlation degree was calculated as the mean value of each correlation coefficient.

DTOPSIS with EW methodStep 1, with m treatments,and each treatment determining n indicators, a matrix can be constructed:

Step 2, normalize matrix A to matrix B. If it is a positive indicator:

3. Results

3.1. Climatic condition

Generally, mean air temperatures during the growing seasons of winter oilseed rape followed the same trend across 2014-2015, 2015-2016 and 2016-2017 as the longterm average (Fig. 2). Over the three years of study, the annual precipitation and monthly and seasonal distribution pattern differed from each other and from the long-term average (Fig. 2).

The annual precipitation was 271, 213 and 380 mm,respectively for 2014-2015, 2015-2016 and 2016-2017,while the long-term average is 278 mm. In 2014-2015,rainfall in the early period (from September to November) was lower, and in the late period (from March to May) was much higher compared to the long-term average. In 2015-2016,total rainfall was about 65 mm lower than the long-term average, and the precipitation was similar and lower in the early and late periods in comparison with long-term average.The precipitations in the early and late periods were both higher than the long-term average, and resulted 100 mm more precipitation in 2016-2017. Thus, 2016-2017 could be considered as a wet year compared to overall conditions.

Fig. 2 The monthly total precipitation (bars) and mean air temperature (lines) during the growing seasons of winter oilseed rape in 2014-2015, 2015-2016 and 2016-2017, and the longterm (10 yr) average at the experimental site.

3.2. Analysis of variance (ANOVA)

The results of ANOVA showed that the effects of planting pattern (P), nitrogen level (N), P×N, and N×Y (year) on all variables were significant (P＜0.05) (Table 2). Similarly, the influences of Y and P×Y on all indices except seed oil content and seed protein content, and the impact of P×N×Y on all indicators except chlorophyll content, photosynthetic rate,seed oil content, and seed protein content were significant(P＜0.05).

3.3. Changes in the evaluation indicators

To visualize the variability of each indicator across different planting patterns, N amounts and growing seasons, scatter diagrams for each indicator in each treatment and growing season are presented in Fig. 3. The RFMF planting pattern significantly increased all indicators in comparison to FP when the N level and growing season were held constantly.Application of N also markedly increased all the indicators except for seed oil content when the planting pattern and growing season did not change, but when N fertilizer reached higher levels, some indicators, such as yield, oil production and water use efficiency decreased. Chlorophyll content, photosynthetic rate, nutrient (nitrogen, phosphorus and potassium) uptake, and seed oil and protein content in each treatment did not vary significantly with changing growing seasons.

3.4. PCA method

Two principal components, the eigenvalues of which were greater than 1, were selected. The cumulative contribution rates reached higher than 98% across the three seasons(Table 3). Then, the eigenvectors of the indicators of winter oilseed rape under the two principal components were calculated (Table 4). Table 5 presents the comprehensive scores of the effects of different planting patterns and N application levels on growth of winter oilseed rape. FP N0 always ranked the last, and RFMF N240, RFMF N300 and RFMF N180 consistently ranked the first,second, and third, respectively, across the three growing seasons.

r tewandT), a(Eirationsp EWU 1)m-1 ma-(kg h ＊＊ ＊＊ ＊＊ ＊＊ ＊＊ ＊＊ ＊＊tra npovality, e ETm)n(m ＊＊ ＊＊ ＊＊ ＊＊ ＊＊ ＊＊ ＊ctio 1)Oil p d(kg huaro ＊＊＊＊u ad qield, see protein t (%)ednten Seco＊ ＊＊ns＊＊ ＊＊＊＊ns＊＊ ＊＊＊ns＊p ta ke, yt uutrien o iled t (%)Senten co＊ ＊＊ns ＊＊ns ＊ns ot n sason icators, shose 1)Yieldag h(k＊＊ ＊＊ ＊＊ ＊＊ ＊＊ ＊＊ ＊＊al ind in ggrow17 icsiologhy 206-01d 2 p ta keK uptakeP u 1)a-(kg h1)a-(kg h＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊M), p an r (A Dattery m 1 6205-015, 201 d d-2 u nroovegabA) for the 1 420cross thee ad rap OVee nce (ANils p ta keN uetic ratethotosyn 1)a-(kg hl m-2 s-1 g-1)mo Ph(μ 1)yllgphroloCh t (m gntenco＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊＊ns＊＊ns.ceerenifft d r otein f variaf w alysis o E) o 1)Ma-AD(kg h An ncy (WU ble 2fficie Ta e eus tio nf varias ourceSo ttern (P)＊＊＊＊)(Nte para tingen Plan Nitrog ar (Y)＊＊＊＊＊＊＊＊＊＊Ye Y NYYN×P×P×N×P×n ifica no sigs, nctively; n, respe0.01P＜nd5 a.0＜0t Ps aceerenifft dnificanig＊＊, snd＊ a

3.5. GCDA method

The indicators of winter oilseed rape in 2014-2015 were taken as a sample to calculate the correlation degree of the different treatments. First, the comparison and reference sequences of different planting patterns and N application levels on all indicators were determined (Table 6). Second, the dimensionless values of different treatments on the indicators were calculated (Table 7). Correlation coefficients of the reference and evaluation indices were then calculated (Table 8). Lastly, the correlation degrees of different planting patterns and N amounts on the indicators of winter oilseed rape were calculated (Table 9). In this method, RFMF N240 and RFMF N300 consistently performed the best across the 2014-2015, 2015-2016 and 2016-2017 growing seasons.

3.6. DTOPSIS with EW method

The adjacent degrees to the ideal solution of different planting patterns and N application were determined and are shown in Table 10. Like the PCA results, RFMF N240, RFMF N300 and RFMF N180 were consistently the top three treatments, and FP N0 ranked the lowest among all treatments across the three growing seasons.

4. Discussion

4.1. Planting pattern

Two planting patterns, RFMF and FP, were used in this study. Compared to FP,the RFMF pattern significantly improved the growth, physiology, nutrient uptake,yield, quality, and water use efficiency of winter oilseed rape. This might be due to that RFMF pattern could collect more rainwater and conserve more soil water for rapeseed crops (Gu et al. 2016). In our study, soil moisture under RFMF pattern was consistently higher than that under the FP pattern across the three seasons (data not shown). In addition, soil temperature during the early growth stage, especially during winter, was 0.6-3.0°C higher under RFMF than under FP. This would also promote root development in the early growth stage, reserve more sufficient root mass in winter, and thus accelerate growth of rapeseed crops more quickly in spring (Rathke et al. 2006; Gu et al. 2016). Crops, such as winter wheat (Triticum aestivum) (Yin et al. 2016), maize (Zea mays) (Mo et al. 2017; Liu et al. 2018), and potato (Solanum tuberosum) (Qin et al. 2013; Liang et al. 2018) also achieved larger canopy, higher yield, better quality, and higher resources use efficiency when grown under the RFMF pattern.

4.2. N application

Fig. 3 Different treatments of shoot dry matter, chlorophyll content, photosynthetic rate, nitrogen uptake, phosphorus uptake,potassium uptake, yield, seed oil content, seed protein content, oil production, evapotranspiration, and water use efficiency of winter oilseed rape in 2014-2015, 2015-2016 and 2016-2017 growing seasons. RFMF, ridge film mulching and furrow planting;FP, flat planting. N0-N300, 0, 60, 120, 180, 240, and 300 kg N ha-1, respectively. Bars mean SE (n=5).

Table 3 Eigenvalues and cumulative contribution proportions of principle components of the indices of winter oilseed rape

Table 4 Eigenvectors of the indices of winter oilseed rape under the two principal components (E1 and E2) in three growing seasons

N is an essential nutrient for crop growth and is one of the main components of many important organic compounds in plants. It is also one of the scarcest nutrients in the soil.Thus, applying N fertilizer to the soil is typically required for a crop to achieve high yield. The application method and amount can vary with regions, crops and planting patterns. Oilseed rape is a heavy N consumer, requiring about 6 kg N to produce 0.1 t rapeseeds (Rathkeet al. 2006; Guet al.2018). In the present study, applying N to the field accelerated the growth of winter oilseed rape, increased the shoot dry matter accumulation and improved the final yield, oil production and water use efficiency. Previous studies of the relationship between N and rapeseed growth or harvest, including Rathkeet al. (2006), Suet al. (2014), Sielinget al. (2017), and Guet al.(2017) obtained similar results. However, shoot dry matter,nutrient uptake, yield, oil production, and water use efficiency of winter oilseed rape decreased when N application was excessive (higher than 180 kg N ha-1under FP pattern and 240 kg N ha-1under RFMF pattern). Previous studies in wheat (Qiet al. 2009), maize (Moet al. 2017), potato (Renset al. 2018),and castor bean (Ricinus communis) (Xueet al. 2017) found similar trends. In this study, we also observed that increasing N application decreased seed oil content, but increased seed protein content (Fig. 3). Rathkeet al. (2006) and Zhaoet al.(2007) reported similar findings. Seed yield was increased by applying N fertilizer at the expense of decreased seed oil content; however, oil production was also the highest at N240(1 190 kg ha-1) under RFMF and at N180 (914 kg ha-1) under FP pattern. These two treatments led to a much higher seed yield compared to other treatments for each planting pattern.

Table 5 Comprehensive scores of the effects of different planting patterns and nitrogen application on winter oilseed rape in three growing seasons

se r u 1)cym-1 m effiatecien 0.54(kg h9.5±Wa 5.0±0.12 0.31 0.15 5.6±3.7±0.16 0.11 7.3±4.2±0.21 0.27 9.0±5.6±0.39 0.54 9.5±7.0±0.26 0.21 8.2±6.7±0.20,80 6.1±20, 1 15 20 4-01 e in 2 d rap ee ils r o te in tion pira ns m)otra(m Evap n 22 0.0±38 15 0.3±29 19 9.1±31 16 0.7±31 23 4.8±34 31 3.5±32 30 3.0±35 28 2.7±33 15 9.3±35 45 8.4±33 33 0.8±37 25 8.5±34 22 0.0±38 ctio 1)rog h dua-Oil p(k Protein t (%)33 8±1 27 32 9±64 41 0±52 33 6±73 52 5±60 74 0±97 52 9±79 64 8±1 21 88 1±1 00 33 8±1 27 45 6±95 72 4±1 09 45 2±87 nten nt f w co 9741975463490.2.2.3.4.3.2.5.4.3.5.5.2.2±0±0±0±0±0±0±0±0±0±0±0±0±0.4.8.1.2.3.6.9.3.5.9.2.4.8 25 19 19 21 20 22 21 24 23 24 24 25 24 0, 1, 6, 0 00 N3 0-g. N ntin la t p P, fla; F ting plan rrow fu nd rs o to ica n ind n o atio plic ap en nitrog te on Oil c)(%1)Yieldam g h(k ssiuke ta Po nd s a rn atte orus 1)taaup (kg h 6652432877052 g p ntin la t p eren ifff d s o ce en qu se ce d referen.9.9.9.8.4.0.9.7.0.8.1.0.4±0±0±0±0±1±1±0±0±1±0±1±1.1.1.6.3.5.0.1.8.0.9.3.5.5 45 87 2±3 20 45 96 2±1 44 44 95 0±1 18 42 82 2±1 74 41 41 43 14 8±1 45 10 4±2 36 40 92 4±1 99 40 78 7±2 98 5.4.3.4.4.3.7.3.7 9.5±±9 1350±6±7±8±5.6.2.2.4.7.8 36 70 52 89 40 7 10 3±2 50 39 87 2±3 20 731.6±±49.8±811.5 11948.7±.6 13 38 38±1in 37 052 10 8±2 49 6.0 0.3±10 10 3±2 84 5.4 14 1±2.72 32 9.5±±1.1 1399 g a osph Ph ke ta 1).8.3.8.5.4.7.8.2.9.8.6.3.1 a-±2±2±2±3±2±3±2±3±3±2±3±2±4 upg h.4.1.3.5.8.9.2.1.1.4.4.8.6(k47 20 15 26 21 33 28 41 en Nitrog ke ta 1)aupg h(k6.3±±8 35 8.2.4.8.0.8.4.2.2.1±8±6±3±8±7±6 47 37±69.40.4 46 36 etic th otosyn te Ph ra l m-2 s-1)1139(μ 1.13 9±.5.7.9.9.4.5.3.8 0.91 1±28 0.79 5±54 0.87 5±44 0.97 6±73 0.92 0±62 1.36 8±99 0.79 0±805.6±±18.2.4 0.94 8±ulch film m ge F, rid FM ts; R en tm trea 12 all the 11826.3±±6r the 1.13 9±.6 1.50 0±.2de 1180un 0.87 6±1.43 6±m(xù)o.9.7.4.4.2.9.7.4.1.9.9.2.1 27 15 14 19 17 22 19 25 22 27 21 27 yll ph ro lo Ch ry t nten co pe ra ed ilse 21r o g g-1)1315006501706(m.1.1.1.1.1.1.1.1.1.1.1.2.1±0 2.83±0 1.18±0 0.93±0 1.73±0 1.39±0 2.04±0 1.63±0 2.32±0 2.04±0 2.83±0 2.02±0 2.72±0 1.97 dd an arison te un in f w ator o rotter mp 1)1636886211302 ic h ind ely.ctiv ac oveg a-Ab mag h 47 9±(k06 36 9±116 78 35 5±5 67 32 5±7 56 44 2±6 62 35 2±9 33 25 2±7 60 33 7±35 30 7±108 14 47 9±06 50 9±118 63 35 4±96 41of e 5±108 42 pe Co ble 6 Ta t1)en tm Trea X0 N0 MF RF N0 FP 0 N6 MF RF 20 0N1 N6 FP MF RF 80 20N1 N1 FP MF RF 40 80N2 N1 FP MF RF 00 40N3 N2 FP MF RF 00 N3 FP alue m v ha-1, res ximu 5).0 kg N D (n=±S ma30 an, thend 1) X0 me 0, a re s a 24 lu Va

4.3. Determination of the optimal cultivation method

Finding an optimal cultivation method for winter oilseed rape is very important to achieve high yield and benefit the economy of certain regions. Three methods, PCA, GCDA and DTOPSIS with EW, were used to evaluate the growth,physiology, nutrient uptake, yield, quality, and water use efficiency of winter oilseed rape cultivated under different planting patterns and N application amounts in the present study, and finally to determine an appropriate cultivation method for winter oilseed rape in northwestern China.PCA, GCDA and DTOPSIS with EW all showed that RFMF N240 consistently obtained the highest score across the 2014-2015, 2015-2016 and 2016-2017 seasons, indicating that RFMF N240 is the optimal cultivation method for winter oilseed rape in northwestern China.1)RFMF, ridge flm mulching and furrow planting; FP, flat planting. N0-N300, 0, 60, 120, 180, 240, and 300 kg N ha-1, respectively.

1 5204-01e in 2d rapeeilsr oteinf wrs otodicainonationplicapennitrognds arnatteg pntinlat perenifff ds oaluenless vnsioDimeble 7Ta sete cien r uWa 1)cymeffi1 mg h 10.52 a-(k 69928979 0.38 0.58 0.44 0.76 0.58 0.94 0.73 10.70 532 0.86 0.64 tio npirans m)otra(m E va pn 10.76 40871966167 0.84 0.81 0.90 0.85 0.92 0.87 0.94 0.89 0.97 0.91 1 ctio 1)Oil p d(kg huaro 10.50 87639533 0.40 0.57 0.47 0.75 0.62 0.95 0.78 10.74 862 0.85 0.68 P ro te inntteOil con t (%)ntenco)(%10.77 6092107533 0.75 110.97 0.82 0.80 0.89 0.86 0.95 0.92 0.98 0.956 10.97 99210697053 0.93 0.92 0.91 0.89 0.90 0.88 0.88 0.85 0.85 0.83 1)Yielda-(kg h 10.45 09458332 0.36 0.54 0.45 0.73 0.62 0.93 0.78 10.78 085 0.88 0.72 mssiu ke taPo 1)taaup (kg h 10.36 3936390849 0.25 0.50 0.37 0.64 0.52 0.80 0.67 0.99 0.71 10.71 o ru sosphPh ketaup 1)a-(kg h 10.42 43905571 0.32 0.55 0.46 0.71 0.59 0.86 0.74 10.78,80, 1200, 1, 6, 000N30-g. Nntinlat pP, fla; Ftingplanrrowfundg ainulchfilm mgeF, ridFMts; Ren 0tm 972 0.98 0.77 en(k Nitrog gta h a-1)keup 10.34 0726174540 0.24 0.47 0.38 0.63 0.53 0.85 0.69 0.99 0.710 10.69 e ticthotosyn te P hyllphroloCh ratntenco mo10.56 l m-2 s-1)(μg g-1)(m 16578772 10.41 0.51 0.69 0.61 0.81 0.70 0.90 0.79 79111601 0.32 0.61 0.49 0.72 0.57 0.82 0.72 10.78 246 10.71 0.97 0.75 416 0.96 0.69 unr drooveg atte1)a-Abdry mg h(k10.61 33383766 0.51 0.68 0.59 0.84 0.68 0.93 0.73 10.78 011 0.99 0.76 t1)entmTrea X0 N 0MFRF N 0FP 0N6MFRF 20 0N1 N 6FP M FRF 80 20N1 N 1FP M FRF 40 80N2 N 1FP M FRF 00 40N3 N 2FP M FRF 0 0N3FP tre a12all ther thedeunperaedilser oteinator o e ly.f wic ctivh ind p eacof e h a-1, re salue) kg Nm v 0 0ximu0 (N3ma 3 0, the nd 1) X00, a 24 1 5204-01e in 2d rapilseeinter of wices on indatioluvad eancef referennts officieoen crrelatioCoble 8Ta s ete cien r uWa 1)cymeffitionnspira(kg h 1 0.41m12193559 a-m)otra(m E va pn 0.38 0.39 0.35 0.44 0.36 0.65 0.37 10.33 363 0.34 31114857196 0.58 0.70 0.59 0.77 0.55 0.76 0.44 0.74 0.34 0.85 0.46 0.3 31 ctio 1)Oil p d(kg huaroProtein t (%)28424616 0.40 0.38 0.38 0.37 0.43 0.38 0.68 0.42 10.36 950 0.33 0.36 n te nco 5171977360 0.59 0.60 0.60 0.61 0.62 0.62 0.69 0.68 0.76 0.762 10.88 n tteonOil c)(%10.94 72052513637 0.81 0.80 0.67 0.68 0.51 0.59 0.33 0.49 0.33 0.51 1)Yielda-(kg h 54743484 0.37 0.37 0.36 0.36 0.41 0.38 0.59 0.42 10.40 124 0.39 0.39 mssiu ke taPo 1)upg htaa-(k0.34 1773133384 0.33 0.34 0.33 0.34 0.33 0.33 0.33 0.90 0.34 10.38 o ru sosphPh ketaup 1)a-(kg h 47463893 0.36 0.35 0.37 0.36 0.39 0.36 0.42 0.38 10.41 e ly.ctivpeha-1, resg N0 k30nd, a40 2, 2 110 0.85 0.44 en(k N itro geticth gta ha-1)3337376547 keup otosyntera l m-2 s-1)0.33 0.33 0.33 0.33 0.33 0.33 0.40 0.34 0.90 0.336 10.36 Ph mo0.42(μ 98494597 0.43 0.46 0.44 0.50 0.44 0.51 0.43 10.40 453 0.73 0.42 yllphroloCh tntenco g g-1)(m 0.36 29408776 0.35 0.40 0.38 0.39 0.35 0.35 0.36 10.34 011 0.65 0.37 dunroovegAb rattedry m 1)a-(kg h 16570999 0.46 0.43 0.45 0.43 0.54 0.42 0.60 0.37 10.40 248 0.88 0.42 t1)entmTrea N 0MFRF N 0FP 0N6MFRF 20 0N1 N 6FP M FRF 80 20N1 N 1FP M FRF 40 80N2 N 1FP M FRF 00 40N3 N 2FP M FRF 0 0N3FP 8 0, 1200, 1, 6, 000N30-g. Nntinlat pP, fla; Ftingplanrrowfundg ainulchfilm mge, ridMF1) RF

Table 9 Correlation degrees of different planting patterns and nitrogen application on the indicators of winter oilseed rape in three growing seasons

Table 10 Adjacent degrees to ideal solution of different planting patterns and nitrogen application in winter oilseed rape in three growing seasons

5. Conclusion

Under the given conditions, planting pattern and N application both significantly affected the growth, physiology,nutrient uptake, yield, quality, evapotranspiration, and water use efficiency of winter oilseed rape. The present study used the three methods of PCA, GCDA and DTOPSIS with EW to comprehensively determine the optimal planting and fertilization strategy. The results of the three evaluation methods all indicated that RFMF N240 was the optimal cultivation method in this study of winter oilseed rape in northwestern China. This finding is of vital significance and will help to maximize yield, optimize quality and improve resource use efficiency of winter oilseed rape cultivated in arid and semi-arid regions.

Acknowledgements

This research was supported by the Special Fund for Agroscientific Research in the Public Interest, China (201503125 and 201503105) and the Fundamental Research Funds for the Central Universities, China (2452018089).