Effects of a controlled-release fertilizer on yield, nutrient uptake, and fertilizer usage efficiency in early ripening rapeseed (<em>Brassica napus</em> L.)^{<a href="#FN01" rid="FN01" class=" fn">*</a>}

1. Introduction

Nitrogen (N), phosphorous (P), and potassium (K) are critical nutrient elements for crop plant growth and development, and applying fertilizers, especially N to crops, is a valuable agronomic practice (Zhao et al., 2013). However, excessive fertilization can result in nutrient inefficiencies and excessive losses of N and P in the field environment, and also impact soil, water and air quality, human health, and biodiversity (Goulding et al., 2008). Nitrate nitrogen (NO₃^{-N) contaminations of both the soil and water are caused by over usage of fertilizers especially N, and large NO}₃^{-N accumulation can reduce the N usage efficiency (NUE) of crops (Zhu and Chen,} 2002; Ju et al., 2004). Also, excessive usage of different fertilizers, especially N, can increase lodging, causing a reduction in the yield and quality of the crops (Ozer, 2003). Therefore, interventions to increase fertilizer usage efficiency (FUE) and reduce nutrient inputs, especially N, are of significant importance for reducing environmental loading and lowering the costs of agricultural production (Wang et al., 2011).

Controlled-release fertilizer (CRF) is a good alternative to soluble fertilizer (SF) to increase FUE and minimize nutrient losses, especially N, in the field environment (Zhao et al., 2013). The coating material types of CRF play a key role in gradually releasing the nutrient (Li et al., 2012), and the most important parameters for controlling nutrient release include the thickness of the coating membrane, followed by temperature, granule radius, soil microbial activity, etc. (Du et al., 2008; Kaplan et al., 2013), and they are controlled to match the nutrient requirement of the plants (Dai et al., 2008; Yan et al., 2008; Ni et al., 2009). Due to a lack of experience with the field performance of CRF and its high relative cost, current grower acceptance is limited (Zhao et al., 2013). However, the greatest benefits of changing from SF to CRF include not only increased profitability, but also reductions in the environmental pollution of crop production. In some countries, it has primarily been applied to nursery stock (Azeem et al., 2014), and there have been few investigations in the field on the performance of crops grown with CRF (Noellsch et al., 2009).

Rapeseed (Brassica napus L.) is an important agricultural crop cultivated for its oil, which can be used as an edible product or for various industrial applications (Malagoli et al., 2005a; 2005b). As one of the major economic crops in China, rapeseed provides more than 40% of the plant oil produced, with about half of the available cropping area being used for rapeseed crops (Wang et al., 2013; Hussain et al., 2014). Along with the acceleration of rural urbanization, farmer employment, and non-agricultural production, there has been a large decline in both the quantity and quality of labor engaged in agricultural production in China, to some extent restricting the development of rapeseed production (Guan, 2006). Under the current environment, the establishment of the CRF method is more important in improving rapeseed production and increasing the edible oil supply in China. Despite its high capacity to remove nitrate from the soil, rapeseed is characterized by a very low N recovery in the reproductive tissues under field conditions (Malagoli et al., 2005a). There currently are no published research data on the fertilizer rate of CRF response to early ripening rapeseed in southern China.

The objectives of this study are: (1) to estimate N, P, and K responses for rapeseed performances at maturity, including seed yield, yield components, and nutrient accumulation; (2) to investigate the effects of CRF on the developmental dynamics of nutrient uptake during the growing season, in terms of dry matter production, nutrient translocation, and N, P, and K usage efficiencies; and (3) to choose a reasonable application rate of CRF for rapeseed in the region. Our results could help increase the understanding of nutrient development dynamics and seed yield responses for CRF, and provide some suggestions to improve the nutrient management of rapeseed by light and simple fertilization.

2. Materials and methods

2.1. Experimental design

In connection with early ripening rapeseed testing, Xiangzayou 1613 (provided by the Hunan Branch of National Center for Oil Modified) planting was done on a farm in Huilongpu Town, Ningxiang Country, Hunan Province in China (23°21′ N, 112°46′ E). The soil, classified as a red-yellow soil and evolved from the red earth platform of the quaternary period, was considered to be highly suitable for crop production. The soil pH was 6.1. The average organic matter content in the tillage layer was 48.89 g/kg, the available N, P, K, and boron (B) were 169.16, 15.43, 59.08, and 0.31 mg/kg, respectively and the total N, P, and K were 2.73, 0.54, and 16.38 g/kg, respectively.

The methods of soil analysis were referenced from Soil and Agricultural Chemistry Analysis written by Bao (2005). CRF, a coated compound fertilizer, was offered by the Hunan Xingxiang Biological Science and Technology Co., Ltd., Xiangtan City, Hunan Province, China and was used in the experiment, with SF being used as a control. The contents of N, phosphorus pentoxide (P₂O₅), and potassium oxide (K₂O) in each SF and CRF were 12%, 6%, and 7%, respectively, with the boric fertilizer being 15 kg/hm ^{(B, 10.8%).}

Field experiments were conducted during 2011–2013 as a random complete block design with three replications. The area of each plot was 20 m^{. There were eleven treatments (Table} ​(Table1),1), as SF1/CRF1 (3750 kg/hm^{), SF2/CRF2 (3000 kg/hm), SF3/CRF3 (2250 kg/hm), SF4/CRF4 (1500 kg/hm), SF5/CRF5 (750 kg/hm), and CK (no fertilizer). All fertilizers were applied at a basal dosage. Direct seeding of rapeseed was performed in October, 2011/2012. After the emergence of seedlings, the density was thinned to 37.5×10 plants/hm by 2 times, and the harvest time was in May, 2012/2013. Herbicides were not applied to the field experiment and weeds were controlled with hand weeding. Field management, in general, was the same as with the rapeseed field.}

2.2. Plant sampling and N, P and K content determination

To measure the N, P, and K uptakes of the aboveground, five plants were collected for treatment in the wintering period, flowering stage, and harvest time, respectively. At maturity, five plants were manually harvested for treatment. The branch number per plant, pod number per plant, and seed number per pod were counted. Drying and weighing the harvest were measured to the community order production. The aboveground dry matter was determined by oven-drying the samples at 80 °C until a constant weight was achieved. Then the dry matter weight and 1000-seed weight were determined. Subsequently, samples were manually separated into the vegetative and seed portions. The seeds and straw were ground using a cyclone sample mill with a mesh size of 0.5 mm. Then the seed and straw N, P, and K concentrations were measured using the Kjeldahl method, vanadium-molybdenum-yellow photometric method and flame spectrophotometer method, respectively. Seed oil content was determined by using a Soxhlet apparatus. The following parameters were calculated (Dordas, 2009): (1) N, P, and K accumulation (kg/hm^{)=N, P, and K content (%)×dry matter accumulation (kg/hm); (2) N, P, and K translocation amount (kg/hm)=N, P, K accumulation of straw at the flowering stage (kg/hm)−N, P, and K accumulation of straw at maturity (kg/hm); (3) N, P, and K translocation rate (%)=N, P, and K translocation amount (kg/hm)/N, P, and K accumulation at the flowering stage (kg/hm)×100%; (4) N, P, and K contribution rate (%)=N, P, and K translocation amount (kg/hm)/N, P, and K accumulation of seed at maturity (kg/hm)×100%; (5) N, P, and K usage efficiency (%)=(N, P, and K accumulation (kg/hm)−N, P, and K accumulation at the CK area (kg/hm))/N, P, and K fertilizer rate (kg/hm)×100%; (6) N, P, and K harvest index (%)=N, P, and K accumulation of seed at maturity (kg/hm)/N, P, and K accumulation of aboveground at maturity (kg/hm)×100%; and (7) seed profit (CNY/hm)=seed yield (kg/hm)×quarter price (CNY/kg)−fertilizer rate (kg/hm)×quarter price (CNY/kg), with the quarter price of seed yield being 4.50 CNY/kg, CRF 1.45 CNY/kg, and SF 1.40 CNY/kg.}

2.3. Statistical analysis

Statistical analyses were performed using the analysis of the variance (ANOVA) in the general linear model procedure of SPSS (Ver. 11, SPSS, Chicago, IL, USA). Results are presented as the mean of the two seasons of the experimentation, because the trends of these parameters were consistent between seasons. The least significant differences (LSDs) between the means were estimated at the 95% confidence level. Unless indicated otherwise, significant differences among different plants are given at P<0.05; the LSD was used to compare adjacent means arranged in order of magnitude.

3. Results

3.1. Seed yield and seed profit

As shown in Table ​Table2,2, seed yield was significantly affected by fertilization treatments (P<0.05), and CRF had higher seed yield than SF by an average of 17.33%, with a lower fertilizer rate per kg yield by an average of 14.20%. CRF4 achieved maximum yield (2066.97 kg/hm^{), and the lowest was observed in SF1 (664.63 kg/hm). SF2, SF3, SF4, and SF5 were higher in seed yield than CK by 63.5%, 160.2%, 159.5%, and 113.6%, respectively. In addition, CRF1, CRF2, CRF3, CRF4, and CRF5 were higher in seed yield than CK by 25.9%, 89.9%, 172.2%, 191.6%, and 154.5%, respectively. CRF4 and CRF5 greatly reduced the N input, while further increasing more yield production. Thus, CRF provides a strategy for environmentally sustainable increases in seed yield.}

Table 2

Seed yield and fertilizer rate per kg yield of early ripening rapeseed using different fertilizer treatments

Treatment	Seed yield (kg/hm)			Fertilizer rate per kg yield (kg/kg)
	SF	CRF	Increase with SF (%)	SF	CRF	Decrease by SF (%)
CK	708.97fg
SF1/CRF1	664.63g	892.67efg	34.31	5.642	4.201	25.55
SF2/CRF2	1159.33def	1346.47cde	16.14	2.588	2.228	13.90
SF3/CRF3	1844.50ab	1929.97ab	4.63	1.220	1.166	4.43
SF4/CRF4	1839.40ab	2066.97a	12.37	0.815	0.726	11.01
SF5/CRF5	1513.93bcd	1804.57abc	19.20	0.495	0.416	16.11

CK: no fertilizer; SF1/CRF1: 3750 kg/hm^{; SF2/CRF2: 3000 kg/hm; SF3/CRF3: 2250 kg/hm; SF4/CRF4: 1500 kg/hm; SF5/CRF5: 750 kg/hm. The means denoted by different letters indicate significant differences among fertilization treatments (}P<0.05)

Seed yield output minus the fertilizer input represents the net profit for each treatment. The average net profits of SF1 to SF5 were −2259.2, 1017.0, 5150.3, 6177.3, and 5762.7 CNY/hm^{, respectively, and the average net profits of CRF1 to CRF5 were −1420.5, 1709.1, 5422.4, 7126.4, and 7033.1 CNY/hm, respectively. Due to excessive fertilizer use, SF1 and CRF1 had negative profits and higher profits were achieved in SF4, SF5, CRF4, and CRF5. Compared with SF4 and SF5, the profits of CRF4 and CRF5 increased by 949.1 and 1270.4 CNY/hm, respectively. CRF4 as the optimum treatment achieved the best yield (2066.97 kg/hm) with a fertilizer rate of 1500 kg/hm, for a net profit of 7126.4 CNY/hm.}

Taking the fertilizer rates (X₁ for SF, X₂ for CRF) as the independent variables and seed yields (Y₁ for SF, Y₂ for CRF) as the dependent variables, the correlation analysis of the two-year experiment was performed to establish regression equations of fertilizer rates on seed yields as shown below (Fig. ​(Fig.1):1):

;

.

Fig. 1

Seed yield as a function applied of early ripening rapeseed

The high R ^{in both equations indicates a close relationship between seed yield and fertilizer rates. According to the equations, the best fertilizer rates of SF and CRF were 1802.12 and 1843.04 kg/hm, respectively, achieving the maximum yield (1839.24 and 2038.35 kg/hm, respectively) and net profit (5753.62 and 6500.16 CNY/hm, respectively). CRF increased the maximum seed yield and net profit by 10.83% and 12.98%, respectively, compared with SF. Therefore, the targeted use of CRF in red-yellow soil can increase seed yields and profitability.}

3.4. N, P, and K uptakes

There appeared to be an “S” shaped curve in the total dry matter accumulation (data not shown). The trends of nutrient accumulation and dry matter accumulation were basically consistent, which was suggested by the N and P accumulations during the different growth periods: flowering stage>harvest time>wintering period (Fig. ​(Fig.2),2), while a relatively small increase was noted from the flowering stage to the harvest time, which appeared in the K accumulation. The N accumulation was significantly affected by the fertilizer rate (P<0.05), while the P and K accumulation responses to the fertilizer rate were similar. Namely, increasing the fertilizer rates resulted in significantly higher N, P, and K accumulations in the rapeseed. The N, P, and K accumulations of CRF4 and SF3 in each group had outstanding performance during different growth periods.

Fig. 2

N, P, and K accumulation of early ripening rapeseed using different fertilizer treatments

SF1/CRF1: 3750 kg/hm^{; SF2/CRF2: 3000 kg/hm; SF3/CRF3: 2250 kg/hm; SF4/CRF4: 1500 kg/hm; SF5/CRF5: 750 kg/hm; CK: no fertilizer. WP: wintering period; FS: flowing stage; HT: harvest time}

According to Fig. ​Fig.2,2, following the field application of SF, the N accumulation rapidly increased during the first phase (i.e., the stage prior to flowering) and then slowly increased after the flowering stage. A one-off application of SF can lead to a large loss during the plant growth, which causes insufficient nutrient supply in the late stage (Liu et al., 2011). However, the N accumulation increased relatively constantly with the increased application of CRF throughout the growth period, and caused obvious delays in leaf senescence. At the flowering stage, the ranks of N accumulation among all treatments were CRF3>CRF4>CRF5>CRF2>CRF1 and SF3>SF4>SF2>SF1>SF5. On the contrary, SF suffered from large N deficiencies at the flowering stage till harvest time.

At harvest time, nutrient accumulation first increased and then decreased with an increase in the fertilizer rate. The N, P, and K accumulations in CRF at harvest time were higher than those in SF by an average of 13.70%, 14.01%, and 10.33%, respectively. The SF suffered from nutrient deficiency at the flowering stage and at harvest time. CRF4 achieved maximum N accumulation (177.76 kg/hm^{) and the lowest was observed in SF1 (105.96 kg/hm). SF1, SF2, SF3, SF4, and SF5 were higher in N accumulation than CK by 109.78%, 192.16%, 193.33%, 172.92%, and 118.14%, respectively. In addition, CRF1, CRF2, CRF3, CRF4, and CRF5 were higher in N accumulation than CK by 150.47%, 187.86%, 229.50%, 251.93%, and 142.33%, respectively. There was a significant relationship between the N accumulation at maturity and seed yield (SF:} y=9.9331x+129.57, R^=0.4868, P<0.05; CRF: y=10.122x+127.25, R^=0.6515, P<0.05) as shown in Fig. ​Fig.3.3. It indicated that increasing the N accumulation appeared to be a necessary procedure to maximize seed yield in rapeseed.

Fig. 3

N accumulation as a function applied of early ripening rapeseed

3.1. Seed yield and seed profit

As shown in Table ​Table2,2, seed yield was significantly affected by fertilization treatments (P<0.05), and CRF had higher seed yield than SF by an average of 17.33%, with a lower fertilizer rate per kg yield by an average of 14.20%. CRF4 achieved maximum yield (2066.97 kg/hm^{), and the lowest was observed in SF1 (664.63 kg/hm). SF2, SF3, SF4, and SF5 were higher in seed yield than CK by 63.5%, 160.2%, 159.5%, and 113.6%, respectively. In addition, CRF1, CRF2, CRF3, CRF4, and CRF5 were higher in seed yield than CK by 25.9%, 89.9%, 172.2%, 191.6%, and 154.5%, respectively. CRF4 and CRF5 greatly reduced the N input, while further increasing more yield production. Thus, CRF provides a strategy for environmentally sustainable increases in seed yield.}

Seed yield output minus the fertilizer input represents the net profit for each treatment. The average net profits of SF1 to SF5 were −2259.2, 1017.0, 5150.3, 6177.3, and 5762.7 CNY/hm^{, respectively, and the average net profits of CRF1 to CRF5 were −1420.5, 1709.1, 5422.4, 7126.4, and 7033.1 CNY/hm, respectively. Due to excessive fertilizer use, SF1 and CRF1 had negative profits and higher profits were achieved in SF4, SF5, CRF4, and CRF5. Compared with SF4 and SF5, the profits of CRF4 and CRF5 increased by 949.1 and 1270.4 CNY/hm, respectively. CRF4 as the optimum treatment achieved the best yield (2066.97 kg/hm) with a fertilizer rate of 1500 kg/hm, for a net profit of 7126.4 CNY/hm.}

Taking the fertilizer rates (X₁ for SF, X₂ for CRF) as the independent variables and seed yields (Y₁ for SF, Y₂ for CRF) as the dependent variables, the correlation analysis of the two-year experiment was performed to establish regression equations of fertilizer rates on seed yields as shown below (Fig. ​(Fig.1):1):

;

.

Fig. 1

Seed yield as a function applied of early ripening rapeseed

The high R ^{in both equations indicates a close relationship between seed yield and fertilizer rates. According to the equations, the best fertilizer rates of SF and CRF were 1802.12 and 1843.04 kg/hm, respectively, achieving the maximum yield (1839.24 and 2038.35 kg/hm, respectively) and net profit (5753.62 and 6500.16 CNY/hm, respectively). CRF increased the maximum seed yield and net profit by 10.83% and 12.98%, respectively, compared with SF. Therefore, the targeted use of CRF in red-yellow soil can increase seed yields and profitability.}

3.2. Oil yield

Oil contents of rapeseed were not significantly affected by the various fertilizer applications (P>0.05; Table ​Table3).3). The highest oil content appeared in the CK treatment (42.88%). The oil contents of the SF and CRF treatments were relatively the same (37.97%–41.62% and 38.77%–41.82%, respectively). When the fertilizer rate increased, the oil content of the rapeseed gradually declined, namely higher fertilizer rates usually reduced the oil content. On the other hand, oil yield significantly varied with different fertilization treatments (P<0.05). The trends of oil yields and seed yields were consistent, namely oil yield of low fertilizer rate treatments (CRF3, CRF4, and CRF5) were significantly higher than those of high fertilizer rate treatments (CRF1 and CRF2) (P<0.05). CRF4 achieved maximum oil yield (848.08 kg/hm^{), followed by CRF3 (776.43 kg/hm) and the lowest oil yield was observed in SF1 (252.36 kg/hm). In addition, oil yields of CRF were significantly higher compared to SF by an average of 17.29%.}

3.3. Yield components

As shown in Table ​Table4,4, there were statistical differences between SF and CRF in rapeseed at the mature stage for the plant height, first branch number, branch height, pod number, stem dry weight, and pod dry weight, but no significant differences for seed number per pod or 1000-seed weight (P>0.05). With the increase of the fertilizer rate, plant height, first branch number, branch height, pod number, stem dry weight, and pod dry weight were all shown as a trend of the first increase and then decrease. In combination with the SF, CRF applications reflected significant improvements of the rapeseed in first branch numbers (4.63%), pod numbers (12.32%), stem dry weight (4.76%), and pod dry weight (4.96%), especially for the CRF4 treatment. Lodging is a common problem with rapeseed, and occurs primarily with taller varieties and under conditions of a higher soil N content (Ozer, 2003). Branch heights of SF were increased by an average of 2.43% compared to CRF caused by rapid nutrient release during the early growth stage, which made it easier for lodging to occur in the later stage.

Number of pods per plant is commonly a major determinant of rapeseed yield and this characteristic is dependent on the number of flowers produced by each plant. In this study, pod numbers showed a significant increase with seed yield (SF: y=10.337x−243.49, R^=0.8098, P<0.05; CRF: y=10.736x−310.51, R^=0.7281, P<0.05), as shown in Table ​Table4.4. CRF4 achieved maximum pod numbers (211.27 per plant), and the lowest was observed in CK (83.00 per plant). SF1, SF2, SF3, SF4, and SF5 exhibited higher pod numbers than CK by 53.82%, 60.16%, 134.14%, 150.72%, and 72.45%, respectively. In addition, CRF1, CRF2, CRF3, CRF4, and CRF5 were higher in pod numbers than CK by 101.29%, 84.34%, 136.23%, 154.54%, and 114.57%, respectively. It was shown that increasing the fertilizer rates usually caused larger increases in the pod numbers of rapeseed plants.

3.4. N, P, and K uptakes

There appeared to be an “S” shaped curve in the total dry matter accumulation (data not shown). The trends of nutrient accumulation and dry matter accumulation were basically consistent, which was suggested by the N and P accumulations during the different growth periods: flowering stage>harvest time>wintering period (Fig. ​(Fig.2),2), while a relatively small increase was noted from the flowering stage to the harvest time, which appeared in the K accumulation. The N accumulation was significantly affected by the fertilizer rate (P<0.05), while the P and K accumulation responses to the fertilizer rate were similar. Namely, increasing the fertilizer rates resulted in significantly higher N, P, and K accumulations in the rapeseed. The N, P, and K accumulations of CRF4 and SF3 in each group had outstanding performance during different growth periods.

Fig. 2

N, P, and K accumulation of early ripening rapeseed using different fertilizer treatments

SF1/CRF1: 3750 kg/hm^{; SF2/CRF2: 3000 kg/hm; SF3/CRF3: 2250 kg/hm; SF4/CRF4: 1500 kg/hm; SF5/CRF5: 750 kg/hm; CK: no fertilizer. WP: wintering period; FS: flowing stage; HT: harvest time}

According to Fig. ​Fig.2,2, following the field application of SF, the N accumulation rapidly increased during the first phase (i.e., the stage prior to flowering) and then slowly increased after the flowering stage. A one-off application of SF can lead to a large loss during the plant growth, which causes insufficient nutrient supply in the late stage (Liu et al., 2011). However, the N accumulation increased relatively constantly with the increased application of CRF throughout the growth period, and caused obvious delays in leaf senescence. At the flowering stage, the ranks of N accumulation among all treatments were CRF3>CRF4>CRF5>CRF2>CRF1 and SF3>SF4>SF2>SF1>SF5. On the contrary, SF suffered from large N deficiencies at the flowering stage till harvest time.

At harvest time, nutrient accumulation first increased and then decreased with an increase in the fertilizer rate. The N, P, and K accumulations in CRF at harvest time were higher than those in SF by an average of 13.70%, 14.01%, and 10.33%, respectively. The SF suffered from nutrient deficiency at the flowering stage and at harvest time. CRF4 achieved maximum N accumulation (177.76 kg/hm^{) and the lowest was observed in SF1 (105.96 kg/hm). SF1, SF2, SF3, SF4, and SF5 were higher in N accumulation than CK by 109.78%, 192.16%, 193.33%, 172.92%, and 118.14%, respectively. In addition, CRF1, CRF2, CRF3, CRF4, and CRF5 were higher in N accumulation than CK by 150.47%, 187.86%, 229.50%, 251.93%, and 142.33%, respectively. There was a significant relationship between the N accumulation at maturity and seed yield (SF:} y=9.9331x+129.57, R^=0.4868, P<0.05; CRF: y=10.122x+127.25, R^=0.6515, P<0.05) as shown in Fig. ​Fig.3.3. It indicated that increasing the N accumulation appeared to be a necessary procedure to maximize seed yield in rapeseed.

Fig. 3

N accumulation as a function applied of early ripening rapeseed

3.5. N, P, and K usage efficiencies

With the increase of the fertilizer rate, N, P, and K usage efficiencies had a downward trend, and N, P, and K usage efficiencies of the CRF were increased by an average of 9.74, 2.28, and 8.18 percentage points, respectively, compared to SF (Table ​(Table5).5). CRF5 received the highest NUE (79.88%), followed by CRF4 (70.69 %), while the lowest NUE was observed for SF1 (12.28%). Under the same fertilizer rate, CRF1, CRF2, CRF3, CRF4, and CRF5 increased in NUE by 4.61, 1.57, 6.76, 22.17, and 13.59 percentage points compared with SF1, SF2, SF3, SF4, and SF5, respectively. On the other hand, the trends of P, K, and N usage efficiencies were consistent at maturity. P usage efficiency was in the range of 2.09%–23.79% for SF and 3.16%–28.54% for CRF, while K usage efficiency was in the range of 20.02%–74.07% for SF and 24.16%–89.42% for CRF. Therefore, CRF could significantly improve FUE for early ripening rapeseed and reduce the loss of nutrient. Corresponding increases in N, P, and K harvest index were observed with using CRF compared to SF and the values were increased with increasing the fertilizer rate.

The differences of the N, P, and K translocation rates and the N, P, and K contribution rates in each fertilization treatment were relatively small, while the P contribution rate showed a statistical decline trend with decreasing the fertilizer rate between CRF and SF, suggesting that the P accumulation of the rapeseed plant was caused more by the absorption of the exogenous P supplied. The N accumulation of CRF in rapeseed at the seedling stage accounted for an average of 62.86% of the total N accumulation throughout the growing season, while the corresponding proportion was higher than that of SF by 11.62 percentage points for plants. CRF3 achieved its maximum N translocation amount (79.69 kg/hm^{). SF1, SF2, SF3, SF4, and SF5 were higher in the N translocation amount than CK by 75.34%, 85.84%, 167.85%, 137.63%, and 99.91%, respectively. In addition, CRF1, CRF2, CRF3, CRF4, and CRF5 were higher in N translocation amount than CK by 126.07%, 135.57%, 263.88%, 221.23%, and 238.54%. Consequently, the N supply to the pods was achieved primarily by N mobilization from the vegetative parts.}

4. Discussion

4.1. Optimum application rates for CRF in the field

The application of slow and controlled release fertilizers reduces environmental pollution in terms of hazardous gaseous emissions and water eutrophication (Azeem et al., 2014) and simultaneously enhances FUE which is an important presupposition for the increased production profit (Gaju et al., 2011). Due to its relatively high price compared with SF, use of CRF may be limited unless its use is practical, profitable, and reduces the environmental N loss (Noellsch et al., 2009; Zebarth et al., 2009). In practice, the use of slow and controlled release fertilizers and/or stabilized fertilizers is primarily increased in greenhouses, golf courses, and professional lawn management, as well as by consumers (home and garden) and landscape gardeners (Arrobas et al., 2011).

Even if CRF use becomes economical, the widespread acceptance by growers will likely be limited as a result of grower concern about field performance. Some researches for increasing rice (Oryza sativa L.) yield by a single basal application of CRF were attributed to a greater soil available N supply, superior development of root systems, better nutrient absorption capacity, slower senescence, and enhancements of lodging resistance at the later stages (Carreres et al., 2003; Tang et al., 2007). Therefore, choosing the appropriate application rate for rapeseed is critical for the successful field application of CRF.

In this present study, with the increasing of the fertilizer rate, the seed yield of rapeseed first increased and then decreased. The application of CRF for seed yield got more than that of SF by an average of 17.33% with a lower fertilizer rate per kg yield by an average of 14.2% (Table ​(Table2).2). CRF4 achieved the highest yield (2066.97 kg/hm^{), followed by CRF3 (1929.97 kg/hm). Seed profit for CRF was significantly higher than that for SF (Table} ​(Table2).2). CRF4 achieved the highest profit rate, 7126.4 CNY/hm^{, followed by CRF5 (7033.1 CNY/hm). Compared with SF4, CRF4 increased by 12.37% in seed yield and decreased the fertilizer rate per kg yield by 11.01%. According to the regression equations of the fertilizer rate on seed yield, the best fertilizer rates for SF and CRF were 1802.12 and 1843.04 kg/hm, respectively, achieving maximum yield (1839.24 and 2038.35 kg/hm, respectively), and net profit (5753.62 and 6500.16 CNY/hm, respectively). Thus, a rate of 1500 kg/hm in CRF has been shown to be adequate for early ripening rapeseed production.}

4.2. Yield components of CRF and SF

N increases its yield by influencing a number of growth parameters such as branches per plant and flowers per plant and by producing more vigorous growth and development. It has been well documented that increasing the N rates produces more lodged plants (Ozer, 2003). This obviously has an indirect effect since high N rates promote the formation of more pods and seeds but also decrease in stem stability. Cheema et al. (2001) also showed that the number of pods per plant increased with increasing rates of N. CRF can significantly increase pod numbers and seed yield, while improving the ratio of seed/stem (Yu et al., 2012). Wang et al. (2013) showed that the coated fertilizer or CRF not only increased the seed yield of rapeseed, but led to effective pod numbers, seed numbers per pod, 1000-seed weight and FUE, and in addition, also improved the resistance to cold, disease, and lodging of rapeseed.

In this present study, the yield differences measured for CRF and SF were primarily due to the changes in first branch numbers, pod numbers, stem dry weight, and pod dry weight. The results also revealed that the number of pods increased with an increase in the applied fertilizer rate, and the major cause of reduced yields in SF is decrease in the number of pods per plant. The pod number reflects a significant increase with the seed yield (SF: R^{=0.8098; CRF:} R^{=0.7281). Compared with SF, although farmers like using fertilizer which costs less, CRF can achieve a greater profit, and is attributed to increased soil availability of N (data not shown), superior development of the root systems, better nutrient absorption capacity, delayed senescence, and enhanced lodging resistance.}

4.3. Improvement of FUE by CRF

FUE has become a critical measure of sustainable agriculture. A successful plant growth supported by a high N uptake rate will be decisive for reaching high seed yield (Barłóg and Grzebisz, 2004). There was a close and positive relationship between nutrient uptake and seed yield, especially in N. N fertilization treatment may significantly affect the strategy of rapeseed plants adaptation to N availability. A successful plant growth supported by a high N uptake rate will be decisive for reaching high seed yield (Barłóg and Grzebisz, 2004), but it frequently leads to reduction of NUE after the increasing of the N uptake. Higher NUE is an important prerequisite for increased profitability, either through greater yields or reduced N losses (Liu et al., 2008; Gaju et al., 2011). Malagoli et al. (2005a) showed that the N requirements for seed filling were satisfied primarily by N mobilized from vegetative parts (about 73% of the total N in pods). The mobilization of endogenous N from these leaves was prolonged and concomitant with N accumulation in the pods. How to promote N transport from the vegetative organs to seed in the late growing stage and improve N seed physiological efficiency, so as to achieve the unity of high yield and high NUE of rapeseed on the basis of a certain amount of N uptake, is becoming a hot issue of current research.

In this present study, there was a significant relationship identified between N accumulation at maturity and seed yield (SF: R^{=0.4868; CRF:} R^{=0.6515). It was noted that NUE, post-heading N uptake, the contribution of N translocation to the total N in the seed, the partial factor productivity of N as well as seed yield were significantly influenced by CRF. After the flowering stage, N translocation and remobilization of different organs became the major metabolism, rather than N absorption by the entire plant (Zuo et al.,} 2014). At maturity, dry matter, nutrient uptake, and their harvest indexes of rapeseed plants varied among the different fertilization treatments, with a significant fertilization being identified with all these parameters. Prolonged N uptake from the soil with CRF resulted in delayed plant senescence. N requirements for seed filling were satisfied primarily by N mobilized from vegetative parts of CRF (average of 62.86% for the total N in the pods), and the endogenous N flow showed that there was a net transfer of N to the pods by the stem and leaves.

With increasing the fertilizer rate, N accumulation increased and N harvest index gradually reduced. The proportion of N accumulated in seeds at the flowering stage compared to the total N at maturity increased gradually. Under high N fertilization, the N harvest index became less, indicating that the appropriate amount of N fertilizer can promote nutrients transporting from other organs of the plant to seed, while further increasing N being only limited by the seed capacity of the plant itself makes it hard for the output of nutrient accumulation in vegetative organs, which leads to the harvest index decreasing. Furthermore, the NUE of different fertilizer rates for CRF was significantly higher than that for SF (P<0.05). These results suggest that the CRF used in this study was effective for the agricultural production of rapeseed, and also indicate that the CRF4 treatment corresponded to the optimum application rate of CRF for the rapeseed fields studied in southern China.

5. Conclusions

In this study, a single basal application of CRF obviously contributed to a greater supply of soil available N, thus creating a suitable nutritional environment for vigorous growth of rapeseed. Seed yield of early ripening rapeseed was significantly higher for CRF treatments than for SF treatments, while no significant differences in seed yield were found between CRF/SF (3750 kg/hm^{) and CK. These results indicated that 1500 kg/hm was the optimum application rate for CRF/SF in rapeseed fields in southern China. CRF and the abilities of N, P, and K fertilizer applied to improve N, P, and K usage efficiencies resulted in increased seed yield as well as the expected environmental and economic benefits. In addition, the yield increases afforded by CRF were partly due to higher nutrient accumulation and FUE. It is necessary to determine the effects of the frequency and timing of CRF applications on rapeseed in the future.}