One-Time Nitrogen Fertilizer Application Using Controlled-Release Urea Ensured the Yield, Nitrogen Use Efficiencies, and Profits of Winter Wheat

Abstract

One-time fertilization with controlled-released fertilizer (CRF) is a promising way for reducing labor cost, increasing nitrogen use efficiency (NUE) and alleviating environmental pollution in winter wheat (Triticum aestivum L.) cultivation. However, CRF release characteristics are related to various factors such as soil properties, temperature and precipitation, and further study is needed in developing suitable fertilizer formulas adapting to local conditions. In this study, five types of CRF were used for a one-time application in a two-year field experiment conducted at two sites with two wheat cultivars. Their effects on soil inorganic nitrogen (SIN) content, nitrogen uptake, wheat growth and grain yield were investigated. The results indicated that nitrogen supply in the CRF-60/80 treatments not only ensured the tiller differentiation at the early growth stage, but also provided adequate SIN after the jointing stage, thereby facilitating the dry matter accumulation and improving post-anthesis photosynthates accumulating in grains. When compared with conventional split fertilization, the CRF-60 and 80 treatments improved the NUE by 9.7–12.1%, and boosted farmers’ economic efficiency by 282.4–327.2 CNY ha⁻¹. According to our research, a one-time application of CRF-60 and CRF-80 could meet the needs of the two-peak nitrogen demand of wheat in Jiangyan and Yanghzou respectively, therefore increasing NUE and having low labor costs for wheat fertilizer application.

1. Introduction

Under the increasing pressure of global food security, agriculture faces critical challenges to achieve higher yields and better quality with lower environmental costs [1]. As an important staple crop within and beyond China, wheat (Triticum aestivum L.) grain yield has increased dramatically mainly due to the improvement in breeding and crop cultivation technologies [2]. China has reached a wheat grain yield of 5.74 tons per hectare in 2020, which is 1.65 times the world average [3]. As the key contribution to crop productivity, nitrogen (N) plays a dominant role in this increment. However, the accumulation of N from wheat cultivation has led to energy waste and environmental pollution in the hydrosphere, atmosphere, and pedosphere. To reduce the reactive N loss and improve N fertilizer use efficiency, farmers in China were directed to split the N fertilizer three to four times to match the need of the crop’s N requirement [4]. Whereas, the inadequate labor resources due to urbanization makes the increased use of fertilizer topdressing a heavy economic burden for farmers [5]. Therefore, it is in urgent need and of vital importance to develop new fertilizer practices with both high N-use efficiency and low labor cost.

Many N management strategies have been developed to address this issue, such as the deep placement of N fertilizer, stabilized fertilizers, urease inhibitors, and slow- and controlled-released fertilizer [6,7,8]. Among these improvements, controlled-release urea (CRU) has been considered as an effective and promising way. CRU, such as sulfur-coated urea (SCU), polymer-coated urea (PCU), polymer coating of sulfur-coated urea (PSCU), etc., is encapsulated to release N in soil solution gradually to match the crop’s nutrient needs along with the crop duration [9,10]. The effects of CRU on reducing active N loss and labor input, improving crop yield and N use efficiency have been widely reported in different crops [11,12,13,14], but these effects vary greatly depending on the nutrient release patterns in different kinds of CRU [15,16]. The “burst” N release of SCU in an unsuitable time has often been detected [16]. In the fertilizer market, the price of CRU products with stable controlled release performance, such as polyurethane-coated urea, was much higher than SCU. The high cost-to-benefit ratio cannot persuade farmers to use them in large area cereal crop production [17]. Therefore, bulk blending urea (BBU) by blending the CRU with conventional urea could be a wise choice, because it would not only decrease the cost of fertilizer but also solve the side effect of short-term N immobilization caused by straw retention [18]. A one-time application of BBU should be the most ideal way to improve NUE and grain yield and minimize the labor input of fertilization. Field experiments have been conducted on rice and maize, suggesting this novel application can achieve similar or higher yield, preserving the soil fertility and also being labor-saving [14,19,20]. Although much effort has been made to clarify the effects of CRU on wheat cultivation, there is still limited information regarding the effect of a one-time application of BBU consisting of CRU and conventional urea on soil inorganic nitrogen content, nitrogen uptake and utilization and grain yield. When compared to rice, wheat has a longer growth period, during which the temperature and precipitation fluctuate greatly [13]. Therefore, it would be more difficult to determine a suitable fertilizer formula that releases N in sync with the demand of winter wheat.

It is generally assumed that there are two peaks of N demand for winter wheat: at the seedling stage and after jointing, and the demand is much higher in the latter [13,21]. Conventional urea from the one-time basal application of BBU could supply the N needed for wheat tiller differentiation. So, the critical point is to choose the CRU with the appropriate release longevity to meet the N needs of wheat after the jointing stage. The N release rate from CRU is dependent on soil properties such as temperature, water content, and pH [22]. In winter, the release rate of CRU is very low due to the low temperature. According to Yao et al. (2019) and Zheng et al. (2016), the CRUs with a controlled release longevity of 100 days in 25 °C water release less than 20% of total N in the first 100 days in a wheat field. Soil texture, which is often neglected in previous studies, is also very important to the release rate of CRU [22,23,24]. Soil texture can influence CRU release by affecting the soil moisture holding capacity and soil temperature [25,26]. Moreover, farmlands with different soil textures vary greatly in the size of soil particles after plowing, which will strongly affect the contact between soil and fertilizer, thus affecting the water absorption effect and release rate of CRU granules. Therefore, it is necessary to select a suitable CRU for the one-time application according to the meteorological conditions and soil types in a certain area.

The middle and lower reaches of the Yangtze River are important areas of winter wheat production in China. Considerable field trials conducted in this area have focused on the effects of CRU on wheat yields and environmental pollution, but few studies have explored the suitable BBU formula for one-time fertilizer application, to achieve the goal of simplified and less labor-intensive wheat cultivation [13,16,23,27,28,29]. Hence, the cumulative release rate of CRU, the temporal variation of soil inorganic N, the N uptake, utilization and yield of wheat, and the economic benefit accounting were investigated in this study. The objectives of the present research were to (a) determine if it is possible to supply the N demand and optimize the NUE for wheat during the whole growth stages with the one-time application of BBU consisting of CRU and conventional urea, and to (b) derive the best options for high efficiency and simplify fertilization methods with minimal labor input in different soil types in the middle and lower reaches of the Yangtze River in China.

2. Materials and Methods

2.1. Experimental Sites and Materials

The field experiments were conducted during two consecutive wheat growing seasons (2017 to 2019) at the Yangzhou Agricultural Experimental Station, Yangzhou City (32°39′ N, 119°42′ E) and Heheng Agricultural Experimental Station, Jiangyan City (32°61′ N, 120°14′ E) in Jiangsu province, China. At both sites, winter wheat was grown in a rotation with rice (Oryza sativa L.), and in both areas the cultivation system has been a rice–wheat rotation for more than five years. The test soils at the Yangzhou site and Jiangyan site were fluvo-aquic soil and paddy soil, and the soil textures were sandy loam and clay loam, respectively. The basic physical and chemical properties for top layer soil samples (0–20 cm) in 2017 are listed in

Table 1. Soil properties at different experiment sites.

Experimental Sites	Soil Texture	FAO Soil Taxonomy	pH	Total N	SOM	Olsen-P	Available K
				g kg−1	g kg−1	mg kg−1	mg kg−1
Yangzhou	Sandy loam	Eutric fluvisols	6.10	1.40	15.23	52.17	141.46
Jiangyan	Clay	Cumulic anthrosols	6.72	1.96	31.72	62.54	165.26

. Both experimental sites were located in the middle and lower reaches of Yangtze River, with a humid subtropical monsoon climate. The annual mean air temperature in Yangzhou site in 2018 and 2019 is 16.7 and 17.1 °C, respectively, and the precipitation is 1288.0 and 830.1 mm, respectively. The annual mean air temperature in Jiangyan site in 2018 and 2019 is 16.8 and 16.6 °C, respectively, and the precipitation is 1185.7 and 837.8 mm, respectively. The temperature and precipitation data of both sites during the winter wheat growth season are provided in Figure 1. The winter wheat cultivars used in this study were ‘Zhenmai 12’ (Z12) and ‘Yangmai 23’ (Y23), both of which are widely cultivated medium–strong glutens wheat in this area. Five types of polyurethane-coated CRUs (N 44%) with different release longevities (40, 60, 80, 100 and 120 days) manufactured by Shandong Maoshi Ecological Fertilizer Inc were used. This polyurethane-coated CRU is a type of PCU that is coated with a thermosetting polyurethane material on the outside. The fertilizers used in the controlled-released fertilizer (CRF) treatments of this study were mixtures of CRU and conventional urea with the N ratio of 6:4.

2.2. Experimental Design

A split-plot design with triple replicates was used in this two-year field experiment at both sites. The main plots were assigned the varieties of winter wheat (Z12 and Y23), while N fertilization modes were assigned to the subplot, including a 4-time split application with conventional urea as a control (CK) (40% at basal, 20% at four-leaf stage, 20% at jointing stage and 20% at booting stage) and a one-time basal application with CRF of different release longevities (40 days (CRF-40), 60 days (CRF-60), 80 days (CRF-80), 100 days (CRF-100), and 120 days (CRF-120)). The N rate applied in this study was 225 kg ha⁻¹, which was considered to be the optimum N rate in local wheat cultivation. Fertilizers of 112.5 kg ha⁻¹ P₂O₅ (as superphosphate) and 112.5 kg ha⁻¹ K₂O (as KCl) were also applied before sowing. A blank control treatment with no N application (N0) was also set for NUE calculation. Detailed information about the N fertilizer was listed in

Table 2. Fertilizer information in the experimental design.

Fertilization Treatment	N Application Level	N Fertilizer Application Method	N Source Type	Release Longevities of CRU
CK	225 kg ha−1	four-split application	CU	-
CRF-40		one-time basal application	40% CU + 60% CRU	40 days
CRF-60				60 days
CRF-80				80 days
CRF-100				100 days
CRF-120				120 days
N0	0 kg ha−1	-	-	-

Note: CU, conventional urea; CRU, controlled release urea.

.

There were 42 plots at each experiment site; plots were arranged randomly with a size of 30 m² (5 × 6 m). Wheat seeds were sown with a row space of 25 cm and a density of 225 × 10⁴ plant ha⁻¹ on 3 November 2017 and 1 November 2018. Pathogens, insect pests, and weeds in wheat fields were controlled with common chemical treatments.

2.3. Samplings and Measurements

2.3.1. Measurement of the N Content and Release Characteristics of CRU

The N release characteristics of CRU were determined according to the method described in “State Standard of the People’s Republic of China—Slow Release Fertilizer” [30]. An amount of 10.00 g of fertilizer samples were weighted and sealed in a nylon mesh bags (100 mesh, 10 cm length and 3 cm width). Each mesh bag was put into a glass bottle with 200 mL of deionized water, and was incubated at 25 °C. Four replicates were set for each fertilizer. The water in the glass bottle was periodically sampled and analyzed with the Kjeldahl method [31] for N contents at 5 day intervals during the first 40 days, and at 10 day intervals until 120 days, or until the cumulative N dissolution rate of CRU reached 80%. After each sampling, water in the glass bottles were replaced with new deionized water.

2.3.2. Soil Inorganic N Content

Soil samples in each experimental site were taken at the main growth stages of wheat (the tillering, over-wintering, re-greening, jointing, anthesis, and maturity stage). Five sub-samples were collected and mixed in each plot, soil samples were sealed in plastic bags and stored in 4 °C before testing. After sampling, fresh soil samples were immediately extracted with 0.01 M CaCl₂ solution. Soil NO₃⁻-N and NH₄⁺-N concentrations were determined with the continuous flow analyzer (AA3, Seal Analytical, Germany). Soil inorganic N content listed in this study was the sum of NO₃⁻-N and NH₄⁺-N contents.

2.3.3. Grain Yield Determination

Two 1 m² (four one-meter rows) areas in each plot were randomly selected to determine the grain yield and yield components. The spike numbers in this area were counted before the wheat was artificially harvested. In each 1 m² sample area, 50 consecutive spikes were selected to determine the number of kernels per spike. The 1000-grain weight (TGW) and grain yield were weighted with the grain moisture adjusted to 13%.

2.3.4. Nitrogen Accumulation

20 plants were sampled from each plot at the main growth stages (the tillering, re-greening, jointing, anthesis, and maturity stage) and separated into different organs. All samples were placed in a 105 °C oven for 20 min for enzyme deactivation before being dried at 80 °C to a constant weight. Dried samples were weighted, ground and sifted through a 2-mm sieve. Nitrogen concentrations were then measured with the micro-Kjeldahl method described by Douglas et al. [32]. Parameters were calculated with the following formulas:

Nitrogen accumulation (kg ha⁻¹) = Nitrogen concentration × dry weight

N Agronomic Efficiency (AE_N, kg kg⁻¹) = (Grain yield in N application treatment − Grain yield in N0 treatment)/N application rate

N Partial Factor Productivity (PFP_N, kg kg⁻¹) = Grain yield in N application treatment/N application rate

N Apparent Recovery Efficiency (ARE_N, %) = (Total plant N uptake in N application treatment − Total plant N uptake in N0 treatment)/Total amount of applied N

2.3.5. Analysis of Relative Chlorophyll Content and Net Photosynthetic Rate

The relative chlorophyll contents of flag leaves were detected and quantified with a SPAD 502 chlorophyll meter (Minolta Camera Co., Ltd., Osaka, Japan) at the anthesis and milking stages. Simultaneously, the net photosynthetic rate (Pn) of flag leaves was measured using a LI-6400 portable photosynthesis system (LI-COR Inc., Lincoln, RI, USA). The measurements of the SPAD and Pn values were taken between 9:00 and 11:00 a.m. on sunny days.

2.4. Statistical Analysis

Data were analyzed by the analysis of variance (ANOVA) with SPSS for Mac (Version 27.0., IBM Corp, Armonk, NY, USA). The statistical model included sources of variation due to location, year, wheat variety, fertilizer treatment and their interactions. The means were compared by the Duncan’s test at the 0.05 probability level (p < 0.05). Figures were drawn with SigmaPlot program (SigmaPlot 14.0. Systat Software, San Jose, CA, USA).

2.5. Economic Evaluation

The economic evaluation is explained in details in below:

Total input = Fertilizer cost + Fertilization labor cost + Other cost

Net income = Grain yield × Grain price − Total input

3. Results

3.1. N Release Behavior of CRU and the Response of Soil Inorganic N Content

Figure 2 shows the N release behaviors of CRFs at 25 °C in water, the release curves indicated an “S”-type release pattern. In the first slow-releasing stages (10, 15, 25, 30 and 40 days for CRF-40, CRF-60, CRF-80, CRF-100 and CRF-120 respectively), CRFs released 17.7, 14.9, 18.8, 16.9 and 16.8% to the total N, respectively. In the following 15, 20, 30, 50 and 60 days, which were the rapid release stages, CRFs released 58.8, 56, 56.1, 59.3 and 61.3% of the total N respectively. Subsequently, the CRF release rates decreased, and total release rates reached 88.1, 89.5, 89.2, 87.8 and 88.9% respectively, meeting the standards of Chinese Industry Evaluation Standard for Slow and Controlled Release Fertilizers (GB/T 23348-2009), which regulates that total amount of fertilizer nutrients released during the specified release period cannot be less than 80%.

The diverse release patterns of CRFs significantly affect the dynamic of soil organic N content across the wheat growth seasons (Figure 3). A declining tendency was observed in the inorganic N contents of the topsoil layer in all the fertilizer treatments throughout the wheat growing period. At the tillering stage, soil inorganic N contents in CRF-100D and 120D were relatively lower than that of CRF-40D, 60D and 80D treatments. The controlled-release effect of different CRF fertilizers was reflected through the diverse decrease magnitude of inorganic N contents. CRFs with release period of 100 and 120 days did not demonstrate their impact of N release on soil inorganic N content until the jointing stage (125 days after seeding). Afterwards, soil organic N contents in all the treatments decreased dramatically at the maturity stage. In the control treatment, the variation of soil organic N content during wheat growth stages was more substantial than that in CRF treatment. The high soil inorganic N level resulting from the basal fertilizer dropped rapidly during the vegetative growth stage. Afterwards, the soil inorganic N content significantly increased after the topdressing of urea at the jointing stage, and then decreased quickly at the anthesis and maturity stages. Soil fertility variation in field trail sites led to higher soil inorganic N status in the Jiangyan site than that in Yangzhou site across all treatments. Moreover, the increase of soil inorganic N content in CRF-100D and 120D treatments at the anthesis stage was gentler in the Jiangyan site than in Yangzhou site, probably due to the buffering effect of a larger N pool.

3.2. Shoot Biomass and N Accumulation in Wheat

During different growth periods, fertilizations affected shoot biomass accumulations to varying degrees (

Table 3. Biomass accumulation of wheat shoot in different CRF treatments in Yangzhou and Jiangyan sites during growth seasons in 2017–2018 and 2018–2019.

Cultivar	Fertilization	2017–2018				2018–2019
		Jointing (kg·ha−1)	Anthesis (kg·ha−1)	Maturity (kg·ha−1)	DM after Anthesis (kg·ha−1)	Jointing (kg·ha−1)	Anthesis (kg·ha−1)	Maturity (kg·ha−1)	DM after Anthesis (kg·ha−1)
Yangzhou
Y23	CK	6171.73 a	11,394.59 a	18,237.22 a	6842.63 a	6931.93 a	12,798.12 a	20,482.46 a	7684.35 a
	CRF-40	6151.78 a	10,952.72 ab	15,738.98 c	4786.26 d	6909.83 a	12,734.78 a	19,237.83 ab	6503.05 bc
	CRF-60	6135.32 a	11,329.17 a	18,258.02 a	6928.85 a	6899.90 a	12,741.01 a	20,533.34 a	7792.32 a
	CRF-80	5895.34 ab	10,835.39 b	17,049.90 b	6214.51 b	6783.08 ab	12,680.98 a	19,955.84 ab	7274.86 b
	CRF-100	5776.75 b	10,312.03 c	15,518.55 c	5206.52 c	6363.33 b	12,192.34 b	18,366.04 b	6173.70 c
	CRF-120	5745.14 b	10,182.54 c	15,261.59 c	5079.05 c	5922.85 c	11,930.53 b	17,916.67 c	5986.14 c
Z12	CK	6281.47 a	11,759.37 a	18,897.10 a	7137.73 a	7094.20 a	13,011.59 a	20,909.39 a	7897.80 a
	CRF-40	6251.08 a	11,322.49 ab	17,020.56 bc	5698.07 c	7061.65 a	12,394.13 b	19,288.29 ab	6894.16 b
	CRF-60	6253.96 a	11,602.02 a	18,716.86 a	7114.84 a	7025.99 ab	13,034.25 a	21,027.39 a	7993.14 a
	CRF-80	5992.73 b	11,406.63 ab	17,947.67 b	6541.05 b	6501.51 b	12,692.75 ab	19,971.32 ab	7278.57 ab
	CRF-100	5845.52 b	10,769.47 b	16,577.47 c	5808.00 c	6181.04 bc	12,025.94 bc	18,477.25 b	6451.31 bc
	CRF-120	5794.63 b	10,417.14 c	16,004.41 c	5587.27 d	6029.10 c	11,664.66 c	17,886.70 c	6222.04 c
Jiangyan
Y23	CK	6795.55 ab	12,370.46 a	19,697.41 a	7326.95 a	7218.14 a	13,139.74 a	20,922.33 a	7782.59 ab
	CRF-40	7010.39 a	12,022.55 a	18,805.59 b	6783.04 b	7260.53 a	12,451.53 b	19,476.59 b	7025.06 c
	CRF-60	6886.65 a	12,250.40 a	19,458.77 a	7208.37 a	7224.28 a	13,040.35 a	20,713.55 a	7673.19 ab
	CRF-80	6820.57 a	12,173.35 a	19,567.27 a	7393.92 a	7088.26 ab	13,033.29 a	20,949.53 a	7916.24 a
	CRF-100	6660.54 b	11,332.74 b	17,775.97 c	6443.24 b	6787.45 b	12,286.54 b	19,272.06 bc	6985.52 c
	CRF-120	6496.32 b	11,663.22 b	17,676.30 c	6013.08 c	6708.44 b	12,232.36 b	18,538.86 c	6306.50 d
Z12	CK	6727.08 a	12,156.89 a	19,359.19 a	7202.30 ab	7285.31 a	13,165.69 a	20,965.65 a	7799.96 a
	CRF-40	6728.51 a	11,469.12 b	17,837.43 b	6368.31 c	7307.40 a	12,839.43 b	19,970.84 b	7131.42 b
	CRF-60	6741.50 a	12,310.57 a	19,295.73 a	6985.16 ab	7288.23 a	13,092.73 a	20,859.52 a	7766.79 a
	CRF-80	6631.47 a	12,206.34 a	19,818.08 a	7611.74 a	6915.44 ab	13,067.26 a	20,915.92 a	7848.65 a
	CRF-100	6313.46 b	11,089.51 b	17,568.49 b	6478.98 c	6482.22 b	12,343.87 b	19,363.66 b	7019.79 b
	CRF-120	6176.55 c	10,359.10 c	15,760.24 c	5401.14 d	6183.46 c	12,115.78 b	19,107.80 c	6992.02 b

Note: CK, conventional urea with a four-split application; CRF-40, CRF-60, CRF-80, CRF-100, CRF-120, one-time basal application with the mixture of conventional urea and CRU with the release longevities of 40, 60, 80, 100 and 120 days respectively. Mean values followed by different lowercase letters within column in each variety and site were significantly different at 0.05 probability level.

). At both experimental sites, biomass accumulations at the jointing stage were higher in CK and CRF treatments with controlled-release longevities of 40, 60 and 80 days when compared to those in CRF-100 and CRF-120 treatments. At the anthesis and maturity stages, wheat in the CRF-40 treatment did not accumulate biomass as much as that in the CRF-60 and CRF-80 treatments. After anthesis, dry matter accumulation was higher in the CK and CRF-60 treatments in Yangzhou site, while in Jiangyan site, wheat in the CK, CRF-60 and CRF-80 treatments had higher dry matter accumulation than that in other treatments.

Furthermore, fertilizer applications had a range of effects on the uptake of total N and on N accumulation at different growth stages, and these effects varied with the experimental site (Figure 4). In terms of total N uptake, wheat in the CRF-60 and CK treatments displayed higher values than other treatments in Yangzhou site. In Jiangyan site, wheat accumulated the highest amount of N in the CRF-80 treatment, but this was not significantly different from that of CK and CRF-60 treatments. The total N uptake of wheat decreased significantly with the extension of the controlled-release period. However, the total N uptake of the CRF-40 treatment was lower than CK in all varieties and sites, mainly due to its low N accumulation during the jointing to maturity stage. For CRF-100 and CRF-120 treatments, insufficient N supply caused the low accumulation during the re-greening to anthesis stage, leading to a small biomass base for the N accumulation and transfer after anthesis.

3.3. Wheat Post-Anthesis Photosynthetic Characteristics

After anthesis, the chlorophyll content (SPAD) and net photosynthetic rate (Pn) of wheat flag leaves decreased rapidly, and the effects of different fertilization treatments on them were consistent (Figure 5 and Figure 6). Wheat photosynthetic characteristics in the CK, CRF-60 and CRF-80 treatments were higher than other treatments at the anthesis stage. Soil nitrogen availability in different fertilization treatments affects the performance of post-anthesis photosynthesis in wheat flag leaves. Following post-anthesis senescence of the leaves, SPAD and Pn values declined rapidly, but the effects of the fertilization treatment varied according to the experimental site. These indicators in Yangzhou site were significantly higher in CRF-60, CRF-80, CRF-100, and CRF-120 treatments than in CK and CRF-40 treatments, while in Jiangyan, the SPAD and Pn values of the CRF-60 treatment were significantly lower than those of the CRF-80, CRF-100, and CRF-120 treatments at the milking stage.

3.4. Grain Yield and Its Components

At both experimental sites, one time basal fertilizer applications were able to obtain comparably high yields to the control treatments in two years (

Table 4. Grain yield and yield components in different CRF treatments in Yangzhou and Jiangyan sites in 2017–2018 and 2018–2019.

Cultivar	Fertilization	2017–2018				2018–2019
		Grain Yield (kg·ha−1)	Spikes (×104 ha−1)	Grains per Spike	TGW (g)	Grain Yield (kg·ha−1)	Spikes (×104 ha−1)	Grains per Spike	TGW (g)
Yangzhou
Y23	CK	7600.00 a	443.84 a	43.0 a	41.22 b	8536.13 a	451.19 a	45.9 a	43.64 b
	CRF-40	6076.89 c	410.18 b	41.6 b	40.84 b	7935.00 b	432.47 ab	42.9 bc	43.66 b
	CRF-60	7632.96 a	439.60 a	44.1 a	41.64 b	8584.18 a	455.42 a	44.9 ab	44.05 b
	CRF-80	7442.24 ab	437.90 a	38.8 b	41.44 b	8242.50 ab	446.51 a	43.3 bc	44.88 ab
	CRF-100	6329.80 c	411.37 b	37.1 c	41.96 b	7491.25 c	417.62 bc	42.4 c	45.56 a
	CRF-120	6081.92 c	385.65 c	37.5 c	43.69 a	7140.00 d	404.66 c	42.8 bc	45.72 a
Z12	CK	7880.43 a	427.73 ab	39.2 a	53.56 ab	8719.59 a	472.34 a	38.7 ab	52.05 bc
	CRF-40	6985.13 c	412.85 ab	34.9 c	52.88 ab	7646.25 b	440.37 ab	36.5 c	51.31 c
	CRF-60	7868.92 a	438.84 a	41.2 a	51.04 c	8840.31 a	470.92 a	39.2 a	52.57 b
	CRF-80	7257.84 b	408.40 b	38.5 ab	51.13 c	8076.18 b	440.37 ab	37.0 bc	52.49 b
	CRF-100	6705.32 c	406.17 b	35.6 bc	52.06 bc	7473.75 c	406.58 b	36.3 c	54.24 ab
	CRF-120	6440.30 d	393.95 b	34.3 c	54.37 a	7197.75 d	399.32 b	36.0 c	55.57 a
Jiangyan
Y23	CK	8144.50 a	439.42 ab	39.9 ab	44.53 bc	8650.98 a	432.06 a	46.8 a	44.65 c
	CRF-40	7690.71 b	430.34 b	37.5 bc	42.91 d	7965.12 b	429.13 a	43.9 bc	45.07 c
	CRF-60	7937.62 ab	431.51 b	41.4 a	43.99 c	8449.47 a	440.10 a	44.5 bc	45.34 bc
	CRF-80	8166.44 a	445.88 a	42.7 a	43.79 c	8743.33 a	444.69 a	46.0 ab	45.37 bc
	CRF-100	7293.77 c	402.82 c	33.6 d	44.95 b	7907.64 b	396.54 b	44.3 bc	45.95 b
	CRF-120	7100.17 c	399.62 c	36.5 cd	45.84 a	7446.64 c	385.67 b	42.9 c	46.90 a
Z12	CK	8116.48 ab	479.82 ab	35.6 a	52.38 b	8790.00 a	500.03 a	39.3 a	52.91 b
	CRF-40	7203.11 c	468.82 b	34.1 ab	52.54 b	8063.72 c	488.16 a	36.4 c	52.92 b
	CRF-60	7931.76 b	481.41 a	34.0 ab	52.69 b	8575.02 b	509.88 a	37.8 b	52.96 b
	CRF-80	8344.70 a	489.07 a	35.0 a	52.22 b	8718.97 a	516.96 a	38.8 ab	53.43 b
	CRF-100	7130.57 c	454.77 c	32.3 bc	53.08 b	8234.13 c	448.35 b	34.9 d	53.75 b
	CRF-120	6202.95 d	435.37 d	31.5 c	54.12 a	8030.00 c	412.00 c	33.5 e	55.60 a

Note: TGW, 1000-grain weight; CK, conventional urea with a four-split application; CRF-40, CRF-60, CRF-80, CRF-100, CRF-120, one-time basal application with the mixture of conventional urea and controlled release urea with the release longevities of 40, 60, 80, 100 and 120 days respectively. Mean values followed by different lowercase letters within column in each variety and site were significantly different at 0.05 probability level.

). The grain yield of Y23 in the CRF-60 and CRF-80 treatments did not differ significantly from CK at either site. For the Z12 variety, CRF-60 in Yangzhou and CRF-80 in Jiangyan increased the yields when compared with the other one-time fertilizer treatments. In comparison with the two-year average yield of CK, the CRF-60 treatment increased the yields of Y23 by 0.5% and Z12 by 0.7% in the Yangzhou site, and CRF-80 treatment increased yields of Y23 by 0.7% and Z12 by 0.9% in the Jiangyan site (

Table 5. Two-year average grain yield of wheat and change relative to CK from 2017 to 2019.

Cultivar	Fertilization	Yangzhou		Jiangyan
		Average Yield (kg·ha−1)	% Change Relative to CK	Average Yield (kg·ha−1)	% Change Relative to CK
Y23	CK	8068.1	-	8397.7	-
	CRF-40	7005.9	−13.2	7827.9	−6.8
	CRF-60	8108.6	0.5	8193.5	−2.4
	CRF-80	7842.4	−2.8	8454.9	0.7
	CRF-100	6910.5	−14.3	7600.7	−9.5
	CRF-120	6611.0	−18.1	7273.4	−13.4
Z12	CK	8300.0	-	8453.2	-
	CRF-40	7315.7	−11.9	7633.4	−9.7
	CRF-60	8354.6	0.7	8253.4	−2.4
	CRF-80	7667.0	−7.6	8531.8	0.9
	CRF-100	7089.5	−14.6	7682.4	−9.1
	CRF-120	6819.0	−17.8	7116.5	−15.8

). The two-year average yield of the other treatments was significantly lower than that of the CK treatment.

Additionally, location-specific variations were observed in yield components (Table S1, TWG F_Location = 79.656, p < 0.01; Grains per spike F_Location = 38.811, p < 0.01; Spikes F_Location = 130.173 ** p < 0.01). CRF-60 in Yangzhou and CRF-80 in Jiangyan along with the control treatment had the highest spike number and grains per spike, while the 1000-grain weights (TGW) were lower than the other treatments. Based on these results, the steady and increased yields of wheat by the CRF treatment depend primarily on the spike number and grains per spike.

3.5. Nitrogen Use Efficiency and the Benefit-Cost Analysis

Fertilizations had significant effects on wheat PFP_N, AE_N and ARE_N, but the trends among treatments varied by location (

Table 6. Nitrogen use efficiencies of wheat under different fertilizer treatments in Yangzhou and Jiangyan.

Cultivar	Fertilization	Yangzhou			Jiangyan
		PFPN (kg kg−1)	AEN (kg kg−1)	AREN (%)	PFPN (kg kg−1)	AEN (kg kg−1)	AREN (%)
Y23	CK	35.86 a	19.62 a	49.23 b	37.32 a	20.67 a	51.33 b
	CRF-40	31.14 c	14.90 c	30.56 de	34.79 c	18.14 c	40.10 c
	CRF-60	36.04 a	19.80 a	54.33 a	36.42 b	19.77 b	49.97 b
	CRF-80	34.85 b	18.61 b	41.05 c	37.58 a	20.93 a	56.32 a
	CRF-100	30.71 d	14.47 c	31.66 d	33.78 cd	17.13 d	36.73 c
	CRF-120	29.38 d	13.14 d	29.83 e	32.33 d	15.68 e	21.98 d
Z12	CK	36.89 a	20.40 a	50.86 b	37.57 a	20.64 a	51.43 b
	CRF-40	32.51 c	16.02 c	27.48 d	33.93 c	17.00 c	37.31 c
	CRF-60	37.13 a	20.64 a	56.05 a	36.68 b	19.75 b	52.02 b
	CRF-80	34.08 b	17.58 b	39.40 c	37.92 a	20.99 a	57.63 a
	CRF-100	31.51 d	15.02 d	23.87 e	34.14 c	17.21 c	34.95 c
	CRF-120	30.31 d	13.81 e	18.48 f	31.63 d	14.70 d	27.92 d

Note: Means followed by different lowercase letters in the same column were significantly different by Duncan’s test (p < 0.05).

). The PFP_N and AE_N of both wheat cultivars were highest in the CK treatment, as well as the CRF-60 treatment in Yangzhou and the CRF-80 treatment in Jiangyan. The PFP_N and AE_N were significantly lower than those in CK at both experimental sites in the other treatments with suboptimal controlled-release longevities (CRF-40, 100 and 120). The average of two wheat cultivars, ARE_N of the CRF-60 treatment in Yangzhou and the CRF-80 treatment in Jiangyan were 10.28 and 10.89% higher than that of the CK treatment, respectively.

The total input, fertilizer cost, fertilization labor cost, and net income are listed in

Table 7. Economic evaluation of wheat production with different fertilizer treatments in Yangzhou and Jiangyan (Averages for two wheat varieties over two years).

Site	Fertilization	Total Input	Fertilizer Cost	Fertilization Labor Cost	Other Cost	Net Income	% Change Relative to CK
		(CNY ha−1)
Yangzhou	CK	8649.7	2289.7	600	5760	9355.2 a
	CRF-40	8530.9	2561.9	150	5760	7281.9 c	−22.2%
	CRF-60	8530.9	2561.9	150	5760	9637.6 a	3.0%
	CRF-80	8530.9	2561.9	150	5760	8588.4 b	−8.2%
	CRF-100	8530.9	2561.9	150	5760	6928.1 c	−25.9%
	CRF-120	8530.9	2561.9	150	5760	6301.1 d	−32.6%
Jiangyan	CK	8649.7	2289.7	600	5760	9886.3 a
	CRF-40	8530.9	2561.9	150	5760	8535.5 b	−13.7%
	CRF-60	8530.9	2561.9	150	5760	9619.7 a	−2.7%
	CRF-80	8530.9	2561.9	150	5760	10,213.5 a	3.3%
	CRF-100	8530.9	2561.9	150	5760	8339.5 b	−15.6%
	CRF-120	8530.9	2561.9	150	5760	7357.0 c	−25.6%

Note: As follows is a description of the economic benefits of each treatment: on the basis of the average fertilizer price in China during 2017–2019, the total cost of N, P, K fertilizer is 2289.7 CNY ha⁻¹, the cost of controlled release urea is 0.8 CNY kg⁻¹ higher than regular urea, resulting in a 272.2 CNY hm⁻² higher cost in CRF treatments. The labor cost for each fertilizer application was 150 CNY ha⁻¹. With four applications for CK treatment, fertilizer labor cost was 600 CNY ha⁻¹. The other management costs, including seeds, machinery, pest control, and the rest of labor costs, totaled 5760 CNY ha⁻¹. The price of wheat grain was calculated at 2.2 CNY kg⁻¹. Means followed by different lowercase letters in the same column were significantly different by the Duncan’s test (p < 0.05).

. The fertilizer cost for CRF treatments were higher than CK due to the extra cost of CRU than conventional urea. There was a reduction of 118 CNY ha⁻¹ in total revenue for CRF treatments due to lower fertilization labor costs resulting from the reduced topdressings. CRF-60 treatment in Yangzhou and CRF-60, 80 treatments in Jiangyan can meet the same net income standard as CK. Therefore, the mean annual net income from the CRF-60 treatment in Yangzhou and the CRF-80 treatment in Jiangyan has increased by 3% (282.4 CNY ha⁻¹) and 3.3% (327.2 CNY ha⁻¹) respectively, when compared with the control treatment (Table 6). The one-time application of fertilizer on wheat can obtain greater net returns with a lower labor cost.

4. Discussion

It is generally recognized that wheat has two crucial peaks in its nitrogen requirements: at the seedling stage and after jointing stage, and the demand for the latter is much higher than the former due to the small size of the wheat seedlings before the re-greening stage [13,24]. However, in the middle and lower reaches of the Yangtze River, it is very common to return straw directly to the field in the rice and wheat rotation system, with the dynamic of soil C:N ratio being changed dramatically [33]. Therefore, a large amount of available nitrogen in the soil would be consumed during the decomposition of straw [34]. Consequently, the proportion of basal fertilizer should not be too low in order to ensure that sufficient wheat tillers will sprout before winter in order to build a sufficient population of wheat to produce high yields and endure the winter. Numerous studies conducted in this region have found that basal fertilization with a rate of 33% to 40% of the total N application significantly improves wheat yields and quality [35,36,37]. In this study, 40% of conventional urea was used to blend with CRU, providing N requirements in the early stage of wheat fertility.

Numerous experiments have demonstrated the effectiveness of CRFs in improving crop production agonomically and environmentally [16,20,24,27,38,39]. Differing from rice and maize cultivation, wheat undergoes a long period of decreasing and increasing temperatures during its reproductive period, and the soil water content varies considerably [16,27]. So, CRF formulations suitable for a one-time application must be carefully considered according to the meteorological, soil and other conditions of the region [40]. In this study, the release pattern of the polymer-coated CRU was typical sigmoidal (Figure 2), and a set of CRUs with different controlled release longevities were tested in two experimental sites with distinct soil types. According to the rainfall and temperature data, the inter-annual variation is larger than the inter-location variation. Yet the differences in soil inorganic N content among fertilizer treatments varied by location. Soil texture affects the moisture content and temperature in the soil, as well as the contact between soil particles and fertilizer particles, thus playing a very important role in the release of nutrients from CRF [41,42]. When compared to Yangzhou, the soil particles of Jiangyan were finer, leading to higher water content, and a faster process of water absorption and nutrient release by CRF. Additionally, the soil indigenous fertility in Jiangyan was higher, and soils with more clay particles were able to absorb nitrogen from fertilizer; the soil buffering effect resulted in more moderate profiles of soil inorganic N content in CRF treatments in Jiangyan (Figure 3). N accumulation in wheat occurs continuously and stage-specifically throughout the growth cycle [13,43]. Due to the small size of wheat seedlings, the N uptake was low during the early growth phase, but sufficient N supply is essential in order to cultivate sufficient tillers. In this study, a mixture of 40% conventional urea in CRFs resulted in no significant difference in nitrogen accumulation among treatments at the tillering stage (30 days after seeding). Soils in the CRF treatments with long controlled release longevities (CRF-100 and CRF-120) did not show a shortage of N supply until the regreening stage (104 days after seeding). From jointing to anthesis, nitrogen is obtained in great quantities, and the supply of nitrogen at this stage is therefore critical for wheat growth [13,44]. As shown in this study, N accumulation was highest with the CRF-60 and CRF-80 treatments in Yangzhou and Jiangyan at the anthesis stage, respectively, which was related to the availability of nitrogen in the soil during the same period. The results suggest that the nitrogen release from different CRUs coincided with the second peak in wheat nitrogen requirements at each of the two experimental sites, and that the nitrogen accumulation in wheat was higher than with CK with split fertilization.

CRU has been widely reported to reduce N loss, enhance crop yield, and increase NUE [19]. According to Zheng et al. [24] and Ma et al. [13], the yield of wheat increases by 11.9 and 12.6% respectively. Nevertheless, large yield increases cannot be achieved by relying solely on fertilizer application changes in highly productive fields. Although only 0.5–0.9% relative yield increase compared to conventional four-spilt fertilizer application was achieved in this study, a 9.7–12.1% increase was detected in ARE_N. Nitrogen uptake and utilization capacity of wheat plants increased by the CRF application, while the grain protein content of wheat increased significantly (data not shown). Some researchers suggested that one-time application of CRU cannot gain yield benefit [13,45,46]. Those were probably explained by the poorly matched controlled release fertilizer type or the improper mixing ratio with conventional urea, thus leading to a lack of synchronization between fertilizer N release and wheat N demand. In this study, the yield trend was consistent between fertilizer treatments across wheat cultivars despite the differences in rainfall between years. This demonstrates that under the irrigation conditions of wheat fields in the middle and lower reaches of the Yangtze River, the process of polymer-coated CRU particles absorbing water and releasing nitrogen was generally stable and was not strongly affected by rainfall.

The formation of wheat yields is based on dry matter accumulation, and in particular, the post-anthesis strongly effects the grain yield and harvest index [16,47]. Ma et al. [13] reported a significant correlation (r = 0.9567 **) between post-anthesis dry matter accumulation and yield. The results in this study consistent with these view: wheat of CRF-60 in Yangzhou and CRF-80 in Jiangyan obtained high dry matter accumulation, especially post-anthesis dry matter accumulation, which also performed well in obtaining a high grain yield. Numerous studies have demonstrated that CRU increases the chlorophyll content and photosynthetic activity in wheat flag leaves, which can prolong the photosynthetic functional period of wheat, slow down post-flowering senescence, and increase root growth [42,43,47,48]. In this study, SPAD and Pn values decreased after anthesis in flag leaves. Consistent with the results of previous studies, among the CRF treatments, especially those of long periods of controlled release, SPAD and Pn values were significantly higher than those observed for CK at the milking stage (Figure 5 and Figure 6). It facilitated the accumulation of photosynthetic products after anthesis. Apart from that, dry matter translocation was also important to the construction of high yield. In treatments of CRF-60 in Yangzhou and CRF-80 in Jiangyan, soils provided appropriate N to produce adequate tillers, and wheat stored sufficient dry matter for translocating to grains after anthesis, which also improved the post-anthesis photosynthesis, and promoted the accumulation of photosynthates in wheat grains.

Economic evaluation was very important considering farmers’ interest. The high cost of CRU may be offset by saving the labor cost of top dressing [45,48]. In our study, 118.8 CNY ha⁻¹ was saved by the total input in CRF treatments than conventional cultivation, while the economic return was higher in CRF-60 or CRF-80 treatments due to the higher yield. This led to a 3 and 3.3% higher income relative to CK. Based on our results, a one-time application of the BBU with CRF-60 in a sandy loam area such as Yangzhou, and BBU with CRF-80 in a clay loam area such as Jiangyan, would be a profitable option for high-yield and high-efficiency wheat cultivation in Jiangsu province.

5. Conclusions

With the same N application rate (225 kg ha⁻¹), the one-time application of CRF-60 in Yangzhou and CRF-80 in Jiangyan were preferable choices considering their higher economic incomes, yield and nitrogen use efficiencies. In Yangzhou and Jiangyan experimental sites, CRF-60 and CRF-80 treatments stood out from CRF treatments for providing most synchronous soil available N to fit the two crucial N requirement peaks of wheat. Thus, the wheat in these treatments obtained higher post-anthesis dry matter accumulation, better post-anthesis photosynthesis performance, higher nitrogen absorption and higher grain yield.