Year-Round Production of Cotton and Wheat or Rapeseed Regulated by Different Nitrogen Rates with Crop Straw Returning

Abstract

Double direct seeding of cotton (with wheat or rapeseed) is a new method for cotton-growing regions in the Yangtze River Basin to adapt to the development of mechanization. It would help to reduce manual labor, optimize the amount of nitrogen fertilizer to be used, reduce the physical and chemical production costs, and improve the benefits of cotton fields. We selected five counties from the major cotton-producing areas of Hubei Province for three consecutive seasons, from winter 2020 to spring 2022. The experimental sites used no tillage with straw returning to the field, double direct seeding, late sowing at high density, and one-time fertilization to study the effects of different nitrogen fertilizer rates on the yield characteristics of cotton, wheat, and rape and calculate the economic benefits of the two cultivation modes under different nitrogen fertilizer input levels through parameters such as land-use efficiency, production efficiency, and profitability. In both cotton–wheat and cotton–rapeseed cropping systems, the number of bolls per plant in cotton was the lowest in the N165 (90 cotton + 75 wheat/rape kg ha⁻¹) treatment. The cotton yield was the highest at N247.5 (135 cotton + 112.5 wheat/rape kg ha⁻¹) in the cotton after the wheat system and N412.5 (225 cotton + 187.5 wheat/rape kg ha⁻¹) in the cotton after the rape system. The yield of wheat and rape increased with the increase in the levels of nitrogen fertilizer, with the N165 treatment showing the lowest values. With an increase in nitrogen fertilizer, the harvest index of wheat first maximized and then started decreasing. The harvest index in wheat was the highest at N247.5 (135 cotton + 112.5 wheat/rape kg ha⁻¹) and N330 (180 cotton + 150 wheat/rape kg ha⁻¹), whereas, in rape, it increased with nitrogen fertilizer application, with the highest value at N495 (270 cotton + 225 wheat/rape kg ha⁻¹). Economically, the expenses and income of both cotton–wheat and cotton–rape systems increased as nitrogen fertilizer increased. The net profit and benefit ratio first increased and then decreased with increasing nitrogen fertilizer, with N247.5 (135 cotton + 112.5 wheat/rape kg ha⁻¹) scoring the maximum values for both of these parameters. The land-use efficiency and production efficiency increased with the increase in nitrogen fertilizer, and the production efficiency of the N165 (90 cotton + 75 wheat/rape) treatment was significantly lower than that of the other four treatments. The profitability increased first and then decreased with the increase in nitrogen fertilizer, with the N247.5 (135 cotton + 112.5 wheat/rape) treatment showing the highest profit. The production cycle of cotton–rape was slightly shorter than that of cotton–wheat, and the system productivity was also lower. The expenses and land-use and production efficiency of the rapeseed system were lower than those of wheat, while the gross income, net profit, and productivity of the cotton–rape system were higher than those of cotton–wheat. The application of nitrogen fertilizer in the cotton–wheat double-cropping system under straw return can achieve the maximum net profit, production ratio, and yield at the low nitrogen level of N247.5, (135 cotton + 112.5 wheat/rape kg ha⁻¹). Due to the price advantage of rape, the net profit, production ratio, and income of the cotton–rape production system are higher than those of the cotton–wheat production system.

1. Introduction

Double cropping in cotton fields is a leading cropping system in the Yangtze River region of China. The traditional double-cropping system consists primarily of wheat (rape) and cotton relay intercropping, with cotton planted after wheat (rape). This model improves the multiple cropping index and economic yield by taking advantage of the long intergrowth period of two crops [1], but because the cotton production cycle is more than 220 days and the bolling period is more than 90 days, focusing on maturity is difficult. With the transformation of China’s economic development, the acceleration of urbanization, the relocation of labor, the rising cost, and the decline of international cotton prices, the traditional system of double cropping in cotton fields can no longer meet the requirements of modern agricultural development due to the high labor input in the cotton season and the low degree of mechanization [2,3]. The fully mechanized production method with early-maturing varieties, wheat (rape)–cotton double direct seeding, increasing density and reducing nitrogen, simplifying fertilization, and mechanical harvesting has become the new double-cropping system in China’s Yangtze River Basin [4].

Based on the cultivation strategy of reducing input without reducing output, Yang Guozheng proposed allocating 60% of the total cost to labor, 60% of the material cost to fertilizer, and 60% of the fertilizer cost to nitrogen fertilizer [5]. The new double-cropping system reduces labor input through complete mechanization while increasing production efficiency. Optimizing nitrogen fertilizer input reduces the production system’s materialized input and improves the overall benefits of a cotton field.

Nitrogen is the most important fertilizer for cotton, wheat, and rape. Appropriate nitrogen fertilizer dosage can improve yield while saving money and increasing the gross planting income. Previous studies have shown that inadequate application of nitrogen fertilizer reduces the output and quality of cotton, wheat, and rape; excessive application of nitrogen fertilizer causes cotton to grow too much or rape and wheat to mature too late [6,7,8]. However, research on overall cotton–wheat and cotton–rape double-cropping production systems is limited. The cost, profit, and income ratios determine the effect of nitrogen fertilizer on yield and cotton planting benefits. The selection of the best winter crops for cotton fields and their comprehensive advantages are among farmers’ top priorities. Therefore, we conducted an experimental study on the appropriate nitrogen application rate for cotton–wheat and cotton–rape double-cropping production systems at five locations over three crop seasons in the main cotton-producing areas of Hubei Province. The study can provide theoretical guidance for the production of double-cropping cotton fields in the Yangtze River Basin.

2. Materials and Methods

2.1. Experiment Site and Cultivar

The field experiments were conducted in 2020–2022 with Jinghuamian 116 (G. hirsutum L.), Xiangmai 25 (wheat), and Zhongyouzha 19 (rapeseed) on the five different sites distributed in Hubei Province shown in Figure 1. Cotton repeats every year at five locations; wheat and rape repeat every two years at five locations. Cotton is directly sown by rotary tillage, and wheat and rape are directly sown with no tillage and with total crop straw returning to field. Xiangyang is yellow cinnamon soil, Jingzhou is loam soil, Huanggang is fluvo-aquic soil, Jingmen is sandy soil, and Wuchang is clay.

2.2. Experiment Design

The recommended amount of local nitrogen fertilizer for cotton is 180 kg ha⁻¹ and 150 kg ha⁻¹ for wheat/rape. A randomized complete block design was employed with three replicates. Five nitrogen amounts were 165 kg·ha⁻¹ (N165), 247.5 kg·ha⁻¹ (N247.5), 330 kg·ha⁻¹ (N330), 412.5 kg·ha⁻¹ (N412.5), and 495 kg·ha⁻¹ (N495). The practical allocation of nitrogen fertilizer application amount was as follows: 90 kg·ha⁻¹ (N1), 135 kg·ha⁻¹ (N2), 180 kg·ha⁻¹ (N3), 225 kg·ha⁻¹ (N4), and 270 kg·ha⁻¹ (N5) in cotton season; 75 kg·ha⁻¹ (N1), 112.5 kg·ha⁻¹ (N2), 150 kg ha⁻¹ (N3), 187.5 kg·ha⁻¹ (N4), and 225 kg·ha⁻¹ (N5) in wheat or rape season. These are marked as N165 (90 cotton + 75 wheat/rape), N247.5 (135 cotton + 112.5 wheat/rape), N330 (180 cotton + 150 wheat/rape), N412.5 (225 cotton + 187.5 wheat/rape), and N495 (270 cotton + 225 wheat/rape) in the text.

Fertilizers, as provided by calcium superphosphate (12% P₂O₅) for 54 kg P₂O₅ ha⁻¹, potassium chloride (59% K₂O) for 180 kg K₂O·ha⁻¹, urea (46.3% N) for five amounts, and borate (10% B) for 1.5 kg B ha⁻¹ in cotton and rape season, were mixed evenly and buried 10 cm deep between cotton rows in bed 10–14 days after squaring in cotton season and applied as base fertilizer with seeds when planting in wheat and rape season. The fertilization time in Wuchang, Jingzhou, Jingmen, Huanggang, and Xiangyang was 29 July, 15 July, 14 July, 12 July, and 20 July, respectively, in cotton season.

2.3. Field Management

The cotton plant density was 7.5 × 10⁴ plants ha⁻¹ with a row-to-row space of 76 cm. The plot size was 30.4 m² (10 m × 3.04 m) with four rows in two beds. The sowing time of Wuchang, Huanggang, Jingmen, Jingzhou, and Xiangyang experimental sites was 11 June, 30 May, 1 June, 27 May, and 30 May, respectively, in 2021.

The seed amount of wheat for planting is 225 kg ha⁻¹, and the seed amount of rape is 6.75 kg ha⁻¹. Wheat and rapeseed are directly sowed; the sowing time in Wuchang, Jingmen, Huanggang, Xiangyang, and Jingzhou was 31 October, 3 November, 9 November, 6 November, and 6 November, respectively, in 2020, and the sowing dates in Wuchang, Jingzhou, Jingmen, Huanggang, and Xiangyang were 29 October, 28 October, 24 October, 4 November, and 29 October, respectively, in 2021.

In the winter of 2020–2021, the harvest time of wheat in Wuchang, Jingmen, Huanggang, Xiangyang, and Jingzhou was 18 May, 20 May, 18 May, 22 May, and 18 May. The harvest time of rape in Wuchang, Jingmen, Huanggang, Xiangyang, and Jingzhou experimental sites was 20 May, 17 May, 9 May, 22 May, and 16 May. In the cotton season of 2021, the cotton harvest time in Wuchang, Jingzhou, Jingmen, Huanggang, and Xiangyang experimental sites was 29 October, 28 October, 24 October, 4 November, and 29 October. In the winter of 2020–21, the harvest time of wheat in Wuchang, Jingzhou, Jingmen, Huanggang, and Xiangyang was 15 May, 15 May, 13 May, 15 May, and 20 May. The harvest time of rape in Wuchang, Jingmen, Huanggang, Xiangyang, and Jingzhou experimental sites was 10 May, 13 May, 6 May, 10 May, and 8 May.

Other field management methods were used according to local conventional practice.

2.4. Data Collection

2.4.1. Wheat Yield Traits

At the mature stage of wheat, the whole shoot of wheat up to 1 m² was taken for calculating the number of spikes and spike grains and 1000-grain weight. The harvest index was calculated based on the dry weight and grain weight of 1 m² wheat stalk. The actual yield was calculated based on the actual wheat harvested in the whole plot.

2.4.2. Rapeseed Yield Traits

At the maturity stage, the above-ground part of the whole rapeseed plant was taken up to 1 m² to obtain density, the number of rape pods, the number of grains per pod, and the weight of 1000 grains. The actual yield was calculated based on the actual rapeseed harvested in the whole plot.

2.4.3. Cotton Yield Traits

Fifteen successive and uniform plants in one row from each plot were determined for the investigation of cotton yield traits. According to the survey data on 15 September, the number of bolls was calculated (the number of big bolls + 0.333 × the number of rotten bolls + 0.75 × the number of small bolls). Fifty normal cotton bolls were taken to test the boll weight and lint percentage. The actual yield was calculated based on the actual seed cotton harvested in the whole plot.

2.4.4. Economic and Statistical Analysis

The annual economic benefits of wheat–cotton and rape–cotton were evaluated by land-use efficiency, production efficiency, economic coefficient, and benefit. Land-use efficiency is the ratio of tillage time to the time of the whole year. Production efficiency is determined by dividing the total output of the planting system by the total number of days in a year. Economic coefficient is also called harvest index (HI), which is obtained by dividing the total output per unit area of the planting system by the total dry matter accumulation per unit area. Profitability is obtained by dividing the net income by the cultivation time. The calculation formulas of land-use efficiency, production efficiency, economic coefficient, and benefits are as follows [9,10]:

$$Land use efficiency(\%)=\frac{D1+D2}{365}\times 100$$

D1 and D2 are the total days of wheat (rape) and cotton cropping seasons, respectively.

$$Production efficiency\left(kgh{m}^{-2} da{y}^{-1}\right)=\frac{Y1+Y2}{D1+D2}\times 100$$

Y1 and Y2 are the yields of wheat (rape) and cotton, respectively, and D1 and D2 are the total days of wheat (rape) and cotton cropping seasons, respectively.

$$Harvest Index(\%)=\frac{\mathrm{Yield}}{Biomass}\times 100$$

Benefit–cost ratio is the ratio of gross income to total expenses of wheat (rape) and cotton.

$$Benefit cost ratio=\frac{\mathrm{Gross} \mathrm{income}}{Total expenses}$$

Y1 is the grain dry weight of wheat (rape), Y2 is the cotton seed cotton weight, and Bio1 and Bio2 correspond to the dry weight of the total biomass of wheat (rape) and cotton.

$$Profitability\left(RMB\text{¥} da{y}^{-1}h{m}^{-1}\right)=\frac{\mathrm{Total} \mathrm{net} \mathrm{income} \mathrm{of} \mathrm{cropping} \mathrm{system}}{Total cropping duration}$$

The economic analysis was carried to determine the profitability of cotton, wheat, and rapeseed relay cropping, including all the expenses from sowing to harvesting of both crops. The benefit calculation is based on the Mukherjee formula. In 2021, the purchase price of cotton seed cotton was 9 CNY/kg, the purchase price of wheat was 2.7 CNY/kg, and the purchase price of rapeseed was 7 CNY/kg. The cost accounting basis in the production system is shown in

Table 1. Management and input costs of cotton production under five rates of nitrogen (N) fertilizer between 2020 and 2022 at five different sites in Hubei Province.

Input			Price	Costs CNY·h−1
				Cotton	Wheat	Rape
Fixed cost (labor, herbicide, mechanics, P, K fertilizer)	Labor		100 CNY/d	6750	1500	1500
	P Fertilizer 12% P2O5		0.8 CNY/kg	450	450	450
	K Fertilizer 59% K2O		4.4 CNY/kg	300	300	300
	Herbicide (CNY/hm−2)			2250	150	150
	Seed	Cotton	40 CNY/kg	22.5	225	6.75
		Wheat	10 CNY/kg
		Rape	110 CNY/kg
	Sowing			1500	600	600
	Harvest			2250	900	900
Variable cost (N fertilizer)	N165		3.5 CNY/kg	195	165	165
	N247.5		3.5 CNY/kg	300	240	240
	N330		3.5 CNY/kg	390	330	330
	N412.5		3.5 CNY/kg	495	405	405
	N495		3.5 CNY/kg	600	495	495

. Multi-year multi-point analysis of variance was performed on the data with SPSS 17.0, and Duncan’s multiple comparisons were used for comparison of means. Benefit–cost ratio (BCR) was estimated using prevailing average cost of inputs and value of produce [11,12,13].

3. Results

3.1. Yield Traits of Cotton, Wheat, and Rapeseed

3.1.1. Yield Traits of Cotton after Wheat

Actual density and boll weight did not differ significantly between nitrogen fertilizer treatments, but there were differences in boll number, cotton percentage, seed cotton, and lint yields (

Table 2. Seed cotton yield and yield components of cotton in different rotation cropping systems at different nitrogen rates.

Treatment	Cotton after Wheat				Cotton after Rapeseed
	Bolls per Plant	Boll Weight g	Lint Percentage %	Seed Cotton Yield kg·ha−1	Bolls per Plant	Boll Weight g	Lint Percentage %	Seed Cotton Yield kg·ha−1
Nitrogen
N165	12.6 b	5.8 a	42.3 a	2950.1 b	12.8 b	6.0 a	42.0 a	3183.2 b
N247.5	13.7 a	5.9 a	42.3 a	3378.6 a	14.4 a	5.9 a	41.5 a	3335.3 a
N330	14.5 a	5.9 a	41.5 b	3296.1 a	13.0 ab	6.0 a	41.5 a	3299.0 a
N412.5	14.3 a	5.9 a	41.5 b	3326.3 a	13.5 ab	6.1 a	41.6 a	3351.2 a
N495	14.5 a	5.9 a	41.5 b	3306.3 a	14.2 a	6.1 a	41.3 a	3283.8 a
Location
Huanggang	19.6 a	6.6 a	40.8 c	3821.8 a	17.7 a	6.9 a	40.4 c	3631.1 b
Jingmen	11.1 c	6.3 b	41.0 c	3776.9 a	12.1 b	6.4 b	39.8 c	3906.4 a
Jingzhou	17.0 b	5.7 d	43.5 a	3283.2 b	17.5 a	5.7 c	44.1 a	3322.0 c
Wuchang	11.3 c	4.8 e	41.7 b	2600.9 c	10.5 c	4.8 d	42.2 b	2214.6 d
Xiangyang	10.7 c	5.9 c	42.1 b	2774.7 c	10.1 c	6.4 b	41.5 b	3178.3 c
Source of variation
N×L	**	ns	ns	*	*	ns	ns	*

Note: N: nitrogen treatment, L: location; **, * represent significance at 0.01 or 0.05 probability level, respectively; ns: no significant difference. Different letters represent significance at 0.05 probability level.

). The number of bolls showed no significant difference between treatments at N247.5–N495, while the N165 treatment showed a considerable decrease. Lint content decreased slightly with increasing nitrogen fertilizer, with N165 and N247.5 showing the highest values and no difference between treatments N330–N495. There was no significant difference in seed cotton yield and lint percentage between N247.5–N495 treatments. The yield trend showed a rise followed by a dip as nitrogen fertilizer application increased, even though the differences between treatments N247.5–N495 were non-significant. The N165 treatment resulted in a considerable decrease in seed cotton yield, whereas the N247.5 treatment showed the highest values. Yields and their composition varied significantly among locations. The Huanggang site had the highest number of bolls per plant (19.6), followed by Jingmen (17.0), and Xiangyang had the lowest (10.7). The weight of a single boll varied from 4.8 to 6.6 g, with Huanggang (6.6 g) claiming the first spot and Wuchang (4.8 g) the last. Jingzhou had the highest lint percentage (43.5%), followed by Xiangyang and Wuchang (42.1, 41.7%), and Huanggang and Jingmen had the lowest (40.8, 41.0%). Huanggang and Jingmen had the highest cotton yield, followed by Jingzhou and Wuchang, with Xiangyang having the lowest.

In the wheat–cotton system, the interaction between nitrogen fertilizer and location has a significant impact on the number of bolls per plant, lint percentage, and seed cotton yield but has no significant impact on boll weight.

3.1.2. Yield Traits of Cotton after Rapeseed

There were no significant differences in actual density, boll weight, and lint content with different nitrogen application rates, but the number of bolls per plant showed differential data across treatments (Table 2). The number of bolls was highest at N247.5 and N495, almost equal at N330 and N412.5, and lowest at N165, with irregular variations. Cotton yield was minimum at N165, while the values for N247.5–N495 did not show any statistically significant difference. The variation in seed cotton yield across treatments exhibited no discernible pattern. However, the changing trends of seed cotton yield at different locations were statistically significant. The seed cotton yields in Huanggang first increased and then decreased with the increase in fertilizer amount. The N412.5 and N165 treatments had the highest and the lowest yields, respectively. In Jingmen, there was no statistically significant difference between the N247.5–N495 treatments, but the N165 treatment showed significantly lower values than the other treatments. Jingzhou seed cotton yield first increased and then decreased as fertilizer application increased. The N247.5 treatment had the highest cotton output, while the N412.5 and N495 treatments had the lowest. Wuchang seed cotton yield followed a similar pattern, with no significant differences. Differences between treatments N165–N495 in Xiangyang were also non-significant. In the cotton–rape cropping system, the interaction between nitrogen fertilizer and location had a significant impact on the number of bolls per plant and seed cotton yield but had no significant impact on boll weight and lint percentage.

3.1.3. Wheat Yield Traits

Analysis of the variance of the yield components of each treatment (

Table 3. Grain yield and yield components of wheat and rapeseed at different nitrogen rates.

Treatment	Wheat					Rapeseed
	Spike Number /Ten Thousand·ha−1	Spike Grains	Thousand -Grain Weight/g	Yield /kg·ha−1	HI /%	Plants /Ten Thousand·ha−1	Number of Rape Pods	Number of Grains /Pods	Thousand -Grain Weight/g	Yield /kg·ha−1	HI /%
Year
2020–2021	316.7 b	35.0 b	40.4 a	3747.8 b	46.2 b	47.7 a	120.2 b	22.8 a	3.8 a	1888.6 b	26.9 a
2021–2022	363.6 a	39.6 a	40.6 a	4634.7 a	49.3 a	44.2 b	199.9 a	22.7 a	3.2 b	2163.5 a	27.0 a
Nitrogen
N165	323.3 b	36.8 a	40.5 a	3562.6 c	46.9 b	50.3 a	142.6 b	21.6 b	3.5 a	1671.1 c	26.3 b
N247.5	341.4 ab	36.9 a	40.7 a	4187.1 b	49.8 a	46.0 ab	152.8 b	22.9 a	3.5 a	2011.2 b	26.4 b
N330	338.7 ab	37.6 a	40.4 a	4333.3 ab	49.3 a	45.7 ab	165.4 a	22.9 a	3.5 a	2095.6 ab	27.1 ab
N412.5	349.1 a	37.7 a	40.9 a	4412.1 a	46.6 b	47.2 a	167.8 a	22.9 a	3.6 a	2158.2 ab	27.1 ab
N495	348.2 a	37.6 a	40.1 a	4461.1 a	46.2 b	40.6 b	171.7 a	23.5 a	3.6 a	2194.5 a	27.7 a
Location
Huanggang	288.3 c	32.6 d	41.1 b	4740.8 a	65.9 a	76.9 a	80.8 d	23.7 ab	4.2 a	2047.6 b	28.9 b
Jingmen	278.5 c	32.7 d	37.8 d	4199.0 b	51.8 b	47.5 b	296.9 a	18.6 c	3.7 b	2004.8 b	29.2 b
Jingzhou	478.3 a	42.7 a	41.6 b	3518.9 d	39.8 c	25.4 c	130.4 c	22.9 b	3.5 d	1850.6 c	18.7 d
Wuchang	358.3 b	40.1 b	42.9 a	3765.2 c	39.7 c	27.7 c	142.1 bc	24.8 a	3.9 c	1824.0 c	30.5 a
Xiangyang	297.4 c	38.5 c	39.1 c	4732.2 a	41.5 c	52.2 b	150 b	23.7 ab	2.3 e	2403.4 a	27.5 c
Source of variation
N×*L	ns	ns	ns	**	ns	ns	**	ns	ns	ns	ns
L×Y	**	**	**	**	**	**	**	**	**	**	**
Y×N	ns	ns	ns	**	ns	ns	ns	ns	**	ns	ns
N×L×Y	ns	ns	ns	*	ns	ns	ns	ns	ns	ns	ns

Note: N: nitrogen treatment, Y: year, L: location; **, * represent significance at 0.01 or 0.05 probability level, respectively; ns: no significant difference. Different letters represent significance at 0.05 probability level.

) revealed that the number of spikes, yield, and harvest index were significantly different among the nitrogen fertilizer treatments but showed consistent values for the thousand-grain weight. The number of effective panicles increased with the increasing amount of nitrogen fertilizer applied, with the W5 treatment having the most and the W1 treatment having the least. Both values exceeded 3 million/hm² but were less than the high-yield level across Hubei Province. The number of effective panicles varied more across the five locations. Xiangyang had the smallest value (1.193 million/hm²), while Jingmen had the largest (4.601 million/hm²), followed by Jingzhou, Xiangyang, and Wuchang (4.391 million/hm², 4.259 million/hm², 3.739 million/hm²). The number of grains per panicle increased with the increase in nitrogen fertilizer under the different nitrogen fertilizer treatments, where the W5 treatment was the highest and the W1 treatment was the lowest. The grain yield showed no significant difference between treatments W2–W5, while the W1 treatment was significantly lower than other treatments.

There are notable differences between locations, with Wuchang having the largest (47.6), followed by Xiangyang and Jingzhou (40.4, 39.6), and Huanggang and Jingmen being the smallest (33.4, 33.9). Thousand-grain weights are relatively stable, and there is no significant difference among the treatments, but there are differences among different locations. The Xiangyang trail showed the maximum value (42.3 g), and the Jingzhou point showed the minimum (37.5 g). The actual yield increased first and then decreased with the increase in nitrogen fertilizer; the value at the N412.5 treatment was the highest, and that at the N165 treatment was the lowest. We observed significant differences in values across the different experimental sites. The actual yields of Huanggang, Jingmen, and Xiangyang were similar, followed by Wuchang (4421.4 kg/hm²), and Jingzhou had the lowest output (3122.2 kg/hm²). The theoretical yield and actual yield showed similar trends. The harvest index showed a rising trend followed by a drop with the increasing nitrogen fertilizer amount. The W1 treatment had the smallest harvest index, and the W2 treatment had the largest. The W2–W4 treatments were similar, and the W1 treatment was significantly lower than the other treatments. There were significant differences between locations, with Huanggang (68.2) having the highest harvest index, and Jingzhou and Jingmen (36.8, 37.2) having the lowest.

The interaction between nitrogen fertilizer and location has a significant impact on yield but has no significant impact on the spike number, spike grains, thousand-grain weight, and harvest index. The interaction between nitrogen fertilizer and years has a significant impact on yield but has no significant impact on the spike number, spike grains, thousand-grain weight, and harvest index. The interaction between location and year has a significant impact on the spike number, spike grains, thousand-grain weight, yield, and harvest index. The interaction of nitrogen fertilizer, location, and year has a significant impact on yield but has no significant impact on the spike number, spike grains, thousand-grain weight, and harvest index.

3.1.4. Rapeseed Yield Traits

Except for theoretical yield and thousand-grain weight, there were significant differences among treatments in the number of seedlings, pods, grains, actual yield, and harvest index (Table 3). The variation in the number of seedlings was irregular, with N165 being the greatest and N495 being the smallest. Plant number and grain number showed significant differences with small fluctuations among treatments. The number of grains per pod first increased and then decreased as the nitrogen fertilizer increased, with treatment N165 being the lowest. Treatment N412.5 was the highest, and N165 was significantly lower than the other treatments. Thousand-grain weight increased with increasing nitrogen fertilizer, with N412.5 and N495 having the highest values and N165 being significantly lower than the other four treatments. Theoretical yield initially increased and then decreased with increasing nitrogen fertilizer. The actual yield increased steadily with the application of nitrogen fertilizer, and the low nitrogen level of N165 prompted a significantly lower yield than for the other four treatments.

The interaction between nitrogen fertilizer and location has a significant impact on the number of rape pods but has no significant impact on the number of grains per pod, thousand-grain weight, yield, and harvest index. The interaction between nitrogen ferti-lizer and years has a significant impact on the thousand-grain weight, but there is no significant impact on the number of rape pods, the number of grains per pod, yield, and harvest index. The interaction between location and year has a significant impact on the number of rape pods, the number of grains per pod, thousand-grain weight, yield, and harvest index. The interaction between nitrogen fertilizer, location, and year did not sig-nificantly affect the number of fruits, the number of grains per pod, thousand-grain weight, yield, and harvest index.

3.2. Economic Benefit of Double-Cropping Cotton Field

The economic analysis revealed that the fixed input in the cotton season was CNY 15,330 per hectare under different nitrogen fertilizer treatments. The variable investment depends on the amount of nitrogen fertilizer used, with N165–N495 fertilizer costing CNY 682.5, CNY 1050.0, CNY 1365.0, CNY 1732.5, and CNY 2100.0 per hectare, respectively (

Table 4. Economic analysis of cotton–wheat and cotton–rape rotation cropping system production at different nitrogen rates.

Treatment	Cotton CNY·h−1				Wheat/Rape CNY·ha−1				Cropping System Profitability CNY·ha−1
	Fixed Cost	Variable Cost	Expenses	Gross Income	Fixed Cost	Variable Cost	Expenses	Gross Income	Total Expenses	Gross Income	Net Profit	Benefit–Cost Ratio
Cotton–wheat
N165	15,330.0	682.5	16,012.5	26,550.9	7080.0	577.5	7657.5	9619.0	23,670.0	36,169.9	12,499.9	1.53
N247.5	15,330.0	1050.0	16,380.0	30,407.4	7080.0	840.0	7920.0	11,305.2	24,300.0	41,712.6	17,412.6	1.72
N330	15,330.0	1365.0	16,695.0	29,664.9	7080.0	1155.0	8235.0	11,699.9	24,930.0	41,364.8	16,434.8	1.66
N412.5	15,330.0	1732.5	17,062.5	29,936.7	7080.0	1417.5	8497.5	11,912.7	25,560.0	41,849.4	16,289.4	1.64
N495	15,330.0	2100.0	17,430.0	29,756.7	7080.0	1732.5	8812.5	12,045.0	26,242.5	41,801.7	15,559.2	1.59
Cotton–rape
N165	15,330.0	682.5	16,012.5	28,648.8	5572.5	577.5	6150.0	11,697.7	22,162.5	40,346.5	18,184.0	1.82
N247.5	15,330.0	1050.0	16,380.0	30,017.7	5572.5	840.0	6412.5	14,078.4	22,792.5	44,096.1	21,303.6	1.93
N330	15,330.0	1365.0	16,695.0	29,691.0	5572.5	1155.0	6727.5	14,669.2	23,422.5	44,360.2	20,937.7	1.89
N412.5	15,330.0	1732.5	17,062.5	30,160.8	5572.5	1417.5	6990.0	15,107.4	24,052.5	45,268.2	21,215.7	1.88
N495	15,330.0	2100.0	17,430.0	29,554.2	5572.5	1732.5	7305.0	15,361.5	24,735.0	44,915.7	20,180.7	1.82

). Thus, the total input of N165–N495 was CNY 16,012.5, CNY 16,380.0, CNY 16,695.0, CNY 17,062.5, and CNY 17,430.0 per hectare, respectively. The gross income in N165 was significantly lower than N247.5–N412.5, with the maximum earnings in N247.5 (CNY 30407.4 per hectare). In the wheat and rape season, the fixed input cost of wheat was CNY 7080.0 per hectare, while that of rape was only CNY 5572.5 per hectare. The differentiated input cost for N165–N495 was CNY 577.5, CNY 840.0, CNY 1155.0, CNY 1417.5, and CNY 1732.5 per hectare, respectively. The total input is the sum of variable and fixed costs. The gross income increased with increasing nitrogen fertilizer application, with the N165 treatment showing significantly less income than N247.5–N495. The gross income range in cotton–wheat in N247.5–N495 was CNY 11,305.2–12,045.0, with N165 only accounting for CNY 9619.0. In the cotton–rape system, the income in N165 was CNY 11,697.7 per hectare and ranged from CNY 14,078.4 to 15,361.5 per hectare in N247.5–N495.

The land-use rate, production efficiency, and yield of the cotton–wheat and cotton–rape production systems were also evaluated (Table 4). The total expenses of cotton–wheat increased with the increasing amount of nitrogen fertilizer applied, ranging from CNY 23,670.0 to 26,242.5 per hectare. The total income of the N165 treatment was significantly lower than that of the N247.5–N495 treatments, with the N412.5 treatment providing the largest share. The N165 treatment at CNY 36,169.9 per hectare provided the lowest gross income. The net profit increased initially and then decreased with the increase in nitrogen fertilizer application. The net profit of N247.5 (17,412.6 hm^–2) was the highest, while that of N165 (12,499.9 hm^–2) was the lowest and significantly lower than that of N247.5–N412.5 (15,559.2–17,412.6 hm^–2). The benefit–cost ratio followed a similar pattern to the net profit ratio, with N247.5 (1.72) accounting for the greatest value. In the cotton–rapeseed system, the total expenses increased with the increase in nitrogen fertilizer, ranging from CNY 22,162.0 to 24,735.0 per hectare. The gross income of the N165 treatment was significantly lower than that of the N247.5–N495 treatment, with the N412.5 treatment giving the largest output. The net profit of N247.5 (21,303.6 hm^–2) was the largest, while that of N165 (18,184.0 hm-²) was significantly lower than that of N247.5–N412.5 (20,180.7–21,303.6 hm^–2). The benefit–cost ratio and net profit followed the same pattern, with N247.5 (1.93) having the biggest ratio.

The data (

Table 5. Land-use efficiency, system productivity, and profitability of cotton–wheat and cotton–rape rotation cropping system at different nitrogen rates.

Treatment	Cotton Duration /days	Winter Crop Duration /days	Total Duration of Cropping System /days	System Productivity /kg·ha−1	Land-Use Efficiency /%	Production Efficiency kg·ha−1 day−1	Profitability CNY·day−1
Cotton–wheat
N165	148.0	188.4	336.4	6512.7	92.2	19.4	37.2
N247.5	147.7	188.7	336.4	7565.7	92.2	22.5	51.8
N330	148.5	189.1	337.6	7629.4	92.5	22.6	48.7
N412.5	149.4	190.4	339.8	7738.4	93.1	22.8	47.9
N495	149.3	190.5	339.8	7767.4	93.1	22.9	45.8
Cotton–rape
N165	152.1	179.6	331.7	4854.3	90.9	14.6	55.7
N247.5	152.6	179.8	332.4	5346.5	91.1	16.1	62.6
N330	152.4	180.2	332.6	5394.6	91.1	16.2	61.5
N412.5	153.5	181.1	334.6	5509.4	91.7	16.5	61.0
N495	153.5	181.7	335.2	5478.3	91.8	16.3	59.2

) further revealed that the cotton production cycle of the cotton–wheat system is 148–149.3 days, the wheat production cycle is 188.4–190.5 days, and the total production cycle is 336.4–339.8 days. The system productivity increased consistently with the increase in nitrogen fertilizer application, with the treatment of N165 (6512.7 kg·ha⁻¹) accounting for the lowest value. The land-use rate of N165–N495 showed little difference, ranging from 92.2 to 93.1%, while the production efficiency increased with the increase in nitrogen fertilizer application. Among them, N165 (19.4 kg·ha⁻¹ day⁻¹) was significantly lower than N247.5–N495 (22.5–22.9 kg·ha⁻¹ day⁻¹). The profitability increased initially and then decreased with increasing nitrogen fertilizer application, with the value of N247.5 (51.8 day⁻¹) being the largest and N165 (37.2 day⁻¹) the lowest. In the cotton–rape system, the cotton production cycle lasts 152.1 to 153.5 days, the rape production cycle lasts 179.6 to 181.7 days, and the total production cycle lasts 331.7 to 335.2 days. The productivity of the cotton–rape system increases with the increase in nitrogen fertilizer application, and the treatment of N165 (4854.3 kg·ha⁻¹) is significantly lower than that of N247.5–N495. The land-use efficiency varies from 90.9 to 91.8%. The production efficiency increases with increasing nitrogen fertilizer application, and N165 (14.6 kg ha⁻¹ day⁻¹) is significantly lower than N247.5-N495 (16.1–16.5 kg ha⁻¹ day⁻¹). The profitability first increases and then decreases with increasing nitrogen fertilizer application, N247.5 being the most profitable.

When comparing and analyzing the comprehensive advantages of cotton fields, the production cycle of the cotton–rapeseed system is slightly shorter than that of cotton–wheat, while the seasonal production of cotton is equal in both systems. The total productivity of cotton–rape is lower than that of cotton–wheat, but since the cost of rapeseed is lower than that of wheat, the productivity of the cotton–rape system is higher than that of cotton–wheat.

4. Discussion

Different nitrogen fertilizer applications significantly affected cotton, wheat, and rape season yields in the cotton–wheat and cotton–rape production systems. Late sowing, dense seeding, and a single nitrogen fertilizer application during the cotton season at the N247.5–N495 level resulted in a stable yield; however, a further reduction in nitrogen treatment (N165) significantly affected the cotton season yield. The appropriate nitrogen fertilizer dosage for traditional spring sowing and transplanting cotton in the Yangtze River Basin is 300 kg·ha⁻¹, which can be reduced to 225 kg·ha⁻¹ with late sowing and high planting density without affecting the yield [14,15]. The optimum nitrogen application rate for cotton grown under continuous cotton–rape cropping was between 250 and 270 kg·ha⁻¹ [16], while for direct seeding cotton after rape, it was 228 kg·ha⁻¹. The apparent utilization rate, partial productivity, and agronomic utilization rate of nitrogen decreased significantly with the increase in nitrogen application rate [17]. A study on a one-time fertilization mode at the flowering stage revealed that under the high density of late sowing in cotton fields with medium fertility, reducing the amount of nitrogen fertilizer to 180 kg/hm² had no adverse effect on yield and quality traits [18]. The field experiment demonstrates that a nitrogen fertilizer rate of 120–240 kg·ha⁻¹ in summer direct-sowing cotton results in a stable yield [19]. In our study, both cotton–wheat and cotton–rapeseed production models allowed a reduction in nitrogen fertilizer of up to 135 kg·ha⁻¹ without any significant change in cotton yield, but further reducing the nitrogen fertilizer to 90 kg·ha⁻¹ resulted in a decrease in cotton yield.

However, wheat yields increased as the nitrogen fertilizer rate increased under late sowing and no-tilling conditions with the return of straw to the field. Earlier, wheat–maize research revealed that returning straw has no significant effect on wheat yield, but the amount of nitrogen fertilizer significantly affects the wheat yield. With increasing nitrogen fertilizer application, wheat yield first increases and then decreases [20], thus making it necessary to consider the risk of lodging and yield reduction with extremely high nitrogen fertilizer application. The optimum nitrogen application amount for local winter wheat is 168 kg·ha⁻¹ [21,22]. Rapeseed yield also increased with the increase in nitrogen application when the density of late sowing was high with no tillage and return of straw to the field. Increasing nitrogen fertilizer can significantly increase rapeseed grain yield, but the rate will slow down, which is consistent with previous research [23]. Nitrogen fertilizer can significantly increase the protein content in rapeseed while reducing the oil content, so different nitrogen fertilizer rates are required in rapeseed for specific purposes [24]. The interaction among nitrogen, location, and year has different impacts on the yield traits of cotton, wheat, and rapeseed; it especially showed significant differences in yield between different locations and years. Because the lint percentage and boll weight are the characteristics of cotton varieties, the interaction between nitrogen fertilizer and the environment has no significant impact on the two yield traits but has a more significant impact on boll number. The yield differences between different locations mainly come from the significant impact of the interaction between nitrogen fertilizer and location on boll number. The interaction between year and location has a significant impact on wheat yield and its component factors, while the interaction between nitrogen fertilizer and location, as well as the interaction between nitrogen fertilizer, year, and the three, only have a significant impact on yield traits. Similarly, the interaction between year and location has a significant impact on the yield and its components of rapeseed. The interaction between nitrogen fertilizer and location only has a significant impact on the number of rape pods, while the interaction between nitrogen and year has a significant impact on the thousand-grain weight of rapeseed. The interaction of the three has no significant impact on the yield traits of rapeseed. Although yield increased with increasing nitrogen fertilizer content in the wheat and rape seasons, after considering the production cost and benefit analysis, the net profit and output ratio of the production system first increased and then decreased with increasing nitrogen fertilizer application, with the N247.5 treatment giving the highest profit and the N165 treatment being equivalent to the N495 treatment. Considering the effects of different nitrogen fertilizer treatments on the benefits of cotton fields based on the land-use rate, production efficiency, and yield of the cotton–wheat and cotton–rape systems, we found that the land-use rate and production efficiency of the cotton–wheat and cotton–rape systems increase with increasing nitrogen application. Further, the production efficiency of both cropping systems at N165 was significantly lower than at N247.5–N495, and the production efficiency of the cotton–rape system at N412.5 was the highest (16.5%). Analyzing the economic advantages of the cotton–rape and cotton–wheat systems at the N247.5 nitrogen level revealed that the yield of rapeseed (2011.2 kg ha⁻¹) was lower than wheat (4187.1 kg ha⁻¹). The corresponding production efficiency of the cotton–rape system was 39.8% lower than that of cotton–wheat, even though the net profit in the cotton–rape system was higher. Cotton–rape had a total output corresponding to 44096.1 ha⁻¹ while that of cotton–wheat was 41,712.6 ha⁻¹; however, the corresponding net profit, output ratio, and yield of cotton–rape are, respectively, 22.3%, 12.2%, and 20.8% higher than cotton–wheat. The main reason is that the seed cost of wheat is higher than that of rape, but the product price is lower. Therefore, it is necessary to consider the cost and benefits comprehensively when making planting decisions.

The object of this study is the production systems of cotton–wheat and cotton–rapeseed. The law of diminishing returns explains from the perspective of cotton field benefits that nitrogen input can achieve the highest economic benefits at a lower level of N247.5 (135 cotton + 112.5 wheat/rape), which provides reference for making decisions on double-cropping production in cotton fields. However, due to the continuous increase tendency of rapeseed and wheat yield with an increasing nitrogen fertilizer rate in this production system, which is inconsistent with the law of diminishing returns, further in-depth research is needed on this phenomenon.

5. Conclusions

Double direct seeding of cotton–wheat and cotton–rape is a new model to adapt to the development of mechanization in the cotton region of the Yangtze River Basin in China. This method helps to reduce labor input, optimize the amount of nitrogen fertilizer, further lower the materialized cost of the production system, and improve the benefit of a cotton field. This study analyzes the comprehensive benefits of cotton–rape and cotton–wheat double-cropping systems from the perspective of a cotton field. Although there was no significant difference in seed cotton yield between the N247.5–N495 treatments, the seed cotton yield under N165 was significantly lower than that under the N247.5–N495 treatments. The yield of rape and wheat increased with the increase in nitrogen application, as did the total output, land-use rate, and production efficiency in both cropping systems. Environmental factors have a significant impact on the yield of cotton, wheat, and rapeseed, which are variation among different locations due to differences in soil and climate, but the interaction among the environment, year, and nitrogen has no significant impact on yield. However, the net profit, output ratio, and yield increased first and then decreased, with the N247.5 treatment exhibiting the highest levels for these metrics. A comparison of the cotton–wheat and cotton–rape cropping systems based on current seed and product prices revealed that the comprehensive benefits of cotton–rape outweigh those of cotton–wheat. Planting decisions, therefore, must consider the cost of seeds and fertilizers and the market price of products to achieve better overall benefits for cotton fields.