Analysis and Closing of the High-Production-Maize Yield Gap in the Semi-Arid Area of Northeast China

Abstract

A mulched drip fertigation system is an effective way to improve maize yield, but at present, the efficiency of nutrient delivery and water use are generally low. Therefore, this study conducted optimization field experiments to identify the main factors limiting the delivery of water and fertilizer, including regulations on nitrogen (N) fertilizer, irrigation conditions, planting density and maize varieties, in the semi-arid area of Northeast China. As part of an effort to close the yield gap for maize, an optimized system (DS) for optimal crop, nutrient and water management was designed to improve the agronomic and economic performance of maize farming in the area. The application rate of N fertilizer was 240 kg ha⁻¹; the base fertilizer was applied four times (once at the jointing stage, twice at the belling stage and once at the silking stage); the rates of application of phosphorus (P) and potassium (K) fertilizer were 90 kg P₂O₅ ha⁻¹ and 90 kg K₂O ha⁻¹, respectively; the irrigation amount was 270 mm ha⁻¹; the maize variety Fumin 985 was planted at a density of 80,000 plants ha⁻¹ in DS; the grain yield of DS reached 13.8 Mg ha⁻¹, 93% of the yield potential. DS yielded an economic benefit of 18,449 yuan ha⁻¹, which was significantly higher than the economic benefit of 13,818 yuan ha⁻¹ achieved under farmers’ practices (FP). Furthermore, the utilization rates of N, P, K, and water were significantly improved under DS. In conclusion, DS increased production potential, with high efficiency in nutrient delivery and water use and low losses of nutrients and water. The crop, fertilizer, and water management of DS provided a technological system to simultaneously improve crop production and resource-use efficiency in the semi-arid area of Northeast China.

1. Introduction

Maize is the food crop with the highest yield in the world and the largest planting area in China, accounting for about 30 percent of the total planting area in Northeast China, and it plays a crucial role in agricultural production [1,2,3]. At present, there are three main factors restricting the yield of maize in China, one of which is the shortage of agricultural resources. The production conditions of corn have not been improved. Water shortage is a critical matter in agriculture in China; hence, most of China’s maize production relies on rain-feeding systems, and even in areas with irrigation, the rate of utilization of water resources is very low due to the use of outdated technology, among other reasons [4]. Another issue is that germplasm resources are scarce, and excellent varieties with high yield, high quality, multiple resistance and wide applicability are lacking. One of the main technical means by which to increase yield of maize is to increase the planting density, but there is a lack of density-tolerant varieties in China [5]. The third is unreasonable application of fertilizer. Excessive and insufficient fertilizer application coexist, which at once limits increases in maize yield, leads to a low rate of fertilizer utilization, and causes environmental hazards [6].

In recent years, international attention has been paid to the study of differences in crop yield. At present, the potential yield of maize per unit area in the world is 10.4 t ha⁻¹. However, the actual yield of maize is only 5.5 t ha⁻¹, and climate, nutrients, water, and crop variety are the most commonly cited yield-limiting factors [7]. Meng et al. (2013) analyzed the yield differences across China’s main maize-producing areas and concluded that there is still substantial room for improvement in our current maize production [8]. Closing the yield gap (the difference between yield potential and the actual output from farmers) can greatly increase food production. Yield potential refers to the maximum yield that can be achieved in a specific area through optimal management and without stresses such as water, nutrients, pests, and diseases [9]. In order to improve crop yield, it is usually necessary to increase investment in agricultural resources in production, but the utilization rate of resources by crops is usually low. There have been studies showing that optimizing management of water and fertilizer, choice of varieties, and cultivation measures are usually effective in increasing crop yield. For instance, in the North China Plain, the N application rate was controlled at 113–180 kg N ha⁻¹, and the maize yield reached 12,000 kg ha⁻¹ [10]. In western Liaoning, the maize yield of Liaodan 565 reached 11,980 kg ha⁻¹, while that of Zhongnong 4 was only 5692 kg ha⁻¹ [11]. In Fufeng Shanxi Province, China, under the same conditions of N fertilizer application, when the Zhengdan 958 variety was planted, the grain yield of maize with a planting density of 75,000 plants∙ha⁻¹ reached 10,177 kg ha⁻¹, while a planting density of 45,000 plants∙ha⁻¹ yielded 8769 kg ha⁻¹ [12].

Traditional water- and fertilizer-management measures such as flood irrigation and one-time basal fertilizer application not only cause serious waste of resources, but also limit the increase in crop yield [13]. In recent years, mulched drip fertigation systems (drip irrigation under mulch) have been widely used due to their advantages of simultaneously increasing crop yield and improving resource utilization [14]. Mulching can reduce water evaporation and evapotranspiration, reduce soil erosion, and increase soil temperature for early spring maize, thereby increasing crop yield and water-use efficiency [15]. Drip irrigation and fertilizer application combine water and nutrients and apply them simultaneously to the root layer of crops, reducing water and nutrient waste and significantly improving utilization efficiency [16]. Studies have shown that mulched drip fertigation systems can significantly increase crop yields in water-limited areas with rainfall less than 400 mm [17]. However, the current mulched drip fertigation system has not fully utilized the potential of this technology to increase production and efficiency. The maximum grain yield cannot be obtained by blindly increasing N fertilizer in a mulched drip fertigation system. Under water-stress conditions, high N content inhibits the growth and development of maize, resulting in low efficiency of water and fertilizer use [18]. There is a threshold of irrigation and fertilizer use in mulched drip fertigation systems; after the threshold is reached, an increase in irrigation and fertilizer application has an inhibitory effect on crop growth [19]. In years with abundant precipitation, the advantage of drip fertilizer-application technology is not highly significant, and its high cost was difficult to balance through improvements in maize yield [20]. Non-optimal use of nutrients and water limits crop-yield potential and resource-use efficiency. Therefore, it is necessary to design optimized management practices for crops, nutrients, and irrigation om mulched drip fertigation systems, addressing production and efficiency issues. Synergistically improving crop yield and nutrient-use efficiency is an inevitable requirement for the development of mulched drip fertigation systems.

Therefore, this study carried out optimization experiments focused on N fertilizer regulation, irrigation conditions, planting density, and maize varieties. The best conditions provide technical support for the construction of a high-yield and high-efficiency mulched drip fertigation systems. Field experiments were carried out using the optimized technical indicators of each factor to develop a high-yield and efficient system for maize planting in mulched drip fertigation systems. The optimized planting system and the traditional water and fertilizer integrated planting system used by farmers were compared by comprehensively evaluating the maize yield, economic benefits, and overall efficiency of nutrient utilization. The purpose of this study was as follows: to demonstrate an improved maize- and N fertilizer-management system to improve maize yield and N-use efficiency; to demonstrate a comprehensive mechanism that optimizes nutrient, water, and crop management; and to propose a comprehensive design system for evaluating maize yields, economic benefits, and nutrient-use efficiency of the novel system compared with farmers’ practices in the region.

2. Materials and Methods

2.1. Experimental Site

The field experiment was conducted in Yangjia Village (45°17′ N, 124°87′ E) (Dawa Town, Ningjiang District, Songyuan City, Jilin Province, China) in 2020–2021. Yangjia Village has a temperate continental monsoon climate, with cold winters and hot summers, and a frost-free period of 135–145 days. The average temperature of the maize growing season in this area is 17–21 °C, and the annual rainfall is about 355 mm. Most of the rainfall occurs in the peak growing period of maize (May to September). The climate conditions (sunshine duration, precipitation, etc.) did not differ significantly between the two years of the experiment, and there were no natural disasters such as wind and insect disasters. The soil at the experiment site is chernozem sandy loam; its physicochemical properties which are shown in Supplementary Table S1.

2.2. Experimental Design

2.2.1. N Application Rate Experiment

The experiment on N application rates was conducted over two consecutive years, from 2020–2021. There were 5 treatments: N0, N90, N150, N210, N270, and the N application rates were 0, 90 kg ha⁻¹, 150 kg ha⁻¹, 210 kg ha⁻¹, and 270 kg ha⁻¹, respectively. The maize variety was Fumin 985, and the planting density was 80,000 plants ha⁻¹. 30% N fertilizer was applied with base fertilizer, and the remaining N fertilizer was applied at levels of 30%:20%:10%:10% in the jointing stage, belling stage, silking stage, and grain-filling stage, respectively. Additionally, 50% phosphorus (P) and potassium (K) fertilizer were applied with base fertilizer, and the remaining 50% was applied with topdressing at the jointing stage. The application rates of the P and K fertilizers were the same for each treatment: 90 kg P₂O₅ ha⁻¹, and 100 kg K₂O ha⁻¹. The N, P, and K fertilizers were applied as urea, calcium superphosphate, and potassium sulfate, respectively. The amount of irrigation water was set according to the water demand for maize production: 20 mm of water was applied at the stage before sowing, the ten-leaf stage and silking stage, and 70 mm of water was applied at the six-leaf, silking, and grain-filling stages. Each plot had a separate drip-irrigation system that controlled fertilizer application and irrigation individually. The drip-irrigation pipeline was laid in the middle of the double row of the ridge. The inner diameter of the dripper belt was 16 mm, and the distance between the drippers was 30 cm. An 18-L container was installed behind the pressure-control valve, and the amount and time of each irrigation and fertilizer application were controlled by adjusting the switch of the valve.

2.2.2. N Fertilizer Management Experiment

The N-management experiment was conducted in 2020, with 4 treatments: N1, N2, N3, and N4, in which N fertilizer was applied twice, three times, four times, and five times, respectively. The specific fertilizer-application measures for each treatment in the experiment are shown in Supplementary Table S2. The amounts of N, P, and K fertilizers used in each treatment were the same across treatments (210 kg N ha⁻¹, 90 kg P₂O₅ ha⁻¹, and 100 kg K₂O ha⁻¹). The base fertilizer was applied in a strip-application method before sowing, and the integrated drip-irrigation technology with water and fertilizer was adopted for top dressing. The N fertilizer was urea (N46%); the P fertilizer was water-soluble mono ammonium phosphate (N-P₂O₅-K₂O:12-61-0); and the K fertilizer was water-soluble potassium chloride (K₂O 60%). Irrigation measures were the same as those used in the N application-rates experiment.

2.2.3. Maize-Planting-Density Optimization Experiment

The maize planting-density-optimization experiment was carried out in 2020, with 5 density treatments: 60,000, 70,000, 80,000, 90,000, and 100,000 plants ha⁻¹. The tested maize varieties were Fumin 985, Fumin 108, Youdi 919, and Xianyu 335. Fertilizer application and irrigation measures were the same as those used in N application-rates experiment.

2.2.4. Experiment for Verification of the Design System

The experiment used for verification of the design system incorporated two treatments: the optimized design system (DS) and farmers’ practices (FP). Under the optimized design system, nutrient- and water-management measures were chosen based on target yield and crop demand. Maize varieties and planting densities were based on the results of the maize planting-density optimization experiment. In the farmers’ practices treatment, maize variety, planting density, nutrient application, and water-management practices were determined by majority answers in survey results.

The plot areas for the N application-rates experiment, the N-fertilizer-management experiment, and the maize planting-density-optimization experiment were 30 m², and that for the verification experiment was 200 m², with three replicates arranged in random blocks. Other field management was carried out according to the production field. The experiment adopted the double-row cultivation mode with large ridges, and artificial mulching was used after fertilizer application and sowing. The drip-irrigation pipeline was laid in the middle of the double row of the ridge; the inner diameter of the dripper belt was 16 mm; and the distance between the drippers was 30 cm. Planting was carried out when the ground temperature of the top 5 cm of soil was stable and at least 8 °C. Wide and narrow rows were used for planting. After heavy compaction, a 2BJHM-2 multi-functional water-saving film-mulching seeder was used to complete fertilizer application, sowing, laying of drip-irrigation pipes, spraying herbicides, and applying the covering film and covering soil. Underground well water was used as the water source, and an independent fertilizer-application tank was connected to each plot. During fertilizing, the amount of fertilizer required by each plot was added to the appropriate fertilizer-application tanks. The fertilizer-application tank was filled with drip water, and the amount of fertilizer to be used was added and stirred thoroughly to dissolve it completely. The lid of the tank was fastened, and the tank was allowed to sit for about 30 min before fertilizer application. No obvious weeds, diseases, or insect pests were observed during the experiment.

2.3. Sample Collection and Analysis

During the jointing stage, silking stage, grain-filling stage, and mature stage of maize growth, 4 adjacent plants with uniform growth were sampled. Samples were taken on the 7th day after irrigation/fertilizer application. The aboveground parts were dried, weighed, and then ground into powder to calculate the aboveground biomass and nutrient content. The samples from the silking stage and the mature stage were divided into three parts: leaves, stalks, and grains. The plant samples were all placed in an oven at 75 °C for drying and weighing, then pulverized to measure the plant N content by the Kjeldahl method. During the harvest period, the harvest area of each yield plot was 12 m², and the yield was converted to the standard yield with 14% moisture content. At the same time, the harvest density, number of ears, number of grains per ear, and thousand-kernel weight were measured at each point of harvest.

2.4. Computational Methods and Data Analysis

Harvest index = grain yield/aboveground biomass × 100% [21];

N partial-factor productivity (NPFP, kg kg⁻¹) = grain yield in N-application plots/N rate [21];

P partial-factor productivity (PPFP, kg kg⁻¹) = grain yield in P-application plots/P rate [22];

K partial-factor productivity (KPFP, kg kg⁻¹) = grain yield in K-application plots/K rate [22];

Agronomic N efficiency (ANE, kg kg⁻¹) = (grain yield in the N-applied plots -grain yield in the N-free plots)/N fertilizer application rate [21];

Recovery N efficiency (RNE, %) = (crop N uptake in N-applied plots- crop N uptake in N-free plots)/(N fertilizer application rate × 100) [21];

Water productivity (WP, kg ha⁻¹ mm⁻¹) = grain yield/irrigation amount [21].

All these experimental data were statistically analyzed using Excel 2010. SAS 9.4 software was used for two-factor statistical analysis. The least-significant-difference method was used for significance comparisons between treatments by the F test in one-way ANOVA, and multiple comparisons of means (p < 0.05) were conducted with a Fisher’s protected least-significant-difference (LSD) test. Origin software version 8.5 was used to create the graphs.

3. Results and Analysis

3.1. Effects of Different N Application Rates on Maize Yield and Yield Factors

N fertilizer application significantly affected maize yield, number of grains per ear, thousand-kernel weight, and harvest index (

Table 1. Effects of different N fertilizer application rates on metrics of maize yield.

Year	Treatments	Grain Per Ear (No.)	Thousand-Kernel Weight (g)	Yield (Mg∙ha−1)	Harvest Index
2020	N0	382 c	287 b	7.6 c	0.39 b
	N90	495 b	295 a	10.9 b	0.43 a
	N150	501 b	297 a	11.7 a	0.42 a
	N210	571 a	301 a	12.6 a	0.43 a
	N270	539 a	317 a	12.5 a	0.43 a
2021	N0	319 c	253 c	6.9 c	0.40 b
	N90	497 b	269 bc	10.2 b	0.45 a
	N150	549 a	278 ab	11.9 a	0.44 a
	N210	569 a	299 a	12.9 a	0.48 a
	N270	559 a	309 a	13.1 a	0.47 a
Source of variation
Year (Y)		**	**	**	NS
Fertilizer N rate (N)		**	**	**	**
Y × N		**	NS	NS	NS

Values are means. Different letters in the same column indicate significant differences among samples (p < 0.05, ANOVA). ** indicates significance at p < 0.01; NS: not significant. N0, N90, N150, N210, and N270 represent N fertilizer used at levels of 0, 90, 150, 210, and 270 kg ha⁻¹, respectively.

). Compared with N0, N application increased the yield by 43–66% in 2020 and 48–90% and 2021 (p < 0.05). The N210 treatment was associated with the highest maize yield. There was no corresponding increase in maize yields even with an increase in the N application rate (N270) in 2020, while in 2021, the yield was higher in the N270 treatment compared to the N210 treatment. Compared with N0, the application of N fertilizer increased the number of grains per ear and the thousand-kernel weight. The interaction between N application rate and year had no significant effect on thousand-kernel weight, maize yield, or harvest index. The results indicate that the effect of N fertilizer application on yield and harvest index had similar effects in different years. The data were integrated into quadratic-equation simulation with one variable, and the results showed that the total N application rate for the highest yield in 2020 was 238.5 kg ha⁻¹ and that the associated maize yield was 12.6 Mg ha⁻¹ (Figure 1). Additionally, the N application rate was about 271.3 kg ha⁻¹; the highest maize yield was 13.1 Mg ha⁻¹ in 2021; and the two-year-average optimal N application rate was 254.9 kg ha⁻¹.

3.2. Effects of Different N Application Rates on Dry-Matter Accumulation and N Uptake by Plants

Both the N application rate and the interannual period had a significant effect on the dry-matter accumulation of maize at each growth stage (

Table 2. Effects of N fertilizer application and year on maize shoot dry matter.

Source of Variation	Shoot Dry-Matter Weight
	Jointing	Belling	Silking	Filling	Mature
Year (Y)	**	**	*	**	**
Fertilizer N rate (N)	**	**	**	**	**
Y × N	*	NS	*	**	NS

* indicates significance at p < 0.05; ** indicates significance at p < 0.01; NS: not significant.

). The interaction between year and N application rate had a significant effect on dry-matter accumulation at the jointing stage, silking stage, and grain-filling stage. The aboveground-biomass accumulation under N-application treatments was significantly higher than used the no-N-application treatment (Figure S1). Compared with N0, the biomass of plants in the N-application treatments increased by 14.7–54.7% and 45.1–108.4% in 2020 and 2021, respectively. The dry-matter accumulation of maize increased rapidly after the jointing stage and slowed down after the grain-filling stage (Figure S1). When the N application rate was increased above 210 kg∙ha⁻¹, dry-matter accumulation also increased. However, when the N application rate increased to 270 kg ha⁻¹, dry-matter accumulation remained stable. Applying N fertilizer can improve the accumulation of dry matter in maize, but the over-application of N fertilizer was not conducive to increasing the accumulation of dry matter. The amount of N uptake by maize was significantly affected by the amount of N fertilizer applied (Figure S2) (p < 0.05). The two-year-average results showed that the N uptake of the N210 and N270 treatments was significantly greater than those of the N90 and N150 treatments at the silking and mature stages. The N accumulation in maize at the different growth stages increased with the increase in N application rate; however, when the N application rate exceeded 210 kg ha⁻¹, the plant N accumulation did not increase further.

3.3. N Use Efficiency under the Different N Application Rates

The rate of N recycling by season was significantly affected by inter-annual variation, N application rate, and the interaction between the inter-annual variation and N application rate (

Table 3. Effects of N fertilizer application rate on N use efficiency.

Year	Treatment	PFPN (kg∙kg−1)	AEN (kg∙kg−1)	REN (%)
2020	N90	121.1 a	36.7 a	51.1 a
	N150	78.0 b	27.3 b	48.3 b
	N210	60.1 c	23.8 b	46.1 b
	N270	46.3 d	18.1 c	33.7 c
2021	N90	113.3 a	36.7 a	63.5 b
	N150	79.3 b	33.3 a	81.6 a
	N210	61.4 c	28.6 b	78.4 a
	N270	48.5 d	23.0 c	62.9 b
Year (Y)		NS	NS	**
Fertilizer N rate (N)		**	**	**
Y × N		NS	NS	**

Values are means, different letters in the same column indicate significant differences among samples (p < 0.05, ANOVA). ** indicates significance at p < 0.01; NS: Not significant. PFPN, N partial-factor productivity; ANE, Agronomic N efficiency; RNE, Recovery N efficiency. N0, N90, N150, N210, and N270 represent N fertilizer being used at a level of 0, 90, 150, 210, and 270 kg ha⁻¹, respectively.

). Moreover, the N partial-factor productivity (NPFP) and agronomic N efficiency (AEN) were significantly affected by the N application rate. With the increase in the N application rate, the NPFP, AEN, and the recovery N efficiency (REN) each season showed a significant downward trend, and the NPFP dropped from 121.1 kg kg⁻¹ to 46.3 kg kg⁻¹ (2020) and from 113.3 kg kg⁻¹ to 48.5 kg kg⁻¹ (2021). The AEN dropped from 36.7 kg kg⁻¹ to 18.1 kg kg⁻¹ (2020) and from 36.7 kg kg⁻¹ to 23.0 kg kg⁻¹ (2021). There was no significant difference between the N150 treatment and the N210 treatment in the two-year-average REN, but the value was significantly higher than that in the N270 treatment (p < 0.05).

3.4. Effects of Different N Fertilizer Application Frequency on Maize Yield, Yield Component Factors, and Economic Benefit Analysis

When the same amount of N was applied, different application frequencies had significantly different effects on maize yield (

Table 4. Effects of frequency of N fertilizer application on maize yield metrics.

Treatment	Grain Per Ear (No.)	Thousand-Kernel Weight (g)	Yield (Mg∙ha−1)
N1	434.1 b	31.5 a	10.4 b
N2	485.6 a	32.5 a	12.4 a
N3	479.1 a	31.9 a	11.9 a
N4	473.5 a	32.0 a	11.7 a

Values are means; different letters in the same column indicate significant differences among samples (p < 0.05, ANOVA). N1, N2, N3, and N4 represent N fertilizer topdressing carried out twice, three times, four times, and five times, respectively.

). Compared with the yield from N1, the maize yield in N2, N3, and N4 increased significantly, by 12.5–19.2%. Among these treatments, the N2 treatment had the highest maize yield. The maize yield had a downward trend with increasing fertilizer application frequencies (N3, N4 treatments). In addition, the different N fertilizer application frequencies affected the number of grains per spike. Compared with the N1 treatment, the N2, N3, and N4 treatments had significantly increased grain numbers per ear, with the N2 treatment having the greatest number, but there was no significant difference between the N3 and N4 treatments. As a result, N2 was noted as representing the optimal fertilizer application frequency to increase the number of grains per ear, thereby increasing the grain yield. The difference in economic cost for the different fertilizer-application frequencies was the result of water, electricity, and labor costs (

Table 5. Effects of different frequencies of N fertilizer application on economic benefits (Yuan).

Treatment	Fertilizer	Drip Irrigation Equipment	Water, Electricity and Labor	Machinery, Seed and Pesticide	Maize Output	Net Return
N1	2050	1350	650	3630	20,826	13,146
N2	2050	1350	760	3630	24,878	17,088
N3	2050	1350	890	3630	23,854	15,934
N4	2050	1350	1060	3630	23,436	15,346

N1, N2, N3, and N4 represent N fertilizer topdressing carried out twice, three times, four times, and five times, respectively.

). N2 had the greatest economic benefit because it was associated with lower economic input and the highest maize yield, and its net benefit was as high as 1154–3942 yuan ha⁻¹.

3.5. Effects of Different Varieties and Densities on Maize Yield

Fumin 985, Fumin 918, and Youdi 919 produced yields that were significantly higher than those of Xianyu 335 (Figure 2). The planting density had a significant impact on the maize yield, and the maize yield was the highest when the planting density was 70,000 plants ha⁻¹. Increasing the planting density above 70,000 plants ha⁻¹ was associated with a decrease in maize yield. The results showed that the highest yield can be obtained when the two-year-average application rate of N fertilizer for two years was 254.9 kg ha⁻¹ and that carrying out topdressing three times yielded the highest economic benefits. Fumin 985, Fumin 918, and Youdi 919 performed the best in terms of yield, and the planting density of 70,000 plants ha⁻¹ resulted in the highest yield. Based on the above results, a high-yield system was designed.

3.6. The Optimized Design System of Maize High-yielding under Mulched Drip Fertigation Systems

Based on the above analysis (Section 3.5), optimizing N fertilizer management (N application rate and frequency) and optimizing the choice of varieties and planting density could effectively increase maize yield. Therefore, the DS was established to integrate the above findings in field verification tests. The N application rate for the DS was set at 240 kg ha⁻¹, and according to the N demand established for high-yield maize, the fertilizer was applied as follows: basal fertilizer, 70 kg; jointing stage, 50 kg; belling stage, 70 kg; silking stage, 50 kg. According to the amounts of P and K recommended for the region [23], the application rates of P fertilizer and K fertilizer were set to 90 kg P₂O₅ ha⁻¹ and 90 kg K₂O ha⁻¹, respectively, and the application ratio was set at 60%:40%, to be applied as base fertilizer and in the belling stage, respectively. The water consumption of maize for a yield of 12 Mg ha⁻¹ in Northeast China was 450–550 mm [24]. Based on the local average rainfall of 250–350 mm, the final optimal irrigation amount was set to 270 mm ha⁻¹. The maize jointing stage and flowering stage constitute the most period of highest water demand for crops [25]; thus, the crops were irrigated with 70 mm ha⁻¹ in those two periods. In addition, 40, 30, 30, and 30 mm ha⁻¹ water were applied, respectively, after the sowing stage, after the belling stage, 10 days after the silking stage, and after the grain-filling stage,. The local high-yield variety, Fumin 985, was selected as the maize variety. The planting density was finally set at 80,000 plants∙ha⁻¹ after comprehensive consideration of both the potential to maximize yield and lodging risk.

The fertilizer application, irrigation measures, maize varieties, and planting density practiced by farmers were determined by the general results from a local survey [26]. The amount of N applied by farmers was 270 kg ha⁻¹ and was normally applied as follows: basal fertilizer, 135 kg; jointing stage, 81 kg; belling stage, 54 kg. Additionally 100 kg P₂O₅ ha⁻¹ and 105 kg K₂O ha⁻¹ were applied in a one-time base application. The irrigation volume was 360 mm ha⁻¹, and irrigation was mainly carried out after the sowing and jointing stages. The maize variety was Hengdan 188, and the planting density was 60,000 plants ha⁻¹.

3.7. Comprehensive Evaluation of the Optimized Design System Compared with Farmers’ Practices

The yield of maize in DS was 13.8 Mg ha⁻¹, which was 23.2% higher than the yield under FP (11.2 Mg ha⁻¹,

Table 6. Maize yield metrics for the DS and FP.

Treatment	Ear Number 104 ha−1	Grains Per Ear (NO.)	Thousand-Kernel Weight (g)	Yield (Mg∙ha−1)	Harvest Index
DS	7.8	494.1	341.6	13.8	0.48
FP	5.9	521.3	319.8	11.2	0.47

DS, optimized design system; FP, farmers practice.

). The economic net benefit was calculated as the difference between total income and total cost. The economic inputs of DS and FP were 9205 yuan ha⁻¹ and 8660 yuan ha⁻¹, respectively (

Table 7. The average net economic profits of the DS and FP.

Factors	DS	FP
Yuan ha−1
Fertilizers	3150	3520
Irrigation
Electricity	650	515
Water	295	345
Seed	1040	850
Drip-irrigation equipment	1350	1350
Herbicides and pesticides	260	260
Land preparation	760	760
Sowing and harvest	1150	940
Labor	550	120
Total	9205	8660
Gross revenue from yield	27,654	22,478
Returns	18,449	13,818

DS, optimized design system; FP, farmers practice.

), and the total incomes were 27,654 yuan ha⁻¹ and 22,478 yuan ha⁻¹, respectively. The net benefit of DS was 33.5% higher than that of FP. The items with the highest economic cost in the two systems were fertilizer and drip-irrigation equipment, and the sum of the two costs accounted for 48.9% and 56.2% of the total investment, respectively (Table 7).

Compared with FP, DS was associated with significantly improved NPFP, PPFP, and KPFP (Figure 3). The NPFP, PPFP, and KPFP increased by 38.5%, 36.5%, and 43.5%, respectively. Additionally, the WP of DS was 51.2 ha⁻¹ mm⁻¹, which was significantly higher than that of FP (31.2 ha⁻¹ mm⁻¹). Comparatively, DS increased WP by 64.1%.

4. Discussion

At present, the most severe global agriculture challenge is how to increase crop yields and resource utilization efficiency while reducing resource consumption and environmental pollution [27]. A comprehensive mulched drip fertigation system for maize was designed by optimizing crop management (variety and planting density) and resource management (the amount and period of nutrient and water application). The results showed that the maize yield reached 13.8 Mg ha⁻¹ under DS, and NPFP, PPFP, and KPFP were 57.5 kg kg⁻¹, 153.3 kg kg⁻¹, and 153.3 kg kg⁻¹, respectively. Moreover, the DS increased WP to 51.2 kg ha⁻¹ mm⁻¹. The yield and AEN of maize were comparable to that of American irrigated maize (15.1 Mg ha⁻¹; 57 kg kg⁻¹ PFPN) under optimal ecological production conditions. The results of this study revealed that DS can simultaneously increase crop yield and resource-use efficiency, thus providing a proven solution to the current agricultural problems in the semi-arid area of Northeast China.

Several studies have been conducted to ascertain how high yields can be achieved; however, most of these studies were conducted under the local optimal natural ecological conditions, and large amounts of fertilizer and irrigation were invested regardless of economic costs and environmental risks. For example, in [25], super-high-yielding maize grown with an input of N fertilizer as high as 743 kg ha⁻¹ achieved an average yield of only 16.6 Mg ha⁻¹ [28]. The PFPN of these high-yielding systems was only 20 kg kg⁻¹, which was much lower than the 57 kg kg⁻¹ in this study. In addition, high-yielding maize systems generally require 590–840 mm of irrigation during the entire growth period for higher yields [29].

Increasing the planting density of density-tolerant varieties has significantly increased maize yields over the past few decades [30]. In this study, maize yields were highest when the planting density was 70,000 plants ha⁻¹, which was greater than the farmers’ 60,000 plants ha⁻¹. There was a downward trend in maize yields and a significant increase in the risk of lodging when the planting density continued to increase. At the same time, growing degree days also had an important impact on crop growth and development [31]. Medium- and long-growing maize varieties had higher yield potential than short-growing varieties, but too-long growing periods also increased the risk of frost in the later stages of growth [32].

This study found that compared with one-time fertilizer application, fractional fertilizer application could significantly improve maize yield. Similarly, when the N fertilizer application rate was 315 kg N ha⁻¹ in Fengxian Shanghai, reducing the amount of base fertilizer and adding fertilizer once during ear development increased maize yield by 6% [33]. When the N fertilizer application rate was 300 kg N ha⁻¹ in Tai’an, Shandong Province, compared with one-time basal fertilizer, applying N fertilizer three times during the maize growth period increased the yield by 13% [34]. When the N fertilizer application rate was 180 kg N ha⁻¹ in Langfang Hebei, compared with one-time N fertilizer application, applying N fertilizer three times and four times increased the yield by 7.7% and 10.9%, respectively [35]. Fractional fertilizer application can match the soil nutrient supply with the crop nutrient demand as much as possible, which not only effectively controls the growth of crops in the early stage of growth, avoiding excessive growth in the early stage, but also ensures that the later growth stages receive sufficient fertilizer and avoids premature crop aging due to fertilizer application. From the analysis of the metrics of yield, it can be seen that the increase in the number of ears and the thousand-kernel weight of the maize are the direct drivers of the increase in yield. In Linfen, Shanxi Province, when the N fertilizer application rate was 225 kg N ha⁻¹, topdressing maize during the grain filling stage resulted in a 7.3% increase in the number of spikes, a 12.3% increase in the thousand-kernel weight, and a 6.4% increase in the grain yield. Topdressing at the big-bell-mouth stage and the grain-filling stage increased the number of spikes by 11.3%, the thousand-kernel weight by 16.2%, and the grain yield by 19.4% [36]. Fractional fertilizer application can better match the supply of exogenous nutrients with crop nutrient demand. Especially in the later stages of growth, the sufficient supply of nutrients is important; for example, it significantly improves the grain-filling rate of corn in the reproductive period, thereby improving the thousand-kernel weight of maize [37]. Similarly, Jiang et al. (2013) also found that topdressing in the later stage of maize growth resulted in increased thousand-kernel weight [38].

Chinese farmers believe that obtaining high yields requires high inputs of resources (nutrients and irrigation water), and the relatively low prices of fertilizers and irrigation water also drive farmers to use large amounts in production [39]. However, these conventional strategies do not improve crop yields. In this study, compared with FP, the amount of fertilizer applied under DS was reduced by 12% and the amount of irrigation water was reduced by 25%; however, the DS increased yield significantly, by 23%. The over-application of water and fertilizer in FP resulted in lower yield and thus in a reduction of economic benefits by 4631 yuan ha⁻¹ compared with the DS.

The optimal resource-input-management strategy is to minimize environmental risks while ensuring sufficient resource supply to crops [40]. In this study, the N application rate was optimized according to the theoretical N requirement of maize. Fractional fertilizer application was used to meet the N requirements of crops at each growth stage, especially during the filling period. Concerning irrigation management, studies have shown that the water requirements of crops vary greatly across growth stages [41]. The period in which maize is most sensitive to water stress is the flowering period [42], and a water deficit in this period will significantly reduce the harvest index of maize [25]. In DS, more irrigation was allocated to the jointing stage, the silking stage, and the grain- filling stage to ensure sufficient water supply during the critical growth periods, thereby increasing the grain yield [43].

Although DS has the advantages of high yield and high efficiency, there are still some shortcomings [44]. For example, the production of drip-irrigation equipment and plastic mulch requires a large amount of energy, and the residual drip-irrigation tape and mulch also cause soil and environmental pollution. After the maize has been harvested, a large amount of plastic film remains in the soil and furrows, and additional labor costs are required to clean and recycle the residual plastic film [45]. In addition, the cost of drip irrigation under mulch is usually 1350 yuan ha⁻¹, making it the second-most-costly input after chemical fertilizers. The government has invested a large amount in subsidies into the technology supporting water and fertilizer integration, but without government subsidies, the further development of this technology will become a challenge. Economic income is the most important driving force for farmers considering new agricultural technologies [46]. Only a relatively high level of income can enable farmers to better accept new agricultural technologies. Therefore, those involved in the development of water- and fertilizer-integration technology should first fully consider the potential of such technology to increase production in the region.

5. Conclusions

The amount and frequency of N fertilizer application significantly affected maize yield. In this study, the total N application rate was 240 kg ha⁻¹, and the best yield performance was seen when topdressing was applied three times. This result indicates that increasing the amount of N applied does not increase the yield, but it does significantly reduce the N use efficiency. Similarly, topdressing with N fertilizer more than three times did not increase the maize yield correspondingly but did reduce the economic benefit. In summary, mulched drip fertigation systems for high-yield maize optimize crop and resource management and can increase maize yield, reduce resource utilization and increase farmers’ economic benefits. In this study, the specific parameters of DS were as follows: the application rate of N fertilizer was 240 kg ha⁻¹; the base fertilizer was applied four times (once at the jointing stage, twice at the belling stage and once at the silking stage); the application rates of P and K fertilizer were 90 kg P₂O₅ ha⁻¹ and 90 kg K₂O ha⁻¹, respectively; the irrigation amount was 270 mm ha⁻¹; and the maize variety Fumin 985 was planted at a density of 80,000 plants ha⁻¹. DS resulted in a maize yield of 13.8 Mg ha⁻¹ and economic benefits of 18,449 yuan ha⁻¹, significantly better than the 11.2 Mg ha⁻¹ maize yield and economic benefits of 13,818 yuan ha⁻¹ under FP. Moreover, DS significantly improved NPFP, PPFP, KPFP and WP. The above findings provide suggestions for achieving significantly higher yield and the more efficient use of resources for agricultural development in water-limited areas.