Nitrogen fertilization affects maize grain yield through regulating nitrogen uptake, radiation and water use efficiency, photosynthesis and root distribution

Introduction

In the past four decades, global maize production has greatly increased (FAO, 2018) mainly due to application of nitrogen (N) fertilizers. Worldwide, N fertilizer has widely been excessively applied to achieve higher grain yields (Cui et al., 2009; Meng et al., 2016; Liang et al., 2020). For example, the average dose of N fertilizer applied by the farmers was greater than 300 Kg ha⁻¹ (288 ± 113 Kg ha⁻¹), which far exceeds the optimal N rates for maize demonstrated in field experiments (Zhang et al., 2015a; Chang et al., 2014; Yang et al., 2017). N fertilizer was applied in excess (350−600 Kg ha⁻¹ year⁻¹, Mueller et al., 2013) in an attempt to maximize yields in the North China Plain. However, excessive application of N fertilizer has negative effects on crops, greatly reduces N use efficiency (NUE), and causes significant nitrate leaching losses (more than 50% N to the environment) and contamination of groundwater (Erisman et al., 2013; Wang et al., 2014, 2019a; McBratney & Field, 2015; Ahmad et al., 2018; Suchy et al., 2018). Reducing N input rates from this level to “moderate” levels in maize fields may improve NUE, maintain a fair level of maize grain yield (Zhao et al., 2010; Dai et al., 2015, Qiang et al., 2019), and display less negative environmental impacts. In order to implement reduced N input rates, it is necessary to assess the extent to which the N fertilizer application rate is consistent with crop N requirements to maximize resource utilization and maintain relatively high grain yields (Robertson & Vitousek, 2009; Zhang et al., 2015b).

Radiation interception and radiation use efficiency (RUE) form the basic framework for analyzing crop yield constraints. Variations in crop biomass due to abiotic factors may be attributed to intercepted photosynthetically active radiation (IPAR), RUE, or the combination of both IPAR and RUE (Fletcher et al., 2013). Reduced leaf area under low-N conditions is accompanied by a reduction in radiation interception (Massignam et al., 2012). Low-N conditions primarily decrease the photosynthetic rate per unit area (Vos, Van Der Putten & Birch, 2005), indicating that a low N effect on both leaf size and photosynthetic capacity may affect the final grain yield. Understanding how maize production and resource use are affected by varying N application rates will inform improvement of N fertilizer management to achieve optimal grain yield and resource use efficiency. The appropriate amount of N fertilizer input can improve utilization of precipitation during the crop season (Dai et al., 2015; Herrera et al., 2016; Yang et al., 2017; Qiang et al., 2019). Previous studies of the effect of N fertigation on grain yield and resource use efficiency primarily describe the effects of single resource utilization at the leaf or plant level (Brown, Jamieson & Moot, 2012). A preceding study described the relationship between maize water use efficiency (WUE) and NUE in pot conditions (Wang et al., 2019b). However, few comprehensive studies on the effects of N fertilizer on the utilization of radiation, water, and N resources in field crops have been performed.

Enhancing photosynthesis is critical to promoting crop yield (Mu et al., 2017). In order to understand the underlying mechanisms and differences in photosynthetic capacity, it is necessary to determine the relevant photosynthetic parameters of ear leaves (Li et al., 2013; Zhang et al., 2017; Lamptey et al., 2017). Root morphology and distribution also play key roles in the acquisition of soil resources such as nutrients and water (Ristova & Busch, 2014; Lynch, 2013; Yu et al., 2014). Mi et al. (2010) proposed the ideotype root architecture of “high yield and high N efficiency” in maize, which provided references for root research. In field conditions, the complexity of root sampling has limited efforts to understand the effects of N on roots to shoots. Such research has been conducted under pot conditions (Wang et al., 2019b); thus, it is not possible to assess variation due to differences in light and temperature conditions as well as nitrate N leaching, as occurs in field conditions. Limited knowledge about shoot and root traits related to maize grain yield, NUE, RUE, and WUE under field conditions was investigated. Therefore, exploring the response of photosynthetic parameters and root development to grain yield reduction due to N reduction will provide an important reference for management of N fertilizer inputs.

In the current study, we determined that N reduction results in maintenance of photosynthetic activity and root development to maximize grain yield and radiation, water, and N use by maize crops. The objectives of this study were to (1) investigate the effects of N reduction on water, radiation, and N use efficiencies in maize crops, and (2) determine the effects of N reduction on photosynthetic activity and root development in maize crops under semi-arid conditions. Our findings provide key data to support enhanced maize production and resource use efficiency.

Materials and Methods

Field experiments

A 4-year field experiment was conducted on the same land at the agricultural experimental station of Northwest A&F University in Yangling (34°20′N, 108°04′E, elevation, 455 m), Shaanxi province, China in 2014, 2015, 2016, and 2017. The experimental site experienced an annual average daylight of 2,150 h, an annual average temperature of 12–14 °C, and an average annual precipitation of 581 mm. The annual mean temperature and total precipitation of the experimental area are shown in Fig. 1. The soil of the experimental site is classified as a dark loess soil, and the former crop was winter wheat. Before sowing, soil chemical properties were analyzed in the top 60 cm of soil for organic matter content, N, phosphorus (P), and potassium (K) (Table 1).

Table 1:

Soil chemical properties of the top 0–60 cm layers in the experimental fields.

Layer (cm)	Organic matter (g Kg−1)	Total nitrogen (g Kg−1)	Alkaline nitrogen (mg Kg−1)	Available phosphorus (mg Kg−1)	Available potassium (mg Kg−1)
0–20	15.43	1.203	67.38	12.19	161.17
20–40	14.02	0.987	44.89	9.03	117.21
40–60	13.46	1.069	45.98	6.96	102.67

DOI: 10.7717/peerj.10291/table-1

The research work was carried out with a randomized complete block design with three replications. The total sub plot size was 39 m² (7.8 m long and 5 m wide). The row-to-row spacing was maintained at 60 cm and plant-to-plant spacing at 25 cm. The maize seeds were planted manually, and in each hill, two seeds were sown at a depth of 5 cm in the middle of June during each growing season. Plants were thinned manually for normal plant population densities in the area of (67,500 pl. ha) at V3 (three-leaf stage) (Ritchie & Hanway, 1982). Plots were kept free of weeds, insects, and diseases; and the weeding was controlled by hand and hoe during each growing season. The N treatments applied in this study were: (1) no N (N₀, 100% reduction from N₃₀₀); (2) 150 Kg N ha⁻¹ (N₁₅₀, 50% reduction from N₃₀₀); (3) 225 Kg N ha⁻¹ (N₂₂₅, 25% reduction from N₃₀₀); and (4) 300 Kg N ha⁻¹ (N_300, the traditional N dose applied by farmers in the Loess Plateau of China). Fertilizer N was sourced from urea (46% N), evenly split in the fractions of 1/2 at pre-sowing and side-banded deep (5 cm) into the soil on the sowing rows of 1/2 at the twelve-leaf stage. A total of 150 Kg of phosphorus (calcium superphosphate, P₂O₅ 16%) ha⁻¹ and 150 Kg of potassium (as potassium sulfate, K₂O 45%) ha⁻¹ were applied 5 days before sowing. Irrigation was applied at the twelve-leaf stage (75 mm). The amount of irrigation was controlled by a water meter (Zhejiang Ningbo Water Meter Co., Ltd., Ningbo, China).

Results

Photosynthetic parameters and LAI

High external nitrogen application rates and year had significant effects on Pn, Gs, and Ci, but the interaction between N application rates and year did not display significant correlated (Fig. 2). In both 2016 and 2017, the N₀ and N₁₅₀ treatments resulted in lower values of Pn and Gs when compared with N₂₂₅ and N₃₀₀; however, the effect of N₂₂₅ was relatively greater than that of the N₃₀₀ treatment at R6. This finding indicates that N fertilizer inputs can increase Gs and improve the photosynthetic capacity of maize crops. Conversely, the Pn and Gs of the N₃₀₀ treatment were reduced compared with N₂₂₅. Pn of N₁₅₀, N₂₂₅, and N₃₀₀ increased ranging from 16% to 81% across growth stages compared with N₀ averaged of 2 years. N application rates significantly affected 1-Ci/Ca (p < 0.05). The 1-Ci/Ca of the N₀ treatment was significantly lower than that of N₃₀₀ in all measurements (Fig. 2).

Overall, the LAI was unaffected (p = 0.55) by N application rates, with an average of 2.6 and 3.7 at the V8 and VT stages (Fig. 2). N application rates significantly affected LAI at the R3 and R6 stages. Compared with N₃₀₀, the LAI of N₂₂₅, N₁₅₀ and N₀ decreased by 1%, 6%, 11%, respectively, and 1%, 7%, and 13%, respectively, at the R3 and R6 stages.

Discussion

An optimal rate of N application is expected to produce a balance between crop demand and N supply and ensure maximum crop production while conserving resources and protecting against environmental damage (Ciampitti & Vyn, 2011; Peng, Li & Fritschi, 2014). Our results showed that during the 2017 growing season, even when photosynthesis was significantly affected by a 50% reduction in N application (N₁₅₀), grain yield was not significantly decreased. In 2016, photosynthetic parameters decreased further for the N₁₅₀ and N₀ treatments, and the grain yield also decreased significantly. Over 4 years, we observed no significant reduction in grain yield for the N₂₂₅ treatment (Table 2), which even displayed a grain yield higher than that of the N₃₀₀ treatment, although the difference was not significant. Our research indicated that the N dose could be reduced by at least 25% without compromising grain production. Our results are similar to those of Li et al. (2007) and Lamptey et al. (2017), who reported an optimal N fertilization range for summer maize of 200 to 300 Kg N ha⁻¹. Under the N₀ and N₁₅₀ treatments, dry matter accumulation was limited, and the difference between N₃₀₀ and N₂₂₅ treatments was not significant. The reduction in crop yield induced by N reduction can be explained by various factors, as described below.

Our experiments also addressed the mechanism by which maize yield is affected by N application rates. We found that for most years, IPAR was not affected by N fertilizer application rates, a finding similar to that described in previous studies (Vos, Van Der Putten & Birch, 2005; Massignam et al., 2012). N application rate significantly affected RUE (p < 0.05). For N₀ and N₁₅₀, the RUE was significantly less than that of N₃₀₀. However, the RUE of the N₂₂₅-treated crop was greater than that of the N₃₀₀-treated crop. In this case, lower grain yield driven by lower N treatment corresponded to lower RUE. These results indicate that under such production conditions, N application rate mainly affects grain yield by affecting RUE rather than IPAR. Vos, Van Der Putten & Birch (2005) also found that maize tends to sacrifice specific leaf N and RUE while maintaining leaf area (small changes to LAI (Fig. 2) in comparison with the large decrease in Pn (Fig. 2) at low N application rates). Therefore, in the condition of high IPAR, improving the RUE may represent a valid mechanism to achieve high maize grain yield.

The actual ET involves two components: crop transpiration and soil evaporation. N application can increase ET during the reproductive period due to high leaf transpiration under high N conditions (Lamptey et al., 2017; Rudnick et al., 2017). The relatively low sensitivity of IPAR to N supply in maize may also be consistent with the low sensitivity of soil evaporation to N supply. In this case, the lower grain yields associated with the N₀ treatments corresponded to lower ET and WUE values. In addition, WUE of the N₃₀₀-treated crop was lower than that of the N₂₂₅-treated crop. This is due to a higher sensitivity to soil water by the plant at higher N application rates. An increase in N application rates is usually accompanied by a decrease in NUE (Ju et al., 2015). Significant effects of N were also observed on N use efficiency (NUE including AEN, REN, NHI, and INE) in our study. Conversely, reducing the N application rate can create a balance between crop demand and N supply (Lamptey et al., 2017). The desired N concentration of the plant, under the N₀ treatment, has a high INE value, indicating that the plant N concentration or yield are low and that the amount of N in the plant absorbed from the soil is small. N accumulation increased with increasing N application, but N accumulation in the grain was not significantly different between N₂₂₅ and N₃₀₀ treatments, and the value of NHI at N₂₂₅ was greater than that at N₃₀₀, which indicated that increased N in the plant did not transfer to the grain, resulting in excessive N absorption and residual N in the vegetative organs.

The plasticity of root morphology allows it to respond to soil mineral nutrients (Peng, Li & Li, 2012; Yu et al., 2014). We found that root dry weight reached a maximum with the N₂₂₅ treatment, which suggested that the relationship between N input and the root system is not linear and positive; N input may even have a negative impact on root growth and development. In the present study, application of N fertilizer promoted growth of roots in the 0–60 cm soil layer and increased the proportion of roots in this layer, indicating that N application improved growth of the upper layer roots. In the late growth stage (after VT), the N₀ treatment exhibited a negative effect on the root dry weight and proportion in the 30–60 cm and 60–90 cm soil layers, indicating that N deficiency would be detrimental during the accelerated aging of deep layer roots. Not only are root morphology and nutrient absorption closely related, but the spatial distribution of roots is also closely connected to crop growth and productivity (Mi et al., 2010; Lynch, 2013). In both years, the root system exhibited the optimal distribution under the N₂₂₅ treatment, with a higher root length density in the observed soil layer, resulting in larger and deeper infiltration scales. Slower root senescence in the N₂₂₅ treatment is also a major contributor to N rate–induced increases in grain yield. Studies have shown that the relative stability of the deep root environment is beneficial in promoting the buffer capacity of the root system in adverse soil environments and achieving high grain yield and resource use efficiency (Chen et al., 2010; Mi et al., 2010; Wasson et al., 2012; Saengwilai et al., 2014). The results of our study during the 2016 and 2017 growing seasons showed that excessive N (N₃₀₀) application negatively affects early deep root growth compared with N₂₂₅. High external N input also appeared to generate an overall inhibitory effect on later root growth. There are many reasons for the observed reduction in crop yield. Slight reductions in crop yield induced by excessive N application may be due to negative impacts on root growth during the early growth stage or may be caused by differing mechanisms of aging leading to N loss and relative N deficiency during the reproductive stage. N deficiency induces root thinning and increases longitudinal expansion by promoting root growth in the deep soil, while high N inhibits vertical expansion of roots (Trachsel et al., 2013; Mu et al., 2015). This study explained the effect of excessive N and N deficiency on yield from the perspective of root morphology and growth.

The role of N in grain formation is mainly explained by photosynthesis, and N reduction usually negatively impacts photosynthetic performance in maize (Massignam et al., 2012; Olszewski et al., 2014). Although the N₁₅₀ treatment significantly reduced Pn in the 2017 growing season, the grain yield for the N₁₅₀ treatment was not significantly reduced, indicating that the plant’s transient photosynthetic parameters were more sensitive than the dry mass factors in responding to changes in N application rates. Our results illustrate the physiological basis for utilizing the 225 Kg ha⁻¹ N rate to improve the stress resistance of summer maize plants in the Loess Plateau. For the N₃₀₀ treatment, decreased Pn was due to lower Gs, which may be due to N-associated increased sensitivity of plant to soil water and results in lower WUE than the N₂₂₅ treatment at the R6 stages. In the low-N treatment, maize plants showed reduced leaf rolling under soil water stress compared with the high N treatment and thus obtained higher WUE (Wang et al., 2019b).

At different measurement dates, the maximum value of Gs was not always associated with the N₃₀₀ treatment (Fig. 2). In general, plants’ water requirements are expected be greater under high N conditions. Thus, high-N fertilizer plots are more likely to be water-deficient if the soil moisture is inadequate, which aggravates the plant’s stomatal limitations (1-Ci/Ca) and resulting in reduction of Gs and Pn. Similar findings have been described in wheat (Zhang et al., 2017). Previous studies also pointed out that under soil water-stress, ABA (as the signal carrier) transmitted to the shoot, reducing the Gs and increasing stomatal limitations in the initial drought (Larcher, 2003; Li & Xu, 2014; Yan et al., 2017). The inadequate soil moisture of the N₃₀₀ treatment is not serious enough to affect non-stomatal factors restricting photosynthetic carbon assimilation. The Pn of the N₂₂₅ treatment was similar to that of the N₃₀₀ treatment during the VT stage, but it was less than that of the N₂₂₅ treatment in R6. Although the N₂₂₅ treatment showed a higher photosynthetic rate in the R6 than the N₃₀₀ treatment, the ratio did not result in a significantly higher yield, which may be caused by the small contribution of high photosynthetic capacity to a grain yield (Acciaresi et al., 2014). In addition, the photosynthetic capacity may also be related to the differences in N nutrition characteristics among different maize genotypes (Chen et al., 2014; Li et al., 2015). Thus, while the application rate of 225 Kg N ha⁻¹ was may still be high, the current N rate was relatively effective in improving resource use efficiencies. Under production conditions, large amounts of N input are often used as an “insurance” against higher yields to ensure further increases in maize production (Huang et al., 2007; Chen et al., 2012). However, this behavior results in a significant reduction in NUE (Chen et al., 2010) and a slight reduction of RUE and WUE. Previously, Chen et al. (2016a) and Mu et al. (2018) reported that N is used primarily for cell morphogenesis, that N in leaves is N redundant, and excess N is mainly stored in soluble protein and light-harvesting pigment-protein complexes. Therefore, in our study, the grain yield of N₂₂₅ treatment did not display significantly different results compared to the N₃₀₀ treatment, and the root system and photosynthetic capacity showed certain advantages of N₂₂₅ treatment. From these findings, we can conclude that it is achievable to improve resource use efficiencies while ensuring grain yield. Actually, maize genotypes and soil moisture also affect GY and physiological characteristics (Ciampitti et al., 2013). In the current experiment, these factors are not taken into consideration. In future studies, we will focus on the effects of genotype with the aim of maximizing GY and resource use efficiency.

Conclusions

Decreased grain yield due to N reduction was driven by reduced radiation utilization efficiency and WUE; the impact of radiation interception and total water evapotranspiration were relatively small. An application rate of 225 Kg N ha⁻¹ could be used as a reference for optimal N application in the Loess Plateau of China. This N application rate optimized the eco-physiological responses of the plant, a finding which was confirmed by measuring photosynthetic activity and the root system. This response to optimizing N input resulted in higher grain yield, RUE, WUE, and NUE. Reducing N application rates has strong recoverability in maize production and can maximize the capture and utilization of resources, increasing the maize grain yield.

Supplemental Information