Optimum Cultivation Model Increased the Grain Yield of Ratoon Rice and Nitrogen Agronomic Efficiency by Improving Root Morphological Traits and Stubble Character of the Main Rice

Abstract

The present study investigated the effects of different integrated cultivation treatments on the yield, nitrogen agronomic efficiency (AE_N), and growth performance of a ratoon rice system. A two-year field experiment was carried out during 2021–2022 using a randomized block design. Cultivation treatments included the farmers’ practice model (control, CK), high-yield and high-efficiency model (T1), enhanced-high-yield and high-efficiency model (T2), and the no-N model (N0). The grain yield, yield components, root traits, and stubble characteristics were determined. The results showed that T2 treatment significantly increased the grain yield and AE_N by 53% and 236%, respectively, compared with CK. Similarly, the stubble weight, non-structure carbohydrate contents, and the root attributes were significantly improved under T2 treatment. The correlation analysis suggested that ratoon rice yield and AE_N was significantly and positively correlated with root weight (p < 0.01) and the stubble parameters including stubble weight (p < 0.01), starch content (p < 0.01), and soluble sugar content (p < 0.01). Additionally, a significant and positive correlation between root weight and stubble weight (p < 0.05) was determined in our study. Consequently, our work highlights the benefits of integrating planting density, nitrogen/water management, and other practices to achieve high AE_N and grain yield of the ratoon rice by improving root trait and stubble characteristics of the main rice.

1. Introduction

Rice is the staple food crop for more than 3 billion people in the world and the most important food crop in China [1]. Due to rapid global population growth and climate change, there will be a dramatic impact on crop production that may threaten food security [2]. By 2030, China needs to increase rice production by about 20% to meet domestic demand [3]. Although lately, rice production has increased year by year, with the development of society and economic progress, rice production is facing difficulties such as a shortage of rural labor and low economic returns [4]. Ratoon rice is the second rice crop produced from the stubble left behind after the harvest of the main crop [5]. According to reports, the annual yield of ratoon rice is 55% to 71% higher than that of single-crop rice. The problems of resource waste and environmental pollution caused by the pursuit of a super high yield through the high input of single-crop rice were solved. In addition, the growing season of regenerative rice was 44 to 48 days shorter than that of double-cropping rice, which solved the problem of declining planting area caused by labor shortage and low production efficiency of double-cropping rice [6], which has aroused wide amounts of attention and has been expected to become one of the major cropping systems in China. It has been reported that there is more than 1 million ha of ratoon rice cropping in southern China [7]. Moreover, the grain yield of ratoon rice has been greatly improved in recent years, and highest yield reached 6.0 t ha⁻¹ in Qichun County, Hubei Province, and 7.5 t ha⁻¹ in Miluo County [8,9], Hunan Province, in small-scale demonstrations [10].

The yield of rice is closely related to the degree of root development, and the root length, root weight, and absorption area of various roots can reflect the root configuration to a certain extent [11]. The growth and yield formation of ratoon rice is highly affected by the growth traits of main season rice, such as root and stubble traits. The root system of the main season rice is critical for the growth and development of ratoon rice [12]. Zheng et al. reported that the generation of ratoon tillers depends on the residual root system of the main crop [13]. The panicle numbers and grain yield of the ratoon rice showed significantly positive correlation with the activity of root system at the maturity of the main rice [14]. On the one hand, the high root activity, large leaf area index, and high grain weight were suggested to be the ideal varietal traits for rice ratooning [15]. On the other hand, the cultivation practices that could increase root weight and root activity were also beneficial for the growth, development, and the grain yield of ratoon rice [16]. In addition to the root traits, the morphological and nutrient status of the rice stubble after the harvest of the main crop harvest is another important factor affecting ratoon yield. In the distribution of photosynthates, about 10–20% of the photosynthate in the leaf was retained in the leaf and stem at the full heading stage in the main season, and mainly in the leaf. Only about 5% was retained in the stem sheath. More than half of the photosynthate stored in rice stubble would be used by ratoon rice [17]. Xu et al. suggested that the stubble dry weight is an important indicator to predict the ratooning ability of hybrid rice [18]. In the early growth phase of regenerative rice, the germination and growth of regenerative tillers are mainly attributed to the nutrients stored in the straw of the main season [19], and the nutrients stored in rice straw are positively correlated with ratooning ability [20]. Meanwhile, the non-structural carbohydrate that stored in the stubble was regarded as the ordinary source for the generation of axillary bud during ratoon season [21]. The research by Xu and Xiong indicated that the increase in the non-structural carbohydrate of the rice stubble increased both generation speed and the total amount of the auxiliary buds during ratoon season [22]. In summary, the abovementioned findings suggest that improvement in both root traits (root weight and root activity) and stubble characteristics (stubble weight and its non-structure carbohydrate) is essential for further improving the productivity of ratoon rice, and such improvement should be achieved either by cultivation technology innovation or variety breeding.

The cultivation technology for ratoon rice has been rapidly developed in recent years, and several high-yield and high-efficient cultivation techniques in ratoon rice system have been reviewed by Xu et al. [17]. Our previous research suggested that the use of integrated practices rather than single agronomic practice was more efficient in achieving the realistic genetic potential and high nitrogen use efficiency [23]. Therefore, based on the integrated practices (planting density, nitrogen/water management and other practices) referenced by Qin et al. [23], field experiments were carried out over two years on two hybrid rice varieties (i) to identify an integrated agronomic practice to improve the root trait and stubble character in the main rice; (ii) to identify an integrated agronomic practice to enhance the grain yield of the ratoon rice; (iii) and to estimate the critical factors for enhancing ratoon rice yield among these root traits and stubble characteristics.

2. Materials and Methods

2.1. Ratoon Rice Planting

An integrated cultivation experiment was conducted in Liuyang County, Hunan Province, China (28°09′ N, 113°37′ E, 43 m a.s.l.). The climate at the site was a subtropical humid monsoon with a mean annual precipitation of 1171.6 mm and average daily temperature of 17.3 °C; the maximum and minimum daily temperature were 37.7 °C and −4.7 °C, respectively. Detailed information about daily temperature and solar energy are provided in Figure 1. The soil at the site was clay with a pH of 6.3, 1.84% organic C, and 0.11% total N.

2.2. Experimental Design and Field Management

We used a randomized block design in 2021–2022 with three different integrated cultivation treatments, including the farmers’ practice model (control, CK), high-yield and high-efficiency model (T1), enhanced-high-yield and high-efficiency model (T2), and the no-N model (N0). Each plot measured 5.0 m × 8.0 m, and there were four replicates of each treatment. Detailed information about the treatments and other management practices is provided in

Table 1. Planting density, nitrogen management, and other measures among different cultivation models.

Cultivation Model	Nitrogen Application	Planting Density	Other Measures
No-fertilizer model (N0)	0 kg N ha−1	25 hills m−2 2–3 seedling hills−1	45 kg P ha−1, 90 kg K ha−1
Farmer’s practice model (CK)	150 kg N ha−1 (7:3:0)	25 hills m−2 2–3 seedling hills−1	30 kg P ha−1, 60 kg K ha−1
High-yield and high-efficiency model (T1)	120 kg N ha−1 (5:3:2)	30 hills m−2 4–5 seedling hills−1	45 kg P ha−1, 90 kg K ha−1, Zinc sulfate 5 kg ha−1
Enhanced-high-yield and high-efficiency model (T2)	150 kg N ha−1 (4:3:3)	32 hills m−2 5–6 seedling hills−1	Ridge tillage, organic fertilizer 1.8 t ha−1, 50 kg P ha−1, 100 kg K ha−1 (5:5), Zinc sulfate 5 kg ha−1

.

Urea for N, calcium superphosphate for P, and potassium chloride for K were used as fertilizer throughout the experiments. In the main rice, urea was applied in three stages: initial, mid-tillering, and panicle initiation. P, zinc sulfate, and organic fertilizer, as well as 50% of the K fertilizer, were applied as initial fertilizers prior to rice transplanting, and the remaining 50% of K fertilizer was applied at panicle initiation. The initial fertilizer application (2 days before rice transplanting) was incorporated into the soil by plowing to a depth of 10–20 cm, and top-dressing was spread at the soil surface. In the ratoon rice, except for the N0, nitrogen quantity of the other models uniformly in the ratoon rice was 105 kg N ha⁻¹.

Seedlings were raised in nursery beds, and 25-day-old seedlings were manually transplanted between March 25 and April 20. Pesticide and herbicide management was performed according to local practices.

2.3. Measurements and Methods

2.3.1. Measurement of Aboveground Total Dry Weight, and Yield and Its Components

At maturity, 10 hills were sampled diagonally from a 5 m² harvest area to determine aboveground total dry weight, HI, and yield components.

Plants were separated into straw and panicles. Straw dry weight was determined after oven-drying at 70 °C to constant weight. Panicles were hand-threshed, and filled spikelets were separated from unfilled spikelets by submerging them in tap water. Three 30 g subsamples of filled spikelets and three 3 g subsamples of unfilled spikelets were collected and counted. Dry weights of the rachis and filled and unfilled spikelets were measured after oven-drying at 70 °C to constant weight. Aboveground total dry weight was the total dry matter of the straw, rachis, and filled and unfilled spikelets.

Panicle numbers were counted in each hill to determine the number of panicles per m². Spikelets per panicle, grain-filling percentage (100 × filled spikelet number/total spikelet number), 1000-grain weight (The weight of 1000 filled spikelets), and HI (100 × filled spikelet weight/aboveground total dry weight) were calculated. Grain yield was determined from a 5 m² area in each plot and adjusted to a standard moisture content of 0.14 g H₂O g⁻¹.

2.3.2. Measurements of Nitrogen Agronomic Efficiency

Nitrogen agronomic efficiency (AE_N, kg N kg⁻¹) was calculated as the difference in yield between an N-application plot and N0 divided by the amount of N applied.

2.3.3. Measurements of Root Morphology Traits

At heading, two hills with uniform plants were selected for each variety. The remaining roots in soil were carefully collected by handpicking. The roots from each pot were combined, washed, and then scanned using a scanner (Epson Expression 1680 Scanner, Seiko Espon Corp., Tokyo, Japan). The scanning images were analyzed via a WinRHIZO root analyzer system (Regent Instruments Inc., Sainte Foy, QC, Canada) to determine root length and root volume. Root biomass was determined after oven-drying at 70 °C to a constant weight. Specific root length was calculated as the ratio of root length to root biomass.

2.3.4. Measurements of Soluble Sugar, Starch Content in the Stubbles

The determination method is determined by anthracene copper colorimetry, with reference to the method detailed by Yoshida et al. [24].

Take 1 mL of soluble sugar (starch) extraction solution into 10 mL of heat-resistant test tube, slowly add 5 mL of anthrone reagent into the test tube, and shake it. After 10 min of color development, determine the absorbance value at 620 nm wavelength, and convert it into the concentration of soluble sugar and starch according to the following formula:

(1)

Soluble sugar concentration (%) = [(C × V/a)/(W × 1000)] × In the formula: C—the mg value of glucose obtained from the standard curve; V—total volume of sample extract (mL); a—Amount of extraction solution taken during color development (mL); W—sample weight (g)

(2)

Starch concentration (%) = [(C × V/a) × 0.9/(W × 1000)] × In the formula: C—the mg value of glucose obtained from the standard curve; V—total volume of sample extract (mL); a—amount of extraction solution taken during color development (mL); W—sample weight (g); 0.9 is the conversion coefficient of glucose to starch.

2.4. Data Analysis

Analysis of variance (ANOVA) and linear regression were performed using Statistix 8 software (Analytical Software, Tallahassee, FL, USA). Means were compared using the least significant difference (LSD) test at a significance level of 0.05. Tables and figures were created using Microsoft Excel 2018 and Origin 2021, respectively.

3. Results

3.1. Yield and Its Component of the Main Rice and Ratoon Rice

The grain yield of the enhanced-high-yield and high-efficiency model (T2) were 8.6 t ha⁻¹ and 6.2 t ha⁻¹ for main and ratoon rice, respectively, which were 48.1% and 51.1% higher than that of the farmers’ practice model (CK, Figure 2). The grain yield analysis suggested that the variances in grain yield were closely associated with the differences in panicle numbers. When averaged across years, the panicle numbers in T2 treatment were increased by 50.1% and 34.5% compared with the CK and T1 treatments, respectively.

3.2. Nitrogen Agronomic Efficiency of the Ratoon Rice

The nitrogen quantity of those models (except for N0) in ratoon rice was consistently at 105 kg N ha⁻¹. The average nitrogen agronomic efficiency (AE_N) of the T2 was 30.6 kg N kg⁻¹ of the variety ‘JLY1468’ and 27.4 kg N kg⁻¹ of the variety ‘YY4949’, respectively. These numbers were 2.4 times higher than that of ‘JLY1468’ and ‘YY4949’ compared to CK (Figure 3). There was an interaction between treatments and year among the ratoon rice (p < 0.01, Figure 3). In 2022, the nitrogen agronomic efficiency of the models was higher than that of 2021.

3.3. Root Morphological Traits in the Flowering Stage

Root length, root volume, and root weight per m² of the T2 were significantly higher than that of the CK. There was a significant difference in specific root length between the varieties. The average specific root length of T2 in the variety ‘YY4949’ was 59 m g⁻¹ for the main rice and 64 m g⁻¹ for the ratoon rice, respectively, which was significantly higher than that of the CK. The average specific root length of CK in the variety ‘JLY1468’ was 31 m g⁻¹ for the main rice and 31 m g⁻¹ for the ratoon rice, respectively, which was significantly higher than that of T2 (Figure 4).

Root length, root volume and root weight per m² in the main rice was higher than or equal to that of the ratoon rice. However, the specific root length of the T2 in the ratoon rice was higher than that of the main rice (Figure 4).

3.4. Stubble Weight of the Main Rice and Its Non-Structure Carbohydrate Contents

The average stubble weight of the T2 were 547 g m⁻² of the variety ‘JLY1468’ and 434 g m⁻² of the variety ‘YY4949’, respectively, showing a significant increase of 40.4% and 13.4% compared with CK (Figure 5). A significant difference was observed between the T2 and the T1 (p < 0.01), and the CK (p < 0.01). In addition, T2 had higher non-structural carbohydrate contents, including starch and soluble sugar. The starch content of T2 was 35.7% for ‘JLY1468’ and 30.3% for ‘YY4949’, while the soluble sugar content for T2 was 23.8% for ‘JLY1468’ and 16.8% for ‘YY4949’, which was higher than that of the CK (Figure 5).

3.5. Ratoon Rice Yield and AE_N Associated with Root and Stubble Parameters

Ratoon rice yield and AE_N were significantly and positively correlated with root weight (Figure 6, p < 0.01). Further analysis showed that a significant positive correlation was observed between spikelets per panicle and root parameters, including root length (p < 0.01), root volume (p < 0.01), root weight (p < 0.01), and specific root length (p < 0.01). Additionally, there was a significant correlation between 1000-grain weight and root length (p < 0.01), root volume (p < 0.05), and specific root length (p < 0.01). However, panicle per m² was significantly negatively correlated with root length (p < 0.01) and specific root length (p < 0.01).

Ratoon rice yield and AE_N was positively associated with stubble parameters, including stubble weight (p < 0.01), starch (p < 0.01), and soluble sugar (p < 0.01). Further analysis showed that a significant positive correlation was observed between panicle per m⁻² and stubble parameters, including stubble weight (p < 0.01) and starch (p < 0.01). Moreover, there was a significant and positive correlation between spikelets per panicle and soluble sugar (p < 0.01).

Therefore, root weight, stubble weight, and their non-structure carbohydrate contents were the most important factors influencing ratoon rice grain yield and AE_N. Additionally, there was a significant positive correlation between root weight and stubble weight (p < 0.05), while significant positive correlation was observed between root weight and soluble sugar content (p < 0.01).

4. Discussion

The grain yield is determined by the panicle per unit land area (pool size), the filling rate of spikelets and the 1000-grain weight, and the pool size is generally considered to be the main determinant [25]. Pool size can be increased by increasing the panicle per unit land area and/or the spikelets per panicle. In rice production, N is usually applied at the early vegetative stage to promote tillering and increase the number of panicles per unit land area, while N topdressing at the panicle differentiation stage is necessary to increase the number of spikelets per panicle [26]. The integrated agronomic practice of T2 mainly includes reducing basal nitrogen while increasing panicle nitrogen, adopting high planting density and increasing seedling number per hills, and implementing ridge tillage (Table 1), compared with CK. The increased seedling density could have ensured higher biomass accumulation in the T2 even with less basal N, compensating for the negative effects of reduced N; this finding is in line with the studies by Huang et al. and Zheng et al. [27,28]. The N top-dressing levels in T2 were higher than in the CK, meeting the N demand for plant growth during the later growth stages, which has also been reported in previous studies [26,29]. Increasing panicle nitrogen could provide more photoassimilates during the grain-filling stage, resulting in reduced translocation of stem reserves before pre-heading. As suggested by Xu and Xiong [22], considering that the main season rice was the basis of the ratoon season, in the main season of rice, the input of photosynthate stored in the previous period into the panicles was reduced, so that more photosynthate was left in the stem and sheath during harvest, which could provide important nutrients for regenerated bud growth; moreover, the residual nutrients in the main season rice stubble could promote bud germination and growth, leading to the enhanced grain yield of ratoon rice.

A strong root system is necessary for high biomass accumulation in hybrid rice, as it provides sufficient inorganic nutrients. The construction of such a system is closely linked to air aeration in soil [30]. Ridge tillage soil had lower bulk density, higher redox potential, loose soil texture, better air permeability, increased porosity and gas rate, and improved physical soil properties. With high redox potential and sufficient oxygen content, on the one hand, the rhizosphere microorganisms are active, which is helpful for the desorption and utilization of minerals and reduces the toxic effect of reducing substances on seedlings. On the other hand, it coordinates the relationship between water, fertilizer, air, and heat in the root growth environment, promotes the activities of microorganisms, improves the soil environment for root growth, and is beneficial to the individual growth and development of rice and the establishment of an efficient population [31,32]. Moreover, it affects the development of crop root systems, leading to a significant increase in total root biomass and volume, absorption capacity, and root-to-shoot ratio. These effects lay the foundation for achieving high yields [33]. Our results show that root length, volume, and weight per square meter were significantly higher in the T2 than in the CK for the main rice. However, there was a significant difference in specific root length between the varieties. The average specific root length was 47 m g⁻¹ for ‘YY4949’ and 26 m g⁻¹ for ‘JLY1468’, indicating that the root diameter of ‘YY4949’ was thinner than that of ‘JLY1468’. Nonetheless, there was no correlation between root weight and specific root length (p > 0.05), potentially because the root length of ‘YY4949’ was longer than that of ‘JLY1468’.

In this study, our findings indicate that the weight of rice stubble increased by an average of 26.9%, and the non-structural carbohydrate content increased by an average of 26.6% in the T2 treatment, which was significantly higher than that of the CK treatment. The root weight per m² of the T2 treatment was also significantly higher than that of the CK treatment, indicating effective improvement in the inheritance characteristics of the main rice. The data on the grain yield of the ratoon rice and nitrogen agronomic efficiency numbers (AE_N) showed an average increase of 49.4% and 2.4 times, respectively, in the T2 treatment compared with the CK. Moreover, our correlation analysis revealed a significant and positive correlation between the root weight, stubble weight, starch content, and soluble sugar content of the stubble parameters with ratoon rice yield and AE_N (p < 0.01). Additionally, a significant and positive correlation between root weight and stubble weight (p < 0.05) was observed. Our study underscores the benefits of integrating planting density, nitrogen/water management, and other practices to improve root traits and stubble characteristics of the main rice and achieve high AE_N and grain yield of the ratoon rice.

5. Conclusions

In summary, the average annual grain yield of the enhanced-high-yield and high-efficiency model (T2) was 14.2 t ha⁻¹ of the variety ‘JLY1468’ and 15.5 t ha⁻¹ of the variety ‘YY4949’, respectively, which were 47.7% and 51.0% higher than that of the farmers’ practice model (CK). The average agronomic nitrogen use efficiency (AE_N) of T2 was 25.4 kg N kg⁻¹ of the variety ‘JLY1468’ and 29.0 kg N kg⁻¹ of the variety ‘YY4949’, respectively, which were 3.5 times higher than that of CK for both varieties. The annual grain yield and AE_N of ratoon rice in T2 were higher than CK due to the increase in root length, root volume, root weight, and rice stubble weight. Therefore, optimizing the cultivation mode can improve the characteristics of roots and rice stubble, thereby enhancing the yield and AE_N of ratoon rice.

However, the relationship between root and rice stubble characteristics is unclear when changing the cultivation mode. Further research is needed to study the information of increasing root weight and stubble weight, particularly before pre-heading.