Nitrogen Budget in a Paddy-Upland Rotation Field with Soybean Cultivation

Abstract

To reduce the over-production of rice, the paddy-upland rotation system, which alternates every few years between paddy rice cultivation and upland crop cultivation in drained (converted) paddy fields, is now commonly practiced in Japan. Recently, depletion of available soil nitrogen (N) and a subsequent decline in soybean yield in converted upland fields with repeated rotation have been reported in northern Japan. To evaluate the N budget in the paddy-upland rotation field with soybean and rice, a 6-year lysimeter experiment was conducted. In the rotation system, a considerable loss of N occurred in both the upland soybean and paddy rice cultivation periods (−11.9 and − 2.3 g N m−2 y−1, respectively). To mitigate the N loss in the rotation system, N supply from organic matter application is required. The effects of applying different types of organic matter (leguminous green manure, hairy vetch, and livestock manure compost) on the N budget in soybean cultivated fields were investigated. Compared to the N loss in the control plot without organic matter application, the N loss was mitigated in the hairy vetch plot, and N accumulation occurred in the livestock manure compost plot (−13.7, −3.5, and +11.8 g N m−2 y−1, respectively).

1. Introduction

In Japan, rice production has been restricted for more than 40 years due to declining rice consumption. The area planted to paddy rice during summer, which was more than 3 million ha in the 1960s, has continued to decline since 1970, reaching 1.58 million ha in 2019 (Figure 1) [1, 2]. As a countermeasure, crop rotation in shifting the cultivation of paddy fields to crops other than staple food rice (crop rotation) has been implemented in earnest since 1970. As of 2019, 18% of the total paddy area was planted with crops other than paddy rice in the summer (Figure 1).

Figure 1.

Trends in crop cultivation in paddy fields during the summer season in Japan. Source: [2].

One of the systems of crop rotation is “paddy-upland rotation,” in which paddy fields are planted with rice and upland crops alternately for one to several years. Although there are no statistics on how much of the total area is under the paddy-upland rotation, the rotations have become a major cultivation system for crop rotation. In Japan, a three-crop in two-year rotation system: paddy rice in summer, followed by wheat or barley from autumn to next early summer, and then soybean cultivation in summer has been conducted. On the other hand, in northern Japan, where it is relatively cold, a rotation system with annual cropping of paddy rice and upland crops such as soybean has also been conducted.

As mentioned above, soybean is an important rotational crop cultivated in paddy fields in Japan. Of the total area under soybean cultivation in Japan, about 80% is planted in paddy fields, and 90% in the Tohoku region of northern Japan, a major paddy field area (Table 1) [2]. The area of soybean cultivated in paddy fields is 115,900 ha, or 29% of the total area of cultivation with crops other than paddy rice shown in Figure 1 (403,000 ha).

Region	Soybean cultivation area (×10^3 ha)			Percentage of paddy field (%)
	Paddy	Upland	Total
Hokkaido	18.4	20.7	39.1	47
Tohoku	32.7	2.4	35.1	93
Hokuriku	11.7	0.7	12.4	94
Kanto-Tozan	7.7	2.2	9.9	78
Tokai	11.4	0.5	11.9	96
Kinki	9.3	0.2	9.4	98
Chugoku	4.0	0.4	4.3	92
Shikoku	0.5	0.0	0.5	94
Kyushu-Okinawa	20.3	0.7	21.0	97
Total	115.9	27.6	143.5	81

Table 1.

Soybean cultivation area by agricultural region in Japan (2019). Source [2].

While the average yield of the world’s major soybean-producing countries is approaching 3 Mg ha⁻¹, the yield in Japan has remained low at 1.55–1.65 Mg ha⁻¹ [3, 4]. Shimada [4] pointed out that there are many factors contributing to the low soybean yield in Japan, but one of the main factors is the inhibition of N₂ fixation due to wet damage and drought stress in paddy-upland rotation fields.

Soybean assimilates N from atmospheric N₂ by symbiotic N₂ fixation in root nodules [3]. The contribution of atmospheric N₂ to the N accumulation of soybean is highly variable and depends largely on the surrounding environment such as oxygen and moisture. Yoneyama et al. [5] reported that the average percentage of soybean N accumulation derived from N₂ fixation in Japan was 50%. Ohyama et al. [3] reported that the percentages of soybean N accumulation derived from N₂ fixation in rotated paddy fields in Niigata, Japan ranged from 59 to 75%, whereas soybean plants require a large amount of N compared to other crops because of the large protein accumulation in their seeds (about 35–40%). In order to meet this high N requirement, N derived from N₂ fixation in root nodules alone is not sufficient; soybean should also absorb significant amounts of N from the soil. Then, most of the accumulated N in soybean could be removed from the field as harvested grain. Therefore, there is a possibility that N output from the soybean cultivated field exceeds the N input to the field, and thus the N loss could occur. Therefore, the N budget of a converted paddy field with soybean cultivation could be negative, indicating N loss from the field.

Recently, depletion of available soil N followed by a decline in soybean yield in a repeated paddy-upland rotation field has been reported in northern Japan [6]. Nishida et al. [7, 8] reported a decrease in available soil N with an increase in upland frequency (i.e., the number of years in soybean cultivation per total cultivation years) in fields with paddy-upland rotation in Akita, Tohoku region, northern Japan (Figure 2). This indicates that soybean cultivation reduces the soil N fertility of paddy-upland rotation fields. They also reported that when the upland frequency exceeded 60%, the amount of available soil N was less than the minimum value of suitable concentrations of available soil N in the paddy field (80 mg kg⁻¹) [9]. And that the concentration of available soil N could be increased by repeated cattle manure compost application at the rate of 2–3 kg m⁻². A similar trend of declining available soil N has been reported in various parts of Japan in recent years [10, 11, 12, 13].

Figure 2.

Relationship between upland frequency and available soil nitrogen (N). ***P < 0.001, **P < 0.01. CMC, cattle manure compost. Broken lines indicate the minimum (80 mg kg⁻¹) and maximum (200 mg kg⁻¹) values of the suitable range of available soil N in paddy fields [9]. Modified from [8].

Takahashi et al. [14] reported that soil N fertility of a converted paddy field could be a major controlling factor for soybean yield when moisture injury is not severe. In maintaining soil N fertility in paddy-upland rotation fields with soybean cultivation, it is necessary to manage N during soybean cultivation based on the N budget. In general, the N budget in a rice paddy field is considered to be neutral [15] or slightly positive (accumulation) [16], suggesting that the N loss from the rotated paddy field occurs mainly during soybean cultivation. However, the N budget in rice paddy fields could vary widely depending on field management practices [17]. Therefore, in the paddy-upland rotation fields, the N budget including that during paddy rice cultivation should be evaluated and measures to improve the budget should be considered.

2. Detailed nitrogen budget measurement in a paddy-upland rotation field with soybean cultivation

Detailed N budget in a paddy-upland rotation field was evaluated for 6 years (3 years for upland soybean, then 3 years for flooded paddy rice) in a lysimeter plot at the Akita Prefecture Agricultural Experiment Station, located in the Tohoku region, northern Japan [18, 19]. The lysimeter was filled with soil collected from a rice paddy field on gray lowland soil, which is a major paddy soil in this region (soil texture: clay loam). No organic matter other than soybean and paddy rice residues was applied, and the crop was grown according to the guidelines for soybean and paddy rice cultivation in Akita Prefecture [20, 21]. Soybean (cv. Ryuho) was cultivated from early June to early October. Paddy rice (cv. Yumeobako or Akitakomachi) was cultivated from late May to mid-September. The major N flows of inputs (fertilizer, bulk N deposition, irrigation, and symbiotic N₂ fixation in soybean) and outputs (harvested grain, leaching, surface drainage, and N₂O emission) were measured (Figure 3). Symbiotic N₂ fixation in soybean was measured using the relative ureide method [22, 23]. Other N flows were estimated from literature values. The N budget was calculated by the difference between the total input N flow and the total output N flow. Positive and negative values indicate net N accumulation and loss in the field, respectively. Nitrogen budgets were measured for 3 years in soybean and rice cultivation, respectively, and averaged to give annual values.

Figure 3.

Outline of major nitrogen (N) flows in soybean upland field and rice paddy field. NH₃, ammonia volatilization; N₂, dinitrogen; N₂O, nitrous oxide. N₂ emission via denitrification in upland was not considered in this study. Modified from [19].

The average yields of soybean and rice were 341 and 519 g m⁻², respectively. The yield of soybean was much higher than the average of Akita Prefecture (about 140 g m⁻²) [20], whereas the yield of paddy rice was lower than the target value of Akita Prefecture (570 g m⁻²) [21]. It could be due to severe damage by insects in the third year (422 g m⁻²) [19].

Among the N inputs during soybean cultivation, symbiotic N₂ fixation by nodule accounted for the majority of the inputs, about over 80% (Figure 4). The percentages of N accumulation derived from N₂ fixation for 3 years ranged from 60 to 69% [19]. On the other hand, the N input from fertilizer was about 6% of the total. This is due to the fact that in soybean cultivation in converted paddy fields, the amount of N fertilizer applied is set as low as 0–2 g N m⁻² to prevent over-luxuriant growth [20]. The major components of N output were harvested grain and leaching (74 and 25%, respectively). During the soybean cultivation under upland conditions, the annual N budget was negative (−11.9 g N m⁻² y⁻¹), indicating net N loss from the field.

Figure 4.

Comparison of the nitrogen (N) flows and budgets in soybean and rice cultivated fields. Positive and negative values indicated N input and output, respectively. The N budget was calculated by subtracting N output from input. NH₃, ammonia; N₂, dinitrogen; N₂O, nitrous oxide. Modified from [19].

A review of N balance in soybean cultivation reported that the mean value of partial N balance, which is calculated from the difference between the input due to N₂ fixation and the output due to harvest, is close to neutral (−0.4 g N m⁻² growing season⁻¹) [24]. However, the value is likely to be negative if a detailed N balance is obtained, taking into account N losses such as leaching, as in this study. Similar to the present study, the N budget including N flow due to water movement such as leaching in a converted paddy field with soybean cultivation in Shiga, central Japan, was negative (−5.4 to −4.0 g N m⁻² growing season⁻¹) [25].

During the paddy rice cultivation, the major input N flow was fertilizer application (63%), whereas the major output N flows were harvested grain and leaching (49 and 29%, respectively; Figure 4). Although less than soybean cultivation, the N budget during paddy rice cultivation was also negative (−2.3 g N m⁻² y⁻¹), indicating N loss from the field. The N loss during the paddy rice cultivation could be due to the limitation of N fertilization to paddy fields converted from upland fields. In Akita Prefecture, to avoid over-luxuriant growth and lodging due to the increased N uptake, it is recommended that basal N fertilization decreases by 100% and by 50–70% in the first and second years after conversion, respectively [21].

In paddy-upland rotation fields including soybean cultivation, the field N budget in both upland soybean and paddy rice is negative, and soil N fertility is likely to decrease due to repeated rotation [6]. It will be essential to take measures to improve the N budget in paddy-upland rotation fields to maintain soil N fertility and crop productivity.

3. Cultivation managements for the mitigation of N loss from paddy-upland rotation fields with soybean cultivation

As mentioned above, soybean productivity in Japan is low and needs to be improved in the future. As soybean yields increase, a corresponding increase in N loss from the field is expected [24]. Hence, measures to improve the N budget to maintain the soil N fertility become more important. To improve the N budget in the field, N inputs need to be increased.

To increase N₂ fixation, which is a major N input in soybean cultivation fields, control of groundwater level in converted paddy fields has been reported to be effective [26, 27]. Deep placement application of slow-release fertilizers [3, 28, 29, 30, 31, 32, 33, 34, 35] has been proposed as a fertilization method that does not inhibit N₂ fixation in soybean. Nitrogen supply by application of organic matter is also effective in improving the N budget. Organic matter not only supplies nutrients but also affects the physical, chemical, and biological properties of the soil to promote soybean growth. The application of livestock manure compost (LMC) has been reported to increase soybean production by improving N availability and N₂ fixation in soybean nodules [11, 36, 37, 38, 39]. Green manure cultivation of N-fixing legumes is also effective in improving the N budget. In a converted paddy field, cultivation of the leguminous green manure hairy vetch (HV) before soybean cultivation has been shown to promote soybean growth by increasing N supply to the crop and improving soil physical properties [11, 40, 41, 42]. A standard application rate of 2 kg m⁻² of LMC [20] is expected to supply about 20 g N m⁻² of N to the field. Nitrogen supply by HV cultivated before soybean cultivation ranges from 10 to 20 g N m⁻² [43, 44]. These N supplies are sufficient to compensate for the N losses during soybean cultivation described above (−11.9 g N m⁻² y⁻¹, [19]). However, increased N supply to the field may alter other N flows, such as inhibition of N₂ fixation in nodules and increased leaching and N₂O emission. Therefore, there is a need to quantitatively evaluate the effect of organic matter application on the N budget in soybean cultivation in converted paddy fields.

The preliminary results of an experiment for different types of organic matter (HV and LMC; Figure 5) application conducted in lysimeter plots at the Center of Field Education and Research, Faculty of Bioresource Sciences, Akita Prefectural University are reported. The results of this study are for a single year, the first year of conversion from paddy. Three lysimeter plots were filled with soil collected from a rice paddy field on gley lowland soil (Fluvic Gleysols in WRB), one of the major paddy soils in this region (soil texture: heavy clay). The three plots were designated as a control plot with no organic matter application except for crop residue, HV, and LMC plots. In the HV plot, HV was sown and cultivated after paddy rice cultivation in the autumn of the previous year, cultivated until before soybean sowing in early June, and then plowed into the soil. In the LMC plot, 2 kg m⁻² of cow dung-based LMC was applied and incorporated into the soil before soybean sowing. Soybean (cv. Ryuho) was cultivated from early June to early October. No chemical N fertilizer was applied to all plots. Soybean cultivation and N budget measurements were conducted basically as in Section 2 [18, 19].

Figure 5.

Hairy vetch (left) and livestock manure compost (right).

The effect of organic matter application on the N flows and budgets in a converted soybean field is shown in Figure 6. The soybean yields in the HV and LMC plots (358 and 343 g m⁻², respectively) were higher than that in the control plot (314 g m⁻²) (data not shown). Although soybean in the HV plots grew more vigorously than in the other two plots, damage to harvested grain by insects was significant, and the difference in grain yield excluding damaged grains among the plots was small. The total grain yield including damaged grains, which was used in the calculation of N budget, in the control, HV, and LMC plots was 414, 550, and 443 g m⁻², respectively.

Figure 6.

Effect of organic matter application on the nitrogen (N) flows and budgets in a converted soybean field. Positive and negative values indicated N input and output, respectively. The N budget was calculated by subtracting N output from input. HV, hairy vetch; LMC, livestock manure compost; N₂, dinitrogen; N₂O, nitrous oxide.

The N inputs from HV and LMC application were 18.0 and 25.1 g N m⁻², which were higher than the respective N input from soybean N₂ fixation. Symbiotic N₂ fixation by soybean nodules was lower in the HV plot (7.9 g N m⁻²) and higher in the LMC plot (12.1 g N m⁻²) compared to the control plot (10.2 g N m⁻²). The percentages of N accumulation derived from N₂ fixation for the control, HV, and LMC plots were 30, 19, and 34%, respectively. Hairy vetch has a low C/N ratio of around 10 [43, 44], indicating rapid mineralization of its N after incorporating into the soil. Therefore, inorganic N could inhibit soybean nodule growth and N₂ fixation activity [45]. On the other hand, LMC may not inhibit N₂ fixation because of its slow decomposition (mineralization). The lower amount and percentages of N accumulation derived from N₂ fixation than the values shown in Section 2 (Figure 4, [19]) may be due to the fact that the study site was located in a reclaimed land and soil N fertility was high [46]. Therefore, soil N uptake via soybean roots could be high and dependence on symbiotic N₂ fixation could be low.

The major components of N output were harvested grain and leaching (83–87% and 12–13%, respectively). The N output by harvested grain in the HV plot (29.7 g N m⁻²) was higher than that of the control and LMC plots (23.0 and 24.1 g N m⁻², respectively) because of the higher total grain yield. The increase in leached N by organic matter application (1.5 and 0.2 g N m⁻² y⁻¹ for the HV and LMC plots, respectively) was much smaller than that of the amount of applied N. The leaching may have been low because the texture of studied soil is heavy clay and its drainage is poor.

The N budget in the control plot without organic matter application was negative (−13.7 g N m⁻² y⁻¹; Figure 6), and was similar to the value reported previously (−11.9 g N m⁻² y⁻¹; Figure 4) [19]. The N loss during soybean cultivation was mitigated by HV application, and N accumulation occurred by LMC application (−3.5 and +11.8 g N m⁻² y⁻¹, respectively). The application of HV increased N inputs, but only mitigated the N loss because it suppressed symbiotic N₂ fixation and increased N output by harvested grain and leaching. The application of LMC did not suppress symbiotic N₂ fixation, but rather promoted it. It was shown that different types of applied organic matter had different effects on the N budgets in converted paddy fields.

Application of organic matter during soybean cultivation in paddy-upland rotation fields can improve the N budgets. For effective use of hairy vetch, it may be necessary to consider management practice to avoid inhibition of N₂ fixation and to reduce the environmental load such as leaching and N₂O emission. Unlike chemical fertilizers, the N supplied to plants by organic matter such as LMC continues for several years [47]. Nitrogen derived from organic matter applied during upland crop (soybean) cultivation is expected to affect the N budget during subsequent paddy rice cultivation. In the future, it will be necessary to evaluate the effect of organic matter application on the N budget in the entire paddy-upland rotation system, including paddy rice cultivation. Furthermore, accumulation of soil N due to continuous application of organic matter [6, 48, 49, 50] is considered to affect the field N budget. Therefore, it is necessary to evaluate the effects of organic matter application on the N dynamics and budget in the field and on crops on a long-term basis.

4. Conclusions

In paddy-upland rotation fields including soybean cultivation, the field N budget in both upland soybean and paddy rice is negative, and soil N fertility is likely to decrease due to repeated rotation. Application of organic matter is an effective measure to improve the N budget in paddy-upland rotation fields to maintain soil N fertility and crop productivity. The effect of organic matter application on the N budget depends on the type of organic matter. Further evaluation of organic matter management practices and their impact on the N budget is needed.

Acknowledgments

We are deeply grateful to the staff members of the Akita Prefectural Agricultural Experiment Station (especially Mr. Kazuki Sekiguchi and Mr. Keiji Sasaki) and the Center of Field Education and the Research, Faculty of Bioresource Sciences (present: Agri-Innovation Education and Research Center), Akita Prefectural University for their support in the management of the experimental fields. We also thank Ms. Emiko Sato, Ms. Keiko Hatakeyama, Ms. Tomoko Suzuki, and the students of the Laboratory of Soil Science, Faculty of Bioresource Sciences, Akita Prefectural University for their great help with the field survey and laboratory analyses. We also thank Ms. Hiroko Sato (Akita Prefectural Livestock Experiment Station) for providing the livestock manure compost.

Conflict of interest

The authors declare no conflict of interest.