White clover living mulch controlled only by mowing supplies nitrogen to corn

Abstract

A field experiment was conducted to investigate the effects of white clover (Trifolium repens L.) living mulch controlled only by mowing on the nitrogen (N) nutrition and yield of silage corn (Zea mays L.). Eight treatments were tested: one living mulch treatment without N application and seven conventional cultivation treatments with different N applications (0–200 kg ha⁻¹). White clover was sown in the living mulch treatment in August. The next May, white clover shoots in the living mulch treatment were clipped and left in the field. After tilling the conventional cultivation treatments, corn was sown. At the time when the corn was sown, the N uptake of the white clover shoots was 173 kg ha⁻¹. Measurements at 40 days after the sowing of corn (40 DAS) revealed that the N concentration of corn shoots grown in the living mulch treatment was lower than that in the conventional cultivation treatment without N application. At harvest (123 DAS), compared with that in the conventional cultivation treatment without N application, the N uptake of corn shoots in the living mulch treatment increased by 31.8 kg ha⁻¹. Based on the N uptake and the yield of corn, the fertilizer N equivalencies of living mulch were 62 and 70 kg ha⁻¹ at harvest, respectively. These results indicate that living mulch controlled only by mowing increases the N uptake of corn and may reduce the N application required for corn production.

Key words:

• Corn
• living mulch
• nitrogen
• white clover

1. INTRODUCTION

Living mulch cultivation is a technique in which a cover crop is planted before the main crop to function as a living groundcover for the growing season (Hartwig and Ammon 2002). Living mulch offers several benefits for crop cultivation, including soil erosion prevention (Hall et al. 1984; Wall et al. 1991), an increase in soil organic matter (Kumwenda et al. 1993) and weed growth suppression (Ilnicki and Enache 1992; Miura and Watanabe 2002; Uozumi et al. 2004).

There are many reports on the effect of the nitrogen (N) supply to the main crop from legume dead mulch, a kind of cover crop that is killed by herbicides at the sowing of the main crop (Smith et al. 1987; Fageria et al. 2005). On the other hand, the effect of living mulch on the N nutrition of the main crop is not consistent. Duiker and Hartwig (2004) reported that crown vetch (Coronilla varia L.) living mulch supplied the main crop with N through biological N fixation. On the contrary, it has been reported that the N uptake of a main crop grown with living mulch is lower than that under conventional conditions (White and Scott 1991; Garibay et al. 1997). Therefore, we decided to further investigate the effect of leguminous living mulch on the N supply of the main crop.

In some cases, living mulch competes with the main crop and reduces its yield (Feil et al. 1997; Garibay et al. 1997). In previous studies, cover crops used as living mulch were suppressed by herbicide application or banded cultivation to avoid competition with the main crop (Teasdale 1996). On the other hand, Uozumi et al. (2004) reported that white clover (Trifolium repens L.) living mulch can be controlled by mowing only once at the time of sowing corn (Zea mays L.) without chemical herbicides. This type of living mulch can reduce the chemical or mechanical input compared to conventional living mulch and, as a result, is a highly cost- and labor-saving and environmentally friendly technique. However, there is no report of the N fertilizer equivalency of white clover living mulch controlled only by mowing without herbicides.

The objectives of this study were (1) to clarify whether or not white clover living mulch supplies N to silage corn and (2) to estimate the N fertilizer equivalency of white clover living mulch to silage corn, when white clover living mulch growth is controlled only by mowing at the time of sowing corn without herbicides.

2. MATERIALS ANDS METHODS

2.1. Experimental design and cultural practices

This study was conducted from 2008 to 2009 at the National Agriculture and Food Research Organization (NARO) Tohoku Agricultural Research Center, Morioka, Japan (39°42'N, 141°09'E). The average temperature and average precipitation in Morioka were 9.3°C and 1182.8 mm, respectively. Six soil samples (0–15 cm) were randomly obtained from the field at the time of sowing of white clover (August 26, 2008). The soil was light clay (Pachic Melanudands) with a pH of 6.4, total carbon of 100.1 g kg⁻¹, total N of 6.9 g kg⁻¹, ammonium-nitrogen (NH₄-N) of 11.9 mg kg⁻¹, nitrate-nitrogen (NO₃-N) of 22.3 mg kg⁻¹ and a phosphate absorption coefficient of 20.3 g phosphorus pentoxide (P₂O₅) kg⁻¹. The soil had a high phosphorus (P)-fixing capacity because it originated from volcanic ash.

The experiment was arranged in a randomized block design with three blocks. One living mulch treatment and seven conventional cultivation treatments were randomly arranged within each block: white clover living mulch without N fertilizer application, conventional cultivation without N fertilizer application (N0), and conventional cultivation with N fertilizer application at the rate of 25 kg N ha⁻¹ (N25), 50 kg N ha⁻¹ (N50), 75 kg N ha⁻¹ (N75), 100 kg N ha⁻¹ (N100), 150 kg N ha⁻¹ (N150) or 200 kg N ha⁻¹ (N200). The recommended application rate of N for silage corn in Japan is 150 or 200 kg N ha⁻¹. Each treatment plot measured 3 m × 4 m.

On August 18, 2008, dolomite and multi-phosphate were applied to all treatments at the rate of 2000 kg ha⁻¹ for soil improvement. On August 26, 2008, white clover (T. repens L. cv. Huia) was sown in the living mulch treatment at the rate of 20 kg ha⁻¹. On May 26, 2009, white clover shoots in the living mulch treatment were clipped and left on the field. On the same day, we tilled by rotary tillage in the conventional cultivation treatments. On May 28, N as ammonium sulfate was applied as indicated above in N25, N50, N75, N100, N150 and N200. P at the rate of 300 kg P₂O₅ ha⁻¹ as superphosphate and potassium (K) at the rate of 300 kg potassium oxide (K₂O) ha⁻¹ as potassium chloride were applied to all the treatments. We applied P and K at higher levels than the recommended levels in Japan (150 or 200 kg P₂O₅ ha⁻¹ and 150 or 200 kg K₂O ha⁻¹) to mask the effect of living mulch on the P and K nutrition of corn. Fertilizers were applied as surface banding. On the same day, corn (Z. mays L. cv. 31N27: RM 125) was sown with a strip tillage seeder at the rate of approximately 75,000 plants ha⁻¹. Chemical herbicides (0.6 kg ha⁻¹ atrazine + 1.0 kg ha⁻¹ metolachlor) were applied to the conventional cultivation treatments to suppress weed growth. No herbicides were applied to the living mulch treatment.

2.2. Plant analysis

On May 29, 2009 (at the time of sowing corn) and on July 7, 2009 [at 40 days after sowing (DAS) of corn], the dry weight of white clover shoots in a 0.5 m × 0.5 m sampling quadrat was measured from each plot in the living mulch treatment. On July 7, 2009 (at 40 DAS), five corn plants were sampled randomly from each plot to investigate corn N nutrition status. For yield determination at final harvest, 3 m² of corn shoots was harvested on September 28, 2009 (at 123 DAS). All plant samples were oven-dried at 70°C for 48 h, weighed and ground. The N concentration of the samples was determined by the dry combustion method (Committee of Soil Environment Analysis 1997). The N uptake of the plant shoots was determined by multiplying the N concentration and dry weight of the shoots.

2.3. Statistical analysis

Regression analyses were used to determine the relationship between the N application rate as the predictor variable and the N concentration of corn shoots on July 7, 2009 (at 40 DAS) and the yield, N concentration of corn shoots and N uptake of corn shoots on September 28, 2009 (at harvest) as the outcome variables. Regression analyses were calculated based on mean treatment values for these parameters in the conventional cultivation treatments.

3. RESULTS AND DISCUSSION

Monthly average air temperature, precipitation and sunshine hours from August 2008 to September 2009 are shown together with the 30-year averages (1971–2000; air temperature and precipitation) or the 15-year average (1986–2000; sunshine hours) from the NARO Tohoku Agricultural Research Center in Fig. 1. The air temperature, precipitation and sunshine hours during the experimental period were similar to their respective averages except for the precipitation in July 2009. Therefore, we considered the experiment to be conducted under typical climate conditions.

Figure 1 Monthly average air temperature, precipitation and sunshine hours during the experimental period and their respective 30-year averages (1971–2000; air temperature and precipitation) or 15-year average (1986–2000; sunshine hours) at NARO Tohoku Agricultural Research Center.

In the spring of 2009, white clover covered the ground. This was sufficient to suppress weed growth without chemical herbicides. Root nodules were formed on white clover roots. On May 29, 2009, the dry weight of white clover shoots was 4416 kg ha⁻¹, which was comparable to that reported by Uozumi et al. (2004), and the N concentration was 3.92%. The amount of biomass N of the white clover shoots was 173 kg ha⁻¹. After clipping the shoots, white clover regrew and covered the ground again. On July 7, 2009 (at 40 DAS), the dry weight of white clover shoots was 2398 kg ha⁻¹ and the N concentration was 4.03%. The amount of biomass N of the white clover shoots was 95.8 kg ha⁻¹ and it was half of the biomass N of white clover shoots at the time of corn sowing.

On July 7, 2009 (at 40 DAS), in the conventional cultivation treatments, there was a quadratic relationship between the N application rate and N concentration of corn shoots (R² = 0.831, P < 0.05; Fig. 2A). On the other hand, the N concentration of corn shoots grown in the living mulch treatment was 2.22%, which was lower than that in N0 (3.42%) and the adequate concentration range (3.5–5%) at 30–45 days after corn emergence reported by Lockman (1969). Therefore, we believe that the white clover competed with corn for N, and the corn grown in the living mulch treatment was N-deficient at this stage. On September 28, 2009 (at harvest), in the conventional cultivation treatments, there were linear relationships between the N application rate and N concentration of corn shoots (R² = 0.867, P < 0.01; Fig. 2B) and between the N application rate and N uptake of corn shoots (R² = 0.924, P < 0.001; Fig. 3), and a quadratic relationship between the N application rate and yield of corn shoots (R² = 0.749, P = 0.0633; Fig. 4). Compared with the values in N0, the N uptake and yield of corn shoots in the living mulch treatment were higher by 31.8 kg ha⁻¹ and 1.75 t ha⁻¹, respectively. Based on the N concentration, N uptake and yield of corn, the fertilizer N equivalencies of living mulch were estimated to be 53, 62 and 70 kg ha⁻¹, respectively. Therefore, we believe that the white clover living mulch supplied N to corn after an early stage and might reduce the N application rate for corn production.

Figure 2 Nitrogen (N) concentration of (A) corn shoots on July 7, 2009 (at 40 DAS) and (B) corn shoots on September 28, 2009 (at harvest) in living mulch and conventional cultivation treatments (n = 3). Vertical bars represent the standard error of the mean (SEM). The dotted line represents the fertilizer N equivalencies of living mulch based on the N concentration response line.

Figure 3 Nitrogen (N) uptake of corn shoots grown in living mulch and conventional cultivation treatments on September 28, 2009 (at harvest) (n = 3). Vertical bars represent the standard error of the mean (SEM). The dotted line represents the fertilizer N equivalency of living mulch based on the N uptake response line.

The N uptake of white clover shoots at the time of sowing corn (173 kg ha⁻¹) was comparable to that of hairy vetch (Vicia villosa Roth.) (130–200 kg ha⁻¹; Smith et al. 1987). However, the fertilizer N equivalencies of white clover living mulch were lower than those of hairy vetch dead mulch (75–100 kg ha⁻¹; Smith et al. 1987). Unlike the dead mulch, living mulch used as a cover crop thrived even after the sowing of the main crop. In this study, at 40 DAS, the biomass N of white clover shoots was half of the biomass N of white clover shoots at the time of sowing corn. We think that part of the biomass N of white clover shoots at the time of sowing corn was utilized for the regrowth of white clover after corn sowing. Accordingly, the fertilizer N equivalencies of white clover living mulch were lower than those of hairy vetch dead mulch.

The increases in the corn yield from N fertilizer application were small and the yield in N0 was decent (Fig. 4). Therefore, the soil fertility in this field was high. The increase in the corn yield by living mulch was also small and we assume that this was due to the high soil fertility. The effect of N supplied by white clover living mulch may depend on the soil fertility.

Figure 4 Yield of corn shoots grown in living mulch and conventional cultivation treatments on September 28, 2009 (at harvest) (n = 3). Vertical bars represent the standard error of the mean (SEM). The dotted line represents the fertilizer nitrogen (N) equivalency of living mulch based on the yield response curve.

Wilson and Hargrove (1986) reported that the percentage of N released from crimson clover residue under no-tillage conditions (37%) was lower than that under conventional tillage conditions (60%) at 4 weeks after killing the cover crop. Furthermore, the N uptake of corn in minimal tillage is less than that in plow tillage without N application because of the low mineralization of N (Meisinger et al. 1985). Thus, we surmise that in this study the N concentration of corn shoots in living mulch was lower than that in conventional cultivation at the early stage due to: (1) the competition between corn and white clover for N because of the regrowth of white clover, as mentioned above; (2) the delay in decomposition of white clover shoots under the no-tillage conditions, and (3) a smaller mineralization of N in the white clover living mulch because of no-tillage. However, at harvest, the fertilizer N equivalency based on the N concentration of corn was 53 kg ha⁻¹ (Fig. 2B). Wilson and Hargrove (1986) reported that the difference between conventional tillage and no-tillage in the percentage of N released from crimson clover residue decreased with time. Therefore, we think that the release of N from white clover shoots increased with time.

This study showed that white clover living mulch controlled only by mowing at the time of sowing corn supplied N to corn. Based on the N uptake and yield of corn, the fertilizer N equivalencies of living mulch were calculated to be 62 and 70 kg ha⁻¹ at harvest, respectively. In our previous study, we found that white clover living mulch increased the yield of corn by facilitating arbuscular mycorrhizal fungus colonization and improving the P nutrition of corn without P application (Deguchi et al. 2007). Therefore, white clover living mulch may reduce N and P chemical fertilizer application rates for silage corn without herbicides. White clover living mulch cultivation is a useful technique for low-input sustainable agriculture and organic agriculture.

ACKNOWLEDGMENTS

We thank Mr. Yoichi Kodate for technical assistance.