Soil tillage properties in clayey upland fields after conversion from rice paddies and the effects of soil tilth on soybean (Glycine max) growth

The self-sufficiency rate of soybeans in Japan has been determined to be only 5-7%. More specifically, in the Hokuriku region comprising Niigata, Toyama, Ishikawa, and Fukui Prefectures, soybean yields are low and unstable. The low productivity in the Hokuriku region may be due to soil characteristics in addition to climatic factors. Paddy fields occupy 89% of the croplands in this region and almost all soybean cultivation takes place in upland fields converted from rice paddies. Typical cropping systems employ rotation between rice paddy cropping and short-term upland soybean cropping using rotation patterns such as rice-rice-soybean-rice-rice-soybean. Furthermore, one-third of the farmland in this region contains clayey soils with low soil drainage and water retention capacity; obtaining ideal soil tilth and suitable water management for soybean growth is difficult under these environmental conditions. Knowledge of the physical characteristics of the soils in which upland crop-rice paddy field rotation systems are used and the effects of these soils on soybean growth is needed to attain high and stable yields in this region. The first objective of this study was to investigate the relationship between soybean yield and soil characteristics in the farmlands of the Hokuriku region to identify limiting factors that prevent high soybean yields. The second objective was to determine the effect of the transformation of iron oxides on soil microstructure and tillage properties. The third objective was to evaluate the effects of using recently developed machinery designed to achieve high seedling establishment and mitigation of water damage by tilling, ridge seeding, and compressing soil in a single process on seed imbibition and nitrogen (N) accumulation during the dry season. Soil properties and soybean growth were investigated in 33 upland fields converted from rice paddy fields in Joetsu City, Niigata Prefecture. Variations in soybean yield could be attributable specifically to soil characteristics because the subject fields were concentrated in a small plateau area with a homogeneous climate, planted with the same cultivar (Glycine max Merr. cv. Enrei), and managed by the same farmer. Fields equipped with underdrains showed significantly (P < 0.01) higher yields and a lower soil water content than fields without underdrains: the mean yields were 420 g /square m for fields with underdrains and 330 g /square m for fields without underdrains. Based on the coefficient of determination, 48% of the yield variance in the fields with underdrains was accounted for by differences in 100-seed-weight (Table 3). A significant correlation was also observed between 100-seed-weight and the amount of mineralized N in the soil (P < 0.01). In contrast, pod number accounted for 82% of the yield variance in fields without underdrains (Table 3). The yields from fields without underdrains were determined mainly by the initial growth during the period from seedling establishment to the determination of pod number. The relatively high yields from the fields with underdrains demonstrates a strong relationship between soybean 100-seed-weight and soil mineralized N. A comparison of paddy soil (Typic Endoaquepts) that contained 38% clay (mainly smectite) and 1.3% free iron oxide with a model substance that was a mixture of smectite with 85g of iron per kilogram of iron oxide revealed that the transformation of soil iron oxide affected the soil microstructure of upland crop-paddy rice rotation soils. The sediment volume (SV) of paddy soil decreased when the matric potential of the soil was less than -1.5 MPa and increased again with flooding after drying. The amount of reduced iron associated with flooding indicates that the increase in SV was dependent upon the soil reduction history and not on the state of reduction. The effect of soil drying on SV was reproduced in the model substance when the matric potential of the sample was less than -1.5 MPa (Fig. 8). Furthermore, drying reduced the SV of smectite that contained iron oxide more than smectite without iron oxide. When samples were reduced by the addition of sodium ascorbate, the SV increased only for the smectite that contained iron oxide (Fig. 10). An analysis of pore distribution and scanning electron micrographs showed that the addition of iron oxide decreased the volume of 1-micro m diameter pores and increased the volume of 100-nm diameter pores (Fig. 11). These results indicate that the aggregation of layered silicates in the presence of iron oxide caused by drying was a contributing factor to the decrease in SV and that the decreased volume could be restored by the reductive dissolution of iron oxides in the soil. Free iron oxides have a poor crystalline order in paddy fields. Poorly ordered iron has properties that allow it to react with some species of anions and change the soil microstructure by reductive dissolution. These chemical properties are unique to upland fields converted from rice paddy fields. A new method for evaluating the status of free iron oxide crystallinity was developed in this study. Free iron crystallinity was defined as the amount of iron extracted over 120min in 1M sodium acetate buffer (pH 3.0) at a solution to soil ratio of 100: 1 (Fe sub(ac)). The soil Fe sub(ac) decreased in proportion to the length of time after conversion from rice paddy to upland cultivation. The Fe sub(ac) was correlated significantly with phosphate retention properties (Fig. 15) and the iron reducibility of soils under submerged conditions (Fig. 16), but did not correlate with the amount of acid oxalate-extracted iron, which is generally used to extract amorphous free iron. The relationship between the transformation of iron oxide crystallinity and soil tillage properties was analyzed in upland crop-paddy rice rotation fields. Soil friability increased with time (0-5 years) after the conversion from paddy field to upland field (Fig. 20). The amounts of dithionite-citrate-extractable free iron and oxalate extractable iron did not change over time after conversion, but the Fe sub(ac), phosphate retention, and ferrous iron content under flooded conditions decreased gradually (Fig. 19). These results imply that the iron oxide crystallinity increased with time after conversion, and that this resulted in a decrease in reactivity with phosphate and reductive dissolution under flooded conditions. During the first year after conversion to an upland field, the Fe sub(ac) decreased gradually, but irregularly, and the changes in the ratio of ferric to ferrous iron in the Fe sub(ac) fraction did not show an obvious trend (Fig. 21). After 6 years as an upland field, the addition of organic matter before flooding increased the content of reduced iron after conversion to a paddy field. A statistically significant relationship was observed between the ferrous iron content and SV and the proportion of clods < 2 mm in size after paddling (Fig. 22). The following model is proposed for the change in soil microstructure in upland crop-paddy rice rotation fields (Fig. 23). The soil microstructure is altered in converted upland fields by the process of drying after drainage. In this soil environment, free iron oxides interact with layered silicate to form a unique microstructure. Because free iron oxides are easily reduced immediately after oxidation, flooding for a short period such as after a period of rainfall could reduce iron oxides and cause the soil microstructure to be unstable. The iron oxides and soil microstructure that are formed become more stable with time after the paddies are converted to upland fields. Soybean seed imbibition is poor in the heavy clay of upland fields after conversion from rice paddy fields due to the severe conditions of drought in the soil. These conditions are due to low water availability caused by physical properties of the soil and planting during late May to early June when the soil is very dry. The effects of soil compression on soil drying and soybean seed imbibition were examined to identify methods for improving seed imbibition. Tillage promoted soil drying. The soil water content remained higher in soil compressed by a seeder (6 kPa) to a depth of 50 mm than in uncompressed soil, and the rate of soybean seed imbibition increased significantly (Fig. 26). Furthermore, an extremely dry layer was present at a depth of 10-40 mm in the uncompressed soil. Soybean seeds are typically sown at a depth of 20-30 mm, placing them in the dry layer. Such a dry layer was not observed in compressed soils (Fig. 25), which indicates that the movement of soil water from lower to upper layers was inhibited in seedbeds with no compression due to small contact areas of soil clods. In conclusion, soil compression facilitates imbibition by promoting the movement of water from lower soil layers, and soil compression immediately after tillage promotes seed imbibition for seeds sown at depths of 20-30 mm. These results also imply that machine tilling and compressing the soil in a single process is an effective approach to improve soybean seed imbibition. Sowing on elevated ridges reduces water damage to soybean plants cultivated in upland fields converted from rice paddy fields. Therefore, the effect of ridge tillage (RT) on soybean N accumulation was investigated. The amounts of plant N derived from N2 fixation in nodules, from soil, or from fertilizer were compared between RT and conventional tillage (CT) in two replicate fields during 2002-2003. Both fields were upland fields converted from rice paddies (Typic Hydraquents). The main difference between the two fields was the presence or absence of field underdrains. The amounts of rubidium (Rb) and potassium (K) that accumulated in the shoots were also determined as an indicator of root distribution in the soil. The yields in the two fields were higher with RT by 106 and 129%, respectively, than with CT. An increased pod number and seed weight were the major factors contributing to the increased yield. An analysis of variance indicated that N2 fixation by nodules and N absorption by roots increased significantly with RT until the R1 (flowering) stage. The amounts of Rb and K that accumulated in the shoots indicate that the roots were distributed more abundantly in the upper soil layers with RT than with CT. Consequently, RT resulted in reduced water damage during the part of the rainy season that overlapped with the flowering stage. N accumulation from N2 fixation through the R7 (maturity) stage was significantly higher with RT than with CT. RT was an effective method for increasing N2 fixation by nodules in poorly drained upland fields converted from rice paddies. In conclusion, improvements to the initial stage of soybean growth are required to obtain high and stable yields from soybean grown in poorly drained heavy clay soil in the Hokuriku region. One of the best practices for improving initial growth is the modification of soil tilth in seedbeds, and high soil friability is desirable to permit the modification of tillage practices. The soil microstructure, which affects soil friability, changes gradually during the period after conversion from paddy field to upland field. This study shows that the transformation of iron oxide in upland crop-paddy rice rotation soil affected the soil microstructure, and a new method for evaluating the transformation of iron oxides was presented. This study also demonstrates that seeding and compressing the soil immediately after tilling promotes imbibition and that RT can mitigate water damage during the initial growth stage and increase N accumulation in plants. These results support the use of machinery that tills, ridges, seeds, and compresses the soil in a single process as one of the best approaches to improve soil tilth and the initial growth of soybeans.