Chapter1: Introduction

Nitrogen (N) is one of the essential nutrients in agriculture. Over the past decades, synthetic N fertilizer application contributed to significant increase in food production. However, intensive use of N fertilizer has caused deterioration in water quality and losses in biological diversity. The recent N fertilizer price rise due to increase in fossil fuel prices has jeopardized economic viability at farm level. To meet increasing demand for food, better N management strategy is opted for in agricultural production.

To improve N use, emphasis has been placed on enhancing congruence of N application with crop demand under synthetic fertilizer based management, Conventional Management (CNM). The other approach is to use the natural internal N cycling within agricultural farm ecosystems as N source. In other words, no synthetic fertilizer is added. This ecosystem based approach is referred to as Agroecological Management (AEM).

In my thesis, I examined possible ways to improve N use of both AEM and CNM in lowland rice (Oryza sativa) farming. Regarding AEM, there is little research available on lowland rice cultivation. Therefore, I took field science approach, on-farm research, to investigate soil N supply and Physiological Nitrogen Use Efficiency (PNUE) under AEM in comparison with CNM. To improve the efficiency of N application under CNM, Japan was examined as a case study in which there have been improvements in N application. I also investigated lowland rice farming in Benin, where CNM has recently become popular, in order to identify factors retarding N use efficiency under less favorable conditions.

Chapter 2. Soil N supply under Agroecological Management

In this chapter, I investigated soil N supply and its uptake by rice plants under AEM in comparison with CNM. The research was conducted in Tochigi prefecture, Japan for 2010 and 2011. The principal study fields were AEM-99, where AEM has been practiced since 1999, AEM-09, where AEM has been practiced since 2009, and CNM. For AEM, human N inputs were rice straw and rice bran. For CNM, slow release synthetic N fertilizer was used with supplemental application of cow manure. Apart from soil fertility management, all the other cultural practices were kept in the same manner.

Estimated soil N supply in AEM-99 was 19.3 g N m-2 and 28.1 g N m-2 in 2010 and in 2011, respectively. In both years, AEM-99 supplied sufficient N to satisfy N demand by the rice plants, as shown by the N uptake of rice as 11.4 g N m-2 in 2010 and 16.7 g N m-2 in 2011. Comparison between AEM-99 and AEM-09 indicated lower soil N supply under AEM-99. Components in AEM maintaining soil N supply were rice straw, rice bran, spring and winter weeds, and soil drying effect before submergence. In AEM-99, a large difference of soil N supply was observed between 2010 and 2011. It may be attributed to higher soil drying effect with less rainfall during the winter before submergence in the latter year.

Chapter 3. Physiological N Use Efficiency under Agroecological Management and Conventional Management

With the same field and research design as in chapter 2, I investigated differences in the use of N absorbed by rice plants grown under AEM and CNM. To examine differences in yields and physiological development among the fields, plant growth characteristics, leaf senescence, and yield components were measured in 2010 and 2011.

Under the record high temperature in summer 2010, PNUE for filled grains was higher in AEM-99, 48.4 g (g N)-1, than in CNM, 35.6 g (g N)-1. Difference in the growth characteristics between AEM-99 and CNM implies that higher PNUE in AEM-99 was most likely caused by relatively high N contents in leaves at heading, vigorous root development, and slow leaf senescence during the ripening phase. Although crop management was unchanged between 2010 and 2011, rice plants in AEM-99 were more susceptible to lodging in 2011 than in 2010. This was most likely attributable to high N uptake estimated from leaf color and growth during the vegetative stage in 2011. PNUE in AEM-99 was also higher than in CNM. However, PNUE in 2011 was lower than that in 2010. It was caused by lodging and high N concentration in panicles which may have resulted from larger amount of mineralized N supply in 2011.

Chapter 4. Transition to N efficient Conventional Management in lowland rice farming in Japan

CNM is the dominant soil fertility management for lowland rice in Japan. In this chapter, I reviewed the literature on lowland rice farming to examine how the transition to more N efficient CNM took place in Japan.

Between 1990 and 2007, national average Nitrogen Application Efficiency (NAE) for lowland rice increased from 50 kg (kg N)-1 to 70 kg (kg N)-1. Three agronomic factors that improved this efficiency were varietal change to high palatability rice with less N content in grains, development of slow release fertilizer and more efficient application method, and improved crop management practices eliminating yield reduction. During the period, socioeconomic changes in the Japanese agro-food system took place. Relative advantages of N efficient practices that emerged from these changes may have facilitated rice farmers' adoption of these practices.

Chapter 5. Nitrogen Application Efficiency in lowland rice farming in Central-Southern Benin

The field survey study was conducted in Benin during the 2010-2011 dry season. I investigated the variation of yields, NAE, and factors affecting the variation based on the crop management information and soil properties.

Rice yields ranged from 1.3 to 7.8 Mg ha-1 with average yield of 4.8 Mg ha-1. NAE ranged from 17.2 kg (kg N)-1 to 191.4 kg (kg N)-1 with the average NAE of 55.3 kg (kg N)-1. Multiple regression analysis revealed that 75% of the variation in yield could be explained by seven factors: water stress, residue management, plowing method, rat damage, N application rate, field slope, and sand content in the soil. The study indicated that insufficient rice management skills decreased NAE among the low yielding plots. This phenomenon was pronounced in the newly developed irrigation scheme. Among high yielding plots, sand content was identified to explain low NAE. Declining in yields despite high N apply implies soil N loss.

Chapter 6. General discussion

In this chapter, I synthesized the findings from previous chapters. Insufficient N supply has been reported as the major limitation of AEM in previous researches. However, this study demonstrated that AEM in lowland rice cultivation could supply sufficient N to rice plants. Under AEM, soil N supply was based on internal N cycling through recycling N in rice straw and rice bran, retaining N by weeds during the fallow period, and facilitating dry soil effect. Capacity of N cycling in supplying sufficient N to rice plants suggests that AEM is a viable soil fertility management option without posing further burden on terrestrial N deposition. Under high temperature, growth characteristics of AEM might have increased heat tolerance against spikelet sterility. The study also demonstrated that high fluctuation of soil N supply under AEM due to climatic condition. Sensitivity to environmental factors emphasizes the importance of management skill in controlling soil N mineralization. It warrants further research in understanding the mechanism of N mineralization in paddy field especially when temperatures exceed 30℃.

Under less favorable condition, such as in the case of Benin, findings of this study imply that there are several ways to improve N use. The primary importance lies in avoiding yield loss through farming management improvement. Efficient N application method like site specific N management could also facilitate recovery of N applied as synthetic fertilizer. Furthermore, strengthening internal N cycling within agricultural ecosystems could provide alternative or supplementary N source, which would lead to efficient use of available N on farm. It would be especially important for resource poor farmers who have difficulties in purchasing synthetic fertilizer.

背景および目的

窒素は,農業生産において最も重要な土壌養分である.従来,化学合成肥料は食料の大幅増産に貢献したが,その一方で過剰の窒素が環境中に放出されて,水汚染などの環境問題を生じた.また,化学肥料窒素に依存した農業が,原油価格高騰に脆弱であることも最近の事象が示している.今後,長期的には化石燃料資源の制約が強まる中で,増え続ける食料需要に応えるためには,環境への窒素負荷を減らしつつ農作物の窒素吸収量を増やすことが重要な課題である.申請者は,水稲栽培における窒素の「農業生態的管理」と「慣行的管理」の二つのアプローチについて,化学合成窒素への依存を減らし,農業生態系内の窒素利用効率を高めることで,上記課題解決に資するために,本研究を行った.

農業生態的管理における土壌窒素供給

2010年と2011年の2年間,栃木県下都賀郡野木町所在の農家水田で現地実験を行った.実験には,化学肥料を用いずに稲わらと米ぬかの施用を主体とする農業生態的管理を1999年以来継続してきたAEM-99区,同じ農業生態的管理を2009年から始めたAEM-09区,そして緩効性窒素肥料と牛ふん堆肥を施用する慣行的管理のCNM区の3区を設け,土壌窒素の供給と水稲による窒素の吸収を調べた.調査の結果,農業生態的管理を10年以上継続したAEM-99区では,CNM区に匹敵する十分量の窒素が土壌から供給されるが,農業生態的管理を始めて間もないAEM-09区では窒素供給が不十分であることが分かった.なお,2011年は2010年よりも土壌窒素供給量が50%ほど多く,前年の作柄や気象の年々変動に伴って土壌窒素発現量が大きく変動することも分かった.

水稲の窒素利用効率に及ぼす農業生態的管理の効果

上記と同じ農業生態的管理および慣行的管理を行った水田における,水稲(品種:コシヒカリ)の生育・収量を,圃場での測定と植物サンプリングによって調べた.水稲の生育は,CNM区と比べてAEM-99区では,初期生育が緩慢だが生育後期は葉面積が大きく葉の枯れ上りが少ない,いわゆる「秋まさり」傾向を示した.いっぽう,農業生態的管理を始めて間もないAEM-09区は,初期生育が劣るばかりか後期の生育もCNMとAEM-99区に及ばなかった.収量調査の結果,記録的な猛暑年であった2010年には,どの区でも稔実と登熟の減退が見られたが,AEM-99区では稔実歩合が他の区よりも有意に高かった.その結果,2010年の精モミ収量はAEM-99区がCNM区を上回り,従って精モミベースの窒素利用効率はAEM-99区がCNM区よりも高かった.いっぽう,AEM-09区は収量が他の区よりも明らかに低かった.2011年も,AEM-99区の窒素利用効率がCNM区よりも高いことは変わらなかったが,その値自体は2010年よりも明らかに低く,2011年に多発した倒伏の影響がうかがわれた.

日本の水稲栽培にみる慣行施肥管理の窒素利用効率向上

統計資料の分析と文献レビューにより,日本の水稲栽培における慣行施肥管理の窒素利用効率(=玄米収量kg /施肥窒素kg)が,1990年の50 kg/kg Nから,2007年の80 kg/kg Nへと向上した理由を調べた.その結果,施肥技術の改良,良食味品種への転換,栽培管理の改善による収量損失減少の3つが主な要因と考えられた.

西アフリカ ベナンにおける水稲栽培の窒素利用効率

2010年から2011年にかけての乾季に,ベナン国中南部所在の6つの小規模灌漑区の総計59水田において,水稲への施肥量,収量,土壌の性質などを調べた.その結果,モミ収量は1.3t/haから7.8 t/haと大きな変異を示し,窒素利用効率(モミ収量kg/施肥窒素量 kg)は17 kg/kg Nから191 kg/kg Nとさらに大きな変異を示した.収量と施肥窒素量の間には相関が無く,従って施肥以外の要因によって窒素利用効率が低下している場合のあることが推察された.実際にも,虫害や野鼠害,病害などによる減収が,低収量の原因となっている場合が確認された.その他の場合にも,砂質土壌や土壌肥沃度の低下が,低収量ひいては低窒素利用効率をもたらしていると考えられた.

以上の研究結果は,水稲栽培において化学窒素肥料に依存しない農業生態的管理により,慣行的管理と遜色ない窒素供給が可能であること,それをアフリカの水稲栽培に生かす可能性があることを示す一方で,土壌窒素発現が年々変動するといった課題も明らかとした.

このように,本論文が水稲栽培の窒素利用効率を向上させるために,農業生態的管理における土壌窒素供給を中心に,実際の圃場で解明したことは,学術上,応用上貢献するところが大きく,よって審査委員一同は本論文が博士(農学)の学位論文として価値あるものと認めた.