窒素を多量施用した農地からの環境負荷窒素化合物の放出に関する研究 : 施肥量の多い畑地からの一酸化二窒素発生と牛ふん堆肥を多量施用した飼料イネ栽培水田からの窒素の浸透流出

Nitrous oxide （N_2O） is a greenhouse gas that has a high global warming potential and a long　atmospheric lifetime. The Intergovernmental Panel on Climate Change （IPCC） pointed out that the main　cause of the increase in N_2O concentration in the troposphere is agriculture. Therefore, it is necessary to　find ways to decrease N_2O emission from agricultural fields. However, N_2O emission has large temporal　and spatial variations. The third report from the IPCC illustrated the large range of N_2O emission（ from 2　to 21 Tg‒N yr＾‒2）. Long-term field monitoring and elucidation of the cause of N_2O emission are required for　accurate estimations of global N_2O emission. In this study, we measured N_2O fluxes in Gray Lowland soil　and Andosol during the snow-free season （from April to November） for either three or six years. The　objectives of this study were to evaluate the seasonal patterns and amounts of N_2O emission, and to　compare our results with values obtained in previous studies.　　In Gray Lowland soil from 1995 to 2000, N_2O and nitric oxide （NO） fluxes from the soil to the　atmosphere ranged from 0.00 to 1.86 mg‒N m＾‒2 h＾‒1 and from 0.00 to 3.30 mg‒N m＾‒2 h＾‒1, respectively. In the　fertilized plot of the Gray Lowland soil, the highest N_2O emissions were observed around harvesting time,　from August to October with a high rainfall frequency, as well or better immediately after fertilization in　May. In contrast, the NO flux increased immediately after fertilizer application. In the non-fertilized plots　of Gray Lowland soil, NO and N_2O flux did not increase immediately after fertilizer application, but only　N_2O flux did increase around harvesting time. The seasonal patterns of soil nitrate （NO_3^‒） and ammonium （NH_4^+） levels and the ratio of N_2O/NO flux indicated that the main process responsible for N_2O production after fertilization was nitrification and that the main process responsible for N_2O production around harvest time was denitrification. The increase in N_2O flux was enhanced by the addition of water from rainfall and of organic matter from onion planting. A significant correlation could be observed between N_2O and carbon dioxide （CO_2） flux. The cumulative N_2O flux during the snow-free season for six years ranged from 0.35 to 1.56 g‒N m^‒2, and about 70% of this flux occurred near harvesting time, from August to October. Therefore, it is necessary to monitor N_2O flux during the entire growing season in order to estimate the annual N_2O emission. In Andosol from 1998 to 2000, the N_2O and NO fluxes ranged from 0.00 to 6.42 and from 0.00 to 0.94 mg‒N m^‒2 h^‒1, respectively. N_2O flux increased markedly after only the first heavy rainfall each year, and it was higher than the N_2O flux that occurred immediately after fertilizer application. This seasonal pattern of N_2O flux from row was similar to the pattern from the furrow, even though no chemical fertilizer was applied to the furrow. The highest N_2O flux was observed after heavy rain, and an increase in NO flux was recognized only from the row. Seasonal fluctuations in NO_3^‒ and NH_4^+ concentrations in soil and in the ratio of N_2O/NO flux suggested that N_2O and NO fluxes occurring after fertilizer application（ mid-May to early July） were mainly produced by nitrification and that the N_2O emitted after heavy rain was mainly produced by denitrification. The cumulative N_2O flux during the snow-free season ranged from 0.83 to 2.33 g‒N m^‒2 over a three-year period. This flux was relatively high compared with those reported worldwide. In contrast, reported cumulative N_2O fluxes from agricultural Andosols in Japan are typically lower than those from other agricultural soils in Japan and around the world. Therefore, the results of our study suggest that high N_2O emissions may occur from Japanese agricultural Andosols with a shallow ground water level. The Gray Lowland soil and the Andosol, as described previously, had a different soil structure, especially with regard to the distribution of macropores and cracks. The influence of this difference on the production and emission of N_2O was investigated. N_2O concentration profiles were measured in two soils during the snow-free season, and N_2O flux in the soil through to a depth of 0.3 m was calculated using the gradient method （using Fick's law）. This flux was compared with the N_2O flux from the soil to the atmosphere using the chamber method. In the Gray Lowland soil, the N_2O concentration above 0.4 m increased with an increase in soil depth. In the Andosol, there were no distinctive N_2O concentration gradients in the topsoil when the N_2O flux did not increase. However, the N_2O concentration at a depth of 0.1 m increased significantly, and this concentration was higher than the concentration below 0.2 m when the N_2O flux increased significantly. The N_2O concentration profiles were thus different between these two soils. The contribution ratios of the N_2O produced in the top soil （0‒0.3 m depth） to the total N_2O emitted from the soil to the atmosphere in the Gray Lowland soil and the Andosol were 0.86 and 1.00, respectively. This indicates that the N_2O emitted from the soil to the atmosphere was produced mainly in the top soil. However, the contribution ratio of the subsoil to the N_2O emitted from the Gray Lowland soil was higher than that of the Andosol. There was a significant positive correlation between the N_2O flux in the soil through to a 0.3 m depth and the flux from the soil to the atmosphere in only the Gray Lowland soil. These results suggest that N_2O production in the subsoil of the Gray Lowland soil could have been activated by NO_3^‒ leaching through macropores and cracks, and subsequently, the N_2O produced in the subsoil might have been rapidly emitted to the atmosphere through those macropores and cracks. The soil carbon content of subsoil in the Gray Lowland soil was higher than that in the Andosol. Denitrification was prompted by an increase in the soil organic carbon; therefore, it is believed that the high carbon content and macropores in the Gray Lowland soil caused the high concentration of N_2O in the subsoil. In Japan, the annual N_2O emission from upland fields for various periods ranged from 0.01 to 0.87 g‒N m^‒2, with most measured values being less than 0.1 g‒N m^‒2. About 80 % of the measured N_2O emission worldwide was less than 0.5 g‒N m^‒2. The cumulative emissions in this study from the Gray Lowland soil and the Andosol were relatively high, compared with those reported worldwide. This suggests that the increases of the N_2O flux in the study fields after heavy rain and harvesting were equal to or higher than the increase that occurred immediately after fertilizer application. The conclusion of this study was showed in the following text. In the Gray Lowland soil and the Andosol, the N_2O emission derived from denitrification, from summer to autumn, was larger than that from the nitrification occurring immediately after fertilizer application. This might be due to the seasonal pattern of rainfall, i.e., no distinct rainy season in the early summer （immediately after fertilizer application） and most rain occurring in September （nearly harvesting）. The differences in the seasonal patterns and the amount of N_2O emission between the Gray Lowland soil and the Andosol might be due to the differences in the depth of the N_2O production in the soil and the N_2O mobility from the soil to the atmosphere. The annual N_2O emissions from both types of soils were relatively high, compared with those reported worldwide. We investigated the effects of heavy application of composted cattle manure on the leaching of nitrogen from small lysimeter paddies, where forage rice was cultivated from April 2003 to March 2007. Nitrogen leaching increased with manure application when adequate rainfall occurred after the application of cattle manure during investigated period. The amount of nitrogen leaching from the paddy to which 18 kg m^‒2 （18M-plot） manure was applied was higher than this amount from the paddy without manure application （0M-plot）. Although the dry matter yield of forage rice increased in the 18M-plot, the losses of nitrogen was high, and the excessive input caused nitrogen to accumulate in the soil. It was determined that heavy application of manure increased the environment load.