【作者】 李成芳；

【导师】 曹凑贵；

【作者基本信息】 华中农业大学 ， 生态学， 2008， 博士

【摘要】 氮（N）是作物生长必需的营养元素，然而过量的N肥施用不仅带来肥效的降低，而且导致一系列不良的环境反应，研究稻田N素的转化及动态规律，对提高N的有效利用及其降低N的环境危害具有重要意义。本研究通过野外采样、室内分析测定、室内培养试验和田间小区实验，利用静态箱技术、自制测渗计技术和密闭式酸吸收法等手段，对稻鸭和稻鱼共作生态系统稻田N的矿物固定与固定态铵的释放；稻田土壤可溶性有机N的生态学效应；田面水N库与土壤N库相互转化；作物生长与N素转化的联系及其系统N素损失进行了较为系统的研究。主要研究结果如下：1、相对于常规稻作，由于鸭和鱼的存在，稻鸭、稻鱼共作降低了土壤pH，显著提高了0．02-1mm的砂粒和细砂粒含量，显著降低了<0．002mm粘粒的含量，使得稻田土壤水稳性团聚体数量增加，土壤团聚化程度加强，因此改善了土壤质地。稻鸭、稻鱼共作提高了土壤全N、全P含量，增幅分别为4％～7％和4％～13％，显著地增加了土壤NH₄⁺、速效P含量，而对土壤NO₃^-影响不大。2、土壤固定态铵受鸭、鱼活动，施肥和水稻生长多种因素影响。施肥促进土壤对铵的固定，土壤固定态铵含量随着土壤交换性NH₄⁺和pH增加而增加；相对于常规稻作，鸭和鱼的存在显著地提高了土壤固定态铵含量，其增幅为4％～5％；土壤固定态铵含量与>0．2mm．的砂粒含量和<0．002mm的粘粒含量成极显著或显著相关；水稻吸N量与土壤固定态铵呈显著负相关，与土壤固定态铵的释放量呈显著正相关，水稻吸收促进土壤固定态铵的释放。因此，相对于常规稻作，稻鸭、稻鱼共作提高了土壤N含量和水稻吸N量，降低了土壤固定态铵的释放，为创造了一个对作物潜在有效的N库。3、在水稻全生育期，土壤微生物量N表现为先升后降，并于成熟期有所回升；土壤脲酶、脱氢酶和蛋白酶活性表现为先升后降，过氧化氢酶活性变化不大。相对于常规稻作，由于鸭子和鱼的活动及其生活粪便作用，稻鸭、稻鱼共作显著提高了土壤微生物量N含量、土壤脲酶活性、脱氢酶活性和蛋白酶活性，其增幅分别为7％～8％、8％～13％、13％～17％和10％～14％。相关分析表明，土壤微生物量N与土壤速效N、土壤全N、全P和水稻吸N量之间不相关，土壤脲酶和脱氢酶活性与土壤速效N负相关，土壤酶活性与土壤全N、全P不相关，土壤脲酶、脱氢酶和蛋白酶活性与水稻吸N量呈显著相关，土壤微生物量N与土壤酶活性不相关。4、稻田土壤SON是水稻吸收、微生物吸收和N淋失的交互作用的综合反映。在水稻全生育期，土壤SON与土壤无机N呈显著正相关，与水稻吸N量呈显著负相关；而由于水稻对N的吸收和N的下渗淋失，土壤SON与土壤微生物量不相关。相对于常规稻作，由于鸭和鱼的存在，稻鸭、稻鱼共作生态系统土壤SON含量显著降低了7％-12％。DON是稻田渗漏水的主要N形态。相对于常规稻作，稻鸭、稻鱼共作减少了土壤SON的潜在淋失。5、NH₄⁺是田面水无机N素的主要形态；同时，相对于处理常规稻作，稻鸭稻鱼共作显著降低田面水pH，显著提高NH₄⁺浓度，而TN浓度有所增加，而对NO₃^-无明显影响；稻鸭共作显著提高了DO浓度，而稻鱼共作显著降低了DO浓度。田面水中NH₄⁺/TH在施肥后第3天达到最大，随后降低，而所有处理NH₄⁺／TN的均值相当，表明稻鸭、稻鱼共作不会增加TN中NH₄⁺的比重，因此不会提高氨N为形态的相对流失潜力。渗漏水中NO₃^-是无机N淋失的主要形态；相对于处理常规，稻鸭、稻鱼共作渗漏水NO₃^-和TN浓度降低，而渗漏水NH₄⁺无明显变化。稻鸭、稻鱼共作肥料N潜在淋失率分别为2．72％、2．58％，低于处理常规稻作（2．99％），表明稻鸭、稻鱼共作可以减少施入N肥潜在的下渗淋失，同时稻鱼共作减少N肥淋失的效果好于稻鸭共作。6、由于鸭子和鱼的存在，相对于常规稻作，稻鸭、稻鱼显著提高稻田田面水总P浓度、溶解P浓度和土壤速效P含量及水稻植株对P的吸收，而土壤全P有所增加。在水稻全生育期，稻鸭、稻鱼共作系统田面水总P浓度、溶解P浓度、土壤全P和速效P含量在施肥后达到最大值，此后随水稻的生长逐渐降低，表明施P肥后一周是控制稻田P流失的关键时期；同时，对P的环境效应分析表明，在稻鸭、稻鱼共作期间，要注意避免农田排水和防止因降雨引起的田面水外溢。此外，由于鸭子和鱼的活动提高了土壤有效养分含量，降低化肥的施用量，进而降低了化肥损失所造成的环境危害。7、各处理N₂O排放具有类似的变化模式；N₂O排放峰值出现在施肥后2星期和稻田落干期。与常规稻作相比，由于鸭鱼的存在，稻鸭共作生态系统N₂O释放量显著增加为8％-13％，稻鱼共作生态系统N₂O释放量则显著减少了4％-5％。在稻田淹水期，N₂O排放与温度变化不相关，与土壤有效N和pH相关；稻田排干后，N₂O排放与温度变化、土壤有效N和pH不相关，表明在稻田淹水期，N₂O排放受土壤氧化-还原层的硝化-反硝化作用影响，而在稻田落干期，N₂O排放受土壤温度、水分含量和pH共同影响。稻鸭共作提高了系统N₂O的增温效应，而稻鱼共作降低了系统N₂O的增温效应。8、尿素的施用促进NH₃的挥发，尿素施用后一周NH₃挥发量占总挥发量的79％-87％，说明了施肥后一周是控制NH₃挥发的关键时期。对影响NH₃挥发的因子分析表明，NH₃通量与土壤pH、田面水pH、田面水NH₄⁺浓度呈显著或极显著相关，与温度和土壤脲酶活性不相关。相对于常规稻作，由于鸭子和鱼的存在降低了田面水pH，稻鸭、稻鱼共作降低了稻田NH₃挥发。因此，稻田养鸭养鱼能够降低N肥的损失率，进而提高N肥的利用率。9、施肥和水稻吸N是影响N平衡最主要的因素，NH₃挥发、降雨和灌溉水也是影响N平衡的重要因素，而N₂O释放、N淋失和鸭和鱼N的输出对平N衡的贡献不大。与常规稻作不同，稻鸭和稻鱼共作N平衡为正，表明了鸭和鱼的存在加速了土壤有机养分的周转，显著地提高了水稻N的输出。更多还原

【Abstract】 Sampling on fields, laboratory incubation and determination and field experiments were conducted to study dynamics and availability of soil fixed ammonium, ecological effects of soil soluble organic N, interactions among different formations of N in floodwater and soil, N losses in the ecosystems and correlations between N transformation and crop growth by a static chambers technique, a closed acid trap method, and a field lysimeter and so on. The main results are summarized as follows.1. Compared with conventional rice paddies （CK）, due to the presence of ducks and fish, integrated rice-duck ecosystem （RD） and rice-fish ecosystem decreased soil pH, and significantly increased the contents of 0.02-1 mm clays and significantly decreased the contents of <0.002mm clays, which enhanced the amounts of water-stable aggregates in soil, and thus improved the texture of soil.Compared with CK, RD and RF significantly increased the contents of total N and total P, exchangeable NH₄⁺, and available P, and had no effects on the content of NO₃^-.2. Fertilization and rice growth are of influencing factors to control fixation of soil ammonium. Application of N fertilizer promoted fixation of soil ammonium while rice growth boosted the release of fixed ammonium. During rice growth, most of recently fixed ammonium was released while native fixed ammonium was not released.Compared with CK, RD and RF significantly increased the content of fixed ammonium. Exchangeable NH₄⁺, pH and texture in soil affected the content of soil fixed ammonium. Close correlation existed between soil NH₄⁺, pH and fixed ammonium content at the significant level of p<0.01; Soil temperature （at 5cm depth） was not associated with fixed ammonium content; Soil fixed ammonium content was positively related to >0.2mm clay content and negatively related to <0.002mm clay content at the significant level of p<0.05 or p<0.01, and not associated with 0.02-0.2mm and 0.002-0.02mm clay content. The study also showed that N uptake of rice was significantly negatively related to the content of soil fixed ammonium and positivelyrelated to the release amount of soil fixed ammonium.Compared with CK, RD and RF increased the content of soil N and N uptake of rice, and decreased the release of soil fixed ammonium, and thus created a more significant N sink for added fertilizer.3. The experimental results showed that during rice growth, soil microbial biomass nitrogen （MBN） content increased firstly after transplantation, and subsequently decreased and slightly increased at rice autumn. Moreover, the soil enzymes activities ascended at the early stage and then declined slightly except that catalase activity slightly changed. Linear regression analysis showed that no close correlations existed between soil MBN and soil enzymes activities, soil nutrients （soil available N （NH₄⁺ + NO₃^-）, total N and total P） and N uptake of rice, and between soil enzymes activities and soil total N and total P. Moreover, Soil urease and dehydrogenase activities were closely related to soil available N, and N uptake of rice was closely related to soil urease, dehydrogenase and protease activities.Compared with CK, RD and RF significantly enhanced soil MBN content and urease, dehydrogenase and protease activity but not affected catalase activity.4. Soil soluble organic nitrogen （SON） constituted a predominant reservoir of soluble N in paddy fields. Correlation analysis indicated that during rice growth SON was significantly positively related to SIN （p<0.01）, and significantly negatively related to N uptake of rice. Due to N-uptake by rice and N leaching, soil SON was not closely related to MBN. The results also showed that due to the presence of ducks and fish, compared with CK, RD and RF significantly decreased the content of soil SON.In the forepart periods of rice growth, contents of different formations of N all were highest. Dissolved organic N （DON） was the main formation of N in percolation water in paddy fields; moreover, statistical analysis indicated that compared with CK, RF significantly decreased the leakage of SON while RF slightly decreased the leakage of SON.In conclusion, soil SON was regulated by the interactions of N uptake of rice plants, microbial uptake and leaching in paddy fields during rice growth.5. NH₄⁺ was the predominant form of nitrogen in floodwater while NO₃^- was the predominant form of nitrogen in percolation water; moreover, fertilization could increase content of all form N in floodwater and percolation water. Compared with CK, RD and RF significantly increased the content of NH₄⁺, slightly enhanced the content of TN in floodwater, and significantly decreased pH in floodwater and didn’t affect the content of NO₃^- in floodwater. Moreover, RD significantly increased the concentration of DO in floodwater while RF significantly decreased the concentration of DO in floodwater. Furthermore, RD and RF decreased the concentrations of NO₃^- and TN in percolation water and didn’t change the concentration of NH₄⁺ in percolation water. In the 3rd day after fertilization, the ratio of NH₄⁺/TN was the most, and thereafter decreased. In addition, mean NH₄⁺/TN ratios of different treatments were almost equivalent.The results also indicated that the potential maximal radios of N fertilizer leaching in RD and RF were 2.72% and 2.58%, respectively, lower than 2.99% in CK, which demonstrated that RD and RF could reduce leakage of N fertilizer and the effect of reducing.Leakage of N fertilizer was better in RF than in RD.6. The results indicated that the P concentrations of soil and floodwater reached peaks immediately after P fertilizer applied, and then decreased in 1 week after fertilization. Compared with CK, RD and RF could significantly enhance total P concentrations of floodwater, dissolved P concentrations of floodwater, available P concentrations of soil and P uptake of rice plants, and slightly increase total P concentrations of soil. Moreover, close relationships existed between dissolved P concentrations of floodwater, available P concentrations of soil and P uptake of rice plants. Analysis on environmental impacts of floodwater P showed that it was a pivotal time in 1 week after fertilization to control P losses by leaching from paddy fields. Moreover, during ducks and fishes reared, drainage of water from paddy fields and leaching due to runoff should be controlled. Since RD and RF could enhance contents of soil available nutrient, leading to decrease the consumptions of fertilizer, which reduced detrimental impact on the environment.7. The results showed that with the same amount of urea applied as basal fertilization, N2O emission fluxes from these treatments followed a similar seasonal variation trend. Our experimental data showed that in 2 weeks after urea application and after drainage peaks of N₂O emission flux occurred. Compared with CK, N₂O emissions in RD significantly enhanced, which were 1.08-1.13 times of that in CK, and N₂O emissions in RF siginificantly decreased, which were 94%⁹5% of that in CK. Correlation analysis indicated that during the flooding seasons, N₂O emission flux was not correlated with temperature, but significantly related to soil inorganic nitrogen （SIN） （p<0.01） and soil pH （p<0.01）. After drainage, N₂O emission flux was not correlated with temperature, SIN and soil pH. These observations showed that during the flooding season N₂O emission might be influenced by nitrification and denitrification due to the presence of oxidized and reduced soil layers. On the other hand, after drainage, N₂O emission might be affected by the interreactions of soil pH, water content and temperature during Nitrification.We evaluated the integrated global warming potentials （GWPs） of integrated rice-duck cultivation system based on N₂O emission, which showed RD could enhance the GWP based on N₂O while RF could reduce the GWP based on N₂O Compared with CK.8. NH₃ volatilization was measured by a closed acid trap method in the ecosystems. The results indicated that urea application could promote volatilization of NH₃. It is the key time to control volatilization of NH₃ in the first week of urea application. Correlation analysis showed that volatilization flux of NH₃ was closely related to pH in soil and floodwater and NH₄⁺ in floodwater, and not related to temperature and soil urease activities.Compared with CK, due to a drop of pH in floodwater by the presence of ducks and fish, RD and RF reduced volatilization of NH₃. Analyses on N losses from paddy fields indicated that most of nitrogen was lost by NH₃ volatilization in paddy fields; moreover, RD and RF could decrease fertilizer N loss rates, and thus promote fertilizer N use efficiency.9. The water balance analysis indicated that the amount between water input and output was imbalance due to seepage. Moreover, the N balance analysis showed that fertilization and N-uptake by rice were the main factors of N balance while N outputs via NH₃ volatilization and N input via precipitation and irrigation were also significant. In contrast to CK, due to the presence of ducks and fish, the apparent N balances were positive in RD and RF, suggesting that the presence of ducks and fish accelerated the turnover of soil organically bound nutrients and thus significantly increased N output from rice uptake.The results also showed that the presence of ducks and fish did not reduce the total yields of rice grain.更多还原