【作者】 傅志强；

【导师】 黄璜；

【作者基本信息】 湖南农业大学 ， 作物信息科学， 2007， 博士

【摘要】 研究水稻植株通气组织与甲烷（CH₄）排放相关性为培育高产低甲烷排放的水稻品种提供理论与实践依据，同时也为通过栽培措施影响水稻植株通气组织来达到减排甲烷的目的提供理论基础。特别是近年来，随着水稻直播栽培技术的提高，直播稻面积越来越大；高产水稻品种的选育十分重视，超级稻品种的推广面积逐年增加。所以对直播稻以及高产水稻品种的甲烷排放观测也很重要，为高产水稻品种、直播稻的大面积推广提供环境学支撑具有重要的现实意义。2005～2006年早、晚共4季观测了39个不同水稻品种以及移栽、撒播、条播、穴播4种不同栽培方式的CH₄排放通量，计算了CH₄季节排放总量与季节排放均值，分析和比较了不同品种、不同栽培方式下的甲烷排放差异；测定了不同水稻品种、不同栽植方式下的水稻植株植物学特征；观测了稻鸭共栖复合生态系统中不同养鸭数量的稻田土壤特性及甲烷排放通量。主要结果如下：1不同水稻品种之间CH₄排放差异显著早稻常规稻品种湘早籼12号CH₄排放量是常规稻郴早糯1号的1.42倍；杂交早稻金优706是威优402的1.17倍；晚稻常规品种余赤231-8是湘晚籼12号的1.35倍，杂交晚稻T优259是金优207的1.24倍。早、晚稻常规稻CH₄排放量极差分别为8.31g／m²、4.85 g／m²，杂交稻分别为2.96 g／m²、3.18g／m²；常规稻CH₄排放变幅大于杂交稻。不同类型之间CH₄排放存在显著差异，早、晚稻常规稻CH₄排放量均值分别高出杂交稻3.66g／m²、0.68g／m²，杂交早稻比常规稻降低了18.4％，杂交晚稻比常规稻减少了3.1％。因此，种植杂交稻CH₄排放量低于常规稻。2水稻植株通气组织与稻田CH₄排放密切相关2005年早、晚稻观测了植株地上部茎秆通气组织与甲烷排放通量，分析了通气组织与CH₄排放相关性。结果表明：早稻期间，植株第1节间直径、第2节间茎秆维管束面积、茎秆壁厚以及叶鞘横切面积／节间横切面积、第3节间叶鞘横切面积／节间横切面积、茎壁横切面积／节间横切面积与CH₄排放量相关性达到显著性水平；晚稻期间：株高与CH₄排放量达到0.001的极显著性水平；第3节间叶鞘横切面积／节间横切面积、茎秆维管束面积／茎秆横切面积、第2节间维管束面积／节间横切面积、第1节间维管束总面积／茎壁横切面积、叶鞘维管束面积／茎壁横切面积均与CH₄排放量达到0.01极显著水平；此外，第1～5节间还有17个数量指标与CH₄排放量达到0.05显著性水平。根据上述相关数量特征确定聚类指标，对早、晚稻品种甲烷排放量进行模糊聚类分析，早、晚稻品种聚类结果与实测结果吻合度分别达到87.5％和90.0％。对水稻植株与CH₄排放相关的因子进行主成分分析，得出株高、维管束、气腔、茎壁等4个主因子，以这4个因子对水稻品种甲烷排放进行聚类，与实测结果吻合度达到83.3％。分析2006年不同栽培方式的甲烷排放通量与水稻根系特性相关性，分蘖盛期的根系生物量、体积、白根数量、根系活力、根系吸收总面积以及活跃吸收面积均与甲烷排放通量呈显著正相关关系。3稻叶气孔特性影响稻田CH₄排放早稻品种剑叶气孔密度、剑叶气孔总数量与拔节～孕穗期甲烷排放通量呈显著正相关；晚稻品种常规稻倒二叶气孔总个数与CH₄排放通量均值、倒三叶气孔总个数、上三叶气孔总个数与拔节～孕穗期甲烷排放通量相关性分别达到0.05、0.05、0.01水平；剑叶气孔总面积与抽穗～齐穗期甲烷排放通量均值呈显著负相关，倒二叶气孔总面积与抽穗～齐穗期甲烷排放通量均值呈显著正相关。直播稻剑叶气孔密度、气孔总数量与抽穗期甲烷排放通量呈显著负相关；不同栽培方式下的叶面积指数存在显著差异，CH₄排放通量与叶面积指数呈显著正相关关系。因此，稻叶气孔特性与甲烷排放通量之间存在一定程度的相关性，影响到稻田甲烷排放；同时也证实了王明星等认为稻叶气孔是甲烷排放通道的观点。4不同栽植方式下的CH₄排放量差异显著基于本田生长期早、晚稻直播与移栽甲烷排放数量比较，直播稻甲烷排放季节均值低于移栽稻，但直播稻甲烷排放季节总量高于移栽稻。2005年早稻移栽稻比直播稻低1.24g／m²，减少了8.5％；2005年晚稻移栽稻比直播稻低1.02g／m²，减少了3.7％；2006年早稻创丰1号抛栽稻甲烷排放量比直播稻减少7.5％，超级稻抛栽比直播要低3.8％；2006年晚稻移栽分别比撒播（80苗／m²）、条撒（80苗／m²）、穴播（2苗／穴）减少了28.9％、54.3％、34.1％。不同撒播密度、不同穴播密度甲烷排放量存在显著差异，而且密度越大，甲烷排放量越大。栽培密度对CH₄的排放通量影响很大。2006年晚稻移栽稻单位经济产量甲烷排放量为37.62g CH₄／kg稻谷，直播稻超出移栽稻17.7％。2006年早稻超级稻直播单位稻谷产甲烷量高于抛栽稻4.84 gCH₄／kg稻谷，常规稻创丰1号直播高于移栽稻3.48 gCH₄／kg稻谷，常规稻不论直播与移栽也都要高于超级稻。基于相同面积相同时间直播与移栽CH₄排放数量比较，早稻甲烷排放总量及季节排放均值排序为常规稻直播＞常规稻抛栽＞超级稻直播＞超级稻抛；晚稻为直播＞移栽。杂交稻与常规稻直播无论早稻还是晚稻单位稻谷的甲烷产生量要高于抛（移）栽稻，常规稻要高于超级稻。综合分析水稻生产与甲烷排放，撒播密度、穴播密度分别以80苗／m²、4苗／穴为最优选择。5不同养鸭数量的稻田CH₄排放量差异显著在水稻不同生育时期，20只鸭／667m²稻田水体溶解氧含量与常规稻田相比，早稻增加了2.2％～68.7％，晚稻增加了11.1％～110.8％；产CH₄细菌种群数量20只鸭／667m²早稻田土壤低于对照38.9％～51.6％；晚稻田土壤低于对照69.8％～90.2％。与对照比，早稻土壤还原物质总量、活性还原物质以及活性有机还原物质降幅分别为0.322～1.118、0.239～0.689、0.203～0.548 cmol kg^-1；晚稻降幅分别为0.171～0.628、0.220～0.591、0.157～0.433 cmolkg^-1。早稻期间，10只鸭／667m²、15只鸭／667m²、20只鸭／667m²处理甲烷排放量分别比对照降低了18.2％、26.1％、28.1％；晚稻期间，降低了17.7％、27.6％、34.4％。环境经济效益分析表明，养鸭数量越多，经济效益越高。10只鸭／667m²、15只鸭／667m²、20只鸭／667m²分别比常规稻作增加纯收入为：2644元／hm²、2830元／hm²、3006元／hm²；环境经济收益分别增加2986、3197、3390 yuan／hm²。但从水稻生产的角度来考虑，20只鸭／667m²影响到了水稻生长发育，比常规稻作显著地减少了产量，而10只鸭／667m²处理比常规稻作增加了产量；15只鸭／667m²对水稻产量影响不大，与10只鸭／667m²处理相比增加了环境经济效益。稻田应以水稻生产为主体，在不影响水稻产量的前提下考虑增加其它附加值，提高经济效益。因此综合考虑水稻生产、生态与经济效益，稻田生态种养模式中养鸭数量以15只鸭／667m²为最优选择。6建立了基于稻鸭共栖系统土壤特性的CH₄排放定量模型（1）土壤还原物质与CH₄排放相关性：CH₄排放通量与还原物质总量间的线性回归方程为：Y=2.1138X+1.7552，相关性达到0.05显著水平（R=0.8397）；与活性还原物质总量、活性有机还原物质总量回归方程分别为Y=3.6161X-0.3435，Y=5.9046X+1.9865，相关性都达到0.01显著水平（R=0.9250，0.9208）。表明CH₄排放通量与土壤还原性物质含量呈显著正相关。（2）水体溶解氧含量与CH₄排放相关性：以早、晚2季不同养鸭数量处理下的CH₄排放通量季节均值为依变量，以田面水中溶解氧含量季节均值为变量，其回归方程为：Y=-1.5276X+14.7707，相关性达到0.001显著水平（R=-0.9390），二者呈真实线性负相关关系。（3）土壤产甲烷细菌与CH₄排放相关性：早稻期间，CH₄排放通量与产甲烷细菌数量间线性回归方程为：Y=11.4949+0.5088X（R=0.7798）；晚稻期间，CH₄排放通量与产甲烷细菌数量间的线性回归方程为：Y=12.9218+0.0558X（R=0.8037），相关性均达到了0.001显著性水平。表明CH₄排放通量与土壤产甲烷细菌种群数量呈显著正相关。稻鸭共栖复合系统中通过鸭子的活动增加水体溶解氧含量、减少产甲烷菌种群数量、降低土壤还原性物质含量，从而显著减少了甲烷排放。养鸭数量越多，溶氧含量越高，产甲烷菌数量及还原性物质减少幅度越大，导致土壤中CH₄气体生成量减少，氧化量增加，CH₄排放通量减少。更多还原

【Abstract】 CH₄ is one of the main greenhouse gases, with importance second only to CO₂. Ricepaddies are an important human-made ecosystem for the global CH₄ budget. Methane transportcan occur via diffusion, gas bubble release and via the aerenchyma of rice plants. Of the threepathways, Over 90％CH₄ emitted to the atmosphere through aerenchyma transport, which isconsidered the most important when rice plants are involved in the ecosystem. Therefore, it iscrucial to research the correlation between methane emission from paddy soil and plantaerenchyma tissues. To understand the relationship is useful for breeding rice varieties with highyielding and low methane emission, as well as by culture measures influence rice plantaerenchyma to reduce methane emission from paddy soil. In addition, with the area of directsowing rice and high yielding rice cultivars increasing, especially super hybrid rice cultivarsbreed and spread quickly, it is vital to measure their methane emission flux and it is full of praxissignification to provide the environmental basis for direct sowing rice and super hybrid ricedevelopment.CH₄ of 39 rice cultivars, including the conventional rice, hybrid rice and super hybrid ricewere determined using the static chamber technique and calculated the amount of methaneemission in the whole season and averaged seasonal methane emission flux. Then methaneemission from the different culture methods were conducted, including translating rice （TR）,broadcast sowing rice （BSR）, drill rice （DR） and hole sowing rice （HSR） and the differences ofmethane emission were analyzed. In the rice-duck complex ecosystem, the optimal number ofducks was confirmed to gain the objective of increasing the economic benefits and reducing theenvironmental cost on the base of comprehensive consideration the rice production, economicand environmental benefits. The main results were as follows.1. The differences of methane emission among rice cultivars were significant. Methane ofthe conventional early rice Xiangzaoxian No.12 was 1.42 times higher than that of ChenzaonuNo.1 and for hybrid early rice, Jinyou No.706 was 1.17 times higher than that of Weiyou No.402.The highest methane emission of late rice cultivars was Yuchi231-8, which was 1.35 timeshigher than that of Xiangwanxian No. 12, and for hybrid late rice, Tyou259 was 1.24 times higherthan that of Jinyou No.207. Methane emission differences were significant between theconventional rice and hybrid rice as the same as between non-glutinous rice and Indicia type rice. Methane emission amount of early and late conventional rice were by 3.66g/m², 0.68 g/m² morethan that of hybrid rice, respectively. Therefore, CH₄ emission of hybrid rice is lower than that ofthe conventional rice.2. Aerenchyma tissues of plant culms above ground were observed and the correlationshipsbetween aerenchyma characteristics and methane emission were analyzed. The results showedthat the significant relativities existed in methane emission and many quantitative indexes ofplant tissues, such as the highest internodes plant diameter, stem vascular bundles area, stem wallthickness and the area ratio of sheath transverse section and internodes transverse section of thesecond internodes during early rice season. During late rice season, the correlation between plantheight and methane emission was significant at 0.001 level. The area ratio of sheath transversesection and internodes transverse section, the area ratio of stem vascular bundle and stem walltransverse section of the third internodes, the area ratio of all vascular bundles and internodestransverse section of the second internodes, the area ratio of all the vascular bundles and stemwall transverse section, the area ratio of all the sheath vascular bundles and stem wall transversesection of the highest internodes, which were significant correlation with methane emission at0.01 levels. In addition, 17 quantitative indexes from the highest internodes to the fifthinternodes were significant with methane emission at 0.05 levels.With above mentioned correlative indexes, fuzzy clustering analysis method was applied tothe methane emission amount of rice cultivars of early and late rice. The consistent degrees ofclustering results with measured results of early and late rice were 87.5％, 90％, respectively. Itwas showed the clustering indexes were appropriate in distinguishing methane emission of ricecultivars to high, medium and low type. The principal component analysis method was applied tothe correlation factors of plant tissues and methane emission, the main factors of plant height,vascular bundles, air chambers and stem wall were determined and by the four main factors,clustering the cultivars with methane emission, the consistent degree was 83.3％comparing withthe observed results.To further analysis on the correlation between methane emission and rice rootcharacteristics, root biomass, volume, number of white roots, root activity, root total absorbingarea and active absorbing area all presented the significant positive correlation with methaneemission. Therefore, it can be concluded that rice root characteristics impacted methane emissionvia rice plant aerenchyma system largely.3. Analyzing on the correlation between rice leaf stoma and methane emission flux, theearly rice flag leaf stomata density and the stomatal number were significant positive correlationwith methane emission flux at the booting stage and for late rice cultivars, the number of thesecond leaf stoma of the conventional rice were significant correlation with average seasonal methane emission flux at 0.05 levels, and the number of the third leaf stoma, with the bootingstage methane emission flux at 0.05 level, the number of up three leaves stoma with bootingstage methane emission flux at 0.01 level.To late rice cultivars, total flag leaf stomata area was significant negative correlation withmethane emission flux at the full heading stage.However, total stomata area of the second leafwas significant positive correlation with methane emission flux at the full heading stage. TheDSR flag leaf stomata density and the total number of stoma presented significant negativecorrelation with methane emission flux at the early heading stage. Therefore, it was affirmativethat rice leaf stomata characters were correlative with methane emission and leaf stomatainfluenced methane emission from paddy field through plant aerenchyma tissues. This resultproved the viewpoint that rice leaf stoma opening and closing impacted the daily variationpattern of methane emission.4. Based on the methane emission from paddy field during rice growth stages,thedifferences of amount methane emission were significant between direct sowing rice（DSR） andtranslating rice（TR）, as well as among three direct sowing methods. There existed manydifferences in sowing date, growing days in the field and growth stages between DSR and TR.So, difference of methane emission existed in the same days or in the same stage, which resultedin without comparison in methane emission between DSR and TR at the same time. In thisexperiment, methane emission fluxes of different cultural methods were measured at the intervalof three days and total amount methane emission calculated by the Inner-integral-sum method.Then the average values in the whole season were figured out by the growth days. It was feasibleto compare their amount of seasonal methane emission and average seasonal methane emissionflux of DSR and TR.The results showed that the seasonal average CH₄ flux of direct sowing rice were lower thanthat of translating rice. However, amount CH₄ emission of DSR was more than that of TR.Amount CH₄ emission of TR reduced by 1.24 g/m² comparing with DSR and by 1.02 g/m² in2005 early and late rice respectively. Amount CH₄ of the conventional early TR was low by 7.5％,and super hybrid TR was reduced by 3.8％in 2006 early rice. Amount CH₄ emission of TR werereduced by 28.9％, 54.3％, 34.1％comparing with the broadcast sowing rice with the density of 80seedlings/m² （BSR）, drill rice with the density of 80 seedlings/m² （DR）, hole sowing rice with 2seedlings per hole（HSR）, respectively. The significant differences also existed in differentbroadcast rice densities or hole sowing densities, moreover, more density, higher methaneemission flux. In addition, methane emission of unit rice yielding of DSR was more than TR. In2006 late rice, methane emission of unit rice yielding of DSR was 37.62 g CH₄/kg rice, andenhanced 17.7％comparing with that of TR. During 2006 early rice, methane emission of unit rice yielding of the conventional DSR enhanced by 4.84 g CH₄/kg rice more than that of TR, forsuper hybrid rice, DSR increased by 3.48 gCH₄/kg rice higher than that of TR and theconventional rice was higher than the translating rice either DSR or TR method.Based on the methane emission of the DSR and TR in the same area and days, during earlyrice season, Comparing the total amount of methane emission and average CH₄ emission, theconventional DSR was the bggest and second to the conventional TR, next to the super hybridDSR and the super hybrid TR was the less.The methane emission of unit rice yielding for DSRwas higher than the TR and the conventional rice was higher than the super hybrid rice.Synthetically considering rice production and methane emission, the density of 80 seedlings/m²and 4 seedlings/hole were the optimal patterns.5. CH₄ emission fluxes among different number of ducks existed significant differences inthe rice-duck ecosystem. Dissolution oxygen content of the water body and soil were increasedas result of ducks activities and looking for food, which resulted in number of methanogenicbacteria lower significantly at the maximum methane emission stage. A t the same time, ducks inthe fields controlled the inefficacy tiller numbers, reduced number of withered leaves. Therefore,methane emission flux reduced significantly with the conventional culture rice. Number ofmethanogenic bacteria under 20 ducks per 667m² treatment paddy soil was reduced by 38.9％～51.6％in early rice and by 69.8％～90.2％in late rice comparing with that of non-ducks paddysoil. Comparing with the non-ducks paddy soil, during early rice, the redox matter, the activeredox matter and the active organic matter were lower at the range of 0.322～1.118,0.239～0.689, 0.203～0.548 c mol kg^-1, and for the late rice 0.171～0.628, 0.220～0.591, 0.157～0.433c molkg^-1, respectively. Therefore, amount methane emission of 10ducks per 667m², 15ducks per667m², 20ducks per 667m² were reduced by 18.2％, 26.1％, 28.1％during early rice and for laterice, were17.7％, 27.6％, 34.4％comparing with that of non-ducks rice soil, respectively.Moreover, ducks were more in the field, dissolution oxygen content in the water body werehigher, but number of methanogenic bacteria in paddy soil and soil redox matter content werelower, which resulted in lower CH₄ emission flux.Environmental and economic benefits analysis showed that more ducks in the rice-duckecosystem resulted in more economic benefits. The net incomes of 10 ducks per 667m², 15 ducksper 667m², 20 ducks per 667m² increased by 2644 yuan/hm², 2830 yuan /hm², 3006 yuan/hm², respectively and environmental economic benefits increased by 2986 yuan/hm², 3197yuan/hm², 3390 yuan/hm² more than that of non-ducks paddy fields, respectively. Comparingwith the conventional culture rice yielding, the treatment of 20 ducks per 667m² impacted ricegrowths and development, reduced rice yielding significantly, and that 10 ducks per 667m²treatment enhanced rice yielding. 15 ducks per 667m² treatment did not influence rice growth and development, at the same time its environmental and economic income was more than 10ducks per 667m². Rice paddy must give priority to rice production and add other affixationvalues in order to increase economic benefit in the premise of ensuring rice production.Therefore, synthetically considering rice production, environmental and economic benefits, 15ducks per 667m² was the optimum pattern in the rice-duck ecosystem.6. This thesis was analyzed the correlation between soil characteristics and methaneemission in the rice-duck ecosystem. Firstly, the significant correlation existed between the soilredox matter content and methane emission flux. The linear regression equation between the totalredox matter（X） and methane emission flux（Y） was Y=2.1138X+1.7552 and their relativity weresignificant at 0.05 level（R=0.8397）, and for soil active redox matter（X）, active organic redoxmatter（X） were Y=3.6161X-0.3435, Y=5.9046X+1.9865, respectively, and their relativities weresignificant at 0.01 levels（R=0.9250, 0.9208）, respectively.Secondly, the significant negative correlation between water body dissolution oxygencontent and methane emission flux. The regression equation between methane emission（Y） anddissolution oxygen content（X） was Y=-1.5276X+14.7707, and their relativity were significant at0.001 level （R=-0.9390）.Thirdly, the significant positive correlation existed between methane emission flux andnumber of soil methanogenic bacteria. During early rice, the regression equation between CH₄emission flux （Y） and number of methanogenic bacteria（X） was Y=11.4949+0.5088X, and laterice was Y=12.9218+0.0558X, and their relativities were significant at 0.001 level （R=0.7798,0.8037, respectively）. During early and late rice, the number of soil methanogenic bacteria variedconsistently with methane emission flux variation from paddy soil. Therefore, soil methanogenicbacteria were important factor to control methane production and emission.The results showed that the ducks activities added the dissolution oxygen content of waterbody, reduced number of methanogenic bacteria and soil redox matter content, which resulted inlower methane emission flux significantly. Withal more ducks, higher dissolution oxygen content,lower number of methanogenic bacteria and redox matter content. Therefore, amount of CH₄production reduced and methane oxidation increased, then methane emission flux was lower.更多还原