【作者】 朱咏莉；

【导师】 吴金水；

【作者基本信息】 中国科学院研究生院（教育部水土保持与生态环境研究中心） ， 土壤学， 2005， 博士

【摘要】 稻田生态系统碳循环对大气温室气体的吸收/排放以及全球气候变化起着不可忽视的影响。该生态系统与大气间CO₂交换特征及其影响因素是碳循环研究的重要内容,其不仅是碳循环过程机理和调控机制以及模型模拟的需要,而且可以为估算和评价稻田生态系统碳源/汇强度及其对大气CO₂浓度变化的贡献提供科学依据。以我国亚热带丘陵区稻田生态系统为研究对象,采用涡度相关技术对该系统与大气间CO₂交换通量进行了连续观测,通过分析能量平衡的闭合程度对涡度相关法观测稻田生态系统通量数据的可靠性进行了评价;分析了稻田生态系统CO₂通量在不同时间尺度上（日、季节）的变化特征及其影响因素;估算出了稻田生态系统与大气间CO₂的年交换通量;并对箱式法在稻田生态系统CO₂吸收/排放通量观测中的应用作了初步的探讨。通过研究,取得的主要结论有以下几个方面:（1）采用涡度相关技术对亚热带丘陵区稻田生态系统CO₂通量进行观测的过程中发现该系统能量平衡存在不闭合现象。在假定常规的有效能量（Rn-G）测定正确的前提下,涡度相关法测定的湍流通量（LE+H）结果偏低。能量平衡比率在5～8月份平均为0.85,表现为能量不闭合程度较高;1～4月份和9～12月份平均为0.92,表现为不闭合程度较低;年能量平衡比率为0.87,平均不闭合程度为13%。这表明该方法在稻田生态系统通量观测中的可靠性相对较高。（2）稻田生态系统与大气间CO₂的交换（NEE）具有明显的日变化规律。在水稻生长季节,白天稻田生态系统以吸收CO₂（为负值）为主,夜间则表现为排放CO₂（为正值）。光辐射和温度是影响CO₂通量日变化的主要环境因子。白天CO₂通量对光量子通量密度（PPFD）变化的响应过程可以用直角双曲线方程进行描述。随PPFD的增加,CO₂通量（绝对值）呈增加趋势,但当PPFD>1000μmol/m²/s时,CO₂通量变化比较稳定。在水稻不同生育期,CO₂通量对光强的响应存在较大差异,其中以水稻生长旺盛期的光能利用效率（a）和最大光合速率（Pmax）最高。早稻和晚稻生长期CO₂通量对PPFD的响应存在较大差异,晚稻各生育期的a值明显高于早稻。（3）在摩擦风速（U*）大于0.1 m/s的情况下,稻田生态系统夜间呼吸（Reco,包括植株呼吸和土壤呼吸）速率随温度的升高呈指数增加。5 cm土层温度（T5）可以作为与Reco进行拟合的温度指标。早稻生长季Reco对温度变化的响应明显较晚稻生长季敏感。此外,排水对稻田CO₂通量变化也产生影响,其导致稻田生态系统夜间排放CO₂量增加,白天净吸收CO₂量减少。土壤湿度是排水期稻田生态系统CO₂通量变化的关键影响因素。（4）稻田生态系统CO₂通量具有明显的季节变化动态。叶面积指数（LAI）是影响CO₂通量季节变化的重要因素,其与水稻日光合吸收CO₂总量（GPP）之间呈显著的正相关关系。水稻生长季净吸收的CO₂总量与其生物量的变化趋势相一致。晚稻生长季稻田生态系统从大气中净吸收CO₂量（NEE绝对值）约为319 g C/m²,明显高于早稻生长季的净吸收量（232 g C/m²）。（5）稻田生态系统CO₂通量的年变化特征表现为6～9月份较高,1～5月和10～12月较低的对称分布。光合有效辐射（PAR）和日平均气温（Ta）是影响CO₂通量年变化的主要环境因子,二者与CO₂通量之间的关系可以用二元线性方程进行拟合。稻田生态系统年GPP、Reco和NEE分别为5861.3 g/m²、3385.8 g/m²和-2475.6 g/m²,表明亚热带区域稻田生态系统是大气CO₂的汇,年净吸收CO₂强度接近2.5 kg/m²。年NEE的计算结果明显受到U*临界值和与夜间呼吸通量进行拟合所选取的温度指标的影响。（6）提出了对透明箱法观测的CO₂通量进行计算的指数一级动力学拟合方法（ER）。该方法可以有效地解决常用的线性拟合方法（LR）在植被同化速率较强的晴天条件下净吸收通量计算结果偏低的问题,这表明ER法可以作为箱式法观测包括植被同化过程在内的CO₂通量的计算方法。（7）箱式法观测的稻田CO₂排放通量与气温和土壤温度（0、5、10和15 cm）之间均存在显著的指数关系。CO₂累计排放量与水稻地上生物量存在极显著的正相关关系,可以用幂函数表示。稻田生态系统CO₂-C净收支随水稻移栽后天数呈幂函数关系增加。更多还原

【Abstract】 Carbon cycle in paddy ecosystem strongly affects the uptaking /emitting of greenhouse gases and global climate change. CO₂ exchange between the ecosystem and the atmosphere is a key part of carbon cycle.Measurements of the net exchange of carbon dioxide between paddy ecosystem and the atmosphere not only benefit to understand well the mechanism of carbon cycle and its modeling and evaluating, but also help to determinate the annual carbon source or sink strength of the paddy ecosystem and to assess its contribution to the budget of CO₂ in the atmosphere, especially in subtropical region.In the research, CO₂ fluxes from paddy ecosystem in subtropical hilly region were measured continuously using eddy covariance technique. The objectives were to assess the accuracy of eddy covariance method, investigate the variation of CO₂ fluxes on daily and seasonal temporal scales, analyze the relationship between CO₂ fluxes and environmental factors, and to quantify the annual net ecosystem exchange （NEE） from the paddy ecosystem. Moreover, application of chamber method in the observation of CO₂ fluxes from paddy fields was investigated.The main conclusion and innovations were showed in the following:（1） Energy imbalance was found during the observation and measurement of CO₂ fluxes using eddy covariance technique in paddy ecosystem in subtropical hilly region. The sum of sensible and latent heat （H+LE） might be underestimated if （Rn-G） was accurate. The energy balance ratio （EBR） of （H+LE） to （Rn-G） was averagely 0.85 from May to Aug, which was lower than that of other periods （0.92） during a year, that is, the extent of energy balance closure was relatively lower from May to Aug. The annual EBR was 0.87 on average with a mean imbalance of 13%. It showed that eddy covariance technique could be reliablely applied in the observation and measurement of CO₂ fluxes.（2） The NEE between paddy ecosystem and the atmosphere was the balance of photosynthesis and respiration processes. A notable diurnal pattern of CO₂ fluxes was observed, with uptaking CO₂ from the atmosphere （negative value） during the daytime and emitting CO₂ to the atmosphere （positive value） in the nighttime. Photosynthetic photon flux density （PPFD） and temperature were the main factors for the daily change of CO₂ fluxes. A rectangular hyperbolic light-response function could be used to describe the relationship of CO₂ flux and PPFD. The absolute values of CO₂ fluxes increased with the increment of PPFD. When PPFD was higher than 1000μmol m^-2 s^-1, light saturation was observed. The carbon dioxide fluxes response differently to light in different growing stages of rice. In the blooming stage, the quantum yield （a） and the maximum rate of photosynthesis assimilation （Pmax） were higher than that in tillering and ripening stages. Moreover, these light response parameters in late rice growing season were general higher than that in early rice growing season.（3） In nighttime, respiration from soil and plants （ecosystem respiration, Reco） with U* （friction velocity） >0.1 m s-1 changed exponentially with the increase of soil temperature at the depth of 5 cm （T5）. Reco during the early rice-growing season was more sensitive to temperature than that during the late rice-growing season. Moreover, CO₂ fluxes were affected by drainage. When the paddy was drained, net CO₂ uptake from the atmosphere in daytime was less, and in nighttime, CO₂ emission was greater than when the paddy was flooded. Soil moisture was proved to be the dominant factor for controlling CO₂ emission during the drainage period.（4） Leaf area index （LAI） was a critical influential factor of the seasonal pattern of daily CO₂ flux. A significant positive correlation was found between LAI and the daily gross primary production （GPP）. In addition, the cumulative GPP was consistent with the change of rice total biomass. During the growing season of late rice, the absolute value of NEE was about 319 g C m^-2, which was higher than that during early rice growing season (about 232 g C m^-2).（5） The annual trend of daily values of GPP, Reco and the absolute value of NEE behaved higher from Jun. to Sep. and lower during the other monthes. Photosynthecially active radiation （PAR） and mean daily air temperature （Ta） were two main influential factors for controlling the annual trend of GPP and NEE. The response of GPP and NEE to PAR and Ta could be described by binary linear functions, respectively. The annual GPP, Reco and NEE in paddy ecosystem were 5861.3 g CO₂ m^-2, 3385.8 g CO₂ m^-2 and -2475.6 g CO₂ m^-2, respectively. It showed that paddy ecosystem in subtropical region was a sink of atmospheric CO₂ with a net absorbing rate 2.5 kg m^-2 a-1. However, the estimated annual NEE was strongly affected by friction velocity threshold and the reference temperature used in the respiration-temperature function. （6） A new method with an order kinetics equation （ER） was put forward to simulate the change rate of CO₂ concentration versus measurement time and to calculate CO₂ flux observed by closed chamber method. This method could solve the lack of linear regression method （LR） to a great extent, which was used very often now but usually underestimated CO₂ flux in sunny daytime. ER method was proved to be a new feasible means to observe and calculate CO₂ flux including photosynthesis process.（7） CO₂ emission rates measured by closed chamber method appeared exponential relationships with air and soil temperatures at the depth of 0, 5, 10 and 15 cm, respectively, which was consistent with the result from the eddy covariance measurement. There was a very marked power function relationship between CO₂ cumulative emission amount and rice biomass. CO₂-C net uptaking from atmosphere increased with rice growing by power function.更多还原