【作者】 沙晨燕；

【导师】 陆健健；

【作者基本信息】 华东师范大学 ， 生态学， 2011， 博士

【摘要】 湿地生态系统在全球碳循环中起着重要作用。有研究表明,湿地在低氧环境中促进碳累积的同时产出温室气体——甲烷(CH4)和二氧化碳(CO2),湿地的碳源和碳汇功能近来成为全球气候变化研究关注的重点问题。对于保护和修复湿地等生态工程措施与湿地温室气体排放量之间的关系还不明确,本文以河滨地区不同类型的自然、半人工和人工河滨湿地为研究对象,尝试在不同的水文模式和植被生长状况下,研究不同湿地类型的CH4和CO2的时空排放规律及对比,探讨影响CH4和CO2排放量的主要因子及其可能的调控途径,并以此提出了针对湿地碳排放的河滨恢复湿地建造与管理建议和措施。研究内容及结果如下：研究开展于美国俄亥俄州哥伦布市Olentangy河河滨冲击平原上的Wilma H. Schiermerier Olentangy河湿地研究中心(ORWRP),研究湿地类型包括：人工种植植被的肾形淡水草泽(湿地1)、自然生长植被的肾形淡水草泽(湿地2)、河流分岔半人工牛轭湖(牛轭湖)和自然河道旁低洼湿地(河道边湿地)。于2008年11月至2009年10月,运用静态箱-气相色谱法对四种类型河滨湿地进行CH4和CO2时空动态变化的排放的测定。其中：间歇性淹没地区中气体的采集使用静态集气箱,在长久淹没区域中使用浮力箱。在实验湿地和半人工牛轭湖的采样点沿着入口至出口呈纵向分布,从干湿交错区至淹没深水区呈横向分布。在ORWRP中4种不同类型河滨湿地的甲烷排放量具有显著的时空异质性,甲烷排放速率(中值)的范围为：自然湿地(0.33-85.7 mg-CH4-C m-2 hr-1)>人工湿地(0.02-20.5 mg CH4-C m-2 hr-1)>半人工湿地(-0.04-0.09 mg CH4-C m-2hr-1)。湿地1、湿地2、半人工湿地和自然河道旁湿地土壤的CO2排放通量的中值(平均值)分别为9.8(19.2)、13.5(20.6)、24.7(36.0)和33.7 (40.3) mgCO2-Cm-2h-1。在湿地1、湿地2和河道边湿地中的甲烷排放量与土壤温度显著性相关,相关系数分别为R2=0.88、R2=0.86和R2=0.85；湿地1、湿地2和牛轭湖湿地CO2通量与土壤温度相关性显著,相关系数分别为0.63、0.54和0.67,土壤温度与CH4和CO2的季节排放模式相关。在不同类型湿地中,土壤含水率与甲烷排放量具有一定的相关性(R2=0.39)；与二氧化碳排放通量具有显著负相关性(R2=0.72)。不同类型湿地采样点中土壤碳含量与其相应的CH4和CO2排放量之间关联度都较高(R2=0.82,R2=0.69)。在同一区域淡水河滨湿地中,自然湿地的CH4和CO2排放通量均大于恢复湿地,并且不同类型湿地CH4和CO2排放的空间异质性是由于洪水冲击频率、土壤状况、地下水位及净初级生产力等因素决定的。补偿河流分岔半人工牛轭湖湿地中CH4排放通量非常低,全年的平均值仅为0.09 mg CH4-Cm-2 hr-1,这与牛轭湖湿地显著的干湿季节,较低的土壤碳含量及其大型植被的缺乏有关。牛轭湖湿地深水区和交错区域中的甲烷排放量分别为0.03(0.06)和0.03 (0.12) mg m-2 hr-1,无显著性差异(P=0.593)；深水区的土壤呼吸通量中值(平均值)(15.5 (20.9) mg CO2-C m-2 h-1)明显小于交错区(38.7(63.0) mg CO2-C m-2 h-1) (p<0.005)。在两个实验湿地中,淹没深水区比干湿交错区有更高的甲烷排放量(P=0.000)。对于二氧化碳排放通量而言,深水区、交错区以及高地的土壤二氧化碳排放量的中值(平均值)分别为-2.2(-10.5)、41.8(53.7)和75.2(93.1) mg CO2-C m-2 h-1 (P=0.000)。并且对人工湿地中甲烷排放速率和二氧化碳排放速率显著性相关(R2=0.52,p<0.05)。建造生态系统适合的水文条件,在湿地的重建和修复项目中也是关键因子之一。通过设计具有既长又浅型状和湿-干-湿交替水文条件的人工湿地,不仅可应用于其它的河流冲积平原的修复,也可在不同的气候和水文条件下检验其生态服务功能。湿地1和湿地2有相似的环境因子,自然生长植被的人工湿地CH4排放通量仍明显大于人工种植植被湿地的排放通量,其年排放量分别为68 g CH4-C m-2y-1和114 g CH4-C m-2 y-1 (p=0.047),这是由于湿地2的累积净初级生产力较湿地1高,和湿地1相对于湿地2有着较高的植被生物多样性。通过2004年至今的监测可得出,实验湿地中的甲烷排放量在过去的5年中持续增长,这是由于湿地中植被的残留物增多,湿地土壤中更多的碳含量。因此发现植被的净初级生产力对湿地甲烷排放影响较大,在恢复湿地中,可以通过种植不同种类植物,增加物种多样性,这样的设计和管理措施可以有效地减少CH4和CO2排放。更多还原

【Abstract】 Wetlands are important ecosystems involved in global carbon cycle. Wetlands are both producers and consumers of the greenhouse gases. Controlling methane (CH4) and carbon dioxide (CO2) emissions from temperate zone wetlands created and restored for habitat replacement and water quality improvement is important. The objective of our study is to estimate and compare temporal and spatial patterns of methane emissions from wetlands and riparian ecosystems with different vegetations types and hydroperiods; to filter out the environmental parameters controlling net CH4 and CO2 emissions. For the study took place in the same area, it is better for the quantitative study of CH4 and CO2 emissions.In the research conducted for this dissertation, investigations were carried out in riparian wetlands at the Wilma H. Schiermeier Olentangy River Wetland Research Park (ORWRP) in Columbus, Ohio, USA. The 20 ha ORWRP includes several wetlands that are differing in vegetation type and water conditions, which are human-planted experimental freshwater marshes (wetland 1), naturally colonized experimental freshwater marshes (wetland 2), a river diversion oxbow (oxbow) and a riverside bottomland (riverside). A non-steady-state chamber design was used for gas sampling, with permanent chamber bases located in dry and shallow water zones, and a portable floating chamber deployed in deeper, permanently inundated zones. Sampling locations were chosen according to hydrology and different types of wetlands. Gas fluxes were measured from November 2008 to October 2009. Plots were distributed along longitudinal (from inflow to outflow) and transverse (from shallow transition edges to deepwater open water zones).At ORWRP, CH4 and CO2 emissions varied remarkably in both temporal and spatial terms. The range of median value methane emissions:riverside (0.33-85.7 mg-CH4-C m-2 h-1)＞wetland 1 and wetland 2 (0.02-20.5 mg CH4-C m-2 h-1)＞oxbow (-0.04-0.09 mg CH4-C m 2 h-1); The median (average) values of CO2 emission rates for wetland 1, wetland 2, oxbow and riverside were 9.8(19.2),13.5(20.6),24.7(36.0) and 33.7(40.3) mg CO2-C m-2 h-1 respectively. Soil temperature had a significant relationship with CH4 emissions in wetland 1 (R2=0.88), wetland 2 (R2=0.86) and riverside (R2=0.85), while the relationship was not significant between CH4 emissions and soil temperature in oxbow siteSoil temperature had a significant relationship with CO2 emissions in wetland 1(R2=0.63), wetland 2 (R2=0.54) and oxbow (R2=0.67) as well. There was a negative relationship between CO2 emissions and soil water content in different types of wetlands (R2=0.72). Nature wetlands have the higher CH4 and CO2 emission rates than created wetlands in river riparian zone here. Overall, our results showed that the edge of a river in a bottomland hardwood forest had the much higher CH4 and CO2 emissions than did created river diversion marshes. The spatial variation of the different types of riverine wetlands is caused by a combination of flood frequency, sediment organic carbon content, groundwater fluxes, and wetland productivity.Methane fluxes from the created oxbow were extremely low, with no more than 0.09 mg CH4-C m-2hr-1. The oxbow had distinct wet and dry seasons led to the low CH4 emissions. Another reason for the low CH4 production at the oxbow may be related to its low soil C content and the lacks of emergent vegetation. There was a seasonal pattern of CO2 emissions. The CH4 emission rates of open water zone and transition zone of oxbow were not different (p=0.593), were 0.03 (0.06) and 0.03 (0.12) mg m-2 hr-1. The CO2 emission effluxes was significantly lower in open water zone (15.5(20.9) mg CO2-C m-2 h-1) than in transition zone (38.7(63.0) mg CO2-C m-2 h-1). For the two created wetlands, there were significantly high rates of CH4 and CO2 emissions from deep water zones compared to transition zones during steady-flow conditions (p=0.000).Thus, hudrologic dynamics must be carefully planned in created and restored wetlands. It would be worth replicating this wetland design with in its long and shallow shape and wet-dry-wet pulsing conditions throughout the world’s river floodplains.When the two experimental wetlands were compared, the natural-colonizing wetland has more methane emissions than human-planted wetland (p=0.047), which were 114 g CH4-C m-2 y-1 and 68 g CH4-C m-2 y-1. The reason may be due to its history higher net primary productivity and the higher biodiversity of wetland 1. We also found that from 2004 to 2009, mean annual methane emissions for wetland 1 increased from 16 g CH4-C m -2 y-1to 68 g CH4-C m-2 y-1, and for wetland 2, from 31 g CH4-C m-2 y-1to 114 g CH4-C m-2 y-1, maybe for the cumulative productivity and higher carbon content in soil. Comparison among the nature-planted and human-planted wetlands suggested that methane emissions from created freshwater riverine wetlands greatly depend on the NPP and hydrology. Riparian created wetlands can be designed to emit less CH4 and CO2 gas possibly by providing the proper vegetation development.更多还原