【作者】 王萍；

【导师】 国庆喜；

【作者基本信息】 东北林业大学 ， 生态学， 2009， 博士

【摘要】 集成生物圈模型（IBIS）是目前最复杂的基于动态植被模型的陆面生物物理模型之一,它集成了大范围的生物物理、生理以及生态过程,通过一种综合考虑各主要过程之间作用的模块化结构将陆地表面生物物理、陆地碳通量和全球植被动态表达出来,并且这种模式框架能够直接与大气环流模式（AGCMs）进行耦合。该模型从设计上就体现了植被既在分钟至世纪尺度上受到大气化学和气候变量的影响,同时也在同样时间尺度上反馈于大气化学和气候。本文选择IBIS模型作为研究对象,对模型进行了相应的改造并结合气象、植被数据等相关资料,将模型应用于东北林业大学帽儿山森林生态系统定位研究站,并利用植被的野外实测数据对模拟结果进行了验证,对次生林碳通量的模拟中,IBIS模型模拟的CO₂通量与实测通量之间相关性比较理想。在此基础上,将模拟从局部地点扩大到区域尺度上,对2004-2005年间大小兴安岭森林生态系统碳通量,包括净第一性生产力（NPP）、土壤呼吸、净生态系统生产力（NEP）等的时空分布格局进行了较为全面的定量估算,这不仅对于评估大小兴安岭森林生态系统碳收支非常关键,而且为最终建立适用于我国森林生态系统的耦合模型提供基础。通过研究,主要得到以下几方面的认识:大小兴安岭森林植被年均NPP值为494.7gC·m^-2·a^-1,年吸收0.06Pg的大气碳。从研究区内不同植被类型单位面积NPP平均水平来看以温带软阔叶林最高,为577.2gC·m^-2,从年NPP的总量来看,以温带针阔混交林植被的年NPP最高,为17.001×10⁶tC·a^-1,占NPP总量的27.16%。研究区年均NPP的空间分布主要受热量条件的影响,大兴安岭地区基本上呈现出由北向南增加的趋势,小兴安岭地区除单位面积年均NPP大于1.1KgC·m^-2·a^-1在小兴安岭北部孙吴和逊克地区分布外,基本上呈现出均匀分布的趋势。大小兴安岭森林植被年均土壤呼吸总量39.262×10⁶tC·a^-1相当于0.04PgC·a^-1,约占东北森林年均土壤呼吸总量(0.19PgC·a^-1)的21%。单位面积年均土壤呼吸值为317.3gC·m^-2·a^-1,是东北森林平均水平558.2gC·m^-2·a^-1的57%。从土壤呼吸的平均水平来看,以温带软阔叶林为最高为412.5gC·m^-2·a^-1,从年土壤呼吸的总量来看,以寒温带落叶针叶林植被的年土壤呼吸最高,为10.947×10⁶tC·a^-1,占土壤呼吸总量的27.88%。大小兴安岭森林生态系统土壤呼吸空间分布的基本特点是:大兴安岭地区呈现出由北向南部减少的趋势,小兴安岭地区基本上呈现出均匀分布的趋势。大小兴安岭森林单位面积年均NEP值为313.4gC·m^-2·a^-1,每年净固定碳量为38.779×10⁶tC相当于0.04PgC,约占东北森林年均NEP总量(0.15PgC·a^-1)的27%。从不同植被类型单位面积NEP的平均水平来看,以温带针阔叶林最高,为370.8gC·m^-2·a^-1,从年NEP的总量来看,以温带针阔混交林植被的年NEP最高,为11.016×10⁶tC·a^-1,占NEP总量的28.25%。研究时段内大小兴安岭森林表现为一个碳汇,大兴安岭地区除单位面积年均NEP大于0.6KgC·m^-2·a^-1在松岭区和加格达奇分布外,基本上呈现出由北向南增加的趋势;小兴安岭地区除单位面积年均NEP大于0.6KgC·m^-2a^-1在小兴安岭北部孙吴和逊克地区分布外,基本上呈现出均匀分布的趋势,其单位面积年均NEP值在0-0.6KgC·m^-2a^-1之间。研究区研究时段内单位面积年均NEP空间变化显著,大兴安岭地区除加格达奇和松岭区显著减少外,这一时段内的NEP的增长区和减少区呈现交错分布;小兴安岭地区除孙吴和逊克地区减少外,其余地区都呈增加趋势。整体上看,全区在波动中有所增加,72.9%（1247个站点）的地区植被年均NEP呈增加趋势,27.1%（452个站点）的地区年均NEP减少。经分析主要因为该时段内,加格达奇和松岭区降水量的下降导致NPP减少明显,同期的土壤呼吸量却显著增加造成NEP减少。IBIS的模拟结果通过与实测资料和其他模型的对比以及森林清查数据的验证表明:IBIS的模拟结果基本落实在实测值的范围之内,说明模拟结果比较准确。当然,限于目前区域时空尺度陆地碳循环模拟研究的水平和发展阶段以及获取模拟所需数据的困难性,研究工作还需要从模型的参数化、区域尺度的观测和数据获取转换等方面得以改进和完善。最后指出将宏观的格局分析与微观的机制研究有效地结合起来是未来碳循环研究的一个重要的发展方向,在本文工作的基础上,经过完善和改进的IBIS模型可以进一步与区域气候模式（大气环流模式）相耦合,在陆面生物物理一大气相互作用、区域气候变化机理及陆地生态系统碳循环等研究领域具有广泛的应用前景。更多还原

【Abstract】 The Integrated Biosphere Simulator （IBIS） is one of the most sophisticated models in simulating terrestrial biosphere processes based on dynamic vegetation schemes. IBIS is designed to integrate a variety of terrestrial ecosystem phenomena within a single, physically consistent model that can be directly incorporated within AGCMs. To facilitate this integration, the model is designed around a hierarchical, modular structure and uses a common state description throughout.For the purposes of validating IBIS, climate and vegetation database have been set up, and the IBIS, which has been improved.With supports of database and IBIS model, we applied it in Mao’er mount forest ecosystem orientation research station of Northeast forestry university, comparisons between simulated secondary forests carbon fluxes and actual measurement data showed that the model was capable of simulating CO₂ flux. On the basis of this analysis, simulation would extend to regional scale form local spot. Carbon fluxes, such as net primary production （NPP）, soil respiration and net ecosystem production （NEP） from 2004 to 2005 have been quantified in the studying area. This study aimed to not only evaluate forest ecosystem carbon budgets in study areas and provided a basis for establishing coupling model which was applicable to forest ecosystem of China. It estimated results have suggested that:In this study, using the inventory and meteorological data in Daxing’anling and Xiaoxing’anling of northeastern China during 2004 - 2005 and IBIS, we quantified the net primary production （NPP） and its spatio-temporal distribution in this region. The mean NPP was 494.7gC·m^-2·a^-1,absorbing 0.06PgC·a^-1 from the atmosphere in the whole region. The highest vegetation type of the annual NPP per unit area was temperate softwood forest, which was 577.2gC·m^-2.The highest vegetation type of the total annual mean NPP was temperate mixed coniferous and broad-leaf forest (equivalent to 17.001×10⁶tC·a^-1), which accounted for 27.16%.The spatial distribution of the mean NPP was affected by heat regime, which increased from north to south in Daxing’anling region, and well-distributed in Xiaoxing’anling region except Sunwu and Xunke.Based on datasets and IBIS, the terrestrial soil respiration and its spatial-temporal change in Da and Xiaoxing’anling during 2004-2005 were studied. As a result, a total annual mean soil respiration was 39.262×106tC·a^-1 (equivalent to 0.04PgC·a^-1) and the annual soil respiration per unit area was 317.3gC·m^-2·a^-1.The highest vegetation type of the annual soil respiration per unit area was temperate softwood forest, which was 412.5gC·m^-2·a^-1.The highest vegetation type of the total annual mean soil respiration was cold temperate deciduous coniferous forest (equivalent to 10.947×106tC·a^-1), which accounted for 27.88%. The distribution of annual mean soil respiration was obvious, which was decreased from north to south in Daxing’anling and well-distributed in Xiaoxing’anling.Based on datasets and IBIS, the terrestrial net ecosystem production （NEP） and its spatial-temporal change in Da and Xiaoxing’anling during 2004-2005 were studied. As a result, theannual NEP per unit area was 313. 4gC·m^-2·a^-1 and a total annual mean NEP was 38.779×10⁶tC （equivalent to 0.04PgC）, which accounted for 27% of the Northeast forest’s total. The highest vegetation type of the annual NEP per unit area was temperate mixed coniferous and broad-leaf forest, which was 370.8gC·m^-2·a^-1.The highest vegetation type of the total annual mean NEP was temperate mixed coniferous and broad-leaf forest (equivalent to 11.016×l0⁶tC·a^-1), which accounted for 28.25%.The forest ecosystem in study areas was carbon sink in the period. The distribution of annual mean NEP was obvious, which was increased from north to south in Daxing’anling and well-distributed in Xiaoxing’anling. The spatial-temporal change of annual NEP in study areas was distinct. During 2004-2005, the annual NEP of Xunke, Sunwu, Jagedaqi and Songling areas were decreased and the annual NEP of other places in study areas was increased. By analysis, quantity of heat and precipitation leaded to this result.On the whole, though a series of numerical simulation in forest ecosystem of China, it could be concluded that IBIS may be a good choice for carbon simulation in China, with its good performance. Meanwhile, improvements and advancements are still needed for the model, especially incorporated within AGCMs. Combined macro and micro methods is the further study priorities on forest carbon cycling simulation.更多还原