【作者】 姜纪峰；

【导师】 黄耀； 延晓冬；

【作者基本信息】 南京农业大学 ， 环境科学， 2006， 硕士

【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，对模型的源程序进行编译和调试，并改进了模式的输入和输出格式，进而将模式分别应用于中国区域的不同类型典型生态系统，对模型模拟碳和水热通量的能力进行检验。 首先，通过应用该模型对国际CEOP计划半干旱区基准站之一的吉林通榆观测站（44°25′N，122°52′E）草地和农田生态系统2003年全年的CO₂和水、热通量变化进行模拟，并将结果与涡度相关法测定的观测值进行了对比分析，以检验IBIS模型在半干旱区的模拟能力。对比结果表明：除CO₂通量模拟结果不够理想外，IBIS模型较好地模拟了通榆观测站的感热通量和潜热通量。在对草地的CO₂、感热和潜热通量模拟中，决定系数R²分别为0.416、0.762和0.825；对农田的模拟中R²分别为0.507、0.727和0.838（其中所有相关分析均通过了0.05以上显著水平的信度检验）。同时，经年变化分析得出，IBIS模型分别低估CO₂通量（偏低9.29％和3.73％）和潜热通量（偏低4.63％和3.48％），对感热通量则是高估（偏高9.90％和11.98％）。对照农田和草地，从总体上看，模型对通榆观测站农田生态系统的模拟效果要好于退化草地。 在此基础上，运用中国陆地生态系统通量观测研究网络（ChinaFLUX）三处森林微气象站的数据资料对IBIS模型模拟CO₂和潜热通量的能力做进一步的检验，结果表明：IBIS模型对潜热通量的模拟效果要远远好于对CO₂通量的模拟，同时模型对两者在长白山等三处站点的模拟均要偏高。具体来讲，在长白山站的模拟情况较为理想，CO₂和潜热通量的模拟-观测决定系数分别为0.326和0.821，模拟的CO₂和潜热通量更多还原

【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 purpose of validating IBIS in semi-arid region and eventually improving the model’s capability for specific land cover types in China, a simulation for the whole 2003 year was conducted over cropland and degraded grassland underlying surfaces in Tongyu field observation station （44°25’N, 122°52’E） in Jilin Province of China, which is one of the reference sites of international Coordinated Enhanced Observing Period （CEOP）. Measurements of turbulent flux of latent and sensible heat, and NEE were performed using the eddy correlation technique （referred to from here on as EC）. Model inputs include the site ecosystem characteristics and main meteorological variables. Comparisons between simulated land surface fluxes and observed EC measurements show that the model is capable of reproducing CO₂, sensible and latent heat fluxes indicated by correlation coefficients exceeding the significant level of 0.05. In general, CO₂ flux and sensible heat flux have obvious diurnal and seasonal variation both at the grassland and the cropland, while the latent heat flux, which is related to evapotranspiration, only has diurnal variation during the growing season. Meanwhile, all fluxes at the cropland are larger than those at the degraded grassland, especially in the growing season. The model generally estimates lower annual CO₂ （underestimating by 9.29% in the grassland and 3.73% in the cropland） and latent heat flux （with corresponding values 4.63% and 3.48%）, and greater annual sensible heat flux （overestimating by 9.90% and 11.98%, respectively） than measured by EC. And the simulations for the cropland are better than those for the grassland, with more reasonable model-measurement agreements.更多还原