Knowledge Resource Center for Ecological Environment in Arid Area
| 基于过程的生态系统服务模型开发与应用 | |
| 其他题名 | Development and Application of an Integrated Multiscale Process-based Ecosystem Services Model |
| 张丽云 | |
| 出版年 | 2017 |
| 学位类型 | 博士 |
| 导师 | 徐明 |
| 学位授予单位 | 中国科学院大学 |
| 中文摘要 | 生态系统以服务流的形式源源不断的向人类输送各种各样的服务，这些服务与人类福祉息息相关。生态系统服务是生态过程的最终表现形式，因此，从生态过程出发，追踪生态系统服务的流动，有助于我们客观准确的评估生态系统服务价值，从而更加重视生态系统的保护和资源的合理开发。本研究以生态系统过程模型为基础，通过改进模型对土壤冻融过程的模拟，参数敏感性分析及关键参数优化，建立了以生态系统服务评估为目标的生态系统服务模型，实现了对生态系统固碳释氧、空气净化、淡水资源和水文调节服务的综合模拟。并利用包括土壤温度、森林资源清查和通量塔碳水通量等多源数据，对模型进行了验证。最后，以青海省生态系统为对象，模拟了青海省森林、草地、荒漠、湿地四大类生态系统的各项服务的时空动态，为青海省的可持续发展提供依据。研究结果如下：（1）通过全局敏感性分析，发现过程模型生理生态参数中，叶子和细根年周转率（LFRT）、Rubisco酶活叶氮量（FLNR）、平均比叶面积、细根与新叶碳分配比，以及叶片碳氮比等五个参数的敏感性最大，总敏感性指数分别达0.46、0.22、0.19、0.19和0.16。各参数的主敏感性指数均小于总敏感性指数，约占总敏感性指数的一半，说明模型各参数之间存在明显的交互作用。在各参数中，以参数LFRT的主敏感性和总敏感性最大，分别在0.24-0.32和0.41-0.49之间，说明这个参数最为敏感；参数FLNR的敏感性其次。（2）根据青海省生态系统的特点，分为常绿针叶林、落叶阔叶林、灌木林、草地、荒漠和湿地等六类植被型，采用马尔可夫链-蒙特卡罗方法，进行关键参数LFRT和FLNR进行优化。发现参数LFRT和FLNR在森林、草地、湿地和荒漠生态系统中都能被NEE和ET实测数据较好的约束，各参数的后验估计均呈近似正态分布。常绿针叶林、灌木林、草地、荒漠和湿地植被型中，优化后的参数LFRT依次为0.358 yr-1、0.925 yr-1、0.9465 yr-1、0.748 yr-1和0.989 yr-1。优化后的参数FLNR由大到小依次为落叶阔叶林、常绿针叶林、灌木林、草地、湿地和荒漠植被型中，优化后的参数FLNR依次为8.258%、6.159%、5.597%、4.539%、4.570%和4.477%。（3）利用优化后的生态系统服务模型进行评估，发现青海省森林、草地、荒漠和湿地四大生态系统提供的服务中，以水文调节量最大，为1719.13×108 t·yr-1，提供的淡水资源量其次，为482.90×108 t·yr-1，固碳释氧量和空气净化量相对较低，分别为5016.74×104 t·yr-1和91.57×104 t·yr -1。与其他生态系统相比，单位面积森林生态系统的固碳释氧量和空气净化量较高，湿地的水文调节和淡水资源服务密度最高。（4）不同年份间，青海省生态系统服务呈现较大的波动。从时间动态来看，1998-2012年间净化空气、水文调节和淡水资源服务量随时间呈波动上升趋势，分别以2010年、2007年和2009年最高。固碳释氧量的时间动态则有很大不同，表现为随着时间呈略有下降的趋势，以1999年固碳释氧量最高。各年间，以淡水资源服务的变化幅度最大，变异系数为39.71%。固碳释氧其次（37.84%），空气净化和水文调节的波动相对较小，变异系数分别为9.32%和8.46%。荒漠生态系统的固碳释氧、空气净化和水文调节服务量在各年间波动最为剧烈，变异系数分别为72.99%、14.61%和10.53%。森林生态系统提供的淡水资源服务的年际波动大于其他生态系统，达58.52%。（5）从空间分布上，青海省生态系统单位面积提供的固碳释氧、空气净化、水文调节和淡水资源服务的空间格局较为一致，表现为从东南向西北逐渐增加的趋势。随着海拔的升高，各生态系统服务指标的变化趋势不尽相同。固碳释氧服务呈先增加后降低的趋势，表现为3500-4000m中高海拔区域的服务量最大。森林和荒漠生态系统净化空气量表现为随着海拔的增高先增加后降低的趋势，而草地和湿地生态系统则逐渐降低。淡水资源量随海拔逐渐增多，高海拔区域的淡水资源量最高。水文调节服务为随着海拔升高逐渐增大，到中高海拔后趋于平缓。 |
| 英文摘要 | The importance of natural ecosystem to human well–being cannot be overstated. Ecosystem services are closely related to ecosystem processes, especially the carbon and water cycling. Therefore, the amounts of ecosystem services could be assessed from relevant ecosystem processes. The knowledge of ecosystem biogeochemical cycles is very useful in ecosystem services valuation. The process of soil freezing and thawing in cool area is a major factor that influences the vegetation. However, most of the ecological process models do not include this process in the model structure. Thus, these models may be less precise when they were applied in cool region. Therefore, we developed a frozen soil model to describe the freezing and thawing process on soil, and coupled with the process-based model. Then, we established a process based ecosystem services model. This model can be used to estimate four ecosystem services, i.e., carbon sequestration, air purification, freshwater provision, hydrological regulation. To improve the performance of the ecological model, we also conducted the parameters sensitivity analysis and the key parameters optimization. The multi-source data, including soil temperature, forest biomass carbon storage, net ecosystem productivity and evapotranspiration, were used to validate the model. Subsequently, we estimated the ecosystem services for the forest, grass, wetland and desert ecosystem in Qinghai Province using the model. Moreover, we were able to estimate the temporal dynamics and spatial distribution for each ecosystem services index via this model. Therefore, this study could provide scientific support for decision making with regarding to the sustainable development of Qinghai Province. The results showed that:(1) With the global parameters sensitivity analysis, we found the tested variables were highly sensitive to the following eco-physiological parameters: annual leaf and fine root turnover fraction, fraction of N in Rubisco, canopy average specific leaf area, new fine root C: new leaf C, C:N ratio of leaves. The main sensitivity indexes were 0.46, 0.22, 0.19, 0.19 and 0.16, respectively. All of the main sensitivity indexes for the parameters were smaller than the corresponding total sensitivity indexes. The main sensitivity indexes were only a half of the total sensitivity indexes, indicating that the parameters of this model had a strong interaction effect. The parameter of annual leaf and fine root turnover fraction (LFRT) was the most sensitive parameter, which has a main sensitivity indexes ranged from 0.24 to 0.32, and a total sensitivity indexes ranged from 0.41 to 0.49. The parameter fraction of N in Rubisco (FLNR) was the second sensitive parameter.(2) Different vegetation types have very different growth characters. In process model, these characters were reflected by lots of ecological parameters. To optimize these parameters locally, we divided the vegetation types in Qinghai into six types and optimized their key parameters respectively. The parameter inversion was made for two key ecological parameters (LFRT and FLNR) using the Markov Chain Monte Carlo method. The parameters were found to be well constraint by the observed NEE and ET. The posteriori estimates for the two parameters in the six vegetation types were generally fitted the normal distribution. The optimized LFRT for evergreen coniferous forest, shrubs, grass, desert and wetland were 0.358yr-1、0.925yr-1、0.9465yr-1、0.748yr-1 and 0.989yr-1, respectively. The optimized FLNR for deciduous broadleaf forest, evergreen coniferous forest, shrubs, grass, desert and wetland were 8.258%, 6.159%, 5.597%, 4.539%, 4.570% and 4.477%, respectively.(3) In the ecosystem of Qinghai Province, the hydrological regulation service was larger than other services, which was 1719.13×108 t·yr-1. It was followed by freshwater provision service by 482.90×108 t·yr-1. The carbon sequestration and oxygen release service was ranked the third, and the air purification service was the least, with values of 5016.74×104 t·yr-1and 91.57×104 t·yr-1, respectively. Among different ecosystems, the forest ecosystem had the largest carbon sequestration and air purification density, while the wetland ecosystem had the largest hydrological regulation and freshwater provision services per unit area.(4) The ecosystem services of Qinghai Province varied much among different years. From 1998-2012, the services of air purification, freshwater provision and hydrological regulation increased gradually, with the highest value found in 2010, 2007 and 2009, respectively. However, the carbon sequestration and oxygen release service tend to decrease, with the highest value found in 1999. The interannual variation was also different in ecosystem services. The coefficient of variation was the largest in freshwater provision service (39.71%), following by the variation of carbon sequestration and oxygen release service (37.84%). In contrast, the variation of air purification and freshwater provision services were quite small, which were only 9.32% and 8.46%, respectively. Among the ecosystems, the desert had the largest interannual variation, with a coefficient of variation of 72.99%, 14.61% and 10.53% for carbon sequestration, air purification and hydrological regulation, respectively, while the forest ecosystem had the largest variation in freshwater provision than other ecosystem, with a value of 58.52%.(5) The spatial distribution of the ecosystem services was found to increase from southeast to northwest. The four ecosystem services were similar in the spatial distribution. However, when the elevation gradient was considered, the spatial distribution of the four ecosystem services was very different. With the elevation increase, the carbon sequestration service increased firstly and then decreased, and saw the maximum value in the gradient at the elevation of 3500-4000m. The variation of the air purification service of forest and desert followed the same trend with the carbon sequestration service. However, the air purification service of the grass and wetland decreased with elevation increase. The freshwater provision service generally increased with the elevation. Besides, the air purification service was found to increase to a mid-high elevation, and then tend to be leveled off. |
| 中文关键词 | 生态系统服务 ; 生态过程模型 ; 生态系统服务模型 ; 青海省 |
| 英文关键词 | ecosystem services ecological process model ecosystem services model Qinghai Province |
| 语种 | 中文 |
| 国家 | 中国 |
| 来源学科分类 | 生态学 |
| 来源机构 | 中国科学院地理科学与资源研究所 |
| 资源类型 | 学位论文 |
| 条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/287862 |
| 推荐引用方式 GB/T 7714 | 张丽云. 基于过程的生态系统服务模型开发与应用[D]. 中国科学院大学,2017. |
| 条目包含的文件 | 条目无相关文件。 | |||||
| 个性服务 |
| 推荐该条目 |
| 保存到收藏夹 |
| 导出为Endnote文件 |
| 谷歌学术 |
| 谷歌学术中相似的文章 |
| [张丽云]的文章 |
| 百度学术 |
| 百度学术中相似的文章 |
| [张丽云]的文章 |
| 必应学术 |
| 必应学术中相似的文章 |
| [张丽云]的文章 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
