【作者】 徐晓甫；

【导师】 陶建华；

【作者基本信息】 天津大学 ， 环境科学， 2013， 博士

【摘要】 随着海岸带地区经济的快速发展和开发活动的加剧，近岸海域生态环境面临着巨大的压力。近岸海域水生态环境特性的研究，对认识海洋生态过程和改善近岸海域水环境状况具有重要的意义。由于近岸海域生态环境的复杂性、动态性和时空特异性，其生态环境研究面临诸多挑战。本文以半封闭的渤海湾天津近岸海域为背景，系统地讨论了近岸海域生态环境因子的时空分析、综合评价和模拟预测方法，并在此基础上对天津近岸海域生态环境特性进行了研究，建立了天津近岸海域生态环境空间决策支持系统，可为今后天津近岸海域的生态环境研究和管理提供技术支持，促进近岸海域生态环境的健康发展。首先，针对近岸海域生态环境的动态性和时空特异性，对近岸海域生态环境的时空分析方法和近十年来天津近岸海域富营养化的时空分布特性进行了研究。结果表明，富营养化及相关环境因子年际变化基本呈现先下降、后回升的趋势，空间上呈现近岸高、远海低的特征，时空趋势变化主要受陆源排放的影响。其次，面对常规生态环境质量评价方法中存在的反映信息不全面、容易失真等问题，本文研究了多级灰关联综合评价法，并对近年来天津近岸海域生态环境质量进行了评价。结果表明，天津近岸海域环境质量整体不容乐观，四个功能区水质差异明显，天津汉沽农渔业区和东南部农渔业区污染最为严重。然后，针对传统生态动力学模型在处理海洋生态环境模拟中不确定性问题的局限性，建立了结构方程模型（SEM）和贝叶斯网络（BN）相结合的不确定性推理模型，并利用该模型对天津近岸海域的浮游植物生物量的动态变化进行了模拟研究，以探索和验证环境因子与叶绿素间的因果关系，模型取得了良好效果。最后，基于天津近岸海域生态特性的研究，建立了天津近岸海域生态环境空间决策支持系统，为天津近岸海域生态环境提供了综合的研究平台。在此过程中，对传统的数据管理和模型集成方式进行了改进，提出生态空间数据一体化管理机制和模型插件式集成模式，为高效管理天津近岸海域生态空间数据、提高研究平台的可扩展性和生态模型的复用率提供了新的思路。更多还原

【Abstract】 With the development of economy and the increase of pollutant emission alongcoastal areas, the coastal ecological environment is facing tremendous pressure. Thestudy on coastal ecological characteristics is of great significance for theunderstanding of marine ecological processes, protecting coastal aquatic ecosystemand marine management.Due to the complexity of the coastal eco-environment, the study of coastalecological environment faces with many challenges. In order to promote the healthydevelopment of the Tianjin coastal ecological environment, we studied the ecologicalcharacteristics of Tianjin coastal area systematically and developed the Tianjin coastalecological environment spatial decision support system.Firstly, the spatiotemporal distribution characteristics of eutrophication in Tianjincoastal waters were analysed using the properly selected temporal-spatial analysismethods. The results indicate that the annual variation of eutrophication showsdownward trend in the earlier stage and upward trend later, and shows the spatialvariation trend of decreasing from near-shore to off-shore. The spatial and temporaltrends mainly influenced by the land-source discharges.Secondly, in order to overcome the shortcomings of the traditional ecologicalenvironment quality evaluation method, such as reflecting the one-sided informationand tending distortion, this study proposed a multi-level gray relationalcomprehensive evaluation method to evaluate the ecological environment quality ofTianjin coastal waters. The result shows that the overall quality of the environment isnot optimistic, and the environmental qualities of the four functional areas aresignificantly different from each other.Thirdly, in view of the limitations of the traditional eco-hydrodynamics modelsto treat the uncertainty in the eco-environment, an uncertainty reasoning model, basedon linking structural equation modeling (SEM) with Bayesian network (BN), wasdeveloped. The SEM-BN model was applied to study the phytoplankton dynamics inTianjin coastal area, with satisfied results being obtained.Finally, based on the characteristics of Tianjin coastal ecological environment,the Tianjin coastal ecological environment spatial decision support system was developed to provide an integrate research platform for studying Tianjin coastalecological environment. Meanwhile, the traditional data management and modelintegration approach was improved for efficient management of ecological data andecological models.更多还原