基于CA与Multi-Agent的滇池流域土地利用变化模拟研究

发布时间：2019-01-02 09:38

【摘要】：土地利用/土地覆盖变化(Land-Use and Land-Cover Change，简称LUCC)是研究全球环境变化与资源可持续发展的主要内容，对区域水循环、大气循环、环境质量、气候变化及陆地生态系统等具有重要影响。人们在利用土地以促进社会经济发展过程中直接引起了土地利用和土地覆盖的剧烈变化，对生态环境、生态效益产生了巨大的影响。近几十年来，随着昆明市经济发展和城市规模的扩大，加重了滇池流域生态环境压力，水体受到污染，导致湖泊严重富营养化，滇池面临着严重水环境污染的困境。而国内外大量研究表明，，流域内不合理的土地利用和不科学的土地管理会引发土壤侵蚀，进而导致水土流失和营养物随地表径流流失造成大面积的非点源污染，因此研究土地利用变化具有重要的生态意义。 LUCC的复杂性决定了土地利用研究必须采取复杂系统的理论方法，因此本研究选择采用复杂系统理论支持下的多智能体(Multi-Agent)、元胞自动机模型（Cellular Automata，CA）和GIS技术相结合，从流域尺度上对土地利用/土地覆盖变化进行模拟预测，并对其空间格局变化特征进行分析，对于寻求如何合理发展流域生态环境和保护流域生态安全都具有重要的意义。此模型利用GIS获取所需要的各种空间变量，并使用人工神经网络（Artificial Neural Network，ANN）代替复杂的转换规则获取模型参数，大大简化CA模型的结构和转换规则的定义。多智能体体现了“人地关系”中“人”的主动性，而CA更加侧重“地”的作用，它们相辅相成，使得模型模拟更加符合现实规律。模型以滇池流域2002年的土地利用类型原始数据动态模拟了该流域2013年土地利用类型，以点对点精度和Lee-Salle指数来衡量模拟的精度。经过反复的参数率定，最终模拟结果点对点精度达到70.0729%，Lee-Salle总指数也超过了0.5，满足了精度的要求。精度达标后，模拟预测了滇池流域2024年土地利用空间格局与结构。模拟结果对加强滇池流域的保护，协调流域经济发展，优化流域资源配置具有重大意义，并为流域可持续发展提供了决策参考。
[Abstract]:Land use / land cover change (Land-Use and Land-Cover Change,) is the main research content of global environmental change and sustainable development of resources. Climate change and terrestrial ecosystems have important impacts. In the process of land use to promote social and economic development, land use and land cover are changed dramatically, which has a great impact on ecological environment and ecological benefits. In recent decades, with the economic development of Kunming and the expansion of the city scale, the ecological environment pressure and water pollution in Dianchi Lake Basin have been aggravated, resulting in serious eutrophication of lakes, and Dianchi Lake is facing the dilemma of serious water environment pollution. A large number of studies at home and abroad have shown that unreasonable land use and unscientific land management in the watershed will lead to soil erosion and resulting in a large area of non-point source pollution caused by the loss of soil and water and nutrients along with the loss of surface runoff. Therefore, the study of land use change has important ecological significance. The complexity of LUCC determines that the study of land use must adopt the theory method of complex system. Therefore, this study chooses multi-agent (Multi-Agent) and cellular automata model (Cellular Automata,) supported by complex system theory. CA) and GIS technology combined to simulate and predict land use / land cover change in watershed scale, and to analyze the spatial pattern change characteristics of land use / land cover change. It is of great significance to find out how to develop the ecological environment and protect the ecological security of the watershed. In this model, GIS is used to obtain the required spatial variables, and artificial neural network (Artificial Neural Network,ANN) is used to obtain the model parameters instead of the complex transformation rules, which greatly simplifies the structure of the CA model and the definition of the transformation rules. Multi-agent embodies the initiative of "human" in "man-land relationship", while CA pays more attention to the role of "land". They complement each other and make the simulation of the model more in line with the practical law. Based on the original data of land use types in Dianchi Lake Basin in 2002, the model dynamically simulates the land use types of Dianchi Basin in 2013. The accuracy of the simulation is measured by point-to-point accuracy and Lee-Salle index. After repeated parameter rate determination, the simulation results show that the point-to-point accuracy is 70.0729 and the total index of Lee-Salle is higher than 0.5, which satisfies the requirement of accuracy. The spatial pattern and structure of land use in Dianchi Lake Basin in 2024 were simulated and predicted after reaching the standard of precision. The simulation results are of great significance for strengthening the protection of the Dianchi Lake basin, coordinating the economic development of the basin and optimizing the allocation of the basin resources, and providing a reference for the sustainable development of the basin.
【学位授予单位】：云南师范大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P208;TP18;F301.2

【相似文献】