生态与农村环境学报 ›› 2021, Vol. 37 ›› Issue (6): 740-750.doi: 10.19741/j.issn.1673-4831.2020.0653

• 自然保护与生态 • 上一篇    下一篇

基于Ann-CA-Markov模型的生态空间预测模拟:以重庆市万州区为例

幸瑞燊1,2,3, 周启刚1,2,3,4   

  1. 1. 生态环境空间信息数据挖掘与大数据集成重庆市重点实验室, 重庆 401320;
    2. 重庆财经学院讯飞人工智能学院, 重庆 401320;
    3. 重庆工商大学环境与资源学院, 重庆 400067;
    4. 重庆工商大学公共管理学院, 重庆 400067
  • 收稿日期:2020-08-11 出版日期:2021-06-25 发布日期:2021-06-24
  • 通讯作者: 周启刚 E-mail:zqg1050@126.com
  • 作者简介:幸瑞燊(1994-),男,土家族,重庆市人,硕士,从事环境规划与管理研究。E-mail:18883175032@163.com
  • 基金资助:
    重庆市技术创新与应用发展重点研发项目(cstc2018jszx-zdyfxmX0021);重庆市基础研究与前沿探索项目(cstc2018jcyjAX0497)

Predictive Simulation of Ecological Spatial Evolution Based on Ann-CA-Markov Model: A Case Study of Wanzhou District, Chongqing

XING Rui-shen1,2,3, ZHOU Qi-gang1,2,3,4   

  1. 1. Chongqing Key Laboratory of Spatial Data Mining and Big Data Integration for Ecology and Environment, Chongqing 401320, China;
    2. Iflytek School of Artificial Intelligence, Chongqing Finance and Economics College, Chongqing 401320, China;
    3. College of Environment and Resources, Chongqing Technology and Business University, Chongqing 400067, China;
    4. College of Public Administration, Chongqing Technology and Business University, Chongqing 400067, China
  • Received:2020-08-11 Online:2021-06-25 Published:2021-06-24

摘要: 生态空间是提供生态服务功能及保护国家生态安全的重要区域,对未来生态空间的模拟预测可以为生态环境保护的政策制定及国土空间的优化管控提供参考依据。以万州区为研究区,使用万州区2000、2006、2012和2018年4期土地利用、数字高程模型(DEM)、道路交通、河流、行政中心、生态红线和自然保护地等数据,界定万州区的生态空间类型为林地、草地、水域和未利用地,对万州区的生态空间变化特征进行分析,并构建Ann-CA-Markov模型对万州区2024年生态空间进行模拟预测,同时对生态空间的生境质量进行评估。结果表明,该研究构建的Ann-CA-Markov模型具有较高的模拟精度,以元胞自动机模型(CA)为主体,结合Markov和Ann模型,有效解决了CA模型的不足,既能够对生态空间类型的转移数量和概率进行预测,又能够处理好影响因子与生态空间变化之间的复杂的非线性关系,从而对生态空间分布进行预测,其模拟的精度达0.983 6。万州区生态空间面积呈现持续增加趋势,但是生态空间斑块的破碎度、分化程度及面积均匀化程度越来越大,生态空间稳定性逐步下降,且生态空间的生境质量呈现逐年下降趋势。因此,在保证生态空间数量规模增加的同时,要兼顾生态环境保护与生态空间的优化管控,以保证生态空间的稳定性及生境质量。

关键词: 生态空间, 预测, Ann-CA-Markov模型, InVEST模型, 生境质量

Abstract: Ecological space is an important area for providing ecological service functions and protecting national ecological security. Simulation and prediction for future ecological space can provide reference for the formulation of policies for ecological environment protection, the optimization and control of national space. Taking Wanzhou District as the research area, the data of land use, digital elevation model (DEM), roads, rivers, administrative centers, ecological red lines, and natural protected areas of the District in 2000, 2006, 2012 and 2018 were used to define the types of ecological space of the District. According to the data, the types of ecological space of Wanzhou District include forest land, grassland, water areas, and unused land. The characteristics of the variations of the ecological spaces were analyzed to build the ANN-CA-Markov model for the simulated prediction of the ecological space of Wanzhou District in 2024, and the quality of the habitats of the ecological space of was also evaluated. The results show that the ANN-CA-Markov model has relatively high simulation accuracy by using the cellular automata model (CA) as the main body, which is the effective solution to the deficiency of the CA model. As the simulation is able to deal with impact factors and ecological space changes, a complex nonlinear relation forecasting the distribution of ecological space with the model has higher simulation accuracy (precision reaching 0.9 836) towards transferring quantities and probability forecasts of the ecological spaces. The Wanzhou District has shown a trend of continuous increase in fragmentation degree, differentiation degree, and area uniformity degree, while the stability of the ecological space is gradually declining. The habitat quality of the ecological space is presenting a declining trend year by year. Therefore, while ensuring the increase of the quantity and scale of ecological space, both ecological environment protection and ecological space optimization and control should be taken into account to ensure the stability of the ecological space and habitat quality.

Key words: ecological space, predictive, Ann-CA-Markov, InVEST model, habitat quality

中图分类号: