地理学报 ›› 2011, Vol. 66 ›› Issue (11): 1518-1530.doi: 10.11821/xb201111008

• 土地利用 • 上一篇    下一篇

复杂系统理论与Agent模型在土地变化科学中的研究进展

余强毅1,2, 吴文斌1,2, 唐华俊1,2, 杨鹏1,2, 陈仲新1,2, 陈佑启1,2   

  1. 1. 中国农业科学院农业资源与农业区划研究所, 北京100081;
    2. 农业部资源遥感与数字农业重点开放实验室, 北京100081
  • 收稿日期:2011-05-04 修回日期:2011-09-02 出版日期:2011-11-20 发布日期:2011-11-20
  • 通讯作者: 唐华俊(1960-), 男, 研究员, 博士生导师, 主要从事土地利用变化、遥感和GIS应用方面研究。E-mail: hjtang@mail.caas.net.cn
  • 作者简介:余强毅(1986-), 男, 博士研究生, 主要研究方向为土地变化模拟。E-mail: yuqiangyi@163.com
  • 基金资助:

    国家重点基础研究发展计划项目(2010CB951504); 国家自然科学基金项目(40930101; 40971218); 国际科技合作项目(2010DFB10030)

Complex System Theory and Agent-based Modeling: Progresses in Land Change Science

YU Qiangyi1,2, WU Wenbin1,2, TANG Huajun1,2, YANG Peng1,2, CHEN Zhongxin1,2, CHEN Youqi1,2   

  1. 1. Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, China;
    2. Key Laboratory of Resources Remote Sensing and Digital Agriculture, Ministry of Agriculture, Beijing 100081, China
  • Received:2011-05-04 Revised:2011-09-02 Online:2011-11-20 Published:2011-11-20
  • Supported by:

    National Basic Research Program of China, No.2010CB951504;National Natural Science Foundation of China, No.40930101; No.40971218; Key Project of InternationalCooperation, No.2010DFB10030

摘要: 基于复杂系统理论与Agent 模型,分别从理论与方法的角度梳理并总结了当前土地变化科学的研究进展。复杂系统理论为研究土地变化所处的“耦合的人类—自然系统”及其复杂性问题提供理论支撑;而基于Agent 的模拟作为研究复杂系统的重要工具,为传统土地变化模拟提供新的方法支持。当前,ABMs与土地变化模型相结合(ABM/LUCC),无论是模型构建还是模型应用均取得了显著进展。然而,其也存在诸多问题,尤其是理论与应用脱节,使很多建模者容易陷入“为建模而建模”的误区。此外,现有的ABM模型设计较为随意,模型间差异太大,不利于跨区域比较研究的开展。因此,虽然ABM是认识复杂系统的有效方法,但是建模之前离不开对系统复杂性的深入研究;基于Agent 的土地变化模拟的意义在于充分表达“人类—自然”系统的综合复杂关系,从而合理的预测土地变化(结构与功能) 的动态过程,解释并评价土地变化的可能影响。

关键词: Agent模型, 人类—自然系统, 土地利用/土地覆盖, 土地变化科学, 土地系统

Abstract: Based on complex system theory and agent-based models (ABMs), the paper summarizes recent progresses in land change science from the perspective of theory and methodology respectively. Complex system theory is the theoretical basis for carrying out researches on the complex land change issues in the "coupled human and natural systems"; while ABMs, one of the key tools for complex system studies, introduces innovating perspective to traditional land change modeling. The integration of ABMs and land use/cover change models (ABM/LUCC) has achieved several important breakthroughs recently; however, some of the crucial issues remain unsolved, such as the problem of "theory divorced from practice" and the deficiency in cross-site comparison studies. As for current problems, the authors finally have discussion and draw a conclusion that firstly, ABMs should be constructed by the support of complex system theory. Moreover, the natural ABM/LUCC are supposed to explore the comprehensive human-natural interactions in land systems, to predict land system dynamics, and to analyze the consequences of land system change.

Key words: agent-based modeling, coupled human and natural systems, land use and land cover change, land systems