地理研究

地理研究 ›› 2005, Vol. 24 ›› Issue (6): 869-877.doi: 10.11821/yj2005060006

• 论文 • 上一篇    下一篇

拉萨地区土地利用变化情景分析

除多1, 张镱锂2,3, 郑度2   

  1. 1. 西藏高原大气环境科学研究所, 拉萨 850000;
    2. 中国科学院地理科学与资源研究所, 北京 100101;
    3. 中国科学院青藏高原研究所, 北京 100085
  • 收稿日期:2005-01-18 修回日期:2005-05-24 出版日期:2005-12-15 发布日期:2005-12-15
  • 通讯作者: 张镱锂,研究员,博士生导师。E-mail:zhangyl@igsnrr.ac.cn
  • 作者简介:除多(1969-),男,西藏白朗县人,博士,高级工程师。主要从事环境遥感与GIS应用。
  • 基金资助:

    国家自然科学基金资助项目(40361001,90202012);国家重点基础研究发展计划(2005CB422006);中国科学院知识创新项目(KZCX3-SW-339)资助

Land use change scenario in Lhasa district using Markov chain model

CHU Duo1, ZHANG Yi-li2,3, ZHENG Du2   

  1. 1. Tibet Institute of Plateau Atmospheric and Environmental Sciences, Lhasa 850000, China;
    2. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China;
    3. Institute of Tibetan Plateau Research, CAS, Beijing 100085, China
  • Received:2005-01-18 Revised:2005-05-24 Online:2005-12-15 Published:2005-12-15

PDF (PC)

赞

被引次数

187

摘要：

根据西藏拉萨地区1990年、1995年和2000年3个时点的土地利用数据,应用马尔科夫过程模型分析了未来20年内拉萨地区的土地利用情景变化,并与90年代制定的拉萨地区土地利用规划面积进行了对比研究。研究结果:1)10年间,土地利用类型变化最广泛的是牧草地。变化方向主要由牧草地向耕地、园地、林地、居民点及水域转变,其中变成林地的面积最大,为2338.25hm2(占变化面积的94.093%);2)拉萨地区未来20年中土地利用类型发展趋势是耕地、牧草地、水域和未利用土地面积将进一步减少,林地、园地和居民点面积将进一步增加;3)土地利用规划面积与基于马尔科夫模型的土地利用变化情景分析结果比较吻合,马尔科夫过程模型对制定该区域土地利用规划具有重要的参考价值;4)由于土地利用变化是一个复杂的过程,不仅受到众多自然因素的影响,而且受到未来土地利用政策、社会经济发展、区域内大型工程项目及其他人类活动等不确定因素的影响,从而不同土地利用类型之间的转移概率会发生变化,使得基于马尔科夫过程模型预测的精度有一定的局限性。

关键词: 土地利用变化, 情景分析, 马尔科夫模型, 拉萨地区

Abstract:

Markov model was used to represent the land-use change data for several reasons.Firstly,land use change is not unidirectional in nature.A given parcel of land theoretically may change from one category of land-use to any other at any time.Markovian analysis uses matrices that represent all the multi-directional land-use changes between all the mutually exclusive land-use categories.Land use changes and scenarios in Lhasa district were analyzed in this paper using Markov model.The main conclusions are as follows: (1) Land use changes in Lhasa district during the last ten years were related to this period's integrated agricultural development project.In 10 years from 1990 to 2000,186.53 ha of cultivated land were converted into forest land,accounting for 54.857% of the total cultivated land area converting into the other land use categories;and 130.05 ha were converted into residential area,being 38.247%.The remaining limited area was converted into the other land use categories such as rangeland and water body. (2) The most extensive changes of land use from 1990 to 2000 in Lhasa district was rangeland conversion.Rangeland was converted into cultivated land,horticultural land,forestland,residential area and water body.Among them,2333.33 ha of forestland were converted from rangeland,or 94.093 %,the largest of all in area.These land use changes were related directly to the integrated development and construction project carried out in the central Tibetan Plateau in the mid 1990s.Because of the implementation of biological and engineering measures consisting of large scale of forestation and construction of field safeguarding forest,since the 1990s,remarkable effects in soil erosion control have been obtained through improving surface vegetation and land cover conditions. (3) Markov chain model prediction indicated that the general trends of land use changes,in Lhasa district in future were that cultivated land,rangeland,water body and unused land will decrease and forest,horticultural land and residential area will increase. (4) It should be pointed out that since there are many factors impacting land use changes,in particular in a short period of time,policies,regulations,important regional projects and human activities and other uncertain factors will affect regional land use changes,hence causing certain errors in prediction accuracy of land use changes in Lhasa district. (5) The simulated areas of land use changes in the future in Lhasa district by Markov mode are very close to the planning areas of land use changes of Tibet,therefore,land use change scenario based on the Markov model has an important value for making land use plan.

Key words: land use change, scenario analysis, Lhasa district, Markov chain model