摘要
针对土地利用动态模拟与预测能为区域可持续发展提供重要参考依据的问题,该文以遥感影像为基础数据,基于人工神经网络元胞自动机(ANN-CA)模型对精河流域绿洲进行土地利用变化动态模拟与情景预测。结果表明:①在过去的26年间,耕地和建设用地居土地利用类型增幅前两位,分别为4.67、3.88,林地和未利用地居降幅前两位,则分别为-0.76,-0.05;②至2026年,3种情景下建设用地的变化幅度均较为明显,显示城镇用地在未来绿洲土地利用发展变化过程中占据主导地位;③研究区未来土地利用过程中,在无任何限制条件下的趋势发展情景必然会造成区域内较大的土地利用格局破碎度,但两种生态保护情景作为修正和补充,可以从空间上优化预测结果。
The dynamic simulation and prediction of land use can provide an important reference for regional sustainable development.Based on remote sensing image,this paper simulates and predicts land use change dynamics and scenarios of oasis in Jinghe River basin using ANN-CA model.The results show that:①In the past 26years,cultivated land and construction land occupied the first two places in the increase of land use types,4.67and 3.88,while woodland and unused land occupied the first two places in the decline,respectively,-0.76and-0.05;②The construction land under the 3scenarios will change drastically to the year of 2026,which shows the dominant position of urban land in the process of development and change of oasis utilization in the future;③In the process of land utilization in the research area in future,the trend development scenario without any restrictions will inevitably result in the greater fragmentation of the land utilization pattern in the region,but the two ecological protection scenarios can be used to optimize the prediction results from the space as a revision and supplement.
作者
于辉
毋兆鹏
YU Hui;WU Zhaopeng(College of Geographical Science and Touris,Xinjiang Normal University,Urumqi 830054,China;Xinjiang Laboratory of Lake Environment and Resources in Arid Zone,Urumqi 830054,China)
出处
《测绘科学》
CSCD
北大核心
2019年第12期86-94,共9页
Science of Surveying and Mapping
基金
国家自然科学基金项目（41761113）
新疆师范大学“十三五”校级重点学科地理学招标课题资助项目
