基于MAS-LCM的沙漠化空间模拟方法研究
发布时间：2018-12-13 12:45
【摘要】：以干旱区典型城市磴口县为研究区,利用1995—2015年每隔5年的Landsat TM影像通过遥感解译获取研究区20年的各等级沙漠化空间分布,利用GIS空间分析和重心迁移模型分析沙漠化景观时空变化趋势。并以2010年沙漠化分类数据为基期年数据,利用Logistic元胞自动机(Cellular automata-Markov,CA-Markov)模型(简称LCM)并引入多智能体系统(Multi-agent system,MAS)模型修正转移规则,预测2015年沙漠化分类情况及其空间分布格局。研究结果表明:磴口县20年间重度及极重度沙漠化面积减小,轻度沙漠化景观面积逐渐增大,其中2015年的非沙漠化景观达到37.09%,各类型沙漠化重心远离磴口县城,呈现良好态势。引入MAS模型的CA-Markov预测模型能够显著提升模型的模拟精度,所预测的2015年数据结果 Kappa系数达到0.62,高于CA-Markov模型模拟结果,能较好预测干旱区沙漠化分布情况,为沙漠化监管与治理提供了技术支持。
[Abstract]:Taking Dengkou County, a typical city in arid area, as the research area, the spatial distribution of desertification in the study area for 20 years was obtained by remote sensing interpretation using Landsat TM images every 5 years from 1995 to 2015. The spatial and temporal trends of desertification landscape were analyzed by GIS spatial analysis and barycenter migration model. Based on the annual data of desertification classification in 2010, the Logistic cellular automata (Cellular automata-Markov,CA-Markov) model (LCM) and the multi-agent system (Multi-agent system,MAS) model are introduced to modify the transfer rules. The classification of desertification and its spatial distribution pattern in 2015 are forecasted. The results show that in Dengkou County, the area of severe and extremely severe desertification decreases in 20 years, and the landscape area of light desertification gradually increases. The non-desertification landscape reaches 37.09 in 2015, and the center of various types of desertification is far away from Dengkou County. Present a good situation. The CA-Markov prediction model with MAS model can significantly improve the simulation accuracy of the model. The Kappa coefficient of the predicted data in 2015 is 0.62, which is higher than that of the CA-Markov model, which can better predict the desertification distribution in arid areas. It provides technical support for the regulation and control of desertification.
【作者单位】： 北京林业大学精准林业北京市重点实验室;北京明德立达农业科技有限公司;
【基金】：国家自然科学基金项目(41371189) “十二五”国家科技支撑计划项目(2012BAD16B00)
【分类号】：X171;X87
本文编号：2376547
[Abstract]:Taking Dengkou County, a typical city in arid area, as the research area, the spatial distribution of desertification in the study area for 20 years was obtained by remote sensing interpretation using Landsat TM images every 5 years from 1995 to 2015. The spatial and temporal trends of desertification landscape were analyzed by GIS spatial analysis and barycenter migration model. Based on the annual data of desertification classification in 2010, the Logistic cellular automata (Cellular automata-Markov,CA-Markov) model (LCM) and the multi-agent system (Multi-agent system,MAS) model are introduced to modify the transfer rules. The classification of desertification and its spatial distribution pattern in 2015 are forecasted. The results show that in Dengkou County, the area of severe and extremely severe desertification decreases in 20 years, and the landscape area of light desertification gradually increases. The non-desertification landscape reaches 37.09 in 2015, and the center of various types of desertification is far away from Dengkou County. Present a good situation. The CA-Markov prediction model with MAS model can significantly improve the simulation accuracy of the model. The Kappa coefficient of the predicted data in 2015 is 0.62, which is higher than that of the CA-Markov model, which can better predict the desertification distribution in arid areas. It provides technical support for the regulation and control of desertification.
【作者单位】： 北京林业大学精准林业北京市重点实验室;北京明德立达农业科技有限公司;
【基金】：国家自然科学基金项目(41371189) “十二五”国家科技支撑计划项目(2012BAD16B00)
【分类号】：X171;X87
【相似文献】
相关期刊论文 前4条
1 胡汝骥;姜逢清;王亚俊;;中国干旱区不能成为“我国新粮仓”[J];干旱区研究;2010年02期
2 张虎生;对干旱区可持续发展的思考[J];环境科学动态;1998年02期
3 阎金凤,陈曦;基于GIS的干旱区LUCC分析和模拟方法探讨[J];干旱区地理;2003年02期
4 王让会,马映军;干旱区山盆体系物质能量及信息的耦合关系——以塔里木盆地周边山地系统为例[J];山地学报;2001年05期
相关会议论文 前1条
1 黄培yP;;干旱区开发引起柽柳植丛扩展现象的研究[A];复合生态与循环经济——全国首届产业生态与循环经济学术讨论会论文集[C];2003年
相关硕士学位论文 前1条
1 刘春静;中亚干旱区植被退化及典型流域生态环境遥感监测[D];山东农业大学;2016年
，本文编号：2376547
本文链接：https://www.wllwen.com/shengtaihuanjingbaohulunwen/2376547.html
