首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于GIS的森林资源空间数据挖掘方法研究——以紫金山为例
引用本文:李明阳,刘方,徐婷,姜文倩.基于GIS的森林资源空间数据挖掘方法研究——以紫金山为例[J].西北林学院学报,2012,27(3):180-186.
作者姓名:李明阳  刘方  徐婷  姜文倩
作者单位:南京林业大学森林资源与环境学院,江苏南京,210037
基金项目:国家自然科学基金(30972298,31170592,50978054)
摘    要:空间数据挖掘,可以从森林资源空间数据库中抽取未清楚表现出来的隐含的知识和空间关系,为森林空间规划和经营措施设计提供科学依据。以南京紫金山国家森林公园为研究对象,以2002年森林资源二类调查空间数据库、2002年遥感卫星SPOT5数据包为主要信息源,选取与风景林美景度有关的林分、地形、人类干扰等10个环境因子,在ArcGIS 9.3平台上,进行空间热点探测、趋势面分析、地理加权回归、C5.0决策分析。研究表明:高美景度风景林主要分布在南坡的中上部,而低美景度风景林主要分布在紫金山森林公园外围;风景林美景度在空间分布上呈现出一种圆心偏东的同心圆状分布格局;风景林美景度与林分生长状况正相关,与所处位置的海拔高度正相关,与距离公路、与居民点距离等人为因素的干扰强度负相关;C5.0决策分析归纳的3个示例规则所揭示的知识与空间热点分析、趋势面分析、地理加权回归的结论保持一致,4个主要输入变量的重要性依次为:平均胸径(0.33)>海拔(0.25)>与公路距离(0.23)>与居民点距离(0.21)。

关 键 词:森林资源  空间数据挖掘  地理信息系统  紫金山

Mining of Spatial Data of Forest Resources Based on GIS-A Case Study of Zijin Mountain
LI Ming-yang , LIU Fang , XU Ting , JIANG Wen-qian.Mining of Spatial Data of Forest Resources Based on GIS-A Case Study of Zijin Mountain[J].Journal of Northwest Forestry University,2012,27(3):180-186.
Authors:LI Ming-yang  LIU Fang  XU Ting  JIANG Wen-qian
Institution:(College of Forest Resources and Environment,Nanjing Forestry University,Nanjing,Jiangsu 210037,China)
Abstract:Spatial data mining can extract hidden knowledge and unclear spatial relationship from spatial datasets of forest resources,and to provide scientific basis for forest spatial planning and decision making of management measures.Zijin Mountain National Forest Park in Nanjing was selected as the study area,while level two forest survey spatial dataset in 2002 and remote sensing image package of SPOT5 in the same year were chosen as the main information sources for data mining.Ten environmental factors which were closely related with scenic beauty value of scenic forest,such as stand,terrain and human disturbance were generated to do spatial hot spot detection,surface trend analysis,geographically weighted regression(GWR) and C5.0 decision tree analysis on the platform of ArcGIS 9.3.The results showed that:(1) scenic forest with high scenic beauty value was mainly distributed on the middle and upper slope of the mountain,while the forest with low value was located around the outskirts of the park;(2) the pattern of spatial distribution of scenic beauty value resembled concentric circles with eastward central point;(3) scenic beauty value was positively correlated with forest growth status and altitude of the scenic forest,negatively correlated with the distance to the settlement and road;(4) the result of C5.0 decision analysis was consistent with spatial hot spot analysis,trend surface analysis and GWR,while the importance of the four major factors was in the order of: average DBH(0.33)>altitude(0.25)>distance to road(0.23)> distance to settlement(0.21).
Keywords:forest resource  spatial data mining  GIS  Zijin Mountain
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号