【作者】 王淼；

【导师】 赵鹏祥；

【作者基本信息】 西北农林科技大学 ， 森林经理学， 2011， 硕士

【摘要】 本研究以黄龙山林区1986年,1997年和2008年三期TM影像为信息源,应用遥感图像处理技术对遥感图像进行图像融合、波段组合、几何精校正和图像裁剪等图像预处理;同时,以1986年,1997年和2006年三期黄龙山森林资源二类调查数据为依托据,建立林地类型解译标志,执行监督分类,获取三期林地类型信息;利用ArcGIS软件计算林地转移矩阵并进行数学统计,提取并分析黄龙山林区林地时空变化特征;利用黄龙山林区DEM数据提取海拔高度、坡度、坡向等地形因子,并据此分析自然控制因子对林地变化的影响;定性分析了国家林业政策法规对黄龙山林区林地变化的影响。同时结合黄龙县1986年-2008年国民经济统计资料,选取合适的社会经济因子,利用统计学方法对黄龙山林区林地变化的社会经济驱动力做出定量分析,并构建了社会经济驱动力模型。本研究内容成果归纳如下:(1)经过计算OIF指数,发现TM543波段的组合既保证了影像含有最大的信息量,也可使波段之间信息的冗余度达到最小,同时,R(5)G(4)B(3)波段的组合也符合人体肉眼对色彩的分辨能力,便于遥感影像的林地类型判读解译。(2)根据黄龙山林区三期遥感影像光谱特征和研究目的,将黄龙山林区林地类型划分为有林地、灌木林地、疏林地和其他林地。研究表明1986年-2008年黄龙山林地经历了较大幅度的波动变化。主要表现为有林地和其他林地面积持续增加;灌木林地、疏林地面积逐年减少。林地内部存在相互转化,主要转化方向为灌木林,疏林地向有林地转化。相比于1986年-1997年,1997年-2008年黄龙山林区林地的转化速率在逐渐减缓,有趋于向稳定状态发展。对黄龙山林地时空变化分析表明,林地变化率顺序由大到小的顺序为有林地>灌木林地>疏林地>其他林地。并以500m×500m格网对各类林地相对变化率进行制图表达,结果见图5-2。(3)黄龙山林区位于陕北黄土高原东南部,在全国生态环境建设总体布局中属“黄河上中游地区”,具有重要的生态区位。国家的天然林保护工程、退耕还林工程和封山育林政策在黄龙山林区林地时空变化过程中发挥了重要的政策法规驱动力作用。(4)通过对黄龙山林区各类林地构建经济驱动力模型发现:黄龙县国民生产总值与黄龙山林区有林地及其他林地变化有较大相关性,灌木林地面积变化主要受社会固定资产总投入影响,农业人口比重是黄龙山林区疏林地变化的重要影响因素。更多还原

【Abstract】 The temporal and spatial changes of different woodland in Huanglong Mountain in 1986, 1997, and 2008, were detected based on the data of remote sensing information and vegetation resources. We adopted the advanced remote sensing image processing techniques such as image fusion, band combinations, geometric corrections, and cropping to improve the spectral and spatial quality of the satellite images. Collected the data of vegetation resources in the three periods, and established the supervised classification and interpretation signs to obtain the information of woodland. Additionally, in order to examine the effect of potential topographic factors (i.e. elevation, slope, and aspect) on the variation of woodland, the Digital Elevation Model (DEM) performed in ArcGIS to analysis the temporal and spatial characteristics of vegetation in the study area. Moreover, we analyzed the role of National Forest Laws and Policies on the variation of forestlands in Mt. Huanglong. Using the regression analysis to choose te optimal social-economic factors depending on the National Economic Statistics during 1986-2008, and then establish the driving force model for the variation of forestlands in Huanglong Mountian. Our results showed that:(1) the OIF index indicated that the band combination 543 provides a plenty of information about remote sensing image, and has minimum redundant information among different bands. Moreover, the colors of the R (5) G (4) B (3) bands combination are suitable for human eyes, which is the best for distinguishing different woodland types.(2) The results showed that there was a significant fluctuation of woodland in Huanglong Mountain. The forestedland and other types of woodland were increasing constantly, while the area of shrubland and open forest land was shrinking. There were internal conversions between different kinds of wood land, and the main trend were that the shrubland and open forest land transferred to forestedland. The conversion rate of forestry resources in Huanglong Mountain was slow down from 1997 to 2008 compare to it was from 1986 to 1997.(3) Mt. Huanglong is located at Southeastern Loess Platwau of Shaanxi province, which belongs to“the upper and middle Yellow river region”in the overall layout of the National Ecological Environment Construction Planning. Thus, The Natural Forest Protection Project, Grain for Green Project, and Closing Mountain are effective measures to drive the temporal and spatial variation of forestlands in Huanglong Mountain.(4) The socio-economic driving force model demonstrated that the forestedland has high relevance to the GDP increasing and other types of forest variation, while the total investment in fixed assets and the proportion of agricultural population were respectively of high relevance with the area change of shrub forest and open forest land in Huanglong Mountain.更多还原