山区县域景观格局尺度效应研究

发布时间：2017-12-31 08:31

本文关键词：山区县域景观格局尺度效应研究　出处：《山东农业大学》2012年硕士论文　论文类型：学位论文

【摘要】：景观格局分析是景观生态学研究的重要组成部分，对于生态过程的理解至关重要。通过对景观格局进行评价和动态演化分析可以反映出一个范围或者地区的景观格局、土地利用模式等方面的问题。任何一种景观格局变化最终都必然落实在一定的时空范围内，因而自然也就离不开尺度，具有明显的尺度相关性。通过分析同一景观格局在不同尺度下的特征及变化规律，发现研究区景观的尺度效应，从而更加精确的揭示其景观格局特点。本研究首先在建立栖霞市空间数据库的基础上，对研究区内景观类型进行划分，获取了景观格局空间信息；然后利用景观格局分析软件和地理信息系统软件对研究区的景观格局的尺度效应以及动态演化特征的进行定量化分析，主要包括：粒度效应分析；中性景观模型与真实景观的一致性分析；基于Lacunarity指数的多尺度景观格局分析；基于三维景观格局指数的景观格局分析。研究结果表明：（1）由于分析粒度的选取将直接影响景观格局分析的准确性，本研究探寻了景观格局指数随粒度变化的特征。通过对景观格局图进行“粗粒化”，分析了尺度对土地覆被类型面积、景观格局指数的影响，并探寻研究区的最佳分析粒度。研究表明：栖霞市景观格局具有明显的“粒度效应”，并呈现出3种变化趋势：1）指数值呈现有规律的上升或下降，并出现一个或多个较为明显的拐点，即尺度转折点。2）指数值在某一尺度下发生突变，而在其他尺度域内保持稳定。3）指数值变化无规律。不同景观格局指数随粒度变化表现出不同的敏感程度，较敏感的是斑块数、斑块密度和最大斑块指数。（2）SimMap中性景观模型，形成模拟景观的图幅大小、景观内各类型数目以及各类型之间的比例等项与真实景观相同，，通过不同景观指标对中性模型系列的反应，看其在多大程度上代替真实景观。研究表明：模拟景观的部分景观格局指数在一定范围内与真实景观的指数相同，即模拟景观只能在一定范围内代替真实景观，不能完全取代；景观格局指数具有“一值多形”问题，其应用具有一定的局限性。（3）孔隙度指数可以在多个尺度上对景观空间异质性特征进行量测，分析景观的尺度效应，能更好地分析景观要素变化的空间结构规律。通过对栖霞市1987年和2003年的景观格局特征进行分析，揭示了栖霞市16年来景观格局的变化特征及其尺度依赖性。栖霞市2003年各景观类型的空隙度指数值及曲线形状较1987年发生了较大的变化，即2003年空隙度指数较1987年有不同程度的减小，且各景观类型的大小排序及曲线形状不同，说明16年来栖霞市景观格局空间异质性发生了较大的变化。通过对比1987年和2003年的空隙度曲线发现，观测尺度ln（r）≈4.5（r≈90m）为特征尺度，景观具有等级结构。（4）通过比较分析传统二维景观格局指数和三维景观格局指数对山区景观格局及其动态变化定量描述的差异，发现三维类型面积、平均斑块大小和边缘密度与其相应二维指数差异明显，三维景观形状指数、平均斑块分维数、香农多样性指数和香农均匀度指数与其相应二维指数差异不显著，而三维聚集度与二维聚集度则无差别。这主要由于三维景观格局指数的计算过程采用的是斑块表面面积和表面周长，而二维景观格局指数采用的是各斑块的投影面积和投影周长，所以在描述山区景观面积、密度、边界等指标时三维景观格局指数将相对精确，但在测定景观形状、多样性和聚散性等指数时，则与传统的二维景观格局指数差异不显著。总之，三维景观格局指数能够加入地形特征，并且对景观格局动态变化的反映相对明显。
[Abstract]:Landscape pattern analysis is an important part of landscape ecology, is essential for the understanding of ecological processes. The landscape pattern analysis can reflect the landscape pattern of a range or the evolution of regional land use pattern and dynamic evaluation, and other aspects of the problem. Any kind of landscape pattern changes will be implemented in a certain time and space therefore, nature also cannot do without scale, scale has obvious relevance. Through the analysis of the characteristics and changes of landscape pattern in the same under different scales, found that the scale effect of landscape in the study area, so as to more accurately reveal the landscape pattern characteristics.
In this paper, based on the spatial database of Qixia City, division of the study area of landscape types and landscape space information were obtained; then the scale effect of landscape pattern analysis software and geographic information system software in the research area of landscape pattern and quantitative analysis, the dynamic evolution characteristics include: particle size effect analysis; consistency analysis of neutral landscape model and real landscape; landscape pattern analysis based on multi scale Lacunarity index; landscape pattern analysis of three-dimensional landscape pattern index based on the result of the study:
(1) the analysis of particle size selection will directly affect the accuracy of landscape pattern analysis, this study explores the characteristics of the landscape pattern index with the size change. Through drawing "landscape pattern coarsening", analyzes the scale of land cover type area, landscape pattern index, and find the optimum particle size analysis research the research showed that the landscape pattern in Qixia city has obvious "size effect", and presents 3 trends: 1) index value regularly up or down, and one or more obvious inflection point, is the turning point of the scale.2 index value) in a certain scale.3 mutation, while maintaining stability in other scales) index variations. Different landscape index changes with the size showed different sensitivities, sensitive is the number of patches and patch density, largest patch index.
(2) SimMap neutral landscape model, map size to form the simulation landscape, such as the proportion between the number of various types of landscape and various types of the same as the real landscape, through the reaction of different landscape index on neutral model series, see instead of real landscape in a large extent. The study shows that the index of landscape pattern simulation the landscape in a certain range and real landscape index is the same, namely the simulation landscape replace the real landscape only in a certain range, can not completely replace; landscape pattern index has a value of polymorphic ", has a certain limitation of its application.
(3) porosity index on multiple scales of landscape heterogeneity measurement, analysis of landscape scale effect, can better analyze the spatial structure of landscape change. Through the analysis of characteristics of landscape pattern in 1987 and 2003 in Qixia City, revealing the Qixia city for 16 years and its variation characteristics the dependence of the scale landscape pattern of different landscape types in Qixia city in 2003. The porosity index and the curve shape is changed in 1987 2003, compared with 1987 the porosity index decreased, and the size of the sort and curve of each landscape type in different shapes, 16 years ago, the heterogeneity of landscape space Qixia city has changed greatly. Through the comparison between 1987 and 2003 the gap degree curve, observation scale ln (R) = 4.5 (r = 90m) as the characteristic scale, landscape has a hierarchical structure.
(4) by comparing the traditional two-dimensional and three-dimensional landscape pattern index, landscape pattern index quantitative description of pattern and its dynamic variation of mountain landscape, found the 3D type area, mean patch size and edge density corresponding to two-dimensional index differences, three-dimensional landscape shape index, fractal dimension, diversity index and uniformity of Shannon Shannon the index of the corresponding two-dimensional index had no significant difference, while the three-dimensional and two-dimensional aggregation aggregation showed no difference. This is mainly due to the calculation process of 3D landscape pattern index using the plaque surface area and surface perimeter, and the landscape pattern index is used in the two-dimensional projection and the projection area of patch perimeter, so in mountainous area landscape area, density, boundary indicators such as the index of three-dimensional landscape pattern will be relatively accurate, but in the determination of landscape shape, diversity and poly dispersion index When compared with the traditional two-dimensional landscape pattern index, the difference is not significant. In short, the three-dimensional landscape pattern index can be added to the terrain characteristics, and the dynamic change of landscape pattern is relatively obvious.

【学位授予单位】：山东农业大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F592.7;F224

【相似文献】