| The terrain of Nanping city, Fujian is dominated by mountain hills, forest resources are distributed in areas with inconvenient traffic and complex terrain. The traditional forest resources field survey method has a large amount of work, and the time period is long, which seriously restricts the research of forest vegetation landscape characteristics. Based on the forest resource survey data, the thesis researched the spatial pattern of forest vegetation landscape in Nanping city with the remote sensing image. In order to reflect the spatial structure and distribution pattern of main forest vegetation in the study area quickly and scientifically, we analyzed the spatial distribution of forest resources in the study area systematically from macro scale. According to the Nanping City forest distribution situation, forestry production and management needs and considering the differences in external and internal structure of the forest landscape, we chose the land use types and classes of dominant tree species of forest resources survey for the forest landscape classification.Using the spatial analysis function of GIS and the principles and methods of Landscape ecology, the paper conclude the regularity of landscape pattern distribution after studying the landscape spatial pattern of Nanping city and the relationship between landscape and environment factors using the processing and analysis tools of GIS and FRAGSTATS. The study revealed the spatial variation and scale effect of forest spatial structure from the perspective of Quantitative Ecology, provided the basic data and theoretical guidance for the dynamic monitoring of forest resources, ecological restoration, environmental assessment and so on, and also provided a reference for the optimization of the spatial structure and structure of forest resources.The contents and the methods of the study including:(1) Using the latest remotely-sensed imagery to update the data of secondary investigation of forest resources, fix the boundaries and related factors. Classification of forest landscape was based on the updated data of forest resources and DEM. ArcGIS was used as technical supporting systems for data processing. As a result, we will get the distribution maps of forest landscapes of the study area. They were will be used for the base data for further studying on forest landscape of study region.(2)There is a high forest cover in the study area. Forested landscape which covers 72.94% of the total study area is the main landscape types and dominated the structure, function and dynamics of the overall landscape of the region. 8 classes forest landscape distributes inequality: Coniferous forest landscape throughout the study area accounted for 36.12% of the total forest area of ecological landscape. Showing the concentration of contiguous distribution, coniferous forest landscape is the dominant landscape type and the matrix in the area of forest landscape in the region. Referring to areas, the other landscape classes were ranged as follows: non forest > broadleaf forest > coniferous forest > bamboo forest > immature forest > fruit trees > shrubs > non-forest land. Areas of each type were markedly different: non forest, broadleaf forest and bamboo forest area ratio reached 15%, are important parts of the forest landscape in the study area and play an important role in maintaining the forest landscape stability and diversity.(3) From the point of view of landscape scale feature types: connectivity of broad-leaved and coniferous forest landscape are better than other types and the largest patch index of non-forest landscape is much higher than other landscape types, shows that the landscapes are strongly influenced by human activities, which are controlled by a few large patches pattern. The area of fruit trees, shrubs, immature forest and non-forest land are relatively small due to the impact of human harvesting, planting and plaque. Patch distribution of coniferous forest and broadleaf forest landscape are relatively gathered, so the two types have a low fragmentation. No woodland, shrub and immature forest landscape's average patch area are small, have a sporadic distribution, and due to the larger average patch area, patch density of broad-leaved forest landscape is small; non-forest landscape has the largest number of plaques and plaque density, and influenced by human activity, has the higher the degree of fragmentation; coniferous forest and non-forest's edge density and landscape shape index are significantly higher than other landscape types, mostly affected by human activities, and are closely related with the large number of operators planted fir, pine and other coniferous forest tree species as well as a variety of production and business activities in non-forest land.(4)With the support of ArcGIS, the elevation, slope and aspect were extracted from DEM, then the distribution maps of forest landscape of the study area were overly with DEM. The result indicate: Forest landscape types show significant changes with altitude regular vertical distribution: 200 m below, non-forest land is in the absolute superiority; within the range of 200-500 m altitude, coniferous forest landscape occupies is the most; 500 m above, broadleaf forest, coniferous forest and bamboo forest landscape have larger areas than other types. Landscape types are different in the area with different slope, the non-forest landscape is mainly in the area with the smallest slope; above steep slopes, there is almost no non-forest landscape. Landscape types in different slope tends to be slightly different but bamboo have the largest area in the north slope; coniferous forests, uniform, broad-leaved forest, fruit trees, shrubs, immature forest and non-forest land in the west slope have the largest area; however, woodland distributes more in the east slope. |
PDF Full Text Request