Font Size: a A A

Huai Basin Landscape Pattern Dynamic Process Research

Posted on:2014-02-13Degree:DoctorType:Dissertation
Country:ChinaCandidate:P F WuFull Text:PDF
GTID:1220330398494850Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
Landscape ecology as an emerging discipline, is in a stage of rapid development, and the theory and methods constantly perfect. Landscape pattern, as an important research field of landscape ecology, plays an important role in landscape ecology study. In this paper, Yanhuai Basin and its surrounding areas is the study area, as important ecological barrier in northwest of Beijing, there has serious ecological and environmental problems, such as drought, lack of vegetation. So, study on landscape pattern dynamics of this region has important practical significance.In this paper, six landsat images were selected as basic data to interpret landscape maps of study area with combining method of visual interpretation and decision tree, and to study landscape pattern dynamics of study area. This study mainly includes four aspects:(1) Dynamic study of landscape types’ size and structure, which focuse on analysis of the proportional relationship between various landscape types, as well as their mutual conversion during dynamic changes;(2) Dynamic study of landscape types’ spatial structure, which focuse on spatial distribution uniformity changes of landscape types;(3) Identification of key landscape types during landscape pattern dynamic process, in which, transfer matrix is regard as a network to handle, and complex network method is used to identify key landscape types during landscape pattern dynamic process;(4) Study on spatial expansion process of agricultural landscape, the key landscape types were identified in the third part, this study studied the spatial expansion process of farmland landscape. A new landscape expansion index is proposed on the basis of previous studies, and the original and expansion patch area are the parameters of this index. This index can not only quantitatively indicate spatial expansion size of patches, but can also accurately identify the spatial expansion pattern of patches.Results of this study are as follows:1. The dynamics of landscape types’ area and structuralOn the time scale of proximity30years, forest landscape, farmland landscape and grassland landscape are the major componentes of study area landscape. Average proportion of forest landscape area cover the overall landscape area is39.00%, and average proportion of farmland landscape area cover the overall landscape area is37.43%, and average proportion of grassland landscape area cover the overall landscape area is9.61%. The trend of forest landscape proportion decreased after increased, and the trend of orchard landscape, building landscape, farmland landscape and traffic landscape proportion increased, and grassland landscape, water landscape and other landscape proportion presents a decreasing trend.In patch structure aspect, forest landscape and other landscape present a decreasing trend in overall patches number, and the largest patch area and mean patch area have an increased thend, patch shape also tends to be regular. The increase of patches number, and the decrease of the largest patch area and mean patch area of grassland landscape and water landscape, presents a fragmentation trend, and the patch shape tends to be irregular. Orchard landscape, farmland landscape, building landscape and traffic landscape as artificial landscapes are roughly have an expansion process, and patch numbe, the largest patch area and mean patch area are showing an increasing trend, and patch shape tend to be regular.2. Dynamic change of landscape types’spatial structureLorenz Curve and Gini coefficient in economics are applied to the study of spatial distribution uniformity dynamic of landscape types in this part. The results show that, in nearly30years, Gini coefficient of farmland landscape is always the smallest, which indicate that the distribution of farmland landscape is always the most uniform in study area. Four times in1978,1987,1993and1998, Gini coefficient of orchard landscape is always greater than that of other landscapes, which indicate that orchard landscape is the most uneven distribution landscape type during these three periods. In2004and2009, Gini coefficient of grassland landscape is bigger than that of orchard landscape, and its Gini coefficient is the largest among landscape types, which is the most uneven distribution landscape type during2004and2009. In nearly30years, Gini coefficientthe of farmland landscape, building landscape, orchard landscape and forest landscapes whole show decreasing trend, which indicate that their spatial distribution have a trend to uniform. Gini coefficient of grassland landscape and other landscape have a trend to increase, which indicate that their spatial distribution have a trend to uneven in study area.3. Identification of key landscape types during landscape pattern dynamic processIn this study, transition matrix is regard as a network for processing, so as to identificate the key landscape types during landscape pattern dynamic process with complex network method in sociology. The results showed that, grass landscape and farmland landscape are the key landscape types in landscape pattern dynamic process in early time (1978-1987and1987-1993); the role of grassland landscape in overall landscape system is constantly decreases because of continuing degradation, so grassland landscape is not the key landscape type and farmland landscape is the key landscape type during landscape pattern dynamic process in medium-term (1993-1998and1998-2004); forest landscape is the key landscape type during landscape pattern dynamic process in late time (2004-2009).4. Study on spatial expansion process of farmland landscapeIts spatial expansion process is studied in this study that is the key landscape type identified in the third part of this study. A new landscape expansion index was proposes on the basis of previous studies, and the original and expansion patch area were the parameters of this index. This index can not only quantitatively indicate spatial expansion size of patches, but can also accurately identify the spatial expansion pattern of patches (adjacent expansion pattern and external expansion pattern). During nearly30years, the values of landscape expansion index are mainly concentrated in the interval of (-1,-0.8), which indicates that expansion pattern of farmland patches are mainly large number and small-scale adjacent expansion pattern. And the values of landscape expansion index have a decreasing trend, indicating that the expansion scale decreased.
Keywords/Search Tags:landscape pattern, landscape structure, complex network, landscape expensionindex, Yanhuai Basin
PDF Full Text Request
Related items