Dissertation > Excellent graduate degree dissertation topics show

Loess Plateau in southern land use change analysis of land use spatial distribution pattern analog

Author: WangMeiXia
Tutor: RenZhiYuan
School: Shaanxi Normal University
Course: Physical Geography
Keywords: Southern Loess Plateau 1000 River Basin Land use Binary Logistic Regression
CLC: F301
Type: Master's thesis
Year: 2011
Downloads: 126
Quote: 0
Read: Download Dissertation

Abstract

Land is the basis of human survival and development, China is the world's land resources on more serious damage to the country, the southern region of the Loess Plateau is one of the most serious damage to land resources, densely populated area, industrial centralized distribution and The policies of the 1990s the implementation of returning farmland to forest and grassland policies proposed and implemented in the late 20th century, large-scale development of the western region, as well as accelerate the development of urbanization are more obvious influence land use patterns in the area. Analysis of land use change in the law and its drive mechanism, the area of ??land use and spatial distribution pattern simulation can reveal not only the law of the spatial distribution of land resources, and the rational use and effective management of land resources in the area to provide guidance for other similar to the area to provide a reference. This article rely on the Ministry of Education Humanities and social sciences key research base of major projects, \data-based mathematical statistical methods, remote sensing and geographic information system technology as an important support, With the Erdas Imagine9.2 software, software that ArcGIS9.3, SPSS 18.0 software File Convert software such as remote sensing image data interpretation, data extraction and classification, data rasterization, spatial overlay analysis and calculations, data format conversion, regression modeling and inspection, and then analyzes the region's large-scale as well as the typical small scale land use spatial and temporal evolution, revealing the driving force on the land use change mechanism, based on the use of relevant software and models to simulate the regional land use spatial pattern. The main conclusions are as follows: (1) 1980 to 2005 the number of changes of each land use type in farmland and construction land in the southern region of the Loess Plateau, where land for construction of the annual rate of change is also the largest. Each type of land use land resources total share size sort always cultivated land gt; grass gt; woodland gt; construction land gt; waters gt; unused land, which increased the proportion of construction land up to the rest of the land use types accounted for the proportion of a slight change. During this time the transfer direction of the main types of land use is arable land to construction land transfer, the transfer of arable land to grassland, and grass to the transfer of arable land and woodland to grassland. (2) 1000 River Basin from 1980 to 2005 the area of ??each land use type, from the time change, where arable land has been showing a decreasing trend, the area of ??other types of land use types, increases and decreases in different years, but the building land growing trend. 1000 River Basin in the stage of land use structure is not changed greatly. Overall the proportion of cultivated land and grassland larger rarely followed by the proportion of woodland, water and land for construction. From space, the various types of land use types have switched in and out. Integrated the direction of the transfer of the land use types in each time period, and found that the transfer of land-use types occurs mainly between farmland and woodland, grassland, woodland and grassland, farmland and construction land. (3) by means of the the SPSS18.0 software established southern Loess Plateau and typical area of ??each land use types and the driving force between the binary logistic regression model and simulate the spatial distribution of land use pattern. Inspection found that the simulation results are better by the ROC curve of the simulation results, which the ROC values ??of the various types of land use in the southern region of the Loess Plateau: arable 0.870, 0.850 of woodland, grassland 0.779 waters 0.918, land for construction 0.847, unused land 0.984. 1000 River Basin ROC values ??of various land use types were: the arable 0.942 woodland 0.953, lawn 0.894 waters 0.973, construction land 0.979. It can be found that, compared to the southern Loess Plateau region, higher goodness of fit of the small-scale simulation of the 1000 River Basin. But there a localized simulation results are not consistent. (4) by binary logistic regression analysis, we found that different driving factors of land use types in the spatial distribution of different degree of influence, including the use of space throughout the Loess Plateau in southern land distribution factor and slope factor affecting the earth GDP. 1000 River Valley small scale land use spatial distribution affect the largest is the land GDP factor and major transportation routes. The main innovation of this paper: (1) To avoid the large scale within the data space of the Loess Plateau in southern autocorrelation binary logistic regression analysis results, the paper using a random sampling method, a random sample of 5% of the sample as involved in the regression analysis. (2) previous studies focused on the dynamic simulation of the change in the number of land-use, the study with the binary logistic regression analysis method focuses on the simulation of the spatial pattern of land use, and during the simulation, considering the natural, social and economic drivers of impact of changes in land use analysis of the spatial distribution of land-use impact mechanism.

Related Dissertations

CLC: > Economic > Agricultural economy > Agricultural economic theory > Land Economics
© 2012 www.DissertationTopic.Net Mobile