| Landscape structure and composition change with time flowing, which is always the key topic of landscape ecology research. The categories, shapes, size, quantities and spatial pattern of landscape patches interact with all natural and human processes, and those processes affect local ecological process. With the change of the production mode of human society, the influence of human activities on the ecological environment is increasingly severe. After the Reform and Opening Up, China experienced a rapidly development. However, the economical gaps between regions are increasingly intensified. In order to reverse this situation, China’s Western Development Program was undertaken at 2000. In addition, people realized that environment is as important as economic development. Thus the Grain to Green Project started the pilot in 13 provinces at the same time to balance the development and environmental protection. Qiandongnan Prefecture with high biodiversity is one of the most important national forest regions. It’s also the hometown for 33 ethnic minorities. Based on this, it was chose as the key development area and the first pilot of Grain to green project. Therefore, it has significant theoretical and practical value to study landscape pattern change by multi- subject cooperates to help environmental protection and decision making.The land use map of Qiandongnan Prefecture were extracted from the remote sensing images of Landsat ETM+/TM in 1993,1999,2009 and Landsat8 in 2013. Based on land use maps, DEM, forestry maps and the related statistic data, this study analyzed the change of farmland and forest before and after the Project of Returning Cropland to Forest. The landscape change and its driving factors were also analyzed this this thesis. The results and conclusions are as follows:(1) The results showed that the forest was decreasing and the farmland was increasing from 1999 to 2013. According to the spatial distribution map of forest and farmland, the forest in the north and northwest, where were the key area of the Project of Returning Cropland to Forest, was increasing and the farmland in the south and southwest is decreasing. However, the forest in the south and southeast was decreasing. Which means that forests were turning back to farmland while the implementation for reforestation. The results of analysis the of land use changing over 25°showed the land use changing character during the 1999 and 2009 had the same trends of the whole area. However, the farmland decreased and forest increased during the 2009 and 2013. It means that the Project of Returning Cropland to Forest achieved some success. During the time period of 1999-2013, forests over the altitude of 1000m declined; in contrast, farmland, built-up and grassland increased. The NDVI over these area shows the vegetation became better. Although the highest value of NDVI dropped, the lowest value rose and the area of medium coverage grew bigger. This situation is more obvious in the northern and northwestern part of Qiandongnan.(2) During the two periods of 1993-1999 and 1999-2009, the main transformation of landscape was from forest and grassland to farmland. From 1993 to 1999, with the small patches merged to larger ones because of the disappearance of small patches of natural forest and grassland, the patch number decreased dramatically. While the mean patch size and AWMSI were increased. The Shannon’s diversity index, Shannon’s evenness index and Edge density were decreased. That means the segmentation of landscape were reduced and the layout became more concentrated. From 1999 to 2009, the mean patch size was continually increased and all the other landscape indices were decreased. The transformation from forest to farmland occurred in the southern area and the transformation from grassland to farmland happened in the northern part. During the period of 2009-2013, the forest kept stable and the farmland decreased slightly. The most notable change was the increasing of residential area. The landscape indices showed that the patches in Qiandongnan Prefecture became more complex and multiple.(3) Topography, climate change and extreme weather were chose as the biophysical factors, and population, GDP, industrial structure and policy were chose as the socioeconomic drivers to explore the relationships between driving factors and landscape pattern changes. Landscape pattern change is the result of the combination of natural and human factors. However, the human activities are the main factor causes the landscape change, especially to the forest and residential area. The results of the correlation analysis shows that population density has a significant negative correlation with the forest and significant positive correlation with residential area. Population density has a weak negative correlation with edge density. |
PDF Full Text Request