| Coastal protection forest is the only forest ecological project in coastal area which is influenced greatly by both human and natural factors. And now, the problem of the construction of coastal protection forest reflects in both micro and macro aspects. Macroscopically, increase and decrease of forest caused by man-made factors, cutting or regeneration, exploit activities in coastal area, for instance. And microscopically, stability of the ecosystem would also decided by the structure of the forest. How to optimize the forest macroscopically, and improve the stability microscopically is what we are striving to solve.According the considerations above, this thesis analysed the dynamics of the land utilization both in quantitative and landscape perspective based on remote sensing and geographic information system, cooperated with field survey and lab analysis. Some pre-existing achievements were also considered. We also tried to optimize the different components of the forest system. An Indexes System was established to evaluate the seven configuration patterns summarized. The optimum proportion was obtained on the condition that the maximum benefits of the forest were achieved. Results were showed as follows:(1) Land utilization of the continental facies of pingtan island were extracted based on remote sensing images. The three remote sensing images were TM/ETM/ALOS, obtained in 1996/2002/2008 respectively. Topographic condition is complex there, and can hardly obtain ideal result from spectral signature aspect, only. So, ancillary data were used to improve the classification precision, and an ideal result was obtained. All the three classification precisions are above 85%, and the Kappas are also above 0.8. So, it can be greatly improved by cooperating topographic factors when the landform is not so good.(2) Classification system catered to coastal protection forest was established. After classification, three maps were overlapped pairwisely. Three indexes were chosen to describe the dynamics of the land utilization. It changed more drasticly in the year 1996 to 2002 than that 2002 to 2008. The area of coastal protection forest increases or decreases along with the other six kinds especially farmland and other types. In the wind gap area, it was also influenced by exploit activities. (3) Factors that influenced the coastal protection forest were analysed. By the establishment of the classification system, transform matrixes were analysed, and also the changes on different terrain factors, slope, aspect, elevation. It showed that human activities were the key factors; of the three terrain factors, elevation influenced more than the other two, slope ranked the second. When the elevation above 100 m and the slope above 15°, forest changed more; wind was not the main factor.(4) In landscape perspective, several indexes were selected on the consideration of their ecological meanings. Dynamics of landscape were analysed on both class and landscape level. It showed that the degree of the landscape fragment was on the trend of increase.(5) Based on the different ecological problems caused by terrain factors, wind and human activities, we divided the region into four parts, and because of their no connection, nine subregions are nominated。(6) The methods to optimize the three components of coastal protection forest were analysed respectively based on the site condition and functions they should play. Forest behind the shelter belts was studied in particular. Seven main configuration modes were summarized and evaluation system was established. The optimum proportion was calculated on the condition that the maximum benefits of the protection forest achieved by analytic hierarchy process, cooperating field survey and lab analysis, and some pre-existing achievements were also considered. The optimum proportion of the seven types:Mixed Forest of Casuarina equisetifolia and Acacias, Mixed Forest of Casuarina equisetifolia and eucalyptuses, Mixed Forest of Casuarina equisetifolia and shrub, Mixed Forest of Casuarina equisetifolia and other intruduced trees, Mixed Forest of Casuarina equisetifolia and several other tree species, Agro forestry of fruit trees and Casuarina equisetifolia, Mixed Forest of Casuarina equisetifolia and pines were 22%, 17%, 8%, 16%, 12%, 17%,9% respectively. |