Spatial Distribution Change Pattern Of Forests In China Based On Remotely Sensed Data

Forest is an important part of the biosphere, which is crucial in maintaining theecological balance. Distribution changes of forest, such as the increase and decrease offorest area, types of succession, will have an impact on the environment. New forest,defined as the forest section turned from non-forest, is a vital part in the spatial changepattern of forest pattern. Generally, detection of forest change mainly follows themethod of change detection, which compares data of two different time to find thepossible emerge of new forest. However,for example, the disturbance of afforestation,where the afforested area usually lead to a change from non-forest to forest and simpledetection method is not available. In this paper, we developed a new method toincorporate the MODIS classification data and NDVI remote sensing data as the datasource, and used the method of statistical histogram to get the NDVI time series offorest, and then after the classification using decision tree, the forest in China in therecent10years at the spatial resolution of500m was obtained. After that, we introduceddata of forest distribution and MODIS NDVI data, combined with the national forestresources inventory data as validation data, we employed approaches of changedetection, estimation of vegetation cover fraction and allocation of pixel area, andfinally, we got the forest distribution at national scale of500m and the spatial patternchange of new forest. The result shows that:(1) extracting forest information fromNDVI characteristic curve method can well reflect the forest distribution of China;forest spatial evergreen forest distribution mainly distributed in the temperate regionsof northern Yunnan, Hainan and other southern subtropical zone and Greater KhinganRange, nearly11years of forest pixels in the pixel average proportion of the total33.72%, forest distribution range increased, increasing the area are mainly located inCentral China, North China.(2) China new forests are mainly distributed in thenortheast of Inner Mongolia and Sichuan Basin, from2002to2011Chinese new forestcoverage has expanded each year, the new forest in forest land in the proportion rose from10%in2002to44.2%in2011.(3) NDVI characteristic curve of forest methodcan well extract the forest information classification, and the accuracy of extractionresult is better than MODIS classification data...