基于TM/ETM+数据的南京三区域城市森林干扰指数及分析

以南京紫金山、幕府山和老山为研究对象,利用1992、1995、1998、2001、2003、2005、2007和2011年8期的Landsat TM /ETM+数据进行缨帽变换,通过对变换后的各分量进行归一化操作进而建立森林干扰指数模型,然后进行3个区域森林干扰指数分级操作,最后借助森林资源二类调查数据以及Google Earth影像的目视解译结果,对发展的森林干扰指数分析方法进行了验证。验证结果表明:基于Landsat TM/ETM+数据而发展的森林干扰分析方法是有效且可靠的。南京3个区域的森林干扰在1992—2001年间变化不明显,2001—2005年干扰上升明显,2005年之后下降趋势明显。空间上,幕府山森林受干扰最强,老山林场次之,紫金山最小。每个区域干扰强度的分布也各有特点,但相同的是区域周边的干扰指数明显大于中心地区。驱动南京城市森林干扰时空变化的因素主要包括人口增长、经济开发活动及景区游览等。

Assessing forest disturbance patterns over the three forested areas of Nanjing using multi-temporal TM/ETM+ imagery

Taking the three areas of Zijin, Mufu and Laoshan Mountains located in Nanjing as the case study, using the Landsat TM/ETM+ observations dated in 1992, 1995, 1998, 2001, 2003, 2005, 2007 and 2011, indices including brightness, greenness and wetness derived from the tasseled cap transform were obtained first, followed by the establishment of the forest disturbance index via a normalization approach. Ultimately, grading the forest disturbance severity was made and the forest disturbance analyses were in part validated by using the forest resources inventories coupled with the high spatial resolution Google Earth imagery. Results showed that the forest disturbance analysis methods developed from Landsat TM/ETM+ imagery in the current work were effective and reliable after an intensive validation. Forest disturbance intensity remained almost unchanged during the period 1992 to 2001, giving way to an increase in forest disturbance severity over the time period 2001 to 2005, connecting to a declining trend after 2005. Additionally, the average disturbance values of the three regions differed from each other, Mufu Mountains with the strongest forest disturbance and Zijin Mountains the lowest, which adequately reflects the differences in forest management purposes and approaches. Distribution of forest disturbance severity of the regions also had their own characteristics, but the patterns that higher disturbance severities were observed along the boundaries of the three regions, with lower disturbance levels located in the central portions of the regions, were same. Ultimately, the driving forces responsible for the differences in observed forest disturbances were identified as demographic expansion, mining events, forest logging and forest insects and disease and forest eco-tourism.