【摘要】 利用遥感技术,提出了综合遥感指数、非监督分类和监督分类的自动识别滑坡体的方法:先构建4个遥感指数(植被指数SAVI、水体指数MNDWI、建筑指数NDBI和云指数NDCI)合成新影像图,用SAVI指数阈值分割新影像图确定无植被区,最后对无植被区进行非监督分类和监督分类区分滑坡体的方法。以TM/ETM+影像为数据源,应用此方法提取了地震重灾区——汶川县的滑坡体,分类精度达93%,进而研究了滑坡体空间分布特征并对其造成的生态服务价值损失进行了评价。结果表明,滑坡体面积随着坡度的增加呈现先增后减的趋势,30～40°之间的滑坡体面积最大;地震共产生滑坡体206.5km2,汶川-茂县,映秀-北川两大断裂带之间的区域受地震影响最大,成为滑坡灾害最为严重的区域,其中映秀镇15.9%的乡镇面积被滑坡体所掩埋;地震造成全县151.08km2林地、16.13km2草地和5.11km2耕地丧失,崩塌的滑坡体填充的河流面积达3.45km2;各类生态系统服务总价值损失22646万元,其中林地生态系统服务价值损失最大,占89.8%,虽然耕地价值损失所占比重不大,但滑坡灾害使人均耕地减少45.6m2,加剧了该县耕地面积短缺的局面。更多还原

【Abstract】 In this paper,we present a method for assessing landslides and associated ecosystem service losses based on TM/ETM+ remote sensing techniques.The method includes three major steps.The first step is to calculate the four indexes,Soil-Adjusted Vegetation Index(SAVI),Modified Normalized Difference Water Index(MNDWI),Normalized Difference Built-Up Index(NDBI)and Normalized Difference Cloud Index(NDCI),and layerstack them as a new image.The second step is to use SAVI for identifying non-vegetation areas from the new image.The final step is to execute once unsupervised and supervised classification respectively.We applied this method in the spatial analysis of the landslides in Wenchuan County after the 8.0 earthquake of May 23,2008.The result shows that the area of a landslide increases with land-slope when the slope is smaller than 30-40 degrees,whereas it decreases with slope when the slope is greater.The affected areas of the earthquake-triggered landslides in Wenchuan County are summed to 260.5 km2.The most seriously damaged areas are located in between two major fault zones,such as the town of Yingxiu where 15.9% of its area was affected by land sliding.County-wide,it is estimated that 151.08 km2 of forestland,16.13 km2 of grassland,5.11 km2 of farmland and 3.45 km2 of river channel were destroyed by the earthquake.The associated cost due to the loss of various ecosystem services is about 226.5 million Yuan,89.8% of which is resulted from the loss of forestland.Although the area of the lost farmland is relatively small,it exerts additional stress on the farmland shortage by decreasing the average ownership by 45.6 m2 per person in Wenchuan County.更多还原