尼泊尔地震触发滑坡识别和雪崩形变分析

Nepal Earthquake Triggered Landslides Recognition and Deformation Analysis of Avalanches' Region

  • 摘要: 2015-04-25尼泊尔Gorkha附近发生了Mw 7.8级强震,引发尼泊尔及中国西藏地区大量的山体滑坡和雪崩灾害,造成了严重的人员伤亡与经济损失。基于Landsat 8影像和亚像素相关性匹配技术识别形变失相关区域,并结合目视解译提取震害滑坡与雪崩沉积区域,统计分析了所提取的滑坡数目、面积等与地形因子的关系,同时对珠峰雪崩高发区的冰川形变进行了定量分析。实验结果表明,提取形变场中的形变失相关区域并结合目视解译验证是一种识别震害滑坡与雪崩沉积区域的有效方法;统计结果显示滑坡体随坡度分布呈现高斯特性,多发生在坡度大于30°的地形,易发于河谷两侧,且高坡度更容易触发单体面积大的滑坡体。该研究为利用Landsat 8影像数据进行震害滑坡与雪崩沉积区域提取、雪崩高发区的冰川流速监测预警工作提供了新视角。

     

    Abstract: On April 25th 2015, an Mw 7.8 earthquake occurred near Gorkha, Nepal, which triggered substantial landslides and avalanches in Tibet and Nepal causing severe casualties and economic losses. In this study, based on Landsat 8 images and sub-pixel correlation technology, we acquire horizontal deformation fields of the earthquake region. Combined with the visual interpretation method, we detected the landslides and debris deposition areas of avalanches by extracting the corresponding decorrelation noises. We determined the correlation between the landslides and topography factors. Experimental results show that the extraction of decorrelation noises caused by geohazards from the deformation field, when combined with validation of visual interpretation method, is an effective method for locating the regions effected by landslides and avalanches. Furthermore, results show that landslides are more likely to occur at slope > 30° terrain, both sides of the valley, and high slope is more easily trigger larger single landslides. This study provides a new perspective on detection of earthquake-triggered landslides and debris deposition areas of avalanches, monitoring and early warning of glaciers' velocity based on the Landsat 8 satellite imagery.

     

/

返回文章
返回