|Alternative Title||Landslide susceptibility mapping in Baxie River Basin|
|Place of Conferral||兰州|
|Keyword||巴谢河流域 滑坡 易发性评价|
基于遥感解译与野外调查，查明了巴谢河流域滑坡的位置、数量、分布及规模等，总结了流域内滑坡的类型及分布特征。通过遥感、地形与地质等多源数据集成，实现滑坡影响因素的提取，选用斜坡单元作为评价单元，定量地分析了滑坡与其影响因素间的关系。从全区249处滑坡中随机选择70%的样本数据用于构建EBF、频率比及确定性系数3种评估模型，而剩下30%的滑坡点用于模型的验证，绘制出滑坡易发性分区图。同时利用成功率曲线与预测率曲线对评价结果进行精度检验，探讨了模型的可信程度。获得的主要研究成果如下：(1) 流域内滑坡规模以大、中型为主，厚度以中、深层居多，沿河流沟谷线性展布；河流两岸滑坡呈不对称发育，且在其两岸沟口处多发育大型滑坡；在黄土厚度大的中、下游地段多分布有大型、深层滑坡；同时显现出同一地点复活性较强的特性；滑坡在时间上具有多期性，后期发生的滑坡常改造前期形成的滑坡。(2) 滑坡分布与地形地貌关系密切，高程2000~2200m、坡度15°~30°、起伏度100~200m、阳坡坡向、切割深度50~100m及沟谷密度1~2.5km/km2的范围内，滑坡尤为发育；滑坡与地层岩性、植被、人类活动等有较好的对应关系，其中岩性为马兰黄土及泥岩、NDVI在0.2~0.3与距离道路600m的范围内滑坡易于发生。(3)把区内滑坡易发性程度分为5个级别：极低易发、低易发、中易发、高易发、极高易发。通过计算得到，EBF模型、CF模型、FR模型成功率曲线下方面积分别为0.8038、0.7924、0.8088，预测率曲线下方面积分别为0.8056、0.7922、0.7989。可以看出，3种模型得到的滑坡易发性区划图对滑坡空间分布的反映与预测较为相似，曲线下方的面积都很接近，但综合考量成功率与预测率来看，EBF模型的评价结果要略优于FR模型、CF模型。
Baxie River Basin is a typical area of landslide intensive development in Linxia Basin, Longxi area, Gansu Province. The landslides are widely distributed, activities are frequent, threaten the safety of life and property of local people. In order to effectively formulate the macro-control policy of regional landslide hazard, we need to make regional zoning for landslide hazard. The application of GIS has greatly improved the working efficiency of landslide susceptibility evaluation and provided scientific basis for landslide disaster prevention and control.
Based on the field investigation and remote sensing interpretation, the location, quantity, distribution and scale of the landslide in the Baxie River Basin were identified, and summarized the types and distribution characteristics of landslides in the basin. This paper extracted landslides conditioning factors through data integration of remote sensing, topography and geology. Slope unit was used as the evaluation unit, a total of 249 landslides were detected, 70% of landslides were randomly selected to construct evidential belief function (EBF), frequency ratio and certainty factor models, while the remaining 30% were used for the model validation. The landslide susceptibility maps were generated. At the same time, success rate curve and prediction rate curve were applied to validate the accuracy of the landslide susceptibility map, and the credibility degree of the models were discussed.(1) The landslides linear distributed along the river valley, most of which are medium-deep seated and configure a large-medium scale. The landslides on both sides of the river show asymmetric development and the large landslides are more likely occur at the gully entrance. The large, deep landslides are mostly distributed in the the middle and lower reaches of thick loess. At the same time, it shows low stability after sliding in the same place. Landslides in Baxie River Basin were formed in several periods, during which the later formed always remould the earlier formed back and forth.(2) The distribution of landslides is closely related to the landform and physiognomy, presenting a high concentration in the ranges of elevation 2000~2200m, slope angle 15°~30°, relief amplitude 100~200m, sunny slope, cutting depth 50 ~100m and gully density 1~ 2.5km/km2. In addition, the distribution of landslides correlates well with the lithology, vegetation covering and human activities, especially with the lithology of Malan loess and mudstone, Normalized Difference Vegetation Index of 0.2~0.3 and scope that 0~600m distance from the roads.(3) The final landslide susceptibility map was classified into five susceptibility regions: very low, low, moderate, high, and very high. The area under the success rate curve for the EBF,frequency ratio and certainty factor models are 0.8038, 0.7924 and 0.8088 by Calculation. The area under the prediction rate curve for the EBF, frequency ratio and certainty factor models are 0.8056, 0.7922 and 0.7989. It can be seen that the landslide susceptibility maps obtained by the three models is similar to the response and prediction of the spatial distribution of the landslide, and the area under the curve is very close. However, the comprehensive evaluation of success rate and prediction rate, EBF model performs slightly better than other models.
|陈卓. 巴谢河流域滑坡易发性评价[D]. 兰州. 兰州大学,2017.|
|Files in This Item:||There are no files associated with this item.|
|Recommend this item|
|Export to Endnote|
|Similar articles in Google Scholar|
|Similar articles in Baidu academic|
|Similar articles in Bing Scholar|