中文摘要：

汶川地震造成泥石流形成条件的改变,其次生灾害堰塞湖的危害已开始显现。如何对震后潜在性泥石流堰塞湖进行判识,成为迫切需要解决的问题。选取岷江上游映秀至汶川段为研究区,通过分析震后泥石流形成条件的变化、典型泥石流堰塞湖的危害及松散物质储量,选取潜在性泥石流堰塞湖的判识指标,利用模糊物元可拓模型,建立潜在性泥石流堰塞湖的综合判识模式。通过判识,研究区形成泥石流堰塞湖可能性高的一级支沟有17条,主要集中分布在映秀镇至草坡乡段,此段将是今后受堰塞湖危害的高危地段。

英文摘要：

The Wenchuan earthquake caused numerous landslides and collapses that provided abundant loose solid materials for future mobilization as debris flows. The formation conditions for debris flows were changed consequently. Debris flows will be very active for a long time in the affected area. Their secondary disaster, barrier lake, is beginning to appear. Therefore, it is urgent to identify those debris flow gullies which are poten- tial to form barrier lakes following the environmental changes caused by the earthquake. This paper selected the upper Min River from Yinxiu to Wenchuan as study area, analyzed the changed formation conditions of de- bris flows and the damages of a typical debris flow barrier lake, and interpreted the loose solid materials from TM satellite images and aerial photographs. Then, the identification factors were selected and classified based on the factor independence analysis. At last, the fuzzy matter-element extension theory was used to construct a model to identify potential barrier lakes. The potential future increase in debris flow activity relates directly to the great increase in loose materials. The interpretation results indicated that the area of the landslides and col- lapses caused by the earthquake is 39.40 km^2, accounting for 6.97% of the total study area. Assuming 55% of the landslide and collapse area were their deposition area and the average deposition depth was 10 m, the vol- ume of the loose solid materials can be estimated as 2.13 ×10^8m^3 which was sufficient for debris flow forma- tion. Based on the factor independence analysis, the volume of the loose solid materials per unit area, the dis- charge ratio between the debris flow gully and the main river, the slope gradient of debris flow channel, the main river width, and the included angle between the debris flow gully and the main fiver were determined as identification factors. According to the multi-factor comprehensive identification, among the 55 first-grade gul= lies, 17 gullies have high probabilities for pot