基于MaxEnt模型的怒江大峡谷泥石流易发性评价和成因

为进行泥石流易发性分区及环境因子的敏感性评价，选取降水、高程、坡度、坡向、径流量、岩性、断层、归一化植被指数、土壤侵蚀模数等9个环境因子，基于最大熵(MaxEnt)模型对187个泥石流沟道进行易发性评价。结果表明：导致泥石流易发的主导因素依次是径流量、高程、降水、岩性、断层、坡度、归一化植被指数、土壤侵蚀模数、坡向。依据泥石流灾害易发概率进行易发性分区发现，贡山—福贡段泥石流易发性最强，模型AUC值为0.974,标准差为0.010,模型分析结果与实际结果一致。与研究区山洪沟道9个环境因子进行对比评价发现，泥石流对环境因子的敏感性要比山洪复杂。通过能量分析的方法证明泥石流环境因子的敏感性复杂的原因，表明流域汇水能量对灾害的形成贡献率最大。

Assessment and Causes of Debris Flow Susceptibility in the Nvjiang Grand Canyon Based on Maxent Model

In order to study the division of debris flow vulnerability and the sensitivity evaluation of environmental factors, 9 environmental factors, including precipitation, elevation, slope, aspect, runoff, lithology, fault, normalized differential vegetation index and soil erosion modulus, were selected to evaluate the vulnerability of 187 debris flow gullies based on the maximum entropy (Maxent) model. The results showed that the main factors leading to debris flow were runoff, elevation, precipitation, lithology, fault, slope, normalized differential vegetation index, soil erosion modulus and slope aspect. Based on the probability of debris flow hazards, the vulnerability zoning was carried out. It was found that the Gongshan Derung and Nu Autonomous County-Fugong section had the highest vulnerability, with an AUC of 0.974 and a standard deviation of 0.010. The results of the model analysis were consistent with the actual results. Compared with mountain torrents, the sensitivity of debris flow to environmental factors is more complex than that of mountain torrents. The reason of the complex sensitivity of environmental factors of debris flow is proved by the method of energy analysis. Runoff energy is shown to have the highest contribution to the formation of disasters in the Salween River Grand Canyon Debris.