The Real-Time Probability Analysis And Refined Risk Analysis And Control Of The Landslide In Shen Zhen Area

Shenzhen located at the south coastal of China, where the stratigraphic lithology distribution is quite uneven. The rainfall from April to September every year induces many landslides. The landslide leads to great potential threat for the persons and the property. At the same time it influences the local economy development. Therefore, the study on the landslides trigerred by rainfall is important. It is necessary to establish a reasonable and effective risk analysis and assessment system.This research uses the mathematical statistics, linear regression and finite element simulation , combining with the reliability theory to study the landslide triggered by rainfall in Shenzhen. Then the probability of landslide based on real-time rainfall forcast is analyzed and a set of reasonable and effective risk assessment model is established. The main contents are as following:1. The landslide and rainfall data in shenzhen are collected from 2006 to 2009, the threshold of rainfall to cause landslide is analyzed based on the collected date. Results indicate that the threshold is 100mm. when the rainfall is larger than the threshold, the landslide would prone to be triggered and the probability of the landslide is calculated in the next section.2. Based on the software Seep/W and Slope/W, the pore-water pressure uw and the total normal stressσn on the slip surface are calculated when the rainfall reaches the threshold. Then the matric suction (ua-uw) and the net normal stress (σ-ua) are selected as the random variables, the probability of the landslide is calculated by the reliability theory. The real-time probability can be calculated according to the real-time rainfall information.3. By compariting different methods of calculating sliding distance for the landslide, the effective and reasonable influence factors of landslide are found, which are the slope angle and the slope height. Then based on the multiple linear regression method, the formula of calculating the sliding distance for the landslide is obtained by analyzing the collected data.4. Based on the actual situation in Shenzhen, the personnel vulnerability risk is studied. In addition, the economic losses caused by the landslide are calculated.5. Based on the risk assessment results, A warning system of landslide is proposed, particularly for populations in the landslide area. Since the warning is based on the real-time rainfall and only affects the specific landslide involved areas, it is more effective.