Correspondence should be addressed to Deng Jianhui Unfavorable hydrodynamic evolution is considered as the major cause leading to reservoir slope instability and is often modeled by numerical method. However, this simulation is seldom checked by systematic field instrumentation. Taking the opportunity of filling the Three Gorges Reservoir, a system was established in Xietan landslide to monitor reservoir water level, subground water level, seepage pressure, rainfall and deformation, etc, The monitored data during reservoir filling shows that: (1) The water level rise in the bank lags behind the reservoir filling and the lag time depends on the bank permeability; (2) rainfall-induced subground water rise and its lag time is closely correlated to hourly rainfall, indicating that it is not feasible or sufficient to use daily rainfall for analysis; (3) the effect of inverse seepage during reservoir filling on stability is ephemeral and reservoir filling is the major cause leading to bank instability.

The East Open Pits of Daye Iron Mine were closed in year 2000 and the ore beyond the pit limit was planned to be mined by underground ways. In order to facilitate the smooth transition to underground mining, hanging-wall ores in the old pits were scheduled to be mined. Xiangbishan pit is one of the four mining spots selected. In 1979 when the pit was mined to around 36 m a.s.l., a landside of around 70000 m3 occurred in its North Slope, leading to the closure of the pit. The pit was then backfilled to 60m a.s.l. to maintain the slope stability. The hanging-wall ore to be mined is just located below the old slide, so whether the slide will be reactivated and how to control its deformation and stability are the key issues concerned. Based on geological investigation and numerical analysis, it was predicted that the slide would be reactivated and a control measure using the intact ore body as a retaining wall was thus suggested. During mining, displacements were monitored to assure safety. By July, 2003, 383000 tons of ore were successfully mined. Though the observed displacement reached 8795 mm, no major failure happened. The final slope is 110 m high with an angle of 70.

Chinese historic documents recorded that on June 1, 1786, a strongM=7.75 earthquake occurred in the Kangding-Luding area, Sichuan, southwestern China, resulting in a large landslide that fell into the Dadu River. As a result, a landslide dam blocked the river. Ten days later, the sudden breaching of the dam resulted in catastrophic downstream flooding. Historic records document over 100,000 deaths by the flood. This may be the most disastrous event ever caused by landslide dam failures in the world. Although a lot of work has been carried out to determine the location, magnitude and intensity of the 1786 earthquake, relatively little is known about the occurrence and nature of the landslide dam. In this paper, the dam was reconstructed using historic documents and geomorphic evidence. It was found that the landslide dam was about 70 m high, and it created a lake with a water volume of about 50 106 m3 and an area of about 1.7 km2. The landslide dam breached suddenly due to a major aftershock on June 10, 1786. The peak discharge at the dam breach was estimated using regression equations and a physically based predictive equation. The possibility of a future failure of the landslide seems high, particularly due to inherent seismic risk, and detailed geotechnical investigations are strongly recommended for evaluating the current stability of the landslide.

总结了1994～2000年问，作者在三峡工程永久船闸高边坡的第一期和第二期位移反分析方面的土要研究成果，包括理论与工程应用、边坡岩体变形机制及其数值模拟方法等。认为位移反分析理论研究很重要，但是由于岩体变形的非连续性，工程应用上应该加强岩体的宏观变形行为及其数值模拟方法的研究。就三峡工程的块状结构边坡而言，岩体的变形主要由松动区的变形构成，在变形分析中将松动区等效为一种弱化的线弹性材料，模拟其宏观变形行为是合适的。

滑带土抗剪强度是滑坡稳定性分析和防治工程设计中十分重要而又难于确定的参数之一。因此，基于临界状态假定的二维反分析方法得到了广泛应用。但是，自然界发生的滑坡基本上呈三维形态，其主滑方向有时变化也较大，使得有必要研究一种滑带土抗剪强度的三维反分析方法和滑坡稳定性的三维评价方法。笔者建议了一种基于强度折减概念的滑带土抗剪强度反分析方法，即通过逐步折减滑动面的强度参数，使滑动面的塑性区完全贯通，此时，塌滑体处于极限状态，所用强度参数即为滑带土的平均抗剪强度参数。从洪家渡水电站1#塌滑体计算成果来看，反演的滑带土摩擦角较二维反演值低4.1°，反映了滑坡体的三维效应，验证了所建议方法的可行性。

结合乌江洪家渡1#塌滑体加固设计工程，探讨了滑动面已知的条件下滑坡稳定性的三维分析方法。首先，模拟了塌滑体在天然状态下处于极限平衡状态时的滑动方向，并使用其应力成果计算了塌滑体的三维安全系数；其次，建议了基于强度折减概念的三维加固安全系数计算方法：最后，提出了加固设计的进一步优化方案。计算结果表明，塌滑体的滑动方向与局部地形等高线近于垂直。主滑方向变化较大，自后缘的NE15方向逐步过渡到中部的NS向，最终过渡到前缘的NE30方向，以致很难作出能表征主滑方向的地质剖面。基于应力计算成果的安全系数计算公式不能充分考虑滑坡体的抗滑潜能，也不能充分反映滑坡体的滑动方向，安全系数偏大。基于强度折减概念的三维加固安全系数计算方法具有有限元等方法的优点，较好地克服了上述局限性，可以合理地评价加固后的滑坡稳定性。

On 13 July 2003, a landslide with a volume of approximately 20

基于对茅坪滑坡体1993～2000年的降雨、水库水位、变形等监测资料的综合分析，初步认为茅坪滑坡体复活是其东侧白岩危岩体中不合理的采煤活动引起的。采煤活动诱发的危岩体开裂和崩塌，加强了滑坡体与危岩体的水力联系，间接扩大了滑坡体的受雨面积。因此，从滑坡体变形的影响因素上讲，降雨（特别是强降雨）的影响较大，水库蓄水和库水位波动的影响范围主要集中在滑坡体前缘。

介绍了使用CSD-60型滑动变形计在三峡工程永久船闸北坡进行的现场监测试验及其有关成果。成果分析表明：（1）边坡岩体的变形具有明显的分区性，主要变形区被定义为边坡岩体松动区；（2）松动区岩体的变形主要由结构面的“张开位移”构成，“张开位移”平均占测孔的累积变形的66.7％；（3）岩体的变形量与开挖强度关系密切；（4）监测区段边坡是稳定的。

首先分析了三峡工程船闸边坡有限元计算和现场监测的位移成果，说明岩石边坡位移反分析必须考虑岩体的松动因素，本构模型可以使用弹性模型；其次介绍了松动区的有关物理力学性质，以及在计算模型中的考虑方法；最后以基于BP网络和遗传算法的位移反分析模型反演了永久船闸边坡岩体及其松动区的弹性模量。