| As landslides occurred frequently and caused large economical lose and casualties, it has become one of the major geo-hazards. Landslides and the secondary disaster caused by landslides, such as landslide tsunami, landslide dam and lake, etc, have made the situation even worse, especially in the last30years, on account of the expansion of the growth of population and land use. China is one of those countries, affected by geo-hazards seriously, especially by landslides. In the last5years, there were about20,000geo-hazard events occurred in each year by average and made an average annual casualties1273persons and direct economical lose4,143,000,000Yuan. According to the reports of National Geological Disaster Bulletin from2005to2012, the number of landslide events represents about65%of the number of geo-hazard events. The maximal percentage is in2006with86%. Besides, based on the recent report in2012, the number of geo-hazard events in2012is14,322and caused direct economical lose5,280,000,000Yuan. In this year, the number of landslide events took76%of the number of the geo-hazard events.Because of the proper environmental conditions in Three Gorges area, geo-hazards occurred frequently in the history. Since June,2003, the operation of the Three Gorges Dam impoundment has been started, the probability of the occurrence of geo-hazards has increased. According to a report in2010, there were5386landslides had been detected and recorded. And so far, although there were887geo-hazards had been performed prevention and mitigation measures, there are still a large amount of landslides with potential risk. These landslides can only be under manual monitoring. Therefore, on the view of the important political, economical, social status and urban development needs in Three Gorges area, it is meaningful to do research on the development regularities of landslides in this area, and to study on hazard and risk from landslides instability and reactivation by the impoundment of the reservoir water level and the fast urban development. The results would make sense to protect the properties and live, and to promote urban development in this area.Wanzhou district, Chongqing city locates in the middle of the Three Gorges Reservoir area. The proper geological environment conditions provides the formation chances of landslides. Thus, in Wanzhou district, landslides occurred frequently and distributed all around this area. In this thesis, amount of work has been carried out in Wanzhou district and based on the widely acceptable theory in landslide hazard and risk research. For instance, landslides development regularities in this area has been made effort to and landslide hazard and risk research in different scale have been studied on.At present, there are numerous works have been carried out on landslide development regularities, but there are still some difficulties have not been gone through yet, such as:how to confirm the integrity and the accuracy of landslide records, how to increase the reliability of landslide special-temporal prediction results, how to define the timeliness of landslide development characteristics, how to uniform the standard on landslide failure type classification, how to simulate a real environment to a rational simulation analysis on landslides, and so on. Besides, there are also some problems need to be solved on the study of landslide hazards and risk. For instance, how to define the research period when doing landslide risk analysis, how to express the spatial dynamics of landslide hazard and risk, how to represent the partial failure probability of creep landslides when using traditional failure probability calculation methods, how to improve the reliability of landslides risk results, how to strengthen landslide risk management, and so on.In this thesis, some of the above difficulties will be discussed. Based on landslides data collection in Wanzhou district, landsides development regularities, landslide susceptibility in regional scale and landslide failure probability zoning on site scale will be focused on in this thesis. By the end of this study, the results will be given as follows:(1) According to the geological environment conditions and the morphological characteristics of landslides, landslide development regularities will be studied in Wanzhou district. This study is based on the latest investigation records and performed in two main aspects. One is on the view of the geological environment where exist landslides and the other is on the view of the morphological characteristics of those landslides. The analysis on the former aspect includes the characteristics of landslides distribution on different elevation, topography, hydrological conditions, landslides boundary conditions, landslide composition and forming age, and so on. In addition, the analysis on the latter aspects includes landslides morphological characteristics and landslide sizes characteristics, such as landslide maximum length, average width, area and volume. Base on the above landslides characteristics statistics, the landslide development regularities in Wanzhou district is aimed to summarized.(2) Based on the size distribution of landslides in Wanzhou district, the relationship between different landslide size index are planned to be analyzed. The results in this thesis show, in Wanzhou district, both reservoir water connected landslide or unconnected landslides, the correlation between landslide maximum length and landslide average width is not obvious but commonly, few landslide developed with both length and width in very large value. Besides, there is a certain correlation between landslide area and its maximum length or average width. The larger the area of landslide is, the longer or wider the length or width of landslide will be. However, when the value of the area becomes larger, the distribution of landslides becomes discrete and the correlation between landslide area and its maximum length or average width decline. In addition, the correlation between landslide volume and its maximum length or average width is similar to that between correlation between landslide area and its maximum length or average width. Whereas, when the value of the volume becomes larger, the discrete distribution is more obvious and the correlation is also lower than those between landslide area and its maximum length or average width. What's more, the correlation between landslide volume and area is relatively high. The larger the landslide volume is, the larger the landslide area develops. But, with the increasing of landslide size, the corresponding number of landslides reduce and there shows a discrete distribution and decrease correlation between landslide volume and area.(3) From the perspective of landslide frequency density-magnitude distribution, landslide size development regularities has been analyzed. Landslides data in this thesis has been cataloged into three groups, there are group one with total landslide in Wanzhou district, group two with reservoir connected landslides in Wanzhou district and group three with reservoir unconnected landslides in Wanzhou district. The landslide frequency density-magnitude distribution function and curve will be fitted. The results demonstrate, the three-parameter inverse gamma density distribution function fits the landslide frequency density-size distribution well, as well as the noncumulative exponential distribution function fits landslide frequency density-size distribution on large size. Additionally, the size (such as width, area and volume) of landslides in Wanzhou district tend to concentrated in a certain size range, especially of the reservoir unconnected landslides. This regularity proves that for those landslides connect with reservoir are more likely to develop in larger size (such as width, area and volume). Besides, by comparing the landslide frequency density distribution regularity in different places and triggered by different events, it proved that, this kind of landslide distribute regularity is universal. The formation of landslides are controlled by geological environmental conditions and the external factors, and landslides develop concentrated in a certain size range, the more complete the landslide data is the closer the landslide frequency density-size distribution curve will be fitted to the initial distribution of completed landslide inventory data.(4) Based on clustering model and combined with geological environment data, landslide distribution and landslides frequency density-magnitude fitting curve, landslide hazard analysis will be studied in Wanzhou district. Firstly, there are7indicators, including strata age, elevation distribution and so on, are chosen to perform landslide susceptibility analysis. In this thesis, landslide susceptibility is based on the location of those landslides in Wanzhou district, with raster data statistics, the indicators landslide susceptibility rating map will be get in this step. Secondly, based on the combination of entropy weight method and AHP method, the weight of each indicator will be calculated. In this thesis, the results show that, the weight of strata age and terrain slope are higher than other indicators for landslide susceptibility. Thirdly, cluster model and information model are used to zone landslide susceptibility map in Wanzhou district. This step implements cluster model algorithm programming and optimizing. The result demonstrates although the accuracy of information model is higher than cluster model, the result from information model is too conservative and the result from cluster model is more closer to the actual situation. In the end, combined with landslides distribution map and landslides frequency density-magnitude fitting curve in this area, landslide susceptibility map will be modified and then the landslide hazard map will be get. Over all, the results show the zones with higher hazard rate in the southern part of the river in Wanzhou district distribute wider than the northern part of the river, the zones with higher hazard rate distribute concentrate in the main city and along the upper part of the river bank while along the downstream section of the river, the zones with higher hazard rate distribute scatteredly.(5) By analyzing the distribution of the elements at risk in Wanzhou district, landslide risk zoning will be carried out in this area. In this thesis, the elements at risk include4categories, there are population density, buildings, traffic lines and land resources. The landslide vulnerability zoning result demonstrates, the vulnerability of elements at risk is generally high in the main city of Wanzhou district. Vulnerability distribution of the population is consistent with hazard distribution of landslides. The highest vulnerability distribution of the buildings are concentrated along the junction of Yangtze river and Zhuxi river in the main city and the rest higher vulnerability distribution of the buildings locate on the area with high hazard grade and spread sporadically. The vulnerability of road traffic lines shows the characteristic of section shape distribution. The higher the use level of the traffic lines type, the higher the resilience of the traffic lines should be and under the same intensity of hazard, the smaller the vulnerability of the traffic lines will get. The maximum vulnerability of land resources locate around the main city, and the other higher vulnerability zones generally consistent with hazard distribution of landslides. Besides, based on the analysis of the economic distribution and population density distribution in Wanzhou district, landslide risk zoning map has been gained. The result shows, whether the economical risk or population risk of landslide, the maximum risk value locates around the main city in Wanzhou district.(6) Based on LURR model to analyze the response relationship between landslide displacement and the triggering factors, such as rainfall and the water level of reservoir. Combined with GPS surface displacement monitoring data of San Zhouxi landslide, rainfall records and the water level scheduling information, the responding relationship between deformation of landslides and triggering factors is aimed to be found out. The results in this thesis show, LURR model works well in this analysis and the load/unload response ratio can well represent landslide stability and the characteristics of landslide deformation in different time period. The whole San Zhouxi landslide is stable, but some parts of it are less stable. Rainfall makes contribute to the deformation of San Zhouxi landslide, but the changes of the reservoir water level influences the deformation of this landslide more than rainfall and generally, the front part of this landslide responds first to the changes of the reservoir water level and upper part of this landslide responds later, this shows the traction deformation characteristics of San Zhouxi landslide.(7) Based on the fractal theory and using box-counting method, the fractal dimension of the moving traces of landslide surface monitoring points have been calculated. After that, based on IDW method, the fractal dimension zoning map of San Zhouxi landslide are obtained. Details are, firstly, according to San Zhouxi landslide GPS monitoring data, the moving traces of these monitoring points are analyzed and made as trajectory curves. Secondly, divided the trajectory curves by two ways, one way is by annually, the other way is by the water impoundment cycles. Thirdly, to process and convert the displacement trajectory curve segments into grayscale image. Then, using box-counting method to calculate the fractal dimension for each grayscale image. Finally, based on IDW method to obtain fractal dimension zoning map for each period. The results show, the whole San Zhouxi landslide is more active when during the drop period of the reservoir water level than during the rise period of the reservoir water level. Besides, from a local point of view, parts of this landslide are influenced by the changes of the reservoir water level and with different effects. In general, the deformation on right side of the landslide are serious than on the left side of the landslide, the deformation on the lower part of landslide are serious than on the upper part of the landslide.(8) Based on landslide failure probability calculation by GeoStudio and landslide deformation fractal dimension maps, San Zhouxi landslide failure probability zoning will be studied on. In this thesis, landslide failure probability analysis is based on GeoStudio simulation software. The function module SEEP/W will be used to make transient simulation of the ground water level of San Zhouxi landslide under different reservoir water scheduling cycles and different years. This ground water level simulation will consider the combined effect of rainfall and water level. After this simulation, the function module SLOP/W will be used to calculate landslide failure probability according to different simulated ground water level in different calculation periods. Then, by comparing the reliability of the simulation results and the safety factor values from landslide stability calculation in a certain calculation period, one landslide failure probability result will be chosen to stand for this corresponding calculation period. In addition, in order to modify the real boundry of the partial deformed zone of this landslide, the geological enviroment conditions and the feild survey records of San Zhouxi landslide have been combined with the obtained landslide fractal demension zoning map for analysis. Landslide partial deformation degree map will be obtained from this step and then, based on this landslide deform situation, landslide failure probability get from last step and location of the simulation profile, landslide partical failure probability will be analyzed. The results in this step shows that the failure probability zoning map represents the deformation situation of San Zhouxi landslide well, the failure probability on the right side of the landslide is higher than on the left side of the landslide. And generally speaking, the failure probability of San Zhouxi landslide, when during the drop period of the reservoir water, is higher than during the rising period of the reservoir water. Besides, by analyzing the cracks on the surface of San Zhouxi landslide, the straight and vertical to the main slip direction cracks maybe formed because of reservoir water level fluctuation and those curved and from vertical to parallel to the main slip direction cracks are more likely formed because of the drop affection of reservoir water level.(9) Based on the analysis on the distribution of elements at risk on San Zhouxi landslide, landslide risk map of this landslide is aimed to get. In this part, time scale of landslide risk is considered. The landslide risk map based on two period, one is during the rising period of reservoir water level and the other is during the dropping period of reservoir water level. The elements at risk in this case includes buildings, indoor population and agriculture and weeds land. The results show, when during the drop period of reservoir water, San Zhouxi landslide will under higher landslide risk than during the rise period of rese economic risk, when during the drop period of the reservoir water level is higher than during the rise period of the reservoir water level. In the former period, the range of the buildings' economic risk is2,800to101,000Yuan, while in the later period, the range of the buildings' economic risk is1,600to2,8000Yuan. Besides, for the indoor population risk, some of them are under higher risk during the drop period of the reservoir water level than during the rise period of the reservoir water level. In the former period, the indoor population risk is8persons and in the later period, the indoor population risk is4persons. In addition, for other agricultural and weeds land economic risk, during the drop period of reservoir water the range of the economic risk is933to2,160Yuan/mu, while during the rise period of reservoir water the range of the economic risk is107to933Yuan/mu. |
PDF Full Text Request