(1. School of Civil Engineering and Architecture, Central South University, Changsha, Hunan 4 10075.
2. Guizhou Provincial Traffic Planning Survey and Design Institute, Guiyang, Guizhou 55000 1)
Rainfall infiltration is one of the main factors inducing the instability of soil-rock mixture slope, and this kind of problem has always been concerned by people, but the research on this problem is not systematic and in-depth. In order to deeply understand the instability mechanism of soil-rock mixture landslide induced by rainfall infiltration and study some important characteristics of slope characteristics changing with time, a typical soil-rock mixture slope in Guizhou section of Shangrui Expressway was selected for artificial rainfall simulation test and comprehensive on-site monitoring. The monitoring results show that the sliding deformation zone of soil-rock mixture slope under the influence of rainfall infiltration is between 0 and 4 m below the slope surface, and the slope surface deformation is the largest, which gradually decreases from the slope surface to the deep part of the slope. In the first two hours of rainfall, the average infiltration rate was 86%. After that, due to the increase of surface runoff, the infiltration rate gradually decreased with time, and after a period of time (6 hours), the infiltration rate decreased to a relatively stable value (50%). Rainfall infiltration leads to the increase of pore water pressure in soil, and due to the decrease of effective stress, the soil absorbs water and softens, which leads to the decrease of shear strength of slope soil. This double effect of rainfall infiltration may be one of the main reasons for the instability of soil-rock mixture slope induced by rainfall.
Artificial simulated rainfall; Slope soil-rock mixture; Field monitoring of rainfall infiltration
With the vigorous development of China's capital construction, the strategic focus of national construction has shifted to the western region, and the engineering construction will inevitably encounter loose accumulation media composed of residual slope deposits, colluvial deposits and proluvial. Its material composition is mainly gravel or rubble, gravel or rubble mixed with soil and other earth-rock mixtures, with disordered material structure, poor sorting, poor intergranular binding force and strong water permeability. It is not only different from the general rock mass, but also different from the general soil mass. It is a special geological mass between the soil mass and the rock mass, which is called soil-rock mixture [1]. Soil-rock mixture slopes are classified according to the material composition of slopes, and belong to the same classification level as soil slopes and rock slopes, and are widely distributed throughout the country and even the world [2]. A lot of research has been done on the mechanism of soil landslide and rock landslide at home and abroad, and a complete set of research results has been obtained. Because of its unique properties, such as the complexity of material composition, the irregularity of structural distribution, the difficulty of sample collection, etc., it has brought great difficulties to our research, and the research results obtained are limited [3], so it is necessary to conduct special research and analysis on the soil-rock mixture landslide.
A large number of statistics show that the main inducing factor of soil-rock mixture slope instability is rainfall [4,5]. The landslide above Pier 3 of Pingxi Bridge on Sansui Expressway in Sansui County, Guizhou Province is a typical landslide of soil-rock mixture induced by continuous heavy rainfall in April and early May, 2003, which killed 35 people. The problem of slope instability under the influence of rainfall has always been concerned by people [6 ~ 8], but the research on this problem is not systematic and in-depth enough. In order to reveal the formation and evolution law of soil-rock mixture landslide induced by rainfall, in April 2005, a typical soil-rock mixture slope at the exit of Qinglong Tunnel in Guizhou section of Shangrui Expressway was selected for artificial rainfall simulation test and in-situ comprehensive monitoring. During the test, combined with comprehensive field monitoring, the formation conditions, deformation and displacement characteristics and failure and sliding laws of soil-rock mixture slope under the condition of rainfall infiltration are analyzed, which provides a theoretical basis for better preventing or controlling such geological disasters in the future.
1 test center
1. 1 Determination of test points
The Shanghai-Ruili Expressway under construction is an artery connecting the east and west of China. On April 2, 2005, on the basis of comprehensive investigation of the section from Zhenning to Shengjingguan of Shangrui Highway in Guizhou Province, according to the borehole geological data, the external shape of the slope and the surrounding environment, the K85 +650 -690 accumulation layer at the entrance of Qinglong Tunnel was selected as the artificial rainfall test site. Remove vegetation and other sundries in the area first, and then brush the slope according to the gradient of 1: 2.5. In order to prevent atmospheric rainfall and moisture in the surrounding soil from infiltrating into the test area and affecting the accuracy of the test, when it rains, the test area is covered with colored strips.
1.2 soil properties
In order to find out the basic physical properties of the soil in the test area and the engineering geological characteristics of the slope soil layer, basic physical and mechanical tests and special drilling investigations were carried out. See table 1 for its physical and mechanical properties. Particle analysis test * * * was carried out in group 15, and the average particle grading curve of soil sample was drawn in figure 1. The average gradation characteristic values in the figure are: clay (< 0.005mm) content 0.95%, silt (0.05~0.005 mm) content 8.88%, gravel (> 5 mm). The uneven coefficient Cu is 12.3 1, which indicates that there are many particle size series in soil samples, and the coarse and fine particle sizes are quite different. The curvature coefficient Cc of particle grading curve is 1.59, and the grading is excellent.
Table 1 Basic physical indexes of natural soil
Figure 1 natural soil particle grading curve
Borehole survey data show that the overlying stratum in the test area is mainly Quaternary eluvial layer (QDL+EL) with a thickness of 10~30m ~ 30m and an average depth of 20m. It is a gravel soil layer, partially mixed with loam, with loose structure and slightly wet. The bedrock is the upper Permian Longtan Formation (P2l) coal measures stratum, which is composed of argillaceous siltstone, carbonaceous mudstone and silty mudstone. The experimental area is located in the mid-level mountain area, with simple hydrogeological conditions, mainly atmospheric precipitation, which is greatly influenced by seasons. The groundwater in the test area is mainly bedrock fissure water, which is buried deeply. During the survey, no groundwater was found in the borehole. The excavation depth of this test is 6m, and the sliding surfaces are all within 5m. Therefore, the test soil layers are all soil-rock mixtures above the groundwater level. See Figure 2 for the engineering geological profile of the test area.
Figure 2 Engineering Geological Profile
(1) the original grounding wire; ② Ground line after slope brushing (test area); ③ Lower limit of strong weathering zone;
Qdl+El- Quaternary eluvial layer; P2l—— Coal Measure Strata of Longtan Formation of Upper Permian
2 Arrangement and burial of instruments
The experimental area is 10m× 10m, and the slope ratio is 1∶2.5. The test area after embedding the instrument is shown in Figure 3. There are 9 boreholes in the test area, including 3 boreholes with inclinometer, 6 boreholes with pore water pressure gauge, and 3 boreholes with 12 pore water pressure gauge and inclinometer. Isolation belts with a width of 0.3m and a depth of 0.5m are excavated on the left and right sides of the test area. The left and right sides of the test area are isolated from the surrounding soil with iron sheets with a height of 1m to prevent rainwater from infiltrating into the surrounding soil. A collecting channel with a width of 0.5m and a depth of 1m was built in the lower part of the experimental area, and the possible sliding area was led out to connect with the collecting channel. Except one side near the slope, the other two sides of the water collecting tank are protected by cement to avoid rainwater loss. The water collecting tank is a square tank with a length, a width and a depth of 2m. In order to prevent rainwater from leaking, the water collecting basin needs to be protected with cement. Excavate a 5m×4m×2m reservoir in the upper right of the test area, which is built with bricks and protected with cement. Fig. 4 shows the layout of monitoring points, and fig. 5 shows the layout of measuring points in L 1 longitudinal section.
Figure 3 Test area after embedding the instrument.
Figure 4 Layout of Monitoring Points
The data unit is m.
Fig. 5 l 1 layout of measuring points in longitudinal section
The data unit is m.
2. 1 Slope crack monitoring
The slope cracks are measured by simple measurement method, and the main crack width of landslide is measured by steel tape measure during surface inspection.
2.2 inclinometer monitoring
The inclinometer consists of inclinometer, inclinometer and digital reader. When measuring, the inclinometer is inserted into the inclinometer tube, and the sliding value of the inclinometer tube, that is, the landslide body, is instantly reflected on the reader through the wire. In this experiment, the inclinometer is 100 made by American Sinco company, with a sensitivity of 8s, an accuracy of 6mm/30m and a measuring range of 0 ~ 53. The inclinometer adopts the high-precision ABS inclinometer produced by Lusheng Geotechnical Material Factory in Jintan City, with an outer diameter of 70mm, an inner diameter of 59mm and an outer diameter of 80mm, and each section is 2m long. Three inclinometers are buried in different positions of the slope at point I, as shown in Figure 3, with a buried depth of11m.
2.3 Pore water pressure monitoring
The pore water pressure sensor of soil is measured by KYJ-30 vibrating string pore water pressure gauge produced by Jintan Civil Engineering Instrument Factory, with the range of 0 ~ 200 kPa. KYJ-30 vibrating string pore water pressure gauge is suitable for drilling and installation, measuring the pore water pressure in the building and the temperature of the buried point. At the same time, it is equipped with ZXY-2 vibrating string frequency tester. The measurement range is: frequency f = 500 ~ 5000 Hz, frequency modulus display value f = F2× 10-3, measurement accuracy: 0.008 Hz, resolution: 0. 1 Hz, sensitivity: received signal ≥300μV, duration.
12 piezometers are buried in different positions of the slope * * *. In Figure 3, two piezometers are buried in each hole of L 1 and L3. L 1 pore water pressure measuring hole is 4m deep, and the buried depth of pore water pressure probe is 1m and 3m. L3 Pore water pressure measuring hole is 5m deep, and the buried depth of pore water pressure probe is 2m and 4m respectively.
2.4 rainfall intensity surface runoff monitoring
The total rainfall in the experimental area is recorded by the flowmeter on the main water supply pipe of the artificial rainfall simulator, and then the rainfall per unit time period is divided by the area of the experimental area1100m2, so that the rainfall intensity per unit time period can be obtained. Surface runoff is collected by the water collecting tank below the test area, and then recovered by the water pump to the reservoir above the test area. The surface runoff per unit time is measured by a flowmeter connected to a water pump.
3 artificial rainfall simulation
3. 1 Self-made artificial rainfall simulation device
Referring to SR field artificial rainfall simulator [9] developed by Institute of Water Conservancy and Soil Conservation of China Academy of Sciences, a special artificial rainfall simulator was made. The device consists of water pump, water meter, control valve, water pressure meter, nozzle, main pipe, branch pipe, two-way pipe, three-way pipe and four-way pipe. The main pipe and branch pipe are composed of short pipes with the length of 1m or 2m, and are assembled by two-way pipes, three-way pipes or four-way pipes. The rainfall intensity of 30 ~ 120 mm can be generated by adjusting the inlet pipe and the control valve on the return pipe. The coverage of the artificial rainfall simulator is 10m× 10m, and its schematic diagram is shown in Figure 6.
Fig. 6 Schematic diagram of artificial rainfall device
The data unit is m.
3.2 Monitoring period and frequency of artificial rainfall simulation test
After the buried instruments are coordinated and stable with the surrounding soil, measure the initial reading of each instrument. The starting and ending time of artificial rainfall simulation test is from April 25th to April 29th, 2005 10: 00. The hourly rain intensity is 60mm/h, and it stops every 2 hours1h. When the rain stops, record all monitoring readings. Record the pore water pressure, slope cracks, deep displacement, actual rainfall intensity and surface runoff of each measuring point every 3 hours. If it is observed that the slope will be unstable, increase the observation density appropriately.
4 Analysis of test results
4. 1 Slope crack monitoring
During the test, the displacement of the slope is not large. On April 30th, 2005 16: 30, a tensile microcrack was found at the rear edge of the slope, with a width of 1 ~ 2mm and a length of 3m.
4.2 inclinometer monitoring
The inclinometer data of each hole are sorted, analyzed and plotted, and the ZK3 hole is taken as an example to illustrate. Fig. 7 shows the variation of cumulative combined displacement of ZK3 in horizontal direction with hole depth. As can be seen from the figure, the displacement deformation zone basically occurs in the range of 0 ~ 2.5 m below the surface, and the displacement decreases with the increase of depth, and the slope deformation is the largest, and the maximum combined displacement reaches 7.67 mm
Fig. 7 Variation of horizontal displacement of ZK3 with hole depth
Fig. 8 shows the relationship between horizontal displacement of ZK3 characteristic point and accumulated rainfall intensity. As can be seen from the figure, the displacement of characteristic points increases gradually with the increase of accumulated rainfall intensity, and this deformation is a relaxation deformation that decreases gradually from slope to slope. The displacement at 0.5m is equivalent to twice the displacement at 1.5m, but there is almost no displacement at 4m, and the slight change of numerical value is caused by measurement error.
Fig. 8 Horizontal displacement and accumulated rainfall intensity of ZK3 characteristic point
Fig. 9 is the relationship curve between cumulative combined displacement and cumulative rainfall intensity at each measuring point. As can be seen from the figure, with the increase of cumulative rainfall intensity, the displacement of soil gradually increases, with the largest displacement in the middle of the slope, the second at the foot of the slope and the smallest at the top of the slope. The maximum combined displacements of ZK 1-ZK3 orifice are 3.36 mm, 10.37 mm and 7.67 mm respectively.
Fig. 9 Horizontal displacement and accumulated rainfall intensity of orifices at each measuring point.
4.3 Pore water pressure monitoring
Fig. 10 is a graph showing the variation of pore water pressure in R2 section with time, where B 1, B2, B3 and B4 respectively represent the pore water pressure when the buried depth of R2 section is 1m, 2m, 3m and 4m. As can be seen from the figure, at the initial stage of rainfall infiltration, the soil permeability is strong and the pore water pressure is low. With the progress of rainfall, pore water pressure increases sharply and reaches a stable value. It can also be found from the figure that the pore water pressures at 1m and 2m are close to 0, and the pore water pressures at 3m and 4m are 16.2kPa and 19.2kPa on average, which are equivalent to 1.65m and 1.96m water column pressures. The reason is that the rainfall intensity used in the test is large, and when the soil is saturated, the permeability decreases and the drainage is not smooth, forming a temporary stagnant water layer, which is about 4 m. This conclusion has also been verified by the results of inclinometer, and the sliding surface here is located 3.5m below the slope. The existence of stagnant water layer is extremely unfavorable to the stability of soil-rock mixture slope. Firstly, the formation of stagnant water layer leads to the increase of pore water pressure in soil and the decrease of effective stress, which leads to the decrease of shear strength of soil; Secondly, the formation of stagnant water layer makes the original unsaturated soil fully absorb water and soften, which also leads to the decrease of shear strength of soil. This double effect of rainfall infiltration may be one of the main reasons for the instability of soil-rock mixture slope induced by rainfall.
Figure 10 R2 Section Pore Water Pressure Variation Curve with Time.
Figure 1 1 shows the change of pore water pressure with time at the same depth (3m), and A3, B3 and C3 respectively represent the pore water pressure at the top, middle and bottom of the slope with 3m depth. As can be seen from the figure, the pore water pressure gradually increases from the top of the slope to the foot of the slope, and the pore water pressure at the foot of the slope is the largest, and the pore water pressure at the top of the slope is basically zero.
Fig. 1 1 curve of pore water pressure with time at the same depth (3m)
4.4 Rainfall intensity and surface runoff monitoring
The curve in figure 12 shows the relationship between the average hourly rainfall infiltration percentage and time during rainfall, which is calculated according to the measurement results of rainfall intensity and surface runoff. It can be seen that the average infiltration rate is 86% two hours before rainfall, and it gradually decreases with time after two hours due to the increase of surface runoff. After 6 hours, the infiltration rate dropped to a relatively stable value (50%), and after 6 hours of artificial simulated rainfall, half of the rainfall became surface runoff. The decrease of rainfall infiltration rate may be due to the water absorption saturation of slope soil and the closure of original open cracks.
4.5 Potential Sliding Surface Shape
The monitoring depth of the inclinometer is from the nozzle of the inclinometer to the inside of the slope 1 1m, and the monitored sliding surface depth is also the distance from the nozzle to the sliding surface, and the nozzle is also at a certain distance from the slope. The exposed part of the inclinometer should be subtracted from the actual sliding surface depth, and the sliding surface positions of ZK 1-ZK3 are 4.2m, 3.2m and 2.2m below the slope respectively. The position of the sliding surface can be determined by combining the position of the sliding surface monitored by the inclinometer with the staggered cracks at the front edge and the tensile cracks at the rear edge of the landslide. The position and shape of the sliding surface in L2 section are shown in Figure 13.
Figure 12 Average hourly rainfall (infiltration) and rainfall infiltration percentage
Figure 13 L2 Sliding Surface Shape
The data unit is m.
5 conclusion
A field monitoring system is equivalent to a full-scale experimental device, and its monitoring results are of great scientific and practical significance for studying and mastering the evolution law of landslide, the mechanism and behavior of disasters and the safety state of slopes. Through artificial rainfall simulation test and field comprehensive monitoring of soil-rock mixture slope, the following points are obtained:
(1) Under the influence of rainfall infiltration, the soil-rock mixture slope is mostly shallow relaxation failure, and the sliding deformation zone is within the range of 0 ~ 4m below the slope surface. The deformation is the largest on the slope and decreases gradually from the slope to the deep part of the slope.
(2) Two hours before rainfall, the average infiltration rate was 86%. After that, due to the increase of surface runoff, the infiltration rate gradually decreased with time. After a period of time (6h), the permeability decreased to a relatively stable value (50%). The decrease of rainfall infiltration rate is due to the water absorption saturation of slope soil, which closes the original open cracks.
(3) Under the action of heavy rainfall, the slope soil is saturated with water, the pores in the soil are partially closed, the permeability is reduced, the drainage is not smooth, and a temporary stagnant water layer is formed near the sliding surface. The existence of stagnant water layer is extremely unfavorable to the stability of soil-rock mixture slope. Firstly, the formation of stagnant water layer leads to the increase of pore water pressure in soil and the decrease of effective stress, which leads to the decrease of shear strength of soil; Secondly, the formation of stagnant water layer makes the original unsaturated soil fully absorb water and soften, which also leads to the decrease of shear strength of soil. This double effect of rainfall infiltration may be one of the main reasons for the instability of soil-rock mixture slope induced by rainfall.
(4) The soil-rock mixture slope in the experimental area received 4500mm rainfall in nearly 4 days and nights, which greatly exceeded the actual rainfall intensity, and the average infiltration rate reached 50%. However, the slope has only been deformed, but not collapsed, which shows that the failure condition of the accumulated slope is not only rainfall, but also related to the slope rate and geological conditions.
refer to
[1] You Xinhua. Stochastic structure model of soil-rock mixture and its application. Doctoral thesis of North Jiaotong University, 200 1
[2] Mid-week. Study on fluid-solid coupling characteristics and prediction of soil-rock mixture landslide. Doctoral thesis of Central South University, 2006.
Wu Jingkun, Fang Qi, Cai Jungang, et al. Expert system method for stability evaluation of landslide in accumulation layer Chinese journal of geological hazard and control,1994,5 (2): 8 ~16.
Luo, Li Hailing, Ge Xiurun, et al. Study on landslide disaster and landslide drainage effect under rainfall conditions. Geotechnical mechanics, 2000,21(3): 231~ 234
Zhan Liangtong, Bao, etc. In-situ monitoring of unsaturated expansive soil slope under rainfall infiltration. Geotechnical Mechanics, 2003,24 (2):151~158.
[6] Lumb P. Influence of rainstorm on slope stability. Sym on soil in Hong Kong. Hong kong, 1962, 73~87
[7] Finley PJ, Phil R, Maguire P.K. Relationship between landslide occurrence probability and rainfall. Can Geotech J .， 1997，34(6):8 1 1~824
Dai Fuchun, Li Fuchun. Frequency-volume relationship and prediction of rainfall-induced landslides. Engineering Geology, 200 1, 59(3/4):253~266.
Chen Wenliang, Tonkley. SR type field artificial rainfall simulator. Research on soil and water conservation, 2000, (12):106 ~110.