Nat Hazards (2015) 75:2545–2557 DOI 10.1007/s11069-014-1442-7 ORIGINAL PAPER
Influence of seasonal melt layer depth on the stability of surrounding rock in permafrost regions based on the measurement Shiwei Shen • Caichu Xia • Jihui Huang • Yan Li
Received: 4 April 2014 / Accepted: 14 September 2014 / Published online: 7 October 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract During tunnel construction in permafrost regions, control of the temperature field and depth of the seasonal melt layer of surrounding rock are the key factors that determine the stability of surrounding rock and the security of a structure. Based on the monitoring data on the temperature of the surrounding rock at two sections of Jiangluling Tunnel, the distribution regularities of the temperature field of the surrounding rock in different depths from the tunnel wall are obtained. These findings are used to determine the changing rules of the depth as well as the increasing speed of the depth of seasonal melt layer with time. At the same time, the influence of the surface of the accumulated temperature of surrounding rock to seasonal melt layer is analyzed, and the influence of the rule of the depth of seasonal melt layer on the stability of surrounding rock and arch crown settlement is determined. Results show that the temperature of the surrounding rock and the depth of the seasonal melt layer are controlled by engineering geological conditions of the surrounding rock and the temperature inside the tunnel. The temperature of the surrounding rock decreases with the increase in depth from the tunnel wall. Correspondingly, the depth of seasonal melt layer increases along with the temperature inside the tunnel. It is a positive correlation between seasonal melt layer and the surface of the accumulated temperature of surrounding rock, and with the increase in the depth of seasonal melt layer, the arch crown settlement increases, whereas the stability of the surrounding rock decreases. Keywords Tunnel Permafrost region Temperature monitoring Depth of seasonal melt layer Stability of surrounding rock S. Shen C. Xia (&) J. Huang Department of Geotechnical Engineering, Civil Engineering College, Tongji University, Shanghai 200092, China e-mail:
[email protected] S. Shen College of Construction Engineering, Jilin University, Changchun 130026, Jilin, China Y. Li Electric Power Design and Research Institute of Guangdong Province, Guangzhou 510663, Guangdong, China
123
2546
Nat Hazards (2015) 75:2545–2557
1 Introduction China is the third largest permafrost country after Russia and Canada. Permafrost takes up 22 % of the country’s national territory area. Recently, over 70 % of a constructed highway and railway tunnel engineering in the permafrost area was damaged by frostrelated phenomena such as icy road, cracked lining, and icy vault, which strongly affected the stability of surrounding rock and operation security. The implementation of the ‘‘Develop the West’’ strategy means that many new highways and railways will be constructed in Qinghai and Tibet. These engineering projects will improve the higher requirements of tunneling design and construction technology. One of the core problems in tunnel construction is the influence of seasonal melt layer on the stability of surrounding rock during the warm season. During tunnel construction in permafrost regions, engineering of geological conditions of surrounding rock as well as lining construction time change the depth of the seasonal melt layer and influence the stability of surrounding rock. Scholars have extensively studied the melt layer of surrounding rock in permafrost tunnels. Numerical solution methods on phase-change heat transfer temperature field have been proposed (Bonacina et al. 1973). Nonlinear finite element analysis has been used to study the phase-change heat transfer temperature field (Comini et al. 1974). The approximation formula (Shamsundar 1982), which was proposed to determine the freeze–thaw properties of soil around a circular cooling tube, was used to discuss the freeze–thaw properties of surrounding rock. Research has also been conducted on the changing rules of freeze–thaw circle of Kunlunshan Tunnel (Huang 2003), and the temperature distribution of the surrounding rock and the inside and outside of the thermal insulation layer has been examined. The depth of freeze–thaw circles under the action of oil temperature has been discussed based on the China–Russia oil pipeline engineering (Xu 2009), and has been modified under conditions of surface melting and increasing ground temperature. The disturbance laws of temperature field and the depth of freeze–thaw circles have also been studied under different conditions (Jia et al. 2005), such as exposure time of bare hole and the time of lining construction. In addition, an analytical solution of the tunnel temperature field in permafrost was developed in view of the lining and thermal barrier (Xia et al. 2010), and the depth of the thermal barrier after optimization was calculated. The re-freezing of the melt circle in Kunlunshan Tunnel was analyzed based on the differential equation of heat balance control (Zhang et al. 2002, 2003). The soil temperature field and melt circle of heat oil pipeline in northeast China were calculated and compared according to phase change in the governing differential equations of transient temperature field (He et al. 2008). The influence of melt circle depth on tunnel stability has been analyzed (Sheng 2011). Tunnel security measures in the construction process have been proposed. The thermal effects of rock freezing from excavation to completion of tunnels were discussed based on a field test (Zhang et al. 2007). This paper analyzes the temperature regularity of surrounding rock at different depths, the change in regularity of the depth, and the increased speed of seasonal melt layer over time. The analysis will be based on the temperature monitoring data of typical sections in a permafrost tunnel. The paper will then discuss the effects of the depth of seasonal melt layer on arch settlement and surrounding rock stability, which have important theoretical significance and application values on tunneling design and construction in permafrost regions.
123
Nat Hazards (2015) 75:2545–2557
2547
2 General situation of the tunnel Jiangluling Tunnel at the Gonghe to Yushu Highway in Xinghai County, Qinghai Province, is given as an example. The tunnel starts at YK329?680 and ends at YK332?525. The tunnel area has a typical continental semiarid climate, which is characterized by a long cold winter and many blizzards, a short summer, and plenty of rain. Its oxygen content is 40 % lower than that on sea level and has a large temperature difference between day and night. In this area, no absolute frost-free period occurs, and the frost period is 7 months in a year. The annual average temperature is -4.2 °C, and the extreme minimum temperature is -48.1 °C. The average annual precipitation is 369.2 mm, which occurs mostly from May to September. The average evaporation is 1,372 mm, and the maximum snow depth is 16 cm. The maximum freeze depth is 277 mm (Li 2013). The physiognomy of the tunnel area is alpine periglacial erosion, and the average altitude is over 4,280 m. The stratum of the tunnel area is fairly homogeneous. The main lithology is interbedded silty mud shale and slate, which contains a small amount of sand crystallization quartzite, limestone, and slate. Shale and slate have mud opaque structures and thin bedding structures, which have weak calcite cementation. The layers are too weak to crack and are softened upon contact with water. Many folds, squeeze faults, and cracks are present between the strata, which fracture the rock mass. The quaternary overburden spreads over the imports and exports of the tunnel, the surface of the mountain, the valley, and the piedmont diluvial fan of the slope. The major ingredients of the quaternary overburden are silty clay and gravel soil, which have different depths. The upper limit of permafrost is 1.2–2.2 m, and the lower level is 28–42 m. Figure 1 shows the profile diagram of the right tunnel of Jiangluling Tunnel.
3 Influence factors of the depth of seasonal melt layer In the construction of the tunnel in permafrost, the depth of seasonal melt layer is the key factor of the stability of surrounding rock, which is controlled by many factors.
classification of surrounding rock segmentation of tunnel
YK329+680
YK329+760 YK329+820
YK330+000 permafrost section
YK331+000 normal temperature section
YK332+525
floor level of permafrost The starting point of Jiangluling tunnel
4520 4500 4480 4460 4440 4420 4400 4380 4360 4340 4320 4300 4280 4260 4240 4220 4200
The ending point of Jiangluling tunnel
Altitude(m)
YK332+000 permafrost section
Fig. 1 Profile diagram of Jiangluling Tunnel
123
2548
Nat Hazards (2015) 75:2545–2557
3.1 Temperature inside the tunnel Because of the difference in atmospheric temperature in summer and winter, and the heat generated by construction, the permafrost will melt and lead to the changing of depth of seasonal melt layer. According to the experience of the tunnel construction in permafrost, the temperature inside the tunnel should be controlled in the range of ±5 °C, and the measures are as follows: 1. 2. 3.
Keep ventilation during the construction; Shorten the explosive time before the first lining; Reduce the using time of the cooling machine, such as internal combustion engine.
3.2 Hot melt disturbance in blasting process Due to the high sensibility to the changing in temperature, in the process of blasting construction in permafrost, the heat transfer and disturbance should be reduced, and the specific measures are as follows: 1. 2. 3.
Partition blasting should be used at peripheral holes, and the damping holes are set between adjacent hole; NaCl solution should be used to padding each blasting and damping hole; Extrapolation quantity should be severe designed in peripheral holes, in order to avoid the hot melt disturbance during secondary blasting.
3.3 Hydration heat of concrete The hydration heat of the sprayed concrete and moulded lining is a significant factor of the seasonal melt layer. Therefore, reducing the heat release during the hydration reaction is the effective method to reduce the depth of seasonal melt layer. The concrete measures are as follows: 1. 2. 3.
Reasonable mineral composition of clinker, such as C2S, should be used; The fineness of grinding of cement and grain composition of concrete should be adjusted; Mixed materials should be mixed moderately, such as coal ash and slag.
4 Monitoring design of the sections During tunnel construction in permafrost, the mechanical properties of surrounding rock are significantly different from the normal temperature condition because of freeze–thaw circles. During the construction process, temperature changes in the surrounding rock change the depth of seasonal melt layers. In the cold season, increased frozen depth is beneficial to the stability of surrounding rock. However, in the warm season, the tunnel lining is loaded by frost heave force. When seasonal melt layer is present, frost heave force may threaten the stability of the structure and the operation security of the tunnel. Therefore, during tunnel construction in permafrost, we shall monitor the temperature of surrounding rock and control the depth of seasonal melt layer to avoid freeze injury of the tunnel lining.
123
Nat Hazards (2015) 75:2545–2557
2549
Fig. 2 Arrangement of monitoring points on surrounding temperature. PS The unit of size marking in the figure is cm
sprayed concrete molded concrete insulating layer lining concrete surrounding rock 6×50
4×25
11 10 9 8 7 6 5 4 3 21 400
Fig. 3 Installation of the temperature sensor in the surrounding rock
According to the engineering geology investigation data, the floor level of permafrost in Jiangluling Tunnel is 28–42 m. The boundary of the permafrost and the nonfrozen earth is 50–70 m extend from the crosspoint of designed elevation of the tunnel and the floor level of permafrost. Based on it, we determine the temperature of the surrounding rock and monitor the temperature of surrounding rock at the YK329?760 and YK329?820 sections in the right tunnel of Jiangluling Tunnel. The arrangement of the monitoring points is shown in Fig. 2 (Li 2013). The half range of the tunnel is given as an example. Eleven temperature sensors are present at different depths in the surrounding rock. The distance between sensor nos. 1–5 is 25 cm and that between nos. 5–11 is 50 cm. The precision of the temperature sensors is 0.1 °C, and the cable casing, which is installed in the high-precision temperature sensors, is embedded into the 4-m-long PVC tube. Adhesive tape is used to seal the tube at both sides. At the embedding location, a u 60-mm drill is used to drill the 4.1-m-deep hole. Then, a 4-m-long steel tube is driven with the PVC tube, the end of which is sealed. Figure 3 shows the photographs of sensor installation in the field.
5 Temperature monitoring of surrounding rock With the YK329?760 and YK329?820 sections of the right tunnel of Jiangluling Tunnel used as examples, the monitoring data are analyzed. The YK329?760 section is above the
123
2550
Nat Hazards (2015) 75:2545–2557
lower limit of the permafrost, which belongs to the permafrost region. The YK329?820 section is below the lower limit of the permafrost, which belongs to the transition region between the normal temperature and the permafrost regions. To determine the adverse influences of temperature change in the surrounding rock on tunnel stability, we monitor the temperature of the surrounding rock during the warm season. 5.1 Monitoring data analysis of YK329?760 section The time range of monitoring at YK329?760 section is from June 8, 2012 to October 20, 2012. The curve of time and the temperature of the surrounding rock is shown in Fig. 4. Figure 4 indicates the following: 1. 2.
3.
4.
The temperature of the surrounding rock gradually decreases with increasing depth from the tunnel wall. In the range of 0–0.75 m from the wall, the temperature of the surrounding rock always declines. From the 36th to the 85th day, the temperature is relatively flat and then declines slightly. In the range of 1.0–2.0 m from the wall, a rising trend in temperature is observed on days 1–10, a spike occurs on days 5–10, and then slowly declines. The equilibrium temperature appears after 25–30 days and slowly declines after 85 days. At 2.5 m from the wall, the temperature is not time-varying. At 3.0 m from the wall, the temperatures at three monitoring points always stay below 0 °C, which shows that the surrounding rock in this range is frozen.
The monitoring location in which negative temperature occurs is considered the depth of the seasonal melt layer. Figure 5 shows the curve of time–seasonal melt layer at the YK329?760 section. In Fig. 5, we can see that the depth of seasonal melt layer shows the following pattern: rapid increase to slight increase to stable. After 85 days, the depth of the seasonal melt layer is stable at about 2.93 m. Based on the rule about the depth of the seasonal melt layer along with time, we can see that the seasonal melt layer has the same characteristics as the temperature of the surrounding rock, that is, the temperature of the surrounding rock and the depth of the seasonal melt layer are stable after 85 days. 16.00 14.00
at the wall
12.00
0.25m away from the wall 0.5m away from the wall
10.00
0.75m away from the wall
8.00
1.0m away from the wall
6.00
1.5m away from the wall
4.00
2.0m away from the wall 2.5m away from the wall
2.00
3.0m away from the wall
0.00 -2.00
3.5m away from the wall
0
20
40
60
80
100
120
140
4.0m away from the wall
Time (d)
Fig. 4 Temperature–time curve of the surrounding rock at different depths at the YK329?760 section
123
Nat Hazards (2015) 75:2545–2557
2551
3
0.07
Depth of seasonal melt layer (m)
0.05 0.04
2.6
0.03 2.4
0.02 0.01
2.2
Increase speed of the seasonal melt layer (m/d)
0.06
2.8
0.00 2
0
20
40
60
80
100
120
140
-0.01
Time (d) depth of seasonal melt layer increase speed of the seasonal melt layer Fig. 5 Depth/increasing speed of seasonal melt layer–time curve of the surrounding rock at the YK329?760 section
5.2 Monitoring data analysis of YK329?820 section The monitoring time range at the YK329?820 section is from May 24, 2012 to December 4, 2012. The curve of time and the temperature of the surrounding rock is shown in Fig. 6. In Fig. 6, we can see that at 0.5 m from the wall, the temperature of the surrounding rock declines with time. In the range of 0.75–1.50 m from the wall, the temperature increases for 15 days, stabilizes gradually, and declines slightly after 145 days. At 2.5 m from the wall, the temperature of the surrounding rock is stable, but the slight increase– decline process is still present.
at the wall
16.00
0.25m away from the wall
Temperature of surrounding
14.00
0.50m away from the wall
12.00
0.75m away from the wall
10.00 1.0m away from the wall
8.00 1.5m away from the wall
6.00 2.0m away from the wall
4.00 2.5m away from the wall
2.00
3.0m away from the wall
0.00 -2.00
3.5m away from the wall
0
40
80
120
Time (d)
160
200 4.0m away from the wall
Fig. 6 Temperature–time curve of the surrounding rock at different depths at the YK329?820 section
123
2552
Nat Hazards (2015) 75:2545–2557
Depth of seasonal melt layer (m)
0.3
5
0.2 4 0.1 3 0 2 1 0
-0.1 40
80
120
160
200
Increase speed of the seasonal melt layer (m/d)
0.4
6
-0.2
Time (d) depth of seasonal melt layer tendency of depth of seasonal melt layer increase speed of seasonal melt layer tendency of increase speed of seasonal melt layer
Fig. 7 Depth/increasing speed of seasonal melt layer–time curve of the surrounding rock at the YK329?820 section
The contrast between Figs. 4 and 6 shows that the temperatures of the surrounding rock at the YK329?760 and YK329?820 sections show the same overall trend. However, the remarkable difference is that the temperatures at the YK329?820 section are higher than those at the YK329?760 section. By comparing the monitoring data of the surrounding rock temperature from Figs. 4 and 6, it can be seen that the temperature range is -0.7–3.5 °C at YK329?760 section, and it is in the range of 1.0–4.3 °C at YK329?820 section, the difference in temperature at the two sections is about 0.8–1.7 °C. In addition, after 40 days, the temperatures are above 0 °C at each depth at the YK329?820 section, which indicates that the surrounding rock is not frozen 4 m from the wall. The monitoring location in which negative temperature occurs is considered the depth of the seasonal melt layer. Figure 7 illustrates the curve of time–seasonal melt layer at the YK329?820 section. In Fig. 7, we can see that the depth of seasonal melt layer displays sustainable growth over time at the YK329?820 section. The depth is over 4.0 m after 35 days with basic constant growth rate, which means that the depth of seasonal melt layer never shows a stable trend.
6 Influence of depth of seasonal melt layer on the stability of surrounding rock 6.1 Influence of surface of accumulated temperature on the depth of seasonal melt layer The surface of accumulated temperature of surrounding rock means the integration of the curve of surface temperature of surrounding rock and time in a certain time, the formula is as follows: Z t2 Tdt ð1Þ X¼ t1
where t1 and t2 are the time interval points of the accumulated temperature calculation; T is the surface temperature of surrounding rock, which is a function of time.
123
2553
depth of seasonal melt layer (m)
3
800 700
2.8
600 500
2.6
400 2.4
300 200
2.2
100 2
0
20
40
60
80
100
120
140
surface of accumulated temperature of surrounding
Nat Hazards (2015) 75:2545–2557
0
temperature of surrounding rock surface of accumulated temperature of surrounding rock
4.5
1200
4.0
1000
3.5
800
3.0 600 2.5 400
2.0
200
1.5 1.0
0
40
80
depth of seasonal melt layer
120
160
200
surface of accumulated temperature of
depth of seasonal melt layer (m)
Fig. 8 Influence of accumulated temperature of the surrounding rock surface to freeze–thaw circle depth at YK329?760 section
0
surface of accumulated temperature
Fig. 9 Influence of accumulated temperature of the surrounding rock surface to freeze–thaw circle depth at YK329?820 section
According to the data of surface temperature of surrounding rock, the influences of surface of accumulated temperature of surrounding rock on the depth of seasonal melt layer at typical sections are calculated by formula (1). The influence of YK329?760 and YK329?820 section is shown in Figs. 8 and 9. It can be seen that the depth of seasonal melt layer and the surface of the accumulated temperature of surrounding rock have a positive correlation. The increasing speed of the depth of seasonal melt layer will slow down, when the surface of accumulated temperature of surrounding rock has a slow increasing speed. Figure 8 shows that from the 15th day, the increasing speed of surface of the accumulated temperature is decreased, and the slope of the depth of seasonal melt layer–time curve is also decreased. After the 85th day, the surface of the accumulated temperature of surrounding rock and the depth of seasonal melt layer are tending toward stability. On the other hand, within 33rd day, the surface of the
123
2554
Nat Hazards (2015) 75:2545–2557
accumulated temperature of surrounding rock presents the linear growth characteristics, and at the same time, the depth of seasonal melt layer never shows a stable trend. 6.2 Influence of depth of seasonal melt layer on arch crown settlement 6.2.1 YK329?760 section The monitoring points are arranged at the arch crown and at both of its sides. The arch crown settlements were monitored starting on June 24, 2012. The curve is shown in Fig. 10. Figure 8 shows that although the data fluctuate slightly, the maximum of the settlement is 10.5 mm, and we cannot see an increasing trend in the monitored time range. This finding is due to the short monitoring time and the fact that the arch crown settlement is in the controlled range, which shows that the surrounding rock has high stability. 6.2.2 YK329?820 section The monitoring points are arranged at the arch crown and at both of its sides. The arch crown settlements were monitored starting on May 8, 2012. The curve is shown in Fig. 11. Figure 11 indicates the following: 1.
2. 3.
4.
The trends on the three monitoring points are similar, namely from largest to smallest, beginning from the left point, the arch crown, and the right point. The main reason behind this pattern is that the left side of the tunnel face has the highest degree of breakage and has weaker geological engineering conditions than the other areas. The monitoring data generally increase with time, and the arch crown settlement gradually increases. At the beginning of monitoring, the data fluctuated slightly. Within 15 days, the arch crown settlement rapidly increased. On the 15th day, the wall lining is constructed, and the surrounding rock is sealed off. Hence, the monitoring curve drops back down during the 15- to 18-day period. After 20 days, the arch crown settlement increased with time. From the point of final settlement, the minimum settlement in the three monitoring points is 95 mm and the maximum is 220 mm. Both results are beyond the allowed value range of the highway tunnel design specification. Thus, the tunnel exhibits large deformation.
Arch crown settlement (mm)
According to the conditions of the surrounding rock and construction, the main reasons for this finding are as follows: (1) the strength of the surrounding rock is reduced after 12 10 8 6 4 2 0 -2 0 -4 -6
left point Arch crown right point 2
4
6
8
10
12
14
Time (d)
Fig. 10 Arch crown settlement–time curve of the surrounding rock at the YK329?760 section
123
Nat Hazards (2015) 75:2545–2557
2555
Time (d)
Arch crown settlement (mm)
-50 0
50
100
150
0 50 100 150
left point Arch crown right point
200 250
Fig. 11 Arch crown settlement–time curve of the surrounding rock at the YK329?820 section
melting in the permafrost tunnel, and the development of the seasonal melt layer is uncontrolled during construction, which leads to the large depth of the seasonal melt layer, and (2) the time span between excavation and lining lasts over 1 month; therefore, the effective support structure system is invalid. 6.3 Influence of depth of seasonal melt layer on tunnel stability From the geological engineering conditions of the YK329?760 and YK329?820 sections, we can see that the tunnel is in the permafrost region, which is characterized by strong weathering silty mud shale with a small amount of slate, which has low strength and strong weathering. The basic quality index of surrounding rock is 190–230. Underground water in the tunnel consists mainly of drops and linear water. After excavating the tunnel, the surrounding rock is placed under the action of repeating freeze–thaw circles, which deform the wall and reduce the stability of the surrounding rock. At the same time, strong weathering of mud shale powder quartz vein is observed near the YK329?840 section, which fractures the surrounding rock and results in drops of water in rain form. Influenced by this process, the surrounding rock at the YK329?820 section is more tattered than it is at YK329?760. According to the engineering geological investigation of Jiangluling Tunnel, the basic quality indicators of the surrounding rock at YK329?760 and YK329?820 sections([BQ]) are 228 and 195. On the other hand, the YK329?820 section is below the lower limit of permafrost. Therefore, the temperature inside the tunnel at the YK329?820 section is higher than that at the YK329?760 section, which leads to lower stability of permafrost. In fact, with the influence of the depth of the seasonal melt layer, the arch crown settlement of YK329?820 section shows large deformation characteristics and is still not convergent after 4 months. To maintain the stability of the surrounding rock, secondary blasting is performed to address the large deformation.
7 Conclusion Based on the measured temperature of surrounding rock in Jiangluling Tunnel, the influence of the depth of seasonal melt layer on the stability of surrounding rock and arch crown settlement is analyzed. The conclusions are as follows:
123
2556
1.
2.
3.
4.
5.
6.
Nat Hazards (2015) 75:2545–2557
The temperature of surrounding rock and the depth of seasonal melt layer are controlled by geological engineering conditions of surrounding rock and air temperature in the tunnel. For instance, in the YK329?820 section, although it is farther from the entrance than the YK329?760 section, a broken quartz vein is present nearby, which degrades the rock mass and hydrogeological conditions. At the same time, the temperature of surrounding rock at the YK329?820 section increases remarkably because of the arrangement of heating construction. From the temperature field distribution rule of surrounding rock at different depths, the temperature decreases as a whole along with increasing depth from the tunnel wall. When the rock is 3 m from the wall, the temperature of surrounding rock is always frozen with the increasing depth. However, the temperature displays a slight increase– decrease trend with high air temperature in the tunnel. After 40 days, the temperatures of surrounding rock at different depths are all above 0 °C, which indicates that the surrounding rock is not frozen. A comparison of the changing rule of the depth of seasonal melt layer with time at the two sections indicates that the depth of seasonal melt layer at the YK329?760 section finally stabilizes at 2.93 m. The depth of seasonal melt layer at the YK329?820 section shows sustainable growth and is greater than 4 m after 35 days. At the same time, the depth has a constant growth rate, which shows that the depth of the seasonal melt layer will not stabilize. According to the analysis of the field monitoring data in Jiangluling Tunnel, before the construction of the first moulded lining, the depth of seasonal melt layer should be controlled within 3.0 m, in order to ensure the stability of the permafrost. From the relationship between the depth of seasonal melt layer and the surface of the accumulated temperature of surrounding rock, they have a positive correlation. The increasing speed of the depth of seasonal melt layer will slow down, when the surface of accumulated temperature of surrounding rock has a slow increasing speed. Along with the increased depth of the seasonal melt layer, arch crown settlement increases with time. Therefore, to avoid large deformation due to the excessively large depth of seasonal melt layer, the lining should be constructed as soon as possible to ensure rapid formation of an effective closed mechanical system.
Acknowledgments This work was supported by the National Science-Technology Support Program of China (No. 2014BAG05B05), the Western Traffic Science and Technology Projects (20113184901070), and China Postdoctoral Science Foundation Project (2014M551452).
References Bonacina C, Comini G, Fasano A et al (1973) Numerical solution of phase change problems. Int J Heat Mass Transf 16(6):1852–1882 Comini G, Guidice SD, Lewis RW et al (1974) Finite element solution of nonlinear heat conduction problems with special reference to phase change. Int J Numer Methods Eng 8(6):613–624 He S, Jin H, Yu W et al (2008) Numerical analysis of the thaw circle characteristics of soil about underground heated pipeline in permafrost region. Oil Gas Storage Transp 27(1):21–25 Huang S (2003) Study on the active ring of permafrost in the Kunlunshan tunnel on the tunnelling. J Glaciol Geocryol 25(Supp. 1):100–103 Jia X, Li W, Zhu Y (2005) Analysis of stability for surrounding rock caused by depth of thawing range during plateau permafrost tunnel construction. Chin J Rock Mech Eng 24(supp. 2):5693–5697 Li Y (2013) Study on the technique of controlling surrounding rock thawing depth and the law of tunnel lining deterioration. Shanghai: Tongji University, Master Thesis
123
Nat Hazards (2015) 75:2545–2557
2557
Shamsundar N (1982) Formulae for freezing outside a circular tube with axial variation of coolant temperature. Int J Heat Mass Transf 25(10):1614–1616 Sheng Z (2011) Analysis on thawing cylinder depth’s influence on the tunnel stability with several-year Frozen earth. Railw Constr Technol 2:35–38 Xia C, Zhang G, Xiao S (2010) Analytical solution to temperature fields of tunnel in cold region considering Lining and insulation layer. Chin J Rock Mech Eng 29(9):1767–1773 Xu B, Ding J, Mi W et al (2009) Research on calculation method of buried pipe foundation melting laps in permafrost of Sino–Russian oil pipeline. Recent development of research on permafrost engineering and cold region environment, pp 208–214 Zhang X, Lai Y, Yang F et al (2002) Numerical analysis on freezing–thawing situation of tunnel’s surrounding rock in cold regions. J China Railw Soc 24(4):92–96 Zhang X, Su X, Lai Y et al (2003) Analyzing the effect of tunneling on temperature of permafrost in the Kunlun mountains, Tibetan Plateau. J Glaciol Geocryol 25(6):621–627 Zhang D, Wang M, Ren S (2007) Experimental study on temperature and response of seasonal active layer of tunnel’s surrounding rock in permafrost region on Qinghai–Tibet Plateau. Chin J Rock Mech Eng 26(3):614–619
123
