- Email: [email protected]
Analysis on thermal stability and deformation stability of CMR
Cold Regions Science and Technology 110 (2015) 215–222
Contents lists available at ScienceDirect
Cold Regions Science and Technology journal homepage: www.elsevier.com/locate/coldregions
Analysis on thermal stability and deformation stability of CMR Yaling Chou a,b,⁎, Yu Sheng c, Ji Chen c, Xingqiang Chen a,b a b c
Key Laboratory of Disaster Prevention and Mitigation in Civil Engineering of Gansu Province, Lanzhou University of Technology, Lanzhou, Gansu 730000, China Northwest Center for Disaster Mitigation in Civil Engineering of Ministry of Education, Lanzhou, Gansu 730000, China State Key Laboratory of Natural Gansu, Lanzhou Institute of Geology, Chinese Academy of Sciences, Lanzhou, Gansu 730000, China
a r t i c l e
i n f o
Article history: Received 15 February 2014 Received in revised form 25 August 2014 Accepted 23 October 2014 Available online 29 October 2014 Keywords: Chai–Mu Railway Engineering measures Shady–sunny slope Permafrost embankment Thermal stability Deformation stability
a b s t r a c t In permafrost regions, embankment stability has been destroyed with longitudinal cracks primarily caused by transverse asymmetric settlement. And shady–sunny slope effect is a principal factor which might result in embankment asymmetric settlement. Based on field data of some typical sections along the Chai–Mu Railway (simply called CMR) with obvious shady–sunny slopes, which is located on the Qinghai–Tibet Plateau, the ground temperature characteristics and settlement deformation properties of the embankment have been analyzed. This paper has studied how the different engineering measures, including rubble ventilation embankment, thermal pipe revetment embankment, rubble and thermal pipe revetment joint embankment, did influence the shady– sunny slope effect. The results show that: (1) Both thermal pipe revetment embankment and rubble ventilation embankment have perfect effect on cooling the roadbed in permafrost regions. However, for thermal pipe revetment embankment, the sunny–shady phenomenon and asymmetric settlement of embankment have been avoided if thermal pipe has been strengthened in sunny side. For rubble ventilation embankment, the sunny–shady effect has been hardly eliminated even though the rubble has been strengthened in sunny side. Although some composite embankments, such as the rubble and thermal pipe revetment joint embankment, have been successfully applied in the Qinghai–Tibet Railway, it is difficult to completely eliminate the sunny– shady slope problem along the CMR because of the complicated engineering geological conditions and the bad permafrost environments. (2) Along the CMR, in all common embankments — the observed results show that the settlement deformation is within 30 cm that is within the settlement standards of the National II Class Railway. Even the settlement deformation of thermal pipe subgrade and rubble ventilation embankment was able to meet the National I Class Railway standards — 20 cm. (3) On the basis of the above, some corresponding countermeasures and advice are proposed. First, rational use of the synthetic application of engineering measures must be taken. For example, let the structural mechanics measures and temperature controlled measures unify altogether to prevent longitudinal cracks. Second, the specific work conditions and details such as engineering geological condition, permafrost environments as well as other local factors including the slope orientation have to be taken into account in order to put forward rational engineering measures and accurate design parameters. Finally, it is essential to protect and to comprehensively evaluate the permafrost environment, which is the ultimate basis of permafrost engineering. © 2014 Elsevier B.V. All rights reserved.
1. Introduction In permafrost regions, construction of embankments has changed the thermo-physical features of natural ground surface and atmosphere, influenced thermodynamic and dynamic stability of frozen soil layers, and complicated temperature fields of frozen soil embankment (Ma et al., 2005; Qi et al., 2007; Sun et al., 2004; Yu et al., 2007). When the two embankment slopes are south-facing and north-facing, respectively, which is called shady–sunny slope in Chinese, the difference of thermal boundary conditions such as solar radiation and surface turbulent flow will induce transverse asymmetry of temperature field, ⁎ Corresponding author. Tel.: +86 931 4967297; fax: +86 931 4967271. E-mail addresses: [email protected], [email protected] (Y. Chou).
http://dx.doi.org/10.1016/j.coldregions.2014.10.004 0165-232X/© 2014 Elsevier B.V. All rights reserved.
and this will lead to a multitude of problems in permafrost roadbeds. And there extensively exists these kinds of phenomena in both warm and low temperature permafrost regions (Liu et al., 2002). The thermal difference of the shady–sunny slope results in there is an unbalanced freeze–thaw condition between both sides of the embankment, which leads, inevitably, to transverse differential settlement. As time goes by, the differential settlement will be aggravated, and finally the embankment stability is destroyed with longitudinal cracks caused by asymmetric settlement of permafrost (Wang and Dou, 2004; Wang et al., 2006). Based on observed field temperature data in Beiluhe on the Qinghai–Tibet Plateau, Sheng et al. (2005) analyzed the thermal difference between the shady and sunny slopes, and pointed out that the thermal difference can lead to development of embankment longitudinal cracks. Lai et al. (2004) recommended that the embankment
216
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
disease that results in uneven settlement with longitudinal cracks can be eliminated by paving ripped-rocks of different thickness on the sunny and shady slopes, respectively. Cheng (2003) showed that some local factors including slope orientation can change permafrost distribution and may cause some engineering damages such as longitudinal cracks. The investigation was done by analyzing both observational geo-temperature and deformation data of the embankment in experimental section K369+100 along the National Highway 214 on the Qinghai–Tibet Plateau. Also, the fundamental cause of the embankment transverse asymmetrical thermal regime and asymmetrical deformation was discussed (Chou et al., 2009). The longitudinal cracks are a very common permafrost embankment disease, which has attracted the attention of more and more researchers (Mao et al., 2010; Pei et al., 2006; Wu et al., 2002; Xu, 2010; Zhang et al., 2003). According to field observation, the longitudinal cracks are distributed mainly over the sunny side of high embankments in unstable permafrost regions. Moreover, the longitudinal cracks developing along the sunny side covers more than 70%. And the longitudinal cracks appear not only in common embankments, but also in reinforced embankments, such as thermal pipe embankment, ripped-rock revetment embankment, insulation embankment and so on (Su et al., 2004; Wang et al., 2003). Through field investigation, Wu et al.(2005) pointed that a better thermal stability situation will be developed under conditions where the temperature near the permafrost table beneath the rubble embankment has largely dropped in cold permafrost regions. But in warm permafrost regions, even the permafrost table has highly raised the annual thermal imbalance of permafrost and temperature will rise, and it is disadvantageous for the thermal stability of permafrost. In the warm permafrost region, only the embankment shady side can be better cooled by adopted rubble layer but in the cold permafrost region, the rubble ventilation embankment not only has cooled the whole subgrade, but also has played a certain role in restraining the sunny–shady effect (Chen et al., 2011). Liu and Li (2008) predicted the temperature field
of embankment with crushed rock protection in the future 50 years by applying the transient thermodynamics FEM method. The computation results show that the temperature difference between the sunny slope and the shady slope has existed at all times, but the difference on the same days and months has varied little with year. Therefore, how the different engineering measures have influenced the shady– sunny slope effect is an important subject worth studying. The Chaidaer Muli Railway, simply called the CMR, is the first local railway with a total length of 142 km in northeastern Qinghai Province, China. It starts in Chaidaer County (100°25′55″E, 37°35′40″N) and ends in Muli Country (99°11′00″E, 38°08′38″N). Fig. 1 gives the geographical location of the CMR. Topographically, it passes through the Datong River valley, the Datong Mountains and the front edge of the Tuolai Mountain in the central-eastern Qilian Mountains. The railway broadly runs from southeast to northwest. The elevations of this region range from 3600 m to 4100 m and the average annual air temperatures vary from about − 2.4 °C to − 5.8 °C. According to the meteorological data, the minimum air temperature is −40 °C and the maximum air temperature is 17 °C, respectively. There is rich rain and snow, and swamps as well as marshes are both developed. The annual precipitation along the CMR is approximately 500 mm·a−1. The vegetative coverage is more than 60% in most of the permafrost regions, and the wetlands are widely distributed. About 77 km of all the CMR runs across wetlands. The permafrost is discontinuous and unstable with average annual ground temperature ranging from − 1 °C to 0 °C. The ice contents in most permafrost regions are also very high. During the construction of CMR, based on the experience of the Qinghai–Tibet Railway, a number of engineering measures were taken to alleviate the sunny–shady effect. This paper will study the ground temperature characteristics and the settlement deformation properties of the embankment through analyzing the observation data of some typical embankment sections along the CMR. Also, that the positive cooling engineering measures, including the rubble ventilation embankment, the thermal pipe revetment embankment, the rubble
Fig. 1. The geographical location of the CMR.
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
and thermal pipe revetment joint embankment, make an appreciable difference to the shady–sunny slope effect is mainly discussed. This work will provide a scientific basis for embankment construction and maintenance in permafrost regions on the Qinghai–Tibet Plateau. 2. Monitoring content Engineering geological conditions, permafrost characteristic, engineering measures, embankment strike and height, and the embankment section shape of the CMR have been investigated. The temperature monitoring probes are laid out within some typical sections. In order to observe the engineering effect of different measures, different ground temperature zones and different monitoring sites have been given a comprehensive consideration when the monitoring sections were laid out. The construction of monitoring sections began on October, 2007. And until July, 2008, the main monitoring equipments and apparatus have completely finished. By early 2010, the whole observation time has been not enough up to 2 annual cycles. As far as the effect of different engineering measures is concerned, the classification of measures is discussed, respectively. Meanwhile, the engineering measures and the effects are deeply analyzed. The principal strike of CMR runs from southeast to northwest, furthermore there are more sections close to run from east to west. Since the effect of the shady–sunny slope is obvious, we select some typical sections to comparatively analyze how the different engineering measures have inhibited the shady–sunny slope phenomenon. In this paper, temperature of slope means the temperature at shallow ground depth (0.5 m) under the sunny slope and the shady slope, respectively. And we refer to the settlement deformation of the sunny shoulder and the shady shoulder as settlement displacement of two slopes, respectively. The data acquisition apparatus is composed of thermal-susceptible resistance sensors and Fluke IV multimeters with a 0.1 Ω precision. The space between two probes is ranging from 0.5 m to 2.0 m. The frequency of observation is once every 15 days. The embankment heights range from 2.0 m to 5.3 m and the fill is fine-grained and gravelly soil. The road surface width is between 6.7 m and 7.9 m, and the slope gradient is 1:1.5. All the monitoring sections are on the slope wetlands, and some sections are asymmetric in transverse profile. The sections DK39+800, DK40+000, DK74+000, DK74+500, DK75+000, DK94+340, DK94+660, DK94+900, DK99+100, DK99+200, DK99+355, DK114+730, DK114+800, DK123+150 and DK123+250 are all laid out with temperature probes. In order to save space, only a
217
sketch of observation of the transverse section and temperature holes of section DK114+730 (Fig. 2) is given here, which is the rubble and thermal pipe revetment joint embankment. The temperature probe positions of other sections are similar to section DK114+730. The length of rubble ventilation embankments and rubble berm embankments reaches 53 km. The rubbles in the ventilation embankment are obtained near the CMR, which are made up mainly of moderately weathered schist and gneiss interlaid with a little pebble. The particle size of rubbles is mainly from 5 cm to 30 cm. 3. Monitoring results 3.1. The rubble ventilation embankment 3.1.1. The temperature difference For the south to north strike, the solar radiations that the two embankment slopes have absorbed are essentially the same. So the permafrost temperature field under embankment is substantially symmetric, and there will be no serious disease because the shady– sunny slope problem is not obvious. But the monitoring sections of the rubble ventilation embankment, such as section DK40+000, section DK74+000, section DK94+900, section DK123+150 and section DK123+250, their strikes are from 260° to 345°. In following description for convenience, the south to north strike is defined as 0°, which is the exact thermal symmetric embankment. And the east to west strike is called 90°, which means the most remarkable shady–sunny slope effect. According to the rule, the strikes of the 5 sections (DK40+000, DK74+000, DK94+900, DK123+150 and DK123+250) are 40°, 60°, 15°, 80° and 80°, respectively. The annual mean ground temperatures of the 5 sections are −0.65 °C, −0.55 °C, −1.31 °C, − 1.18 °C and −1.10 °C, respectively. The shallow ground temperatures of section DK74+000, section DK94+900 and section DK123+150 at a depth of 0.5 m under slope surface are shown in Fig. 3. For section DK74+000 and section DK123+150, we find two obvious phenomena: 1) the shallow ground temperatures (0.5 m) on shady slopes are always lower than those of the sunny slopes; 2) between the two slopes, the maximum temperature difference appears in winter and there has been a little temperature difference in summer. In all the sections, the shady–sunny slope effect of section DK94+900 is the weakest, and the temperature curve crosses each other so that it is difficult to discriminate the temperature difference between two slopes only through Fig. 3. The results of the data analysis indicate that the mean annual ground temperature of section DK94+900, section DK74+000 and
Fig. 2. Sketch of observation transverse section and temperature holes of section DK114+730 (unit: m).
218
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
than that of the shady foot by 0.47 m. And in section DK123+150, the permafrost table of the sunny foot is deeper than that of the shady foot by 1.92 m. In 2009–2010, there is little permafrost table difference between the sunny shoulder and shady shoulder in section DK94+900. However, the permafrost table difference of section DK123+150 between the sunny shoulder and shady shoulder reaches 0.4 m. The strike difference of the rubble ventilation embankment shows that there is still a sunny–shady slope problem if the sunny side has not adopted special measures to eliminate the sunny–shady slope phenomenon. Therefore, how the rubble ventilation embankment completely gets rid of the sunny–shady slope problem will need to be studied further. For instance, the sunny–shady slope phenomenon may be eliminated or weakened by building a much thicker rubble or laying thermal pipe on the sunny side. According to the specific work condition, the rational and accurate design parameters should be put forward.
(a) Section DK74+000
(b) Section DK94+900
(c) Section DK123+150 Fig. 3. Shallow ground temperatures of 0.5 m under slope surface in rubble embankment sections.
section DK123+150 at a depth of 0.5 m on the sunny slope is higher than that on the shady slope by over 2.23 °C, 4.69 °C and 7.13 °C, respectively. The perturbations that the sunny–shady slope affects on the permafrost table are the most fundamental performance of the thermal asymmetric embankment, which might destroy the embankment stability with longitudinal cracks. Table 1 shows the artificial permafrost table of different thermometer holes in the rubble embankment sections. Under conditions without any other external factors, the temperature difference has increased with the embankment strike increasing, and the difference of the permafrost table between the sunny side and shady side has increased accordingly. In 2008–2009, in section DK94+900 the permafrost table of the sunny foot is higher
3.1.2. The settlement deformation difference The embankment settlement deformation is related to many factors, such as engineering geological condition, corresponding measures, construction quality and so on. By analyzing the settlement deformation, we describe whether the engineering geology investigation and the design parameters are accurate or not, as well as whether the embankment construction is qualified. Moreover, for embankments in permafrost regions, the settlement deformation depends largely upon the physical and mechanical properties of the underlying permafrost. Generally speaking, permafrost degradation inevitably results in the settlement deformation increasing. So the key of embankment treatment is to control the settlement of the bearing stratum. Also, analysis on the settlement deformation can also help us study the cooling effect of the rubble ventilation embankment further. The following will discuss the settlement deformation of the rubble embankment settlement in high-temperature permafrost regions and in low-temperature permafrost regions, respectively. Section DK39+800 and section DK40+000 are both in hightemperature permafrost regions, and the annual average air temperature here is between − 2.1 °C to − 2.9 °C. Fig. 4 gives the settlement deformation curves of section DK39+800 and section DK40+000. Section DK39+800 is an ordinary embankment without any measures (as the contrast section). The curves have features such as:(1) the settlement deformation of the sunny shoulder is more than that of shady shoulder, and this rule is in agreement with the abovementioned temperature difference; (2) the settlement deformation of the rubble embankment section is greater than that of the contrast section; and (3) there is a slight frost heave in the freezing season. Since the sunny slope has a high temperature, the permafrost table under the sunny side is much deeper than that under the shady side, and a greater settlement deformation has occurred in the sunny side. However, feature (2) does not agree with the temperature field. The ground temperature of section DK40+000 is lower than that of section DK39+800, but the settlement deformation of section DK40+000 is greater than that of section DK39+800. According to the engineering geological conditions, there are both thick layers of humus in the ground under the two sections. There is humus with a thickness of 2 m under section DK39+800 and humus with a thickness of 3 m under section DK40+000, respectively. Therefore, the rubble embankment section has
Table 1 The drilling table in rubble embankment sections (unit: m). Section
DK40+000 DK74+000 DK94+900 DK123+150 DK123+250
2008–2009
2009–2010
Sunny foot
Shady foot
Sunny shoulder
Shady shoulder
Sunny foot
Shady foot
Sunny shoulder
Shady shoulder
−2.23 −3.82 1.72 2.90 1.64
– −1.45 2.19 0.98 1.72
– – 1.28 1.54 1.43
−1.26 – 1.50 1.35 0.77
−2.20 −3.89 – 2.80 2.15
– −1.37 1.89 0.98 1.32
– – 1.52 1.55 1.53
−0.94 – 1.51 1.15 0.77
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
219
3.2. Thermal pipe revetment embankment
Fig. 4. Settlement deformation changing with time in rubble embankment sections.
poor geological condition that results in the settlement deformation of section DK40+000 which is greater than that of section DK39+800. From engineering measures, 2 m of soft soil in section DK39+800 has been dug. Meanwhile, the rubble with a thickness of 1.2 m has been filled on the original earth surface in section DK40+000, and the soft soil has not been dug. The soft soil is located near the permafrost table, which is more vulnerable to thaw under the engineering disturbances, so there is a big settlement after construction in section DK40+000. The digging of soft soil much reduces the settlement deformation, so the settlement deformation of section DK39+800 is relatively small. As section DK40+000 is concerned, if the rubble can lift the permafrost table to the ground surface, the compression deformation of the seasonal active layer is avoided. But in fact, the permafrost table is −0.94 m below the natural ground. That is to say, a large deformation of section DK40+000 takes place as a result of seasonal active layer thawing. In low-temperature permafrost regions, the settlement deformation curves are analogous to those in high-temperature permafrost regions. Section DK123+150 and section DK123+250 are both in lowtemperature permafrost regions, and the annual average air temperature here is about − 3.5 °C. Fig. 5 gives the settlement deformation curves of section DK123+150 and section DK123+250 (the contrast section). There is a larger settlement deformation in the sunny side and smaller settlement deformation in the shady side. Comparing Figs. 4 and 5, it is not difficult to find that the settlement deformation in hightemperature permafrost regions is much larger. Through analyzing all the sections, by the beginning of 2010 the amount of deformation has been within 10 cm–20 cm in high-temperature permafrost regions, but the amount of deformation in low-temperature permafrost regions has reached 2 cm–8 cm. Both in high-temperature permafrost regions and in low-temperature permafrost regions, the amounts of settlement deformation in rubble embankment sections were able to meet the National I Class Railway standards — 20 cm.
3.2.1. The temperature difference As a highly-effective heat conduction set, thermal pipes are commonly used to enhance thermal stability of building foundation in permafrost regions. In order to eliminate the sunny–shady slope problem, the thermal pipes have stood in two ranks on sunny side, but there is a row of pipes on the shady side. The longitudinal spacing of the thermal pipe is 3 m. On the sunny side, one row of thermal pipes is situated at the embankment foot, and another row of thermal pipes lie beside the revetment foot. There is 3 m between the two rows of thermal pipes. On the shady side, there is one row of thermal pipes beside the revetment shoulder. Section DK74+500, section DK99+100 and section 99+200 are thermal pipe revetment embankments, whose strikes are 70°, 35° and 35°, respectively. The annual mean ground temperatures of the three sections are − 0.32 °C, − 0.89 °C and − 0.99 °C, respectively. Fig. 6 shows shallow ground
(a) Section DK74+500
(b) Section DK99+100
(c) Section DK99+200
(d) Section DK99+355 Fig. 5. Settlement deformation changing with time in rubble embankment sections.
Fig. 6. Ground temperatures under slope surface changing in thermal pipe embankment sections.
220
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
temperatures of slope in thermal pipe revetment embankments. Section DK99+355 is a common landfill embankment, as a contrast section, whose strike is 35° and the sunny–shady slope effect is weaker. We can see: (1) in the thermal pipe revetment sections, the temperature of sunny side is lower than that of the shady side during most time of the year. But in contrast section DK99+355, the temperature of the sunny side is higher than that of the shady side during most time of the year. And the occurrence of this phenomenon is apparently related to the cooling effect of the thermal pipe; (2) the annual temperature amplitude of the thermal pipe section is greater than that of the contrast section. The annual temperature amplitude of the thermal pipe section is 40 °C, but the annual temperature amplitude of the contrast section is not beyond 20 °C. Judging from the processes of temperature curves, the temperature amplitude over a short period in the contrast section is relatively small, and the temperature follows a relatively smooth curve. Meanwhile, the temperature amplitude over a short period in the thermal pipe section is relatively larger, and the temperature curve occurs in jumps. The temperature amplitude difference between the two types of section is not in evident in the thermal pipe, which is related to covering conditions of the slopes. There is generally rubble masonry pavement in the thermal pipe section. The covering layer of the revetment is thin, and whose thermal conductivity is high, so the temperature wave is easy to transfer in the deep ground. However, the contrast section DK99+355 has been adopted by grassed slope and the covering layer is much thicker. The moisture content of grassed covering layer is high, which has seriously hindered the temperature wave from propagating. Consequently, there is an extreme difference in temperature amplitude between the two types of sections. 3.2.2. The settlement deformation difference Based on the above analysis, the thermal pipe revetment embankment can effectively cool the permafrost embankment and can weaken or remove the sunny–shady slope problem. Moreover, theoretically the thermal pipe is normally effective in decreasing the permafrost embankment deformation. Fig. 7 gives settlement deformation changing with time in section DK99+100 and section DK99+200, and the settlement characteristics are:(1) there is a certain fluctuation in the settlement curve, which is sinking in the thaw season and is a slight frost heave in the freezing season;(2) the thermal pipe revetment is a comparatively good method of inhibiting settlement deformation and asymmetric settlement. In summary, the settlement deformations of the two sections are both within 10 cm. Even the settlement deformation of section DK99+200 is less than 3 cm. Also, the embankment settlement deformation stops increasing in the later stages and the embankment has stabilized. It was obvious that the amounts of settlement deformation were able to meet the National I Class Railway standards — 20 cm.
(a) Section DK 99+100
(b) Section DK99+200 Fig. 7. Settlement deformation changing with time in thermal pipe embankment sections.
higher than that of the shady slope (Fig. 8). The temperature curves of two slopes present that the rubble and thermal pipe revetment joint embankment have not completely eliminated the sunny–shady slope problem, which is chiefly due to many factors. Firstly, the embankment is higher so that it needs more filler, which results in much bigger thermal disturbance on permafrost. Also, the higher the embankment, the bigger the slope area, the more solar radiation is absorbed by the
3.3. The rubble and thermal pipe revetment joint embankment There is a small amount of application of the rubble and thermal pipe revetment joint embankments in CMR, which are mainly used for sections in ice-rich permafrost regions where the engineering geological condition are very poor. 3.3.1. The temperature difference The embankment strikes of sections DK114+730 and DK114+800 are both 65°, whose sunny–shady slope problem is very serious. The annual mean ground temperatures of the two sections are − 1.36 °C and −1.28 °C, respectively. During the first design, the rubble ventilation embankment has been only used. Afterwards there happens sideslip in the sections, so the thermal pipe is strengthened in the rubble ventilation embankment. Here, the contrast section is not laid out. Due to the influence of engineering, the temperature curves of the sunny slope and the shady slope intersect each other at an early stage. But after the early stage, the temperature of the sunny slope is always
(a) Section DK114+730
(b) Section DK114+800 Fig. 8. Ground temperatures under slope surface in rubble with thermal pipe embankment sections.
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
sunny slope. Secondly, the thermal insulation of rubble ventilation has hindered the cooling effect of the thermal pipe. Thirdly, the local geography, geological environments and climate conditions resulting in the cooling effect of the joint embankment are not ideal. From these analyses, we realized that the main way of heat disturbance on permafrost has not been made clear, even these powerful measures that have been proved effective in the Qinghai–Tibet Railway (another railway on the Qinghai–Tibet Plateau), such as the rubble and thermal pipe revetment joint embankment, cannot completely eliminate the sunny–shady slope effect in CMR. The problem acts as a strong warning to permafrost embankment design, and so many factors have to be taken into account. The rational use of the synthetic application of engineering measures must be taken. For example, let structural mechanics measures and temperature controlled measures unify altogether to prevent longitudinal cracks. 3.3.2. The settlement deformation difference Fig. 9 gives settlement deformation changing with time in section DK114+730 and section DK114+780, whose distribution is consistent with the shallow ground temperature. The settlement characteristic is: (1) the settlement deformations of the two sections are both small, which is within 5 cm; (2) the settlement deformations are smooth and steady, and there is little fluctuation in the settlement curve, which means that there is sinking in the thaw season and a slight frost heave in the freezing season; and (3) the settlement deformation of the sunny side is greater than that of the shady side. Although settlement deformation difference between two shoulders is very minute, this contrast relationship agrees with the temperature. The embankment settlement deformation is little increasing with time and the embankment has slowly stabilized. 4. Analysis and discussion There are many monitoring sections along CMR, and these sections differ greatly in geological conditions and permafrost environment. Limited by the length of the thesis, it would have been virtually impossible to research all the sections. Therefore, through the measuring data we make a summary and evaluate the whole railway stability. In order to study the engineering effect of the rubble ventilation embankment, the thermal pipe revetment embankment, as well as the
221
rubble and thermal pipe revetment joint embankment, the paper makes a comprehensive estimation from the cooling effect and settlement deformation. In low-temperature permafrost regions, the common embankment can lift the permafrost table under the embankment itself and the shady side, but the permafrost table under the sunny side has descended. In high-temperature permafrost regions, the permafrost table under the shady side has lifted and that under the embankment itself and the sunny side has descended in the common section. So in high-temperature permafrost regions, the sunny–shady slope phenomenon is serious, which considerably varies from annual average air temperature, embankment strike and height, geometrical shape of section, permafrost engineering geological condition and so on. Both thermal pipe revetment embankment and rubble ventilation embankment have perfect effect on cooling the roadbed in permafrost regions. However, for thermal pipe revetment embankment, the sunny–shady phenomenon and asymmetric settlement of embankment have been avoided if the thermal pipe has been strengthened in the sunny side. For rubble ventilation embankment, the sunny–shady effect has been hardly eliminated even though the rubble has been strengthened in the sunny side (for example, rubbles on sunny slope are much thicker than those on the shady slope) largely because the design parameters are not accurate. Theoretically, the rubble and thermal pipe revetment joint embankment performs better than any single measure, but in fact it cannot completely eliminate the sunny–shady slope effect. There is a small amount of application of rubble and thermal pipe revetment joint embankment in CMR, whose engineering effect is most severely disturbed by many factors and the evaluation results are subject to much disturbance. Along the CMR, in all common embankments — the observed results show that the settlement deformation is within 30 cm of the settlement standards of the National II Class Railway. Even the settlement deformation of thermal pipe subgrade and rubble ventilation embankment was able to meet the National I Class Railway standards — 20 cm. Moreover, the settlement deformations of the rubble and thermal pipe revetment joint embankments are within 5 cm. The running time of all sections is not long enough, so the cooling effect of all measures has not yet reflected in the long run. The engineering effect should be further researched. Also, the permafrost environment and the engineering geological conditions along the CMR are typical of both complexity and inconstancy. So, the influence of how the local factors, such as slope-exposure, snow cover, vegetation and swamp distribution, act on permafrost embankment stability cannot be ignored in the next step. 5. Conclusion
(a) Section DK114+730
(b) Section DK114+780 Fig. 9. Settlement deformation changing with time in rubble with thermal pipe embankment sections.
(1) Both thermal pipe revetment embankment and rubble ventilation embankment have perfect effect on cooling the roadbed in permafrost regions. However, for the thermal pipe revetment embankment, the sunny–shady phenomenon and asymmetric settlement of embankment have been avoided if the thermal pipe has been strengthened in the sunny side. For the rubble ventilation embankment, the sunny–shady effect has been hardly eliminated even though the rubble has been strengthened in the sunny side largely because the design parameters are not accurate. For example, rubbles on the sunny slope are not thick enough than those on the shady slope. Although some composite embankments, such as the rubble and thermal pipe revetment joint embankment, have been successfully applied in the Qinghai–Tibet Railway, it is difficult to completely eliminate the sunny–shady slope problem along the CMR because of the complicated engineering geological conditions and the bad permafrost environments. (2) Along the CMR, in all common embankments — the observed results show that the settlement deformation is within 30 cm that is within the settlement standards of the National II Class Railway. Even the settlement deformation of thermal pipe
222
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
subgrade and rubble ventilation embankment was able to meet the National I Class Railway standards — 20 cm. (3) On the basis of the above, some corresponding countermeasures and advice are proposed. First, the synthetic application of engineering measures must be taken. For example, let the structural mechanics measures and temperature control measures unify altogether to prevent longitudinal cracks. Second, the specific work conditions and details such as engineering geological condition, permafrost environments as well as other local factors including the slope orientation have to be taken into account in order to put forward the rational engineering measures and accurate design parameters. Finally, it is essential to protect and to comprehensively evaluate the permafrost environment, which is the ultimate basis of permafrost engineering. Acknowledgments This work was supported by the National Natural Science Foundation of China (50908111), by the Science Fund Project of GanSu Province (1014RJZA026), by the Young Outstanding Teacher Fund of Lanzhou University of Technology (Q200911), by the China Postdoctoral Science Foundation (20100470108), and by the open fund of the State Key Laboratory of Permafrost Engineering (SKLFSE201003), and by the 77 fund of Construction (TM-QK-1101). References Chen, Ji, Song, Ruifang, Sheng, Yu, Dong, Xianfu, Zhang, Luxin, 2011. Cooling effect of measures for rubble subgrade convection of Chaidaer–Muli Railway. J. Railw. Eng. Soc. 5 (Ser.152), 40–44 (55). Cheng, Guodong, 2003. The impact of local factors on permafrost distribution and its inspiring for design Qinghai–Xizang railway. Sci. China Ser. D 33 (6), 602–607. Chou, Yaling, Sheng, Yu, Wei, Zhenming, 2009. Temperature and deformation differences between southern and northern slopes of highway embankment on permafrost. Chin. J. Rock Mech. Eng. 28 (9), 1 (896–1903). Lai, Yuanming, Zhang, Shujuan, Zhang, Luxin, Xiao, Jianzhang, 2004. Adjusting temperature distribution under the south and north slopes of embankment in permafrost regions by the ripped-rock revetment. Cold Reg. Sci. Technol. 39 (1), 67–79. Liu, Xinlong, Li, Ning, 2008. Study on the difference between the southern and northern slopes of the embankment with crushed rock protection in permafrost region. J. Xi, an Technol. Univ. 28 (5), 487–492.
Liu, Yongzhi, Wu, Qingbai, Zhang, Jianming, Sheng, Yu, 2002. Deformation of highway roadbed in permafrost regions of the Tibetan Plateau. J. Glaciol. Geocryol. 24 (1), 10–15. Ma, Wei, Cheng, Guodong, Wu, Qingbai, 2005. Thoughts on solving frozen soil engineering problems in the construction of QingHai–Tibet railway. Sci. Rev. Technol. (1), 23–28. Mao, Xuesong, Hou, Zhongjie, Wang, Weina, Lu, Lu, 2010. Formation mechanism and numerical simulation of longitudinal cracks in wetland section of Qinghai–Tibet highway. Chin. J. Rock Mech. Eng. 29 (9), 1915–1921. Pei, Jianzhong, Dou, Minjian, Hu, Changshun, Wang, Binggang, 2006. Forming mechanism of embankment longitudinal cracks in permafrost regions. J. Glaciol. Geocryol. 28 (1), 117–121. Qi, Changqing, Wu, Qingbai, Shi, Bin, 2007. Analysis of thermal state of permafrost under high embankment along Qinghai–Tibet Railway. Chin. J. Rock Mech. Eng. 26 (Supp.2), 4518–4524. Sheng, Yu., Wei, Ma, Zhi, Wen, Mingyi, Zhang, 2005. Analysis of difference in thermal state between south faced slope and north faced slope of railway embankment in permafrost region. Chin. J. Rock Mech. Eng. 24 (17), 3197–3201. Su, Yi, Xu, Zhaoyi, Wang, Lianjun, 2004. The influence of different roadbed engineering construction on the roadbed crack of Qinghai–Tibet railway. J. Eng. Geol. 12 (03), 292–297. Sun, Zengkui, Wang, Lianjun, Bai, Mingzhou, Wei, Qingchao, 2004. Finite element analysis on temperature field of Qinghai–Tibet Railway embankment on permafrost. Chin. J. Rock Mech. Eng. 23 (20), 3454–3459. Wang, Tiehang, Dou, Mingjian, 2004. Analysis of transverse thermal difference of embankment in permafrost region. Coal Geol. Explor. 32 (1), 45–47. Wang, Yinsheng, Jia, Haifeng, Li, Yong, Du, Shun, 2003. Preliminary analysis on cracking of subgrade in Qingshuihe experimental section of Qinghai–Tibet railway. Chin. J. Rock Mech. Eng. 22 (sup2), 2647–2651. Wang, Xiaojun, Han, Wenfeng, Jiang, Fuqiang, Niu, Huanjun, Chen, Wenwu, Wu, Xiaopeng, Liu, Baosheng, 2006. Discussion and analysis of settlement of rubble ventilation embankment and crack of ordinary embankment in permafrost regions. Chin. J. Rock Mech. Eng. 25 (9), 1904–1911. Wu, Qingbai, Liu, Yongzhi, Shi, Bin, Zhang, Jianming, Tong, Changjiang, 2002. Advance research on frozen engineering permafrost region along Qinghai–Xizang plateau highway. J. Eng. Geol. 10 (1), 55–61. Wu, Qingbai, Zhao, Shiyun, Ma, Wei, Liu, Yongzhi, Zhang, Luxin, 2005. Monitoring and analysis of cooling effect of block-stone embankment for Qinghai–Tibet Railway. J. Geotech. Eng. (12), 1386–1390. Xu, Anhua, 2010. A discussion about the relation between longitudinal embankment cracks and road trend in permafrost regions. J. Glaciol. Geocryol. 32 (1), 121–125. Yu, Qihao, Pan, Xicai, Cheng, Guodong, He, Naiwu, 2007. Study on main influential factors on ventilated embankment and corresponding measures. Chin. J. Rock Mech. Eng. 26 (Supp.1), 3045–3051. Zhang, Luxin, Yuan, Sicheng, Yang, Yongping, 2003. Mechanism and prevention of deformation cracks of embankments in the permafrost region along Qinghai–Xizang railway. Quat. Sci. 23 (6), 604–610.
Contents lists available at ScienceDirect
Cold Regions Science and Technology journal homepage: www.elsevier.com/locate/coldregions
Analysis on thermal stability and deformation stability of CMR Yaling Chou a,b,⁎, Yu Sheng c, Ji Chen c, Xingqiang Chen a,b a b c
Key Laboratory of Disaster Prevention and Mitigation in Civil Engineering of Gansu Province, Lanzhou University of Technology, Lanzhou, Gansu 730000, China Northwest Center for Disaster Mitigation in Civil Engineering of Ministry of Education, Lanzhou, Gansu 730000, China State Key Laboratory of Natural Gansu, Lanzhou Institute of Geology, Chinese Academy of Sciences, Lanzhou, Gansu 730000, China
a r t i c l e
i n f o
Article history: Received 15 February 2014 Received in revised form 25 August 2014 Accepted 23 October 2014 Available online 29 October 2014 Keywords: Chai–Mu Railway Engineering measures Shady–sunny slope Permafrost embankment Thermal stability Deformation stability
a b s t r a c t In permafrost regions, embankment stability has been destroyed with longitudinal cracks primarily caused by transverse asymmetric settlement. And shady–sunny slope effect is a principal factor which might result in embankment asymmetric settlement. Based on field data of some typical sections along the Chai–Mu Railway (simply called CMR) with obvious shady–sunny slopes, which is located on the Qinghai–Tibet Plateau, the ground temperature characteristics and settlement deformation properties of the embankment have been analyzed. This paper has studied how the different engineering measures, including rubble ventilation embankment, thermal pipe revetment embankment, rubble and thermal pipe revetment joint embankment, did influence the shady– sunny slope effect. The results show that: (1) Both thermal pipe revetment embankment and rubble ventilation embankment have perfect effect on cooling the roadbed in permafrost regions. However, for thermal pipe revetment embankment, the sunny–shady phenomenon and asymmetric settlement of embankment have been avoided if thermal pipe has been strengthened in sunny side. For rubble ventilation embankment, the sunny–shady effect has been hardly eliminated even though the rubble has been strengthened in sunny side. Although some composite embankments, such as the rubble and thermal pipe revetment joint embankment, have been successfully applied in the Qinghai–Tibet Railway, it is difficult to completely eliminate the sunny– shady slope problem along the CMR because of the complicated engineering geological conditions and the bad permafrost environments. (2) Along the CMR, in all common embankments — the observed results show that the settlement deformation is within 30 cm that is within the settlement standards of the National II Class Railway. Even the settlement deformation of thermal pipe subgrade and rubble ventilation embankment was able to meet the National I Class Railway standards — 20 cm. (3) On the basis of the above, some corresponding countermeasures and advice are proposed. First, rational use of the synthetic application of engineering measures must be taken. For example, let the structural mechanics measures and temperature controlled measures unify altogether to prevent longitudinal cracks. Second, the specific work conditions and details such as engineering geological condition, permafrost environments as well as other local factors including the slope orientation have to be taken into account in order to put forward rational engineering measures and accurate design parameters. Finally, it is essential to protect and to comprehensively evaluate the permafrost environment, which is the ultimate basis of permafrost engineering. © 2014 Elsevier B.V. All rights reserved.
1. Introduction In permafrost regions, construction of embankments has changed the thermo-physical features of natural ground surface and atmosphere, influenced thermodynamic and dynamic stability of frozen soil layers, and complicated temperature fields of frozen soil embankment (Ma et al., 2005; Qi et al., 2007; Sun et al., 2004; Yu et al., 2007). When the two embankment slopes are south-facing and north-facing, respectively, which is called shady–sunny slope in Chinese, the difference of thermal boundary conditions such as solar radiation and surface turbulent flow will induce transverse asymmetry of temperature field, ⁎ Corresponding author. Tel.: +86 931 4967297; fax: +86 931 4967271. E-mail addresses: [email protected], [email protected] (Y. Chou).
http://dx.doi.org/10.1016/j.coldregions.2014.10.004 0165-232X/© 2014 Elsevier B.V. All rights reserved.
and this will lead to a multitude of problems in permafrost roadbeds. And there extensively exists these kinds of phenomena in both warm and low temperature permafrost regions (Liu et al., 2002). The thermal difference of the shady–sunny slope results in there is an unbalanced freeze–thaw condition between both sides of the embankment, which leads, inevitably, to transverse differential settlement. As time goes by, the differential settlement will be aggravated, and finally the embankment stability is destroyed with longitudinal cracks caused by asymmetric settlement of permafrost (Wang and Dou, 2004; Wang et al., 2006). Based on observed field temperature data in Beiluhe on the Qinghai–Tibet Plateau, Sheng et al. (2005) analyzed the thermal difference between the shady and sunny slopes, and pointed out that the thermal difference can lead to development of embankment longitudinal cracks. Lai et al. (2004) recommended that the embankment
216
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
disease that results in uneven settlement with longitudinal cracks can be eliminated by paving ripped-rocks of different thickness on the sunny and shady slopes, respectively. Cheng (2003) showed that some local factors including slope orientation can change permafrost distribution and may cause some engineering damages such as longitudinal cracks. The investigation was done by analyzing both observational geo-temperature and deformation data of the embankment in experimental section K369+100 along the National Highway 214 on the Qinghai–Tibet Plateau. Also, the fundamental cause of the embankment transverse asymmetrical thermal regime and asymmetrical deformation was discussed (Chou et al., 2009). The longitudinal cracks are a very common permafrost embankment disease, which has attracted the attention of more and more researchers (Mao et al., 2010; Pei et al., 2006; Wu et al., 2002; Xu, 2010; Zhang et al., 2003). According to field observation, the longitudinal cracks are distributed mainly over the sunny side of high embankments in unstable permafrost regions. Moreover, the longitudinal cracks developing along the sunny side covers more than 70%. And the longitudinal cracks appear not only in common embankments, but also in reinforced embankments, such as thermal pipe embankment, ripped-rock revetment embankment, insulation embankment and so on (Su et al., 2004; Wang et al., 2003). Through field investigation, Wu et al.(2005) pointed that a better thermal stability situation will be developed under conditions where the temperature near the permafrost table beneath the rubble embankment has largely dropped in cold permafrost regions. But in warm permafrost regions, even the permafrost table has highly raised the annual thermal imbalance of permafrost and temperature will rise, and it is disadvantageous for the thermal stability of permafrost. In the warm permafrost region, only the embankment shady side can be better cooled by adopted rubble layer but in the cold permafrost region, the rubble ventilation embankment not only has cooled the whole subgrade, but also has played a certain role in restraining the sunny–shady effect (Chen et al., 2011). Liu and Li (2008) predicted the temperature field
of embankment with crushed rock protection in the future 50 years by applying the transient thermodynamics FEM method. The computation results show that the temperature difference between the sunny slope and the shady slope has existed at all times, but the difference on the same days and months has varied little with year. Therefore, how the different engineering measures have influenced the shady– sunny slope effect is an important subject worth studying. The Chaidaer Muli Railway, simply called the CMR, is the first local railway with a total length of 142 km in northeastern Qinghai Province, China. It starts in Chaidaer County (100°25′55″E, 37°35′40″N) and ends in Muli Country (99°11′00″E, 38°08′38″N). Fig. 1 gives the geographical location of the CMR. Topographically, it passes through the Datong River valley, the Datong Mountains and the front edge of the Tuolai Mountain in the central-eastern Qilian Mountains. The railway broadly runs from southeast to northwest. The elevations of this region range from 3600 m to 4100 m and the average annual air temperatures vary from about − 2.4 °C to − 5.8 °C. According to the meteorological data, the minimum air temperature is −40 °C and the maximum air temperature is 17 °C, respectively. There is rich rain and snow, and swamps as well as marshes are both developed. The annual precipitation along the CMR is approximately 500 mm·a−1. The vegetative coverage is more than 60% in most of the permafrost regions, and the wetlands are widely distributed. About 77 km of all the CMR runs across wetlands. The permafrost is discontinuous and unstable with average annual ground temperature ranging from − 1 °C to 0 °C. The ice contents in most permafrost regions are also very high. During the construction of CMR, based on the experience of the Qinghai–Tibet Railway, a number of engineering measures were taken to alleviate the sunny–shady effect. This paper will study the ground temperature characteristics and the settlement deformation properties of the embankment through analyzing the observation data of some typical embankment sections along the CMR. Also, that the positive cooling engineering measures, including the rubble ventilation embankment, the thermal pipe revetment embankment, the rubble
Fig. 1. The geographical location of the CMR.
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
and thermal pipe revetment joint embankment, make an appreciable difference to the shady–sunny slope effect is mainly discussed. This work will provide a scientific basis for embankment construction and maintenance in permafrost regions on the Qinghai–Tibet Plateau. 2. Monitoring content Engineering geological conditions, permafrost characteristic, engineering measures, embankment strike and height, and the embankment section shape of the CMR have been investigated. The temperature monitoring probes are laid out within some typical sections. In order to observe the engineering effect of different measures, different ground temperature zones and different monitoring sites have been given a comprehensive consideration when the monitoring sections were laid out. The construction of monitoring sections began on October, 2007. And until July, 2008, the main monitoring equipments and apparatus have completely finished. By early 2010, the whole observation time has been not enough up to 2 annual cycles. As far as the effect of different engineering measures is concerned, the classification of measures is discussed, respectively. Meanwhile, the engineering measures and the effects are deeply analyzed. The principal strike of CMR runs from southeast to northwest, furthermore there are more sections close to run from east to west. Since the effect of the shady–sunny slope is obvious, we select some typical sections to comparatively analyze how the different engineering measures have inhibited the shady–sunny slope phenomenon. In this paper, temperature of slope means the temperature at shallow ground depth (0.5 m) under the sunny slope and the shady slope, respectively. And we refer to the settlement deformation of the sunny shoulder and the shady shoulder as settlement displacement of two slopes, respectively. The data acquisition apparatus is composed of thermal-susceptible resistance sensors and Fluke IV multimeters with a 0.1 Ω precision. The space between two probes is ranging from 0.5 m to 2.0 m. The frequency of observation is once every 15 days. The embankment heights range from 2.0 m to 5.3 m and the fill is fine-grained and gravelly soil. The road surface width is between 6.7 m and 7.9 m, and the slope gradient is 1:1.5. All the monitoring sections are on the slope wetlands, and some sections are asymmetric in transverse profile. The sections DK39+800, DK40+000, DK74+000, DK74+500, DK75+000, DK94+340, DK94+660, DK94+900, DK99+100, DK99+200, DK99+355, DK114+730, DK114+800, DK123+150 and DK123+250 are all laid out with temperature probes. In order to save space, only a
217
sketch of observation of the transverse section and temperature holes of section DK114+730 (Fig. 2) is given here, which is the rubble and thermal pipe revetment joint embankment. The temperature probe positions of other sections are similar to section DK114+730. The length of rubble ventilation embankments and rubble berm embankments reaches 53 km. The rubbles in the ventilation embankment are obtained near the CMR, which are made up mainly of moderately weathered schist and gneiss interlaid with a little pebble. The particle size of rubbles is mainly from 5 cm to 30 cm. 3. Monitoring results 3.1. The rubble ventilation embankment 3.1.1. The temperature difference For the south to north strike, the solar radiations that the two embankment slopes have absorbed are essentially the same. So the permafrost temperature field under embankment is substantially symmetric, and there will be no serious disease because the shady– sunny slope problem is not obvious. But the monitoring sections of the rubble ventilation embankment, such as section DK40+000, section DK74+000, section DK94+900, section DK123+150 and section DK123+250, their strikes are from 260° to 345°. In following description for convenience, the south to north strike is defined as 0°, which is the exact thermal symmetric embankment. And the east to west strike is called 90°, which means the most remarkable shady–sunny slope effect. According to the rule, the strikes of the 5 sections (DK40+000, DK74+000, DK94+900, DK123+150 and DK123+250) are 40°, 60°, 15°, 80° and 80°, respectively. The annual mean ground temperatures of the 5 sections are −0.65 °C, −0.55 °C, −1.31 °C, − 1.18 °C and −1.10 °C, respectively. The shallow ground temperatures of section DK74+000, section DK94+900 and section DK123+150 at a depth of 0.5 m under slope surface are shown in Fig. 3. For section DK74+000 and section DK123+150, we find two obvious phenomena: 1) the shallow ground temperatures (0.5 m) on shady slopes are always lower than those of the sunny slopes; 2) between the two slopes, the maximum temperature difference appears in winter and there has been a little temperature difference in summer. In all the sections, the shady–sunny slope effect of section DK94+900 is the weakest, and the temperature curve crosses each other so that it is difficult to discriminate the temperature difference between two slopes only through Fig. 3. The results of the data analysis indicate that the mean annual ground temperature of section DK94+900, section DK74+000 and
Fig. 2. Sketch of observation transverse section and temperature holes of section DK114+730 (unit: m).
218
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
than that of the shady foot by 0.47 m. And in section DK123+150, the permafrost table of the sunny foot is deeper than that of the shady foot by 1.92 m. In 2009–2010, there is little permafrost table difference between the sunny shoulder and shady shoulder in section DK94+900. However, the permafrost table difference of section DK123+150 between the sunny shoulder and shady shoulder reaches 0.4 m. The strike difference of the rubble ventilation embankment shows that there is still a sunny–shady slope problem if the sunny side has not adopted special measures to eliminate the sunny–shady slope phenomenon. Therefore, how the rubble ventilation embankment completely gets rid of the sunny–shady slope problem will need to be studied further. For instance, the sunny–shady slope phenomenon may be eliminated or weakened by building a much thicker rubble or laying thermal pipe on the sunny side. According to the specific work condition, the rational and accurate design parameters should be put forward.
(a) Section DK74+000
(b) Section DK94+900
(c) Section DK123+150 Fig. 3. Shallow ground temperatures of 0.5 m under slope surface in rubble embankment sections.
section DK123+150 at a depth of 0.5 m on the sunny slope is higher than that on the shady slope by over 2.23 °C, 4.69 °C and 7.13 °C, respectively. The perturbations that the sunny–shady slope affects on the permafrost table are the most fundamental performance of the thermal asymmetric embankment, which might destroy the embankment stability with longitudinal cracks. Table 1 shows the artificial permafrost table of different thermometer holes in the rubble embankment sections. Under conditions without any other external factors, the temperature difference has increased with the embankment strike increasing, and the difference of the permafrost table between the sunny side and shady side has increased accordingly. In 2008–2009, in section DK94+900 the permafrost table of the sunny foot is higher
3.1.2. The settlement deformation difference The embankment settlement deformation is related to many factors, such as engineering geological condition, corresponding measures, construction quality and so on. By analyzing the settlement deformation, we describe whether the engineering geology investigation and the design parameters are accurate or not, as well as whether the embankment construction is qualified. Moreover, for embankments in permafrost regions, the settlement deformation depends largely upon the physical and mechanical properties of the underlying permafrost. Generally speaking, permafrost degradation inevitably results in the settlement deformation increasing. So the key of embankment treatment is to control the settlement of the bearing stratum. Also, analysis on the settlement deformation can also help us study the cooling effect of the rubble ventilation embankment further. The following will discuss the settlement deformation of the rubble embankment settlement in high-temperature permafrost regions and in low-temperature permafrost regions, respectively. Section DK39+800 and section DK40+000 are both in hightemperature permafrost regions, and the annual average air temperature here is between − 2.1 °C to − 2.9 °C. Fig. 4 gives the settlement deformation curves of section DK39+800 and section DK40+000. Section DK39+800 is an ordinary embankment without any measures (as the contrast section). The curves have features such as:(1) the settlement deformation of the sunny shoulder is more than that of shady shoulder, and this rule is in agreement with the abovementioned temperature difference; (2) the settlement deformation of the rubble embankment section is greater than that of the contrast section; and (3) there is a slight frost heave in the freezing season. Since the sunny slope has a high temperature, the permafrost table under the sunny side is much deeper than that under the shady side, and a greater settlement deformation has occurred in the sunny side. However, feature (2) does not agree with the temperature field. The ground temperature of section DK40+000 is lower than that of section DK39+800, but the settlement deformation of section DK40+000 is greater than that of section DK39+800. According to the engineering geological conditions, there are both thick layers of humus in the ground under the two sections. There is humus with a thickness of 2 m under section DK39+800 and humus with a thickness of 3 m under section DK40+000, respectively. Therefore, the rubble embankment section has
Table 1 The drilling table in rubble embankment sections (unit: m). Section
DK40+000 DK74+000 DK94+900 DK123+150 DK123+250
2008–2009
2009–2010
Sunny foot
Shady foot
Sunny shoulder
Shady shoulder
Sunny foot
Shady foot
Sunny shoulder
Shady shoulder
−2.23 −3.82 1.72 2.90 1.64
– −1.45 2.19 0.98 1.72
– – 1.28 1.54 1.43
−1.26 – 1.50 1.35 0.77
−2.20 −3.89 – 2.80 2.15
– −1.37 1.89 0.98 1.32
– – 1.52 1.55 1.53
−0.94 – 1.51 1.15 0.77
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
219
3.2. Thermal pipe revetment embankment
Fig. 4. Settlement deformation changing with time in rubble embankment sections.
poor geological condition that results in the settlement deformation of section DK40+000 which is greater than that of section DK39+800. From engineering measures, 2 m of soft soil in section DK39+800 has been dug. Meanwhile, the rubble with a thickness of 1.2 m has been filled on the original earth surface in section DK40+000, and the soft soil has not been dug. The soft soil is located near the permafrost table, which is more vulnerable to thaw under the engineering disturbances, so there is a big settlement after construction in section DK40+000. The digging of soft soil much reduces the settlement deformation, so the settlement deformation of section DK39+800 is relatively small. As section DK40+000 is concerned, if the rubble can lift the permafrost table to the ground surface, the compression deformation of the seasonal active layer is avoided. But in fact, the permafrost table is −0.94 m below the natural ground. That is to say, a large deformation of section DK40+000 takes place as a result of seasonal active layer thawing. In low-temperature permafrost regions, the settlement deformation curves are analogous to those in high-temperature permafrost regions. Section DK123+150 and section DK123+250 are both in lowtemperature permafrost regions, and the annual average air temperature here is about − 3.5 °C. Fig. 5 gives the settlement deformation curves of section DK123+150 and section DK123+250 (the contrast section). There is a larger settlement deformation in the sunny side and smaller settlement deformation in the shady side. Comparing Figs. 4 and 5, it is not difficult to find that the settlement deformation in hightemperature permafrost regions is much larger. Through analyzing all the sections, by the beginning of 2010 the amount of deformation has been within 10 cm–20 cm in high-temperature permafrost regions, but the amount of deformation in low-temperature permafrost regions has reached 2 cm–8 cm. Both in high-temperature permafrost regions and in low-temperature permafrost regions, the amounts of settlement deformation in rubble embankment sections were able to meet the National I Class Railway standards — 20 cm.
3.2.1. The temperature difference As a highly-effective heat conduction set, thermal pipes are commonly used to enhance thermal stability of building foundation in permafrost regions. In order to eliminate the sunny–shady slope problem, the thermal pipes have stood in two ranks on sunny side, but there is a row of pipes on the shady side. The longitudinal spacing of the thermal pipe is 3 m. On the sunny side, one row of thermal pipes is situated at the embankment foot, and another row of thermal pipes lie beside the revetment foot. There is 3 m between the two rows of thermal pipes. On the shady side, there is one row of thermal pipes beside the revetment shoulder. Section DK74+500, section DK99+100 and section 99+200 are thermal pipe revetment embankments, whose strikes are 70°, 35° and 35°, respectively. The annual mean ground temperatures of the three sections are − 0.32 °C, − 0.89 °C and − 0.99 °C, respectively. Fig. 6 shows shallow ground
(a) Section DK74+500
(b) Section DK99+100
(c) Section DK99+200
(d) Section DK99+355 Fig. 5. Settlement deformation changing with time in rubble embankment sections.
Fig. 6. Ground temperatures under slope surface changing in thermal pipe embankment sections.
220
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
temperatures of slope in thermal pipe revetment embankments. Section DK99+355 is a common landfill embankment, as a contrast section, whose strike is 35° and the sunny–shady slope effect is weaker. We can see: (1) in the thermal pipe revetment sections, the temperature of sunny side is lower than that of the shady side during most time of the year. But in contrast section DK99+355, the temperature of the sunny side is higher than that of the shady side during most time of the year. And the occurrence of this phenomenon is apparently related to the cooling effect of the thermal pipe; (2) the annual temperature amplitude of the thermal pipe section is greater than that of the contrast section. The annual temperature amplitude of the thermal pipe section is 40 °C, but the annual temperature amplitude of the contrast section is not beyond 20 °C. Judging from the processes of temperature curves, the temperature amplitude over a short period in the contrast section is relatively small, and the temperature follows a relatively smooth curve. Meanwhile, the temperature amplitude over a short period in the thermal pipe section is relatively larger, and the temperature curve occurs in jumps. The temperature amplitude difference between the two types of section is not in evident in the thermal pipe, which is related to covering conditions of the slopes. There is generally rubble masonry pavement in the thermal pipe section. The covering layer of the revetment is thin, and whose thermal conductivity is high, so the temperature wave is easy to transfer in the deep ground. However, the contrast section DK99+355 has been adopted by grassed slope and the covering layer is much thicker. The moisture content of grassed covering layer is high, which has seriously hindered the temperature wave from propagating. Consequently, there is an extreme difference in temperature amplitude between the two types of sections. 3.2.2. The settlement deformation difference Based on the above analysis, the thermal pipe revetment embankment can effectively cool the permafrost embankment and can weaken or remove the sunny–shady slope problem. Moreover, theoretically the thermal pipe is normally effective in decreasing the permafrost embankment deformation. Fig. 7 gives settlement deformation changing with time in section DK99+100 and section DK99+200, and the settlement characteristics are:(1) there is a certain fluctuation in the settlement curve, which is sinking in the thaw season and is a slight frost heave in the freezing season;(2) the thermal pipe revetment is a comparatively good method of inhibiting settlement deformation and asymmetric settlement. In summary, the settlement deformations of the two sections are both within 10 cm. Even the settlement deformation of section DK99+200 is less than 3 cm. Also, the embankment settlement deformation stops increasing in the later stages and the embankment has stabilized. It was obvious that the amounts of settlement deformation were able to meet the National I Class Railway standards — 20 cm.
(a) Section DK 99+100
(b) Section DK99+200 Fig. 7. Settlement deformation changing with time in thermal pipe embankment sections.
higher than that of the shady slope (Fig. 8). The temperature curves of two slopes present that the rubble and thermal pipe revetment joint embankment have not completely eliminated the sunny–shady slope problem, which is chiefly due to many factors. Firstly, the embankment is higher so that it needs more filler, which results in much bigger thermal disturbance on permafrost. Also, the higher the embankment, the bigger the slope area, the more solar radiation is absorbed by the
3.3. The rubble and thermal pipe revetment joint embankment There is a small amount of application of the rubble and thermal pipe revetment joint embankments in CMR, which are mainly used for sections in ice-rich permafrost regions where the engineering geological condition are very poor. 3.3.1. The temperature difference The embankment strikes of sections DK114+730 and DK114+800 are both 65°, whose sunny–shady slope problem is very serious. The annual mean ground temperatures of the two sections are − 1.36 °C and −1.28 °C, respectively. During the first design, the rubble ventilation embankment has been only used. Afterwards there happens sideslip in the sections, so the thermal pipe is strengthened in the rubble ventilation embankment. Here, the contrast section is not laid out. Due to the influence of engineering, the temperature curves of the sunny slope and the shady slope intersect each other at an early stage. But after the early stage, the temperature of the sunny slope is always
(a) Section DK114+730
(b) Section DK114+800 Fig. 8. Ground temperatures under slope surface in rubble with thermal pipe embankment sections.
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
sunny slope. Secondly, the thermal insulation of rubble ventilation has hindered the cooling effect of the thermal pipe. Thirdly, the local geography, geological environments and climate conditions resulting in the cooling effect of the joint embankment are not ideal. From these analyses, we realized that the main way of heat disturbance on permafrost has not been made clear, even these powerful measures that have been proved effective in the Qinghai–Tibet Railway (another railway on the Qinghai–Tibet Plateau), such as the rubble and thermal pipe revetment joint embankment, cannot completely eliminate the sunny–shady slope effect in CMR. The problem acts as a strong warning to permafrost embankment design, and so many factors have to be taken into account. The rational use of the synthetic application of engineering measures must be taken. For example, let structural mechanics measures and temperature controlled measures unify altogether to prevent longitudinal cracks. 3.3.2. The settlement deformation difference Fig. 9 gives settlement deformation changing with time in section DK114+730 and section DK114+780, whose distribution is consistent with the shallow ground temperature. The settlement characteristic is: (1) the settlement deformations of the two sections are both small, which is within 5 cm; (2) the settlement deformations are smooth and steady, and there is little fluctuation in the settlement curve, which means that there is sinking in the thaw season and a slight frost heave in the freezing season; and (3) the settlement deformation of the sunny side is greater than that of the shady side. Although settlement deformation difference between two shoulders is very minute, this contrast relationship agrees with the temperature. The embankment settlement deformation is little increasing with time and the embankment has slowly stabilized. 4. Analysis and discussion There are many monitoring sections along CMR, and these sections differ greatly in geological conditions and permafrost environment. Limited by the length of the thesis, it would have been virtually impossible to research all the sections. Therefore, through the measuring data we make a summary and evaluate the whole railway stability. In order to study the engineering effect of the rubble ventilation embankment, the thermal pipe revetment embankment, as well as the
221
rubble and thermal pipe revetment joint embankment, the paper makes a comprehensive estimation from the cooling effect and settlement deformation. In low-temperature permafrost regions, the common embankment can lift the permafrost table under the embankment itself and the shady side, but the permafrost table under the sunny side has descended. In high-temperature permafrost regions, the permafrost table under the shady side has lifted and that under the embankment itself and the sunny side has descended in the common section. So in high-temperature permafrost regions, the sunny–shady slope phenomenon is serious, which considerably varies from annual average air temperature, embankment strike and height, geometrical shape of section, permafrost engineering geological condition and so on. Both thermal pipe revetment embankment and rubble ventilation embankment have perfect effect on cooling the roadbed in permafrost regions. However, for thermal pipe revetment embankment, the sunny–shady phenomenon and asymmetric settlement of embankment have been avoided if the thermal pipe has been strengthened in the sunny side. For rubble ventilation embankment, the sunny–shady effect has been hardly eliminated even though the rubble has been strengthened in the sunny side (for example, rubbles on sunny slope are much thicker than those on the shady slope) largely because the design parameters are not accurate. Theoretically, the rubble and thermal pipe revetment joint embankment performs better than any single measure, but in fact it cannot completely eliminate the sunny–shady slope effect. There is a small amount of application of rubble and thermal pipe revetment joint embankment in CMR, whose engineering effect is most severely disturbed by many factors and the evaluation results are subject to much disturbance. Along the CMR, in all common embankments — the observed results show that the settlement deformation is within 30 cm of the settlement standards of the National II Class Railway. Even the settlement deformation of thermal pipe subgrade and rubble ventilation embankment was able to meet the National I Class Railway standards — 20 cm. Moreover, the settlement deformations of the rubble and thermal pipe revetment joint embankments are within 5 cm. The running time of all sections is not long enough, so the cooling effect of all measures has not yet reflected in the long run. The engineering effect should be further researched. Also, the permafrost environment and the engineering geological conditions along the CMR are typical of both complexity and inconstancy. So, the influence of how the local factors, such as slope-exposure, snow cover, vegetation and swamp distribution, act on permafrost embankment stability cannot be ignored in the next step. 5. Conclusion
(a) Section DK114+730
(b) Section DK114+780 Fig. 9. Settlement deformation changing with time in rubble with thermal pipe embankment sections.
(1) Both thermal pipe revetment embankment and rubble ventilation embankment have perfect effect on cooling the roadbed in permafrost regions. However, for the thermal pipe revetment embankment, the sunny–shady phenomenon and asymmetric settlement of embankment have been avoided if the thermal pipe has been strengthened in the sunny side. For the rubble ventilation embankment, the sunny–shady effect has been hardly eliminated even though the rubble has been strengthened in the sunny side largely because the design parameters are not accurate. For example, rubbles on the sunny slope are not thick enough than those on the shady slope. Although some composite embankments, such as the rubble and thermal pipe revetment joint embankment, have been successfully applied in the Qinghai–Tibet Railway, it is difficult to completely eliminate the sunny–shady slope problem along the CMR because of the complicated engineering geological conditions and the bad permafrost environments. (2) Along the CMR, in all common embankments — the observed results show that the settlement deformation is within 30 cm that is within the settlement standards of the National II Class Railway. Even the settlement deformation of thermal pipe
222
Y. Chou et al. / Cold Regions Science and Technology 110 (2015) 215–222
subgrade and rubble ventilation embankment was able to meet the National I Class Railway standards — 20 cm. (3) On the basis of the above, some corresponding countermeasures and advice are proposed. First, the synthetic application of engineering measures must be taken. For example, let the structural mechanics measures and temperature control measures unify altogether to prevent longitudinal cracks. Second, the specific work conditions and details such as engineering geological condition, permafrost environments as well as other local factors including the slope orientation have to be taken into account in order to put forward the rational engineering measures and accurate design parameters. Finally, it is essential to protect and to comprehensively evaluate the permafrost environment, which is the ultimate basis of permafrost engineering. Acknowledgments This work was supported by the National Natural Science Foundation of China (50908111), by the Science Fund Project of GanSu Province (1014RJZA026), by the Young Outstanding Teacher Fund of Lanzhou University of Technology (Q200911), by the China Postdoctoral Science Foundation (20100470108), and by the open fund of the State Key Laboratory of Permafrost Engineering (SKLFSE201003), and by the 77 fund of Construction (TM-QK-1101). References Chen, Ji, Song, Ruifang, Sheng, Yu, Dong, Xianfu, Zhang, Luxin, 2011. Cooling effect of measures for rubble subgrade convection of Chaidaer–Muli Railway. J. Railw. Eng. Soc. 5 (Ser.152), 40–44 (55). Cheng, Guodong, 2003. The impact of local factors on permafrost distribution and its inspiring for design Qinghai–Xizang railway. Sci. China Ser. D 33 (6), 602–607. Chou, Yaling, Sheng, Yu, Wei, Zhenming, 2009. Temperature and deformation differences between southern and northern slopes of highway embankment on permafrost. Chin. J. Rock Mech. Eng. 28 (9), 1 (896–1903). Lai, Yuanming, Zhang, Shujuan, Zhang, Luxin, Xiao, Jianzhang, 2004. Adjusting temperature distribution under the south and north slopes of embankment in permafrost regions by the ripped-rock revetment. Cold Reg. Sci. Technol. 39 (1), 67–79. Liu, Xinlong, Li, Ning, 2008. Study on the difference between the southern and northern slopes of the embankment with crushed rock protection in permafrost region. J. Xi, an Technol. Univ. 28 (5), 487–492.
Liu, Yongzhi, Wu, Qingbai, Zhang, Jianming, Sheng, Yu, 2002. Deformation of highway roadbed in permafrost regions of the Tibetan Plateau. J. Glaciol. Geocryol. 24 (1), 10–15. Ma, Wei, Cheng, Guodong, Wu, Qingbai, 2005. Thoughts on solving frozen soil engineering problems in the construction of QingHai–Tibet railway. Sci. Rev. Technol. (1), 23–28. Mao, Xuesong, Hou, Zhongjie, Wang, Weina, Lu, Lu, 2010. Formation mechanism and numerical simulation of longitudinal cracks in wetland section of Qinghai–Tibet highway. Chin. J. Rock Mech. Eng. 29 (9), 1915–1921. Pei, Jianzhong, Dou, Minjian, Hu, Changshun, Wang, Binggang, 2006. Forming mechanism of embankment longitudinal cracks in permafrost regions. J. Glaciol. Geocryol. 28 (1), 117–121. Qi, Changqing, Wu, Qingbai, Shi, Bin, 2007. Analysis of thermal state of permafrost under high embankment along Qinghai–Tibet Railway. Chin. J. Rock Mech. Eng. 26 (Supp.2), 4518–4524. Sheng, Yu., Wei, Ma, Zhi, Wen, Mingyi, Zhang, 2005. Analysis of difference in thermal state between south faced slope and north faced slope of railway embankment in permafrost region. Chin. J. Rock Mech. Eng. 24 (17), 3197–3201. Su, Yi, Xu, Zhaoyi, Wang, Lianjun, 2004. The influence of different roadbed engineering construction on the roadbed crack of Qinghai–Tibet railway. J. Eng. Geol. 12 (03), 292–297. Sun, Zengkui, Wang, Lianjun, Bai, Mingzhou, Wei, Qingchao, 2004. Finite element analysis on temperature field of Qinghai–Tibet Railway embankment on permafrost. Chin. J. Rock Mech. Eng. 23 (20), 3454–3459. Wang, Tiehang, Dou, Mingjian, 2004. Analysis of transverse thermal difference of embankment in permafrost region. Coal Geol. Explor. 32 (1), 45–47. Wang, Yinsheng, Jia, Haifeng, Li, Yong, Du, Shun, 2003. Preliminary analysis on cracking of subgrade in Qingshuihe experimental section of Qinghai–Tibet railway. Chin. J. Rock Mech. Eng. 22 (sup2), 2647–2651. Wang, Xiaojun, Han, Wenfeng, Jiang, Fuqiang, Niu, Huanjun, Chen, Wenwu, Wu, Xiaopeng, Liu, Baosheng, 2006. Discussion and analysis of settlement of rubble ventilation embankment and crack of ordinary embankment in permafrost regions. Chin. J. Rock Mech. Eng. 25 (9), 1904–1911. Wu, Qingbai, Liu, Yongzhi, Shi, Bin, Zhang, Jianming, Tong, Changjiang, 2002. Advance research on frozen engineering permafrost region along Qinghai–Xizang plateau highway. J. Eng. Geol. 10 (1), 55–61. Wu, Qingbai, Zhao, Shiyun, Ma, Wei, Liu, Yongzhi, Zhang, Luxin, 2005. Monitoring and analysis of cooling effect of block-stone embankment for Qinghai–Tibet Railway. J. Geotech. Eng. (12), 1386–1390. Xu, Anhua, 2010. A discussion about the relation between longitudinal embankment cracks and road trend in permafrost regions. J. Glaciol. Geocryol. 32 (1), 121–125. Yu, Qihao, Pan, Xicai, Cheng, Guodong, He, Naiwu, 2007. Study on main influential factors on ventilated embankment and corresponding measures. Chin. J. Rock Mech. Eng. 26 (Supp.1), 3045–3051. Zhang, Luxin, Yuan, Sicheng, Yang, Yongping, 2003. Mechanism and prevention of deformation cracks of embankments in the permafrost region along Qinghai–Xizang railway. Quat. Sci. 23 (6), 604–610.