Nat Hazards (2015) 75:2589–2605 DOI 10.1007/s11069-014-1444-5 ORIGINAL PAPER

Degradation characteristics of permafrost under the effect of climate warming and engineering disturbance along the Qinghai–Tibet Highway Hui Peng • Wei Ma • Yan-hu Mu • Long Jin • Kun Yuan

Received: 16 July 2014 / Accepted: 17 September 2014 / Published online: 1 October 2014 Ó Springer Science+Business Media Dordrecht 2014

Abstract Based on the monitoring data from 13 typical monitoring sites along the Qinghai–Tibet Highway, the degradation characteristics of the permafrost under asphalt pavement and natural ground surface were analyzed with considerations of climate warming and engineering disturbance. Results indicated that the mean annual thawing indexes (MATI) and mean annual freezing indexes (MAFI) of asphalt pavement ranged from 895 to 2,540 °C days and from 290 to 1,097 °C days, respectively, while the MATI and MAFI of natural ground ranged from 144 to 1,550 °C days and from 127 to 1,544 °C days, respectively. In warm seasons, average temperatures of asphalt pavement were 0.76–8.58 °C higher than that of natural ground, while in cold seasons, average temperatures of asphalt pavement were 0.22–4.19 °C lower than that of natural ground. Both natural permafrost table and artificial permafrost table were continuously declining through 1995–2011. Under the effect of climate warming, the active layer thickness (ALT) increased about 0.44 m, with an average increasing rate of 3.42 cm a-1 in cold permafrost regions [the mean annual ground temperature lower than -1.0 °C (MAGT \ -1.0 °C)], while in warm permafrost regions (MAGT [ -1.0 °C), the ALT increased about 0.68 m, with an average increasing rate of 5.72 cm a-1. Under the effect of engineering disturbance, the ALT increased 1.38 m in cold permafrost regions, with an average increasing rate of 12.28 cm a-1, while in warm permafrost regions, the ALT increased 1.32 m, with

H. Peng  W. Ma (&)  Y. Mu  K. Yuan State Key Laboratory of Frozen Soil Engineering, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Science, Lanzhou 730000, People’s Republic of China e-mail: [email protected] H. Peng e-mail: [email protected] H. Peng  L. Jin  K. Yuan China Communications Construction Company First Highway Consultants Co., Ltd, Xi’an 710000, People’s Republic of China H. Peng  K. Yuan University of Chinese Academy of Science, Beijing 100049, People’s Republic of China

123

2590

Nat Hazards (2015) 75:2589–2605

an average increasing rate of 11.18 cm a-1. Meanwhile, changes in permafrost temperature under asphalt pavement were different from that under natural ground. The warming rate in permafrost under asphalt pavement at 6, 10 and 15 m depths was 0.024, 0.022 and 0.02 °C a-1, respectively, while the three values under natural ground were 0.016, 0.013 and 0.013 °C a-1. From these results above, it can be concluded that influences from climate warming on permafrost degradation in warm permafrost region were greater than that in cold permafrost region, and influences from engineering disturbance on permafrost degradation in warm permafrost region were less than that in cold permafrost region. Keywords Qinghai–Tibet Highway (QTH)  Permafrost degradation  Active layer thickness  Permafrost temperature

1 Introduction The Qinghai–Tibet Plateau (QTP) is the highest and the most extensive plateau in the world (Liu and Chen 2000; Wu et al. 2011). Plateau permafrost is the most widely distributed high-altitude permafrost (Cheng and Jin 2013), and the area on the Plateau is estimated at 1.3 9 106 km2 (Nan et al. 2005). Compared with high-latitude permafrost, the plateau permafrost has the characteristics of higher ground temperature, thinner thickness and poorer thermal stability (Ma et al. 2013). Plateau permafrost is more sensitive to climate warming and engineering disturbances and thus is a sensitive indicator of climate warming (Zimov et al. 2006). According to the meteorological data by Intergovernmental Panel on Climate Change (IPCC), the global temperature is continuously raising (IPCC 2013). The years from 2001 to 2010 are the warmest 10 years since the modern meteorological measurements started at 1850 (WMO 2013), and the average global temperature reaches the highest level in the last 10,000 years since the end of Ice River stage (Shaun et al. 2013). The continuously raising of climate results in significant changes of permafrost environment and engineering properties (Osterkamp 2007; Harris et al. 2003), which causes a great reduction to distribution area of the QTP permafrost (Cheng and Wu 2007; Wu and Zhang 2010; Zhao et al. 2010). Over the past 40 years, the distribution area of permafrost on the QTP has already reduced from 1.26 9 106 to 1.5 9 106 km2 (Cheng and Jin 2013). Embankments constructed in permafrost inevitably disrupt the surface energy balance, resulting in the rise of permafrost temperature and permafrost table (Sheng et al. 2001; Wu et al. 2006). The significant rise in permafrost temperature is caused by the QTH because of the heat-absorbing effect of asphalt pavement (Wu et al. 2003, 2007a, b). From 1996 to 2007, the mean annual permafrost temperature at 6 m depth increased from 0.2 to 0.96 °C, with an average of 0.44 °C during the past 12 years (Wu and Niu 2013). And the artificial permafrost table increased from 0.16 to 2.6 m, with an average of 1.67 m (Wu et al. 2010a, b). Based on the monitoring data of 13 typical monitoring stations, the long-term permafrost degradation characteristics beneath asphalt pavement and natural ground were analyzed simultaneously by the freezing–thawing time, freezing–thawing depth and permafrost temperature. Herein, the effects of climate warming and engineering disturbance to thickness and temperature of permafrost in different permafrost regions are investigated. These results were used to providing more reliable long-term sequence for permafrost degradation at Qinghai–Tibet Engineering Corridor.

123

Nat Hazards (2015) 75:2589–2605

2591

2 Data and method The 13 monitoring sites extend from Xidatan, the northern boundary of permafrost distribution, to Anduo, the southern boundary of permafrost distribution, along the QTH (Fig. 1) with a distance of roughly 550 km. Each site was established with four observation boreholes, including natural borehole, roadbed borehole, left road shoulder borehole and right road shoulder borehole. Soil temperatures from all boreholes were measured at 0.5–10/15 m. All measurements were made using a string of thermistors with an interval of 0.5 m in depth. The thermistors type is 105 T, and its observation accuracy is 0.01 °C, which were connected to CR1000 data acquisition instrument made by Campbell Company. These thermistors were made by the State Key Laboratory of Frozen Soil Engineering, Chinese Academy of Science. The data were collected twice a month, in the beginning and middle of each month. These 13 sites were established earliest in 1995 and latest in 2003; thus, the observation durations of them vary from 9 to 17 years. Table 1 provides detailed information on geographical locations and observation period at each site. In the following analyses, the permafrost table is defined as the maximum seasonal thaw depth, or it means the deepest depth of 0 °C isotherm penetrated. The natural permafrost table is formed on natural conditions, and the artificial permafrost table is formed under human engineering activities. And the permafrost regions are divided by the mean annual ground temperature (MAGT). The MAGT of warm permafrost regions is higher than -1.0 °C, and the MAGT of cold permafrost regions is lower than -1.0 °C (Liu and Zhang 2012).

3 Results 3.1 Air temperature increase on the QTP Permafrost is the product of the heat exchange between ground and atmosphere, and its thermal regime is enslaved to geological, geography and climate. Thus, permafrost has an extensive response and feedback to climate changes. Based on the climatologically data from 1981 to 2010 (Fig. 2), the linear rate of climate is 0.04–0.05 °C (Wang et al. 2012; Yin et al. 2012; Li and Wu 2005). The continuously raising of air temperature results in significant increasing of permafrost temperature, which causes a significant degradation of permafrost on the QTP. 3.2 Characteristics of permafrost freezing–thawing process along the QTH 3.2.1 Shallow soil temperature Embankment constructed in permafrost disrupted the surface energy balance because of the changes of surface micro-topography, vegetations and thermo-physical of shallow soil layers. As a consequence, the permafrost temperature and thermal regime were changed under asphalt pavement. Shallow soil temperature at 1.0 m depth showed a significant difference between asphalt pavement and natural ground because of the heat-absorbing effect and impermeability (Figs. 3, 4). The annual amplitude of shallow soil temperature under asphalt pavement was greater than that under natural ground. In warm seasons,

123

2592

Nat Hazards (2015) 75:2589–2605

Fig. 1 Monitoring sites along the Qinghai–Tibet Highway

Table 1 Geographical data and information of 13 monitoring sites on the Qinghai–Tibet Plateau Sites

Areas

Longitude

Latitude

Altitude (m)

MAGT (°C)

Observation period

XT

Xidatan

35°430 0500 N

94°040 2900 E

0

4,516

-0.51

2003–2011

KL

Kunlun Mts.

35°39 08 N

94°030 2800 E

4,724

-2.01

2003–2011

XS

Xieshui River

35°310 2200 N

93°450 2600 E

4,588

-1.49

2002–2011

CM

Chumaer River

35°230 4900 N

93°320 0100 E

4,490

-0.56

1996–2011

KX

Kekexili

35°120 1600 N

93°060 3800 E

4,716

-1.62

1995–2011

FH

Fenghuo Mts.

34°410 2400 N

92°530 3000 E

4,900

-3.21

1995–2011

TR

Tuotuo River

33°520 4800 N

92°130 4800 E

4,572

-0.03

2003–2011

KM

Kaixin Mts.

33°570 2100 N

92°200 2300 E

4,627

-0.82

2003–2011

TG

Tanggula Mts.

32°420 3000 N

91°520 1600 E

4,951

-1.20

2003–2011

ZR

Zhajiazangbu River

32°300 2800 N

91°320 0300 E

5,002

-1.11

1998–2011

TJ

Touerjiu Mts.

32°290 3300 N

91°490 1700 E

5,077

-0.79

2002–2011

JR

Jiebuqu River

32°250 1700 N

91°440 4500 E

4,900

-0.15

1998–2011

AD

Anduo

32°230 0000 N

91°420 2700 E

4,800

-0.16

2002–2011

00

shallow soil temperature under asphalt pavement was higher than that under natural ground because of the strongly heat-absorbing effect. However, shallow soil temperature under asphalt pavement was lower in cold seasons. The reasons were that the air temperature was lower with the reducing of solar radiation strength, which caused a weaker effect of heat absorbing. Meanwhile, the faster snow melting and water evaporation on asphalt pavement resulted in shallow soil temperature lower because of the heat dissipation effect of asphalt pavement.

123

Nat Hazards (2015) 75:2589–2605

2593

Fig. 2 Climate change in QTP from 1980 to 2010 (Li and Wu 2005)

Fig. 3 Change of shallow soil temperature in warm permafrost regions

3.2.2 Thawing indexes and freezing indexes The freezing and thawing indexes are defined as the cumulative number of degree days for a given time period. In this paper, the freezing/thawing indexes are decided as the sum of the daily air temperature exceeding/falling below 0 °C, and they can be computed based on the following equations (Jiang et al. 2008):

123

2594

Nat Hazards (2015) 75:2589–2605

Fig. 4 Change of shallow soil temperature in cold permafrost regions

MAFI ¼

Zt1 t0

jT j dt;



T\0 C;

MATI ¼

Zt1 T dt;



T [ 0 C:

ð1Þ

t0

Thawing indexes and freezing indexes are important indices for evaluation on distribution area of permafrost and seasonally frozen ground (King et al. 2006). The regions are covered by permafrost when thawing indexes were lower than freezing indexes. And the regions are covered by continuous permafrost with smaller thawing indexes and bigger freezing indexes (Frauenfeld et al. 2007). Table 2 shows the statistical result of shallow soil temperature, MAFI and MATI at 0.5 m depth under asphalt pavement and natural surface. The MAFI and MATI of asphalt pavement are different from that of natural surface because of the heat-absorbing effect of asphalt pavement. Under asphalt pavement, the MATI were 895–2,540 °C days and the MAFI were 290–1,097 °C days. Under natural surface, the MATI were 144–1,550 °C days and the MAFI were 127–1,554 °C days. The shallow soil temperature under asphalt pavement and natural surface is different in different seasons (Table 2). In warm seasons, the shallow soil temperature under asphalt pavement was 0.76–8.58 °C higher than that under natural surface. And in cold seasons, the shallow soil temperature under asphalt pavement were 0.22–4.19 °C lower than that under natural surface. The reasons were explained in Sect. 3.2.1. The temperature ratio is defined as the specific value of temperature in warm seasons and that in cold seasons, which means the characteristics of annual fluctuation. The annual fluctuation under asphalt pavement is bigger than that under natural surface (Table 2). The temperature ratio under asphalt pavement was 1.23–2.18 and that under natural surface was 0.42–1.24.

123

Nat Hazards (2015) 75:2589–2605

2595

Table 2 Shallow soil temperatures, MATIs and MAFIs along the QTH Areas

XS CM FH TR KM TG ZR TJ JR AD

Shallow soil temperature in cold seasons (°C)

MAFI (°C days)

Temperature ratio

Shallow soil temperature in warm seasons (°C)

MATI (°C days)

Asphalt pavement

8.11

1,817

-5.19

-760

1.56

Natural surface

3.24

504

-4.97

-1,066

0.65

Asphalt pavement

9.72

2,540

-4.95

-1,070

1.96

Natural surface

4.28

800

-4.66

-904

0.92 1.23

Location

Asphalt pavement

7.50

1,472

-6.08

-1,052

Natural surface

2.73

364

-6.47

-1,544

0.42

Asphalt pavement

9.95

2,407

-4.95

-616

2.01

Natural surface

7.85

1,550

-6.53

-1,151

1.20

Asphalt pavement

9.91

2,262

-6.07

-866

1.63

Natural surface

2.52

381

-1.88

-433

1.34

Asphalt pavement

8.89

1,849

-6.85

-1,097

1.30

Natural surface

2.05

292

-2.82

-669

0.73

Asphalt pavement

7.97

1,621

-5.15

-825

1.55

Natural surface

3.63

560

-4.87

-1,089

0.75 2.18

Asphalt pavement

3.97

895

-1.82

-290

Natural surface

3.21

577

-3.85

-774

0.83

Asphalt pavement

8.74

1,946

-4.70

-590

1.86

Natural surface

3.87

695

-3.01

-580

1.29

Asphalt pavement

9.81

2,430

-3.41

-427

2.88

Natural surface

1.23

144

-0.51

-127

2.41

3.2.3 Characteristics of Thawing–freezing process From 1998 to 2011, the start date of thawing was continuous brought forward and that of freezing was continuous delayed in past 14 years, which caused a significant increase in permafrost melt days (Fig. 5). Under the effect of climate warming, the thawing periods were 5.5 months and the freezing periods were 6.5 months, with a difference value of 1 month in warm permafrost regions. And in cold permafrost regions, the thawing periods were 4.5 months and the freezing periods were 7.5 months, with a difference value of 3 months. Under the mutual effect of climate warming and engineering disturbance, the thawing periods were 7–8 months and the freezing periods were 4–5 months, with a difference value of 3 months in warm permafrost regions. And in cold permafrost regions, the thawing periods were 7 months and the freezing periods were 5 months, with a difference value of 2 months. The thawing periods were brought forward 1–2 months in warm permafrost regions and that were brought forward 2.5 months in cold permafrost regions, which was caused by engineering disturbance. The depth of the maximum seasonal penetration of the 0 °C isotherm was also continuous increased in past 14 years (Fig. 5) both under asphalt pavement and natural surface, and the increased amplitude of maximum seasonal thawing penetration depth was 1–2 m. But the depth under asphalt pavement was deeper than that under natural surface. In warm permafrost regions, the depth under asphalt pavement was over 8 m and was \6 m under natural surface, with a difference value of exceeding 2 m. In cold

123

2596

Nat Hazards (2015) 75:2589–2605

a

b

c

d

Fig. 5 Thawing–freezing process of permafrost in different permafrost regions. a Thawing–freezing process of permafrost in region of Jiebuqu River (in warm permafrost region), b thawing–freezing process of permafrost in region of Chumaer River (in warm permafrost region), c thawing–freezing process of permafrost in region of Zhajiazangbu River (in cold permafrost region), d thawing–freezing process of permafrost in region of Kekexili region (in cold permafrost region)

permafrost regions, the depth under asphalt pavement was 5–8 m and was \4 m under natural surface, with a difference value of 1–4 m. The maximum seasonal thawing penetration depth continuous increasing resulted in detachment of frozen ground under embankment because of the climate warming and engineering disturbance, which caused residual thawed layers. The residual thawed layers lied in the depth of 5–11 m under asphalt pavement in warm permafrost regions and showed the expanding trend (Fig. 5a). 3.3 Changes characteristics of permafrost table along the QTH 3.3.1 Depth changes of permafrost table ALT on the QXP has a sensitive response to climate warming, and it changes consistently with changes in air temperature. Continuous soil temperature monitoring data from 13 sites

123

Nat Hazards (2015) 75:2589–2605

2597

Fig. 6 Depth changes of permafrost table in different permafrost regions. a Depth change of artificial permafrost table, b depth change of natural permafrost table

indicated that ALT experienced a significant inter-annual variation 1995–2011 (Fig. 6). Under the effect of climate warming, the freezing and thawing process within the active layer showed predominant change, which caused a continuous decrease in the depth of permafrost table. In cold permafrost regions, the decreasing amplitude of permafrost table ranged from 0.1 to 1.06 m, with mean decreasing amplitude of 0.44 m. In warm permafrost regions, the decreasing amplitude of permafrost table ranged from 0.14 to 1.5 m, with mean decreasing amplitude of 0.68 m. The decreasing amplitude of permafrost table was greater after embankment constructed because of the engineering disturbance and heatabsorbing effect. Under the mutual effect of climate warming and engineering disturbance, the decreasing amplitude of permafrost table ranged from 0.97 to 2.32 m in cold permafrost regions and ranged from 0.34 to 2.22 m in warm permafrost regions. And under the effect of engineering disturbance, the decreasing amplitude of permafrost table ranged from 0.34 to 2.22 m, with mean decreasing amplitude of 1.38 m in cold permafrost regions. In warm permafrost regions, the decreasing amplitude of permafrost table ranged from 0.55 to 2.24 m, with mean decreasing amplitude of 0.68 m. Comparing with the cold permafrost regions, the effect extent of climate warming in warming permafrost regions was higher, and the effect extent of engineering disturbance in warming permafrost regions was lower. The differences of decreasing amplitude of permafrost table in different permafrost regions were related to the degradation process of permafrost. The degradation of permafrost would be downward while the maximum season thawing penetration depth exceeding the maximum season freezing penetration depth, which caused detachment of

123

2598

Nat Hazards (2015) 75:2589–2605

frozen ground and residual thawed layers. Geothermal gradients between permafrost and surrounding melting soil layer resulted in changes of permafrost degradation. Permafrost was melting to laterals because the geothermal gradients of permafrost were lower (Jin et al. 2006). The correlation expressions of permafrost table depth with time were given in Table 3. The degree of correlation under asphalt pavement was different from that under natural surface. Under asphalt pavement, the depth decreasing rate of artificial permafrost table ranged from 11.68 to 20.14 cm a-1, with an average rate of 16.39 cm a-1. Under natural surface, the depth decreasing rate of natural permafrost table ranged from 1.82 to 12.88 cm a-1, with an average rate of 4.67 cm a-1. The depth decreasing rate of artificial permafrost table was 4.6 times than that of natural permafrost table. The decreasing rates of permafrost table were differences in different regions (Fig. 7). Under the effect of climate warming, the average decreasing rate of natural permafrost table was 5.72 cm a-1 in warm permafrost regions and was 3.42 cm a-1 in cold permafrost regions. The average decreasing rate of artificial permafrost table in warm and cold permafrost regions was 16.9 and 15.7 cm a-1, respectively. Under the effect of engineering disturbance, the average decreasing rate was 11.18 cm a-1 in warm permafrost regions and was 12.28 cm a-1 in cold permafrost regions. In warm permafrost regions, the effect on permafrost caused by climate warming was greater, comparing to that caused by engineering disturbance that was lower. 3.3.2 Temperature changes of permafrost table The temperature at the top of permafrost (TTOP) is an important indicator of the thermal regime of the active layer (Li et al. 2012). TTOP at 13 sites along the QTH exhibited an increasing trend (Figs. 8, 9), but the factors of surface vegetation, soil moisture content, surface color and so on had considerable influence on TTOP each other. Therefore, the combined effects of larger thermal conductivity and smaller surface albedo increased the variation rate of TTOP. The TTOP under asphalt pavement was higher than that under natural surface because of the changes in ground surface and heat-absorbing effect after embankment constructed. The maximum difference values of TTOP changes between asphalt pavement and natural surface were 1.2–2.3 °C (Fig. 8) in warm permafrost regions and that were 3.5–6.0 °C (Fig. 9) in cold permafrost regions. Because of the heatabsorbing effect of asphalt pavement, it had influenced not only the TTOP but also annual average ground temperature of permafrost. In different permafrost regions, the differences of influences caused by asphalt pavement were different. In cold permafrost region, at the depth of natural permafrost table, the soil under the asphalt pavement was still in freezing state. It could maintain the dynamic balance of freezing–thawing process. But in warm permafrost region, the soil was in thawing state. It needed to be continuously decreased to maintain the dynamic balance. 3.4 Variations of permafrost temperature along the QTH Permafrost temperature at the monitoring sites indicated that permafrost temperature at 6 and 10 m both experienced a significant inter-annual variation from 1995 to 2011 (Figs. 10, 11, 12, 13). At the 13 sites, permafrost temperature showed an increasing trend, but the inter-annual fluctuations of MAGT were difference in different permafrost depth and surface conditions. Generally speaking, at the same depth, the inter-annual fluctuations of MAGT under asphalt were higher than that under natural surface. And in the same

123

Y = 0.1473x - 287.85

Y = 0.2014x - 398.47

Y = 0.1559x - 301.93

ZR

TJ

JR

Y = 0.2019x - 396.83

Y = 0.1915x - 377.26

CM

TG

Y = 0.1168x - 226.20

Y = 0.1324x - 259.55

XT

XS

Regression equation of artificial permafrost table

Areas

0.9379

0.9510

0.8922

0.8369

0.9628

0.8767

0.9120

Correlation coefficient (R2)

0.1559

0.2014

0.1473

0.1915

0.2019

0.1324

0.1168

Decreasing rate

Table 3 Correlation expressions of permafrost table depth (m) associated with times (a)

Y = 0.1288x - 253.36

Y = 0.0203x - 38.021

Y = 0.0234x - 44.222

Y = 0.0379x - 74.032

Y = 0.0597x - 116.05

Y = 0.0178x - 34.013

Y = 0.0167x - 30.404

Regression equation of artificial permafrost table

0.9460

0.2442

0.4299

0.8333

0.7211

0.2568

0.2536

Correlation coefficient (R2)

0.1288

0.0330

0.0246

0.0379

0.0488

0.0402

0.0182

Decreasing rate

Nat Hazards (2015) 75:2589–2605 2599

123

2600

Nat Hazards (2015) 75:2589–2605

Fig. 7 Degradation rate of permafrost table under asphalt pavement and natural surface

Fig. 8 Changes of permafrost temperature in warm permafrost regions

surface conditions, the inter-annual fluctuations of MAGT at 6 m depth were higher than that at 10 m depth. The response of permafrost to climate warming and to engineering disturbance had an obvious difference. Engineering disturbance was short-term and quick-effect processes on permafrost, whereas climate warming was a long-term and slow-effect process on permafrost. Permafrost temperature showed an increasing trend at all the observed sites, but the changes in permafrost temperature under asphalt pavement were different from that under natural ground. The warming rate in permafrost under asphalt pavement at 6, 10 and 15 m depths was 0.024, 0.022 and 0.02 °C a-1, respectively. And the three values under natural ground were 0.016, 0.013 and 0.013 °C a-1.

123

Nat Hazards (2015) 75:2589–2605

2601

Fig. 9 Changes of permafrost temperature in cold permafrost regions 2 y = 0.0256x - 50.026 R² = 0.2669

1.5

XT

Temperature(°C)

1 y = 0.042x - 83.689 R² = 0.3781

0.5 0

y = 0.0213x - 42.486 R² = 0.57

XS FH

y = 0.0536x - 107.34 R² = 0.8665

y = 0.0282x - 56.658 R² = 0.877

-0.5

TG ZR

y = 0.017x - 35.44 R² = 0.1404

-1

JR

-1.5 -2 1995

asphalt pavement 1997

1999

2001

2003

2005

Year(a)

2007

2009

2011

2013

Fig. 10 MAGT at 6 m depth under asphalt pavement

In different permafrost regions, the increasing trends of permafrost temperature at the same depth were differences (Figs. 14, 15). At the depth of 6, 10 and 15 m under asphalt pavement, the mean increasing rate was 0.018, 0.025 and 0.019 °C a-1 in warm permafrost regions. And the three mean increasing rates were 0.032, 0.019 and 0.02 °C a-1 in cold permafrost regions. At the depth of 6, 10 and 15 m under natural surface, the mean

123

2602

Nat Hazards (2015) 75:2589–2605 0.4 0.2

y = 0.0302x -60.396 R² = 0.8152

Temperature(°C)

0

XT

y = 0.013x -26.172 R² = 0.8572

XS

-0.2

y = 0.0225x -45.501 R² = 0.9576

CM

-0.4 -0.6

KM

y = 0.0271x -54.662 R² = 0.9878

ZR

y = 0.0262x -53.029 R² = 0.9153

y = 0.0159x -32.776 R² = 0.9349

-0.8

JR

asphalt pavement -1 1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

Year(a) Fig. 11 MAGT at 10 m depth under asphalt pavement

Fig. 12 MAGT at 6 m depth under natural surface

increasing rate was 0.016, 0.014 and 0.012 °C a-1 in warm permafrost regions. And the three mean increasing rates were 0.016, 0.012 and 0.014 °C a-1 in cold permafrost regions.

4 Discussions and conclusions The long-term thermal effect of asphalt pavement on permafrost is influenced not only by engineering but also by climate changes. However, it is difficult to determine the

123

Nat Hazards (2015) 75:2589–2605

2603

0.4 y = 0.0075x - 15.221 R² = 0.8906

Temperature(°C)

0

XT

-0.4

y = 0.0238x - 48.384 R² = 0.9587

-0.8

-1.2

KM

y = 0.0143x - 29.508 R² = 0.9079

TG

y = 0.0102x - 21.514 R² = 0.4197

y = 0.0215x - 44.26 R² = 0.8922

-1.6

y = 0.0099x - 21.599 R² = 0.8891

natural surface -2 1995

XS

1997

1999

2001

2003

2005

2007

2009

2011

ZR JR

2013

Year(a) Fig. 13 MAGT at 10 m depth under natural surface

Fig. 14 The mean increasing rate of permafrost temperature at the different depths under natural surface

Fig. 15 The mean increasing rate of permafrost temperature at the different depth under asphalt pavement

123

2604

Nat Hazards (2015) 75:2589–2605

contributions of climate change and thermal effect of asphalt pavement (Wu et al. 2010a, b). At the same time, there is a large difference in response of warm and cold permafrost to engineering disturbance and climate warming due to phase changes of ground ice (Wu et al. 2007a, b). Under the effect of engineering disturbance and climate warming, ALT and permafrost temperature increased and had extensive temporal and spatial differences along the QTH. Based on the temperature data at 0.5 m depth of 13 monitoring sites from 1995 to 2011, changes in permafrost temperatures under asphalt pavement were different from that under natural surface. In warm season, the average temperatures of asphalt pavement were 0.76–8.58 °C higher than that of natural ground. While in cold season, the average temperatures of asphalt pavement were 0.22–4.19 °C lower than that of natural ground. The MATI and MAFI of asphalt pavement ranged from 895–2,540 to 290–1,097 °C days, respectively. The MATI and MAFI of natural ground ranged from 144–1,550 to 127–1,544 °C days, respectively. Because of the effect of asphalt pavement heat absorbing, the ALT in the permafrost exhibited a remarkable increasing trend in the past 17 years. The decreasing amplitude in depth of natural permafrost table ranged from 0.1 to 1.06 m in cold permafrost and that ranged from 0.14 to 1.5 m in warm permafrost regions. The mean decreasing rate in depth of natural permafrost table ranged from 1.82 to 12.88 cm a-1, with an average rate of 4.67 cm a-1. The decreasing amplitude in depth of artificial permafrost table ranged from 0.97 to 2.32 m in cold permafrost regions and that ranged from 0.34 to 2.22 m in warm permafrost regions. The mean decreasing rate in depth of artificial permafrost table ranged from 11.68 to 20.14 cm a-1, with an average rate of 16.39 cm a-1. Permafrost temperature at the monitoring sites indicated that permafrost temperature at 6, 10 and 15 m experienced a significant inter-annual variation from 1995 to 2011. The warming rate in permafrost under asphalt pavement at 6, 10 and 15 m depths was 0.024, 0.022 and 0.02 °C a-1, respectively. And the three values under natural ground were 0.016, 0.013 and 0.013 °C a-1. In present paper, long-term characteristics of permafrost temperature increase and permafrost table decline were investigated under the effects of climate warming and engineering disturbance, but without the consideration of local factors on these processes. Further studies are required into the permafrost change process under the effects of local factors change. At the time, to protect the permafrost beneath roadway, there were many cooling measures, such as ventilation pipes, heat pipes, block-stone embankment and so on, used in the QTH and the Qinghai–Tibet Railway. However, it needs to study more on the effects of these cooling measures as the air temperature continuously rising. Acknowledgments This work was support by the National Key Basic Research Program of China (973 Program) (No. 2012CB026106), National Key Technology Support Program (No. 2014BAG05B01), West Light Program for Talent Cultivation of Chinese Academy of Sciences (for doctor Y. Mu).

References Cheng G, Jin H (2013) Groundwater in the permafrost regions on the Qinghai–Tibet Plateau and its changes. Hydrogeol Eng Geol 40:1–11 (in Chinese) Cheng G, Wu T (2007) Responses of permafrost to climate change and their environmental significance, Qinghai–Tibet Plateau. J Geophys Res 112(F2). doi:10.1029/2006JF000631 Frauenfeld OW, Zhang T, Mccreight JL (2007) Northern Hemisphere freezing/thawing index variations over the twentieth century. Int J Climatol 27(1):47–63. doi:10.1002/joc.1372

123

Nat Hazards (2015) 75:2589–2605

2605

Harris C, Vonder Mu¨hllb D, Isaksen K et al (2003) Warming permafrost in European mountains. Glob Planet Change 39:215–225. doi:10.1016/j.gloplacha.2003.04.001 IPCC (2013) Working group I contribution to the IPCC fifth assessment report (AR5). Climate change 2013: the physical science basis. Final Draft Underlying Scientific-Technical Assessment [R/OL] Jiang F, Hu R, Li Z (2008) Variations and trends of the freezing and thawing index along the Qinghai– Xizang Railway for 1966–2004. J Geogr Sci 18:3–16. doi:10.1007/s11442-008-0003-y Jin H, Zhao L, Wang S, et al (2006) Ground temperature characteristics and degradation style about permafrost along the Qinghai-Tibet Highway. Sci China Ser D Earth Sci 36(11):1009–1019 (in Chinese) King L, Herza T, Hartmanna H et al (2006) The PACE monitoring strategy: a concept for permafrost research in Qinghai–Tibet. Quat Int 154(155):149–157 Li S, Wu T (2005) The relationship between air temperature and ground temperature in the Tibet Plateau. J Glaciol Geocryol 27:627–631 (in Chinese) Li R, Zhao L, Ding Y et al (2012) Temperature and spatial variations of the active layer along the QinghaiTibet Highway in a permafrost region. Chin Sci Bull 57(30):2864–2871 Liu X, Chen B (2000) Climatic warming in the Tibetan Plateau during recent decades. Int J Climatol 20:1729–1742 Liu S, Zhang J (2012) Review on physic-mechanical properties of warm frozen soil. J Glaciol Geocryol 34:120–129 (in Chinese) Ma W, Mu Y, Li G et al (2013) Responses of embankment thermal regime to engineering activities and climate change along the Qinghai–Tibet Railway. Sci Sin Terrae 43:478–489 (in Chinese) Marcott SA, Shakun JD, Clark PU et al (2013) A reconstruction of regional and global temperature for the past 11,300 years. Science 339:1198–1201. doi:10.1126/science.1228026 Nan Z, Li S, Cheng G (2005) Prediction of permafrost distribution on Qinghai–Tibet Plateau in the next 50 and 100 years. Sci China Ser D Earth Sci 48:797–804 Osterkamp TE (2007) Characteristics of the recent warming of permafrost in Alaska. J Geophys Res Earth Surf (2003–2012), 112:F02S02. doi:10.1029/2006JF000578 Sheng Y, Zhang J, Liu Y et al (2001) Thermal regime in the embankment of Qinghai–Tibet Highway in permafrost regions. Cold Reg Sci Technol 35:35–44 Wang P, Tang G, Cao L et al (2012) Surface air temperature variability and its relationship with altitude and latitude over the Tibet Plateau in 1981–2010. Progress Inquisitiones DE Mutatione Climatis 8:313–319 (in Chinese) World Meteorological Organization (2013) The global climate 2001–2010: a decade of climate extremessummary report, WMO-No. 1103 Wu Q, Niu F (2013) Permafrost changes and engineering stability in Qinghai–Xizang Plateau. Chin Sci Bull 58. doi:10.1007/s11434-012-5587-z Wu Q, Zhang T (2010) Changes in active layer thickness over the Qinghai–Tibetan Plateau from 1995 to 2007. J Geophys Res Atmos 115:D09107. doi:10.1029/2009JD012974 Wu Q, Shi B, Liu Y (2003) Study on interaction of permafrost and highway along the Qinghai–Xizang Highway. Sci China (Ser D) 46:97–105 Wu Q, Cheng G, Ma W et al (2006) Technical approaches on permafrost thermal stability for Qinghai–Tibet Railroad construction. Int J Geomech Geoengin 1:119–127 Wu Q, Dong X, Liu Y (2007a) Responses of permafrost on the Qinghai–Tibet Plateau, China, to climate change and engineering construction special section. Arct Antaret Alp Res 39:682–687 Wu Q, Liu Y, Yu H (2007b) Analysis of the variation of permafrost under ordinary embankment along the Qinghai–Tibet Railway. J Glaciol Geocryol 29:960–968 (in Chinese) Wu Q, Zhang T, Liu Y (2010a) Permafrost temperatures and thickness on the Qinghai–Tibet Plateau. Glob Planet Chang 72:32–38 Wu Q, Zhang Z, Liu Y (2010b) Long-term thermal effect of asphalt pavement on permafrost under embankment. Cold Reg Sci Technol 60:221–229 Wu Q, Zhang T, Liu Y (2011) Thermal state of the active layer and permafrost along the Qinghai–Xizang (Tibet) Railway from 2006 to 2010. Cryosphere Discuss 5:2465–2481 Yin Y, Wu S, Zhao D et al (2012) Impact of climate on actual evapotranspiration on the Tibetan Plateau during 1981–2010. Acta Geogr Sin 67:1471–1481 (in Chinese) Zhao L, Wu Q, Marchenko SS et al (2010) Thermal state of permafrost and active layer in Central Asia during the International Polar Year. Permafr Periglac Process 21:198–207 Zimov SA, Schuur EA, Chapin FS (2006) Permafrost and the global carbon budget. Science 312:1612–1613

123