Degradation of permafrost in the Xing'anling Mountains, northeastern China

PERMAFROST AND PERIGLACIAL PROCESSESPermafrost and Periglac. Process. 18: 245–258 (2007)Published online 7 June 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/ppp.589
Degradation of Permafrost in the Xing’anling Mountains,Northeastern China

Huijun Jin,1* Qihao Yu ,1 Lanzhi Lu ,1 Dongxin Guo ,1 Ruixia He ,1 Shaopeng Yu ,1,2 Guangyou Sun 3,4

and Yingwu Li 1,5

1 State Key Laboratory of Frozen Soils Engineering, Cold and Arid Regions Environmental and Engineering ResearchInstitute, Chinese Academy of Sciences, Lanzhou, China2 Department of Geography, Harbin College, Harbin, China3 Northeast Institute of Geography and Agro-Ecology, Chinese Academy of Sciences, Changchun, China4 Institute of Resources and Environments, Capital Normal University, Beijing, China5 Railway Permafrost Observatory at Yituli’he, Inner Mongolia Autonomous Region, China

* CoFroztal aScienE-ma

Copy

ABSTRACT

Permafrost in the Da and Xiao Xing’anling Mountains in northeastern China is warm, thin and sensitiveto climatic warming. In the 1970s, the southern limit of permafrost (SLP) was empirically correlated tothe �1 to 08C isotherms of mean annual air temperature (MAAT) in the western part of the DaXing’anling Mountains, to about 08C in the northern part of the Songnen Plain, and to 0 to þ18C in theeastern part of the Xiao Xing’anling Mountains. Climate warming and deforestation have led topermafrost degradation as shown by deepening of the active layer, thinning permafrost, rising groundtemperatures, expanding taliks and the disappearance of permafrost patches. The present position of theSLP was estimated using the �1.0 to þ1.08C MAAT isotherms for 1991–2000. Compared to the SLP inthe 1970s, areas of sporadic discontinuous and isolated patchy permafrost have decreased by90,000–100,000 km2, or 35–37% of their total areal extent (260,000–270,000 km2) in the 1970s.Recent field observations along the Hei’he to Bei’an Highway, the proposed Mo’he to Daqing CrudeOil Pipeline route and the Hai’lar to Daqing Highway confirm these changes. Continuing northwardshifting of the SLP is likely to occur during the next 40–50 years under a warming of 1.0–1.5 8C,reducing the permafrost areal extent to an estimated 35% of that in the 1970s and 1980s. Copyright #2007 John Wiley & Sons, Ltd.

KEY WORDS: Xing’anling Mountains; southern limit of permafrost; permafrost degradation; climatic warming;

deforestation

INTRODUCTION

Research on permafrost in the Da and XiaoXing’anling Mountains (together known as the

rrespondence to: Huijun Jin, State Key Laboratory ofen Soils Engineering, Cold and Arid Regions Environmen-nd Engineering Research Institute, Chinese Academy ofces, 326 West Donggang Road, Lanzhou 730000, China.il: [email protected]

right # 2007 John Wiley & Sons, Ltd.

Xing’anling Mountains) in northeastern China beganin the early 1950s. During the 1960s and 1970s,numerous regional investigations together with localstudies for specific purposes, such as water supply,road construction or coal-mining, were conducted inorder to meet the needs of economic development,engineering design and construction, and environ-mental management (Zhou and Wu, 1965; Guo and Li,1981; Guo et al., 1981; Dai, 1982; Zhou and Guo,1982; Wang, 1982; Northeastern China Permafrost

Received 30 June 2006Revised 5 March 2007

Accepted 5 March 2007

246 H. Jin et al.

Research Taskforce, 1983; Guo and Huang, 1989;Peng and Cheng, 1990; Lu et al., 1993; Tong, 1993;Gu and Zhou, 1994; Zhou et al., 1996).

Degradation of permafrost in northeast China hasbeen of considerable concern during the past decade(Gu and Zhou, 1994; Chen and Yin, 1996; Zhou et al.,1996; Wang et al., 1996). Many major engineering andenvironmental projects, such as the proposed China-Russia Crude Oil Pipeline from Skovorodino, Russiato Daqing, China, a hydropower project near theHeilongjiang (Amur) River, express highways andhigh-speed railways, necessitate an understanding ofthe status and the future of permafrost in theXing’anling Mountains over the next 50–100 years.This paper reviews the evidence and reasons forpermafrost degradation, and offers a preliminaryassessment of the present status and the future trendsof permafrost conditions within the Xing’anlingMountains.

STUDY REGION

The Da and Xiao Xing’anling Mountains are locatedin the northernmost (468000 to 538330N) part ofnortheastern China, and extend for about 900 km fromsouth to north, and 1400 km from east to west(Figure 1). The Da Xing’anling Mountains range inelevation from 1000 to 1400–1500 m in the middlesection from A’ershan to Yi’ershi and to the source ofthe Zhuo’er River. They then decrease northward, to500–600 m in the section from Mangui to Gulian.Large quantities of eroded debris from the steepereastern slopes have been carried by deeply incisedrivers to the Nenjiang River Plain which lies betweenthe two mountain ranges. In contrast, the XiaoXing’anling Mountains are gentler in topography,with meandering river channels, elevations of 500–600 m, and few peaks higher than 800 m.

The Xing’anling Mountains comprise Late Palaeo-zoic granites, and Cenozoic volcanic formations, withsporadic Palaeozoic metamorphic and sedimentaryrocks. Since the onset of the Quaternary, theXing’anling Mountains have been slowly and differ-entially uplifting, and have been subject to long-termerosion and leveling. The slope deposits on the upperparts of the mountains are predominately angulargravels or sandy clays with gravels, generally only1–2 m in thickness, and with a layer of humic topsoil0.1–0.2 m in thickness. However, in piedmont areas,intermontane basins and valleys, deposits from debrisflows, fluvial and slope processes are up to 10–15 mthick. In typical soil profiles, peat and humic soils areunderlain by clayey sands and sandy clays with sandsand gravels.

Copyright # 2007 John Wiley & Sons, Ltd.

The region is characterised by a temperate continentalmonsoon climate with long, dry, cold winters and short,moist, hot summers. Mean annual air temperatures(MAATs) range from 0 to þ18C in the south to �5 to�68C in the north, and the range of monthly average airtemperature similarly increases from 358C to 528C. Theaverage annual precipitation declines from 500–700 mm in the southeast to less than 200 mm in thenorthwest. Rainfall in summer accounts for 80–90% ofthe annual precipitation. MAATs in the Xing’anlingMountains are 1–38C and 3–58C cooler respectivelythan in the adjacent Xinjiang Uygur AutonomousRegion of China and the Republic of Mongolia atsimilar latitudes. This explains the southward protrusionof the Eurasian southern limit of permafrost (SLP) intothis region (Guo et al., 1981; Northeastern ChinaPermafrost Research Taskforce, 1983).

The prevailing Siberian High Pressure in wintercauses an extensive atmospheric temperature inversion(Nekrasov and Klimovskii, 1978) which affects thedevelopment and distribution of permafrost in theregion. Within the Xing’anling Mountains, the depth ofthe inversion ranges from 500 to 1000 m and its thermalgradient is 108C km�1 in Mo’he, 88C km�1 in Nenjiang,and 58C km�1 in Haila’er (Meteorological Institute ofChina Meteorology Administration, 1961, 1969).

In the northern part of the Xing’anling Mountains,the vegetation comprises cold to temperate mountainneedle-leaf forest, dominated by Larix gmelinii.Needle and broad-leaf mixed forest and forest steppegradually appear when moving southwards, and thevegetation is eventually replaced by agricultural landat the southern margin of the study region. Wherepresent, dense forests, underbrush and a moss-peatmat greatly reduce insolation at the ground surface,and cause thermal offsets of 3 to 48C (Nekrasov andKlimovskii, 1978; Gu and Zhou, 1994), favouring theoccurrence and preservation of permafrost.

PRINCIPAL CHARACTERISTICSOF PERMAFROST

Permafrost displays a latitudinal zonation and strongregional variations. The areal extent increases north-wards from 0–20% to 60–75% of the terrain, and meanannual ground temperatures (MAGTs) decrease from0 to �1 to �1.5 to �2.58C, and even to as low as�4.28C. The observed thickness of permafrostgenerally varies from a few metres to 50–75 m, andcan exceed 130 m, as observed in the Gulian Coalmine(Table 1) (Northeastern China Permafrost ResearchTaskforce, 1983). Permafrost distribution ranges fromisolated patches of permafrost in the south to

Permafrost and Periglac. Process., 18: 245–258 (2007)

DOI: 10.1002/ppp

Fig

ure

1S

tud

yar

eam

apsh

ow

ing

per

maf

rost

dis

trib

uti

on

and

chan

ges

of

the

sou

ther

nli

mit

of

per

maf

rost

inth

eX

ing

’an

lin

gM

oun

tain

s.

Copyright # 2007 John Wiley & Sons, Ltd. Permafrost and Periglac. Process., 18: 245–258 (2007)

DOI: 10.1002/ppp

Permafrost Degradation in Xing’anling 247

Tab

le1

Su

mm

ary

per

maf

rost

dat

afr

om

the

Da

and

Xia

oX

ing

’an

lin

gM

ou

nta

ins

area

(rev

ised

fro

mZ

ho

uet

al.

,2

00

0).

Lo

cati

on

Lat

itu

de

(8N

)L

on

git

ud

e(8

E)

Ele

vat

ion

(m)

MA

AT

(8C

)P

erm

afro

stth

ick

nes

s(m

)M

AG

T(8

C)

and

bo

reh

ole

no

.

Lu

og

uh

e5

382

00

12

783

70

80

0N

/AN

/A�

2.0

to�

2.5

Xil

inji

(Mo

he)

5380

50

12

680

20

67

0�

5.0

66

–7

0�

1.3

Hu

ola

Bas

in5

285

70 –

5280

30

12

185

20 –

12

280

40

56

0–

74

0�

4.8

2.0

–5

0�

0.5

to�

1.0

(Eas

t)7

0–

12

0�

1.0

to�

2.9

(Wes

t)L

inzh

on

g5

285

50

N/A

70

7N

/A6

1.5

�1

.9Z

hao

hu

i5

285

20

N/A

72

6�

5.4

41

�1

.1

Am

u’e

r5

285

00

12

381

10

74

7�

5.4

62

.0,

47

–5

0�

�1

.9(C

K0

)�

1.7

(CK

39

)2

1.5

11

0–

12

0�

�0

.1(C

K3

8)

�4

.2(C

K3

7)

45

–5

0� ,

11

0–

12

0�

�1

.3(I

-IC

K2

)�

3.8

(I-I

CK

3)

2.5

km

no

rth

of

Man

gu

i5

280

20

12

280

60

88

0�

4.8

80�

12

0–

13

0�

20�

�2

.1(I

-IC

K4

),�

1.7

(II-

IIC

K1

)�

4.2

(II-

IIC

K2

)�

0.1

(II-

CK

3)

#2

Tu

nn

elin

Cu

ilin

g5

184

00

N/A

10

27

�4

.6N

/A�

1.3

Ho

ng

tu5

183

80

N/A

81

7�

4.6

N/A

�2

.9N

iu’e

r’h

e5

183

20

N/A

98

8N

/AN

/A�

3.3

Ch

un

’yo

u5

181

20

N/A

72

0N

/AN

/A�

0.7

De’

erbu

’er

5180

50

12

180

40

11

23

�5

.44

0–

54

.4�

1.1

Gen

’he

5084

10

12

183

10

98

0�

5.4

N/A

�1

.3Y

i’tu

’li’

he

5083

80

12

182

90

99

0�

5.2

40

�0

.8Ji

aged

aqi

5082

30

12

480

70

38

2�

1.4

10�

�0

.1H

ei’h

e5

081

50

12

782

70

16

7þ

0.9

0.5

–1

.5�

0.1

toþ

0.3

Wu

’erq

i’h

an4

983

30

12

182

40

70

0N

/A3

0�

�0

.4S

un

wu

4982

60

12

782

10

23

50

.21

.0–

3.0

�0

.1Y

akes

hi

4982

40

12

084

20

66

7�

2.9

N/A

�0

.2

No

te:

MA

AT¼

mea

nan

nu

alai

rte

mp

erat

ure

;M

AG

T¼

mea

nan

nual

gro

und

tem

per

ature

;N

/A¼

no

tav

aila

ble

.�

Est

imat

edth

ick

nes

so

fp

erm

afro

stb

ased

on

gro

un

dte

mp

erat

ure

mea

sure

men

ts.

Copyright # 2007 John Wiley & Sons, Ltd. Permafrost and Periglac. Process., 18: 245–258 (2007

DOI: 10.1002/ppp

248 H. Jin et al.

)


widespread discontinuous and continuous permafrostin the north.

Permafrost in the Xing’anling Mountains is affectedby local climate, geology and geography. Forexample, in the northern part, tussocks and mosslayers are dense, and unconsolidated deposits arethicker (8–12 m) on the shaded slopes of intermontanebasins, wetlands or lowlands, and low river terraces.As a result, permafrost can be as thick as 60–80 m, andoccasionally exceeds 100 m. In contrast, on treeless, orsparsely vegetated sunlit slopes with thick surficialdeposits, permafrost is either thin (<20–30 m) orabsent (Guo et al., 1981; Northeastern ChinaPermafrost Research Taskforce, 1983). On shadedslopes with denser vegetation, the reduced insolationcauses permafrost conditions intermediate betweenvalley bottom lowlands and sunlit slopes.

An important regional characteristic is that, as in theOuter Xing’anling (Stanovoy) Mountains and theTrans-Baikal of Russia, permafrost is thicker andcolder in lower topographic positions than at higherelevations (Shpolianskaya, 1978; Guo et al., 1981).This likely relates to the extensive and stable winteratmospheric air temperature inversion, thick surficialdeposits, dense vegetation and the widespread distribu-tion of wetlands in the valley bottoms and lowlands.

Figure 2 Maximum seasonal thaw depths (m) and mean annual air(YPO) at 1218200E, 508560N, 731 m and the Tuli’he Meteorological Staof YPO, with similar vegetation and topography. The dashed lines are


EVIDENCE OF PERMAFROSTDEGRADATION

Permafrost degradation in the study regions is eviden-ced by deeper active layers, thinning and warmingpermafrost and a reduction in permafrost extent.

Deeper Seasonal Thaw Penetration

During the 1960s–70s, active layer depths in theAmu’er area in the northern part of the Da Xing’anlingMountains within wetlands with 20-cm-thick mosslayers were 50–70 cm (Zhou and Wu, 1965; Dai, 1982;Zhou and Guo, 1982). During the 1978–91 period, thelatter had increased by 32 cm, and the annual averagesoil temperatures at a depth of 20 cm had increased by0.88C (Third Survey and Design Institute of RailwayMinistry of China, 1978; Gu and Zhou, 1994). Activelayer thicknesses increased further to 90––120 cmduring the 1990s (Wang et al., 1996).

In the middle part of the Da Xing’anling Mountains,thaw penetration depths declined during the early1980s at the Yituli’he Permafrost Observatory (YPO)(1218200E, 508560N, 730.5 m) in spite of risingMAATs (Figure 2). This may be due to recoveryfollowing surface disturbance during drilling. During

temperature (MAAT) (8C) at the Yituli’he Permafrost Observatorytion (1218410E, 508290N, 733 m in elevation) about 15 km southeastlinear interpolation of the missing data during 1998–2001 at YPO.


DOI: 10.1002/ppp

250 H. Jin et al.

the 1990s, maximum thaw penetration increasedslightly from 1.0 to 1.2 m between 1981 and 1989,while the MAAT changed from �5.5 to �3.08C. From1990 to 1997, maximum thaw penetration declined asa result of cooling MAATs. After 2000, however, theactive layer deepened to more than 1.8 m, whilethe MAATs declined. The reason for this is unclear butthe increase over conditions at the beginning of the1980s is substantial.

Figure 3 Permafrost temperature curves in boreholes CK3 andCK38 in the Amu’er area. The boreholes are 15 m apart and weredrilled 2–3 months before the readings, i.e. in early October; thetemperature readings were taken in late December 1979 and 1991.Ground temperatures were measured with thermistor cables, with aprecision of 0.18C.

Thinner and Warmer Permafrost

Thinning of permafrost has occurred mainly in thevalleys of medium and smaller rivers with relativelythin permafrost. For example, in the Hanjia’yuanzigoldmine areas in the lower parts of the Huma River,the base of permafrost was 5 m below the surfacebefore 1982, but by 1987 it had risen to a depth of3.8–4.0 m. As a result of human activities such asmining, urbanisation and deforestation, permafrosthad disappeared in some sections by 1995 (Wanget al., 1996). In the Xilinji Lumber Storage in theMo’he area in the northern Da Xing’anling Moun-tains, the base of permafrost was at 73 m in 1972. Aftertwo years of pumping for water supply, the base ofpermafrost had risen by 5 m (Yuan, 1989).

Boreholes CK38 and CK3 are 15 m apart at thesame elevation of 740 m on an undisturbed shadedslope in the Amu’er area. They were drilled with asmall diameter (76 mm), slowly rotating auger,without circulating liquids, in early October 1979and early October 1991, respectively. Therefore,thermal disturbances associated with the drillingshould have been minimal when the temperaturereadings were obtained in late December. Measure-ments showed that permafrost existed betweenthe depths of 5–12 m in December 1979, but onlybetween 8–10 m in December 1991 (Figure 3).Permafrost in this small area had thinned by about5 m during the 12 years from 1979 to 1991 at anaverage rate of 0.4 m / year.

Borehole no. 14 at the YPO is in an undisturbedmeadow on the first terrace of the northern bank of theYituli’he River. Tussock mats fully cover the groundsurface. Inactive ice wedges were identified within100 m of the YPO (Peng and Cheng, 1990). In theearly 1980s, permafrost may have exceeded 40 m inthickness, and the MAGT was �0.88C at a depth of20 m. Since then, the mean annual soil temperatures atshallow depths have warmed as a result of thecombined influences of rapid urbanisation, disturb-ance from road construction and use, urban heat-island


effects and climatic warming (Figure 4). Groundtemperatures at 13 m increased by about 0.28Cbetween 1984 and 1997.

Negative geothermal gradients are considered to beindicators of permafrost degradation (Wu, 1983). Insuch cases, mean annual soil temperatures declinewith depth from the permafrost table to a minimumlocated below the depth of zero annual amplitude, andthen begin to increase at greater depths. Borehole no.3, 25–30 m away from a heated building complex inMangui, in the northern Da Xing’anling Mountains,exhibited ground temperature profiles with negativegeothermal gradients (Figure 5). The MAGT was�1.98C at a depth of 20 m in October 1973 and hadwarmed by 0.78C by 1978.


DOI: 10.1002/ppp

Figure 4 Average annual soil temperatures at shallow depths inborehole 14 at the Yituli’he Permafrost Observatory. Measurementswere obtained using thermistor cables fixed in the dry, sealedboreholes.

igure 5 Ground temperatures measured in borehole no. 3,5–30 m from a building complex in Mangui. Measurements wereade with a string of vaseline-retarded Assmann thermometerswered down the dry, sealed borehole.


F2mlo


Expanding Taliks and DisappearingPatches of Permafrost

Many forestry bureau offices and residential areas ofvarious sizes have been established since the 1950sand 1960s as a result of forest resource development inthe northern Da Xing’anling Mountains. Most of thesebuilding complexes are located in the transition zonebetween the mountain slopes and valley bottoms, andon river terraces, where permafrost is generally thinand warm with extensive taliks. Thaw bulbs have formedbeneath building sites, and taliks along the majorrivers have deepened or expanded during the past40–50 years. For example, taliks near the Fuqingshanand Linzhong Forestry Bureau offices along theDalin’he River expanded upslope by 400–500 and500–2000 m, respectively, between 1973 and 1985(Yuan, 1989). Recent preliminary inspection along theproposed China-Russia Oil Pipeline route, andnumerous exploratory boreholes, have indicated thatmany patches of permafrost observed in the Jiagedaqicity area in the 1960s have since vanished.

Shrinking and disappearance of patchy permafrostare both occurring, mainly in the southern parts of theXing’anling Mountains, where the MAGT ranges from�0.5 to þ0.58C, and the thickness of permafrost variesfrom 3 to 15 m. In many sections, permafrost is thinnerthan the depth of zero annual amplitude, whichgenerally results from a positive MAGT. Intensivehuman activities started relatively early in these areas.For example, when the cities or towns in Yakeshi,Jiagedaqi and Da’yangshu in the vicinity of the SLP,were founded in the early 1950s, patchy permafrost wascommon, but after 30–40 years of anthropogenicactivity, much of it has disappeared. A permafrostisland with the permafrost table at 1.7 m was detectednear Jiagedaqi during the railway construction in 1964,but by 1974 permafrost was absent under therail roadbed. In one forestry bureau area in theXiao Xing’anling Mountains, permafrost declined fromthe 1950s onwards as deforestation occurred (Table 2)(Guo, 1990). Given the rate of deforestation and thesubsequent trend of permafrost degradation, permafrostmay now have completely thawed in this area.

The 300-km-long Hei’he to Bei’an Highwaytraverses three subzones of patchy permafrost fromnorth to south: (1) the foothills along the northern slopeof the Da Xing’anling Mountains, (2) the foothills inthe lower Xiao Xing’anling Mountains, and (3) thenorthern edges of the Song’huajiang and NenjiangRivers Plains. The present-day distribution of patchypermafrost revealed by the highway should bereasonably representative of these three sub-zones. Inthe 1970s, the percentage of permafrost immediately


DOI: 10.1002/ppp

Table 2 Correlation between the degradation of perma-frost and deforestation in one of the forestry bureaus inthe Xiao Xing’anling Mountains (Guo et al., 1981).

Year Forestcover (%)

Areal extent ofpermafrost (%)

1957 94 10.51960 76 5.81970 25 2.31980 10 0.1

252 H. Jin et al.

north of the SLP was 10–25% (Guo et al., 1981). Incontrast, drilling and coring, ground penetration radarand hand-dug pits during August–September 2000from Kilometers 42 to 190 (the section most prone topermafrost) encountered only 17 isolated patches ofpermafrost varying from 40 and 500 m in length. Thetotal distance was 3.2 km and most permafrost was lessthan 4 to 5 m in thickness (Wang et al., 2001). It hasbeen deduced that the areal percentage of patchypermafrost along the Hei’he-Bei’an Highway, and inthis general area, is about 1%. Similarly, recentfieldwork by CAREERI geocryologists and PetroChinaPipeline Company geologists has revealed that manyisolated patches of permafrost associated with wetlandssouth of Jiagedaqi and north of Da’yangshu in the1970s have disappeared or shrunk significantly.

Pattern of Permafrost Degradation

Permafrost is more extensive, thicker and colder atlower elevations in the Xing’anling Mountains area.This pattern influences permafrost degradation. Whenhuman influences are similar, permafrost degradesmore rapidly on sunlit (south-facing) slopes, at higherelevations, and in the mountains than on shadedslopes, at lower elevations, and in valleys without opentaliks. This pattern of permafrost degradation was firstrecognised during the extensive engineering surveysof the 1960s–70s (Third Survey and Design Instituteof the Railway Ministry of China, 1978) and has beenrepeatedly confirmed. For example, the upper edge ofpermafrost on the southeastern slope of the northernAmu’er Valley descended by 80–100 m during the1980s to early 1990s (Gu and Zhou, 1994).

FACTORS AFFECTING PERMAFROSTDEGRADATION

Climatic warming and increasing human activitiesduring the last century are both thought to beresponsible for the degradation of permafrost inthe Xing’anling Mountains. Extensive and increasing


human activities for resource exploitation, agricultureand economic development have accelerated theprocesses. In particular, frequent forest fires causedby lightning during very dry seasons and accidentalfires started by human beings have had appreciableeffects on the thermal regimes of permafrost andforested and wetlands environments.

Climatic Change

The principal cause of the observed permafrostdegradation in the Xing’anling Mountains is climaticwarming that has occurred since the end of the 19th

century. Based on decadal average air temperatures(DAATs) from northeastern China, this change can bedivided into three main periods: (1) persistentwarming from the end of 19th century to the 1940s;(2) quasi-stable average temperatures during the1950s to 1970s; and (3) a second persistent warmingfrom 1970 to 2000, and particularly during 1991–2000(Qin, 2002). DAATs from 33 meteorological stationsin Heilongjiang Province and the northeastern part ofthe Inner Mongolia Autonomous Region (IMAR)during 1961–2000 exemplify the latest warmingperiod (Table 3). During 1991–2000, DAATs were0.9–2.28C higher than those for 1961–70, and 25stations (75%) exhibited increases greater than 1.58C.

Analyses of air temperatures from nine meteor-ological stations in the Heilongjiang Province,including Huma, Qiqi’har, Jiamusi and Hailun,between 1951–80 and 1981–2000 indicate thatMAATs during the latter period averaged 1.08Cwarmer than during the former (Pan et al., 2004). Thegreatest warming occurred in winter, with an averageincrease of 1.68C in the Heilongjiang Province, andthe warming centre was located in Huma to thenorthwest, where the increase was 2.28C. The secondgreatest warming occurred in spring, with an overallincrease of 1.38C in Heilongjiang Province, and againthe greatest rise of 1.68C was in Huma. Summer andautumn both exhibited average increases of about0.58C (Pan et al., 2004).

The temperature increase in Heilongjiang Provincefrom 1951–80 to 1981–2000 is the largest in China(Pan et al., 2003) and the neighbouring IMAR has thesecond largest warming of 0.98C. Most of thiswarming extended over the study region, i.e. northernHeilongjiang Province and eastern IMAR.

Forest Fires

The average number of forest fires was 42 per yearfrom 1966–87 (Du, 1987). The most destructive fire in


DOI: 10.1002/ppp

Tab

le3

Dec

adal

aver

age

air

tem

per

atu

res

(DA

AT

s)in

the

Xin

g’a

nli

ng

Mo

un

tain

s,1

96

1–

20

00

.

Sta

tio

nL

atit

ud

e(8

N)

Lo

ng

itu

de

(8E

)E

levat

ion

(m)

DA

AT

s(8

C)

Aver

age

DA

AT

19

60

s

-DA

AT

19

90

s1

96

0s

19

70

s1

98

0s

19

90

s1

96

1–

20

00

Mo

’he

5285

80

12

283

00

43

5�

4.9

�4

.9�

4.0

�4

.0�

4.4

0.9

Hu

ma

5184

30

12

683

90

17

7�

2.1

�1

.8�

0.8

�0

.4�

1.3

1.7

Gen

’he

5084

70

12

183

10

98

0�

5.5

�5

.0�

4.0

�3

.3�

4.5

2.2

Yit

uli

’he

5082

90

12

184

10

73

3�

5.4

�5

.0�

4.3

�3

.9�

4.6

1.5

Da

Xin

gan

ling

5282

40

12

480

70

37

2�

1.3

�0

.7�

0.2

1.5

E’e

rgu

’na

5081

50

12

081

10

58

1�

3.1

�2

.9�

2.6

�1

.7�

2.6

1.4

Hei

’he

(Ai’

hu

i)5

081

50

12

782

70

16

6�

0.5

�0

.30

.50

.90

.21

.4M

anzh

ou

li4

983

40

11

782

60

66

2�

1.3

�1

.3�

0.9

0.1

�0

.91

.4S

un

wu

4982

60

12

782

10

23

5�

1.6

�1

.5�

0.5

0.2

�0

.91

.8C

hen

ba’

erh

uB

ann

er4

981

90

11

982

60

57

7�

3.0

�2

.3�

1.5

�0

.6�

1.9

2.4

Hu

lun

bai

’erm

eng

4981

30

11

984

50

61

2�

2.0

�1

.7�

1.1

�0

.1�

1.2

1.9

Nen

’jia

ng

4981

00

12

581

40

24

2�

0.4

�0

.10

.40

.90

.21

.3B

ok

etu

4885

50

12

181

40

74

0�

1.0

�0

.9�

0.4

0.0

�0

.61

.0R

igh

tC

hen

ba’

erh

uB

ann

er4

884

00

11

684

90

55

60

.30

.40

.91

.90

.91

.6B

ei’a

n4

881

70

12

683

10

27

00

.00

.40

.81

.20

.61

.2L

eft

Ch

enb

a’er

hu

Ban

ner

4881

30

11

881

60

64

2�

0.6

�0

.40

.11

.00

.01

.6X

iao’e

rgou

4880

70

12

384

30

28

8�

1.1

�1

.1�

0.3

0.5

�0

.51

.6K

esh

an4

880

30

12

585

30

23

51

.01

.41

.82

.41

.71

.4Z

alan

tun

4880

00

12

284

40

30

62

.32

.73

.23

.83

.01

.5F

u’y

u4

784

80

12

482

90

16

31

.92

.12

.73

.42

.51

.5Y

ich

un

4784

40

12

885

50

24

10

.30

.61

.41

.61

.01

.3H

ailu

n4

782

60

12

685

80

23

91

.21

.62

.02

.61

.81

.4Q

iqi’

har

4782

30

12

385

50

14

73

.23

.33

.94

.63

.81

.4H

egan

g4

782

20

13

082

00

22

82

.73

.13

.43

.83

.31

.1A

’ers

han

4781

00

11

985

70

10

27

�3

.3�

3.1

�2

.8�

2.1

�2

.91

.2M

ing

shu

i4

781

00

12

585

40

24

91

.92

.32

.83

.52

.61

.6T

ieli

4685

90

12

880

10

21

00

.91

.22

.02

.41

.61

.5Ji

amu

si4

684

90

13

081

70

81

2.7

3.1

3.7

4.1

3.4

1.4

Su

i’h

ua

4683

70

12

685

80

18

02

.02

.22

.83

.42

.61

.4S

uo

lun

4683

60

12

181

30

10

28

2.3

2.0

2.5

3.0

2.4

0.7

Ula

n’h

ot

4680

50

12

280

50

27

64

.14

.35

.15

.74

.81

.6H

arb

in4

584

50

12

684

60

14

23

.53

.64

.24

.94

.11

.4D

on

gw

uq

iB

ann

er4

583

10

11

685

80

83

90

.71

.01

.31

.91

.21

.2

Copyright # 2007 John Wiley & Sons, Ltd. Permafrost and Periglac. Process., 18: 245–258 (2007

DOI: 10.1002/ppp


)

254 H. Jin et al.

1987 burned an area of about 12,000 km2 over thecourse of one month.

The effect of fire varies greatly depending on theduration of the blaze, the types and densities of thevegetation/forests burnt, and whether the burn is onsloping ground or in a wetland. On slopes, therelatively dry organic mat tends to burn (oftensmouldering) almost completely, leaving more-or-lessbare ground exposed. In wetlands, the crowns of thegrasses and the roots of the bushes are often relativelylittle damaged and can begin to recover more quickly,spurred on by the nutrients provided by the ash fromthe fire. Wetlands areas are also less prone to erosionthan slopes.

Viereck (1973) and Pavlov (1980) concluded thatthe direct thermal impact of forest fires on permafrostis minimal due to their transitory nature. But manystudies have shown that subsequent effects can includeincreased near-surface ground temperatures, signifi-cantly increased thaw depths and differential thawsettlements in thaw susceptible soils (Gavrilova, 1969;Heginbottom, 1973; Racine, 1981; Viereck, 1982;Zhou and Liang, 1991; Yoshikawa et al., 2003).

Measurements made in wetlands four to six yearsafter the 1987 forest fire showed that temperatureswere 3.68C warmer at depths of 10–40 cm and 0.4 to2.28C warmer at the permafrost table depth of 1.5 m,and that depths of thaw were greater in burnt thancomparable unburned areas (Zhou et al., 1994). Onslopes, supra-permafrost taliks developed withincreasing thaw and at some sites, the warm, thinpermafrost completely degraded. Temperatures at adepth of 20 cm were 1.0 to 2.58C warmer, those at1.5 m depth were 0.2 to 1.38C warmer, but values at2.5 m were actually colder than those on the unburntslopes. The MAGT was �2.18C at the depth of annualamplitude of 15 m beneath naturally forested land (Guand Zhou, 1994; Zhou et al., 1994). Assuming atypical geothermal gradient, this temperature wouldsuggest a permafrost thickness of approximately 80 min this part of the Xing’anling Mountains.

Human Activities

Forested environments in the Xing’anling Mountainshad been little disturbed prior to the 19th centurybecause of the sparse human population. However,beginning in the late 1930s and 1940s, the populationincreased to more than 600,000 and by mid-century,much of the original forest in northeastern China hadbeen destroyed. Logging has continued northwards atan unprecedented rate, reaching the southern bank ofthe Heilongjiang (Amur) River. Even secondary


forests are sparse near Yakeshi on the Harbin toManzhouli Highway.

Deforestation removes shading provided by thetrees, damages the slow-growing moss-litter insulat-ing organic mat, and allows the growth of different andfast-growing vegetation assemblages, whereas agri-culture totally removes the insulating moss cover.Both practices allow more insolation and windturbulence to reach the ground surface. Sensible andlatent heat losses to the atmosphere may increase butthey are outweighed by greater insolation whichgenerally results in increased heat flow to thepermafrost, the growth of supra-permafrost taliks,the development of thaw bulbs beneath highways andother engineering foundations, and ultimately loss ofpermafrost. As a result, the areal extent of patchypermafrost declined as forest cover diminished (seeTable 2).

Measurements during the summer of 1992 showedthat logged surfaces were 1.7 to 2.08C warmer thanforested surfaces, that soil moisture contents had beenreduced by about 50%, and that the summertemperatures at the depths of 20–30 cm were about58C higher than at similar depths in nearby forests.The consequent increases in the depths of thaw wereabout 20–30 cm (Chen and Yin, 1996).

PRESENT POSITION AND FUTURECHANGES IN THE SLP

SLP in Northeastern China

Permafrost in the Xing’anling Mountains is an integralpart of Eurasian continental permafrost. Russiangeocryologists in the late 19th and early 20th centuriesrecognised that the Eurasian SLP extended into theChina (Zhou et al., 2000). Xin and Ren (1956) refinedthe positioning of the SLP on the basis of geologicaland hydrogeological surveys, and engineering ex-ploration data. Subsequent excavations for roadbedsand bridges revealed many patches of permafrost,south of the established SLP. Field investigationsconducted during 1973 and 1974 included drilling,hand-dug pits and DC-electrical resistivity surveys tocheck uncertain permafrost conditions as suggested bythe-then-available engineering exploration data.These resulted in the location of the SLP whichreflected the combined impacts of geographic zona-tion and local variability (Guo et al., 1981) in the late1970s (see Figure 1).

Compared to the Siberian Plains, the SLP in theXing’anling Mountains protrudes southwards by12–138 latitude (Zhou et al., 2000) due to the higher


DOI: 10.1002/ppp


elevations and the influence of the prevailingSiberia-Mongolian cold air masses, which generallyresults in 10–208C colder temperatures in winter in theXing’anling Mountains than at similar latitudes inEurasia. In summer, the monsoon-induced precipi-tation reduces heating of land surfaces, and con-sequently air temperatures in the Xing’anling Moun-tains are 1–78C cooler than those at similar latitudes inXinjiang Uygur Autonomous Region. The combinedseasonal influences result in much lower MAATs innortheastern China.

In northeastern China, the SLP takes the shape ofthe letter ‘W’, protruding southwards in the Da andXiao Xing’anling Mountains, but northwards in thenorthern part of the Songnen Plain, with a span ofabout 28 latitude. It stretches about 900 km southwardsto the Huang’gang’liang Mountains at 2000 m inelevation, where the lower limit of alpine permafrost,on the basis of field investigations, is estimated to bebetween 1700 and 1800 m.

SLP in the 1970s

The SLP during the 1970s (see Figure 1) was generallylocated between the MAAT isotherms of �1.0 andþ1.08C. More specifically, it occurred between �1and 08C in the western section (west of the DaXing’anling Mountains), around 08C in the middlesection (Songnen Plain and southeastern and south-western slopes of the Da Xing’anling Mountains), andbetween 0 and þ1.08C in the eastern section (southernand southeastern slopes of the Xiao Xing’anlingMountains) (Northeastern China Permafrost ResearchTaskforce, 1983). The variation in the empiricalcorrelation between the SLP and MAAT is believed torelate to active layer moisture contents (Guo et al.,1981). Annual precipitation increases southeastwards,resulting in higher active layer moisture contents inthe eastern section compared to the western section.When the soil moisture contents in the active layer arehigher, the resultant greater thermal inertia leads tolonger time lags in soil temperatures responding tochanges in air temperatures.

SLP in the 1990s

The annual precipitation and soil moisture contentsmay have changed slightly from the 1970s to thepresent, but the broad correlation between the SLP andthe MAAT probably has not altered substantially.Therefore, the same rules of thumb can be used toestimate the present SLP: in the western, middle andeastern sections, it is expected to coincide with theMAAT isotherms for the period 1991–2000 of 0 to


�1.0, around 08C, and 0 to þ1.08C, respectively (seeFigure 1).

In comparison to the 1970s, most of the Hulunbei’erHigh Plateau, the northern Songnen Plain and thesoutheastern Xiao Xing’anling Mountains, whichwere then sporadic permafrost, are now south of theSLP. During the 1970s, the areal extent of permafrostvaried northwards from 5% to 30% of the terrain (Dai,1982). This situation has changed markedly, andpermafrost in the vicinity of and just to the north of theSLP has disappeared (Jin et al., 2000a, 2000b).According to a recent survey, using ground penetrationradar and hand-dug pits, along the Hei’he to Bei’anHighway in the Xiao Xing’anling Mountains, perma-frost underlies only 2% of the area (Wang et al., 2001).Isolated patches of permafrost account for only 0.1%of the terrain in one of the forested areas (Guo, 1990;Wang et al., 2001). Overall, it is estimated that the areaof permafrost has decreased by 90,000–100,000 km2,to 62% of the original area (260,000–270,000 km2) ofsporadic permafrost in the 1970s (Guo et al., 1981).

SLP in the Near Future

Average air temperatures in East Asia are projected toincrease by 0.98C by 2023 and by 1.18C by 2040because of global climatic warming (Wang and Dong,2002). Pan et al. (2004) have predicted that MAATswill increase by 1.98C by 2030 and by 2.48C by 2050in Harbin, Qiqihar, Jiamusi and Mudanjiang, all inHeilongjiang Province, due to climate warming andthe rapid economic development. These projected/predicted increases in the MAATs will eventuallycause a downward degradation of the uppermostpermafrost layer. This degradation will be controlledby summer air temperatures and depths of thaw untilthe temperature at the top of permafrost rises above08C, when a supra-permafrost talik will develop anddownward degradation will continue. The effects ofchanges in air temperature diminish in amplitude withdepth and are increasingly lagged. When theamplitudes of the temperature changes at the surfaceare 10–208C, it takes about a year for the thermalimpact to reach depths of 12.5–14.5 m, 10 years toreach 40–47 m, and about 100 years to affect depths of125–145 m (Brewer, 1958; Garagulia, 1985; Perma-frost Institute, Siberia Branch, USSR Academy ofSciences, 1990; Gavrilova, 1993). The SLP boundary,therefore, will move northwards, but with time lags inthe order of years or decades as the air/ground-surfacetemperatures increase and the permafrost degrades.

If MAATs during the next 40–50 years are 1.0–1.58Chigher than today, many of the current sporadic andisolated patches of permafrost will disappear and the


DOI: 10.1002/ppp

256 H. Jin et al.

SLP could approach the present-day southern boundaryof the zone of the widespread discontinuous permafrost.The latter would then gradually be converted to sporadicor isolated patches of permafrost. Zhou et al. (1996) andJin et al. (2000a, 2000b) estimated that the area ofpermafrost in the Xing’anling Mountains might becomeonly 35% of the estimated 380,000 to 390,000 km2 thatexisted in the 1970s. However, these numbers remainspeculative due to the paucity of site-specific permafrostdata, uncertainties regarding the degree of climaticwarming and changes in precipitation, and the extent offuture industrial, natural resource and agriculturaldevelopment.

CONCLUSIONS

The rate of retreat of the SLP in the Xing’anlingMountains, except in the higher southern DaXing’anling Mountains, markedly accelerated duringthe 20th century, partly due to climatic warming,natural phenomena such as forest fires resulting fromlightning strikes and greatly expanded human activity.

The strongest meteorological warming trend inChina during the 20th century, 1.18C, occurred inHeilongjiang Province. This resulted in deepening ofthe active layer, warming of permafrost, thinning andsometimes total degradation of the thinnest perma-frost, and the shifting of the permafrost zonesnorthward. The SLP moved northwards so that patchyand island permafrost disappeared, discontinuouspermafrost became sporadic, and some of thecontinuous zone became discontinuous.

Climatic warming is expected to continue over thenext 50 years, as are industrialisation, furtherdeforestation and agricultural development. If pro-jected MAAT increases of 1.18C to 2.48C occur by2050, the net effect on the distribution of permafrostwill be continuing northward retreat of the permafrostzones and an estimated reduction of permafrost areasin the Xing’anling Mountains to approximately 35%of those prevailing in the 1970s.

A more rigorous programme of field observations,which had dwindled due to a focus on theQinghai-Tibet Plateau while the Railway Projectwas underway, needs to be implemented in order toavoid engineering and economic mistakes and tominimise environmental disruptions in the Xing’an-ling Mountains areas.

ACKNOWLEDGEMENTS

This study was supported by the Chinese Academy ofSciences (CAS) 100 Talents Project (grant no.


0481811) ‘Stability of linear engineering foundationsin warm permafrost regions under a changing cli-mate’, Key Directional Project (grant no. KZCX2-YW-311) of the CAS Knowledge Innovation Program‘Mechanisms and mitigation of frost heaving and thawsettlement of pipeline foundation soils in permafrostregions’, Pilot Project (grant no. O650445) of the CASCAREERI Knowledge Innovation Program ‘Climaticand environmental changes in cold and arid north-eastern China and their adaptation’ and the Glaciol-ogy and Geocryology Incubation Program (grant no.40225001) of the National Science Foundation ofChina ‘Symbiosis and degradation of permafrostand wetlands ecosystems in northeastern China’.Generous assistance for revising the paper was pro-vided by the Editor, Professor Antoni G. Lewkowicz,and by Alan Heginbottom, Jerry Brown, Max C.Brewer and an anonymous reviewer.

REFERENCES

Brewer MC. 1958. Some results of geothermal investi-gations of permafrost in northern Alaska. AmericanGeophysical Union, Transactions 39: 19–26.

Chen Y, Yin Y. 1996. Impacts of forest industry devel-opment in the Da Xing’anling Mountains on theseasonally thawed layer overlying permafrost. In Pro-ceeding of the 5th Chinese Conference on Glaciologyand Geocryology (Vol. 2). Gansu Culture Press: Lanz-hou; 1087–1091 (in Chinese).

Dai J. 1982. Features of permafrost temperature in north-ern part of the Da-Xing’anling Mountains. Journal ofGlaciology and Geocryology 4(3): 53–63 (in Chinese).

Du X. 1987. Study on types of combustibles in the Ta’heForest Bureau area. Journal of Northeast ForestryUniversity (2): 5–7.

Garagulia LC. 1985. Methods for Forecasting Perma-frost Conditions under the Anthropogenic Influe-nces. Moscow State University Press: Moscow; 20–29.

Gavrilova MK. 1969. Radiation regime. Forestry 1: 16–23

Gavrilova MK. 1993. Climate and permafrost. Perma-frost and Periglacial Processes 4: 100–111.

Gu Z, Zhou Y. 1994. Changes of permafrost in theAmu’er area of the Da-Xing’anling Mountains. InImpacts and Countermeasures of Forest Fire in theDa-Xing’anling Mountains, Zhou YW, Yang ZX (eds).Science Press: Beijing; 156–161 (in Chinese).

Guo D. 1990. Frozen Ground in China. Gansu EducationPress: Lanzhou; 84–95 (in Chinese).


DOI: 10.1002/ppp


Guo D, Huang Y. 1989. The role of geological structure inthe formation of permafrost in the Hola Basin, north-eastern China. Journal of Glaciology and Geocryology11(3): 216–222 (in Chinese).

Guo D, Li Z. 1981. Geological evolution and age offormation of permafrost in northeastern China sincethe late Pleistocene. Journal of Glaciology andGeocryology 3(4): 1–6 (in Chinese).

Guo D, Wang S, Lu G, Dai J, Li E. 1981. Zonation ofpermafrost in the Da- and Xiao-xing’anling Mountainsin northeastern China. Journal of Glaciology andGeocryology 3(3): 1–9 (in Chinese).

Heginbottom JA. 1973. Some effects of a forest fire on thepermafrost active layer at Inuvik, Northwestern Ter-ritories. Geological Survey of Canada, Department ofEnergy, Mines and Resources, Ottawa, Ontario,Canada.

Jin H, Li S, Wang S, Cheng G, Li X. 2000a. Impacts ofclimatic changes on permafrost and cold regionsenvironments in China. Acta Geographica Sinica55(2): 161–173 (in Chinese).

Jin H, Cheng G, Li S, Zhao L. 2000b. Permafrost andclimatic change in China. Global and PlanetaryChange 26: 387–404 (in English).

Lu G, Weng B, Guo D. 1993. Geographical southern limitof permafrost in Northeastern China. Journal ofGlaciology and Geocryology 15(2): 214–218 (inChinese).

Meteorological Institute of China MeteorologicalAdministration. 1961. Historical Synoptic Charts inthe Eurasia Continent. Meteorology Press: Beijing (inChinese).

Meteorological Institute of China MeteorologicalAdministration. 1969. Climatological Data on AirPressures, Temperatures, and Humidity in the HighAtmosphere during the Period from 1960–1969.Meteorology Press: Beijing (in Chinese).

Northeastern China Permafrost Research Taskforce.1983. Major features of permafrost distribution in
northeastern China. In Proceeding of the 2ndChineseConference on Geocryology. Gansu People’s Press:Lanzhou; 36–42 (in Chinese).
Pan HS, Zhang G, Xu NP. 2003. A preliminary analysis ofclimate warming in Heilongjiang Province since the
1980s. Climatic and Environmental Research 8(3):348–355 (in Chinese).

Pan HS, Zhang G, Xu NP, Zu S. 2004. Impacts of climaticwarming on agricultural and ecological environmentsand mitigation measures for crop structures. Heilong-jiang Journal of Hydraulic Engineering 31(4): 6–9 (inChinese).

Peng H, Cheng G. 1990. Ice wedges in the Da Xing’anl-ing Mountains, northeastern China and their implica-tions in paleo-climatology. In Proceeding ofthe 4thChinese Conference on Glaciology and Geo-cryology. Science Press: Beijing; 9–16 (in Chinese).

Permafrost Institute, Siberian Branch, USSR Academy ofSciences. 1990. General Geocryology, Guo D (trans.).Science Press: Beijing; 130–150 (in Chinese).

Qin D. (ed.). 2002. Features and Evolution of Environ-ments in Western China: Assessment of the Evolutionof Environments in Western China. Science Press:Beijing; 41–49 (in Chinese).

Racine CH. 1981. Tundra fire effects on soil and treeplant communities along a hill slope gradient in theSeward Peninsula. Alaska. Arctic 34(1): 71–84.

Third Survey and Design Institute of the Railway Min-istry of China. 1978. Frozen Ground Engineering
Geology. Science Press: Beijing; 55–70 (in Chinese).
Tong B. 1993. Ice wedges in northeastern China.Journal of Glaciology and Geocryology 15(1):41–46 (in Chinese).

Viereck LA. 1973. Ecological effects of river floodingand forest fires on permafrost in the taiga of Alaska. InProceedings, The North American Contribution tothe 2ndInternational Conference on Permafrost, July1973, Yakutsk, USSR. National Academy of Sciences:Washington DC; 60–67.

Viereck LA. 1982. Effects of fire and firelines on activelayer thickness and soil temperatures in interiorAlaska. In Proceedings, 4thCanadian Permafrost Con-ference, 2–6 March 1981; Calgary, Alberta: The RogerJ E Brown Memorial Volume. National ResearchCouncil of Canada: Ottawa, Ontario, Canada;123–134.

Wang B, Sheng Y, Liu J. 2001. Distribution and degra-dation of permafrost along the Hei’he to DaqingHighway in the Xiao Xing’anling Mountains. Journalof Glaciology and Geocryology 22(3): 302–305(in Chinese).

Wang C. 1982. The permafrost table and ground ice innorthern part of the Da Xing’anling Mountains. InProceedings of Chinese Conference on Glaciology andGeocryology (Volume of Geocryology), GeographicSociety of China (ed.). Science Press: Beijing;31–37 (in Chinese).

Wang C, Zhang B, Liu F, Song C. 1996. Marks andphenomena of permafrost degradation in northern


DOI: 10.1002/ppp

258 H. Jin et al.

China. In Proceedings of the 5thChinese Conference onGlaciology and Geocryology, Vol. 2. Gansu CulturePress: Lanzhou; 1063–1070 (in Chinese).

Wang S, Dong G. (eds). 2002. Assessment of theevolution of environments in western China, Session 1.In Features and Evolution of Environments in WesternChina. Science Press: Beijing; 29–66 (in Chinese).

Wu Z. 1983. Principal types of permafrost temperaturesof the Qinghai-Tibet Plateau and computation methodsof thermal parameters for engineering applications. InMonograph on the Research of Permafrost on theQinghai-Tibet Railway. Science Press: Beijing;185–194 (in Chinese).

Xin K, Ren Q. 1956. Distribution of permafrost in north-eastern China. Geology Knowledge 10: 1–6 (inChinese).

Yoshikawa K, Bolton WR, Romanovsky VE, Fukuda M,Hinzman LD. 2003. Impacts of wildfire on the perma-frost in the boreal forests of interior Alaska. Journal ofGeophysical Research 108(D1): 8148. DOI: 10.1029/2001JD000438.

Yuan H. 1989. Degrading permafrost in the Da- andXiao-Xing’anling Mountains. In Proceedings ofthe 3rdChinese Conference on Geocryology. SciencePress: Beijing; 54–57 (in Chinese).


Zhou Y, Guo D. 1982. Major features of permafrost inChina. Journal of Glaciology and Geocryology 4(1):1–19 (in Chinese).

Zhou Y, Liang L. 1991. Impacts of forest fires onpermafrost environments: a review. Advances in Geo-sciences 6(1): 62–66 (in Chinese).

Zhou Y, Wu Z. 1965. Permafrost in the northern part ofthe Da-Xing’anling Mountains. In Monograph onthe geocryological expedition along the Qinghai-Tibet Highway, Appendix. Science Press: Beijing(in Chinese).

Zhou Y, Liang L, Gu Z, Liang F, Zhang Q. 1994. Changes ofthermal regimes of permafrost after forest fires in north-ern Da Xing’anling Mountains. In EnvironmentalImpacts of Forest Fires in the Da Xing’anling Mountainsand their Mitigation, Zhao Q, Zhang W, Zhou Y, Yang Y(eds). Science Press: Beijing: 25–35 (in Chinese).

Zhou Y, Wang Y, Gao X, Yue H. 1996. Temperaturesand distribution of permafrost in northeasternChina under a warming climate. Journal ofGlaciology and Geocryology 18(Supplement):140–146 (in Chinese).

Zhou Y, Qiu G, Guo D, Cheng G, Li S. 2000. FrozenGround in China. Science Press: Beijing; 171–205 (inChinese).


DOI: 10.1002/ppp

Degradation of permafrost in the Xing'anling Mountains, northeastern China

Documents