“recent warming”: ice core evidence from tropical ice cores with emphasis on central asia

Global and Planetary Change. 7 (1993) 145-156 145 Elsevier Soence Pubhshers B V, Amsterdam

"Recent warming'" ice core evidence from tropical ice cores with emphasis on Central Asia

L.G. T h o m p s o n ~, E. M o s l e y - T h o m p s o n a, M. Davis ~, P.N. Lin a, T. Yao b, M. D y u r g e r o v c

and J. Dal a

a Byrd Polar Research Center, The O&o State Umt,erstty. 125 S Oval Mall. Columbus, OH 43210, USA b Lanzhou Instttute of Glactology and Geocryology, People's Repubhc of China

c The Instttute of Geography, Moscow, Russta

(Received March 2. 1992. revised and accepted July 3, 1992)

ABSTRACT

Thompson, L G., Mosley-Thompson, E , Davis, M, Lm, P N, Yao, T, Dyurgerov, M and Dal, J , 1993 "Recent warming" ice core evidence from tropical ice cores w~th emphasis on Central Asia Global Planet_ Change, 7 145-156

Ice cores from the tropics and subtropics, Jn conjunction with those from the polar regions, provide a multtfaceted record (dust, chemistry, stable isotopes, accumulation) of environmental changes which can be viewed both spatially and temporally This paper emphasizes the oxygen isotopic record (8l~O) preserved in cores from the poles to the tropics and assesses the evtdence for global warming In the last 50-100 years From north to south these records include Camp Century, Greenland, Dunde and Guhya Ice Caps, China, Gregonev Ice Cap, Kirghxzia (formerly part of USSR), Quelccaya Ice Cap, Peru and Slple Station and South Pole, Antarctica The central Asian records along with that from Quelccaya provide strong evidence of recent and rapid warming In the troptcs and subtropics For the Dunde Ice Cap, where a long paleochmanc record is available, the warming m this century appears to be unprecedented in the Holocene These tropical and subtropical records contrast sharply with those from polar cores which show little evtdence of a recent warming These data suggest that either the recent warming Is a middle and lower latitude phenomenon or that these high altitude troptcal and subtropical glaciers may be more sensitive to climate changes than the massive polar ice sheets Regardless, the current rapid disintegration of many tropical and subtropical glaciers may result in the permanent loss of numerous umque archwes

Introduction

The patterns and sources of lnterannual, decadal and centennial climatic variability are least well known and constitute a gap in our understanding of the global climate system. Com- parisons of climate model results with systematic regional, continental, hemisphenc and global- scale compilations of high-resolution paleoch- mattc data are needed to fill this gap and to develop more realistic climate models. The Earth's ice sheets and ice caps are recognized as "the best of only a few" libraries of atmospheric history from which past climatic and environmental conditions may be extrapolated. However, much of the climatic activity of significance to

humanity may not be strongly expressed in (or may not reach) the polar ice caps. Fortunately, ice core records also can be recovered from a limited number of high-altitude, low- and mid- latitude "polar" type ice caps (Fig 1) In addition, the high accumulation on these Ice caps and at some locations on the polar ice sheets makes It possible to recover high resolution records for the last 1000 years. Changes m the physical character and chemical composition of the atmosphere may be reconstructed using soluble and insoluble dust, stable isotopes, net accumulation changes and, in some cases, trapped gases. Additionally, non- polar ice core records provide a unique archive of biological changes as inferred from pollen, di- atoms and plant fragments. Such environmental

0921-8181/93/$06 00 © 1993 - Elsevier Science Pubhshers B V All rights reserved

146 L G THOMPSON ET AL

Camp Century ~

Fig 1 Location of the Camp Century, Greenland, Dunde, Guhya and Gregorlev Ice Caps m central Asia, the trop,cal Quelccaya Ice Cap in Peru, the Siple Station and South Pole core m Antarctica

information is not readily available from polar ice cores. Thus, the variety of chemical and physical data extracted from ice caps and ace sheets pro-

vide a multl-faceted record of both the chmatic and environmental h~story of the Earth which is critical to the assessment of the relattve impor-

~180 (~oo) Decadal Averages Camp Century Dunde Quelccaya

Greenland China Peru 77*10'N, 61"08'W 38*06'N, 96°24'E 13"56'S, 70"50'W

-31 -29 -27 -125 -105 -210 -190 - f rO -52

2 i "40 t ,,,0 t -

ar:::i I 580 t (AD) 1500 t

11001

South Pole Siple Star,on Antarct,ca Antarct|ca

75.55's, 84ol5'w -515 -505 -495 -51 -505 -295 -285 ' ',..q. ' m ' ' 1980 ' , t

= t ~ i 1900

r _ ~ l 1820

r - - - i r ~ r I 1740

t ~ l Jim 1660

1580 I

2~ , ,5oo 1420

Fig. 2 Decadal averages of the 6 1 8 0 records m a N - S transect from Camp Century, Greenland (Johnsen et al , 1970) m the north to South Pole Station, Antarct ,ca in the south The shaded areas represent lSOtoplcally less negative (warmer) periods and the unshaded areas represent lsotoplcally more negatwe (colder) periods relative to the mdlvldual record means

'RECENT WARMING" EVIDENCE FROM TROPICAL ICE CORES WITH EMPHASIS ON CENTRAL ASIA 147

tance of such climate system components as vol- camc activity, greenhouse gas concentrations, atmospheric dust loading and solar varlabdlty. Un- derstanding the climate system with the goal of prediction requires a global synthes~s of its history over a period of hme from which the most reliable, dwerse and complete data sets exist. At the same t~me, the longer records encompassing glacial/interglacial fluctuations, while lacking m temporal resolution, are important as they provide a much longer perspective from which the recent (last 1000 years) variations must be viewed

High resolution reconstruction of the last 1000 years is critical for a variety of reasons. (1) it is most relevant to human actlVltles, present and future, as recent climatic and environmental van- abdlty due to human achvitles are superimposed; (2) times of extremes existed within the Holocene such as the "Little Ice Age" period; (3) maximum data coverage exists, (4) multi-proxy reconstruc- tions are feasible; (5) annual (or seasonal) resolution is possible; (6) leads, lags and rates of change w~thin the climate system can be studied directly, (7) the causes of the observed changes are unde- termined; and (8) potential forcing functions may be identified and tested.

Results

The global perspectlt,e

The assimilation of multl-faceted ice core data sets from polar and non-polar regions reveals both temporal and spatial variability of the particular ice core parameter under consideration (i.e., dust, chemistry, gases, isotopes, etc). Here we examine a global array of s lsO records to assess whether the available ~ce core histories provide evidence of recent global warming. In varying degrees, slSO records reflect (1) condensation air temperatures, (2) nature of atmospheric processes between the water vapor source and the deposition site, (3) local conditions such as diffusion and densiflcation which modify the preserved 6tSO signal, (4) the surface elevation of the deposltlonal sxte and (5) the latitude of the site (see Dansgaard et al., 1973 and Bradley, 1985, for reviews). Although the quantitative rela-

tlonShlp between atmospheric temperatures and 61SO and their spatial representativeness are still under discussion, /ilSo continues to be widely applied as a proxy for climate and in particular, for temperature (Dansgaard et al , 1973, Jouzel et al., 1983; Thompson et al , 1986 and Peel et al , 1988)

Figure 2 provides a global perspective of the decadal and centennial variations for the last 500-1000 years from five diverse sites Here isotopically heavier, or warmer periods, are shaded while isotopically lighter, or colder periods, are unshaded There is a large dwers~ty in detail not unhke that expected if five widely dispersed meteorological stations were used to reconstruct global temperatures However, several common features are evident: (1) the isotopic variability is similar in the two northern hemisphere sites in- cluding a rather pronounced - 180-yr oscillation m the Dunde record, (2) the s lSo records from the tropical Quelccaya Ice Cap, Peru and South Pole, Antarctica show strong similarities includ- ing a "Little Ice Age" ( ~ A D 1530-1900) which is characterized by more negative 61SO values; and (3) all non-Antarctic sites are isotoplcally less negative (warmer) during the twentieth century. Of these five records, the Slple Station record is unique, exhibiting less negative S~O and hence presumably warmer conditions during much of the "Little Ice Age" (Mosley-Thompson et al., 1990; Thompson, 1992). Both Antarctic records are lSOtOplcally hghter in the twentieth century, but wlth differing trends for the last 30 years. At South Pole s i l o suggests a continued cooling, while at Siple Station s lSo suggests a warming trend since the 1960's, consistent w~th the warmmg trend in the Antarctic Peninsula from 1960 to 1990 (Mosley-Thompson et al , 1991; Peel et al , 1988)

Central Asia

Two ice cores to bedrock from the Dunde Ice Cap (38006 'N, 96024 'E, 5325 masl), Qihan Moun- tains, on the northeastern margin of the Qing- hal-Tibetan Plateau of China provide a detailed record of Holocene and Wisconsln-Wtirm late glacial stage (LGS) climate changes m the sub-

148 L G T H O M P S O N E T A L

Dunde Ice Cap, China Core 1, 50 Y e a r A v e r a g e s

1 8 8 0 (%0)

- 1 3 - 1 2 -11 - 1 0 - 9

DUNDE ICE CAP, CHINA CORE 3, 50 YEAR AVERAGES

8180(%,) - 1 4 - 1 3 - 1 2 - 1 1 - 1 0 - 9 0 - ~

O..

m

k. c0 O >-

~'=-10.94

Fig 3 Fifty-year averages of 3180 for the last 12,000 years from Core D-1 on the Dunde Ice Cap, China, are compared w]th the projection (shaded) of the most recent 50-yr average, 1937-1987 Vmwed from the long-term perspectwe, 1937-1987 ~s the warmest 50-yr period since the end of the last glacial stage

Q. m v r r < IJJ >-

1000-

2000-

3000-

7000

8000

9000-

10,000-

trop]cs (Thompson et al., 1988b, 1989, 1990; Xie

et a l , 1989; Yao et a l , 1990; Yao and Thompson,

1992) The ice core records provide a long- term

perspective of climate change. In Fig. 3, 50-yr

averages of 8 tSo for the last 12,000 yr are pre-

sented for core 1 D-1 These data, based on the

analysis of 3280 samples, show both colder and

warmer periods since the termlnat]on of the last

glacial stage. The shaded por t ion reflects the

6~80 average for the 50-yr per iod from 1937 to

12,000 ~ '= -1 1 2

Fig 4 F]fty-year averages of 61~O for the last 12,000 years from Core D-3 on the Dunde Ice Cap, China, are compared with the projection (shaded) of the most recent 50-yr average, 1937-1987 Viewed from the long-term perspectwe, 1937-1987 ~s the warmest 50-yr period since the end of the last glacial stage

'RECENT WARMING ' EVIDENCE FROM TROPICAL ICE CORES WITH EMPHASIS ON CENTRAL ASIA 149

1987 and demonstrates that it is ~sotopically, the warmest 50-yr period in the record. However, these 50-yr averages must be viewed cautmusly as the most recent 50-yr average includes 320 samples, while the lower 50-yr averages, are based on 2-5 samples, a result of denslfication and thin- ning due to flow

Core D-3 was also drilled on the Dunde Ice Cap and unhke core D-1 whtch was returned entirely as bottled samples, the lower 80 m were returned to OSU as ice Thus, core D-3 was sectxoned into much smaller samples which pro- vlded a higher tame resolution. The resulting 3]sO profile (Fig 4) represents the 50-yr averages of 7045 samples, more than double that in core D-1. In core D-3 the most recent 50-yr average represents 384 samples while the lower 50-yr averages represent 4-10 samples. Comparison of the two 6180 records demonstrates their reproducibility and provtdes strong evadence that the most recent 50-yr period, as well as the last century (1887- 1984) are isotoplcally the warmest periods m the 12,000-yr record. When viewed from the long- term perspectwe provided by these cores, the most recent warming does indeed appear to be stgnlflcant, exceeding the maximum warmth of

the earher Holocene periods in central Asm. Me- teorological observations from Delingha station 100 km to the southeast, the closest station to Dunde, shows a 1.2°C warming since the mid- 1950s, when the record began

Large-scale variations of 6 ~ O over the past 40-180 yr in central Asia appear to be directly related to similar large-scale trends m temperature. The 5-yr running means of 3~SO from the Dunde Ice Cap are compared in Fig 5 with 5-yr running means of Northern Hemisphere temperatures (Hansen and Lebedeff, 1987). For the period from 1895 to 1985 the correlation coefficient (r) is 0.5 (slgmficant at the 99.9% level). This correlation suggests that the Dunde Ice Cap 6 t80 should serve as a good proxy for larger-scale temperature variations.

The recent 8~80 enrichment, interpreted as a function of recent warming m this region, is not restricted to the Dunde Ice Cap In 1990 two cores, 16-m and 20-m, were drilled on the summit of the Gregorlev Ice Cap in the Tian Shan of Kxrgh~zia (42°N, 78°E, 4660 masl). The prelimi- nary time scale was cahbrated using a d~stinct time horizon provaded by the 1963 Beta peak in core 1 (Fig. 6). The 3~SO records from both cores

(2) =o

co

-6

-9

-10

-11

Oxygen Isotopes vs. Northern Hemisphere Temperatures

1 8 9 5 - 1 9 8 5

- - 5 - Y e a r Runn ing M e a n

Dunde Ice Cap, China, 1987 D-1 Oxygen Isotopes

Northern Hemisphere Temperatures

- 1

- 2

- 3

- 4

~ - -5

4 o

3 u~

2 t~

Q. 1 E

I--- 0

"r-

E ..c ~Z o

- - 12 . . . . I . . . . I . . . . I . . . . f . . . . I . . . . I . . . . I ' ' I ' l . . . . ~ . . . . I . . . . I . . . . f . . . . I . . . . I . . . . I . . . . I . . . . I ' ~ r

1895 1905 1915 1925 1935 1945 1955 1965 1975 1985

Tem0eralure Source Hanaen & Lebedeff, 1B87 Years

F~g 5 F i v e - y e a r r u n n i n g m e a n s o f ~ x 8 0 f r o m D u n d e core and N o r t h e r n H e m i s p h e r e t e m p e r a t u r e s f r o m 1895 to 1985


A

V

"1- I-- 13_ LU C3

GREGORIEV ICE CAP, TIAN SHAN 1990 CORE 1 1990 CORE 2

(dph /kg X 10 -2) ~180(%o) I I I I J I

0- ' 0 m

4

1oi

81%(%o) I I I

[.

3 X=-10 1

- - - - -98

Fig 6 Total /3-radioactw]ty and ~51SO from the 20-m Core 1 and 61~O from the 16-m Core 2 drilled m 1990 from the summit of the Gregonev Ice Cap, Karghizm

(Fig. 6) exhibit a significant enrichment of ]SO (less negative 6~80) in the upper layers, possibly m response to recent atmospheric warming. In addition, the parttal loss of the seasonal ~180 variations, especially in the near-surface layers, arises from both isotopic diffusion and meltwater percolation which are consistent with a recent warming in the vicinity of Gregoriev. A significant amount of melting and runoff were observed dur- mg the 1990 summer season, as well as a few large pools of standing water on flat areas of the ice cap below 4500 m elevation

A recent warming in this area is supported by other evidence. For example, the temperature profile (Fig. 7a) measured on Gregoriev indicates an increase of 2.2°C at 20 m when compared with the 20-m temperature measured in 1962 by Soviet expedttions (Deklkh, 1965; Mikhalenko, 1989). In fact, the temperature increase may be slightly greater as the 1962 measurement was at a s~te 200-m lower than the 1990 measurement. On glaciers where refreezing of meltwater occurs, as

(a)

.~ 5 =, o .:5, 10 n- O

~ 15 "1"

a.

~ 2o

25

GREGORIEV ICE CAP BOREHOLE TEMPERATURES

1962 1990 i i i J i i

- 6 - 4 - 2 0 BOREHOLE TEMPERATURE (°C)

(b)

-7 .0

~.-75 : 3 I..-

r r

~ - 8 o :E ILl I '"

-8_5 1

SURFACE TEMPERATURE 4 0

T=an Shan, USSR (3614) B" '" • /

• " • Imgha, Chin, (3200)

,30 4'0 50 60 7'0 80 YEAR (A D )

g 3 5 ~

3

<

3 o ~ w

2 5 1 9 9 0

(c)

-6

DECADAL AVERAGE ISOTOPE RATIOS

- 8

--~ -10 o a0

co -12

Dunde Ice Cap, China

/2 Gregonev Ice Cap, USSR

-14

1930 4'0 5'0 6'0 7~0 8'0 1990 YEAR (A D )

Fig 7 (A) Borehole temperatures measured m a 20-m borehole at 4400 m elevation during a 1962 Soviet expedition are compared with temperatures measured at 4660 m elevation in 1990 by a joint U S A - U S S R expedition (B) Decadal averages of surface temperatures measured at the USSR Tmn Shan Meteorological Station (3614 m e|evatlon), 6 km from Gre- gorlev Ice Cap and from the Chinese Dehngha meteorologxcal staUon (3200 m elevation), 100 km from the Dunde Ice Cap (C) Decadally averaged 61SO values in two summit cores from the Gregonev Ice Cap, former USSR, and from the Dunde Ice Cap, China.

" R E C E N T W A R M I N G " E V I D E N C E F R O M T R O P I C A L ICE C O R E S W I T H E M P H A S I S O N ( E N FRAI A S I A 151

m the case of the Gregorlev Ice Cap, 20-m temperatures will generally exceed the mean annual air temperature (Paterson, 1981). Therefore, the 2.2°C warming must be considered as an upper l imi t .

Further evidence of recent warmmg in thls region comes from near-surface temperature observations at the Tian Shan Meteorological Sta- tion (3614 m elevation, nearest station to the Gregoriev Ice Cap) and from the Delingha mete-

orological station (3200 m, nearest station to the Dunde Ice Cap). Figure 7b illustrates a warming of 0.5°C over the past 60 yr at the former USSR station and of I 2°C at the Chinese station Fig- ure 7c shows the decadal averages of ,~XO from both the Gregoriev and the Dunde Ice Caps, demonstrating an l~o enrichment of approxi- mately 1%c in precipitation over the last 40 yr. Isotopic enrichment generally indicates an increase m condensation temperature The stmllar-

8t a 0 (%) I I I I

I I I O0 ~ T ~ ~ © "~ o~

~ . , = 1989S

~ 1988S 2 5 __ ~ " 1987S

1986S E 1985S c - - 1984s

5 c 1983s Q. ® 1982s E3 1981S

7_5 1980S

x ---- - -10 6%o

1979S

Gul iya Ice Cap, China 1990 Core 1

8 D ( % 0 ) T o t a l P a r t i c l e s * _,' ~ ( l o 5 ) ( l o 3 ) ? o " " " "

0 0 0 0 0 ~ ~ CO 01 "-4 ~ 0 0 0 0 0 0 ' " ' ' ' ' ' ' ' 0.( o.o.

2 5 ~ 2"= 2_5

¢=RE =

5 ~ 5 c 5.0,

m

i 7 5 '=~ 75 7 5

x = - 6 9 3 x = 31 × 10 ° x = 77 X 10 °

L a r g e P a r t i c l e s * *

0 o o

~) 2.5

E e..

~ 5 o o. G)

75.

C I - (p_eq. 1-1 )

o 0 0 "L " 1 i i i f i i J

~' I ,

I

1990W

1989W 1988W

1987W 2 5

1986W

1985W 1984W 1983W 5 0 1982W 1981W

1980W

1979W 7 5

C o n d u c t i v i t y (/J.S c m - 1 )

0_0 ' ' ' ' ' ' O0 ~' I ' ' ~' ' ' ' '

N O 3- (~eq I - 1)

i c

% ,--r-

2 5 ¸

5 0

S O 4 2 - (~eq 1-1)

A -

7 5 I

x_-- 1 2 6 x = 4 1 x = 1 5 3 ~ = 2 2 9

:(< Total Particles >_ 0 6 3 t o ~ 1 6 . 0 / m ~ m D m m e l e r p e r m l o f S a m p l e ( 1 0 5 )

=k ~ Large Particles >_ 1 5 9 t o ~ 16_O/Jm m D i a m e t e r p e r m l of Sample ( 1 0 3 )

Fig 8 Measurements of 6]So, 6D, total particles (diameter >~ 0 63 p.m m diameter per ml sample), large particles (/> 5 04/.tm in diameter per ml of sample), Cl , NO~- and SO42 and conductwlty m an 8-m core from Site 1, Guhya Ice Cap, Western Kunlun, China

-30 0

A

E a .c

t-~ 10

12

14

l JR

8 mean - 19.4%o

1991

-2O I I

1991 _ _ J

5 q__

,A, I 8 mean-17.4%,

Quelccaya Ice Cap, Peru 818 O%0

1976 1976 Summit Dome Middle Dome

-10 -30 -20 -10 -30 -20 I I I L 11976 _ I I . ~

~ ~ ' ~ p 1 9 7 5

~ 1 9 7 4

1973

1972 , f 1971

f 1970

(B) 1969

-5700

12

( C ) 14

8 mean - 1 9 . 2 % o

16

1976 South Dome

-10 -30 -20 -10 | | t | 0 '

2'

4

6

10

(D) )


0 . . . . 4 n ~ 0

- 5 6 0 0

O3 - 5500 =

E

-5400

- - 5300 Fig 9 Tropical Quelccaya Ice Cap, Peru oxygen Isotope profiles from a core drdled m 1991 on (A) the summit dome and cores drilled in 1976 on (B) the summit dome, (C) the middle dome and (D) the south dome (E) N-S cross-sechon of the Quelccaya Ice Cap dlustratlng the locahons from which the 6180 records m (A)-(D) were drdled and reveals the 130-m rise m the percolation

zone from 1976 to 1991

' R E C E N T W A R M I N G " E V I D E N C E F R O M T R O P I C A L ICE C O R E S W I T H E M P H A S I S ON C E N T R A L ASIA 153

lty of the isotopic enrichment on both the Gre- goriev and Dunde Ice Caps, 1700 km apart, cou- pled with both the local meteorological observations and the 2.2°C increase in borehole temperatures on Gregoriev, strongly suggest a recent warming in this part of central Asia.

Recently, a new drilling program has been initiated on the Guhya Ice Cap (35°17'N, 81°29'E, 6707 masl), located in the western Kunlun Moun- tains (Fig 1), where conventional long-term climatic data are completely absent. An 8-m ice core was collected and analyzed for mtcropartlcle concentrations (MPC) and size distributions, stable isotopic ratios (~180 and ~D) and selected chemical species (Fig. 8) Annual accumulation

over the last ten years has averaged 650 mm H 2 0 equivalent. On Guliya, high MPC is associated with less negative filSo during the summer/fal l seasons. Seasonal variations in 6t80 and 6D, total and large-diameter insoluble particle concentrations and liquid conductivity (soluble dust) are reasonably well preserved and will allow ac- curate dating of the deep cores recovered in 1992. The annual peak concentrations of Cl , NO 3 and SO42- are generally associated with more negative 6180 and 3D which reflect winter precipitation This relationship is strongest for NO 3 variations (The isotopic records reveal an interesting trend with values becoming progressively less negatwe toward 1990, presumably re-

Retreat of Quelccaya Ice Cap Margin 1977-1991

1977

Boulder Ice Margin Boulder Ice Margin

1983 1991

1979 Boulder Ice Margin

Fig 10 Boulder and Quelccaya Ice Cap western margin as seen in 1977, 1978, 1979, 1983 and 1991

1991


flectlng warmer condensation temperatures or a warmer moisture source region). The average 6180 for 1980-1985 IS - 11.46%o, while the average for 1986-1990 is -9 .53%0, a 1 9%0 increase. The 6D values show a similar trend, with an increase of 20%0 since 1985. These lSOtOplC results are very consistent with those from the Dunde Ice Cap on the eastern edge of the Plateau (Thompson et al., 1989, 1990) and the Gregonev Ice Cap. A potentially important relationship exists between the trends over the last ten years In 6180 and CI- , NO 3 and SO4 z-. Between 1980 and 1990 6180 has become progressively less negatwe while CI , NO 3 and SO4 z- concentrations have decreased.

Quelccaya Ice Cap, Peru

A 1500-year record extracted in 1983 from the Quelccaya Ice Cap (Thompson et al., 1986, 1988a) in the tropical Andes of Peru (13°56'S, 70°50'W ', 5670 masl) was recently updated by the analysis of a new core drilled in October, 1991. Figure 9 illustrates the 6180 records from ice cores drilled in 1976 on Quelccaya at three locations: the 5670 masl summit dome (Fig. 9B), the 5540 masl middle dome (Fig. 9C) and the 5480 masl south dome (Fig. 9D). The 1976 records are compared In Fig 9 with results from the upper 15 m of a 21.4-m core drilled on the summit dome in 1991 (Fig 9A). In the 1976 south and middle dome cores the annual 6180 signal is lost below 6 m due to percolation of meltwater. However, the 6180 record from the 1976 summit core shows well- preserved seasonal variations throughout its length with a mean 6180 value of -19.40%o. When the 1976 6180 record is compared with the 1991 6180 record from the same site (Figs. 9A,B) it is evident that (1) the 6180 values have been enriched by 2%0 and (2) annual variations in 6180 are no longer preserved at this site. If the survey to assess the quality of the record preserved in this Ice cap had been conducted in 1991, Quelccaya would have been ehmmated as a possible site for acquisition of a long ice core paleoclimatic history.

The impact of the recent warming on the Quelccaya Ice Cap can be seen in the massive

and accelerating retreat of the margin of the tce cap as illustrated for 1977, 1978, 1979, 1983 and 1991 in Fig. 10. If the current warming trend persists, then many of these unique tropical and subtropical archives of climate and environmental history are in imminent danger of being lost.

Conclusions

Model results of Hansen et al. (1988) suggest that the central part of the Asian continent may be one of the first places to exhibit an unambigu- ous signal of the anticipated "greenhouse warming" as it is far from the mitigating influences of oceans. Certainly the Dunde ice core results suggest that the recent warming on the Tibetan Plateau has been substantial in relation to the Holocene record. Recent radiosonde data from southern India (Flohn and Kapala, 1989) show that, in fact, the average tropospheric temperature has increased nearly I°C since 1965. Wang et al (1991) in model simulations found that trace gases increased the temperature from the surface all the way up to 30 km, with peak warming of > 6°C near 8 km. All the ice core records pre- sented here come between 5 and 7 km. These results would support their findings, indicating large warming of the tropical tropopause. Even- tually more robust temperature-isotope transfer functions must be developed for the Tibetan Plateau and indeed, for all regions of the Earth where ice core ISOtopic records p r o v t d e a proxy indication of temperature.

Ice core climatic and environmental records from low, middle and high latitudes greatly increase our knowledge and understanding of the course of past climatic events, which is essential for prediction of future climatic oscillations, which may or may not be dominated by increasing greenhouse gas concentrations The forcing fac- tors, internal and external, which have operated m the past will continue to operate and influence the course of events (Grove, 1988) The Dunde Ice Cap cores provide the first Ice core record of the Holocene- la te Pleistocene chmate from the subtropics The stable isotope record indicates that the last 60 yr on the Tibetan Plateau have been the warmest period in the entne record.

"RECENT WARMING" EVIDENCE FROM TROPICAL ICE CORES WITH EMPHASIS ON CENTRAL ASIA 155

Other climatically unique periods which are global in scale such as the "Little Ice Age" also must result from large-scale chmatic forcing. The ex- pression of the "Little Ice Age" in any record is quite variable and appears to be more distinct in higher elevation sites, such as Dunde, Quelccaya and South Pole than in the lower elevation polar sites of Camp Century and Siple Station We suggest two reasons for this. F~rst, climatic change is probably a function of elevation. Recent results from Guliya, Dunde and Quelccaya Ice Caps suggest that the recent warming may be expressed more strongly and earher at higher elevation low latitude sites. Secondly, it is likely that subtle changes in climate like the "Little Ice Age" are more clearly recorded farther from the mitigating influence of the oceans.

The tropical and subtropical ice core records may potentially provide long records of E1 Ntfio/Southern Oscillation events and monsoon varlabihty and thus provide Insight to the variations in the magnitude and frequency of these global-scale events. Moreover, a global array of cores can contribute to estabhshing a record of changes in global precipitation over the last 1000 yr and possibly provide records of geographical variations in the concentrations of selected greenhouse gases. These data also make it very clear that some of these unique archwes of climate and environmental history are in imminent danger of being lost forever if the current warmmg trend persists.

Acknowledgements

Supported by the National Science Founda- tion's Office of Chmate Dynamtcs and the Divi- sion of Polar Programs (ATM-8519794, ATM- 8916635, DPP-9014931), the National Oceanic and Atmospheric Administration (NA-89AA-D- AC197), the National Geographic Society (3323- 86 4309-90 and 4522-91), The Ohio State Univer- sity, the Academia Simca of China and the Na- tional Science Foundation of China and the former USSR Academy of Sciences. The efforts of all American, Chinese, Peruvian and Soviet field participants are gratefully acknowledged. We thank John Bolzan for his contribution to the

establishment of time-scales for the Dunde cores and the Polar Ice Coring Office for drilhng the deep ice cores. John Nagy prepared the tllustra- tions and Kathleen Doddroe typed the manuscript This is contribution no. 790 of the Byrd Polar Research Center

References

Bradley, R S, 1985. Quaternary Paleochmatology_ Allen and Unwm, Boston, 472 pp

Dansgaard, W, Johnsen, S.J., Clausen, H B and Gundestrup, N, 1973 Stable ISOtOpe glaciology_ In: Stable Isotope Glaciology Medd Gronl, 197 1-53

Dekikh, A N, 1965 Temperature regime of fiat-top glaciers (using Gregor6v glacier as an example) Glaclologlcal in- vestigations m the Tlan Shah, Grudze, 1965 Lett Tlan Shan Glaclol Stn, 11 32-35 (m Russian)

Flohn, H and Kapala, H, 1989 Changes of tropical sea-air interaction processes over a 30-year period_ Nature, 338 244-246

Grove, J M, 1988_ The Little Ice Age Methuen, London, 498

PP Hansen, J and Lebedeff, S, 1992_ Global surface air temper-

atures update through 1987 Geophys Res Lett, 15(4): 323-326

Hansen, J , Fung, I , Lacles, A.L, Rind, D, Lebedeff, S_, Ruedy, R_, Russell, G. and Stone, P, 1988 Global trends of measured surface air temperature J Geophys Res, 894 92(D8) 9341-9364

Johnsen, S J , Dansgaard, W., Clausen, H B and Langway, C C, 1970 Chmatlc oscillations 1200-2000 A D Nature, 322 43O--434

Jouzel, J , Merhvat, L, Petit, J R and Lonus, C, 1983. Chmatlc information over the last century deduced from a detailed ISOtOpic record m the South Pole snow J_ Geo- phys Res, 88(C4)- 2693-2703

Mlkhalenko, V W, 1989. Peculiarities of the mass-exchange of flat-summit glaciers of internal Tlan-Shan Data Glaclol, Stud, 65 86-92 (in Russian)

Mosley-Thompson, E , Thompson, L G, Grootes, P and Gundestrup, N, 1990 Little Ice Age (Neoglaclal) paleoen- vlronmental conditions at Slple Station, Antarctica Ann Glaclol_, 14 199-204

Mosley-Thompson, E_, Dal, J_, Thompson, L G, Grootes, P M, Arbogast, J K and Paskievltch, J F , 1991. Glaciolog- lcal studies at SIple Station (Antarctica) Potential ~ce core paleochmatlc record J Glaciol, 37(125) 11-22

Paterson, W S B, 1981 The Physics of Glaciers Pergamon Press, Oxford, 380 pp

Peel, D A., Mulvaney, R and Davlson, B M., 1988 Stable isotope/air-temperature relationships m ice cores from Dolleman Island and the Palmer Land Plateau, Antarctic Peninsula Ann Glaclol, 10:130-136

Thompson, L G, 1992 Ice core evidence from Peru and China In R S Bradley and P D Jones (Editors), Chmate since A D 1500 Routledge, Chapman and Hall, London, pp 517-548


Thompson, L G , Mosley-Thompson, E, Dansgaard, W and Grootes, P M, 1986. The Little Ice Age as recorded in the stratigraphy of the tropical Quelccaya Ice Cap Science, 234 361-364

Thompson, L G , Davis, M, Mosley-Thompson, E and Lm, K, 1988a Pre-Incan agricultural activity recorded m dust layers m two tropical ice cores Nature, 336 763-765

Thompson, L G, Mosley-Thompson, E , Wu, X, and Xle, Z , 1988b Wlsconsm/Wurm glacial stage ice in the subtropical Dunde ~ce cap, China GeoJourna[, 17(4) 517-523

Thompson, L G, Mosley-Thompson, E , Davis, M E, Bolzan, J F , Dal, J , Yao, T_, Gundestrup, N_, Wu, X, Klein, L and Xie, Z_, 1989 Pleistocene chmate record from Qlng- hal-Tibetan Plateau ice cores Science, 246 474-477

Thompson, L G, Mosley-Thompson, E , Davis, M E_, Bolzan, J , Dal, J , Klein, L, Gundestrup, N, Yao, T , Wu, X and Xie, Z , 1990 Glacial stage ~ce core records from the

subtropical Dunde Ice Cap, China Ann Glaool, 14 288-297

Wang, W C , Dudek, M P , Xin-Zhang, L and Kaehl, J T , 1991 Inadequacy of effective CO 2 as a proxy in simulating the greenhouse effect of other radmtlvely active gases Nature, 350 573-577

Xie, Z , Wu, X, Yao, T and Thompson, L, 1989 Chmatlc and enwronmental record from ~ce cores in Dunde ice Cap Quat Sci, 2 134--142 (in Chinese with Enghsh abstract)

Yao, T , Xle, Z , Wu, X, Thompson, L G , 1990 The Little Ice Age as recorded in the Dunde Ice Cap Scl Sin., ll(B) 1196--1201

Yao, T and Thompson, L G , 1992 Trends and features of climatic changes in past 5000 years recorded by Dunde ice core Ann_ Glaclol_, 16 21-24

“recent warming”: ice core evidence from tropical ice cores with emphasis on central asia

Documents