s6lfsson

Is6lfsson

ULL, No. 45, 1998

Interpretation of glacier variations in Iceland 1930-1995

Tomas J6hannesson 1 and Oddur Sigurosson

National Energy Authority, Hydrological service Grensasvegi 9, IS-108 Reykjavik

Abstract- Glacier variations in Iceland since 1930 show a clear response to mriations in climate during this period. Most non-surging glaciers retreated strongly during the early half of the mon­itoring period,following the warm climate between1930 and 1940. A cooling climate after I940 led to a slowing of the retreat and many glaciers started to advance around 1970. A warming of the climate since about I 985 has led to an increased number of retreating glaciers in recent years. The mriations of non-surging glaciers in Iceland since I930 appear to be caused primari­ly by variations in temperature as there are no long term variations in precipitation over this pe­riod. Variations of surge-type glaciers are dominated by the surge events. Climate variations do, nevertheless, seem to play a role in obsen'CCllong term variations of some surge-type glaciers al­though they are more difficult to interpret than for the non-surging glaciers.

INTRODUCTION

C )n the basis of the original sources, observations Ill glacier variations in Iceland from 1930 to 1995 l~:tvc been checked, corrected and computerised by .·;q•mosson (this volume). In addition to measurements "' 1 he advance and retreat of the monitored glacier i<lll).!.liCS, this compilation includes descriptions and lo­' ;II ion maps of the glaciers and graphs of the varia­lt()fls at the individual measurement locations. The ptt•scnt paper presents selected records from this new d;tla set and summarises the advance/retreat records 11 'Jill non-surging glaciers in light of measured varia­lt<lllS in the climate of Iceland since 1930.

Icc caps and glaciers cover more than 10% of the • 11 ,·a of Iceland and they receive on the order of 20% "' I he total precipitation that falls on Iceland. When ilw glaciers are close to a steady state, they thus con­It ihutc on the order of 20% to river runoff and infiltra­''' 111 to groundwater aquifers in Iceland. The relative tlllportance of glacier runoff is greater in highland .tn•;ts. Glacier runoff is, therefore, particularly impor-

tant for hydropower purposes, which are of great eco­nomic value for Iceland.

It is estimated that during the next decades the mean surface air temperature of the Earth will rise at a rate between 0.1 and 0.35 oc per decade due to in­creasing concentration of C02 and other trace gasses in the atmosphere (Houghton and others, 1996). If rea­lised, this warming will have pronounced effects on glaciers and ice caps and lead to large runoff changes in glaciated areas. Climatic warming on the order of 0.3 oc per decade can temporarily increase glacier runoff from some Icelandic ice caps and glaciers by more than 50% and lead to a decrease in volume by approximately 40% of some glaciers over the next century (J6hannesson and others, 1995; J6hannesson, 1997; Saolthun and others, in press) .

The advance and retreat of glaciers is an indicator of climate changes and reconstructions of past varia­tions of glaciers from geological evidence or historical observations are an important tool for deductions about past climate. Infmmation about many past cli­mate changes has in part been deduced from records of

1 l'tcscnt affiliation: Veourstofa Islands (Icelandic Meteorological Office), Bustaoavegi 9. IS-150 Reykjavik, Iceland

II >KULL, No. 45, 1998 27

glacier variations. Future climate changes will be mon­itored instrumentally and this will of course give much more quantitative infotmation about the changes than is available about past climate changes. Nevertheless, it is important to monitor glacier variations associated with future climate changes (Haeberli, 1995). The cli­matic signal contained in records of glacier variations has a large area coverage whereas most meteorological data consist of point measurements. Moreover, the glacier records document climate changes that occur in relatively high mountain areas which are often poorly represented in the meteorological data. Finally, it is im­portant for the interpretation of past records of glacier variations to obtain good data on glacier variations as­sociated with climate changes which are well defined by measurements.

The interpretation of climate from records of glacier variations is problematic in a number of ways. The advance and retreat of surge-type glaciers is un­related to climate and records from such glac iers must be eliminated from a data set that is to be used for de­ductions about climate. The response of non-surging glaciers to climate changes is influenced by the re­sponse time of the glaciers, which leads to a delay be­tween a climate change and the associated response of the glacier terminus (cf. J6hannesson and others, 1989).

Glacier variations are not only important because of local hydrological effects or as indications of cli­mate changes. Increased runoff from glaciated areas is important for future global sea level rise that may occur as a consequence of future climate warming

Fig. 1. S6lheimajokull, S-Iceland, has one of the most complete records of glacier front variations since 1930. Photo: Oddur Sigurosson , 30 Oct. 1985. - S6/heimajokull.

28 JOKULL, No. 45, 1998

from records of number of ways.

pe glaciers is un­uch glaciers must to be used for de­se of non-surging tenced by the re­ads to a delay be­jated response of sson and others,

mportant because ndications of cli­m glaciated areas ~vel rise that may climate warming

930. Photo: Oddur

JLL, No. 45, 1998

1 \V:~nick and others, 1996). Recent worldwide compi­Lil iDtls of glacier mass balance (Dyurgerov and kin, 1997a,b) indicate that melting of small glaciers

"'I l·:arth (excluding the ice sheets of Greenland and 111arctica) contributed 0.25 ± 0.1 mm a' to global

•• ·:1 level rise in the period 1961 to 1990.

<;LACIER VARIATIONS IN ICELAND SINCE 1930

Records of glacier variations in Iceland show a ' l.·ar relationship with climate and such records have IH'I'Il used as an indicator of historical climatic varia­''' illS, before regular weather monitoring was started 111 Iceland around the middle of the nineteenth centu­' v 1 1>6rarinsson, 1974). Many glaciers in Iceland start­··d retreating from their Little Ice Age maximum be-

(-820000,700000)

0 100 km leah sEd

(-820000,300000)

l:ig. 2. Location map. - Yfirlitskort.

J0KULL, No. 45, 1998

tween 1850 and 1900 and the rate of retreat became quite rapid after 1930. As the climate gradually cooled after 1940, the retreat of the glaciers slowed down and many glaciers started advancing after 1970 (P6rarinsson, 1974; Bjornsson, 1979). After about 1985 the climate has become warmer and many glaciers have started retreating again.

Figure 1 shows a photograph of the outlet glacier S6lheimajokull, which flows to the south from the ice cap Myrdalsjokull, S-Iceland (Fig. 2), and Figure 3a shows the advance and retreat of S6lheimajokull from 1930/31 to 1994/95. Figure 3a shows a rapid retreat of S6lheimajokull between 1930 and 1960 followed by an advance which has slowed down markedly in re­cent years.

S6lheimajokull is a non-surging glacier which is directly affected by climate changes (Sigurosson and

(-260000,700000)

rnsv 19980313

(-2soooo.3ooooo) 1

29

Jonsson, 1995). Variations of surge-type glaciers are not directly induced by climate changes in the same way as for non-surging glaciers. Surge-type glaciers may, however, be expected to reflect climate changes on time-scales longer than the interval between surges , since accumulated mass balance changes over long time intervals do affect the total size of the glaciers significantly. Figure 3b shows the advance and retreat of Mulajokull (cover photograph), which is a surge­type outlet glacier from the ice cap Hofsjokull, central Iceland (Fig. 2). Mulajokull has in recent decades surged with irregular 7 to 18 year intervals. The surge events in 1954, 1972, 1986 and 1992-1993 are, of course, the most prominent features of the record from Mulajokull , but the longer tetm variations on a time­scale of several decades are qualitatively similar to the variations of S6lheimajokull in Figure 3a.

Figure 4 shows the relative number of advancing and retreating non-surging glacier termini in Iceland from 1930/3 1 to 1994/95 . The figure is based on mea­surements at 19 locations at the following 18 glaciers (Fig. 3 shows the location of the main ice caps ; Sigurosson, this volume, includes location maps of the individual glacier tongues):

• Outlet glaciers from bnefajokull , SE-Iceland: Svfnafellsjokull, Virkisjokull, Falljokull, Kvf<hjokull , Hrutarjokull, Fjallsjokull.

• Outlet glaciers fro m Vatnajokull, SE-Iceland: Morsatjokull , Skaftafell sjokull, Br6kmjokull, Skala-

Ko r-~~~--L-~~~-L~~SL_ ~~~II~ - olheimaj6ku 1il ~ Qj

-ai u c

"' > 1il Q)

> .B :; E ::J

0

a 1930 1940 1950 1960 1970 1980 1990

Years

fellsjokull , Heinabergsjokull , Flaajokull (two loca­tions), Svfnafellsjokull in Hornafjorour.

• Other glaciers: Gfgjokull in Eyjafjallajokull (S­Iceland), GljUfuratjokull (N-Iceland) , Hyrningsjokull in Smefellsjokull (W-Iceland), Nauthagajokull in Hofsjokull (central Iceland), S6lheimajokull in Myr­dalsjokull (S-Iceland).

Only termini which have been measured for sev­eral decades are included in the figure. Monitoring did not start until 1932 or 1933 at many locations and the figure is therefore based on somewhat fewer data series before 1935 (8 to 14 series) than after this time (15 to 19 series) . The measurement periods for the in­dividual glaciers are given in Table 1 in Sigurosson (this volume). Measurements from glaciers which are known to surge are not included because terminus variations of surging glaciers are typically not con­trolled by climate changes . Data from mixed type glaciers, which are affected by surge events, but also seem to react to decadal variations in climate, are not included either due to difficulties in the interpretation of such records. Data from Skeioararjokull and Breioamerkurjokull in Vatnajokull are not included for this reason.

The glaciers represented in Figure 4 have lengths varying from 2 to 29 km and vertical elevation ranges from 740 to over 2000 m. They fall in categories 2 and 3 as defined by Haeberli (1995):

b

1. "The smallest, somewhat static, low-shear-

1930 1940 1950 1960 1970 1980 1990

Years

0 0 (\J

Fig. 3. Yearly advance and retreat of S6lheimaji:ikull, "vesturtunga" (a) , and MU!aji:ikull , "suour", (b) from I 930/3 1 to 1994/95 (histograms) and the cumulative advance/retreat of the glaciers since 1930 (dashed curves). The ticks on the x-axis indicate the beginning/end of the measurement intervals (i.e. the tick marked I 960 is at the end of the in tervall 959/60 and at the begirming of the interval 1960/61 which is in the fall of 1960). - Breytingar a S6lheimajokli,, vesturtunga " (a) og Mulajokli , ,suour" (b), a tfmabilinu 1930/31 til 1994/95 (stoplarit) og heildarbreyting a stoou jokulsporoanna s foan 1930 (slirnir ferlar). Hokin a x-asnum syna upphaf og endi mcelit fmabilanna (p. e. hakio sem merkt er1960 synir endi mceliarsins 1959/60 og upphafmceliarsins 1960/61 haustio 1960).

30 JOKULL, No. 45, 1998

okull (two loca-3ur. ijafjallajokull (S­), Hyrningsjokull authagajokull in majokull in Myr-

neasured for sev­gure. Monitoring any locations and ewhat fewer data 1an after this time Jeriods for the in­: 1 in Sigurosson ~laciers which are Jecause terminus )'pically not con­'rom mixed type e events, but also n climate, are not the interpretation ioanirjokull and are not included

re 4 have lengths I elevation ranges II in categories 2

tatic, low-shear-

1980 1990

1930/31 to 1994/95 e x -axis indicate the and at the beginning

94195 (stoplm·it) og di ma:/it(mabilanna ).

JLL, No. 45, 1998

,,~I ;:! <0 Jl ;< 0 :~> "<:t

"' ' 0 !) ('\j.

' :; 0 ··ll,_Lilll~~:::':'~~~:::':'~~:::':'~~~:':"':~~~_j ''1,

1930 1940 1950 1960 1970 1980 1990

Years

·.lrcss glaciers (cirque glaciers, glaciers reservoirs) rdkct yearly changes in climate and mass balance al­most without any delay."

2. "Larger, dynamic, high-stress glaciers (moun­lain glaciers, glaciers evacuateurs, react dynamically lo decadal variations in climatic and mass balance lorcing with an enhanced amplitude after a delay of ;;l'Veral years."

3. "The largest valley glaciers give strong and most efficiently smoothed signals of secular trends with a delay of several decades."

The figure, therefore, mixes category 2 and cate­!'.ory 3 glaciers to some extent, which is unfortunate, hut a further subdivision of the glaciers is not war­ranted due to the small number of glaciers in the data set, and we have not been able to discern a difference in the response of the glaciers according to size in this small data set.

Figure 4 shows that between 80 and 90% of the I nmini were retreating in the time period 1931 to

1930 1940 1950 1960 1970 1980 1990

:I Years

Fig. 4. Percentage of advancing (dense shading) and retreat­ing (less dense shading) termini of non-surging glaciers in Iceland as a function of time from 1930/31 to 1994/95. No shading indicates stationary tem1ini. Over most of the time period shown, the figure is based on measurements at 15 to 19 locations at 18 termini (somewhat fewer tem1ini in the years 1931 to 1935). The figure is based on the measure­ments of glacier variations in Iceland tabulated in Table 2 in Sigurosson (this volume). The glaciers represented in the figure are listed in the text. - Hlutfal/ jokla sem ganga fram (jJettstrikaa) og hopa (gisin strikun) a arabilinu 1930/31 til1994195. 6strikaa aa sva:ai taknar hlutfal/ kyrrsta:ara jokulsporaa. Einungis eru teknir mea joklar sem ekki em framhlaupsjoklar. Yfir me stan hi uta t(mabilsins er myndin bygga a gognum fra 15 til 19 jokulsporaum (heldur fa:rri a arztmtm 1931 til1935). Nofn joklanna komafram r megimnali. Gognin sem myndin byggir a koma fram f toflu 2 f grein Odds Sigurassonar um joklabreytingar a fslandi ( jJeSSll tolu/Jiaai Jokuls.

1960. Although between 10 and 20% of the glacier termini were typically advancing each year during this period, all the termini, without exception, had re­treated from their 1930 positions by 1960. After 1960, the fraction of advancing termini started to rise, but after 1980 more glaciers have started retreating again. The fluctuations of the Icelandic glaciers are similar to glacier fluctuations in the Alps (Haeberli and oth­ers, 1989; Haeberli, 1995), but glacier fluctuations from other parts of the world can be quite different. For example, the retreat rate of glaciers in the Tian­shan Mountains in China did not slow down or re-

b Years

Fig. 5. Summer temperature (average of May to September) and annual precipitation (sum from October of previous year to September) at Kirkjubxjarklaustur (a) and Reykjavik (b) from 1931 to 1995 (solid curves). Dashed curves are weighted, run­ning 5 year means using the weights (1/9,2/9,3/9,2/9, 1/9) centered about each year.

Sumarhiti (meaaltal mal til september) og arszirkoma (szmzmafra okt6berfyrra drs til september) a Kirkjuba:jarklaustri (a) og f Reykjavik (b) fra 1931 til 1995 (heildregnir ferlar). Slitnir ferlar syna vegia 5 ara meaaltal reiknaa mea va:ginu (I !9,2!9,319,219,1/9)

.JOKULL, No. 45, 1998 31

Yl' l ~ l ' 111 111 111d J!)(,() I I ~ W II S tlu; l ' II SL' l ll ic ·la11d :111d i11 Ill ' lp •, ( ' IIIHI Ii li , 1111/.n.

' I 'll ' I V~ p1111 ~ • ol' I il l ' I ·nnini vmies between the llloll lltll\'d )' lll l' il'l' IOII ' II ·s as ' :111 h' seen from graphs 111' ill · i11divid11al dnln se ri es in Sigurosson (this vol­lllll •) . I u • to the smnll number of glaciers in the data s ·t, w' arc, however, unable to draw any conclusions regarding regional differences of glacier variations in Iceland during the monitoring period.

Figure 5 shows summer temperature (May to September) and yearly precipitation (October previ­ous year to September) in Kirkjubrejarklaustur S-Ice­land (a) and Reykjavik SW-Iceland (b) from 1931 to 1995. Variations in glacier ablation are primarily re­lated to variations in the summer temperature, where­as changes in the accumulation may be expected to correlate with variations in the precipitation record. There is a good con·elation between the temperature records from Kirkjubrejarklaustur and Reykjavik, as is expected, since there is in general a good coiTela­tion between temperature measurements from differ­ent meteorological stations in Iceland (Einarsson, 1991). It can, therefore, be assumed that the tempera­ture records from Kirkjubrejarklaustur and Reykjavik give an estimate of temperature variations that oc­CUlTed on Icelandic glaciers in this time period. The figure shows that the temperature was relatively high from 1931 to 1960, especially in the years 1931 to 1940. After about 1960 the climate cooled markedly, the temperature reached a minimum between 1980 and 1985 and the climate has been slowly warming since then. There are no similar systematic trends in the precipitation records in spite of large short term fluctuations (note the shifted precipitation scale in Figure 5a,b) .

The temperature and precipitation records in Fig­ure 5 indicate that glacier fluctuations in Iceland (cf. Figure 4) were primarily controlled by variations in the summer temperature. The rapid glacier retreat from 1931 to 1960 is related to high summer temper­atures in the decades after 1930. The retreat slowed down and turned to advance in the period 1965 to 1970 when the climate had become cooler and the fraction of advancing glaciers reached its maximum in the period 1975 to 1990 after the summer tempera­ture had reached its minimum around 1980.

32

In spite of the strong coiTelation between glacier va ri ations and summer temperature, which is appar­ent from figures 4 and 5, the figures show that the re­lationship between glaciers and climate is not simple. The decade from 1931 to 1940 is the warmest decade in the instrumental record in Iceland. In spite of this, between 10 and 20% of the glacier termini were ad­vancing in this period, and more than 40% of the ter­mini were still retreating during most of the very cold summers from 1970 to 1980.

REFERENCES

Bjiimsson, H. 1979. Glaciers in Iceland. Jokul/29 , 74-80. Chaohai, Liu and Han Tianding. 1992. Relation between re­

cent glacier variations and climate in Tianshan Moun­tains , Central Asia. Annals of Glaciology, 16, 11-16.

Dyurgerov, M. and M. F. Meier. 1997a. Mass balance of mountain and subpolar glaciers: A new global assess­ment for 1961-1990. Arctic and Alpine Research, 29(4), 379-391.

Dyurgerov, M. and M. F. Meier. 1997b. Year-to-year fluctua­tions of global mass balance of small glaciers and their contribution to sea-level changes. Arctic and Alpine Re­search, 29(4), 392-402.

Einarsson, M. 1991. Temperature conditions in Iceland, 1901-1990. Jokull41 , 1-20.

Haeberli, W., P. MUller, P. Alean and H. Bosch. 1989. Glacier changes following the Little Ice Age - A survey of the in­temational data basis and its perspectives. In Oerlemans, J. , Ed. Glacier Fluctuations and Climatic Change, 77-101. Dordrecht, Kluwer Academic Publishers.

Haeberli, W. 1995. Glacier fluctuations and climate change detection - operational elements of a worldwide moni­toring strategy. WMO Bulletin, 44(1), 23-31.

Houghton, J. T., L. G. Meira Filho, B. A. Callander, N. Har­ris, A. Kattenberg, K. Maskell. 1996. Climate change 1995: The Science of Climate Change. Cambridge, Cambridge University Press. 572 pp.

J6hannesson, T. , C. F. Raymond and E. D. Waddington. 1989. Time-scale for adjustment of glaciers to changes in mass balance. Journal of Glaciology, 35(121), 355-369.

J6hannesson, T. , 0. Sigurosson, T. Laumann, M. Kennett. 1995. Degree-day mass balance modelling with applica­tions to glaciers in Iceland, Norway and Greenland. Journal of Glaciology, 41(138), 345-358.

JOKULL, No. 45 , 1998

between glacier which is appar­how that the re­te is not simple. warmest decade In spite of this, ~rmini were ad-40% of the ter-of the very cold

'ikull29, 74-80. lation between re-Tianshan Moun­

'ii)', 16, 11-16. Mass balance of ~w global assess­' Research, 29(4),

ar-to-year fluctua­glaciers and their ic and Alpine Re-

tions in Iceland,

sch. 1989. Glacier '\. survey of the in­•es. In Oerlemans, wtic Change, 77-Jlishers. 1d climate change worldwide moni-23-31. :allander. N. Har­. Climate change nge. Cambridge,

D. Waddington. :iers to changes in ·5(121), 355-369. ann, M. Kennett. ling with applica-

1 and Greenland. 'i8.

.L, No. 45, 1998

l<lhannesson, T. 1997. The response of two Icelandic glaciers to climatic warming computed with a degree­day glacier mass balance model coupled to a dynamic glacier model. Journal of Glaciology, 43(143), 321-327.

:;igurosson, 0. and T. Jonsson. 1995. Relation of glacier variations to climate changes in Iceland. Annals of Glaciology, 21, 263-270.

Sigurosson, 0. 1998. Glacier variations in Iceland 1930-1995. From the database of the Iceland Glaciological Society. Jokull45, 3-25.

Sa:lthun, N. R., P. Aittoniemi, S. Bergstrom, K. Einarsson, T. J6hannesson, G. Lindstrom, P.-E. Ohlsson, T. Thomsen, B. Vehvili:iinen, K. 0. Aamodt. In press. Climate change impact on runoff and hydropower in the Nordic coun­tries, Oslo, Nordic Council of Ministers.

1>,\rarinsson, S. 1974. Samblio lands og lyos f ellefu aldir (in Icelandic). In Linda!, S. Ed. Saga islands I, 29-97. Reyk­javik, Hio fslenska b6kmenntafe1ag, Sogufelagio.

Warrick, R. A., C. Le Provost, M. F. Meier, J. Oerlemans and P. L. Woodworth. 1996. Changes in sea level. In Houghton, J. T., L. G. Meira Filho, B. A. Callander, N. Hanis, A. Kattenberg, K. Maskell. Eels. Climate change 1995: The Science of Climate Change, 359-405. Cam­bridge, Cambridge University Press.

JOKULL, No. 45, 1998

A grip

Tulkyn a joklabreytingum a Islandi 1930-1995

Ji:iklabreytingar a fslandi sfoan 1930 syna skyr viobri:igo ji:ikla vio loftslagsbreytingum a pessu tfma­bili. Flestir ji:iklar, sem ekki eru framhlaupsji:iklar, hopuou hratt a fyrra helmingi tfmabilsins, i kji:ilfar hlyindanna a fj6roa aratug aldarinnar. Eftir 1940 dr6 tir ji:ikulhi:irfuninni samfara k6lnandi veourfari og margir ji:iklar f6ru ao ganga fram um 1970. Hlynandi veourfar eftir miojan niunda aratuginn hefur leitt til pess ao margir ji:iklar hafa fario ao hopa ao nyju a siO­ustu arum. Svo viroist sem ji:iklabreytingar a fslandi sloan 1930 seu fyrst og fremst afleioing af breyting­um f hitafari a pessu timabili. Langtimabreytingar i tirkomu eru ekki miklar a timabilinu. Breytingar a stoou framhlaupsji:ikla raoast fyrst og fremst af pvi hvemer gangur er f ji:iklunum. Loftslagsbreytingar viroast p6 hafa nokkur ahrif a langtfmabreytingar sumra framhlaupsji:ikla eins og p~r birtast f m~ling­um a sti:iou ji:ikulsporoa. Breytingar a framhlaups­ji:iklum er p6 mun erfioara ao ttilka a grundvelli lofts­lagssi:igu en breytingar a Mrum joklum.

33

34

Ur myndasafni Ing6lfs fs6lfsonar

Ur Orrefum. Morsiiijokull , Skaroatindur ( 1385 m) til hregri. Myndina t6k ensk kona 1925.

Orrefafero III 1938 (sja grein: Feroir um Vatnajokul). Oddur (t.v.) i Skaftafelli ao kveoja Guomund Sigmundsson a Skeioararsandi , Morsardalur i baksyn .

JOKULL, No. 45, 1998