Annals oJ GlacioLogy 22 1996
©
Inte rn a ti o na l Gl ac iologica l Soc ie ty
Glacial cirque fortnation in northern Scandinavia
C ECILlA RI C H ARDSON AN D PER HOU.lLU ND
Deparlmellt of Ph)lsical
G'eograp/~JI ,
Slockholl17 Uni versi{y, S-106 9 1 Stockholm, Su'eden
ABSTRACT. Passglac ia ren is a sm a ll cirq ue glacie r in th e K e bn e ka ise m ass if,
no rth e rn Swed e n . It is fi"oze n to its bed O\'e r mo re th a n 700/0 or its a rea, a nd und e r
prese nt clim a ti c co nditi o ns has littl e effec t o n cirqu e fo rm a ti on . i\l o re fa\'o urab le
co nditi o ns fo r cirqu e g lacier erosio n during the H oloce ne a re of sho rt dura ti o n.
Ass uming imil a r conditi o ns during previo us int erg lac ia ls, it is sugges ted th a t fo rms
suc h as the Passglacia ren cirqu e cb'e1 o ped m a inl y durin g th e initi al ph ases of glac ia ls
wh en th ey we re pa rt of ne two rks of la rge va ll ey glaciers or of a sm a ll wa rm - based
mo untain-centred ice shee t. Passglacia ren has bee n exa min ed in o rd e r to e\'a lu a te its
eros ive capac ity a nd its assoc ia ti o n with th e subg lacia l cirq ue mo rp hology. Th e
m e th od s used a re rad a r surveys a nd direc t ice-tempera ture m eas u re m ents. Erosio n is
restri cted to a sm a ll sec ti o n of th e glacier bed , a t present res ulting in o nl y pa ni al
d ee penin g of th e cirqu e a nd erosio n of th e bac kwa ll . In cold , a rid regio ns " 'ith
ex tensive pe rm a fi"os t, sm a ll cirq ue glaciers are la rge ly fr oze n to th e bed , a nd th e refo re
ca nn o t co ntribute sig nifi cantl y to cirgu e fo rm a ti o n. I n such reg io ns glacia l erosion by
la rge r te mp era te glaciers is mo re lik ely to be th e m aj o r ca use of cirqu e excava ti o n .
INTRODUCTION
Th e a im of this stud y IS to in ves ti ga te th e ge nes is of a
cirqu e in no rth e rn Swed e n occ upi ed by th e g lac ie r
Passg lacia re n. Interpre ta ti o n or la ndfo rms ca n be used
to reco nst r uc t fo rm er cl im a tes . In th ese a ppli ca ti o ns, it is
"e ry impo rta nt to und ersta nd th e elim a ti c conditi o ns
produ cing specifi c la nd forms. Acco rdin g to th e comm o n
o pini o n , cirqu e fo rm s d e\'e lo ped success ive ly during th e
Qu a tern a ry (Enqui st, 19 16; Eva ns. 1977; Vilbo rg. 1977;
Rudberg, 1994) . Co ndi tio ns fo r erosio n by cirq ue g lac ie rs
a re gene ra ll y beli e\'ed to ha \'e bee n mos t fa \'o ura bl e
durin g th e initi a l stages of g lac ia ti o ns wh e n clrqu e
glacie rs we re fo rm ed in th e m o unta ins. Gl acia l cirqu es
a re co mm o n in th e Sca ndin a \'i a n m o unta in ra nge,
espec ia ll y in th e No rwegia n coas ta l zo ne. Th e number
of cirqu es g radu ally d ecreases towa rd s th e eas te rn pa rt of
th e m o unta in ra nge, a nd here th e genes is of cirq ues is
comp lica ted (cr. H o lmlund , 199 1; Rudbe rg, 1992 ). In
this rela tively co n tin en ta l pa rt of th e m o unta in ra nge, th e
clim a to logica l se ttin gs co nt radi c t th e id ea th a t cirqu e
glacie rs a re the m aj o r age n ts fo r cirqu e fo rm a ti o n. Th e
centra l-eas te rn pa n of th e mo ul1lain ra nge is c ha racte ri sed by a rela ti ve ly dr y clim a te a nd ex tensive perm a fr os t (Kin g, 1984). In gen eral, cirqu e g laciers in this
regio n a re ra th er sm a ll a nd have low m ass lurn ove r,
res ultin g in low e rosio n po te nti a l (And rews, 19 72 ) .
Pe rm a fi"os t m ay ex te nd d ow n to th e base o\"C r la rge
pa rts o f thin glac iers . In these zo n es w here th e ice is fr oze n
to th e bed , th e effec t of glacia l erosio n is neg ligi ble.
Th e presen t eros i,'e ca pacity of P assglac ia ren has bee n
evalu a ted by surveys of ice thi ckn ess, ve rti ca l temp era ture
profil es a nd th e a m o unt of cold- based /w a rm-based ice.
Addi tio nal info rm at io n was suppli ed by studi es of glacia l
102
mo rph ology a nd by a radi oca rbo n-d a ted sampl e (i"o m lh e
proglac ia l a rea .
GEOGRAPHICAL SETTINGS
Passg lac ia ren is a cirqu e g lac ier loca ted in th e K eb ne ka ise
m assif in th e no rth eas tern pa rt o f th e Sca ndi na \'ia n
mo unta in ra nge . K e bn e kaise is a hi g h- a ltitud e a rea o f
a lpin e c ha rac ter " 'ith summit heig hts of 1600 2 100 111
a .s.1. (Fi g . I ). Th e regio n is cha rac teri sed by pred omin a ntl y wes te rl y \\'ind s (Enqui st, 19 16 ), a nd , acco rdin g to
th e acc umul a ti o n pa tte rn o n Passg lac iare n , wind-drifted
snow is a maj o r so urce of acc umul a ti on (Fi g . 2) . Th e
ci rq ue g lac ie r has a n a rea of 0.5 km 2 a nd a su rf~lce
inclin a ti o n of a bo Llt 16°. Th e ge ne ra l ice fl ow is directed
toward s th e no rth . N ea r th e centre of th e g lac ier, th e
surface ice \'eloc it y h as bee n m eas ured a t a bo u t 2 m a I .
At prese nt , Passglacia re n fill s its cirqu e a lm os t entirely,
sugges tin g future expa nsio n \\'o uld sta rt to infill t he
d ow nstrea m tro ugh \'a ll ey, Ku o pe n ·agge.
METHODS
Th e ice d e pth a nd th e thi c kn ess of th e co ld pe rm a fros t
laye r we re m a pped using a g ro und - based hi g h-reso luli o n
rad a r. Th e radi o-ech o so und e r is based o n a Hew lell a nd
Pac ka rd Ne twor k A nalyse r (8753 B) (H a mra n a nd oth e rs,
1995 ). In this stud y, two kinds of a n te nn ae we re used ,
A ll gon 7 125. 04.05 .00 a nd A llgo n 7 104.0 1. 05 .00 , \,'ith
frequ ency ra nges of 780- 930 a nd 3 15- 375 MH z, respecti \"C ly. Th e sta n a nd e nd points of th e rad a r pro fil es \l'e re
surveyed with a theod o lite \\' ild T 2 a nd a Geod imetc r
RichardsolZ and H olmlund: Clacial cirqZle jormaLioll
Fig . I. LocaLion map wilh a conLour inLerval oj 100 m. T he 10ca.liolZ oj Passglacidrfll is indica led by all arrow.
the occ urrence a nd intensit y of sca tte red ec hoes w ith in
th e ice (H o lm lund a nd Eriksson , 1989; H a mra n a nd
A a rh o lt , 1993 ) . G eo m et ri c co rrec tio ns we re m a d e
m a nu a lly to o bta in ve rti cal d epth va lu es ri'o m m eas urem e nts perfo rm ed in th e direc ti o n no rm a l to a sloping ice
surface .
\ 'e rti cal ice-tempe ra ture pro fi les we re meas ured by
th crm isLO rs insta ll ed in two bo reho les a lo ng th e cc ntrc
lin e of th e g lac icr (F ig. 3). Th e bo reho les d id no t reac h
the g lac ie r bed . A t th e upper site (A) th ermi stors \I'ere
install ed eve ry 10 m d ow n to a d epth of 43 .5 111 , a nd a t the
lower site (B) eve ry 5 111 d own to 18 .5 m. T o esta bl ish th e
seaso n a l te mp e ra tu re \'a ri a ti o ns, m eas u rem e n ts we re
ca rried o ut in th e pe riod fr o m sprin g (Apri l) to a utumn
(Se pt em ber) .
Fig. 2. PllOtograph of Passg/acidren takm from the east .
T he cirque glacier is localed Oil Ihe easlem side oj a
mOllnlain ridge, wlzich callses a lee-side ~flecl jor Ihe
IJredominall lSV westerl)l winds . PassglaciCiren is low led
ullsLream oj a V -shaped vaLley called Kuopervagge, seen aL
the right oj Lhe phoLograph.
AGA 12A . I n th e ra d a r registra ti o ns, th e tra nsiti o n fr o m
cold to te mperate ice is m a rked by a d istin c t inc rease in
RESULTS
Th e a \'C rage ice d cpth of th c g lac ier is 25 m , a nd th e
m ax imum r ecord cd ice d e pth excee d s 80 m. Th e
max imum a rea o\,e r whi ch th e g lac ie r m ay be wa rm based , a nd thu s erosi\'e, is sho wn o n th e ice d epth m a p in
F ig ure 4. Th e bound a ry fo r th e wa rm -based p a rts of t he
g lacie r is prim a ri ly d ete rmin ed fro m ra d a r surveys. I n th e
upper pa rt of th c acc umu la tio n a rea, w he re ra d a r pro files
103
Richardsol1 alld Holmllllld: Glacial cirque /ormalioll
-'
Cl)
ro
N
1600
.s
1500
Q)
1400
u
.3
:;::;
:::(
1300
~
400
200
0
600
800
1000
Distance (m)
SOOm
~egend
-ada r survey profile s
_____ Indications of warm-based ice
"'---'---'=-""'---
Fig. 5. LOlIgillldinal jJrojlle ~f Pe/S.\glaciiirfll ji olll S0 1l11t 10
north ( a'- a" ill Figure 3) . The dashed lille indicales lite
Ihickness (If Ihe cold j)ennaj'rosl Ir!)'er. A alld B lIlark Ih e
borehol e lJOS il iOlls.
a'- - a " Position of longitudinal profile
, A Temperature profile , vertical
Fig . 3. Lowliolls 0/ boreholes alld radar jnqfiles.
ACI'ordillg 10 Ihe radar Sltn'P')!S, Illf warlll -based ire is
jiald!),. A alld lJ lIIark l/ie jJosilioll.\ oI boreholej Ior
measuremelll.I' o/ l'erlical lemjJemlllre jHoJiles. The da shed
lille ( a'- a" ) shows Ihe fmilion o.J Ihe longillldilla/ /Jro/ile
seen ill Flj;ure 5.
a rc spa rse , th e boundary is de lin ea ted by the bergsc h r und. At least 70 % o f the areal exte nt of Passg lacia re n
is fro ze n to th e bed . R a d ar sun'eys indi ca te that the
wa rm -based ice is pa tch y ( Fi g . 3 ). A lo ng itudin a l pro fi le
or Passglaciiiren and the thickness or the co ld perma(I'os t
la ye r <He show n in Fi gu re 5. The cirqu e fo rm has an
O\'erd eepe ning \\'ith a he ig ht diITerellce up to the rock
threshold of 10- 30 m. Th e permafrost is deepest a t th e
g lac ier fr on t, reac hing abo ut 50 m clown in th e ice. Th e
p e rm afros t d epth decreases to \\'a rd s the acc umula tio n
area , \\'hi c h basica ll,' co nsists 0 [' tem pe ra te ice . Th e
tempera ture profiles ( Fig. 6) sho\\' th a t at 15 18 m d epth
there is a tempera ture difference of approximately 0 .5' C
bet \\'een loca ti o ns A a nd B.
Bedroc k is exposed a t the \\'este rn pa rt of the cirqu e
thres hold , while the easte rn proglacial area is cm-e red by
ti ll. Glacial st riae ill (I'ont o f the g lac ier indica te former
ice-Oo\\' dircctions o r 340- 350 . Orga ni c matcrial \\'a s
found be nea th till 60 m in front of the g lac ie r terminus. I t
was presel'l'ed in a sma ll cavity in th e stoss sid e of a la rge
Oa t- topped bou lder (2- 3 m \ Th e bo uld e r lI'as immediate l\' o\'e rl a in by 0.1111 of st ra tifi ed sec\ il11cnts or silty-tosaneh- rrac ti ons, in turn cO\'C recl b,' 0.+ m 0[' ti ll . Th e
si ig h tl y in c li ned sed im e n t la reI's liTre cu t n IT a t the boulder
s LII' f~lce . A ra di oca rbo n el a tin g o r th e o rga ni c m a teri a l
yielded a n un ca li brated age o r 3260 ± 80yea rsB P (lab.
no. Ua -3795 ) . Three m e ltll'a ter strea ms drain Pa ssg l aci~i r en. On ly the lI'estCrJlmost strea m co nta in s sed iment , while the o th e r tll'O car ry clear lI·a ter.
DISCUSSION
Eros ion potential of PassglacHiren through tiD1.e
The gen era l tempera ture distributi o n \\'ith in the g lac ier is
uncomp lica ted. P assg lac ia ren is pa rt ly warm -ba sed and
partly cold-b ased , an d ca n thu s be class ifi eel as a subpolar
0
\
-10
I..c:
-20
A
Cl.
Q)
"0
Q)
.!2
-30
-40
-50
-3.0
Fig . .f. Ice dejJlh maj) /01' Passg/aciiirell wilh a cOlllollr
illterval q/ /0111. Tlte dashed area slw1.f.'s Ihe I'slil11aled
IIWlimll1l1 ellflll oI warlll-based ice. The a/TolL' ill l/ie
celllreil j){nl q/ Ih e glacier indicales lite gmeral ice-flolt·
direclioll.
] 04
-2.0
-1.0
0.0
Temperature (0C)
Fig. 6. r"erlical lemjJeralllre jJrqfiles , lIleasllred ill Augusl
/993. Seasonallel7ljJeralure varialiolls l('ere obsen'ed dowlI
10 15 m. The lellljJeralllre dijJerellce belwcfII .,1 and B al
equal deplhs illdialles a horizonlal lemjJeralure gradiflll
bell('em Ihl' a(ClllIllllalioll area (llId Ih e ablalioll area. This
gradielll is also rejlecled b), the Ihickll ess oI Ih e cold
jJl'rl7laj'rosl I(~)'er wllirh decreases lo wards Ihe aCCllllllllalioll
area.
R ic/wrrhol/ alld H o/m/III/d,' G /acial (irqllf jimllalioll
g la cier , Temperate ice is res tri c led to th e d eepes t p a rt 0 ['
th e g laci e r \\'ith a di splace ment toward s the accumu lation
area , Th e displacement is also re nec ted in a horizo ntal
lemperature g ra di en t betwee n the acc umul a tion a rea a nd
th e Cl bl a ti on area ( Fi g , 6 ) , Thi s lem pera ture d iSlri bu ti o n
is pro ba bl y due to the ro rm a ti o n of temperate ice in th e
acc umulati o n a rea, Durin g summ er, surface me lt\\'a ler
pe ne tra tes th e firn a nd subsequent ly rcfreezes, Thi s
releases late nt heat \I'hi ch ra ises th e lempera ture of lh e
firn a nd ice , I n th e a bl at io n zo ne, thi s process ca n occ ur
o nl y in early summ e r whell th e ice is cOI'e red b>' SIl O\l',
sin ce so li d ice is imp erm ea ble to mell\l'a ter. \\,i th a
tcmpe ra ture di stribution similar to th e present, P assg lac ia re n ha s little impact o n th e cirque m o rph o logy , Th c
eros io n m al' o nlv res ult in part ia l d ee pe nin g or th e cirqu c,
Assuming a co nsla nt precip itation, changes in areal
eX l e nt o r ice thi c kn ess or a g lacier h an' critical
co nseq uences for its erosil'e ca pac i l I', As mellli o ned , a
redu c li o n in surface a rea of a g lac ier res ults in sm a ll e r
m ass l u rn o \'e r and d ecrea sed e rosil'e ca paci l ~I , Fu rth e rmore , \I'hen a sub-pol a r g lac ier beco m es thinn er. lh e
a m ou nl o f' warm - based ice decreases and, el'entua ll y, th e
g lacier ma y becom e co mpl e tel\' fi'oze n lO th e bed , Th e
radi oca rbon d a te impli es th a l prior to 3260 ± 80 IIC
years BP. P assg l ac i ~i re n was sm a ll er lhan. or p oss ibl~ '
simi lar to. ils prese nt size, This age is in lin e \I,ith th e
le rminati o n or th e H olocene h yps ithcrm a l, a climat ic
optimum initi ated aro und 8500 BP ( ~ esj e a nd K l"<l mme .
199 1), Sed im ent studi es o r proglac ia l lakes sugges t lhat
one o f th e m ore pron o unc ed g lac ie r aeh 'a nces in
Sca ndin a \'ia occurred about 3200-2800 BP (Karl c n ,
1988 ), Acco rdin g to th e sam e stud y, so me Scandimtl'ian
g lac ie rs di sa ppeared durin g a warm peri od in th e ea rl y
H o loce ne a nd did no t re-fo rm until a bo ul 2000 BP, Th e
inre rr ed redu ced size or P assg lacia re n durin g lh e
H o loce ne clim a ti c o ptimum imp li es a d ec reased cros il'C
capac ity ol th e g lacier durin g so m e peri od prior to
3260 ± 80 yea rs BP,
R ec urrent co ld peri od s, I\'ith g lac iers substa nti a l'"
large r th a n a t prese nt , h al'C prel'a il ed for m ore lh a n 90%
o f th e tim e durin g th e pa sl 2 '\1 a (P oneI', 1989 ) ,
P assg l ac i ~i r e n ca nn o l g ro \\' mu c h larger than al presc nt
until it sta n s to infill th e d o wnstrea m I'a ll ey , Thus, it ma y
ha\'e bee n sm a ll eno ug h to be cl ass ifi ed as a cirqu e g la cier
lor less th a n 10% of th e past 2 !.I a, i, e, 200000 yea rs, I t is
poss ibl e th a l th e cirqu e g lac ie r co mpl e tel y di sa ppea red a t
so me stage o f th e H o loce ne , as \\'ell as during pa n s of'
prn'io us int erg la cia ls, Its pe riod o r existence m ay thus
ha l'e bee n sig nifi ca ntl y sho rter th a n 200000 wa rs,
Cryogenic d enudati o n processes, \\'hi ch m ay ac t o n
ice-free surlaces, pro ba b ly innu e nce th e rate ol head\\'a ll
re trea l. H o\\'el 'e r, these processes a re mos t proba bl y no t
capa bl e o f enhan cin g th e bedrock d ep ression durin g
peri ods wh en th e P assg lac ia ren c irquc is fr ee fi'o m ice,
PassglacHiren's contribution to the cirque
excavation
Cirqu e fo rmatio n tim e ca n be es timat ed based o n e rosio n
rate a nd a reco nslructed preglac ia l topography (L a rse n
a nd ~ I angerud, 198 1), F o r th e P assg lac ia ren c irqu e, t\\-O
reconstructions or preglac ia l to pog raph y ha w bee n
perfo rmed, Th e maximum reco nstructi o n ass um es a
sm oo th ed co ntinu a ti o n or adjacen l slopes , and lh e
mlt1lmum recons tru c ti o n assumes a minor co nc,1I' it\' a t
th e I'alle\, sid c, The present cirq ue lorm co rresponds to a
maximum bedroc k d eepe nin g 0 1' 2 10 m a nd a minilllum of'
120 Ill,
:\ eith er the prese nl no r past erosio n ra les lor Passg lacia ren a re kn O\Ul, Ncverthe less, lh c ran ge o r poss ible
\'a lu es ca n be nalTo\\'cd by loo kin g a l e ros ion ratcs of
o th er Scandinal'ia n g lac iers , Th e suspended sedim e nt
transport rrom Slorg l ae i ~i r e n , a I'a ll ey g lac ier near
Passg lac ia ren Fi g, I , co rrespo nds to an eros io n ra te or
0 ,6 111 m a I (Sc hn eider and Bro nge, in press ), I n a slu e'" ol
se\'era l prog lac ia l melt\\'aler strea ms in \! o r\\-ay, th e
sedim ent transport equ a lled eros io n ra les or O,2- 0,6 mm
a I (O strem and o th ers, 1970 ), Th e ('t'osion rale of' a
Yo un ge r On'as ci rq ue g lac ie r loca ted o n the :\o r\\'Cgia n
coas t has been calc ulal ed as 0 ,.5 0,6 mm a I ( La rsc n a nd
1\ Ia nge rud, 198 1), C si ng the tl\'O reco nstructed preglacial
su rfaces for the Passg laciaren cirqu e. and a cons ta nt
eros ion ra te o f' O,6 mm a I, a m a inl y \\'arm - based cirqu e
g lac ie r \\'ould require 200000- 350000 years lO d el'C lop
th e prese nt cirCJu c form , The typ ical eros io n ra te ol'e r
tim e has probab l!' bee n sig nifi ca nt'" lo\\'e r. Furthermore.
it is like '" th a l th ere hal'e been periods \I,hen lit e c irqu e
g lac ie r was co mpl e te ly lroze n to the bed , Us in g an crosion
ra le of' 0 ,3 mm a I, a time period as long as +00000700000 yea rs is required, Cons id e rin g the length of'time
durin g the Qu a terna r y tlt a l th e cirqu c g lac ier ma>' hal'e
ex isted , these ca lcul ated c irqu e formation limes slro ng'"
sugges t lh a t th e cirqu e g lac ier has n Ol bee n th e princ ipa l
age nl excG\,-at in g til e cirque,
Cirque forInation under alternative glacial
conditions
A glac ier \I'itlt th e ca pac it y to eXCL] I'<1 le a c irqu e form
mu st be \\'arm -based a ll ol'('r th e cirqu e, Situations \I,he n
thi s co nditi o n might be fulfill ed in th e stud y a rea a rc ( I )
when the climate is o f a pronounced maritime ll'pe
producin g a hi g h'" d>'n a mi c cirqu e g lac ier, o r (2 ) II'hen
lh e ice co, 'e rin g th e c irqu e co nstitutes part or th e
acc umul atio n a rea of' a la rge r g la cie r. Al present. th e
po lar fronl y ie ld s hi g h cycloni c ac ti v it y a t th e la titud es o l
no rth e rn Scandinal'ia, \I,ilh frequent 10 \1' press ures
co ming fi'om th e \\,est. D esp il e thi s, th e eastern pa n o r
lh e m o untain ra nge has a rcla li,-e ly dn' co ntin c nta l
cli mate, There arc no indi ca tion s 0 1' a pronounced
maritim e climal e in th e stud y area durin g th e H o lo('e ne, 1n th e case \\'here the ci rq ue co nsti tu les pan of the
acc umul a ti o n a rea of a la rge r g lac ie r, this situ at io n must
ha l'e occ urred se\'e ra l tim es , \" hen the d O\l'nstrea m I'a ll ey
Ku ope J'\'agge ( Fi gs I and 2 ) \\'as excal'ated, th e Passg lac ia re n c irqu e mu sl hm'e bee n part oi' th e acc umulati o n
a rea of' lh is la rge g lac ier , Th e mo rph o logica l simil a rili es
bet\\'een g lac ia l cirques a ncl ol'e rcl ee penings a re rem a rk a bl e , H oo ke ( 1991 ) d esc rib es a poss ible co mm o n
mec ha nism ror th e d el'C lop m e nt o r the l\l'O rcatures,
Acco rdin g lO Po rte r ( 1989 ), th e <ll'erage Fenn osca ndian Plei slocT ne g lac ier ex te nsio n cor respo nd s to a
mountain-centred ice shee l o r ice field . rather than a
la rge, mo re eas terl y-cc ntred ice shee t cOl'e rin g th e \\'ho le
o f Scandina\'ia, R ece nl num e ri ca l-m ode l experim ent s of
th e lasl g laciation sugges t th a l during th e first lhird or th e
105
Rlchardson and Holmlund: Glacial cirque jorma/ion
\Veich se li an the m o unta in areas were covered by a sm a ll
ice sh ee t (H o lmlund a nd Fas too k, 1995) . Th e ice-flow
pat te rn of suc h a n ice shee t o r ice fi eld is co n troll ed by bed
topography. The ice surface o f a mounta in -ce ntred ice
sh ee t is probably at a lowe r a ltitud e than that of an ice
shee t close to a maximum exte nt , g ivin g wa rm e r
co nditi o ns a t the ice surface . A lso, a smal l ice shee t wi ll
be m o re influ enced b y th e cyclo ni c acti\"ity of the
\\·es tcrli es, favouring surface m eltin g in th e summer a nd
th e form a ti o n of te mperate ice. Durin g th e initi a tion of
th e last a nd earli er g lacial s, small te mpera te, or basicall y
warm-based , erosi'·e ice shee ts probably cove red the
Scandinavian mountain range. Large ice sh eets reach
hi g h enou g h a ltitud es fo r po la r ice to form , which
e,·e ntu a ll y ma y d ec rease th e erosion ra tes a t th e bed. In
case of e rosion b y a large ice shee t, cirqu e form s wou ld
probab ly be d emoli shed because th e ice-flow pattern is
controlled by the geo m etry of the ice bod y ra th er than by
bed topogra ph y.
If wc accep t th e co nce pt o f a m ean Pl eistoce ne iceshee t ex tent simila r to th e one sugges ted by Porter ( 1989),
we m av a lso accep t lon g period s o f glacia l eros ion by
networks of large valley glaciers, ice fi elds or sm a ll
mountain-centred ice shee ts. It is th e refor e p lausib le th a t
these la rger glacier types ha ve bee n the major eroding
age nts for th e g lac ia l ci rq ue fo rm a ti o n on th e eastern sid e
of th e Scand in avia n mountain range .
CONCLUSIONS
Under clim a ti c co nditions such as th e present, glacie rs
like Passglac ia ren , which is la rge ly fro zen to the bed , have
a limited erosive ca pacity and a res tri c ted effec t on cirque
m o rph o logy . If th e cirque g lac iers a re warm-based to
so m e exte nt , th ey ma y ha,·e some imp act on the
develop m e nt of overdeepenings. Th e large-sca le form of
th e il1\"Cs ti gated cirqu e probab ly deve loped prima ril y
when th e cirqu e constituted p a rt of a n acc umul a tion a rea
of a la rge r g lacier. Durin g th e Quatern a ry there have
been periods more favo urabl e for glacial erosion in thi s
a rea than the prese nt. Such peri od s occ urred durin g
initi a tion s of glacia ls, at tim es whe n the stud y area was
cove red by netwo rks of la rge vall ey g lacie rs or b y a sm a ll
mountain-centred ice sheet. Du e to th e low a ltitud e· of
these g lac ier types, co mpared to a la rge ice sheet, th ese
glac iers a re influ enced by th e cyclo ni c act i,·ity of the
westerlies, fa,·our ing form a tion of temperate ice . Th e
,·alle y g la ciers a nd sma ll mountain -ce ntred ice shee ts may
thu s have bee n tem pera te or a t leas t basica ll y warmbased.
Concern ing glacia l cirqu es on th e eastern sid e of the
Scandina,·ian mounta in ra nge, we ag ree with the ,·iew
that cirque forms were m ainl y form ed in assoc iation with
g lacia ti o ns. However, \\·e be lie\"e th a t they were not
primar il y form ed by sm a ll cirqu e g lac iers, but ra th e r b y
106
large erosive glacier co mpl exes o r small ice sheets . Cirque
form s are thus not good proxy data for climatic reconstru c tions.
ACKNOWLEDGEMENTS
Thi s stud y was fund ed by Th e Swedish Natural Science
R esea rc h Council (NFR ), th e Swedi sh Society for
Anthropology a nd Geography (SSAG ) a nd the Universiti es of Luncl and Stockho lm. We are very grateful to
].-0 . Nas i und for frui tful cl isc ussion s a nd c ri ti ca l
co mm ents on the ma nu sc ript. Th a nks to P . J a nsson and
'vV. K a rl e n for id eas a nd va luabl e co mm e nts. \Ve wou ld
a lso lik e to ac kn ow led ge the staff a t th e Tarfa la R esearch
Station fo r support a nd assistance during th e field -wo rk .
J. Boygle kindl y improved the English and H. Drake
ass isted in drawing th e illustration s.
REFERENCES
AndIT I\"s, J. T. 1972. Glacier pOll'er, mass ba la nces, I·elocities and
erosion a l potential. .(. GeomorjJhol., Supp lemc ntband 13, I 17.
Enqui st, F. 1916. Der Einfluss des \\·indes a uf die Verteilung dcr
G lclsc hc r . .\fedd. CptJS(tia L·nil'. Gtol. /11.11., 14, I 108.
E\"an s,
T. S.
1977.
co ncen lrat ion
151 175.
\ \ "or lcl-\v icl c ,"arialions
or c irqu c
in
the
direction
and
a ll d glacier aspects. C{logr. Ann. , 59A (3--4 ).
Halllran, S. -E. and E. Aarholl. 1993. Glacier study using lI·a l·enumbcr
domain sy nth e ti c aperture radar. Radio Sci., 28 (4 ), 559- 570.
H a mran, S. -E. , D. T. Gjcss in g, J. Hj elmstad a nd E. Aarholl. 1995.
G ro und penetra ting sy nthetic pulse racial': d yna mi c range and modes
01" operation. J. . 1/)/;1. (;eo/}/I)'s., 33. 7- 14.
H olm lund. P. 1991. Ci rqu es at In\\" a ltitud es need not nccessa rily halT
bcen Cu t by small glacicrs. Geogr . .'11111 .. 73A ( I ), 9 16.
H o lmlund , P. and 'd. Eriksson. 1989. Th e cold surface la ye r o n
Sto rglaciiiren. Geogr . .'11111. , 71A (3- 4) , 241 244.
H olmlund , P . and J. Fastook. 1995. A time dependent model o f the
\\' eichscl ian ice shee t. QJlal. /111., 27 , 53- 58.
H ooke. R . Le B. 1991. Posili ,·e ICed bac ks assoc ia ted with eros io n or
p;lae ia l cirques and ol·c rdce pc nin gs. Geo!. Soc. ,·1111. Bull., 103 (8),1104
1108.
K arlcn. \\ .. 1988. Scandinal·ian glacia l and climatic flu c tuat ions during
th e H olocenc. Qjllll. Sci. Rev ., 7 (2), 199 209.
Kin g. L. 1984. Zonation and ecology of h igh mounta in permafrost in
Sca ndinavia. (;eogr . .·11111., 68A (3), 131 139.
Larscn. E. and J. i\ la ngerud. 1981. Erosion rate of a Youngc r Dryas
c irquC' g la c ier at K nlkcllcs. western Norway. AUI1. G/ociol., 21 153-158.
Ncsje, A. and :-'1. KI·a lllllle. 1991. H o loeenc g lacicr and climati c
"a ri a tions in western Norway: e\·iclc ncc for ea rl y H o loccnc g lacier
demise a nd multiple \feoglacial (""CIllS. Ge%g)'. 19.610- 612.
O stre m. G ., T . Ziegier and S. R . Ekm an. 1970. Slamtransportundersbkelser i norske bre-eil"e r 1969 . . \"orges r·assdragj·- og Eleklrisilelsl'esell.
r·lI.isdrag,rlireklorlllel. Hydr%,gi;k . 11,de/illg. RajJjJorl 6/70.
Po rt cr, S. C. 1989. Some geo logical impli catio ns 0 1" al"CTagc Quaternary
glacial conditions. Qj/al. ReJ .. 32 (3) . 245 261.
Rudbcrg, S. 1992. i\ l ulti plc g laciat io n in Sca ndin al· ia
see n in p; ross
Illorpho logl" or not? Gcogr . .·11111 .. 74A (2 3),23 1 243.
Rudbcrg, S. 1994. Glacial cirques in Sca ndinal·ia . .:\ or. Geogr. Tidsskr.,
48 (4), 179- 197.
Schneider. T. and C. Bronge. In press . Suspended sediment transpo rt in
the Storglaciiircn d,."ina ge basin. Geogr. AI/II.
\ ·ilbo rp;, L. 1977. Th e ci rquc fo rlll s 01" SII·cd ish Lap land. Geagr .. 11111 ..
59A (3- 4 ).89- 150.