J Ol/ mal
of Claciolog)"
VoL. 42, . \ 0. NJ , 1996
Visible and near-infrared reflectivity during the ablation
period on Peyto Glacier, Alberta, Canada
PAU L M . C UTL ER- A."\TD
D. ScorI'
M UNRO
D e/Jar!men! oJ Ceogmj)/l)" Cllil1ersilJ! oJ T orolllo, Eril/{Iale College, l\/ississouga, Oll!ario L5L JC6. Canada
~I od els fo r calc ul a ting g lac ier mass ba la n ce a r e se nsiti\'e to surface
refl ectivit y va ri a tion . Fi eldwo rk carri ed o ut on Pey to Gl ac ie r, Alberta, C a nad a,
contributes to the d a ta se t a \'a il a bl e fo r ice-refl ec ti\'ity p a ra m eterizatio n in such
mod els. H e mi sph eri ca l re fl ec ti\'it y in th e \'isibl e a nd nea r-infra red pa rts o f th e so la r
spec trum was obtain ed fo r roc k, snow a nd three co ntras tin g glacier surfaces to
exa mine tempo ral a nd sp a ti a l va ri a ti ons. Glac ier-i ce nea r-in fr a red refl ec ti\'ity di spl ays
onl y min o r spa ti a l \'a ri a ti o n (0.12- 0. 17) in co mp a riso n with th e visibl c ra nge (0.230.40 ) , th e la tter bein g influ e nced prim a ril y by surface im purit y conte nt. S urface
rou g hn ess is of min o r impo rta nce co mpa r ed with impuriti es . T emporal va ria ti o n of
refl ectivi ty \,'as wea k a t a ll g lacier-i ce a nd roc k loca ti o ns; si ig h t \'a ri a ti o ns o bse n 'ed
II"cre du e to cha nges in eith e r so la r ze nith a ngle or cl o ud a m o unt. Snow refl ec ti vit y
di spl ayed pronounced diurn a l asy mm e tr y an d a large r res p o nse to cl oud co\·e r. Th e
minim al te mpora l \'a ri a ti o n in glacie r-ice re fl ecti\'ity sim plifi es its pa ra m e te ri zati on.
Thi s b e ll£1\"io ur is a dditi o n a ll y useful fo r sa tellite-b ased m eas urem e nts o f th e
refl ec ti\'it y fi el d on la rger g laciers, as im ages o bta in ed \\' ithin a 6 h wind o \\' ce n tred
on sola r n oo n a re likely to yield \'a lu es whi c h a re within 2 3% of da il y m ea n ya lu es.
ABSTRACT.
INTRODUCTION
Th e a blat io n a reas of mid-l a ti tud e glaciers co ntribute a
di spro po r tio n a tely la rge vo l um e of m e ltwa te r to to ta l
summ e r disc h a rge in proglac ial stream s. Differe n tia l
refl ec ti o n o f in cid ent sola r radi a ti on by th e va ri ety of
sur faces enco untered o n \'a ll ey g lacie r s exerts a n
importa n t te m po ral a nd sp a ti a l con tro l o n me! twa ter
genera ti o n . \\' hilst snow re fl ec ti \'ity has recei\'ed ex tensi\'e
exa min a ti o n , in terms of bo th measurem e nt a nd modellin g (see r e\'ie\l's by M e li o r, 1977; W a rre n , 1982; Dozier,
1987 ) , ex posed glac ier ice rem ain s und e rex pl o red in thi s
rega rd. l\Ie teorological im'es tiga tions o n g laciers have
in clud ed th e m eas ure m e nt of ice re fl ec ti\'it y, with
consid e ra bl e \'a ri a ti on in th e res ults o bta in ed (Andrews,
1964; H o lm g ren , 197 1; W e nd lc r a nd W ell e r , 1974) . In
lig ht of th e se nsiti\'ity of c urrent glac ier m ass-ba la nce
simul a ti o ns to re fl ecti vity (O erl ema ns a nd H ooge nd oo rn ,
1989; Munro, 199 1), es tim a ting a single \'a lu e from such
studi es to re present th e e n tire a bl a ti on a rea o f a new a rea
\I'ill p rodu ce e rroneo us predi c ti ons of'd a ily m ass loss.
Fi eld wo rk on Peyto G lacier, Alberta, Ca n ad a, was
desig ned to qu a nti fy the spa ti al a nd tempo ra l ra nge of
refl ec ti\'it y in a tem pe rate g lacier se tting . Sn o w refl ec ti \'ity
para m e te ri za ti ons in so m e c urrent g lo b a l circ ul a ti o n
• .\101'" a t th e D epa rtme nt o f Geology a nd Geo ph ysics,
U ni ve rsit y of Minn eso ta, 3 10 Pill sbury Dri \'e SE,
~ finn eapo li s, T\fN 55455 -02 19, U .s.A.
m odels sepa ra te th e \'isible and nea r-in frared wa ve ba nds
( ~I a rs h a ll a nd W a rren, 1987 ) because of their different
responses to surface ph ysical pro p e rti es . Values fr om Pe)'to
Glacier we re o bta in ed for th e sp ec tra l \I'a \'eba nd s of 0.350.695 {lm (\'isibl e ) a nd 0.695- 2.8 {lm (nea r-infi'a red ) .
BACKGROUND
~1 eas ur e m e nt s o f' broa d - ba nd re nec ti vit y a re o ft e n
condu cted o\,e r th e short- wa \'e spec trum fro m 0 .35 to
2.8 Jlm , a nd a rc commonl y re f'e rred to as a lb ed o (p).
H emisph eri cal re nec ti \'i t)' o \'C r th e broad \' isi bl e a nd
nea r- infra red spec tral ba nds (PVl S a nd PNIH ) is rep o rted in
this pa per. Th e te rminology ado pted here origin a tes fr om
th e more precise d efiniti ons of Siegel a nd H owell ( 1972 ) .
Ea rl y m eas ure m en ts of P o n mid-l atitud e glaciers in
th e Austri a n Alps we re ca rried o ut by Sa ub er e r a nd
Dirmhirn ( 1952 ) , Sc heibbn er a nd ~J a hrin ge r ( 1968 ) a nd
W agner ( 1979 ) . T a b le I includ es res ults of th ese studi es in
a compa riso n o f th e quoted ra nges o f P for diffe re n t sno\\'
a nd ice surfaces. Grenfell a nd P e rovich ( 1984) provid ed
\'a lu es fo r spec tra l refl ec ti\'ity o f g lacier ice be twee n 0.38
a nd 2.5 Jlm fo r Ath a basca Gl acie r, Ca nad a . !-.1unro a nd
Y o ung ( 1982 ) m a d e some preli m in a ry in ves tiga tio n s of P
o n PeyLO Gl ac ie r.
Th e fac to rs influ e ncing refl ec ti vity a re those rel a ted to
th e surface itse lf a nd th ose co nce rning th e in c id e nt so lar
radi a ti on. An in c rease in g ra in-size res ults in red uced
refl ec ti\'ity, most e ffect ively in th e nea r-infra red sp ec tral
regio n (W a rre n , 1982 ) . Bo hre n a nd Besc hta ( 1979 )
333
J Ollmal OJ Claciology
T able 1. Examples of rejJorted ranges of lotal hemisjJherical reflectivity over glacier sU1Jaces. SllIface descrijJtions are
ada/Jted ]ram those lIsed by the autli ors cited in the lable
Wagller (1979 )
Sou rce
Scheib b71e1" and _Halzri71ger
,\larkin ( 1961)
Sauberel" and Dirmlzim
( 1952)
Frall;:-Josef Land
SOllnblickglet scller
( /968)
Hinlereistemer
Location
N ew snow (> 0.0 5 m thick )
N ew snow « 0.05 m thick)
Clean old snow
Impure old sno""
Clea n firn
Im pu re firn
Clean ice
l\1edium impure ice
Impure ice
Sonnblickglelscher
0.90- 0.95
0.60- 0.90
0.62 0.68
0.46- 0.59
0.25- 0.30
0.15- 0.20
d em onstrated tha t for a give n g ra in-size p is, a t m os t,
weakl y d epend ent on density. \Va ter alone does no t
increase radi a tio n a bsorpti on, beca use its refrac ti ve ind ex
is simil a r to th a t of ice (Berge n and o th ers, 1983) , but th e
efTec tive gra in-size is increased b y wa ter in inter stices
be tween grain s (C o lbeck , 19 79 ), a nd this redu ces p.
Impurity co ntent ac ts mainl y on visibl e refl ecti vity; la rge r
con ce ntra ti o ns I-edu ce \'a lu es o f p (Wa rren , 1982 ) .
In creasing surface ro ughness, thro ug h th e d evelo pm em
of mi crotopog ra phi cal fea tures such as sas trugi , has th e
sam e effec t. Slop e a ngle and as pec t interac t \\-ith in cident
sola r-radiati on geo metry to al ter its effec tive a ng le of
a pproach to th e surface, thus a ltering p (M a nn stein ,
1985) . A depend ence of refl ec ti\'i t y on sola r zeni th a ng le,
has bee n d emon stra ted in num e ro us studi es ove r snow
a nd ice (Sa uber er a nd Dirmhirn , 1952; Hubl ey, 1955;
Sch eibbn er a nd M a hrin ger, 1968; Ya ma nouchi , 1983;
M cGuflie a nd H end erson-Sell ers, 1985; Wendl er a nd
K ell ey, 1988 ), wh e re refl ecti\'ity in creases as in cr eases,
beca use of enh a nced spec ul a I' refl ec tion of th e direc tbea m component o f radi a ti on.
Th e spec tra l compositi on of in cid ent sola r ra di a tion is
a l tCl-ed by th e presence of clo ud s as water a nd ice
preferenti ally a bsorb nea r-infra red radia tion, thus increas ing th e proportion of visible li ght. Thi s enh a n ces p
beca use snow a nd ice refl ec t m o re visibl e th a n nea rinfrared radi a tion (W a rren, 1982 ) . Additi ona ll y, th e
difTuse frac tio n of th e incid ent flu x increases with cl o ud
cover. The diurna l varia ti on of p is smoo thed as th e
diffuse fi'ac ti on in creases, with va lu es for > 50° redu ced
but th ose for < 50° in creased , a r es ponse which h as bee n
well d oc um ented for wa ter (1'\ unez a nd oth ers, 19 72 ) .
e,
e
e
0.88- 0. 98
0.85-0 .95
0.65- 0.70
0. 70- 0. 88
0.50- 0.70
0. 35- 0. 50
0. 50- 0.63
0.1 5- 0. 30
0. 30- 0.45
0.20- 0. 30
0.06- 0. 20
e
0.60- 0. 65
0.70- 0.90
0.46- 0.70
0.30- 0.60
0.50- 0. 65
0.18- 0.30
0. 30- 0. 46
0.20- 0. 30
0.09- 0. 20
Impurity com ent was determin ed by filtering a melted
sa mpl e o f known volum e through a pre-weighed qu a rtz
membra ne filter ( 1.5 JiI1l pore spacing) . Th e sampl e was
th en dri ed for 24 h a t 110°C and reweighed to yield
impurity content. Sn o w d ensity was meas ured using
sta ndard tec hniqu es (0s trem and Bru gm a n, 1991 ) .
V a lu es fo r ice we re obta ined from a SIPRE-type co rin g
a uger (83 mm di a mete r ) penetra ting th e topmos t 0.250.30 m o f th e surface. Ea ch sa mple was split a t th e
disco ntinuity between th e wea th ering cru st (MLiller a nd
K ee ler , 1969) and th e m ore dense ice below, permittin g
sepa ra te d ensiti es to b e ealcul a ted. Surface roughn ess at
eac h site was es tim a ted b y laying out 5 m long poles o n
th e ice a nd meas urin g th e distance from pole to surface a t
0.1 m intervals. Fi ve hundred valu es we re coll ec ted a long
eac h o f two tra nsec ts a ligned a t rig ht-a ngles to on e
a no the r, a nd sta nd a rd d evia tion a bout the mea n dista nce
was used as an ind ex o f ro ughn ess .
Four pyranometers (Kipp and Zonen , ~I o d e l CM 11 ,
H oll a nd ) we re used to o btain hemisph erical reilec ti vi ty.
Th ese were mounted in two bac k-to-bac k pairs on adj ace nt
horizo nta l supporting a rms 1.6 m above the surface . On e
pair reco rd ed in co min g a nd refl ec ted so la r radi a tion over
th at pa rt of the spectrum whi ch passes through clear g lass
(0.35- 2. 8 ~lm ) , thus capturin g all but neglig ibl e amounts of
solar radi a tion (K ) fro m th e full sho rt-wave spectrum.
Hemisph erica l R G695 filters on th e second pair permitted
onl y wa velength s grea ter th an 0.695 ~lm to pass through ,
thus capturing near-infra red solar radi a ti on (K Nm). Th en
PV lS a nd PNIR a re es ri m a ted by:
0.095
PVlS
1
=
p)J (\ c1A
0.35
1 0.095
K., c1A
PROCEDURES
K T- K h m
K 1 - K lNfR
(l a )
0.3~
.) 8
Fi ve sites in th e a bl a tion zone o f Peyto Glacie r we re
ex amin ed , eac h fo r a pprox im a tely I week of th e 1990
summ er. Three loca tions wer e c hose n to r e prese nt
extremes of cl ea n, dirt y a nd ro ug h ice (excludin g
cr evassed a reas ) . Th e rem a inin g two sites wer e on
m elting snow a nd roc k. A d eta il ed assessment o f surface
properti es was ca rri ed ou t fo r eac h si te (T a bl e 2) .
334
r-
p>, K ).. dA
J 0695
PNm
= /2.8
K >, dA
(lb)
J 0695
wh ere A refers to wa vcl eng th and a rrows d enote directions of flu x.
Culler and ,Ill/lira: R eflecliviO'
011
Pl')ilo Clacier
Table 2. Properties oJ sll/dJ'-sile Jlllfaces
Sile
i\Ieas ure ment p eriod
EIe\'a tion
Aspect
Inclin ation
Surface impurity co ntent l
:"l aximum 2
IVI i n i m um :;
Densit y
-t
Surface
Substrate'')
R oug hn ess 6
Ea st- west
North- so u th
G ra in-size
Unils
SIIOU'
D irly ice
Cleall ice
J-/u11Zl7wck} ice
IV!oraine
Juli a n days
m
203- 210
2500
032
09
2 10- 217
2267
350
04
2 17-223
2265
350
02
223- 232
2244
i
107 ,287
30- 48
198- 203
2240
3 15
07
gl
gl
0. 199 ± 0.040
kg m
kg m
3
3
mm
mm
mm
I From th e top 10 mm.
:"1 ean of three hig hes t \·a lu es .
3 ~I ean of th ree lowes t \'a l u es .
I To a d e pth of 609mm .
5 L aye r iml1l ed i a t e l ~' be low s urface samp le
of 0.3 m.
2.9 11 ± 1. 354 1.390 ± 0.626 2.47 5 ± 1. 976
1.598 ± 0.253 0 .1 5 1 ± 0.088 0. 197 ± 0.079
515 ± 13
569 ± 12
671 ± 2 1
822 ± 28
63 0 ± 3 1
834 ± 23
613±41
834 ± 27
30
27
0.75- 1.00
89
44
5- 30
11 3
72
5-30
163
30
5- 20
6
2
to
50
44
5-400
Standard d el'ia ti o n of 500 readin gs in east west a nd
no rth- so uth tra nsec ts.
Aspects represe nt two do min a nt slopes o f th e hu m m oc ked surfa ce, 180 0 a pan.
a depth
It lI'as impo rt ant to ma int a in consta nt se nso r he ig ht in
o rder to eliminate changes in th e fi eld o f' v iew of the
il1\'C rted ins trum ents. A " fl oa tin g" supp o rt sys tem \\'as
thus ad o pted in lI'hi ch a vertica l mas t rested on a sma ll
fl a t rock (0.0 5 m 2 in a rea ) o n th e glac ier surface . A bolt
embedded in th e roc k held th e pole ce ntred, whil e three
g uy wires pro \'id ed Ih e necessa ry stabilit y. Th e sys tem
performecl II'C I\ thro ug ho ut th e season , ",i th minima l
adjustm e nt. Th e pyranometers lI'Cre a lig ned suc h that
midd ay shad ows were cas t north of' th e ce ntre o f th e
il1\'e rt ed -p y ra no me ter I·iell·i ng c i rcI e.
Th e pe rce nta ge of' total re fl ec ted so la r ra di a ti o n ( ~, )
ori g in a tin g fr o lll a circle o f' ra diu s a ca n be calc ul a ted
fr om in strum e nt height Zo (Schll-crdtfeger, 1976 ) :
(2)
Senso rs Illoun ted at 1.6 m lI'o uld therefore rcce ilT 9 1% of'
th eir sig na l fr o m within a c irc le of radiu s 5 m. Aeco rcling h', o bse n 'a ti o ns o f surface physical pro p e rti es lI'ere
ma d e within thi s bo undary whenel'Cr possibl e.
Th e p y ra nolllet er sig na ls were sca nn ed at 5 s intel'\',ti s
by a cl ata logger (Campbcl l Scient ifi c. l\fodcl 2 1:\ ,
tJ.S.A. ) II' hi c h reco rd ed 15 min mea n I·a lu es . A number
o f' adjustm e nts lI'ere ma d e to th e data :
(i) A filt er fa c tor of 1.049 was a ppli ed to th e sen sors fitt ed
lIith R G695 filt e rs to correc t fo r sola r rad ia ti o n a bso rb ed
b y th e g lass.
(ii ) A co rrec tion facto r of 1.0 I , b ased on measu re m e nts of
sha d ed a r ea o l 'e r a ran ge o f sola r ze nith ang les, \\'as
app li ed to da ta fr om the i l1\'Crted se nsors.
(iii ) D a ta o btain ed durin g prec ipita tio n pe rio d s were
di sca rd ed, sin ce th e p resence o f raindrops on th e ill\'e rted
g lass hemispheres ma v ha l'e produ ced erroneo us readings.
Considera ti o n o f potentia l e rrors in pyra nometri e
m eas urement s. trea tin g th em as ind epend ent a nd add itil'C , res ults in a worst- case e rro r of ±5.+% for th e
re!l eC[il'itl' meas ure m e nts. ~I cG urfi e a nd H end c rson -Se llC l'S ( 1985 ) po i nt o u t th a t this ra n g e is broad e noug h to
mask man y or th e s ubtle I'a ri atio ns in rell ee ril'it), th a t a rc
so ug ht. Thi s is, howeve r, not Lhou g ht to be a pro bl e m sin ce
th e errors represe nt rh e mea n de\'iat ion from the co rrec t
rea din g, th e d irec ti o n o r whi ch is unlikely to cha nge ol'er
th e da y, or fi 'om o nc dal' to th e next. Expe rim ents
perform ed on th e illI'erted i nstru m ent s, to iI1\'CS ti ga te
poss ibl e a lterati o n in ca libra ti o n co nsta nt du e to ove rturning, showed no sig nifica nt erTeer upon th e ca librat io n.
D ata from th e up\l'a rd-facin g unlilt ered pyranomete r
lI'(' re used, in co njun c tion wit h p y rh eli ometri c m eas urem e nts or direc t rad ia ti on (Eppl y, l\I ode l E6 , U.S.A. ) to
find the difru se radiati on, D, b y residu a l. Thu s, a
s tatisti ca l rela ti o nship \\'as fo und b e tll'een g loba l r ad ia ti o n , a nd th e diffu se fr ac ti on:
~ =(\+ (3 ](1
]( 1
10 co s e
(3)
whe re 1o, th e so la l' co nstant, is take n to be 1353 \ Vm 2
(M on teith , 1973 ) a nd is th e so lar zen ith a ng le. Our d a ta
y ie ld ed a = 1.1 7 1 a nd (3 = - 1. 225, lI'ith a coeffi c ient o f
d e te rmination equal to 0.82 . It sh o uld be noted that th e
dirT'use fr ac tion is assum ed to be equ a l in the I'isibl e and
n ea r-infrared s pec tral region s in la ter data ana lys is.
Commo nl y th e d i fru se-rad ia ti on p ea k drifts fu rther in to
th e nea r-infrared as cl o ud increases (\\'e iss a nd Norman.
1985 ).
e
335
J ournal oJ Claciology
RESUL TS AND DISCUSSION
A strong distinction between P VIS a nd P NIR was o bse n 'ed
und er clea r ski es a t a ll three g lac ier-i ce locati o n (Fi g.
l a ) . The la rger va lu es for PVIS ag ree with pre\ 'iously
observed contras ts be twee n th e two waveba nd s ove r snow
surfaces . Th e la rge surface impurity con te nt a t the dirtyice site is res ponsibl e for the relatively greate r difTe rence
be tween PVIS a nd PNlR a t this site. R efl ec ti\'ity di spl ayed
limit ed te mpor a l va ri a ti on, w hil st spa ti a l va ri a ti on,
indi cated by co mpa rin g th e clea n-ice a nd dirt y-ice
r es ul ts, was mar ked ly la rger in th e \'isibl e s pec tral
ra nge . Th e prese nce of surface roughn ess elem ents a t
th e humm oc ky-ice site, crea ting mi croto pographi c shading, is app a rently less influenti a l th a n impurity co ntent.
E vid ence for asy mm etr y of P VIS a nd PKIR a ro und solar
noo n, ca used by ph ysical cha nges of th e surface ove r the
passage of a d ay (mel ting), was onl y sli ght, a lth ough
afte rnoo n \'a lues tend to be lovver th an mid-m o rning
va lues a t th e hummoc ky- and dirty-ice sites ,
Snow co\'e r a nd rock are a t th e ex tremes of th e ra nge
of o bserved reOec ti\'ity behavio ur (Fig, I b), Th e g rea ter
diurnal va ria tio n o f re fl ecti vi ti es occurred a t th e sn o w site,
As no completely clear d ays occ urred during th e sn ow
meas urements per iod , a composite trace \\'as plolled b y
co mbin ing record s fro m two days with ove rl apping clea r
pe ri o d s. Here asy mm e try around so lar noo n is m o re
pro n o un ced , with redu ced a ftern oo n valu es att ributa ble
to surface metamo r p hos is a nd slope geo metry; the slop e
dips a t 9° wes t (280 °) , whi ch fa\'o urs g rea ter refl ec tio n in
th e m orning. I n co ntrast to the glacier surfaces. reOecti o n
of so la r rad ia ti on fro m roc k is grea ter in th e near-infra r ed
ra nge. N on-lin ea riti es in the traces a re probab ly ca used
by sh a ding effec ts from indi\'idu a l rock elements, As
exp ec ted fro m a n op a qu e medium , d iurn a l \'aria tio n is
negligible.
I n o rd er to stud y th e infl uence of cl o ud cover on P VIS
a nd PNIR, refl ec ti viti es reco rd ed in th e zenith -angle ra nges
3So< < 4S o, and SOO< < 60° we re plo tted against th e
diffuse fraction o f in cid ent so lar rad ia ti on, D/ ]{ 1, a n
approac h used by N un ez and oth ers ( 1972 ) o\'er wa ter.
As n o ted ea rlier, th e p rese nce of clo ud should subdu e
reOec ti vity for > SOO, whi lst th e co n verse is ex pec ted fo r
< SOO.
Th e res ul ts of thi s a n a lysis (Fig. 2 ) for dirty ice, in th e
ze nith- a ngle ra nge 3S o< < 4S o, sh ow th a t neith e r
wave b a nd is pa rti cu la r ly se nsitive to in creasing d iffi,lse
frac t io n, alth oug h th ere is a risin g trend whi ch is
qu a ntifi ed by r eg r essin g refl ec tivit y aga in st d iffu se
frac ti o n of in cid ent g lo ba l radi a tion (T a ble 3). Fo r dirt y
ice in th e zenith -a ng le r a nge 3So< < 4S o, PV1S ri ses by
0, 033 a diffuse frac tion goes fro m 0 to I , wh ereas P V1S
incr eases by onl y 0.011 . R eOecri\'ity fo r a ll glacier-i ce sites
sh ows a similarl y w eak res ponse to in creased diffu se
fr ac ti o n, alth ough of th e two wave ba nd s th e near-in fra red
shows the greater se n siti \'ity, rt sho uld be no ted th a t th e
sta nd a rd error di sp layed in T able 3 is o ften >50% of the
cha nge in refl ectivity from purely direct to purel y diffuse
co ndi tions,
e
e
e
e
e
e
1°·5
LOCAL APPARENT TIME (h)
a
rO.4
~--------~~~--~--~~----~ 1
0 .8
a.
>
SNOW,
~::
P y I5
oJ
I-
,.; ~
>
i=
(.)
W
...J
uW
a:
' :_"_R:__:_K_,_P_N
_I_R___
0 .2
ROCK , PylS
-.........
I
o+---r----c----,----r---,------,----r-,-------,--.....---.-----+r
1
°
0600
0900
1200
1500
1800
LOCAL APPARENT TIME (h)
>
~
I
~ oJ
t; i
crQ E D
L£ ~ §JrJ
o
O r
0.2
} 0.2
W
· 0 : -fJ_
_ t-_o
I_-o" ~-_O_'_O_:-o-.~-:-·--·-:,:_1_0
"-oT
;__
I
__O
+l: '
D/K +
Fig, 2. The responses of P VIS (open squares) and PNIR
( closed circles) aL the dirLy-ice site Lo increasing diffuse
}i'acLion oJ incident solar radiation Jor 3S' < < 4S ,
e
b
Fig , I. (a) D iurnal variation oJ PVlS and P NlR on clear
days aL the dirly-ice ( D.ICE) , clean-ice (G. ICE ) and
11lI1111110cked-ice ( H ICE) sites , (b) Diumai varia tion oJ
P,'IS a7ld P~ IR 011 clear days at t/ie snow and rock sites,
336
A s th eory predi cts (C houdhu ry a nd C hang, 198 1) ,
snow refl ec ti vity und er cl ear ski es (D / J{ 1-> 0) is gr ea ter
for th e la rge r of th e two e ra nges, so th ere is a m o re
dra m a ti c response o f snow refl ec ti vity to cloud (Fig . 3 ) ,
Cutler and JlIlIllro: R eJ7ectivi{l' on P,,-vto Glacier
TabLe 3. Results oj Linear regression al1a[)I.'iis oj rej7ectil'il)' vs diffuse fraction oj incidellt solar radialion
Site
Rej7ectivity
I.y/)e
Range oj ()
(0)
Sam/Jle size
PVlS
35- 45
50- 60
P:-IlR
P\' lS
Dirty ice
Clean ice
Slandard
error
ReJ7ectivi{JI at
71
53
0.0 11
- 0.002
0.008
0 .01 3
0.2 34
0. 253
0.245
0.250
35- 45
50- 60
71
53
0.033
0.03 3
0.008
0.0 12
0. 11 5
0.125
0. 147
0.158
35- 45
50- 60
11 5
54
0.000
- 0.00 3
0.009
0.0 15
0.406
0.417
0.406
0.4 13
35- 45
50 60
11 5
54
0.023
0.057
0.0 14
0.0 18
0.164
0.160
0.1 87
0.2 16
35- 45
50- 60
144
74
0.021
- 0.0 12
0.0 11
0.0 19
0.3 78
0.4 12
0.398
0.40 1
35- 45
50- 60
144
74
0.033
0.027
0.0 16
0.021
0.1 54
0.1 6 1
0.186
0. 188
35- 45
50- 60
61
45
0.049
- 0.01 4
0.0 14
0.022
0.728
0.779
0.777
0.766
PmR
35- 45
50- 60
61
45
0.184
0.166
0.0 16
0.0 17
0. 396
0.399
0.580
0.565
PV IS
35-45
50 60
38
31
- 0.006
- 0.0 19
0.0 11
0.0 18
0.171
0. 182
0. 165
0. 164
P;"\T lR
35- 45
50- 60
38
31
0.008
- 0.004
0.0 10
0.013
0.240
0.247
0.248
0.243
PN IR
Humm ocked Ice
P VIS
P:-IlR
Snow
PV lS
R oc k
0 .9 · - -
~
>
f:
(,)
L1J
...J
LL
L1J
er:
- - - - - ' - - - --
o,~
0 .8 l
Cl.
~
.~-.
~.
q.
D .!
- - - - - - - 0 .9
=- -
- .!:
.. .
?
0
...: ~c: .
......
0 .7
~
0
9 1'"
•
• U
T'
i .
~
Lo.8
7i
,
rf-
1
0 7
"
.
o.J
o.J
0.4
·.}- O.6
Dj K l= O
at
Dj K l= 1
On e factor clri\'ing cha nges in reOect i\'ity uncler
in c reasing cloucl is diminished a lte ra ti on of th e a ngle of
app roach o f in c icl e n[ so la r ra di a tion. It is therefore
interesting [ 0 exa min e th e rela tion ship betwee n refl ec ti\'it )' a nd cos () und e r sunn y ski es (Fi g . 4 ), co ntras tin g th e
b e h a\'iour of' P VlS a nd P KIR a t th e cl ea n-i ce and th e snow
sites (clean ice w as chose n bcca use surface impurity
co ntcnt \\'as th c closest ma tc h w ith th e snow site ) .
Predomin a ntl y s unn y sky pre\'a ils in th c se lec ted ra n ge
- 0 .5
1 t~
-+ ~'--____,--____,---------+I 0 .4
o
Refleclivi~v
Regression
slojJe
0 .2
0 .4
0 .6
0.8
. .., .
....
-----~
1
D/K +
Cl.
----'-
. ......
, o.s
~
~ 0.6
0 . 61
(,)
~O . 4
L1J
....I
LL
0 . 4-
L1J
a:
j
CLEAN ICE
C.
0 . 2-
Again PNlR is th e \'a ri ab le more se nsitive to in creasll1g
diffuse fraction, a nd P VlS provid es a good exa mpl e o f'h ow
th e [\\' 0 ze nith-angle ran ges ma y converge at D j J( 1 = 1
(as sugges ted by Nunez a nd o th ers, 1972 ) . These res ults
indi ca te that m ost of th e ri se in p und er in creas ing cloucl is
clu e to th e r ise o f re Oec ti v ity in th e nea r-infrarecl pa rt of'
th e spec trum .
1
~
SNOW
E
---r
l'
., •• ::::.,
0 . 8l
>-
Fig. 3. The resjJollse oj P Vl S (cloSfd squares ) and PNIR
(close d circles) at the metal1lorjJhosed snoU' site to
increasing di/JlIse fractio ll oj illcidfllt solar mdialion j01'
35"< () < 4:;-0, and eqlliNdentres/Jonmjor 50 < () < 60°
(a/Jell 5,.vmbols) .
--
A
A
A
A
#;,~t;.~b;,~~~!~
0 ~1~~~~~~--___
o
0 .2
0.4
0 .6
0 .2
- 0
0. 8
cos e
Fig. 4. The r eS/J 01ISe of PVlS (s quares) al/d P NJR
( triangles ) 10 ill creasing solar zenith angle at the sites jor
cleal/ ice (O/Jell 5,.J1mbols) al/d SIIOW (closed ~ ) ' II/bols ) .
337
J ournal of Glaciology
of cl o ud co nd i tion s (0- 0. 2 cl o ud ), thus mIlllmlZll1g th e
inf1u ence of cl o ud on
As ex p ec ted from th e r ela tivel y
g rea ter respo nse of th e snow surface to cl o ud , th e snow
ref1 ec ti vity displ ays th e grea ter se nsiti\'ity to cos e.
Direc t compa riso n o[ our res ults with da ta se ts from
oth er glaciers is hind ered by in a d equa te desc ripti o ns of
surface prope rti es in pas t in ves ti ga ti ons ofref1 ec tivit y. Th e
broad-ba nd n a ture o[ our d a ta a lso ma kes compa risons
difficult beca use th e majorit y of m eas urements availa bl e
are spec tral ref1 ec tiviti es [or na rrow \\·a\'e ba nd s. N everth eless , comp a ri so n o[ o ur d a ta with so m e sp ec tra l
refl ec ti viti es pu blished for glacier ice (fi g. 5) sho ws th a t
P VlS [or dirty ice (sola r noon \'a lue ) is slightl y greate r th a n
valu es in th e visibl e range reported by Zeng and oth ers
( 1984) for wh a t th ey term " ho neycomb glacier ice" . Th eir
surface is inte rpre ted as ice with multipl e cryoconite hol es
conta ining m o raini c debris.
e.
0 +-- - - - - - - - - - - - - -- -
o
0 .5
C
0
CONCLUDING REMARKS
Q C
1 .5
2.5
WAVELENGTH ( fl m)
Fig. 5. Comparison oj broad-band visible and nearinfrared rejlectiviliesfrom clean and dirl} glacier ice ill this
slud} wil/z sj)eclral rejleclivil)l data jrom olher glacier-ice
slllJ aces . The division between lite visible and nearinfra red bands is indicaled al lite tOll oj lite fig ure.
O\'era ll , it seem s th a t th e clea nes t ice fo und o n Peyto
Glacier is surpassed on som e gla ci ers. Compa ri son o f P VIS
fro m th e c1 ea nest pa rt of Pey to Glacier wi th sp ectral
refl ec ti\'iti es o f g lacier ice (Fi g . 5 ) sugges ts th a t surfaces
exa mined b y Grenfell and Perovich (1984 ), Z e n g a nd
o th ers ( 1984) a nd H a ll a nd o th ers ( 1990) co nta ined fewer
impurities . Th e va lu es of Z eng a nd oth ers a re [o r a blu eice ba nd , whilst H all a nd o thers o nl y desc ribe th eir site as
being in th e a blati o n a rea .
Compari so n of values in the near-infra red band is
compli ca tcd b y th e ra pid dec rea se in spec tra l ref1 ec ti\'iti es
a t 1.4 pm. Additi o nall y, th e sp ec tra l distributio n of sola r
radi a ti o n m ay b e slightl y weig hted towa rd s th e sh o rter
wa \'eleng th s (Monteith , 1973; W eiss a nd ~ o rm a n , 1985).
T a kin g this into con sid era ti o n , th e da ta of Gre nfell a nd
PerO\'ic h, wh en weighted acco rdin g to th e ex p ec ted
di stribution o [ in co ming so la r near-infra red ra dia ti on
(Monteith , 197 3), yield a valu e of 0 .20 for th e w a v elength
ra nge equiva le nt to PNfR , whi c h compa res well with 0.1 8
[or P NIrl a t th e cl ean-i ce site .
338
For eo mp a n so n o f sno\l' refl ec ti\'ities with th e
lite rature, values a re ta ken from sol a r n oo n und er cl ea rsky condi tions (P VlS = 0 .71 , PNIR = 0 .42 ) . Sn olV surface
den sity o btain ed I d b efo re th e ref1 ec tivit y meas urem e nts
was 536 ± II kg m- 3 . L a bo ra tory tes ts w ere conduc ted b y
O 'Bri e n a nd \Iuni s ( 1975 ) for sno\\' (d ensity 525 kg m 3)
ove r th e waveleng th ran ge 0.6- 2.5 pm. Th e ra nge 0 .60.695 pm li es in the \'isible region as d efin ed by th e
prese nt stud y. In this ra nge, O 'Brien and J\Iuni s observed
refl ec ti viti es of 0.85 ; substanti all y hi g h er th a n 0.71 fo r
P VlS . Thi s is a ttributed to impuriti es in th e na tural surface
on P ey lO Glacier , as well as g r ea te r rou g hn ess . A
com para bl e snow surface is reported by Grenfell a nd
oth e rs ( 1981 ) wi th a d ensity of 480 kg m 3 and a n
impurity content < 500 ppm by m ass . Visibl e sp ec tral
refl ec ti\'iti es ra nged from 0.72 to 0 .78. More surface du st
con cen trati on (2000 ppm by mass wh e n converted from
g I I in T a bl e 2), as w ell as possible influ ence from g lac ie r
ice und er th e rela tively thin snowp ac k (0. 3 m thi c k) ,
co uld aceo unt fo r the lower P VIS valu e of 0.71 o n P ey to
Gl ac ie r .
A di stincti ve feature of roek refl ec tivity a t P ey to
Gl acier base camp w as Pl\lR > P VIS, a reversa l from th e
sno w a nd ice surfaces . This beha\'iour was a lso o bse n 'ed
b y Rott a nd Sogaa rd ( 1987 ) during sp ec tral scans of icem a rg in al loca ti on s in w es tern Greenland. Fo r "gra ve l a nd
roc ks cove red in lichens", th eir reflec ti vities ra nged fr om
0.1 3 to 0.1 6 in th e vi sibl e range, a nd 0.18 to 0. 34 in th e
nea r-infra red (co mp a red with P VIS = 0.16 a nd P :-.1 JR =
0.25 on Pey to Glacie r moraine a t sol a r noon ).
Th e results f1"om three ex treme typ es o f ice in this stud y
(c lean , dirt y a nd hummocked ) clarify th e fa ct tha t
impurity co ntent is th e d omin a nr influence o n sp a ti a l
\'ariability of he misph e ri cal ref1 ectivit y. Th e rela tively
min o r difference in va lu es betll'een sm ooth a nd humm ock ed iee surfaces furth er e mph as izes th e w eaker
influ en ce of rou g hness compa red with impurity con ce ntra tio n. Ass uming that th e three selec ted sites do represe nt
e xtre m e eases o f cl ea n , dirt y, rou g h a nd sm o oth
(exc luding cre\'assed a reas ), th e d a ta suggest that n ea rinfra red refl ec ti vi ti cs va ry onl y slightl y O\'er th e e n tire
a bl a tio n a rea of P ey to Glacier. Th us, m os t of th e spati a l
\"a ri a ti o n in m e ltw a te r ge nera ti o n will be du e to
a bsorpti on of visibl e sola r radi a tion. On th e oth er h a nd ,
th e O\'era ll high e r abso rption in the nea r-infi"ared b a nd
pro vid es th e stro n ge r co ntribu tio n to tota l meltwa te r
ge n e ra ti on by radiati ve melt.
Diffe re nces b e t wee n cl ea n ice (sm oo th es t ) a nd
hummocked ice (rou g hes t) indi cate th a t th e d evel o pm e n t of mi croto pogr a ph y during th e a bl a ti on season m ay
redu ce ref1 ectiviti es . Th e Im hummoc k reli ef d eve lo ps
O\'er a number o f melt seasons. Lowe ring of reflec tiviti es
b y up to 0. 1 du e to inc reased shadin g th e refo re represe nts
a co nserva ti ve uppe r co nstraint [or a singl e summe r.
Th e diurn al varia bility of refl ec ti vity a t a ll ice sites
w as sm all in compa ri son to th e vari a tion between ice
ty p es . In contras t to th e grea ter flu c tuation of P VlS for
diffe rent loca ti o ns, it was PNIR whi c h displayed la rger
res pon ses to tempo r a l c ha nges in a tm os ph eri c a nd surface
Cutler alld
co nditi o ns (a lso tru e at th e snow site) . This bi as h as also
bee n no ted in broad-band p a rameteri za tions o f sn Ol1'
re fl ectil'ity ca rri ed out by Marsh a ll and W a rre n ( 1987 )
a nd M arks a nd Dozier (1992 ) .
The redu ced refl ecti\'iti es durin g intensive surface
m elt o nJu lia n da y 2 18 (Fi g. 6) , illu stra te th e co ntributi on
of meltwa ter to th e low temporal varia bility 0[' g lacier-ice
refl ec tiviti es on P ey to G lacier. M elting ice crysta ls a t th e
surface ha\'e less a bilitl' to spec ul a rl ), refl ec t in cid e nt solar
ra di a tio n th a n fr oze n crys ta ls. Additi o nally, water-fill ed
intersti ces be twee n crys ta ls inc rease th e density 0 [' th e
surface laye r, and thus red uce PNIR .
0 . 5 --~--~--~------~--~--~--------~-+ 0.5
. ,.j
>-
~
i=
0.4
VISIBLE
0.3
0.3l
(,)
w
Li
NEAR - INFRARED
0.2'1
~ " ~
o
i
0600
0900
~ I::
1200
1500
JIl III/TO:
R4lerliviLJ! 011 Peyto GLacier
Th e d omin a nt influ e n ce of impurity co nten t on spa ti a l
I'ariation of PVIS ra ises a qu es tion regardin g the impac t o f
impurity acc umula tion ove r tim e, a nd its effec t on m ass
bala n ce. Th e cleallesl patc h of ice ava il a ble possessed low e r
PVIS th a n fo und o n so m e g laciers, wh ic h su ggests th at the
depositi o n a l climate m ay play a role in influ encing m ass
balance. Nega ti ve m ass ba lances have b eco me incr easingly co mmull on P eyto Glac ier in rece nt yea rs: se\'e n
nega tive years during th e 1975- 85 p e ri od (Letregu ill y
a nd R ey n a ud , 1989 ) . Th ese h ave prev io usly b ee n
associated with increasi ng temperat ures a lone, but the
possibility th a t acc umula ting impuriti es h a \'e 10ll'ered th e
refl ec ti \"i ty should a lso b e co nsid ered .
YVhil st th e lack of d iurn al varia tion in refl ec ti vity a t
the ice a nd moraine sites is in itself un exc iting, th is
beh avio ur is useful fro m th e standpoint of rem ote se nsin g .
\ Vi th in a 6 h windo'vV cen tred on so lar noon , suita b ly
processed sate llite im age ry is likely to b e a reli ab le so urce
ofre fl ec ti\'ity estim a tes for ice. as these a re within 2-3% o f
d a il y m ean I'alu es, th o ug h a 4 h window is mOIT a ppro pri a tc fo r snow . Suc h va l ues ca n be used as a base to
whi c h su c h factors as so lar zenith- a ng le d e pend ence ca n
be a dcl ed in mode lli ng th e net sho rt-wal'C radi a ti o n
reg i m e o f th e glac ier su rface .
ACKNOWLEDGEMENTS
1800
LOCAL APPARENT TIME (h)
Fig. 6. The im/Jact oJ high-melt condiliolls Jor ] uliall da)!
218 (lighl jJLol ) all reflecl il'iO', illustrated b)' comparison
wilh Ihe JoLLowillg d(~J' oJ lower mell (bold /llot ) . Sk)'
cOlldiliolls all dCI)' 218 were /Jledominallt(J' clear, wilh 0.3
CllI1l11I11S cloud at all times. The lIerl dCI)' lcas clear
Ihroughoul .
Th e timing o f th e peak melt\\'a ter disc ha rge a t a
g lacier LOngue wi ll be influ enced b y th e diurn a l va ri a ti o n
o f refl ec til·ity. Our results sugges t that peak m eltwater
ge ne ra ti on in th e a b lation a rea sho uld co in cid e wit h so la r
n oo n, a lth o ug h o th e r facLO rs suc h as in creased a ft e rn oon
a ir temperatures , c lo ud prese nce a nd ka ta ba ti c wind s II'ill
co m pli ca te th e si tu a tion.
Th c meas ure m e nts from Peyto G lac ier pose a proble m
fo r ma ss-b a lance m od ell ers in th a t sp a ti a l hetc roge n eity of
refl ec ti vity in th e \'isibl e ran ge mak es it d ifTic ult to
es ta bli sh co rrect "representative" re fl ec til'iti es fo r a ltit udin a l zo nes. O e rl e m a ns ( 1990 ) d emo nstrated th e poss ibili ty of a III m ri se in equilibrium -l in e a ltitud e du e to a
d ec rease of o nl y 0.03 in sho rt- wave refl ec tivity . Yet
diffe re nces of up to 0. 15 II'ere see n b e twee n th e dirty- and
clean-i ce loca tion s, sires with less th an 5m ele \'a tion
diffe rence whic h wo uld thu s be g rouped within th e sa m e
a ltitudin a l zo ne. On c me th od of a lle\'ia ting thi s prob lem
mi g ht be LO ado pt a gridded mod el, II'hereby a reas of
h o moge neo us refl ec ti vity a rc i n co rpo ra tedinto sin gle
el e m ents. Clearly, thi s ap proac h wou ld require d e ta il ed
kn o wl edge of refl ecti\'ity ol'C r th e e ntire glacier a t regu la r
inte n 'als, a feat whi c h is prac ti cal w i th sa tellite im age r y, if
it is suita bly pro cessed to acco unt fo r bi -direc tiona l
refl ec tan ce.
R esearc h fundin g was provided by th e Natural Sciences
a nd En g in ee rin g R esea rch Counc il of Canada. Th e
a uth o rs would lik e to th a nk Pa rks Canada fo r permi ss io n
to \\'o rk o n Pe)'LO G lac ie r, a nd th e National H ydrology
R esea rc h I nstitute (J'\HR 1), Enviro nm e nt Canada fo r
all oll·i ng us to use th ei r base cam p fac ili ty a t th e glac ie r.
Th e co ntinuin g NHRI progra mm e of m ass-b a la n ce
assess m en t a t Pey to i: do in g mu c h to e nco urage
assoc ia ted resea rch work such as th a t wh ic h is desc ribed
here. Sugges ti ons by R. LeE. Hooke o n a n ea rl y draft o f
th e m a nusc ript res ulted in a number o f improl·emenls.
REFERENCES
Andrc\\'s, R . H . G. 196+. .llclt·{)lolog)' alld heal balallte oJ Ihe abialioll area.
SlImlller 19GO ( L OiI'fI' Ire
ll 'hile Glacier. CIII/{/dilll! .·" rlic . lrrilljJflago
S'lalioll: 7920. \ '. 9U 39' 11'. 208111.) . ~I on lrca l. Que" l\ l cGi ll U ni,·crs it ,. Uaco bsl'n ~rcGill Arctic Researc h Expedition 1959 1962.
~lc teoro l ogy I .. \ xel H eibl'rg hland Researc h R cports.
Bergcll.,J.D. , B. ,\ . Hu tc hi nso n, R .T. ~1c~ l ilkll. AD. OZIl1CIll a nd
G. ,J . GOltrricd. 1983. Obse r \'a ti ons on the relation or the short\\'a,'c
rcflecti"ily or recl'nt ly deposit ed snO\l to its p h ysica l properties. J.
Clilllall' . Ipp/. .I leleorol. , 22 , 2 , 193 200.
Bo hre n , C. F. and R . L. Besehta. 1979. Sno\\'pack a lbedo and sno\\'
de nsit y. Cold RI·g. Sci. T eclllwl" b I .47 50.
C ho udhur)" B.J. and A. T. C. C ha ng;. 1981. On th e :tngu lar "ariatio n or
so lar re fl cC'lancc oI'sno\\'.]. (;1'0/)h),5. i?e.l" 86 Cl ). 465 +72.
Co lb('ck, S. C. 1979. Gra in clusters in \lC t sno\\'. ]. Colloid IIIICljilce, 'ci"
72 3), 37 1- 38+.
Dot ic l·. ,J . 1987. R cccntrcsea rc h ill snO\l hvd rol ogy. RCI' . C;eojJhp .. 25 (2 ),
153 161.
Grcnrell. T. C. and D. K . Pe rm·ieh. 198+. Spcc tral a lbedos orsca iec and
in c id clll solar irrad iance in the southern Beaufort Se;). J. C;eo/)IO'.I.
RCJ ., 89 (C3 ), 3573 3580.
G rc nlc ll , T. C., D. K . Pe rm'ie h and J. A. Og re n . 1981. SpL'l'lril l a lbedos
or an a lpin e sno\\'pack. Cold Reg. Sri. Fer/lilO/., 4 (2 ). 121 127.
Ha ll, D . K .. R . A. Bindsc h ad ler. J. L. Foster. 1\. T. C. Chang and H.
S idd a li ngai a h. 1990. Compariso n or ill .Iilu a nd s<l tcl lilc-dcr i\'cd
rc fl cCta nccs or Forbindcl s G lac ier. Greenland. 1111. ]. i?l'Il1ole Selllillg,
11 3 . ·193 50,,,
339
J ournal of Glaciology
H olm gren , B. 197 1. Climale alld energy exchallge 011 a sub-polar ice ((IP ill
summer. Arctic Illslilule of .V'ortlt Ameri((l Daoll fslalld Expedilioll 19611963. Part E. Radialioll climale . Uppsa la, U ppsa la Uni vc rsitc l.
( ~l e t eo ro l og i s ka InstitUlione n . M eddcla nde 111. )
H uble)" R . C. 1955 . ~ I eas ure m e n ts o f diurnal va ri a tio ns in snow a lbedo
on L em o n C reek G lacier. Alas ka. ]. Clariol .. 2 ( 18 ) . 560- 563 .
Let reguill y, A. a nd L, Reyna ud . 1989 . Spa tia l palle rns of mass-ba lance
nuctu a tio ns of :\onh Am erica n glaciers. J. Glaciol., 35 ( 120), 163 168 .
McGufli e, K. a nd A. Hend erso n-Scllers. 1985 . Th e diurnal hystcresis of
sno\\' a lbed o.]' GlaClol .. 31( 108 ) , 188 189.
~ [ a nn s t ei n , H . 1985. Th e int er p re ta tion of albedo meas uremellls on a
sno,,' cO\'c red slope. Arrh . ,lfeleO/'o/. Ceoph)'S. Hioklimalol .. Ser. B, 36 ,
73-8 1.
l\l arkin , \' . A. 196 1. AI'bedo led ni kov Zemli Fra ntsa-I osifa i ego 1'01' v
radi a tsio n nom rez him e a rkh ipcl aga [Th e albedo o f the f ranz J ose r
Land g laciers a nd its role in th e ra d ia ti on regim e of t h e a rchipelago],
Akad . . \ 'auk SSSR . .lIezhd. !lom . PrOl'. .lIe~dh. Ceojiz . Coda, Sbomik
Statl!)'. Ser. fX . Rode! Programlf!.)' .II CG ( Gi)'alsiologiya) 6. 11 1- 122 .
l\ l arks, D. a nd J . Dozier. 1992 . C lim a te and energy excha nge a t the
snow su rface in th e alpi ne regio n of thc Sierra "icva da. 2. Snow cover
energy ba la nce . Il'ater Resou,.. Res., 28 ( 11 ), 3043 305 4.
:v[ arshall. S. E. a nd S. G. W a rre n. 1987 . Parame te riza tio n of sno,,'
a lbedo fo r clim a te models. IlIlematiollal Associalio/l qJ H)'Clrological
Sciellces Pubiicalioll 166 Sympos ium a t \ 'a nco uve r 1987 - LGlge Scale
Effecls qJ Seasollal S/lOW Cover ), 43 50 .
l\ l ellor, ~ 'I. 1977. Engin eering pro perti es of sno\\ . J. Claciol ., 19 (81 , 15 66 .
~1 0 nt e ith , J . L. 1973. Prillri/)Ip", if e/l vironmelllal physics . L o nd on , Arnolcl.
l\lull er, F. a nd C . M . Keele r. 1969 . Errors in sho rt-term a blati on
meas urem ents on melting ice surfaces. J. Clariol., 8 (52), 9 1- 105.
l\ l unro, D. S. 199 1. /\ surface ene rgy excha nge mode l o f glacier melt a nd
nct m ass ba la nce. 1nl. ]. Climalol., 11 (6), 689 700 .
~'i unro, D . S. a nd G. J. Yo un g. 1982 . An opera tion al net short\\,a\'e
radi a tio n model [or glacier b asin s. Water Reso"r. Res., 18 (2) ,220 23 0.
Nu nez, ~[ . , J. A. Davies a ncl P ..1 . R ob inson. 1972. Surfa ce albedo a t a
towe r sit e o n La ke Ont a ri o. BOlluc/ClI)'-La)'er ,'I1eleO/'o/' , 3 ( I), 77 86 .
O 'Bri en. H . W. a nd R, H . l\l un is. 1975 . R ed ancl nea r-infra red spec tra l
renccta nce of sno,,·. In Ra ngo, A. , ed. Operalional applicatiolls oJ salellile
sllowcover o/)eraliolls. Washing to n , D C, \Ta tiona l Aerona utics a ncl
Space Admin istra tion , 345- 360 . (NASA Spec. P ub. 391. )
O eriem a ns, .1 . 1990. A model Jor the sllIjace balance qJ ice masses: Part I.
/11/);lIe glaciers . Utrec ht , Utrech t U nivcrsity. I nsti tute for f.,[ ar·in c a nd
Atmospheri c R esea rch. (1\'10 U R esea rch Report 90- 1.)
O erlemans. J. a nd N . C . Hooge nd oo rn. 1989. ~ l ass -b a l a n cc g ra di ents
a nd cl im ati c ch a ngc . J. Glaciol .. 35 ( 12 1) ,399-405.
0 strem, G. and ~1. l\ 1. Bru!l:ma n, 199 1. Glacier mass balallce measuw/lenls.
A manual Jor field work. A guide Jor j)CI"so/lnei with limited backgrounds ill
glariolog)'. Th ird edilioll. Sas katoo n , S as k., El1\'ironm cnt Ca nada .
Na ti onal H Y'd ro logy Research I nstitut e. (N HR I Sciencc R eport 4.)
R Oll . H . ancl H . Sogaa rcl. 1987 . Spec tra l renec tanc e of snow-em'c reel
a nd snow -rree ter ra in in ,,'es t e rn Gree nl a nd . ,('. Glelscherkd.
Clazialgeol., 23 (2 ) , 11.'i- 12 1.
Sauberer, f . a nd 1. Dirmhirn . 1952 . D cI' Stra hlungsha usha lt hori zonta lc r G lctscherfiac hen a uf dr m llOh e n So nll bli ck. Ceogr. Anll., 34 (34),26 1 290.
Scheibb ner, F. a nd v\'. ~ r a h r in ge r. 1968. Di e Albedo c1 er Sonnblickgletscher unci ihre zcitli chen Va riati o nen. Arch . Meleorol. Geo/)h)'5.
Bioklimatol .. Ser. B , 16 (2 3). 174- 194.
Sch" 'ercltfcgc r, P. 1976. PI,rsical princij)le; qJ micromeleorologicalllleasurements.
Am sterd am, Else\·ier. (Deve lopm ents in Atmosph eric Science 6. )
Siege!. R. and J. R . H owcll . 1972 . Thermal radialioll heal /ran~fer . i\ e,,'
York , l\ l cG ra w-H il l.
Wagner, H. P. 1979. Strahlu ngshausha ltsuntersuchungen an einem O stal pengletscher Il"a hrencl cler Hauptabl ationsperiode. .'Irch . .l/eleorol. Geoph)'s.
Bioklimalol .. Ser. B , 27 (4). 325 333 .
\\'a rren, S. G. 1982. Opti ca l properties of snow. Rev. Geoph),s. S,)aCe PI,ys.,
20 ( I ). 67 89 .
W eiss . A. and J. M. Norm an. 1985. Pa rt itioning solar radiat io n int o
di rec t ancl diffuse visible and nea r-i n(i'a red compon ents. Ilgricullural
and Foresl ,IIeleorology, 34, 205 2 13.
W end ler, G. a nd J. K ell e)'. 1988 . On th e a lbedo of snow in Ant a rcti ca: a
contri buti on to I. A .G .O . J. Claciol., 34 ( 11 6), 19-25 .
W endl er, G. a nd G. Weller. 1974 . A hea t-ba lance stud y o n ~ !c Ca ll
G lacier, Broo ks R.a ngc. Al(l sktl: a co ntributi on to th e I nrcrna tional
H yd rologica l D ec ad c. J. Claciol., 13 (67 ), 13 26.
Ya ma nouchi. T. 1983 . Va ri a ti ons of incid en t so lar nux and snow a lbedo
on the sola r ze ni th angle and cl o ud covcr, at l\ l izuh o S ta tion.
Antarctica. J. .Ifeleorol. Soc. J p", 61 (6 ) , 879- 893 .
Z cng. Q , ~ l. Cao, ~r. Feng. F . Li a ng, X . C hen and F. S hcng. 1984.
Study of' spec tra l renec tion cha racte ri stics of Sl1 o\\', ice a ncl wa ter of
northwes t Chin a . Sri. Sin .. Ser.B, 27 (6 ) , 647 656 .
NI S received 24 April 1995 alld accepted in revised fo rm 28 December 1995
340