Portland State University Portland State University
PDXScholar PDXScholar
Geology Faculty Publications and Presentations Geology
4-1-1985
The Effect of Glaciers on Stream=ow Variations The Effect of Glaciers on Stream=ow Variations
Andrew G. Fountain
Portland State University
, andrew@pdx.edu
Wendell V. Tangborn
Follow this and additional works at: https://pdxscholar.library.pdx.edu/geology_fac
Part of the Geology Commons, and the Glaciology Commons
Let us know how access to this document bene<ts you.
Citation Details Citation Details
Fountain, A. G., and W. V. Tangborn (1985), The Effect of Glaciers on Stream=ow Variations, Water Resour.
Res., 21(4), 579-586, doi:10.1029/WR021i004p00579
This Article is brought to you for free and open access. It has been accepted for inclusion in Geology Faculty
Publications and Presentations by an authorized administrator of PDXScholar. Please contact us if we can make
this document more accessible: pdxscholar@pdx.edu.
WATER RESOURCES RESEARCH, VOL.
21,
NO.
4,
PAGES 579-586, APRIL
1985
The Effect
of
Glaciers
on
Streamflow Variations
ANDREW
G.
FOUNTAIN
Project Office-Glaciology,
U.S.
Geological Survey, Tacoma, Washington
WENDELL V. TANGBORN
Hymet Company, Seattle, Washington
The effect of temperate glaciers on runoff variations
is
examined for the North Cascade Mountains of
Washington
State. The principal influences of glaciers on streamflow are often unexpected contributions
to streamflow volume, a delay of the maximum seasonal
flow,
and a decrease in annual and monthly
variation of runoff. The delay of maximum
flow
is
caused by temporary englacial storage of spring
meltwater and
by
peak meltwater production occurring in midsummer. The englacial storage,
for
one
case,
is
54% of the potential May runoff.
An
algorithm
is
presented that calculates the coefficient of
va~at~on
of runoff
for
a~y
arbitrary glacier cover. The results suggest that a minimum in year-to-year
vanatlon occurs for basms about
36%
glacierized.
On
a month-to-month basis, maximum variation
occurs in July and August for basins with
less
than
10%
glacier cover but
is
a minimum for basins with
glacier covers greater than
30%.
INTRODUCTION
Glaciers are frozen reservoirs of water. The residence time
of potential runoff ranges from a
few
hours to thousands of
years, depending on the size and type of glacier, the season of
precipitation occurrence, and location on the glacier of initial
precipitation deposition. The release of water from storage
greatly affects the local hydrologic cycle by contributing to
streamflow in otherwise low
flow
periods. Understanding
these effects on basin runoff
is
of both theoretical and practi-
cal importance.
For
example, hydroelectric power generation
hinges on the quantity and timing of the water supply, and the
efficient operation of the facility depends on the prediction of
these factors. The prediction of runoff
is
not easy in simple
basins, and the presence of glaciers greatly complicates the
situation. The study of glacier runoff also concerns water flow
through glaciers, which influences glacier movement
[Pater-
son,
1981].
In the state of Washington there are more than 1,000
gla-
ciers [Post et al., 1971], and as the hydroelectric potential and
other water uses
in
the state become fully realized, the impor-
tance of glacier runoff will increase.
One of the first to deal
with the influence of glaciers on runoff in the Pacific North-
west was
Henshaw [1933] who found a lower annual (year-to-
year) variation of runoff from glacierized basins.
Meier and
Tangborn
[1961] demonstrated a longer delay in the summer
runoff maximum from increasingly glacierized basins. They
also elucidated the effect of glaciers in compensating for the
year-to-year variation in precipitation, whereby a glacier will
produce more meltwater in a warm and dry year than in a
cool and wet one, thus reducing the annual fluctuations of
runoff. The occurrence, in July and August, of peak runoff
from a Pacific Northwest glacier was shown by
Meier [1969]
to coincide with a seasonal minimum in cloud cover during a
period of high insolation and low snow/ice albedo. The
summer-to-summer (May-Sept) variation of runoff was
com-
pared with the basin fraction of glacier cover by Krimmel and
Tangborn
[1974]; they showed that the minimum variation in
Copyright
1985
by the American Geophysical Union.
Paper number 4Wl53l.
0043-1397/85/004W-1531$05.00
runoff occurs
at
about 30% glacier cover. Meier and Tangborn
[1961], Meier [1969], and Krimmel and Tangborn [1974] all
discussed the independence of annual runoff variation, in
par-
tially glacierized basins, from annual precipitation variation.
Although these studies have examined the influence of
gla-
ciers on runoff
in
the Pacific Northwest, it
is
the purpose of
this paper to examine more fully the delay of runoff caused by
glaciers and their effect
on
monthly and annual streamflow
variation.
APPROACH
The salient features of glacial influence on basin runoff can
be revealed by comparing runoff
data
from glacierized and
nonglacierized basins
[Meier and Tangborn, 1961]. The criti-
cal assumption in this comparison
is
that the precipitation
characteristics of the glacierized and nonglacierized basins are
similar.
For
this reason, the study
is
limited to a relatively
small, well-defined region in the Pacific Northwest. Most of
the glacierized basins in the state of Washington were identi-
fied
by using two reports: Post et
al.
[1971] and Rasmussen
and Tangborn
[1976]. Several additional basins were identi-
fied
from topographic maps. The nonglacierized basins were
chosen in the same mountainous region as the glacierized ones
and were selected on the basis of their area, altitude, and
location. The most desirable unglacierized basins were those
that had area and altitudes similar to the glacierized ones. The
source
of
runoff
data
was the National Water
Data
Storage
and Retrieval System,
WATSTORE [U.S. Geological Survey,
1975].
All
records were examined to eliminate those influenced
by water diversion and regulation.
Data
from the compiled
stations are presented in Table
1.
To
supplement this data set
and to serve as an independent check,
data
were used from
climatically similar Southeast Alaska (Table
2).
REGION DESCRIPTION
The study centered on the
North
Cascades region in the
state of Washington, which for the purpose of this report
is
bounded on the northern edge by the Canadian border, by
Mount Rainier on the southern edge, and extends from the
eastern face of the North Cascade Range to the lowlands
along Puget Sound (Figure
1).
Washington
is
the most gla-
579
580
FOUNTAIN
AND
TANGBORN:
GLACIER
INFLUENCES
ON
STREAMFLOW
TABLE
1.
Drainage
Basins
in
Washington
State
and
Pertinent
Data
Mean
Mean
Basin
Percent
Annual Annual
Station
Period
Elevation, Area,
Glacier
Runoff, Coefficient
ID
of
Record
m
km
2
Cover
m
of
Variation
Nisqually
River
825
1942-1982
1225
344.5
5.00
2.02
0.164
Puyallup
River
920
1909-1982
1250
240.4
8.70
1.97
0.135
Carbon
River
939
1966--1978
1220
197.4
5.70
1.96
0.173
South
Fork
Cedar
River
1140
1945-1982
1067
15.5
0
2.20
0.197
Cedar
River
1150
1946--1982
985
105.4
0
2.29
0.195
Rex River
1155
1946--1982
1024
34.7
0
2.71
0.193
South
Fork
Skykomish
River
1330
1903-1982
1158
919.4
0.43
2.37
0.214
North
Fork
Skykomish
River
1345
1929-1982
1127
1385.6
0.40
2.55
0.223
Wallace River
1350
1929-1982
811
49.2
0
3.04
0.189
Sultan
River
1375
1934-1971
951
193.0
0.10
3.68
0.171
Woods
Creek
1410
1946--1972
191
146.1
0
0.95
0.185
Middle
Fork
Snoqualmie
River
1413
1961-1982
1131
398.9
0.25
2.87
0.216
North
Fork
Snoqualmie
River
1420
1930-1982
975
165.8
0
2.70
0.196
South
Fork
Snoqualmie
River
1434
1961-1982
1033
107.7
0
2.52
o.i21
South
Fork
Stillaguamish River
1610
1928-1981
793
308.2
0
3.11
0.204
Squire
Creek
1650
1950-1969
771
51.8
0.76
3.20
0.167
Pilchuck
Creek
1685
1929-1982
393
134.7
0
1.86
0.196
Big Beaver
Creek
1720
1940-1969
1341
163.7
0.39
1.56
0.183
Ruby
Creek
1735
1949-1969
1737
533.5
1.00
1.19
0.177
Thunder
Creek
1755
1931-1982
1768
271.9
14.20
2.02
0.138
Stetattle
Creek
1775
1914-1982
1524
57.0
2.30
2.88
0.187
Newhalem
Creek
1781
1961-1982
1262
72.3
0.41
2.14
0.212
South
Fork
Cascade
River
1811
1957-1982
1902
6.1
53.30
3.83
0.092
Salix
Creek
1812
1961-1981
1643
0.2
0
3.15
0.200
Cascade
River
1825
1929-1980
1341
445.5
4.2
2.06
0.180
Sauk
River
1860
1918-1982
1127
393.7
1.00
2.58
0.202
Alder River
1960
1943-1971
390
27.7
0
1.14
0.218
North
Fork
Nooksack
River
2050
1938-1982
1311
271.9
6.10
2.54
0.151
South
Fork
Nooksack
River
2090
1934-1982
914
266.8
0
2.47
0.176
Steheken
River
4510
1911-1982
1564
831.4
3.40
1.52
0.216
Kachess River
4760
1904-1978
1106
164.7
0
1.58
0.255
Cle
Elum
River
4790
1904-1978
1323
525.8
0.30
1.58
0.205
Standard
U.S. Geological Survey
station
ID's
can
be
reconstructed
by
multiplying
the
given
ID
by
100
and
adding
12,000,000.
cierized state of the contiguous United States, and the North
Cascades contain more than
90°/(J
of the state's glaciers. The
topography varies from broad, flat, lowlands to deep, narrow
valleys in the alpine region.
The climate of the western slope of the Cascades
is
maritime
with typically mild winters of high precipitation and cool,
drier summers. Snow accumulation of
8-10 m near an altitude
of
2,000 m
is
common.
On
the eastern side the climate
is
more
TABLE
2.
Drainage
Basins in
Southeast
Alaska
and
Pertinent
Data
Mean
Mean
Basin
Percent
Annual
Annual
Station
Period
Elevation, Area,
Glacier
Runoff, Coefficient
ID
of
Record
m
km
2
Cover
m
of
Variation
Harding
River
220
1952-1981
732
174.6
9
3.71
0.100
Cascade
River
260
1918-1973
964
59.6
13
3.73
0.112
Long
River
340
1917-1973
732
84.2
22
4.81
0.159
Dorothy
Creek
400
1930-1968
946
39.4
16
3.20
0.128
Carlson
Creek
440
1952-1961
671
62.9
10
4.77
0.156
Sheep
Creek
480
1919-1973
580
11.8
2
3.64
0.167
Gold
Creek
500
1918-1981
732
25.3
8
3.77
0.161
Lemon
Creek
520
1952-1973
1046
31.3
67
4.32
0.124
Mendenhall
River
525
1966--1981
994
220.4
66
4.56
0.136
Montana
Creek
528
1966--1976
458
40.1
3
2.31
0.150
Lake
Creek
538
1964-1973
357
6.5
0
1.75
0.170
West
Creek
562
1963-1977
1037
111.9
26
2.64
0.136
Perseverance
Creek
600
1932-1969
409
7.3
0
4.49
0.131
Sawmill
Creek
880
1921-1957
732
101.0
3
4.24
0.170
Baranof
River
980
1916--1974
610
82.9
12
4.44
0.123
Takatz
Creek
1000
1952-1969
702
45.3
19
5.03
0.095
Hasselborg
Creek
1020
1952-1968
366
145.6
1
1.95
0.116
Standard
U.S. Geological Survey
station
ID's
can
be
reconstructed
by
multiplying
the
given
ID
by
100
and
adding
15,000,000.
FOUNTAIN
AND
TANGBORN:
GLACIER
INFLUENCES
ON
STREAMFLOW
581
~,tl"JJ.
,,l
~"~
STUDY
AREA
Q I
,I
I
/
Southeast
----Alaska
---
..
-,:--
--------··---r·-
;
~
\
as~
1 \
1
·,...
Montana i
--·-·-·>
L,
t··
0
,
·~
~·-·-·--·-·-·-1
cu
reg/
'·-·~·
\
a ! Idaho .
·-·-
J.-r-·---.1
i
Wyoming
(1,)
.
·-·-·-·-·-·-·
...
<!
i i
()
! j i..·-·--·-·-·-··
f
Nevada
i I
''·,
i
Utah
I Colo
·,.,
i !
·,
. I
·,.,
L--·-·-·-,-·~·-·-·-·-
.,
j •
·,·,.
!'; I
" ;
't
Fig.
l.
Study region in the North Cascade Mountains of the State of Washington and position of Southeast Alaska
where supplemental basins are located. Each basin used in this paper
is
outlined in the study area and includes the station
number which refers to Table 1; hatching indicates a basin includes glaciers.
continental with colder winters, hotter summers, and much
less precipitation. Maximum precipitation
in
both regions
occurs during the winter, and the minimum occurs in midsum-
mer. Because most of the winter precipitation at higher alti-
tudes occurs
as
snow, the maximum runoff from glacierized
basins
is
delayed until spring.
GLACIER CONTRIBUTION TO RUNOFF
The significance of glacial runoff to the volume
of
runoff
from a basin
is
dependent on the "health" and size
of
the
glaciers. A glacier in equilibrium neither grows nor shrinks
and thus has little effect on the annual volume of streamflow.
If
the glacier experiences a positive net mass change, however,
some potential runoff
is
stored
as
ice, which diminishes the
streamflow. This effect normally has little noticeable impact
because a positive glacier balance usually occurs during a year
of high precipitation, which produces a generous water supply.
Runoff resulting from a glacier's net mass loss augments local
streamflow, which can be especially important during periods
of drought.
The relative effect of a glacier's mass loss on basin runoff
can be calculated by multiplying the glacier mass balance
(averaged over the glacier area) by the fraction of the basin
glacierized and dividing by the mean specific runoff (volume
divided by basin area) from an unglacierized basin of similar
582
FOUNTAIN
AND
TANGBORN:
GLACIER
INFLUENCES ON STREAMFLOW
4.0
(I)
a:
w
._
3.0
w
~
~
u:
II.
2.0
0
z
::l
a:
z
<C
w
1.0
~
PERCENT OF GLACIER COVER
Fig.
2.
The percent increase in specific runoff from a glacierized
basin over a nonglacierized basin of similar precipitation character-
istics as a function
of
mean specific runoff and percentage
of
basin
glacierized, assuming an annual net glacier mass balance
of
-1.0
m
water equivalent; shaded region indicates the extent of glacierized
basins which are gauged in the North Cascades.
precipitation characteristics. The unglacierized basin's runoff
is
used as an approximation of what the subject basin would
yield in runoff were it not for the presence of the glacier. The
results of this calculation, percent increased runoff, are shown
in Figure 2 as a function
of
percent glacier cover and mean
runoff from an unglacierized basin, given a glacier annual
mass balance value of negative 1 m water equivalent (WEQ).
The shaded region concerns the North Cascades where no
currently gauged basin exceeds 53% glacier cover.
If
a basin
has a
20% glacier cover and the unglacierized basins nearby
have a mean annual runoff
of
2
m,
then the annual runoff
from the glacierized basin would
be
increased by about 10%
for a 1-m
WEQ
glacier mass loss. Most of this meltwater
occurs during July and August when the skies are
less
cloudy
and precipitation
is
low [Meier, 1969]. The unglacierized
runoff, however,
is
low during these months and constitutes
often
less
than 15% of the annual runoff.
If
glaciers are pres-
ent and cover 20%
of
the basin area, then the increase in
runoff for the summer period
will
be
as much as 50%.
PEAK
RUNOFF
DELAY
The delay of the peak summer
flow
is
shown by plotting the
monthly fraction of mean annual runoff for basins of different
glacier cover (Figure
3).
A monthly runoff record for a hypo-
thetical basin of 100% glacier cover was generated by sub-
tracting an adjusted value of the monthly discharge from a
basin of
0% glacier cover, Salix (station
1812),
from the dis-
charge from a basin of 53% glacier cover, South Cascade
(1811),
and dividing
by
the area of the glacier. The adjusted
value
is
the ratio between the ice
free
area of South Cascade
and the area of
Salix. Salix basin
is
located adjacent to South
Cascade basin and both are similar in altitude and precipi-
tation accumulation. The unglacierized basin (station
1140)
shows a bimodal annual distribution of runoff caused by a
winter maximum in precipitation that occurs as rain at the
lower altitudes of the basin and a spring maximum from
melt-
ing
of
the snowpack at higher altitudes.
For
unglacierized
basins the maximum runoff occurs in May.
For
glacierized
basins, however, maximum runoff occurs later and later as the
percentage
of
glacier cover increases. The delay
is
caused, in
part, by later snow melting with increasing altitude. A plot
(not shown) of the time of peak seasonal
flow
occurrence from
an unglacierized basin as a function of basin mean altitude
shows this relationship to be linear and of small influence in
comparison with the parabolic relationship between time of
maximum seasonal runoff and percent glacier cover (Figure
4).
The reasons for a glacier-caused delay in runoff can
be
found
by
examining the elements of runoff.
In the North Cascades, the components of summer runoff
are snowmelt and precipitation; if glaciers are present, icemelt
and water released from internal glacier storage
[Stenborg,
1970; Tangborn et al., 1975; Larson, 1978] also contribute.
The principal components of runoff from glacierized basins
are snowmelt
[Rasmussen and Tangborn, 1976] and icemelt,
whereas precipitation and the release of water from internal
storage are relatively small contributions to summer stream-
flow.
The quantity of meltwater from
an
unglacierized basin
is
limited by the snowpack volume. In contrast, a glacierized
basin has a relatively unlimited meltwater source, and the
quantity of melt produced
is
only limited by the available
energy. The variation in meltwater production
is
controlled
by
solar radiation, cloud cover, and snow-ice albedo, and the
combined effectiveness
of
these factors for meltwater pro-
duction reaches a maximum in the North Cascades in late
July and early August
[Meier, 1969]. The snowpack generally
disappears in much of the North Cascades by late June,
al-
ready having made its significant contribution to runoff. A
glacier, however, continues to melt
at
an increasing rate until
midsummer. This peak melt rate in midsummer
is
one cause
for
the late appearance of maximum runoff from glacierized
basins.
Another delay mechanism
is
the temporary internal liquid
storage of meltwater in the glacier.
Tangborn et
al.
[1975]
calculated that in May of
1970,
0.20 m of potential specific
runoff was stored in
South Cascade Glacier. The specific
30
.
20
100%
GLACIERS
(ESTIMATED)
u.
u.
10
0
z
0
::>
a:
..J
30
<
::>
z
20
14.2% GLACIERS (STATION
1755)
z
_ _,...___,
<
u.
10
0
._
0
z
w
(.)
30
a:
w
0..
20
0% GLACIERS
(STAT
ION
1140)
10
..J
<
~
~
Q
30
~
...
~.
E~rER~
TURj:E
21
o:
:~~w~
~ ~
20
. : PRECIPITATION :
!:!::10 .
~·
~
~
. . .
~
£I:
0 I
ffi
o..
0 N D J F M A M J J A S -
1
0
~
0..
Fig.
3.
Monthly fraction of the annual specific runoff for basins of
various glacier covers. The monthly fraction of precipitation and
mean monthly temperature (Snoqualmine
Pass, Washington) are in-
cluded for comparison.
