Jourmd of Hytb,'dogy 141 (1993) 5-31 5

Elsevier Science Publishers B.V., Amsterdmn

[21

The role of permafrost and seasonal frost in the

• o

hydrology of northern " ' , ~ " ~ xt~+~. America

Ming-Ko Woe a n d T h o m a s C. W i n t e r b
"Dppartment Of Geography, !?4cMaster Unirersity, Hamilton, Ont. LSS 4KI. Canada
bU.S. Geological Survey. Box 25046, Mail Stop 413, Oeuver Federal (.'enter, Dr,nver CO 80225, USA

ABSTRACT

Woo, M.-K. and Winter, T.C., 1993.The role of permafrost and seasonal frost in the hydrology of northern
wetlands in North Amerie:~ ! Hydrol., 141:5 31.

Wetlands are a common landscape feature in the Arctic, Subarctic, and north "i'emperate zones of
North America. In all three 2ones, the occurrence of sea,cc,nal frost results in similar surface-water processes
in the early spring. For example, surface ice and snow g:nerally melt before the soil frost thaws, causing
meh wa|er to I'~owiulo depressions, over the land surtacc and at times, across tow topographic divides.
However, evapolranspiration and ground-water movement differ among ~he three climatic zones because
they are more affected by per,nafrost than seasonal frost, l'be water source for plants in the Arctic is
restric,ed to the small volume of subsurface water lying above the permafrost. Although this also is true
in the Subarctic where permafrost exists, where it does not, plants may receive, and possibly reflect, more
regional ground-water sources. Where permafrost exists, the inter;'ction of wellands with subsurface water
is htrgely restricted to shallow local flow systems. But where permai'rost is absenL iu paris of the Subarctic
and all of the Temperate zone, wetlands may have a complex interaction with ground-water-flow systems
of all magnitudes.

INTRODtJCT1ON

W e t l a n d s a r e a c o m m o n l a n d s c a p e f e a t u r e in t h e A r c t i c a n d S u b a r c t i c
z o n e s o f N o r t h A m e r i c a , a s well a s in t h e glac~iated p a r t o f th,e T e m p e r a t e z o n e
( W i n t e r a n d W o o , 1990). W e t l a n d s h a v e p h y s i c a l , c h e m i c a l , a n d b i o l o g i c a l
c h a r a c t e r i s t i c s t h a t reflect t h e i r h y d r o g e o l o g i c a n d c l i m a t i c s e t t i n g s . T h e
t i m i n g , m a g n i t u d e , a n d f r e q u e n c y o f , ~ o m e h y d r o l o g ; c p r o c e s s e s in w e t l a n d s
in t h e A~ctic, S u b a r c t i c , a n d n o r t h T e m p e r a t e z o n e s m a y v a r y c o n s i d e r a b l y
b e c a u s e o f t h e v a r i e t y o f c l i m a t i c a n d p h y s i o g r a p h i c s e t t i n g s in t h e s e z o n e s .
However, frozen ground, either permafrost, seasonal frost, or bodl, may be
a s i g n i f i c a n t f a c t o r a f f e c t i n g t h e h y d r o l o g y o f w e t l a n d s in t h e s e z o n e s .

Correspondence to: T.C. Wimer, U.S. Geological Survey, Box 250,i6, Mail Stop 413, Denver
Federal Cecter, Denver. CO 80225, IJSA.

0022-t~i94J93/$06.00 ~,i) J993 . Etsevier Sc~e~acc Pt~blishers B.V. All rights reserved
11 M.-K. WOO AN[) T,(?. WINTER

The purpose of this paper is to review and evaluate the effect of frozen
ground on wetland hydrology in the Arctic, Subarctic, and north Temperate
zones of North America. The hydrologic processes discussed are: (1) surface
runoff; (2) evapotranspiration; (3) interaction with subsurface water.

PHYSIOGR..eHY, CLIMATE, AND FROST OF THE ARCTIC, SUBARCTIC, AND

NORTI-I TEMPERATE ZOI':2S

For the purposes of this paper the Arctic is defined by the limit of
continuous permafrost and the Subarctic is defined by the limit of discon-
tinuous permafrost (Fig. 1). The term Boreal, as used here, refers to the
southern fringe of the Subarctic. Both the Arctic and Subarctic include
mountains, central lowland plains, Canadian Shield, and Hudson Bay
lowlands. In addition to these physiograpiaic types, the Arctic has an extensive
area of coastal lowlands. The north Temperate zone considered here is limited
to non-mountainous areas north of 42-45 ° latitude that have fi'ozen soil in
winter. Physiographic types include the Canadian Shield and Ehe area of
glacial drill, that rims the Canadian Shield (Fig. 1).
The climatic factors of interest are precipitation and the difference between
precipitation and evaporation. Precipitation for the northern half of Norlh
America east of the Rocky Mountains' increases relatively uniformly in a
southeasterly direction from 200 mm in the Arctic islands to 1600 mm along
the Atlantic coast (Fig. 2). On average, snow covers the ground for more than
70% of the year in the Arctic, 50-70% of the year in the Subarctic, and
30-50% of the year in the north Temperate zone (Fig. 2). The similarity of
snow-cover distribution and climatic zones can be seen by comparing Figs. 1
and 2.
In the Arctic, precipitation exceed~ open-water evaporation by 0.-300 mm
(Fig. 3). In the Subarctic, precipitation s!igbtly exceeds ol~cn.-wa~er evaporation
in, and near, Alaska, bul between the Rocky Mountains and the Canadian
Shield in bolh the Subarctic and norlh Temperate zones, the reverse occurs.
Continuing east in both zones, precipitation increasingly exceeds evaporation
to values greater than 500 mm at the Atlantic coast.

Occurr~'nce o["frost
The duration and depth of frost are controlled Imgely by climate. Alll'~oush
the gc~teral distribution of permafros! is indicaled in I:ig. I, a mimber of
gtound~teni|~erature measurements, supplemellted by depih-lo-!¥ost measure-
merits using a steel rod, in several North American wetlands provide specific
data on the exlcnt of frost in organic ferrain (Table 1). (Locations of specific
sites mentioned in this paper are shown in Fig. 4.} Arctic wetlands, as
TItE HY )RO O( Y I~IF NI')RTIIERN WETLANDS IN NI:)RT}I AMERICA

BO° 70 ~ 80o

50 °

40 °

30 °

130° 120° 110° 100° 90 ~ 80~ 70 °

EXPLANATION
A Arctic M Mour,~tns. south of subarctic ~ Dc~sertic
S Subarctic G Glact~ terrane (eastern) K s~a~ol¢
C C~nadianShield (temperate) p Glacial terrane (prairie) ~ C~ddunes
R mv~fine

Fig. I P h y s i o g r a p h y o l N o r l h A m e r i c a , a n d d i s t r i h u l i o n of general h y d r o g e o l o g i c types o ' / l a k e s a n d

w e l l a n d s , {From Winter and Woo, 1991L)
M.-K WOO AND I'.C. WINTER

% f
,', I

/ ! i? I~
I , / / / ¢1 f . t # "

I pl 1 ( , / t
prleipn~lon / ; /I ~0o --.'

dl.db~dol~) / (I/: ~ ~ " / J

EXPLANATION " A I i "! / '~/

i 5 0 ~
AVERAGEPERCENTO~
TIM~GROUNDHAS /+ t ',:, :, --0",t /
SNOWCOVER
- - ,4~-. ~ MILANANNUAL
P~ECWITATION,IN
MILLIMI~TERS
' ~ I \ I, ,, :" ,~
I

Fig, 2. Mean annuai precipitation for North America east oflhe Cordillera, il~ millimelers '~mm) (modilied
from UNESCO, 1978J, and average pereenlage of lime during a year Ihat Ihc ground has ~now cover, based
on the period 1966,-1980. (Modi[]ed from Dewey ai~d I-tetra, 19827

exemplified by tile Lone Gull (renalned Kiggavik) site west of Baker Lake in
centrnl Keewaiin (l:ig. 5(A)), are underlain by continuous permafrost, and the
near-surface zone (or tile active layer) is thawed only for 3-4 months each
year. Ground temperatures may fall below - 2 0 ° C in winter, and the summer
thaw depths are less than 0,5 m. In the Subarctic, organic terrain commonly
includes permafro,~t patches (Fig+ 5(B)) (Brown, 1977). trl arcas of no
l~rmal'rost, wetlands tend to have thick seasomd frost (Fig, 5(C)), Depth of
seasonal t]~ost is leas! ill file Tcrnperate zone, and some wetlands ir~ this zone
may not t¥ceze at all during mild winters (Fig. 5(D)).
In areas of no permal¥ost, wetlands commonly have a thinner layer of frost
than mineral-soil terrain (Brown and Williams, 1972; Price, 1983). The
|HI{ H~t t)R(II,I)¢LS~ (;'1 NOR'I'I|EI~,NWI~I'LANI)S ~N N(>[{Ilt AMI~RI(.A

60 ° ?0 ° 80*
r

50*

40*

.'i¢0"

20 °

- - 4 0 - - f'RE.CIPITA'IqON M~NUS EVAPORAT,ON, "~p~.~'. iq) ~ . ~ j ~

130 ° 120 ° 110 ° 100 ° 90 ~ 80 ° 70 °

Fig. ~,, I)isiribulion of the diHerence belween precipil;it~,)~ and open-water evap¢ ratio ~ in centimeters.
(I);lla f~'oin Wil~lcr alld Woo, 1990.)

reasons for this phenomenon are: (t) heat toss is more rapid from mil~eral soils
because their thermal conductivities are usually higher than the lichen- arid
moss-covered pea-~s in wetlands; (2) much latent heat dissipation is required
to freeze water in saturated wetland soils, However, in spring, these two
l~ctors operate in reverse to retard thawing of organic terraiq. Freezing and
~ ~ ~ ~',.~ = o= b.: ..~ ..,= ~ .= .~ ~ .,=

.P,

NN
.g

Z Z Z Z Z Z Z Z Z Z

o oo
F,
gKK
5

•~ - ~ - ~ . ~ ~ ~ ~ ~ ~ ~ ~

==== oo

el)

~3 E~

=.=~ .:I =~
-5 o

!
-g

.~ ~ ; !

[!- ~ ~. ,..:. ,.,, v: z =

~ e e
~,., Z Z
'|Ilk tl;H)RO[ ()(;Y ()t NOI{]fIFRN WFI'I ANDS iN NOR rlt ~M|:RICA

Limit of Arctic
Limit of Subarctic
E

[-)~- ogoton~k ...'e" , GO

C Lone C~J (Htggavlk)

D W~sldngton Creek
E Little Buffalo P~vet
F Churchgl
G Wlnisk Rl'~r ,I
Klnole Lake • r/r, )
I Kwataboabegar~ Rlver , 5
I Wasl~uc~a~ ~ w r
; " Schelferellle
L S l a ~ Lake
N Candle Lake
N Nlpawtn Park
O St. Denis
i P Cottonwood Lcke area
, Q Ros~.au Rlw:r
I ~ M a r ~ l ] Forest
S Beverly Swamp
T Met Bleue

0 500 MILES

0 5go KILOMETERS

r i g . 4 L o c a t i o n s o. specific sites m e n t i o n e d in this paper.

thawing proce:s:;es in peatIands are highly asymmetric; icy peat conducts heat
at a rate several times faster than unfrozen peat of equivalent moisture content
(Farouki, 1981, pp. 50-54). Thus, heat loss by conduction is more rapid in
winter than he~t gain in frozen soil by conduction in spring.
The formatic n of seasonal frost is strongly affected by the ground co~er of
12 M ,K WOO AND I C WINTFR

A. L o n e Gull, Keewatin abov~ freezing

1/1 Water level Temperatures are
04• I .............................., , / o f Water Fable not shown
• /
i o.= F snow ~ . s~te /
| ......... flooded ._../~"
I£ 9.0 b.............7"--T ' v - c - \ '-~x,iLi.....................~ ; ~ ...............

IX ~ x ~ _ ~ ~ ~ temperature
-0.6 ~ - - - ~ - - ~ r - , r
MAY JUNE JULY AUG.
1983

B. Washington Creek, Alaska

9.6 - ~ - - ~ M ~ - ~

0.4 - /~ ~
192 ~ j Snow
LLI " - \
9•9 ................ 20 10 I. 0-fi . . . . . . . . ~5.-1.0. -10 :_.10 _.:_1_5_. _ -10 -5

ILl -9.4 - kineofequal ~') \ [

"~ I temperature ! ) ~] ,

1974 1975
Fig. 5. Ground temperature, soil frc,~t, and snow cover at specific sites in the Arctic (A) (from Roulel an(
Woo, 1986), Subarctic (B) (fi'om Kane et al., 1975), and north "temperate zones (C) (from Kingsbury an¢
Moore, 1987) and (D) (from Smilh and Woo, i986). Beverly Swamp, Ontario (D), did not have soil frost.

wethmds. The insulating effect of snow on ereeze and thaw of organic terrain
was demonstrated by Outcalt and Nelson (1984) usipg a one-dimensional
model Field data fi'tm~ a st, barctic wetland verilied the significarme of ihe
insulating clrect of ,Hmw on t¥ost penetration in wetlands (Moore, 1987).
Vegetation covet" may shade and insulate a wetland to retard thawing (Cowell
et al., 1978), but the effect on the thermal regime is probably small. For
example, Brown (1976) failed to detect temperature differences in the top I m
of forested and strip-cut parts of a bog in northern Minnesota.
The convection of heat by surface wate.r aid ground water w'celerates
thawing of seasonal frost, in |he Boreal zone. FitzGibbon (1981) and
Swanson and Rothwell (1986) lbund that heat exchange between ground
water and fi'ozen peat was responsible for enhanced thawing from beneath the
frost layer, In some cases in the north Tempe 'ate zone. winter flooding of
wetlands can prevent formation of frost (Woo and Valverde, 1982)• However,
I'HE HYI)Rt)I.QL;Y O : NORTHERN W|'TI'I.ANI)S IN Nt)R] H AMERICA

C. Scheffei~,,'ille, Q u e b e c
08 I-

0.6

0.4

0.2
O3
tic 0~ -2,5 -15 ~10
. 0.0

~; 0.2
i
Z
0.4 ~-'--'-"~, ~ -- .. pore ice ~ / / ]

k~ 0.6
121 -- Segregated
,ce-- " / / !
o.a

-1.0 1

-1.2 i Unt'rozen Peat

-1.4 ~ - - - 1 q • r q • r---~
OCT. NOV, DEC. [ .JAN. FEB MAR. APR, MAY JUNE JULY
198~t 1985

D. Beverly Swamp, Ontario

Water level
Snow \ or Water Table ~ S ....

o _ / I
2 \

~u \ \ '~ \ \ \ kJ'--~4~" i~ \ '\6 \ !

-~-10 - ~ 8
~'- \ \ 10 Jr !

[ \ ~ / ' k temperature I I
.20 b~-d~J~h' FEB.
MA--"--~7-A~-n~A~v~Su.~ J~'L~---~&-U--d%E~T~
OCT~5~. r;~N.
1982 1983 1984

Fig. 5 (continued).
14 M-I<. WOO A N D T ( " WINTER

(,t} . . . . . . . . AIR T[~MPERATLIRE

0~
I<> "r'ESURFACE
MPE~ATUREOF
WATER
/[ I

fj I SEDIMENTS HALF I [ [[ J. l
/ T~.~'~g ~ ~ /

a / #<'! ,~ E

J
f
.15 [. . . . . . . . . . . a...................... x......................... 1. . . . . . . . . . . . . . . . . . . . . . . . . . 1 ......... ~ ..................
OCT, NOV, DEC, ! JAN, FEB. MAR APR.
1987 1988

Fig. 6. Temp~:rature o f a i E , water. Rlld s e d i m e n t 0.5 m below t h e b o t t o m o f w e l l a n d PI in t h e C o l l o n w o o d

L a k e area, N o r t h D a k o t a , d u r i n g the w i n t e r o f 1 9 8 7 - 1 9 8 8 .

i~'~some small prairie wetlands in the north Temperate zone, the open water
and the upper few decimeters of sediment may freeze completely, as is the case
at a semipermanent wetland in the Cottonwood Lake area, central North
Dakota. At this site, a probe containing thermistors at several levels needed
to be pounded through an ice lens in the sediments after the pond ice thawed
every spring. Furthermore, sediment coring on 13 March 1984, indicated that
the sediment was frozen to 30cm below the pond bottom. Following the
open-water season of 1987, the thermistor probe was left in place over the
winter, A graph of air temperature, surface-water tempeiaturc, and sediment
tenal~)erature 0.5 m I~elow the sedimcnt~-,waler interface indicate~ the progression
of tcniperature cllange of all three front I October 1987 to 30 April 1988 (Fig.
6). At the time the water fi'oze at the end of November, the sediment tem-
perature was 6°C warmer. The sediment temperature continued to decline all
winter to a minimum of about t .5°C at the end of March. The rapid warm-,up
of air in April resulted in a similar rapid warm-up of the l)ond and sediment
lenlperatures. The sedmletll telllperature agahl htgged Ihe increase m water
telnperalut'~ {~y a s nluch as 5 ° ( , ~,

F*'o<Yt;gnd s;r~mnd ice

The penetralio~ of freezing temperatures into saturated soils Freezes the

TIlE H Y D R O L O G Y O F NOR1 q)'-'RN WETLANDS IN N O R T H AMERICA

moisture in situ. In unsaturated wetland soils, upward migration of moisture

to the freezi~, g front will produce segregation ice (Ad Hoc Study Group on Ice
Segregatior~ and Frost Heave, 1984). This transfer of moisture can continue
in the peat even after soil temperatures fall below 0°C (Juusela, 1967). The
occurrence of ice as veins or as fillings in soil pores has been observed in frozen
wetlands (FitzGibbon, 1981; Kingsbury and Moore, 1987). In addinon,
ground-water seepage m~rmally freezes as it encounters the cold air at the
ground surface, creating a layer of ice (or aufeis), so:metimes mixed with the
snow, on wetland surfaces (Kane and Slaughter, 1972; Price, 1983).
The presence of an ice cover or the abundance of ice in the pores of wetland
soils greatly reduces the rate of melt-water infiltration. Few measurements of
the hydraulic conductivity of fwzen wetland soils have been made (Slaughter
and Kane, 1979), but for mineral soils the hydraulic conductivities are orders
of magnitude lower in frozen sails ~han in unfrozen ~'~oils(Burt and Williams,
1976). It is to be expected that frozen wetland soils will curtail the entry of
water, although Kane and Bredthauer (1981) demonstrated ex!)erimentally
the infiltration of melt water into frozen organic materials. Ho,'cever, if the
frozen soil is very cold, infiltrating watec will freeze, and seal the ~oil pores to
fimher infiltration.

HYDROLOGIC PROCESSES

S.t:/a~'e ,'u.off
The general pattern of surface runoff for southecn Canada and ~he northern
United States was mapped by Hare (1980) as a runoff ratio (mean annual
runoff/mean annual precipitation). For the Subarc6c and north Temperate
zones, the ratio increases from iess than 0.1 in the interior plai~:s to greater
than 0.7 at the North Atlantic coast (Fig 7). Although annual ~-t~noffcharac-
teristics are important knowledge as background information, seasonal runoff
patt,~rns are of greater importance to understanding wetland hydrology.
Beck:use studies of the hydrology of wetlands are not comm~:~n, only a few site
studies are available to use as examples for '.he three zones ~,f interest in this
paper,

Freest and smJace runoff"

Most wetlands subject to frost have continual snow accumulation without
a significant mid-winter melt. Warm spring temperatures melt the bulk of the
snowpack within a few days or weeks, while the soils are still frozer~. A large
suppb of melt water entering frozen wetlands quickly exceeds their ~urface
storag,~" capacity and generates extensive scwface ttows across we'dands.
16 M-K WOO A N I ) T . ( ' . WINTER

7,

i,'i~ ? tltl~loll h m o , dcrl~ed ;1~, Ih,,~ itllhl ii1 II1~,i111;lllllll~II llHtOJ]' [t'~ [Ileal} lll}lltlit[ precipilalion, (Modified
II01"II|I~II'¢,1981k1

Widespread flooding of wetlands during the melt period has been reported ~'~r
the Arctic (Ryden, 1977; Woo, 1988L the Suharclic (q'hom. 1072}. and the
Boreal part of lhe Std~trcfic zqme (Woo and I)iCcnz., 1989), I:lot'~dingof llat
wctLmd te~°r~fin cm~ obliterate the dislitlcliofl bclWCelt ch;mn¢led flow.
dcptesqon storage, and overland flow, Patterns of Stlrl]ice flow change as
st~ov,- and ice melt progress. Flow may even cross indistinct drainage divides
(Fig, 8), carrying wili~ it slush, ice, and organic debris (Thorm t972; Woo and
Heron. 1987bk
TIlE IIYDR(3LOGY OF NOR'fltlLRN WEll_ANDS IN NORTH AMERICA

APRIL 22 ..--- -- il APR,L23 o.m. . .... ]

0~0m _f_1:4_\" :,:
I,>
(Li a
o..>-<-
J 2
-.-:1
- " I

1 / ]
' /S{L":-?Y \'-~
.X<---~ ," _- 2 . -

I APRIL 23 (p.m. . .... -~ r-] ~, ~, APRIL2~ _:7 -

I~ . i -~- 4-.

II y\ "-

.<>~;.-- o, #,

--I APRIL ~2~-29,-_- EXPLANAi !ON

El-:} MARSH"O~R~SSION
" W O O D E DRAISED gEACI I
[ J AND LEVEES

[-"~ OVERLAND FLOW ZONE

K

DIRECTION OF FLO~,&'
////
.L a STREAM CHANNEL IN
SNOW-FREE PERIOD

Fig, 8. Surfiice llow I'rl~m s n o w m e l t a c r o s s a w e l l a n d ~ e a r J a m e s Bay. ( D a t a f r o m W o o :rod H e r o n , 1987b.}.

In tile north Temperate zone, surface runoff m the spring can be greatly
affected by seasonal frost. Snow melt and spring rains commonly occur before
wetland soils lhaw, In addition, ice commonly forms on wetlands when the
instllafing snowpack melts and nighttime temperatures arc below lceezing.
t~ M - K WOO AND I'(" WINTiR

19"19 I~80 1981 1902 1983 1984 1985 1986 1987 198R 1989 1990
YEAR

Fig. 9. Altitude o f the water surface of wetland PI in the C o t t o r w o o d Lake area, N o r t h D a k o t a ,

1979- 199(k

These processes result in overland flow to stream valleys and to closed

depressions. Because the water does not infiltrate, streamflows and water
levels in depressions generally reach their highest levels of the year (Fig. 9). In
this respect, for most years, spring runoff in north Temperate wetlands is no
different than spring runoff in the Arctic and Subarctic because seasonal frost
prevents infiltration during the spring melt in all Ihrec zones. However, frost
is not a necessary condition for spring flooding. Flooding results whenever a
water supply exceeds the storage capacity of a wetland. At Beverly Swamp,
Ol~tario, ,vhere the soil seldom fi'eezes (Fig. 5(D)) but where a snow and ice
cover is commonly present at the time of melt, surface flow in spring may
sweep across most parts of the wetland.

Streamflow
Streams in Subarctic and norlh Temperate welhmd lerrain generally have
low tlow duri~g winter. (Streams in the Arctic generally do not flow at all
during win|or.) Dt ring the time of frost growth, gradual drainage and freeze
up of ground ware|' takes place, with little replenishment from snow melt (Fig.
5,,_)).
t[~ "...... : ...... oect;r 11; ",~,,~t.~.u~
E^e~,pt,o.~ " *~-
' ' a ~ | " ' ~ ' ~ " occupyirig . . .3LPLItItl;.I
tllt~." . . . . ~"--~l part of "--
tlllff
Temperate zone, where winter rain and intermittent snowmelt sustain a high
THE HYDROLOGY OV NORTHERN WF-TLANDS IN NORTH AMERICA 19

water table in winter (Fig. 5(D)), or where fens are fed by a large ground-
water-flow system (Verry and Boelter, 1978).
In northern cliff, ares, streamflows are commonly high in spring. This
applies to streams draining wetlands in the Arctic (Brown et al., 1968; Ryden,
1977), Subarctic (Price, 1987; Woo, 1988), and north Temperate zones
(Wimer et al., 1967; Bay, 1969). Hydrographs of three northern Ontario rivers
draining wetland basins ~,f different sizes in the Subarctic are shown in Fig.
10. The smallest river, Wa~hkugaw River (basin area 175 km 2) has negligible
flow during winter, estimated to be less than 0.00t m 3 s -~ (Woo and Heron,
1987b). The annual peak flow of 1984 occurred dm'ing the melt period when
the entire wetland was still underlain by seasonal frost. Subsequent rainstorms
in summer produced high flows but their magnitudes were lower, and their
duralions were shorter, than flows during spring. The Kwataboahegan River
(basin area 4250 km 2) had winter baseflow of about 1.5 m 3 s -~ , but flow rose
sharply in the melt season to peak a~ 662 m 3 s -~, which occurred about the
same time as the spring peak of the Washkugaw River. Streamflow responses
to summer rainfall were dampened by !arger wetland storage in the Kwata-
boahegan Basin compared with the much smaller Washkugaw Basin. The
Winisk River (basin area 50 000 km 2) had a similar pattern of discharge as the
Kwataboahegan River, even though the cec,troids of the two basins are 500
km apart. These examples indicate that streamflow in wetlands of different
size but having similar environments respond in the same way to the spring
melt when the wetlands are uuder the influence of frost, snow, and ice
conditions.
These same flow characteristics are exhibited by' streams draining wetland
terrain farther south in the north Temperate zone. For example, the Roseau
River of northern Minnesota drains part of an extensive area of wetlands in
,~he Glacial Lake Agassiz plain. A !0--year b.ydrograph indicates spriog peak
flows commonly are greater than 28 m 3 s- ~, and the stre~ m discharge recedes,
with minor peaks, throughout the remainder of the year to less than 0.5 m 3
s ~ (Fig. 1 t). The flow duration curve for this river indicates some release ot~
water from storage at low flow conditions, but the amount of water released,
including that released from ground-water storage, is not large (Fig. 12).

Evapotranspiration

Research on evaporative energy fluxes is limited, but such studies have been
conducted for a range of cold environments (Addison, 1977; Munro, 1979;
Roulet and Woo~ 1986; Lafteur and Rouse, 1988). The formation of f?ost and
the persistence of snow cover lhat accompany cold temperatures lirait the
length of the active evapotranspiration ~:eason (Dewey and Helm, 082).
20 M,-K. WOO A N D T,('. W I N I E R

l~13Alld M~,'OI1)II-ISVM
~H1 UO..IONOOa8 '~3,:l
SI:EI131N 31Bf"13 NI '391:IVHOSIEI
o m o

ea

...... .:I

==

-, . . .~,~

's

'E

"t3

.o

S'IJ3A|~INV~3HVOBV~VM)t ON~f )tSINI/'IA3H1 UO:J

GNO33S ~3¢JSEI313~ 311~rl3 Ni '3~)~VH3SIQ
YDROLOGY t~F NORTHFRN ~,VI~TI.ANDSIN ~O~.TH ~tMERtCA 2

IIIIIII I I "-'7111111 I [IIIIIFI-~-- ]1111111 I [11111 I ~-~

~ .~,~='-..

?
M.I E
O

I
g
d

.E

. . . . . . . . , I

=
ONOO:3S 1::13dSB3131N OlSf"l~) NI '3~)UVHOSIO A~IIVEI-NV31N
~2 M~K WOO AND T ( W~NTE|~.

I / " Roseau River at Ross (1929-621

m
~ 2.8
n

w
i-

:D .2,~ --'
O
z_

° \

O0i 01 I ~' 5 10 a0 50 I0 9;) 95 98 999 9999

PERCENT OF TIME DISCHARGE WAS
EQUALED OR EXCEEDED
Fig. 12, Durat~o/1 curve of annual discharge of the Roseau River at Ross, Minnesota, 1929-1962, (D~la
from Winter ~,t aL, 19673

Based oxl the duration or snow cover (Fig. 2), the mean period or active
evaporation decre;,ses £ron'~ 70% of a year in the north Temperate zone to less
than 30% of a year in the Arctic.
Annu~I evaporation decreases poleward because evaporation seasons are
short and radiation energy input is low in the Arctic and Subarctic. Frost at
I'IIE IIYDROI.O(IY ()~ NOI~.'i-IIFRN WETLANI)S IN NORTH AMERICA 23

shallow depth throughout the thaw season requires much lateat heat t o melt
the ice. This retards warming of the soil above and consequently creates a
steep temperature gradient between the soil surface and the frost layer. This
large temperature gradient uses about 18% of the net radiation to heat soil in
areas of permafros~ compared with about 10% in the Temperate zone
(Halliwell and Rouse, 1987). The remaining net radiation available is par-
titioned into evaporatioa and warming of the air. Under cool conditions, the
vapor-pressure deficit in the atmosphere and evaporative demand is generally
low. Rouse et al. (1987) found that the Bowen ratio (the ratio of sensible to
evaporative heat flux) decreases with air temperature over Subarctic coastal
marshes. At high latitudes where the summer air temperature is lower than it
is farther south, more of the r~et radiation may go to sensible heat than to
evaporation.

Interaction of' wetlands with subsu~J~,ce water

In the Arctic, shallow subsurface wa ~er consisting of soil water and shallow
ground water lies above tke permafros~ and deep ground water lies beneath
the permafrost. The permafrost serves as a continuous confinins bed because
of its minimal hydraulic conductivity. Uf,,on melting of seasonal frost, the
most active zone is the shallow subsurface water which is usually less than 1
m deep (Table 1), with tl'.e water table bein~ close to the land surface (Fig.
5(a)). Any additional source of water, either from a rainstorm or from lateral
influx, results in a rapid rise of the water table, sometimes to the wetland
surface, generating overland flow (Roulet and Woo, 1988).
In Subarctic latitudes, the advance of the freezing front may be faster than
the decline of the water table, which can significantl3, alter ground-water-flow
patterns. Price (1983) fotmd that the water table in a bog in central Sas-
katchewan, fell faster than the advance of the freezing i'~-ont; but for a fen, the
arrival of frost in the saturated wetland was accompanied by a rise in the
hydrostatic pressure (Fig. 13(A)). When the ground water emerged through
the frost to the fen surface, flooding and icing followed (Price and FitzGibbon,
1987). At a coastal marsh in northern Ontario, piezometc.'r measurements
during summer and winter along a 0.6 km transect showed ttiaat the descent
of the freezing front was uneven, being faster and deeper beneath the
windswept, snow-free, opep marsh than in the snow-covered, therefore
insulated~ wooded area. The formation of the fl ost bulb blocked the ground-
water flux, forcing ground water to move laterally beneath the wetiand (Price
and Woo, 1988) (Fig. 13(B)). At anofiler site in northern Ontario, uneven
241 M,-K, W O O A N D T ('. WlNTI~I{

A,
-l 3
Discharge from I
BOG 17 4 krn~ basin t~ ] 2~ FEN Possibility of
I ico' ...... ~on I
E
....
0.0 r
F
',,.
-.I- '~
~_. 6"0 !--'--~'~,~x. . . . . . . . . Ss e- ~ao n - a i . . . . . . . .
ud I FrostZ~ frost
:~ 6.=~ tabl~ ",_.

3E 0.4~- piezometrie~ ~ . . . . . . .
l surface / after Price 11993and 19871
06 ~ v ~ Dec~u~ ~e~ m ~
1981 1982 198t 1982

.. S . SUMMER

wooded area ............ r.~'G'~ . . . . . . . . o p a n m a r s h . . . . . .

et.
uJ
F- 0
LU " ~'''''"" '''-o ..-----d..- oeat_ , '

1
Z

,.°.~..=°o°. ~.....-,,-,4~ ° . .......... . ....... oo°°°

°.. ........... °-
clay
u.I
r~ 3

WINTER
/
I'-~'=%~7:~TE~L'-~7 " ............... ~. ~--~ . . . . . ........... : ~ : ": -= = -: ~: "::-; ~:~t
H
W
1
z_
4 o.g j~.b . ~ ... _

I.U
3 ............... "T. . . . . . . . . . . . . . . . . T- . . . . . . . . . . . . . . . . . 7 ................... ~..........................T " " n ................. r
0 100 200 300 400 500 600
METERS

Fit~, I~, Frt~s! ~ I 1 , ; I C,iiC~I1 a ~ d t, ul~,~ul'fa~ce f l o w p=iUcr=~s m =l w e t l a n d in c e n t r a l S , ' t s k a I c h c w a n ( A ) a n d a

'I'~|E IIYI)ROIoOGY OF NOR'I'IIERN WETLANDS IN NORTH AMERI('A 25
thawing of t]~e seasonal frost in a wetland led to a reversal of ground-water-
flow direction until the frost dissipated in summer, at which time the normal
flow gradient was re-established (Woo and Heron, t987a).
During spring ~n the Subarctic, seasonal frost reduces infiltration of water
from snowmelt and early spring rains; therefore, most of it runs over the land
surface or through the very shallow part of the subsurface much as in the
Arctic. After seasonal frost melts, surface water and soil water can inte,:act
with ground water whe~'e permafrost is absent. These areas of no permafrost
are referred to as 'windows' in the permafrost-confining bed. Understanding
of ground-water-flow sys~ms associated with surface water or wetlands is
limited because of the difficulty and expense of developing instrumented well
fields, particularly in areas t~at have discontinuous confining beds. Thus, a
numerical model was used tt~ evaluate a hypothetical surface-water body
underlain by a discontinuous confining bed, The following analysis was
developed for a situation where the confining bed lies deeper than the usual
depth to permafrost, but the results allow a discussion of flow-system concepts
and could have direct application where permafrost is deep (Winter, unpub-
lished data, 1989).
The model setting consists of surface water underlain successively by
low-permeability sediments of limited extent, a relatively permeable porous
medium (K~) across the entire basin, a dis¢:ontinuous confining bed (K2), and
an aquifer across the entire basin having the highest permeability (Ka) (Fig.
14). The position of the window in the co,~fining bed, with respect to the
surface water, has a more significant effect on the flow field associated with the
surface water than the size of the window. F~r" example, assuming a !.7 m
downward hydraulic gradient between the surface water and the aquifer, and
a window slightly smaller than the surface water lying directly beneath the
surface water, the flow field indicates seepage into the surface water along its
edges, but outward seepage through a large area of the central part of the
surface-water bed (Fig. 14(A)).
If the window in the confining bed is smaller, for example, only 8% of the
total length of the ground-water basin compared with the 20% in the previous
setting, the flow field is only slightly different to that in ta~.eprevious setting
(Fig. 14(B)). The area of seepage along the edge of the surface water is larger
and the area of seepage out of the surface-water bed in the ~'enter is smaller
than that shown in Fig. 14(A). In contrast, if the window in the confining bed
does not underlie the surface water, as in Fig. 14(C), the surface water is not
affected by the downward gradient from the surface water to the aquifer
- - there is no seepage from the surface water at all.
Based on these few numerical simulations, implications with ~-espect to
Subarctic wetlands include: (1) wetland water has the greatest probebility of
26 M .to. W(IO A N D I ( WlNIt.I~.

A .....
(.~ EXPLANATION
- :'35- -- L!NE OF EOL!AL HYDRAULIC
HEAD, ;NI'ERVAL IS VARIABLE
"230 ,~l.... S T A G N A r f O N POINT

OIRECI'ION OF GROUND-WATER
, , ,. . . . . .. 4 FLOW
LAKE SEDIMENTS
I I ~ / \ I '
[ ~ UNIT K 1
K=I
CONFINING BED, K 2
K=lxlO 3

B ,:~d '.:.:,2 AO'.!!~R K~

K~ l x ? 0 3
&lo /C "~" "]
. -,~F ~ 0".~.:----:~o----~.-," ' ~ ~1 K= hvdlauhc corldtlctiwty 3
Anisotropy of all units is 10

[q','_,~J,~Z,7,~,t.L 5.~ ,t:,..~_ _L.', ??.~'.~C,',LL',~,

c . . . . ~

r.,~ ~ ~ ~, ",~ / ~ ~.~

' I t ~ " -

&1.i.'~'.' " ' ' L'm~.,L._'2"~'~L,Z~L2~5 ..........

F i g . ~4. N t l m e r i ~ a l simul~| lhms o f |lypolltcticM scllili~ ~]lowinl~, g. mttnd-waf er ltow' a a s o c i a l c d w i l h a lake

•4epara|¢d l]lt++l tl tle~tp +ltlllif,21 + by a di~l~t+lll+IlllOtl~ ~A+ltlltti|l[ t~ed

inter;~.clh~/~, wi~l'i regional ground-water-flow systems if a window in the

permafrost underlies the wetland; (2) the size of the window in the permafrost
does not significantly affect the general patterns of seepage in and out of the
wetland; (3) these simulations are of a setting that has only a downward
hydraulic gradient from the welland to tile hydrologic uni! underlying the
permafrost. If an upwal'd gradienl exisfs there is likely to be signilican, flow
fro|~i the deeper ground water to the welland.
In the north Temperate zone, the presence of seasonal frost results ha early
spring runoff that behaves similarly to early spring runoff in the Arctic and
Subarctic. That is, snowmelt and early spring rains run off the uplands to
| t l i ' ItYI)R()I ~.)(iY O | ' NI)P,| III'RN \x,'t?'I I AN|)S IN NOR'Ill AN{ERI('A 27

A April 1 T8
METERS Y5 ~,~
585 ~ 2't
T1 PI 73 Ta c~ 5 ~ ~ -
575 ~
15

835-1 Dlre~,on ef ground waler ftow

B May 19
M~T r qS

C June 30
METERS

f~ATUM ~S ARBITRARY 0 150 ~¢]ETERS VERTICAL EXAGGERATION x 5

Fig. 15. Altitudes of wedand walcr levels and water-tai~le levels for IAJ 1 April, (B) 19 May, and (C) 30
June 1980 in the Cottonwood Lake area, North Dakota. Wetlands T8 and T5 are g~ound-water recharge
areas. Beneath tile hills hetween these twn wetlands, water-table highs d~ not underlie land-surface highs.
Reversals ill llt~w between wetlands and ground water occtlr from early to late spring in the are~ between
well l0 and wetland PI. A water-lable high comrnoaly underlies the land~u,-face high near well 16. (Data
from Lal~augh el aL, 1987.)

depressions commonly occupied by streams, lakes, and wetlands. However, in

the case of seasonal fi'ost, once the frost disappears, water in these land-
surface depressions can interact fi'eely with the entire underlying ground-
water system. Tim importance of this wocess is that much of the recharge to,
and discharge from, the ground-water system is focused in depressions during
spring (Lissey, 1971). The Cottonwood Lake area, North Dakota, p,rovides an
example of the types of ground-water-flow systems that can exist, in glacial
terrain because of focused recharge and discharge (Fig. !5). This site ~,,.~dicates
the importance of small depressions H~at hold ~pring runoff water for only a
2~ M K WOO A N t } f { ' WtNIER

R,'w weeks or months to ground.water recharge. Although seasonal fl'ost has

an important role in concentrating water in depressions, it is not the only
mechanism for causing concentration of water. Many geologic heterogeneities
in the subsurlhce also can eoncentrale water ill depressions, such as low-
permeability layers in soil horizons and cross-bedding in dunal deposits.

DISCUSSION

Field observations and existing literature show that fi'ost is a common

phenonlenon in wetlands of the north Temperate, Subarctic, and Arctic zones.
Even though the duration and thickness of the frost layer increases with
latitude, some of the hydrological responses are similar for all three climatic
zones. For example, retardation of infiltration enhances overland flow in
spring. Other hydrologm processes, such as evaporation, and ground-water-
flow patterns may be considerably different in the three zones. The extent to
which wtrious hydrological proce,,~ses are affected may be considered as a
continuum rather than as discrete changes in the wetland hydrological charac-
teristics in the different zones.
The occm°rence of permafrost and seasonal frost in wetland and non-
wetland areas has potentially similar effects on snowmelt infiltration and
runoff. Kane and Stein (1983), Granger et al. (1984), and Woo and Marsh
0990) described varying quantities of infiltration into frozen soils in non-
wetland areas, depending on the characteristics of soil material and the
amount of ice in the pores. When the yield of snow melt water exceeds
infiltration, overland flow begins on the frost-laden slopes. Roulet and Woo
0988) described overland flow from non-wetland slopes draining towards the
wetlands in the valley bottoms in a continuous permafrost environment. Woo
et al. (1996) rioted overland flow from frozen slopes toward a prairie slough.
Although the processes ate similar, the magnitude and duration of overland
flow in wethmd area.,: are likely to be greater than in uplands because the
higher degree of saturation ill the thtt could produce more ice to f}ll the soil
pores and retard infiitratkm, and the gentler gradients cause slower drainage
of surface runolL

REFERENCES

Addison. P.A., 1077~Studieson evapo|ra|l~pilalio~and energy budge:Ison Troeh~vc I a~whmtl.

i¢~: l...C~ Bt~:~ (Editor). Tru~iovc Lowland, I)~:von Island. Canada: A High Arctic
E~:o~y,~em~Uni~'er~ilyof Alberta, Edmonton. pp. 281-300.
Ad |{oc Study (koup on tee Segrcgaiion and From ~-{~avc,i984. ice Segregation and Frost
Heaving U,S, National Research Council°National Academy. Washington. DC, 72 pp.
Bay. RR. t969. Runoff from smMlpeafland watersheds J. Hydrol., 9: 90--102,
IUl~ IIY|)ROI,O(.~Y GI: NOR |HERN WI ILANDS IN NOR'HI AMERICA 29

Brown. J., Ding~.~m, S.L and Lewellen, R.I,, 1968. Hydrology of a drainage basin on the
Alaska coastal p~ain. U.S. Army CRREL Research Report 240, t8 pp, Hanover, N.H.
Brown, J.M., 1976. Peat ~emperature regime of a Minnesota bog and the effect of canopy
removal. J. Appl. Ecol., 13: 189-194.
Brown, R.J.E., 1977. M,askeg and permafrost. In: N.W. Radforth and C.O. Brawnier (Editors),
Muskeg and tile Nortb,ern Environment in Canada. University of Toronto, Toronto, pp.
148-i63.
Brown, R.J.E. and Williams, G.P., i972. The freezing of peatiand. National Resource Cour, cil
of Canada, Division of Bu~,~ding Research Technical Paper 381, NRC12881, 29 pp.
Burt. T.P. and Williams, P.J.. 1976. Hydraulic conductivity in frozen soils. Earth Surf.
Processes, 1: 349-360.
Cowell, D~W., Jeglum, J.K. and Merriman, J.C., 1978. Preservation of seasonal frost in
peatlands, Kinoje Lakes, southern Hudson Bay Lo~ia~ld. ln: Proc, 3rd Int. Conf. on
Permafrost. Edmonton, Alberta. National Research Council of Canada, Ottawa, Ont., pp.
454-459.
Dewey, K.F. and Helm, Jr., R., 1982. A digital archive of northern hemisphere snow cover,
November 1966 through December 19'80. Bull. Am, Meteoro]~ Sac, 63: 1132-114t.
Dredge, L.A., 1979. Thaw depths and pez,qaafrost in polygonal pea! terrain, Hudson Bay
Lowland, Manitoba. Current Research, P~rt C, Geologic~d Survey of Canada, Paper 79-ic,
pp. 27-30.
Farouki, O.T., 1981, Thermal properties of soil~-. U.S. Army CRREL Monogr. 81-1, 136 pp.
FitzGibbon, J.E., 1981. Thawing of seasonally ?cozen ground in organic terrain in central
Saskatchewan. Can. J. Earth Sci., 18: 1492-1496.
Granger, R.J., Gray~ D.M, and Dyek, G.E., 1984. ~nowmelt infiltration into frozen prairie
soils. Can. J. Earth Sci., 21: 669-677.
Halliwe!!, D.H. and Rouse, W.R., 1987. Soil heat flux in permafrost: characteristics and
accuracy of measurements. J. Climatol., 7: 571-584.
Hare, F.K., 1980. Long-term annual surface heat and water balances over Canada and the
United States south of 60°N; Reconciliation of precipitati,~m, runoff, and temperature fields.
Atmos. Ocean, 18: 127-153.
Juusela, T., 1967. Some results of field observations on the frc~t phenomenon on peat soil. J.
Hydro[, 5: 269-278.
Kane, D.L. and Bredlha uer, S.R., 1981. Subarctic snowmclt runoff genera tion. In: Proc. of the
Specialty Conf. on the Northern Community. ASCE, New York, pp. 591-601.
Kane, D.L. and Slaughter, C.W., 1972. Seasonai regime and hydrological significance of stream
icings in central Alaska. In: The Role of Snow and Ice in Hydrol,~gy, Proc., Banff Symp.,
!AHS Publ. No. 107, pp. 528-540.
Kane, D.L. and Stein, J , 1983. Water movement into seasonally frozen: soils. Water Resour.
Res,, 19: t547-1557.
Kane, D,L., Lathin, J.N. and Taylor, G.S., 1975, Heat and mass transfer i~ cold region soils.
Institute of Water Resources Research Report IWR-65, University of Alaska, Fairbanks.
Kingsbury, C.M. and Moore, T R., 1987. The freeze-thaw cycle of a subarc'Sc, fen, northern
Quebec, Canada. Arct. Alpine Res., 19: 289-295.
LaBaugh, J.W., Winter, T.C. and Swanson, G.A., 1987. Geohydrology and chemistry of
selected prairie wetlands in the Cottonwood Lake area, Stutsman County, N~,rth Dakota,
!979-!982 U.S Geo!. Sur~., Prof. Pap. !,13!. 26 pp.
LafleuL P.M, and R o u t , W.R., 1988. The influence of surface cover and climate ~ energy
partitioning and ew~por~llion in a subarctic wciland, Boundary-Layer Meteorol., ~:,4:327-
347.
30 ~.]-K w ~ ) A N D r u WI ~ l | r

1,issey. A,, 1971. Depression-foeu cd Transient Groundwater Fiow Patterns in Maniloba.

Geological Associaiion of Canada, Ottawa, Ore., pp. 333-341.
Moc,~'e, T.R., 1987. Thermal regime of peatlands ill subarctic caslcrn Canada. Can. J. Earth
Sei,, 24: 1352-d359.
Mtu!~ro, D.S., 1979. Daytit :e energy exchange and evaporation fi'om a wooded swamp. Water
Rcs~l,,r. Res., 1 5 : t 2 5 9 !265.
Outcalt, ,kk and Nelson, I-., 1984. Computer simulation of buoyancy and snow-cover effects
in palsa dynamics. Arct. Alpine Res., 16: 259-263.
Price, LS., f983. The effect of hydrology on ground freezing in a watershed with organic terrain.
In: Prn~. Fourth Int. Conf. on Permafi'ost, Fairbanks, Alaska, National Academy Press,
Washil~,g~on D.C., pp. 1009-1014.
Price, J.S., 1987. The inlluence of wetland and mineral terrain types on snowmelt runoff in the
sabarclic. Can. Water Rcsour. J., 12: 43-52.
Price, J.S. ,'~nd FilzGibbon, J.E., 1987. Groundwater storage-streamltow rclalions during
wi.~er to a subarctic wetland, Saskatchewan. Can. J. Earth Sci., 24: 2074-2081.
Price, J.S. and Woo, M.K., 1988. Sludies ofa subarc~'(c coastal marsh, I. Hydrology. J. Hydroi..
103: 275-292.
Roulct, N.T, and Woo, M,K., 1986, Wetland and bt-,~ e,,'.q~oration in tlw low Arctic. Arct.
Alpine Res., 18: 73-,91.
Roltlel, N,T, and Woo, M,K., 1988. llydrology t~f a wetland in tlae continuous pcrmal?ost
region. J, Ilydrol,, 89: 73-91.
Rouse, W.R., 1983. Active layer energy exchange in wet and dry tundra of the Hudson Bay
Iowhmds. In: Proc. of the Fourth Int. Conf. on Permafrost. Fairbanks, Alaska, Nalional
Academy Press, Washington, D.C,, pp. 1089-1094.
Rouse, W.R., Hardilk S.G. lind Lafleur, P., 1987. The energy balance in the coastal environ-
ment of James Bay alld Hudson Bay during the growing season. J, Climatol., 7: 165-179.
Rydell, B.E,, 1977. Hydrology of Truelove Lowland. In: L.C. Bliss (Editor), Truelove Lowland,
Devon Island, Canada: A High Arctic Ecosystem. klniversity of Alberta. Edmonton, pp.
107-136.
Slaughter, C.W. and Kane, S,I,, i979. Hydrologic role of shallow organic soils in cold climate.
In: Pr:~c. Canadian Hydrrqogy Syrup. '79: Cold Climate Hydrology. Vancouvcr, B.C.,
National Research Council of Canada, Ottawa, Onl., pp. 380-389.
Smilh, SL, lind Woo. M.K.. 1986. Ground and water temperature in a middatitudc swamp.
Can. Water Re,';our~ J,, I: 76-88,
Swanson. L.E. and Rothwell, R.IL, 1986, Thawing of ground frosl oll "t drained and undrained
boreal w~;,tl|tnd,~~]le~ hi: K,L. K,~ute (lktitor), Cold Regions Hydrology Syrup, American
Water Resources Assnciatio,~, Minneapolis, MN, pp. 231-236,
Thorn, f~,G., 1972. The role of sprin~ thaw in string bog genesis. Arctic, 25: 236-239.
UNESCO, 1978. World Water Balance and Water Resources of lhe Earth, !nc!udh~g AHas of
World Water Balance. UNESCO, Paris, 663 pp.
Verry, E.S. lind Boelter, D,H., 1978. Peatland hydrology. In: Wetland Frame,ions and V~dues:
the S1ate of Our Understanding~ Proc. Nation.lll Syrup. on Wellamb,, 7 I11 Novemhc~' 1978.
Lake Bilenlt Vista, F t . Anterical| Water Re'¢OUl'Ce~,As~{ociallo~,LMh/neap~lis, MN, pp.
1t89~402,
Wilit~c~v,~Ry,N.J~ 196(i, Enviror~men~ of the Cape Thompson Regi(~n_ Ala~k~L 1_'S. Atomic
E~ergy Commission, 1250 pp,
Winte,'. T,C. and Woo, M.-K,, 1990. Hydrology of lakes and wetlands, in: M,G. Wolman and
t.t,C. R i g ~ (Editors), Surface Water Hydtotogy, The Geology of North America, Vol,0-1.
Ge~!ogical g~)ciely of America. Boulder, CO, pp. i 59 ~187,
I'IIE ItYI)ROI.O(~'Ir O~: NORT!t[~RN WE't LANDS IN NORTH AMERICA 31

Winter, T.C., IV~,aclay, R.W. and Pike, G.M., 1967. Water resources of 'he Roseau River
watershed, no~thwestern Minnesota. U,S. Geoiogicai Survey Hydrologic investigations
Atlas HA-241, t~ree sheets.
Woo, M.K., 1988. V~L~tland runoff regime in northern Canada. In: Proc. 5th Int. Conf. on
Permafrost, Trondheim, Norway, Trondheim Tapir, Trondheim, Vol. I. pp, 644-648.
Woo, M.K. and diCenz~, P.D., 1989. Hydrology of small tributary streams in a subarctic
wetland. Can. 3. Earth Sci., 26: 1557-1566.
Woo, M K . and Heron, R., t987a. Effects of forests on wetland runoffduring spring. In: Forest
Hydrology and Watershed Management, Proc. Vancouver Syrup. IAHS Publication No.
167, pp. 297-307,
Woo, M.K. and Heron, R., 1987b. Breakup of small rivers in the subarctic. Can. J. Earth Sci.,
24: 784-795.
Woo, M.K. and Marsh, P., 1990. l~sponse of soil moisture change to hydrological processes
in a continuous permafrost envirenment. Nord. Hydrol., 21: 235--252.
Woo, M.K. and Valverde, J., 1982. Ground and water temperatures of a forested mid-!at!rude
swamp. In: Proc. Hydrology Sym,~,. "82: Hydrological Processes of Forested Areas.
Frederiction, N.B., National Researck Council of Canada, Ottawa, Ont, pp. 301-3i2.
Woo M,K., Rowseil, R.D, and Clark, R.G. 1990. Climate impacts on the hb.i!lrology of prairie
wetlands. In: G. Wall (Editor), Proc. Syrup. on the Impacts of Clir:tate Change and
Variability on the Great Plains. University of Waterloo, Department of G,;ography Publica-
tion Series, Occasional Paper No. 12, pp. 23'9-266.