The Hydrological Functions
of a Boreal Wetland
Christopher Spence, National Hydrology Research Centre, Saskatoon
Xiu Juan Guan, University of Saskatchewan, Saskatoon
Ross Phillips, University of Saskatchewan, Saskatoon
Background
• The catchments that drain boreal stream
•
•
•
networks exemplify heterogeneous
conditions.
Within this heterogeneity, wetlands are of
critical importance to the catchment
hydrology, because they are often situated at
the outlet of headwater basins.
While research has illuminated the runoff
generation processes in these wetlands, an
investigation seeking to understand the
dynamic of hydrological function is absent.
The questions addressed in this study were:
–
–
–
From where is the majority of water in a
wetland collected, specifically one at the
bottom of a headwater catchment?
Where and how does a wetland in such a
landscape position tend to store the water it
collects?
When discharging water, is a wetland in such
a landscape position predominantly
transmitting or contributing?
DRAFT – Page 2 – August 16, 2010
Study site and methods
• The wetland under study was a 3.3 ha fen at
•
the bottom of a 9.4 km2 catchment draining
Moss Creek, a tributary of Baker Creek.
The water budget was estimated for the period
from April 14 – July 16, 2008 following:
∆S=P + M + Qi – Qo - ET
• Piezometer nests were installed at locations to
•
Lake
690
k
ee
Cr
Climate tower
Piezometers and wells
Ablation line
Soil moisture, temperature
Streamflow
s
os
M
•
•
monitor intra-site groundwater hydraulic
gradients.
Precipitation, groundwater and surface waters
were sampled for chemical analysis, including
major ions, specific conductance, and the
stable isotopes of 18O and 2H.
If ∆S>Qo, storing; ∆S<Qo, discharging
If Qi>Qw, transmitting; Qi<Qw, contributing
W9
W7
N
DRAFT – Page 3 – August 16, 2010
50
0
m
100
Vital
Lake
P, PMand
andMET
(mm)
(mm)
Water budget
40
30
PP (mm)
M
M(mm)
20
ET
E (mm)
10
0
400
I-690(mm)
Lake 690 inflow
Q-surface
Outflow (mm)
internal
Internalrunoff
runoff(mm)
Q (mm)
300
200
100
0
400
Σ∆ S (m m )
300
200
100
0
14-Apr
28-Apr
12-May
26-May
9-Jun
DRAFT – Page
4 – August 16,
2010
23-Jun
7-Jul
Intra-wetland groundwater flux
DRAFT – Page 5 – August 16, 2010
Stable isotopes
DRAFT – Page 6 – August 16, 2010
Stable isotopes
-12
-14
-
δ18HCO
O (‰)
3
-16
-18
-20
-22
Lake 690 outflow
wetland outflow
W9
W7
bedrock runoff
-24
-26
14-Apr
28-Apr
12-MayDRAFT –26-May
Page 7 – August9-Jun
16, 2010
23-Jun
7-Jul
21-Jul
Ion chemistry
Lake 690 outflow
wetland outflow
W9
W7
bedrock runoff
EC (µS/cm)
Ca
SO4-2
14
12
10
8
6
4
2
0
35
30
25
20
15
10
5
0
300
250
200
150
100
50
0
14-Apr
28-Apr
12-May
26-May
9-Jun
DRAFT – Page 8 – August 16, 2010
23-Jun
7-Jul
21-Jul
Streamflow separation
• The ion chemistry of rain, W7 groundwater and outflow permit the
assumption that Qo after June 14 was only a product of water stored
in the wetland such that:
18
18
f gl + f gw = 1
δ Oo − δ Ogw
f gl = 18
δ Ogl − δ 18Ogw
• Where f is the fraction of water in Qo, and the subscripts gl, gw and
•
o, refer to isotopic signatures from groundwater from piezometer
W7, ground from piezometer W9 and Qo water.
The value of fgl averaged 91% of Qo after June 14 implying Qo is
composed of water that is only transient in the wetland.
DRAFT – Page 9 – August 16, 2010
Collection
• The wetland collects the
•
•
majority of its water from
the upstream watershed;
not the immediately
adjacent hillslopes.
Synchronicity may be
important.
Rain and bedrock runoff
from adjacent hillslopes
become important after
inputs from the upper
watershed decrease.
Previous hydrological
process studies suggest it
is more likely to be the
latter.
DRAFT – Page 10 – August 16, 2010
Storage
• The wetland tends to store the
•
•
water it collects in two separate
zones; one near the bisecting
stream and another on the
fringes of the wetland.
Piezometric gradients and
hydrochemistry suggest there is
little exchange between the
two.
Water stored in the central zone
appears to be only transient in
the wetland.
DRAFT – Page 11 – August 16, 2010
Discharge
• The wetland is predominantly a transmitter of water.
• It is predominantly a contributor only after upstream sources are
reduced.
DRAFT – Page 12 – August 16, 2010
Two phase functioning
DRAFT – Page 13 – August 16, 2010
Conclusions
• This wetland is primarily a transmitter of water and what water it did
•
•
supply was altered little by intra-wetland geochemical processes.
That only a fraction of the wetland contributed water to the outlet
would suggest that it is very important for source areas to be
properly simulated for coupled hydrology-biogeochemical models to
be successful.
Landscape position influenced the hydrological function of this
wetland and should therefore be considered in hydrological model
parameterization.
DRAFT – Page 14 – August 16, 2010
Acknowledgements
• Newell Hedstrom, Erin Shaw,
•
•
Dave and Rhonda Phillips, and
Dave Fox provided assistance
in the field.
All the staff of the Yellowknife
office of the Water Survey of
Canada.
Our funding agencies.
DRAFT – Page 15 – August 16, 2010