Parameterizations of organic-covered permafrost soils in land surface and hydrological models
Y. Zhang1 , S.K. Carey1, W.L. Quinton2 and J. R. Janowicz3
of Geography and Environmental Studies, Carleton University, Ottawa, Ontario, Canada
2Cold
60
0
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
-1.2
-1.4
40
60
(d)
40
60
0
20
40
60
(e)
Wolf Creek
North Facing Slope
20
40
60
20
40
60
Days after observed
thawing-start
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
-1.2
-1.4
(f)
Wolf Creek
North Facing Slope
0
50
100
150
Wolf Creek South
Facing Slope
Wolf Creek
South
Facing
Slope
(g)
0
0.0
-0.1
-0.2
-0.3
-0.4
Granger
Creek
20
0
Five thawing/freezing algorithms
Five infiltration algorithms
Three soil thermal conductivity parameterisations
Three soil hydraulic property parameterisations
Three unfrozen water parameterisations
Three parameterisations of ice impedance to hydraulic conductivity
20
(c)
Granger
Creek
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
-1.2
-1.4
Wolf Creek Alpine Tundra
60°34’N, 135°09’W, 1615 m
Organic layer Depth: ~0.03 m
Permafrost Table: >0.2m
0
(h)
0
50
100
150
200
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
-1.2
-1.4
200
Days after observed
freezing-start
Figure 8: Observed ground
thawing and freezing depths
and simulations with three soil
thermal conductivity
parameterizations.
Granger Creek North Facing Slope
60°33’N, 135°11’W, 1338 m
Organic layer Depth: ~0.35 m
Permafrost Table: >0.4 m
61°31’N, 135°31’W, 1175 m
Organic layer Depth: 0.23 m
Permafrost Table: N.A.
Ice impedance
factors
Simulated with Run1
Simulated with Run2
Figure 3: Observed
ground thawing and
freezing depths at
Wolf Creek north
facing slope and
simulations by five
algorithms and three
sets of model run
20
16
12
8
4
0
-4
-8
0.0
(a)
Farouki, 1986; Flerchinger and Saxton.
1989
Clapp and Hornberger equations
CH-PARA
Clapp and Hornberger, 1978
Brooks and Corey equations
BC-PARA
Brooks and Corey, 1964
VG-PARA
van Genuchten, 1980
A power function
UFW-PF
Anderson and Tice, 1972
A segmented linear function
UFW-SL
Goodrich, 1978
A water potential-freezing point depression function
UFW-WP
Cary and Mayland, 1972; Zhao and
Gray, 1997
Exponential ice content function
EP-Ice
Zhao and Gray, 1997
Squared ice content function
SQ-Ice
Soulis and Seglenieks, 2008
Linear ice content functions
LN-Ice
Bloomsburg and Wang, 1969;
Flerchinger and Saxton, 1989
Tair
Ts at 0cm
ET
24
20
16
12
8
4
0
-4
-8
-12
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
(b)
200
(c)
(a)
(c)
HMSA
(e)
FD-DECP
Run1
Upper boundary and soil moisture were from observation; Bottom
boundary was prescribed
Run2
Upper boundary were from observation; Bottom boundary and soil
moisture were prescribed
Run3
50 100 150 200 250 300 3500
Observation
Snowmelt
Rainfall
30
50 100 150 200 250 300 3500
0.8
20
15
0.6
0.4
5
(e) 10 cm
0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
0.2
0.0
0.8
0.6
0.4
(f) 20 cm
0.2
200
0.0
0.8
150
150
0.6
100
100
0.4
50
(g) 30 cm
50
0
0
(d)
200
0.0
0.8
200
150
150
100
100
50
50
0.6
0.4
(h) 40 cm
20 40 60 80 100 120 140
0.2
80
(b)
Simulated with Run1
Observed thawing depth
Simulated with Run3
Simulated with Run2
Observed freezing depth
0.0
12
-0.2
(b)
T at soil surface
8
-0.4
-0.6
4
ATIA
-0.8
0
-1.0
-4
T at 0.7m
-1.2
-8 (a)
-1.4
soil depth
0.0
0.0
-0.2
(d) -0.2
(c)
-0.4
-0.4
HMSA
TDSA
-0.6
-0.6
-0.8
-0.8
-1.0
-1.0
-1.2
-1.2
-1.4
-1.4
0.0
0.0
-0.2
-0.2
(f)
(e)
-0.4
-0.4
FD-DECP
-0.6
-0.6
FD-AHCP
-0.8
-0.8
-1.0
-1.0
-1.2
-1.2
-1.4
-1.4
0
0.0
0
20 40 60 80 100 120 140
50 100 150 200 250 300 3500
Observation
Snowmelt
Rainfall
16
14
12
10
8
6
4
2
0
80
(d)
(e) 5 cm
0.0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
0.4
40
40
0.2
20
20
0.1
0
80
0
80
0.0
0.3
(f) 15 cm
0.4
60
60
40
40
0.2
20
20
0.1
0
0
0.0
20 40 60 80 100 120 140
0.3
0
20
40
60
80 100 120 140
Days from Mar. 31 to Aug. 31, 1999 at Wolf Creek alpine site
Days from Mar. 25 to Aug. 31, 2005 at Scotty Creek Peat Plateau
Inputs sources
Figure 5: Observed ground thawing,
infiltration, runoff and soil water content
at Scotty Creek and simulations by three
infiltration algorithms
50 100 150 200 250 300 350
Tair
Ts at 5cm
ET
GA-SHAW
ML-CLASS
IT-TOPO
60
(c)
Figure 4: Observed
ground thawing and
freezing depths at
Wolf Creek south
facing slope and
simulations by five
algorithms and three
sets of model run
Days after September 20, 1996
60
0
0
0
0.2
(a)
Thawing / freezing depth (m)
Thawing / freezing depth (m)
Thawing / freezing depth (m) Temperature(oC)
Observed thawing depth
Observed freezing depth
0.0
-0.2
(b)
-0.4
-0.6
-0.8
-1.0
-1.2
-1.4
0.0
(d) -0.2
-0.4
TDSA
-0.6
-0.8
-1.0
-1.2
-1.4
0.0
-0.2
(f)
-0.4
-0.6
-0.8
FD-AHCP
-1.0
-1.2
-1.4
Tair
Ts at 0cm
ET
24
20
16
12
8
4
0
-4
-8
-12
50 100 150 200 250 300 350
Days after August 25, 2002
50 100 150 200 250 300 350
GA-SHAW
ML-CLASS
IT-TOPO
25
10
0
All inputs and boundary forcing were from observation
-0.4
0
ATIA
-0.2
-0.4
-0.6
-0.8
-1.0
-1.2
-1.4
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
-1.2
-1.4
-0.1
-0.2
-0.3
-0.4
Days after April 20, 1997
surface water(mm/d)
De Vries
Simulated with Run3
T at soil surface
T at 0.7m soil depth
FD-AHCP
FD-DECP
-0.3
Days after September 1, 2004
0
De Vries’s formulations
-0.4
0.0
(f)
(e)
-0.1
-0.2
24
20
16
12
8
4
0
-4
-8
-12
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
-0.9
-1.0
80
Observation
Snowmelt
Rainfall
10
(a)
8
6
4
0.3
0.4
0.3
0.2
0.2
0.1
0.1
-5
-4
-3
-2
-1
0 -5
-4
-3
-2
-1
0
0.4
0.3
0.2
0.4
(e) 0-30cm soil depth
at Wolf Creek
North Facing Slope
(f) 30-60cm soil depth
at Wolf Creek
North Facing Slope
0.3
0.2
0.1
0.1
0.0
-2.0
-1.5
-1.0
-0.5
0.0-2.0
-1.5
-1.0
-0.5
0.4
0.3
0.2
(g) 0-40 cm soil depth
at Wolf Creek
South Facing Slope
0.0
0.0
0.4
(c) 40-60 cm soil depth
at Wolf Creek
South Facing Slope
-2.0
-1.5
-1.0
-0.5
0.0-2.0
-1.5
-1.0
Soil temperature (oC)
-0.5
0.3
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0-8 -7 -6 -5 -4 -3 -2 -1 0
(c) 5cm soil depth
0.2 at Wolf Creek forest site
(d) 30 cm soil depth
at Wolf Creek forest site
0.4
0.3
0.2
0.1
0.1
0.0
0.4
0.0
-18 -15 -12 -9
-6
-3
(e) 5 cm soil depth
at Wolf Creek alpine site
0.3
0-12 -10 -8
-6
-4
-2
0
(f) 15 cm soil depth
at Wolf Creek alpine site
0.4
0.3
0.2
10
-6
-4
-2
0-12 -10 -8
-6
-4
-2
(b)
Granger Creek
Upper organic layer
1
0.1
Granger Creek
Upper organic layer
0.01
0.001
100
(c)
10
(d)
1
0.1
0.01
Scotty Creek
Upper organic layer
0.2
0.4
0.6
0.8
0.0
0.2
0.4
0.6
0.8
10-3
10-4
10-5
10-6
10-7
10-8
10-9
10-10
10-11
10-12
10-13
10-14
0.001
Scotty Creek
Lower organic layer
0.001
0.0
-8
10-2
(a)
0.0
-10
BC-Para
CH-Para
Observation
VG-Para
100
0.1
0.0
0.0
0.0
Figure 9: Observed unfrozen
water content and simulations
using three unfrozen water
parameterizations.
0.1
0.0
0.1
0.2
0.1
0.0
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.2
0.3
0.1
BC-Para
CH-Para
(b) 30cm soil depth
at Scotty Creek
0.2
1.0
Observation
VG-Para
BC-Para
CH-Para
1e-5
1e-6
1e-7
1e-8
1e-9
1e-10
1e-11
1e-12
1e-13
1e-14
1e-15
1e-16
0.01
0.1
1
10
100
water pressure head (m)
Volumetric water content (m3 m-3)
1e-4
Water potential
effect only
With LN-Ice
With SQ-Ice
With EP-Ice
K-CLASS
1e-17
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
ice fraction (qi/qs)
0
Soil temperature (oC)
Figure 10: Observed unfrozen water
content and simulations using water
potential-soil temperature
relationship coupled with three water
retention parameterizations
Figure 11: Soil water
retention data measured at
several organic soils and best
fitting curves with three
common parameterisations
Figure 12: Observed average
unsaturated hydraulic conductivity
of 16 soil samples from Wolf Creek
and fitting curves with three
common parameterisations
Figure 13: Reductions to
hydraulic conductivity by several
ice impedance
parameterisations
2
(b)
(c)
0.6
0.2
0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
-0.9
-1.0
80
0.0
(f) 15 cm
0.4
0.3
0.2
0.1
0.0
(g) 30 cm
0.4
60
40
40
0.2
20
20
0.1
0
80
0
80
0.0
0.3
(h) 80 cm
0.4
60
60
40
40
0.2
20
20
0.1
0
0.3
0
0
20 40 60 80 100 120 140
0.0
0
With carefully chosen parameters, all three evaluated soil hydraulic parameterisations could achieve similar soil
water retention curves and conductivity curves. However, van Genuchten’s method gives smooth and
continuous curves over all soil moisture ranges, while Brooks-Corey and Clapp-Hornberger’s methods have to
be capped by maximum and minimum values on saturate, extremely dry or frozen soil conditions.
The lowered water potential imposed by soil freezing alone could reduce the frozen soil hydraulic conductivity to
the order of magnitude of observation, the various employed ice impedance factors may not be necessary for
frozen organic soils.
REFERENCES
Details of all the references listed in the lower left table could be found in the following two publications:
Zhang Y., S.K. Carey, W. L. Quinton, 2008. Evaluation of the algorithms and parameterizations for ground thawing and freezing simulation in permafrost regions.
Journal of Geophysical Research, 113, D17116, doi:10.1029/2007JD009343: 1761-1775.
Zhang Y., S.K. Carey, W. L. Quinton, J. R. Janowicz and G. N. Flerchinger, 2009. Simulation of infiltration into organic-covered permafrost soils (to be submitted)
Acknowledgment
20 40 60 80 100 120 140
Days from Mar. 31 to Aug. 31, 1999 at Wolf Creek forest site
Figure 6: Observed ground thawing,
infiltration, runoff and soil water content
at Wolf Creek alpine site and simulations
by three infiltration algorithms
The segmented linear function (UFW-SL) for unfrozen water content is the easiest to be parameterised with
minimum observation data, while water potential–freezing point depression relation (UFW-WP) is the best choice
for coupled numerical simulation of soil thermal and moisture transfers.
0.4
60
(d)
De Vries’ parameterisation is recommended as the best method to parameterize the thermal conductivity in
permafrost soils.
Paired observation values of unfrozen soil water content and soil temperature could be an effective data set to
derive soil hydraulic parameters of permafrost soils.
GA-SHAW
ML-CLASS
IT-TOPO
(e) 5 cm
Soil liquid water content (m3 m-3)
Verseghy 1991; Dai et al., 2003; Woo et
al., 2004; Yi et al., 2006
-0.4
0.0
Thawing / freezing depth (m)
Common-Johansen
-0.6
-0.8
50 100 150 200 250 300 350
Figure 2: Observed
ground thawing and
freezing depths at
Granger Creek and
simulations by five
algorithms and three
sets of model run
surface water (mm/d)
Commonly used set of Johansen’s formulations
van Genuchten equations
Unfrozen water
content
parameterisations
Zhao and Gray, 1997; Pomeroy et al.,
2007
FD-AHCP
50 100 150 200 250 300 3500
Temperature(oC)
Complete-Johansen Farouki, 1986
0
Thaw depth (m)
ZHAO-IN
Complete set of Johansen’s formulations
-0.2
-0.4
FD-DECP
-0.6
-0.8
Thaw depth (m)
Zhao and Gray’s parametric infiltration algorithm
-0.8
0.0
(f)
(e)
-0.2
-0.4
-0.3
Infiltration(mm)
Gray et al., 1985; Pomeroy et al., 2007
-0.8
0.0
-0.3
Soil liquid moisture content (m3 m-3)
Zhang et al., 2000; Peckham et al., 2009
GRAY-IN
-0.6
Runoff (mm)
Mein and Larson, 1973; Verseghy, 1991
IT-TOPO
Gray’s empirical infiltration algorithm
-0.2
-0.4
Thawing / freezing depth (m)
ML-CLASS
Instantaneous infiltration algorithm in Topoflow
(d)
Thawing / freezing depth (m) Temperature(oC)
Modified Mein and Larson algorithm for non-uniform soils
(d) 30cm soil depth
at Granger Creek
(c) 20cm soil depth
at Granger Creek
0.4
0.5
(a) 10cm soil depth
at Sotty Creek
Both analytical algorithms modified for non-uniform soil from Green-Ampt and Mein-Larson methods could
simulate the infiltration into organic covered frozen and unfrozen soils reasonably well as long as soil thermal and
hydraulic properties are appropriately parameterised and soil thawing depth is accurately represented.
Simulated with Run1
Observed thawing depth
Simulated with Run3
Simulated with Run2
Observed freezing depth
0.0
12
T at 0.7m soil depth
8
(b)
-0.1
4
T at soil surface
ATIA
-0.2
0
-4
-0.3
-8
(a)
-12
-0.4
0.0
0.0
(d) -0.1
(c)
-0.1
TDSA
HMSA
-0.2
-0.2
surface water (mm/d)
Flerchinger et al., 1988; Flerchinger and
Saxton. 1989
GA-SHAW
-0.8
0.0
TDSA
HMSA
-0.6
o
Goodrich, 1978; Smirnova et al., 2000;
Nicolsky et al., 2007
Modified Green and Ampt algorithm for non-uniform soils
Observed thawing depth
Observed freezing depth
0.0
(b)
-0.2
ATIA
-0.4
-0.6
(c)
-0.2
-0.4
Temperature( C)
Finite difference numerical scheme with the Apparent Heat Capacity FD-AHCP
Parameterization
Figure 1: Observed
ground thawing and
freezing depths at
Scotty Creek and
simulations by five
algorithms and three
sets of model run
Infiltration(mm)
Verseghy, 1991; Foley et al., 1996;
Shoop and Bigl, 1997;Dai et al., 2003
Simulated with Run3
Soil liquid water content (m3 m-3)
Hayashi et al., 2007
FD-DECP
Simulated with Run1
Simulated with Run2
24
T at 0.7m
20 (a)
16
soil depth
12
T at soil
8
surface
4
0
-4
0.0
Runoff (mm)
Woo et al., 2004; Yi, et al. 2006
HMSA
Finite difference numerical scheme with the Decoupled Energy
Conservation Parameterization
Thawing / freezing depth (m) Temperature(oC)
TDSA
Hayashi’s Modified Stefan Algorithm
-8 -7 -6 -5 -4 -3 -2 -1 0 -8 -7 -6 -5 -4 -3 -2 -1 0
0.3
Numerical models with an apparent heat capacity treatment gives the most accurate simulation of ground
thawing/freezing depths in permafrost sites as long as appropriate time and spatial resolutions are configurated
and accurate ground surface temperature is supplied.
Thawing / freezing depth (m) Temperature(oC)
Two Directional Stefan Algorithm
0.5
Observation
VG-Para
0.5
0.4
All the empirical and semi-empirical algorithms to quantify ground thawing depth and water infiltration are
subject to site specific parameter calibration, thus are not suitable for land surface and regional hydrological
models that normally operate across various site conditions
Wolf Creek North Facing Slope
RESULTS
Temperature(oC)
Soil hydraulic
parameterisations
Nelson and Outcalt, 1987; Hinkel and
Nicholas, 1995
Thaw depth (m)
Soil thermal
conductivity
parameterisations
Key References
ATIA
Infiltration(mm)
Infiltration
Algorithms
Abbreviations
Accumulated Thermal Index Algorithm
Runoff (mm)
Thawing
and
Freezing
Algorithms
Algorithms / Parameterisations
0.1
CONCLUSIONS
Wolf Creek South Facing Slope
(WC_SFS) 61°18’N, 121°18’W, 280 m
Organic layer Depth: N.A.
Permafrost Table: N.A.
TESTED ALGORITHMS AND PARAMETERISATIONS
Categories
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
(b) 30 cm soil depth
at Sotty Creek
0.2
hydraulic conductivity (m s-1)
40
(a) 10cm soil depth
at Sotty Creek
0.3
hydraulic conductivity (m / s)
Thawing depth(m)
Wolf Creek Forest Site
60°36’N; 134°57’W, 750 m
Organic layer Depth: ~0.1 m
Permafrost Table: N.A.
20
UFW-SL
UFW-PF
0.4
water pressure head (m)
0
Scotty Creek Peat Plateau
61°18’N, 121°18’W, 280 m
Organic layer Depth: ~3 m
Permafrost Table: >0.7m
Observed freezing depth
0.0
(b)
-0.2
-0.4
-0.6
Scotty Creek
-0.8
Scotty Creek
0.0
-0.1
-0.2
-0.3
-0.4
Observation
UFW-WP
Observed thawing depth
(a)
unfrozen soil water content (m3m-3)
Close to one-third of the earth’s surface is underlain with permafrost and much of the permafrost
terrain is covered with a surface organic layer of various depths. The need to improve
mathematical representation and parameterization of cold region processes in land surface and
hydrological models have been well recognized in recent decades. However, progress has been
hindered by (a) the complexity and variability of the soil system associated with thawing/freezing
processes and organic cover and (b) the shortage of high quality field data due to the technical
and logistic difficulties imposed by the harsh environments. Large variations exist in the
parameterizations of thermal and hydrological processes in current land surface and
hydrological models. Many of them were developed and validated in soil and climate conditions
different from those in permafrost regions. In this study, efforts have been made to examine the
most important thermal and hydraulic parameterizations and their effects on the simulations of
ground thawing/freezing and infiltration/runoff processes against detailed measurements
obtained at six field sites in Canada’s discontinuous permafrost region. The tested algorithms and
parameterizations include:
0.0
-0.2
-0.4
-0.6
-0.8
Volumetric soil water content (m3m-3)
Complete-Johansen
De Vries
Common-Johansen
MODEL TESTING SITES
unfrozen soil water content (m3m-3)
INTRODUCTION
Regions Research Centre, Wilfrid Laurier University, Waterloo, Ontario, Canada
Programs Branch, Yukon Department of Environment, Whitehorse, Yukon Territory, Canada
Freezing depth(m)
3Environmental
Volumetric soil water content (m3m-3)
1Dept.
Figure 7: Observed ground thawing,
infiltration, runoff and soil water content
at Wolf Creek forest site and simulations
by three infiltration algorithms
This study was supported by the Canadian Foundation for Climate and Atmospheric Sciences (CFCAS) via the project: Improved processes and
parameterisation for predictions in cold regions (IP3), International Polar Year (IPY) and the Natural Sciences and Engineering Research Council of Canada
(NSERC). Some of the modelling source codes were contributed by G.N. Flerchinger, L.E. Goodrich, M. Hayashi, M. Mollinga, D. Verseghy, M-K. Woo and S. Yi.
Some field data were provided by M. Hayashi, M. MacDonald and N. Wright. We greatly appreciate the inspirations and comments made by C.R. Burn, J.
Pomeroy, R. Soulis and Yu Zhang.