Snow Hydrology and Modelling
in Alpine, Arctic and Forested Basins
John Pomeroy
and collaboratorsRichard Essery (Edinburgh), Chris Hopkinson (CGS-NS), Rick Janowicz (Yukon Env), Tim Link (Univ Idaho), Danny Marks (USDA ARS), Phil Marsh (Env Canada), Al Pietroniro (Env Canada), Diana Verseghy (Env Canada), Jean Emmanual Sicart (IRD France
and Centre for Hydrology Faculty, Researchers and StudentsTom Brown, Kevin Shook, Warren Helgason, Chris DeBeer, Pablo Dornes, Chad Ellis, David Friddell, Warren Helgason, Edgar Herrera, Nicholas Kinar, Jimmy MacDonald, Matt MacDonald, Chris Marsh, Stacey Dumanski, Brad Williams, May Guan
Mountain Snow
Snow depth in January Snow depth in June
summer snow water reservesvast water reserves in winter snowpack
Study Elements• Processes
– Snow accumulation, structure and observation – Turbulent transfer to snow – Radiation effects on snowmelt under tundra shrubs and evergreen forests
• Parameterisations – Blowing snow over complex terrain– Irradiance in complex terrain – longwave from terrain, shortwave shadows – Forest snow interception, unloading and sublimation– Sub-canopy snowmelt– SCA Depletion in complex terrain,– Contributing area for runoff generation in snowmelt period
• Prediction – Wind and atmospheric modelling over complex terrain– Level of spatial complexity necessary in models– Regionalisation of CLASS parameters– Snow modelling contribution to MESH– CRHM
• Arctic and sub-arctic snow hydrology, Wolf Creek & Trail Valley Creek• Alpine snow hydrology, Marmot Creek• Montane forest snow hydrology, Marmot Creek
Blowing Snow in Complex Terrain
Inter-basin water transfer
Transport of snowto drifts
Supports glaciers,late lying snowfields,hydrologicalcontributing areas
Granger Basin, Wolf Creek,
Yukon Territory
NF
SF
LiDAR used to develop
topography and vegetation DEM
Wind Direction
SS ETdt
dSWE
Essery and Pomeroy, in preparation
0 500 1000 1500 2000 2500 30000
500
1000
1500
2000
2500
3000
0 500 1000 1500 2000 2500 30000
500
1000
1500
2000
2500
3000
Computer simulation of wind flow over mountains
Windspeed Direction
3 km
Granger Basin, Wolf Creek, Yukon
3 km
Simulation of Hillslope Snowdrift
Marmot Creek Research Basin
Bow River valley
Kananaskis River valley
x x
x
xx x
x
CRHM Mountain Structure
Alpine Hydrological Response Units
North Face
South Face(top)
South Face
(bottom)
Forest
Snow Transport
Snow Deposition
Sublimation
RidgeTop
Solar Radiation
Wind Direction
SourceSink
Winter Snow Redistribution Modelling
Winter Snow Redistribution and Sublimation
0%50%100%150%200%250%
0
200
400
600
800
Forest SF bottom SF top Ridgetop NF Transect
SWE/
Snow
fall
SWE (
mm
)
SWE SWE/Snowfall
0%
25%
50%
75%
050
100150200
Forest SF bottom SF top Ridgetop NF Transect Blo
win
g Sn
ow
Subl
imati
on/
Snow
fall
Blo
win
g Sn
ow
Subl
imati
on
(mm
)
Blowing Snow Sublimation Sublimation/Snowfall
Point Evaluation of Snowmelt Model2008 2009
-25
-15
-5
1-Apr 15-Apr 29-Apr 13-May 27-May 10-Jun 24-Jun
Date (2008)
Sno
w te
mp
(°C
)
Simulated active layer TMeasured active layer TSimulated lower layer TMeasured lower layer T
0
300
600
900
1-Apr 15-Apr 29-Apr 13-May 27-May 10-Jun 24-Jun
Date (2008)
Dep
th (
mm
)
Simulated SWE (mm)Measured SWE (mm)
0
1
2
3
1-Apr 15-Apr 29-Apr 13-May 27-May 10-Jun 24-Jun
Date (2008)
Dep
th (
m)
Measured depth (m)
Simulated depth (m)
0
1
2
3
1-Apr 15-Apr 29-Apr 13-May 27-May 10-Jun 24-Jun
Date (2009)
Dep
th (
m)
Measured depth (m)
Simulated depth (m)
0
300
600
900
1-Apr 15-Apr 29-Apr 13-May 27-May 10-Jun 24-Jun
Date (2009)D
epth
(m
m)
Simulated SWE (mm)
Measured SWE (mm)
-15
-10
-5
0
1-Apr 15-Apr 29-Apr 13-May 27-May 10-Jun 24-Jun
Date (2009)
Sno
w te
mp
(°C
)
Simulated active layer TMeasured active layer TSimulated lower layer TMeasured lower layer T
0
0.05
0.1
0.15
0.2
0.25
015
030
045
060
075
090
010
5012
0013
5015
00
SWE (mm)
f (S
WE
)
SWEmeasurements
Theoreticaldistribution
Frequency Distributions of SWE from LiDAR Depths and Measured Density
N facing slope
0
0.05
0.1
0.15
0.2
0.25
0.3
SWE (mm)f (
SW
E)
SWEmeasurements
Theoreticaldistribution
0
0.05
0.1
0.15
0.2
0.25
0.3
SWEmeasurements
Theoreticaldistribution
S facing slope
SWE distribution within HRU fit log-normal density distribution
Snowcovered Area from Oblique Terrestrial Photographs, Aerial Photographs and LiDAR DEM
Snow-covered Area Depletion Modelling
0
0.5
1
10-May 20-May 30-May 9-Jun 19-Jun 29-Jun 9-Jul
Date (2008)
SC
A fr
act
ion
Fully distributed Uniform Variable snowmelt Variable SWE dist. Observed
Observed – using oblique photographyUniform – spatially uniform SWE distributions and applied melt rates for each HRUVariable SWE dist. – each HRU has a distinct distribution of SWEVariable snowmelt – each HRU has a distinct melt rate appliedFully distributed – each HRU has a distinct distribution of SWE and applied melt rate
Four HRU (NF, SF, EF, VB) with modelled melt applied to SWE frequency distributions.
0
0.5
1
10-May 20-May 30-May 9-Jun 19-Jun 29-Jun 9-Jul
Date (2007)
SC
A fr
act
ion
Fully distributed Uniform Variable snowmelt Variable SWE dist. Observed
Snowmelt Runoff Intensity by HRU
0
0.5
1
25-Apr 5-May 15-May 25-May 4-Jun 14-Jun 24-JunDate (2008)
Are
a fr
act
ion
0 - 5 mm/day5 - 10 mm/day10 - 20 mm/day>20 mm/day
East facing slope
0
0.5
1
25-Apr 5-May 15-May 25-May 4-Jun 14-Jun 24-JunDate (2008)
Are
a fra
ctio
n
0 - 5 mm/day5 - 10 mm/day10 - 20 mm/day>20 mm/day
South facing slope
0
0.5
1
25-Apr 5-May 15-May 25-May 4-Jun 14-Jun 24-JunDate (2008)
Are
a fra
ctio
n
0 - 5 mm/day5 - 10 mm/day10 - 20 mm/day>20 mm/day
North facing slope
0
0.5
1
25-Apr 5-May 15-May 25-May 4-Jun 14-Jun 24-JunDate (2008)
Are
a fra
ctio
n
0 - 5 mm/day5 - 10 mm/day10 - 20 mm/day>20 mm/day
Overall alpine basin
Visualisation of Snowmelt Runoff Intensity
26-Apr 29-Apr
02-May 05-May
0-5 5-10 10-20 bare forest cliffMelt rates (mm/day)
Early Snowmelt Period - 2008
Snow Interception & Sublimation
Net Radiation to Forests: Slope Effects
South Face Clearing
North & South Face Forests
North Face Clearing
Date (M/D/YY)
10/1/07 11/1/07 12/1/07 1/1/08 2/1/08 3/1/08 4/1/08 5/1/08 6/1/08 7/1/08 8/1/08
SW
E [
kg m
-2]
0
50
100
150
200
level
30o north-sloping
30o south-sloping
Forest Snow Regime on Slopes
Date (M/D/YY)
10/1/07 11/1/07 12/1/07 1/1/08 2/1/08 3/1/08 4/1/08 5/1/08 6/1/08 7/1/08 8/1/08
SW
E [
kg m
-2]
0
20
40
60
80
100
level
30o north-sloping
30o south-sloping
Open slopes highly sensitive to irradiationdifference, forests are not
HRU Delineation• Driving meteorology:
temperature, humidty, wind speed, snowfall, rainfall, radiation
• Blowing snow, intercepted snow
• Snowmelt and evapotranspiration
• Infiltration & groundwater• Stream network
Model Structure
Model Tests - SWE
Streamflow Prediction 2006
Mean Bias = -0.13 all parameters estimated from basin data
Streamflow Prediction 2007
Mean Bias = -0.068 all parameters estimated from basin data
Conclusions• Appropriate process based models driven by
enhanced remote sensing and good observations can be used to achieve adequate hydrological prediction in the alpine.
• Model process and spatial structure must be appropriate to the complexity of the energy and mass exchange processes as they operate on the landscape.
• It is possible to test for the most appropriate structure for balance between model complexity and predictive ability.