Le système de modélisation
environnementale communautaire MEC
V. Fortin1, P. Pellerin1, A. Pietroniro2, B. Davison3
1
RPN: Numerical Weather Prediction Research
2 NWRI: National Water Research Institute
3 HAL: Hydrometeorology and Arctic Lab
1
The idea behind MEC: make the surface
scheme an independent component from
the atmospheric model
Atmospheric
model
with its own
surface scheme
Atmospheric
model
“On-line”
mode
“Off-line”
mode
SVAT
Hydrological model
with its own land surface scheme
Hydrological
model
2
Increasing the spatial resolution of
the land-surface scheme
• We anticipate a large impact on the prediction of
cold processes
• We hope to be able to show that this also leads
to better atmospheric forecasts
3
Using the offline mode to predict
snow water equivalent
• SWE predictions in Southern Quebec (2001-05)
GEM
@ 24 km
«offline»
mode
300
200
100
0
0
100
200
300
Observé (mm)
(b) ISBA
400
Simulé (mm)
Simulé (mm)
CLASS and ISBA
@ 0.1º
(a) CLASS 3.0
400
400
300
200
100
0
0
100
200
300
Observé (mm)
400
4
Using the offline mode to predict
snow water equivalent
• SWE predictions in Southern Quebec (2001-05)
(b) Modèle de neige du CEHQ
Simulé (mm)
400
GEM
@ 24 km
300
200
100
0
«offline»
mode
100
200
300
Observé (mm)
(a) CLASS 3.0
400
300
200
100
0
0
100
200
300
Observé (mm)
400
(b) ISBA
400
Simulé (mm)
Simulé (mm)
CLASS and ISBA
@ 0.1º
0
400
300
200
100
0
0
100
200
300
Observé (mm)
400
5
Combining MEC predictions with
obs. to obtain a SWE analysis
• SWE analysis in Southern Quebec (9-04-2002)
450
ÉEN prédit par MEC (mm) .
400
GEM
@ 24 km
350
300
250
200
150
100
50
0
0
«offline»
mode
50
100
150
200
250 300
350
400 450
450
52
400
51
350
50
CLASS and ISBA
@ 0.1º
500
ÉEN estimé par le CEHQ aux stations nivométriques (mm)
300
49
250
48
200
47
150
46
100
45
50
280
285
290
295
6
CLASS status within MEC
• CLASS 3.2 driver just completed
– mosaic resolving land cover heterogeneity
implemented
– subdivision of the third layer not implemented
• A number of bugs have been uncovered in
MEC during the development of the driver
– we expect to solve these problems by the end
of the calendar year
• In the meanwhile, CLASS 3.0 can be used
7
Moving towards MEC 2
GEM 3.2.2
GEM x.x.x
MEC 2
PHY 4.4
(CLASS 3.0)
MEC 1.0
Operations
PHY y.y
(CLASS 3.3)
MEC 1.1
IP3
MEC 1.2
NAESI
HEPEX
Development
Research
8
Timeline
• FY 2006 – 2007: Research & Development
– Develop a high-resolution land-data assimilation
system to update the prognostic variables of the landsurface scheme using surface and satellite
observations
– Evaluate the impact of two-way coupling of the
atmoshperic model with a higher resolution landsurface model on atmospheric and environmental
predictions
• FY 2008: Parallel NWP and EP system
– Start using MEC for environmental and weather
prediction at CMC in parallel with the existing system
– Determine if we really make improved predictions
9