Transcript PowerPoint-presentasjon

The WWRP-THORPEX IPY Cluster
Coordinator: Thor-Erik Nordeng
Norwegian Meteorological Institute (met.no), Oslo, Norway
(Fronted by David Burridge)
The IPY-THORPEX Cluster
10 individual projects
(see WMO Bulletin Oct. 2007)
The objectives of the IPY-THORPEX Cluster are:
– Achieve a better understanding of small scale
weather phenomena
– To improve the understanding of physical/dynamical
processes in polar regions
– Explore use of satellite data and optimised
observations to improve high impact weather
forecasts
– Utilise improved forecasts to the benefit of society,
the economy and the environment
Meteorologisk Institutt met.no
The WWRPTHORPEX
IPY cluster
WWRP-THORPEX IPY Cluster
(T.E. Nordeng, coordinator)
ARCMIP
Arctic Regional Climate
Model Intercomparison Project
(K. Detholf, Alfred-Wegener Institute)
STAR
Storm Studies of the Arctic
(J. Hanesiak, U Manitoba)
GFDex
Greenland Flow
Distortion experiment
(I. Renfrew, U. East Anglia)
TAWEPI
Norwegian IPY-THORPEX
Thorpex Arctic Weather
and Environmental
Prediction Initiative
(Ayrton Zadra,
Environment Canada)
(J.E. Kristjansson, U Oslo)
GREENEX
(H. Olafsson, Iceland & DLR)
Impacts of surfaces fluxes
on severe Arctic storms, climate change
and coastal orographic processes
(W. Perrie, BIO Canada))
T-PARC
Concordiasi
THORPEX Pacific Asian
Regional Campaign
(D. Parsons, NCAR)
Use of IASI data
(F. Rabier, Meteo-France)
Greenland Jets
(A. Dombrack, DLR)
Meteorologisk Institutt met.no
Polar lows
Meteorologisk Institutt met.no
Topogographically induced jets
(light grey is strong wind)
Meteorologisk Institutt met.no
Lee waves under capping inversion
(strong downslope wind and turbulence)
Meteorologisk Institutt met.no
Channeling
(Sandvik and
Furevik, 2002)
Meteorologisk Institutt met.no
Challenges – initial conditions
Model improvements
Use of satellites difficult
Few traditional observations
RMS error of mslp forecasts with the Norwegian limited area model system
(HIRLAM) over a two year period; the Barents Sea in red and the North Sea
in blue.
Meteorologisk Institutt met.no
How to improve NWP (in Polar regions)
• Better understanding of physical
processes  improve the models
• Use more observations
• probability forecasts
Meteorologisk Institutt met.no
Meteorologisk Institutt met.no
Targeting
Strategy:
• compute sensitivity area before the
actual forecast starts
• go there (by plane)
• drop sondes
Meteorologisk Institutt met.no
Norwegian
IPY-THORPEX
(J.E. Kristjansson,
U of Oslo)
GFDex
Greenland Flow
Distortion experiment
(I. Renfrew, U. East Anglia)
TAWEPI
Thorpex Arctic Weather
and Environmental
Prediction Initiative
(Ayrton Zadra,
Environment Canada)
Examples from the
WWRP-THORPEX
IPY cluster
Concordiasi
(F. Rabier, Meteo-France)
Meteorologisk Institutt met.no
The targeted sondes improve
the forecast of the polar low at landfall
CONTROL forecast
TARGETED forecast
Verification:
ECMWF analysis
13
Meteorologisk Institutt met.no
Targeting During GFDex
(Emma Irvine, Suzanne Gray and John Methven (University
of Reading) + David Walters (Met Office))
• 5 cases:
– 24 February
– 26 February
– 01 March
– 03 March (NULL)
– 10 March
• 5 -11 targeted dropsondes per flight
• Data transmitted to GTS in real-time and assimilated
into Met Office operational forecast
(The flight on 1st March is the green track on the diagram.)
Meteorologisk Institutt met.no
• Four targeted observing flights were conducted during GFDex,
around southern Greenland and Iceland
• Targeted sonde data was used by the data assimilation system
to modify the background state and influence the forecast via
analysis increments
• The forecast improvement is small compared to the forecast
error for the same period; targeted observations have both
improved and degraded the forecast
• The 1st March case showed that modification of the upper-level
PV anomaly by the inclusion of targeted sonde data led to
forecast improvement propagating into the Scandinavian
verification region with a developing polar low
Meteorologisk Institutt met.no
Planned targeting experiment (Concordiasi)
1)
Determination of sensitive area
1)
Depending on the track and/or swath of IASI and AIRS sensors
Track of IASI the 7th October 2007.
The colour gives the hour of the passage.
2)
Also depending on the predicted sensitive area at 18hUTC. (Ex with VORCORE data)
Predicted sensitive area valid on the
2007/10/07 at 18Z, initialized at 00Z and
optimized for the 2007/10/09 at 00Z.
Balloon trajectories start on the 2007/10/07 at
00Z and reach sensitive areas at 18Z. The blue
shading shows mean wind speed at 50 hPa on
that period (ECMWF operational forecast).
The navy dashed curve shows the limits of sea
ice as in ECMWF system.
2)
Targeting of sondes in these area
Meteorologisk Institutt met.no
Ensemble prediction
• Estimate the forecasted pdf (probability density
function) rather than single deterministic approach
• Assumption: # of perturbed forecasts large enough
to cover the whole ”true” pdf.
method
• run a number of integrations from a number of
(optimally) perturbed initial states
• combine results from a number of models
 Use spread as a measure of uncertainty
Meteorologisk Institutt met.no
Downscaling LAMEPS with high resolution model (UM – 4 km)
Flight 3: 4 March 10.15-13.30 UTC
(Silje Sørsdal (Master thesis, UiO, Norwegian IPY-THORPEX)
Meteorologisk Institutt met.no
Probability of wind at 925hPa>25m/s
T+42h (12UTC 04.03)
LAMEPS
UM-EPS
•Comparing with observation data from flight 3(flight time 10.15-13.30).
•Black contours are std.dev of MSLP.
Meteorologisk Institutt met.no
Collaboration: Status of extended regional model at CMC*
Polar extension of CMC’s
regional NWP model
• global, rotated, variable-resolution
lat-lon grid
• core: 15-km resolution
• 58 hybrid vertical levels, top 10 hPa
• timestep: 7.5 min
Implementation plans
• 4 runs (48-h forecasts) per day
• to replace current operational
regional model
• probable implementation in the
winter of 2009/2010
Fig.: Grid of CMC’s next regional model
(Note: Only every 5 grid-point is shown)
_____________________________________________________________________________
* Project partly funded by IPY-LIEP. Grid parameters kindly provided by A. Patoine (CMC).
Meteorologisk Institutt met.no
TAWEPI subproject 1:
Coupling snow and ice
Coupling flowchart
Y.-C. Chung, S. Bélair, J. Mailhot
Goal
To investigate snow and sea ice evolution in
the Arctic Ocean by a coupled snow/sea ice
system
Methods
Sequentially couple models:
• 1-D, multi-layer, offline sea ice model in
Meteorological Service of Canada (MSC) operational
forecasting system
• 1-D, multi-layer snow model SNTHERM (Jordan,
1991)
• 1-D, blowing snow model, PIEKTUK (Déry, 2001)
SHEBA
Surface Heat Budget of the Arctic
Ocean (SHEBA) Datasets
• Multi-year ice floe, drifted more than 1400 km in the
Beaufort and Chukchi Seas
• Measurements for one year from October 31, 1997
Meteorologisk Institutt met.no
- Sensitivity analysis of snow depth


Wind effect and error related to new snow
density should be considered in winter
During ablation period, uncertainties in albedo
affect stored energy & grain size, retarding or
accelerating spring snow melt
temporal evolution
of snow depth
- Temporal evolution



The model predicts snow depth well after
considering erosion due to blowing snow
The model system captures accurately the start
of snow melt (5/29) and intensive snow melts
until snow depletion (6/24 ~7/12)
The model predicts the ice thickness very well
before snow depletion. The underestimation after
snow depletion is caused mainly by the error of
the ice model
- Vertical structure

Temperature, grain, density, thermal
conductivity, etc.
temporal evolution
of ice thickness
vertical structure of
snow temperature
Meteorologisk Institutt met.no
TAWEPI (Canada)
Subproject 1: Coupling snow and ice
Subproject 2: Polar-GEM clouds
Subproject 3: Sea-ice modelling
Subproject 4: Sensitivity studies in the
Arctic using singular vectors
Subproject 5: Hyperspectral IR
assimilation in the Arctic
Subproject 6: GEM IPY Analyses
Meteorologisk Institutt met.no
Outcome of the THORPEX IPY cluster
Data for improving physical parameterization in NWP
models, -clouds, microphysics, surf fluxes
Improved assimilation techniques for high latitudes
with emphasis on satellites data
Increased understanding on the effect of the use of
ensemble simulations for high latitudes
Increased understanding on the effect of targeting in
high latitudes
Increased understanding of dynamics of high latitude,
particularly high impact weather phenomena
Demonstration of the effect of new instruments
Demonstration of the effect of increased Arctic and
Antarctic observations for local and extratropical
NWP forecasting.
Meteorologisk Institutt met.no
IPY legacy
• We call for an immediate, high-level
and sustained focus on polar prediction
services, stimulated, led and
coordinated by WMO, as the best way
to integrate and synthesize the IPY
observational efforts and to
communication and maximise the
impact of IPY science
(David Carlson, IPY IPO, July 2009 - in preparation)
Meteorologisk Institutt met.no
Thank you for your
attention
with special thanks to
scientists of the
THORPEX IPY cluster
projects and others who
contributed to this summary
Meteorologisk Institutt met.no