| Info
|
|
|
Ice modelling activities within ICEMON.
Ice models are important as input to the numerical weather prediction (NWP) models,
and are regarded as the future of ice charting.
The activites presented are not limited to services available in the ICEMON portfolio.
|
|
|
IRIS and HELMI are two models that are running in pre-operational state
at the Finnish Institute of Marine Research. They are developed in cooperation with other
partners.
The IRIS model is a dynamical ice model that describes the ridging of the ice,
and is coupled to SAR data through a feedback system.
The main objectives with the IRIS model is to:
develop ice modelling and SAR interpretation so that ridging parameters are obtained
to include the parameters into systems of ice information delivery
to relate the parameters to the trafficability of ships
and finally to apply the enhanced ice information in ship route selection.
The HELMI model is a multi-category sea-ice model that describes the motion of the ice,
ice concentration and ridging properties.
Both models have a resolution of 5km and is issued 5-7 days a week during the ice season on the
Baltic Sea.
View animation (1200 kb) of observed and modelled (HELMI) ice concentration for the 2004 winter.
View animation (893 kb) of modelled (HELMI) deformed ice thickness for the 2004 winter.
View example products on the ICEMON website
|
Figure 1a. IRIS model ridge density.
|
|
HIROMB (High-Resolution Operational Model for the Baltic) is a three-dimensional
baroclinic ocean forecasting model developed at SMHI (Swedish Meteorological and
Hydrological Institute). HIROMB is used in operational services and research.
HIROMB is driven by a N Atlantic storm surge model, surface fluxes from an atmospheric
model and fresh water inflow from a hydrological model.
HIROMB is supplied with an ice module, producing results in form of ice concentration,
ice thickness, ridge density and ridge height. The ice model has viscous-viscoplastic dynamics.
SMHI is a partner in the IRIS project described above.
|
Figure 1b. HIROMB model output.
Click to view full size animation.
|
|
MI-IM(Meteorological Institute Ice Model), a new state-of-the-art dynamic-thermodynamic sea
ice model has been developed based on the elastic-viscous-plastic (EVP)
dynamics of Hunke and Dukowicz and the thermodynamics of Mellor and Kantha.
The ice model has been coupled with both ocean models
MI-POM
and MICOM and the atmosphere models HIRHAM and MM5.
Simulations utilising the new coupled systems are in progress.
Some major features of MI-IM are:
To calculate the thermodynamic fluxes, the ice model operates with three layers in the vertical.
A snow layer, which is working as an insulator and as an absorber of incoming radiation
in the melting period. Then two ice layers of equal thickness.
MI-IM receives sea surface- temperature, salinity and currents from an
ocean model. Variables needed for atmospheric input are mean sea level pressure,
10 m wind, 2 m air temperature, 2 m dew point temperature, cloud coverage and precipitation rate.
In addition to ice concentration, ice thickness and ice velocity, MI-IM calculates the heat and fresh water flux at the interfaces between the atmosphere and ocean and between the ice and ocean.
The text is taken from the met.no website text on MI-IM.
Read more about the MI-IM on met.no website
|
Figure 2. MI-IM four day forecast of ice concentration.
Click image to view full size
|
|
Coupled sea ice - ocean models at the Canadian Ice Center
CIOM east coast, POM on a 5-20km grid (BIO)
CIOM nested grid (BIO)
Gulf of St Lawrence, GF8 on a 5km grid (IML)
Arctic Ocean, MOM v2.2 on a 50km grid (IOS 2004)
Archipelago, MOM v2.2 on a10-20km grid (IOS 2005)
CIOM Great Lakes (CIS 2005)
Existing Products include ice drift, ice concentration and ocean currents. New products will include
forecast ice charts, ice drift charts and regular ice charts.
Some main components of the coupled sea ice - ocean model are Zhang-Hibler momentum solver,
viscous-plastic rheology, multi-category ice thickness distribution function,
Thorndike ice thickness redistribution and zero layer thermodynamics (no snow).
Under development are a particle in cell module with momentum solved on regular grid,
a spherical coordinate system, ice advected as lagrangian particles carrying level and
deformed ice thickness and concentration, viscous plastic rheology and an
improved ice thickness redistribution scheme.
The text is extracts from presentations given by Tom Carrieres at International Ice Chart Working Group meetings in St Petersburg
2003 and Hamburg 2004, and the ICEMON S2 Methods compendium prepared by Jan Askne.
Read more about the ice models at the website of the Canadian Ice Service
|
Figure 3. Prognosis chart from the CIS.
Click image to view full size
|
|
TOPAZ System is an ocean forecasting system that includes sea ice.
It uses the following numerical models HYCOM + KPP (U. Miami - LANL,USA), sea ice thermodynamics model and
sea ice dynamics model (EVP, Hunke & Dukowicz 1997). There are plans to
use a new HYCOM version (MPI) and a multi-category sea ice model.
The system uses the following data:
Altimetry, SST (CLS, F)
Sea Ice concentration (NSIDC, USA)
SLA, SST and Ice concentration is assimilated through an Ensemble Kalman Filter (Evensen 1994, 2003).
The model outputs relevant to sea ice are concentration and thickness. The plan is to add ice drift
and ice temperature.
It covers the Arctic and Antarctic with 18-30km resolution. It is possible to downscale to regions,
e.g. the Barents Sea with 4-5km resolution.
The text is extracts from a presentation given by Knut Arild Lisæter at the International Ice Chart Working Group meeting
Hamburg 2004.
Visit the TOPAZ website where the latest forecast can be downloaded
|
Figure 4. TOPAZ ice concentration forecast.
Click image to view full size
|
|
|
Modelling activities outside ICEMON
FOAM is a model being developed at UK Met Office. The motiviation are to
improve operational forecasts of sea-ice and ocean, improve seasonal and decadal forecasts,
improve NWP forecasts by using better sea-ice data and reduce uncertainties in sea-ice models by detailed
intercomparison of models & data.
The FOAM system will consist of
Operational daily 5-day forecasts of temperature, currents & sea-ice
Driven by atmosphere, assimilates satellite & in situ observations
Relocatable high resolution nested model capability
Hindcast capability (back to 1997)
The system will use sea-ice model developed for HadGEM1 and derived from the CICE model (Hunke & Lipscomb, LANL).
It uses zero layer thermodynamics (Semtner 1976),
elastic viscous plastic rheology (Hunke & Dukowicz 1997)
and ice thickness distribution from (Lipscomb 2001)
Principal data sources for assimilation are SSM/I (ASI) , QuikSCAT (CERSAT), and SSM/I (NSIDC).
The QuikSCAT product is a part of the ICEMON portfolio, which makes UK Met Office an important user
for ICEMON.
Information is taken from a presentation given by John Stark at the International Ice Chart
Working Group meeting in Hamburg April 2004.
Go to the FOAM website at UK Met Office for real time data and information
Access archived data from the FOAM model at the NERC Environmental Systems Science Centre
|
Figure 5. 1/3° operational April 2001
|
|
|
| |
|
This page was last modified: August 30 2004 12:49:42. |
Training
View animation (1200 kb) of observed and modelled (HELMI) ice concentration for the 2004 winter.
View animation (893 kb) of modelled (HELMI) deformed ice thickness for the 2004 winter.
