The main source of observation coverage for regular monitoring of both sea ice and sea surface temperature (SST) is satellites. For SST state of the art products based on AVHRR imagery (Andersen et al., 1998, 1999, Eastwood et al., 1999) are available from EUMETSAT Ocean and Sea Ice Satellite Application Facility (OSISAF). These observations give a good data coverage in cloud-free areas. These data can be complemented with similar infrared imagery from MODIS and AATSR, and less cloud-dependent SST-retrievals are available from the AMSR-E instrument.

Passive radiances from the SSM/I instrument on the DMSP series of satellites enables retrieval of sea ice concentration and some information on ice type. This is a main source for production of low to medium resolution daily sea ice charts as well as long term monitoring of sea ice trends for the last decades. More recently passive sensors like AMSR-E and SSMIS give valuable observations for sea ice monitoring, and plans for future passive microwave radiometry ensures data continuity.

Active microwave instruments with low or medium resolution such as scatterometer data from the ERS-2 satellite as well as the Seawinds sensor on Quikscat provide medium resolution information on sea ice with good data coverage, and the new ASCAT sensor on the METOP satellite will ensure data continuity in the future. SAR data gives active microwave sea ice cover information with higher resolution. High-resolution information from AVHRR and MODIS give supplementary coverage in cloud-free areas.

In addition to the information on ice edge, concentration and type available from satellite sensors, there has recently been progress in deriving ice drift information from satellite data such as microwave radiances and from scatterometer observations. SSM/I data can be processed to give ice drift (for instance Kwok et al., 1998 and Martin and Augstein, 2000). Recently there has been advances in combining passive radiometry with scatterometer observations to produce multi-sensor ice drift products (Liu et al., 1999 and Ezraty and  Piollé, 2004). Another recent approach is to derive ice drift from resolution enhanced scatterometer observations (Haarpaintner, 2006). There is also a potential for deriving ice drift information on higher resolution from AMSR-E and AVHRR.

It is important to understand the differences between ice drift products at different resolutions and the error magnitudes and properties of the different products for best possible use of the ice drift data. This both necessitates assessing the error properties and characteristics of the satellite instruments, the retrieval algorithms as well as understanding the real small-scale ice drift motion spectrum to understand the differences caused by differences in spatial and temporal coverage of the observations.

Subtask 1: Satellite sea ice real-time data flow and algorithms

1.    Set up processing system. Implement sea ice algorithms in an operational framework at met.no and set up data flow for obtaining near-real-time input of sea ice data available from elsewhere. A main focus will be on sea ice drift algorithms.

Subtask 2: Intercomparison, validation and improvement of ice drift algorithms

1.    Compare satellite ice drift data to modelled ice drift and surface ice drift data available from Task 1.8 and 1.10 and other ice drift data sources during IPY.


Subtask 3: Multi-sensor ice drift product, optimal input to ice forecasting model

1.    Assess the representativeness problem of ground reference data: point measurements vs. satellite measurements.
2.    Low/medium satellite measurements vs. high-resolution satellite measurement.
3.    Develop multi-sensor product.
4.    Define optimal product for assimilation in sea ice model.