Field campaigns are needed to obtain information for both model development and to verify emission models and retrieval algorithms.
1. Scientific Studies and Algorithm DevelopmentStudies on sea ice emissivity focussed on frequencies below C-band are scarce and the same applies to availability of sea ice brightness temperature data. Emissivity models have been developed for higher frequencies, but they have not been verified at L-band.
2. Campaigns (More urgent than Category 1)
3. Synergy with Other Sensors (Not prioritised against 1 and 2)
2.1. Scientific Studies and Algorithm Development
(A) Emission Behavior of Sea Ice (Most urgent topic in Category 1)
(B) Detection of Thin Ice
The relatively long wavelength associated with L-band may be useful for mapping ice up to 0.5 meters, depending on ice salinity and surface temperature
Another reason is that the growth of thin sea ice is associated with a brine flux to the underlying ocean and this in turn has a profound effect on the water mass properties of the ocean. It is believed that the brine generated from new sea ice growth is an important driver of the global thermohaline circulation
- The possibility of discriminating thin sea ice and determining ice thickness should be investigated based on progress in the development of ice emission models
- Attention should be paid to identifying the weather and seasonal conditions and ice parameters (salinity, temperature) under which information on thin ice can be obtained
- Thin sea ice algorithms should be developed and verified.
(C) Detection of Melt Ponds
The largest source of error in mapping Arctic sea ice concentration during the melt season is the presence of melt ponds on sea ice.
Ice concentration estimates as much as 20 % too low may result from this, due to the poor capability of present algorithms, relying on the frequency range of SSM/I, to discriminate melt ponds from open water
The salinity of water in melt ponds is much lower than that of water between the floes. Therefore, the sensitivity of the L-band SMOS instrument to water salinity has the potential to distinguish between melt ponds and the high salinity sea water. This may, however, require observations at both L-band and shorter wavelenghts.
- The feasibility of using L-band data to discriminate melt ponds should be investigated
- The optimum frequencies for this algorithm should be determined, based on the availability of data from other spaceborne microwave radiometers in the near future
- The developed algorithms should be verified.
(D) Determination of Ice Temperature
Based on experiments conducted previously with radiometers operating at C-band and higher frequencies, C-band is the best frequency range for determining the temperature of the radiating portion of sea ice.
SMOS may contribute to ice temperature observations, because the penetration depth at L-band is much larger than at C-band.
Additionally, L-band observations are less sensitive to various surface features that may cause errors in temperature measurement.
- The capability of L-band radiometry to determine sea ice temperature should be investigated.
- Additional observations at C-band should be used.
- The developed algorithms should be verified.
(E) Snow Accumulation over Ice Sheets
Low-frequency measurements provide good penetration for snow over ice sheets.
The capability of L-band radiometry to obtain information on snow layering in deep snow areas should be investigated.
Experimental L-band data on sea ice and snow are needed urgently, accompanied with extensive high-quality ground truth
In general, airborne campaigns are preferred due to their capability to collect data on various sea ice types and snow scenes over relatively large areas
Ground-based radiometers are useful for collecting long time series of data.
- Airborne data collection should be started with available conventional L-band radiometers
- Later, two-dimensional interferometric radiometers should be used in order to gain experience on deriving the brightness temperature maps of cryospheric targets from their visibility functions
- Comparison of results obtained simultaneously with traditional and interferometric radiometers is necessary in order to verify the interferometric data
Airborne campaigns should be conducted in order to collect data to support activities discussed in Category 1 A through D. Good ground truth data are essential.
(A) Sea Ice (Most urgent activity in Category 2)
Campaigns should be conducted under various weather and seasonal conditions in order to cover all typical ice/snow conditions
Measurements should be conducted over the incidence angle range of SMOS using both vertical and horisontal polarisation
Airborne campaigns should be conducted in order to support activities discussed in 1 E
(B) Snow over Ice Sheets
Data collection flights should be carried out over deep snow areas. Ground truth activities under summer conditions should utilise existing research camps.
The test sites should be large enough to allow comparison of airborne data with those from the spaceborne AMSR sensors. The possibility of combining these experiments with sea ice experiments should be considered.
(A) Scientific Studies
Multichannel (polarisation, frequency) radiometer data have traditionally been used to maximise information on a target.
The feasibility of using SMOS data jointly with other datasets (AMSR, etc.) for cryospheric applications should be investigated. For example, detection of melt ponds and determination of ice temperature may require joint use of L-band and C-band data.
Airborne radiometer measurements of cryospheric targets should be conducted simultaneously with L-band radiometers and higher-frequency receivers operating at the AMSR frequencies (6.9 to 89 GHz). This would allow determination of synergy between various frequencies and polarisations.