GCOS GOOS WCRP/OOPC IX/3
SMMR/SSM/I brightness temperature record with some care. The tricky part is
comprehensive inter comparison and validation. Few groups appear to have the motivation or
resources to perform truly comprehensive inter comparison and validation studies, which
would be a necessary part of deciding which one algorithm (or combination of algorithms) is
best for climate monitoring purposes.
It is also necessary to critically assess the passive microwave record in the context of
extended GCOS climate monitoring principles (in terms of its stability, homogeneity and
continuity). Given that I have had trouble finding out whether or not there are stability issues
with the passive microwave data, I suggest this sort of information could be better presented.
Climatically speaking, the 25 years of the passive microwave is insufficient by itself to study
decadal variability, so homogeneity over the whole data record must also be assessed. In order
to do this, we need a good understanding of how charts are and were created and exactly how
this relates to a passive microwave concentration field.
A lot of historical chart data for the Arctic have become available over the last five
years or so, which is a major step forward. There has been less improvement in the Antarctic,
but there are fewer data to find and those that are around are likely point observations which
will need digging out of archives and assembling or reconstructing. The first meeting of the
JCOMM Expert Team on Sea Ice agreed to prepare historical information for the Southern
Ocean and a report is due this autumn. In the meantime, AARI have useable data in their
archives for the 1950s and 1960s and I believe NSIDC plan to digitise these shortly.
Assuming the Expert Team survey identifies some potentially fruitful data sources, OOPC
should encourage funding for digitization of historical Southern Ocean sea ice.
The usefulness of passive microwave retrievals in summer time is seriously limited by
the instruments' inability to see through melt water and wet snow on top of the ice. More
work on finding ways around this should be encouraged (by approaching funding bodies).
This research is best done by remote sensing experts. However, the OOPC should first check
the status of the EUMETSAT Ocean and Sea Ice SAF project as they appeared to be
intending to go in the right direction in this regard.
Uncertainties in passive microwave sea ice concentration retrievals are estimated in a
limited fashion for some algorithms and given in the literature, so encouragement should be
given to derive comprehensive fields of uncertainty. The inability of most algorithms to
retrieve thin ice should also be reflected in these estimates, as this is important for monitoring
of total ice extent. However, complete and meaningful uncertainty estimates will not be
available until a best (combination of) algorithm(s) is identified, the stability and
homogeneity of the record is assessed and we can retrieve more accurate concentrations at
times of melt. It is also necessary to encourage production of error estimates for historical
chart data sets, so these can be meaningfully compared with satellite retrievals and their
Progress will likely be best made by liaising with the lead author team on the
Observed Cryosphere chapter of IPCC 4AR when it is announced. OOPC/IPCC could use
their combined influence to try and encourage focused work (aimed at solving the
aforementioned problems, which I see as the main challenges facing any progress in being
able to say anything new about sea ice change in the next IPCC report) from data set