OVERSEAS TRAVEL REPORT
IOCCG Meeting, Capetown, South Africa, 19-21 January 1998.
John Parslow
I attended the 3rd annual
meeting of the International Ocean Colour Coordinating Group in
Capetown from 19-21 January. The IOCCG was originally established
under the auspices of the Intergovernmental Oceanographic Commission
and the Committee on Earth Observing Satellites, to oversee and
coordinate ocean colour missions, research and applications. The
Committee includes a mix of representatives of space agencies
involved in ocean colour missions, and scientific experts. It
is supported by funds from the space agencies. Because of some
administrative difficulties, the committee has been re-established
under SCOR (Scientific Committee for Oceanographic Research) and
continues to receive the support of the space agencies. My travel
to the meeting was partly supported by the EOC. I've reported
below some highlights of the meeting of potential interest to
CSIRO and the Australian Ocean Colour Working Group.
The IOCCG continues to maintain a web
site (http://www.ioccg.org) which provides useful information on current
and upcoming ocean colour missions, conferences, research programs,
etc, with links to relevant homepages. The IOCCG also uses the
remote sensing publication backscatter as a publication
outlet, and welcomes articles and news on ocean colour activities.
The IOCCG would welcome a brief description of the AOCWG for inclusion
in the web-site, with a link to an AOCWG home page. It would also
welcome a longer article on the AOCWG for inclusion in backscatter.
Perhaps we should try to prepare such an article once we have
an AOCWG web-site established. The IOCCG web-site also includes
a jobs vacant page, which we may find useful for recruiting.
An update was given on current and planned
ocean colour missions, including MOS (Germany / India), SeaWiFS,
MODIS and MISR (NASA), MERIS (ESA), OCTS, GLI and S-GLI (Japan),
POLDER and POLDER-2 (ESA), OSMI (Korea), OCI (Taiwan) and OCM
(India). Among the items of interest:
DLR (Germany) has released a CD-ROM with
ocean colour results from the first 2 years of the MOS mission.
(I have a copy of this CD.) MOS is an imaging spectrometer with
a 200 km swath and 500 m pixels. It is possible to directly downlink
MOS at HRPT ground stations, although a special hardware box must
be obtained from India at a cost of at least US$65k. As far as
I know, no-one in Australia is working with MOS. DLR is keen to
have Australians involved.
Janet Campbell reported on the current
status of the SeaWiFS mission. She described the basis of the
default at-launch chlorophyll algorithm (OC2), which is based
on a data set of 911 points involving normalized water-leaving
radiance and chlorophyll. This is a standard 2-band ratio algorithm.
More complex semi-analytic algorithms such as Carders algorithm
are included in SeaDAS, but are regarded as experimental, with
possible application to CASE II waters.
Reprocessing of the SeaWiFS data was
started in January at NASA, using improved navigation and geometry,
as well as revised radiometric calibration. Because of the revised
geometry, this reprocessing needs to start from Level 0, not Level
1A. The in-flight radiometric correction was apparently based
on a quite small set of observations of water-leaving radiance.
This emphasizes the value of accurate ground-truth data.
It is worth noting that SeaWiFS has a
bilinear gain, and does not saturate over land or cloud. It therefore
provides highly accurate narrow-band spectral data over land.
These are being used at NASA to produce an NDVI product. I'm not
aware of any research in Australia on terrestrial applications
of SeaWiFS. There is some interest in vicarious calibration of
SeaWiFS and other ocean colour sensors, including MERIS, over
land targets. NASA and ESA are interested in the Australian land
calibration sites.
The website maintained by Mark Abbott
for MODIS was highly recommended.
MERIS is an ocean colour instrument to
be launched by ESA on ENVISAT. It has interesting spectral resolution
(15 bands), and options for high spatial resolution (300 m) in
coastal waters. The data distribution policy for MERIS is unclear,
but data will not necessarily be freely available, except to participating
investigators. There is a current announcement of opportunity
from ENVISAT (on the ENVISAT homepage) for pilot projects using
MERIS data, with letters of intent due in April. I think the AOCWG
should consider responding to this announcement.
NASDA has reprocessed the (historical)
OCTS global data sets using a Mark 3 algorithm. Cloud detection
is much improved, and this has removed the "spottiness"
from the global images. They have an extensive set of ground truth
data, obtained through collaboration with Japanese fisheries agencies,
as well as international collaboration. These data indicate good
performance up to 2 mg m-3 chlorophyll, but their algorithm appears
to saturate at higher chlorophyll levels. I was informed by Tanaka
from NASDA that he has approval to go ahead with an agreement
between NASDA and CSIRO/AOCWG for us to receive, process and distribute
the OCTS data for the Australian region.
GLI is intended for launch on ADEOS-II,
with earliest launch in June 2000. An AO for this mission has
been delayed. SGLI is intended for launch on ADEOS-III, but the
launch date is unclear. NASDA is facing budget cuts, and an increased
commitment to the space station.
Korea plans to launch an ocean colour
satellite OSMI in late 1999. They are looking for international
collaboration on ground truth and applications. OSMI is an imaging
spectrometer, with 2.6nm bands from 400 to 900 nm. It can record
only 6 bands, but will have capability for on-orbit selection
of bands. Korea is seeking suggestions from the ocean colour community
about possible experimental uses of the band selectivity.
OCI is an ocean-colour satellite to be
launched by Taiwan. It has a low-inclination orbit, and is not
sun-synchronous, which may allow some interesting applications
in tropical waters.
OCM is a CCD-based imaging spectrometer
with 350 m resolution and SeaWiFS bands to be launched by India
at the end of 1998.
Andre Morel presented a draft report
from an international workshop on minimum requirements for future
ocean colour sensors, which was funded by the IOCCG. This report
is intended to provided guidelines for basic operational ocean
colour sensors which might be included on future operational satellite
missions, such as NPOESS. The workshop report contains an excellent
summary of current and future sensors, and the current state of
knowledge for atmospheric correction and in-water algorithms.
I have an electronic copy, which I could forward to those who
are interested.
SIMBIOS is a NASA research project to
develop the methods and capability to produce consistent products
from multiple ocean colour sensors. It has a strong emphasis on
intercalibration of sensors and field instruments, and collaboration
on field measurements. Only US scientists are funded, but the
project welcomes participation by foreign scientists. It is an
open on-going project - foreign scientists can apply to participate
at any time. SIMBIOS will fund US scientists to participate in
foreign ground truth activities, and supports international working
group meetings, round-robins, etc. I think participation in SIMBIOS
could be a good way to upgrade our field optics campaigns, and
the AOCWG should consider developing a proposal to join SIMBIOS.
A global set of chlorophyll data is being acquired at NASA GSFC
in a SeaBASS database. Again, AOCWG could consider contributing
data to SeaBASS.
SIMBIOS is developing techniques to merge
data from different sensors, but does not have a mandate or funding
to produce merged global or regional data sets. The IOCCG concluded
that these data sets would be needed, and established a working
group to look into it.
A proposal was presented for an international
ocean colour cruise, which would provide both training and opportunities
for collaboration and intercalibration. The cruise is to be carried
out on a Russian ship, and consist of ca 6 3-week legs over a
6 month period. Cost for participants, including ship-time, was
estimated at ca $200/d, which is relatively inexpensive. It would
probably take place in 2000. A working group was established to
pursue this further.
There was discussion about future ocean
colour training courses for developing countries. A highly successful
training course was held in South America in 1997, and options
were considered for a training course in 1998 in South-east Asia
or Africa. There has been no decision as yet on a location. I
didn't offer Australia as a location, as I didn't feel I could
make the necessary time available to organise and host it. However,
if someone else is interested, please let me know.
The IOCCG has been given the task of
developing a project on Ocean Biology, as one of the 6 protype
projects under the IGOS program organised by the Strategic Implementation
Taskforce of CEOS. A draft project description was circulated.
There is potential for links from this project to another prototype
project, the Global Ocean Data Assimilation Experiment (GODAE),
run by Neville Smith.
Reports were presented on ocean colour
remote sensing in the Benguela Region, and on the repeat Atlantic
Meridional Transect run by Jim Aiken of Plymouth.
There was discussion about in-flight
intercalibration of ocean colour sensors, and it was agreed a
small workshop will be held in 1998 to assess the performance
of solar diffusers on current sensors.
Under the agenda item on validation,
it was agreed that there is an ongoing need for both validation
of atmospheric correction (ie accurate measurements of water-leaving
radiance), and consistent measurements of chlorophyll a. Jim Aiken
reported that, using on-board calibration, they had been able
to maintain extremely precise calibration of radiometers on their
cruises. The IOCCG has set up a working group to look at the development
of a consistent global chlorophyll database to be available for
validation of all sensors.
Reports were provided on cal/val activities
in Europe, and also on software development. Upgrades are planned
to provide capabilities for processing OCTS and MODIS data in
SEADAS. The European JRC may provide a MOS processing module for
SEADAS.
The IOCCG spent considerable time discussing
the number and specification of planned ocean colour sensors.
There is a perception at CEOS that there is redundancy in the
planned sensors, and that some missions should be scrapped. At
present, there is no clear statement of the number and specification
of ocean colour sensors needed to meet user requirements, and
no clear distinction between operational and experimental missions
in the planned sensors. CEOS and the agencies expect the IOCCG
to provide advice on this issue, and the IOCCG has formed a small
working group to prepare a report.
The IOCCG has adopted a membership policy
which involves rotation of the scientific members off the committee
after serving 3 year terms.
The next meeting has been tentatively planned to coincide with the Ocean Optics Meeting in November 1998. I have previously circulated a message from Joan Cleveland about the Ocean Optics meeting.