1. The Change Points of GLI data definition (Mr. Inoue, Fujitsu Ltd)
2. Data distribution concept for PIs (Mr. Ito, JAXA)
3. Reformat GLI data from MODIS data (Dr. Honda)
4. Other discussion

List of Attendees:

Dr. Genya Saito	genya@niaes.affrc.go.jp
Dr. Hiroshi Murakami	murakami@eorc.JAXA.go.jp
Dr. Masanobu Shimada	shimada@eorc.JAXA.go.jp
Dr. Kohei Arai	arai@is.saga-u.ac.jp
Dr. Akira Mukaida	akira@restec.or.jp
Dr. Yasushi Mitomi	mitomi@restec.or.jp
Dr. Takashi Nakajima	nakajima@eorc.JAXA.go.jp
Mr. Masakatsu Nakajima	nakajima.masakatsu@JAXA.go.jp
Dr. Kedar Mahapatra	Kedar@scc.u-tokai.ac.jp
Dr. Nguyen Duong	duong.nd@hn.onn.on
Dr. Koji Kajiwara	koji@ceres.cr.chiba-u.ac.jp
Dr. Tomohiko Oishi	tomo@scc.u-tokai.ac.jp
Dr. Motoaki Kishino	kishino@postman.riken.go.jp
Dr. Kanoko Muramatsu	muramatu@ics.nara-wu.ac.jp
Mr. Shuichi Tanahashi	tanahashi@ssd.se.fujitsu.co.jp
Dr. Daren Lu	ludr@sun.ihep.ac.cn
Dr. Roland Doerffer	doerffer@gkss.de
Dr. Masa Moriyama	matsu@cis.nagasaki-u.ac.jp
Dr. Michel Verstraete	michel.vestraete@jrc.it
Dr. Yoshiaki Honda	yhonda@rsirc.cr.chiba-u.ac.jp
Dr. Vu Saito	vsaito@eorc.JAXA.go.jp

• Chairman Dr. Honda explained the morning's 3 main topics and passed over to Mr.Inoue.

• Mr. Inoue started by giving the outline of his presentation, which covered the following:
  • GLI level definition (Levels 1B, 2A and 2, Level-2Map, Level-3 binned, Level-3STA map)
  • GLI standard products flow
  • Products and algorithm codes
  • List of standard Level-2 products and outline of Level-2A/segment products
  • Level 2-A_OA concept, Level 2-A_LC concept, Atmospheric Segment Data Concept
  • Level-2 topics: atmosphere, ocean, land, cryosphere, and product specification
  • Level-3 binned topics: atmosphere, ocean, and cryosphere
  • Level-3STA map topics: atmosphere, ocean, land, and cryosphere
  • Volume estimation for each level (data size per month)
  • Bit configuration
  • Definition of solar/satellite zenith, azimuth angle
  • GLI channel configuration (figure available on website)

Q: (Dr. Duong) Is there any fixed standard for byte data?
A: (Mr. Inoue) I think it is fixed. The byte order is a universal data format from HDF library so if you have that library you don't need to worry about byte order.

Q: (Dr. Honda) How long will you hold onto temporal files?
A: (Mr. Inoue) After we confirm the validity of the data we will delete it as soon as possible, but it will be held for 3 - 6 months for validation.

Q: (Dr. Asanuma) When you create level 3 product, will you continue to hold levels 1 and 2 on database? (In case of OCTS there is no intermediate product and we have to refer to original data, which takes time)
A: (Mr. Inoue) Yes, we will keep intermediate product and you don't have to go back to level1.

Back to index

• ADEOS-II Science Project Manager, before starting explanation, Mr Ito reported on the launch schedule status. Space Activities Commission is to review and make a decision soon (present launch schedule is for Japanese Fiscal Year 2001). He then went on to present the ADEOS-II mission operation planning, relating to how data will be acquired.
  • AMSR, GLI and SeaWinds and operation parameters
  • operation requirements of GLI for atmosphere, land, ocean and cryosphere
  • observation request flow
  • mission operations overview
  • mentioned EOC design and capability presentations
  • tilt operation
  • facility scaling design
  • asked that attendees register their names and details to get further information.
  • ADEOS-II GLI Data Distribution Policy (investigation)

Q: (Dr. Makoto Kuji ) I am not a PI of ADEOS, and want to obtain OCTS data - is it possible to get it freely?
A: (Dr. Vu Saito) In principle, data order should be made through EORC/Order Desk.
Q: (Dr. Kuji) Do you have the organization in JAXA to accept such orders?
A: (Mr. Ito) ADEOS was used to establish physical parameters and JAXA has learnt from experience with ADEOS and we need a new structure/mechanism. The EOC system has been updated to be more user-friendly. We have a new function for PIs during validation, a concept for the better handling of data to research communities. So in principle, we can provide such data if it can be defined within ADEOS-II project data. However, our capabilities are not currently able to deal with a large number of requests.

Dr. Teruyuki Nakajima's Comment: There are 2 issues - some people want to order OCTS data spot by spot and this can easily be released by EORC. However, a discussion on procedure is needed for those who want a large volume of data. Also, GLI project may analyze OCTS data (due to delay of launch) to test GLI data system (OCTS or MODIS) and PIs of GLI can automatically use that data.

Q: (Dr. Arai) Do we need a password to access data through the WWW? And how soon will temporal files be deleted?
A: (Mr. Ito) There are 2 systems, EOC for general users (one year after launch), where data is defined in technical documents, EORC is for PIs and priority is on Cal/Val for algorithms and products. The website will be limited by resources and is not yet defined in detail - we would like to ask for everyone's co-operation in doing that. Existing planning may or may not be OK, please comment.
Q: (Dr. Arai) Is Level-1 data available?
A: (Mr. Ito) Not yet completely decided but EORC is priority for PIs. EOC service will be provided for general user one year after launch.

Q: (Dr. Honda) How do PIs send orders to you? Via email?
A: (Mr. Ito): It was agreed to send orders by either fax or email.

Q: (Dr. Honda) How do PIs recognize whether their request has been accepted or not?
A: (Mr. Ito) There are 2 or 3 steps to deal with requests and final decision will be posted on the web site for access by all.

Q: (Dr. Arai) What about the availability of Level-1A products? These are necessary for calibration scientists.
A: (Mr. Ito) How much and when is not yet decided, but we are ready to provide to calibration PIs.

Back to index

• Dr. Honda presented his studies on the topic above focusing on the issues of resource data (MODIS) and purposes of simulated GLI data.
  • survey of the demand quantity of reformat GLI data
  • presented questions concerning about area, time, timing of data acquisition, acquisition interval, purpose, medium
  • processing process
  • request of the answer: please submit the demand of reformat GLI data to the leader of each area by January 5, 2000.
  • reformat data generation: asked for candidates for this project from each group
• Dr. Honda moved to the discussion session

(Dr. Teruyuki Nakajima) There are various issues - firstly how to get ancillary data, it would be a good strategy to get it before GLI launch, to encourage the entire system. Not a data service, but data coming into EOC, therefore an issue for EORC. i.e. a research system needs to be established
(Dr. Honda) You mean from data in EORC?
(Dr. Teruyuki Nakajima) This will be a huge effort, so JAXA has to do on behalf of PIs. Also, there is a cost issue for changes in GLI system, e.g. for different wavelengths or response function, which are now not included in GLI system development. There is a need to prepare linkup data. A third issue is that there needs to be a capability for switching off some channels if required.
(Dr. Honda) Mr. Inoue of Fujitsu has a plan to do this I think...
(Dr. Teruyuki Nakajima) I think this is an issue for JAXA, not for Fujitsu
(Mr. Ito) We have been negotiating with JMA and others, for ancillary data provision, and have to set up a new schedule for scientific activities etc following rescheduling of launch. We will take into account your recommendations, but can't give details or firm dates now.
(Dr. Honda) The Data Management WG will discuss this and we want to do our best to incorporate Dr. Teruyuki Nakajima's suggestions. I think we should discuss the ancillary data problem with the Cal/Val WG also, and maybe set up a joint WG meeting early next year.
(Dr. Teruyuki Nakajima) I think that GLI team leaders could discuss this and a joint WG may not be necessary.
(Dr. Honda) We have a Data Management WG meeting early next year anyway, please put this in your diary.
(Dr. Teruyuki Nakajima) Could you add OCTS to this, and OCTS and MODIS are the only existing datasets?
(Dr. Honda) That is hard work - we will discuss OCTS data. We are now trying to archive it, but that alone has taken more than 100 days
(Dr. Teruyuki Nakajima) Ongoing projects, but in different format, and we need to coordinate with MODIS project or visiting scientists can't touch it. Need to make these projects visible to PIs
(Dr. Honda) I will add that point to discussion point of WG. There is huge data, so distribution is a headache, so please do not request huge areas - just CD-ROM size is easy to distribute (global dataset is 30 Gigabyte data) If you have any discussion points, please tell us, I would like to open for discussion.
(Dr. Duong) I think Dr. Teruyuki Nakajima's ideas for MODIS data and conversion to actual GLI data is good, we need some kind of conversion
(Dr. Honda) We want to consider this, but we have limited facilities, so are not sure if we can do that. If not so many orders, and if you accept not complete conversion, it may be possible. I don't have confidence that we can make a complete conversion.
(Dr. Duong) Reformat MODIS data to GLI format, and if not capacity for conversion, please provide a lookup table.
(Dr. Honda) Of course we can provide a lookup table.

• Dr Honda finished the discussion session with a request for attendees to fill out a circulated paper with name and contact details for automatic data distribution, thanked everyone for their attendance and asked that any requests for reformatted GLI data be made to team leaders. Please do not make unnecessary requests.