I have been at the first "co-location" meeting for the Climate Change Initiative, where ESA gather representatives of all the CCI projects (11, comprising 10 variable teams and the climate modelling users group), and thereafter at the EUMETSAT conference in Cordoba. This post notes a few useful conversations for the SST CCI team.
1. Rainer Hollman (CM SAF and Cloud CCI). We discussed Fundamental Climate Data Records from AVHRR Global Area Coverage. A project the CM SAF is involved in (SCOPE CM) will create AVHRR GAC FCDR with some homogenisation applied, plus the CM SAF cloud products, by the end of this year. Both because of local downloading, and the homogeneous cloud mask, this may be a practical alternative source for AVHRR GAC.
2. Carol-Anne Clayson (Florida State U). Carol-Anne uses the SSMI records to retrieve ocean-atmosphere fluxes, by determining inputs to bulk flux parameterisations, including a surface temperature term. I encouraged Carol-Anne to take part in the Round Robin exercise, by adding SSMI to the multi-sensor matches. This would be a practical way of making comparisons with infra-red based SSTs.
3. Don Grainger (Oxford and Cloud CCI). One of Don's students is extending their aerosol retrieval to include infra-red channels. SST then gets added to the state vector. Again, I encouraged Don to get involved in the Round Robin exercise, where the SST obtained by their scheme can be compared with our SSTs.
4. Steve Ackermann (Wisconsin). In their MODIS cloud work, they also take and apply the official MODIS SST algorithm. I explained that MODIS could be added to the project's MMD to create some useful retrieval and cloud mask inter-comparisons. Steve is keen on this: they have systems for pulling out MODIS data for a given lat and lon and time already, so overhead would be modest.
Thursday 23 September 2010
Thursday 16 September 2010
Met Office Telecon 2
10 September 2010, by skype/phone, Nick Rayner, Simon Good, Chris
Merchant
Topics: WP 10130 on User Requirements / ESA Collocation meeting
SG gave a progress report on the User Requirements questionnaire. 767
people have been emailed so far, with about ~100 bounces. Have been
asked to respond by 17 September. A further batch will be added to
include a greater diversity of countries. 19 responses by 9 September,
the nature of which SG summarized. Importantly, 5 our of 19 have
agreed to the follow-up discussion session, 12 to review the URD and 8
to receive test data -- looks a promising level of engagement.
NR has piloted interactive discussions. CM commented the notes of the
discussions so far make very interesting reading, and suggesting
prompting users about their requirements for uncertainty information
if it didn't naturally arise.
Actions/decisions:
1. NR/SG will supply CM a slide or two schematically describing the
questionnaire process for CM's presentation at the first collocation
meeting next week.
DONM: Fri 1 Oct 10.00
Discussions at University of Leicester
Discussions at UoL on 3 September
Present: Chris Merchant, Gary Corlett and John Remedios
Project month / Action
2 & 3 / Specify match-up data sets and write experimental code for
collating multi-sensor matchups, do ToC for PVP
4 / Circulate specification and visit Brockmann Consult
5 / Write Product Validation Plan (while BC start prototype code for
MMD)
6 / Define MMD tools for extraction etc
7 / verification of MMD (while BC start to code tools)
Aim is for a working MMD by end January 2011.
Gary will be involved in Envisat post-orbit drop validation from 2 Nov
for ~few weeks.
We had a preliminary discussion about MMD tools. Elements we foresee
are:
By end January 2011:
- extraction tools, including metadata to set up extraction options
- NWP matcher and extracter, including providing files in correct
formats to run RTTOV-10 and diurnal models
- tools to add fields (e.g. model outputs) to MMD that will then be
available to other users
By RRDP:
- RRDP input tool (to accept new MDs and alternate algorithm results)
For system prototype:
- MD generator
Thinking about the nature of the matches, there may be a sparsity
problem for comparisons across the full range of sensors. So, in
addition to the previously agreed hierarchy (where there is always an
ATSR observation) we agreed that triplets of Metop+SEVIRI+PMW should
also be collected.
The MMD will cover dates as follows:
ATSR/AVHRR: 1991 - 2010
ATSR/AVHRR/Metop/SEVIRI/AMSRE/TMI: 2003 - 2010
As part of the Round Robin exercise, we will invite addition of MDs
from other sensors in a standard format, to be associated into the MMD
using the same tools. Priorities: MODIS, GOES, MTSAT. The mechanism
for submitting results of alternate algorithms to the Algorithm
Selection exercise will be an MD in the same standard format.
The Data Access Requirements Document will be drafted during October,
led by Karen Veal. Final version requires review of URD and PSD. DARD
is likely to comprise a page describing data, giving URL/info for
third parties to obtain the data. For MMD data, which we will
redistribute (in RRDP/CDRP), we need to seek a license to do so, and
the DARD will have notes on status regarding this (and whether the
next party is licensed to redistribute further). It would be helpful
if ESA had a standard letter requesting data for all CCI projects.
ACTION: GC to ensure this is proposed during 1st CCI collocation
meeting.
CIRIMS radiometry data may be useful to include in MD. ACTION: CM to
discuss CIRIMS data with And Jessup at Seattle SST meeting.
Actions/decisions:
1: GC to ensure ESA letter requesting data is proposed during 1st CCI
collocation meeting.
2: CM to discuss CIRIMS data with And Jessup at Seattle SST meeting.
3. (Not exactly CCI) CM to provide QWG with 3 or 4 pages on how new
coefficients were generated.
Present: Chris Merchant, Gary Corlett and John Remedios
Project month / Action
2 & 3 / Specify match-up data sets and write experimental code for
collating multi-sensor matchups, do ToC for PVP
4 / Circulate specification and visit Brockmann Consult
5 / Write Product Validation Plan (while BC start prototype code for
MMD)
6 / Define MMD tools for extraction etc
7 / verification of MMD (while BC start to code tools)
Aim is for a working MMD by end January 2011.
Gary will be involved in Envisat post-orbit drop validation from 2 Nov
for ~few weeks.
We had a preliminary discussion about MMD tools. Elements we foresee
are:
By end January 2011:
- extraction tools, including metadata to set up extraction options
- NWP matcher and extracter, including providing files in correct
formats to run RTTOV-10 and diurnal models
- tools to add fields (e.g. model outputs) to MMD that will then be
available to other users
By RRDP:
- RRDP input tool (to accept new MDs and alternate algorithm results)
For system prototype:
- MD generator
Thinking about the nature of the matches, there may be a sparsity
problem for comparisons across the full range of sensors. So, in
addition to the previously agreed hierarchy (where there is always an
ATSR observation) we agreed that triplets of Metop+SEVIRI+PMW should
also be collected.
The MMD will cover dates as follows:
ATSR/AVHRR: 1991 - 2010
ATSR/AVHRR/Metop/SEVIRI/AMSRE/TMI: 2003 - 2010
As part of the Round Robin exercise, we will invite addition of MDs
from other sensors in a standard format, to be associated into the MMD
using the same tools. Priorities: MODIS, GOES, MTSAT. The mechanism
for submitting results of alternate algorithms to the Algorithm
Selection exercise will be an MD in the same standard format.
The Data Access Requirements Document will be drafted during October,
led by Karen Veal. Final version requires review of URD and PSD. DARD
is likely to comprise a page describing data, giving URL/info for
third parties to obtain the data. For MMD data, which we will
redistribute (in RRDP/CDRP), we need to seek a license to do so, and
the DARD will have notes on status regarding this (and whether the
next party is licensed to redistribute further). It would be helpful
if ESA had a standard letter requesting data for all CCI projects.
ACTION: GC to ensure this is proposed during 1st CCI collocation
meeting.
CIRIMS radiometry data may be useful to include in MD. ACTION: CM to
discuss CIRIMS data with And Jessup at Seattle SST meeting.
Actions/decisions:
1: GC to ensure ESA letter requesting data is proposed during 1st CCI
collocation meeting.
2: CM to discuss CIRIMS data with And Jessup at Seattle SST meeting.
3. (Not exactly CCI) CM to provide QWG with 3 or 4 pages on how new
coefficients were generated.
Friday 10 September 2010
SST CCI Calendars
The calendar to which SST CCI meetings and telecons can be added by anyone in the project (who has associated their e-mail address to a Google account) is:
http://www.google.com/calendar/ical/v6f4dq21fq5c1dill1eao8u46o%40group.calendar.google.com/public/basic.ics
A calendar with the planned durations of work packages within the project is
http://www.google.com/calendar/ical/j22otpbcof0vl4bngcv8l96jvs%40group.calendar.google.com/public/basic.ics
http://www.google.com/calendar/ical/v6f4dq21fq5c1dill1eao8u46o%40group.calendar.google.com/public/basic.ics
A calendar with the planned durations of work packages within the project is
http://www.google.com/calendar/ical/j22otpbcof0vl4bngcv8l96jvs%40group.calendar.google.com/public/basic.ics
Met Office Telecon 1
The main purpose of this experimental blog is to allow the SST CCI team to have access to discussions and decisions they were not directly involved in. So ... here are notes from a telecon I had with colleagues in the Hadley Centre.
13 Aug 2010 by skype, Chris Merchant, Nick Rayner, Simon Good
Main focus: WP10130
NR explained that SG has identified ~1000 users on the basis of citations of SST CDR papers over last 5 to 10 years. >400 are from US, so they may be subselected to avoid unbalance. Will send a named e-mail to invite to take online survey. Briefing material will give user a choice between an extremely brief summary and something with some scientific detail. Questionnaire is being developed and likely distributed within project for comment around 20th Aug.
Regarding interactive sessions, SG explained that Web-X appears to be ideal. NR will pilot the project presentation for the interactive sessions at the technical kick off (TKO) at Edinburgh. Sessions will be trialled within Hadley Centre around end of August.
CM confirmed that WP20123 will be initiated by discussions with Gary Corlett at a side meeting in Edinburgh at TKO.
Actions/decisions:
1. NR/SG to send CM User Requirements Document at outline stage for review.
2. CM to distribute links to project Google calendars.
3. CM to try to get his calendar system to send out invitation to next telecon.
DONM: telecon agreed for 10 am 10 Sep by skype.
13 Aug 2010 by skype, Chris Merchant, Nick Rayner, Simon Good
Main focus: WP10130
NR explained that SG has identified ~1000 users on the basis of citations of SST CDR papers over last 5 to 10 years. >400 are from US, so they may be subselected to avoid unbalance. Will send a named e-mail to invite to take online survey. Briefing material will give user a choice between an extremely brief summary and something with some scientific detail. Questionnaire is being developed and likely distributed within project for comment around 20th Aug.
Regarding interactive sessions, SG explained that Web-X appears to be ideal. NR will pilot the project presentation for the interactive sessions at the technical kick off (TKO) at Edinburgh. Sessions will be trialled within Hadley Centre around end of August.
CM confirmed that WP20123 will be initiated by discussions with Gary Corlett at a side meeting in Edinburgh at TKO.
Actions/decisions:
1. NR/SG to send CM User Requirements Document at outline stage for review.
2. CM to distribute links to project Google calendars.
3. CM to try to get his calendar system to send out invitation to next telecon.
DONM: telecon agreed for 10 am 10 Sep by skype.
An experiment in team communication
This blog is an experiment in team communication. I'm leading a European Space Agency project called the Sea Surface Temperature Climate Change Initiative (SST_cci) which involves the collaboration of eight organisations round Europe (see our main web site). The idea is, briefly, to get better information about how the surface temperature of the oceans has evolved over the last few decades, using satellite data.
I'm going to be having lots of meetings and discussions as I run this project, and I am wondering whether blogging would be a good way to keep the team in touch with the ideas and decisions that arise from all those conversations. There is also a principle of transparency to the project, so having these conversations open to all to view also fits in with that -- not that I expect too many people outside of the team will want to follow the blog, since it will all be rather technical.
Anyway, thought I'd try it and see.
I'm going to be having lots of meetings and discussions as I run this project, and I am wondering whether blogging would be a good way to keep the team in touch with the ideas and decisions that arise from all those conversations. There is also a principle of transparency to the project, so having these conversations open to all to view also fits in with that -- not that I expect too many people outside of the team will want to follow the blog, since it will all be rather technical.
Anyway, thought I'd try it and see.
Subscribe to:
Posts (Atom)