CASU meeting Wednesday 18th January 11:30-13:00 CASU meeting area.
Present: MJI, JRL, AKY, EGS, SC, STH, MNO, JPE
Apologies: NAW, RGM, MR
Agenda:
1. Actions from last meeting
2. Comments on WFAU minutes
3. Hardware update
4. Meetings and telecons
5. Data archives update
6. WFCAM update
7. VISTA update
8. VST update
9. Other processing
10. AOB
This was MNO's last meeting and he was duly thanked for his efforts as part
of the VISTA processing team. His replacement, Joe Findlay (JRF), is due to
start on the 1st Feb.
1. Actions from last meeting
-----------------------------
JRL - still continues to investigate "rolling" master linearity corrections
and will include "average" channels too - honest ! <<<<
SC - completed one Blade's Ubuntu/Debian upgrade - rest ongoing <<<<
STH - progress with VST calibration papers ongoing <<<<
MJI - the reprocessed UDS H-band data was incorporated
JRL - still needs to check if the info in darks.dqc could have detected the
problem VISTA darks on 20110913 <<<<
SC - added apm45 to the web monitoring but its either dead or .......... <<<<
MJI - spec'd, quoted, ordered and installed apm46
SC
SC - apm29 is once more an expensive thermometer
MJI - a VST foibles list has been compiled (see previous minutes and
technical pages)
MJI - acquired learned summary on the future of JCMT submm astronomy and
the plans for continuing UKIRT operations
JRL - generated tape backups of raw VST data
EGS
JRL - figured out why tiles were not generated in two VMC OBs
MNO - generated the burgering missing tiles
EGS - fixed faux pas in VISTA P88 name header on web pages
STH - duly organised the CASU weight-watchers Christmas party
2. Comments on WFAU minutes
----------------------------
...... sin commentarios ...... muchos minutos
3. Hardware update
-------------------
As noted previously, the other half of the VST processing system, aka apm46,
was installed and will shortly host the raw data half of the VST data storage
and processing system. Processed VST data will remain on the apm45 RAID.
Modest hell broke loose with the IPV6 changes to the University network
causing "interesting" problems that took a while to track down and sort
out. The most painful of which involved some corruption of an NFS-mounted
RAID array. The next hurdle will be when the old name servers are finally
turned off ~end of March.
4. Meetings and telecons
-------------------------
VISTA IOT 23rd November: there is a planned VISTA intervention from 10th-18th
December for Hexapod and closed cycle cooler maintenance; the i-band
predictions from 2MASS, and the more reliable nature of the 2MASS catalogue,
are working well; the state of the VIRCAM entrance window (remember the
10% improvement) will be monitored more regularly; the next meeting will
be sometime around May 2012.
VMC skypecon 24th November: minor grumbles about missing tiles for 2 OBs
(reported earlier) and missing OBs (ESO aborted the exposures, hence no files)
otherwise nothing other than science.
MJI went to the VST ATLAS meeting in Durham on 5th December and regaled the
attendees, including 4 from WFAU, with a list of VST foibles, as well as
an overview of the CASU processing. He took with him an overview of what
VST data for ~200 band-merged ATLAS fields currently yields. This, plus
the fact that CASU are processing all the VST public survey data, generated
some interesting discussion about who wants to do what and with whom.
MJI did not attend the future of UK Submillimetre Astronomy Workshop at IfA
on 12th-13th December 2011 but did get to talk with Gary Davies about
the future of JCMT and UKIRT, and hence continuing WFCAM survey options.
Apparently SCUBA-2 is working well on JCMT which is good news for continuing
access to UKIRT, and the Korean interest in WFCAM time is extending to
include another Institution. WFCAM UHS survey observations should start in
June when the UKIDSS surveys (bar UDS) complete towards the end of May.
JPE went to the VMC meeting in Bologna, January 9th-11th, and reported that
lots of interesting science was ongoing including attempts by the VMC PI
to measure accurate proper motions of Cloud populations with respect to an
extragalactic quasar-based reference frame. Although not that many quasars
are published in that region there is apparently a good possibility of
selecting out quasar candidates from the VMC photometry.
Upcoming meetings include: the Third VVV Science Meeting, 22nd-24th March
2012, Viña del Mar, Chile and "Calibration & Standardization of Large Surveys
& Missions in Astronomy & Astrophysics" conference to be held April 16th-19th,
2012, at Fermilab in Batavia, Illinois, USA. NAW will going to the former
to find out what is going on, particularly with variables, in the VVV survey
and will report back. STH expressed an interest in going to the Fermilab
meeting, but is already committed elsewhere - a pity given the amount of work
CASU have done in calibrating large area surveys.
5. Data archives update
------------------------
Everything is up to date, and no transfer problems in spite of IPV6 changes.
6. WFCAM update
----------------
Ingest, checking and processing all going fine. Data to end of December
have been checked and flagged for copying.
Our bluff about processing 3 camera data was finally called and due to
insatiable user demand two nights with camera#3 data completely missing
(8th, 9th January) were processed using a cunning workaround. Quite what
downstream consequences will ensue remains to be seen. Since these nights
were used for non-survey data projects we remain ever hopeful.
Various requests for fast flat file access were received and dealt with
(including data from the 3 camera nights).
Mike Read found an interesting feature of a much unloved (and hardly used)
aperture correction (#2), in that it occassionally defaults to the same
value as the correction for aper3 when the pipeline is having trouble
computing it reliably. This turned out to be a consequence of a recent
improvement (sic!) for computing more reliable aperture corrections in
crowded fields !! Unfortunately this defaulting is in the generic classifier
code used for WFCAM, VISTA and VST and occurs often enough that it needs
fixing - even though its impact on the science is likely to be negligibly
small.
All (2005-2011) the processed WFCAM data was fixed over the Chistmas period
as was the VST data. VISTA will be more subtle to fix due to the grouting
and the best approach for this is being much pondered.
After a request from Eric Gaidos, the illumination correction tables,
available through the web pages, were updated.
7. VISTA update
----------------
Receipt of data, ingest, checking and processing has yet again been going
smoothly. Raw data up to 29th December has arrived and data up to the end
of November have been processed through Phase I of the pipeline.
October data was released to PIs and WFAU in early December. This inlcuded
some non-public survey Stripe 82 PI data.
JRL and EGS will finish pondering and fix the apcor2 problem for VISTA
data before anybody else notices. <<<<
No news is probably good news from the point of view of the next installment
of ESO Phase 3 deliverables. We still await the product format standard and
Phase 3 instructions for catalogue submission.
MJI recalled that Valentin had reported a strange problem with IRACE
reporting the wrong values of DIT and NDIT aka essentially DITxNDIT was
wrong, with the only obvious symptom being "unusual" sky levels. EGS will
investigate further. <<<<
EGS copied another bunch of VVV data up to end of September to USB disks
(~4TB) and shipped them off to Maren while she was over in Germany.
JPE showed some results from Joe Findlay's thesis for predicting the size
and placement of ghost images from bright stars in VISTA data. This could
be useful for flagging regions of suspect photometry.
8. VST update
--------------
We discovered at the end of November that the ESO raw data archive had been
mistakenly hiding the Public Survey data taken from the 15th October from the
outside world. Consequently what we thought was completed processing to end
of November turned out not to be and had to be done all over again - which
it has been.
Since the hiatus however, internet transfers of VST data have been running
smoothly, hopefully with no large missing chunks of data this time. Raw data
up to the beginning of January has been copied over and ingested. apm46 will
now host the raw data repository and apm45 will remain the processed data
store. As of mid-November the ESO archive VST raw data is now Rice-compressed
(.fz) which makes life a bit easier.
Meanwhile the VST web pages, processing and survey progress and PI access are
all up and running thanks to EGS.
JPE noted that the SADT restriction on only using cardinal directions for
VST observations has now been removed. Murphy's Law therefore predicts
arbitrary PA orientations will be wending their way here shortly.
We are still working on different ways of characterising the scattered
light more quantitatively whilst trying to work the corrections into a
similar pipelining scheme as that used for WFCAM and VISTA. Meanwhile
we have at least computed the required corrections for the August,September
period using a cunning mixture of APASS and 2MASS standards on the many
ATLAS pointings taken during this period (see the VST technical web pages
for some stunning plots).
We will next investigate using available SDSS photometry in the SA fields as
a supplement to this approach.
We have also done a few test processings of one of the VST GTO programmes
partly to assess issues with mosaicing the whole array for projects that
are using tiles as their unit deliverables.
Some extra Chilean time to extend the ATLAS U-band exposure times x2 has
been awarded. Malheureusement, the obvious expedient of just extending the
OB's to include 4x60s exposures is way too simple and we will probably end
up with completely separate lots of U-band exposures for later combination !
The VPHAS observations started in January. Constraints on unguided OB
contents, and the odd observing pattern suggest processing these optimally
will not be straightforward either. <<<<
Dusting off the crystal ball, we note that the "the first Phase 3 submission
for VST PS will be one year **after** the official start of operation i.e.
October 15, 2012, and will include the data products for the OBs executed in
the first semester of VST operations plus the dry runs." - Magda
9. Other processing - HAWKI, INT WFC, MegaCam, Subaru, VST, CTIO
-------------------- - UVEX, IPHAS, and Kepler processing
No news yet again.
10. AOB/news
------------
Lots of CASUhelps and even replys but no prizes this month - so rollover to
next meeting.
The CASU standard grant was submitted end of November into a deafening pool
of silence.
Continuing actions
------------------
JRL - investigate benefits of a "rolling" master linearity correction or
even an average channel one
SC - ongo the rest of the Ubuntu/Debian upgrade for Blades
STH - progress evermore the VST calibration papers writeups
JRL - check if the info in darks.dqc could have detected the problem darks
New actions
-----------
SC - resuscitate apm45's communications with web network access monitoring
JRL - figure out a simple way to fix the apcor2 problem for VISTA data and
EGS make it so
EGS - trawl the VISTA DCQ database for peculiar sky brightnesses
MJI - contact the VPHAS team and discuss processing options