File contents

6th August 2008  APM comfy chairs  11:30-13:05

Present:          RWA, MJI, JRL, PSB, MR, DWE, EGS, NAW
Apologies:        STH, RGM

Agenda
------

1.  Actions from last meeting
2.  Comments on WFAU minutes
3.  Recent meetings
4.  Data archives update
5.  Optical/NIR processing
6.  WFCAM update
7.  VISTA update
8.  AOB


Minutes
-------

1.  Actions from last meeting

STH  has not finished checking that the Plone astrometry technical pages 
     are correct - ongoing                                                 <<<<

MJI  has tried to acquire the transmission curves for the VIRCAM N118 
     filter but life is too short ......

JRL  has revived the changes log web page for the WFCAM pipeline. This 
     can be found on the internal pages at
     http://casu.ast.cam.ac.uk/surveys-projects/wfcam/software-changes/
     
MJI  has found out more regarding the VIRCAM commissioning plans including
     updated schedules and work plan
     
STH  has not included the graphic showing the size of the photometric radial 
     distortion correction in the Plone technical pages - ongoing          <<<<
     
MJI  has checked the connections on an LCD TV to see if displaying from a
     laptop works ok. The external width of a 37" TV is about 40", which 
     is the largest that will fit comfortably on the tables planned for the
     meeting area.
     
MJI  has tried to order some suitable tables for the meeting area, but has
     encountered the University Department of Bureaucracy - ongoing        <<<<

MJI  has obtained the VISTA schedule (see above)

STH  have investigated photometry in regions of high nebulosity in Eta-Carina
MJI  data from HawkI and obtained some impressive results

JRL  has not produced a web page of HawkI publicity for CASU - ongoing     <<<<

MJI  has updated the pipeline toolkit on the Solaris cluster

MJI  has investigated ways of demonstrating photometric corrections for flat 
     file users and has decided to provide a simple programme that starts 
     from a catalogue FITS file and outputs fully calibrated and corrected 
     magnitudes in an ASCII file.

JRL  deferred deciding on a flipping jitter telecon with JAC for now since
     several other items have since cropped up (see later)

JRL  has written a script to flag summit-rejected MSBs during raw WFCAM data 
     ingestion.  MR has yet to include this in the ingestion script.       <<<<

MJI  has implemented trial MSBTid logs for 08A, this will be used in anger
     for 08B data.


2.  Comments on WFAU minutes

Not many comments other than we noted the usual saga with the air 
conditioning - good job its cold in Edinburgh; and the announcement of the
release of DR4, sans GPS and UDS, on 4th July.


3.  Recent and upcoming meetings

MJI had a day out in London on July 8th attending the meeting where STFC 
presented the results of the Programmatic Review.  

In addition to EGS and NAW, MR and JRL will attend the Cambridge VOTC 
workshop in early September.                  

JRL pointed out the upcoming ADASS late autumn meeting in Quebec, ALL to 
consider submission of abstracts deadline is August 15.                    <<<<

MJI also noted the planned Science from UKIDSS-II meeting at the RAS on
15-17 December.


4.  Data archives update

The remaining photographic plates left at the IoA are being rationalised
into one office prior to being removed to the off-site storage facility.

With the rearrangement and installation of more computer kit in the computer
room, we have a DVD/CD-ROM cabinet too many to comfortably fit in the space.
RWA will remove the JKT DVD/CD-ROMs from the computer room and free up one
cabinet which will then be removed to the IoA annex. The JKT data is 
already on disk and backed up on LTO tape.                                 <<<<

MJI asked if the internet transfers were included in the ING archive backups.
Although these are backed up in monthly batches as they arrive it was felt
they should now be included in the general ING archive backup.
EGS will implement a backup scheme to include the internet transferred 
INT/WHT data on LTO4s now that we have two independent LTO4 library units  <<<<

PSB already found a suitable supplier of LTO4 tapes and will order some
more now he has got his order book back.                                   <<<<

JRL reported that he is in the middle of revamping the UKIRT archive to 
fix a few long standing problems with headers and legality of PHUs.

AAT data continues to arrive in roughly monthly chunks where the transfer
is initiated manually, but its such a small amount of data that its not
worth doing any other way at this stage.

All of the WFCAM raw data is now back online after the fault with the 
recalcitrant RAID unit was finally traced and fixed - its SP was F'd.

Raw data tape backups for all of 08A have been made and are sitting in 
a modest pile under MJI's desk.  LTO2's only store ~5 nights of compressed
data per tape (hence the large pile) so we will switch to LTO4's for 08B 
where we expect to get ~20 nights on a tape.  Although 08A's tapes make
a great footstool they need removing to the out-of-building store.         <<<< 


5.  Optical/NIR processing

JRL, STH and MJI had a meeting with a posse of Exeter PhD students and
postdocs lead by Mark McCaughrean to discuss the CASU-processed HAWK-I
JHKs data on Eta Carina.  In addition to the beautiful 2D images, and
exquisite photometry in nebulous regions we also managed to improve the
astrometry systematics to < 20 mas over the whole array using our
usual ZPN WCS.  JRL is investigating further improvements to the 2D
processing and when that is finalised we will switch to generating
catagloues from mosaic tiles of the 4 dither stacked detectors per 
pointing, rather than from individual detectors, which should help with 
the non-uniform exposure maps caused by the large dither patterns used.

NAW, EGS and MJI attended an EGAPS meeting in Warwick on 24th June. EGS 
and MJI reported there that we have finished processing all the IPHAS data 
up to end of 07B and are still awaiting final decisions on the U-band and 
HeI-band filter offsets for the UVEX equivalent.


6.  WFCAM update

MJI reported on an email received from AJA about the possibility of 
overcoating the WFCAM secondary mirror with gold rather than silver.
If this goes ahead it shouldn't have much impact on the throughput
in any of the bands at the ~1% level, but it would make sense to double
check the colour equations.

The idea of a telecon with JAC before WFCAM 08B commences was discussed.  
There are several issues that could profitably be dealt with including:
a resume of the jitter flipping strategy, offset skies and data 
self-description; pipelines running at JAC; technical updates; fast
flat file access and so on.

MJI to sort out topics and a suitable time/date around the beginning 
of September for a telecon with JAC. Attendees from our side likely to be 
MJI, STH, JRL and MR, no doubt at Jim's place.  In return Jim will 
provide the refreshment.                                                   <<<<

As reported earlier the reluctant raw data disk was finally fixed, and
in the same burst of creative energy, this rarely used disk was switched 
over from its own fibre channel to a shared channel fibre loop, thereby 
freeing up a fibre channel for a recently acquired 14 TByte RAID system.

The GPS astrometry glitches affecting ~200 files over the 05A-07B period 
noted in the previous minutes were fixed and have been re-transferred
by WFAU.

There have been a one or two examples of 08A data (eg. Orion for GCS) where 
a strange sky subtraction decision was made by the pipeline.  These were
fixed manually after they were reported to us by NCH.  MJI and JRL discussed
a couple of minor changes to the special case object masking sky creation 
strategy to prevent this happening in the future.  JRL will implement.     <<<<

Phil Lucas is going to supply a list of GPS data within regions of high
nebulosity which would benefit from recataloguing using the recently 
developed nebulosity filtering option.  This methodology is undergoing
final trials and will eventually be incorporated as an option within the 
cataloguing software.

MJI and MR have been exploring some software updates in the general toolkit 
software that allow things like internal dribbling of interleaved frames 
to minimise the "hedgehog" PSF problem; use of frame magnitude zero-points
to monitor and correct for variable throughput when mosaicing; and different 
interpolation options for stacking and mosaicing.  The latter, in particular, 
will be needed for VISTA image processing.


7.  VISTA update

The big news is the VISTA first on-sky photons on VIRCAM.  The first FITS
file with sky images on was received at CASU on 2nd July via network from
Edinburgh off a disk hand-carried from Paranal.  Since then a steady stream
of test data and on-sky data has been received, some directly in small 
doses from Paranal, but the majority via hand-carried disks.

As expected there were the usual teething problems with WCS's and various
telescope/camera parameters and observing templates, but these have quickly 
been sorted.  Sufficiently so, such that recent frames have included enough 
calibrations and the odd on-sky dither sequence to enable assorted 
preliminary science checks to be made.

As expected the WCS is well described by the ZPN projection and our
measured distortion coefficient is within about +/-1 of the predicted value 
of 42.  Although we are awaiting remaining optical tweaks before doing  
further analyses, after calibrating some dither stacks against 2MASS the 
systematic astrometric residuals are already at the 50 mas level over the 
whole array.

A dither sequence in the Eta Carina region has sufficiently good image
quality (~1 arcsec) and sufficient bright stars being shifted around
the arrays, to enable a thorough search for persistence and cross-talk
artefacts.  So far, in spite of our best efforts, we have been unable to
find any trace of cross-talk and cannot see any significant trace of 
persistence at a level that would require correcting. Although early days 
this is extremely promising and makes the 2D image processing aspects more
straightforward than they might otherwise have been.

Having real on-sky data is helping enormously with the DRL development and
fine tuning but we are still desperate for more calibration data to really
nail the properties of the camera, especially linearity sequences and 
twilight flats.

In parallel with this we have also been acquiring and installing some 
VISTA data processing and storage kit.  This had already involved assorted 
changes to the air conditioning placement, rearrangement of what was the 
storage room and installation of yet more mains supply - this time in
the form of 9 (8+1 spare) 16-amp commando sockets.  The latter are primarily
needed for the recently installed Sun 20-core Blade 6000 system, which will 
form the backbone of the VISTA data processing here, and for future upgrades 
for the GAIA data processing kit.  In addition to this hardware, we have
also procured and installed a 48TB Sun X4500 disk system and another
LTO4 tape library unit, both installed in the VISTA rack together with the 
Blade 6000.

Other recent upgrades include the installation of 10 G-bit switches in 
both computer rooms and, as reported before, both rooms are now on the IoA 
10 G-bit backbone.  


8.  AOB

NAW noted that VST science commissioning is scheduled to start mid-2009 
with the public science surveys likely to start by the end of that year.

EGS raised the issue of releasing the CASU cataloguing software and the
rest of the pipeline toolkit, which among other things would raise the 
profile of the group. MJI was worried, as always, about the amount of 
implicit extra support and maintenance that this would entail but he and
EGS will look further into this and present a cunning plan at the next 
meeting.                                                                   <<<<

MJI and NAW summarised the current state of play of the CASU rolling grant
saga and also our involvement in the PLATO and PrepSKA applications.


Continuing actions
------------------

STH  check that Plone astrometry technical pages are correct

STH  include graphic showing the size of the photometric radial distortion
     correction in the Plone technical pages.                              

MJI  continue ordering some suitable tables for the meeting area

JRL  produce web page of HawkI publicity for CASU


New actions
-----------

MR   install JRL's script to flag summit-rejected MSBs during raw WFCAM 
     data ingestion

ALL  consider submission of abstracts for ADASS 2008 (deadline August 15)

RWA  remove JKT CD ROMs from computer room and rationalise storage to free
     up a cabinet

EGS  implement backup scheme for internet transferred INT/WHT data and
     swtich to LTO4 tapes

PSB  order some LTO4 tapes

RWA  move 08A WFCAM raw data backup tapes to Obs. store cupboard

MJI  discuss topics and suitable time/date for telecon with JAC

JRL  make minor changes to object masking sky estimation strategy

MJI  assess and if feasible come up with a plan for software release 
EGS