[ ESO ] Working Group Recommendations 

EIS Working Group
Draft Minutes of the Meeting

ESO Garching, January 21/22 1998
Draft 3: March 24, 1998


Attendance

The meeting was attended by M. Albrecht (part time), D. Baade(part time), J. Bergeron, S. Charlot, G. Chincarini, S. Cristiani, L. Da Costa, S. D'Odorico, R. Gilmozzi (part time), J. Krautter, K. Kuijken, K. Meisenheimer, Y. Mellier, F. Paresce (part time), P. Quinn (part time), M. Ramella, A. Renzini, H. Rottgering, R. Saglia, P. Schneider.


Agenda

J. Krautter welcomed all participants. The proposed agenda of the meeting was approved as follows:

1.
EIS progress report and further strategy
2.
Data release policy
3.
Strategy of the DEEP survey
4.
VLT test camera
5.
Public surveys with the 2.2m
6.
Requirements for the 16k camera for the 2.5m telescope
7.
Discussion of concerns and worries in the member states

Minutes were taken by S.Cristiani.


EIS progress report and further strategy

L. Da Costa reported on the progresses of EIS. The favorable weather has allowed completing 1.2 and 3.2 sq.deg (V and I respectively) in patch A; 1.5, 1.5, 1.6 sq.deg. (BVI) in patch B with a typical seeing of 1''. Observations in patch C are about 2/3 complete up to now with good seeing ($\sim 0.8''$). Observations in patch D also started. As recommended in the previous meeting, it has been tried to re-observe the areas in patch A affected by poor seeing.

SURVEY AREA (sq.deg.)

(January 98)

Patch B V I Used Total
A   1.2 3.2 4.4  
B 1.5 1.5 1.6 4.6  
C     4.0 4.0  
D     1.6 1.6  
  1.5 2.7 10.4 14.6 17.4

Most of the technical problems previously encountered have been solved or at least significantly improved. In particular CAMAC problems and pointing do not appear to be an issue any longer. Misalignment of the optics and ghosts of bright stars have been detected and may complicate the reduction process. The overall optical performance of EMMI appears however to be much more stable and reproducible than in the past.

Among the completed tasks L.Da Costa listed:

At present the photometric performances are estimated to be: a $1 \sigma$ limit at an isophote of 25.6 mag/ sq.arcsec in I, 26.3 in V and 26.8 in B. This translates roughly into 22.5-23.0 I limiting mag ($5 \sigma$) for a single frame, corresponding to a contamination level in the counts of about $10\%$. The expected accuracy of the photometry is 0.05 mag internal and 0.1 mag absolute. The astrometry has an internal accuracy of 0.03 arcsec, sufficient for future multi-slit spectroscopy at the VLT.

The presently ongoing activities encompass:

Future tasks, to be completed by July 98, include:

As a future strategy the WG recommends to complete patch C and D in I (64 observing blocks required) and patch B in U (1.5 sq.deg. down to mag U=24). For the latter part it is recommended to request 24 hours of imaging with SUSI2 in fall 98 (period 62).

Data release policy

The guidelines for the distribution of the data products of the EIS survey have been discussed at length. The importance of a release of the data as quick and wide as possible - within the possibilities of the present technology and ESO manpower - has been stressed. The interface to the user will be the Archive, to which the EIS team delivers the data.

The following products are envisaged:

1.
Catalogs of sources: no problem is expected for their full availability on the web. The catalogs will contain: $\alpha$, $\delta$, x, y, magnitudes (SEXtractor auto mags), mag errors, major and minor axes and position angle of the image, a star/galaxy classifier and a description of the context, i.e. info on the PSF, the $1 \sigma$ limiting mag and other observational parameters.
2.
Sub images: cut-outs of $10' \times 10'$ will be provided through an image server in a protected form so to limit the access to the ESO community. A set of about 300 cutouts of cluster candidates will be prepared on DATs already in advance of the foreseeable request from the community.
3.
Reduced frames within a limit of $\sim 2$ Gby per archive request, i.e. about 100 frames, will be distributed through the archive. The reduced frames will contain world coordinates allowing direct astrometry and zero-points for photometry. For comparison it should be noted that 1 patch is typically made of 600 frames. Compressed/rebinned versions of the coadded images will also be available.
4.
Large number of raw frames, coadded images, full set of reduced frames. The large amount of the data (e.g. about 100 Gby the coadded images, 90 Gby the calibrated frames) with all probability will limit their use to a few institutes in the ESO community. It is therefore envisaged that these data sets will be distributed on DLTs (e.g. 3 DLTs for the coadded images) to the institutions in the ESO community that will request them, possibly upon payment of a fee covering the expenses for the storage media. An inquiry will be made in collaboration with the ESO webmaster to survey the community and build up a statistic of what kind of data products will be of interest to the users, in particular to quantify the number of institutes interested in having a copy of the full/coadded data. The coadded sections will be subdivided in images of 4K$\times$4K size.

The deadline for the release of the above described products is Jul 31, 1998. In the meantime there will be two preliminary releases: in March 1998 reduced and calibrated I frames in patch A with astrometric parameters and common photometric zero-point upon request. The corresponding preliminary catalogs will also be released. In May 1998 similar data for the V band in patch A and B, V, I, in patch B will be released.

It has been stressed that also the tools developed in the framework of EIS should be considered a product of the survey, in particular in view of the future surveys to be carried out at the 2.2m and other telescopes. Another product available for the community will therefore be a portable survey pipeline to produce routinely coadded images and catalogs. The WG group thinks that the development of a portable survey pipeline by ESO will be an important support for future Wide Field Surveys carried out by individual persons or institutions of the European Community. It should be noted however that ESO does not have the manpower to produce a fully portable software and a certain degree of hardware dependence has to be expected.

Strategy of the DEEP survey

S. D'Odorico reported on the status of SUSI2: at the end of February the tests at the telescope will be completed, the 2 chips available show good QE ($60\%$ at 320nm, $\gt 90\%$ at 440nm), excellent RON ($\simeq 5 e^-$), and satisfactory cosmetics (about 3 and 12 hot-spots, respectively). They are mounted with a gap of 6 arcsec and independent astrometry has to be carried out on each of them.
A.Morwood reported on the status of SOFI: recent tests at the NTT show that with a seeing of 0.6 arcsec and 1h exposure time in a $2 \times$ PSF FWHM aperture the following S/N=5 limits can be reached: J=22.9, H=21.9, K=20.9. The ``flatness'' of the array is $5\%$ without flat fielding, $0.07\%$ can be reached with FF. The FOV of $5\times 5$ arcmin should be compared with the ISAAC largest field of $2.5\times 2.5$ arcmin.

In the first quarter of 1998 commissioning data on a deep field will be obtained with SUSI2 and SOFI. Test reduction with the EIS pipeline will start in April.

For EIS Deep 7 nights with SUSI2 (5 + 4 1st half) and 6 nights with SOFI (4 + 4 1st half) are scheduled in period 61. The WG recommends to carry out the EIS deep observations in the direction including the HDF-S region according to the following strategy: 3 fields of $5\times 5$ arcmin corresponding to the STIS, WFPC2 and NICMOS areas should be observed in UgrIJK. The recommended limiting magnitudes are U=26.5, g=26, r=26, I=26, J=23.5 and K=21.5. The field corresponding to the WFPC2 images should be observed down to a deeper limiting magnitude in the J (24) and K (22) bands. Should the weather not be favorable in Period 61, the WFPC2 field has the highest priority.

The data volume will be of 58 Gby, the data rate = 0.08 Mb/s. 5000 science frames will be produced and processed by the EIS pipeline, the same developed for EIS wide.

The WG recognizes the difficulties generated by the present choice of the HDF-S. In particular the presence of a close-by bright star appears to be detrimental to any deep imaging with NTT and VLT/FORS. The working group unanimously passed the following resolution:
``The ESO Working Group for Wide Field Surveys urges STScI to change the location of the Hubble Deep Field South. Because of the nearby bright star ($d \sim 5'$) it will be impossible to obtain deep ground-based optical images in the UBVRI bands. This applies both to imaging with the NTT and FORS at the VLT. Any star brighter than 10th magnitude should be away by at least 15'.''

The WG recommends that the request for the period 62 should cover both a field including HDF-S fields and one additional field at high galactic latitudes to be observed in UgrIJK. A total of 8.5 nights are required.

VLT test camera

A. Renzini briefly presented the characteristics of the VLT Test Camera (TC), recalling in particular its FOV (80''), similar to that of one CCD chip of the WFPC2. The VLT science verification with the TC is now planned for the whole dark time of August 1998, hence conveniently placed for observing the HDF-S fields. Various observations, complementary to HST observations, are threrefore possible. For example, observe NICMOS fields thus providing deep optical observations complementing the IR from HST; Observe one WFPC2 chip with the TC providing information on low surface brightness details for which HST is photon starving, and allowing to combine HST and VLT data for which specific capacity exists at the ST-ECF. The use on the TC of the set of SUSI2 intermediate band filters is also possible. Renzini asked the WG to discuss these various possibilities and possibly set priorities.

The EIS WG considered observations with the VLT test camera a very useful complement to EIS. WFPC2 deep fields will be observed with it. The EIS WG recommends that the VLT test camera should carry out observations also of the EIS deep fields, in particular of the HDF-S and of the NICMOS field. The images should be obtained in the same SUSI2 filters and become immediately public.

Public surveys with the 2.2m

A. Renzini reported on the status of the wide field imager at the 2.2m (WFI@2.2). The delivery of the detectors is on schedule, and the instrument will be installed at the telescope in September 1998. ESO aims at allowing the scientific use of the new facility starting November 1, 1998, though delays cannot be excluded due to late delivery of the detectors. In any event, it is expected that the WFI@2.2 will become available at least for the second half of Period 62. The next call for proposals (to be issued on March 1, 1998) will include the WFI@2.2. Taking into account collecting area, field of view, throughput and QE of the CCDs, the camera will be a factor $\sim 6$ more efficient than NTT/EMMI. The 2.2m telescope will be the first telescope of its class to be fully dedicated to wide field imaging, wordwide, thus offering a unique opportunity to the ESO community. It is also important to notice that the data flow rate from the WFI@2.2 will be several times higher than that of all other optical instruments at ESO telescopes on La Silla (the WFI@2.2 will have 67 million pixels, against 34 million pixels of all other optical instruments together). All in all, in 1999 the data flow from La Silla will be 4-5 times higher than in 1998.

Under these circumstances it appears that at least during 1999 the bottleneck for wide field imaging will no more be represented by the available telescope time (and corresponding raw data), but by the ability to properly and timely process this tremendously increased data flow rate. The experience that has been accumulated with EIS will be precious to start coping with this problem, offering a working survey pipeline able to process frames producing coadded, mosaiced, photometrically and astrometrically calibrated images and object catalogs. However, the EIS survey pipeline needs to be significantly upgraded in order to be able to deal with images made of 8 slightly disjoint subimages, one for each chip of the WFI@2.2. An evaluation of the resources necessary to this upgrade is currently under way.

Renzini emphasized that a relatively modest incremental effort is required if one dwells on the expertise now accumulated by the EIS Team. A public pilot survey with the WFI@2.2 in Period 62 would then provide the focus for the pipeline upgrade, with also the aim of making the pipeline portable towards the community, the phylosophy of ESO public surveys being that every products (including software tools) are made publicly available. Besides producing the immediate advantages for the VLT science of such a survey, the goal will therefore be to spread within the community the capability of dealing with large numbers of very large format images such as those produced by the WFI@2.2 (and later by the 2.5m telescope on Paranal, the VST), and to efficiently produce surveys, no matter whether public or private. Instrumental to make the survey pipeline practically portable is the wide participation in the EIS Team of astronomers and scientists from the community. Otherwise, the production of a fully documented, platform independent pipeline would require resources that are not available, or an impractically long time. Instead, the return to the home institution of well trained EIS Team members will be essential for a most efficient use of these new facilities. Institutes and groups interested in an extensive use of the WFI@2.2 should be encouraged to participate in the EIS Team activity through the EIS Visitor Programme, while planning for the necessary hardware investment at home. For this latter aspect, concern was also expressed for the capability of the community to take prompt advantage of the WFI@2.2 during 1999.

Finally, Renzini emphasized how this effort for the WFI@2.2 is essential to prepare for a prompt scientific use of the VST in 2001, when the new telescope will mark another quantum jump in the wide field imaging capability (a factor of $\sim 12$ compared to the WFI@2.2!). A long discussion followed on the policies to be followed for future surveys. Various issues were treated also in a closed session restricted to the non-ESO members of the WG and resulted in the following results.

Definition of a public survey: a survey that the WG recognizes of compelling interest for the ESO community, in particular in view of the maximization of the scientific outcome of the VLT, and is not likely to be covered by ``private'' proposals. A public survey makes the data available immediately, has to contain in the proposal a detailed description of the products and the distribution policy. Successful examples of public surveys are EIS, the HDF or the recent radio surveys FIRST and NVSS.

Collaboration between ESO and institutes in the community is highly recommended. Interchange of people and expertise between ESO and the institutes has proved to be highly effective (e.g. the DENIS experience transported in the EIS).

A call for ideas to the ESO community should be issued as soon as possible using channels of communication as broad as possible. The answers to the call should be collected by the WG which will study the ways to stimulate collaboration, integrate the scientific aims into general interest programs in the framework of the ESO possibilities in terms of manpower, budget, data distribution and contribute with own ideas in order to maximize the scientific output of the survey. The ideas for public surveys should reach the WG with some advance (about 2-3 months before the deadline for normal proposals) for discussion and integration of the goals. In this way the WG will be able to improve its task of sensing what are the feelings of the community and define a baseline for public surveys with hopeful collaboration between ESO and the community.

The competition among all the possible survey proposals should be encouraged: a public survey proposal, after being endorsed by the WG, should go through the OPC and compete with the other proposals. The WG recommends that there should be neither special rights like e.g. a special presentation to the OPC nor special priorities in the scheduling because of the publicity. Scheduling should be done on the basis of scientific priorities only. The scientific priorities will be, of course, determined by OPC. In future, the Chairman of the OPC should be ex officio a member of the WG.

The WG recommends that up to 1/3 of the total time available at the 2.2m telescope is dedicated to public surveys.

As a pilot project for the first quarter of 1999 the WG recommends the completion of the EIS survey in the originally planned bands with the addition of a long wavelength filter (the z band $\sim 900$nm) especially suited for observations in grey time. The WF camera of the 2.2m should also be used to reach a fainter limiting magnitude in the I band and to observe a deep field in order to explore what are the limiting capabilities of this facility.

Requirements for the wide field camera for the 2.5m telescope

Renzini reported that an Announcement of Opportunities will soon be issued by ESO in order to procure from the community a wide field camera for the 2.5m telescope (VST) to be located on Paranal. In the formulation of the AO ESO will take into account any recommendation concerning the scientific requirements for the camera that the WG may wish to express, e.g. sampling, filter set, tip/tilt, ADC, driftscan capability, etc., and their relative priorities.

The characteristics presently foreseen for the camera at the VST, have been illustrated: at least a 16k x 16k Camera, a square of about 23 cm side of CCDs with pixel size of 0.23", slightly under-sampling in conditions of very good seeing. Alternatively an 18K, 0.21" camera could be foreseen. There were some doubts whether a drift scan mode acquisition would be possible because of the expected misalignment of the individual CCDs.

The WG gives top priority to image quality. Specifically, items like atmospheric dispersion corrector, small pixel CCD's (perhaps an 18K, 13.5 micron pixel design), tip/tilt or other fast image tracking, are considered to be very important. A further suggestion would be to go to even smaller (and more!) pixels in order to be able to fully exploit the very best seeing.

Filters were mentioned by several WG members, but no clear recommendation has emerged. One possibility could be to install a second filter wheel for user-provided interference filters. Filter exchange should be avoided whenever possible. It was also mentioned the possibility to install negative masks in order to block bright objects. Drift scan capabilities, if feasible, are considered to be very useful.

The WG also points out the great interest of wide field IR imaging, possibly in campaigns of several months alternated with optical campaigns.

Discussion of concerns and worries in the member states

The feedback of the community as sensed by the WG member has been discussed at length. The relationship public versus private surveys, the overall implications in terms of global efficiency in terms of scientific results, operations, data reduction have been examined in the framework of the general relationship between ESO and the various institutes in the member countries.

The WG recognizes the importance and usefulness of public surveys in general and that EIS represents a significant investment of telescope time in a period in which the pressure of the ESO community on the La Silla facilities is particularly high. The WG was well aware of this when endorsing the survey and takes full responsibility for this choice.

The WG appreciates the efforts of the EIS team to publicize the EIS characteristics, goals and data products. For the future it is recommended to maintain these efforts and extend them to the general concept of public surveys, which should be widely advertised. The collaboration between ESO and interested groups in the community should also continue to be fostered. In this respect, the above described ``call for ideas'' will be a useful way of sensing the reactivity of the community and stimulating future interactions.

Send comments to <eisadm@eso.org>
Last update: Oct 30, 2000 		[Science Activities][ESO][Index][Search][Help][News][Archive]