ESO VISTA Public Surveys — a Status Overview

Telescopes and Instrumentation

ESO VISTA Public Surveys — A Status Overview

1 500 Figure 1. Overview of Magda Arnaboldi P85+ Marina Rejkuba1 the VISTA time alloca- 1 400 tion in hours. P85+ Jörg Retzlaff P86 included dry runs which 1 Nausicaa Delmotte 300 were executed before Reinhard Hanuschik1 P87 the official start of VISTA Michael Hilker 1 200 operations. 1 P88 Wolfgang Hümmel 100 Gaitee Hussain1 Valentin Ivanov 1 0 P89 Alberto Micol1 VV V VH S 1 VM C VIDE O

Mark Neeser VIKING 1 Monika Petr-Gotzens Open tim e Ultra-VISTA Thomas Szeifert 1 Fernando Comeron 1 six VISTA PSs from the extended period imager VIRCAM is provided by Emerson Francesca Primas 1 85 to period 89 (P85–P89). et al. (2006). Observation preparation Martino Romaniello 1 and submission for public surveys is The implementation of the PSP recom- done on a half-yearly basis, following the mendation on time to survey completion standard ESO service mode procedures. 1 ESO implied a revision of the standard VLT The PIs of VISTA public surveys and their model of scheduling observing time in teams submit Observation Blocks (OBs) service mode because the VISTA public — the standard basic observing units. The ESO policies for public surveys survey programmes are all essentially The deeper surveys (Ultra-VISTA, VIDEO, include regular monitoring and reviews at the highest priority of execution. As an VIKING and VMC) have typically ~ 1-hour- of the progress of the surveys to en- added complexity, normal programmes long OBs, while the wide surveys (VHS sure their legacy value and scientific were scheduled alongside public surveys and VVV) have OBs lasting between 5 competitiveness. The review process is and Chilean programmes from P87. and 30 minutes. The survey with the larg- carried out on the basis of reports by Because the former have top priority and est number of submitted OBs is VVV: the ESO survey team and the survey PIs essentially cover all observing condi- because of the multi-epoch observations submitted to the relevant Public Survey tions, normal (open time) programmes the total number of OBs prepared and Panel (PSP) — the VISTA PSP in this must also be assigned a similar priority if executed will be of order 30 000. Figure 2 case. A summary of the time allocation they are to be executed at the telescope. shows the number of hours of execution to VISTA surveys, the service mode Starting from P88, normal programme of OBs each month for all VISTA surveys. observations and the current progress observations are not carried over regard- The time charged for each OB execution after two years of telescope operation less of their ranking, to avoid extra car­ry­ is only counted for successfully com- is provided. Furthermore the content overs hindering the progress of the pleted observations and Figure 2 shows of the data products delivery by the ­surveys. To limit the impact of open pro- the completed OB hours that are com- survey teams to the ESO Science grammes on the progress of the VISTA puted by the observation preparation Archive Facility is described. PS the following policy has been imple- software (P2PP3). While in the first year mented since P89: open time proposals of the operations there was still some can access restricted right ascension fine-tuning of the operational procedures Phase 1 — time allocation to surveys ranges, but should carefully justify the and overheads, now the overheads com- scientific goals and why these scientific puted by version 3 of the Phase 2 Pro- Scheduling of the VISTA Public Surveys goals are not achievable through the posal Preparation system (P2PP3) are (PS) is done in line with the VISTA PSP scheduled PS. very similar to the actual time needed to recommendation that every effort should execute the observations. be made to complete the surveys within five years. Realistic expectations on what Phase 2 — service mode observations The histogram in Figure 2 shows the can be executed in the next semester for public surveys effect of technical interventions on are also taken into account, based on the speed of execution of the public sur- ex­perience gained from the previous VISTA public surveys are ambitious vey OBs, with the main impact falling semester(s) of VISTA operation. The time ­programmes which entail very large allo- in the period 3 April 2011 to 4 May 2011, allocation to surveys is based on the cations of time, of order of 1800 to which coincided with the recoating of updated observing request by the PS 3400 hours, with consistent observing the primary (M1) and secondary (M2) mir- principal investigators (PIs) and takes into conditions. For a summary of the scien- rors with aluminium due to fast degra­ account the carryover time needed to tific goals of the VISTA public surveys, dation of the original silver coating, and complete the observations submitted in their area coverage and photometric an extended recovery and maintenance the previous period. Figure 1 provides characteristics, see Arnaboldi et al. (2007). of M1. an overview of the time allocation to the An overview of VISTA and of its wide-field

The Messenger 149 – September 2012 7 Telescopes and Instrumentation Arnaboldi M. et al., ESO VISTA Public Surveys — A Status Overview

250.0 Figure 2. Histogram of VISTA public survey completion rates the number of hours of execution of OBs each month for all VISTA sur- Here we present the current fraction of 200.0 veys up to 1 April 2012. completeness for each VISTA PS and the

r) Quality control grade A expected time to completion, based on (h is assigned to OBs com- the current efficiency of execution. The 150.0 pleted fully within io n requested constraints absolute completion rate is based on the- (i.e. sky transparency, oretical OB execution times as computed ecut 100.0 image quality, airmass, by the P2PP execution time module and ex Moon illumination, etc.), the time requested by the PI of each sur-

OB while grade B is assigned to those OBs vey for the total duration of the project, 50.0 for which one of the as described in the Survey Management constraints is violated Plan (SMP). The absolute completeness up to ~ 10%. fraction is given by the ratio of the com- 0.0 pleted OB time divided by the time re­­ 11 11 11 11 11 11 12 12 10 10 10 10 10 10 09 r- r- v- y- r- quested in the SMPs. The completion y- v- v-

Jul- Grade A Jul- Jan- Ma No Sep- Jan- Jan- Ma Ma

Ma rates are calculated for the observations No Sep- Ma No Grade B up to the end of Period 88 (end of March In order to enable the survey teams to area definition file generated with SADT. 2012), hence they cover the first four prepare large numbers of similar OBs In this way hundreds of similar OBs ­periods of VISTA operations. In these four and for ESO user support and science that adopt the same observing strategy periods, the effective telescope time operations to verify and execute them and point to different areas in the sky is about 1.75 years, as there were three in the most efficient manner, new tools are generated automatically, thus pre- months of technical downtime (mirror have been developed (Arnaboldi et al., venting typing errors and saving time. In coating, M1 intervention and technical 2008). We list them briefly: addition to an interface with the SADT, downtime to correct the focal plane tilt). this tool also allows the survey PIs to These technical interventions affected The Survey Area Definition Tool (SADT) define the survey observing strategy via mostly the VVV and Ultra-VISTA projects, was developed by the VISTA consortium scheduling containers. These concepts as they took place in those months (led by J. Emerson) and delivered as can be easily encoded in an automatic when either the Milky Way Bulge or the part of the United Kingdom in-kind con- ranking algorithm that suggests to the COSMOS fields are visible. tribution to joining ESO. The main role observers which OB is to be selected out of the SADT is to efficiently mosaic the of many thousands of similar survey OBs. The absolute completion rates of the survey area and to identify the necessary The OT is used on Paranal for selecting ­surveys at the end of P88 are listed in active optics and guide stars for each the observations to be executed; it has Table 1. These can be used to extrapo- tile. The SADT has been further improved the VISTA-specific ranking algorithm late the time necessary to complete following the experience gained during (Bierwirth et al., 2010) as well as report- the VISTA public surveys, assuming that VISTA science verification, dry runs, and ing functionality. the observations continue at a similar the first period of observations (P85+). pace. From these assumptions, the VHS The latest improvement was to introduce Survey Monitoring and Visualisation survey can be completed in about five the 2MASS catalogue transformation Tool (SVMT) is a new tool developed to years, while VVV, Ultra-VISTA, VIDEO and to I-band magnitudes, which is now used facilitate the Phase 2 review of thousands VMC are all expected to take more than for guiding and active optics on VISTA. of survey OBs and to enable monitoring seven years. These numbers are lower This is important, as it was noticed that of the survey observations. The main limits as the requested observing time in other catalogues suffered from low reso- functionalities are the display of the sur- the SMPs was computed using overhead lution, containing binary stars or galax- vey areas and the OB status for selected values that later proved to be underesti- ies, and some defects (e.g., containing areas or survey runs, and the visualisa- mates. As the survey projects progress, spurious sources due to scratches on tion of the distribution on-sky for the sub- and once one of the approved surveys is the plates from which these surveys were mitted survey data products. Basic completed, the speed of completion is derived) which caused execution time ­statistical functionalities have been im­ple­­ expected to increase for the other sur- losses due to the choice of unsuitable mented, such as calculation of the sur- veys. The VIKING survey was set to the guide and/or active optics stars. vey area and accumulated exposure lowest priority in P89 (October 2011– times for which the observations have March 2012), as this survey currently fails Phase 2 preparation tool for surveys been completed or the data products to comply with the Phase 3 policies of (P2PP3) and Observing Tool (OT) for submitted. This tool is used by the User data products submission for the ESO public survey support have resulted from Support Department and Archive Sci- public surveys. Additional time allocation the evolution of the main ESO observa- ence Group, for the purpose of Phase 2 for the public surveys is pending the tion preparation and execution tools. The and Phase 3 reviews, respectively, and completed delivery of the agreed data PS teams prepare the observations using for overall survey progress monitoring by products. P2PP3 which takes as input the survey the ESO Survey Team (EST).

8 The Messenger 149 – September 2012 Table 1. Overview of the absolute completeness of the VISTA public surveys after 1.75 years of are condensed into a few terabytes of VISTA observing time (effective). data products each year. Because of the legacy value of the public survey data Period VVV VMC VIKING Ultra-VISTA VIDEO VHS products, ESO’s policy is to ensure their 01.04.2010–30.03.2012 24 % 22% 32% 29% 25% 38% long-term archival storage and public accessibility in order to promote their wide scientific use by the astronomical VISTA science operations Catalogue 2 (GSC2) optical catalogue community at large, beyond those scien- with the 2MASS near-infrared catalogue, tific goals initially identified by the sur- The VISTA telescope and its near-infrared and by extending the latter catalogue to vey teams. In order to bring this about, camera VIRCAM are in normal operation: include synthetic optical magnitudes. ESO has set in place a whole new pro- science data are being taken, and deliv- The high elongation of the measured point cess, called Phase 3, which represents ered to the users. Currently, the typical spread function (PSF) detected during the final step in the execution of ESO technical downtime is 2–3%, to be com- early telescope operations is now mostly large programmes and public surveys. pared with 10–12% of weather down- under control and resolved. Occasionally time. During the commissioning of VISTA elliptical images are recorded, typically The new Phase 3 infrastructure was and VIRCAM, the overheads for telescope due to azimuth axis oscillations that deployed on 10 March 2011; it supports pointing, guiding and camera setup appear at seldom and irregular intervals. the reception, validation and publication turned out to be ~ 30% larger than the of data products from the public survey overheads estimated by the VISTA con- Several optimisation steps were intro- projects and large programmes to the sortium. Several actions were then imple- duced into the instrument and telescope ESO Science Archive Facility; for a mented to reduce the various sources operation to increase the efficiency of description of the Phase 3 tools and the of overhead. Furthermore, the telescope observations: relaxing the declination tol- infrastructure see Arnaboldi et al. (2011). efficiency is about 10% lower than ex­­ erance of the AO correction and par­ Data products must be consistent pected because the original silver mirror allelising the instrument and telescope with the data standards documented by coating was replaced with aluminium setups to decrease the overheads. Finally, the Archive Science Group at ESO. in April 2011, due to the fast degradation the calibration plan was further revised, These data standards are required to of the earlier coating. In Table 2 we pro- giving an extra 40 minutes per night for characterise the level of data reduction vide the zero-point measurements in survey OB execution. Further hardware and calibration, to track provenance, early operations with silver coating (dat- improvements are not realistic; software which allows ESO to monitor survey pro- ing from October 2009), the last meas- developments and further improvement gress, and finally to support the query urements for silver (from March 2010) and of operations procedures may effectively for specific data products via the ESO the zero point with the new aluminium improve progress speed, but only by a archive interfaces. In the context of the coating (measurements from July 2011). few percent. Phase 3 process, the data products returned by the survey teams are regu- The ESO policies for public surveys larly monitored with respect to the actu- Table 2. Photometric zero-point measurements entail that raw data become public as ally executed observations. The timeline (Vega magnitudes) in early operations (October soon as they are ingested in the ESO defining the requested delivery dates 2009) with silver coating, last silver coating values (taken March 2010) and with the new aluminium ­Science Archive Facility (SAF). Following by type of data product is given by the coating (July 2011). the Memorandum of Understanding Phase 3 policies. for VISTA, the Cambridge Survey Unit Filter October March July (CASU) receives a copy of all the raw For the VISTA data, a number of prod- 2009 2010 2011 data collected by VISTA for public sur- uct types have been identified: images, Ks 23.03 22.88 22.96 veys and normal programmes. PIs of weight maps and source lists. As high- H 23.87 23.61 23.76 VISTA public surveys do not automatically level data products, the resulting source J 23.78 23.43 23.66 receive a copy of the raw data, but can catalogues from the PS projects repre- Y 23.45 22.97 23.33 access the pipeline data products either sent particularly important resources. Z 23.82 23.22 23.58 from CASU or the Wide Field Astronomy The catalogues are different from the Unit (WFAU) archive in Edinburgh. Raw source lists, which are per-tile products data are delivered regularly upon request and can be downloaded as entire FITS During current standard operations, typi- to the Ultra-VISTA team. tables. The catalogue contents will cal VIRCAM/VISTA technical downtime be searchable via a dedicated query is due either to the M2 unit or active interface in the ESO SAF, which should optics (AO) and autoguiding (AG) related Phase 3 – data products from public sur- become available in the second half problems. The AO and AG problems are veys of 2012. Basic functionalities will be sup- partially caused by the incompleteness ported to allow the archive user to carry of the catalogues used by SADT to select The raw data collected at the survey tele- out searches by position as well as guiding stars. The catalogue issue was scopes for the different projects amounts by non-positional source parameters for improved by replacing the Guide Star to about 1.5 TB per month, which, in turn, sources in any area of the sky visible

The Messenger 149 – September 2012 9 Telescopes and Instrumentation Arnaboldi M. et al., ESO VISTA Public Surveys — A Status Overview

Table 3. Overview of the ESO Phase 3 data releases resulting from the VISTA public surveys.

Survey Submission Date of Release Content Filter Sky Type of Data Total Total Publication Date Observations Coverage­ Products Volume Number Date (sq.deg) of Files VVV 03.05.2011 Feb 2010–Sep 2010 Contiguous patch of ZYJHKs ~ 520 Tiles 1.5 TB 7980 25.07.2011 Bulge and Disc region (348 tiles) Single-band source lists including multi-epoch data in Ks

VIDEO 03.05.2011 Nov 2009–Feb 2010 XMM-LSS field YJHKs 1.5 Tiles 24 GB 291 25.07.2011 Single-band source lists

VMC 08.09. 2011 Nov 2009–Nov 2010 Two tiles in the LMC: YJKs 3 Stacked tiles and 8.1 GB 1256 25.09.2011 one overlapping 30 pawprints Doradus and the other Single-band and band- the South Ecliptic Pole merged source lists region

VHS 15.09.2011 Nov 2009–Sep 2010 VHS DES – YJHKs ~ 19 10 Tiles 3.9 TB 96474 17.10.2011 120 s in JHK Pawprints VHS ATLAS – Single-band source lists 60 s in YJHK VHS GPS – 60 s in JK

Ultra- 06.10.2011– Dec 2009–Apr 2010 Deep imaging of the YJHKs, 1.5 Stacked images, 87 GB 19 15.02.2012 VISTA 30.01.2012 COSMOS field NB118 ­SExtractor catalogues including w-selected multi-band catalogue

VIKING 10.10.2011 Phase 3 data submission to be closed by PI – – 19 GB 6276 –

from the La Silla Paranal Observatory, for further scientific selection and investiga- tion on the user’s computer.

Table 3 provides an overview of the ESO Phase 3 data releases resulting from the VISTA public surveys and Figure 3 shows a map with the sky coverage. A detailed description of the data releases is available1. These release pages are based on the information provided by the PIs and give an overview of the obser­ vations, a description of the data reduc- tion and calibration, additional release notes illustrating the scientific quality of the data and any additional useful infor- mation regarding the content of the release. The data products from each VISTA public survey are available via the query form2 or by following the link at the bottom of each release description page.

Figure 3. The VISTA public survey data products released through the Phase 3 process in 2011/2012 cover almost 2500 square degrees of the southern hemisphere. Colour coding: VHS–green, VVV–red, VMC–yellow, VIDEO–pink, UltraVISTA–purple. Entire survey footprints are shown in light blue.

10 The Messenger 149 – September 2012 The first public data release from the actively carrying out their observations References VISTA public surveys took place in and achieving their scientific goals: as Arnaboldi, M. et al. 2011, The Messenger, 144, 17 December 2011. Since then, more than documented by the nine ESO press re­­ Arnaboldi, M. et al. 2008, The Messenger, 134, 42 five thousand files and 2.0 TB of VISTA leases based on VISTA data since Janu- Arnaboldi, M. et al. 2007, The Messenger, 127, 28 data products have been downloaded ary 2010 and the 23 refereed papers from Bierwirth, T. et al. 2010, SPIE, 7737, 19 from the ESO SAF by the community. The ESO public surveys thus far. In order to Emerson, J., McPherson, A. & Sutherland, W. 2006, The Messenger, 126, 41 most requested and downloaded data increase community awareness of the product type is the source list (in FITS scientific research opportunities from the table format). VISTA public surveys, ESO is organising Links a workshop on 15–18 October 2012 enti- 1 Description of the Phase 3 data release: tled “Science from the next generation http://www.eso.org/sci/observing/phase3/data_ 3 Conclusions imaging and spectroscopic surveys” .The releases.html astronomical community is invited to 2 Query form for data products from each VISTA VISTA survey operations have now access the opportunities provided by the public survey: http://archive.eso.org/wdb/wdb/adp/ phase3_vircam/form reached a steady state and nearly 80% large wealth of data products delivered 3 Homepage of the workshop “Science from the next of the night-time on VISTA is used to by the public surveys and to use them to generation imaging and spectroscopic ­surveys”: ­successfully execute survey observa- explore science goals beyond those origi- http://www.eso.org/sci/meetings/2012/sur- tions. The VISTA public survey teams are nally identified by the VISTA PS teams. veys2012.html ESO/INAF-VST

This large field colour image of the nearby edge-on spiral galaxy NGC 253 was composed from VLT Survey Telescope (VST) images in g-, r- and i-bands. NGC 253 is a member of the Sculptor group at a distance of about 3.5 Mpc and as the nearest starburst galaxy shows signs of disc disruption and outflowing ionised gas. The dust absorption, well seen in this image, obscures the central star- burst and possibly an active galactic nucleus. More details are available in Release eso1152.

The Messenger 149 – September 2012 11