OmegaCAM: The 16k x 16k Survey Camera for the VST

26/03/2003

OmegaCAM: The 16k x 16k Survey Camera for the VST

Observing and data reduction a Virtual Survey System

Edwin A. Valentijn

26/03/2003

Paranal

July 2004

26/03/2003

VLT Survey Telescope-VST– Alt-AZ - Cassegrain– aperture 2.610 m– corrected FOV 1.47 degree

– lens corrector: U - z– Atmospheric disp. correct.: B -z

– f/5.5– scale 14.266 arcsec/mm– CCD pixel size: 15 um– 0.214 arcsec/pixel

– image quality: 80% EE– two-lens: 1.70 pixel – ADC: 1.77 - 2.18 pixel

26/03/2003

VST factory - Napoli

26/03/2003

Detectors• Science array 1 x 1 degree, 32 CCDs

– 15 m pixels – 0.21 arcsec/pixel– Marconi (former EEV) 2k x 4k– 16k x 16k pixels

• Auxiliary CCD’s – 4 CCDs– For guiding– Image analysis

26/03/2003

Filters• Primary set: Sloan u’, g’, r’, i’, z’ high throughput interference• Johnson B, V, Stromgren-v• Segm H up to ~12000 km/s 658.8 665.5 672.2 678.9/ 10.7 nm

– 1100, 4200, 7300, 10400km/sec / 4900 km/sec• Composite u’, g’, r’ ,i’ in four quadrants• Segm Ly alpha z=2-3 372, 400, 450, 507nm / 8 nm • Night sky leak CWL=851.8nm - 877.8nm /13nm

26/03/2003

Wide Field Imaging Science

• Provide targets for VLT• ~60% of time through ESO’s OPC• Individual programs

– Supernovae, Lensing, Kuiper belt objects, Gamma ray, bursts, Microlensing, Brown dwarfs, High proper motion objects, Galactic halo objects, Quasars, AGNs

• Sky Surveys• Long term archival research (10 yr mission)• Science Cases

– Finding exceptional single, rare objects– Statistics on large samples of objects

26/03/2003

Large Data Volume• Wide-field imaging instruments, vast amounts of

data– E.g.: VST = Southern sky (30 min exp, 300 nights/y) in 3

years. Large amount of data! 100 Tbyte of image data and Tbytes of source list data

• Science can only be archive-based

• Handling of the data is non-trivial– Pipeline data reduction– Calibration and re-calibration– Image comparisons and combinations– Working with source lists– Visualization

ESOcompliant}

26/03/2003

Concepts for solutionVirtual Survey System

• Environment that provides systematic and controlled– Access to all raw and calibration data– Execution and modification reduction/calibration pipelines– Execution of source extraction algorithms– Archiving reduced data and source lists, or regenerates these

dynamically– Can be federated to link different data centers

• Dynamical archive continuously grows, can be used for – small or large science projects– generating and checking calibration data– exchanging methods, scripts and configuration

• Key functionality– Link back from source data to the original raw pixel data and

calibration files

26/03/2003

How to use this• Deep multi-color fields

– No need to take all data in one campaign– Combine data of particular quality, assess results– Select sources, visualize interesting ones, …

• 1-in-1,000,000 events spurious or not?• Large homogeneous surveys

– E.g. weak lensing maps, cluster searches, star counts• Variability (source list - or pixel based)

– Proper motions (asteroids, nearby stars)– Flux variations

• Monitor instrument (calibration files)• Planning observations

– View quality of existing data– Build on what already exists, add more filters, more

exposure time, better seeing, …

26/03/2003

Keys -Solution

• Procedurizing– Data taking at telescope for both science and

calibration data– Full integration with data reduction– Design – Data model (classes) defined for data reduction and

calibration– View pipeline as an administrative problem

26/03/2003

Observing Modes Dither

• Dither matching max. gap between arrays ~400 pixels– N pointings (N=5 is standard) – nearly cover all gaps in focal plane and maximizes sky coverage– Very complex context map– couple the photometry among individual CCDs.

–Dither with N = 5

26/03/2003

Observing Modes Jitter

• Jitter matching the smallest gaps in CCDs ~5 pixels– optimizes for maximum homogeneity of the context map – observations for which the wide CCD gaps are not critical– all data from single sky pixel originates from single chip

26/03/2003

Observing ModesStare and SSO

• Stare reobserving fixed pointing positions multiple times– main workhorse monitoring instrument and optical

transients.

• SSO observing Solar System objects– non-siderial tracking and the auto guiding switched off.

26/03/2003

Strategiesscheduling observing modes

• Standard– Single observations (one observing block)

• Deep– Long, multiple integrations– Selected atmospheric conditions– Several nights

• Frequent– Monitors same field– Timescales from minutes to months (overriding)

• Mosaïc– Maps areas of sky > 1o

26/03/2003

Calibration procedures

Sanity checks

Quality controlCalibration procedures

Image pipeline

Source pipeline

26/03/2003

Science Observations

Photometric pipeline

Bias pipeline

Flatfield pipelineImage pipeline

Source pipeline

26/03/2003

Share the loadAstroWise Survey System • Processing

– Hardware • Beowulf processors – 32 (most cases)• Multi Terabyte disks (10 – 100)

– Data reduction• Derive calibration• Run image pipeline (1 Mpx/s)

• Archiving– Storage

• Images (100’s Tbyte), Calibration files (10 Tbyte)• Source parameters (1-10 Tbyte)

– Federate (network speed)• 5 Mb/s (24 hours/day) full replication • 200 Mb/s no replication, on-the-fly retrieval

26/03/2003

Concepts of federation• Federation maintained by a single database- Oracle9i• Full history tracking

– of all input that went into result – providing on-the fly reprocessing

• Dynamical archive - Context as object attributes– Project: Calibration, Science, Survey, Personal– Owner: Pipeline, Developer, User– Strategy: Standard, Deep, Freq (monitoring), Mosaïc– Mode: Stare, Jitter, Dither, SSO– Time: Time stamping VO interface

• Software standards– Classes/data model/procedures– 00 – inheritance/ persistency– Python scripts/ c-libraries USER Python

26/03/2003

26/03/2003

26/03/2003

Observing proposals• Garanteed time NOVA 10%• Call – 25 April--- see www.omegacam• Super clusters- distant clusters• Galactic structure

– Weak shear, microlensing– Bulge

• 2dF, 100 Sq Degree, 10000 Sq Deg• Deep field • Lorentz center July 2003• Fall 2004