Wide-field monitoring strategy for the study of fast optical - - PowerPoint PPT Presentation

Wide-field monitoring strategy for the study of fast optical transients

S.Karpov, G.Beskin, S.Bondar, V.Plokhotnichenko, G.Greco, A.Guarnieri, C.Bartolini, A.Piccioni Presented by G. Greco Deciphering the Ancient Universe with Gamma-Ray Bursts 19-23 April 2010, Kyoto, Japan

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Fast Variability of the Sky: historical perspective

H.Bondi, «Astronomy of the future», 1970

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Fast Variability of the Sky: what is «fast» and what is «slow»?

As a rule, fast optical transients have unpredictable localizations, both in time and on the sky

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Gamma-Ray Bursts:

• pen questions about optical emission
• When does it start and when does it end?
• Transition from prompt emission to afterglow

– several hundreds of afterglows, but only about ten prompts

• Temporal variability

– gamma is highly variable down to 10-4 s, what about optics?

• Relation to gamma emission

– are they correlated? – what is the temporal lag between them? who is the first?

• Prompt emission from the short bursts

– afterglows are basically the same. what about prompts? All this require the detection of very first moments of the burst and, obviously, high temporal resolution of observations

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Gamma-Ray Bursts: lessons from the Naked-Eye Burst

• Peaked at V~5.3 m
• Fast optical variability

– ~9 seconds — four peaks – ~1 second — around last peak

• Simultaneous start and end
• 0.82 correlation with 2 s optical

delay

• Rules out large subset of

theoretical models, like External Shock and Inverse Compton

• nes

Naked-Eye Burst demonstrated the importance of high temporal resolution in optical study of GRBs

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Catching the transient of unknown localization: different ways to be the first

How to catch the short transient of unknown localization?

• Listen to Swift, and then move fast

– Typical strategy of robotic alert-based systems. – A lot of instrument, a lot of afterglows. What about the prompt?

• Listen to Swift, but look the same direction

– Several wide-field monitoring systems around the world – Several upper limits (~10m) for the moment of the burst – and finally — the Naked-Eye Burst!

• Be completely on your own

– Routine monitoring of wide areas of the sky – Automatic detection and classification of transients

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Wide-Field Monitoring: requirements for alert-based observations

The faster you repoint — the better

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Wide-Field Monitoring: efficiency of assisted observations

You need only to look al the transient position when the satellite detects it Upper limits for the prompt flux are results too So, the shorter the focus - the better

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Wide-Field Monitoring: efficiency of independent observations

Exposure shorter than the event Exposure longer than the event

Short transients require short exposures (preferably equal to its duration)

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Wide-Field Monitoring: requirements for a general-purpose system

• Need wide field of view

– the shorter the focus the better

• Need good detection limit

– the larger the diameter the better

• Need high temporal resolution

– short exposures and fast read-out – low read-out noise

• Need real-time processing software

– real-time detection and classification of transients

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Wide-Field Monitoring: systems currently in operation

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

FAVOR & TORTORA systems:

• verview

FAVOR (FAst Variability Optical Registrator) camera — SAO RAS, since 2003 Built in collaboration with IPI and IKI (Moscow), supported by CRDF grant

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

FAVOR & TORTORA systems:

• verview

TORTORA - Telescopio Ottimizzato per la Ricerca dei Transienti Ottici Rapidi Two-telescope complex:

• independent detection
• automatic study

La-Silla, Chile mounted on REM since 2006 Team: SAO RAS, IPI, Bologna University, REM

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

FAVOR & TORTORA systems: technical details

Objective Diameter: 150 mm Focal length: 180 mm D/F: 1/1.2 Field of view: 17x24o Image Intensifier type: S20 diameter: 90 mm amplification: 120 downscale: 4.5/1 Q.E.: 10% CCD type: SONY 2/3'' IXL285 size: 1388х1036 exposures: 0.128 — 10 sec scale: 50''/pixel limit: ~11.5m for 0.13с Data flow rate — 20 Mb/s, per night— 600 Gb, ~200.000 frames

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Real-Time Data Processing:

• verview

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Real-Time Data Processing:

• verview
• Data flow rate is 20.5 Mb/s = 160 megabit/s
• 7.5 frames per second, 1388x1036 pixels each
• Single frame processing is ten times slower!

–

• bject detection / SExtractor - ~0.5 s

– PSF photometry of ~1000 objects - ~0.5 s – Classification of ~1000 objects - ~0.3 s

• Solution

– differential imaging for real-time detection and classification of transients – complete data storage for at least one day – detailed post-factum study of selected interesting events

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Real-Time Data Processing: differential imaging

mean value subtraction dispersion normalization

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Real-Time Data Processing: decision scheme

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Real-Time Data Processing: decision scheme

is elongated? is moving? is in catalogues? METEOR SATELLITE KNOWN OBJECT SATELLITE NEW TRANSIENT Object we just detected

yes yes yes no no no

and all this is done in only 0.4 seconds

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Real-Time Data Processing: meteors and satellites

• ~100 satellite passes per night

– ~20 transient-like flashes per night – 2.5% of tracks are unidentified satellites! – 50-500 points per one pass — good quality of the trajectories

• ~100 meteors per night

– typical duration of 1-3 frames – real-time detection and logging – day-time processing by dedicated software

• Hough transform — determination of direction
• photometry along the track — start/end, light curve

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Conclusions

• Wide-field monitoring is inevitable for detecting fast
• ptical transients of unknown localization

– GRBs, meteors, satellites, debris

• High temporal resolution is necessary for short or fast

moving events

– short bursts – Naked-Eye Burst

• Data processing for such monitoring is easy

– detection and basic classification in 0.4 seconds

• Selection of optimal hardware parameters is not so easy

– simple and cheap, and not very efficient – complex and clever, and expensive

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Wide-Field Monitoring Systems: perspectives

Next things to do:

• Increase field of view
• Improve the detection limit
• Keep the high temporal resolution (~0.1 s)
• Acquire some spectral information

– Multicolor data – Low-res spectra

• Measure the polarization

We have such a project now – the Mega-TORTORA

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Mega-TORTORA: detector and objective

Andor iXonEM+888 Back-illuminated EMCCD 1024x1024 pixels, 13um pixel size QE up to 95% Frame rate from 8.9 till 310 per second Read-out noise < 1e- CANON EF 85 f/1.2 L USM II

D=70mm, F=85mm 9.0 x 9.0 deg FOV with 31'' pixel

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Mega-TORTORA: real-time data processing

Standard 1U rackmount PCs One PC per channel Data acquisition and storage for several nights (~1 Tb) Real-time data processing Already existing FAVOR software, well tested since 2003 Transient detection and classification in 0.4 s

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Mega-TORTORA: basic 3x3 unit

B V R P1 P3 P2  Nine objectives in gimbal suspension each  Independent pointing  Three fast-installable color filters ( for synchronous BVR photometry)  Three fast-installable polarization filters (for synchronous mesurement

• f 3 Stokes parameters)

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Basic 3x3 unit: modes of operation

Wide-Field Monitoring Filters Installation Narrow-Field Follow-Up

Mode transition in ~0.3 seconds

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Basic 3x3 unit: performance

• Wide-Field Monitoring mode

– ~730 sq.degrees FOV – ~13m limit (S/N=5) in white light (or B band) for 0.1 s

15.2m for 10 s, 18m for 1000 s

– detection and classification of transients in ~0.4 seconds

• Narrow-Field Follow-up mode

– insertion of color and / or polarimetric filters and repointing towards one object in ~0.3 s – ~81 sq.degrees FOV – limits in different regimes:

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Basic 3x3 unit: time scales

• Wide-Field Monitoring – 1000 s

– frame co-addition to increase the limit

• Narrow-Field Follow-up – 15 ms (3 ms for very bright OT)

– region of interest + binning to increase temporal resolution

Andor iXonEM+888

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Mega-TORTORA: complete 8 units system



72-objective system



Cover large sky area (2100 deg2 for 3x3x8)



Can reach 17.2m-19.7m in narrow field mode (for 0.1 — 10 s exposure)



Will see the light of GRBs once per month (at least we hope so) Prices



Objective — 2000 E

 CCD — 45000 E 

Rackmount PC — 1000 E



Fork mounting — 26000 E

 500 kE for basic 3x3 unit 

5 ME for complete 8 units system

  

—



Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Mega-TORTORA: targets sky survey twice per night

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Mega-TORTORA: its place among the others

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Fast Variability of the Sky: who and how fast (second try)?

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Wide-Field Monitoring: Mega-TORTORA prototype

• 9 Canon (85/1.2) objectives with polarization and color (BVR)

filters

• celostate mirrors for fast repointing
• image intensifiers (GaAs, QE = 30% at 4500A) and

ICX285AL CCDs

• ~100 square degrees per channel, ~900 in total
• 12.5m B limit for 0.13 s in monitoring mode, 15m for 10s

13.5m B limit for 0.13 s in narrow field mode, 16m for 10s It will cost about 180 Ke We have money to begin!

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Mini-Mega-TORTORA: elements for several channels

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Mini-Mega-TORTORA: schematics

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

Wide-Field Monitoring: Rise of the Machines

• Naked eye (since prehistoric times)
• Human-operated instruments (since 1700s)
• Robotic telescopes (since 1990s)

– instruments for discovery of transients – telescopes for detailed study of transients

• The Transformer Telescope (we will inform you later)

– different regimes — different modes of operation – single instrument for both discovery and detailed study – self -organizing (intelligent?) system of large number of identical units (like SKA)

Beware invasion of small robots working together!

Wide-Field Monitoring Strategy for the Study of Fast Optical Transients Kyoto, 2010

See you in nine years!

Naked-Eye Burst