Stellar Intensity Interferometric Capabilities of IACT Arrays* - - PowerPoint PPT Presentation

Stellar Intensity Interferometric Capabilities

• f IACT Arrays*

Dave Kieda Nolan Matthews University of Utah Salt Lake City, Utah *for VERITAS and CTA collaborations

Photon Bunching & Intensity Interferometry

D= Correlation Distance

3 SII Imaging → Basic Intensity Interferometry

𝐽"

Measures correlation in intensity fluctuations (not amplitude!).

• R. Hanbury Brown, J. Davis and L. R. Allen,

MNRAS 137 (1967) 375.

B

R 𝜇

4

𝐽"

Measures correlation in intensity fluctuations (not amplitude!).

Some Pains, But Large Gains:

• Relatively insensitive to

atmospheric turbulence ( turbulence ~ kHz, sampling ~100’s MHz ).

• Km Baselines possible,
• perate at blue wavelengths
• >sub milli-arcsecond resolution
• No need to maintain steady optical

interference path between focal planes.

• > higher order intensity

correlations possible

• Loss of phase information:

→ Phase recovery needed.

4 SII Imaging → Basic Intensity Interferometry

𝜇

B

5

Signal to Noise Ratio (SNR) in Intensity Interferometry

Hanbury Brown 1974; Twiss 1969

Example Calculation:

• 109 p.e./sec (mv = 1.0, A=100 m2

= 0.3 )

• 1013 Hz (for 10nm filter)
• 250 MHz
• 1 hour observation
• dual polarization
• Assume unresolved = 1.0

SNR = 200 (assuming ideal system, limited to bright sources) Spectral Flux Density (ph s-1 m-2 Hz -1) Photo-electron Rate Optical Bandwidth Electronic Bandwidth Integration Time Normalized Visibility Light collection area q.e. of detectors

How to Improve?

• Multi-Channel

Intensity Interferometer: 𝑂%&

• improvement,

(Trippe, 2015)

• Redundant baselines:

𝑂()*

• improvement.
• Fast optics & correlators

(> 1GHz).

2nd-order time coherence g(2) & Fourier Image plane

+,-,./ +,-/+,./ = g(2)(u, v, t ) = 1 + ½g(1) (u, v, t)½2 For II: experimental time resolution ∆𝑢 ~ 1 nsec blackbody coherence time 𝑢% ~ 4 ∆5 ⁄ ~ 10 psec

g(2)(0,0,0) = 1 + 𝜗 ~ 1 + 10-4 small non-Gaussian fluctuations => Need large photon counts: 10+ m mirrors

g(2)(0,0,0)

g(1) (u, v, t) : first-order coherence =1 for [u, v, t=0]

g(4) u, v, 0 = E I l, m eI"JK *LMNO d𝑚 𝑒𝑛

• I l, m describes the image size

and brightness distribution (Van Cittert-Zernike Theorem 1934,1938) 𝐽 𝑚, 𝑛 𝑛 𝑚

Lab measurement of g(2)(0, 0, t ) of simulated star/thermal light Matthews, Kieda & LeBohec , accepted in J Opt (2017) Reconstructed SII laboratory images stellar disk (left) & binary system (right) Matthews, Kieda & LeBohec , accepted in J Opt (2017)

Potential SII at Optical Telescope Arrays

• Excellent instruments for SII:
• Large photon collection area

(~10 m diameter mirrors)

• Optically isochronous (< 5 ns)

VERITAS IACT Array Future CTA/pSCT Array

1-2 km

100 m to km baselines (milli-arcsec resolution) VLTI- Paranal

100 m

• J. Holder and S. LeBohec, Ap. J. 649 (2006) 399
• D. Dravins et al., New Astronomy Reviews 56, 5 (2012) 143

VERITAS Camera 499 PMT pixels Dual polarization SII pixel (replaces 3 Center PMTs)

Potential VERITAS SII Augmentation

WR-SWITCH Module GPS Timecode Generator

10 MHz 1 PPS

Single Mode Fiber 50m - 2 km (80 km max) 10 GB Ethernet

SII Data Quality Monitor plastic fiber plastic fiber 120 ft double shielded RG 223

Telescope 2 Telescope 3 Telescope 4 Telescope 1 Telescope N

WR-SWITCH Module GPS Timecode Generator

10 MHz 1 PPS

Single Mode Fiber 50m - 2 km (80 km max) 10 GB Ethernet

SII Data Quality Monitor plastic fiber plastic fiber 120 ft double shielded RG 223

Telescope 2 Telescope 3 Telescope 4 Telescope 1

Typical SII Augmentation

Standalone telescope connected

• nly by fiber optic (White Rabbit, 10G)

Telescope N Replace with Custom board in camera?

Simple VERITAS interferometer simulation

*`ghost images’ caused by incomplete sampling of Fourier plane *reflection symmetry of ghost images caused by loss of phase information Fourier image Plane sampling (vertical) Simulated baselines 100 𝜈 𝑏𝑠𝑑𝑡𝑓𝑑

• S. Lebohec et al. 2009

http://www.cta-observatory.org/ 1-2 km baselines 20-100 telescopes

Simulated observations of binary stars with different sizes. (mV = 3; Teff = 7000 K; T = 10 h; Dt = 1 ns; l = 500 nm; Dl = 1 nm; QE = 70%) Already changes in stellar radii by only a few micro-arcseconds are well resolved. Better sampling of Fourier image plane-> no ghost images D.Dravins, S.LeBohec, H.Jensen, P.D.Nuñez:, CTA Consortium Optical intensity interferometry with the Cherenkov Telescope Array, Astropart. Phys. 43, 331 (2013)

Simulated Fourier image Planes Reconstructed Binary images Input Binary images

Prototype is Currently under construction at VERITAS Observatory (Fall 2017 commission)

f = 5.6m, D= 9.6m,FOV= 8° Pixel= 6mm (×11,328SiPMs) PSF

D68= ~0.04 –0.08° (pix= 0.06°)

CTA-US Schwarzschild-Couder Telescope

• V. Vassiliev, S. Fegan, P. Brousseau:

Wide field aplanatic two-mirror telescopes for ground-based g-ray astronomy Astropart.Phys. 28, 10 (2007)

Schwarzschild-Couder two-mirror IACT telescope

Wide field of view, excellent spot size RMS spread in arrival time of rays at focal plane as a function of field angle. .

On-axis: photon timespread <0.2 nsec rms >>improved g(2)(t) >>reduced observation time 2-4Ghz sampling + SiPM (QE-0.9). SNR = 200 -> SNR =2400 !

IACT Observatories are Excellent Instruments for SII Imaging

• Digitizing electronics/White Rabbit synchronization
• Standalone telescopes connected by inexpensive fiber optics.
• 1-10 km baseline separations now achievable
• Offline (post-observation) photon correlations
• Polarization data allows higher S/N, baseline noise estimation
• Demonstrated correlation using of pipelined FPGA for near real-time processing
• Higher order correlations may contain additional image information (Ofir & Ribak 2006)
• Near Term Implementation
• VERITAS Telescopes (Potential deployment/use in 2017-2018)
• 100-200m baselines: 100 𝜈 arcsec imaging possible
• Longer Term Development
• VLTI visible light feed /single board implementation?
• CTA/SCT-telescope array implementation : 1-2 km baselines, <100 𝜈 arcsec imaging?