Transcript Image quality - Queen`s University Belfast

ROSA
A high-cadence synchronized multi-camera solar
imaging system
Dr. Mihalis Mathioudakis
Physics and Astronomy, Queen’s University Belfast
ROSA : Rapid Oscillations in the Solar Atmosphere
Outline
•
History (SECIS – RDI)
•
Science examples
•
Improving image quality
Post-observing correction (Speckle, PDS)
•
The proposed instrument – Tests
•
Observing modes
•
Associated instruments
•
Summary
SECIS - RDI
•
SECIS (Solar Eclipse Coronal Imaging System)
(RAL Ken Phillips, QUB)
Fast mode impulsively generated wave in a loop (6 s)
Williams, Phillips et al. MNRAS 2001
Williams, Mathioudakis et al. MNRAS 2002
•
RDI (Rapid Dual Imager)
Oscillation induced along a flare ribbon (40 – 70 s)
(BBSO - NSO, Sac Peak)
McAteer et al. ApJ 2005
High frequency oscillations in the lower atmosphere (15 – 30s)
Andic, Jess, Mathioudakis in preparation
EIT/Loop image
Intensity variations along the loop
Williams, Mathioudakis et al. MNRAS 2002
200 arcsecs
NOAA 9591 – C9.6 in Hα
RDI at Big Bear Solar Observatory
McAteer et al. ApJ 2005
C9.6 flare – Period of 52sec
McAteer et al. ApJ 2005
Ha blue wing

Oscillatory power
15 – 30 sec (60 – 30mHz)
50 arcsec

RDI at DST Sac Peak
50 arcsec
RDI was funded by a Royal Society Instrument Grant
Andic, Jess, Mathioudakis in preparation
Multi-wavelength
McAteer et al. ApJ (2003)
The need for synchronised imaging
Krijger, Rutten et al A&A (2001)
Krijger, Rutten et al. A&A 2002
The need for high cadence
Allred et al. ApJ (2005)
G-Band
Image credit : SST - MPS
Image quality – The problem
•
•
Atmospheric turbulence
Fried’s r0 – diameter of refractive index fluctuations
r0 = 0.114 ((λ cosz) / 550))0.6 m
r0 = 11 cm (λ = 550nm , z = 0)
Spatial resolution of a ground based telescope limited to that of a
telescope with diameter r0
The largest telescopes have the same image quality as an 11cm telescope
(if no image correction is applied)
•
Choose an observing site with a large r0
•
Time scale of atmospheric fluctuations : t = r0 / v
Wind speed v = 11 mph , t = 20 msec, (moves by its own diameter)
Act quickly – Exposure times of a few msec at most!
Speckle pattern
•
Remember : Seeing is equivalent to many small telescopes
observing the same object but affected differently by
atmospheric turbulence
Speckle reconstruction
•
•
•
Image of a source in an ideal telescope in the absence
of atmosphere is shaped by diffraction
The Imaging Equation
i (x) = o (x) ‫ ٭‬p (x) (1)
i - observed intensity/image of the source
o - actual/true image of the source
p - PSF describing instrument and seeing
x - angular position
Following the FT of (1)
I (u) = O (u) • P (u)
P (u) – is the Optical Transfer Function (OTF)
u – spatial frequency
Speckle reconstruction
G-band
Andic, Jess, Mathioudakis in preparation
Improving image quality
•
For Speckle to work you need
Very short exposures. Freeze the seeing for each exposure
(<20ms)
Very high cadence. A sequence of images (50-100) over
timescales that solar features remain unchanged (< 10 s). Bad
seeing requires more images
Great demand on camera read out speeds
Signal to noise can be very low in narrow band images
National Solar Observatory (NSO/NSF)
Sacramento Peak

Altitude : 2800m

Very good seeing for short periods
(morning)

Dunn Solar Telescope 0.76m
41m above ground + 67m underground

ASP/DLSP/SPINOR (vector magnetograms)
IBIS, HSG, UBF, High Order AO

PPARC approved solar facility

20 days per year for UK led proposals
Camera - Computing
•
iXon+ 1004 X 1002 CCD
Andor/Texas Instruments
•
Max Frames per sec : 32 (full CCD)
200 (125 x 125)
•
1.8TB/day/CCD (8 hours observing)
•
Fast local disks (15K RPM)
•
LTO2/3 tape autoloaders
ROSA – Hardware tests
Ha center before and after reconstruction
Doppler velocities – Narrow band filters
• Construction of blue (λ – Δλ) and red (λ + Δλ) wing images.
• The intensity difference between the images provides a
Doppler shift. In a symmetric profile there is no difference in
intensity.
Image credit : IBIS group
Fe I velocity map
Image credit : IBIS group
Magnetic Fields
Zeeman effect – Polarization

Longitudinal case
B  to the line of sight

Transverse case
B  to the line of sight

Splitting proportional to the
magnetic field

Components are polarized
Magnetic Fields
•
Δλ = 4.67 x 10-13 g λ2 B//
where B// is the line of sight component of B
•
UBF (Universal Birefringent Filter) and a
Wollaston prism
•
Images of opposite circular polarization
Summary
•
ROSA has been funded £450K (SRIF3 and PPARC)
•
Hardware tests completed (lab & telescope)
•
Software tests (November 2006)
•
Delivery in late 2008 at DST/NSO
•
Common user instrument
Time through TAC
The DST is a PPARC approved facility
•
Strong interest from the UK community
•
20 days per year for UK proposals (any instrument)
•
The solar microscope
Advanced Technology Solar Telescope (ATST)
The need for high resolution
•
Photospheric photon mean free path and pressure scale height
0.1’’ = 70 km
•
Magnetoconvection coupled with atmospheric dynamics
•
Small scale structures
Umbra dots – Spicules – Bright points
•
Flux Tubes – Buidling blocks of the magnetic photosphere
Flux Tubes and Wave Generation
Flux Tubes & Coronal Loops – How are they linked ?
•
•
•
•
Physical processes take place in very small scales (10-20 km)
Implications on stellar activity
Advanced Technology Solar Telescope
ATST
Aperture : 4m
FoV : 5’
•
•
0.35 – 35 µm
0.03’’ @ 0.5 µm
0.08’’ @ 1.6 µm
First light in 2012
Haleakala, Hawai
Altitude : 3,080m
•
•
•
•
•
•
•
Broad-band imager - Visible & NIR spectropolarimeters - Visible tunable
filter - NIR tunable filter - IR spectrograph - Vis/NIR high dispersion
spectrograph
Design challenges : Energy removal, AO, scattered light, detectors