Cool Stars 14 - Submitted Abstract # 333
This version created on 05 October 2006

Measuring and Modeling the detection limits in high contrast
coronagraphic images


Remi Soummer, AMNH
Sasha Hinkley, Columbia University/AMNH
Andre Ferrari, Universite de Nice, France
Claude Aime, Universite de Nice, France
Ben Oppenheimer, AMNH
Anand Sivaramakrishnan, AMNH/Stony Brook
Bruce Macintosh, LLNL
Douglas Brenner, AMNH
Russell Makidon, STScI
James (Cornell) Lloyd
Lewis C Roberts, Boeing
Marshall Perrin, UC Berkeley

The achievable contrast (dynamic range) in coronagraphic high contrast
images, is a critical parameter which determines the physical
parameter space for faint companion detection (mass, separation).  On
the ground, the scientific goals are essentially the detection and
study of giant planets.  With Extreme Adaptive Optics and
Coronagraphy, the current on-sky limitations are due to residual
quasi-static speckles, which are not easily calibrated and are the
dominant noise contribution.  Future ground-based instruments (Gemini
Planet Imager, VLT-Sphere) will include active calibration systems
(speckle nulling) and spectrographs (Integral Field Units) to help
push the dynamic range further.  In space, although the problem is
very different and the goals more ambitious (TPF), the dynamic range
of a coronagraphic instrument will also be mainly limited by
hard-to-calibrate, slow-varying speckles.  Similar calibration
approaches (active or passive) will be required to reach the detection
of terrestrial planets.  Theoretical statistical models of the noise
can help us understand the properties of the residual speckles in such
high contrast images.  Models have now been developed and verified on
real data to include the effects of static, quasi-static and residual
atmospheric wavefront errors.  Because of the generality of the
physics involved, these models can be applied to both for ground-based
or space-based instruments, depending on their parameters.  In the
case of actual coronagraphic images, like those obtained with the Lyot
Project coronagraph, understanding of the detection limits is
necessary to set upper-mass limits for non-detections and understand
the performance of the instrument for further improvement.  Detection
limits and upper-masses of Vega's hypothetical companion are
discussed.  

----------------------------------