Transcript Poster - Astronomy - University of Florida

SPIE Astronomical Telescopes and Instrumentation, San Diego, CA,
th
27
June –
nd
2
July 2010
Poster 7737-92
Characterization of the mid-IR image quality at Gemini South
Dan Li, Charles M. Telesco, and Frank Varosi
Department of Astronomy, University of Florida, 311 Bryant Space Science Center, P. O. Box 112055, Gainesville, FL-32611, USA
ABSTRACT
To help the prospective observer take full advantage of the mid-IR capability of Gemini South, we characterize a key aspect of the mid-IR performance of the 8meter telescope at Gemini-S, namely, the appearance and stability of its delivered mid-IR image profiles, with the goal of demonstrating that it can be used with a
level of precision not used before. About 2000 images obtained with T-ReCS (a facility mid-IR camera at Gemini-S) between late 2003 and early 2009 were used for
our image quality (IQ) analysis. All targets are flux standards and recorded at one or more of the four bands Si-2 (8.74 μm), N (10.36 μm), Si-5 (11.66 μm), and Qa
(18.3 μm). A non-linear least squares fitting of three profile models (Lorentzian, Gaussian, and Moffat) was performed on each image, and the FWHM, ellipticity,
position angle and Strehl-ratio (SR) were measured from the fitted profile. We find that the long-time-scale image quality is quite stable in terms of profile width
or ellipticity, though short-time-scale variation is evident. We also examined the correlation between image quality and many ambient parameters and confirmed
the interdependence between the image quality in the Qa band and the ambient humidity. The ellipticity of the profile was analyzed statistically as well. The
average profiles for different filters can be used as important references in the future when a high-quality profile reference is not available during an observation.
Figure 1 (left). An example of how
critical the PSF reference is in the
detection of a barely resolved
emitting region of ζ Lep. The
azimuthally
averaged
intensity
profiles at 18.3 μm of ζ Lep
(diamonds) and the reference PSF
star (dots) show a small difference in
their FWHM, which is indicated by
the two vertical lines. The intrinsic
half-width of the deconvolved disk is
0.14”±0.02”, or 3.0±0.4 AU
(Moerchen et al. 2007).
Figure 4 (above). A direct comparison between
FWHM and SR, which are both used to evaluate
the IQ in this study, shows that SR is a more
sensitive measure of the IQ than FWHM when one
is nearly diffraction-limited. The FWHM does not
change as noticeably as the SR does when the IQ
is approaching the theoretical limit.
Figure 5 (above). A weak correlation between the
ambient humidity and IQ (in terms of SR)
observed in Qa images. This suggestion that the
IQ in Qa band is more sensitive to the high
humidity than other filter is consistent with the
fact that there is strong water absorption within
the Qa passband.
Figure 2 & 3 (above). The temporal variation of the FWHM and
Strehl ratio (SR). The vertical dashed lines indicate the beginning
of each calendar year, whereas the horizontal dashed lines
represent the diffraction-limited FWHM. For most images, the
image quality is close to the diffraction-limit, and the overall IQ
for all four filters has been quite stable since 2006, despite the
significant short-time-scale variances. We checked many ambient
parameters found in the FITS headers to see if any of them could
account for the large dispersion in these IQ measurements, and
confirmed only one correlation between IQ and humidity in Qa
band (see Fig. 5.).
Dan Li June, 2010
SUMMARY
1) In terms of PSF size and SR, the image quality of GeminiS with T-ReCS has been quite stable since 2006;
2) SR is a better measure of image quality than is FWHM,
especially when one is nearly diffraction-limited;
3) The ambient humidity is weakly, but noticeably,
correlated with the quality of the Qa (18.3 μm) images;
4) There is a statistical correlation between the SR and the
ellipticity of the image profile, and observers should be
cautious with non-zero ellipticities measured from the midIR images;
5) The average PSF profiles, which can be used as artificial
references for future observations, are computed for four
filters and two PSF models.
Figure 6 & 7 (above). A statistical
correlation between ellipticity and SR,
which can help us evaluate the reliability
of an ellipticity measurement (i.e.,
whether a measured, non-zero ellipticity
is due to the real physical extension of
the celestial object, or due to an
instrumental imperfection). For example,
we can look at the images with a small
range of SR (e.g., Si5 images with
0.35<SR<0.45, as shown in Fig. 6) and
plot their ellipticity histogram (Fig. 7),
which fits a Gaussian very well.
Assuming that there is a new Si5 image with SR=0.4, and ellipticity=0.2, barely
outside the 3σ according to Fig. 7, one can conclude that it probably reflects the
real elongation of the object, since we seldom see an image with a similar SR having
such a high ellipticity in our image database of flux standards. Similarly, the
“standard” ellipticities of other filters and SR ranges can be computed as
summarized in Table 2 (above right).
Figure 8 (above). Averaged Moffat (left panels) and Gaussian (right panels) profiles.
The dotted profiles are the best PSFs (i.e., with the smallest FWHM) in the
database. Observers can use these profiles as empirical references for planning or if
there is no high-quality PSF star available during an observation. They give us a
general idea of the actual resolution that the Gemini-South and TReCS can achieve
routinely.
ACKNOWLEDGMENTS
This study is supported by NSF grant AST-0738883 to C. Telesco, and based on observations obtained at the Gemini
Observatory (acquired through the Gemini Science Archive), which is operated by the Association of Universities for
Research in Astronomy, Inc., under a cooperative agreement with the NSF on behalf of the Gemini partnership: the National
Science Foundation (United States), the Science and Technology Facilities Council (United Kingdom), the National Research
Council (Canada), CONICYT (Chile), the Australian Research Council (Australia), Ministério da Ciėncia e Tecnologia (Brazil)
and Ministerio de Ciencia, Tecnología e Innovación Productiva (Argentina). The authors wish to acknowledge the kind
assistance of Gemini staff astronomers Scott Fisher, Tom Hayward, and Rachel Mason.