Astrophysics Luncheon Seminar
The Precision Projector Laboratory: Detector Characterization and Astronomical Emulation
Presented by Charles Shapiro
Jet Propulsion Laboratory, California Institute of Technology
Monday, March 9, 2015
12:00 noon in 169-336
"Weak" gravitational lensing refers to the way large-scale (extragalactic) gravitational fields slightly distort the observed shapes of galaxies. By averaging over many galaxy images, one can map gravitational fields and therefore the distribution of matter in the Universe, which is sensitive to the properties of dark matter and dark energy. As ongoing and future weak lensing missions create larger weak lensing maps, pushing galaxy counts from the millions into the billions, the drastically improved statistical power puts increasingly strict limits on the allowed galaxy shape measurement errors due to a telescope's detector (image sensor).
In order to assess the impact of detectors on weak lensing measurements, Caltech Optical Observatories and JPL have jointly formed the Precision Projector Lab (PPL). Our principal instrument - an Offner-based re-imaging system (a.k.a. "the projector") - casts precisely controlled images onto UV-VIS-NIR detectors. Measuring these images allows us to characterize detectors and quantitatively understand non-idealities that are inaccessible by conventional detector tests (i.e. darks and flats). The projector emulates astronomical data such as stars, galaxies, or spectra in order to assess the impact of detector errors on scientific applications (e.g. shape measurement, astrometry, photometry, spectroscopy, guiding). It is designed to be versatile and stable with a simple point-spread-function (PSF).
I will discuss the various capabilities and applications of the PPL and how it has been leveraged by multiple projects. I will also present preliminary results from our investigation of a Teledyne Hawaii-2RG infrared detector similar to detectors planned for the Wide Field Infrared Survey Telescope (WFIRST). We place upper limits on shape correlation biases induced by the detector that are two orders of magnitude smaller than the expected gravitational signal from WFIRST.