Volume 555, July 2013
|Number of page(s)||14|
|Published online||27 June 2013|
Apodization in high-contrast long-slit spectroscopy
Closer, deeper, fainter, cooler
1 Astrophysics group, School of Physics, University of Exeter, Stocker Road, Exeter EX4 4QL, UK
2 Aix Marseille Université, CNRS, LAM (Laboratoire d’Astrophysique de Marseille) UMR 7326, 13388 Marseille, France
3 Space Telescope Science Institute, 3700 San Martin Drive, Baltimore MD 21218, USA
Received: 2 March 2012
Accepted: 11 May 2013
The spectroscopy of faint planetary-mass companions to nearby stars is one of the main challenges that new-generation high-contrast spectro-imagers are going to face. However, the high contrast ratio between main-sequence stars and young planets makes it difficult to extract a companion spectrum that is not biased by the signal from the star. In a previous work we demonstrated that coupling long-slit spectroscopy (LSS) and classical Lyot coronagraphy (CLC) to form a long-slit coronagraph (LSC) allows low-mass companions to be properly characterized when combined with an innovative a posteriori data analysis methods based on the spectral deconvolution (SD). However, the presence of a slit in the coronagraphic focal plane induces a complex distribution of energy in the Lyot pupil plane that cannot be easily masked with a binary Lyot stop, creating strong diffraction residuals at close angular separation. To alleviate this concern, we propose to use a pupil apodization to suppress diffraction, creating an apodized long-slit coronagraph (ALSC). We show that this concept allows looking at a closer separation from the star, at deeper contrast, which enables the characterization of fainter substellar companions. After describing how the apodization was optimized, we demonstrate its advantages with respect to the CLC in the context of SPHERE/IRDIS LSS mode at low resolution with a 0.12′′ slit and 0.18′′ coronagraphic mask. We performed different sets of simulations with and without aberrations, and with and without a slit to demonstrate that the apodization is a more appropriate concept for LSS, at the expense of a significantly reduced throughput (37%) compared to the LSC. Then we performed detailed end-to-end simulations of the LSC and the ALSC that include realistic levels of aberrations to obtain several datasets representing 1 h of integration time on stars of spectral type A0 to M0 located at 10 pc. We inserted the spectra of planetary companions at different effective temperatures (Teff) and surface gravities (log g) into the data at angular separations of 0.3′′ to 1.5′′ and with contrast ratios from 6 to 18 mag. Using the SD method to subtract the speckles, we show that the ALSC brings a gain in sensitivity of up to ~3 mag at 0.3′′ over the LSC and that both concepts are essentially equivalent for separations larger than 0.5′′. The gain at small separation is the result of suppressing of the bright Airy rings that are difficult to estimate at very small angular separations because of the point spread function chromaticity. The improved sensitivity is confirmed by extracting the simulated companions spectra from the data and comparing them to libraries of models to determine their Teff and log g. Using a restoration factor that quantitatively compares the input and output spectra, we show that the ALSC data systematically leads to better quality spectra below 0.5′′. In terms of Teff, we demonstrate that at small angular separations the limit with the ALSC is always lower by at least 100 K, inducing an increase in sensitivity of a factor up to 1.8 in objects’ masses at young ages. Finally, for the determination of log g, we show that the ALSC provides a less biased estimation than the LSC.
Key words: instrumentation: adaptive optics / instrumentation: high angular resolution / techniques: spectroscopic / methods: numerical / planetary systems
© ESO, 2013
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.