The Supernovae Integral Field Spectrograph: keys to high-precision spectro-photometry

HAL Id: in2p3-00171064

http://hal.in2p3.fr/in2p3-00171064

Submitted on 5 Aug 2009

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of

sci-entific research documents, whether they are

pub-lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diffusion de documents

scientifiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

The Supernovae Integral Field Spectrograph: keys to

high-precision spectro-photometry

C. Buton, Y. Copin, E. Gangler, G. Smadja, E. Pecontal, G. Rigaudier, P.

Antilogus, S. Bailey, R. Pain, R. Pereira, et al.

To cite this version:

C. Buton, Y. Copin, E. Gangler, G. Smadja, E. Pecontal, et al.. The Supernovae Integral Field

Spec-trograph: keys to high-precision spectro-photometry. Semaine de l’Astrophysique Française (SF2A

2007), Jul 2007, Grenoble, France. Société Française d’Astronomie et d’Astrophysique, 2007.

�in2p3-00171064�

SF2A 2007

J. Bouvier, A. Chalabaev, C. Charbonnel (eds)

THE SUPERNOVAE INTEGRAL FIELD SPECTROGRAPH : KEYS TO

HIGH-PRECISION SPECTRO-PHOTOMETRY.

C. Buton

, on behalf of the Nearby Supernova Factory, Y. Copin

, E. Gangler

, G. Smadja

, E.

Pecontal

_{, G. Rigaudier}

_{, P. Antilogus}

_{, S. Bailey}

_{, R. Pain}

_{, R. Pereira}

_{, C. Wu}

_{, G. Aldering}

_{, C.}

Aragon

, S. Bongard

, M. Childress

, S. Loken

, P. Nugent

, S. Perlmutter

, K. Runge

, R. C.

Thomas

_{, B. A. Weaver}

_{, C. Baltay}

_{, A. Bauer}

_{, D. Rabinowitz}

_{and R. Scalzo}

Abstract.

The Nearby Supernova Factory aims at discovering and observing a sample of type Ia supernovae through the dedicated Supernovae Integral Field Spectrograph, currently in operation since 2004 on Mauna-Kea UH telescope. To reach the targeted spectro-photometric accuracy, attention has been focused on various aspects of the calibration procedure, including: estimate of the night photometricity, derivation of the mean atmospheric extinction over the extended optical domain (320-1000 nm), its modeling in terms of physical components (Rayleigh and aerosol scatterings, ozone absorption and telluric lines) and its variability within a given night. Point-source extraction from the IFS datacube also requires a detailed knowledge of the atmospheric-induced point spread function (PSF). The overall accuracy of the calibration chain is estimated on reference flux standard stars.

1 Introduction

The Nearby Supernova Factory (SNfactory, Aldering et al. 2002) is an international project dedicated to the study of the nearby thermonuclear supernovae. Based upon the NEAT search for the target discovery and the dedicated 3D-spectrograph SNIFS for the follow up, the goal is to study the evolution of a vast sample of Sne Ia at 0, 03 < z < 0.08 from −15 to +50 days on the extended optical range (320 − 1000 nm). This will allow to probe in detail the local Hubble diagram, the SNe Ia physics, the SNe host galaxy correlations, etc.

The dedicated Supernova Integral Field Spectrograph (SNIFS, Lantz et al. 2004) mounted on the UH 2.2m telescope (Mauna Kea) comprises an IFS covering a 6” x 6” field from 300 to 1000 nm (blue and red channels with micro lens array), a photometric channel, covering 4’ x 5’ around the IFS (monitoring of the atmospheric transmission) and an integrated guiding and focusing channel.

This contribution describes two steps of the calibration chain (Copin et al. 2007): the correction for the atmospheric extinction, and an empirical derivation of the point spread function which is needed because the field is too small and one wants to extrapolate the data to have a precise measurement of the flux.

2 Mean atmospheric extinction

In order to obtain the mean atmospheric extinction, one can break up the transmission of the atmosphere in this manner:

1

Institut de physique nucl´eaire de Lyon, Universit´e de Lyon, CNRS/IN2P3 UMR 5822, 4, rue E. Fermi 69622 Villeurbanne cedex, France

2

Centre de Recherche Astronomique de Lyon, France

3

Laboratoire de Physique Nucleaire et de Haute Energies de Paris, France

4

Lawrence Berkeley National Laboratory, Berkeley, CA

5

Yale University, New Haven, CT

c

Soci´et´e Francaise d’Astronomie et d’Astrophysique (SF2A) 2007

2 SF2A 2007

Tz(λ) =

S(λ)

S⋆_{(λ) × T (λ)} (2.1)

where S is the observed spectrum, S⋆ _{the intrinsic spectrum, T the telescope transmission and T}

z the

atmo-spheric transmission.

To account for the various extinction components, Tz can be parametrised as:

Tz= 10−0.4KZ with KZ= κ(λ)z + κT(λ)zα (2.2)

where κ(λ) and κT(λ) represent the different atmospheric extinction contributions which consists of four

com-ponents (Adelman et al. 2003, see Fig. 1):

• Rayleigh scattering consists on a power law scattering by the air molecules (λ−4).

• Ozone absorption which is concentrated to an altitude from 10 to 35 km . Two bands are to be considered, Huggins (UV) and Chapuis (visible), with two large peak of diffusion around 573 and 602 nm.

• Aerosol scattering is due to the particles and the various pollutants of human source discharged into the

atmosphere (λα

, −1 < a < 4).

• Telluric lines (see Fig. 1) are absorption lines dues to O2 and H2O of the Earth atmosphere. The telluric

correction is proportional to zα _{(0.5 < α < 1 for H}

2O lines and α ≃ 0.5 for O2 saturated lines) (see

Fig. 1).

As the telluric lines have a different behavior with respect to z, we need to treat in a separate way the correction of the extinction and the correction of the telluric lines (see Eq. 2.2).

0.4 0.5 0.6 0.7 0.8 0.9 Wavelength[micron] 0.0 0.1 0.2 0.3 0.4 Ex tin c tio n [m a g / a ir m a s s ] Mean_extinction. fits a .x^(;4)+b.O3+c.x^d Rayleigh,a=0.007 Ozone,b=1.102 Aerosol,c=0.057d=1.612 CFHTextinctioncurve a = 0.54 + /`0.07 0.00 0.05 0.10 0.15 0.20 0.25 0.30 `1.80 `1.75 `1.70 `1.65 `1.60 `1.55 Lin e 1 a = 1.01 + /`0.25 0.00 0.05 0.10 0.15 0.20 0.25 0.30 `2.6 `2.5 `2.4 `2.3 `2.2 `2.1 `2.0 Lin e 2 a = 0.58 + /`0.02 0.00 0.05 0.10 0.15 0.20 0.25 0.30 log(secz) `1.05 `1.00 `0.95 `0.90 `0.85 L in e 3 a = 0.68 + /`0.34 0.00 0.05 0.10 0.15 0.20 0.25 0.30 log(secz) `2.6 `2.5 `2.4 `2.3 `2.2 `2.1 `2.0 `1.9 L in e 4 Telluric absorption 5500 6000 6500 7000 7500 8000 8500 9000 0.0 0.2 0.4 0.6 0.8 1.0 F lu x

Fig. 1.Left: Computed SNIFS extinction (blue curve) in comparaison with the original Mauna Kea extinction (green

curve, Boulade O. 1987, B`eland et al. 1988); white points represent telluric lines, red curve represent Rayleigh component, blue curve represent ozone component and purple curve represent aerosol component. Right: Telluric exponent α for each telluric line computed from 10 stars of the night 06 014.

The function κ can be estimated from several exposures taken in the same night at different airmasses. Thus, Several κ allows to derive an average atmospheric extinction.

3 PSF radial profile

The point spread function (PSF) describes the response of an imaging system to a point source. For ground based optical telescopes, atmospheric turbulence (seeing) usually dominates the contribution to the PSF as it is the case for SNIFS.

The radial profile of the point spread function cannot be parametrised by a single gaussian function, a pair of gaussians, nor a Moffat function (Trujillo et al. 2001). It was found however that our data could be reproduced by the sum of a Gaussian for the core, and a Moffat function for the wings (see Fig. 2). An additional constant contribution is needed for the sky, as seen in equation 3.1:

P SF(r) = nG× exp( r 2 2σ2) + nM×  1 + r 2 α2 −β + sky (3.1) 0 2 4 6 8 10 12 Radius[Spaxels] 103 102 101 100 101 Flu x [ Ar bitr ar yu ni ts ] 06_164_040_003_17_B,HR4963,z=1.107,e=1.012 Dataset Gaussian+Moffat+sky:Chi2=102728.96 Gaussian:s=1.79,n=6.73 Moffat:a=3.06,b=2.67,n=6.17 0 2 4 6 8 10 12 Radius[Spaxels] 10d3 10d2 10d1 100 101 Flu x [ Ar bitr ar yu ni ts ] 06_134_072_002_17_R,SS433,z=1.061,e=1. 373 Dataset Gaussian+Moffat+sky:Chi2=375603.32 Gaussian:s=4.08,n=17.09 Moffat:a=2.04,b=2.39,n=16.43 sky:0.0448

Fig. 2.Left: PSF radial profile fitted by a gaussian+moffat function for the short exposure star (∼ 1 sec) HR4963.

Right: same as left with a sky component for the long exposure (∼ 300 sec) SS433.

To reduce the number of parameters involved, we have investigated their correlations with the radius α of the Moffat function. They are shown in Fig. 3 for a sample of 330 stars.

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 alpha 1.0 1.5 2.0 2.5 3.0 3.5 4.0 be ta Correlations ShortExposuresB LongExposuresB ShortExposuresR LongExposuresR a=0.41+/Ò0.08,b=1.38+/Ò0.18 a=0.34+/Ò0.13,b=1.74+/Ò0.32 a=0.42+/Ò0.09,b=1.41+/Ò0.17 a=0.35+/Ò0.09,b=1.63+/Ò0.22 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 alpha 0.0 0.5 1.0 1.5 2.0 2.5 s ig ma ShortExposuresB LongExposuresB ShortExposuresR LongExposuresR a=0.26+/Ò0.08,b=0.50+/Ò0.18 a=0.31+/Ò0.13,b=0.36+/Ò0.32 a=0.25+/Ò0.09,b=0.42+/Ò0.17 a=0.21+/Ò0.09,b=0.56+/Ò0.22 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 alpha 0 2 4 6 8 10 12 Ga u s sia n/ Mo ffa t ShortExposuresB LongExposuresB ShortExposuresR LongExposuresR a=Ò0.30+/Ò0.08,b=1.71+/Ò0.18 a=Ò0.13+/Ò0.13,b=1.32+/Ò0.32 a=0.03+/Ò0.09,b=1.37+/Ò0.17 a=0.03+/Ò0.09,b=1.41+/Ò0.22 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 sigma 0 20 40 60 80 100 Nu m be r ofs ta rs median=1.0410 median=0.8980 LongExposures ShortExposures 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 alpha 0 20 40 60 80 100 Nu m be r ofs ta rs median=2.3183 median=1.7632 LongExposures ShortExposures 1.0 1.5 2.0 2.5 3.0 3.5 4.0 b eta 0 20 40 60 80 100 120 Nu m be r o fs ta rs median=2.4923 median=2.1504 LongExposures ShortExposures $2 0 2 4 6 8 10 gaussiannorm/moffatnorm 0 20 40 60 80 100 120 140 160 180 Nu m be r o fs ta rs median=0.9793 median=1.0665 LongExposures ShortExposures

Fig. 3.Correlations and distributions of parameters for 330 stars observed by the SNFactory collaboration in 2006 (0

correlations fixed). The correlation parameters are different for long and short exposures but are the same for red and blue channel.

It is seen in Fig. 4 that when these correlations are imposed as linear relations, we are left with three free parameters in Eq. 3.2 (sky, α and N).

P SF(r) = N × η(α) × exp( r 2 2σ(α)2) + N ×  1 + r 2 α2 −β(α) + sky (3.2)

where η(α) = Gaussian norm_{Moffat norm} .

.

4 SF2A 2007 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 sigma 0 5 10 15 20 25 30 35 40 Nu m be r o fs ta rs median=0.8887 median=0.7973 LongExposures ShortExposures 0.5 1.0 1.5 2.0 2.5 3.0 3.5 alpha 0 10 20 30 40 50 60 Nu m be r o fs ta rs median=2.0505 median=1.5653 LongExposures ShortExposures 1.0 1.5 2.0 2.5 3.0 3.5 b eta 0 10 20 30 40 50 60 70 80 Nu m be r o fs ta r s median=2.4801 median=2.0878 LongExposures ShortExposures = 2 0 2 4 6 8 10 gaussiannorm/moffatnorm 0 20 40 60 80 100 Nu m be r o fs ta r s median=1.1210 median=1.2555 LongExposures ShortExposures

Fig. 4.distribution of the parameters for a sample of data truncated by sigma < 1.0. It is interesting to point out that

the parameter beta for the short exposures is significantly smaller than for the long exposures (Short N. et al. 2003).

After having imposed the correlations of σ, β and η with alpha, we obtain ’constrained’ distributions for these parameters . It is observed in particular that in the sample with good seeing (σ < 1”, β parameter is especially well measured with good seeing), the parameter beta is significantly smaller for the short exposures (see Fig.4). A dependence of the power law on the exposure time is indeed expected from the work of Nicholas Short (Short et al. 2003) .

4 Conclusion

We have measured with the SNIFS instrument of the SNfactory collaboration an accurate extinction curve at Mauna Kea, and derived a correction for telluric lines. These are important steps in the calibration chain of the SNfactory collaboration. Starting from preliminary studies, a good empirical description of the PSF was found, which can be used in the Point Source Extractor of the collaboration.

References

Trujillo, I., Aguerri, J. A. L., Cepa, J., & Guti´errez, C. M. 2001, The effects of seeing on S´ersic profiles -II. The Moffat PSF, mnras, 328, 977

Aldering, G. et al. 2002, Overview of the Nearby Supernova Factory, in Survey and Other Telescope Technologies and Discoveries, ed. Tyson, J. A.; Wolff, S. Proceedings of the SPIE, 4836, 61

Copin, Y. et al. 2007, this volume

Adelman, S. et al. 2003, Round table on instrumentation and data processing, Modelling of stellar atmo-spheres,Astronomical Society of the Pacific, 210, 337

Lantz, B. et al. 2004, SNIFS: a wideband integral field spectrograph with microlens arrays, in Optical Design and Engineering, ed. Mazuray, L., Rogers, P. J.; Wartmann, R. Proceedings of the SPIE, 5249, 146

B`eland, S., Boulade, O., & Davidge, T. 1988, Bulletin d’information du telescope Canada-France-Hawaii, 19, 16 Boulade, O. 1987, Bulletin d’information du telescope Canada-France-Hawaii, 17, 13

Cuillandre, J.-C. et al. 2002, CFHT’s SkyProbe: a real-time sky-transparency monitor, in Observatory Operations to Optimize Scientific Return III, ed. Quinn, P. J. Proceedings of the SPIE, 4844, 501

Short N., Fitetson W., Hale D., Townes C. 2003, Interferometry for optical astronomy, Proceedings of SPIE Vol. 4838(2003).