World-leading science with SPIRou - the nIR spectropolarimeter / high-precision velocimeter for CFHT
X. Delfosse, J.-F. Donati, D. Kouach, G. Hébrard, R. Doyon, E. Artigau, F. Bouchy, I. Boisse, A.S. Brun, P. Hennebelle, T. Widemann, J. Bouvier, X. Bonfils, J. Morin, C. Moutou, F. Pepe, S. Udry, J.-D. do Nascimento, S.H.P. Alencar, B.V. Castilho, E. Martioli, S.Y. Wang, P. Figueira, N.C. Santos, SPIRou Science Team
SSF2A 2013
L. Cambr´esy, F. Martins, E. Nuss and A. Palacios (eds)
WORLD-LEADING SCIENCE WITH SPIROU -THE NIR SPECTROPOLARIMETER / HIGH-PRECISION VELOCIMETER FOR CFHT
X. Delfosse , J.-F. Donati , D. Kouach , G. H´ebrard , R. Doyon , E. Artigau , F. Bouchy , I.Boisse , A.S. Brun , P. Hennebelle , , T. Widemann , J. Bouvier , X. Bonfils , J. Morin , C.Moutou , F. Pepe , S. Udry , J.-D. do Nascimento , S.H.P. Alencar , B.V. Castilho , E.Martioli , S.Y. Wang , P. Figueira , N.C. Santos and the SPIRou Science Team Abstract.
SPIRou is a near-infrared (nIR) spectropolarimeter / velocimeter proposed as a new-generation instru-ment for CFHT. SPIRou aims in particular at becoming world-leader on two forefront science topics, (i) thequest for habitable Earth-like planets around very- low-mass stars, and (ii) the study of low-mass star andplanet formation in the presence of magnetic fields. In addition to these two main goals, SPIRou will be ableto tackle many key programs, from weather patterns on brown dwarf to solar-system planet atmospheres,to dynamo processes in fully-convective bodies and planet habitability. The science programs that SPIRouproposes to tackle are forefront (identified as first priorities by most research agencies worldwide), ambitious(competitive and complementary with science programs carried out on much larger facilities, such as ALMAand JWST) and timely (ideally phased with complementary space missions like TESS and CHEOPS).SPIRou is designed to carry out its science mission with maximum efficiency and optimum precision.More specifically, SPIRou will be able to cover a very wide single-shot nIR spectral domain (0.98-2.35 µ m)at a resolving power of 73.5K, providing unpolarized and polarized spectra of low-mass stars with a ∼ UJF-Grenoble 1/CNRS-INSU, Institut de Plan´etologie et d’Astrophysique de Grenoble (IPAG) UMR 5274, 38041 Grenoble,France UPS-Toulouse / CNRS-INSU, Institut de Recherche en Astrophysique et Plan´etologie (IRAP) UMR 5277, Toulouse, F31400France Institut dAstrophysique de Paris, CNRS, Universit´e Pierre et Marie Curie, 98bis Bd. Arago, 75014 Paris, France D´epartement de physique and Observatoire du Mont-M´egantic, Universit´e de Montr´eal, C.P. 6128, Succursale Centre-Ville,Montr´eal, QC H3C 3J7, Canada Aix Marseille Universit´e, CNRS, LAM (Laboratoire dAstrophysique de Marseille) UMR 7326, 13388, Marseille, France Laboratoire AIM Paris-Saclay, CEA/Irfu Universit´e Paris-Diderot CNRS/INSU, 91191 Gif-sur-Yvette, France LERMA (UMR CNRS 8112), Ecole Normale Sup´erieure, 75231, Paris Cedex, France Paris Observatory, LESIA UMR 8109, Meudon, France LUPM, Universit´e Montpellier II, CNRS, UMR 5299, Place E. Bataillon, 34095, Montpellier, France Canada-France-Hawaii Telescope Corporation, 65-1238 Mamalahoa Hwy., Kamuela, HI 96743, USA Observatoire Astronomique de l’Universit´e de Gen`eve, 51 Ch. des Maillettes, 1290 Sauverny, Versoix, Suisse Departamento de F´ısica Te´orica e Experimental (DFTE), Universidade Federal do Rio Grande do Norte (UFRN), CP 1641,59072-970 Natal, RN, Brazil Departamento de F´ısica - ICEx - UFMG, Av. Antnio Carlos, 6627, 30270-901 Belo Horizonte, MG, Brazil Laborat´orio Nacional de Astrof´ısica/MCT, Rua Estados Unidos 154, 37504-364 Itajub´a, MG, Brazil Institute of Astronomy and Astrophysics, National Taiwan Univ., Taiwan Centro de Astrof´ısica, Universidade do Porto, Rua das Estrelas, 4150-762, Porto, Portugalc (cid:13)
Soci´et´e Francaise d’Astronomie et d’Astrophysique (SF2A) 2013 a r X i v : . [ a s t r o - ph . I M ] O c t SF2A 2013Requirement ValueSimultaneous Spectral Range full coverage from 0.98-2.35 m (YJHK bands)Resolving Power >
70K (goal 75K)RV Precision < >
100 per 2 km/s pixel in 1 hr at J=12 & K=11;bright limit: H < <
1) - faint limit: H ∼ >
70% & >
90% for 15 min & 1 hr visit respectivelySky Coverage up to airmass 2.5 (zenithal distance 70 ◦ ) Table 1.
Summary of SPIRou scientific requirements
The science programs SPIRou proposes to tackle are forefront (first priorities for most research agencies world-wide), ambitious (competitive and complementary with science programs carried out on much larger facili-ties, e.g., ALMA/ESO and JWST/NASA) and timely (ideally phased with complementary instruments, e.g.,TESS/NASA, CHEOPS/ESA and JWST/NASA). SPIRou plans to concentrate on two main scientific goals.The first one is to search for and characterize habitable exo-Earths orbiting very-low mass stars (vLMSs) usinghigh-precision radial velocity (RV) measurements. This search will expand the initial, exploratory studies car-ried out with visible instruments (e.g., HARPS/ESO) and will survey in particular large samples of stars mostlyout of reach of existing instruments. In particular, carrying out a new large- scale survey at nIR wavelengthswill boost the sensitivity to habitable exo-Earths by typically an order of magnitude on planetary mass (withrespect to existing instruments). SPIRou will also work in close collaboration with space- and ground- basedphotometric transit surveys like TESS/NASA, CHEOPS/ESA and ExTrA ∗ to identify the true planets amongthe candidates they will discover.The second main goal is to explore the impact of magnetic fields on star and planet formation, by detectingfields of various types of young stellar objects (e.g. class-I, -II and -III protostars, young FUor-like protostellaraccretion discs) and by characterizing their large-scale topologies. SPIRou will also investigate the potentialpresence of giant planets around protostars and in the inner regions of accretion discs. In particular, this studywill vastly amplify the initial exploration surveys carried out at optical wavelengths within the MaPP (MagneticProtostars and Planets) and MaTYSSE (Magnetic Topologies of Young Stars and the Survival of close-in massiveExoplanets) CFHT Large Programs (LPs). It will also ideally complement the data that ALMA/ESO has juststarted collecting on outer accretion discs and dense prestellar cores. SPIRou will also be able to tackle manyadditional exciting research topics in stellar physics (e.g., dynamos of fully convective stars, weather patternsof brown dwarfs), in planetary physics (e.g., winds and chemistry of solar-system planets), galactic physics (e.g.stellar archeology) as well as in extragalactic astronomy. We detail these goals below, giving in the main casesthe typical samples that need to be explored and their observational properties. SPIRou is designed to carry out its science mission with maximum efficiency and optimum precision. Morespecifically, SPIRou will be able to cover a very wide single-shot nIR spectral domain (0.98-2.35 µ m) at aresolving power of 73.5K, providing unpolarized and polarized spectra of low-mass stars with a ∼
15% averagethroughput and a RV precision of 1 m/s. Table 1 list the main scientific requirements of SPIRou, needed tocarry out most science goals detailled in Sec 3 and 4.The very-wide simultaneous spectral range, including in particular the K band, is crucial to SPIRou. Itmaximizes the instrument efficiency, both for the exoplanet programs - the K band totaling ∼
40% of the RVcontent in a full YJHK spectrum for an average M dwarf (see Fig 1) - and for the magnetic fields and star/planetformation themes - the relative spectropolarimetric weight of the K band reaching 60-70% given the increase of ∗ ExTrA is a recently funded ERC project whose aim is to detect transiting Earth-like planets around M-dwarfs from the ground.As a low resolution multiple-object spectrograph, ExTrA will allow extremely accurate photometry in narrow wavelength bands cience with SPIRou 3the Zeeman effect with wavelength as λ . Moreover, the K band is the only window to access class I embeddedprotostars, a key stellar sample to be explored for the first time with SPIRou. The wide spectral coverage andspectropolarimetric capabilities are also unique to SPIRou; other nIR RV instruments currently under planning(Carmenes on Calar Alto, IRD on Subaru, HZPF on the HET) do not cover beyond the H band nor include apolarimeter. The spectral resolution is also very important, not only to maximize the velocimetric efficiency,but also to ensure high enough spectropolarimetric sensitivity to Zeeman signatures; with a spectral resolutionof at least 73.5K, SPIRou is nearing optimal performances. Fig. 1.
RV rms photon-noise error (in m/s) vs rotational broadening ( v sin i ) for a M7 dwarf (2,700 K) and for thedifferent nIR bands (color curves), assuming a spectral resolution of 75K and a peak S/N of 160 per 2 km/s pixel. ARV precision of 1 m/s can be reached at low v sin i ’s. The K band (orange) is the main contributor to the RV precisionand contributes almost as much as all other bands (yellow). Without the K band, a twice longer exposure time would berequired to reach the same RV precision. Note that this estimate is likely pessimistic by potentially as much as a factorof 2, nIR synthetic spectra systematically under-estimating the strength of many molecular and even atomic features. The resulting instrument concept proposed for SPIRou is a direct heritage from previous successful instru-ments built by various members of the SPIRou project team: HARPS at the 3.6-m ESO telescope (Mayor et al.2003), ESPaDOnS at CFHT (Donati 2003) and SOPHIE at 1.93-m OHP telescope (Bouchy & Sophie Team2006; Bouchy et al. 2013). More specifically, SPIRou includes a cryogenic high-resolution spectrograph inspiredfrom the evacuated spectrograph of the HARPS velocimeter, a Cassegrain unit derived from the ESPaDOnSspectropolarimeter, a fiber-feed evolved from those of ESPaDOnS and the HARPS/SOPHIE velocimeters, anda Calibration/RV reference unit largely copied from those of SOPHIE and HARPS.2.1.1 The Cassegrain unitThe Cassegrain unit consists of 2 modules mounted (on top of each other) at the Cassegrain focus of thetelescope. The upper Cassegrain module includes an ADC correcting the entrance beam for the atmosphericrefraction and a tip-tilt module stabilizing the entrance image to better than 0.05” rms; this module alsoincludes a calibration wheel allowing to inject light from the calibration unit into the instrument. Beginningwith a circular instrument aperture of diameter 1.3”, the lower Cassegrain module mainly includes an achromaticpolarimeter made of two 3/4-wave dual ZnSe Fresnel rhombs coupled to a Wollaston prism, splitting the beaminto 2 orthogonal linear-polarization states. The 2 beams emerging from the beamsplitter are injected into 2 SF2A 2013separate fibers at polarimeter output2.1.2 The fiber link and pupil slicerThe fiber link conveys the light from the twin orthogonally polarized beams coming out of the Cassegrainpolarimeter into the cryogenic spectrograph. This link consists of a dual 35-m circular fluoride fiber custom-made with purified material to ensure a throughput of >
90% over the entire spectral range of SPIRou; thisfiber link also includes a pupil-slicer at spectrograph entrance to minimize injection losses without affectingthe spectrograph resolution. The last section of the fiber link includes a triple 90 µ m octogonal fiber (2 for thescience fibers and 1 for a simultaneous RV reference) ensuring a high scrambling of the near-field image is atleast 1000.2.1.3 The cryogenic spectrographThe high-resolution ´echelle spectrograph is bench-mounted, protected by one active and 3 passive thermalshields, and enclosed within a cryogenic dewar. Thanks to a dual-pupil design and an off-the-shelf commercialR2 ´echelle grating (w/ 23.2 gr/mm), the spectrograph can record the entire spectral range on a Hawaii 4RGdetector (15 µ m square pixels). The pixel size translates into an average spectral bin of 2.28 km/s and thespectrograph features a non-Gaussian instrumental profile yielding a spectral resolving power of 73.5K. Theoptical design of the spectrograph ensure a high total average throughput of 45% (detector included). Thespectrograph is cooled down to 80 K and thermally stabilized at a rms level of ∼ < < ∼
10 mK rms to ensure that spectral lines do not drift by more than0.25 m/s rms throughout one night. This thermalized FP unit (and possibly even the Th/U lamps) could bereplaced in the future with a nIR tunable laser comb (stable to < The SPIRou project team gathers a number of partners from different institutes and countries. More specifically,the team includes several institutes from France (IRAP and OMP in Toulouse, IPAG in Grenoble, OHP andLAM in Marseille, plus an extended science team from IAP / LESIA / CEA / LERMA / IAS / LUTH /LATMOS based in Paris and surroundings), from Canada (UdeM / UL in Montr´eal and Qu´ebec City, NRCin Victoria), from Switzerland (Geneva Observatory), from Taiwan (ASIAA in Taipei), from Brazil (LNA inItajuba, plus additional science contribution from UFRN / UFMG in Natal and Belo Horizonte), from Portugal(CAUP in Porto) and from CFHT.
One of the 2 main goals of SPIRou is to search for, and to characterize, exo-Earths orbiting low-mass stars- with a particular interest for planets located in the habitable zone (HZ) of their host stars. The study ofexoplanetary systems is one of the most exciting areas of astronomy today. Identifying habitable Earth-likeplanets and searching for biomarkers in their atmospheres is among the main objectives of this new century’sastronomy, motivating ambitious space missions (e.g., JWST, TESS, CHEOPS, EChO, PLATO). Among thevarious techniques developed to detect exoplanets, two are very efficient and complementary. Whereas RVstudies look for Doppler shifts induced by orbiting planets in the spectrum of their host stars, giving access tocience with SPIRou 5the planet mass, long-term photometric monitoring searches for regular occultations caused by planets transitingthe visible stellar disc, yielding the planet radius. For exoplanets detected with both techniques, one can estimatetheir densities and thus constrain their bulk compositions. Provided host stars are bright enough, one can evenprobe the outer atmosphere of transiting planets using transit spectroscopy, opening the new research field ofexoplanetology (Charbonneau et al. 2007).In this context, much interest has recently been focused on low-mass M dwarfs, around which habitablesuper-Earths are much easier to detect. To be considered potentially habitable, planets must be within theproper range of orbital distances where liquid water can be stable on their surface. This constraint also imposeslimits on the atmospheric pressure at the planet surface, and thus indirectly on the planet mass. The range oforbital distances for HZs also strongly depends on the mass (and thus on the temperature) of the host star,with lower temperatures moving HZs closer in. Habitable exo-Earths around M dwarfs are thus expected toproduce much larger RV wobbles (4 × to 8 × for M4 and M6 dwarfs, respectively) compared to the same planetorbiting a Sun-like star. A 1 m/s RV precision is sufficient to detect habitable telluric planets around M dwarfs- the much shorter orbital periods (of order of weeks) vastly decreasing the timescale over which observationsmust be collected; this is how the first likely- habitables super-Earths were discovered (Udry et al. 2007; Mayoret al. 2009; Delfosse et al. 2013; Bonfils et al. 2013b).Photometric transits are also much deeper for M dwarfs as a result of their smaller radii - by 11 × and 45 × forM4 and M6 dwarfs, respectively. A prime goal of the coming years is to discover Earths or super-Earths whoseatmosphere can be scrutinized and characterized with space missions (such as JWST and/or EChO) in the nextdecade. Since atmospheric characterization primarily requires as deep an atmospheric transit as possible on theone hand, and as bright a star as possible on the other hand (in the nIR, where absorption from atmosphericmolecules mostly concentrates), M dwarfs are optimal targets for this quest (Rauer et al. 2011). Today, only ahandful of very-bright transiting systems have been discovered up to now - most being giant gaseous planets -but many more are expected with forthcoming space missions like TESS or PLATO.Last but not least, statistical properties of planets around M dwarfs (compared to those around Sun-like stars)can provide key information on planetary formation, and in particular on the sensitivity of planet formation toinitial conditions in the protoplanetary disc (e.g., Ida & Lin 2005). That M dwarfs vastly dominate the stellarpopulation in the solar neighborhood and are likely hosting most planets in our Galaxy only makes this studyeven more crucial. Based on high-precision RV measurements, the SPIRou planet search we propose will greatly expand the currentexploratory studies carried out with existing visible velocimeters (e.g., HARPS@ESO, SOPHIE@OHP) by givingaccess to a large sample of stars inaccessible with existing instrumentation. The SPIRou planet search will inparticular build upon the success of the pioneering HARPS RV survey of M dwarfs (Bonfils et al. 2013a), whichdemonstrated that super-Earths with orbital periods <
100 d are more numerous around M dwarfs than aroundSun-like stars, with an occurrence frequency close to 90%; moreover, preliminary results suggest that about halfof these super-Earths are located in the HZs of their host stars. With existing velocimeters such as HARPS,RV measurements with a precision of 1 m/s are possible for only the ∼
100 brightest M dwarfs. This is clearlyinsufficient, either to have a realistic chance of detecting several transiting habitable super-Earths or to achievea proper statistical survey of rocky exoplanets around M dwarfs. Given their low temperatures, red and browndwarfs are much more accessible at nIR wavelengths (see Fig 2). In addition to a 1 m/s RV precision and a highthroughput, SPIRou offers the widest simultaneous nIR spectral coverage (0.98-2.35 µ m) yet available on anytelescope, making it optimally suited for carrying out efficient, systematic RV exoplanet surveys of M dwarfs.SPIRou will also crucially contribute to the forthcoming extensive photometric surveys of transiting planetsaround M dwarfs, either from space (e.g., TESS, CHEOPS, PLATO) or from the ground (e.g., ExTrA). Spec-troscopy is indeed mandatory to discard false detections (e.g., background eclipsing binaries), to establish theplanetary nature of all transiting objects detected around low-mass dwarfs through photometric monitoring andto measure their mass from RV measurements. A high-precision velocimeter working in the nIR will thus beessential to monitor all candidates detected with ground and space photometers around bright M dwarfs, andin particular around late-M ones, hardly accessible to velocimeters working in the visible. A nIR spectrographwill also usefully contribute to the quest for close-in transiting exo-Earths around bright M dwarfs through asystematic survey prior to any photometric observations.More specifically, SPIRou will contribute to exoplanet science along 3 main avenues, that we foresee as theprime exoplanet themes of the SPIRou planet search. SF2A 2013 Fig. 2.
Photon distribution (per 2 km/s velocity bin size) for a M6 (3,100 K, red) and M8 (2,300 K, green) dwarfs at10 pc (derived from NextGen models (Allard et al. 1997). M6 and M8 dwarfs respectively produce ∼
30 and ∼ ∼
300 super-Earths, is certainly very promising. Since(i) most Earths and super-Earths detected with TESS will orbit around M dwarfs, and (ii) less than ∼
30% ofthem will be accessible to optical RV follow-ups (Deming et al. 2009), SPIRou will be the best RV instrumentto monitor in the nIR the ∼
150 best candidates visible from CFHT, to confirm or reject their planetary natureand to determine their masses.Monte Carlo simulation show that with ∼
60 visits per star and with S/N ∼
160 spectra per visit SPIRou hasthe capacity to validate and characterize planets of Earth Mass, orbiting mid-M dwarfs with a period of ∼
30 d.This observational effort requires a total of 150 CFHT nights.3.2.2 RV survey of a large sample of M-dwarfsAs TESS will majoritarly operate on 27-d windows of continuous monitoring for most stars, the majority ofplanet candidates showing at least 2 transits will have periods <
20 d and will not be located in the HZ oftheir host stars. For planets with longer periods, and in particular for those located in the HZ, RV-drivenplanet searches will be more efficient. Our Monte Carlo simulations demonstrate that (see Fig 3), with a surveyfocussing on ∼
600 M dwarfs (requiring 600 CFHT nights for >
60 visits per stars), SPIRou could potentiallydetect ∼
450 new exoplanets, ∼
300 being less massive than 5 M ⊕ ; among the latter sample, ∼
50 would beorbiting in the HZ and ∼
15 would be transiting, while ∼ η ⊕ , the faction of habitable planets in the Solar neighborhood, with an accuracy of < P l ane t m a ss ( M ea r t h ) Fig. 3.
Planets found with a 600-target survey according to our Monte Carlo simulation. Filled circled indicate detectedplanets, open circles undetected ones and red circles (both filled and open) represent transiting planets. Blue lines shownotional limits for the habitable zone, both in mass and temperature. Most planets with > ⊕ in the habitable zoneare detected, including one transiting. Interestingly, a sample of sub-M ⊕ planets with T eq >
350 K is also detected. with JWST or EChO.3.2.3 Occurrence frequency of planets around M dwarfsBy expanding the sample by 10 × (with respect to the existing optical surveys of M dwarfs) and thus by bringinga 3 fold improvement in the statistics of planet properties, the SPIRou observations outlined in the 2 first itemsof our planet search will provide much more reliable constraints on planet formation models. Moreover, byextending the RV monitoring on a selected sample of M dwarfs and on a larger time span, SPIRou will likelyreveal additional bodies in most systems at larger period. This extended monitoring, not included in the firstpart of RV of our planet search, will be carried out on the ∼
350 most interesting M dwarfs with detected planets/ systems and will require an additional amount of 250 CFHT nights to achieve 40 more visits per star.
The other main goal of SPIRou is to explore the impact of magnetic fields on star and planet formation, bydetecting and characterizing magnetic fields of various types of young stellar objects (e.g., classical T Tauristars, embedded class-I protostars, young protostellar accretion discs). This quest will expand the pioneeringsurveys carried out in the framework of the study with optical spectropolarimeter, mainly [email protected] how Sun-like stars and their planetary systems form comes as a logical addition to the directobservation of exoplanets. Within the last decades, this research field underwent major observational andtheoretical advances, for instance by clarifying the crucial role of magnetic fields, not only on the gravitationalcollapse of giant molecular clouds (e.g., Hennebelle & Fromang 2008), but also on the formation of accretiondiscs and pre-stellar cores (e.g., Hennebelle & Teyssier 2008) from which stars and their planetary systems areborn.At an age of ∼ ∼
15 cTTSs, the large-scale magnetic topologies that link low-mass protostars to their accretion discs, andto demonstrate that this topology strongly relates to the internal structure of the protostar, and thus to bothits age and mass (e.g., Donati et al. 2010). When the protostar is young enough and has a low-enough mass tobe fully-convective, its large-scale magnetic topology is dominated by a strong dipolar-like field roughly alignedwith the stellar rotation axis - thereby providing a quantitative explanation of the physical star/disc couplingmechanism through which the protostar is strongly spun down (e.g., Zanni & Ferreira 2013). These observationsare however still rather sparse as a result of the relative faintness of cTTSs (at optical wavelengths); moreover,younger class-I protostars (with ages < Fig. 4.
The Orion nebula as seen in visible (left) and IR (right) light (ESO/VISTA). In the IR, dust cocoons aroundyoung stars are much more transparent.
With a much higher magnetic sensitivity than ESPaDOnS, thanks to both the increased nIR brightnessof protostars (especially in K, see Fig 4) and the enhanced Zeeman effect at larger wavelengths, SPIRou willprovide a much deeper and more systematic access to large-scale fields of class-I and -II protostars. Morespecifically, it will allow (i) to survey a 5-10 × larger sample of cTTSs than the very limited one currentlyaccessible with ESPaDOnS, and (ii) to extend for the first time this study to the brightest class-I protostarsthanks to the K band coverage. The suggested SPIRou survey will ideally complement ALMA observations ofpre-stellar (class-0) cores and of their magnetic fields, and will thus bring one of the key missing pieces in ourunderstanding of star / planet formation.cience with SPIRou 9SPIRou will also have the power to detect hot Jupiters orbiting around more-evolved class-III protostars (theso-called “weak-line T Tauri” stars or wTTSs) and thus to verify whether close-in giant planets are either muchmore or much less frequent around low-mass protostars than around mature, Gyr-old Sun-like stars. Theseobservations will thus yield a direct observational test of the formation and migration of hot Jupiters, allowingto estimate the relative fraction produced through disc migration (acting during the formation stage) and thatattributable to interactions / scattering (occurring much later). Finally, SPIRou will also be able to observethe innermost regions of protostellar accretion discs, out of reach of ALMA, to detect and characterize theirmagnetic fields and to identify the potential presence of migrating hot Jupiters (e.g., Donati et al. 2005; Powellet al. 2012). Fig. 5.
JHK magnitude histograms of class-I (left) & -II (right) protostars in Taurus (up) and ρ Oph (bottom). Class-Iprotostars are mostly brighter than 12 in K only (as a result of obscuration, especially in ρ Oph), whereas a significantfraction of class-II protostars are accessible in all 3 bands.
With a survey carried out on 5 of the most accessible star forming regions (e.g., Taurus/Auriga, TW HyaAssociation, ρ Ophiuchius, Lupus, Orion Nebula Cluster; see Fig 5), SPIRou can detect for the first time thelarge-scale fields of ∼
50 embedded class-I protostars, bringing yet unknown information on how magnetosphericaccretion operates at so early a step in the formation process. In addition to this, SPIRou will be able tomonitor ∼
200 class-II and -III protostars (cTTSs and wTTSs), expanding the pioneering ESPaDOnS survey by5-10 × into full-scale surveys of magnetic protostars and their close-in giant planets. This survey will requirea total of 250 CFHT nights; on top of this, monitoring a small sample of ∼
10 protostellar accretion discs willrequire an additional 50 nights.
In addition to the two main goals, SPIRou will be a very innovative and efficient instrument for tackling manymore science themes. A few of them are briefly outlined below.0 SF2A 2013
Using the spectropolarimetric data collected for the exoplanet survey of M dwarfs, SPIRou can also studythe large-scale dynamo fields of fully convective dwarfs. These magnetic fields are indeed the main source oftheir activity and therefore a potential drawback for the habitability of their planets (Lecavelier des Etangset al. 2012; Vidotto et al. 2013); studying dynamos of fully convective bodies can also be very informative onmagnetic fields of Earth-like exoplanets (e.g., Christensen et al. 2009; Reiners & Christensen 2010) and usefullycomplement direct sensitive radio observations (e.g., LOFAR, Zarka 2010), with the ultimate aim of workingout whether magnetic fields of exoplanets can improve their habitability.Published studies of large-scale fields of fully-convective dwarfs have already demonstrated that these mag-netic topologies are very sensitive to the aspect ratio of the convective zone and even suggest, for very-low massdwarfs, a bistable behavior of the underlying dynamo processes potentially similar to that invoked for planetarydynamos (Morin et al. 2008, 2010, 2011). By comparing magnetic topologies of M dwarfs with theoretical pre-dictions and results of numerical simulations, and by trying to generalize these results to planetary dynamos,one should ultimately be able to better understand the physical processes capable of amplifying and sustaininglarge-scale magnetic fields in both fully-convective dwarfs and planets (e.g., Schrinner et al. 2012). Using datafrom the exoplanet survey, SPIRou will provide a thorough census of magnetic topologies of M dwarfs (and inparticular of fully-convective ones) that will usefully guide theorists towards more realistic, generalized dynamomodels in better agreement with observations of both stellar and planetary large-scale fields.
SPIRou can also very efficiently contribute to atmospheric studies of telluric or giant planets, whether or notthey belong to the solar system.In the case of the solar system, SPIRou will be able to carry out spatially-resolved, detailed spectroscopicstudies of the chemical composition (at both low and high altitudes), of the wind dynamics and of the auroralemission of planetary atmospheres. These studies will allow in particular to better understand the complexinteractions between atmospheric volatiles, planetary interiors, surfaces and climates (e.g., B´ezard et al. 2009);they can also accurately estimate wind velocities at different atmospheric locations and altitudes, as well astheir temporal variability (e.g., Widemann et al. 2008). Finally, auroral emission (and polarization) can informon potential links between planetary atmospheres and magnetospheres. Thanks to its wide spectral domain(including the K band), to its high RV precision and to its spectropolarimetric capabilities, SPIRou will be ableto very significantly contribute to chemical and dynamical studies of solar-system planet atmospheres.Exoplanet atmospheres are obviously much more elusive and tricky to detect and to characterize. In theparticular case of transiting exoplanets, atmospheres can be scrutinized either by transmission during a planetarytransit, or by occultation during a planetary eclipse. For close-in planets (and in particular hot Jupiters), itcould be possible to detect from the ground the spectral contribution of the star-lit side of the planet, bymonitoring the Doppler shift (induced by the planet orbital motion) of specific atmospheric species (e.g., CO,Snellen et al. 2010). SPIRou will thus be able to contribute to this quest in a original way, thanks to its widespectral domain and to the K band in particular (that includes a number of key atmospheric markers).
Studying the atmospheres of ultra-cool L and T brown dwarfs (BDs) is yet another obvious research field forSPIRou. Despite huge modeling efforts invested since the discovery of BDs, many questions remain open - likelyrelated to the complex physics of BDs atmospheres and in particular to mechanisms of dust clearing throughspecific weather patterns occurring in their atmospheres (e.g., Radigan et al. 2012). The disc-averaged spectralenergy distribution (SED) of ultra-cool BDs may indeed not be representative of any single region, and thuscannot be modeled using a unique set of physical parameters (e.g., temperature, dust-settling, grain properties).Evidence that this is likely the case comes from the fact that ultra-cool BDs often exhibit photometric variabilityat a level of a few % up to a remarkable 25% (Artigau et al. 2009). This variability is apparently due to acombination of rotational modulation and intrinsic evolution on short timescales, likely caused by weather-likeclearings in the dust- cloud deck (e.g., Littlefair et al. 2006). Unravelling the physics of these weather patternsand distinguishing between the several theoretical scenarios requires observations capable of localizing the dustclouds in BD atmospheres and following their rapid evolution with time.cience with SPIRou 11Doppler imaging through high-resolution nIR spectroscopy is a very attractive and viable approach to mapweather patterns of BDs; most BDs are indeed rapid rotators and often exhibit rotational modulation, bothphotometrically and spectroscopically, providing ideal conditions for Doppler imaging. Though not yet appliedto stars cooler than mid-M due to their intrinsic faintness at optical wavelengths, Doppler imaging of BDs inthe nIR is perfectly feasible with SPIRou, opening a new window for studying chemical inhomogeneities in theiratmospheres. SPIRou will be able to monitor ∼
10 L and T BDs among the best suited for this experiment, fora total amount of 30 nights.
The amount of observing time required to complete the two main science goal is large (1300 nights). In thiscontext, SPIRou only makes sense if coupled to a SPIRou Legacy Survey of 500 CFHT nights on a timescaleof ∼ SPIRou has successfully passed the preliminary design phase (PDR) in October 2012 and is now in the finaldesign phase. Provided SPIRou is selected by CFHT and succeeds the upcoming final design review (FDR), itshould be installed on the 3.6-m CFH telescope in early 2017 after the following phases : • early 2014: design validation (FDR) • • • References
Allard, F., Hauschildt, P. H., Alexander, D. R., & Starrfield, S. 1997, ARA&A, 35, 137Artigau, ´E., Bouchard, S., Doyon, R., & Lafreni`ere, D. 2009, ApJ, 701, 1534B´ezard, B., Tsang, C. C. C., Carlson, R. W., et al. 2009, Journal of Geophysical Research (Planets), 114, 0Bonfils, X., Delfosse, X., Udry, S., et al. 2013a, A&A, 549, A109Bonfils, X., Lo Curto, G., Correia, A. C. M., et al. 2013b, A&A, 556, A110Bouchy, F., D´ıaz, R. F., H´ebrard, G., et al. 2013, A&A, 549, A49Bouchy, F. & Sophie Team. 2006, in Tenth Anniversary of 51 Peg-b: Status of and prospects for hot Jupiter studies, ed.L. Arnold, F. Bouchy, & C. Moutou, 319–325Bouvier, J., Alencar, S. H. P., Harries, T. J., Johns-Krull, C. M., & Romanova, M. M. 2007, Protostars and Planets V,479Charbonneau, D., Brown, T. M., Burrows, A., & Laughlin, G. 2007, Protostars and Planets V, 701Christensen, U. R., Holzwarth, V., & Reiners, A. 2009, Nature, 457, 167Delfosse, X., Bonfils, X., Forveille, T., et al. 2013, A&A, 553, A8Deming, D., Seager, S., Winn, J., et al. 2009, PASP, 121, 952Donati, J.-F. 2003, in Astronomical Society of the Pacific Conference Series, Vol. 307, Solar Polarization, ed. J. Trujillo-Bueno & J. Sanchez Almeida, 41Donati, J.-F., Paletou, F., Bouvier, J., & Ferreira, J. 2005, Nature, 438, 466Donati, J.-F., Skelly, M. B., Bouvier, J., et al. 2010, MNRAS, 409, 1347Hennebelle, P. & Fromang, S. 2008, A&A, 477, 9Hennebelle, P. & Teyssier, R. 2008, A&A, 477, 25Ida, S. & Lin, D. N. C. 2005, ApJ, 626, 10452 SF2A 2013