The SPIRou wavelength calibration for precise radial velocities in the near infrared
M. J. Hobson, F. Bouchy, N. J. Cook, E. Artigau, C. Moutou, I. Boisse, C. Lovis, A. Carmona, X. Delfosse, J.-F. Donati, SPIRou Team
AAstronomy & Astrophysics manuscript no. main © ESO 2021February 5, 2021
The SPIRou wavelength calibration for precise radial velocities inthe near infrared (cid:63) , (cid:63)(cid:63) M. J. Hobson , , F. Bouchy , N. J. Cook , E. Artigau , C. Moutou , , I. Boisse , C. Lovis , A. Carmona , X.Delfosse , J.-F. Donati , and the SPIRou Team Aix Marseille Univ, CNRS, CNES, LAM, Marseille, Francee-mail: [email protected] Millennium Institute for Astrophysics, Chile Observatoire Astronomique de l’Université de Genève, 51 Chemin des Maillettes, 1290 Versoix, Switzerland Institut de Recherche sur les Exoplanètes (IREx), Département de Physique, Université de Montréal, C.P. 6128, Succ. Centre-Ville,Montréal, QC, H3C 3J7, Canada Univ. de Toulouse, CNRS, IRAP, 14 avenue Belin, 31400 Toulouse, France Canada-France-Hawaii Telescope Corporation, 65-1238 Mamalahoa Hwy, Kamuela, HI 96743, USA Univ. Grenoble Alpes, CNRS, IPAG, 38000 Grenoble, FranceReceived 13 May 2020, accepted 18 January 2021
ABSTRACT
Aims.
SPIRou is a near-infrared (nIR) spectropolarimeter at the CFHT, covering the YJHK nIR spectral bands (980 − Methods.
We make use of a UNe hollow-cathode lamp and a Fabry-Pérot étalon to calibrate the pixel-wavelength correspondence forSPIRou. Di ff erent methods are developed for identifying the hollow-cathode lines, for calibrating the wavelength dependence of theFabry-Pérot cavity width, and for combining the two calibrators. Results.
The hollow-cathode spectra alone do not provide a su ffi ciently accurate wavelength solution to meet the design requirementsof an internal error of < .
45 m s − , for an overall RV precision of 1 m s − . However, the combination with the Fabry-Pérot spectraallows for significant improvements, leading to an internal error of ∼ .
15 m s − . We examine the inter-night stability, intra-nightstability, and impact on the stellar RVs of the wavelength solution. Key words.
Astronomical instrumentation, methods and techniques – Instrumentation: spectrographs
1. Introduction
The spectroscopic method of exoplanet detection, also known asthe radial velocity (RV) method, has proven itself extremely pro-ductive, with over 700 exoplanets discovered by this method .The detection of exoplanets through spectroscopy relies on theextremely precise measurement of tiny shifts of the stellar spec-tral lines. In order to make these measurements, a precise wave-length solution is in turn required. (cid:63) Based on observations obtained at the Canada-France-Hawaii Tele-scope (CFHT), which is operated from the summit of Maunakea bythe National Research Council of Canada, the Institut National desSciences de l’Univers of the Centre National de la Recherche Scien-tifique of France, and the University of Hawaii. The observations at theCanada-France-Hawaii Telescope were performed with care and respectfrom the summit of Maunakea which is a significant cultural and his-toric site. Based on observations obtained with SPIRou, an internationalproject led by Institut de Recherche en Astrophysique et Planétologie,Toulouse, France. (cid:63)(cid:63)
Table A1 is only available in electronic form at the CDS via anony-mous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http: // cdsweb.u-strasbg.fr / cgi-bin / qcat?J / A + A / See e.g. The Exoplanets Encyclopaedia ( http://exoplanet.eu/ ),the NASA Exoplanet Archive ( https://exoplanetarchive.ipac.caltech.edu/ ). Historically, two main instrumental approaches have beenused for wavelength calibration: iodine cells and hollow-cathode(HC) lamps. Each of these calibrators has advantages and disad-vantages. With iodine cells, the calibrating spectrum is imprinteddirectly on the stellar spectrum, allowing for a fully simultane-ous calibration (e.g. Marcy & Butler 1992; Butler et al. 1996).However, the iodine lines span only a small portion of the op-tical spectrum (around 510 to 620 nm, Fischer et al. 2016), anda high signal-to-noise ratio (S / N) is required to model the linespread function. With HC lamps (primarily ThAr in the visi-ble), calibration exposures must be taken before the stellar ob-servations, and a separate fibre is required to monitor the in-strument drift from the time of calibration (e.g. Baranne et al.1996). The advantage of these lamps is that the thorium linescover a much larger wavelength domain than the iodine lines,and - as they are not superimposed on the stellar spectrum -fainter targets can be observed. An overview of the main spec-trographs operating in the visible, their wavelength calibrators,and measurement precisions, is given in Fischer et al. (2016).For iodine cells, the precision is limited to ≈ − (Fischeret al. 2016, Spronck et al. 2015). As for the ThAr HC lamps,Lovis & Pepe (2007) were able to achieve a groundbreaking20 cm s − precision for the wavelength solution of the HARPSspectrograph through the creation of an improved line list. Forthe near-infrared (nIR), however, HC lamps alone limit preci- Article number, page 1 of 12 a r X i v : . [ a s t r o - ph . I M ] F e b & A proofs: manuscript no. main sion to above 1 m s − (Halverson et al. 2014), and the fill gasesemit bright lines that saturate the detectors (Quirrenbach et al.2018).In order to increase the precision of the wavelength solu-tion and enable the detection of smaller planets, Fabry-Pérot(FP) étalons and laser frequency combs (LFCs) have recentlybegun to be incorporated into wavelength solutions. Fabry-Pérotétalons provide lines that are evenly spaced in frequency, butwhose wavelengths need to be derived by anchoring to an abso-lute calibrator, such as an HC lamp (e.g. Bauer et al. 2015). Asan example, the HARPS spectrograph can achieve 10 cm s − pre-cision over one night through the incorporation of the FP lines(Wildi et al. 2011). Likewise, the CARMENES spectrograph in-corporates FP observations into the wavelength calibrations forboth its visible and nIR arms; for the nIR arm especially, it is anecessity due to the less densely populated emission lines andstrongly saturated gas lines in the HC spectra (Quirrenbach et al.2018). Laser frequency combs, on the other hand, provide evenlyspaced (in frequency) lines with wavelengths that are known tovery high accuracy (e.g. Murphy et al. 2007); their main limita-tion is the wavelength span they can cover (e.g. Co ffi net et al.2019, who describe the HARPS LFC, which currently spansaround three quarters of the HARPS domain). Some nIR spectro-graphs currently use LFCs as their primary calibrators, such asthe Habitable Zone Planet Finder (HPF Halverson et al. 2014)and the InfraRed Doppler (IRD) spectrograph (Kokubo et al.2016; Kotani et al. 2018); however, the wavelength coverage ofthese spectrographs is smaller than that of SPIRou, not extendingas far into the red. Finally, LFCs are also much more costly thanFP étalons, and are still maturing as a technology (HPF and IRDboth employ HC lamps and FP étalons as back-up calibrators).An LFC covering the 1.0 - 2.2 micron range has recently beeninstalled on SPIRou and is in a testing and optimisation phase,but is not currently part of the standard calibrations (Donati et al.submitted).In this article, we describe the development of the SPIRouwavelength solution, using a UNe HC lamp and an FP étalon.Section 2 presents the SPIRou wavelength calibrators and thedatasets selected for testing. Section 3 describes the di ff erent ap-proaches tested. The performances of the di ff erent wavelengthsolution scripts are analysed in Sect. 4. Finally, we summariseand conclude in Sect. 5.
2. Data
SPIRou (SpectroPolarimetre InfraRouge) is an nIR spectropo-larimeter, mounted on the 3.6 m Canada-France-Hawaii Tele-scope (CFHT), which began operations in February 2019. Anoverview of the optical and mechanical design is is given in Ar-tigau et al. (2014), Donati et al. (2018) and Donati et al. (sub-mitted) . SPIRou’s nominal spectral range, on which the designwas optimised, is the 980 − − R ≈
70 000 ± × See also the project website, http://spirou.irap.omp.eu ,and the CFHT instrument page, a given polarisation state, either circular or linear), and one forcalibration.The SPIRou calibration unit has been previously described inBoisse et al. (2016), and Perruchot et al. (2018). The calibrationmodule permits the illumination of science, calibration, or allchannels simultaneously, by any of the following: a cold source(Black Acktar surface at -25 ◦ C) for observation of faint starswithout simultaneous calibration; a white lamp (Tg) for blazemeasurement; one of two HC lamps, UNe or ThAr, for wave-length calibration; an FP étalon for wavelength calibration andsimultaneous drift monitoring; or a reserve port intended for fu-ture upgrades or visitor instruments.The wavelength calibrators are the HC lamps and the FPétalon. Figures 1 and 2 show the central regions of raw SPIRouspectra of the UNe lamp and the FP étalon, respectively. Eachof these calibrators has advantages and disadvantages. Either ofthe HC lamps can provide an absolute calibration, through iden-tifiable catalogued lines; however, the lines are unevenly spacedand vary significantly in flux. Meanwhile, FP lines do not havefixed absolute wavelengths but must be anchored to another cali-brator; however, once a first absolute calibration is obtained fromthe HC lamp, the multitude of evenly spaced FP lines across theentire detector enable a refinement of the wavelength solution.
Fig. 1.
Raw SPIRou spectrum of the UNe HC lamp (zoom to the centralregion). Light from the lamp is being fed to both the science and cal-ibration fibres. Around eight orders can be seen, centred around 1560nm. Three ’ghost’ lines are also visible, corresponding to contaminationfrom strongly saturated lines in other orders.
SPIRou aims for a target RV precision of 1 m s − . In orderto achieve this precision, the error budget requires for the wave-length solution an internal error of < .
45 m s − . At the start of SPIRou commissioning, the only available cat-alogue of UNe lines in the nIR was that of Redman et al.(2011) (hereafter R11), covering the 850 − − − ). This catalogue wastherefore used in the initial development of the HC wavelength Article number, page 2 of 12. J. Hobson et al.: The SPIRou wavelength calibration for precise radial velocities in the near infrared
Fig. 2.
Raw SPIRou spectrum of the FP étalon (zoom to the same centralregion as Fig. 1). Light from the lamp is being fed to both the scienceand calibration fibres. Around eight orders can be seen, centred around1560 nm. solution during validation tests. In 2018, however, Sarmientoet al. (2018) (hereafter S18) published an updated catalogue ofU lines in the 500 − − ) for a total of 13554 lines in acombined R11 + S18 catalogue.For the ThAr lamp, the only catalogue that covered theSPIRou domain at the time was that of Redman et al. (2014)(median uncertainty: 0.12 pm, translating to a median RV uncer-tainty per line of 7 m s − ). The UNe catalogue has far more linesthan the ThAr catalogue, even before the incorporation of thelines from S18 (9767 vs 1587 in the SPIRou wavelength range).The laboratory tests performed at Toulouse, which are de-scribed in Perruchot et al. (2018), show that the UNe spectrahave around four times more identifiable lines than the ThArspectra, resulting in consistently more accurate and stable wave-length solutions throughout the tests. Additionally, the ThArspectra show many more, and more strongly, saturated lines thanthe UNe spectra in the SPIRou wavelength range. Saturated linesare useless for fitting a wavelength solution, since their cen-tres cannot be precisely determined. They will also ’bleed’ intoneighbouring orders and contaminate them. Additionally, persis-tence, which is a known problem for CMOS detectors such as theH4RG (Artigau et al. 2018; Bechter et al. 2019), is stronger forsaturated lines, contaminating subsequent observations. There-fore, the UNe lamp was adopted as the primary absolute wave-length calibrator. To showcase the performance of the di ff erent methods devel-oped, we applied all the scripts on the calibrations of a two-week Fig. 3.
Comparison of wavelengths for common lines between the cata-logues of Redman et al. (2011) (R11) and Sarmiento et al. (2018) (S18).The blue points represent the di ff erence in wavelengths, the black linesthe S18 error bars. Points without error bars indicate that S18 reportedno uncertainty for that line. The di ff erences in wavelength are generallywithin the error bars when known, and always small. SPIRou run in February 2019 . The selected calibrations consistof pairs of 1 HC (UNe) spectrum and 1 FP spectrum. The manyinstrument changes over the comissioning and early science runsprevent us from making longer-term comparisons.
3. Methods
The wavelength solution is derived for each night as part of AP-ERO (A PipelinE to Reduce Observations); APERO is main-tained and version-controlled on github . For the remainder ofthis work we will refer to the SPIRou specific part of APERO asthe SPIRou data reduction system (DRS). The work presentedhere was carried out primarily with version 0.5.000, which wasreleased on 10 May 2019. The main di ff erences with the currentdevelopment version are noted in Sect. 5, and will be describedin a forthcoming paper (N.J. Cook et al. in prep).A full description of the DRS is beyond the scope of thiswork, and will be done in N.J. Cook et al. (in prep). However, webriefly summarise the calibration processing sequence for ver-sion 0.5.000: pre-processing to remove certain detector e ff ectssuch as the amplificator crosstalk; generation of a dark calibra-tion; creation of a map of the bad pixels, including two smallholes and a scratch on the detector; localisation of the orders forfibres A, B, and C; mapping of the slit profile across the detector;creation of the blaze profiles for all extracted fibres; generationof the wavelength map for all extracted fibres. Two main methods were developed and tested for the wavelengthsolution based on the HC lamp. The principal conceptual dif-ference between them lies in the way in which the HC lines All SPIRou calibrations are publicly available via the Cana-dian Astronomy Data Centre, at Located at https://github.com/njcuk9999/apero-drs
Article number, page 3 of 12 & A proofs: manuscript no. main are identified. The first, method HC1, is analogous to the SO-PHIE / HARPS wavelength solution (Baranne et al. 1996): Foreach line in the catalogue, the region where it should be locatedis selected and a Gaussian fit is attempted, with poor fits beingdiscarded. In the second, method HC2, Gaussians are fitted toevery peak in the HC spectra, and the best match to the cata-logue is identified. The detailed structure of the full routine is asfollows.
Data identification and reading:
The input files are verifiedvia FITS header keys to be extracted two-dimensional spectra(E2DS) corresponding to an HC lamp. The E2DS files con-tain the extracted spectrum for a single fibre in the format of a50 × Calibration set-up:
The calibration files that are closest in timeto the input file(s) are copied from the calibration database (in-cluding the previous wavelength solution). The previous wave-length solution is read and checked for compatibility with thecurrent parameter set-up (degree of the polynomial fits em-ployed, as this varied during development). The correct HC linecatalogue (UNe or ThAr) is read in.
Identification of the HC lines:
The lines are identified followingone of the procedures outlined above:
Method HC1:
For each line in the catalogue, a 4 σ region inwavelength (where σ is the expected width of the line, computedfrom the central wavelength and the average spectral resolution)around its expected position is selected, based on the previouswavelength solution. A Gaussian fit to the region is attempted,and kept if the σ of the fitted Gaussian is less than 4 km s − (todiscard very broad, flat lines whose centres will be imprecise),and the amplitude is less than 1e8 (to avoid fitting on saturatedlines; we note that the amplitude is dimensionless as the E2DSfiles are blaze-corrected within the routine). Method HC2:
For each spectral order, a 13-pixel-wide win-dow is moved along the order in four-pixel shifts. If the max-imum flux is at least at a 2 σ level above the local RMS andis within four pixels of the centre of the segment, a Gaussianfit is attempted. The Gaussian is kept if the residual of the fitnormalised by the peak value is between 0 and 0.2 (to ensurethe fit is a good representation of the spectrum in that region),and the Gaussian σ in pixels, assuming a FWHM of 2 pixels,is between 0.7 and 1.1 (to discard narrow cosmic rays or broadblended lines). Once all the peaks in the order are identified, the20 brightest (which are generally the least likely to be spurious,and most likely to correspond to catalogued lines) are selectedand the closest catalogue line to each one is identified. Then, allpossible three-line combinations are used to fit a second-orderpolynomial to the order, test wavelengths are calculated for alllines using the polynomial fit, and the velocity o ff set for eachfrom its identified catalogue line is calculated. The polynomialwith the most lines within 1 km s − of the catalogue is held tobe the correct identification, and all peaks within 1 km s − of thecatalogue are kept.The parameter values were set and refined over the courseof the pipeline development in order to obtain the most accurate and stable wavelength solutions. For method HC1 in particular,the SOPHIE / HARPS parameters were used as first guesses, andmodified as necessary. The fitted Gaussians to the HC lines havea median FWHM of 2 . .
01 nm at the blue end of the spectrum to0 .
05 nm at the red end.
Fitting the solution:
With the HC lines identified, a fourth-orderpolynomial fit is performed for each order between the pixel po-sitions given by the centres of the Gaussians, and the cataloguedwavelengths. In method HC2, continuity of the polynomial coef-ficients across the orders is also imposed at this step (attempts toimpose similar cross-order continuity for method HC1 resultedin less stable solutions, so they were discarded).
Littrow solution check:
The Littrow check is a verification ofthe cross-order continuity of the wavelength solution, followingthe same principles as in Cersullo et al. (2019). It is evaluatedfor several pixel positions (currently every 500 pixels along eachorder in the dispersion direction). For each position x, a fourth-order polynomial fit is performed between the inverse echelleorders and the fractional wavelength contribution at x for eachorder (normalised by the wavelength for the first order), with aniterative step to remove the largest outlier. For an ideal spectro-graph, the residuals to these fits would be zero. The polynomialfits, the residuals, and the minimum, maximum, and rms valuesof the residuals are stored. An example is shown in Fig. 4.
Fig. 4.
Example Littrow solution check, showing the residuals of cross-order fits between the inverse echelle orders and the fractional wave-length contribution. Each line corresponds to a di ff erent pixel position.It should be noted that for the bluest orders, there is very little flux atpixels 500 and 3500. Extrapolation of the reddest orders:
For the last two orders,very few HC lines are catalogued (13 in the last order, wave-length range 2438-2516 nm, and 68 in the penultimate, wave-length range 2362-2437 nm, compared to an average of 300 forthe rest) and even fewer are identified (around ten in the lastorder and 25 in the second-to-last for method HC2). This meansthat the fitted solution often fails or is highly unstable. Therefore,for these orders it is not fitted from the HC lines, but extrapolatedfrom the Littrow solution: The cross-order polynomial fits are
Article number, page 4 of 12. J. Hobson et al.: The SPIRou wavelength calibration for precise radial velocities in the near infrared used to generate pixel-wavelength pairs at the Littrow evaluationpositions, and these values are used in turn to fit a fourth-orderpolynomial for each spectral order. A possible alternative wouldbe to use telluric lines to fit these two orders; however, workby Figueira et al. (2010) using CRIRES suggests the precisionwould not be better than 5-10 m s − . Quality controls:
The structure of an E2DS file (one order perrow, wavelength increasing along the order) means that alongeach column, the wavelength must be increasing - that is, eachpixel of order N must have a smaller wavelength value than thecorresponding pixel of order N +
1. A first rough quality check,therefore, verifies that this is in fact the case (when this fails, itgenerally points to a problem with the slit shape or order iden-tification calibrations). A second quality check is applied to theminimum, maximum, and rms values of the Littrow check resid-uals: The minimum and maximum must be within 0 . − ,the rms within 0 . − . Logging statistics:
The mean and rms of the deviation from thecatalogued lines in m s − is logged, as is the total number of linesused to fit the solution. Most importantly, the internal precisionof the solution is determined, as the rms of the residuals of thefit to the catalogued lines, divided by the number of lines used togenerate the fit. Saving the solution:
The wavelengths per pixel generated fromthe polynomial fits for each order are saved as an E2DS file. Thecoe ffi cients of the polynomial fits are stored in the header. If thequality controls were all passed, the new wavelength solution iscopied to the calibration database, and to the header of the inputHC E2DS spectra.Performance tests (discussed in more detail in Sect. 4.2)showed that while the first method has slightly better internalaccuracy than the second, it is much less stable night-to-night.In any case, neither method reaches the expected internal accu-racy of < .
45 m s − budgeted for the SPIRou wavelength solu-tion. The first method is also very sensitive to drifts from the ini-tial wavelength solution (this was particularly highlighted by theearthquake in May 2018, which caused a detector shift of severalpixels), to which the second method is more robust. The secondmethod was therefore adopted for the HC wavelength solution,and is the only one available in version 0.5.000 of the DRS.During the SPIRou validation and commissioning tests, it be-came clear that the HC lamps alone did not provide a su ffi cientlyaccurate and stable wavelength solution, with internal accuracymeasurements of ∼ − compared to the 0 .
45 m s − accu-racy demanded by the SPIRou error budget for an overall 1 m s − RV precision. There are likely several contributing factors to thislack of accuracy: the low flux levels in the edges of the bluestorders, the low number of lines found for the reddest orders, im-precision in the catalogue wavelengths, among others. The nextstep, therefore, was to combine the HC lamps with the FP étalonspectra.
The design of the SPIRou FP étalon is described in Cersullo et al.(2017). It has a finesse value of F = − λ m = dm , (1)where λ m is the wavelength of the line of (integer) line num-ber m , and d is the e ff ective cavity width of the interferometer.In an ideal FP interferometer, d is constant (and known since itis set by the manufacturer); therefore, once the wavelength λ m , r is known for a specific reference line, its line number m r canbe determined. Then, for any other line, the line number can bedetermined simply by counting from the reference line, and itswavelength calculated with Eq. 1. However, in real FP étalons,the cavity width d is not constant. Bauer et al. (2015) showed itto be wavelength-dependent, due to a varying penetration depth;photons of di ff erent energy (i.e. di ff erent wavelength) will pene-trate the soft coating to di ff erent depths. This wavelength depen-dence needs to be calibrated in order to allow the use of the FPlines in a wavelength solution.Here, again, two methods were developed. The first, methodFP1, follows the approach described by Bauer et al. (2015): Wefirst use the HC lines to generate a rough wavelength solution,from which we obtain first-guess FP wavelengths; these first-guess wavelengths are in turn used to fit the cavity width. Thesecond, method FP2, is based on the one developed by C. Lo-vis for ESPRESSO (private communication). Fractional FP linenumbers are assigned to the HC lines, which are then used to fitthe cavity width directly. The overall structure of the algorithmis as follows. Data identification and reading:
The input files are verified viaFITS header keys to be E2DS files corresponding to an HC lampand the FP étalon, respectively. The data and header are read, andthe lamp and fibre are identified. Fibre correspondence betweenthe HC and FP files is checked. If more than one HC file is given,the median of the frames is obtained and used as input for the restof the routine. Currently, providing more than one FP file is notsupported. This option was chosen as allowing for multiples ofeach file type increased set-up complexity, and the FP spectra aregenerally bright everywhere while the HC lines can be weaker.
Calibration set-up:
Previous calibration files are copied fromthe calibration database (including the blaze file and the previouswavelength solution). The previous wavelength solution is readand checked for compatibility with the current parameter set-up (order of the polynomial fits employed). The correct HC linecatalogue is read in.
Generation of a first-guess wavelength solution:
The HCspectra are used to generate a wavelength solution, using MethodHC2 from Sect. 3.2. Quality controls are applied to this first-guess solution.
Article number, page 5 of 12 & A proofs: manuscript no. main
Incorporation of the FP lines:
The FP lines are identified andGaussians fitted to them: For each order, the highest value isidentified, and a Gaussian fit attempted on a 7-pixel box aroundit. Fits that do not fail and are centred within ± . .
01 nmat the blue end of the spectrum to 0 .
05 nm at the red end. Withthe FP lines’ pixel positions known, the initial wavelength solu-tion generated from the HC spectra alone is used to fit first-guessFP wavelengths from the FP line pixel positions. Using the FPequation and an input cavity width value, the FP line numberis obtained for the last peak of the reddest order. We initiallyused the manufacturer’s value of d = .
25 mm before coating,meaning 2d = . = . Wavelength dependence of the cavity width:
With the FPline numbers identified, the wavelength dependence of the cavitywidth is dealt with using one of the two methods named above:
Method FP1:
Using the line numbers and the first-guesswavelengths of the FP lines (derived from the first-guess HCsolution previously generated), a cavity width is calculated foreach peak. A ninth-order polynomial is fitted to the cavity width(Fig. 5, top points and fit), and corrected FP line wavelengths arecalculated from this polynomial fit.
Method FP2:
First, a fourth-order polynomial is fitted to theFP line numbers as a function of pixel positions per order. This fitis used to generate fractional line numbers for the ’best’ HC lines(selected as the lines at blaze values of more than 30%, and withvelocity o ff set from the catalogue less than 0 .
25 km s − ). Usingthese fractional line numbers and the catalogue wavelengths, thecavity width is calculated for each HC line using the FP equation.Ninth-order polynomials are then fitted to these cavity width val-ues (Fig. 5, bottom points and fit) as a function of both line num-ber and wavelength, and corrected FP line wavelengths are cal-culated from the fit. Optionally, a previous cavity width fit can beread in; in this case we assume only an achromatic change (i.e.a shift) may have taken place, and correct the fit for this shiftby subtracting the median of the residuals between the newlycalculated cavity widths for the HC lines and the previous fit. Fitting the solution:
The HC and FP lines are combined. Afourth-order polynomial fit is performed between the pixel po-sitions, given by the centres of the Gaussians, and the line wave-lengths (catalogued for the HC lines, generated from the cavitywidth fit for the FP lines).
Calculating the FP RV:
The RV of the FP spectrum is cal-culated using the cross-correlation function (CCF) method,through cross-correlation with an FP mask. It is stored in orderto give an FP RV zero-point, from which the drift of the spec-trograph for a later observation of a star with simultaneous FP can be measured (by subtracting the RV of the wavelength so-lution’s FP from the RV of the simultaneous FP). Currently, weassume the drift between the HC exposure that gives the abso-lute zero-point and an immediately subsequent FP exposure tobe negligible, given the stability of SPIRou. In the future, a pos-sible next step would be to employ HC-FP and FP-HC exposures(i.e. exposures with the HC lamp on the science fibres and the FPétalon on the calibration fibre or vice versa) to calibrate this drift,as done in, for example, ESPRESSO or HARPS.
Littrow solution check, quality controls, logging statistics:
These three steps are analogous to those of the HC solution.
Saving the solution:
Similar to the HC solution. If the qual-ity controls were all passed, the header of the input FP E2DSspectra is also updated. The HC and HC-FP solutions are storedunder di ff erent file names, so that the HC-FP solution does notoverwrite the HC solution. Saving results tables:
Two tables are stored. The first logs thestatistics of the Littrow solution quality check. The second is alist of all lines used for the solution, containing the order, wave-length, di ff erence in velocity of the final fit from the input linevalue, weight, and pixel position for each line. Fig. 5.
Variability of the FP cavity width with regard to the input valuefor methods FP1 (FP lines) and FP2 (HC lines), o ff set by 0.5 µ m forvisibility (top), and residuals to the fits, o ff set by 0.05 µ m for visibil-ity (bottom). The results from the two methods are very similar. Formethod FP1, some outliers (poorly fitted FP lines) can be seen, whilefor method FP2 the number of lines drops o ff towards low line numbers(i.e. high wavelengths) and no lines are selected for the reddest order.For scale, a cavity width o ff set of 0.25 µ m corresponds to a velocityo ff set of 6 .
12 km s − , using the FP equation and the initial cavity widthvalue of 2d = . As will be discussed in Sect. 4.3, the two methods are com-parable, though method FP2 has slightly better accuracy and sta-bility. In version 0.5.000 of the DRS, only method FP1 was avail-able for general use as method FP2 was still in development. Inforthcoming versions both will be o ff ered as options in the finalwavelength solution algorithm.
4. Performance tests and validation
To test the performance of the di ff erent wavelength solution gen-eration methods, we ran all scripts on the calibrations of a two- Article number, page 6 of 12. J. Hobson et al.: The SPIRou wavelength calibration for precise radial velocities in the near infrared week SPIRou run in February 2019. We used three di ff erent UNeline catalogues: the R11 catalogue, a combination of the R11and S18 catalogues (with the S18 wavelengths kept for match-ing lines), and a selection of the most stable lines (i.e. those con-sistently identified for di ff erent HC frames) with updated, moreaccurate wavelength values. This selection of lines was derivedusing method FP2. First, the cavity width was fitted from the bestHC lines (at blaze values of more than 30%, and with velocityo ff set from the catalogue less than 0 .
25 km s − ) for each of 50HC exposures taken during commissioning (between May andNovember 2018). For each of these exposures, the wavelengths,fractional line numbers and calculated cavity widths of the linesused for the fit were saved. Then, all the lines were combined,and their wavelengths and cavity widths were fitted together togenerate a very accurate cavity width fit (Fig 6, top panel). Us-ing this accurate cavity width fit, it can be seen that while foreach catalogue line the di ff erent cavity widths measured for eachexposure cluster together, these clusters can be significantly o ff -set from the overall fit (Fig 6, bottom panel). This would meanthat for these o ff set values, the catalogue wavelengths are inac-curate. Therefore, each measured line’s wavelength was recal-culated from its fractional line number and the FP equation. Fi-nally, each line that was selected in at least two exposures wasassigned an updated wavelength, as the median of its recalcu-lated wavelengths. The stable lines catalogue is therefore not justa selection of the best lines from the others, but a new cataloguewith updated wavelength values for each line. The full updatedcatalogue is presented in Appendix 5. The wavelength solution is the last step of a long calibration se-quence. This made its development particularly challenging aschanges upstream frequently meant large modifications in thedata that proved destabilising for the wavelength solution, es-pecially in the earliest versions of the DRS. As the DRS hasevolved, this has somewhat ameliorated. Nevertheless, the pre-vious calibrations can still impact the wavelength solution. Inorder to explore the performance of the wavelength solutionsalone, therefore, the tests presented in the rest of this section areall carried out with the HC and FP files extracted using a singleset of input calibrations. We found that fixing the input calibra-tions introduces large drifts between the wavelength solutionsfor di ff erent nights, of the order of ∼ − ; however, theyare easily calibrated out by subtracting the night-to-night me-dian RV di ff erence. This removes sensitivity to long-term drifts,without a ff ecting the analysis of the wavelength solution. To test the performance of the HC wavelength solutions, we ranboth methods on the UNe spectra taken as part of the afternooncalibrations for the two-week SPIRou run in February 2019, forall three wavelength catalogues. The spectra were reduced witha single set of calibrations. We obtained one solution per nightfor each method and catalogue. Table 1 summarises the results.The ’internal error’ row is a median of the internal accuracies re-ported for the solutions obtained with the corresponding methodand catalogue. The ’local night-to-night variation’ represents themedian di ff erence between consecutive nights’ solutions (com-puted as the median of the pixel-by-pixel absolute drift-correctedRV di ff erences, with the drift corrected by subtracting the over-all median), and can be thought of as the typical uncertainty of Fig. 6.
Construction of a stable lines catalogue. Top: Overall cavitywidth fit (black line) from the HC lines (dots) for fifty exposures. Bot-tom: Zoom showing how the values for each line cluster together, butcan be o ff set from the main fit. For scale, a cavity width o ff set of 0.2 µ m corresponds to a velocity o ff set of 4 .
88 km s − , using the FP equa-tion and the initial cavity width value of 2d = . a single spectral line. The ’HC lines used’ row is the median ofthe number of HC lines that were identified and used to fit eachwavelength solution. The line count is performed per order, soany lines used for more than one order will be counted twice.Method HC1 has better internal accuracy than method HC2,but is less stable from one night to the next. As noted in Sect.3.2, its sensitivity to the input wavelength solution was par-ticularly highlighted by the earthquakes su ff ered by the CFHTduring SPIRou validation, on 3 and 4 May 2018. The multi-pixel displacement meant all lines were significantly shifted withregard to the search windows defined from the previous (pre-earthquake) solutions. This required the creation of additionalalgorithms to identify the pixel shifts and generate shifted first-guess solutions, in order for method HC1 to be able to run. Con-cern over this sensitivity was in fact one of the driving motiva-tions for the development of method HC2, which (since it iden-tifies all peaks in the spectrum and then generates a best matchto the catalogue) is more robust to such shifts.Regarding the catalogues, adding the lines from S18 does notseem to create a substantial change in accuracy or stability. Most Article number, page 7 of 12 & A proofs: manuscript no. main
Table 1.
Summary of HC wavelength solution performances.
Method HC1 Method HC2R11 catalogue Global internal error 1 .
88 m s − .
87 m s − Local night-to-night variation 16 . − . − HC lines used 5607 4770R11 + S18 catalogue Global internal error 1 .
88 m s − .
95 m s − Local night-to-night variation 15 . − . − HC lines used 6186 5310Selected lines Global internal error 1 .
46 m s − .
21 m s − Local night-to night variation 13 . − . − HC lines used 2363 2124of the lines added have fairly low relative intensities reported byS18, so they are likely small and their Gaussian fits may be lessprecise. The ’stable lines’ catalogue provides somewhat moreaccurate and stable wavelength solutions for both methods. Inany case, neither method reaches the required internal precisionof 0 .
45 m s − with any catalogue. We used the same two-week SPIRou run in February 2019, pro-cessed with a single set of calibrations, to test the performancesof both the combined HC-FP wavelength solution methods. Togenerate the first-guess HC solution, we applied method HC2 inboth cases. Once again, we tested the three di ff erent wavelengthcatalogues. Table 2 summarises the results, with the same rowsas for Table 1.In this case, the two methods are very comparable, thoughmethod FP2 has somewhat better internal accuracy and stabil-ity. Rather perplexingly, for the combined HC-FP solutions the’stable lines’ catalogue provides the least night-to-night stabil-ity! This is particularly evident for method FP2, where the HClines are used to fit the cavity width directly. Since this cataloguewas generated from a cavity width fit using multiple HC expo-sures, each reduced with the corresponding nightly calibrations,this may perhaps be a derived e ff ect of the previous calibrations’instability. Nevertheless, in all cases the internal accuracy is ex-cellent, and the night-to-night stability is much improved com-pared to the HC solutions.As was described in Sect. 3.3, for method FP2 there is anoption to read in a previous cavity width fit and correct it fromany achromatic shift, instead of generating it anew. The reason-ing behind this is that the chromatic dependence is an intrinsicproperty of the soft coating; while it may evolve slowly over thelifetime of the instrument, it is not expected to change from onenight to the next. An achromatic shift, meanwhile, would cor-respond to a change in the physical separation of the FP, whichcould be caused by pressure or temperature changes.We tested the implementation of this option, redoing theanalysis with an initial cavity width read in. We found a me-dian internal error of 0 .
14 m s − , and a median night-to-nightvariation of 0 . − , regardless of the catalogue used. This im-plies that a substantial part of the night-to-night variability forthe combined HC-FP solutions is in fact coming from the cavitywidth fit. An example of a night-to-night comparison is shownin Fig. 7, for the solutions computed with method FP2 usinga fixed cavity width, for the nights of 21 st and 22 nd February2019, respectively. The night-to-night variations are significantlyreduced compared to the solutions obtained with a free cavity
Fig. 7.
Night-to-night variations of the FP wavelength solution: di ff er-ence (in RV space) between the solutions generated with method FP2for the 21 and 22 February 2019, using a fixed cavity width fit. The me-dian value of the absolute RV di ff erences is 0 . − . The drift inducedby the fixed set of prior calibrations has been corrected. width fit, with the redder orders driving most of the remainingvariability. To verify the intra-night stability of the HC-FP wavelength so-lution, we computed solutions using sequences of 1 HC and 10FP frames taken continuously over a 14h period. We adoptedtwo test configurations: fixing the HC frame and varying the FPframe, and fixing the FP frame and varying the HC frame. Foreach of these, we obtained HC-FP wavelength solutions usingmethod FP2. We then computed the RV of an FP frame (takingit as an artificial ’star’) using the di ff erent wavelength solutions.The resulting RV variations (corrected for the spectrograph driftby using the FP calibration fibre CCF, as described in Section3.3) are shown in Fig. 8. For the case of a fixed HC frame andvarying FP frames, the RV variations show a slow downwardtrend, which is probably due to the intrinsic drift of the FP étalonwith respect to the absolute wavelength reference. For the caseof a fixed FP frame and varying HC frames, the RV variations arevery small, with an amplitude comparable to the photon noise. Article number, page 8 of 12. J. Hobson et al.: The SPIRou wavelength calibration for precise radial velocities in the near infrared
Table 2.
Summary of HC-FP wavelength solution performances.
Method FP1 Method FP2R11 catalogue Global internal error 0 .
18 m s − .
12 m s − Local night-to night variation 3 . − . − HC + FP lines used 23612 21662R11 + S18 catalogue Global internal error 0 .
18 m s − .
12 m s − Local night-to night variation 3 . − . − HC + FP lines used 23691 21874Selected lines Global internal error 0 .
18 m s − .
13 m s − Local night-to night variation 4 . − . − HC + FP lines used 22990 20466Fixed cavity width Global internal error NA 0 .
14 m s − Local night-to night variation NA 0 . − Fig. 8.
Radial velocity variations over time for a CCF RV of an FPframe (taking it as an artificial ’star’), computed using wavelength so-lutions with a fixed HC frame and varying FP frames (top), or varyingHC frames and a fixed FP frame (bottom). The RVs have been drift-corrected.
Ultimately, the interest of an accurate and stable wavelength so-lution for a spectrograph lies in being able to measure preciseRVs. In the SPIRou DRS, RVs are measured by the CCF method- that is, by cross-correlating a binary stellar mask with the ob- served spectrum. This cross-correlation is first performed orderby order; these are then summed together, and a Gaussian fitis performed, the final RV being measured as the centre of theGaussian. The RVs per order and the combined RV are all stored.If the star is observed with simultaneous FP on the calibration fi-bre, the instrumental drift is computed by the same recipe, cross-correlating the FP spectrum to a binary FP mask.To evaluate the impact of the wavelength solution on theRVs, we selected observations of two stars, which we shall referto as star A and star B, chosen as they are the brightest targetsobserved close to the centre of the February 2019 run. For eachstar, we computed the RVs changing the input wavelength solu-tion. To generate the wavelength solutions, we adopted methodFP2 and a fixed cavity width fit as this was shown to provide themost accurate and stable set of wavelength solutions. We usedthe CCF computation from the SPIRou DRS to obtain the RVs.Both stars were observed with simultaneous FP calibration, sothe drift was also computed from the FP CCFs, as described inSection 3.3. The results are summarised in Table 3. The standarddeviation of the di ff erences from the RV obtained for 13 Febru-ary (taken as the reference point) is 0 .
67 m s − for star A, and0 .
40 m s − for star B. We observed and corrected for large drifts(median value 11 . − ); these drifts are induced by the fact wehave fixed a single set of prior calibrations for extracting all thefiles, which creates o ff sets between the wavelength solutions, asdescribed in Sect. 4.1. The physical origin of these day-to-daydrifts may lie in many di ff erent factors: The instrument is knownto be sensitive to vibrations; the room in which the FP is locatedis not stabilised; there are known jumps at each thermal cycle ofthe cryostat; among others.Although the overall variations are small, it is worthwhile toinspect the CCFs in more detail. Fig. 9 shows the drift-correcteddi ff erence in CCF RVs per order for each star. Gaps correspondto orders for which no CCF could be calculated (generally due toa very low or null atmospheric transmission). It is clear that someorders are far more variable than others, and may be driving theRV di ff erences. It is hard to say whether this is due to solely tothe wavelength solution (since, for instance, Fig. 7 does not showsignificantly higher night-to-night variability in these orders), orwhether there are also factors due to the CCF at play (e.g. smallerspectral lines in these orders whose fit could be more impactedby small shifts in wavelength solution).In particular, for star A the RVs for echelle orders 71 (cen-tral wavelength 1097 nm) and 79 (central wavelength 984 nm)have standard deviations of ∼ − , while the rest are below ∼ − . Likewise, for star B, the RVs for echelle orders 58(central wavelength 1348 nm) and 71 have standard deviations Article number, page 9 of 12 & A proofs: manuscript no. main
Table 3.
Summary of CCF RV di ff erences using di ff erent wavelength solutions. Star Wavelength RV di ff (all RV di ff (sel.sol. night orders) [m s − ] orders) [m s − ]star A 13 Feb — —14 Feb 0.89 0.8915 Feb 0.20 0.2016 Feb 0.22 0.2217 Feb 0.68 0.6818 Feb 0.26 0.2519 Feb 0.36 0.3521 Feb -0.39 -0.4022 Feb 0.07 0.0723 Feb 0.09 0.0924 Feb -0.54 -0.5525 Feb 1.93 -0.78 σ RVdi f f σ RVdi f f ∼
12 m s − and ∼ − respectively, while the rest are below ∼ − . For order 58 for star B, in particular, closer analysisshows the Gaussian fit to the CCF is clearly poor, explaining thehigh RV o ff set and dispersion. Recalculating the RVs excludingthe worst two orders for each star, the di ff erences are generallyslightly reduced (last column of Table 3), with a standard devia-tion of 0 .
50 m s − for star A and 0 .
39 m s − for star B, though theimpact is not large.There are many factors that may contribute to the remainingnoise. The photon noise is at the ≤ .
10 m s − level, so is un-likely to be a major contributor. The spectrograph drift correc-tion may be playing a role (over 12 days, the drift is of the orderof ∼
50 m s − . In standard operations with daytime calibrations,these drifts would not appear; as we fixed a single set of calibra-tions, they need to be removed separately). For the wavelengthsolution, inaccuracies in the UNe catalogues and fit instabilitiesmay increase the noise. Several detector-related e ff ects may alsobe contributing, such as cosmic rays, bad pixels, the known per-sistence on H4RG detectors, or intra-pixel response variations.
5. Conclusions and future perspectives
We have implemented and tested di ff erent methods of generat-ing a wavelength-pixel correspondence, using either HC lampsalone or the combination of HC lamps with an FP étalon. TheHC lamps alone did not provide su ffi cient accuracy, being at thelevel of ∼ − internal error, while the error budget pre-vision was of < .
45 m s − . The combined HC-FP solutions, onthe other hand, have an excellent internal error of ∼ .
15 m s − .The stability from one night to the next is complicated by thedependence of the wavelength solution on the previous calibra- tions, especially on the slit determination. Fixing all prior cali-brations produces a noticeable ( ∼ − ) but constant RVo ff set between solutions; when this o ff set is removed, the night-to-night variations are greatly diminished. We analysed the im-pact of changing the wavelength solution on the RV calcula-tions, finding that the calculated RVs remain fairly consistentwith ∼ . . − dispersions, and that the drift computationis e ffi cient at removing the RV o ff set between wavelength solu-tions computed with a fixed set of previous calibrations.This article is based primarily on the current stable versionof the DRS, 0.5.000. Significant changes have been planned forfuture versions, which the DRS team is currently working on.Several of these changes are either directly on the wavelengthsolution, or are expected to impact it, such as the implementationof a set of ’master’ calibrations that are not expected to changeon a nightly basis, but only per run or even per thermal cycle.One of these master calibrations will be a master wavelength so-lution, with nightly calibrations measuring only the o ff set fromthis master solution. Another problem to be dealt with is poten-tial variations between solutions obtained for the science and cal-ibration fibres. In version 0.5.000, these are computed indepen-dently, which can lead to cases such as the wavelength solutionfailing quality controls for one fibre but not another. Upcomingversions will anchor the calibrations together to avoid this prob-lem.Wavelength solutions combining HC and FP exposures areimplemented for CARMENES (Bauer et al. 2015; Caballeroet al. 2016) and have recently been tested for HARPS (Cersulloet al. 2019). In both cases, these wavelength solutions are shownto be suitable for reaching a 1 m s − overall RV precision. Weanticipate that this will also be the case for SPIRou. Article number, page 10 of 12. J. Hobson et al.: The SPIRou wavelength calibration for precise radial velocities in the near infrared
Fig. 9.
Drift-corrected CCF RV di ff erences per order for star A (left) and star B (right), using wavelength solutions from di ff erent nights. Thestrong variations of echelle order 58 for star B are due to a poor CCF fit. Acknowledgements.
The authors wish to recognise and acknowledge the verysignificant cultural role and reverence that the summit of Maunakea has alwayshad within the indigenous Hawaiian community. We are most fortunate to havethe opportunity to conduct observations from this mountain. This work was sup-ported by the Programme National de Planétologie (PNP) of CNRS / INSU, co-funded by CNES. We acknowledge funding from ANR of France under contractnumber ANR-18-CE31-0019 (SPlaSH). This research made use of matplotlib,a Python library for publication quality graphics (Hunter 2007); SciPy (Joneset al. 2001–); IPython package (Pérez & Granger 2007); Astropy, a community-developed core Python package for Astronomy (Astropy Collaboration et al.2018, 2013); NumPy (Van Der Walt et al. 2011); Astroquery (Ginsburg et al.2019); ds9, a tool for data visualisation supported by the Chandra X-ray Sci-ence Center (CXC) and the High Energy Astrophysics Science Archive Center(HEASARC) with support from the JWST Mission o ffi ce at the Space TelescopeScience Institute for 3D visualisation. We thank all SPIRou partners for theirfunding contributions to the SPIRou project, whose construction cost (includ-ing reviews and travels) reached a total of (cid:27) / CNRS,CFI, CFHT, LNA, CAUP and DIAS. We are also grateful for generous amountsof in-kind manpower allocated to SPIRou by OMP / IRAP, OHP / LAM, IPAG,CFHT, NRC-H, UdeM, UL, OG, LNA and ASIAA, amounting to a total of about75 FTEs including installation and ongoing upgrades. This work has been car-ried out within the framework of the National Centre of Competence in ResearchPlanetS supported by the Swiss National Science Foundation.
References
Artigau, É., Kouach, D., Donati, J.-F., et al. 2014, in Proc. SPIE, Vol. 9147,Ground-based and Airborne Instrumentation for Astronomy V, 914715Artigau, É., Saint-Antoine, J., Lévesque, P.-L., et al. 2018, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 10709,Proc. SPIE, 107091PAstropy Collaboration, Price-Whelan, A. M., Sip˝ocz, B. M., et al. 2018, AJ, 156,123Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558,A33Baranne, A., Queloz, D., Mayor, M., et al. 1996, A&AS, 119, 373Bauer, F. F., Zechmeister, M., & Reiners, A. 2015, A&A, 581, A117Bechter, E. B., Bechter, A. J., Crepp, J. R., & Crass, J. 2019, arXiv e-prints,arXiv:1908.11429Boisse, I., Perruchot, S., Bouchy, F., et al. 2016, in Society of Photo-Optical In-strumentation Engineers (SPIE) Conference Series, Vol. 9908, Ground-basedand Airborne Instrumentation for Astronomy VI, 990868Butler, R. P., Marcy, G. W., Williams, E., et al. 1996, PASP, 108, 500Caballero, J. A., Guàrdia, J., López del Fresno, M., et al. 2016, in Societyof Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol.9910, Proc. SPIE, 99100ECersullo, F., Co ffi net, A., Chazelas, B., Lovis, C., & Pepe, F. 2019, A&A, 624,A122 Cersullo, F., Wildi, F., Chazelas, B., & Pepe, F. 2017, A&A, 601, A102Co ffi net, A., Lovis, C., Dumusque, X., & Pepe, F. 2019, A&A, 629, A27Donati, J.-F., Kouach, D.and Moutou, C., Doyon, R., et al. submittedDonati, J.-F., Kouach, D., Lacombe, M., et al. 2018, SPIRou: A NIRSpectropolarimeter / High-Precision Velocimeter for the CFHT, 107Figueira, P., Pepe, F., Melo, C. H. F., et al. 2010, A&A, 511, A55Fischer, D. A., Anglada-Escude, G., Arriagada, P., et al. 2016, PASP, 128,066001Ginsburg, A., Sip˝ocz, B. M., Brasseur, C. E., et al. 2019, AJ, 157, 98Halverson, S., Mahadevan, S., Ramsey, L., et al. 2014, Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 9147, Thehabitable-zone planet finder calibration system, 91477ZHunter, J. D. 2007, Computing In Science & Engineering, 9, 90Jones, E., Oliphant, T., Peterson, P., et al. 2001–, SciPy: Open source scientifictools for Python, [Online; accessed ]Kokubo, T., Mori, T., Kurokawa, T., et al. 2016, Society of Photo-Optical Instru-mentation Engineers (SPIE) Conference Series, Vol. 9912, 12.5-GHz-spacedlaser frequency comb covering Y, J, and H bands for infrared Doppler instru-ment, 99121RKotani, T., Tamura, M., Nishikawa, J., et al. 2018, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 10702, Proc. SPIE,1070211Lovis, C. & Pepe, F. 2007, A&A, 468, 1115Marcy, G. W. & Butler, R. P. 1992, PASP, 104, 270Murphy, M. T., Udem, T., Holzwarth, R., et al. 2007, MNRAS, 380, 839Pérez, F. & Granger, B. E. 2007, Computing in Science and Engineering, 9, 21Perruchot, S., Hobson, M., Bouchy, F., et al. 2018, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 10702, 1070265Quirrenbach, A., Amado, P. J., Ribas, I., et al. 2018, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 10702, Proc. SPIE,107020WRedman, S. L., Lawler, J. E., Nave, G., Ramsey, L. W., & Mahadevan, S. 2011,ApJS, 195, 24Redman, S. L., Nave, G., & Sansonetti, C. J. 2014, ApJS, 211, 4Sarmiento, L. F., Reiners, A., Huke, P., et al. 2018, A&A, 618, A118Spronck, J. F. P., Fischer, D. A., Kaplan, Z., et al. 2015, PASP, 127, 1027Van Der Walt, S., Colbert, S. C., & Varoquaux, G. 2011, Computing in Science& Engineering, 13, 22Wildi, F., Pepe, F., Chazelas, B., Lo Curto, G., & Lovis, C. 2011, Societyof Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol.8151, The performance of the new Fabry-Perot calibration system of the ra-dial velocity spectrograph HARPS, 81511F
Article number, page 11 of 12 & A proofs: manuscript no. main
Appendix A: Updated wavelength catalogue
In this appendix, we present our updated list of UNe lines, as de-scribed in Sect. 4. The original wavelength values come from thecatalogues of Redman et al. (2011) or Sarmiento et al. (2018).Only the first few lines are shown; the full catalogue is availablein electronic form at the CDS.
Table A.1.
Updated list of UNe line wavelengths
Original wavelength [nm] Updated wavelength [nm]965.3596 965.35960878965.5902 965.59063720965.9620 965.96233723967.0700 967.07066522967.1665 967.16724282 · · · · · ·· · · · · ·