4. Auxiliary system files

Apart from the raw data (e.g. bias, dark, flat, lamp, scientific frames) acquired during the observing runs, the reduction process requires in input Auxiliary system files.

SIPGI is released with a set of Auxiliary system files necessary to the reduction. These auxiliary files can be grouped in two main classes: the instrument files and the calibration files.

Important

Together with the DRS, the Auxiliary system files are the core of SIPGI; these files are used to properly calibrate and extract spectra from frames acquired with different instrument configurations.

Both instrument and calibration files are appended to raw files by the built-in data organizer during the import.

4.1. Instrument files

The instrument files are of three types:

1. Paf files

2. Grism tables

3. CCD tables

4.1.1. Paf files

The Paf files are ASCII files that depend on the instrument configuration and on the mask type, long slit (LS) or multi-object spectroscopy (MOS). They contain the coefficients of the models used by SIPGI to locate and wavelength calibrate 2D spectra. These geometrical models are:

4.1.1.1. The Optical Model

The Optical (OPT) Model assigns the mask slit position (provided by the mask constructor in millimeters) to the position (in pixels) of the grism central wavelength (λ_ref, see grism table) on the detector. Hereafter this position will be referred as slit reference position. The model takes into account the known optical distortions of the instrument. The OPT model does not depend on the grism in use, since it just locates the slit position on the detector ignoring any information about dispersion.

The model is defined by a pair of global 2D polynomials of the same degree: X describing distortions along the \(x\) axis (i.e. along the cross dispersion direction) and Y along the \(y\) axis (i.e. along the dispersion direction):

\[x_{[pix]} = \sum_{i=0}^2 \sum_{j=0}^2 X_{i,j} x_{[mm]}^i y_{[mm]}^j\]

\[y_{[pix]} = \sum_{i=0}^2 \sum_{j=0}^2 Y_{i,j} x_{[mm]}^i y_{[mm]}^j\]

The Paf file contains the set of \(X_{i,j}\) and \(Y_{i,j}\) coefficients.

4.1.1.2. The Curvature Model

The Curvature (CRV) Model describes the 2D spectra displacement with respect to the ideal dispersion direction, perfectly aligned along the pixels grid. Due to optical distortions on real data, spectra are not perfectly aligned along the pixels and these distortions change within the FOV. The CRV Model describes these deviations (in pixel) of the spectral trail wrt the theoretical dispersion direction.

For each slit, one mono dimensional polynomial (of order N) is used to describe the displacement \(\Delta c\) along the cross dispersion direction, starting from the slit reference position (located by the OPT Model) (see Fig. 4.1)

\[\Delta c_{[pix]} = \sum_{i=0}^N a_{x,y,i} \Delta d_{[pix]}^i\]

\(\Delta d\) is the displacement with respect to the slit reference position in pixel.

Fig. 4.1 The figure shows the distorted 2D spectrum (orange area) wrt the ideal horizontal dispersion direction on the detector. The yellow region indicates the position of the slit. The CRV Model estimates the amount of distortion \(\Delta c\) as a function of \(\Delta d\), i.e. the distance from the slit.

Each slit has its own set of \(a_{x,y,i}\) coefficients because each slit is in a different position in the FOV.

SIPGI uses a global model (the CRV Model) to describe the coefficients variation across the FOV. In particular, the local \(a_{x,y,i}\) coefficients depend on the computed global model (\(A\)) at the slit reference position on detector:

\[a_{x,y,i} = \sum_{h=0}^N \sum_{k=0}^M A_{i,h,k} x_{[pix]}^h y_{[pix]}^k,\]

where N and M are the order of \(x\) and \(y\) polynomial, respectively. The Paf file contains the set of \(A_{i,h,k}\) coefficients.

4.1.1.3. The Inverse Dispersion Solution Model

The Inverse Dispersion Solution (IDS) Model describes the wavelength to pixel relation. Once the slit positions are identified on the detector (by the OPT Model) and the tracing of the 2D dispersed spectra is reconstructed (by the CRV Model) the wavelength calibration of the 2D spectra can be carried out. The IDS Model moves along the tracing curves described by the combination of the OPT and CRV Models and it assigns the expected wavelength values to each pixel along this curve.

The IDS Model mathematical description is quite similar to the CRV Model: for each slit, one mono dimensional polynomial (of order N) locates the wavelength position wrt λ_ref:

\[\Delta d_{[pix]} = \sum_{i=0}^N b_{x,y,i} (\lambda - \lambda_{ref})^i,\]

where \(\Delta d\) is the displacement of λ in pixels with respect to λ_ref.

Since each slit is in a different position in the FOV, and distortions change within the FOV, each slit has its own set of \(b_{x,y,i}\) coefficients. A global 2D polynomial (\(D\)) is used to describe the variation of \(d_{x,y,i}\) across the FOV:

\[b_{x,y,i} = \sum_{h=0}^N \sum_{k=0}^M B_{i,h,k} x_{[pix]}^h y_{[pix]}^k,\]

where N and M are the order of \(x\) and \(y\) polynomial, respectively. The Paf file contains the set of \(B_{i,h,k}\) coefficients.

Note

For a given instrumental configuration, the OPT, CRV and IDS Models clearly differ whether they have to describe LS or MOS observations. Despite this, if the instrument is stable, once the three Models have been calibrated on a set of LS(MOS) data, the same Models can be used to describe every LS(MOS) set of data acquired with the same configuration.

SIPGI comes with a set of default Paf files (see table below) covering the standard MODS and LUCI instrument configurations, both LS and MOS. For LUCI observations, the provided Paf files refer to the nominal central wavelength of the configuration (i.e. \(\lambda_{ref}\) in the formula above) unless otherwise specified. If data were acquired with a central wavelength different from the nominal one, the user can “adjust” the Paf file as described in Adjust First Guesses paragraph.

Table 4.1 The Paf files released with SIPGI. Column 1: Instrument; Column 2: Grism; Column 3: Dichroic; Column 4: λ _ref; Column 5: Camera; Column 6: Mask type; Column 7: Paf file name. The “#” could be 1 or 2 to indicate files for the two different LBT arms.

Instrument	Grism	Dichroic	λ ref	Camera	Bin	Mask	Paf file
MODSR	G670L				1x1	LS	MODS#R_LS_G670L_Red.paf
MODSR	G670L				1x1	MOS	MODS#R_MOS_G670L_Red.paf
MODSR	G670L_dual	Dual			1x1	LS	MODS#R_LS_G670L_Dual.paf
MODSR	G670L_dual	Dual			1x1	MOS	MODS#R_MOS_G670L_Dual.paf
MODSB	G400L				1x1	LS	MODS#B_LS_G400L_Blue.paf
MODSB	G400L				1x1	MOS	MODS#B_MOS_G400L_Blue.paf
MODSB	G400L_dual	Dual			1x1	LS	MODS#B_LS_G400L_Dual.paf
MODSB	G400L_dual	Dual			1x1	MOS	MODS#B_MOS_G400L_Dual.paf
MODSR	G670L				2x2	LS	NO DATA AVAILABLE
MODSR	G670L				2x2	MOS	NO DATA AVAILABLE
MODSR	G670L_dual	Dual			2x2	LS	MODS#R_LS_G670L_Dual_2x2.paf
MODSR	G670L_dual	Dual			2x2	MOS	NO DATA AVAILABLE
MODSB	G400L				2x2	LS	NO DATA AVAILABLE
MODSB	G400L				2x2	MOS	NO DATA AVAILABLE
MODSB	G400L_dual	Dual			2x2	LS	MODS#B_LS_G400L_Dual_2x2.paf
MODSB	G400L_dual	Dual			2x2	MOS	NO DATA AVAILABLE
LUCI	150Ks		2.17 μm [1]	N1.8		LS	LUCI#_LS_150Ks_2.15_1.8.paf
LUCI	150Ks		2.17 μm [1]	N1.8		MOS	NO DATA AVAILABLE
LUCI	200H+K		1.17 μm	N1.8		LS	LUCI#_LS_200H+K_1.17_1.8.paf
LUCI	200H+K		1.17 μm	N1.8		MOS	LUCI#_MOS_200H+K_1.17_1.8.paf
LUCI	200H+K		1.93 μm	N1.8		LS	LUCI#_LS_200H+K_1.93_1.8.paf
LUCI	200H+K		1.93 μm	N1.8		MOS	LUCI#_MOS_200H+K_1.93_1.8.paf
LUCI	210zJHK		0.95 μm [2]	N1.8		LS	LUCI#_LS_210zJHK_0.95_1.8.paf
LUCI	210zJHK		0.95 μm [2]	N1.8		MOS	NO DATA AVAILABLE
LUCI	210zJHK		1.25 μm	N1.8		LS	LUCI#_LS_210zJHK_1.25_1.8.paf
LUCI	210zJHK		1.25 μm	N1.8		MOS	LUCI#_MOS_210zJHK_1.25_1.8.paf
LUCI	210zJHK		1.65 μm	N1.8		LS	LUCI#_LS_210zJHK_1.65_1.8.paf
LUCI	210zJHK		1.65 μm	N1.8		MOS	LUCI#_MOS_210zJHK_1.65_1.8.paf
LUCI	210zJHK		2.20 μm	N1.8		LS	LUCI#_LS_210zJHK_2.20_1.8.paf
LUCI	210zJHK		2.20 μm	N1.8		MOS	LUCI#_MOS_210zJHK_2.20_1.8.paf

[1] (1,2)

Nominal wavelength 2.15

[2] (1,2)

Nominal wavelength 0.97

4.1.2. Grism tables

The Grism tables are FITS files that depend on the instrument configuration. They supply SIPGI with:

• the reference lambda in Angstrom (\(\lambda_{ref}\)), i.e. the central wavelength of the configuration (WLEN_CEN keyword); through the Paf file SIPGI associates the slit position on the frame with the \(\lambda_{ref}\) position;

• the tracing limits (in pixels) wrt the slit position (LLEN_LO and LLEN_HI keywords). Starting from the location of the slit position on the detector, the tracing of the 2D spectra will be performed in the range [slit_position - LLEN_LO, slit_position + LLEN_HI];

• the 2D spectra extraction limits: WLEN_START, WLEN_END keywords (in Angstrom);

• the spectral sampling of 2D and 1D spectra: WLEN_INC keyword (in Angstrom).

Note

The spectral tracing is performed in the range [slit_position-LLEN_LO,slit_position+LLEN_HI](pixels). The Grism tables provided with SIPGI are such that extraction range [WLEN_START, WLEN_END] (Å) is included in the [slit_position-LLEN_LO,slit_position+LLEN_HI] range (pixels) according to the first guesses solution contained into the Paf file.

SIPGI supplies the user with a set of Grism tables covering all the standard instrument configurations. The list of these Grism tables with their default extraction limits and spectral sampling is reported below. The extraction limits reported in the LUCI Grism tables are tuned on real data and on the nominal central wavelength (see \(\lambda_{ref}\) in table below) of the specific configuration. For MODS data, the extraction limits correspond to those reported in the MODS webpage. The default sampling is the nominal dispersion of the configuration.

Table 4.2 The Grism Tables released with SIPGI. Column 1: Instrument; Column 2: Grism; Column 3: Dichroic; Column 4: λ _ref; Column 5: Camera; Column 6: Grism table name; Column 7: Extraction limits in Angstrom; Column 8: Spectral sampling.

Instrument	Grism	Dichroic	λ ref	Camera	Bin	Grism Table file	Extraction limits	Spectral sampling
MODSR	G670L				1x1	grismTable_G670L.fits	5000-10000 Å	0.85 Å/px
MODSR	G670L_dual	Dual			1x1	grismTable_G670L_Dual.fits	5400-10000 Å	0.85 Å/px
MODSB	G400L				1x1	grismTable_G400L.fits	3100-6000 Å	0.52 Å/px
MODSB	G400L_dual	Dual			1x1	grismTable_G400L_Dual.fits	3200-5900 Å	0.52 Å/px
MODSR	G670L_dual	Dual			2x2	grismTable_G670L_Dual_2x2.fits	5400-10000 Å	1.70 Å/px
MODSB	G400L_dual	Dual			2x2	grismTable_G400L_Dual_2x2.fits	3200-5900 Å	1.04 Å/px
LUCI	150Ks		2.17 μm	N1.8	N1.8	grismTable_150Ks_2.17_1.8.fits	19800-23200 Å	2.59 Å/px
LUCI	200H+K		1.17 μm	N1.8	N1.8	grismTable_200H+K_1.17_1.8.fits	9600-13680 Å	2.16 Å/px
LUCI	200H+K		1.93 μm	N1.8	N1.8	grismTable_200H+K_1.93_1.8.fits	15100-23300 Å	4.32 Å/px
LUCI	210zJHK		0.95 μm	N1.8	N1.8	grismTable_210zJHK_0.95_1.8.fits	8830-10060 Å	0.64 Å/px
LUCI	210zJHK		1.25 μm	N1.8	N1.8	grismTable_210zJHK_1.25_1.8.fits	11745-13170 Å	0.76 Å/px
LUCI	210zJHK		1.65 μm	N1.8	N1.8	grismTable_210zJHK_1.65_1.8.fits	15490-17400 Å	1.01 Å/px
LUCI	210zJHK		2.20 μm	N1.8	N1.8	grismTable_210zJHK_2.20_1.8.fits	20400-23600 Å	1.63 Å/px

If the user is not satisfied with these values, he/she can modify them during the reduction (see Adjust First Guesses and Reduce Observations). The Grism table includes also the wavelength in Angstrom of a set of sky lines covered by the spectral configuration. These lines will be used to check and refine directly on scientific data the wavelength calibration obtained from lamp frames (see Reduce Observations).

During the import of raw data, the built-in data organizer retrieves from the FITS header the information on the instrument configuration and append to each file the suitable Paf file and Grism table. For LUCI observations with central wavelength different from the nominal one, SIPGI appends the files related to the nominal configuration. As said before, if the instrument is stable and central wavelength is the nominal one, SIPGI can “automatically” locate the spectra and wavelength calibrate them using the information stored in the Paf files. Nonetheless, instruments are not perfect, optical distortions can change on a night base for several reasons. Furthermore masks could be not perfectly aligned in the FPU or the central wavelength might not be the nominal one. For these reasons, SIPGI offers the users the possibility to check and refine the default Paf file on real data (see Adjust First Guesses).

4.1.3. CCD tables

The CCD tables are FITS files containing information related to the CCD such as gain and readout noise organized by quadrants. The CCD table will also store the list of bad pixels on the detector if provided by the user (pixels enumeration starts from 1) (see Append Bad Pixel Image).

SIPGI comes with the following CCD tables:

Instrument	Readout mode	Bin	CCD Table file
MODS#R		1x1	MODS#R_ccdTable.fits [3]
MODS#B		1x1	MODS#B_ccdTable.fits [3]
MODS#R		2x2	MODS#R_2x2_ccdTable.fits [3]
MODS#B		2x2	MODS#B_2x2_ccdTable.fits [3]
LUCI#	LIR		LUCI#_LIR_ccdTable.fits [4]
LUCI#	MER		LUCI#_MER_ccdTable.fits [4]
LUCI1	DCR		LUCI1_HAWAII_DCR_ccdTable.fits [5]
LUCI1	MER		LUCI1_HAWAII_MER_ccdTable.fits [5]

[3] (1,2,3,4)

see MODS LBT pages for more details

[4] (1,2)

see LUCI LBT pages for more details.

[5] (1,2)

These CCD tables are provided for old data acquired with HAWAII-2 HdCdTe detector. See old LUCI manual (Table 8) for more details.

During the ingestion of raw frames, the built-in data organizer appends also the CCD table to the raw files, together with Paf file and Grism table.

4.2. Calibration files

The calibration files are those necessary for calibration purposes. SIPGI has two types of calibration files, those necessary for the creation of the sensitivity function (Spectro-photometric standard, Stellar templates) and those necessary for the wavelength calibration (Lines catalogs).

Note

In this SIPGI version, MODS and LUCI calibration files have wavelengths in vacuum. In case the original files were defined in air wavelengths (see e.g. Stellar template), the conversion from air to vacuum wavelength has been performed by the PyAstronomy library using the Edlen (1953) and Ciddor (1996) prescriptions. All the MODS spectra released by LBT spectroscopic data reduction center have wavelengths in air.

4.2.1. Spectro-photometric standard

The Spectro-photometric standards are ASCII files describing the spectrum of standard stars. SIPGI comes with a set of Spectro-photometric standard files listed in Appendix A. For MODS observations, the sensitivity function is derived by comparing the wavelength-calibrated 1D spectra of the observed standard star with the corresponding Spectro-photometric standard (for more details see Create Sensitivity).

Caution

Users are strongly encouraged to check the resolution of the provided Spectro-photometric standards and to verify they are suitable for their scientific purposes (see Appendix A). If this is not the case, user can upload in SIPGI a customized Spectro-photometric standard file through the Custom Auxiliary Files panel.

4.2.2. Stellar templates

The Stellar templates are ASCII files with the theoretical spectra of stars of different type and class. SIPGI is released with a series of Pickles (Pickles, 1998) Stellar templates listed in Appendix A. For LUCI observations, the sensitivity function is derived comparing the wavelength-calibrated 1D spectra of the observed telluric with the theoretical template of a star of the same spectral type and class. Before comparing real data with the template, SIPGI scales the last one to nominal fluxes/magnitude of the observed telluric. The nominal magnitude for a significant sample of tellurics are listed in the SIPGI Hipparcos Catalog.

The SIPGI Hipparcos Catalog is provided by SIPGI and contains the spectral type/class and the nominal 2MASS-J, -H, -K magnitude for a sample of 112404 Hipparcos stars (Zacharias et al. 2004).

Caution

Users are strongly encouraged to check the resolution of the Stellar templates provided and to verify they are suitable for their scientific purposes (see Appendix A). If this is not the case, user can upload in SIPGI a customized Stellar template through the Custom Auxiliary Files panel.

4.2.3. Lines catalogs

The Lines catalogs are FITS files containing the wavelengths (in Angstrom) of the main bright isolated emission lines of calibration lamps. These catalogs must cover the whole spectral range of interest in order to provide the most accurate wavelength calibration of the spectra. For this reason the content of these catalogs depends on arc lamps, grism and camera. These files are appended to lamp frames by the built-in data organizer.

Starting from the theoretical IDS Model stored in the Paf file, SIPGI is able to predict and show the position of the Lines catalogs lines on the lamp frames (see Adjust First Guesses). This will allow the user to visually check and refine the lambda-pixel relation and hence to perform the wavelength calibration in an easier and quicker way. Together with the lamp Lines catalogs, SIPGI has also sky Lines catalogs that list the wavelengths of the brightest sky emission lines, to be used if the wavelength calibration has to be performed directly on scientific frames. SIPGI comes with the following Lines catalogs:

Table 4.3 The Lines catalogs released with SIPGI. Column 1: Instrument; Column 2: Grism; Column 3: Camera; Column 4: arc element(s); Column 5: Lines catalog name.

Instrument	Grism	Camera	Lines	Line Catalog
MODS	G400L/G670L		Ar	lineCat_Ar.fits
MODS	G400L/G670L		Hg	lineCat_Hg.fit
MODS	G400L/G670L		Kr	lineCat_Kr.fits
MODS	G400L/G670L		Ne	lineCat_Ne.fits
MODS	G400L/G670L		Xe	lineCat_Xe.fits
MODS	G400L/G670L		Ar Ne	lineCat_ArNe.fits
MODS	G400L/G670L		Hg Ne	lineCat_HgNe.fits
MODS	G400L/G670L		Kr Xe	lineCat_KrXe.fits
MODS	G400L/G670L		Ar Hg Ne	lineCat_ArHgNe.fits
MODS	G400L/G670L		Ar Kr Xe	lineCat_ArKrXe.fits
MODS	G400L/G670L		Hg Kr Xe	lineCat_ArHgKrXe.fits
MODS	G400L/G670L		Ar Hg Kr Xe	lineCat_ArHgKrXe.fits
MODS	G400L/G670L		Ar Kr Ne Xe	lineCat_ArKrNeXe.fits
MODS	G400L/G670L		Hg Kr Ne Xe	lineCat_HgKrNeXe.fits
MODS	G400L/G670L		Ar Hg Kr Ne Xe	lineCat_ArHgKrNeXe.fits
MODS	G400L/G670L		Sky lines	lineCat_Sky.fits
LUCI	150Ks	1.8	Ar	lineCat_Ar_150Ks_1.8.fits
LUCI	150Ks	1.8	Ne	lineCat_Ne_150Ks_1.8.fits
LUCI	150Ks	1.8	Ar Ne	lineCat_ArNe_200H+K_1.8.fits
LUCI	150Ks	1.8	Sky lines	lineCat_sky_150Ks_1.8.fits
LUCI	200HK	1.8	Ar	lineCat_Ar_200H+K_1.8.fits
LUCI	200HK	1.8	Ne	lineCat_Ne_200H+K_1.8.fits
LUCI	200HK	1.8	Xe	lineCat_Xe_200H+K_1.8.fits
LUCI	200HK	1.8	Ar Xe	lineCat_ArXe_200H+K_1.8.fits
LUCI	200HK	1.8	Ne Xe	lineCat_NeXe_200H+K_1.8.fits
LUCI	200HK	1.8	Ar Ne Xe	lineCat_ArNeXe_200H+K_1.8.fits
LUCI	200HK	1.8	Sky lines	lineCat_sky_200H+K_1.8.fits
LUCI	210zJHK	1.8	Ar	lineCat_Ar_210zJHK_1.8.fits
LUCI	210zJHK	1.8	Ne	lineCat_Ne_210zJHK_1.8.fits
LUCI	210zJHK	1.8	Xe	lineCat_Xe_210zJHK_1.8.fits
LUCI	210zJHK	1.8	Ar Ne	lineCat_ArNe_210zJHK_1.8.fits
LUCI	210zJHK	1.8	Ar Xe	lineCat_ArXe_210zJHK_1.8.fits
LUCI	210zJHK	1.8	Ne Xe	lineCat_NeXe_210zJHK_1.8.fits
LUCI	210zJHK	1.8	Ar Ne Xe	lineCat_ArNeXe_210zJHK_1.8.fits
LUCI	210zJHK	1.8	Sky lines	lineCat_sky_210zJHK_1.8.fits

SIPGI Lines catalogs are obtained using these references:

• Argon lines from IRAF (see argon.dat);

• Xenon lines from IRAF (see xenon.dat);

• The NIST data base;

• The HO lines atlas provided by: Rousselot et al. (2000); lowd_ir_ohlines.dat; ir_ohlines.dat; Osterbrock et al. (1996)

4.3. Mask files

The Mask file is the file that describes the mask for MODS data. The description of the mask is necessary to SIPGI to locate the slit(s) position on the CCD. LUCI does not require these kind of files since this information is stored in the FITS header of raw frames. The Mask files for standard MODS long-slit (LS) observations are released with SIPGI. SIPGI comes with these LS Mask files:

Mask file	Slits Number	Slit aperture
LS5x60x0.3.mms	5	0.3 arcsec
LS5x60x0.6.mms	5	0.6 arcsec
LS5x60x0.8.mms	5	0.8 arcsec
LS5x60x1.0.mms	5	1.0 arcsec
LS5x60x1.2.mms	5	1.2 arcsec
LS60x1.0.mms	1	1.0 arcsec
LS60x5.mms	1	5.0 arcsec

For MODS MOS observations, the user must supply SIPGI with the Mask file of its observations (see the Add mask definition files). The mask to upload is the file produced by the LBT mask preparation software.

4.4. Extinction table

The Extinction Table is an ASCII file containing the model extinction curve for LBT. It is provided by LBT (see MODS page).