J/A+A/684/A38 HARPS instrumental profile (Milakovic'+, 2024)
A new method for instrumental profile reconstruction of high-resolution
spectrographs.
Milakovic' D., Jethwa P.
<Astron. Astrophys. 684, A38 (2024)>
=2024A&A...684A..38M 2024A&A...684A..38M (SIMBAD/NED BibCode)
ADC_Keywords: Line Profiles
Keywords: instrumentation: spectrographs - methods: data analysis -
techniques: spectroscopic
Abstract:
Knowledge of the spectrograph's instrumental profile (IP) provides
important information needed for wavelength calibration and for the
use in scientific analyses. This work develops new methods for IP
reconstruction in high-resolution spectrographs equipped with
astronomical laser frequency comb (astrocomb) calibration systems and
assesses the impact that assumptions on the IP shape have on achieving
accurate spectroscopic measurements. Astrocombs produce ∼10000 bright,
unresolved emission lines with known wavelengths, making them
excellent probes of the IP. New methods based on Gaussian process
regression were developed to extract detailed information on the IP
shape from these data. Applying them to HARPS, an extremely stable
spectrograph installed on the ESO 3.6m telescope, we reconstructed its
IP at 528 locations of the detector, covering 60% of the total
detector area. We found that the HARPS IP is asymmetric and that it
varies smoothly across the detector. Empirical IP models provide a
wavelength accuracy better than 10m/s (5m/s) with a 92% (64%)
probability. In comparison, reaching the same accuracy has a
probability of only 29% (8%) when a Gaussian IP shape is assumed.
Furthermore, the Gaussian assumption is associated with intra-order
and inter-order distortions in the HARPS wavelength scale as large as
60 m/s. The spatial distribution of these distortions suggests they
may be related to spectrograph optics and therefore may generally
appear in cross-dispersed echelle spectrographs when Gaussian IPs are
used. Empirical IP models are provided as supplementary material in
machine readable format. We also provide a method to correct the
distortions in astrocomb calibrations made under the Gaussian IP
assumption. Methods presented here can be applied to other instruments
equipped with astrocombs, such as ESPRESSO, but also ANDES and G-CLEF
in the future. The empirical IPs are crucial for obtaining objective
and unbiased measurements of fundamental constants from
high-resolution spectra, as well as measurements of the redshift
drift, isotopic abundances, and other science cases.
Description:
The file provided here contains the numerical values describing the shape of
the HARPS IP as well as the most likely hyperparameter values for the
Gaussian process (GP) describing those shapes (see text for details).
The file was derived by applying the methods described in the accompanying
paper to the astrocomb spectrum taken on 2018-12-07 at 00:12:50.196 UTC,
observed in fibre A of the HARPS spectrograph.
To aid table understanding, explanations of some columns contain the names
of variables as they appear in the accompanying manuscript. References to
appropriate equations are also given.
Python code with examples of file use can be found at
https://zenodo.org/doi/10.5281/zenodo.10492989
(Instrumental Profile of the HARPS spectrograph, Milakovic' & Jethwa 2024)
The file is in Flexible Image Transport System (FITS) format and contains
five Header Data Units (HDUs), of which only four contain data:
0. an empty HDU.
1. 'pixel_gp' -- a binary table HDU.
Contains the most likely hyperparameter values for the GP describing
the IP shapes in pixel space and the cosrespodning training data.
It also contains the most likely hyperparameters for the secondary GP
(describing the empirical variance on the training data) and the
training data for that secondary GP.
Consists of 24 columns and 528 rows. Each row corresponds to one
detector segment. The columns have the following descriptions:
---------------------------------------------------------------------
Format Units Label Explanations
---------------------------------------------------------------------
I4 --- order Ordinal number of the echelle order in
the HARPS pipeline 'e2ds' file (0--72)
I4 --- optord Echelle order number (89--161)
I4 --- segm Segment number (1--16)
F4 pix ledge Segment left edge
F4 pix redge Segment right edge
F8(600) pix data_x GP training data, Δ(x)
F8(600) --- data_y GP training data, \psi
F8(600) --- data_yerr Error on data_y, σ_\psi
F8(40) pix sct_x Secondary GP training data, bin coord.
F8(40) --- sct_y Secondary GP training data, S^2
F8(40) --- sct_yerr Error on sct_y
I4 --- numlines Number of data points in data_x
F8 --- logL Log likelihood of the trained GP
F8 pix shift Shift applied to data_x, Eq.(20)
F8 --- mf_amp Mean function amplitude, A
F8 pix mf_loc Mean function location, µ
F8 pix mflogsig Log of the mean function std.dev., τ
F8 --- mf_const Mean function offset, y_0
F8 --- gplogamp Log of the GP amplitude, a
F8 pix gplogscale Log of the GP scale, l
F8 --- logvaradd Log of additive variance, σ_0
F8 --- sctlogamp Log of the secondary GP amplitude, a_g
F8 --- sctlogscale Log of the secondary GP scale, l_g
F8 --- sctlogconst Log of the secondary GP constant, C_g
2. 'pixel_model' -- a binary table HDU.
Contains the numerical values describing the most likely IP shapes
in pixel space.
Consists of 9 columns and 528 rows. Each row corresponds to one
detector segment. The columns have the following descriptions:
---------------------------------------------------------------------
Format Units Label Explanations
---------------------------------------------------------------------
I4 --- order Echelle order ordinal number in
HARPS pipeline e2ds file (0--72)
I4 --- optord Echelle order number (89--161)
I4 --- segm Segment number (1--16)
F4 pix ledge Segment left edge
F4 pix redge Segment right edge
F8(601) pix x IP shape x-coordinates
F8(601) --- y IP shape y-coordinates, ψ
F8(601) --- scatter Error on y
I4 --- numlines Number of points in the training data
3. The same as HDU number 2, but for IP models in velocity space.
Consists of 24 columns and 528 rows. Each row corresponds to one
detector segment. The columns have the following descriptions:
---------------------------------------------------------------------
Format Units Label Explanations
---------------------------------------------------------------------
I4 --- order Ordinal number of the echelle order in
the HARPS pipeline 'e2ds' file (0--72)
I4 --- optord Echelle order number (89--161)
I4 --- segm Segment number (1--16)
F4 pix ledge Segment left edge
F4 pix redge Segment right edge
F8(600) km/s data_x GP training data, Δ(x)
F8(600) --- data_y GP training data, \psi
F8(600) --- data_yerr Error on data_y, σ_\psi
F8(40) km/s sct_x Secondary GP training data, bin coord.
F8(40) --- sct_y Secondary GP training data, S^2
F8(40) --- sct_yerr Error on sct_y
I4 --- numlines Number of data points in data_x
F8 --- logL Log likelihood of the trained GP
F8 km/s shift Shift applied to data_x, Eq.(20)
F8 --- mf_amp Mean function amplitude, A
F8 km/s mf_loc Mean function location, µ
F8 km/s mflogsig Log of the mean function std.dev., τ
F8 --- mf_const Mean function offset, y_0
F8 --- gplogamp Log of the GP amplitude, a
F8 km/s gplogscale Log of the GP scale, l
F8 --- logvaradd Log of additive variance, σ_0
F8 --- sctlogamp Log of the secondary GP amplitude, a_g
F8 --- sctlogscale Log of the secondary GP scale, l_g
F8 --- sctlogconst Log of the secondary GP constant, C_g
4. The same as HDU number 3, but for IP models in velocity space.
Consists of 9 columns and 528 rows. Each row corresponds to one
detector segment. The columns have the following descriptions:
---------------------------------------------------------------------
Format Units Label Explanations
---------------------------------------------------------------------
I4 --- order Echelle order ordinal number in
HARPS pipeline e2ds file (0--72)
I4 --- optord Echelle order number (89--161)
I4 --- segm Segment number (1--16)
F4 pix ledge Segment left edge
F4 pix redge Segment right edge
F8(601) km/s x IP shape x-coordinates
F8(601) --- y IP shape y-coordinates, ψ
F8(601) --- scatter Error on y
I4 --- numlines Number of points in the training data
--------------------------------------------------------------------------------
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
harps_ip.fits 2880 10989 The FITS file containing the HARPS IP,
as per Description
--------------------------------------------------------------------------------
Acknowledgements:
Dinko Milakovic', dinko(at)milakovic.net
(End) Patricia Vannier [CDS] 05-Feb-2024