Keck LRIS

Overview

This file summarizes several instrument specific settings that are related to the Keck/LRIS (RED and BLUE) spectrograph.

Common Items to LRISb and LRISr

FITS format

Before May 2009, both LRISb and LRISr observations were stored in standard simple FITS format consisting of a single primary HDU without any extensions. Subsequently, the data were stored in multi-extension FITS files, including four extensions, one for each amplifier. To handle both formats, PypeIt defined 2 sets of spectrographs for each LRIS camera, one for the pre-May 2009 data (keck_lris_blue_orig and keck_lris_red_orig) and one for the post-May 2009 data (keck_lris_blue and keck_lris_red). Note that this change in FITS format coincided with the installation of the LBNL detectors (2kx4k) in LRISr (see LRIS RED).

Slit-masks

PypeIt can now incorporate slitmask information in the reduction routine for keck_lris_red, keck_lris_red_mark4, and keck_lris_blue similar to its DEIMOS capabilities (see Slit-mask design matching). That is, if the trace calibration files with mask information are used, PypeIt is capable of using said information to match the traced slit edges to those predicted by the slitmask design, assign to each extracted spectrum the corresponding RA, Dec and object name information, force the extraction of undetected objects at the location expected from the slitmask design, identify serendipitous sources and, subsequently, collate by RA/Dec. See Additional Reading for more information.

Unfortunately, LRIS raw frames do not come ready with slitmask data and thus this information needs to be inserted by the user before processing with PypeIt if they are desirous of incorporating the above mentioned features into their reduction. To do so, the user must first obtain the mask design files and then process them with the software TILSOTUA. Here are the steps to follow:

1. Obtain the mask design files, which include:

   1. Output files produced by autoslit.

      • One file with the ".file3" extension containing milling information.

      • One file with the ".file1" extension containing the object catalog matched to the slitmask slits.

   2. The ASCII object list file fed as input to autoslit to generate the files above.

   Note

   ".file3" is mandatory while the other two files can be optionally excluded to debug TILSOTUA.

2. Process the design files with TILSOTUA : The design files contain the milling blueprint (the BluSlits table). When using the ".file3" design file, TILSOTUA creates a FITS file with the BluSlits table following the UCO/Lick template. If the ".file1" file and the object list are provided, the FITS mask design file will also includes the DesiSlits, ObjectCat and SlitObjMap binary tables, otherwise they will be empty. These tables include information on the slitmask design, the object catalog and the mapping between the two, similar to the binary tables in DEIMOS raw frames. TILSOTUA populates these tables using its xytowcs function (in LRIS_Mask_Coords_to_WCS.py). This function can be run by providing two parameters:

   • the data_input_name, which is either the FITS or ".file3" mask design file (be sure the name includes the extension);

   • the output_file, which is the name of the output file that TILSOTUA will generate. Do not include any extension such as .fits.

   If only the ".file3" file is provided, the calling sequence is:

   from tilsotua import xytowcs
   
   xytowcs(data_input_name="yourmaskname.file3",output_file="yourmaskname_output")
   

   Although the other parameters are optional for xytowcs as a standalone code, users interested in applying the slitmask information to their PypeIt reduction must provide the `obj_file` and `file1` files to ensure that object names are assigned to the extracted spectra.

   If the ".file1" file and the object list are provided, the calling sequence is:

   from tilsotua import xytowcs
   
   xytowcs(data_input_name="yourmaskname.file3",output_file="yourmaskname_output",
           obj_file="yourtargets.obj", file1="yourmaskname.file1")
   

   It is assumed that the entries in file1 and obj_file have unique Name values, i.e., make sure you have a unique identifier for each object. Without this, it is not possible to correctly reconcile the two tables.

3. Add the TILSOTUA-generated slitmask design information to your raw trace FITS files: The user must first verify that TILSOTUA has indeed processed the files correctly. This implies:

   • TILSOTUA has correctly identified the alignment stars (see the QA plot it generates).

   • TILSOTUA has estimated the values of the TopDist and BotDist columns in the SlitObjMap table correctly.

   Once satisfied with the processed FITS file from TILSOTUA, the user can append the binary tables populated by TILSOTUA to the LRIS trace FITS files as additional HDUs, e.g.:

   from astropy.io import fits
   
   tracehdus = fits.open("trace_rawframe.fits")
   autoslithdus = fits.open("yourmaskname_output.fits")
   
   for hdu in autoslithdus[1:]:
       tracehdus.append(hdu)
   tracehdus.writeto("trace_with_maskinfo.fits")
   

If processed correctly, PypeIt should now fully utilize its arsenal of slitmask processing tools to reduce and coadd spectra with the WCS information incorporated.

Flexure

There is substantial flexure in the LRIS instrument and the default settings attempts to characterize both the spectral and spatial effects.

See the Flexure Correction notes if you wish to turn either of these off.

LRIS BLUE

This section provides information on both keck_lris_blue_orig and keck_lris_blue. When not specified, the information applies to both.

Default Settings

See KECK LRISb (keck_lris_blue) and KECK LRISb (keck_lris_blue_orig) for a listing of modifications to the default settings. You do not have to add these changes to your PypeIt reduction file! This is just a listing of how the parameters used for LRISb differ from the defaults listed in the preceding tables on that page. Moreover, additional modifications may have been made for specific setups, e.g, for different grisms, or different slitmasks, etc. You can see a list of all the used parameters in the keck_lris_blue_XXX.par file generated by PypeIt at the beginning of the reduction.

Calibrations

Wavelength calibration

Arcs

We recommend that you turn on most of the standard arc lamps, including those slated for the red side.

These are:

Ne,Ar,Cd,Kr,Xe,Zn,Hg

The archived solutions expect most of these lamps. FeAr lamp can also be used, since a FeAr line list is available for reduction process, although it appears to be less useful to obtain good wavelength solutions.

Wavelength Solution

As default, the wavelength calibration is performed using the Full Template algorithm. The templates are created in the same way as done for Keck/DEIMOS (see DEIMOS Wavelength Calibration) and kept in the data/arc_lines/reid_arxiv directory. There are four templates, one per each LRISb grism:

GRISM	template
300/5000	keck_lris_blue_B300_5000_d680_ArCdHgKrNeXeZnFeAr.fits
400/3400	keck_lris_blue_B400_3400_d560_ArCdHgNeZnFeAr.fits
600/4000	keck_lris_blue_B600_4000_d560_ArCdHgKrNeXeZn.fits
1200/3400	keck_lris_blue_B1200_3400_d560_ArCdHgNeZn.fits

PypeIt will automatically choose the right template according to the specific dataset. These templates work for both keck_lris_blue_orig and keck_lris_blue.

Flat Fielding

Pixel Flat

It is recommend to correct for pixel-to-pixel variations using a slitless flat. If you did not take such calibration frames or cannot process them, you may wish to use an archival. This link has the existing ones staged by the PypeIt team.

And then set the following in your PypeIt Reduction File:

[calibrations]
    [[flatfield]]
        frame = path_to_the_file/PYPEIT_LRISb_pixflat_B600_2x2_17sep2009.fits.gz

Warning

Internal flats may be too bright and need to be tested.

Trace Flat

We strongly recommend on-the-sky trace flats through full instrument setup. Aim for 1000 counts per pixel above bias. These are best achieved by taking twilight flats within 15 minutes of sunset/sunrise.

Warning

Internal/dome flats are likely to be too faint in the very blue.

400/3400 grism

If you are using this grism, you are likely aware there are strong ghosts. We have found these complicate edge tracing for dome flats (sky flats appear ok). Therefore, you may need to increase the edge_thresh parameter to 40 for successful performance, i.e.:

[calibrations]
    [[slitedges]]
        edge_thresh = 40

LRIS RED

This section provides information on keck_lris_red_orig, keck_lris_red, and keck_lris_red_mark4. When not specified, the information applies to all.

Detectors

There have been 3 (or is it 4?!) generations of detectors in the LRISr camera. The first detector change (from Tektronik 2kx2k to LBNL 2kx4k) happened in May 2009, concurrently with the change in the FITS file format (see FITS format). To reduce the data taken before May 2009, the user should use the keck_lris_red_orig spectrograph. For data taken with the LBNL detectors, after May 2009, the user should use the keck_lris_red spectrograph. The newest detector was installed around May 2022 and is referred to as the Mark4 detector. To reduce the data taken with this detector, the user should use the keck_lris_red_mark4 spectrograph.

Default Settings

See KECK LRISr (keck_lris_red), KECK LRISr (keck_lris_red_orig), and KECK LRISr (keck_lris_red_mark4) for a listing of modifications to the default settings. You do not have to add these changes to your PypeIt reduction file! This is just a listing of how the parameters used for LRISr differ from the defaults listed in the preceding tables on that page. Moreover, additional modifications may have been made for specific setups, e.g, for different gratings, or different slitmasks, etc. You can see a list of all the used parameters in the keck_lris_red_XXX.par file generated by PypeIt at the beginning of the reduction.

Calibrations

Wavelength calibration

Arcs

We recommend that you turn on most of the standard arc lamps:

Ne,Ar,Cd,Kr,Xe,Zn,Hg

The archived solutions expect most (or all) of these lamps.

Wavelength Solution

As default, the wavelength calibration is performed using the Full Template algorithm. The templates are created in the same way as done for Keck/DEIMOS (see DEIMOS Wavelength Calibration) and kept in the data/arc_lines/reid_arxiv directory. When the first detector change happened (from Tektronik 2kx2k to LBNL 2kx4k) in May 2009 (see Detectors), the pixel size changed from 24 to 15 microns. Because of this, different templates are used for keck_lris_red_orig and keck_lris_red. Luckily, the pixel size did not change when the Mark4 detector was installed, so the same templates are used for keck_lris_red and keck_lris_red_mark4. These are the templates available, one per each LRISr grating:

GRATING	template for keck_lris_red_orig	template for keck_lris_red and keck_lris_red_mark4
150/7500	keck_lris_red_orig_R150_7500_ArHgNe.fits	keck_lris_red_R150_7500_ArCdHgNeZn.fits
300/5000	keck_lris_red_orig_R300_5000_ArCdHgNeZn.fits	keck_lris_red_R300_5000_ArCdHgKrNeXeZn.fits
400/8500	keck_lris_red_orig_R400_8500_ArCdHgNeZn.fits	keck_lris_red_R400_8500_ArCdHgKrNeXeZn.fits
600/5000	keck_lris_red_orig_R600_5000_ArCdHgNeZn.fits	keck_lris_red_R600_5000_ArCdHgKrNeXeZn.fits
600/7500	keck_lris_red_orig_R600_7500_ArCdHgNeZn.fits	keck_lris_red_R600_7500_ArCdHgKrNeXeZn.fits
600/10000	keck_lris_red_orig_R600_10000_ArCdHgNeZn.fits	keck_lris_red_R600_10000_ArCdHgKrNeXeZn.fits
831/8200	keck_lris_red_orig_R831_8200_ArCdHgNeZn.fits	keck_lris_red_R831_8200_ArCdHgKrNeXeZn.fits
900/5500	keck_lris_red_orig_R900_5500_ArCdHgNeZn.fits	keck_lris_red_R900_5500_ArCdHgNeZn.fits
1200/7500	keck_lris_red_orig_R1200_7500_ArCdHgNeZn.fits	keck_lris_red_R1200_7500_ArCdHgKrNeXeZn.fits
1200/9000		keck_lris_red_R1200_9000.fits

PypeIt will automatically choose the right template according to the specific dataset. Note that the 1200/9000 grating was first released in 2013, so it was not available for keck_lris_red_orig.

Known issues

LRISb Slit Edges

When observing in long-slit mode, PypeIt might set the slit incorrectly for detector 2. This may occur if the counts from the flat field are too low (e.g., using internal flats rather than twilight flats with a higher signal in the blue). Therefore, if you use internal flats, be careful to inspect the slits defined by PypeIt as described in Edges.

If the defined slit(s) does not cover the portion of the illuminated detector where your source falls, you can manually define the slit position as described in Missing a Slit.

Here is an example for the PypeIt file:

[calibrations]
    [[slitedges]]
        add_slits = 2:788:10:650
        sync_predict = nearest

This will force a slit onto the detector for reduction.

Multi-slit

The code may identify a ‘ghost’ slit in empty detector real estate if your mask does not fill most of the field. Be prepared to ignore it.

Additional Reading

Here are additional docs related to Keck/LRIS. Note many of them are related to the development of PypeIt for use with LRIS data:

• Automated typing of LRIS (BLUE and RED) frames
• Automated sorting of LRIS frames by instrument configuration
• Slitmask ID assignment and missing slits
• RA, Dec and object name assignment to 1D extracted spectra
• Object position on the slit from slitmask design