zSALT

Measuring redshifts from SALT longslit and MOS observations.

Instructions

1. Clone the git repository onto your device

   git clone https://github.com/crawfordsm/zSALT.git

2. Run the script 'gettemplates.sh' in the directory to download the SDSS templates.

3. Copy the script 'reducedata.py' to the directory that has the data that you want to reduce. Edit this file to point to the zsalt directory, the proposal code, and the bad pixel file map. If you are using MOS data, edit the file to point to the xml file describing the MOS mask and to use mosred instead of specred.

4. The directory where you will run the data should have the format such that in the top level directory there should be the obsdate (e.g. 20131227) and then the raw directory inside that. The raw directory should only include the files that you want reduced so any extra files which are sometimes included in your directory should be removed.

5. Run reducedata.py with

   python reducedata.py 20131227

   Replace the obsdate with the appropriate observations date

6. Reducedata should step through all the tasks needed to produce wavelength calibrated images. There will be one interactive step to identify arc lines.

7. Once complete, inspect the final product in the sci directory that was created. Skylines should be straight and cosmic rays should be cleaned.

8. To extract spectra, use extractobject.py. This can be called just by giving the frame name in which to extract the object. It will then ask for the central row of the object to be extract. This can be repeated for different rows.

   python extractobject.py xmfxgbpP201302010093.fits

   You will either need the full path to extractobject.py or copy it into the directory. This will perform a sky subtraction and extraction of the spectrum.

9. To calculate the redshift, use redshift.py. You will have to give it a template to match and different templates are available for matching. The format of it is:

   python redshift.py [SPECTRUMFILE] [TEMPLATEFILE]

   This will display the spectrum, measure a redshift, and overplot the template at the best fit redshift. This may need to be repeated for different redshifts