HRS Pipeline

HRS data products are distributed automatically every day to PIs, based on the MIDAS HRS pipeline (Kniazev et al. 2016a, 2016b). All available documentation for MIDAS HRS pipeline: (1) how it works, (2) what the various data products are and (3) different studies on its accuracy, is available at http://www.saao.ac.za/~akniazev/pub/HRS_MIDAS/. See also the list of documents towards the end of this page. Note that the MIDAS pipeline currently does not include the High Stability (HS) mode.

Please cite the references above when you use the data products in your publications. Astro Ops is very grateful for all feedback regarding the pipeline products. Please do not hesitate to send suggestions and criticism to salthelp. While the dedicated development work has been completed we anticipate further improvements and tweaks, especially after user feedback.

Accuracy of the pipeline

We have found that the radial velocity accuracy of the automatic HRS MIDAS pipeline products is <300 m/s for the LR, MR, HR modes using the default calibrations (i.e. nothing extra requested by the PIs, arcs and RV standards not taken in the same night necessarily). More information is available under the MIDAS documentation.

Better accuracy might well be achieved with more involved calibrations and analysis methods.

Which data are reduced

Currently all data from the beginning of 2016-2 are reduced and available for HRS PIs. Most of 2016-1 has also been run, and parts of 2015-2. If not already included in the daily deliveries, pipeline data-sets can be accessed simply by using the “Request data” box on the relevant block item on your program’s Web Manager page.

Older data-sets can in principle be run as well, but we request PIs needing specific dates to request them from salthelp, for us to prioritise running past sets through the pipelines. Please note that due to limited resources we cannot promise immediate response; also, some datasets from early 2015, and especially from before October 2014 may prove difficult to reduce accurately.

In summary, if you cannot find the pipeline products of your HRS data after requesting it on the WM, do contact salthelp to inquire about the chances to get them done.

Selected documentation

HRS MIDAS Pipeline by Alexei Kniazev:

  • HRS_pipeline.pdf – general description for the MIDAS HRS pipeline and its Data Structure. If you would like to understand which kind of data you received from this pipeline – read it here.
  • HRS_LRr.pdf – general description of the MIDAS HRS reduction + its implementation for the LR red arm data. All general algorithms are HERE. They are similar for all others modes, so all other documents will refer to THIS ONE
  • HRS_LRb.pdf – MIDAS HRS reduction to the HRS LR blue arm data
  • HRS_HRr.pdf – MIDAS HRS reduction to HRS HR red arm
  • HRS_MRr.pdf – MIDAS HRS reduction to HRS MR red arm
  • HRS_RV.pdf – velocities accuracy with MIDAS HRS pipeline
  • R_HRS.pdf – resolution and FWHM with MIDAS HRS pipeline for different modes
  • R_HRS_tilt.pdf – study for possible implementation of the tilt correction for the HR mode with MIDAS HRS pipeline
  • HRS_SPST.pdf – Spectrophotometric standards for HRS

Unsupported alternative

Alternatively, HRS users have an additional option and may also choose to use PyHRS (Crawford et al. 2016) which is available at https://github.com/saltastro/pyhrs. Note that HRS users have to manually run PyHRS themselves. Unfortunately, PyHRS pipeline is NO LONGER supported after the main developer for the PyHRS pipeline left the SALT project. PIs should feel free to run PyHRS themselves.

Note that the PyHRS pipeline does mainly include LR, MR, HR modes. Possibly, High Stability (HS) mode data can be reduced by running tasks available for the PIs as part of the PyHRS software distribution. Please cite the reference above when you use the data products in your publications.

Further documentation for the PyHRS version, about how it works and what the various data products are, is available at https://github.com/saltastro/pyhrs. Note once again that PyHRS can be run by HRS users themselves using the software distribution available under this link. The Astro Ops team DOES NOT support it anymore and we are not able to help users with this software.