search

e-merlin / eMERLIN_CASA_pipeline

This is CASA eMERLIN pipeline to calibrate data from the e-MERLIN array. Please fork the repository before making any changes and read the Coding Practices page in the wiki. Please add issues with the pipeline in the issues tab.
GNU General Public License v3.0
14 stars 11 forks source link

Avoid short baselines to compute bandpass tables #61

Closed jmoldon closed 6 years ago

jmoldon commented 6 years ago

The triangle Mk2/Lo - Pi - Da give 10-14 km baselines with some huge boradband RFIs that are not very easy to eliminate completely. We can try to set a uvrange > 15km in the bandpass steps to have a more accurate bandpass calibration by avoiding that triangle. Then flagdata(mode='TFCROP') can eliminate broadband RFIs that aoflagger didn't clean completely.

This should work even if some antennas are missing, but we may need to do some quick tests to verify.

jmoldon commented 6 years ago

Not the best option. In the test I conducted, Cm has huge RFI between 1.30 and 1.32 GHz (higher part of IF1 or spw0). That means that with the uvrange cut we need to solve for 5 stations (Mk2, Kn, De, Pi, Da) with only 7 baselines, and the solution is not good enough. The situation would be even worse if any antenna is missing in the observation.

jmoldon commented 6 years ago

Deactivated by default. If we want to do it, we just need to include the uvrange parameter.