This site requires javascript to be enabled to fuction correctly


Knowledgebase: ALMA Observing Tool (OT)
Should I put sources that require multiple tunings into different Science Goals?
Posted by Andy Biggs, Last modified by Sarah Wood on 14 February 2017 07:17 PM

Users may notice that the calibration times for projects that require multiple tunings (multiple source redshifts or spectral scans) are very high, particularly for projects that require multiple executions. This is because SBs in multi-tuning mode observe all the calibrators (amplitude, bandpass and sideband ratio, plus associated system temperature measurements) at each tuning with each and every SB execution. Therefore, the more tunings and SB executions that are required, the more inefficient a multi-tuning project becomes compared to observing each tuning in its own SB.

The additional calibration overheads are illustrated in the three figures below. Each figure shows the total observing time for an example of 8 sources with 8 different redshifts that require 5 different tunings (meaning that three of the sources can share a single tuning). The histograms to the left shows the total observing time when all five tunings are included in a single Science Goal (SG), while the right histogram shows the total observing time when the sources are distributed into 5 separate SGs, one for each tuning. The different colors break out the total execution time devoted to science targets (Time-on-source or TOS) in red, the time for all calibrations in green, and the time for additional overheads in purple.

The top figure shows an example of a short observation in which all five tunings of the eight sources in the combined SG can be observed in a single execution. In this case, it is more efficient to include all tunings in a single SG compared to the five separate executions for when each tuning is put into a separate science goal. 

The next two figures show the results for when the multi-tuning SG requires two executions (middle), or nine executions (bottom). In these cases, it would would take less time if each tuning was set up in its own SG. 

time requred as a function of tuning strategy

Despite this, a user should consider the following before placing each tuning into a separate Science Goal:

  • If the tunings are split into different SGs, they may be observed on different days, with different conditions and different numbers and distribution of antennas. By putting all tunings in the same SG, all sources will be observed under similar observing conditions and with the same number and arrangement of antennas. 
  • Spectral Scans will be split up into different SBs in Phase 2
(0 vote(s))
Helpful
Not helpful

Comments (0)