National Radio
Astronomy Observatory

5.2.5.5 Sub-parameter interpolation

The interpolation sub-parameter controls how spectral channels in the MS are gridded into the output image cube. This is available in all modes except ’mfs’. The options are: ’nearest’, ’linear’, ’cubic’.

For ’nearest’, the channels in spw beyond the first are mapped into the nearest output image channel within half a channel (if any).

For ’linear’, the channels are gridded into the planes using weights given by a linear function of the frequency of the MS channel versus the plane. Each input channel will be mapped to 1 or 2 output planes. For most users, this is the best choice.

For ’cubic’, the channels are gridded using a cubic interpolation function.

’Linear’ and ’cubic’ interpolation methods require that there are two datapoints that sandwich your new, regridded bin. This can introduce edge effects like in the first or last channel or adjacent to flagged channels where data is only available on one side of the spectrum. interpolation=’nearest’ will avoid such edge effects but may not work so well for data with spws that overlap. For mode=’velocity’ or ’frequency’, ’linear’ interpolation usually works best and for mode=’channel’ the ’nearest’ interpolation method is superior. But this could be different for your dataset and you should carefully check your results.

More information about CASA may be found at the CASA web page

Copyright © 2010 Associated Universities Inc., Washington, D.C.

This code is available under the terms of the GNU General Public Lincense