flux error in Duchamp

[sara.shakouri@…]

Hi There,

I'm a Duchamp user. I need the total flux error for the sources found by Duchamp. I could not find in the output file. Is there anything I can do to get it in Duchamp?

[MatthewWhiting]

Hi, Can you tell me what you mean by the total flux error? Do you mean the sum in quadrature of all the individual pixel errors (which, for the assumption of a single noise value sigma would be sigma*sqrt(NumPixels?))?

This quantity should be calculable for a given set of results, assuming you have calculated the statistics (since the standard deviation & the MADFM are reported in the results file).

If you like I can add this calculation to a future release. If there's some other calculation you require, let me know.

[MatthewWhiting]

Some further thoughts: That calculation is not perfect, as it does make these assumptions of independent pixels and constant noise etc. But then those assumptions are made elsewhere, in the reconstruction for instance, so we can probably accept it.

Does anything else require an error estimate?

[MatthewWhiting]

This has been put into the code, and will be available when I release the updated version very soon.

Just to summarise, Duchamp now calculates two errors:

• one on the total flux (F_tot) as: sigma*sqrt(N)
• the other on the integrated flux (F_int) as C*sigma*sqrt(N)*delta_V

where:

• sigma = the noise level in the cube
• N = number of voxels in the object
• C = correction factor for the beam (to go from eg. Jy/beam to Jy)
• delta_V = velocity width of a channel (to go from Jy to Jy km/s - or whatever the spectral units are...)

I'll let you know when the next version is available (just finishing off a few things prior to release).