help with rvsao.sumspec
marco.03 wrote on Jan 09, 2011
Hello. I am new to the forum, and hope you can help clear some doubts I have with the sumspec task of the rvsao package.
I am trying to create a template spectrum from five individual spectra of the same object. For each of these I have computed the barycentric velocity correction with the bcvcorr task. I have set the svel_co parameter to 'file', in order to read the velocity corrections from the image header.
I have noticed that running sumspec corrects all spectra but the first one: the online manual states 'If svel_corr is not set to "none", but velcomp and zcomp are INDEF, data is shifted to the barycentric velocity correction of the first spectrum, if it has one.' If that was true, I would expect all spectra but the first to be shifted by the difference of barycentric velocities between each individual spectrum and the first. Instead, the other spectra are shifted by their absolute value.
I'll try to make myself clearer: the five spectra have their BCV header keywords equal to 27.01576, 27.01193, 27.0079, 27.00367, 26.99925 (km/s).
I would expect spectra from 2 to 5 to be shifted by, respectively, 0.0383, 0.00786, 0.01209, 0.01651 and the composite spectrum to retain the BCV value of the first spectrum. Instead, the composite spectrum has BCV=0 and spetra 2 to 5 are shifted by 27.01, 27.01, 27.00, 27.00.
I assume this means that the first spectrum is shifted in the first place to zero velocity, although the log doesn't say so explicitly (actually the log says that only the other ones are shifted). Is this correct?
thanks in advance for your help
Marco
I am trying to create a template spectrum from five individual spectra of the same object. For each of these I have computed the barycentric velocity correction with the bcvcorr task. I have set the svel_co parameter to 'file', in order to read the velocity corrections from the image header.
I have noticed that running sumspec corrects all spectra but the first one: the online manual states 'If svel_corr is not set to "none", but velcomp and zcomp are INDEF, data is shifted to the barycentric velocity correction of the first spectrum, if it has one.' If that was true, I would expect all spectra but the first to be shifted by the difference of barycentric velocities between each individual spectrum and the first. Instead, the other spectra are shifted by their absolute value.
I'll try to make myself clearer: the five spectra have their BCV header keywords equal to 27.01576, 27.01193, 27.0079, 27.00367, 26.99925 (km/s).
I would expect spectra from 2 to 5 to be shifted by, respectively, 0.0383, 0.00786, 0.01209, 0.01651 and the composite spectrum to retain the BCV value of the first spectrum. Instead, the composite spectrum has BCV=0 and spetra 2 to 5 are shifted by 27.01, 27.01, 27.00, 27.00.
I assume this means that the first spectrum is shifted in the first place to zero velocity, although the log doesn't say so explicitly (actually the log says that only the other ones are shifted). Is this correct?
thanks in advance for your help
Marco
Mike Fitzpatrick wrote on Jan 09, 2011
The RVSAO package is distributed by SAO (CfA) so if you don't get a reply here I'd suggest you contact Doug Mink (dmink@cfa.harvard.edu) directly. I know Doug monitors this site, but unfortunately we don't have much expertise with this specific task.
-Mike
-Mike
marco.03 wrote on Jan 09, 2011
Thank you for your advice. I have tried to contact Doug Mink, but I thought that maybe someone in the community already had some expertise with the package.
I would like to add one detail, for anyone who could help me: cross-correlating one of the spectra with the template, and correcting the spectrum (but not the template, which now has BCV=0) for barycentric velocity I would expect to have a value of the radial velocity close to 0 (am I wrong?). Instead, xcsao yields a value of approx. 26, which seems to indicate that the original spectrum has not been shifted. It shouldn't be a problem due to the template, because it has BCV=0 so changing the velocity correction of the template from 'none' to 'file' has obviously no effect.
Marco
I would like to add one detail, for anyone who could help me: cross-correlating one of the spectra with the template, and correcting the spectrum (but not the template, which now has BCV=0) for barycentric velocity I would expect to have a value of the radial velocity close to 0 (am I wrong?). Instead, xcsao yields a value of approx. 26, which seems to indicate that the original spectrum has not been shifted. It shouldn't be a problem due to the template, because it has BCV=0 so changing the velocity correction of the template from 'none' to 'file' has obviously no effect.
Marco
Last post on Jan 09, 2011