Logscale with SPECPLOT?
cmcclel3 wrote on Dec 08, 2008
I'm trying to plot multiple spectra (for comparison/time evolution) with ONEDSPEC>SPECPLOT. Is there a way to have the fluxes of each one in logscale?
If not, does anyone know of any macros or examples of how to do this (multiple spectra, scaled and shifted) in Supermongo or GNUplot?
Thanks a bundle.
If not, does anyone know of any macros or examples of how to do this (multiple spectra, scaled and shifted) in Supermongo or GNUplot?
Thanks a bundle.
Mike Fitzpatrick wrote on Dec 08, 2008
Are you overplotting the spectra or stacking them?
If stacking, the offsets in intensity are relative, so e.g. you could use the GTOOLS command ":/ytransform log" followed by 'r' to redraw and you would see then entire Y-axis put on a log scale but not the individual spectra with a constant offset. Tasks like IMFUNC or IMEXPR can be used to take the log of the flux and then stacking these images may be more what your after.
If overplotting, there is no offset and so the ":/ytransform" trick may work (see 'help gtools' for details).
-Mike
If stacking, the offsets in intensity are relative, so e.g. you could use the GTOOLS command ":/ytransform log" followed by 'r' to redraw and you would see then entire Y-axis put on a log scale but not the individual spectra with a constant offset. Tasks like IMFUNC or IMEXPR can be used to take the log of the flux and then stacking these images may be more what your after.
If overplotting, there is no offset and so the ":/ytransform" trick may work (see 'help gtools' for details).
-Mike
cmcclel3 wrote on Dec 08, 2008
SPECPLOT must be stacking them, since the ":/ytransform log" trick effectively reduces my five spectra separated by a constant offset to a single line. The problem with using IMFUNC is that it takes the log of both the flux and wavelength of each of my sprectrum.fits files, where I only need the former in logscale.
Mike Fitzpatrick wrote on Dec 08, 2008
See the SARITH task help page .....
Francisco Valdes wrote on Dec 08, 2008
This seems like a useful feature. For the next version of IRAF there will be a "transform" parameter which can be set to "log" to take the log of the spectra on input. If you have any other suggested transforms now would be the time to suggest them.
Yours,
Frank Valdes
Yours,
Frank Valdes
Last post on Dec 08, 2008