Lost and ProFound: Tips and Tricks

Load boring things

library(ProFound)
library(FITSio)

General advice

Pass in an image with a WCS attached (as per Rfits_read_image [Rfits], read.fits [astro], readFITS [FITSio]).

You can always plot the output of profoundProFound to get a handy 3x3 diagnostic plot.

Parameters to focus on

There are a lot of parameters in ProFound. Most do not need to be touched, some need to be touched a lot. Here are the ones to focus on in early experimentation (i.e. usually need to be adjusted away from defaults for a given data set):

• magzero - this needs to be set correctly. It is not read from the header (to many different ways this is written unfortunately).
• skycut - the most important of all, controlling the depth of extraction.
• tolerance - controls how groups are segmented (so does reltol, these two play together in a slightly complex manner)
• box - controls the sky background box size (also see grid). This needs to be set sensibly for the data.

Useful things and suggested settings:

• verbose - nice to hear what is happening
• plot - nice to see what has happened
• sigma - controls how much the image is smoothed before it is analysed. This should not need much fiddling, but it should appropriate for the data (roughly the PSF)
• SBdilate - a relatively recent addition, but it allows natural additional dilation for more extended objects (should be set in the range 1–2 usually)
• fluxtype = ‘Jansky’ is usually more useful than ‘Raw’
• roughpedestal = TRUE for extended sources that dominate the pixels of an image (stops over subtraction)

Some examples:

image=readFITS(system.file("extdata", 'VIKING/mystery_VIKING_Z.fits', package="ProFound"))

profound=profoundProFound(image, magzero=30, plot=TRUE)
profound=profoundProFound(image, magzero=30, plot=TRUE, skycut = 2)
profound=profoundProFound(image, magzero=30, plot=TRUE, tolerance = 10)
profound=profoundProFound(image, magzero=30, plot=TRUE, box = 50)
profound=profoundProFound(image, magzero=30, plot=TRUE, SBdilate = 1)
profound=profoundProFound(image, magzero=30, plot=TRUE, roughpedestal = TRUE)

Multi-band

Some different data:

GALEX_NUV=readFITS(system.file("extdata", 'GALEX_NUV.fits', package="magicaxis"))
VST_r=readFITS(system.file("extdata", 'VST_r.fits', package="magicaxis"))
VISTA_K=readFITS(system.file("extdata", 'VISTA_K.fits', package="magicaxis"))

# Warp to common WCS:
GALEX_NUV_VST=magwarp(GALEX_NUV, VST_r$hdr)$image
VISTA_K_VST=magwarp(VISTA_K, VST_r$hdr)$image

multi=profoundMultiBand(inputlist=list(GALEX_NUV_VST, VST_r$imDat, VISTA_K_VST),
magzero=c(20.08,0,30), detectbands=c('r','K'), multibands=c('NUV','r','K'))

Good but not perfect:

profound=profoundProFound(VST_r, roughpedestal=TRUE, SBdilate=1, plot=TRUE)

Try to find the best parameters for the above galaxy (e.g. particularly play with the tolerance and reltol. Once you have a decent effort, we can fix the remainder:

fixedRGB=profoundSegimFix(list(R=VISTA_K_VST, G=VST_r$imDat, B=GALEX_NUV_VST), segim=profound$segim)

profoundSegimPlot(image=VST_r$imDat, segim=fixedRGB$segim)

We can now feed this back into ProFound for our best final effort:

profound2=profoundProFound(VST_r, segim=fixedRGB$segim, plot=TRUE)