I haven't had much satisfaction from attempts to look at consonants with
formant analysis, so I haven't included any of that here. The
plots that appear below show the time-function in each case.
The plot for "Mark's 'catum'" may be a little misleading because the spikey
appearance of the first vowel makes it look as if it has spikes that are similar to
what I'm calling the plosive "t". I've spent considerable effort
in Gold Wave convincing myself that there really is a plosive "t" there, and the
vowel doesn't sound at all as the plot suggests it might. I even slowed
the reproduction down in Gold Wave to 30% of normal, so that I could clearly see and
hear the plosive "t" occurring at the point where I say it is.
I've included the plots of my own pronunciations to provide background
information on what a plosive "t" looks like in a time-domain plot.
My conclusion from all of this is that Mark clearly has a plosive "t" in
"catamaran".