Interpreting the sub-linear Kennicutt-Schmidt Relationship: The case for diffuse molecular gas

On galactic scales, the star formation rate is observed to be strongly
correlated with the gas surface density.  Accurately measuring the
parameters of this power-law correlation, now known as the
Kennicutt-Schmidt (KS) relationship, is important for building
realistic models of star formation and the ISM.  Recent hierarchical
statistical analysis has shown that there is no "universal" KS
relationship, as the parameters varies between galaxies.  Moreover,
many individual galaxies, as well as the mean trend for two
observational surveys, clearly evince a sub-linear KS relationship.
This suggests that a number of other physical properties besides the
CO traced gas influence the star formation rates.  Further, the
sub-linearity in the KS relationship indicates that CO emission is not
solely originating from star forming gas (e.g. "GMCs") but rather that
CO also traces a diffuse non-star forming diffuse component.  The
presence of such diffuse molecular gas is supported by a number of
other observational results, such as extended CO emission, the
similarity between CO and HI linewidths, and high 12CO/13CO intensity
ratios.  I will discuss how such a diffuse component is a natural
outcome of a hierarchical, turbulent ISM.  Consequently, the inferred
molecular gas depletion time is not constant, but rather increases
with increasing molecular gas surface density, as traced by CO
emission.