Actually, the different lines of the different gases are not so teeny: they form the entire signal!
I tried to find a non-pay picture of a CIRS spectrum and on this page there is a simulation of a 'limb' obervation:
http://www.markelowitz.com/titan.htmCIRS measures the thermal radiation coming off Titan as a function of wavelength. The molecular signals are caused by the fact that different molecules absorb and emit only at very specific wavelengths. The difference in thermal flux for different wavelengths is basically a result of the difference in altitude (and corresponding temperature at those altitudes) at which the thermal radiation originates. At wavelengths of high absorption, the radiation comes from higher in the atmosphere (for non-limb observations). For optically thin limb observations (you can look straight through the atmosphere at space at the horizon) higher absorption by the gases means more radiation is emitted.
In both cases you first need to know the temperature of the atmosphere as a function of altitude before you can derive the amounts of gases present, because the thermal emission is dependent both on the temperature and the amount of absorption by the gases.
Here's an example of looking for seasonal variations:
http://adsabs.harvard.edu/abs/2011epsc.conf...20T