Several points on this subject but this post primarily focuses on performance and compares AM to narrowband FM.
There are multiple reason why AM remains in use for air traffic control; historic, bandwidth, performance etc. Everyone says that FM has a higher signal to noise ratio than AM.
That statement is a half truth, an urban legend that needs to be qualified. There is a vast difference in performance of each depending upon the received signal strength and FM is not always better.
The improvement in S/N ratio for FM in dB over a 100% modulated AM signal can be calculated using the formula dB = 10*Log 3*m^2, where m is the modulation index of the FM signal. The modulation index in FM is calculated by this formula m = deviation / highest mod frequency.
In a typical NFM system like a two-way radio, the deviation is 5 KHz deviation and the highest modulating frequency is 3 KHz so m = 1.667. If you plug 1.667 into the first formula the S/N ratio gain for FM over a 100% modulated AM signal is about 9.2 dB. In an FM broadcast system where the modulation index is 5, the gain is 18.75 dB!.
But there is one caveat, in order to achieve this gain in S/N ratio: all of the significant FM sidebands need to exceed a minimum signal level. (significant sideband is any whose amplitude is 1% or more of the unmodulated carrier. The amplitudes of the significant FM sidebands are not equal and some are very small. ) This is sometimes called the FM improvement threshold but more commonly called the quieting point. As the FM signal level sinks below this threshold, the S/N advantage quickly disappears because the smaller amplitude sidebands drop down below the receiver detection level. The bottom line is that when signals are strong FM will outperform AM by about 9 dB but when signals are weak there is a limit point at which AM then has a higher S/N ratio than FM.
This is because of two factors: 1. FM has multiple sidebands and all of them need to be detectable in order to have the gain in S/N ratio. 2. An FM receiver bandwidth is typically in the 15 KHz range where as an equivalent AM receiver bandwidth can be as narrow as 6 KHz. That fact by itself gives AM an approximate 4.0 dB advantage in noise reduction. This is one of the reasons why single sideband, a form of AM, works best for weak signal Dxing.
To summarize: If the FM signal exceeds the quieting point then it will typically have a higher signal to noise ratio than an equal level AM signal. As the FM signal level drops below this point, the AM signal will eventually have a higher signal to noise ratio.
Also to clarify the noise immunity of AM vs FM, there are theoretically similar limitations between the two systems. There are modulation index conditions in which AM outperforms FM in noise rejection. However in practical systems this isn't the case. Also it is generally considered that atmospheric noise is not a significant factor at VHF frequencies and above.
As far a changing to digital comm as others have mentioned. The Nexcom program is replacing FAA VHF radio equipment nationwide. These are dual-mode equipment, capable of both conventional AM and digital modulation (D8PSK). The digital mode will support both digital voice and data services. From what I have gathered, the digital modes will be first introduced in the super-high sectors because the aircraft that fly in those sectors have more modern systems are therefore more likely to be able to use the digital modes. The MDR (multimode digital radio) equipment is capable of 25 KHz and 8.33 KHz channel spacing
There is a simultaneous but unrelated project underway to replace the UHF comm radio equipment. These use conventional AM and are not digital capable and they are only capable of 25 KHz channel spacing.
Attached are links to data sheets on each.
VHF radio
http://www.acd.itt.com/pdf/CAVU_2100.pdf
UHF radio
http://www.gdc4s.com/documents/D-CM300-10-0706.pdf
D8PSK
http://spacecom.grc.nasa.gov/icnsconf/docs/2002/04/Session_B1-1_Hung.pdf
Dan
wa8vzq@yahoo.com