To answer the OP's original question.
4:2:2 is the standard system used to sample broadcast video for distribution around a broadcast centre. The 4 refers to the sampling rate that is used for the luminance (Y) i.e. the brightness signal, the 2:2 refers to the sampling frequency used for the two colour difference signals (R-Y/Cr and B-Y/Cb) i.e. the colour/chrominance.
This means that the colour information is sampled at half the resolution horizontally as the brightness - which is fine for most things (though it can cause issues with chroma-keying where you need high resolution colour information) as the eye/brain is less sensitive to colour than brightness. However vertically, in 4:2:2, the resolution is the same for both the luminance and chrominance. This has advantages in broadcast applications - particularly in interlaced systems where the vertical resolution is effectively halved on moving information where there is movement between the two fields within a frame... (*) If 8 bit sampling is used then 4:2:2 subsampling from 4:4:4 reduces the number of bits required to code each "pixel" from 32 to 16.
However for DVD, HD-DVD, BluRay and both SD and HD broadcasts to the home, 4:2:0 is used. This reduces the chrominance resolution vertically as well, by sending just one colour signal - either R-Y or B-Y for each line, and alternating them. (Incidentally the French SECAM colour system did this as well). This means that the vertical chroma resolution is now halved as well giving a balanced horizontal and vertical chroma resolution, and also requiring less space to carry - as now only 12 bits per "pixel" are required - as for every 4 luminance samples (in a 2x2 array) there is only one R-Y and one B-Y sample...
What has this got to do with satellite feeds?
Well 4:2:2 satellite links, using a different MPEG2 Profile to 4:2:0 direct to home links, are used by broadcasters for their high quality links from outside broadcast locations back to their network control areas. These links are likely to be at a higher data rate - say 24-34Mbs rather than the 3-6Mbs used to broadcast to us - and the 4:2:2 video is higher quality, and will survive around a broadcast area in quality for longer.
4:2:2 MPEG2 is not usually decodable by domestic SD satellite receivers - hence it appears to be encrypted - but in fact it isn't always. Some HD MPEG2 receivers can decode SD 4:2:2 as can most PCs with DVB-S cards and the right MPEG2 software codec (with 4:2:2 support - like the Elecard ones?)
Bottom line - 4:2:2 at a high data rate is usually used by broadcasters as it is higher quality - and thus when the signal is recompressed to broadcast to us the quality is higher than if 4:2:0 were used.
(News broadcasts may often use 4:2:0 at lower rates though)
(*) In interlaced systems the resolution of a 50Hz SD the luminance signal can drop from 720x576 to 720x288 on fast moving information, and in 4:2:2 you get a resulting 360x288 chrominance signal. Worst case scenario with 4:2:0 is that you get 360x144 resolution chroma on interlaced fast moving content, which is why 4:2:2 is used within broadcast centres. This is because a 720x576/50i signal behaves like a 720x288/50p signal on fast moving information - the trade-off with any interlaced system is that the vertical resolution and motion resolution are not independent.