K_Jo said:
Tell us more, Mr. Science!
Huh! I'll try, but I'm no expert. I know an engineer or two that would be telling me to quit while I'm ahead (I'm not an engineer).
The first part of this is the atmospheric effect I referred to as "ducting." As with any impure liquid, the amount of impurities isn't distributed evenly throughout the volume of the atmosphere. The combination of humidity, conductive dust particles, polarization of ion concentrations, and probably some other stuff can combine to make areas of the atmosphere more conductive than others. The size and shape of the unusually conductive area governs the frequency of signal that will follow it. So some frequencies are more susceptible to a given duct formation than others.
Over a body of water where evaporation causes the water to release ions and salts into the air above it, the ducting effect is much more common than in a dry area, because these things released by the water are (as mentioned before) distributed unevenly.
So you can imagine a big invisible pipe for electromagnetic energy forming in the atmosphere. This is why we (around here, anyway) can frequently hear Jose in Havana on his 5-watt CB radio. That signal is (a) not strong enough and (b) the wrong frequency to normally make it here, but it does because of ducting.
Okay, the second part of the answer is all about power. A radio frequency transmitter develops and transmits a fixed amount of power out through the antenna. In an ideal situation where the antenna is in a vacuum (no atmosphere), the antenna's shape will cause it to radiate the energy in a very predictable fashion. You've probably seen depictions of circles spreading out and away from a radio antenna in cartoons. Not a bad way to show it.
But in the real world, things like ducting and atmospheric absorption start changin the radiation pattern from the antenna, causing areas of greater and lesser energy (referred to as field strength).
In ducting and in atmospheric loss, we have a problem of the power budget.
If the transmitter transmits 5 watts of power, and 3 watts of it cover the best areas for reception, the other 2 watts are used up by ducting and atmospheric loss.
But if ducting and atmospheric loss are really bad someday, and they're stealing 3 watts of energy, you will only have 2 watts left for everything else - because 5 watts is all you get to do the job.
With solar activity and electromagnetic storms, all bets are off. Solar radiation disrupts the ionosphere and radio propagation goes to poop worldwide. But that's another story!