In rural areas with no cell service, AM radio is often the only available form of weather and news. It's also a great form of emergency information broadcasting in the event of a natural disaster.
There might be a silver lining to the slow death of commercial terrestrial radio: Doing away with the polite fiction that commercial terrestrial radio is a valid way to communicate emergency information. Even if the "local" radio stations are active and broadcasting a signal, they're often only local in a way that does not matter when it comes to emergency broadcasts:
> The Minot train derailment occurred just west of Minot, North Dakota, United States, on January 18, 2002, when a Canadian Pacific Railway freight train derailed, spreading ammonia gas across the city, delaying rescue operations. The cause was found to be small fatigue cracks in the rails and joint bars, not detectable by the inspection routines then enforced by Canadian Pacific.
[snip]
> Because it was the middle of the night, there were few people at local radio stations, all operated by Clear Channel with mostly automated programming. No formal emergency warnings were issued for several hours while Minot officials located station managers at home. North Dakota's public radio network, Prairie Public Broadcasting, was notified and did broadcast warnings to citizens.
Public radio succeeded. Commercial radio did not. People have been utterly resistant to learning this lesson. Maybe once commercial radio is dead they'll have no other choice.
Obviously, cell phones can do it. They already do this for a lot of people.
NOAA Weather Radio All Hazards already exists as well, but the receivers should be more widespread. Putting them into cars and TV sets would be a good start, especially with GPS receivers to automatically figure out where they are and, therefore, which alerts to listen to.
Giving more funding to noncommercial radio stations, to guarantee they exist, for one thing, and have people in the booth all the time, for another. America doesn't have an analogue to the BBC so ensuring the local and regional public stations remain usable for this purpose is essential.
> alerts that go out to cell phones probably reach more people than radio.
Only if the disaster is mild enough that cell phones still work. And let's not forget that there are enormous areas of the US where there is no reliable (or any) cell service.
I'm just hoping that for disaster communications, nobody is actually deciding to cut some people off just because there aren't enough of them.
The only solution is multiple solutions. Commercial radio has a threshold of how small a market can be before it isn't worth it any longer, noncommercial radio (including NOAA weather radio stations) can fill in those gaps but require people to have the receivers handy, and more people carry cell phones than radios nowadays but cell service isn't absolutely everywhere. We can improve each of those individual technologies, but no single technology is going to become the only answer.
https://en.wikipedia.org/wiki/Minot_train_derailment
> The Minot train derailment occurred just west of Minot, North Dakota, United States, on January 18, 2002, when a Canadian Pacific Railway freight train derailed, spreading ammonia gas across the city, delaying rescue operations. The cause was found to be small fatigue cracks in the rails and joint bars, not detectable by the inspection routines then enforced by Canadian Pacific.
[snip]
> Because it was the middle of the night, there were few people at local radio stations, all operated by Clear Channel with mostly automated programming. No formal emergency warnings were issued for several hours while Minot officials located station managers at home. North Dakota's public radio network, Prairie Public Broadcasting, was notified and did broadcast warnings to citizens.
Public radio succeeded. Commercial radio did not. People have been utterly resistant to learning this lesson. Maybe once commercial radio is dead they'll have no other choice.