Global 5G wireless deal threatens weather forecasts

November 24, 2019 10:29 PM

"5G transmissions will involve many frequencies, but the key one under discussion is 23.8 gigahertz. Water vapour in the atmosphere naturally produces a weak signal at this frequency, which satellites use to measure humidity. Those data feed into weather forecasts. But if a 5G station is transmitting a signal near the 23.8-gigahertz frequency, a weather satellite might pick it up and interpret it as water vapour. And that bad data could degrade forecasts."
posted by They sucked his brains out! (22 comments total)

Counting down until this appears on Alex Jones as "mainstream media admits 5G causes hurricanes blargle blarg declassified weather weapons."
posted by Scattercat at 12:00 AM on November 25, 2019 [3 favorites]

Just another reason to stay indoors and never put down your phone....
posted by Sheydem-tants at 3:29 AM on November 25, 2019

Weather forecasts are a big waste of time. With all the mega forecast tech they tout, you are better off sticking your head out da window.
posted by DJZouke at 4:48 AM on November 25, 2019

The NOAA knows they have a modeling problem, but they're working on it.
posted by zamboni at 5:21 AM on November 25, 2019 [1 favorite]

Apparently, money trumps physics.
posted by ZeusHumms at 6:26 AM on November 25, 2019 [1 favorite]

Weather forecasts are a big waste of time.

Can't tell if serious or joking, but this is obviously not true. People love to shit on meteorology because it's an imperfect science at best, but the 5- and 7-day forecasts are about 90% and 80% accurate, respectively.

That's a hell of an improvement from not having any idea what the weather is going to do in the future, which was the dominant state throughout much of human history.

I think people do not realize how much global transportation, and thus the economy, relies on weather forecasting and satellite-based weather observation.
posted by Kadin2048 at 6:50 AM on November 25, 2019 [53 favorites]

But the FCC auctioned off the first chunk of the 5G spectrum with minimal protection. The sale ended on 17 April and reaped nearly US$2 billion.
Who reaps this?
posted by MtDewd at 7:10 AM on November 25, 2019 [4 favorites]

Psht! The whirlwind obvs!
posted by evilDoug at 7:23 AM on November 25, 2019 [9 favorites]

Warm money evaporates into the atmosphere, then rises and condenses in the upper atmosphere, releasing heat and intensifying the money updraft. As long as there is enough warm money available the whirlwind will continue to intensify, absent money shear or 5G cellular interference.
posted by Huffy Puffy at 7:28 AM on November 25, 2019 [12 favorites]

Weather forecasts are a big waste of time.

joking, right? how many trillions of dollars and thousands of lives does NWS save annually? and then there's climate change on top of it all?
posted by eustatic at 7:54 AM on November 25, 2019 [4 favorites]

Weather forecasts are a big waste of time.

Over the past decade, the average error in hurricane forecasts at 36 hours before landfall is 50 miles or so. 30 years ago, the average error was more like 175 miles. If that seems like not a big difference, a 50 mile radius from Pensacola, FL has about 900,000 residents. A 175 mile radius has over 4,500,000 residents and includes Tallahassee and New Orleans. One of those can possibly be evacuated; the other -- no way.

The other way to look at the accuracy is that 50 mile error in hurricane tracking that's now 36 hours away (at least one entire day's worth of daylight, hopefully two) used to be available only 12 hours before the storm hit, 30 years ago. The 175 mile error that 30 years ago used to be 36 hours warning is now over 96 hours. That's critical in planning and protection; especially since it means that if warnings are more accurate, people take them more seriously. If you give three warnings a year, and usually none of them happen, then people will start ignoring them. If you give one warning every couple of years and it's usually right, then people will take action.

So there's actually been a lot of improvement in 30 years of forecasting, and it's helped save lives. Oh, hey - here's an interesting article: 5G Coverage Could Set Back Accurate Weather Forecasts By 30 Years
posted by Homeboy Trouble at 10:34 AM on November 25, 2019 [22 favorites]

Tangential but interesting, The weather is big business, and it’s veering toward a collision with the federal government. Basically a look at the challenges NOAA faces in navigating the changing landscape of private data collection and forecast modeling.
posted by peeedro at 10:48 AM on November 25, 2019

Companies that operate 5G networks will have a relatively loose standard from now until 2027. After that, the regulation will get stricter. The idea is to let 5G companies start building networks now, and then to add more protection for Earth observations as 5G transmissions become denser.

This is so incomprehensibly dumb.

Also, is there an actual reason why it's so terribly important to rush into 5G right now? The only people clamoring for it seem to be those who plan to use it to make lots of money and/or further infest our daily life with their technology.
posted by Not A Thing at 11:33 AM on November 25, 2019 [5 favorites]

And nobody thought this through?
posted by acb at 12:05 PM on November 25, 2019 [2 favorites]

Not to be taken literally. I should have clarified what I said some. In the case of severe weather that is life threatening, I watch the weather. So should people in the path of a tornado, hurricane etc. But here in the great NE of the USA, weather prediction is not exactly accurate much of the time. I'm stupid but not that stupid.
posted by DJZouke at 12:39 PM on November 25, 2019 [2 favorites]

The Anti-Science elements in the U.S., particularly the Climate Change deniers promoters, definitely have a disincentive to have accurate weather forecasts, even the shortest range ones. The upcoming Mass Extinction is a major profit center for some.
posted by oneswellfoop at 4:37 PM on November 25, 2019 [2 favorites]

It does make me wonder if having less accurate weather information would make it easier for the US government to misinform and control the public, as severe weather events get worse and more frequent.
posted by They sucked his brains out! at 8:35 PM on November 25, 2019 [2 favorites]

Water vapor is a potent greenhouse gas, and is apparently responsible for about 60% of the warming of Global Warming, but it functions as a multiplier of the effects of CO₂ since the warming due to extra CO₂ is what caused more water vapor in the air in the first place. Then that water vapor causes more warming, which produces more water vapor, and so on, so that water vapor is part of one of the many positive feedback loops that make Global Warming so frightening to me.

I looked around briefly, but I didn't find a definitive claim that the amount of water vapor in the atmosphere is an important measure of the progress and extent of Global Warming as well as a harbinger of more to come, but it seems likely that it would be.
posted by jamjam at 1:04 AM on November 26, 2019 [1 favorite]

But here in the great NE of the USA, weather prediction is not exactly accurate much of the time

Predicting the weather is hard. People demand exact numbers or simple yes/no answers from a probabilistic science, and often, they don't like the answer they get.

Forecast verification is about assessing that accuracy. isn't just a simple matter of being right N% of the time, it's more how wrong are you, how often?

In the US, the National Digital Forecast Database is one of the ways this is tracked. Different forecasts are compared to actual measurements, using a variety of scoring systems defined on the help page. Things with specific values like temperature uses mean average error, probability of precipitation is Brier score, sky cover is Heidke skill score.

Here's a graph of Brier scores for probability of precipitation in October, with regions plotted against time. You switch months and forecast types using dropdowns on the left column. Generally, Eastern seems to do comparatively well for everything except probability of precipitation. Blame the Atlantic Ocean, I guess?
posted by zamboni at 8:46 AM on November 26, 2019 [7 favorites]

I'm surprised the same (time slot?) (spread spectrum?) techniques used for other radio equipment can't be used for water vapor sensing with good 5G rejection.
Anyone knowledgeable care to weigh-in on this?
posted by ZenMasterThis at 9:47 AM on November 26, 2019

If we wanted to reduce contention we'd have to basically have all radios, both provider and client, totally hush up whenever a sensing satellite was going to pass anywhere near overhead. Given that the satellites (that I know of) have beam widths of ~1 degree (say, 20 x 20 km square) and pass overhead about every 100 minutes that'd be ... interesting. Not impossible, but interesting.

I can't speak to how much margin you'd need outside the above numbers. Radio is my thing but propagation up at 22+ GHz is most definitely not.
posted by introp at 11:25 AM on November 26, 2019 [1 favorite]

Elon Musk's communication satellites are fucking up astronomy too
posted by vibratory manner of working at 3:52 PM on November 27, 2019