[♪ INTRO] For all that we complain about weather forecasts,
they’re actually pretty amazing. A 5-day forecast today is about as reliable
as a 24-hour forecast was in 1980. It’s not just about the convenience of having
your umbrella, either. Good weather forecasts save lives. Hurricane forecasts, for example, can give people days instead of hours to get
to safety. So it’s understandable that meteorologists are up
in arms about a big potential threat to forecasting: the way the US Federal
Communications Commission, the FCC, is handling the next generation of cell phone
service, known as fifth-generation or 5G. The issue is that the FCC wants to let cellular
carriers use some radio frequencies that are right next door to the frequencies
measured by weather satellites. Those weather measurements need to be very
sensitive for models to do their work well. So scientists are worried that unless the
FCC changes its guidelines, 5G transmissions could drown out the measurements, setting weather forecasts back decades. 5G technologies promise something we all want:
to make our phones faster and more reliable. One of the biggest changes 5G makes to get
there is to move to new radio frequencies not previously used
for cell service. Different frequency ranges come with different
advantages and disadvantages. With higher frequencies, it’s easier to get faster
speeds and lots of devices on the same network. That’s because each range of cellular frequencies
gets divided up into channels, thin slices of the radio spectrum. Like a pipe carrying water, each channel can
only carry a limited amount of data. The wider a channel’s band of frequencies,
the more data it can carry. This is actually why we use the word bandwidth
to mean capacity. When a cell tower and a mobile device communicate, they take up capacity on one of the available
channels. And in lower frequency ranges, capacity is
a valuable commodity. These frequencies tend to be very in-demand, so the cellular industry gets relatively small
blocks of spectrum to work with. But there’s much less competition for higher
frequencies. So it’s easier to allocate big blocks and
carve them into lots of wide channels. The network can then support more devices,
and more bandwidth for each device. Now, the downside of higher frequencies is, well, there are reasons they’ve historically been
snubbed. For one thing, researchers will tell you it’s
just harder to engineer the equipment. A more fundamental issue is that higher-frequency
radio signals don’t travel nearly as far. They quickly fade as they pass through air, and even more so through trees, and walls,
and rain, and fog. To balance these tradeoffs for different use cases, 5G standards are designed
around 3 frequency ranges: Below 1 gigahertz, for long-distance links
that can tolerate lower speeds; 1 to 6 gigahertz, for a balance of range and
bandwidth; And above 24 gigahertz, for super-fast communication
in areas like cities where small cell towers can be placed every few
hundred meters to compensate for that signal fading. It’s that highest range that’s causing
the meteorology ruckus. In the US, the rights to use each band in each
geographic area are auctioned off by the FCC. As part of the government’s big push to
get 5G off the ground, the FCC recently ran its first few auctions
for 5G bands, including one that starts at 24.25 gigahertz. That band is appealing to telecoms, because
it’s the furthest-traveling and the easiest-to-engineer among the uncrowded
higher frequencies. But it is also uncomfortably close to a critical
frequency for weather forecasting: 23.8 gigahertz. The 23.8 gigahertz channel
is used by weather satellites to measure water vapor in the atmosphere,
a pretty key variable for forecasting. Roger Saunders, a meteorologist with the UK
Met Office, explained to SciShow that that’s because both the ground and
the water vapor in the air are constantly giving off and absorbing radio
waves all over the frequency spectrum. And water vapor happens to absorb and emit
particularly strongly at 23.8 gigahertz. And it’s not a super-intense effect. The water vapor doesn’t absorb that much of what’s emitted by the ground, and it
doesn’t emit too much itself. So in most places, water vapor has only a very
small impact on the signal detected by satellites. But, even though it’s a small signal, it’s
a very sensitive signal: changes in the amount of water vapor make
a noticeable difference in how much 23.8 gigahertz radiation filters
up to the satellites. So if a satellite passes over some point on
Earth and sees an unusually weak or strong 23.8 signal, that
can indicate a change in the amount of water vapor in that column
of atmosphere. If it is water vapor, it’s specifically
low-altitude water vapor, since that’s where the water vapor gives
off the strongest 23.8 signal. This isn’t the only frequency that we can
use to detect water vapor throughout our atmosphere, but scientists
argue that the 23.8 channel is crucial. Not only does it tell us a lot about vapor
at lower altitudes, it’s also important for checking the assumptions behind measurements
at other heights and of other variables. By helping to paint this whole picture, the satellite instruments that observe this
channel slash the error rate of forecasts worldwide. Now, theoretically those measurements should
be unaffected by 5G. Remember, the 5G band starts at 24.25 gigahertz,
which is a different number! The catch is that no radio transmits at one
precise frequency. A transmission on any given channel is more
of a smear across different frequencies, with a peak at the target frequency. So the concern is that some of the off-target
noise from 5G transmitters could bleed over into the 23.8 channel. The FCC does impose limits on how much off-target
noise a transmitter is allowed to produce. But those limits are much higher than almost
every other country outside of the U.S. is recommending for this
band. With lots of 5G radios hollering away, the noise bleeding over from 24.25 into 23.8
could add up to a lot. Scientists from NASA, and NOAA, and even the
US Navy are warning that satellites could pick up this noise and think
it was emitted by water vapor. That would make measuring the 23.8 signal like trying to listen to your friend in the middle
of a concert. A study by NOAA and others suggests that satellites
would lose 77% of certain microwave data, which would
set forecasts back by 40 years. However, it is not totally clear how serious
this risk is. Unsurprisingly, telecom industry advocates
claim everything is fine. They claim, for example, that current satellite
instruments are less susceptible to interference than those
in the NOAA study. Also, 5G transmitters could probably be designed
and deployed carefully enough that they don’t send too much off-target
noise upward toward the satellites. But NOAA hasn’t made its study data public
yet, and until it does, it’s the claims of the cell phone industry
and the FCC against those of NOAA, NASA, and the Navy,
and other scientists. Meanwhile, many scientists are alarmed, and
say the FCC should at least tighten the limits on spillover, even though
that would mean lower-power cell towers. So that’s the science so far behind this
argument. Now it is time for the FCC to come to some
sort of agreement with NOAA, NASA, Congress, and the rest of the
forecast-loving world. Thanks for watching this episode
of SciShow News. If you are interested in more SciShow content,
you might be interested in SciShow After Hours. We make a lot of SciShow videos here on SciShow,
but there are lots of topics that come across our desks that we decide aren’t
quite appropriate for this channel. They might not be family friendly enough for
our general content. But we think they’re fascinating and we
still wanna tell people about them! So we’ve decided to produce short, monthly
podcast episodes just for our patrons. Anyone pledging $4 or more on Patreon will
get to listen to one of our editors regularly astound, amaze, and appall me! I just recorded one actually, and I was made
fairly uncomfortable. There’s a sample episode up right now, which
you can check out for free at patreon.com/scishow. If you think it’s as fun as we do, join
our Patreon community for monthly episodes. [♪ OUTRO]