5G wireless may lead to inaccurate weather forecasts

Rutgers study is the first to model impact of 5G radiation “leakage” on forecasting


Research News


Upcoming 5G wireless networks that will provide faster cell phone service may lead to inaccurate weather forecasts, according to a Rutgers study on a controversial issue that has created anxiety among meteorologists.

“Our study – the first of its kind that quantifies the effect of 5G on weather prediction error – suggests that there is an impact on the accuracy of weather forecasts,” said senior author Narayan B. Mandayam, a Distinguished Professor at the Wireless Information Network Laboratory (WINLAB), who also chairs the Department of Electrical and Computer Engineering in the School of Engineering at Rutgers University-New Brunswick.

The peer-reviewed study was published this month at the 2020 IEEE 5G World Forum, sponsored by the Institute of Electrical and Electronics Engineers. Fifth-generation cellular wireless technology (5G) stems from new, smarter ways to use the higher (mmWave) frequencies for mobile communications. This technology will revolutionize internet communication and telecommunication. It has faster connection times, increases the number of devices that can connect to a network and will be more widely available over the next two to three years, according to IEEE.

The Rutgers study used computer modeling to examine the impact of 5G “leakage” – unintended radiation from a transmitter into an adjacent frequency band or channel – on forecasting the deadly 2008 Super Tuesday Tornado Outbreak in the South and Midwest.

The signals from the 5G frequency bands potentially could leak into the band used by weather sensors on satellites that measure the amount of water vapor in the atmosphere and affect weather forecasting and predictions. Meteorologists rely on satellites for the data needed to forecast weather.

Based on modeling, 5G leakage power of -15 to -20 decibel Watts (a decibel Watt is a unit of power that describes the strength of radio waves) affected the accuracy of forecasting of precipitation (by up to 0.9 millimeters) during the tornado outbreak and temperatures near ground level (by up to 2.34 degrees Fahrenheit).

“It can be argued that the magnitude of error found in our study is insignificant or significant, depending on whether you represent the 5G community or the meteorological community, respectively,” Mandayam said. “One of our takeaways is that if we want leakage to be at levels preferred by the 5G community, we need to work on more detailed models as well as antenna technology, dynamic reallocation of spectrum resources and improved weather forecasting algorithms that can take into account 5G leakage.”

The lead author is Mohammad Yousefvand, a Rutgers electrical engineering doctoral student. Co-authors include Professor Chung-Tse Michael Wu in the Department of Electrical and Computer Engineering, Professor Ruo-Qian (Roger) Wang in the Department of Civil and Environmental Engineering and Joseph Brodie, director of atmospheric research in the Rutgers Center for Ocean Observing Leadership.


From EurekAlert!

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
Ian Magness
September 27, 2020 6:08 am

You can see it coming: “5G network results in lower than actual temperature, wind speed, precipitation etc recordings – need to make adjustments.”

Rich Davis
Reply to  Ian Magness
September 27, 2020 7:19 am

Yes indeed. Fast forward to 2030…

CNN reported today that record high temperatures of 106F are predicted tomorrow in Minnesota, far above normal for late January.

In other news, scientists are baffled by mysterious deaths in New York City, particularly among the homeless, that mimic the effects of frostbite and exposure despite recent temperatures in the upper 90s. An unidentified pollutant, a white crystalline substance which fell in depths of up to three feet are suspected to be a contributing factor. CNN’s science reporter Leonardo DiCrapio explains that this is what should be expected after the Pence Administration has spent the past 6 years denying that CO2 impacts global temperatures. Pence falsely claimed that the substance is actually just snow.

David Brewer
Reply to  Rich Davis
September 27, 2020 9:59 am

Best post ever.

Mark Luhman
Reply to  Rich Davis
September 27, 2020 10:43 am

I missed out on the 105 F in April 1980 in Moorhead Minnesota, on but I was in Alanta at the time for training on a proof machine. I had left Moorhead in mid March snow on the ground. Back home to green grass.

Rich Davis
Reply to  Mark Luhman
September 27, 2020 11:49 am

According to our (properly-adjusted) records, you are probably referring to the spike up to 45F that occurred at that time. Of course that was an early warning of the coming climate emergency.

Jim Gorman
Reply to  Rich Davis
September 27, 2020 1:50 pm


Reply to  Rich Davis
September 27, 2020 7:18 pm

Since no-one knows what snow is, how can they be sure that this white stuff isn’t snow?

Reply to  RoHa
September 27, 2020 8:47 pm

Because it could be black swans whitening their feathers and this has got to stop. Not that black swans really matter of course.

Doug Huffman
Reply to  observa
September 28, 2020 3:28 am

Apropos; will Thanksgiving Day be cancelled this year?

For a thousand days Tom Turkey was sure that this is the best of all possible worlds as Farmer Jones fed him and watered him and adjusted the tanning lamps. Every day as the Sun rose Tom gave thanks. Then came Thanksgiving and Farmer Jones with the AXE.

Beware the Black Swan hidden camouflaged in the fractal complexity of reality. (After N. N. Taleb and Benoit Mandelbrot)

very old white guy
Reply to  Rich Davis
September 28, 2020 7:23 am

I like it Rich.

David A
Reply to  Ian Magness
September 27, 2020 4:32 pm

Or, if they clearly cause a very high error, they will adjust those reading down, but only a small fraction of the high error.

? If 5 g dies this, what dies 4 g do? A 2.3 degree error is 23 tenths. Dies 4 g produce errors if tenths of a degree?

Reply to  David A
September 27, 2020 7:02 pm

David, it is very unlikely the earlier generation technologies have any effect. The reason 5G is of concern is that the frequency band planned is 26-28 GHz, which is right where the weather satellites and radar operate (best frequency for water vapour detection). Once you drop below 3 GHz, the resultant wavelength is way too long and these early technologies are all below this frequency.

Reply to  Observer
September 28, 2020 1:10 am

5G actually has 3 ranges low band which is very close to 4G, mid band is 2.5-3.7 and only high band is 25–39 GHz even in the paper it explains that. The report ignores we have already seen all this before on the 5Ghz band with airport radar and the implementation of Dynamic Frequency Selection (DFS) into the specification.

If it really is a problem the actual experts not climate science retards will implement a similar situation on the carrier equipment where they detect the installations and avoid the frequencies or use time domain encoding to avoid interference.

September 27, 2020 6:23 am

The majority of cellphone weather apps are feeding people crap forecasts to begin with. How we will know the difference?

Reply to  2hotel9
September 27, 2020 8:07 am

Indeed. They’re wrong more than they’re right.

Reply to  Jeff Alberts
September 27, 2020 2:37 pm

I deal with several people, daily, on a recurring basis, who routinely say to me”This is not how my (insert name of retarded app) said the weather was going to be today?” And no matter how many times you waste your precious time ‘splainin’ it to them they still can’t figure it out. Yea, I know a bunch of people with two black eyes and they stumble about dazedly exclaiming”What happened to my eye?”.

Reply to  2hotel9
September 27, 2020 8:55 pm

That’s the beauty of all the weather apps silly. Climate change becomes more certain.

Reply to  observa
September 28, 2020 7:42 am

Yea, the greenunistas keep doing this kind of crap and then cry that no one believes them.

Reply to  2hotel9
September 27, 2020 8:10 am

Thank you.
I was just about to make that very same point.

Reply to  Mr.
September 27, 2020 2:41 pm

I am seriously tired of people staring at screens all the time and still not figuring anything, anything at all, out. If the thing can’t show you what the weather will do in a 12 hours timeframe why keep staring at it?

Doug Huffman
Reply to  2hotel9
September 28, 2020 3:34 am

Again referring to N. N. Taleb; prognosticators have no doxastic commitment to truth and lie for a living, confident of no repercussions, confident of no Damocles’ Sword of Truth.

My weather-widget reports conditions in the Joseph?? box at the airport 2 miles away.

Reply to  Doug Huffman
September 28, 2020 7:55 am

I am a Accuweather subscriber, early morning(5:30 or so) I start the Northeast Snow Ice Rain radar and switch over to Enhanced Infrared, back&forth, as I am reading news(yes, I don’t watch tv “news”), in 10 minutes I know what is going to happen through the day. By 7:00 I am informed on what needs knowing and ready to move on. No cell phone needed. AW gets data from monitors stations less than 20 miles from me, east and west along I 80, way more accurate than KDKA, WTAE or WPXI out of Pittsburgh.

David Guy-Johnson
September 27, 2020 6:49 am

You’ll never notice the difference

Reply to  David Guy-Johnson
September 27, 2020 6:59 am

I won’t, I get my weather forecasting from actual weather forecasters, not some H1B visa overstay working in a cubicle in some office building with no windows. And I don’t pay any attention to the latest pregnant staffers reading a teleprompter at our local TV “news” outlet, they are far beyond clueless.

September 27, 2020 6:51 am

So how did they validate their model? I did read the paper, they did a good job on pointing out the myriad of variables at work as well as the possibility for potential effects, but that’s about it.
The main issue seems to be the number of variables as well as the density of 5G networks in urban as opposed to rural areas, to say nothing of the types of antennae other equipment used.

Kevin kilty
September 27, 2020 7:10 am

This is a lot like the dispute over 5G using some of the 1.5GHz band interfering with GPS — the problem is nil or horrid depending on which community one represents. And dBW is a pretty large unit of leakage.

Curious George
Reply to  Kevin kilty
September 27, 2020 10:43 am

Minus 15 dBW is 30 mW.

Stephen Wilkus
Reply to  Curious George
October 9, 2020 8:21 pm

10 Log10(Watts)=10Log10(10^3 mWatts)=30+10Log10(mWatts)

September 27, 2020 7:13 am

If satellite data for weather forecasts will become affected and unreliable, what about all the other services that depend upon satellite data such as GPS?

Carbon Bigfoot
Reply to  Peter Wilson
September 27, 2020 8:53 am

Pete: This far more important than you GPS:

Reply to  Carbon Bigfoot
September 27, 2020 1:15 pm

I’m not saying it is, but it could be more dangerous and there are other satellite data uses in other services.

Dave Dodd
Reply to  Carbon Bigfoot
September 27, 2020 8:09 pm

“RF radiation has been proven harmful for humans…”

NOPE! Outside of frying seagulls that flew through our ship’s RADAR, there is NO INDICATION that NON-IONIZING RF causes harm to ANY living creature, not even developing embryos in chicken eggs, subjected to strong RF fields!

Reply to  Carbon Bigfoot
September 28, 2020 5:46 am

Suggest you read this article by Kenneth Foster Professor of Bioengineering at the University of Pennsylvania

Also check out the references at http://emfandhealth.com/

Reply to  Carbon Bigfoot
September 30, 2020 11:40 am

Well that was a well written wake up call. The more advanced we get the further away from real caring about humanity. The robots are coming and the owners of the said Bots have no idea that they are next to leave

Steve Richards
September 27, 2020 8:09 am

I am sure that the international specification of 5G transmissions are acceptable to satellite operators.

September 27, 2020 8:21 am

Will this added signal noise make a difference in the ability to predict who is going to experience a PSPS’s?

They, PSPS’s, are getting OLD- number 7 is going to be called in my area later today.

September 27, 2020 8:22 am

“It can be argued that the magnitude of error found in our study is insignificant or significant, depending on whether you represent the 5G community or the meteorological community, respectively,” Mandayam said.
Wow, check our math then you must be a denier! The science is settled, even though we just made it up, off to the gallows!

/Sarc, for the benefit of the climate-credulous, these are not marching orders for the next riot.

Don K
September 27, 2020 8:22 am

This is not something I know (or care) much about, but my impression is that the range of 5G signals is embarrassingly short because the frequencies used don’t penetrate walls, leaves, rain, … anything. — typical range 500m maybe. On a good day. My GUESS is that considerable effort will be expended to focus the power into areas where potential users are, not wasting it on migrating geese and other overhead objects. Also, its limited range may confine its use to a few densely populated areas where its high bandwidth actually makes sense. So maybe there won’t be a problem.

Reply to  Don K
September 27, 2020 12:15 pm

I deliver for Doordash and Uber Eats, and just got a new 5G phone. I’m surprised by the amount of 5G coverage I’ve got in my relatively rural area. Even at my house I can get a half decent signal. Not indoors, though, at least not at my house. I think the nearest tower is too far away for that.

Reply to  Jeff Alberts
September 27, 2020 2:34 pm

According to the paper, 5G has a range of 200m. Range extenders will be a big business.
It’s just the old problem of marketing vs reality.

Reply to  Jeff Alberts
September 29, 2020 11:13 am

Jeff, it’s likely your 5G phone is using the sub-6GHz cellular bands since you are in a “relatively rural area”. It is highly unlikely mm-wave frequency bands will be deployed out there because the ROI isn’t there. MM-wave band cells will be deployed in the cities and thickly settled suburban areas first (low-hanging fruit, i.e. best return on investment).

The mm-wave cell sites have between 200 and 2000 foot coverage, meaning the between 5 and 20 times the number of mm-wave sites will be needed to provide similar coverage to that of the sub-6GHz cell sites (4G/5G).

Reply to  Don K
September 27, 2020 3:01 pm

Oh, there will be a problem, the leftists always find problems.

Jim Whelan
September 27, 2020 9:20 am

WOW! “Peer reviewed!” That and $5 won’t even get you your favorite elitist drink at Starbucks.

September 27, 2020 9:20 am

An old dodgy CRT TV set will bump-off 5G super-fast broadband in a 1/4 mile radius, a faulty microwave oven might be just as good. Just doing some scientific research in the old technology, don’t ya know. 🙂

Boff Doff
September 27, 2020 10:00 am

This “computer modeling”; is it in any way related to the “climate modeling” or “epidemiological modeling” that has proven so helpful in dealing with the political challenges posed by the pesky selfishness of the population of planet earth?
It surely is!
Thank goodness Academia is on the case. After all, who else can we trust?

Reply to  Boff Doff
September 27, 2020 1:59 pm

Does that mean that climate models will become even less accurate?

Is that even possible ?

September 27, 2020 10:01 am

Wouldn’t this be much the same argument that smart meters caused health issues because it used Wifi to transmit wireless meter data through bouncing off all other smart meters back to the main office? Meanwhile, the people complaining about that have a wifi router sitting next to their lazy boy, while they talk on a cell phone next to their brain? At the scale they are talking about for the entire atmosphere, the intensity of radio waves over distance obeys the inverse-square law, which states that intensity is inversely proportional to the square of the distance from a source. Think of it this way: double the distance, and you get four times less power. How is terrestrial 5G at the surface on a fraction of the planets surface going to interfere with the entire atmospheric reading from space?

I don’t know enough about this specific claim to make an informed decision so I hope some real meteorologists will set us straight on this issue. It just seems the source terrestrial 5G signal is so weak at the 24 GHz band when water vapor in the atmosphere emits faint radiation at a frequency of 23.8 GHz. Is that faint radio spectrum on the planet surface in a few percent of the land mass going to make any difference at planetary scale of the entire global atmosphere? Now I am curious as to the real truth of the matter. If it bolsters the claims that it makes it appear cooler/drier in only populated areas, then I could see some scientists latching onto this to make further adjustments if it embellishes the CAGW narrative.

Reply to  Earthling2
September 27, 2020 2:11 pm

I think you’ve hit the nail there. “…when water vapor in the atmosphere emits faint radiation at a frequency of 23.8 GHz.” If they are trying to measure the faint radiation from water vapour to make predictions, then a slight interference from 5G leakage could cause them huge problems.

Reply to  Herbie
September 29, 2020 12:13 pm

I can see another issue, that being any signals that may make it from a 5G cell site would likely be a sidelobe from the antenna array and would be down an additional 20 to 50dB from the main lobe. The main lobe would usually be aimed at the horizon or below. If my understanding is correct, most of the cell transmitters would be running at ~10W aggregate power, but all of that power is split among the various beams servicing a cell phone on each face of the antenna array. (It must be remembered the antennas used in the mm-wave region are phased-arrays that can steer a beam in two dimensions – azimuth and elevation. And they can generate multiple different beams from the same array simultaneously.) So any sidelobe ‘facing’ up into the sky could be as low as 1/100,000th the power of the main lobe. Would that be enough cause the effect being discussed in the paper?

Reply to  DCE
September 29, 2020 9:47 pm

Thats my point, as you explain technically that the 5G signal is directional towards the horizon at a very low power level, at a very high frequency of 24 GHz in this competing case with the clouds emitting at 23.8 GHz. Considering that the 5G cell signal range on the ground where the antenna is being aimed is less than a 1/4 mile, I don’t see this as being able to be an issue. It would be like saying I could pick up a wifi modem connection from 100 miles away, when you can barely see your neighbours signal next door. And at a much lower frequency which would carry further.

If the satellite making the observations had a 300 foot receiving dish aiming at one specific spot in a city, that may be one thing, but it is looking at thousands of square miles as its snapshot of the planet/cloud surface. So I doubt this can have any effect on the 23.8 GHz cloud/satellite data at all. Plus given that 5G won’t even cover a fraction of a percent of the planets surface, it would have no effect on 70% of the earths surface which is ocean, and another 20% that is remote Antartica, Greenland, Siberia and all the deserts in the world. Putting on my skeptical hat, this myth has to be a busted.

Shoki Kaneda
September 27, 2020 10:50 am

Even allowing that 5G may cause inaccuracy in weather forecasting; how could you tell?

john cooknell
Reply to  Shoki Kaneda
September 27, 2020 1:26 pm

My thoughts exactly, how would we know?

September 27, 2020 3:13 pm

This is actually a piece from The Onion isn’t it?

Next, we can expect a Babylon Bee article declaring that the problem was solved when they found out that 20% of the formerly male staff now identify as female.

September 27, 2020 4:44 pm

“The water molecule, in the gaseous state, has three types of transition that can give rise to absorption of electromagnetic radiation:”
Man uses microwave frequencies to light up the water molecule (Microwaves are a form of electromagnetic radiation with wavelengths ranging from about one meter to one millimeter; with frequencies between 300 MHz (1 m) and 300 GHz (1 mm). )
In the 70’s they used this radar transmitter to transfer wireless electrical power over a mile and blow the inverse square law out the window .

Gary Edward Miller
Reply to  jmorpuss
September 27, 2020 5:51 pm

Perhaps a few words are in order from this retired antenna engineer. Sorry, but the inverse square law was not blown out of the water in the NASA demonstration. The efficiency that was claimed dealt with the conversion of the incident RF power on the receive antenna array into DC. You may be referring to the fact that the receive antenna was not in the far field of the 26 m diameter transmit antenna and hence the radiated power did not drop off as 1 over r-squared, but this is a well understood phenomenon that is used in every day applications.

Reply to  Gary Edward Miller
September 28, 2020 2:38 pm

Debunking the Inverse Square Law
Forget everything you’ve learned about the Inverse Square Law – it does NOT apply to focused lighting!

Reply to  jmorpuss
September 28, 2020 3:49 pm

So, this is an ad for a photographic lighting system. Okey dokey, then.

Reply to  2hotel9
September 29, 2020 2:50 pm

Light is a small part of the electromagnetic spectrum, you can focus all frequencies in one spot the same as you can light.
“A parabolic antenna is an antenna that uses a parabolic reflector, a curved surface with the cross-sectional shape of a parabola, to direct the radio waves. The most common form is shaped like a dish and is popularly called a dish antenna or parabolic dish. The main advantage of a parabolic antenna is that it has high directivity. It functions similarly to a searchlight or flashlight reflector to direct the radio waves in a narrow beam,”

Flight Level
September 27, 2020 4:46 pm

Hmmmm. Now I understand why they urgently updated the firmware of all chemtrail control units of our fleet.
(*implicit palmface*)

Al Miller
September 27, 2020 6:51 pm

What you mean here on the wet coast they won’t be able to accurately tell us if it will rain tomorrow?? The shock, it will be just like now…

Gary Edward Miller
September 27, 2020 7:00 pm

Perhaps a few words are in order from this retired antenna engineer. Sorry, but the inverse square law was not blown out of the water in the NASA demonstration. The efficiency that was claimed dealt with the conversion of the incident RF power on the receive antenna array into DC. You may be referring to the fact that the receive antenna was not in the far field of the 26 m diameter transmit antenna and hence the radiated power did not drop off as 1 over r-squared, but this is a well understood phenomenon that is used in every day applications.

James Clarke
September 27, 2020 7:10 pm

Computer modelling, the great research tool of the 21st Century. What once took years of study to screw up can now be screwed up far faster with a computer model and dodgy data.

September 27, 2020 7:22 pm

I don’t know what these Gs are, but I am reliably informed that 5G (whatever it is) will not only cause brain cancer but also COVID 19. Now I find it will cause global warming or something. Is there any disaster it won’t cause?

Reply to  RoHa
September 29, 2020 12:17 pm

You forgot to add ED to the list of problems caused by 5G.

September 27, 2020 11:09 pm

I guess 5G is a trial and will get an upgrade in the future, let’s see

Reply to  jojo
September 28, 2020 1:11 am

Close it’s climate science retards playing in fields they know little about.

Mariano Marini
September 28, 2020 12:27 am

Sorry for my naive question. What about microwave and ice melt or temperature sensors? I know that 700W is needed to warm food in minutes. There are some study about this? How many Watts comes from all the satellites that round the earth? (Even my English is quite naive). Thank you.

Reply to  Mariano Marini
September 28, 2020 2:21 am

A microwave is a close cavity resonator at a specific resonant frequency of water the other is a free radiant emission. The only way we could probably create an analogue and answer your question is to wrap you up in alfoil designed to resonate at all the right frequencies and put all the satellites up your butt and measure what happens. As silly as that answer is it is the exact truth because emission power has no relationship to heating you require absorption to get heating.

It’s the same problem as a bird sitting on a power line with thousands of volts and power flowing under it’s feet. If you want the exact analog look at anyone sitting or touching a Tesla Coil they can have massive emissions passing harmlessly over there body in skin effect.

Mariano Marini
Reply to  LdB
September 28, 2020 6:43 am

Let see if I well understand. If I use a magnetron on plain air there will be no warm to my hand?

Reply to  Mariano Marini
September 28, 2020 4:48 pm

Aside from the magnetron gets actually hot from losses you also have water in your hand that will resonate at that magnetron frequency AKA absorption … So your example is ill formed and posed relative to the satellite question you asked.

Now imagine an ideal magnetron with no losses and change the frequency slightly away from from 2.45 GHz and hang onto it and what happens …. answer nothing.

It’s not rocket science even a layman should get it, you can’t answer you question without specifics about the emission and the possible absorption target.

Mariano Marini
Reply to  LdB
September 29, 2020 12:07 am

Ok. Thanks for your kindly explanations. Verily my naive question rose because I don’t know the exact frequencies used by satellite communications. I looked at Internet and found always that they use “microwave”. Thank a lot.

David Stone CEng
September 28, 2020 3:46 am

This paper is undoubtedly one of those “Modelling” ones that we especially like /sarc.
If the satellite sensors use the same frequency band as 5g then potentially there ccould be some interference. However they should not be using the same frequencies, so there should not be a problem.

A major problem with 5g and millimeter frequencies in general is that water increases the propagation attenuation very significantly, and not at specific frequencies, but broadband due to its conductivity. This could be clouds, high humidity etc, and so 5g cells are quite small so that they still work when it rains.

The only way that the 5g network could change the weather (rather than the measurements) is if the power used were sufficient to heat the bulk of the water vapour significantly. This is much more than 5g uses, and at the required level it would also warm people which generally considered not to be a good idea. The mechanism is somewhat different to a microwave oven, but similar volumetric power levels would be required.

Reply to  David Stone CEng
September 29, 2020 12:21 pm

The absorption of mm-wave radiation isn’t ‘smooth’ and does have dips and peaks at various wavelengths, though the the dips and peaks are usually on the order a few dB. In the end those dips and peaks won’t have a major effect.

September 28, 2020 5:27 am

dbm – is decibels using a milliwatt scale. That is a very common way of stating power levels.

September 28, 2020 6:38 am

The problem with this “research” is the implicit assumption that weather forecasts are accurate to start with.

Not Chicken Little
September 28, 2020 8:40 am

5G will lead to inaccurate weather forecasts? How will they ever know, how will they be able to tell?

Right now in my neck of the woods, fairly often they can’t even get the weather right just 4 hours into the future.

And what’s with the 50% chance of rain??? I just want to know, will it rain, or won’t it? They can’t even tell me that! But their supposedly better-educated, more sophisticated “brothers”, the “climate scientists”, assure us of what’s going to happen 12, 20, 50, 100 years in the future. None of their predictions have come true yet, but…just you wait!

September 28, 2020 10:36 pm

I would probably be more concerned if current forecasts were more accurate. They aren’t bad but I’m not sure they’ve improved despite supercomputers and much more Science in meteorology.

%d bloggers like this:
Verified by MonsterInsights