Arctic Investigation Underway Into Solar Storm Sat-Nav Disruption

another random user writes "Scientists in the Arctic have launched an urgent investigation into how solar storms can disrupt sat-nav. Studies have revealed how space weather can cut the accuracy of GPS by tens of metres. Flares from the Sun interact with the upper atmosphere and can distort the signals from global positioning satellites. The project is under way at a remote observatory on a windswept mountainside in the Norwegian archipelago of Svalbard in the High Arctic. The site was chosen for its isolation from electronic pollution and for its position in relation to the Earth's magnetic field which flows from space down towards the far North."

Comments

"Tens of metres"

[Baloroth]

"Tens of metres" is not exactly very precise, and it makes rather a large difference in precision if this is 20 metres or 99 metres: the first is annoying, the second might severely impair your ability to navigate, although I'd question that a bit. I mean, line of sight usually works, and in storms when it doesn't you really shouldn't be navigating close enough to the ground or a potential collision that even 100 metres off would be a dangerous problem, so am I missing something or is not that big an issue? Annoying, yes, and I can see the issue in S&R (gets a lot colder than the -20 mentioned in TFA that close to the poles, hell it gets colder than that here sometimes), but is there any highly important usage case where it would be an extremely detrimental problem?

I'm also a bit curious why they don't just use DGPS anyways, since that exists and it seems like it would solve the problem quite nicely. Added bonus that it helps even when there isn't a solar storm, and it's even more accurate than regular GPS.

Re:"Tens of metres"

[Charliemopps]

I think that's what they're trying to figure out.

Re:"Tens of metres"

[BitterOak]

it's the difference between landing on the runway or plowing through someone's house if you're landing on instruments.

So yeah, fairly detrimental for the five hundred passengers.

I think planes mostly still use ILS rather than GPS-only for instrument landings. ILS shouldn't be affected in this way by space weather.

Re:"Tens of metres"

[ravenlord_hun]

Some modern GPWS systems actually use digitalized maps of the area, determining possible collisions not only by radar but by "looking" around the map using the GPS coordinates. Pretty sure it'd cause problems there, causing false alerts and not warning other times...

Re:"Tens of metres"

[vokyvsd]

I'm a surveyor, and I use GPS to locate points to within a few hundredths of a foot (a couple centimeters, if you will). So, I don't know if my interpretation is exactly what the article intended, but I saw "tens of meters" and immediately thought "really really bad" and didn't even consider the possibility that the range of variation in "tens of meters" would be significant...

It's interesting how our minds immediately write things off like that... In most other circumstances I think I would have went exactly where you did and asked about the precision.

Something like... if you or I heard that it would cost "several billion dollars" to buy out a particular company, we'd just go "whoa, that's a lot"... but there's a select subset of people who would perk up their ears and say "umm, how much is 'several'?"

Re:"Tens of metres"

[Tastecicles]

I didn't think of that. Of course, you'd be right - GPS isn't a replacement for ILS, it's an augmentation if anything. ILS would still be the primary signal base for instrument landing, since it uses a terminal narrowband beacon which is a: static and b: situated at the end of the runway and pointing directly outward into the glide slope. GPS would be used in situations where vision is obscured, the surrounding terrain is less than ideal (big mountains in the way for instance, or open ocean either end and to the sides of the tarmac - such as Kansai in Japan which is an amazing piece of engineering and which does in fact put up with some of the worst weather on the planet - ferocious crosswinds and downpours like you wouldn't believe, yet it just carries on functioning with nary a break) and you need a little extra help to avoid touching down on the wrong bit of terrain. Accurate GPS will tell you if you're over water, an aviator unit tells you your altitude as well I believe, in AMSL and ATL (I have seen them in sailplane cockpits), all ILS is really, when you boil it down, is an autopilot landing system based on a beacon.

Re:"Tens of metres"

[edibobb]

Differential GPS can reduce that error of "tens of meters" to a few inches, even in a solar storm. A ground station at known location calculates the error and transmits it to the mobile GPS receiver, which adds the error into its location calculations.

Re:"Tens of metres"

[ColdWetDog]

The new system relies on GPS ADS-B

Very cool in the Alaska brush. If GPS went out it would be like tossing Alaska general aviation back 50 years when they crashed into mountains a lot. Now we only crash into mountains on rare occasions.

Re:"Tens of metres"

[riverat1]

The company I worked for helped develop ADS-B and was heavily involved in the Capstone Program. ADS-B essentially forms a network between all airplanes and ground stations equipped to send and receive the signal. An airplane periodically (like every few seconds) broadcasts its location and vector so anyone who can receive the signal can tell where the other plane is and where it's heading in relation to it. The GPS that's a part of it also had terrain maps and will warn you if you're headed for a mountain. I'm pretty proud of the work we did on ADS-B. It's improved the safety of flying small airplanes in Alaska immensely and it's coming to the lower 48 States gradually.

Re:"Tens of metres"

[Anonymous Coward]

In theory it can, but in practice, we can't calculate the correction fast enough during serious storms. The CPOS system normally provides real-time cm-precision positioning in Norway, but it breaks down during some storms. Improving this is a subject of research. See this article for details: http://www.swsc-journal.org/articles/swsc/pdf/2012/01/swsc120026.pdf

Re:"Tens of metres"

[Eyeball97]

I hate to break it to you, but aircraft don't use GPS for landing.

It's the difference between your car telling you to turn left "here" when it actually means the junction 45m up the road. Which admittedly could be an even bigger problem for the special kind of twit who already drives into walls "because the satnav told me to".

Re:"Tens of metres"

[jasnw]

I've been working on ionospheric impacts on GPS damn near since GPS was launched. Comments:

1. Ten meters is indeed not a hell of a lot until you consider things like (a) your average airport runway is likely not much more than 20 meters in width, particularly in areas where there probably isn't other forms of landing assistance and GPS is needed, and (b) while ten meters horizontally can be OK, ten meters vertically can be brutally bad if you're trying to land an airplane in bad weather and are depending on GPS to tell you where you are. You either fly into the ground or "flare" for landing 10 meters too high. Not good in either case.

2. DGPS can help, but only if the ionospheric disruption is such that it throws the DGPS base station off the same amount (and direction) as the users GPS. This is not necessarily a good assumption to make, and is still being researched. The ionospheric scintillation discussed in TFA is a problem for DGPS.

3. This is mainly a problem for non-static GPS uses. Surveryors typically take a long-term GPS position measurement (at minimum several minutes) which will, if done correctly, smooth through the big errors caused by ionospheric disruptions. The main problem is dynamic uses of GPS, like on aircraft, where you can't linger around to integrate up a good position solution.

Bottom line is that GPS isn't quite the "all weather system" it was cracked up to be. Lots of time and effort has gone into trying to resolve ionospheric impacts on GPS over the past several decades. Again, typically not a problem for lots of applications, but a potentially serious problem for dynamic position/velocity GPS users.

Comment removed

[account_deleted]

Comment removed based on user account deletion