Is current signal technology safe?

[whole diary]

[grahame]

From the BBC^» (British Broadcasting Corporation - home page)

Rare solar storms could cause chaos on the UK^▸ (United Kingdom)'s railways by triggering "powerful magnetic disturbances", scientists have warned.

A study at Lancaster University found the celestial events could disrupt the network's electrical systems and cause signals to switch from red to green.

Researcher Cameron Patterson said the "relatively rare" storms needed to be "taken seriously" by the rail industry.

However, Network Rail said it considered the risk to be "very small".

Do we get any warning of solar storms and if so could rail traffic be stopped when there's a significant risk?

Should older technology be kept as a backup system (any excuse for nostalgic pictures!)

[stuving]

I'm not sure "current signal technology" is appropriate here - unless it's a pun on "current". The specific risk that Cameron Patterson is studying for his PhD is to DC^▸ (Direct Current) track circuits, due to currents flowing in the ground underneath. So old technology, and disappearing all the time as axle counters are now preferred. This small risk from geomagnetic disturbances may well count as an argument to hasten their replacement.

[grahame]

I'm not sure "current signal technology" is appropriate here - unless it's a pun on "current". ...

It isn't a pun - this seems to be a case of me reading an article and taking it at face values when I should not have done.  Thank you for correct / educating me.

[broadgage]

Current rail signalling is safe under all ordinary circumstances, it has a good track record.
During a severe solar storm it is potentially unsafe, due to currents induced in long conductors such as rail lines, and signalling wiring.
Only a very limited train service could be safely run under such circumstances. Branches could be operated with only one engine in steam. Longer routes reverting to manual time interval working.

Probably no electric trains as not likely to be any electricity.
Diesel trains very limited by the amount of fuel already in stock at depots, I doubt that re-supply would be available.
Steam is a possibility, but coal and water supply would be the limiting factor.

Rail transport would likely be restricted to food, fuel, and other essentials, maybe one passenger train a day on selected routes.

There might be a days warning of a severe solar storm, but this is reliant on a single satellite that is already past its design life.
Even given a little warning I have little faith in HMG doing anything effective to prepare.

[Electric train]

I am less concerned about the UK^▸ (United Kingdom) railway signalling system risk from a solar storm, my concern would be the National Grid, DNO^▸ (Distribution Network Operator) 132, 33, 11 kV networks, telecom networks G4, G5 broadband etc, data centres.

You also have to consider road transport, cars, vans and trucks are full of electronics which could be venerable.

[broadgage]

I am less concerned about the UK^▸ (United Kingdom) railway signalling system risk from a solar storm, my concern would be the National Grid, DNO^▸ (Distribution Network Operator) 132, 33, 11 kV networks, telecom networks G4, G5 broadband etc, data centres.

You also have to consider road transport, cars, vans and trucks are full of electronics which could be venerable.

Agree that electrical and communications infrastructure is very vulnerable, road vehicles should be ok as they are not connected to any long conductors.
A severe solar storm induces undesired  current on long conductors such as railway lines, electrical infrastructure, pipelines, and even long wire fences.

[stuving]

This is one of those risks where most comments assume no-one knows much or is doing anything about it. In fact the issues are well-known to people who know about them, which includes the engineers most concerned with protecting the infrastructure. And we do have experience and observations for 200 years, even if for most of that period we only have simple magnetic field measurements. But even in space we have over 50 years of instrumented platforms.

The one scary thing is that we do not have a good grasp of the severity of the risk - we can't quantify the likelihood (per year) of a very severe storm, nor the severity of the likely storms, well enough. That's because to estimate at the 1% per year level (i.e. once per century) to anything better than a factor of two accuracy, you need 1000 years of observations and unchanging statistics. The same limitation affects climate-related risks like floods, where we know the statistics are changing. For the sun, we don't know that either way.

But we do, as I said, have numbers, and a lot more observations than the ones usually pulled out of the cuttings library. The biggest magnetic disturbances (measured in the customary way as Dst in nT) are:

1859:     900           
1909:     595
1921:     907
1989:     565 

So the Carrington event wasn't really a huge outlier at all.

[Oxonhutch]

(measured in the customary way as Dst in nT)

I assume (always risky!) the nT are nanoteslas (magnetic flux density) that I am familiar with, but what is 'Dst' ?

[stuving]

(measured in the customary way as Dst in nT)

I assume (always risky!) the nT are nanoteslas (magnetic flux density) that I am familiar with, but what is 'Dst' ?

The peak field is in nT, though I left out the fact that it is negative (the usual field weakens). Dst is, as I said, the customary way of quantifying these events. I think it stands for disturbance-storm-time index, which doesn't help a lot. Or try: "Dst is often interpreted in terms of the westward-directed magnetospheric ring current."