Tuesday, February 7, 2017

Coronal Mass Ejections (again)

Back in 2014 I blogged about one of digital preservation's less well-known risks, coronal mass ejections (CME).  Additional information accumulated in the comments. Last October:
"President Barack Obama .. issued an Executive Order that defines what the nation’s response should be to a catastrophic space weather event that takes out large portions of the electrical power grid, resulting in cascading failures that would affect key services such as water supply, healthcare, and transportation.
Two recent studies bought the risk back into focus and convinced me that my 2014 post was too optimistic. Below the fold, more gloom and doom.

Mark Gilbert's How Space Could Trigger a Future Economic Crisis reports on a new paper in Space Weather:
In four scenarios envisaging the economic impact of a solar storm, the mildest triggers a daily loss to the U.S. economy of $6.2 billion, or 15 percent of daily output; the worst case sees a cost of $41.5 billion, wiping out every dollar the world’s largest economy generates each day.
and:
A study published last month by the Cambridge Centre for Risk Studies estimates that a solar storm would have the potential to wipe between $140 billion to $613 billion off the global economy in a five-year time span, depending on the severity of the impact.
According to a NASA blog post, the probability is 12% per decade:
In February 2014, physicist Pete Riley of Predictive Science Inc. published a paper in Space Weather entitled "On the probability of occurrence of extreme space weather events." In it, he analyzed records of solar storms going back 50+ years. By extrapolating the frequency of ordinary storms to the extreme, he calculated the odds that a Carrington-class storm would hit Earth in the next ten years.

The answer: 12%.
Macroeconomic impact
So there is about an 1-in-8 chance that in the next decade we will face one of the Cambridge scenarios. They divide the economic impact of the severe scenario's CME impacting the US into four areas:
  • Direct Impacts. It takes 5 months to restore power to 95% of the US population.
  • Indirect Supply Chain Impacts. The impact of power outages on international supply chains is bigger than their direct impact.
  • Macroeconomic Impacts. There is a large initial hit to US domestic product, but a fairly rapid recovery as government spends on recovery.
  • Insurance Impacts. For various reasons, insurance companies bear only about 14% of the economic loss, but this still amounts to about 4 time the total catastrophe losses they bear in a normal year.
I've always said that the chief threat to digital preservation is economic; digital information being very vulnerable to interruptions in the money supply. In the context of economic losses of the magnitude envisaged by the Cambridge report, digital preservation systems would be very low on the priority list for recovery funds.

The risk of CME's is one reason Facebook has advanced for their investment in optical storage for cold data. A CME could destroy the electronics in the racks, but it would not destroy the data on the DVDs. Actually, a CME is equally unlikely to destroy the data on hard disk platters, but destroying the drive electronics makes that data very expensive to recover.

16 comments:

David. said...

Annalee Newitz at Ars Technica reports that analysis of ceramic jar handles from the kingdom of Judah reveals large, rapid variations in the Earth's magnetic field:

"Sometime late in the 8th century BCE, there was a rapid fluctuation in the field's intensity over a period of about 30 years—first the intensity increased to over 20 percent of baseline, then plunged to 27 percent under baseline. ... A fluctuation of such intensity could leave the planet far less protected from solar storms that overload electrical grids, destroying transformers and causing widespread blackouts."

David. said...

In the light of recent saber-rattling, Ian Thomson at The Register reports on the possible effects of a North Korean EMP attack on the US:

"Testimony to the US Congressional EMP Commission stated that in the event of a massive EMP attack on the US using multiple high-yield warheads, around 90 per cent of the American population would be dead after 18 months due to famine, disease, and societal breakdown."

That is beyond the North's capabilities, but:

"Baker said that a low-yield device such as that thought to be owned by North Korea, detonated at optimum height, would generate EMP over an area with a diameter of around 1,000 miles. He understandably declined to specify the optimum height, but it's thought to be around 50 to 80 miles up.

If such a device were to be detonated over the most densely populated part of the US, namely the North East, the consequences for power grids, computing centers and telecommunications systems could be catastrophic, he explained."

David. said...

See this comment on my earlier post for the likely effects of EMPs and CMEs on stored data and the hardware needed to read it.

David. said...

Annalee Newitz reports that between Blade Runner and Blade Runner 2049 an:

"EMP detonation. ... not only caused a global blackout, but also erased a bunch of history that had been chronicled in electronic archives. Maybe it erased the memories of replicants, too? Either way, the future of 2049 is missing a huge part of its history and is struggling to rebuild both from a massive global famine and an information black hole."

Am I zeitgeist-compatible, or what?

David. said...

There's a much cheaper way to shut down a nation's power grid than a nuclear weapon. Dan Goodin's Hackers lie in wait after penetrating US and Europe power grid networks explains that at least two groups appear to have penetrated the networks controlling the grids and, in effect, have installed kill switches.

David. said...

A strong (G3-class) geomagnetic storm is forecast for today and tomorrow, caused by a CME.

David. said...

Puerto Rico is a foreshadow of what an EMP or CME event in the US would look like, says Llewellyn King at oilprice.com

David. said...

Measurement of magnetic field and relativistic electrons along a solar flare current sheet by Bin Chen et al describes research potentially leading to a way to predict CMEs. This might provide a couple of days warning of a Carrington-class event.

Tip of the hat to Anton Petrov's explanation on YouTube.

David. said...

At Daily Kos, skralyx's Welcome to Solar Cycle 25! The coronal mass ejection watch is on... is an excellent overview of the increasing risk of a Carrington-level event as the next solar cycle starts.

David. said...

Lily Hay Newman's A bad solar storm could cause an “Internet apocalypse” reports on Solar Superstorms: Planning for an Internet Apocalypse by Sangeetha Abdu Jyothi of UC Irvine:

"Abdu Jyothi's research points out an additional nuance to a blackout-causing solar storm: the scenario where even if power returns in hours or days, mass Internet outages persist.
...
Abdu Jyothi found that local and regional Internet infrastructure would be at low risk of damage even in a massive solar storm, because optical fiber itself isn't affected by geomagnetically induced currents. Short cable spans are also grounded very regularly. But for long undersea cables that connect continents, the risks are much greater. A solar storm that disrupted a number of these cables around the world could cause a massive loss of connectivity by cutting countries off at the source, even while leaving local infrastructure intact."

David. said...

Analysis of tree rings revealed a huge CME in 775CE called the Miyake event. Similar evidence for the Carrington event is missing, suggesting that the Miyake event was 1-2 orders of magnitude greater, i.e. a solar superflare. Now, similar analysis in Tree rings reveal two strong solar proton events in 7176 and 5259 BCE has detected two more Miyake-scale events. CMEs on this scale would devastate our current civilization, so finding three in the last 10K years with potentially more to come is worrying.

Hat tip to Anton Petrov.

David. said...

Matt Rigel's Here Comes the Sun—to End Civilization discusses Scott McIntosh's model of coronal mass ejections:

"McIntosh’s attempt to do that goes back to 2002, when, at the behest of a postdoctoral mentor, he began plotting tiny ultraviolet concentrations on the solar surface, known as brightpoints. “I think my boss knew what I would find if I let a full cycle pass,” he recalls. “By 2011, I was like, holy fuck.” He found that brightpoints originate at higher latitudes than sunspots do but follow the same path to the equator. To him, this implied that sunspots and brightpoints are twin effects of the same underlying phenomenon, one not found in astrophysics textbooks.

His grand unified theory, developed over a decade, goes something like this: Every 11 years, when the sun’s polarity flips, a magnetic band forms near each pole, wrapped around the circumference of the star. These bands exist for a couple of decades, slowly migrating toward the equator, where they meet in mutual destruction. At any given time, there are usually two oppositely charged bands in each hemisphere. They counteract each other, which promotes relative calm at the surface. But magnetic bands don’t all live to be the same age. Some reach what McIntosh calls “the terminator” with unusual speed. When this happens, the younger bands are left alone for a few years, without the moderating influence of the older bands, and they have a chance to raise hell."

David. said...

@sovietologist has an interesting thread on the EMP effects of tactical nuclear weapons:

"TL;DR-if you're close enough to a near-surface nuclear burst to worry about its EMP, you're probably going to be killed by something else--but there are exceptions"

David. said...

Is the ‘internet apocalypse’ nigh? Breaking down the solar-storm science by Kelsey Ables is based on Solar Superstorms: Planning for an Internet Apocalypse by Sangeetha Abdu Jyothi. Her abstract reads:

"Black swan events are hard-to-predict rare events that can significantly alter the course of our lives. The Internet has played a key role in helping us deal with the coronavirus pandemic, a recent black swan event. However, Internet researchers and operators are mostly blind to another black swan event that poses a direct threat to Internet infrastructure. In this paper, we investigate the impact of solar superstorms that can potentially cause large-scale Internet outages covering the entire globe and lasting several months. We discuss the challenges posed by such activity and currently available mitigation techniques. Using real-world datasets, we analyze the robustness of the current Internet infrastructure and show that submarine cables are at greater risk of failure compared to land cables. Moreover, the US has a higher risk for disconnection compared to Asia. Finally, we lay out steps for improving the Internet’s resiliency."

David. said...

Michelle Starr's Giant Solar Storm 14,000 Years Ago Leaves The Carrington Event in The Dust reports that:

"In the rings of ancient, partially fossilized trees, scientists have found evidence of a solar storm at least an order of magnitude more powerful than the Carrington Event.

It took place, they say, some 14,300 years ago, well before there was a technology grid to disrupt. Such events, more powerful than Carrington, appear periodically in the fossil record. But this one is the most powerful ever seen.
...
The concentrations of both of these signatures are consistent with what we call Miyake events. These are incredibly powerful geomagnetic storms – way more powerful than the Carrington event.

Including this new discovery, nine Miyake events have been identified in the last 15,000 years, the most recent being around 774 CE."

David. said...

More about Miyaki events in Sun-like stars produce superflares roughly once per century by Valeriy Vasilyev et al:

"We used photometry from the Kepler space observatory to investigate superflares on other stars with Sun-like fundamental parameters. We identified 2889 superflares on 2527 Sun-like stars, out of 56,450 observed. This detection rate indicates that superflares with energies >10^34 erg occur roughly once per century on stars with Sun-like temperature and variability. The resulting stellar superflare frequency-energy distribution is consistent with an extrapolation of the Sun’s flare distribution to higher energies, so we suggest that both are generated by the same physical mechanism."

The good news is that:

"However, in 2021, another research team found that superflares tend to erupt from closer to the poles of stars, meaning that if the sun erupted with such a flare, there is a good chance it would miss Earth."