Science News Feb 1 (Patreon)

[This is a transcript with references]

Welcome everyone to this week’s science news. Today we’ll talk about changes in the Earth’s core, a quantum computer that folds a mini-protein, radioactive water that the Japanese want to dump into the ocean, NASA’s plans for a nuclear rocket and a transonic plane, how to send data without power, recovering signals from noise, light pollution, and of course, the telephone will ring.

If you trust recent headlines, which you should not, Earth’s core seems to have stopped spinning and might even start spinning into the opposite direction. It must be very tiresome to be the centre of earth day-in, day-out, so I wouldn’t blame it if it were to stop spinning, but this isn’t what’s happening.

We don’t really know what’s in the centre of earth because no one’s been there, but geophysicists have a pretty good idea based on data from earthquakes, magnetic fields, and precision measurements of the strength of gravity. Their best current model says that if you dig down into earth, at about three thousand kilometres depth you get to what’s called the outer core, which is liquid and believed to be mostly iron and nickel.

If you go further down to about five thousand kilometres you get to the inner core. The inner core is made of the same stuff as the outer core, but it’s solid because it’s under pressure. The pressure comes from the weight of the upper layers that pushes in on the core and prevents the metals from melting.

The inner core is roughly 70 percent the size of the Moon and has approximately the same temperature as the surface of the Sun. It floats in the centre of the molten outer core, which is why it doesn’t necessarily spin at the same pace as the mantel. Indeed, previous measurements suggested that the core of earth is spinning somewhat faster than the mantle.

The recent headlines are about a paper that was just published in Nature Geoscience. The authors looked at measurements of earthquakes whose shockwaves travel through the Earth’s core. These waves get redirected by the rotation, so the rotation can be inferred from the measurements.

Most earthquakes happen in the always same regions of the planet, often because tectonic plates are rubbing on each other. This makes it possible to figure out how the earthquake waves that pass through the core change over time. In the paper they now found that around 2009, the core seems to have stopped spinning faster than the mantel of earth. The best fit to the data is currently that it’s now spinning somewhat slower.

So it’s not like the core of earth has stopped spinning or reversed direction. They say it’s stopped spinning faster than the mantle and is now spinning somewhat slower.

What does this mean? Well, you may remember that angular momentum is conserved. This means if the core of earth spins somewhat slower, then something else must spin faster. That something else is the earth’s mantle that we live on.

There are larger contributions to the spin of earth’s mantle which come from the moon and the atmosphere. But if you want to be precise, if the core spins slower, the days get shorter. It’s only a tiny fraction of a second but if you’ve been feeling somewhat tired since 2009, that should explain it. Don’t blame the inner core too much, it’s under a lot of pressure.

A quantum computer has, for the first time, succeeded in simulating the folding behaviour of a protein molecule. The computer, called JUPSI and housed at the Jülich Supercomputing Centre in Germany, has five thousand qubits and is the largest such computer outside the United States.

JUPSI is not a universal quantum computer like Google’s Sycamore or IBM’s Osprey, instead it’s from the Canadian company D-Wave. Their devices specialize in quantum annealing which is basically an optimization mechanism. These computers automatically relaxe into an optimal state. If you want to make a calculation, you set up the computer so that this optimal state is the answer to your question.

This method is very good at tackling problems that have a great many potential solutions, in which finding the lowest energy state produces the answer. And it does that by exploiting quantum tunnelling, superpositions, and entanglement. So it’s a real quantum simulation. Whether you want to call it a quantum computation is somewhat of a semantic issue, it depends on what you mean by calculation. There are some people who say that D-Wave isn’t a proper quantum computer, but then again in the end it doesn’t matter what you call it but what you can do with it.

Folding a protein lattice into its minimum energy structure is a problem that annealing should be able to handle well. In their paper the authors used a very simplified model in which the folding is done in only two dimensions on a lattice. So the protein is basically a chain that you have to put onto a grid. They used a chain up to a length of 64 elements, each of which is supposed to be one amino acid. This simplified problem can and has been solved on conventional computers. The point of the study was not to outdo a conventional computer, but to show that a quantum computer can also do it.

The authors tried two different methods of calculation that the D-wave computer can do, one is a pure quantum calculation, the other one a hybrid version of the quantum and a conventional, classical simulation. Interestingly enough, they found that the pure quantum simulation decreased drastically in performance beyond a chain length of only 14, whereas the hybrid simulation did well up to 64 which was the largest they looked at. I have no idea what this means, but we’ll probably see some follow up studies on that.

One of the authors of the paper said in an interview that we may need up to three further generations of quantum computers before the calculations become relevant for drug research.

Speaking of quantum computers, Chalmers University of Technology in Sweden plans on making a 25 q-bit quantum computer available for researchers and industry, together with a quantum helpdesk.

I wonder how this will work.

Hello, this is the quantum helpdesk, how can I help you?

Have you tried turning it on and off at the same time?

The Japanese government has announced that it’ll be releasing radioactive water into the ocean, and would you believe it, not everyone likes the idea.

In 2011, the Fukushima nuclear power plant was wrecked by an earthquake and tsunami which eventually caused three cores to largely melt. There are currently more than one thousand tanks of water on the site, holding more than a million tonnes of water. The water is a mix of groundwater, coolant, and the seawater that flooded the coastal site. The site continues to produce contaminated water every day and they are running out of storage space.

Before the water is stored, most of the radioactive particles are removed, including cobalt-60, strontium-90 and caesium-137, but this process isn’t perfect, so some traces remain.

Another problem is hydrogen-3, also known as tritium. Since it’s chemically identical to normal hydrogen, it can bind with oxygen and form a type of water that is extremely difficult to separate from normal water. Its half-life is about 12 years. For what radioactive substances are concerned, tritium isn’t all that bad because it doesn’t strongly accumulate in the body. Though this is somewhat like saying that for what diseases are concerned the plague isn’t all that bad. In any case, you don’t want tritium in your tea, and neither do fish.

Releasing the water into the ocean will strongly dilute the radioactive substances, to a level where the added radiation dose is below the natural background level. This plan has been backed by the International Atomic Energy Authority which has closely monitored the decommission of the Fukushima power plant. It seems therefore likely that the Japanese will go ahead with dumping the water in the ocean.

Not so surprisingly, the plan has faced some opposition from interest groups in the surrounding countries. Henry Puna, leader of the Pacific Islands Forum, points to the appalling experiences of people on the Pacific Islands during nuclear weapons testing, fears over further damage to their health, and that many people on these islands rely on the sea for their livelihood. Fishermen have also joined the calls for a rethink, voicing worries about contamination of seafood. And they are probably right to worry, because in the end it doesn’t matter if the seafood is safe to eat so long as customers are afraid it’s not.

Speaking of radiation. Some people don’t like the idea of flying nuclear reactors, and historically space agencies have been among them, but this is about to change. NASA has announced that it’s teaming up with the US military’s advanced projects agency DARPA.

That’s the same agency that previously experimented with mechanical elephants for jungle warfare and tried to recruit telepathic spies. They also in the 1950s came up with the idea of accelerating a spaceship by chucking nuclear bombs out of the back and riding on the blast wave, which, if nothing else, is a pretty good metaphor for making jokes on YouTube.

Luckily their new plan is slightly less dramatic. They want to put a nuclear fission reactor into a rocket to supply heat to a tank of propellant, most likely liquid hydrogen. The hydrogen turns into a gas, expands and rushes out through a rocket nozzle to provide thrust in a good, old-fashioned way that Newton would have approved of. This method could provide more power than a chemical rocket, allowing spacecraft to carry up to three times the payload, and get astronauts to Mars more quickly.

It’s not a new idea. It originated in the 1950s and several ground-based tests were carried out by both the US and Russia, though no nuclear rocket has ever flown. The risk that a rocket failure or collision with orbital debris could distribute radioactive material all over the globe has kept the technology firmly on the ground.

Until now. NASA’s agreement with DARPA aims to put a rocket in orbit by as early as 2027, most likely boosted into space by traditional chemical rockets before turning on its reactor. What could possibly go wrong.

NASA also has plans to revolutionize passenger flight. They just awarded Boeing 425 million dollars to develop a transonic passenger plane, that is, one that flies close by the speed of sound but doesn’t cross it. The new plane will be designed to lessen the impact of air travel on the environment.

The money, which is to be paid over seven years, will go toward developing and flight-testing an airframe that can dramatically reduce the amount of fuel used in passenger flight, and hopefully influence the next generation of single-aisle commercial jets, which are responsible for nearly half of aeroplane emissions worldwide.

One of the technologies being tested is Boeing’s Transonic Truss-Braced Wing concept, which aims to reach speeds up to Mach 0 point 8. The new design, combined with new engines and materials could see fuel savings of 30 percent over current passenger planes. The wings of the new plane have a somewhat different angle, shape, a position on the aircraft that’s higher and further forward than usual, and a supportive truss that comes from the bottom of the fuselage, joining the wing slightly less than halfway along its length.

The demonstrator aircraft is expected to take its first flight in 2028. Let’s hope it’s on time.

Mr President.

That’s right, they found that sheep seem to make democratic decisions.

No, I don’t think that’s Russian propaganda. But maybe it explains why we speak of baa-llot boxes.

You’re welcome.

A group of engineers from the university of Washington have come up with a way to send signals seemingly without power, and no it’s not two cups connected by a wire. It also isn’t just a theoretical speculation, they actually built the thing and sent signals with it.

Doesn’t this violate one or two laws of nature? Not necessarily. Here’s what they did. They used noise to create a signal. You see, an electric circuit that isn’t powered will still create noise, because it’s constantly hit by particles from the environment. And that noise depends on whether the circuit is open or closed. So, the researchers say, rather than powering the circuit and submitting a signal with it, we’ll use a receiver to pick up how the noise changes when you open and close that circuit.

This doesn’t violate any conservation laws because opening and closing the circuit does require energy, if not much, and most of the energy has to be invested at the receiver to pick up the signal. They managed to detect these passive signals from up to 7 meter away, and achieve data transfer rates of up to 26 bps. That’s bits per second and slower than my first dial up modem, so there’s someway to go. But one day I’ll use nothing but a noisy circuit to control the quantum computer that’s flying your transonic plane. And then I’ll wake up.

A group of researchers from China and Saudi Arabia have come up with a new method of recovering data from obscured, blurred, or noisy images. A lot of techniques have been proposed for this before, but theirs excels at doing the job with a low-resolution camera with little computational cost.

For this research, they passed light from a known image through a random diffuser to simulate the effect of cloud. They moved the camera around to collect more information, built a three-D model of the diffuser cloud, and recovered the image after subtracting the effect of the diffusion.

This method doesn’t magically produce a perfect colour photograph from a mass of blur and noise, but it does improve the readability of single letters. Maybe they’ll one day even find a way to decipher my doctor’s handwriting.

Light pollution is getting worse faster than expected, according to a citizen science project whose results were just published in Science. The study collected reports from more than 50 thousand participants all over the world. They found that star sightings decreased faster than brightness measurements from satellites would suggest.

According to the authors, it’s because the spectrum of the emission and the typical angle has changed. Switching to LED lights, they say, has significantly contributed to the problem. Because LED streetlights save energy, they are more widely being used. At the same time, the cheapest LED lights have a large blue component of the spectrum, which is much more disruptive for sleep cycles than the old, yellowish light. There are LEDs that both save energy and have a low blue part of the spectrum, but they are somewhat more expensive, and they are presently basically not being used. I talked about this in more detail in an earlier video.

Light pollution isn’t just inconvenient for star lovers. It’s correlated with sleep problems, both in quality and quantity, especially if the light has a strong blue component. Light pollution disorients and weakens insects and birds with effects that propagate up the food chain, and it makes finding prey difficult for nocturnal animals, such as bats, foxes, and serial murderers.

But don’t worry, the future will be bright.