Science News July 12 (Patreon)

[This is a transcript with links to references.]

Welcome everyone to this week’s science news. Today we’ll talk about a new atomic nucleus, a map of a fruit fly brain, an explanation for the gravity hole in the Indian ocean, a better source of quantum light, how NASA is preparing for a trip to MARS, high resolution climate models, a new type of computer memory, faster data transfer, seaweed, and of course, the telephone will ring.

A student at the University of Jyväskylä, Finland has discovered a new atomic nucleus as part of her master’s thesis.

The new atomic nucleus is an isotope of astatine, that’s element 85 in the periodic table. The reason you’ve never heard of it is that it’s the rarest of the naturally occurring elements. It comes about only in the decay chains of heavier elements, such as uranium or thorium. Astatine itself decays very quickly. The most stable of its isotopes is astatine two hundred and ten that has a half-life of about 8 hours.

Because it decays so quickly, the total amount of naturally occurring astatine in Earth’s crust is just about 1 gram at any given moment. Probably a good thing because you don’t want it to sit on your desk while it’s spitting out alpha particles and turning into polonium.

That said, the stuff isn’t all that difficult to produce, you just need to slam big nuclei into each other at sufficient speed, and every once in a while, you’ll produce some astatine and can investigate it. This is what they did for the new paper. They fired a beam of strontium particles on a silver target, and then used a recoil separator to sort the decay products. They found clear evidence of decay chains that require the production of astatine one hundred and ninety which had never seen before. Quite a discovery for a master’s thesis.

The new isotope has a half-life of about a milli-second after which it quickly decays into radioactive remains. The researchers now plan to use this result to understand Jordan Peterson’s tweets.

A team of researchers from Princeton University has published a 3-dimensional replication of a fruit fly brain.

This is the first time the entire brain of an adult fruit fly has been mapped. It contains more than one hundred and twenty thousand neurons and more than thirty million synapses. The network of connections between the neurons in the brain is called the “connectome”. The scientists have called their project “FlyWire”. The connections are classified by what purposes they fulfil. Because it’s a neuron-level map, scientists are able to test how different parts of the fly’s brain work and relate to one another by activating and deactivating neurons. They’ve even simulated the “eating” and “grooming” behaviours of the fly by activating certain regions of the brain.

Just in March, a team of researchers completed the connectome of a fruit fly larva with about 3000 neurons. With the new work, we’re now at more than one hundred thousand. Their next big goal is the connectome of a mouse which contains about one hundred million neurons.

I’ve you’ve ever fancied tickling a fruit fly, the data are all available online and you can play around with it yourself. I can just see how the evolution of the internet will progress from antivirus software to virtual insect sprays.

Hello

Hi Mark,

Controversial fakebook posts go viral faster? I don’t think this is going to go viral.

You’re welcome!

Two Indian geophysicists claim to have found an explanation for the Gravity Hole in the Indian Ocean.

The gravitational pull on the surface of earth is roughly constant, but not exactly. That’s because strictly speaking the strength of gravity depends on what material is underneath your feet and how much of it. Scientists have known since the 1980s that’s there’s a patch in the Indian Ocean where the gravitational pull is abnormally low. You can see it here in this image. The bluer, the lower the gravitational pull. It’s called the “Indian Ocean geoid low”, aka the “Gravity Hole”, and lots of theories have been put forward for it.

For the new paper that was just published in Geophysical Research Letters, the scientists created a model for the movements of tectonic plates and ran 19 simulations for slightly different scenarios, beginning one hundred and forty million years in the past.

In six of those simulations, the anomaly was similar to the gravity hole we see today. What these six simulations had in common is that the tectonic plates in that area churned up plumes of very low-density magma as they fell into the mantle. Beginning around 20 million years ago, these plumes of magma slowly worked their way up towards the crust, creating a lack of solid mass beneath the ocean floor.

While the results of the simulations are in line with previous models, it’s unclear whether these magma plumes are actually the right explanation. Finding out would require a lot of digging, which isn’t all that easy to do on the ocean floor. Luckily, the anomaly is expected to last a few million years longer, so geoscientists have some time to figure out how to do that.

Researchers from MIT have found a new way to create light with quantum properties.

Research in quantum computing has seen some dramatic shifts in the past few years. The first approaches, which are still dominant today, use tiny superconducting circuits or trapped ions as the units of computing. The problem with these is that they need to be cooled to near absolute zero. One promising avenue to get around this is computing with photons, the quanta of light.

For this to work, not any photons will do, they need to all be very similar to each other, or “coherent” as physicists say, because otherwise they can’t properly interfere with each other. It’s turned out to be difficult to come up with such photon sources that are also cheap and easy to produce.

In the new paper now they grew a type of perovskite nanocrystals. You might have heard of perovskite before, it’s also a big hope for the next generation solar panels. They used a laser to excite the atoms in the crystals, and then they measured the photons that were emitted for the next 10 minutes or so. Their measurements show that the photons are in most cases sufficiently similar so that quantum interference is possible.

This also finally gives us a good definition for quantum light: The photons are single and ready to mingle.

NASA just locked up 4 volunteers who will spend an entire year in a habitat to simulate future Mars missions.

The experiment is called the “Crew Health and Performance Exploration Analog”, CHAPEA, for short. The four volunteers will live in close quarters of about one hundred and fifty square meters. The entire habitat was 3D printed at NASA's Johnson Space Centre in Houston, Texas and comes with a red sand enclosure where the crewmembers can conduct simulated experiments in spacesuits assisted by augmented reality simulations.

The four volunteers passed the same tests required of astronauts before being selected for the mission. They’ll eat freeze-dried food, and their communication with the outside NASA team will be conducted on a 22-minutes delay, which is how long it would take for a message to be sent from Mars to Earth. NASA hopes that this experiment will help them learn more about the psychological impacts and logistical challenges of locking people into close quarters millions of miles away from their home planet.

NASA also plans to put the volunteers into a few ‘surprise’ situations, and I certainly hope that Martians will be involved.

At a conference in Berlin last week, the computer company Nvidia has announced they’re going to support an initiative from climate scientists to create an Earth Visualization Engine.

The climate on earth is changing rapidly, and to figure out how to best adapt and mitigate we need detailed information about what change is going to happen where. Current climate models just aren’t good enough to provide local information because their grid resolution is too low. They are also not good with predicting extreme weather events, especially not their duration.

The major culprit isn’t anything specific about the models. Embarrassingly enough it’s just a lack of computing power. The Earth Visualization Engine is an initiative to remedy this problem, and Nvidia is supporting it with both GPU power and artificial intelligence. Nvidia says that they’ll be able to bring the resolution of climate models down to the kilometres scale and also speed up computation, which will allow larger ensembles and hopefully smaller uncertainties. At the meeting last week they presented some very impressive preliminary results.

And if we wreck the planet entirely at least we’ll have a digital copy to marvel at.

Researchers at the University of Cambridge have created a new type of computer memory that could make computers faster and more energy-efficient.

By the end of the decade, internet and communications technologies are expected to consume about thirty percent of global electricity, so more energy efficiency would certainly be a good idea. We also wouldn’t complain if computers were faster.

The new type of computer memory works by what is called resistive switching. This means that a material can be prodded with an electric current and in responses changes its resistance. The resistance then keeps the information. Materials can do this for example when defects or ions can relocate easily.

Resistive switching can store data when the power supply is removed, so it can be used as permanent storage like your USB stick.  But since it can be switched very quickly it can also be used as a working memory. Materials for resistive switching have been developed previously, but they tended to be unstable and not suitable for technological applications.

In the new work now, the researchers report a simple method to develop a material for the purposes. It’s a mix of hafnium oxide and barium. They demonstrated that it has the properties they desire and already filed a patent for it.

Personally I just hope they won’t make authentication codes any longer than 6 digits because that’s about as much as my memory can hold.

Hello

Hi Elon,

Famous wreck?

Ah, for a moment I thought you meant Donald Trump. Yes, of course I heard of the Titan accident, very tragic.

I think people just like the word “deep”. Deep oceans, deep space, deep fry…

Deep pockets, exactly, lots of famous wrecks down there.

No, I’ll stick with the shallow jokes.

Talk soon.

Researchers from Columbia University have found a way to massively increase data transfer in high-performance computers, one of the main bottlenecks in speeding up computation.

Supercomputers have massive computing power within individual nodes, but the data usually has to be shovelled around between those nodes. That takes time and slows down the entire computation.

The current best way to do this is through fibre-optic cables. But to make that work, the data must first be converted from electrical currents to light and then back again. These conversions eat up time and energy.

In the new study that was just published in Nature Photonics, the researchers have created a microchip that does the job. It uses frequency combs that break up the signal into thirty-two channels of different wavelengths that can act independently. The device is barely a few millimetres in size, which helps to reduce energy consumption. An additional bonus is that these chips are easy to produce with existing technology.

They were able to transfer data from one chip to another one at a rate of a stunning five hundred and twelve Gigabits per second. But computers won’t impress us until they can select all images with bicycles.

A new study from researchers in California found that seaweed isn’t going to save the world.

Scientists have come up with a lot of ideas for how to get the climate back under control. One of them is letting seaweed do the job. Seaweed grows quickly and it binds carbon dioxide, so if you can get it to sink to the bottom of the ocean, potentially this could remove carbon dioxide from the atmosphere for thousands of years. Previous studies have found that it isn’t all that easy to make the stuff actually sink to the bottom of the ocean and stay there, but the new paper now points out another problem. It might be difficult to grow that much seaweed to begin with.

To have any noticeable impact on the climate, the seaweed would have to absorb at least a few Gigatons of Carbon Dioxide a year, that’s a few billion tons. For their calculation the researchers looked at the multiple factors that would limit the growth of the seaweed, most importantly nitrate supply. They found that the amount of seaweed needed to absorb sufficient carbon dioxide would require seaweed farms covering at least one million square kilometres of the most productive waters in the world, in the best case the equatorial Pacific Ocean. If the farms were in other regions, they might need to cover about three million square kilometres, and farmers would have to add nutrients to keep the plants growing.

For comparison, the country of Egypt is about a million square kilometres, so picture an Egypt-sized blob of seaweed farms in the Pacific. Once you get those thoughts of seaweed pyramids out of your head, you’ll begin to see the problem – even if the area farmed for seaweed would be broken into smaller parts, the logistics are unfeasible. Besides, the environmental impacts are rather unforeseeable.

So keep on dusting those solar panels guys.