Mountain Mars

It’s all about the long haul. The Hawaii Space Exploration Analog and Simulation (HI-SEAS) tasked six researchers with spending a , Mauna Loa. They were only allowed leave their habitat to explore the vicinity if they wore space suits. The team didn’t see another soul for an entire year, and all communications were put through a 20-minute delay, to mimic the reality of interplanetary existence.

Within the dome, the team experimented to find out which crops might grow best inside a Martian hut under LED lights and with limited water. They ate the results of these hydroponic experiments – – and also ate fermented foods they made themselves, including yogurt and cheese.

The crew celebrated Earth holidays and had a non-denominational winter celebration. Sheyna Gifford (pictured above), the mission’s chief medical officer, wonders if Christmas will make sense on another planet. “When are they going to do that on Mars given that the calendars don’t match?” She suspects Martian settlers will probably invent new holidays such as “Landing Day”, the date of the first human touchdown on the Red Planet. By the time a colonising spaceship arrives there, the crew will already have developed their own unique culture, Gifford reckons.

White Mars

The average temperature on the Martian surface is about -55 °C, so prolonged exposure to bitter cold will be one of the main challenges of living on  the Red Planet. Last year, a crew of 13 overwintered at Concordia station (pictured) in Antarctica – in part to test the .

All wore activity watches which detect when another person wearing a watch comes close, to explore how their interactions changed over time. Crew member Beth Healey of the European Space Agency, a medical doctor, saw that the group divided into sub-groups based on when they were awake during the extended polar darkness, which lasted over 100 days. Some people tended to have chaotic sleep-wake cycles, while others clung to a traditional 9-to-5 work schedule using cycles of artificial light. “You could really only be friends with people who were up at the same time,” says Healey. Enforcing common mealtimes helped the different groups communicate.

Healey also examined her colleagues’ video diaries to see if the way they talked about their lives changed throughout the mission, and looked at brain scans taken before and after their stay. Findings from this research are yet to be published, but brainwave recordings of crews on a past mission showed that their brains were surprisingly resilient to the prolonged dark and isolation.

Though the winter was brutal, Healey noted in her blog that . “It was strange to see how quickly money can become meaningless,” she wrote, because the base does not use currency. Instead, improvised gifts from crew members became strangely valuable. “A tinfoil penguin and a lampshade made from can ring pulls and stones were among my favourites.”

Desert Mars

In a remote corner of Utah, the Mars Desert Research Station has other Red Planet analogues beaten for sheer barren glamour. “It’s a half-hour drive from the town of Hanksville, and Hanksville has the motto, ‘Where the hell is Hanksville?’” says at the Canadian Museum of Nature in Ottawa, who recently did a stint as a crew biologist at the station.

The facility is owned and run by the , and houses “Utahnauts” in a cylinder-shaped habitat designed to look like something dropped off by a rocket. Whenever they head out through the small airlock, they wear orange jumpsuits and bubble helmets that have a nasty habit of fogging up.

Enough rain falls there to let some small shrubs, fungus and soil microbes grow. Sokoloff’s project was to survey this “Martian” flora. Astronauts won’t be picking sagebrush or mushrooms on Mars, but the soil-dwelling bacteria Sokoloff collected could reflect the survival strategies of Martian microbes, if they exist. The methods used to carefully isolate those extreme life forms from the soil are much like what future Mars botanists must resort to. Some microbes change the structure of the rocks and soils that host them: if similar structural shifts turn up on Mars, it could signify life.

Wet Mars

Mars’s surface gravity is only two-fifths that of Earth’s, and Mars has no breathable air – so what better place to mimic its suffocating floatiness than under the sea on Earth?  This is the premise of the Aquarius Reef Base, a small habitat 19 metres down off the coast of Key Largo, Florida. In July last year, six aquanauts spent 16 days living down there – the 21st instalment of the .

The crew tested equipment inside the habitat, and carried out simulated spacewalks on the seabed. The lessons learned there could be invaluable for future, low-gravity missions, to Mars and beyond.

FEELING blue? A new algorithm can predict depression from photos people post on Instagram and may in future prompt a visit to the doctor.

Andrew Reece at Harvard University and Chris Danforth at the University of Vermont in Burlington surveyed the mental health of 166 people, then set an algorithm to comb through 43,950 photos posted by them on social media, looking for features correlating with depression.

Depressed people, it turned out, tended to post blue-toned or dim photos, or use black-and-white filters. They were also more likely to post photos with faces, but with fewer faces per photo. Danforth says this might mean lots of selfies, part of the focus on self that occurs in depression, or it may mean the person is spending less time with friends or family ().

The algorithm used these features to predict depression 70 per cent of the time – better than the 42 per cent average accuracy of human doctors.

“It’s very inexpensive – it’s something that could be an app on someone’s phone,” says Danforth.

However, it could be tricky for doctors to access this data. “Most physicians are reluctant to view social media profiles of patients due to worries about confidentiality,” says Megan Moreno at Seattle Children’s Hospital.

This article appeared in print under the headline “Instagram posts can hint at depression”

ROOM for one more? Six thousand people somehow crammed into this indoor pool in Suining, south-west China, last week, seeking refuge from temperatures in excess of 40 °C. The chaotic, gaudy scene unfolded at a resort called the Dead Sea. Here the water’s natural concentration of salts – at around 22 per cent – gives swimmers a buoyancy similar to what people experience at the real Dead Sea in the Middle East. The resort also has a 10,000-square-metre artificial surfing pool.

The searing conditions were the result of high-pressure systems trapping hot air over the country. Last week, China issued an orange alert, its second-highest temperature warning, for central and southern areas.

It isn’t just China that is sweltering. Last month, Earth’s average surface temperature was at its highest since records began.

This article appeared in print under the headline “An inflatable rainbow”

It’s a squid-eat-squid world down there. Remotely controlled submersibles have filmed two deep-sea species of squid gulping down members of their own species.

Several species of squid, including giant squid, are known to eat each other.

But we have mainly gleaned this from their stomach contents, and it wasn’t clear if this was normal behaviour, or something they did when they have been caught in a net, says at the GEOMAR Helmholtz Centre for Ocean Research in Kiel, Germany.

“Animals get stressed in a net,” he says. “They probably start doing something called ‘net feeding’ – they feed on anything that is close by.”

Hoving and his colleagues sent down subs to observe two species of the Gonatus genus of squid in an underwater canyon of Monterey Bay, off the coast of California. The squid span much of the water column, from shallow waters to more than 2000 metres down.

Caught cannibalising

The subs frequently saw both species eating their own kind, especially Gonatus onyx. “Up to 40 per cent of the animals examined were feeding on animals of the same species,” says Hoving. “That was quite high – that was surprising.”

The subs filmed the squid doing this in their native habitat, so it could be ordinary behaviour for them, Hoving says.

The squid grow quickly and breed only once, needing lots of food to fuel their metabolism, says Hoving. When other food is scarce, cannibalism could be a good way to get food, he says.

There are other benefits, too. “It eliminates one competitor for food,” he says.

“Clearly cannibalism is important,” says at the University of South Florida in St Petersburg. “They’re doing it on a semi-regular basis or you wouldn’t see it from a submersible.”

Don’t eat too many

But it wouldn’t make sense for the species to overindulge in itself, Seibel says. “A species can’t exist feeding only on itself.”

Using subs with high-powered lights to observe the creatures could have led to the amount of cannibalism being overestimated, Seibel says. It’s possible that squid clutching large prey items – such as their kin – are less able to swim away from the vehicle, meaning that they’d be counted more often than squid eating lighter meals, he says.

To get the most accurate picture of the elusive animals’ behaviour, you have to combine observations from subs with other methods such as netting, he adds. In general, these squid aren’t picky eaters, says Seibel. “They eat just about whatever they can get their tentacles on.”

The trade in bunnies helped power an ancient economy. Teotihuacan, an ancient city in central Mexico, was an advanced metropolis where most people lived in apartment complexes. The city reached its peak between the first century and 550 AD. With about 100,000 residents, it was the largest urban area in the Americas at the time, of a similar scale and sophistication as other ancient centres like Alexandria and Rome.

But until now, it has been a mystery what kinds of animals supported this complex society. “One of the big puzzles for the pre-Colombian Americas has always been the lack of domesticated animals,” says at Boston University.

Other than managing dogs and turkeys, Mesoamericans didn’t appear to have the close relationships with animals that sustained ancient peoples in Africa and Europe.

Now it seems that raising cottontails and jackrabbits may have given the city a reliable source of meat and fur. at the National Autonomous University of Mexico in Mexico City and colleagues have uncovered an apartment compound that seems to have belonged to rabbit breeders and butchers. The team found rooms littered with rabbit bones, as well as obsidian blades for butchering and for scraping skins.

The remains of baby rabbits and a low-walled room that appears to have been a pen indicate that the inhabitants were breeding and rearing the animals, Manzanilla says.

A stone rabbit sculpture on top of a household courtyard temple (see illustration below) suggests that the residents specialised in the rabbit trade.

The carbon within the rabbit bones gave another clue, says at the University of California in San Diego. Animals eating maize and other common Mexican crops like agave cactus tend to have higher levels of an isotope of carbon with an extra neutron.

Analysing the bones showed that up to 74 per cent of the animals’ diet came from human-grown foods rather than wild plants.

“This study does a great job of showing the innovations in this urban society for cultivating their own protein sources,” says Carballo. “It gives you a good idea of what regular folks were up to in this city.”

The rabbits could have served a few different uses, such as a source of meat and fur or ritual purposes, says at the University of North Carolina at Chapel Hill.

at Arizona State University in Tempe says we shouldn’t overestimate the importance of the meat, because the diet of beans and maize available at the time was already a complete protein source. “It’s not as if, ‘oh my gosh, they’re starving if they don’t get some rabbit meat.'” Still, the study gives more evidence that Teotihuacan had a highly organised economy with specialised workers, Smith says.

Palatial housing

The city’s tradespeople, like the rabbit butchers, were well off. Nearly everyone lived in large multifamily apartment buildings that would have matched royal palaces in other ancient cities. “I don’t know of any other ancient society where the bulk of the population lived in such luxury,” he says.

There’s also a conspicuous lack of royal tombs or paintings of powerful leaders amid the city’s abundant murals, says Carballo. This suggests that there were no kings; instead, government was probably a more collective affair.

It seems one of the ancient city’s traditions remains. “Some of the delicacies of the Teotihuacan valley today involve rabbit,” says Carballo. “It continues to be an important food for the area.”

Read more: Huge Mexican pyramid could collapse like a sandcastle

Some of the best lessons come from the school of hard knocks. But some kit is too delicate or expensive to be subjected to this. So researchers have instead taught cheap, expendable drones to pass on their hard-won knowledge to their more precious peers. Getting robots to learn and share general concepts in this way could also make them better at independent decision-making.

Teaching an artificial intelligence to fly an expensive vehicle is risky, since it needs to know what both success and failure look like. “Let’s say you want to train it to fly a really big helicopter,” says at the Jet Propulsion Lab in Pasadena, California. “You need it to crash a lot to get it to learn what a crash is – but that’s often not possible.”

Not wanting to risk a new, expensive drone, Daftry and colleagues took a cheaper vehicle and piloted it through a forest, sometimes taking it between obstacles and sometimes crashing it. Trial and error let the robot figure out how to fly safely by itself.

The researchers then took the drone’s abilities and transferred them to their more expensive craft, which was immediately able to use the second-hand know-how to avoid flying into trees itself.

Dog-brained drones

The trick to passing on an ability that can be adapted to fresh situations lies in the way the first drone learns. Teaching an AI sometimes works like dog training – the robot gets a treat or a slap on the paw depending on its choices, says at the Massachusetts Institute of Technology.

But in this case, the drone first has to learn what’s a treat and what’s a slap – then it writes its own rules based on its experiences. “It’s trying to work out for itself what its own reward should be,” says Roy.

The upshot is that the initial drone picks up general-purpose rules. Rather than learning specifically that it should go left when it sees a brown patch – a tree, say – it learns that a brown patch is bad and must be avoided in any way possible.

The strategy should work for many kinds of robots, says Roy. Getting robots to pass on general concepts they have learned makes them much more independent, he says. “It’s a change in how we think robots should make decisions.”

Reference: ArXiv,

Read more: How drones are learning to find their own way in the world; Moth navigation probed to improve micro-drone flight

Species: Greenland shark (Somniosus microcephalus)
Habitat: deep in the North Atlantic and the cold surface waters of the Arctic

Fish that were alive during the Age of Enlightenment are still swimming strong. A Greenland shark has lived at least 272 years, making the species the longest-lived vertebrate in the world – smashing the previous record held by a 211-year-old bowhead whale. But it may have been as old as 500 years.

“We definitely expected the sharks to be old, but we didn’t expect that it would be the longest-living vertebrate animal,” says of the University of Copenhagen, Denmark.

Living deep in the North Atlantic and the frigid surface waters of the Arctic, Greenland sharks have a stable environment and grow just a few centimetres per year. Despite their slow growth, though, they reach more than 5 metres in length and are often the apex predator in their ecosystem.

Old blue eyes

It was once thought to be impossible to age Greenland sharks. Their skeletons, made of cartilage, lack the calcified growth rings of hard-boned vertebrates. And other fish are aged by measuring calcareous bodies that grow in their ears, but this doesn’t work for sharks.

Instead Nielsen and his colleagues focused on radiation in the sharks’ eyes. Nuclear bomb tests in the 1950s and ’60s blasted radioactive particles into the atmosphere.

Those particles entered food webs all over the world and show up in the form of radioactive forms of carbon in organisms that lived through that period. Because Greenland sharks’ eye lens tissue doesn’t change during its lifetime, it preserves the historic radiation.

After catching a 2.2-metre shark that showed radiation levels indicating it was born in the 1960s and was about 50 years old, the team calculated how fast the sharks grew.

150-year dry spell

The team estimated that one 5-metre animal was at least 272 years old – but could be more than 500 years old (392 +/- 120 years). Another was at least 260 years old, and could be more than 400 years old.

And the female sharks don’t seem to reach breeding age until they are about 150 years old. “They have to wait more than 100 years to get laid – I’m sure they’re not happy about that,” says Nielsen.

Despite the uncertainty in estimating birthdays deep in the past, it’s clear these sharks are centuries-old, says at the Australian Institute of Marine Science.

“This is our best estimate of how old these things are, but I don’t think it’s the final word,” he says. “The key message here is that these things are living a very, very long time.”

Non-vertebrates can live longer, for example, some coral and sponges are thought to live thousands of years. Clams, too, can live for hundreds of years.

Pining for the fjords

Many aspects of the Greenland shark’s life remain unknown, says MacNeil.

We don’t know for sure where shark pups are born, but one hypothesis is that females give birth to live young in Arctic fjords, MacNeil says. It’s also unclear how much climate change will affect the species through changing its cold-water habitats, he adds.

Finding out more would help scientists determine whether the species has a population healthy enough for a fishing industry, says Nielsen.

While there seem to be many young Greenland sharks around, Nielsen believes many breeding-age sharks were harvested for oil around the time of the second world war, and because of the time it takes them to mature sexually, the population will be recovering for another 100 years. The long generation times could also make them vulnerable to habitat disturbances, he says.

Other mysterious deep-sea shark species could also have surprising life spans, says at the University of Miami in Florida. “Every single time you study a deep-sea shark, you make a new discovery,” he says. “This might be just the tip of the iceberg.”

Read more: Mysterious deep-sea sharks biting chunks out of migrating whales; Extreme survival: Meet the immortals

One day your hard drive could just be a pile of plastic. Researchers have coded a word into short chains of plastic molecules, which could be used as a space-saving way to store our mountains of data – or even to reveal counterfeit goods.

DNA has shown some promise in holding millions of bits of information in a tiny volume. But DNA is fragile and hard to write and read. So at the Institut Charles Sadron in Strasbourg, France, has been experimenting with more customisable chains of plastic molecules that can encode information in similar ways.

Also known as polymers, these chains are made up of two kinds of molecules that stand for the 1s and 0s of digital computer code. Previous research has seen data stored in single long chains, but these become harder to read as the length increases, so the storage record stands at just 10 bits.

Lutz’s team had a different approach. “Instead of making very long chains, the idea is to create a library of very short chains,” he says.  As a demonstration, the team wrote the acronym CNRS, the abbreviation for the French National Center for Scientific Research, across six polymers – a 32-bit message when encoded using standard ASCII characters.

The researchers read the message by sorting the chains from shortest to longest using a mass spectrometer and then sequencing the chains by breaking them apart molecule-by-molecule. The shorter chains, less than a nanometre each, are easier to manage than one long one, while their varying length keeps the data in the right order.

“I think it’s an encouraging step,” says at the University of Washington in Seattle.

Stolen goods

To store big chunks of information, researchers will have to scale up the technology to make large libraries of chains, Ceze says. In addition, it would be quicker to write data into the molecules if the chains could be synthesised in parallel rather than one at a time, he adds

While writing and reading these polymers is currently expensive, there are some high-end uses for them. Embedding coded polymers in pricey electronics or artworks could label them with the maker’s or owner’s identity, says Lutz. Reading a sample of the molecule chains would let investigators discover the source of the object, helping them find counterfeits and stolen goods.

Both DNA and synthetic polymers have the potential to shrink the size needed for data storage, says at Boise State University in Idaho. That’s important, since the world’s data is piling up faster than our ability to store it. In 2040, the amount of space needed to keep our data in silicon chips could be as large as a small country, perhaps twice the size of Liechtenstein, he says.

Using several types of molecules would let researchers write in a code that packs more information into shorter sequences than binary 0s and 1s can, says Zadegan. While scientists have introduced two artificial letters into DNA for a total of six letters, it could be simpler to design an entirely new plastic molecule alphabet. “These polymers are perhaps easier to deal with if you want to expand the language,” he says.

Journal reference: Angewandte Chemie International Edition, DOI:

Read more: Plastic fantastic: The quest to create the smartest materials

Mammoths were in crisis at the end of the last ice age, when human hunters began to settle in their habitat. Most mammoths on mainland Asia and North America went extinct over 13,000 years ago.

But a few hardy herds clung to Arctic islands where people did not arrive for several more millennia. One of these populations retreated to the Bering land bridge linking Siberia with North America. As the sea rose, they became stranded on the small, low Alaskan island now called St Paul.

What happened to them remained a mystery until Russell Graham at Pennsylvania State University and colleagues pinpointed that 5600 years ago, they ran out of water. “It’s probably one of the best-dated prehistoric extinctions around,” says Graham. The team reached that precise date using three lines of evidence, including the radiocarbon dating of both mammoth bones and lake sediment containing mammoth DNA (PNAS, ).

“They all coincide on 5600 years ago – that’s wonderful and unique as far as I know,” says Adrian Lister at the Natural History Museum in London.

The lake sediments also gave away the cause of death. Right around the extinction time, the lake plankton switched from deep-water species to shallow-water ones. This pointed to steadily evaporating fresh water. Mammoths, like their living cousins the elephants, needed many litres of water every day to survive. Low supplies would have done them in quickly, says Lister.

The very last mammoths lived on the larger Arctic island of Wrangel, off Siberia, until 4000 years ago, but it’s unclear whether climate or humans finished them off.

This article appeared in print under the headline “America’s last mammoths died of thirst on an Alaskan island”