Dinosaurs were once assumed to have been ectothermic, or cold-blooded, an idea that makes sense given that they were reptiles. While scientists had previously discovered evidence of dinosaur species that were warm-blooded, though what could have triggered this adaptation remained unknown. A team of researchers now think that dinosaurs that already had some cold tolerance evolved endothermy, or warm-bloodedness, to adapt when they migrated to regions with cooler temperatures. They also think they’ve found a possible reason for the trek.

 

Using the Mesozoic fossil record, evolutionary trees, climate models, and geography, plus factoring in a drastic climate change event that caused global warming, the team found that theropods (predators and bird ancestors such as velociraptor and T. rex) and ornithischians (such as triceratops and stegosaurus) must have made their way to colder regions during the Early Jurassic. Lower temperatures are thought to have selected for species that were partly adapted to endothermy.

 

“The early invasion of cool niches… [suggests] an early attainment of homeothermic (possibly endothermic) physiology in [certain species], enabling them to colonize and persist in even extreme latitudes since the Early Jurassic,” the researchers said in a study recently published in Current Biology.

Hot real estate

During the Mesozoic Era, which lasted from 230 to 66 million years ago, proto-dinosaurs known as dinosauromorphs began to diversify in hot and dry climates. Early sauropods, ornithischians, and theropods all tended to stay in these regions.

 

Sauropods (such as brontosaurus and diplodocus) would become the only dinosaur groups to bask in the heat—the fossil record shows that sauropods tended to stay in warmer areas, even if there was less food. This suggests the need for sunlight and heat associated with ectothermy. They might have been capable of surviving in colder temperatures but not adapted enough to make it for long, according to one hypothesis.

 

It's also possible that living in cooler areas meant too much competition with other types of dinosaurs, as the theropods and ornithiscians did end up moving into these cooler areas.

Almost apocalypse

Beyond the ecological opportunities that may have drawn dinosaurs to the cooler territories, it’s possible they were driven away from the warm ones. Around 183 million years ago, there was a perturbation in the carbon cycle, along with extreme volcanism that belched out massive amounts of methane, sulfur dioxide, and mercury. Life on Earth suffered through scorching heat, acid rain, and wildfires. Known as the Early Jurassic Jenkyns Event, the researchers now think that these disruptions pushed theropod and ornithischian dinosaurs to cooler climates because temperatures in warmer zones went above the optimal temperatures for their survival.

 

The theropods and ornithischians that escaped the effects of the Jenkyns event may have had a key adaptation to cooler climes; many dinosaurs from these groups are now thought to have been feathered. Feathers can be used to both trap and release heat, which would have allowed feathered dinosaurs to regulate their body temperature in more diverse climates. Modern birds use their feathers the same way.

 

Dinosaur species with feathers or special structures that improved heat management could have been homeothermic, which means they would have been able to maintain their body temperature with metabolic activity or even endothermic.

 

Beyond the dinosaurs that migrated to high latitudes and adapted to a drop in temperature, endothermy might have led to the rise of new species and lineages of dinosaurs. It could have contributed to the rise of Avialae, the clade that includes birds—the only actual dinosaurs still around—and traces all the way back to their earliest ancestors.

 

“[Our findings] provide novel insights into the origin of avian endothermy, suggesting that this evolutionary trajectory within theropods… likely started in the latest Early Jurassic,” the researchers said in the same study.

 

That really is something to think about next time a sparrow flies by.

 

Current Biology, 2024.  DOI: 10.1016/j.cub.2024.04.051

 

European car manufacturers are currently tripping over themselves to figure out how personal transport and "last mile" solutions will look in the years to come. The solutions are always electric, and they're also tiny. What most companies (bar Citroen, Renault, and Fiat) seem to have forgotten is that we've had an answer to this problem all along: the microcar.

 

The microcar is a singular little thing—its job is to frugally take one person (or maybe two people) where they need to go while taking up as little space as possible. A few have broken their way into the public consciousness—Top Gear made a global megastar of Peel's cars, BMW's Isetta remains a design icon, and the Messerschmitt KR200 is just plain cool—but where did these tiny wonders come from? And do they have a future?

 

Well, without the microcar's predecessors, we may not have the modern motorcar as we know it. Sort of.

 

Let's roll back to the genesis of the car: the Mercedes-Benz Patent Motorwagen. While not a microcar by any means (though it seats only two people and has a tiny engine and only three wheels), it got plenty of people thinking.

 

While Karl Benz was inventing the car and his wife was road-tripping it in 1885, a French inventor named Léon Bollée put his thinking cap on. He was 15 at the time, but it gave him time to be with his thoughts. At that age, he had a keen brain—one that invented a pedal boat of sorts. Bollée was smart, to say the least—he built calculators to help his father's business, one of which won an award at the 1889 Paris Exposition and went on to be patented all over the world.

 

In 1895, Bollée and his father created "La Novelle," a steam-powered trike, and in the same year, Bollée created a gasoline-powered… thing as well. A year later, Bollée founded Léon Bollée Automobiles to mass-produce his tiny cars, dubbing them "Voiturette"—a mashup of the French for automobile (voiture) and the suffix you throw on a word to make it small (ette). Small car, basically.

 

A few years later, Renault (maker of tiny hatchbacks and the gloriously silly Avantime and popularizer of the people carrier in Europe) became a car manufacturer with the release of its descriptively named Voiturette. Louis Renault's small mechanical wonder was built in 1898, and the first was sold on Christmas Eve of the same year to a friend of Louis' father—he liked the fuel economy from its one-cylinder De Dion-Bouton 273 cc 1.75 hp (1.3 kW) engine and the fact that it could get around town with ease.

 

That same night, the story goes, Renault sold a further twelve cars. Over its mere five-year production run, the first Renault went from open-top two-seater to a four-seat covered wagon capable of over 35 mph (56 km/h). Bear in mind that less than a century earlier, Stephenson's Rocket and its almost 30 mph (48 km/h) top speed caused great concern about whether human physiology could withstand such speeds. 35 mph was quite the achievement.

 

Voiturettes and their less "ette" siblings were very successful, but they were a bit too much for some people. That's where the cyclecar came in.

 

First appearing around 1910, cyclecars took small engines—single cylinders, V-twins, the odd four-pot—and attached them to simple, lightweight four-wheeled bodies. To be a cyclecar, a vehicle had to have a gearbox and clutch. A huge industry popped up around them, and for good reason—regular cars were expensive to tax and run, whereas a cyclecar wasn't.

 

A formal decision by the Federation International des Clubs Moto Cycliste gave the cyclecar a classification. Small cyclecars could weigh no more than 331 lbs (150 kg) and have an engine no bigger than 0.75 L. Its tires could have a minimum 55 mm section. Large cyclecars could weigh up to 772 lbs (350 kg) and have a 1.1 L motor, and the smallest section tire it could wear was 60 mm.

 

To say the cyclecar was a hit is something of an understatement. Small manufacturers popped up all over the world to make teeny tiny cars to move the masses gently around their respective countries. They had twee names like Dudley Bug (only 100 made), the Wooler Mule, Izaro, and Cyklonetka. People could move around without getting covered in horse waste, and they weren't charged an arm and a leg for the privilege.

 

As time passed, bigger, more practical, faster cars became more affordable. Henry Ford saw how a cow went in one end of a meatpacking plant and came out the other in bits, then figured out how to mass-produce cars for a fraction of the cost of his competitors. All of a sudden, the cyclecar seemed a bit out of style. Why would consumers pay for a small, wobbly handmade thing from a shed when they could have a Ford Model T, an Austin 7, or a number of others? Almost as quickly as the cyclecar arrived, it was gone. Still, a decade in the limelight is better than nothing at all.

 

Of the many that hopped on the bandwagon at the time, few survived, and if they did, they aren't household names. The UK's Morgan Motor Company managed to make it through—were it not for HFS Morgan's desire to get around the Malvern Hills cheaply, his Three Wheeled Runabout might not have kick-started over a century of quirky British sports cars (a version of which you can still buy today in the Super Three—it's not particularly affordable or practical, but it is fun).

The conditions were ripe for a bubble

The cyclecar was gone but not forgotten. In the middle of the 20th century, war struck, and when it was over, things weren't rosy for those who remained. Economies were trashed, the populace had no money, and nations were rebuilding. The knock-on effect was that, once again, people needed a way to get around on a budget. Meet the darlings of the '50s and '60s: Bubblecars—so named because they looked like little bubbles as they went along.

 

Germany's efforts were many, and some manufacturers who had previously made airplanes found themselves with time on their hands after the war, which is why the Messerschmitt KR175 and Heinkel Kabine have potentially familiar names.

 

Messerschmitt's efforts were sweet things. "KR" stands for "Kebineroller," basically a scooter with a cabin, and the KR200, a follow-up to the cute but flawed KR175, took the world by storm in its own way. It was cheap, at 2,500 DM (about $6,972 today), and over 12,000 were built in its first year (over 41,000 were produced during its near-decade-long run). Its tiny 191 cc 10 hp (7.5 kW) motor gave it a top speed of 56 mph (90 km/h), and hydraulic shocks gave it an oddly smooth ride. Well, as smooth as a 507 lb (230 kg) tube on wheels could be.

 

BMW's front-opening Isetta is perhaps the most famous of the era, though it started out as an Iso instead of a BMW. The Iso Isetta came with a tiny 236 cc 9.5 hp (7 kW) motorcycle engine and looked like a happy little egg. It was made because Iso's owner thought it would be a good idea to get into the tiny car game after making refrigerators, tiny trucks, and scooters. It was a comparative hit in its home territory, but when the Fiat 500c came along and people lost interest in the tiny egg, Iso wanted to work on its Rivolta sports car, so the firm went looking for licensing deals. And boy, did it find them—versions of the Iso Isetta ended up in France, Argentina, Brazil, and Germany.

 

The German Isetta isn't a like-for-like copy of the original. BMW is a German company, so it decided to reengineer much of the vehicle—it had a new engine, for one thing—to make it a touch more palatable for its market. Over its life, it was revised, fettled, and even extended to add extra seats and wheels to make it a family car in the form of the BMW 600. There was a British version of the BMW Isetta, which swapped things around to make it more digestible for the tea-drinking market. The British car came in RHD three-wheeled flavor (other Isettas had two tiny rear wheels), which boosted its popularity thanks to legislation that meant a motorcycle license was all you needed to drive cars with the wrong number of wheels. Very, very technically, it was a motorcycle—just one with a steering wheel (see also: Reliant Robin).

 

While modern microcars aren't as big a deal as they were in the days of yore, you can still find plenty of them if you know where to look. Rural France is often a treasure trove of microcars, as is Spain, where you'll be dazzled by tiny Axiams, Ligiers (yes, the same Ligier as used to race in Formula 1), and more. A delve into their respective websites is a thing of joy. Modern microcars come with exciting wheels, spoilers, rear diffusers, touchscreens, decent sound systems, and other modern conveniences to ensure that your up-to-20-hp motor is as indistinguishable from a regular motor as possible… to a point. A quadricycle's crash regulations aren't as stringent as those for a "proper" car, and you won't find yourself cooing at material quality. For not much money, at age 16, who cares about plastic trim, anyway?

 

Cities like London and Paris can be good hunting grounds, too. Big cities tend to favor the likes of the electric Renault Twizy and Citroen Ami (the latter is a "light" quadricycle, which can irk other road users because it doesn't go exactly 30 mph).

 

The French pair are wonderful examples of the breed. The Twizy went out of production in 2023, but it enjoyed an 11-year run all over Europe. A bizarre pod with outboard wheels, it could seat two people in tandem (a cargo version ditched the rear chair in favor of a load bay), and it had doors that opened upward. It also didn't have windows as standard—there were optional plastic sheets that zipped up for rainy climes. It had over 50 miles (80 km) of range, which made it perfect for city living.

 

Citroen's Ami is still available for sale and is a masterwork of clever cheapness. Its front and rear panels are the same, but with clear lamps up front and red lights at the back. Its doors are the same, too, which means one opens the normal way and the other in "suicide" style. It comes with a USB port for charging your phone, a very noisy fan to demist the windshield, and a hole in the dash to pop a Bluetooth speaker in if you want to listen to something other than the road as you drive. Retailing at less than $11,500 in the UK, it's not cheaper than a used hatchback, but its tiny 5.5 kWh battery will take you to the store and back, so long as there aren't any hills and the supermarket isn't more than a 40-mile (64 km) round trip. The Ami is unrelentingly cool in a cheap and cheerful sense, and it brings joy to anyone who sees one.

 

From the dawn of the motorcar to today, the microcar has fought on. Whether as a cheap means of getting the masses around or as a way to village-hop in rural France, there always seems to be a place for them, even though they're not as popular as they used to be. More microcars will come in time—electrification allows for small batteries with decent punch, and more relaxed crash regulations mean design teams can stretch their fingers in more interesting ways.

 

And, hey, who doesn't like the idea of whirring around town at 28 mph in a tiny plastic Citroen box? Last-mile transportation could still be fun enough to take the long way home.

 

 

In early August 2023, a beekeeper near the port of Savannah, Georgia, noticed some odd activity around his hives. Something was hunting his honeybees. It was a flying insect bigger than a yellowjacket, mostly black with bright yellow legs. The creature would hover at the hive entrance, capture a honeybee in flight, and butcher it before darting off with the bee’s thorax, the meatiest bit.

 

“He’d only been keeping bees since March… but he knew enough to know that something wasn’t right with this thing,” says Lewis Bartlett, an evolutionary ecologist and honeybee expert at the University of Georgia, who helped to investigate. Bartlett had seen these honeybee hunters before, during his PhD studies in England a decade earlier. The dreaded yellow-legged hornet had arrived in North America.

 

With origins in Afghanistan, eastern China, and Indonesia, the yellow-legged hornet, Vespa velutina, has expanded during the last two decades into South Korea, Japan, and Europe. When the hornet invades new territory, it preys on honeybees, bumblebees, and other vulnerable insects. One yellow-legged hornet can kill up to dozens of honeybees in a single day. It can decimate colonies through intimidation by deterring honeybees from foraging. “They’re not to be messed with,” says honeybee researcher Gard Otis, professor emeritus at the University of Guelph in Canada.

 

The yellow-legged hornet is so destructive that it was the first insect to land on the European Union’s blacklist of invasive species. In Portugal, honey production in some regions of the country has slumped by more than 35 percent since the hornet’s arrival. French beekeepers have reported 30 percent to 80 percent of honeybee colonies exterminated in some locales, costing the French economy an estimated $33 million annually.

 

 

All that destruction may be linked to a single, multi-mated queen that arrived at the port of Bordeaux, France, in a shipment of bonsai pots from China before 2004. During her first spring, she established a nest, reared workers, and laid eggs. By fall, hundreds of new mated queens likely exited and found overwintering sites, restarting the cycle in the spring. The hornet’s fortitude—it is the Diana Nyad of invasive social wasps—allowed it to surge across France’s borders into Spain, Portugal, Italy, Belgium, Germany, the Netherlands, the United Kingdom, and Switzerland in only two decades, hurtling onward by as much as 100 kilometers a year.

Suspected stowaway

As the hornet fanned out across Europe, scientists in North America wondered when it might arrive on their side of the Atlantic. Queens sometimes overwinter in crates and containers, allowing them to stow away on ships and be transported long distances. In 2013, researchers cautioned that a yellow-legged hornet invasion at any one point along the US East Coast would have the potential to spread across the country.

 

After the first sighting last summer, Georgia’s agricultural commissioner urged people to report hornets and nests, and warned that the yellow-legged hornet could threaten the state’s $73 billion agriculture industry. American farmers grow more than 100 different crops, including apples, blueberries, and watermelons, that depend on pollinators. Georgia mass-produces honeybees and ships them north to jumpstart spring crops, like Maine blueberries, before local pollinators have awakened.

 

 

Less than two weeks after the first hornet was spotted, scientists found a nest in a tree 25 meters off the ground. In a night operation, while the hornets idled, a tree surgeon climbed to the nest, sprayed it with insecticide, and cut it down. Just a quarter of the full nest was the size of a human torso, and the Georgia Department of Agriculture displayed a chunk, still wrapped around the branch, at a press conference—warning that this was larger than those seen in Europe.

 

“Savannah, Georgia, is primo climate for these guys,” says Otis. It’s a lush, subtropical paradise, giving the insect a long growing season—and a rich hunting ground.

 

For the next several months, Bartlett helped the state agricultural researchers set traps and follow individual hornets to find other nests. By the end of 2023, they’d removed four more. “We think we’ve discovered them at a very early stage, which is why pursuing eradication is very, very plausible,” Bartlett said in November. If not, Georgia and its neighbors could get caught in an endless—and costly—game of whack-a-mole.

Social wasps: Invasive global predators

The yellow-legged hornet and other social wasps, like the common yellowjacket, the German yellowjacket, and the western yellowjacket, have successfully invaded every continent except Antarctica. They’ve been introduced to new areas by global trade, sometimes more than once over several decades.

 

The hornets live in colonies of individuals organized into groups that divvy up the labor of reproduction, foraging, and caregiving. These behaviors, and the insects’ nearly omnivorous appetites, make them among the most successful invaders of new habitats and fiercest aggressors of native fauna. In their endemic ranges, these wasps are eaten by skunks, squirrels, bears, or snagged in flight by kingbirds and tanagers, or attacked by other predatory wasps. But in the absence of predators, their toll can be enormous.

 

In New Zealand’s Nelson Lakes National Park, the beech forests are thick with invasive yellowjackets by early autumn. They sip the sugary secretions of scale insects living on the trees, and will fight the bellbirds, tui, silvereyes, and other birds for it, even slaughtering nest-bound chicks. The densities of the yellowjacket nests—up to 40 nests per hectare and 370 wasps per square meter of tree trunk—are among the world’s highest.

 

“When you walk through the forest, you should smell the sweetness of the honeydew and hear the birds,” says invasive species biologist Phil Lester of Te Herenga Waka—Victoria University of Wellington, coauthor of a review of management strategies for invasive social wasps in the 2019 Annual Review of Entomology. “But with the wasp, you don’t hear the birdsong, you don’t smell the honeydew.”

 

In Hawaii, the western yellowjacket has had dramatic impacts on the island ecosystem. Genetic studies show that the original population came from the Pacific Northwest or Northern California, possibly in a shipment of Christmas trees. It hunts native bees and drains the nectar from the wispy red flowers of the ‘ōhi’a lehua tree, stealing food from other pollinators and curtailing seed production.

 

“They eat everything,” says ecologist Erin Wilson-Rankin of the University of California, Riverside, who has been studying invasive social wasps for nearly 20 years. “They don’t specialize. They’ll eat caterpillars, aphids, flies, the whole gamut of arthropods.”

Controversial tools

People have tried just about everything to get rid of wasps: fire, boiling water, electricity, traps, poison, and brute force. While many poisons do work, they can also harm native insects and other animals. New Zealand has suppressed yellowjacket populations in highly trafficked areas with a selective poison bait called Vespex, but they reinvade elsewhere.

 

Nest destruction can kill hundreds of wasps at once, but it’s dangerous: Yellowjackets can squirt venom into an attacker’s eyes, and stings can be painful or life-threatening. Reiner Jahn, a hornet-buster and research assistant for a local landscape conservation association in Germany, describes the pain of a yellow-legged hornet sting as “digging a hot rusty knife into your flesh.”

 

 

Another approach to managing invasive species is biological control: A different species, often a natural enemy, is transplanted into the ecosystem to take on the role of contract killer. It can do the trick, but the long history of this strategy going awry (think harlequin ladybirds, cannibal snails, small Asian mongoose, cane toads) gives pause.

 

Cajoling foreign predators to take root in new places is another bother. In New Zealand, for example, the government recently approved the release of a non-native hoverfly and beetle to target invasive wasps. In Europe, both species hitch a ride into the hornet nests, feasting on the juvenile hornet grubs and decimating the next generation. But the imported insect predators had to have their seasonal cycles flipped before they could be released in the Southern Hemisphere. After some setbacks, scientists released about 20 hoverflies into the wild on the northern end of the South Island in mid-May.

 

Lester has other ideas: Silencing some of the wasps’ essential genes could reverse their spread. A handful of genetic control technologies are being tested globally to target invasive or harmful insects. For example, the biotechnology company Oxitec aims to combat the spread of dengue and other mosquito-borne diseases by releasing gene-edited male mosquitoes that produce female offspring that die young. (It’s the females that bite and spread disease.) Other researchers are using CRISPR gene editing on a range of agricultural pests to reduce pesticide use and save crops.

 

In 2020, an international group of researchers, including Lester and Wilson-Rankin, sequenced the genomes of three invasive social wasps: the common yellowjacket, the German yellowjacket, and the western yellowjacket. Lester then zeroed in on a gene called ocnus that’s involved in sperm development, with the goal of making sterile males.

 

Like many insect pests, common yellowjackets are haplodiploid, which means that fertilized eggs become female wasps (with two copies of each chromosome), and unfertilized eggs produce males (with only one copy of each chromosome). If a queen mates with a sterile male, the eggs laid would produce only male wasps. Without female worker wasps, the nest would fail. But Lester’s modeling has shown that it would take decades for the mutation to spread across the South Island wasp population. So he continues to look for new genetic targets that might snuff out New Zealand’s invasive wasps more quickly.

 

Many people are unsettled by the idea of releasing genetically modified organisms into the wild, even if it’s to save native species, but the approach carries advantages. The impact would be precise; it wouldn’t poison other animals or insects. It would disperse over large distances and into remote areas. It would also be self-perpetuating, so people wouldn’t have to climb long ladders in protective suits to cut down enormous nests full of angry wasps.

Nest busting

On a hot afternoon in mid-September, Jahn, the German hornet-buster, pulls up to the Metropolitan International School in Viernheim, an industrial town east of the Rhine River. Kids run and jump in the playground, until a teacher ushers them away. High in a tree overhanging the soccer field is a caramel-colored, beach-ball-sized yellow-legged hornet’s nest.

 

“The kids can’t play soccer. I had to close the field because it is too dangerous,” says Oliver Wagner, the school’s facility manager.

 

A whiff of revenge hangs in the air as Jahn and his crew set up. Each is a beekeeper who has lost colonies to yellow-legged hornets or knows someone who has. Jahn extends a telescopic pole fitted with a spray nozzle into the branches. He jabs the nest and blows in a fine powder called diatomaceous earth as chunks of the nest tumble to the ground. Hornets stream out like the air escaping from a punctured balloon.

 

Dusted with the white powder and unable to fly, the inch-long yellow-legged hornets wander through the grass and across the tarp. The crew picks through the nest debris and they tweeze the larger hornets into specimen bottles. When a nest is attacked—whether by a predator or a human—the queen may try to escape, Jahn explains. Find her, and the work is done. This time, she’s unaccounted for.

 

The trick to stopping a yellow-legged hornet invasion is to find the nests and destroy them before hundreds of new queens fly out in the fall to establish their own nests. European Union member states must, by law, control the hornet’s spread, but Germany has strict rules that protect pollinator and native insects and limit what beekeepers and hornet-busters can do. Diatomaceous earth, often used in homes to kill cockroaches and centipedes, has become Jahn’s go-to solution. It sticks to the hornet’s exoskeletons and dries them out but doesn’t spread to other insects.

 

In all of 2023, Jahn destroyed 160 yellow-legged hornet’s nests in his home state of Hesse and 80 in a neighboring state, most brought to his attention by beekeepers. After a few years of nest-busting, he’s given up beekeeping (there’s no more time), and he no longer believes that the yellow-legged hornet can be eradicated in Germany—the country may have waited too long to start removing nests. Still, he says, “it’s easier to do something now than wait until next year.” But by mid-May this year, he’d already fielded calls for 19 new nests, compared with only two by late May last year.

 

Back in Georgia, Bartlett has tracked down the source of the captured yellow-legged hornets. His genetic analysis shows that a single queen arrived from southern China, the Korean peninsula, or Japan in late 2022. He believes the hornets captured last year were the first American-born generation founded by the stowaway queen. Now, the second generation has emerged. “We have been finding queens a little further out than we had hoped. But nothing near the distances they see in Europe,” says Bartlett. As of the end of April, the state had trapped and destroyed 21 queens.

 

Bartlett sees the work as his duty to protect the beekeeping industry, but his hope is that the hornet won’t define his scientific career. Still, he knows he can’t relent. “If we don’t get rid of them, there is very little chance that I’m not going to become the yellow-legged hornet expert in the US.”

 

Hannah Hoag is a science journalist in Toronto. She has written for the Atlantic, Biographic, Nature, Science, the Globe and Mail, the New York Times, and more.

 

When the Brazilian nutritional scientist Carlos Monteiro coined the term “ultra-processed foods” 15 years ago, he established what he calls a “new paradigm” for assessing the impact of diet on health.

 

Monteiro had noticed that although Brazilian households were spending less on sugar and oil, obesity rates were going up. The paradox could be explained by increased consumption of food that had undergone high levels of processing, such as the addition of preservatives and flavorings or the removal or addition of nutrients.

 

But health authorities and food companies resisted the link, Monteiro tells the FT. “[These are] people who spent their whole life thinking that the only link between diet and health is the nutrient content of foods ... Food is more than nutrients.”

 

Monteiro’s food classification system, “Nova,” assessed not only the nutritional content of foods but also the processes they undergo before reaching our plates. The system laid the groundwork for two decades of scientific research linking the consumption of UPFs to obesity, cancer, and diabetes.

 

Studies of UPFs show that these processes create food—from snack bars to breakfast cereals to ready meals—that encourages overeating but may leave the eater undernourished. A recipe might, for example, contain a level of carbohydrate and fat that triggers the brain’s reward system, meaning you have to consume more to sustain the pleasure of eating it.

 

In 2019, American metabolic scientist Kevin Hall carried out a randomized study comparing people who ate an unprocessed diet with those who followed a UPF diet over two weeks. Hall found that the subjects who ate the ultra-processed diet consumed around 500 more calories per day, more fat and carbohydrates, less protein—and gained weight.

 

 

The rising concern about the health impact of UPFs has recast the debate around food and public health, giving rise to books, policy campaigns, and academic papers. It also presents the most concrete challenge yet to the business model of the food industry, for whom UPFs are extremely profitable.

 

The industry has responded with a ferocious campaign against regulation. In part it has used the same lobbying playbook as its fight against labeling and taxation of “junk food” high in calories: big spending to influence policymakers.

 

FT analysis of US lobbying data from non-profit Open Secrets found that food and soft drinks-related companies spent $106 million on lobbying in 2023, almost twice as much as the tobacco and alcohol industries combined. Last year’s spend was 21 percent higher than in 2020, with the increase driven largely by lobbying relating to food processing as well as sugar.

 

In an echo of tactics employed by cigarette companies, the food industry has also attempted to stave off regulation by casting doubt on the research of scientists like Monteiro.

 

“The strategy I see the food industry using is deny, denounce, and delay,” says Barry Smith, director of the Institute of Philosophy at the University of London and a consultant for companies on the multisensory experience of food and drink.

 

So far the strategy has proved successful. Just a handful of countries, including Belgium, Israel, and Brazil, currently refer to UPFs in their dietary guidelines. But as the weight of evidence about UPFs grows, public health experts say the only question now is how, if at all, it is translated into regulation.

 

“There’s scientific agreement on the science,” says Jean Adams, professor of dietary public health at the MRC Epidemiology Unit at the University of Cambridge. “It’s how to interpret that to make a policy that people aren’t sure of.”

 

The food industry, dominated by global conglomerates such as Nestlé, PepsiCo, Mars, and Kraft Heinz, likes to project itself as committed to public health. “Our strategy is all about nutrition, health, and wellness,” Paul Bulcke, the chair of Nestlé, told investors at the company’s annual meeting in April.

 

Innovations in processing over the 20th century not only made food more affordable and accessible, the industry’s advocates note, but also created beneficial products like sugar-free sweeteners and protein-enriched milk.

 

Food processing has allowed the reformulation of recipes to add whole grains and fiber to food while reducing sugar, salt, and saturated fat, said Nestlé in a statement. “We should not lose sight of the vital role it plays in providing safe, nutritious, high-quality, and affordable products all over the world.”

 

In a statement, PepsiCo said it aimed to “improve the core nutritional profile of our products” and use more diverse ingredients in order to “meet many dietary needs and preferences.” Kraft Heinz did not respond to request for comment.

 

Yet, as researchers have learned more about the link between UPFs and poor health outcomes, companies have remained largely silent about these risks, leaving trade bodies that advocate on their behalf to argue loudly against the validity of the research.

 

The UK’s Food & Drink Federation argues there is no legal definition of processed or ultra-processed food and that consumers struggle to understand the difference. A spokesperson said: “Our concern about the concept of ultra-processed food is that it’s not linked to current government dietary guidance nor food safety regulations, which are underpinned by rigorous science and assessed by expert, independent committees.”

 

David Chavern, president of the US’s Consumer Brands Association, says food companies were “trying to bring rationality to the debate.” The research, Chavern says, has an “anti-corporate wrapper around it” and creates a false sense that companies are hiding something from consumers. “The industry views itself as incredibly transparent. There is extensive disclosure about ingredients on packaging,” he adds.

 

For decades, the industry has been quietly pouring money into the world’s leading food and nutritional sciences departments.

 

Researchers studying human nutrition at the UK’s Reading University, for example, received 262,832 pounds in research funding from food giant Mars between 2018 and 2023, according to a recent freedom of information request. PepsiCo provided 61,756 pounds to the researchers over the same period.

 

“We work with the food industry so that we can do more research that has an immediate impact on people’s diet and health,” says Robert Van de Noort, vice-chancellor of the University of Reading. “We want our work to be on the shelves of the supermarket, not just the library.” A Mars spokesperson said the funding was to support a diverse range of science projects, largely supporting Reading’s facility for cocoa research, while PepsiCo said it had funded research into shelf life and product quality, among other things.

 

Anna Gilmore, co-director of the Centre for 21st Century Public Health at the University of Bath, say the ties with scientists help the industry to “manufacture doubt” by funding analysis that exonerates companies or suggests the case against them is not proven.

 

A 2018 review of studies that criticized Monteiro’s Nova system found that the authors overwhelmingly had connections to the UPF industry.

 

Scientists at the US government’s Agricultural Research Service led another study demonstrating it is possible to build a healthy diet with 91 percent of the calories coming from UPFs. The authors had connections to the soy industry, sauces and flavoring company McCormick, and Atkins diet food owner Simply Good Foods.

 

Regulatory bodies also have some of these corporate links. A review of conflicts of interest in UK food regulation found that nine of the 15 members of the government’s Scientific Advisory Committee on Nutrition had received funding from the UPF industry.

 

The SACN concluded last summer that there were “uncertainties around the quality of evidence available” on UPFs, as studies were mainly observational, and that “confounding” factors like energy intake, body mass index, smoking, and socioeconomic status may not have been factored in.

 

“The resistance and counter-attacks are not a surprise,” says Tim Lang, a professor at City University’s Centre for Food Policy who co-led the review. An “epidemiological transition” driven by the food industry has been taking place for decades, he says, in which countries switch away from simple, whole-food diets as they become wealthier.

 

“It is what the food industry has celebrated and trumpeted. And now they are hoist by their own petard. All the things they claimed as success are now flaws,” he says.

 

It is a similar story in the US, where the Kevin Hall study prompted the US government to review the effects of UPFs on public health for potential inclusion in the next round of national dietary guidelines. A lobbyist at a prominent food trade group in Washington tells the FT that keeping UPFs out of these guidelines is the group’s key objective.

 

Current advice in the US is based on individual nutrients, which means companies can formulate foods to meet requirements. Food served in the subsidized US National School Lunch Program, for example, includes processed foods like Kraft Heinz’s Lunchables and PepsiCo’s Walking Tacos.

 

 

“My guess is they won’t say there is strong evidence,” says Aviva Musicus, science director at the US Center for Science in the Public Interest, a nonprofit health-advocacy group. Because the studies used the Nova classification and are based on prospective cohort studies, she notes, they do not show definitive causality.

 

In addition, nine out of 20 members of the Dietary Guidelines Advisory Committee advising on the guidelines had conflicts of interest with food, pharmaceutical, or weight-loss companies or industry groups, according to another non-profit, US Right To Know.

 

One panel member, Fatima Cody Stanford, received tens of thousands of dollars in consulting fees in 2022 from manufacturers of obesity drugs, including Novo Nordisk and Eli Lilly, according to public disclosures.

 

Stanford did not respond to a request for comment, but the US Office of Disease Prevention and Health Promotion noted that all candidates for the panel are vetted thoroughly for conflicts of interest.

 

In 2023, PepsiCo spent millions of dollars lobbying the US government. According to one disclosure from last July, the Doritos and Tostitos maker spent $1.27 million on Supplemental Nutrition Assistance Program (SNAP) purchasing restrictions, the upcoming dietary guidelines, sweeteners, and food labeling, among other issues.

 

Where legislation is passed that has a direct impact on food multinationals, they have often fought back in the courts. In Mexico, companies including Kellogg’s and Nestlé have sued the government over the introduction of front-of-package warning labels and other restrictions including the use of children’s characters in marketing.

 

The labels—black octagons that warn about excess of sugars, sodium, trans fat, saturated fat, and calories in products—were rolled out in 2020. A handful of the lawsuits were accepted by the Mexican Supreme Court and are still being fought.

 

Nestlé said it “supported front-of-pack labeling that helps consumers make informed choices” including government-endorsed labels such as Nutri-Score in some European countries or the traffic light system in the UK.

 

The industry has also successfully framed the issue as one of personal choice. In Brazil, where legislators are considering the inclusion of UPFs in a group of products that would attract a higher excise tax, the industry has argued that regulation could limit consumer options and make food more expensive.

 

It’s a resonant charge in a country where hunger is a major issue, says Paula Johns, co-founder and director of public health advocate ACT Promoção da Saúde (ACT Health Promotion).

 

Manufacturers also argue that the harm caused by their products is a result of a lack of personal willpower or failure to exercise, says Bath University’s Gilmore, and “nothing to do with industry or its UPF products that overwhelm our internal systems that regulate appetite.”

 

Even in public health circles there is little agreement about what shape regulation of UPFs should take.

 

The chief concern among public health experts is that guidance telling people to avoid UPFs risks stigmatizing those reliant on packaged food due to socio-economic circumstances. Some also share the industry’s criticism that the definition is too broad to draw a clear causal line between exposure to UPFs and their effects.

 

The evidence on UPFs points to a “staggering range of health outcomes,” says Cambridge’s Adams. “If you’re going to ask people to eat more minimally processed food, you need to ask how do we support people to cook at home themselves.”

 

“Ultra-processed” is not necessarily a useful term for regulation that requires “real nuance,” said Chris van Tulleken, author of the book Ultra-Processed People, at a recent House of Lords committee hearing. Existing legislation targeting high-fat, sugar, or salt (HFSS) foods would also ensnare many UPFs, he noted.

 

But it was “a very powerful way of describing our terrible diet,” he added. “We have great evidence that there is a single pattern of diet that drives harm—and it is an industrialized, American diet produced by transnational food corporations.”

 

It is in the food industry’s interest to conflate UPFs and HFSS foods, says University of London’s Smith. “We have a definition that works, [they say]—HFSS... That way they can reduce and reformulate, but still use all the UPF ingredients.”

 

What the food companies really fear, he says, is that UPF ingredients such as additives and stabilizers will be targeted for labeling. “Once we start interfering with those, they will be prevented from making food that makes us eat more,” he adds.

 

It is clear that the public is now much more aware of UPFs, and concerned about them. Two-thirds of Europeans now believe that ultra-processed foods are unhealthy and will cause health problems in later life, according to a February survey of 10,000 people in 17 countries, and 40 percent do not trust that the authorities are regulating them well enough. Research by Mintel in the UK has found that 70 percent of UK adults try to avoid ultra-processed foods.

 

“I don’t think even Carlos Monteiro in his wildest dreams expected the public discourse to get so attuned,” says Lang at City University. “The public is running with it. The genie is out of the bottle.”

 

Corporate affairs professionals say food companies are overwhelmed by the level of public concern about UPFs and are scrambling to find an effective way to counter it.

 

“The industry needs to at least try and find a common position it can defend itself from,” says a former food and drinks lobbyist in the UK. “If it doesn’t, this kind of emotional reaction is likely to gather ground, at which point it will be very difficult to stop all of this.”

 

Welcome to Edition 6.45 of the Rocket Report! The most interesting news in launch this week, to me, is that Firefly is potentially up for sale. That makes two of the handful of US companies with operational rockets, Firefly and United Launch Alliance, actively on offer. I'll be fascinated to see what the valuations of each end up being if/when sales go through.

 

As always, we welcome reader submissions, and if you don't want to miss an issue, please subscribe using the box below (the form will not appear on AMP-enabled versions of the site). Each report will include information on small-, medium-, and heavy-lift rockets as well as a quick look ahead at the next three launches on the calendar.

 

 

Firefly may be up for sale. Firefly Aerospace investors are considering a sale that could value the closely held rocket and Moon lander maker at about $1.5 billion, Bloomberg reports. The rocket company's primary owner, AE Industrial Partners, is working with an adviser on "strategic options" for Firefly. Neither AE nor Firefly commented to Bloomberg about the potential sale. AE invested $75 million into Texas-based Firefly as part of a series B financing round in 2022. The firm made a subsequent investment in its Series C round in November 2023.

 

Launches and landers ... Now more than a decade old and with a history of financial struggles, Firefly has emerged as one of the apparent winners in the small launch race in the United States. The company's Alpha rocket has now launched four times since its unsuccessful debut in September 2021, and it is due to fly a Venture Class Launch Services 2 mission for NASA in the coming weeks. Firefly also aims to launch its Blue Ghost spacecraft to the moon later this year and is working on an orbital transfer vehicle.

 

Blue Origin makes successful return to flight. With retired Air Force captain and test pilot Ed Dwight as the headline passenger, Blue Origin's New Shepard spacecraft returned to flight on Sunday morning. An African American, Dwight was one of 26 pilots the Air Force recommended to NASA for the third class of astronauts in 1963, but the agency didn't select him. It took another 20 years for America's first Black astronaut, Guion Bluford, to fly in space in 1983. At the age of 90, Dwight finally entered the record books Sunday, becoming the oldest person to reach space. “I thought I didn’t need it in my life," Dwight said after Sunday's fight. "But I lied!"

 

One chute down ... This was the seventh time Blue Origin, the space company owned by billionaire Jeff Bezos, has flown people to suborbital space, and the 25th flight overall of the company's fleet of New Shepard rockets. It was the first time Blue Origin had launched people in nearly two years, resuming suborbital service after a rocket failure on an uncrewed research flight in September 2022. In December, Blue Origin launched another uncrewed suborbital research mission to set the stage for the resumption of human missions Sunday. There was one issue with the flight, as only two of the capsule's three parachutes deployed. It's unclear how long it will take to address this problem.

 

RFA tests first stage of its rocket. German launch startup Rocket Factory Augsburg announced Sunday that it had begun the hot-fire campaign for the first stage of its RFA One rocket. "We hot-fired a total of four Helix engines, igniting one by one at four-second intervals," the company said on the social media site X. "All engines ran simultaneously for 8 seconds with a total hot-fire duration of 20 seconds. The test ran flawlessly through start-up, steady-state, and shutdown." It's a great step forward for the launch company.

 

Targeting a test flight this year, but ... The test occurred at the SaxaVord Spaceport in the United Kingdom. The RFA One vehicle is powered by nine Helix engines and will have a payload capacity of 1.6 metric tons to low-Earth orbit. The company is targeting a debut launch later this year, but I'm fairly skeptical of that. By way of comparison, SpaceX began test firing its Falcon 9 first stage in 2008, with a full-duration test firing of all nine engines in November of that year. But the rocket did not make its debut flight until June 2010.

 

China expanding commercial spaceport. China is planning new phases of expansion for its new commercial spaceport to support an expected surge in launch and commercial space activity, Space News reports. Construction of the second of two launch pads at Hainan Commercial Launch Site could be completed by the end of May. The first, completed in December and dedicated to the Long March 8 rocket, could host its first launch before the end of June.

 

Fulfilling a mega-need ... However this appears to be just the beginning, as the spaceport could have a total of 10 pads serving both liquid and solid rockets. The reason for the dramatic expansion appears to be increasing access to space and allowing China to achieve a launch rate needed to build a pair of low-Earth orbit megaconstellations, each over 10,000 satellites strong. It is also a further sign of China’s commitment to establishing a thriving commercial space sector. (submitted by Ken the Bin)

 

A spaceport in the Dominican Republic? The Dominican Republic has kicked off a study into the feasibility of setting up a commercial spaceport near the equator, Space News reports. Florida-based Launch on Demand, a launch licensing and technical services specialist, announced a contract May 22 to lead the six-month study with the country's National Intelligence Directorate. The study will focus on the Oviedo area to the south of the island nation.

 

Go south, young man ... The study will examine vertical lift capacity for orbital and suborbital missions, which could enable the country to improve critical satellite monitoring and surveillance capabilities to address illegal migration, drug trafficking, and environmental damage. The Caribbean nation is located at a latitude between 17 and 20 degrees, significantly to the south of Cape Canaveral, at 28.4 degrees.

 

 

Starliner launch delayed until early June. On Wednesday evening, NASA said the earliest Starliner launch opportunity will be Saturday, June 1, with additional launch opportunities June 2, June 5, and June 6. But there's still work to do before NASA gives the green light for the Starliner launch, which has been struggling with a stubborn leak of helium from the capsule's propulsion system, Ars reports. The mission will be the first crewed flight by Starliner and is due to launch on an Atlas V rocket.

 

It's always the valves ... "Work continues to assess Starliner performance and redundancy following the discovery of a small helium leak in the spacecraft’s service module," NASA said late Wednesday. "As part of this work, and unrelated to the current leak, which remains stable, teams are in the process of completing a follow-on propulsion system assessment to understand potential helium system impacts on some Starliner return scenarios." A separate problem with a pressure regulation valve on the spacecraft's United Launch Alliance Atlas V rocket prompted officials to scrub a launch attempt on May 6. (submitted by Ken the Bin)

 

Ariane 6 debut timeline set. The Ariane 6 Launcher Task Force published a new update this week that provided some key status updates on the vehicle's path toward launch, European Spaceflight reports. The most recent milestone on the road to the Ariane 6 debut was reached on May 16, when the flight’s various payloads arrived at the Guiana Space Centre, ready for integration. The rocket’s completed upper composite, consisting of the launcher adapter, the payloads, and the fairing, will be completed and ready for transportation to the launch pad for stacking in June.

 

Flying high in July? ... On June 18, the Ariane 6 rocket's wet dress rehearsal will be conducted. This process involves filling and draining both the core and upper stage propellant tanks and is the last major milestone before the rocket’s first launch attempt. According to this week's update, the date for the debut flight has been narrowed down to sometime in the first two weeks of July. A tentative date for the first launch attempt will be announced at the ILA airshow in Berlin during the week of June 5. (submitted by EllPeaTea and Ken the Bin)

 

Falcon 9 launches next-gen spysats. A SpaceX Falcon 9 rocket on Wednesday launched an undisclosed number of small spacecraft into orbit for the National Reconnaissance Office, Space News reports. The classified mission, designated NROL-146, was SpaceX’s 52nd launch of the year and the Falcon 9’s fifth launch for the NRO. The Falcon 9 lifted off from Vandenberg Space Force Base, California.

 

Sorry, no details ... NROL-146 is the agency’s first deployment of a new imaging satellite constellation built by SpaceX and Northrop Grumman. The NRO has not disclosed how many satellites were launched on this mission or the projected size of the new constellation. Agency officials previously said six launches are planned in 2024 for the NRO’s proliferated architecture of small satellites. (submitted by Ken the Bin)

 

 

Fourth test flight focused on Starship reentry. After three test flights, SpaceX has shown that the world's most powerful rocket can reach space. Now, engineers must demonstrate the company's next-generation Starship vehicle can get back home. This will be the central objective for the fourth Starship test flight, which could happen as soon as early June, Ars reports. Based on some of the latest road closure requests, it's possible SpaceX may attempt to launch Starship on June 1.

 

Some like it hot ... For this upcoming flight, SpaceX officials would like to see the Super Heavy booster for the next test flight, named Booster 11, make a controlled pinpoint splashdown in the Gulf of Mexico just offshore from Starbase. Halfway around the world, the Starship upper stage, known as Ship 29, will try to survive the blistering reentry back into Earth's atmosphere. Starship is dressed in about 18,000 hexagonal heat-absorbing ceramic tiles to protect its stainless-steel structure during reentry, when temperatures peak at about 2,600° Fahrenheit (1,430° Celsius). SpaceX is eager to collect data on how these tiles will perform and what condition they will be in for re-flights.

 

SpaceX Florida plans for Starship become clearer. Recently, both the US Federal Aviation Administration and the US Space Force have initiated "Environmental Impact Statement" projects for launch pads in Florida, LC-39A and SLC-37, for potential use as Starship launch facilities. On the one hand, these environmental reviews often take a while and could cloud Elon Musk's goal of having Starship launch sites in Florida ready for service by the end of 2025, Ars reports. "A couple of years would not be a surprise," said George Nield, an aerospace industry consultant and former head of the FAA's Office of Commercial Space Transportation.

 

Much work to do ... Another way to look at the recent FAA and Space Force announcements of pending environmental reviews is that SpaceX finally appears to be cementing its plans to launch Starship from Florida. These plans have changed quite a bit in the last five years. The result of these reviews will help SpaceX finalize its plans to launch Starship vehicles from both its existing facility in South Texas and one or more launch pads in Florida. In a long feature, Ars explores the work SpaceX has yet to do to deliver on its Starship project—and the role Florida will play in that.

Next three launches

May 24: Falcon 9 | Starlink 6-63 | Kennedy Space Center, Florida | 02:13 UTC

May 25: Electron | Ready, Aim, PREFIRE | Mahia Peninsula, New Zealand | 07:15 UTC

May 27: Falcon 9 | Starlink 6-60 | Cape Canaveral Space Force Station, Florida | 11:30 UTC

 

 

 

Good morning. It's May 24, and today's photo comes from the European Space Agency's new Euclid space telescope.

 

Launched in July 2023, the mission is intended to create a giant map of the Universe, across more than one-third of the nighttime sky. Its big-ticket goal is to help scientists better understand the nature of dark matter and dark energy, which account for the vast majority of the mass in the Universe—but about which we know almost nothing.

 

On Thursday the mission's operators released five images, each of which was taken shortly after the instrument's launch. The image in this post features the Messier 78 object, a star nursery wrapped in interstellar gas some 1,300 light-years from Earth.

 

According to the European scientists, "Euclid peered deep into this nursery using its infrared camera, exposing hidden regions of star formation for the first time, mapping its complex filaments of gas and dust in unprecedented detail, and uncovering newly formed stars and planets. Euclid’s instruments can detect objects just a few times the mass of Jupiter, and its infrared ‘eyes’ reveal over 300,000 new objects in this field of view alone."

 

It's fabulous.

 

Source: ESA/Euclid/Euclid Consortium/NASA, image processing by J.-C. Cuillandre (CEA Paris-Saclay), G. Anselmi

 

Commercialization has not dealt kindly with the Mayan Food of the Gods. Modern chocolate products are filled with sugar and calories, contributing to the obesity epidemic in the West. And the cocoa crop is hardly in great shape; climate change is decreasing production, causing prices to rise; farmers in West Africa have responded by clear-cutting rainforests to plant more cocoa plants. However, researchers at ETH Zurich may have found a path to start addressing both problems, making chocolate that has less sugar and calories and makes more efficient use of the cocoa crop. The Swiss perfected chocolate-making over 200 years ago, so if they say the chocolate is good, it is.

 

Chocolate is traditionally made by mixing dried, roasted, and ground fermented cocoa beans to make cocoa mass. The cocoa mass is then mixed with refined sugar, usually from sugar beets. Instead of sugar, this new Swiss whole fruit chocolate uses the pulp surrounding the cocoa beans along with the inner rind of the cocoa pod husk to make a cocoa gel. When mixed with cocoa mass, this produces chocolate that is higher in fiber and lower in saturated fat than conventional chocolate.

 

The “whole fruit” on its label is certainly more appealing than the air or fish oil that has previously been substituted for cocoa butter to reduce the saturated fat content of chocolate confections. (Extra cocoa butter, or fat isolated from the cocoa bean, is sometimes added to cocoa mass to make the end product smoother and waxier.) The pulp and the cocoa pods are generally discarded, so upcycling them instead of tossing them could reduce the land use impact and global warming potential of cocoa cultivation.

 

Eleven middle-aged women were given four opportunities to taste the chocolate and compare it to the standard formulation. Kim Mishra, the lead author of the study, wrote in an email that “the wholefruit chocolate was rated less sweet than conventional chocolate at the same absolute sugar level. We estimate a sweetening power of the gel between 0.51 and 0.99 of powdered sugar. However, the gel also gave distinct fruity notes and notes of dried fruits.”

 

As he also noted: “The snap and gloss were identical to the conventional chocolate,” which is really important in evaluating and judging the finished product. But “the sweetening gels were perceived as more coarse on the tongue compared to sugar since particle size was bigger.” This also explains why the chocolate made with the gel was relatively less sweet; since the particles are bigger, there are fewer of them and therefore less particle surface area for sugar to dissolve from. Moreover, the sugar in the gel is not crystalline like it is in refined sugar, so it doesn’t dissolve as quickly in saliva.

 

Still, Mishra wrote, he is “confident that we can get there with some more research.”

 

Nature Food, 2024.  DOI: 10.1038/s43016-024-00967-2

 

A team of physicists has discovered that it’s possible to build a real, actual, physical warp drive and not break any known rules of physics. One caveat: the vessel doing the warping can’t exceed the speed of light, so you’re not going to get anywhere interesting any time soon. But this research still represents an important advance in our understanding of gravity.

Moving without motion

Einstein’s general theory of relativity is a toolkit for solving problems involving gravity that connects mass and energy with deformations in spacetime. In turn, those spacetime deformations instruct the mass and energy how to move. In almost all cases, physicists use the equations of relativity to figure out how a particular combination of objects will move. They have some physical scenario, like a planet orbiting a star or two black holes colliding, and they ask how those objects deform spacetime and what the subsequent evolution of the system should be.

 

But it’s also possible to run Einstein’s math in reverse by imagining some desired motion and asking what kind of spacetime deformation can make it possible. This is how the Mexican physicist Miguel Alcubierre discovered the physical basis for a warp drive—long a staple of the Star Trek franchise.

 

The goal of a warp drive is to get from A to B in the time between commercial breaks, which typically involves faster-than-light motion. But special relativity expressly forbids speeds faster than light. While this never bothered the writers of Star Trek, it did irritate Alcubierre. He discovered that it was possible to build a warp drive through a clever manipulation of spacetime, arranging it so that space in front of a vessel gets scrunched up and the space behind the vessel stretched out. This generates motion without, strictly speaking, movement.

 

It sounds like a contradiction, but that’s just one of the many wonderful aspects of general relativity. Alcubierre’s warp drive avoids violations of the speed-of-light limit because it never moves through space; instead space itself is manipulated to, in essence, bring the spacecraft’s destination closer to it.

 

While tantalizing, Alcubierre’s design has a fatal flaw. To provide the necessary distortions of spacetime, the spacecraft must contain some form of exotic matter, typically regarded as matter with negative mass. Negative mass has some conceptual problems that seem to defy our understanding of physics, like the possibility that if you kick a ball that weighs negative 5 kilograms, it will go flying backwards, violating conservation of momentum. Plus, nobody has ever seen any object with negative mass existing in the real Universe, ever.

 

These problems with negative mass have led physicists to propose various versions of “energy conditions” as supplements to general relativity. These aren’t baked into relativity itself, but add-ons needed because general relativity allows things like negative mass that don’t appear to exist in our Universe—these energy conditions keep them out of relativity’s equations. They’re scientists’ response to the unsettling fact that vanilla GR allows for things like superluminal motion, but the rest of the Universe doesn’t seem to agree.

Warp factor zero

The energy conditions aren’t experimentally or observationally proven, but they are statements that concord with all observations of the Universe, so most physicists take them rather seriously. And until recently, physicists have viewed those energy conditions as making it absolutely 100 percent clear that you can’t build a warp drive, even if you really wanted to.

 

But there is a way around it, discovered by an international team of physicists led by Jared Fuchs at the University of Alabama in Huntsville. (The team is also affiliated with the Applied Propulsion Laboratory of Applied Physics, a virtual think tank dedicated to the research of, among many other things, warp drives.) In a paper accepted for publication in the journal Classical and Quantum Gravity, the researchers dug deep into relativity to explore if any version of a warp drive could work.

 

The equations of general relativity are notoriously difficult to solve, especially in complex cases such as a warp drive. So the team turned to software algorithms; instead of trying to solve the equations by hand, they explored their solutions numerically and verified that they conformed to the energy conditions.

 

The team did not actually attempt to construct a propulsion device. Instead, they explored various solutions to general relativity that would allow travel from point to point without a vessel undergoing any acceleration or experiencing any overwhelming gravitational tidal forces within the vessel, much to the comfort of any imagined passengers. They then checked whether these solutions adhered to the energy conditions that prevent the use of exotic matter.

 

The researchers did indeed discover a warp drive solution: a method of manipulating space so that travelers can move without accelerating. There is no such thing as a free lunch, however, and the physicality of this warp drive does come with a major caveat: the vessel and passengers can never travel faster than light. Also disappointing: the fact that the researchers behind the new work don’t seem to bother with figuring out what configurations of matter would allow the warping to happen.

The future of gravity

On one hand, that’s a gigantic letdown. We already have plenty of methods for traveling slower than light (rockets, walking, etc.), so adding one more to the list isn’t all that exciting. Plus, even if we wanted to build this warp drive, the gulf between this hyper-theoretical work and an actual, physical propulsion mechanism is the same as the difference between writing down Newton’s laws and building a Falcon 9.

 

But that doesn’t mean this new development isn’t interesting. We don’t fully understand gravity, and we know that Einstein’s theory is incomplete. One of the signposts that we have to a future understanding of gravity is the fact that general relativity allows for interesting, exotic solutions—like warp drives—that appear to violate other domains of physical understanding.

 

Us physicists like it when all of our theories line up and agree on the nature of the Universe. So if the energy conditions set real limits on physics—limits where things like negative mass don’t just not exist, but can’t exist—then we’d like a physical theory that says that from the beginning, instead of relying on add ons like the energy conditions.

 

Exploring how a warp drive might (not) work, and under what conditions and restrictions, is a step in that direction. For years physicists thought that the energy conditions outlawed all kinds of warp drives, yet the new research shows a possible way around that. What comes next will be a win no matter what; whether we get a fancy superluminal warp drive or not. That’s because whatever comes out of future lines of inquiry along these directions, we’re going to learn more about the force of gravity, and just possibly revolutionize our understanding of it.

 

And who knows what we’ll get once we understand gravity better.

 

Classical and Quantum Gravity, 2024.  DOI: 10.1088/1361-6382/ad26aa

 

A dairy farm worker in Michigan has tested positive for an H5 bird flu virus, marking the second human case of bird flu that has been linked to the unprecedented outbreak of highly pathogenic avian influenza H5N1 virus among dairy cows in the US.

 

According to the Centers for Disease Control and Prevention, the worker only experienced a mild eye infection and has since recovered, much like the first case in a dairy farm worker in Texas. The Michigan worker was being monitored for symptoms due to exposure—the person worked on a dairy farm with H5N1-infected cattle.

 

In a press briefing Wednesday afternoon, CDC Principal Deputy Director Nirav Shah said the person was taking part in an active surveillance system, in which the state's health department sent out a text message every day asking about the presence of any symptoms. The worker's infection was identified after the person responded that there were symptoms.

 

Health officials took swabs of the worker's nose and eye. The CDC, which received the swabs for testing on Tuesday, confirmed that day that the nasal swab was negative while the eye swab was positive.

 

"We found this case because we were looking for this case," Shah said, emphasizing that the CDC and Michigan had been prepared to find the additional human case and expected it.

 

In a separate statement, Natasha Bagdasarian, Michigan's chief medical executive, emphasized that the state has been tracking the situation closely. "The current health risk to the general public remains low. This virus is being closely monitored, and we have not seen signs of sustained human-to-human transmission at this point. This is exactly how public health is meant to work, in early detection and monitoring of new and emerging illnesses."

 

The CDC also said that its risk assessment for the general public remains low. However, the agency added that the infection underscores the risks and needed precautions for people exposed to infected or potentially infected animals.

 

It's unclear how the worker became infected, but the CDC suspects that the person either got raw milk splashed in their eye or rubbed their eyes with contaminated hands. Raw milk from infected cattle has been found to carry extremely high levels of H5N1. It's unclear at this point if the person had access to personal protective equipment, such as a face shield, which the federal government has been working to distribute.

 

For now, the testing by the CDC has only confirmed that the worker has an H5-type influenza infection. Further testing that can type the virus as H5N1, specifically, has not yet been completed. The CDC expects to have genetic sequencing of the virus available in one to two days. At that point, researchers can assess if the virus is the same as what's been seen in cows and if it has acquired any worrying mutations that may allow it to spread more easily among humans.

 

With each new infection, H5N1 has the ability to adapt to new hosts, and experts are anxiously monitoring the outbreak in dairy cows—which was first detected in March—for signs that the virus will gain the ability to spark an outbreak in humans. Dawn O’Connell, assistant secretary for Preparedness and Response within the US Department of Health and Human Services, said in the press briefing Wednesday that, as a precaution, the federal government is moving to finish production of 4.8 million doses of H5N1 vaccine that is well-matched the virus seen circulating in dairy cows.

 

To date, the US Department of Agriculture has recorded 52 dairy cow herds have been infected with H5N1 across nine states. Michigan has reported 19 infected herds.

 

This post has been updated to include new information from today's press briefing and statements.

 

Whales use complex communication systems we still don’t understand, a trope exploited in sci-fi shows like Apple TV’s Extrapolations. That show featured a humpback whale (voiced by Meryl Streep) discussing Mahler’s symphonies with a human researcher via some AI-powered inter-species translation app developed in 2046.

 

We’re a long way from that future. But a team of MIT researchers has now analyzed a database of Caribbean sperm whales’ calls and has found there really is a contextual and combinatorial structure in there. But does it mean whales have a human-like language and we can just wait until Chat GPT 8.0 to figure out how to translate from English to Sperm-Whaleish? Not really.

One-page dictionary

“Sperm whales communicate using clicks. These clicks occur in short packets we call codas that typically last less than two seconds, containing three to 40 clicks,” said Pratyusha Sharma, a researcher at the MIT Computer Science and Artificial Intelligence Laboratory and the lead author of the study. Her team argues that codas are analogues of words in human language and are further organized in coda sequences that are analogues of sentences. “Sperm whales are not born with this communication system; it's acquired and changes over the course of time,” Sharma said.

 

Seemingly, sperm whales have a lot to communicate about. Earlier observational studies revealed that they live a fairly complex social life revolving around family units forming larger structures called clans. They also have advanced hunting strategies and do group decision-making, seeking consensus on where to go and what to do.

 

Despite this complexity in behavior and relationships, their vocabulary seemed surprisingly sparse.

 

Sharma’s team sourced a record of codas from the dataset of the Dominica Sperm Whale Project, a long-term study on sperm whales that recorded and annotated 8,719 individual codas made by EC-1, a sperm whale clan living in East Caribbean waters. Those 8,719 recorded codas, according to earlier research on this database, were really just 21 coda types that the whales were using over and over.

 

A set of 21 words didn’t look like much of a language. “But this [number] is exactly what we found was not true,” Sharma said.

Fine-grained changes

“People doing those earlier studies were looking at the calls in isolation... They were annotating these calls, taking them out of context, shuffling them up, and then tried to figure out what kind of patterns were recurring,” Sharma explained. Her team, by contrast, analyzed the same calls in their full context, basically looking at entire exchanges rather than at separate codas. “One of the things we saw was fine-grained changes in the codas that other whales participating in the exchange were noticing and reacting to. If you looked at all these calls out of context, all these fine-grained changes would be lost; they would be considered noise,” Sharma said.

 

The first of those newly recognized fine-grained changes was termed “rubato,” borrowed from music, where it means introducing slight variations in the tempo of a piece. Communicating sperm whales could stretch or shrink a coda while keeping the same rhythm (where rhythm describes the spacing between the clicks in a coda).

 

The second feature the researchers discovered was ornamentation. “An ornament is an extra click added at the end of the coda. And when you have this extra click, it marks a critical point, and the call changes. It either happens toward the beginning or at the end of the call,” said Sharma.

 

The whales could individually manipulate rubato and ornamentation, as well as previously identified rhythm and tempo features. By combining this variation, they can produce a very large variety of codas. “The whales produce way more combinations of these features than 21—the information-carrying capacity of this system is a lot more capable than that,” Sharma said.

 

Her team identified 18 types of rhythm, three variants of rubato, five types of tempo, and an ability to add an ornament or not in the sperm whale’s communication system. That adds up to 540 possible codas, of which there are roughly 150 these whales frequently used in real life. Not only were sperm whales’ calls built with distinctive units at a coda level (meaning they were combinatorial), but they were compositional in that a call contained multiple codas.

 

But does that get us any closer to decoding the whale’s language?

 

“The combinatoriality at the word level and compositionality at the sentence level in human languages is something that looks very similar to what we found,” Sharma said. But the team didn’t determine whether meaning was being conveyed, she added. And without evidence of meaning, we might be barking up the wrong tree entirely.

Mockingbird songs

“It’s funny how people always want to compare animal communication to language—it is sexy, and it helps to sell papers. But to me, human music is a much richer, much more apt comparison,” said Tecumseh Fitch, an evolutionary biologist at the University of Vienna and author of the book The Evolution of Language. Fitch pointed out that three ingredients are necessary for a language to appear. The first is vocal learning, an ability to learn new sounds. Sperm whales have that, but so do thousands of other species.

 

Then there are syntax and semantics. “Syntax is the only thing this whole study is relevant to because it suggests that sperm whales have a much more complex syntax than we thought before. Semantics, though, the study says nothing about. The authors are very open about it,” Fitch said.

 

One thing that requires vocal learning and syntax but doesn’t have semantics is music. Let’s take a mockingbird song, for example.

 

“What a mockingbird has is a system of small syllables that it combines into larger song fragments, and then any given bird will sing sometimes for hours and combine thousands of these what you might call words into super complex sequences that you might call songs,” Fitch says. But there is no indication, despite lots of research, that those songs are anything like sentences carrying meaning like “I had a breakfast this morning” or “there is a really good tree in my territory.”

 

“No songbird researcher thinks that. Now, if we return to sperm whales, we can ask what is the difference between their communication and a mockingbird song? I mean from the structural point of view, from the point of view of comparing whether it is a language or music, the answer right now is ‘none that we can identify,’” Fitch said.

 

Back in the 1960s, Charles F. Hockett, an American linguist, published a list of the design features of language, basically a set of 13 boxes a communication system has to check to be considered a language. According to Sharma’s study, we can confidently say that sperm whales’ communication checks nine of those 13 boxes. The four still missing include displacement: the ability to talk about things not present in either space or time. There’s also productivity, a capacity to say things that have never been said before, and “duality of patterning,” which is organizing meaningless sounds into meaningful words and sentences. Last, and far from least, is semanticity.

 

“You would need all these design features in order to have a language. But you just can’t have displacement or duality of patterning without knowing anything about the semantics,” said Fitch. But testing if there are semantics in sperm whales’ calls is hardly possible without cooperation from the sperm whales. And getting them to cooperate may take an unexpected turn.

Whale in the loop

One way to figure out semantics in sperm whales’ calls is by doing playback experiments with wild whales—basically, synthesizing codas or coda sequences, playing those near sperm whales, and seeing how they react. But playback experiments have been attempted before, and there is a dark side to doing them with wild animals.

 

“There were experiments where they started to make contact with animals in the wild—with apes. The population exposed to those experiments got violent, and it disrupted the ecosystem,” said Sharma. “In the wild, we have no understanding of what to expect, and the species we are studying is already endangered—sperm whales are classified as such.” In her opinion, playback experiments are a no-go unless there is confidence they would do no harm. And today we’re not so sure.

 

According to Sharma, even collecting passive data on whales like the one gathered by the Dominica Sperm Whale Project is very expensive and takes lots of time—it took over 13 years to record the codas she used in her study.

 

Finally, even if sperm whales do talk Sperm-Whaleish, why would they want to talk to us? “If aliens came to Earth, they would probably have an incentive to talk to humans and [use] communication systems that are more interactive. An alien would draw something and say a word, then a human would draw something. Communication with whales has been very passive,” Sharma said. “This thing they showed in [the show] Extrapolations was like a dream system. I don’t know when we could have something like that—10, maybe even 20 years? But how we get the evidence of casual relationship between the sound and the action is not clear for me at all,” she added.

 

“This study kind of invites you to think sperm whales’ communication is more like language than we thought, as opposed to more like music than we thought. But the kind of questions it is asking—the leading questions, like ‘is there a language?’—a lot of people would like to think it’s true,” said Fitch.

 

Nature Communications, 2024.  DOI: 10.1038/s41467-024-47221-8

 

Tesla fans—and CEO Elon Musk himself—are excited about the prospects for Tesla’s Full Self Driving (FSD) software. Tesla released a major upgrade—version 12.3—of the software in March. Then, last month, Musk announced that Tesla would unveil a purpose-built robotaxi on August 8. Last week, Musk announced that a new version of FSD—12.4—will come out in the coming days and will have a “5X to 10X improvement in miles per intervention.”

 

But I think fans expecting Tesla to launch a driverless taxi service in the near future will be disappointed.

 

During a late March trip to San Francisco, I had a chance to try the latest self-driving technology from both Tesla and Google’s Waymo.

 

During a 45-minute test drive in a Tesla Model X, I had to intervene twice to correct mistakes by the FSD software. In contrast, I rode in driverless Waymo vehicles for more than two hours and didn’t notice a single mistake.

 

So while Tesla’s FSD version 12.3 seems like a significant improvement over previous versions of FSD, it still lags behind Waymo’s technology.

 

However, Waymo’s impressive performance comes with an asterisk. While no one was behind the wheel during my rides, Waymo has remote operators that sometimes provide guidance to its vehicles (Waymo declined to tell me whether—or how often—remote operators intervened during my rides). And while Tesla’s FSD works on all road types, Waymo’s taxis avoid freeways.

 

Many Tesla fans see these limitations as signs that Waymo is headed for a technological dead end. They see Tesla’s FSD, with its capacity to operate in all cities and on all road types, as a more general technology that will soon surpass Waymo.

 

But this fundamentally misunderstands the situation.

 

Safely operating driverless vehicles on public roads is hard. With no one in the driver’s seat, a single mistake can be deadly—especially at freeway speeds. So Waymo launched its driverless service in 2020 in the easiest environment it could find—residential streets in the Phoenix suburbs—and has been gradually ratcheting up the difficulty level as it gains confidence in its technology.

 

In contrast, Tesla hasn’t started driverless testing because its software isn’t ready. For now, geographic restrictions and remote assistance aren’t needed because there’s always a human being behind the wheel. But I predict that when Tesla begins its driverless transition, it will realize that safety requires a Waymo-style incremental rollout.

 

So Tesla hasn’t found a different, better way to bring driverless technology to market. Waymo is just so far ahead that it’s dealing with challenges Tesla hasn’t even started thinking about. Waymo is playing chess while Tesla is still playing checkers.

Tesla is several years behind Waymo

The current excitement around Tesla’s FSD reminds me of the hype that surrounded Waymo in 2018. Early that year, Waymo announced deals to purchase 20,000 I-Pace sedans from Jaguar and 62,000 Pacifica minivans from Fiat Chrysler.

 

But the service Waymo launched in December 2018 was a disappointment. There were still safety drivers behind the wheel on most rides, and access was limited to a handpicked group of passengers.

 

It wasn’t until October 2020 that Waymo finally launched a fully driverless taxi service in the Phoenix area that was open to the general public. And even after that, Waymo expanded slowly.

 

Waymo began offering commercial service in San Francisco in 2023 and is now expanding to Los Angeles and Austin. Today, the company has only a few hundred vehicles in its commercial fleet—far fewer than the 82,000 vehicles it was planning to purchase six years ago.

 

What went wrong? In an August 2018 article, journalist Amir Efrati reported on the limitations of Waymo’s technology. Efrati wrote that “Waymo vans have trouble with many unprotected left turns and with merging into heavy traffic in the Phoenix area.” In addition, “the cars have trouble separating people, or cyclists, who are in groups, especially people near shopping centers or in parking lots.”

 

Efrati’s reporting makes Waymo’s technology in 2018 sound worse than FSD 12.3, but not that much worse. FSD still sometimes struggles with merging in heavy traffic and navigating around groups of pedestrians.

Waymo’s technology improved steadily after 2018. In late 2020, after Waymo launched its fully driverless service in the Phoenix area, I interviewed Joel Johnson, a college student who had taken dozens of Waymo rides and published a number of ride videos on YouTube.

"They've really ironed out stuff like unprotected lefts," Johnson said in one video. "It's definitely improved over time."

 

In an October video, a Waymo car was driving through a Costco parking lot crowded with pedestrians. It waited patiently until they were out of the way, then moved forward confidently.

 

"This amount of pedestrians would have caused whiplash-inducing brake usage in March 2020," Johnson wrote in an on-screen note. "And it would have completely given up in July 2019. No longer!"

The version of FSD I tried in March was clearly not ready for driverless operation. For example, I had to intervene to prevent the Model X from running over a plastic lane divider, a mistake Waymo would not have made in 2020. So while FSD 12.3 seems superior to Waymo’s technology circa 2018, it’s not as good as Waymo’s technology at the end of 2020.

Waymo relies on remote operators

 

In its early years, Waymo did all of its testing with safety drivers. When the software made a mistake, the driver would intervene to prevent a crash and then carefully document the incident. Engineers used data from these near misses to improve Waymo’s software.

 

The better self-driving software gets, the more expensive this testing strategy becomes. If self-driving software makes a mistake every 50 miles, a safety driver might experience several errors in a single workday. But if the software makes a mistake every 5,000 miles, a safety driver might have to drive around for weeks, at company expense, to get a single bug report.

 

It would have been prohibitively expensive for Waymo to continue testing with safety drivers until its software was provably safer than a human driver. So instead, Waymo started deploying driverless vehicles supported by a remote operator.

 

Waymo programmed its driverless vehicles to make extreme caution their default behavior. If a driverless Waymo isn’t 100 percent confident that it’s safe to proceed, it will slow to a stop and ask for remote assistance. The hope is that the software’s average confidence will increase over time and vehicles will need remote assistance less and less often.

 

Waymo says remote operators never directly drive its vehicles. Instead, operators answer questions and give hints to guide the vehicle in the right direction. Here are two examples supplied to me by Waymo:

 

• In this video, a Waymo’s path is blocked by a large truck coming in the opposite direction. The remote operator has the Waymo squeeze into the right lane to give the truck room to pass.
• In this video, a Waymo is approaching an intersection with multiple fire trucks. The Waymo asks the remote operator two questions—“Is the emergency vehicle blocking all indicated lanes” and “Is the road closed?”—that help it move through the scene with confidence.

 

This strategy gets tricky on freeways. If a driverless vehicle asks for help and doesn’t get a timely response, it needs to stop and wait. But that’s hard to do on a freeway while going 70 miles per hour.

 

So although Waymo has tested its technology on freeways (with safety drivers) for more than a decade, Waymo’s driverless taxis don’t use them yet. This makes Waymo’s service less useful.

 

My first Waymo trip in March took me from downtown San Francisco to a McDonald's in the Bayview neighborhood. For Waymo, this was a 28-minute trip on surface streets. An Uber or Lyft driver would have taken the 101 freeway and gotten there in about 15 minutes.

 

Waymo is working to address the problem. In January, the company started testing driverless operations on freeways in the Phoenix area. If testing goes well, Waymo may enable freeway driving for its commercial fleet in the coming months.

 

According to statistics published by Waymo, Waymo’s cautious approach has worked remarkably well, at least from a safety perspective. Over Waymo’s first 7 million driverless miles, its vehicles got into injury-causing crashes about one-fourth as often as comparable human drivers.

Does Tesla have a better approach?

Many Tesla fans see the limitations of Waymo’s current service—avoidance of freeways, reliance on remote operators, and restriction to a few metro areas—as evidence that Waymo’s technology is fundamentally flawed.

 

In a typical tweet last year, a Tesla supporter argued that Waymo and GM-owned Cruise have “developed extremely narrow, very brittle technology that won't scale.”

 

Musk replied, “Yeah, extremely brittle to local conditions & doesn’t scale.”

 

 

A key part of this argument has to do with neural networks. Waymo began as the Google self-driving car project 15 years ago, before the deep learning revolution of the 2010s. The earliest versions of its software probably used hand-coded rules rather than machine learning. Some Tesla supporters seem to assume Waymo is still using the same outdated techniques.

 

In reality, Waymo now makes extensive use of neural networks.

 

For example, in this 2020 article, Waymo described how it uses neural networks for its perception system—the software that identifies and tracks nearby objects. In this February 2024 talk, a Waymo engineer explained how the company uses transformers, the Google-invented architecture behind large language models, to predict the actions of other vehicles.

 

So Waymo might have had a brittle software stack a decade ago, but the company hasn’t been standing still.

 

A big factor that makes self-driving difficult is the “long tail” of unusual situations a driver might encounter in the real world. Wet cement. A bike strapped to a car. A tree on the back of a truck. A woman in a wheelchair chasing a duck.

 

Companies building self-driving technology need to do millions of miles of testing to discover as many of these “edge cases” as possible. And this is one place where Tesla plausibly has an advantage over Waymo.

 

As we’ve seen, Waymo has to pay safety drivers for every mile of supervised testing. In contrast, Tesla has convinced thousands of customers to test its Full Self Driving software for free. Indeed, customers pay thousands of dollars for the privilege!

 

This gives Tesla access to effectively unlimited data. In theory, more data should enable Tesla to efficiently identify edge cases its self-driving software needs to handle. More data should also enable Tesla to train better neural networks.

 

While access to more data is certainly helpful, it’s not a magic bullet. One issue is that the data Tesla collects is unlabeled. Waymo’s safety drivers document each disengagement to help identify flaws in Waymo’s software, but Tesla customers are unlikely to do that.

 

Another issue is that some edge cases are much harder to handle than others.

The first responder problem

Let’s take interactions with police and firefighters as an example. This is a problem that Waymo and GM’s Cruise struggled with a lot last year. I wrote about this in September:

 

According to San Francisco Fire Department records, several Waymo or Cruise vehicles blocked narrow roads, forcing fire trucks to take detours en route to fires. AVs got stuck near firefighting operations, forcing firefighters to work around them as they positioned hoses and ladders. A few AVs parked in front of fire stations, trapping fire trucks inside.

The problem wasn’t that these vehicles were running into firetrucks (Tesla has had a problem with that in the past). Rather, they were so cautious that they were getting stuck. This is why these incidents only became a concern after Waymo and Cruise began driverless operations; a safety driver could have intervened after a few seconds of inaction. It’s also why this hasn’t become an issue for Tesla yet: Every Tesla still has a human behind the wheel who can take over if FSD gets stuck.

 

Most of the time, driving requires following simple, deterministic rules: stay in the center of the lane, avoid hitting other road users, obey stop lights and stop signs, and so forth.

 

But navigating through the scene of a fire or car crash is much trickier. Emergencies can totally disrupt the flow of traffic, forcing drivers to improvise new traffic patterns. Drivers often need a nuanced understanding of what other people are trying to accomplish so that they can avoid getting in their way. If police or firefighters are directing traffic, drivers need to understand their hand signals.

 

In short, navigating the scene of a fire or car crash sometimes requires reasoning skills that are far beyond the capabilities of today’s AI systems. So emergency scenes are likely to remain as a “corner case” for Tesla’s FSD for at least a few more years, just as they still are for Waymo’s software.

 

 

Recognizing this, Waymo backstops its software in a number of ways. Waymo vehicles can request guidance from Waymo remote operators if they aren’t sure what to do. First responders can lean into the car to talk to the remote operator, or they can jump into the car and drive it themselves. City officials can erect a geofence to keep Waymo vehicles away from emergency scenes. And Waymo has provided training to thousands of first responders in Phoenix, San Francisco, and elsewhere on how to interact with its vehicles.

 

These efforts seem to be paying off. In a February story, the San Francisco Chronicle reported that firefighters have been filing fewer reports about misbehaving vehicles since August. That’s partly because Cruise stopped operations in San Francisco in October. But the number of Waymo reports seems to be down as well.

 

If Tesla pushed an FSD update that enabled fully driverless operations, I expect we’d start to see the same kind of stories that we saw about Waymo and Cruise last year: stories about Tesla vehicles driving over hoses, blocking ambulances, ignoring police officers’ instructions, interfering with firefighters’ ladder placements, and so forth.

 

Given the size of its fleet, Tesla could face a backlash from first responders much bigger than the one Waymo and Cruise faced last year in San Francisco. And as Waymo and Cruise discovered, police and firefighters have a lot of political influence.

 

If Tesla is serious about providing a driverless taxi service, it will need the kinds of infrastructure and support services Waymo has been building in recent years. That includes remote operators to intervene when a vehicle gets stuck and personnel to work with local governments. Instead, Tesla has been moving in the opposite direction. The Information reported last month that Tesla was dissolving its policy team.

A service, not just a software product

Today’s FSD is a software product, but a robotaxi service is much more than just software. Here’s a simple example: What happens if a Tesla robotaxi gets a flat tire?

 

If you're driving your car and it gets a flat tire, it’s your problem. If you’re riding in a taxi that gets a flat tire, it’s the driver’s problem. But who changes the tire if there is no driver?

 

Tesla envisions a future where people buy a Tesla and then rent it out via a Tesla-run ride-hailing network. So theoretically, Tesla could say that a flat tire is the owner’s problem. But that wouldn’t work in practice. The owner might be in a meeting or even on vacation. And it wouldn’t be acceptable to leave a car stranded by the side of the road—possibly with a passenger inside—for hours while Tesla tries to contact its owner.

 

So if Tesla wants to get into the taxi business, it will need a staff of mobile technicians to rescue stranded Teslas. These could be Tesla employees or independent contractors, but arrangements need to be in place before the company starts offering service in a particular area. And these same people can be dispatched to rescue Teslas when, inevitably, FSD software gets stuck.

 

And this means that Tesla will likely want to roll out the driverless version of FSD gradually, one metropolitan area at a time. This makes sense for other reasons, too. It would give Tesla time to introduce itself to local officials and offer training to local police and fire departments. And while remote operators and customer service agents need not be local, a phased rollout would give Tesla time to hire people for those roles.

 

I hope it’s clear now why I don’t consider the current limitations on Waymo’s service— geographic restrictions and the use of remote operators—particularly damning. Every taxi service is geographically limited, and there are good reasons for Waymo to roll out its service one city at a time.

The bitter lesson

I’ve talked to a lot of Tesla fans, so I have a pretty good idea what they’d say here: that I’m underestimating how quickly Tesla’s self-driving technology will improve as Tesla throws more data and computing power at the problem.

 

Some Tesla backers like to reference the famous Rich Sutton essay “The Bitter Lesson.” Sutton argued that AI researchers have historically spent too much time time trying to hand-code human insights on the best way to solve a problem like computer chess or image recognition. People ultimately got better results with general-purpose learning algorithms trained on large amounts of data.

 

Sutton wrote his essay in 2019. Since then, the success of large language models has demonstrated Sutton’s insight in a spectacular way. Early AI researchers would try to understand the properties of natural language and then encode their insights into AI systems. These systems didn’t work very well. What worked much better was taking the simple transformer architecture and scaling it up to hundreds of billions of parameters to create LLMs like GPT-4.

 

Musk is betting that this same dynamic will work in Tesla’s favor with self-driving. He has described FSD version 12 as using “end-to-end neural nets.” He is investing billions of dollars in hardware to train those neural networks using vast amounts of data collected from Tesla customers. If you buy Sutton’s argument, you might expect Tesla to jump ahead of Waymo.

 

But I think people are reading too much into Sutton’s argument. Sutton’s point is that large neural networks trained on lots of data tend to outperform hand-coded AI systems. But it doesn’t follow that throwing more data and computing power at any particular neural network will achieve arbitrarily high levels of performance.

 

LLMs are a powerful example of this. LLMs hallucinate. They fail at simple tasks like counting objects and reading analog clocks. LLMs are great for applications where accuracy isn’t that important or where a human being is checking the output after it’s generated. But if you need very high accuracy, they are not a good choice.

 

Self-driving systems do need very high accuracy. And it’s not obvious that an end-to-end neural network with enough data and computing power will necessarily achieve it.

 

Business professor Ethan Mollick has written about the “jagged frontier”: Complex AI systems are often impressively good at some tasks but surprisingly bad at others. Tesla may get really good at navigating freeways, intersections, and traffic circles but make little progress on avoiding wet cement or understanding the hand signals of police officers.

Could Waymo have a Just Walk Out problem?

Waymo’s approach to this problem is to build a mostly automated system that is able to gracefully fall back on human assistance when needed.

 

While this works quite well from a safety perspective, I’ve started to wonder about the economics of it. If Waymo vehicles were constantly asking for remote guidance, Waymo might need to hire so many remote operators that it negates the cost savings that come from not needing a driver.

 

Last month, Amazon announced it was removing its no-checkout technology, called Just Walk Out, from Amazon Fresh grocery stores.

 

 

Like Waymo, Amazon was bullish about its technology in 2018. That year, Bloomberg reported that Amazon was planning to open 3,000 Amazon Go convenience stores based on Just Walk Out technology.

 

But that never happened, and reporting from The Information’s Theo Wayt helps to explain why. Wayt reported last year that Amazon’s technology—like Waymo’s—wasn’t fully automated. Amazon had more than 1,000 workers in India manually verifying customer selections. Wayt says that in mid-2022, “Just Walk Out required about 700 human reviews per 1,000 sales.”

 

Amazon aimed to reduce this figure to 20 to 50 human reviews per 1,000 items, but the company “repeatedly missed” its performance targets.

 

Could Waymo have a similar problem? I don’t know, and unsurprisingly, Waymo declined to comment about the frequency of remote interventions.

 

My best guess is that this will not be a serious issue for Waymo. During the rides I took in March, Waymo’s vehicles drove smoothly and confidently. If they were constantly seeking remote guidance, I would have expected more hesitation and erratic driving.

 

Waymo also—finally—seems to be expanding fairly rapidly. Earlier this month, the company announced it was serving 50,000 trips a week, up from 10,000 weekly trips nine months earlier. It seems unlikely that Waymo would grow that quickly unless management was confident they had a clear path to profitability.

 

Regardless, I don’t think Tesla has discovered a better way to approach the problem. Large, complex neural networks tend to be good at some things but not as good at others. Yet the AI system that controls a two-ton vehicle needs to be very reliable all the time. For the next few years, at least, that will only be possible with human backup.

 

Tim Lee was on staff at Ars from 2017 to 2021. Last year, he launched a new newsletter, Understanding AI, that explores how AI works and how it's changing our world. You can subscribe here.

 

With a zap of electricity from well-placed electrodes on the back of the neck, patients with tetraplegia can regain some modest yet potentially "life-changing" functioning of their hands and arms, according to data from a small clinical trial published Monday in Nature Medicine.

 

The relatively simple stimulation method—which requires no surgery—offers an accessible, more affordable, non-invasive means for those living with paralysis to regain some meaningful function, the researchers behind the trial say. However, the therapy's further potential remains limited given that scientists have yet to fully understand exactly why it works.

 

For the trial, 60 patients with tetraplegia underwent the stimulation therapy over at least 24 sessions during a two-month period. At the end, 72 percent (43 patients) saw clinically meaningful improvements in both strength and functional performance. Further, 90 percent (54 patients) saw improvement from at least one strength or functional outcome. There were no serious adverse events reported.

 

"The most exciting thing for us is that we're seeing effects that improve quality of life," Chet Moritz, a co-author of the study and co-director for the Center for Neurotechnology at the University of Washington, said in a press briefing. "And also, we believe that the stimulation may be causing neuroplasticity or, in a sense, healing part of the damage to the spinal cord injury, such that the benefits persist beyond stimulation."

 

 

The trial, which took place at 14 clinical sites across the US, Canada, and Europe, was a prospective study—not the gold-standard design of a randomized, double-blind placebo control trial. Moritz and his colleagues explained that there were a number of reasons for this. For one, they weren't sure if they could use a sham electrical stimulation for a placebo group. Patients can feel the real electrical stimulation, described as an internal "buzz," and efforts to re-create the feeling, but not the effects of a sham treatment, were uncertain. Also, there were ethical concerns about having people with tetraplegia repeatedly travel to clinical sites and be subjected to potentially uncomfortable treatments with no expected benefit.

 

Still, the researchers behind the study are confident that the gains they saw were not simply placebo effects. For one thing, all the trial participants spent two months in standard rehabilitation therapy before they underwent the stimulation therapy. And their progress in that first phase of the trial was compared to their progress in the second half, in which they received the therapy. The differences were "very dramatic for many of the measures," Edelle Field-Fote, a co-author and Director of Spinal Cord Injury Research at Shepherd Center in Georgia, said in the briefing.

Real improvements

Melanie Reid, a trial participant and journalist with The Times of London who spoke at the press briefing, was quick to report that she also believed the benefits seen were not from a placebo effect. Reid broke her neck 14 years ago in a fall from a horse. She was initially left with little function below her armpits but eventually regained some function of her right hand, while her left remained "useless." She said that with the stimulation therapy, she had regained function in her left hand, allowing her to release a seat belt buckle, scroll on touchscreen devices, and put her hair in a ponytail.

 

In response to a question about placebo effects, Reid recalled an instance during therapy in which she was holding a wide, weighted jar in her hand. With the stimulator on, "I was able to hold it quite comfortably," she said. But then, the stimulator was turned off, and the loss of the stimulation caused her to immediately drop the jar. "It was, to me, a real illustration of the increased power [from the therapy]," she said.

 

And that increased power can make a world of difference, she emphasized. "Everyone thinks that spinal injury, you know, all you want to do is be able to walk again, but if you're a tetraplegic or a quadriplegic, what matters most is working hands," she said. "There's no miracles in spinal injuries, but tiny gains can be life-changing."

 

Another trial participant, Sherown Campbell, echoed Reid's points at the briefing. Campbell broke his neck about 10 years ago during a wrestling match and was subsequently diagnosed with quadriplegia. During the stimulation therapy, he could see real improvements in grasping and pinching tests. But, he also saw improvements at home afterward, with no stimulation. He improved his ability to open jars, grip a steering wheel, and tie balloons. Campbell also reported that, after the therapy, he regained the ability to sweat below the level of his injury, which allows him to control his body temperature in hot weather and during exercise—a significant improvement in quality of life.

Unknowns

While the therapy is making improvements in patients' lives, the potential for it to move past these modest (though meaningful) gains is held back by a lack of understanding of how the therapy works. In the study, the researchers describe the stimulation therapy as being from two electrodes placed along the spine, with one below and one above the point of injury. The electrodes then produce a stimulation at "30 Hz with a 10-kHz carrier frequency overlay, which consisted of 10 pulses with a 10-kHz frequency and 100-μs pulse width." But it's unclear how this stimulation is improving muscle control and, potentially, spurring neuroplasticity.

 

The device is similar to external spinal cord stimulators that are already used to treat chronic pain. But, even with those, researchers are still debating why it helps. The initial idea for the stimulation stems from the "gate control theory" dating back to the 1960s. In this hypothesis, pain impulses from around the body travel to the spinal cord to send a message up to the brain. But, at the spinal cord, there's a gated entry for the signal to continue on. The pain impulses, which travel along smaller fibers, can open the gate. But, if signals from larger, sensory fibers are activated (such as by a stimulator), they can close the gate, blocking the pain signal from reaching the brain. This is why, the theory initially went, sensations of hot, cold, or skin rubbing can also disrupt the feeling of pain. But in the decades since the theory's origin, researchers have come to understand that pain perception is far more complex, potentially involving various types of cells in the central nervous system, immune cell activation, ion channel alterations, epigenetics, and chemical mediators.

 

The researchers behind the trial call for more studies into the mechanisms responsible for the improvements seen in their patients. In the meantime, a startup called Onward Medical is preparing the non-invasive stimulator for the commercial market, calling it the ARC-EX system. In April, Onward reported that it had submitted the device to the Food and Drug Administration for review. Researchers involved in the trial also said they were working on implantable devices that they hoped could achieve greater benefits.

 

When you look at the northern sky, you can follow the arm of the Big Dipper as it arcs around toward the bright star called Arcturus. Roughly in the middle of that arc, you'll find the Northern Crown constellation, which looks a bit like a smiley face. Sometime between now and September, if you look to the left-hand side of the Northern Crown, what will look like a new star will shine for five days or so.

 

This star system is called T. Coronae Borealis, also known as the Blaze Star, and most of the time, it is way too dim to be visible to the naked eye. But once roughly every 80 years, a violent thermonuclear explosion makes it over 10,000 times brighter. The last time it happened was in 1946, so now it’s our turn to see it.

Neighborhood litterbug

“The T. Coronae Borealis is a binary system. It is actually two stars,” said Gerard Van Belle, the director of science at Lowell Observatory in Flagstaff, Arizona. One of these stars is a white dwarf, an old star that has already been through its fusion-powered lifecycle. “It’s gone from being a main sequence star to being a giant star. And in the case of giant stars, what happens is their outer parts eventually get kind of pushed into outer space. What’s left behind is a leftover core of the star—that’s called a white dwarf,” Van Belle explained.

 

The white dwarf stage is normally a super peaceful retirement period for stars. The nuclear fusion reaction no longer takes place, which makes white dwarfs very dim. They are still pretty hot, though, and they're super dense, with a mass comparable to our Sun squeezed into a volume resembling the Earth.

 

But the retirement of the white dwarf in T. Coronae Borealis is hardly peaceful, as it has a neighbor prone to littering. “Its companion star is in the red giant phase, where it is puffed up. Its outer parts are getting sloughed off and pushed into space. The material that is coming off the red giant is now falling onto the white dwarf,” Van Belle said.

Ticking time bomb

And it doesn’t take much littering to make the white dwarf explode. “The material from the red giant will accumulate on the white dwarf’s surface until it forms a layer that’s actually not that thick. Just a few meters—the depth of a deep swimming pool,” Van Belle explained. Most of the material coming off the red giant is hydrogen. And since the red dwarf is still hot, there will eventually be a spark that triggers a runaway nuclear fusion reaction. “That is what causes the explosion,” Van Belle said.

 

The explosion is a nova, which means it doesn’t kill either the white dwarf or the red giant as a supernova would. “Only about 5 percent of the hydrogen layer fuses into heavier elements like helium, and the rest just gets ejected into space. Then the process starts all over again because the explosion isn’t large enough to disrupt the red giant, the donor of all this hydrogen, so it just keeps doing its thing,” Van Belle told Ars. This is why we can predict this event with such precision.

 

“Predictions in astronomy come in two flavors. One is super precise—like the eclipse is going to pass over the city of Houston at exactly 11:35 pm. Other predictions are like this, when we say, ‘well, it’s going to explode sometime between May and September. Maybe a little outside of that window,’” said Van Belle.

 

We can estimate this window because we know what the events leading up to the T. Coronae Borealis explosion looked like the last time it went off in 1946. “The brightness of this object in the sky got a little bit dimmer, a little fingerprint in the light curve. Then it popped off. We have seen the same process with this object recently during the last year or year-and-a-half or so. So that’s why we expect that it is getting to that point,” Van Belle explained.

 

Accordingly, many high-precision instruments at Lowell and other observatories will be zeroed in on T. Coronae Borealis in the coming months to measure the geometry of the expanding fireball to infer the exact physics of the explosion. “But it’s also kind of neat that you don’t need anything if you fancy to go out and see this thing. Last time, in 1946, it hit magnitude 3, and prior to that, in 1866, it was magnitude 2. Magnitude 2 is about as bright as the Northern Star,” Van Belle said.

 

Even neater is that stellar explosions like this bring us way more than fireworks to see. They’ve made your iPhone possible.

The mother of all iPhones

Very energetic events like the T. Coronae Borealis explosion often take light elements like hydrogen and turn them into heavier ones. “This particular kind of object makes most of the lithium we have. Batteries in our phones and other things ultimately came from explosions like this specific one, a recurrent nova,” Van Belle told Ars. The Big Bang, according to Van Belle, formed a bit of lithium, but that didn’t survive till today. The material we have was made in nuclear reactions powering nova explosions.

 

Lithium and other heavier elements (heavier than hydrogen and helium, that is) are ejected into space by the likes of T. Coronae Borealis and ultimately end up in newly formed stars and the disks around them. “That is how lithium ends up in planets such as Earth,” said Van Belle. “This event is not going to call attention to itself by casting shadows on the ground,” he added. But at least we know when to expect it.

 

“When the Betelgeuse supergiant in the Orion constellation explodes, you’ll know it because it will be as bright as the full moon and it will be very hard to ignore. I can say with confidence that it will explode sometime between now and 100,000 years from now. That’s your typical astronomical prediction,” Van Belle said.

 

 

Good morning. It's May 21, and today's photo comes from the James Webb Space Telescope. It showcases the coming together of two massive black holes in the early Universe, just 740 million years after the Big Bang.

 

Each of the black holes has an estimated mass of roughly 50 million times the mass of our star, the Sun. The discovery of this merger so early in the Universe indicates that the growth of these objects in the centers of galaxies occurred very rapidly.

 

Fortunately, Webb and its Near-Infrared Spectrograph are well positioned to observe the fast-moving dense gas characteristic of black holes accreting matter. As they swallow nearby matter, black holes produce highly ionized gas.

 

"Our findings suggest that merging is an important route through which black holes can rapidly grow, even at cosmic dawn," said Hannah Übler of the University of Cambridge. "Together with other Webb findings of active, massive black holes in the distant Universe, our results also show that massive black holes have been shaping the evolution of galaxies from the very beginning."

 

Source: ESA/Webb, NASA, CSA, et. al

 

If things ultimately work out as hoped, brain implants will ultimately restore communication for those who have become paralyzed due to injury or disease. But we're a long way from that future, and the implants are currently limited to testing in clinical trials.

 

One of those clinical trials, based at the University of California, San Francisco, has now inadvertently revealed something about how the brain handles language, because one of the patients enrolled in the trial was bilingual, using English and Spanish. By tracking activity in the area of the brain where the intention to speak gets translated into control over the vocal tract, researchers found that both languages produce consistent signals in this area, so training the system to pick up English phrases would help improve its recognition of Spanish.

Making some noise

Understanding bilingualism is obviously useful for understanding how the brain handles language in general. The new paper describing the work also points out that restoring communications in multiple languages should be a goal for restoring communications to people. Bilingual people will often change languages based on different social situations or sometimes do so within a sentence in order to express themselves more clearly. They often describe bilingual abilities as a key component of their personalities.

 

So, if we really want to restore communication to people, giving them access to all the languages they speak should be a part of it.

 

The new work is designed to make that more likely. Part of a clinical trial called BRAVO (brain-computer interface for restoration of arm and voice), it involved placing relatively simple implants (128 electrodes) into the motor region of the brain—the part that translates intentions to perform actions into the signals needed to trigger muscles to execute them.

 

In terms of speech, this means the neurons needed to convert the desire to say a word into the muscle activity needed to control the mouth and tongue, expel sufficient breath, and tension the vocal cords. This is downstream of the portion of the brain where word choices are made, and English and Spanish presumably differ (which is in turn downstream of where meaning is sorted out, where the two languages might overlap).

 

Hypothetically, if portions of words sound sufficiently similar in terms of how they sound, the muscle control needed to produce that sound would also be similar. So, by tracking neural activity here, they should be able to handle both languages and even detect overlaps between them.

Leveraging language

The process of detecting these signals is rather complex, given that neural activity looks like a noisy series of "spikes," or bursts of voltage changes. Translating those to specific meanings is generally handled by AI systems that are trained to associate certain patterns of activity with specific information (whether that information is "I want to say cat" or "I've seen a cat").

 

So, a lot of work here involved training the software portion of the system to recognize when the participant with the implant wanted to say specific words. This involved him imagining speaking them and the software knowing the word he intended to speak. Using this, the researchers trained the system to recognize 50 English words, 50 Spanish words, and a few that were identical in both languages. Because of things like verb tenses, this worked out to be 178 distinct words.

 

On its own, this wasn't especially effective, with word error rates in the area of 70 percent. But we already have neural networks trained to put words into normal sentences, so the researchers piped this output into GPT2, which identified and ranked likely sentences. That got the per-word error rate down to below 15 percent.

 

Once the system was trained, the participant could use it to participate in conversations and switch languages by choice. The system worked for at least 40 days before the software needed to be recalibrated through additional training, which is better than the reported performance of many other systems.

Word production: They’re all the same

What did this tell us about bilingualism? For one, it supports the hypothesis that this area of the brain is specialized in making sounds and doesn't care about the language they're needed for. While the system would perform better if you told it what language was being spoken first, it did far better than random guessing if it had to figure out the language on its own. And the electrical signals picked up by the implant showed no sign of language specialization, with the authors concluding that there are "no language-specific electrodes."

 

The researchers used the data obtained during the system's training to test aspects of its performance, obtaining results that are also consistent with this idea. For example, pre-training the system on nothing but English words dramatically cut down on the time needed to train it on Spanish afterward, consistent with the two sharing similar activity profiles. The researchers also identified which electrodes were critical for interpreting specific syllables and found that similar patterns of electrode use were best explained by similar-sounding syllables.

 

"This provides compelling evidence that a shared syllable representation can allow data collected in one language to be repurposed for a second language," they conclude.

 

Note the use of "can" and not "will" in that sentence, though. As noted above, bilingual speakers will sometimes switch language mid-sentence, which the current system isn't prepared to handle. And there are lots of languages with features not found in English, such as the clicks of some African languages or the centrality of intonation in Mandarin.

 

Also, note that the system was only trained on a very limited vocabulary. It's likely that this training would help if the vocabulary is expanded, as many new words would share syllables with the existing ones. But ultimately, expanding the range of potential sentences will eventually make it harder for the GPT component to rank potential choices.

 

Despite the limitations, it's a fair bet that the participant in this study, identified only by his nickname of Pancho, is happy to be able to communicate at all.

 

Scientifically, the work largely confirms something you might expect: The portions of the brain that control the muscles needed to make the noises we associate with language aren't especially picky about which language they're handling. Still, it's always a good thing to have evidence in support of a hypothesis, and it's tough to imagine how we'd have gotten this evidence otherwise.

 

Nature Biomedical Engineering, 2024. DOI: 10.1038/s41551-024-01207-5  (About DOIs).

 

Later this week, the US federal agency charged with weather forecasting will release its outlook for the 2024 Atlantic hurricane season at a news conference in Washington, DC. But we already know what the National Oceanic and Atmospheric Administration (NOAA) forecast will say: This year will likely be extremely active in the Atlantic Ocean, Gulf of Mexico, and Caribbean Sea.

 

The Atlantic season formally begins on June 1, and based on current trends, the first named storm may not develop until the middle of the month or later. But make no mistake—when the light switches on later this summer, the season is likely to be a blockbuster.

 

Why? Because the oceans are screaming at us.

 

In the Pacific Ocean, there is an increasingly high likelihood of a La Niña developing during the critical months of August, September, and October, when the Atlantic season peaks. This has a couple of effects in the Atlantic. First, through a combination of weaker trade winds and westerlies over the tropical Atlantic, La Niña tends to create a more placid atmosphere. This leads to less wind shear, which is favorable for the development and strengthening of tropical storms and hurricanes.

 

Secondly, there is a growing amount of data to support the idea that Atlantic storms tend to recurve into the open ocean less during La Niña years, with fewer systems becoming "fish storms" that do not interact with landmasses. Rather, a stronger high-pressure system such as the Bermuda High steers such storms westward into the Caribbean Sea or Gulf of Mexico—and potentially toward the United States.

 

Another signal from the oceans is the extremely warm temperatures in the tropical Atlantic right now. Due at least in part to climate change, the ocean is anomalously warm, with seas closer to August "normals" than what is to be expected in late May.

 

So, if you put together sizzling seas and low wind shear, you get a recipe for lots of hurricane activity.

Blockbuster forecasts

We've already seen this in seasonal forecasts from other outfits. The venerable forecast team at Colorado State University, led by Phil Klotzbach, has predicted an "extremely active" 2024 season with 23 named storms and 11 hurricanes. Each of these values is about 60 percent higher than a typical season.

 

One of the pre-season forecasts that most caught my attention came from the University of Pennsylvania, led by scientists Michael Mann, Shannon Christiansen, and Michael Kozar. They predict an astounding 33 named storms in the tropical Atlantic this year. This would eclipse the previous record of 30 named storms, set in 2020.

 

It might be easy to dismiss the Pennsylvania forecast. In some corners of the Internet, Mann is viewed as a master of climate doomerism for his outspoken views on the perils of a warming world. But if anything, since its first issuance in 2007, Mann's forecast has proven to be conservative. In 2020, for example, the team predicted 20 named storms. So it's a sobering outlook.

 

If you prefer a purely numerical prediction, the European Centre for Medium-Range Weather Forecasts—which produces the most reliable computational weather forecasts in the world—has been issuing seasonal hurricane outlooks for three decades.

 

As the center's time series indicates, the seasonal forecast is far from perfect. (Most hurricane season forecasting, it must be said, is part art and part science.) But it is one of the best forecasts available, and for 2024, it made its highest prediction ever. The European model calls for an "accumulated cyclone energy" forecast twice as high as a normal year. This means that the duration and intensity of tropical activity are expected to double that of a typical season.

 

As always, a busy hurricane season does not guarantee that any part of the United States, Mexico, Central America, or the Caribbean Islands will get hit. Rather, it simply loads the dice, increasing the odds of a strike during a season. As a coastal resident, I don't particularly like those odds.

 

One-quarter of the world’s population is currently water-stressed, using up almost their entire fresh water supply each year. The UN predicts that by 2030, this will climb to two-thirds of the population.

 

Freshwater is perhaps the world’s most essential resource, but climate change is enhancing its scarcity. An unexpected source may have the potential to provide some relief: offshore aquifers, giant undersea bodies of rock or sediment that hold and transport freshwater. But researchers don’t know how the water gets there, a question that needs to be resolved if we want to understand how to manage the water stored in them.

 

For decades, scientists have known about an aquifer off the US East Coast. It stretches from Martha’s Vineyard to New Jersey and holds almost as much water as two Lake Ontarios. Research presented at the American Geophysical Union conference in December attempted to explain where the water came from—a key step in finding out where other undersea aquifers lie hidden around the world.

 

As we discover and study more of them, offshore aquifers might become an unlikely resource for drinking water. Learning the water’s source can tell us if these freshwater reserves rebuild slowly over time or are a one-time-only emergency supply.

Reconstructing history

When ice sheets sat along the East Coast and the sea level was significantly lower than it is today, the coastline was around 100 kilometers further out to sea. Over time, freshwater filled small pockets in the open, sandy ground. Then, 10,000 years ago, the planet warmed, and sea levels rose, trapping the freshwater in the giant Continental Shelf Aquifer. But how that water came to be on the continental shelf in the first place is a mystery.

 

New Mexico Institute of Mining and Technology paleo-hydrogeologist Mark Person has been studying the aquifer since 1991. In the past three decades, he said, scientists’ understanding of the aquifer’s size, volume, and age has massively expanded. But they haven’t yet nailed down the water’s source, which could reveal where other submerged aquifers are hiding—if we learn the conditions that filled this one, we could look for other locations that had similar conditions.

 

“We can’t reenact Earth history,” Person said. Without the ability to conduct controlled experiments, scientists often resort to modeling to determine how geological structures formed millions of years ago. “It’s sort of like forensic workers looking at a crime scene,” he said.

 

Person developed three two-dimensional models of the offshore aquifer using seismic data and sediment and water samples from boreholes drilled onshore. Two models involved ice sheets melting; one did not.

 

Then, to corroborate the models, Person turned to isotopes—atoms with the same number of protons but different numbers of neutrons. Water mostly contains Oxygen-16, a lighter form of oxygen with two fewer neutrons than Oxygen-18.

 

Throughout the last million years, a cycle of planetary warming and cooling occurred every 100,000 years. During warming, the lighter ¹⁶O in the oceans evaporated into the atmosphere at a higher rate than the heavier ¹⁸O. During cooling, that lighter oxygen came down as snow, forming ice sheets with lower levels of ¹⁸O and leaving behind oceans with higher levels of ¹⁸O.

 

To determine if ice sheets played a role in forming the Continental Shelf Aquifer, Person explained, you have to look for water that is depleted in ¹⁸O—a sure sign that it came from ice sheets melting at their base. Person’s team used existing global isotope records from the shells of deep-ocean-dwelling animals near the aquifer. (The shells contain carbonate, an ion that includes oxygen pulled from the water).

 

Person then incorporated methods developed by a Columbia graduate student in 2019 that involve using electromagnetic imaging to finely map undersea aquifers. Since saltwater is more electrically conductive than freshwater, the boundaries between the two kinds of water are clear when electromagnetic pulses are sent through the seafloor: saltwater conducts the signal well, and freshwater doesn’t. What results looks sort of like a heat map, showing regions where fresh and saltwater are concentrated.

 

Person compared the electromagnetic and isotope data with his models to see which historical scenarios (ice or no ice) were statistically likely to form an aquifer that matched all the data. His results, which are in the review stage with the Geological Society of America Bulletin, suggest it's very likely that ice sheets played a role in forming the aquifer.

 

“There’s a lot of uncertainty,” Person said, but “it’s the best thing we have going.”

Searching for water

Person’s results don’t mean that ice sheets were the only contributor to filling up the Continental Shelf Aquifer. It could have been connected to onshore groundwater through past underwater channels through sediments or even filled by rain. There’s also a chance that the aquifer is still connected to onshore groundwater, meaning that it is constantly being supplied with new freshwater.

 

The only way to know for sure is to sample the aquifer. After years of submitting proposals, Person’s team secured a $21 million grant from the Integrated Ocean Drilling Program—a marine research collaboration focused on seafloor sediments—to drill offshore in summer 2025. It will be the first-ever hydrogeologically oriented offshore drilling campaign.

 

“I think it’s fabulous that they’re going to do the drilling,” said Deborah Hutchinson, a geologist emeritus at the United States Geological Survey who was not involved in the research. “If it’s not going to provide the answers, I don’t know what will.”

 

Confirming that ice sheets played a role in forming the aquifer would tell scientists to search for other offshore aquifers near historically glaciated locations. But in the meantime, researchers have been finding them in some areas that weren’t recently covered in ice, highlighting the importance of studying other hypotheses about filling the aquifers.

 

Since the electromagnetic technique was developed, researchers have discovered five more aquifers off the coasts of New Zealand, Malta, Israel, Hawaii, and California. There is also a sixth near New England, which may be connected to the Continental Shelf Aquifer, and others are being studied near Hong Kong.

 

With so many offshore aquifers discovered in such a short amount of time, scientists think the total number is much higher.

 

“It’s going to be likely impossible to map all of the [offshore] groundwater systems in the world,” said Aaron Micallef, a geomorphologist at the Monterey Bay Aquarium Research Institute who has collaborated with Person and studied the New Zealand and Malta aquifers. To better understand all offshore aquifers around the world, he explained, scientists need to “look at case studies that are representative of other places.”

 

Understanding offshore aquifers more generally is crucial to determining whether nations could tap into them as climate change and drought threaten traditional water sources. If the aquifers are actively recharging, however slowly, they could represent a potential water resource. If not, they are an emergency fallback—still useful in a drought-stricken world.

 

Micallef, who is from Malta—currently one of the top 10 most water-scarce countries—said that finding offshore water there was largely “based on hope.”

 

It’s not just water-scarce nations that could benefit but also water-intensive coastal metropolises. Person published a paper in the journal Groundwater in 2010 suggesting that the volume of freshwater in the East Coast aquifer is more than two orders of magnitude higher than all the groundwater pumped annually in the US, making it an immense potential resource.

 

Tapping offshore aquifers wouldn’t be cheap. But neither is a world where we truck drinking water into a city like New York, Person explained.

 

While the science is progressing, we’re still far from drawing water from under the sea. There might be unforeseen environmental threats, technology constraints, or legal difficulties from operating in international waters. Also, undersea aquifers don’t keep out 100 percent of saltwater, meaning they would require some desalination. Compared to pure seawater, however, the energy required to get the salt out of this water would be much lower.

 

Difficulties aside, researchers are united in believing that offshore aquifers could become important. Person likes to borrow a quote from Benjamin Franklin: “The worth of water is realized when the well goes dry,” he paraphrased, “which is basically saying you have to spend whatever money it takes to have enough water, because we can’t live without it.”

 

Hannah Richter is a freelance science journalist and student in MIT's Graduate Program in Science Writing. She primarily covers environmental science and astronomy. 

 

There are a couple of ways to read the announcement from the Federal Aviation Administration that it's kicking off a new environmental review of SpaceX's plan to launch the most powerful rocket in the world from Florida.

 

The FAA said on May 10 that it plans to develop an Environmental Impact Statement (EIS) for SpaceX's proposal to launch Starships from NASA's Kennedy Space Center in Florida. The FAA ordered this review after SpaceX updated the regulatory agency on the projected Starship launch rate and the design of the ground infrastructure needed at Launch Complex 39A (LC-39A), the historic launch pad once used for Apollo and Space Shuttle missions.

Dual environmental reviews

At the same time, the US Space Force is overseeing a similar EIS for SpaceX's proposal to take over a launch pad at Cape Canaveral Space Force Station, a few miles south of LC-39A. This launch pad, designated Space Launch Complex 37 (SLC-37), is available for use after United Launch Alliance's last Delta rocket lifted off there in April.

 

On the one hand, these environmental reviews often take a while and could cloud Elon Musk's goal of having Starship launch sites in Florida ready for service by the end of 2025. "A couple of years would not be a surprise," said George Nield, an aerospace industry consultant and former head of the FAA's Office of Commercial Space Transportation.

 

Another way to look at the recent FAA and Space Force announcements of pending environmental reviews is that SpaceX finally appears to be cementing its plans to launch Starship from Florida. These plans have changed quite a bit in the last five years.

 

The environmental reviews will culminate in a decision on whether to approve SpaceX's proposals for Starship launches at LC-39A and SLC-37. The FAA will then go through a separate licensing process, similar to the framework used to license the first three Starship test launches from South Texas.

 

NASA has contracts with SpaceX worth more than $4 billion to develop a human-rated version of Starship to land astronauts on the Moon on the first two Artemis lunar landing flights later this decade. To do that, SpaceX must stage a fuel depot in low-Earth orbit to refuel the Starship lunar lander before it heads for the Moon. It will take a series of Starship tanker flights—perhaps 10 to 15—to fill the depot with cryogenic propellants.

 

Launching that many Starships over the course of a month or two will require SpaceX to alternate between at least two launch pads. NASA and SpaceX officials say the best way to do this is by launching Starships from one pad in Texas and another in Florida.

 

Earlier this week, Ars spoke with Lisa Watson-Morgan, who manages NASA's human-rated lunar lander program. She was at Kennedy Space Center this week for briefings on the Starship lander and a competing lander from Blue Origin. One of the topics, she said, was the FAA's new environmental review before Starship can launch from LC-39A.

 

"I would say we're doing all we can to pull the schedule to where it needs to be, and we are working with SpaceX to make sure that their timeline, the EIS timeline, and NASA's all work in parallel as much as we can to achieve our objectives," she said. "When you're writing it down on paper just as it is, it looks like there could be some tight areas, but I would say we're collectively working through it."

 

Officially, SpaceX plans to perform a dress rehearsal for the Starship lunar landing in late 2025. This will be a full demonstration, with refueling missions, an uncrewed landing of Starship on the lunar surface, then a takeoff from the Moon, before NASA commits to putting people on Starship on the Artemis III mission, currently slated for September 2026.

 

So you can see that schedules are already tight for the Starship lunar landing demonstration if SpaceX activates launch pads in Florida late next year.

Let’s be real

While the environmental review may not be welcome news for SpaceX, let's be real for a moment. The company has a lot of work to do before Starship can reach the Moon. First and foremost, Watson-Morgan said NASA wants to see SpaceX string together dozens of Starship launches with reliable performance from the rocket's methane-fueled engines. SpaceX must also prove it can refuel Starship in orbit and create a a hospitable crew cabin for astronauts.

 

These are serious engineering problems that will take time to solve, but SpaceX's track record suggests its engineers can solve them. SpaceX also needs time to actually build the new Starship launch pads. While the skeleton of a Starship launch tower is in place at LC-39A, the site doesn't yet have any of the propellant storage or fueling infrastructure needed for launch operations.

 

SpaceX's portion of the Artemis program, while significant, is just one piece of the puzzle. Axiom Space, another NASA contractor, must complete the development of a new spacesuit before astronauts make the first Artemis lunar landing. There's no assurance that NASA's chronically delayed Space Launch System rocket and Orion spacecraft, which will ferry astronauts to the vicinity of the Moon to meet up with the Starship lander, will be ready for a lunar landing mission in the next few years.

 

The schedule for the Artemis III lunar landing in 2026 seems completely unrealistic, which has prompted NASA officials to consider revamping the flight plan for Artemis III to have an Orion crew capsule dock with Starship in low-Earth orbit. This would be achievable well before a lunar landing mission, allowing NASA and SpaceX to reduce risk for the later Moon flight and close a potential years-long gap between Artemis missions, which could threaten political support for the program.

 

 

There is widespread agreement among government and industry officials that the FAA's commercial space office is underfunded to meet the demands of the rapidly growing space industry. Last fall, SpaceX called for Congress to add money to the FAA's budget to double the staff of the commercial space office. The FAA is charged with ensuring public safety during commercial rocket launches, along with compliance with environmental laws.

 

"The fact that it can take years to go through this process is not a good thing in terms of how fast commercial space and other activities would like to be able to go," Nield said.

 

Last fall, a SpaceX official told Ars that FAA licensing was a "critical path item for the Artemis program." Before the second Starship test flight last November, SpaceX waited a couple of months for the FAA to issue a launch license.

 

"In terms of timing, yes, environmental reviews can often be the long pole," Nield said. "I don't think this will end up being the critical factor in the schedule for Artemis missions."

 

Let's imagine a scenario in which SpaceX has proven it can refuel Starship in orbit and has completed an uncrewed landing on the Moon, and the spacesuits, the SLS rocket, and Orion spacecraft are ready for a lunar landing mission. But what if SpaceX lacks FAA approval to launch Starship from Florida or runs into lengthy construction delays?

 

In that scenario, Watson-Morgan said it's "conceivable" that SpaceX could launch all of the Starship tanker and depot missions from two launch pads in Texas. "But that is not the current plan," she said. "The plan is to have LC-39A at KSC up and operational. SpaceX is kind of a just-in-time, rapid-iteration-type company, so they are focused right now on the pads at Starbase."

 

"There are different ways you could do this, and SpaceX has a history of moving pretty fast in putting together places to operate, so we'll see how that comes out," Nield said. "There are a lot of things that have to happen right in order to meet the kinds of schedules that people are aiming toward."

 

But for lunar landing campaigns, NASA's preference is clearly to cycle between Starship launches in Texas and Florida.

 

"On Artemis III, we anticipate using at least two of the launch sites: one at KSC and one at Starbase," Watson-Morgan said.

Shifting plans at LC-39A

SpaceX signed a 20-year lease with NASA for LC-39A in 2014. Since then, the company has launched 83 missions from this pad since 2017 with its operational Falcon 9 and Falcon Heavy rockets. Early last year, SpaceX reshaped the Cape Canaveral skyline by stacking a more than 450-foot-tall (137-meter) Starship launch tower about 1,000 feet (300 meters) from the existing Falcon 9 launch mount at LC-39A.

 

Then, seemingly as quickly as SpaceX put it up, work stalled on the new Starship launch tower for most of last year. Finally, in the last few months, some construction work has resumed at the site. The location of the Starship launch tower close to the existing Falcon 9 launch pad raised some concerns within NASA that an explosion could damage the only launch site capable of accommodating astronaut missions to the International Space Station. In response, SpaceX modified a nearby launch pad for crew and cargo missions.

 

At LC-39A, onlookers have spotted construction crews disassembling the legs already in place for the Starship launch mount. Presumably, SpaceX is redesigning the launch mount and modifying the foundation of the Starship pad for a water deluge system. These changes are based on lessons learned from Starship test flights from Texas.

 

NASA and SpaceX already completed an environmental assessment in 2019 for launching Starship from LC-39A. Since then, the full-scale Super Heavy booster and Starship rocket have launched three times from SpaceX's Starbase facility located at Boca Chica Beach, Texas, just north of the US-Mexico border. A fourth test flight could happen at the end of May or early next month.

 

 

With Starship now flying, SpaceX has a better handle on how it will operate the rocket in Florida. In its notice announcing the pending environmental review, the FAA said SpaceX's concept of operations has "evolved" since 2019. Rather than the previous estimate of 24 Starship launches per year at LC-39A, SpaceX now projects a higher tempo of up to 44 flights per year. SpaceX also proposes launching a more powerful version of the Super Heavy booster and Starship, with up to 35 engines on the first stage and up to nine engines on the upper stage, up from 31 and six.

 

"As they gain experience and start to firm up their plans, it looks like their planned launches don't necessarily fall within that previous scenario," Nield said. "So in terms of size, the exact location, and the launch frequency, that's no longer being covered by the work that had been done previously. What that means is, at the very least, you need to go back and potentially modify the existing assessments."

Construction on hold?

SpaceX plans to land the reusable Super Heavy first stage booster back at the launch site rather than downrange on a drone ship. To do this, the company wants to build a catch tower that will use articulating arms to grab onto the rocket as it slows to a hover just above the ground. This is a separate tower from the Starship launch pad structure currently in place at LC-39A. Other new facilities SpaceX wants to build at LC-39A include a natural gas liquefaction system, an air separation unit for propellant generation, and a water deluge system to dampen the force of liftoff.

 

It's not clear how much of this construction can proceed while the FAA's environmental review remains open. NASA said it has only provided approval for SpaceX to build Starship launch infrastructure at LC-39A that is within the scope of the 2019 environmental assessment.

 

"SpaceX, as a tenant leaseholder on Kennedy Space Center, is not authorized to undertake any construction activity at LC-39A related to Starship launch and landing facilities without prior approval from NASA," NASA said in a statement to Ars. "This EIS will support future decision-making on, among other things, construction activities that may be proposed to occur at LC-39A."

 

An EIS, like those underway at two sites in Florida, is the most thorough type of environmental review the FAA can undertake to comply with the National Environmental Policy Act (NEPA) of 1969. The EIS process takes longer than environmental assessments like the one NASA concluded in 2019 for SpaceX's original proposal to launch Starship from LC-39A.

 

"In consideration of SpaceX’s proposal, NASA, as the land management agency, and the FAA, as the licensing agency, have determined that an EIS is the appropriate level (of review) to address the potential environmental impacts of Starship-Super Heavy operations at LC-39A," a NASA spokesperson told Ars.

 

An FAA spokesperson said the agency expects to issue a draft version of the EIS for LC-39A sometime next year but didn't provide a more specific timeline. This will be followed by a public comment period before the FAA publishes the final report and decides whether SpaceX can launch Starships out of LC-39A.

 

The Space Force projects that a similar ongoing environmental review at SLC-37 will take at least a year and a half. This review started early this year, so the estimated completion date is in September 2025. If the FAA's review follows the same timeline, the end of 2025 seems like a best-case scenario for a decision on LC-39A.

 

However, Nield said the engineers, scientists, and regulators reviewing SpaceX's proposal for LC-39A will have access to the results of the environmental assessment NASA completed in 2019.

 

"I think they're in a much better position than starting from scratch," he said. "At the same time, this is a much bigger effort that is being proposed now, so it's not just a question of 'let's change this number to that, or we'll broaden the number of plants or animals or fish or whatever that might be impacted.'"

Where to build and launch

SpaceX's iterative approach to designing and testing Starship centers on a spiral development model. Essentially, it's a cycle of build, test, find, and fix, where SpaceX assembles a full-size Starship rocket, launches it, finds what went wrong, and then fixes the problems on the next rocket. For this model to work, the company must have a lot of rocket hardware on hand. SpaceX is expanding its sprawling campus of assembly buildings and high bays at Starbase, with several Starships well into production for future test flights. One of these buildings called the Starfactory, will spread across a million square feet to produce components for multiple Starships per week.

 

"This is now going to be a permanent location for us to be producing and launching vehicles," said Kathy Lueders, SpaceX's general manager at Starbase, in a talk earlier this week. SpaceX is also putting in an office building for SpaceX employees at Starbase.

 

"Then, we'll probably be in the process of building another high bay," she said. "We are building a second pad right now, too. All of this is to get us ready to be able to start meeting the production and launch rate that we need to be able to accomplish our missions."

 

SpaceX eventually aims to launch Starships more often than it flies the Falcon 9 rocket, which has launched at an average cadence of once every 2.7 days since the start of the year. With design upgrades, a reusable Starship could deliver more than 100 metric tons of payload to low-Earth orbit.

 

Lueders said SpaceX has invested more than $3 billion at Starbase since 2014. The company is about to start construction of a second launch pad at Starbase next to the existing tower. More than 3,000 SpaceX employees and contractors work at Starbase every day, she said.

 

"We need two launch areas for us to be able to meet our manifest," Lueders said. "Just a single (Moon) landing requires 15 tanker launches, and they need to be done within a certain period of time."

 

 

In her remarks to local officials in South Texas, Lueders said Starbase will remain SpaceX's "workhorse area" for Starship development and testing. Musk has said he envisions the Texas location as a research and development complex and Florida as an operational launch base.

 

"We will also need the Florida base to be able to do the number and sequencing of the missions. We're trying to figure out how to do the mix," Lueders said. "We're still working through this but ... we know Starbase is going to be the home of Starship."

 

There's currently no capacity to build Starships in Florida and no evidence of imminent construction of a Starship factory at Cape Canaveral.

 

The Cape Canaveral spaceport is a draw for almost every US launch company. Starship's arrival will "unlock a lot more potential at places like Cape Canaveral that are already world leaders in terms of space launch," said Rob Long, president and CEO of Space Florida, a state-backed enterprise established to attract aerospace business.

 

Eventually, SpaceX may build a rocket factory on Florida's Space Coast, but if Starships are really going to launch from the Kennedy Space Center in the next couple of years, the company will need to ship the 30-foot-diameter (9-meter) vehicles from the factory in South Texas. Officials haven't said how they will transport boosters and the ship 1,000 miles (1,600 km) across the Gulf of Mexico.

 

Whatever SpaceX decides, Long said Cape Canaveral is uniquely positioned for Starship. With a deepwater seaport, air, road, and rail connections, plus access to space, the region is home to the nation's only "quintemodal" port.

 

"I think that really leans into this thought of normalizing space launch as a true mode of transportation, just like air, land, or sea modes of transportation," he said.

 

SpaceX's shifting plans for Starship in Florida are not a big surprise to anyone who has watched the arc of the company's development. At one time, SpaceX acquired oceangoing oil rigs to convert them into mobile launch and landing pads for Starship. For now, SpaceX has dropped that idea.

 

The three Starship test flights over the last year have each been valuable learning exercises, influencing design changes that SpaceX will incorporate into future launch pads. In that context, SpaceX's decision to temporarily halt construction at LC-39A makes some sense.

 

SpaceX also abandoned a plan to build a Starship launch pad at an undeveloped site, known as Launch Complex 49 (LC-49), on the northern side of the Kennedy Space Center. In late 2021, NASA announced it was beginning the environmental assessment process for LC-49. Early this year, NASA said it suspended those plans.

 

Instead, SpaceX has proposed basing Starship flights at SLC-37, which sits on Space Force property. If SpaceX doesn't get approval to launch there, it proposed a backup option of building a brand new launch pad called SLC-50 just north of SLC-37.

 

With two launch pads at Starbase, plus LC-39A and SLC-37 in Florida, SpaceX would have four active Starship launch pads. This should be enough to support the kind of Starship launch rate SpaceX needs to achieve through the rest of this decade for test flights, lunar missions supporting NASA's Artemis program, and launches of Starlink broadband satellites.

 

Ultimately, if Musk's grandest ambitions for a Mars settlement become real, SpaceX probably needs a lot more launch pads.

 

 

An asteroid that has been wandering through space for billions of years is going to have been bombarded by everything from rocks to radiation. Billions of years traveling through interplanetary space increase the odds of colliding with something in the vast emptiness, and at least one of those impacts had enough force to leave the asteroid Ryugu forever changed.

 

When the Japanese Space Agency’s Hayabusa2 spacecraft touched down on Ryugu, it collected samples from the surface that revealed that particles of magnetite (which is usually magnetic) in the asteroid’s regolith are devoid of magnetism. A team of researchers from Hokkaido University and several other institutions in Japan are now offering an explanation for how this material lost most of its magnetic properties. Their analysis showed that it was caused by at least one high-velocity micrometeoroid collision that broke the magnetite’s chemical structure down so that it was no longer magnetic.

 

“We surmised that pseudo-magnetite was created [as] the result of space weathering by micrometeoroid impact,” the researchers, led by Hokkaido University professor Yuki Kimura, said in a study recently published in Nature Communications.

What remains…

Ryugu is a relatively small object with no atmosphere, which makes it more susceptible to space weathering—alteration by micrometeoroids and the solar wind. Understanding space weathering can actually help us understand the evolution of asteroids and the Solar System. The problem is that most of our information about asteroids comes from meteorites that fall to Earth, and the majority of those meteorites are chunks of rock from the inside of an asteroid, so they were not exposed to the brutal environment of interplanetary space. They can also be altered as they plummet through the atmosphere or by physical processes on the surface. The longer it takes to find a meteorite, the more information can potentially be lost.

 

Once part of a much larger body, Ryugu is a C-type, or carbonaceous, asteroid, meaning it is made of mostly clay and silicate rocks. These minerals normally need water to form, but their presence is explained by Ryugu’s history. It is thought that the asteroid itself was born from debris after its parent body was smashed to pieces in a collision. The parent body was also covered in water ice, which explains the magnetite, carbonates, and silicates found on Ryugu—these need water to form.

 

Magnetite is a ferromagnetic (iron-containing and magnetic) mineral. It is found in all C-type asteroids and can be used to determine their remanent, or remaining, magnetization. The remanent magnetization of an asteroid can reveal how intense the magnetic field was at the time and place of the magnetite’s formation.

 

Kimura and his team were able to measure remanent magnetization in two magnetite fragments (known as framboids because of their particular shape) from the Ryugu sample. It is proof of a magnetic field in the nebula our Solar System formed in, and shows the strength of that magnetic field at the time that the magnetite formed.

 

However, three other magnetite fragments analyzed were not magnetized at all. This is where space weathering comes in.

…and what was lost

Using electron holography, which is done with a transmission electron microscope that sends high-energy electron waves through a specimen, the researchers found that the three framboids in question did not have magnetic chemical structures. This made them drastically different from magnetite.

 

Further analysis with scanning transmission electron microscopy showed that the magnetite particles were mostly made of iron oxides, but there was less oxygen in those particles that had lost their magnetism, indicating that the material had experienced a chemical reduction, where electrons were donated to the system. This loss of oxygen (and oxidized iron) explained the loss of magnetism, which depends on the organization of the electrons in the magnetite. This is why Kimura refers to it as “pseudo-magnetite.”

 

But what triggered the reduction that demagnetized the magnetite in the first place? Kimura and his team found that there were more than a hundred metallic iron particles in the part of the specimen that the demagnetized framboids had come from. If a micrometeorite of a certain size had hit that region of Ryugu then it would have produced approximately that many particles of iron from the magnetite framboids. The researchers think this mystery object was rather small, or it would have had to have been moving incredibly fast.

 

"With increasing impact velocity, the estimated projectile size decreases," they said in the same study.

 

Pseudo-magnetite might sound like an imposter, but it will actually help upcoming investigations that seek to find out more about what the early Solar System was like. Its presence indicates the former presence of water on an asteroid, as well as space weathering, such as micrometeoroid bombardment, that affected the asteroid’s composition. How much magnetism was lost also affects the overall remanence of the asteroid. Remanence is important in determining an object’s magnetism and the intensity of the magnetic field around it when it formed. What we know of the Solar System’s early magnetic field has been reconstructed from remanence records, many of which come from magnetite.

 

Some magnetic properties of those particles might have been lost eons ago, but so much more could be gained in the future from what remains.

 

Nature Communications, 2024.  DOI: 10.1038/s41467-024-47798-0