In 2017, the EHT collaboration used a network of seven pre-existing telescopes around the world to gather data on M87, creating an “Earth-sized telescope.” However, since it is infeasible to cover the Earth’s entire surface with telescopes, gaps arise in the data — like missing pieces in a jigsaw puzzle.

“With our new machine learning technique, PRIMO, we were able to achieve the maximum resolution of the current array,” says lead author Lia Medeiros of the Institute for Advanced Study. “Since we cannot study black holes up-close, the detail of an image plays a critical role in our ability to understand its behavior. The width of the ring in the image is now smaller by about a factor of two, which will be a powerful constraint for our theoretical models and tests of gravity.”

PRIMO, which stands for principal-component interferometric modeling, was developed by EHT members Lia Medeiros (Institute for Advanced Study), Dimitrios Psaltis (Georgia Tech), Tod Lauer (NOIRLab), and Feryal Özel (Georgia Tech). Their publication, “The Image of the M87 Black Hole Reconstructed with PRIMO,” is now available in The Astrophysical Journal Letters.

“PRIMO is a new approach to the difficult task of constructing images from EHT observations,” said Lauer. “It provides a way to compensate for the missing information about the object being observed, which is required to generate the image that would have been seen using a single gigantic radio telescope the size of the Earth.”

PRIMO relies on dictionary learning, a branch of machine learning which enables computers to generate rules based on large sets of training material. For example, if a computer is fed a series of different banana images — with sufficient training — it may be able to determine if an unknown image is or is not a banana. Beyond this simple case, the versatility of machine learning has been demonstrated in numerous ways: from creating Renaissance-style works of art to completing the unfinished work of Beethoven. So how might machines help scientists to render a black hole image? The research team has answered this very question.

With PRIMO, computers analyzed over 30,000 high-fidelity simulated images of black holes accreting gas. The ensemble of simulations covered a wide range of models for how the black hole accretes matter, looking for common patterns in the structure of the images. The various patterns of structure were sorted by how commonly they occurred in the simulations, and were then blended to provide a highly accurate representation of the EHT observations, simultaneously providing a high fidelity estimate of the missing structure of the images. A paper pertaining to the algorithm itself was published in The Astrophysical Journal on February 3, 2023.

“We are using physics to fill in regions of missing data in a way that has never been done before by using machine learning,” added Medeiros. “This could have important implications for interferometry, which plays a role in fields from exo-planets to medicine.”

The team confirmed that the newly rendered image is consistent with the EHT data and with theoretical expectations, including the bright ring of emission expected to be produced by hot gas falling into the black hole. Generating an image required assuming an appropriate form of the missing information, and PRIMO did this by building on the 2019 discovery that the M87 black hole in broad detail looked as predicted.

“Approximately four years after the first horizon-scale image of a black hole was unveiled by EHT in 2019, we have marked another milestone, producing an image that utilizes the full resolution of the array for the first time,” stated Psaltis. “The new machine learning techniques that we have developed provide a golden opportunity for our collective work to understand black hole physics.”

The new image should lead to more accurate determinations of the mass of the M87 black hole and the physical parameters that determine its present appearance. The data also provides an opportunity for researchers to place greater constraints on alternatives to the event horizon (based on the darker central brightness depression) and perform more robust tests of gravity (based on the narrower ring size). PRIMO can also be applied to additional EHT observations, including those of Sgr A*, the central black hole in our own Milky Way galaxy.

M87 is a massive, relatively nearby, galaxy in the Virgo cluster of galaxies. Over a century ago, a mysterious jet of hot plasma was observed to emanate from its center. Beginning in the 1950s, the then new technique of radio astronomy showed the galaxy to have a compact bright radio source at its center. During the 1960s, M87 had been suspected to have a massive black hole at its center powering this activity. Measurements made from ground-based telescopes starting in the 1970s, and later the Hubble Space Telescope starting in the 1990s, provided strong support that M87 indeed harbored a black hole weighing several billion times the mass of the Sun based on observations of the high velocities of stars and gas orbiting its center. The 2017 EHT observations of M87 were obtained over several days from several different radio telescopes linked together at the same time to obtain the highest possible resolution. The now iconic “orange donut” picture of the M87 black hole, released in 2019, reflected the first attempt to produce an image from these observations.

“The 2019 image was just the beginning,” stated Medeiros. “If a picture is worth a thousand words, the data underlying that image have many more stories to tell. PRIMO will continue to be a critical tool in extracting such insights.”

Development of the PRIMO algorithm was enabled through the support of the National Science Foundation Astronomy and Astrophysics Postdoctoral Fellowship.

“Our attraction to people who share our attributes is aided by the belief that those shared attributes are driven by something deep within us: one’s essence,” said lead author Charles Chu, PhD, an assistant professor at the Boston University Questrom School of Business. “To put it concretely, we like someone who agrees with us on a political issue, shares our music preferences, or simply laughs at the same thing as us not purely because of those similarities, but because those similarities suggest something more — this person is, in essence, like me, and as such, they share my views of the world at large.”

This thought process is driven by a type of psychological essentialism that is applied specifically to people’s ideas about the self and individual identity, according to Chu, adding that people “essentialize” many things — from biological categories such as animal species to social groups such as race and gender — and do so in virtually all human cultures.

“To essentialize something is to define it by a set of deeply rooted and unchanging properties, or an essence,” said Chu. “For example, the category of ‘wolf’ is defined by a wolf essence, residing in all wolves, from which stems attributes such as their pointy noses, sharp teeth and fluffy tails as well as their pack nature and aggressiveness. It is unchanging in that a wolf raised by sheep is still a wolf and will eventually develop wolf-like attributes.”

Recently, researchers have begun to focus on the category of the self and have found that just as we essentialize other categories, we essentialize the self, according to Chu.

“To essentialize me is to define who I am by a set of entrenched and unchanging properties, and we all, especially in Western societies, do this to some extent. A self-essentialist then would believe that what others can see about us and the way we behave are caused by such an unchanging essence,” he said.

To better understand how self-essentialism drives attraction between individuals, researchers conducted a series of four experiments. The research was published in the Journal of Personality and Social Psychology.

In one experiment, 954 participants were asked their position on one of five randomly assigned social issues (abortion, capital punishment, gun ownership, animal testing, or physician-assisted suicide). Half the participants then read about another individual who agreed with their position, while the other half read about an individual who disagreed with their position. All participants then completed a questionnaire on how much they believed they shared a general view of the world with the fictitious individual, their level of interpersonal attraction to that person and their overall beliefs in self-essentialism.

Researchers found that participants who scored high on self-essentialism were more likely to express an attraction to the fictitious individual who agreed with their position and to report a shared general perception of reality with that individual.

A similar experiment involving 464 participants found the same results for a shared attribute as simple as the participants’ propensity to overestimate or underestimate a number of colored dots on a series of computer slides. In other words, the belief in an essential self led people to assume that just a single dimension of similarity was indicative of seeing the entire world in the same way, which led to more attraction.

In another experiment, 423 participants were shown eight pairs of paintings and asked which in each pair they preferred. Based on their responses, participants were identified as either a fan of the Swiss-German artist Paul Klee or the Russian painter Wassily Kandinsky. Half of each fan group was then told that artistic preference was part of their essence; the other half was told it had no connection. All were then exposed to two hypothetical individuals, one of whom had the same artistic preference and one who differed. Participants who were told that artistic preference was connected to their essence were significantly more likely to express an attraction to a hypothetical person with the same artistic preferences than those who were told artistic preference had nothing to do with their essence.

A final experiment categorized 449 participants as fans of one of the two artists and then presented them with information about whether using one’s own essence was useful or not in perceiving other people. This time, one-third of the participants were told that essentialist thinking could lead to inaccurate impressions of others, one-third were told that essentialist thinking could lead to accurate impressions of others and the final third were given no information.

As expected, researchers found participants who were told that essentialist thinking could lead to accurate impressions of others were more likely to report attraction to and shared reality with hypothetical individuals with similar art preferences.

Chu said he was most surprised to find that something as minimal as a shared preference for an artist would lead people to perceive that another individual would see the world the same way as they do. Self-essentialist thinking, though, could be a mixed blessing, he warned.

“I think any time when we’re making quick judgments or first impressions with very little information, we are likely to be affected by self-essentialist reasoning,” said Chu. “People are so much more complex than we often give them credit for, and we should be wary of the unwarranted assumptions we make based on this type of thinking.”

Scientists at Scripps Institution of Oceanography at UC San Diego and the Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) report that extremely intense days of rain or snow will be more frequent by the end of this century than previously thought — as often as once every 30 or 40 years in the Pacific Northwest and southeastern United States.

The conclusions come from analyzing a 30-terabyte data set that models temperature and precipitation at scales roughly the size of urban ZIP codes: six kilometers (3.9 miles). Researchers developed the data set, called Localized Constructed Analogs Version 2 (LOCA2), to provide climate information that is useful for local planners. In contrast, most of the existing advanced climate models look at regions that range from 50 to 250 kilometers (30 to 400 miles).

“With this data set, we’re able to look at the impacts of actual weather pattern changes across the United States at an extremely granular level,” said Dan Feldman, staff scientist at Berkeley Lab and the project’s principal investigator. “We see that there is a lot more extreme weather that is likely to happen in the future — and by looking at actual weather patterns, we show that changes in extreme precipitation will actually be more extreme than previously estimated. Land use managers and planners should expect more extremes, but location matters.”

The LOCA2 data set updates a similar analysis conducted in 2016 in advance of the Fourth National Climate Assessment (NCA), which was released in 2018 by the U.S. Global Change Research Program. The NCA is intended to assist the U.S. government with planning for, mitigating, and adapting to changes in climate that will affect the country. The Fifth NCA is expected to be issued later this year.

LOCA2 projections cover the lower 48 states of the United States, southern Canada, and northern Mexico. The data set draws on more than 70 years of weather data and incorporates 27 updated climate models from the Coupled Model Intercomparison Project (CMIP6), the latest iteration of an international effort to simulate climate that includes the “coupling” of natural systems such as the ocean and atmosphere to understand how they will act in concert as climate changes.

“We’ve spent a lot of effort improving the representation of extreme wet days, which is important for understanding both the likelihood of flooding and the availability of water for agricultural, commercial, and residential use,” said David Pierce, a scientist at Scripps Oceanography and the developer of LOCA and LOCA2.

The LOCA2 climate projections are available through the end of the century down to the daily level, and for three different greenhouse gas emissions scenarios known as SSPs, or Shared Socioeconomic Pathways. The three scenarios are a medium level of emissions that is slightly less than current levels (SSP 245), medium-high (SSP 370), and high, where emissions greatly increase (SSP 585). The data set is freely available for planners and decision makers to use.

The projection reinforces what climate scientists have long predicted: Future weather events will become more extreme in a warming world. LOCA2 finds that the heaviest days of rain and snowfall across much of North America will likely release 20 to 30 percent more moisture than they do now. Much of the increased precipitation will occur in winter, potentially exacerbating flooding in regions such as the upper Midwest and the west coast.

“The big picture is clear: it’s getting warmer and wetter,” Feldman said. “This research translates that bigger picture into more practical data for infrastructure and operations planning. With this more detailed look at local impacts, we can help local officials make better-informed decisions, such as how long to make an airport runway, how much resilience to include for constructing buildings or bridges, or where to put crops or culverts.”

The improved set of LOCA2 data was created by better identifying and preserving extreme weather events in the past, training models to more accurately reflect extremes in simulations of the future.

“We undertook a Herculean effort of personnel and computer time not just to produce a bunch of numbers, but to produce local projections that are relevant and useful,” Feldman said. “We do so by recognizing how heat waves and storms have occurred and will occur at the local level, and projecting those forward.”

Seasonal and regional predictions

While the data varies at the local level, researchers found substantial trends across the area covered by LOCA2 at the end of the century.

Across most seasons, a major part of North America will see roughly the same or fewer number of days with precipitation, roughly the same or fewer number of days with light and medium amounts of precipitation, and a large increase in the number of days with the most extreme precipitation (the top 1 percent and 0.1 percent of storms).

“People will be more affected by the really rare and most extreme events, because those are showing the biggest increase,” said Pierce, who is the lead author of the paper on extreme precipitation published in the Journal of Hydrometeorology. “The wettest day you would expect to see in five years, or 50 years, or 500 years — those extreme events are going to be substantially wetter, and that’s a really big issue, because it has implications for flooding and run-off.”

Southern Canada and most of the United States will see increases in extreme precipitation days that occur primarily in winter. The wettest days of precipitation will increase by 20-30 percent, depending on the emissions scenario and how extreme the storm is.

Arizona, New Mexico, and northern Mexico can expect increases in extreme precipitation days that occur primarily in autumn. The wettest days of precipitation increase by 10-30 percent, depending on which emissions scenarios come to be and how extreme the storms are. While the region becomes drier overall, the number of days with extreme precipitation events still goes up, meaning the precipitation that does come will often do so in larger storms.

“It’s quite interesting that you see the same kind of pattern of fewer low- and medium- precipitation days and more extreme precipitation days across pretty much the entire country,” Pierce said. Knowing the changing character of precipitation and the frequency of extreme events is useful in two ways, Pierce added. “One is for building new infrastructure in the future, and one is for understanding impacts upon existing facilities already there.”

Funding for this research was provided by the Department of Defense and Department of Energy through the Strategic Environmental Research and Development Program (SERDP). The NASA High-End Computing Capability (HECC) Program provided resources supporting this work through the NASA Earth Exchange (NEX), Earth Science Division, and the NASA Advanced Supercomputing (NAS) Division at Ames Research Center.

“Soil microbes are beneficial, and we couldn’t live without their cycling of carbon and nutrients, but climate change and drought can tweak that balance, and we have to be aware of how it’s changing,” says Allison of the University of California, Irvine.

Some soil microbes take carbon from decomposing plants and store it in the soil, while others release plant carbon back into the atmosphere. The carbon that ends up in the soil is beneficial in multiple ways. “The carbon in the soil has these reverberating effects out to the rest of the world in terms of the infrastructure in our natural and managed ecosystems,” says Allison. “Carbon-rich soils hold more nutrients, so plants growing in those soils tend to be more productive, and the carbon changes the physical properties of the soil, which prevents erosion.”

“In California now, we have this system where the droughts are more intense, and then the rainfall is more intense,” he says. “So, if you’re losing your soil carbon, when it rains really hard it could carry away your soil and cause erosion, landslides, mudslides, sediments, and all kinds of problems that we’re actually seeing right now.”

The carbon that is released back into the atmosphere is another story. “From a climate mitigation standpoint, what we want is for more carbon to be in plants and soils and less carbon to be in the atmosphere, so the more carbon we can absorb into plants through photosynthesis and the more we can transfer and keep in the soil, the better off we’re going to be in terms of climate change,” says Allison. “That’s why it’s really important to know how the balance of incoming versus outflowing carbon changes with drought, or warming, or any other climate factor.”

Plants and microbes will both be impacted by the increasing frequency of drought, but Allison suspects that microbes will be able to bounce back faster. “Microbes are really adaptable — they can change their physiology, they can change their abundances so that more drought-adapted microbes take over, and they can potentially evolve — so we expect that they are going to resist or bounce back from drought,” says Allison. “All those different processes can happen pretty quickly with microbes, and much more quickly than with plants.”

If more carbon-releasing microbes survive than carbon-sequestering microbes, we could end up with carbon-depleted soils, which would have serious negative implications for plant productivity and future greenhouse gas levels.

We may be able to nudge the balance in the right direction, Allison says, but more research is needed first. “There’s still a lot to be done. Right now, we have data that suggests that when we have drought, something changes that results in carbon loss, but we don’t understand exactly how or why that’s happening, whether drought’s changing the abundance of beneficial plant associated microbes versus the carbon releasing microbes, or if it’s causing the evolution of one of the microbe groups, or if it’s more determined by changes to their immediate physiology,” says Allison.

Some microbes could actually help plants cope with drought. If we knew which microbes were most beneficial to plants, and most likely to retain carbon in soil, we could try to tip the balance in their favor.

“There’s a lot of potential for us to manage or engineer soil microbes,” says Allison. “In agricultural systems, we can look into manipulating the soil or adding beneficial microbes back in. In more natural systems, management would probably be on the plant side: soil microbes are often closely intertwined with plants, so managing the plants can also benefit the microbial part of the ecosystem.”

“We also need more measurements to get a good sense of how drought affects soil carbon change in different ecosystems,” says Allison. “There’s a lot of landscape out there — from the Arctic tundra to the deserts — and we could use more research across those diverse habitats.”

Materials that strongly change their resistivity under magnetic fields are highly sought for various applications and, for example, every car and every computer contain many tiny magnetic sensors. Such materials are rare, and most metals and semiconductors change their electrical resistivity only by a tiny fraction of a percent at room temperature and in practically viable magnetic fields (typically, by less than a millionth of 1 %). To observe a strong magnetoresistance response, researchers usually cool materials to liquid-helium temperatures so that electrons inside scatter less and can follow cyclotron trajectories.

Now a research team led by Professor Sir Andre Geim has found that good old graphene that seemed to be studied in every detail over the last two decade exhibits a remarkably strong response, reaching above 100% in magnetic fields of standard permanent magnets (of about 1,000 Gauss). This is a record magnetoresistivity among all the known materials.

Speaking about this latest graphene discovery, Sir Andre Geim said: “People working on graphene like myself always felt that this gold mine of physics should have been exhausted long ago. The material continuously proves us wrong finding yet another incarnation. Today I have to admit again that graphene is dead, long live graphene.”

To achieve this, the researchers used high-quality graphene and tuned it to its intrinsic, virgin state where there were only charge carriers excited by temperature. This created a plasma of fast-moving “Dirac fermions” that exhibited a surprisingly high mobility despite frequent scattering. Both high mobility and neutrality of this Dirac plasma are crucial components for the reported giant magnetoresistance.

“Over the last 10 years, electronic quality of graphene devices has improved dramatically, and everyone seems to focus on finding new phenomena at low, liquid-helium temperatures, ignoring what happens under ambient conditions. This is perhaps not so surprising because the cooler your sample the more interesting its behaviour usually becomes. We decided to turn the heat up and unexpectedly a whole wealth of unexpected phenomena turned up,” says Dr Alexey Berdyugin, the corresponding authors of the paper.

In addition to the record magnetoresistivity, the researchers have also found that, at elevated temperatures, neutral graphene becomes a so-called “strange metal.” This is the name given to materials where electron scattering becomes ultimately fast, being determined only by the Heisenberg uncertainty principle. The behaviour of strange metals is poorly understood and remains a mystery currently under investigation worldwide.

The Manchester work adds some more mystery to the field by showing that graphene exhibits a giant linear magnetoresistance in fields above a few Tesla, which is weakly temperature dependent. This high-field magnetoresistance is again record-breaking.

The phenomenon of linear magnetoresistance has remained an enigma for more than a century since it was first observed. The current Manchester work provides important clues about origins of the strange metal behaviour and of the linear magnetoresistance. Perhaps, the mysteries can now be finally solved thanks to graphene as it represents a clean, well-characterised and relatively simple electronic system.

“Undoped high-quality graphene at room temperature offers an opportunity to explore an entirely new regime that in principle could be discovered even a decade ago but somehow was overlooked by everyone. We plan to study this strange-metal regime and, surely, more of interesting results, phenomena and applications will follow,” adds Dr Leonid Ponomarenko, one of the leading Nature paper authors.