The research led by Thi Tran, assistant professor of management information systems at Binghamton University’s School of Management, expands on existing studies by offering tools for recognizing patterns in misinformation and helping content creators zero in the worst offenders.

“I hope this research helps us educate more people about being aware of the patterns,” Tran said, “so they know when to verify something before sharing it and are more alert to mismatches between the headline and the content itself, which would keep the misinformation from spreading unintentionally.”

Tran’s research proposed machine learning systems — a branch of artificial intelligence (AI) and computer science that uses data and algorithms to imitate the way humans learn while gradually improving its accuracy — to help determine the scale to which content could cause the most harm to its audience.

Examples could include stories that circulated during the height of the COVID-19 pandemic touting false alternate treatments to the vaccine.

The framework would use data and algorithms to spot indicators of misinformation and use those examples to inform and improve the detection process. It would also consider user characteristics from people with prior experience or knowledge about fake news to help piece together a harm index. The index would reflect the severity of possible harm to a person in certain contexts if they were exposed and victimized by the misinformation.

“We’re most likely to care about fake news if it causes a harm that impacts readers or audiences. If people perceive there’s no harm, they’re more likely to share the misinformation,” Tran said. “The harms come from whether audiences act according to claims from the misinformation, or if they refuse the proper action because of it. If we have a systematic way of identifying where misinformation will do the most harm, that will help us know where to focus on mitigation.”

Based on the information gathered, Tran said, the machine learning system could help fake news mitigators discern which messages are likely to be the most damaging if allowed to spread unchallenged.

“Your educational level or political beliefs, among other things, can play a role in whether you are likely to trust one misinformation message or not and those factors can be learned by the machine learning system,” Tran said. “For example, the system can suggest, according to the features of a message and your personality and background and so on, that it’s 70% likely that you’ll become a victim to that specific misinformation message.”

While other studies have been conducted about using blockchain — a type of shared database technology — as a tool to fight fake news, Tran’s research also expands on previous findings by exploring user acceptability of such systems more closely.

Tran proposed surveying 1,000 people from among two groups: fake news mitigators (government organizations, news outlets and social network administrators) and content users who could be exposed to fake news messages. The survey would lay out three existing blockchain systems and gauge the participants’ willingness to use those systems in different scenarios.

Traceability is one of the nice features of blockchain, Tran said, because it can identify and classify sources of misinformation to help with recognizing the patterns.

“The research model I’ve built out allows us to test different theories and then prove which is the best way for us to convince people to use something from blockchain to combat misinformation,” Tran said.

Tran recently presented his research at a conference hosted by SPIE, the international non-profit dedicated to advancing light-based research and technologies. One paper focused on the machine learning-based framework and another paper dealt with the use of blockchain.

Geologists have found evidence of impacts, such as ejecta (material flung far away from the impact), melted rocks, and high-pressure minerals from more than 3.5 billion years ago. But the actual craters from so long ago have remained elusive. The planet’s oldest known impact structures, which is what scientists call these massive craters, are only about 2 billion years old. We’re missing two and a half billion years of mega-craters.

The steady tick of time and the relentless process of erosion are responsible for the gap, according to Matthew S. Huber, a planetary scientist at the University of the Western Cape in South Africa who studies impact structures and led the new study.

“It’s almost a fluke that the old structures we do have are preserved at all,” Huber said. “There are a lot of questions we’d be able to answer if we had those older craters. But that’s the normal story in geology. We have to make a story out of what’s available.”

Geologists can sometimes spot hidden, buried craters using geophysical tools, such as seismic imaging or gravity mapping. Once they’ve identified potential impact structures, they can search for physical remnants of the impact process to confirm its existence, such as ejecta and impact minerals.

The big question for Huber and his team was how much of a crater can be swept away by erosion before the last lingering geophysical traces disappear. Geophysicists have suggested that 10 kilometers (6.2 miles) of vertical erosion would erase even the biggest impact structures, but that threshold had never been tested in the field.

To find out, the researchers dug into one of the planet’s oldest known impact structures: the Vredefort crater in South Africa. The structure is about 300 kilometers (186 miles) across and was formed about 2 billion years ago when an impactor about 20 kilometers (12.4 miles) across slammed into the planet.

The impactor hit with such energy that the crust and mantle rose up where the impact occurred, leaving a long-term dome. Farther from the center, ridges of rock jutted up, minerals transformed and rock melted. And then time took its course, eroding about 10 kilometers (6.2 miles) down from the surface in two billion years.

Today, all that remains at the surface is a semicircle of low hills southwest of Johannesburg, which marks the center of the structure, and some smaller, telltale signs of impact. The bullseye, caused by the uplift of the mantle, appears in gravity maps, but beyond the center, geophysical evidence of the impact is lacking.

“That pattern is one of the last geophysical signatures that is still detectable, and that only happens for the largest-scale impact structures,” Huber said. Because only the deepest layers of the structure remain, the other geophysical traces have disappeared.

But that’s okay, because Huber wanted to know just how reliable those deep layers are for recording ancient impacts from both a mineralogical and geophysical perspective.

“Erosion makes these structures disappear from the top down,” Huber said. “So we went from the bottom up.”

The researchers sampled rock cores across a 22-kilometer (13.7-mile) transect and analyzed their physical properties, searching for differences in density, porosity and mineralogy between impacted and non-impacted rocks. They also modeled the impact event and what its effects on rock and mineral physics would be and compared that to what they saw in their samples.

What they found was not encouraging for the search for Earth’s oldest craters. While some impact melt and minerals remained, the rocks in the outer ridges of the Vredefort structure were essentially indistinguishable from the non-impact rocks around them when viewed through a geophysical lens.

“That was not exactly the result we were expecting,” Huber said. “The difference, where there was any, was incredibly muted. It took us a while to really make sense of the data. Ten kilometers of erosion and all the geophysical evidence of the impact just disappears, even with the largest craters,” confirming what geophysicists had estimated previously.

The researchers caught Vredefort just in time; if much more erosion occurs, the impact structure will be gone. The odds of finding buried impact structures from more than 2 billion years ago are low, Huber said.

“In order to have an Archean impact crater preserved until today, it would have to have experienced really unusual conditions of preservation,” Huber said. “But then, Earth is full of unusual conditions. So maybe there’s something unexpected somewhere, and so we keep looking.”

A collaboration of scientists from the United States and Senegal focused on doing just that by removing overgrown aquatic vegetation from areas upstream of the Diama Dam in northeastern Senegal. In doing so, they generated positive impacts to the local communities’ health and economies.

“It is rare and gratifying when we can find a potential win-win solution to both human health and livelihoods,” said UC Santa Barbara geography professor David López-Carr, a co-author of a paper that appears in the journal Nature. In it, the researchers provide proof for a hypothesis that agricultural activities, including the use of fertilizers, contribute to parasitic infections by fueling the growth of aquatic vegetation. “The results suggest a simple solution to positively impact society at the intersections of health, society and economy of northern Senegal, with implications for the over 700 million people globally in schistosomiasis endemic areas.”

Since the construction of the Diama Dam in 1986, local farmers have had better access to fresh water to irrigate their fields. However, the presence of the new infrastructure also has increased the prevalence of the schistosoma parasite, a tiny freshwater flatworm commonly found in Africa, South America and Southeast Asia. Nearly 250 million people around the world are estimated to be infected with this parasite.

As far as tropical diseases go, schistosomiasis (also known as bilharzia or snail fever) isn’t immediately fatal or even transmissible between people. But in the long term, the condition is debilitating.

“The disease is most prevalent in poor communities lacking potable water and adequate sanitation,” said López-Carr, an anthropogeographer who specializes in human-environment dynamics in the developing world. Adult worms take up residence in blood vessels and lay eggs in tissue, causing reactions and generally wreaking havoc on organs. Long-term effects include increased risk for cancer and infertility, and those infected are less able to work and go to school, keeping them in the cycle of poverty. “Poor farmers can lose up to half of their yields due to infection,” he said.

Health agencies and organizations have been fighting these infections with drugs that work well, however, the medicine does not prevent reinfection, which can happen as soon as the individual encounters contaminated water. Previous research has also focused on using the snails’ natural predators — prawns — which were cut off by the dam.

In their effort to get ahead of the disease, the collaboration took a close look at the habitat that supports the worms’ intermediate host, a small snail that lives in the Senegal River and its tributaries. They found that a common aquatic plant called Ceratophyllum demersum — also known as hornwort — can hold up to 99% of these snails, with which they have a mutualistic relationship.

Exacerbated by fertilizer runoff from agricultural operations farther upstream, c. demersum and other aquatic plants tend to proliferate in local waterways, which impedes access for daily activities such as cooking, irrigation and washing clothes.

For their experiment, the researchers conducted a three-year randomized control trial in 16 communities, to see if and how much nuisance vegetation removal in about half of the communities would affect the presence of the snails. They measured baseline infection rates, administered antiparasitic drugs, removed the vegetation and then measured reinfection rates in more than 1,400 schoolchildren. In total, the research teams took out an estimated 430 metric tons (wet) of aquatic vegetation from water access points.

“In our randomized controlled trial, control sites — places where we didn’t remove submerged vegetation from water access points — had 124% higher intestinal schistosoma reinfection rates,” López-Carr said. In addition to lowered infection rates where they removed the vegetation, the researchers found that the removed material could be used to feed livestock, or turned into compost for growing crops, lowering costs dramatically and increasing yields for local farmers. In this way, according to López-Carr “the approach yielded an economic incentive to remove nuisance vegetation from waterways and return nutrients from aquatic plants back to the soil and for livestock feed with the promise of severing poverty-disease traps while lowering infectious burden at the same time.”

“A broader benefit is the hope that this example can set for enhancing win-win planetary health research and solutions that improve livelihoods while also reducing infectious morbidity and mortality,” he added.

Having conducted these trials, the researchers hope that this study is implemented elsewhere in other similar regions to replicate the same kind of health and economic outcomes.

And, it might not be just a solution for developing countries. “Perhaps vegetation growth resulting from excess nutrients could also be used as livestock feed in more developed countries as well,” López-Carr said.

Despite the ubiquity of dairy-based foods, many people have some form of lactose intolerance — up to 36% of Americans, according to the National Institutes of Health. As a result, many turn to lactose-free, plant-based alternatives, some of which have added health benefits. For example, one drink under development combines rice flour and coconut water: Rice is hypoallergenic and high in fiber, and coconut water is hydrating and low in calories. To understand how heat treatment might alter this beverage, Jorge Yán?ez-Fernández, Diana Castro-Rodríguez and colleagues wanted to test the formulation against two different high-temperature processing steps.

The team used three versions of the beverage, containing either 2%, 5% or 8% rice flour, with coconut water comprising the rest. These were heated either by pasteurization in a water bath at 140 degrees Fahrenheit or by sterilization in an autoclave at almost 250 degrees Fahrenheit. After these treatments, the team found that the starches in the rice flour gelatinized and underwent the Maillard reaction, producing a slightly darkened color and stickier fluid for all three versions. Additionally, the drinks’ acidities increased, and there were fewer sugars, which may alter the way they taste. The team plans to use these results to inform future research into similar, dairy-free, “functional beverages,” including those that could one day contain probiotic, lactic-acid bacteria.

In nature, poisonous species typically advertise their toxicity, often by producing high contrast colours such as black, white and yellow, like wasps and bees.

Along similar lines, New Zealand’s cyanide-producing stonefly, Austroperla cyrene, produces strong ‘warning’ colours of black, white and yellow, to highlight its threat to potential predators.

In a new study published in Molecular Ecology, University of Otago Department of Zoology researchers reveal that an unrelated, non-toxic species ‘cheats’ by mimicking the appearance of this insect.

Lead author Dr Brodie Foster says by closely resembling a poisonous species, the Zelandoperla fenestrata stonefly hopes to avoid falling victim to predators.

“In the wild, birds will struggle to notice the difference between the poisonous and non-poisonous species, and so will likely avoid both.

“To the untrained eye, the poisonous species and its mimics are almost impossible to distinguish,” he says

The researchers used genomic approaches to reveal a key genetic mutation in a colouration gene which distinguishes cheats and non-cheats.

This genetic variation allows the cheating species to use different strategies in different regions.

However, co-author Dr Graham McCulloch says the strategy, known as Batesian mimicry, doesn’t always succeed.

“Our findings indicate that a ‘cheating’ strategy doesn’t pay in regions where the poisonous species is rare,” he says.

Co-author Professor Jon Waters adds cheating can be a dangerous game.

“If the cheats start to outnumber the poisonous species, then predators will wake up to this very quickly — it’s a bit of a balancing act,” he says.

The Marsden-funded team is assessing how environmental change is driving rapid evolutionary shifts in New Zealand’s native species.

Data collected by remote-sensing satellites is fundamental for many key activities, including aerial mapping, weather prediction, and monitoring deforestation. Currently, most satellites can only passively collect data, since they are not equipped to make decisions or detect changes. Instead, data has to be relayed to Earth to be processed, which typically takes several hours or even days. This limits the ability to identify and respond to rapidly emerging events, such as a natural disaster.

To overcome these restrictions, a group of researchers led by DPhil student Vít Růžička (Department of Computer Science, University of Oxford), took on the challenge of training the first machine learning program in outer space. During 2022, the team successfully pitched their idea to the Dashing through the Stars mission, which had issued an open call for project proposals to be carried out on board the ION SCV004 satellite, launched in January 2022. During the autumn of 2022, the team uplinked the code for the program to the satellite already in orbit.

The researchers trained a simple model to detect changes in cloud cover from aerial images directly onboard the satellite, in contrast to training on the ground. The model was based on an approach called few-shot learning, which enables a model to learn the most important features to look for when it has only a few samples to train from. A key advantage is that the data can be compressed into smaller representations, making the model faster and more efficient.

Vít Růžička explained: ‘The model we developed, called RaVAEn, first compresses the large image files into vectors of 128 numbers. During the training phase, the model learns to keep only the informative values in this vector; the ones that relate to the change it is trying to detect (in this case, whether there is a cloud present or not). This results in extremely fast training due to having only a very small classification model to train.’

Whilst the first part of the model, to compress the newly-seen images, was trained on the ground, the second part (which decided whether the image contained clouds or not) was trained directly on the satellite.

Normally, developing a machine learning model would require several rounds of training, using the power of a cluster of linked computers. In contrast, the team’s tiny model completed the training phase (using over 1300 images) in around one and a half seconds.

When the team tested the model’s performance on novel data, it automatically detected whether a cloud was present or not in around a tenth of a second. This involved encoding and analysing a scene equivalent to an area of about 4.8×4.8 km² area (equivalent to almost 450 football pitches).

According to the researchers, the model could easily be adapted to carry out different tasks, and to use other forms of data. Vít Růžička added: ‘Having achieved this demonstration, we now intend to develop more advanced models that can automatically differentiate between changes of interest (for instance flooding, fires, and deforestation) and natural changes (such as natural changes in leaf colour across the seasons). Another aim is to develop models for more complex data, including images from hyperspectral satellites. This could allow, for instance, the detection of methane leaks, and would have key implications for combatting climate change.’

Performing machine learning in outer space could also help overcome the problem of on-board satellite sensors being affected by the harsh environmental conditions, so that they require regular calibration. Vít Růžička said: ‘Our proposed system could be used in constellations of non-homogeneous satellites, where reliable information from one satellite can be applied to train the rest of the constellation. This could be used, for instance, to recalibrate sensors that have degraded over time or experienced rapid changes in the environment.’

Professor Andrew Markham, who supervised Vít’s DPhil research, said ‘Machine learning has a huge potential for improving remote sensing — the ability to push as much intelligence as possible into satellites will make space-based sensing increasingly autonomous. This would help to overcome the issues with the inherent delays between acquisition and action by allowing the satellite to learn from data on board. Vít’s work serves as an interesting proof-of-principle.’