‘Flash Droughts’ Coming on Faster, Global Study Shows

A map of drought conditions across the United States in mid-July 2012, the peak of a flash drought that decimated crops in the Midwest.
Credit: Richard Heim/ NCEI/NOAA

Just like flash floods, flash droughts come on fast — drying out soil in a matter of days to weeks. These events can wipe out crops and cause huge economic losses. And according to scientists, the speed at which they dry out the landscape has increased.

Researchers at The University of Texas at Austin, The Hong Kong Polytechnic University and Texas Tech University found that although the number of flash droughts has remained stable during the past two decades, more of them are coming on faster. Globally, the flash droughts that come on the fastest — sending areas into drought conditions within just five days — have increased by about 3%-19%. And in places that are especially prone to flash droughts — such as South Asia, Southeast Asia and central North America — that increase is about 22%-59%.

Rising global temperatures are probably behind the faster onset, said co-author and UT Jackson School Professor Zong-Liang Yang, who added that the study’s results underscore the importance of understanding flash droughts and preparing for their effects.

“Every year, we are seeing record-breaking warming episodes, and that is a good precursor to these flash droughts,” he said. “The hope and purpose [of this research] is to minimize the detrimental effects.”

Does this artificial intelligence think like a human?

MIT researchers developed a method that helps a user understand a machine-learning model’s reasoning, and how that reasoning compares to that of a human.
Credits: Christine Daniloff, MIT

In machine learning, understanding why a model makes certain decisions is often just as important as whether those decisions are correct. For instance, a machine-learning model might correctly predict that a skin lesion is cancerous, but it could have done so using an unrelated blip on a clinical photo.

While tools exist to help experts make sense of a model’s reasoning, often these methods only provide insights on one decision at a time, and each must be manually evaluated. Models are commonly trained using millions of data inputs, making it almost impossible for a human to evaluate enough decisions to identify patterns.

Now, researchers at MIT and IBM Research have created a method that enables a user to aggregate, sort, and rank these individual explanations to rapidly analyze a machine-learning model’s behavior. Their technique, called Shared Interest, incorporates quantifiable metrics that compare how well a model’s reasoning matches that of a human.

Shared Interest could help a user easily uncover concerning trends in a model’s decision-making — for example, perhaps the model often becomes confused by distracting, irrelevant features, like background objects in photos. Aggregating these insights could help the user quickly and quantitatively determine whether a model is trustworthy and ready to be deployed in a real-world situation.

Carbon flow through inland and coastal waterways, implications for climate

A recent study by an international team of scientists including Raymond Najjar, professor of oceanography at Penn State, found that the flows of carbon through the complex network of water bodies that connect land and ocean has often been overlooked and that ignoring these flows overestimates the carbon storage in terrestrial ecosystems and underestimates sedimentary and oceanic carbon storage.
Credit: Pixabay

Terrestrial and marine ecosystems have a powerful influence on the Earth’s climate by regulating the level of atmospheric carbon dioxide. A recent study found that the flows of carbon through the complex network of water bodies that connect land and ocean has often been overlooked and that ignoring these flows overestimates the carbon storage in terrestrial ecosystems and underestimates sedimentary and oceanic carbon storage.

Carbon storage by the ocean and by land is usually quantified separately and does not fully consider the land-to-ocean transport of carbon through inland waters, estuaries, tidal wetlands and continental shelf waters — referred to as the land-to-ocean aquatic continuum or LOAC. In a detailed analysis of the LOAC, researchers from Belgium, the United States and France provide a perspective on the global carbon cycle and identify key knowledge gaps that have significant implications for enforcing the carbon calculations that are part of international climate accords. They reported their findings in the journal Nature.

“There are scientists who develop terrestrial models and scientists who develop marine models, and the continuum between them is poorly represented in global earth system models,” said Raymond Najjar, professor of oceanography at Penn State and co-author on the paper. “There are different communities of scientists studying land and ocean. Our assessment indicates that more attention needs to be paid to the LOAC, and that our models need to better represent it.”

Smart but stressful

Intelligent personal assistants accompany people worldwide every day.
Credit: RUB, Kramer

Intelligent personal assistants make everyday work easier. If you use them intensively and over a longer period of time, they can also create stress.

Siri, delete the light! Alexa, what does the weather forecast say? Nowadays, so-called intelligent personal assistants such as loudspeakers with speech recognition are hard to imagine in our everyday life. But do they only have a positive effect on us? What does the long-term human-assistant relationship look like?? Prof. also asked himself these questions. Dr. Sascha Alavi, chair holder at the Sales Management Department of RUB, and his research colleagues Prof. Dr. Valéry Bezençon and Ertuğrul Uysal from the University of Neuchâtel. In their joint study, published in the Journal of the Academy of Marketing Science, economists show that personal assistance systems can also have damaging effects on users in the long term.

“Previous studies have primarily and exclusively dealt with the advantages of intelligent assistance systems, highlighted their benefits for the world of work, for companies, especially from a commercial point of view. We were also interested in the potentially damaging consequences for consumers,” reports Alavi. To this end, he carried out surveys with his colleagues with more than 1,000 users of intelligent language assistants as well as qualitative in-depth interviews with eleven users.

“Tree of life” could help slow climate change

Mauritia palm fruit are vital to the local economy
Credit: Dael Sassoon

Changing the way fruit is gathered from a “tree of life” could have hugely positive environmental and financial impacts in Amazonia, according to a new study.

An international research team, jointly led by the University and the Peruvian Amazon Research Institute (Instituto de Investigaciones de la Amazonia; IIAP) has shown for the first time the widespread harm caused in Peru by cutting down the palm tree Mauritia flexuosa in order to harvest its fruit.

The scientists examined where and why the trees were felled, producing detailed maps and analysis to reveal the extent of the environmental and economic damage caused by cutting down the palms.

Study lead author Gabriel Hidalgo, who conducted the research as a postgraduate student at Leeds’ School of Geography while based at IIAP, said: “Cutting down female palm trees to harvest the fruit has halved the total production of fruit of this palm that is available to local communities.

“This is a clear example of the impact of humans on natural resource levels, in an ecosystem that, on first look, appears undamaged.

“However, changing the way the fruit is harvested can increase both the number of fruit-bearing palms trees, and the value of these Amazonian peatland ecosystems to people.”

Scientists discover genetic variants that speed up and slow down brain aging

Researchers from a USC-led consortium have discovered 15 “hot spots” in the genome that either speed up brain aging or slow it down — a finding that could provide new drug targets to resist developmental delays, Alzheimer’s disease and other degenerative brain disorders.

The research appeared online Tuesday in Nature Neuroscience.

“The big game-changer here is discovering locations on the chromosome that speed up or slow down brain aging in worldwide populations. These can quickly become new drug targets,” said Paul Thompson of USC, a lead author on the study and the co-founder and director of the ENIGMA Consortium. “Through our AI4AD [Artificial Intelligence for Alzheimer’s Disease] initiative we even have a genome-guided drug repurposing program to target these and find new and existing drugs that help us age better.”

ENIGMA is working group based at USC that is exploring a vast trove of brain data and has published some of the largest-ever neuroimaging studies of schizophrenia, major depression, bipolar disorder, epilepsy, Parkinson’s disease, and even HIV infection.

To discover the hot spots, or genomic loci, more than 200 ENIGMA-member scientists from all over the world looked for people whose brains were scanned twice with MRI. The scans provided a measure of how fast their brains were gaining or losing tissue in regions that control memory, emotion and analytical thinking.

Achilles’ heel of dangerous hospital pathogen

A scanning electron micrograph (SEM) of a highly magnified cluster of Gram-negative, non-motile en:Acinetobacter baumannii bacteria; Mag - 13331x. 
Credit: Janice Carr

A team from Research Unit 2251 of the German Research Foundation led by Goethe University has shed light on the structure of an enzyme important in the metabolism of the pathogenic bacterium Acinetobacter baumannii. The enzyme “MtlD" is critical for the bacterium's synthesis of the sugar alcohol mannitol, with which it protects itself against water loss and desiccation in dry or salty environments such as blood or urine. Structural analysis has revealed weak spots where it might be possible to inhibit the enzyme and thus attack the pathogen.

Each year, over 670,000 people in Europe fall ill through pathogenic bacteria that are resistant to antibiotics, and 33,000 die of the diseases they cause. In 2017, the WHO named antibiotic resistance as one of the greatest threats to health worldwide. Especially feared are pathogens that are resistant to several antibiotics. Among them, Acinetobacter baumannii stands out, a bacterium with an extraordinarily pronounced ability to develop multi-resistance and, as a “hospital superbug", dangerous above all for immunosuppressed patients. Acinetobacter baumannii is highly resilient because it can remain infectious for a long time even in a dry environment and thus endure on the keyboards of medical devices or on ward telephones and lamps. This property also helps the microbe to survive on dry human skin or in body fluids such as blood and urine, which contain relatively high concentrations of salts and other solutes.

Finding Planets That Have No Star

 

Most planets orbit a star, but some planets can escape and “go rogue.” But how do astronomers study planets that wander the cold dark of interstellar space?

Join our host, Summer Ash of the National Radio Astronomy Observatory, as she talks about how radio astronomers' study rogue planets.

Source/Credit: National Radio Astronomy Observatory / Amy C. Oliver

ast040522_01

Parasites thrive if hosts survive

The Australian native social parasitic bee Inquilina and its host Exoneura
Credit: Flinders University

Like diseases affecting humans, parasites can wage a deadly evolutionary ‘arms race’ against their hosts. But can hosts and parasites upgrade their weapons at the same rate?

The parasite bee and host species have evolved in complementary ways.

This can be a very unequal battle for two reasons, Flinders University researchers say. If the parasite is too successful it will wipe out its host, and therefore lose its only means of surviving.

At the same time, evolutionary ‘wars’ between hosts and their parasites depend on their rates of evolution; we can think of that as their ability to ‘upgrade their weapons’, says Associate Professor Mike Schwarz, from the College of Science and Engineering.

The Flinders University study examined this conundrum by examining a native Australian social bee (Exoneura) and its social parasite, another bee (Inquilina).

“These parasitic species spend their entire life cycle within the nest of the host species and have extreme adaptations to social parasitism, they are not able to survive without their hosts,” says first author Dr Nahid Shokri-Bousjein in an article in Ecology and Evolution.

The ability of species to adapt to existential challenges depends on their ability to ‘discover’ new strategies via random mutations.

Squid recorded color-matching substrate for the first time

A species of oval squid (locally known as Shiro-ika) from Okinawa is being cultured at OIST’s Marine Science Station. This animal exhibited amazing camouflaging abilities never before recorded in squid. Credit: Ryuta Nakajima / OIST.

While octopus and cuttlefish are famous for their use of camouflage to match the color of the substrate, a third type of cephalopod—the squid—has never been reported displaying this ability. Now, in a study published in Scientific Reports, scientists from the Physics and Biology Unit at the Okinawa Institute of Science and Technology Graduate University (OIST) have shown that squid can and will camouflage to match a substrate as a way of avoiding predators. This work opens up research avenues on how squid see and perceive the world around them. Furthermore, it sheds light on their behavior, and thus could go on to inform conservation initiatives.

“Squid usually hover in the open ocean but we wanted to find out what happens when they move a bit closer to a coral reef or if they’re chased by a predator to the ocean floor,” explained one of the three first authors, Dr. Ryuta Nakajima, OIST visiting researcher. “If substrate is important for squid to avoid predation than that indicates that increases or decreases in squid populations are even more tied to the health of coral reef than we thought.”

Previous studies on cephalopod camouflage have mostly been conducted on cuttlefish and octopus. Squids, as an animal that tends to live in the open ocean, are notoriously hard to keep in captivity and so have been avoided for this kind of research. But, since 2017, the scientists in the OIST’s Physics and Biology Unit have been culturing a species of oval squid in captivity.