Hexbyte Tech News Wired Thousands of Unstudied Plants May Be at Risk of Extinction

Hexbyte Tech News Wired

Pleurothallis portillae is one odd-looking orchid. Sporting a small nub of a flower nestled in a long, bulbous leaf that droops like a pair of string beans, it’s considered fashionably drab by collectors. But its true home is in the remote cloud forest of the Ecuadorian Andes—a region where, according to an algorithm, it’s most likely under threat of extinction.

Plants have long gotten short shrift in conservation circles. Although perhaps a fifth of the kingdom’s species are at risk, according to the UK’s Royal Botanic Gardens, Kew, identifying which ones are on the brink is a somewhat anecdotal business. Less than 10 percent of plant species have been assessed by the IUCN Red List, considered the preeminent global directory of extinction threat. Comprehensive evaluations, which take time and money, end up favoring so-called “charismatic” species, the lions and polar bears that grace glossy donation mailers. That, and the sheer number of known plant species—almost 400,000 of them, spread far across the globe in hard-to-reach places, with thousands more being discovered every year—makes the whole affair a massive, underfunded game of catch-up.

But botanists are drowning in data that could potentially help, says Anahí Espíndola, a professor of evolutionary ecology at the University of Maryland. “We wanted to find a way to speed up the process.” In a study appearing Monday in the Proceedings of the National Academy of Sciences, she and her co-authors use reams of data to predict the status of 150,000 plant species whose vulnerability is currently unknown.

Professors, curators, and citizen scientists have long gone out into the field in search of plants common and rare, returning with meticulous records of their observations that pile up in public databases. Data is available, to varying degrees, for hundreds of thousands of plants. In recent years, all that rough-and-tumble exploring has also generated millions of GPS points referring to locations where individual plants were observed. Espíndola’s team found that if they crunched the numbers available for plants already listed on the IUCN Red List—data on the species’ range, location, and traits, as well as regional climate and geographic indicators—they could build a machine learning model that could predict the status of other species.

The results indicated that between 8 and 30 percent of those unassessed plants were at risk, potentially tens of thousands of species. Even more concerning is the fact that conservation efforts might be overlooking many of those plants because of where they live. Plant protection programs tend to favor areas like Europe, where many research institutions happen to be located, or ecological marvels like Madagascar that attract hordes of botanists. Other regions, they found, like the fog-nurtured biome hugging the southern coast of the Arabian peninsula, harbored a large number of potentially threatened species that hadn’t received nearly as much attention.

IUCN ratings aren’t the end-all-be-all of conservation assessment; but they do have bearing on which areas are protected as biodiversity hotspots abroad, and are fed into databases that extractive industries use for reducing threats to endangered species as they work. “Not having plants in those analyses means that people are working with incomplete datasets,” explains Anne Frances, a botanist who coordinates Red List efforts in North America. “We’re determining key biodiversity areas without a big chunk of the biodiversity being taken into account.”

That’s especially troubling given the foundational role plants have in ecosystems. Studies have suggested plant species are less adept than animals at responding to changes in habitat and climate. And when they go extinct, their disappearance can cause cascading effects through broader ecological networks.

The challenge, Frances says, is one of time and funding. Much of the grunt work of assessing species is carried out by volunteer experts. And by necessity, current conservation strategies tend to focus on large pushes to study a single category of plant–the IUCN recently completed an inventory of the world’s cactuses, for example, and is currently working on trees–or focus on those with salient uses, like medicinal plants and the wild relatives of common food crops.

But Espíndola argues those strategies can end up shuttling resources away from plants most in need of conservation. Machine learning predictions aren’t a replacement for those on-the-ground assessments: Our humble orchid, for example, will require a closer look by botanists working in the Andes, who would tally up individuals, sum up local threats, and evaluate the genetic diversity of the wild populations. But they could be a starting point, a quick way for conservationists to identify plants that need more study. The models could also be taken up by regional conservation groups and fleshed out with more data, yielding crisper, more accurate local assessments.

Perhaps a similar model could be applied to other creatures, like fungi, kickstarting still-nascent efforts to catalogue an overlooked kingdom, says Espíndola. “There are other groups of organisms that are even less attractive than plants.”

More Great WIRED Stories

Read More

Why thousands of AI researchers are boycotting the new Nature journal

Why thousands of AI researchers are boycotting the new Nature journal

Budding authors face a minefield when it comes to publishing their work. For a large fee, as much as $3,000, they can make their work available to anyone who wants to read it. Or they can avoid the fee and have readers pay the publisher instead. Often it is libraries that foot this bill through expensive annual subscriptions. This is not the lot of wannabe fiction writers, it’s the business of academic publishing.

More than 200 years ago, Giuseppe Piazzi, an isolated astronomer in Palermo, Sicily, discovered a dwarf planet. For him, publishing meant writing a letter to his friend Franz von Zach. Each month von Zach collated letters from astronomers across Europe and redistributed them. No internet for these guys: they found out about the latest discoveries from leatherbound volumes of letters called Monatliche Correspondenz. The time it took to disseminate research threw up its own problems: by the time Piazzi’s data were published, the planet had vanished in the sun’s glare.

It was a 23-year-old reader in Göttingen who saved the day. Using Kepler’s laws of planetary motion, Carl Friedrich Gauss calculated the location of what we know today as Ceres. Gauss, who became Germany’s greatest mathematician, and Piazzi shared their learnings freely, but they accepted the need to pay for the work that von Zach undertook. This is the closed-access publishing model.

In my own field of machine learning, itself an academic descendant of Gauss’s pioneering work, modern data are no longer just planetary observations but medical images, spoken language, internet documents and more. The results are medical diagnoses, recommender systems, and whether driverless cars see stop signs or not. Machine learning is the field that underpins the current revolution in artificial intelligence.

Machine learning is a young and technologically astute field. It does not have the historical traditions of other fields and its academics have seen no need for the closed-access publishing model. The community itself created, collated, and reviewed the research it carried out. We used the internet to create new journals that were freely available and made no charge to authors. The era of subscriptions and leatherbound volumes seemed to be behind us.

The public already pays taxes that fund our research. Why should people have to pay again to read the results? Colleagues in less well-funded universities also benefit. Makerere University in Kampala, Uganda, has as much access to the leading machine-learning research as Harvard or MIT. The ability to pay no longer determines the ability to play.

Machine learning has demonstrated that an academic field can not only survive, but thrive, without the involvement of commercial publishers. But this has not stopped traditional publishers from entering the market. Our success has caught their attention. Most recently, the publishing conglomerate Springer Nature announced a new journal targeted at the community called Nature Machine Intelligence. The publisher now has 53 journals that bear the Nature name.

Should we be concerned? What would drive authors and readers towards a for-profit subscription journal when we already have an open model for sharing our ideas? Academic publishers have one card left to play: their brand. The diversity and quantity of academic research means that it is difficult for a researcher in one field to rate the work in another. Sometimes a journal’s brand is used as a proxy for quality. When academics look for promotion, having papers in a “brand-name journal” can be a big help. Nature is the Rolex of academic publishing. But in contrast to Rolex, whose staff are responsible for the innovation in its watches, Nature relies on academics to provide its content. We are the watchmakers, they are merely the distributors.

Many in our research community see the Nature brand as a poor proxy for academic quality. We resist the intrusion of for-profit publishing into our field. As a result, at the time of writing, more than 3,000 researchers, including many leading names in the field from both industry and academia, have signed a statement refusing to submit, review or edit for this new journal. We see no role for closed access or author-fee publication in the future of machine-learning research. We believe the adoption of this new journal as an outlet of record for the machine-learning community would be a retrograde step.

Neil Lawrence is on leave of absence from the University of Sheffield and is working at Amazon. He is the founding editor of the freely available journal Proceedings of Machine Learning Research, which has to date published nearly 4,000 papers. The ideas in this article represent his personal opinion.

Read More