AI could help win ‘race against extinction’ of vital plants, say botanists | Plants
The rise of artificial intelligence and digitalization could be a turning point in the “race against extinction” facing botanists trying to identify and save vital plants before they disappear, according to a landmark study. report From the Royal Botanic Gardens, Kew.
The new technology allows scientists to track how flowering times vary over weeks around the world, quickly identify new specimens, and even extract important genetic data from 180-year-old fungal samples, potentially opening a “genomic gold mine.” Digitization and online access to millions of examples, hitherto only accessible in archives, is also producing new insights, especially in the global south.
Plants and fungi underpin all life on Earth by providing food and medicine, storing carbon, and regulating the climate. However, approximately 40% of the 70,000 plant species evaluated are in danger of extinction, while another 330,000 plant species have yet to be analyzed. It is also believed that there are another 100,000 plant species that have yet to be named by scientists.
Around 2,000 new plant species are recorded each year but RBG Kew’s scientific director Prof Alexandre Antonelli said this was “only scratching the surface”.
This means potential new drugs and sustainable crops are becoming extinct before they are even discovered.
The situation is even worse for fungi; 90% of the estimated 2 million species are still unknown to science, and less than 1% of known species are considered at risk of extinction.
“While documenting and preserving all life on Earth continues to be formidable challenges, digitization and its accompanying technologies make me increasingly hopeful that we will succeed,” Antonelli said.
AI can learn how to identify challenging plants such as sedges and peat mosses whose distinguishing features are microscopic, meaning new or vulnerable species can be detected more quickly. “These AI models can now sometimes detect better than experts, which is incredibly exciting,” he said.
Digitizing images and collection data of plant and fungal specimens also accelerates international collaboration and could open up important but rarely accessed collections in biodiversity hotspots such as Madagascar.
Landy Rajaovelona, a senior botanist at Kew Madagascar, says: ‘Madagascar is one of the world’s most extraordinary biodiversity hotspots. By digitizing [37,000 physical specimens]“We have unlocked a wealth of knowledge spanning centuries and provided invaluable insight into today’s biodiversity.”
RBG Kew now Digitized all 7.4 million samplesThese, including those collected by Charles Darwin, are freely available online. The four-year program involved taking 20,000 high-resolution images per day at its peak. There is in total 145m digital samples It is now online globally, but this is less than 16% of the total held in herbariums, leaving “major blind spots in understanding”, scientists said.
The report also includes a global study that used an AI model trained to detect pox by analyzing 8 million digitized samples. It turned out that flowering shifted by an average of 2.5 days per decade in the last century due to the climate crisis. Changing rainfall patterns and rising temperatures mean some flowers arrive later and others arrive earlier.
This can seriously disrupt ancient relationships between plants and pollinators and other animals that depend on them at certain times of the year. For example, one study using herbarium specimens showed that: 80% of species trees in the Western Ghats region of IndiaThe tree, which is important for timber, was also blooming. By the 1990s, this rate dropped to less than half.
New technology also reveals the genetic secrets of fungi; scientists can now produce high-quality genomes from very old samples, some up to 180 years old. The breakthrough makes historical fungarium samples a “genomic goldmine” for new drugs and prediction of disease outbreaks, the researchers said. Penicillin and statins were obtained from fungi.
“Fungi are very opportunistic and love heat and moisture,” said Dr Esther Gaya, senior research leader at RGB Kew. “As the hot season gets longer in temperate regions, some human pathogens appear to be spreading from warmer places.”
There are concerns about the intensive energy and water use of AI data centers, which are acknowledged by the report’s authors. The Guardian reported in May that data centers currently consume 6% of electricity in the UK and US. OpenAI boss Sam Altman said in February: “Training a person also takes a lot of energy.”
The report, prepared by 400 scientists from 40 countries, warns that digitization and the use of artificial intelligence could increase existing biases and inequalities unless the underlying data is expanded and improved. It called for partnerships between technology companies and environmental organizations, and for governments and funders to invest in plant and fungal collections.
