A groundbreaking international study using artificial intelligence has revealed that climate change is causing a significant shift in flowering times for plants around the globe. By analysing over eight million digitalised plant specimens accumulated over the last century, researchers have determined that flowering is now occurring an average of 2.5 days earlier or later each decade. This alarming trend threatens the essential symbiosis between plants and their pollinators, raising crucial questions about ecological stability in an era of rapid climatic transformation.
The Role of AI in Understanding Plant Phenology
Artificial intelligence has emerged as a powerful tool in ecological research, enabling scientists to sift through vast amounts of data with unprecedented efficiency. The recent study, part of Kew’s “State of the World’s Plants and Fungi” report, highlights the profound implications of climate change on plant phenology — the timing of flowering and fruiting. By employing AI to classify the presence or absence of flowers across an extensive array of specimens, researchers found not only the average shift but also significant regional variations, particularly pronounced in tropical regions.
The findings underscore an urgent need for conservation strategies that account for these shifting timelines. As flowering patterns change, pollinators may struggle to adapt, leading to cascading effects throughout ecosystems that rely on these vital interactions.
Digitisation: A Game Changer for Conservation Efforts
The digitisation of herbarium and fungarium collections is revolutionising how scientists approach biodiversity and conservation. Kew Gardens has successfully transformed 7.4 million specimens into digital records, providing a freely accessible resource for researchers worldwide. This ambitious project, supported by the Environment Department (Defra), involved extensive scanning and cataloguing efforts, allowing for the preservation and sharing of invaluable data that was once confined to physical archives.
Professor Alexandre Antonelli, Kew’s executive director of science, emphasized the transformative potential of these digital assets. “We can harness the information locked in many of these specimens that have been here for centuries,” he said, noting how this information can advance science and conservation globally. The digital platform not only facilitates the discovery of new species but also aids in tracking those that may have gone extinct or are at risk.
New Discoveries Amidst Ongoing Threats
Despite the hope offered by new technologies, the report paints a sobering picture of the current state of plant and fungal diversity. With only 17% of plant species and a mere 0.6% of fungi evaluated for extinction risk, the potential loss of biodiversity is alarming. An estimated 100,000 plant species and two million fungi remain undiscovered. In just the years 2024 and 2025, researchers described 4,600 new plant species and 7,800 new fungi, highlighting the urgent need for continued exploration and documentation.
Antonelli noted that these newly discovered species often face immediate threats, stating, “We are just scratching the surface,” and many newly identified plants and fungi are at risk “from day one.” The integration of AI and digitisation not only enhances our understanding of existing species but also bolsters efforts to identify climate-resilient alternatives to vital crops, such as coffee, amidst changing environmental conditions.
The Future of Biodiversity Research
The implications of this research extend far beyond mere academic interest. Digitised herbarium records have proven instrumental in advancing studies on flowering and fruiting patterns in ecologically sensitive areas like the tropics and the Arctic, where traditional fieldwork can be challenging. The ability to map species loss, safeguard protected areas, and enhance food security is more crucial than ever.
Moreover, scientists are employing digitised records and advanced statistical models to assess whether certain species are genuinely extinct or simply difficult to detect. By unlocking the “dark matter of fungi” — the vast number of fungal species that remain uncharacterised — researchers are also exploring high-quality genomes from specimens that date back as far as 180 years. This exploration could lead to innovative applications, from developing high-protein food alternatives to discovering fungi capable of breaking down plastic waste.
Why it Matters
As our planet faces unprecedented environmental challenges, understanding the intricate relationships between climate change and biodiversity is paramount. The insights gained from AI-driven research not only illuminate the extent of the crisis but also offer pathways for proactive conservation measures. By digitising historical collections and leveraging cutting-edge technology, we stand at a pivotal moment in the fight to preserve our planet’s rich tapestry of life. The time to act is now, for the future of our ecosystems and the myriad species that inhabit them hangs in the balance.