AI Is Unlocking Centuries of Climate Knowledge Hidden in Plant Collections

Artificial intelligence is helping scientists uncover new insights into how climate change 
is reshaping the natural world, using data collected long before the technology itself 
existed.

At the Royal Botanic Gardens, Kew, researchers are applying AI tools to millions of 
digitised plant and fungi specimens gathered over centuries. These collections, some 
dating back hundreds of years, provide a unique record of how species have responded 
to environmental change over time.

The work forms part of Kew's latest State of the World's Plants and Fungi report, which 
marks the publication's tenth anniversary and draws on contributions from more than 
400 scientists across 40 countries. The report paints a picture of both growing scientific 
capability and significant knowledge gaps in our understanding of the natural world.

Around eight million specimens have now been digitised, creating an extensive 
database that can be analysed far more quickly than would be possible through 
traditional methods alone. One recent study examined millions of plant records and 
found that flowering times are shifting by an average of 2.5 days per decade, providing 
further evidence of how climate change is altering ecosystems and natural cycles.
The report also highlights the scale of the challenge facing conservation. Fewer than 
one in five plant species have been formally assessed for extinction risk, while just 0.6% 
of known fungi species have undergone the same evaluation. Scientists estimate that
approximately 100,000 plant species and as many as two million fungi species remain 
undiscovered.

Despite centuries of scientific research, new discoveries continue at pace. In 2024 and 
2025 alone, researchers described around 4,600 new plant species and 7,800 new fungi 
species for the first time.

Against this backdrop, digitisation and artificial intelligence are becoming increasingly 
valuable tools for scientific research. By enabling vast collections of biodiversity data to 
be searched, analysed and compared at scale, they can help researchers identify 
patterns more quickly and improve understanding of how species are responding to 
environmental pressures.

Kew's work demonstrates that some of the most valuable insights into the future of 
biodiversity may already exist within historical collections. The challenge now is 
ensuring that this knowledge can be accessed, analysed and applied quickly enough to 
support conservation efforts in a rapidly changing world