The technology transforming Kew’s research

Using AI to speed up science

Artificial intelligence is currently the hot topic in technology, being incorporated into almost every aspect of life.

At Kew, it’s the type of AI known as machine learning that’s revolutionising the way that we analyse data and helping to protect potentially endangered species.

In our Herbarium, we’re using an algorithm that has been trained to identify plant species based on common features. Digitised images of collections are run through the algorithm, which provides an estimate on how likely it is a specimen has been incorrectly named.

It would be impossible for our taxonomists to double check each of our eight million specimens, but machine learning allows them to focus on the specimens most likely to be misidentified. 

We can also use similar algorithms to find out which plants species are under threat. Using data on the extinction risk factors of various palm species, researchers from Kew trained a number of algorithms to estimate which unassessed species are most under threat.

Combining the algorithm's results with recently published extinction risk assessments, our scientists were able to create an overview of the global risk to 75% of the world’s palm species. They found that over half of palms may be threatened; that's over 1,400 species.

Using this new information, our scientists discovered areas that were priorities for palm conservation, including Madagascar, Jamaica and Vietnam.

As we gather more data and refine the machine learning models, we’ll be able to understand even more about what’s most at risk and how we can act to best protect them.

Collecting plants with paramotors

Peru's coastal desert fog oases, better known as lomas, depend on sea fog as the only source of moisture. These unique ecosystems are incredibly fragile, and under threat from climate change and human impact.

To know what to protect, we have to know what’s there. But how can we collect plants from these remote regions when travelling by car, or even walking across the sand, can cause damage to the ecosystem?

That’s when we look to the skies.

Kew scientists have partnered with the French nonprofit Forest Air and the Brazilian paramotor team Aita (Escola Nacional De Paramotor) to learn more about the lomas without putting it at further risk.

The paramotorists were trained on how to identify target plant species, then how to collect, georeference and preserve them for future study.

Not only was paramotoring a more delicate way of collecting than using off-road vehicles, this new way of collecting reduced CO² emissions by up to two-thirds.

Shooting lasers at trees

Along with soil, trees and other plants are a vital store of carbon. But working out just how much carbon they store can be a time-consuming and damaging process. That’s why we’re utilising the power of LiDAR. 

LiDAR works by firing a laser at an object, then measuring the time it took for the reflected light to return to the receiver. This information allows us to create a highly detailed 3D model of a tree called a point cloud.

Then, using specialised software, we calculate the volume of the trunks and stems which in turn allows us to measure how much carbon is stored in the entire tree.

To learn this in the past, scientists would have to chop down a tree and weigh it. This new technique gets us the same information at no cost to nature.

We’ve used LiDAR both at Kew Gardens and Wakehurst to understand how much carbon is stored in our trees, and it’s being utilised in other biodiversity hotspots around the world.

Technology like LiDAR isn’t just used for scientific research though: it’s also the magic behind some incredible art. 

Of the Oak uses LiDAR scans, along with data from photogrammetry and CT scans of the soil to build a living image of the Lucombe Oak.

Discover the magic of trees and technology when you visit Kew Gardens this summer.