fbpx Skip to content

How Oxford-based Dendra Systems is using AI and drones to combat deforestation?

Dendra Systems
Image credits: Dendra Systems

Oxford-based startup, Dendra Systems uses custom-built drones, ecology and Artificial Intelligence to help the biggest natural resources companies in the world clean up and replant degraded land. The company just launched its third-generation aerial seeding technology and analysis platform. This step takes its ‘sky tractor’ planting and mapping capability to a new level.

With the new technology, it lets companies rehabilitate degraded land 11 times faster and at a one-time cost as traditional methods of remediation. This means that the company can replant up to 60 Ha per day, which is an area equivalent to 85 football pitches.

In 2020, Dendra Systems raised $10M (nearly £7.2M) from leading investors to help continue its work given the global demand for solutions that can be applied at scale.

Puts an end to deforestation

It is reported that nearly 25% of the world’s landmass has been degraded by farming, mining, logging and other industrial processes. As a result, soil carbon and nitrous oxide are released into the atmosphere and make land degradation an important contributor to climate change. As per scientists, unsustainable agricultural practices leading to land degradation could strip almost 95% of the Earth’s land areas by 2050.

This is where Dendra Systems comes to play as it works with natural resources companies including Glencore, BHP and Rio Tinto, public agencies and governments in order to reverse the impact of industrial processes such as farming, mining and large road and rail building projects.

Deploys high-resolution imagery

Dendra Systems works all over the world on behalf of global natural resources companies to restore land. It claims that an upgrade to its technology will accelerate the rate at which it can regain natural woodland habitats for land that has lost all the flora and fauna.

With high-resolution imagery, the Oxford startup will be able to map the land equivalent to 400 football pitches a day per team. Also, the company can identify 120 species from the sky ranging from trees such as Brigalow acacia to perennial flowers like Ptilotus exaltatus. Dendra Systems can also identify animal species from lizard to kangaroo.

Using drones, Dendra helps conservation managers to see the land thousands of times more accurately than traditional satellite imagery. Its Artificial Intelligence technology helps in identifying and mapping all species on every square metre to enable accurate weed management and proactive stewardship.

With ecology-driven data science, conservationists can build an extremely detailed insight into every aspect of the land’s condition. After getting a full picture of the land, Dendra uses bespoke drones that can carry increased payloads for aerial seeding. This method increases the rate at which trees can be planted, and reduces the on-site risks for planters.

Founded in 2014 by Susan Graham, Dendra Systems claims that 10 drones could plant nearly 300,000 trees per day. On the other hand, up to 2,000 trees can be planted a day using the traditional hand planting method.

Susan Graham, founder of Dendra Systems, said: “Dendra Systems was founded to capture data about the land, turn it into insights about the ecosystem so that we can identify early risks like erosion or invasive species before they derail restoration work and then take action at scale. For the first time, Dendra’s Generation 3 technology will dramatically speed up and reduce the cost of the work ecologists and land managers.”

Lucy Roberts, Corporate Head HSEC and Human Rights, Glencore said: “Dendra helps Glencore overcome these challenges by providing unprecedented insights into the condition of the land and ecosystems using ecology-driven data science and artificial intelligence. Complemented by drone-based aerial seeding which increases the rate of planting and mitigates on-site risks. Together, Glencore and Dendra are restoring thriving ecosystems.”

Topics