While forests account for over one-third of Earth’s land surface and are the homes to a biodiverse array of species, we’re still cutting down trees at a rapid rate. Technology itself can’t stop deforestation, but recently, we’ve seen the thesis that artificial intelligence and satellite imagery can play a pivotal role in tracking forest health.
To learn more about how the technology works, the process of forest health management under the hood, and how data science and artificial intelligence can play a role in better forest health tracking, I wanted to speak with someone directly working on this innovation.
That person was Anniek Schouten, the co-founder of Netherlands-based forest intelligence company 20tree.ai. Together, we discuss the challenges and limitations of forest health management through artificial intelligence and why the company is betting on technology disrupting the space.
The process and challenges of tracking forest health with software
While collecting data is a big part of managing forest health, the monitoring process is really just one-third of the process, which consist of mapping, monitoring, and predicting.
According to Schouten, these steps are crucial in ensuring long-term results. And by repeating this process over time, complex patterns can be extracted to “predict future scenarios of our natural resources more accurately,” she told me.
But the challenge of tracking and monitoring millions of hectares of land is that it can result in terabytes of unstructured data. Hence, it would be impossible for a human to make sense of this data. That’s where AI can help perform tasks like distinguishing between different tree species, Schouten shared.
Then, to keep an accurate, up-to-date database of our land, 20tree.ai uses what it calls planet intelligence. The process is comprised of four key steps.
Planet intelligence can make it easier for companies to stay accountable
At 20tree.ai, planet intelligence entails combining satellite imagery, AI, and computing to discover answers to crucial business and sustainability issues. Schouten told me that her company swears by four steps to create interactive and updated land analysis:
- Sourcing, structuring and preparing input data — including satellite and other climate data sources
- Processing the data with deep learning — resulting in a segmentation map of terrains
- Filtering out the relevant data to report
- Visualizing the filtered output to fit the customer process
Beyond having the potential to help businesses make greener decisions, planet intelligence can also play a role in regulation. With the technology, third-parties can make sure companies are keeping their word to sustainable promises.
The use cases of planet intelligence beyond forest health
For planet intelligence companies like 20tree.ai, leveraging technology to analyze land has helped monitor commitments to zero-deforestation and forest health. But the technology’s implications don’t stop there.
For instance, companies can verify vegetation maintenance completion. They can even identifying environmentally ethical areas to build parks.
The use cases are endless — and they can help everyone make sure we’re being accountable to our environmental impact.
Sonia John is a Writer at The Rising mainly covering the intersection of businesses and sustainability. You can pitch her stories at firstname.lastname@example.org.