A researcher in Switzerland has developed an intelligent algorithm that uses satellite and drone images of rainforests to predict where the next sites of deforestation will occur. He presented his research at the climate conference in Madrid.
Computer scientist David Dao is a doctoral student at the Swiss Federal Institute of Technology (ETH) in Zurich. He is a specialist in machine learning and develops intelligent algorithms that can autonomously analyze satellite and drone images. According to an ETH Zurich press release, these can even predict where the rainforest will recede in the near future.
The biggest challenge that Dao had to overcome is that computer algorithms are unable to discern what is forest coverage and what isn’t. While satellites and drones supply countless images of rainforests, they don’t contain any labels that would indicate what is a forest, river or road, for example.
According to Dao, the algorithms instead have to read sequences in order to recognize which areas are forested and whether these areas are shrinking. Sequences are individual images strung together in chronological succession.
When a new road is built through the rainforest, for example, numerous smaller roads form off it over time – and it is along these roads that the forest coverage is destroyed. So by comparing these chronologically sequential aerial views, algorithms can determine how road systems and forest coverage change over time, explains ETH Zurich.
Dao will test his development next year together with Chile’s forestry authority CONAF. His approach may even be able to detect not only overall decline in the rainforest, but also determine which species of trees are most affected.
According to ETH Zurich, the latter is a significant factor in climate change, because “different types of trees store CO2 at different rates, and one approach to forest conservation is to offer location populations financial incentives for retaining trees as CO2 storage rather than clearing the forest”.
David Dao presented his research project at the 25th UN Climate Change Conference in Madrid (COP25) last week.
Image credit: Neil Palmer (CIAT) via Flickr