Image credit: Satellite image (c) 2020 Maxar Technologies
Satellite images processed with the help of computer algorithms devised at the University of Bath are a promising new tool for surveying endangered wildlife.
For the first time scientists (Olga Isupova, University of Bath and Isla Duporge, University of Oxford) have successfully used satellite cameras coupled with deep learning to count animals in complex geographical landscapes, enabling conservationists to take an important step forward in monitoring populations of endangered species. For this research, the satellites Worldview 3 and 4 used high-resolution imagery to capture African elephants moving through forests and grasslands. The automated system detected animals with the same accuracy as humans are able to achieve.
Dr Olga Isupova participated in creation of this collaborative algorithm from the University of Bath. Dr Isupova said the new surveying technique allows vast areas of land to be scanned in minutes, offering an efficient alternative to human observers counting individual animals from low-flying airplanes. A satellite can collect over 5,000 km² of imagery in a few minutes on a cloud-free day, and also eliminates the risk of double counting. Other benefits include removing the risk of disturbing animals during data collection, and ensuring humans are not hurt in the counting process, making data collection logistically easier, particularly in hard-to reach environments. It also makes it simpler to count animals moving from country to country, as satellites can orbit the planet without regard for border controls or conflict.
This work is vital, due to the rapid decline in the population of African elephants over the past century, mostly due to poaching and habitat fragmentation. There are approximately 415,000 African savannah elephants left in the wild, and the species is classified as endangered. “Accurate monitoring is essential if we’re to save the species,” said Dr Isupova. “We need to know where the animals are and how many there are.”
Whilst the study was not the first to use satellite imagery and algorithms to monitor species, it was the first to reliably monitor animals moving through a heterogeneous landscape – a backdrop that includes areas of open grassland, woodland and partial coverage. “This type of work has been done before with whales, but of course the ocean is all blue, so counting is a lot less challenging,” said Dr Isupova. “As you can imagine, a heterogeneous landscape makes it much hard to identify animals.”
The researchers believe their work demonstrates the potential of technology to support conservationists in their work to protect biodiversity and slow the progress of the sixth mass extinction.
“We need to find new state-of-the-art systems to help researchers gather the data they need to save species under threat,” said Dr Isupova. African elephants were chosen for this study for good reason – as well as their conservation status, they are the largest land animal and therefore the easiest to spot. However, Dr Isupova is hopeful that it will soon be possible to detect far smaller species from space. “Satellite imagery resolution increases every couple of years, and with every increase we will be able to see smaller things in greater detail,” she said, adding: “Other researchers have managed to detect black albatross nests against snow. No doubt the contrast of black and white made it easier, but that doesn’t change the fact that an albatross nest is one-eleventh the size of an elephant.”
The research has since been featured on the BBC website, Radio 2, Radio 4 and local BCC stations. The BBC article was then translated to other languages, for example, Persian https://www.bbc.com/persian/science-55747347 and Russian https://www.bbc.com/russian/news-55750006. Dr Isupova and Isla Duporge were then interviewed by Euronews.
As well as appearing on LBC for an interview which you can listen to via the link below, this work has inspired a poem, featured on the website The Poetry of Science .
Read the full paper
The paper Using very‐high‐resolution satellite imagery and deep learning to detect and count African elephants in heterogeneous landscape is published in Remote Sensing in Ecology and Conservation.
The researchers involved in this project were Dr Olga Isupova from the University of Bath, Isla Duporge, Dr Steven Reece, and Professor David W. Macdonald from the University of Oxford, and Dr Tiejun Wang from the University of Twente. The study was designed by Isla Duporge as part of her PhD, which is supervised by the Geospatial department at the University of Twente.