SeaSpotter app could revolutionise marine conservation

Sea lion
© Lou Sanson/WWF
Success Story

A new app is bringing marine conservation and the citizens of New Zealand together. The app has been developed by MAUI63 in collaboration with WWF-New Zealand, and with funding and support from other agencies including Christchurch City Council, Microsoft, DOC and MPI. SeaSpotter will empower New Zealanders to become citizen scientists in the fight to save our threatened marine mammals. 

Nearly half of the world’s cetaceans (whales, porpoises and dolphins) live in New Zealand’s waters. They, and other species, face serious threats from habitat degradation, climate change, pollution, and commercial fishing. In order to effectively protect these incredible species, we need to better understand where they are and how many are left. 

“One of the biggest challenges that scientists face is the lack of data available on many marine mammal species in New Zealand. We are losing biodiversity at an unprecedented rate and, what’s worse, we don’t even know by how much,” says Dr Krista van der Linde, WWF-New Zealand Marine Species Programme Manager.

More than 70% of marine mammals in New Zealand’s Department of Conservation (DOC) Threat Classification system are listed as Data deficient, Threatened or at Risk. Unfortunately, over 50% of these marine mammals are listed as Data Deficient which means we don’t know a lot about them including accurate population sizes. 

“If we want to protect marine mammals, we have to know how many there are, what areas they occupy, and if their populations are declining,” continues van der Linde.

Anyone who spots a marine mammal around our coastlines can upload the photos through the SeaSpotter app. This will help researchers track and gather information on our threatened marine mammal species and allow scientists to use artificial intelligence to mine these images for crucial data, including identifying and tracking individuals to offer new insights into their movements and population trends.

“There are millions of images taken every year of threatened marine mammals by scientists, tourism companies, members of the public and even tourists,” says van der Linde. “They contain a wealth of data that we can extract and analyse to help not only protect marine mammals in Aotearoa but also combat their extinction.”

Through a new scientific field, imageomics, researchers can use AI to extract biological information of these animals directly from the photos. 

AI algorithms can identify individuals using identifying features such as the shape of a whale’s fluke or the dorsal fin of a dolphin. When paired with where and when these images were taken, these photos can aid in conservation by providing population counts, birth and death dynamics, species range, social interactions and interactions with other species, including humans.

“We don’t have the ability to attach satellite tags or to genetically sample every marine mammal we come into contact with, but we do have the ability to take a photo of them. Then we can use AI techniques such as machine learning to analyse the images to provide much of the information we need and can potentially revolutionise marine conservation in the process,” van der Linde says. 

MAUI63 co-founder Tane van der Boon says the data from the project will be publicly available, “One of the unique aspects of this project is the data will be publicly accessible as a rich source of learning for curious scientific minds, innovators, conservations, industry, and policy makers.” 

“One of the unique aspects of this project is that data will be publicly available as a rich source of learning for curious scientific minds, innovators, conservationists, industry and policy makers,’’ says Tane van der Boon, from MAUI63.  “For the first time, this technology has the potential to provide detailed information about marine mammals. As we are storing the photographs of animals in the cloud, it will be available to train AI to identify many types of marine species. The science potential seems limitless to us.”

The successful use of citizen science data in the work, was the inspiration behind the SeaSpotter app. Citizens around Aotearoa sent their photographs and sightings information to which was used to identify unique leopard seals occupying our shores and gave enough information to reclassify them from Vagrant to Resident. 

“This is the main source of data scientists currently have on leopard seals in New Zealand. It was able to show these seals were no longer Vagrant. We knew that if this could be done for leopard seals, then it could also be done for other marine mammal species,” explains van der Linde.