Could AI help save the Vaquita?
Author: Analisa Ruiz, Projects Intern at C Minds
Meet the vaquita. Easily identifiable by the dark patches around its eyes and lips, this marine mammal can be found swimming alone in the Sea of Cortés, located between the northeastern coast of Mexico’s mainland and the Baja California Peninsula. They tend to avoid ships and do not spend much time near the surface, making them difficult to observe. Because of their aversion to humans, it’s possible you have never even heard of them. But if you have, it’s probably because, as of 2006, the vaquita claimed the title as the most endangered cetacean in the world.
Thirteen years later the situation has only gotten worse. How did this happen? And can AI help?
Vaquitas were recognized as critically endangered in 1996, with a little over 600 remaining. This number has decreased rapidly since, and it is estimated that somewhere between 6 and 22 vaquita are left. Many blame this extreme population decline on the use of gillnets, which fishermen use to capture the totoaba fish, another endangered marine species. The vaquita gets entangled in the gillnet and is unable to surface for air. The use of these nets have led to the endangerment of both the vaquita and the totoaba, and have been illegal in Mexico since 2015.
This is where the narrative gets a little complicated. Many of the fishermen using these nets are dependent on fish like the totoaba for income. The bladder of the totoaba fish is considered a culinary delicacy abroad, making the fish extremely valuable in the international market. Mexico’s government has taken many steps to protect the vaquita, including offering financial support to fishermen in the area as a way to compensate for the ban on gillnets, but many continue to hunt totoaba fish. Non-profit organizations like the Sea Shepherd Conservation Society are fighting to protect these endangered species by patrolling the Sea of Cortés and removing gillnets.
The problems encountered in Mexico’s northeastern sea are not isolated. Around the world, the existence of tens of thousands of species are threatened by human activity. Artificial Intelligence (AI) has allowed us to explore new ways of protecting our ecosystem. Below are some examples of how AI is being used across the animal world. These examples are meant to serve as a conversation starter for how technology can be used in general conservation efforts and how these might apply to the vaquita. As non-experts in conservation matter, C Minds welcomes any professionals in the subject to reach out and help us learn more about how technology can be used to protect our species.
Wild Me, a non-profit organization based in Oregon, USA, uses AI to study animals and has partnered with Microsoft’s AI for Earth program to expand and improve their research. The organization uses AI to identify animals in a non-invasive way to determine location, date of sighting, migration patterns, and an animal’s social group. With this information, researchers are able to better understand the species we are trying to protect and provide better insight into how we can protect them. If we were able to use AI to obtain this data about the vaquita, scientists could be able to better inform the public on how many vaquitas truly remain and where they are migrating.
In the Pacific Ocean, researchers from the University of Toulon, France, are using AI to create a dataset of the “soundscape” of the ocean. Just as a landscape maps different geological features, a soundscape is meant to interpret the various sounds under the ocean. This information is particularly important for our understanding of marine mammals, who rely on clear communication channels to find each other and navigate their environment. When there are too many external noises blocking these channels, cetaceans have been observed to eat and mate less. By putting together a soundscape, scientists understand where these important communication channels are located and can suggest quiet zones -- areas free from human activity -- to policymakers that will help in the conservation of these animals. Using this same technology, scientists could locate the remaining vaquitas and create “quiet zones” for them to exist peacefully. Beyond technology, there are interesting best practices at a regulatory level that could help protect these mammals.
In addition to learning more about vaquitas, AI can also be used to predict hunting patterns of poachers. PAWS, a machine learning algorithm, was tested in Uganda and Malaysia to predict where poaching was likely to occur. The technology was based on data from past patrols by rangers and helped both countries protect their wildlife from harm. Similarly, if this technology could be applied in Mexico’s waters, the nation’s navy and vaquita conservation groups would be able to focus on areas predicted to experience totoaba fish poaching and prevent further harm to the sea animals in the region.
So what does this mean for the vaquita? In order to save the species, drastic action is required. Would harnessing the power of AI help accelerate a solution? Or are there other emerging solutions that could change the fate of the vaquita? One thing we do know is that the time to act is now. To take part in vaquita conservation efforts, check out the following sites all focused in vaquita preservation:
C Minds is dedicated to the responsible use of new technologies, both ecologically and socially. We recognize that many of the issues we face do not have easy solutions, but we work to create a world in which safety and fairness are guaranteed for all. We believe a better future is attainable by combining the potential of new technology with the power of collaboration, and we will continue to work with partners across the world to overcome challenges of all kinds.
Let’s come together this Earth Day to celebrate the milestones that have been achieved (see Giant panda no longer Endangered) and recognize the large amount of work ahead of us for a more just and sustainable world.